Extraction of Thorium and Uranium from Chloride Solutions by Tri-N-Butyl Phosphate and Tri-N-Octyl Phosphine Oxide

Journal of NUCLEAR SCIENCE and TECHNOLOGY, 1, No.5, p.155~462 (1964). 155

Extraction of Thorium and Uranium from Chloride Solutions by Tri-n-Butyl Phosphate and Tri-n-Octyl Phosphine Oxide

Kenju WATANABE*

Received November 4, 1963

The extraction of thorium and uranium chlorides by TBP and TOPO was studied. The composition of complexes extracted from the chloride solutions of low acid concentration was established by partition study to be UO2Cl2(TOPO)2,UCl4(TOPO)2, ThCl4(TOPO)2 and UCl4(TBP)2. Composition of the thorium complex in the TBP phase free from hydrochloric acid was revealed by infrared study to be ThCl4(TBP)4. The extraction behavior of thorium chloride by TBP was different from that of U(IV) and Pu(IV) chloride, and the composition of the complex was presumed to be HThCl5(TBP)4 in the extraction from concentrated chloride solution containing hydrochloric acid.

The extraction of thorium, uranium and drochloric acid. The extraction of Pu by TBP plutonium chlorides by tri-n-butyl phosphate from hydrochloric acid is also discussed. (TBP) or tri-n-octyl phosphine oxide (TOPO) has been studied by many workers. The EXPERIMENTAL distribution of uranyl chloride between TBP and hydrochloric acid was reported by Reilly(1), General techniques used were similar to Irving(2), Larsen(3) and Ishimori(4). Naito pro- those in previous studies(4)(10).Measurements posed the composition of the extracted complex were made at a constant temperature of 25dc. to be UO2Cl2(TBP)2(5). According to Iwase(6), Reagents: TBP was purified by the usual in the extraction of uranium tetrachloride method(10)~(13).TOPO was obtained from East- from hydrochloric acid solution by TBP, the man Organic Chemicals and Dojindo & Co., composition of the species extracted is UCl4• Ltd. The TOPO was purified to remove 3TBP. The extraction behavior of thorium acidic impurities: A xylene solution of TOPO chloride between the aqueous and TBP phases was scrubbed three times with 5 % sodium was investigated by Peppard(7), but the extrac- bicarbonate solution, twice with water, twice tion mechanism still remains uncertain. The with 1M hydrochloric acid and three times extraction behavior of Th between TOPO with water, all in succession. Other reagents and hydrochloric acid solutions was studied used were of A. R. grade. by Ross(8), who determined the extracted Radioactive tracers: The extraction of U species to be ThCl4¥HCl¥3TOPO. He also and Th was studied by the use of 237Uand studied the extraction of several metal ions 231Th prepared by g-ray irradiation of natural including uranyl chloride by TOPO(9). In most U and Th with a linear accelerator(14). of the studies mentioned above, excepting the Thorium-234 milked from 238U(15)was also used one by Naito, the metal ions were extracted as thorium tracer. Plutonium-239 was pre- from fairly concentrated acidic solutions, where pared by neutron irradiation of U in JRR- competitive extraction of the metal chloride l(16)(17). and hydrochloric acid was observed. Uranium(IV) was prepared by reduction The present paper endeavors to elucidate with zinc metal; and Pu(IV) by the addition the extraction mechanism by comparing the of 0.2 M ammonium nitrite. extraction of thorium and uranium chlorides by TBP and TOPO from chloride solutions * Japan Atomic Energy Research Institute, Tokai- with that from solutions containing only hy- mura, Ibaraki Pref.

9 156 J. Nucl. Sci. Tech.

