ABSTRACT

Nuclear fuel reprocessing plants generate a large amount of low-level ra­ dioactive wastes, such as condensate waste, which could be treated with the HFMLLE technique under suitable chemical conditions [48c]. Kathios et al. [22] conducted preliminary tests of a microporous hollow-fiber membrane module for the liquid-liquid extraction of actinides to evaluate the feasibility of recovering actinides or processing radioactive wastes. This work involved the analysis of the potential for using these modules for process-scale metal sepa­ rations such as radioactive waste stream cleanup and environmental remedia­ tion. MHF modules were used to remove neodymium (a surrogate for ameri­ cium) from 2 M nitric acid using DHDECMP (dihexyl-ΑζΑ^ -diethylcarbemoyl methylphosphonate) and CMPO («-octylphenyl-A^,A^-diisobutylcarbamoyl methylphosphine oxide) as extractants in diisopropylbenzene. The modules were also used to concentrate the neodymium from the organic phase by backextraction of it into 0.01 M nitric acid. The extractants removed trivalent metal ions from the aqueous phase by forming the complex N d(N 03)3 * 3CMPO or N d(N 03)3 * 3DHDECMP. The results were used to determine the number of modules required to achieve one theoretical extraction stage, that is, one that first allowed mass transfer between the aqueous and the organic phases to reach equilibrium and then allowed complete separation of the two phases to occur before they exited the process. The number of modules per theoretical stage, NTS, is

(45)

where A , is the total membrane surface area of the module based on the internal diameter, ACim is the logarithmic mean concentration difference, is the

logarithmic mean concentration difference in a theoretical stage, A js is the total membrane surface area of all modules based on the internal diameter, Cfi is the solute concentration in the aqueous feed, Cfo is the solute concentration in the effluent streams, and C]* is the solute concentration in aqueous steam in a theoretical stage. The modules used in this analysis were small laboratory-scale versions specifically designed for experimental purposes. In an actual process application, pilot-scale (or larger) modules could be used, each one providing as much as 10 times the membrane area of a laboratory-scale version. An in­ crease in surface area could also be achieved by connecting a number of modules in series. Earlier deficiencies in the design of HF contactors have been removed. Newer modules have an advanced fiber mesh that allows the organic liquid to flow more uniformly throughout the fiber bundle. Also, new shell-side baffle technology has been introduced that may significantly improve the per­ formance.