I OFL. ORNLP 2354 { TEESEEN 3.6 KO .25 1.1.4 16 MICROCOPY RESOLUTION TEST CHART NATIONAL BUREAU OF STANDARDS - 1963 ORN-P, 2384 Conf-660805-3 SEP 22 1956 ORNL - AEC - OFFICIAL MASS ACTION AND NON-IDEALITY IN EXTRACTION BY AMINE SULFATE CFSTI PRICES C. F. Coleman and J. W. Roddy : HC 1/.003 MN_5B Oak Ridge National Laboratory Oak Ridge, Tennessee For presentation at the INTER IATIONAL CONFERENCE ON SOLVENT EXTRACTION CHEMISTRY August 27 - September 1, 1966, Göteborg, Sweden and publication in the Proceedings RELEASED FOR ANNOUNCEMENT IN NUCLEAR SCIENCE ABSTRACTS LEGAL NOTICE The report no prepared as an account of Governmedi sponsored work. Neither the United Suites, por the Commission, nor any person acung on behalf of the Commission: A. Makes any warranty or representation, expressed or implied, with respect to the accu- racy, completeness, or usefulness of the Information ccolained in thiu report, or what the use of any information, apparatus, method, or process discloseu In this report may dot Infringe printely owned rights; or B. Assumes any liabilities with respect to the use of, or for damages resulting from the ure of any informatic.), apparitis, method, or process disclosed in this report. As used in the above, "person acting on behalf of the Commission" includes nay em- plogue or coatraclor of the Commission, or employce of such contractor, to the extent that such employee or contraclor of the Commission, or employee of such contractor prepares, disseminates, or provid: access to, any information pursuant to die employment or contract with the Commission, or do employment with such contractor. Winn.now noiva Research sponsored by the U.S. Atomic Energy Commission under contract with the Union Carbide Corporation. ORNL - AEC - OFFICIAL A - WISHIn-now- MASS ACTION AND NON-IDEALITY IN EXTRACTION BY AMINE SULFATE ORNI - AEC - OFFICIAL C. F. Coleman and J. W. Roddy Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA r-nos!!!! "Research sponsored by the U.S. Atomic Energy Comission under contract with · the Union Carbide Corporation. ABSTRACT Distribution cquilibria involving alkylammonium extractants (e.g., acid and metal ion extractions, amine salt distributions) often appear to deviate significantly from the Mass Action Law, especially in some sulfate systems. The explanation originally offered, based on assumed high aggregation of amine sulfates, was ruled out by finding that that is exceptional rather than general in benzene solutions. So far, a satisfactory general explanation of the apparent anomalies has not been found, Meanwhile, the study of this problem, emphasizing the system tri-n-octylamine - benzene - sulfuric acid - water, has shown the following relations : At water activities 0-0.9, TOA and (TOAH)2804 and their mixtures are monomeric up to ~ 0.5 m; dilute TOAH.HS04 is dimeric, approach- ing tetrameric at ~ 0.5 m; excess H2S04 increases the aggregation farther. Activity coefficients Yroa and Y(TOAH) so decrease smoothly to 0.63 and 0.64 at 0.35 and 0.6 m, resp. ; 'water (for concentration expressed as H20/TOAH) correlates smoothly with TOA/H2S04 ratio. The slope of log Dy vs log Minoag se remains near 1 as water activity is decreased from ~ 1 to < 0.9. It is well known that in many amine extraction systems the exponential dependence of the extraction coefficient on the amine concentration does not . match the apparent ratio of amine to metal ion in the extract species, al- though each may be nearly constant. It was early assumed that such apparent .. . discrepancies were all due to high aggregation of the amine salts in organic media, extracted metal ions entering into the aggregates without changing ORNL - AEC - OFFICIAL . MY - --, . .. .