Infrared measurement: A Koken DS Type 301 spectrometer equipped with a rock salt prism RESULTS AND DISCUSSION was used in the infrared measurements. The techniques were essentially the same as de- 1. TOPO-Chloride Solution System scribed in the reports(5)(18) by Naito, et al. The measurements of the distribution ratio Determination of thorium in TBP phase: In Kd for Th, U(IV) and (VI) were made by the the infrared study, the concentration of Th extraction from solutions in various concen- in TBP phase was determined as follows: trations of both hydrochloric acid and chloride. Thorium in TBP solution of a known volume The chloride solutions was added with 0.1 M was stripped into aqueous solution and precip- hydrochloric acid to prevent hydrolysis. In itated with oxalic acid. Then the Th oxalate these cases, the concentration of metal ions in was ignited to oxide, and weighed. the aqueous phase was maintained at less than

Distribution of U(IV) Distribution of Th between 1% between 1% TOPO- Distribution of U(VI) between TOPO-toluene and Chloride toluene and Chloride 1% TOPO-toluene and Chloride Solutions Solutions Solutions Fig.1 Fig.2 Fig.3

10-4 M. The results obtained are shown in As is shown in Figs. 1, 2 and 3, the Kd Figs.1, 2 and 3. values for Th,U(IV) and (VI) in the extraction In Fig.1, the dotted lines show the results from solutions of low hydrochloric acid con- obtained by the use of TOPO as received, while centration are approximately the same as with TOPO scrubbed with sodium bicarbonate those from the corresponding chloride solutions solution the results are as indicated by the containing 0.1 M hydrochloric acid. In the solid lines. It is likely that the TOPO as region of more concentrated solutions, how- received contains some impurities of acid form ever, the Kd values in the hydrochloric acid seeing that Kd values increase again with system show a maximum, whereas those decreasing acidity in the region of hydrochloric in the chloride solution system do not and acid concentration less than 1 M, while with continue to increase evenly with the chloride TOPO scrubbed with alkali solution Kd con- concentration. The decrease of Kd values in tinues to decrease down to lowest concentra- the region of high acid concentration may be tions. Therefore, in this work, the purified explained as competitive extraction of the TOPO was used to avoid the influence of acidic metal ions and hydrochloric acid, as pointed impurity. out by some investigators(5)(19)~(21). According

10 Vol.1, No.5 (1964) 157 to Naito(5), Kd values for metal ions of same because the complex concentration in the valency extracted in the same extraction organic phase is low. If the measurement is system should have a maximum at nearly the made in the region where little variation of same acid concentration. In fact, in the (fH+) is expected, one can determine ap- extraction by TOPO of thorium and uranium proximately the value of (n-m) from the slope chlorides, both tetravalent, the Kd values for of the curve relating logKd(M) to logCH+. these metal ions assume maximum values at Such curves are reproduced in Fig.4, for the about the same hydrochloric acid concentration case where sodium chloride was added to of 7.5~8 M (Figs.1 and 2), maintain the constant chloride concentration. Mechanism of extraction: In the present From this figure, it is concluded that the systems, the extraction mechanism of metal values of (n-m) are zero for Th, U(IV) and chloride may be expressed by the equation (VI), thus establishing that the complexes of TOPO and these metal ions do not contain hydrogen ion. (1) where M+m indicates a metal ion and S an extracting reagent, while n and m are integers. When the solute concentration in the organic phase is low and it is assumed that the all metal ions in the organic phase contribute to the formation of the complex, the following equation for the distribution ratio, Kd(M) can be derived:

(2) where f and C respectively are the activity coefficient and equilibrium concentration of the corresponding term indicated by the suffix, Variation of Kd Values for Th, U(IV) and Cd0 is the initial concentration of the solvent and (VI) with Acid Concentration Aq. soln. NaCl+HCl, [Cl-]=5M in the organic phase. The extraction mecha- Fig.4 nism can be determined by finding the values of n, (n-m) and m. Since the presence Solvent dependence: When the Kd(M) is of hydrochloric acid causes competitive ex- measured while keeping the concentration of traction of metal ion and acid which com- chloride and hydrochloric acid constant, Eq. plicates the analysis of extraction mechanism, (2) becomes the concentration of acid must be kept as low Kd(M)=const.,(f0d)m,(C0d)m.(4) as possible to simplify the analysis. On the If the measurement is made at low solvent other hand, it is known that hydrolysis of concentration, the value of m is obtained by Th begins from about pH2(22)~(24). Bearing plotting logKd(M) vs. logCd0since the variation these facts in mind, an acidity of 0.1 M was of fd0 is small under this condition. The chosen as experimental condition for the solvent dependence curves for Th, U(IV) and chloride solution system. The amount of (VI) are shown in Fig.5. From the slopes of hydrochloric acid extracted into TOPO phase these lines the value of m is determined to be is very small under this condition(25). 2 for all these metal chlorides. Acid dependence: If the Kd is measured at Chloride dependence: In a similar way, one constant concentrations of the TOPO and chlo- can determine the chloride dependence n by ride, Eq.(2) becomes measuring Kd(M) under constant concentration Kd(M)=const.,(fH)n-m,(CH+)n-m(3) of solvent and hydrochloric acid, and varying where it is assumed that (fcomplex) is constant the concentration of chloride. In this case,

11 158 J. Nucl. Sci. Tech.

U(IV) and (VI) by TOPO are ThCl4(TOPO)2, UCl4 (TOPO)2 and UO2Cl2(TOPO)2. These are for the species extracted in the region of low acid concentration. While Ross(8) reported that the composition of the species of ThCl4 extracted by TOPO was ThCl4,HCl,(TOPO)3, it is recalled that this was done in fairly high acid solution such as 7 M. 2. TBP-Chloride Solution System The distribution ratios for Th, U(IV) and Pu(IV) in the extraction from hydrochloric acid solutions by TBP are shown in Fig.6.

Variation of Kd Values for Th, U(IV) and (VI) with TOPO Concentration Fig.5 Eq.(2) becomes Kd(M)=const.,(fCl-)n,(CCl-)n. (5) It is seen from Fig.3 that in the region of chloride concentration less than 1 M, Kd values for U(VI) vary at a rate of about the second power of chloride concentration. This suggests that the value of n is 2 for uranyl chloride. Precise Kd values for Th and U(IV) were unfortunately not obtained in the region of chloride concentration less than 1 M, where the variation of (fCl-) is expected to be small, because Kd was very low in this region and the values for tetravalent metal ions were Distribution of Th, U(IV) and Pu(IV) seriously influenced even by very small between TBP and HCl or LiCl Solutions amounts of acidic impurity. Thus the chloride Fig.6 dependence for the extraction of Th and These metal ions resemble each other in the U(IV) were not determined directly by the shape of the curve of logKd(M) vs, logCHCl, method. However, since it was already known except that the Kd value for Th does not that (n-m) was equal to zero, the value of have a maximum. The Kd values of these n could be assumed to be 4 for tetravalent metals increase with atomic number. The metal ion in so far as Eq.(1) represented values for U(IV) and Pu(IV) have maxima at the extraction mechanisms of Th and U(IV) about 9~10 M hydrochloric acid concentra- in TOPO-chloride system. tion. The values for Th on the other hand Composition of complexes: It is established have no maximum value in the region by these analyses that the composition of the studied, and those in the system of 20 % TBP complexes formed in the extraction of Th, and hydrochloric acid solutions scarcely differ

12 Vol. 1, No.5 (1964) 159 from those in the system of undiluted TBP chloride solution. In the latter three cases, and lithium chloride solutions containing 0.1 M the Kd values in the extraction from 10 M hydrochloric acid. The dissimilar behavior of chloride solutions decrease gradually with the Th to that of tetravalent U and Pu suggests increase in hydrochloric acid concentration. that the extraction mechanism of Th might An acid dependence curve obtained for Th be different from that of tetravalent U or Pu. in 5 M aqueous chloride concentration is less In fact, the composition of the complex of Th accurate because the values are too low. Thus, extracted into TBP from concentrated chloride it remains uncertain whether the formation solution containing hydrochloric acid differs of thorium chloride-TBP complex is acid depen- from that of U(IV) as described below. dent in moderate chloride concentrations. Mechanism of extraction: In a way similar Solvent dependence: The effect of TBP con- to the case of TOPO extraction, n-m and m centration on the Kd values of Th, U(IV) and were determined from the slopes of the acid Pu(IV) chlorides at constant acid concentration dependence and solvent dependence curves. in the aqueous phase are shown in Fig.8. Acid dependence: Acid dependence curves of Th and U(IV) are shown in Fig.7, together with those for the extraction by TOPO from 10 M chloride solutions. Since the Kd values for U(IV) in the extraction from 5M chloride solutions are nearly constant, it is concluded that n=0 for U(IV). On the other hand, the value of logKd for Th in the extraction from 10 or 8 M chloride solutions is in proportion to the value of logCH+. The solid lines show the results obtained by the use of 234Th trecer, while the broken line shows those with 1.5x10-3 M thorium tetrachloride solution containing 234Th tracer . Macro amounts of Th were added to avoid the formation of radiocolloidal Th in the aqueous phase. The behavior of Th in the system of TBP and chloride solution ap- parently differs from that of U(IV) in the same extraction system, as well as from those of Variation of Distribution Ratio of Th, U(IV) Th and U(IV) in the system of TOPO and and Pu(IV) with TBP Concentration Fig.8. The Kd values for U(IV) and Pu(IV) vary approximately with the second power of TBP concentration, whereas those of Th vary with the third power in the region of low concentration of TBP, and the slope of this solvent dependence curve decreases with TBP concentration in the more concentrated region. In Fig.9 are shown solvent dependence curves of Th and U(IV) in the system of TBP and chloride solution. The value of m for U(IV) is established to be 2. On the other hand, in the case of Th, the values of the slope of the Variation of Kd Values for Th and solvent dependence curves vary between 3 U(IV) in Extraction by TBP and TOPO and 4 according to the variation of acid con- Fig.7 centration of the aqueous phase. The broken