-- them much, but such high aggregation proved to be exceptional rather than viivosiv-gp- viidu general. Subsequently, several systems have been fitted, some very precisely, ORNL - AEC - OFFICIAL by correlations based on models involving specific small aggregates. However, this problem is still a live subject, since there are systems that have re- sisted such fitting, and all the more in that some systems have been well- cis fitted by two different models, so that at least one model must be wrong .. . . ... . --.. although apparently successful. For example: In 1956 Allen fitted sulfuric acid extraction by TOA/benzene on the basis of a deduced aggregation model (1), then found [2] that TOA sulfates in benzene are at most only slightly aygra- gated. Recently Wilson (3) fitted the same extraction data on the basis of a model involving homo- and hetero-dimers, which also fitted Allen's early * . -. * - - * - . -... -=-* aggregation data but was at sharp variance with subsequent extended data. me - -y. (cf. fig. 1). Apparently, both approaches merely provided a sufficient num- ber of suitably-adjustable parameters to fit the data. Again: Lloyd [4] fitted a considerable range of uranyl nitrate extractions by TLA/toluene on the basis of TLA nitrat, monomers and diwers producing two distinct uranyl extract species in which those monomers and dimers remain intact; Scibona (5) fitted the same data on basis of a more satisfying model, involving formal reaction with only the monomers in the partially aggregated TLA nitrate solutions to produce a mixture of two uranyl extract species. (Reference 5 implies that the same approach will suffice for all systems, including the uranyl,ToA sulfate system considered here; but this not possible unless the measured non-aggregation of TOA normal sulfate can be contradicted.) With the shift of emphasis from hydrometallurgy to reactor fuel re- processing and heavy element production, research interest has shifted from sulfate to nitrate and chloride systems. However, for study of the apparent Mass Action anomaly we have returned to the earliest system observed, uranyl ORNL - AEC - OFFICIAL sulfate - TOA/benzene, and here we can so far neither remove nor explain the -- ORNI – AEC - OFFICIAL ONI - AEC - OFFICIAL anomaly. Briefly, log Dy vs log (unbound (TOAH)2SO4) gives slopes near 1 while several different methods indicate 2 - 2.5 moles of (TOAH)2804 (= 4 - 5 moles of contained TOA) bound per mole of uranium (6), but both (TOAH)2S04 and the extract uranium species are monomeric (2). (An anomaly of similar magnitude but opposite direction appears in the system wranyl sulfate - dl-n-decylamine/benzene: There, log Diy ve log (unbound (DDAH)2804) gives slopes nainiy near 1, and 3 moles oi (DDAH)2S04 are bound per mole of uranium (7). However, (DDAH)2804 18 a nearly 40-fold aggregate and the uranium extract species is monomeric (2), so that the expected slope would be less than 0.08.) The following possibilities were considered in attempt to resolve the - - - . anomaly: -. (1) Systematic error in measured Dyr ar ine/u ratio, and/or aggregation number á. (2) Significant increase in ñ under conditions beyond the range first --- -- examined. (3) Significant increase in ñ with changing water content. (4) Change in log D-log concn. slope with changing water content. (5) Aggregation masked by dissociation and/or solvation. (6) Rapid divergence among organic-phase activity coefficients. (7) A cumulative result of relatively small contributions from several effects, fortuitously acting together. All of the specific hypotheses (1 - 6) have been examined except for the .. . ............... activity coefficient of the uranyl•amine-sulfate extract species, under (6). . con .... (Another specific hypothesis, shift of equilibrium due to large increase in - interfacial area, 6418 previously shown to be real and pertinent under some conditions, but not over the range of conditions of present Interest (8).) - - - ORNL - AEC - OFFICIAL ORNL - AEC - OFFICIAL The essential behavior of both the log-log slope and the extract mole ratio have been reconfirmed repeatedly in this system and in many other re- lated systems, so that a significant bias in either of these 18 hardly ORNL - AEC - OFFICIAL credible. The original evaluation of n = 1 by light scattering and viscosity measurements was confirmed and extended