13 160 J. Nucl. Sci. Tech.

line in this figure represents the results ob- for Th increases with decreasing acid concen- tained by the use of carbon tetrachloride as tration in the aqueous phase and approaches 4 diluent. Carbon tetrachloride is a more reliable in the case of aqueous solution containing 0.1 M diluent for determining the value of the slope of hydrochloric acid. solvent dependence curve than other diluents To determine the value of m directly, in- in the region of TBP concentration above 10~ frared examination of this system was also 20%(5). As is shown in Figs.8 and 9, the value tried. In general, the absorption band due to of the slope of the solvent dependence curves the P=0 group at 1,280cm-1 is shifted to the longer wavelength by the formation of a complex with inorganic salt. The intensity of the band due to free TBP decreases with increase in the Th concentration in the TBP phase. Since the decrease of the concentration of free TBP is related to the composition of the complex formed, the composition of the complex can be determined by plotting the decrease of the concentration of free TBP against the increase of the concentration of Th in the TBP phase(5). The infrared spectra obtained for the ThCl4-TBP system are shown in Fig. 10. The samples were prepared by dissolving thorium tetrachloride crystals in the TBP phase to concentrations of 0.074, 0.111, 0.147, 0.221, 0.295 and 0.442 M. The spectra were obtained by capillary samples and their thickness was adjusted so as to give a constant intensity in Variation of Kd Values for Th and the absorption band due to CH3 and CH2 U(IV) with TBP Concentration deformation vibration at 1,470cm-1. The de- Fig.9 crease in the concentration of free TBP with

Fig.10 Infrared Spectra of TBP-ThCl4 System Vol.1, No.5 (1964) 161

the increase in concentration of thorium chloride solution containing a small amount of hydro- in the organic phase is shown in Fig. 11. The chloric acid is determined to be 4 by consider- initial slope is 4.0, so the solvation number m ing the solvation number in the infrared study can be determined in this case to be m=4. and from the fact that the slope of the solvent Figure 12 represents the change of the intensity dependence curves of Th approaches 4 with of absorption due to the formation of the decrease of acid concentration. complex at 1,205cm-1. It is observed that Chloride dependence and the composition of intensity of this absorption band increases in complexes: The value of n, i.e. chloride de- proportion to the concentration of Th in the pendence, was not directly determined for the organic phase, except the value for 0.442 M. same reasons as in the case of TOPO extraction. However, the value of n may be determined indirectly from the values of m and (n-m) in so far as Eq.(1) can be applied to this system. Accordingly, as it was earlier established that n-m=0 and m=4, n is 4 for U(IV), and the composition of the complex formed in the extraction by TBP from chloride solution is determined to be UCl4(TBP)2. This does not agree with the results obtained by Iwase(6), who reported the composition of the complex to be UCl4¥(TBP)3. The results by Iwase were obtained from the solubility of UCl4 into TBP phase from 6 M hydrochloric Variation of Intensity of Absorption due to Free TBP acid solution. In the case of the extraction with Th Concentration of Th by TBP, it was observed from partition Fig.11 study (Fig.7) that the Kd values vary with the first power of the acid concentration in the aqueous phase. Accordingly, it is likely that the composition of the Th complex extracted by TBP from the concentrated chloride solution is HThCl5(TBP)4. Although details of the extraction mechanism of Th in concentrated chloride solution was not clarified by the present work, it is likely that, in the extraction by TBP from concentrated chloride solution, the hydrogen ion contributes to the formation of a Th complex, and that the extraction mechanism of Th is more compli- Variation of Intensity of Absorption cated than that of U. due to the Complex Formation of Thorium Chloride-TBP ACKNOWLEDGEMENT Fig.12 The author wishes to express his gratitude to Dr.T. Ishimori and Dr.K. Naito for their The results obtained by this infrared study invaluable discussion and helpful advice. He is indicate that the composition of the complex indebted to Mrs.E. Akatsu for the prepara- of thorium chloride dissolved in TBP phase is tion of 231Th and 237U tracers and to Miss T. ThCl4(TBP)4. However, it should be noted that, Izumi for her help during the experiment. in this system, no hydrochloric acid or any chloride other than thorium chloride is present. REFERENCES The value of A for the complex of Th (1) REILLY,E.: ANL-5254, (1954). extracted into TBP phase from a chloride (2) IRVING,H., EDGINGTON,D.N.: J.Inorg. Nucl.