by diluent vapor pressure measure- Fi ments, with respect to both H2SO4/TOA ratio (fig. 1) and amine concentration (fig. 2). For TOA, (TOAH)2S04, and their mixtures, ñ remains close to 1. F2 Aggregation develops in proportion to bisulfate content up to ñ = 3.8 for TOAH·HS04 at 0.5 m, then to higher levels only with the incorporation of excess sulfuric acid. The values of ñ also stayed close to 1 over the con- centration range from 0.02 to 0.35 m TOA and 0.6 m (TOAH)2304 (1.2 m contained TOA). (Here ñ is defined in terms of each designated individual species, and of their weighted averages in the mixtures. Alternatively, ñ could be defined in terms of contained TOA radicals or nitrogen atoms, with the obvio:18 re- sulting shift of the points between mole ratios 0 and 1 in fig. 1, or in terms of mean molecular weights without explicit ii.) (TOAH)2504 is known to dissociate slightly into TOA and TOAH• HS04 (1). If attributed entirely to such dissociation, the decrease of nonu.se below 1 in fig. 2 indicates a maximum dissociation of around 5% at 0.4- 0.5 m (TOAH)2SO4, and a maximum in water extraction at H2S04/TOA = 0.5 suggests the same order of limit of dissociation. This is consistent with titration data (1), ruling out the likelihood of significantly large ag- gregation being masked by large simultaneous dissociation. Correspond ingly, model calculations of the 180piestic mole fractions ruled out any likeli- hood that significant aggregation could be masked by benzene solvation. It is of interest to note that the impag.uso, curve in fig. 2 18 closely ORNL - AEC - OFFICIAL parallel to Fomin's cryoscopically determined curve (9) through the range ORNI - AEC - OFFICIAL of overlap, 0.1-0.6 m. Fomin's values are about 1 un't higher, which seems reasonable for their lower temperature. Stoichiometric activity coefficients evaluated for ToA and (TOAH)2804 from the 1sopiestic data represented in fig. 2, by extrapolation of an 08- motic coefficient function, (1-0)/m, to infinite ollution, decreased smoothly and without much curvature from 1 to 0.63 at 0.35 m TOA and to 0.64 at 0.60 m (TOAH)2S04. (In contrast, the same osmutic coefficient function for TOAX•HSO4 could not be satisfactorily extrapolated, so that for it only activity- coefficient ratios (above 0.1 m ETOA) have so far been evaluated.) Since amine salt solutions were partially dried in light-scattering measurements and completely dry in early isopiestic measuremenis, it was obvious to suspect that higher water contents might induce different aggre- gation beliavior during the actual extractions. A technique was developed F3 to permit isopi'estic balancing of benzene solutions in the presence of water vapor from aqueous solutions at fixed activities up to 0.9; å did not change much for any of the TOA species in this range (fig. 3). The water activities in typical extraction systems are somewhat higher. To bridge the remain ag gap, uranium extractions from saturated sodiun sulfate solutions (a, = 0.86) were compared with extractions from 1 M sodium sulfate solutions (a, = 0.96). The log-log plots are parallel (fig. 4), showing that the effect we are 74 seeking does not consist in some abrupt change at highest water activity. The behavior of water itself in the organic phase was also examined, by isopiestic balancing with various TOA-H2S04 solutions in (non-volatile) phenylcyclohexane at fixes water activities from 0.11 to 0.90, and by liquid- liquid equilibration with solutions in benzene at water activities from 0.86 to 1. of the various possible ways of expressing the organic-phase water ORNL - AEC - OFFICIAL concentration, its mole ratio to ionized amine gave the simplest correla- ORNL - AEC - OFFICIAL tion of the corresponding activity coefficients, suggesting that the water may be associated principally with the amine cation. The activity coef- ficients decreased smoothly with decreasing water contents, and increased with increasing bisulfate content. Water contents were also measured in undiluted (TOAH)2804 at fixed water activities, by 180piestic balancing, and from those the activity of the amine sulfate (as solvent) was calculated (fig. 5). The water activity curve deviates considerably from Henry's law, F5 . but neither it nor the amine sulfate activity curve shows any discontinuity or region of critical sensitivity. . Thus, no one of the hypotheses so far considered has been able to resolve ..... . ..... the apparent anomaly in uranyl sulfate extraction by TOA. Unless the correct explanation 18 some thing that we have not yet envisioned, this points toward the last hypothesis (above), « fortuitously cumulative result of several effects acting together. Such an explanation 18 a priori unpleasing, since it would seem to require a considerable degree of coincidence over a consider- able range of conditions within this system and also between this and other systems such as the DDA system mentioned above, but it 18 possible, and at -.present it appears to be the 'nost likely possibility remaining. ORNL - AEC-OFFICIAL References ORNI – AEC - OFFICIAL (1) Allen, Kenneth B., J. Phys. Chem., so, 239-245 (1956). (2) ibid., 62, 1119-1123 (1958). (3) Wilson, A. s., "The Extraction of Sulfuric Aci.d into Benzene by Tri-g- Octylamine," USAEC Report HW-SA-3548 (1965). [4] Lloyd, P. J., and Mason, E. A., J. Phys. Chem., 68, 3120-3129 (1964). (5) Scibona, G., J. Phys. Chem., 70, 1365-1371 (1966). [6] Allen, Kenneth A., J. Am. Chem. Soc., 80, 4133-4137 (1958). [7] McDowell, W. J., and Baes, C. F., Jr., J. Phys. Chem., 62, 777-783 (1958). [8] Allen, Kenneth A., and McDowell, W. J.; J. Phys. Chem., 64, 877-280 (1960). 19) Fomin, V. V.; and Potapova, V, T., 2. Neorg. Khimii, 8, 990-1002 (1963); English trans. Martha Gerrard, O.R.N.L. ----......... ... . - . . . . ORNL - AEC - OFFICIAL !. - - - - i Captions for Figures ORNI - AEC - OFFICIAL F18. 1. Aggregation in the Dry TOA-H2S04-Benzene System vs Sulfate/Amine Ratio. Mean aggregation number n calculated in terms of the nominal compounds. Isopiestic balancing vs triphenylmethane. Fig. 2. Aggregation in the Dry TOA-H2SO4-Benzene System vs Concentration. Mean aggregation numbers ñ and molalities in calculated in terms of the nominal compounds. Ispiestic balancing 18 azobenzene and triphenylmethane. Fig. 3. Aggregation in the TOA-H2SO4-Benzene System vs Water Activity (reference state, pure water). Mean aggregation numbers i calculated in terms of the nominal compounds, at 0.2-0.3 m. Isopieetic balancing vs triphenylmethane, in the presence of water vapor from various saturated aqueous solutions; direct differential vapor pressure mea- surements with and without water. n, AMINE POLYMERIZATION NUMBER . Fig. 4. Reagent Concentration Dependence of Uranyl Sulfate Extraction by TOA at different Water Activities. 0.1 M ETOA in benzene, pre- : equilibrated with specified aqueous solution; ETOA/EU = 100. Equili- brium (TOAH!250g concentration (moles/liter) calculated by material balance fron analytical (ETOA), (H+), and (U) on basis of species UO2SO4.2.3(TOAH,2504 (average) and TOA + (TOAH)2S04 or (TOAH)2SO4 + TOAĦ. HSO4 1 M Na2S04, a = 0.96, pH 1.0 PH 2.0 A Saturated Na2S04, a = 6.86, pH 1.0 0 PA 2.0 A . - - - -- - - -- SI ,, I Fig. 5. Activities in Undiluted .TOA-H2S04-Water System (Organic Phase). Isopiestic balancing, undiluted TOA normal.sulfate vs various saturated aqueous solutions. Reference states: solute, pure water; solvent, pure dry (TOAH)2S04. * ORNL - AEC - OFFICIAL UNCLASSIFIED -; ORNL-DWG 64-8090 CRNI - AEC - OFFICIAL n, AMINE POLYMERIZATION NUMBER FREE-BASE AMINE NORMAL SULFATE BISULFATE AND AND AND NORMAL SULFATE BISULFATE EXCESS H₂SO4 o 0.08 TO 0.16 m TOA O 0.2 TO 0.3 m TOA IN DRY BENZENE 2 h1. junne Suze.. -00 Odovo 1.5 0.5 1.0 MOLE RATIO: SULFATE/AMINE IN ORGANIC PHASE LED:.::.- 73 3 /s in < ! .. 76 10.0 573 Go7 47,5% te .... ORNL AEC - OFFICIAL ORNL - HEC - OFFICII ORNI - AEC - OFFICIAL . . ..؟ و زن-و 7 . . لندن سے - . TO کہا کہ - .