15 162 J. Nucl. Sci. Tech.

Chem., 10, 306 (1959). NAKAMURA,E., KUROYANAGI,T.,ISHIMORI,T.: (3) LARSEN,R.P., SEILS,C.A.: Anal. Chem., 32, to be published. 1863 (1960). ) DYRSSEN,D.:(15 Svensk. Kern. Tid., 62, 153 (1950). (4) Islip/tom,T., WATANABE,K., NAKAMURA,E.: (16) KIMURA,K., ISHIMORI,T., NAITO,K., et al.: J. Bull. Chem. Soc. Japan, 33, 637 (1960). At. Energy Soc. Japan, 2, 328 (1960). (5) NAITO,K, Suami,T.: J. Phys. Chem., 66, (17) WATANABE,K.: ibid., 3, 497 (1961). 989 (1962). (18) NUKADA,K., NAITO, K., MAEDA,U.: Bull. Chem. (6) IwAsE,E., IsoNo, T.: Sci. Pap. I.P.C.R., 53, 13 Soc. Japan, 33, 894 (1960). (1959). (19) GLUECKAUF,E.: Belgisshe Chem. Ind., 23, 1215 (7) PEPPARD,D.F., MASON,G.W., MAIER,J.L.: J. (1957). Inorg, Nucl. Chem., 3, 215 (1956). KAPLAN,L., HILDEBRAND,R.A., ADER,M.: (8) Ross, W.J., WHITE, J.C.: ORNL-2627, (1958). ANL-4521, (1950). (9) idem. ORNL-CF-56-9-18, (1956). (21) MCKAY,H.A.C., ALCOCK,K., SCARGILL,D.: (10) ISHIMORI,T,, UMEZAWA,H., WATANABE,K.: J. AERE-C/R-1514, (1954): J.Inorg. Nucl. Chem., At. Energy Soc. Japan, 1, 299 (1959). 4, 304 (1957). (11) ALCOCK,K., GRIMLEY,S.S., HEALY,T.V., et al.: (22) HIETANAN,S.: Acta Chem. Scand., 8, 1626 Trans. Fraday Soc., 52, 39 (1956). (1954). (12) PEPPARD,D.F., MASON,G.W., MAIER,J.L.: J. (23) KRAUS,K.A.: A/Conf., P/731, 7, 245 (1955). Inorg. Nucl. Chem., 3, 215 (1956). (24) KRAUS,K.A., HOLMBERG,R.W.: J.Phys.Chem., PEPPARD,D.F., DRISCOLL,W.J., SIRONEN,R.J., 58, 325 (1954). et al.: ibid., 4, 326 (1957). (25) ISHIMORI,T., KIMURA,K., FUJINO,T., et al.: J, (14) NAKAMURA,E.:Radiochimica Acta, 2, 66 (1963); At. Energy Soc. Japan, 4, 117 (1962).