ا 77 نه به ده / م ر . - . .. . . . . ORNI - AEC - OFFICIAL ORNI - AEC - OFFICIAL .. .. .. .. .. .. .. .. . .... ORNL - AEC - OFFICIAL ORNI - AEC - OFFICIAL - --- -- - - - - . . - - .. . - 3.. Torna n .... .. ... ... . . ä. MEAN AGGR, 17:21 NUR ........... . . . - . A 1/11 . C 1177 70 21 ; . .. ...... . .. .. .. . . . - ..' -. ORNL - AEC - OFFICIAL ORNI - AEC - OFFICIAL VIDIO. 4. INHO ز - ما mer ORNL - AEC - OFFICIAL ORNI - AEC - OFFICIAL . : ...... ..103 ' : . . . A . O: :.00 D : A ::.. I 0 : 0.001 . 9.01 Odd FREE (TOPH.SOH (1) ::.:.:. .. in wi . - · .: . .... . .. .. ORNL - AEC - OFFICIAL 1 . : .... . . cinsi rene !! micemi... ........... . ORNL - AEC - OFFICIAL -- - - . . . . - - - - - - - 3 - - .. V *- 0.0 TTTT .. . :... tin -..-... : V1OLE PROTON WATER (TOAH), SOM omnia WATER . . . .... . . . V . - . . - : ... ... ..... - .. - :. . .. . - ... ORNL - AEC - OFFICIAL ORNI - AEC - OFFICIAL To be distributed at Conference only Mass Action and Non-Ideality in Extraction by Amine Sulfate C. F. Coleman and J. W. Roddy, ORNL, Oak Ridge, Tenn., U S A ORNI - AEC - OFFICIAL Distribution equibibria involving alkylammonium extractants often appear to deviate significantly from the Mass Action Law, especially in some sulfate systems. Such behavior in some (perhaps all) nitrate and chloride systems can be explained in terms of specific aggregates, but at least some of the sulfate systems have resisted such explanation. E.g., in 200 uranyl sulfate extraction by both tri-n-octylamine (TOA) and di-n-decylamine (DDA) sulfates/benzene, the extract species 18 monomeric in uranium and binds 2 to 3 amine sulfates (4 to 6 amine radicals) 1005 per U. (TOAH)2804 18 monomeric, so that its slope of log D,,-18-(extractant) is expected to be 2 to 3; (DDAH)2804 18 aggregated to ñ ~ 40, so that its slope is expected to be < 0.1; yet both show slopes rather close to 1 gver considerable ranges of con- ditions (1, 2, 3).* The amine/U ratio 18 constant or nearly 80 over a range of amine concentrations and l'loadings." (in contrast, amine nitrate/uranyl 20 nitrate ratio changes significantly with loading.) 0.0 The expected slope of < 0.1 with (DDAH)2804 has IETOA) – 4.6 (UI actually been found when the interfacial area is (0.3 M H,504) (2) limited, but even then only at very low lonic strength (4). The evidence that uranyl 18 monomeric in amine sulfate extracts came first from the observation that Dy is independent of the uranium concentration (aside fcom "loading" effects) in all cases examined. For its complexes with (TOAH)2SO4 and (DDAH)2504 this was confirmed by light scattering, and for the latter also by viscosity and isopiestic measurements (1). The first evidence that (TOAH)2504 is monomeric came from light scattering; it has been confirmed and extended by isopiestic measurements in the present work. The aggregation / number n (calculated in terms of the nominal 8pecies TOA, (TOAH)2S04, TOAH. HSO4, H2S04, and 0.01 0./ their weighted averages in mixtures) is close to 1 from TOA to (TOAH)2304, then rises to nearly 4 at LE TOA] "121 TOAH.HS04 and farther with excess H2S04. The ñ so (0.76 M Hesori ETOA/U :100) defined is a convenient arbitrary formulation of the aggregation. The accompanying figure shows how details change without any change in their signi- ficance if the aggregation is ex- pressed instead as ñ' in terms of TOA radicals (not defined at H2S04/TOA > 1), as ń "mterms of s04 radicals (not defined at H2SO4/TOA < 0.5), or as average molecular weight of unstated species. In each case the dashed line shows the values to be expected if the species 10 were unassociated monomeric TOA, (TOAH)2504, TOAH. HSO4, and H2S0.4. 0.01 0.1 0.001 0.01 0.1 Water concentration from 0 to 0.9 saturated has only slight effect [E DDA)-6[U] [EDDA) *0.94 637 (E DDA/U = 50) De 100% PH? -.- pH 2 . - - on 106 ORNL - AEC - OFFICIAL **000 on i. for any of the species. Increasing amine concentration gives a slight increase in ñ for TOA í ñ ñ 1 + 0.017 m), and a slight decrease for (TOAH)2804 (consistent with less than 5% dis- sociation into TOA + TOAH.HS04 at ~ 0.5 m). For TOAH. HSO4, ñ is close to 2 at 0.01 m, rapidly approaches 4 (n = 2.47 + 8.8 m - 13.9 m², 0.01