KINETIC AND EQUILIBRIUM STUDY OF STRONTIUM REMOVAL BY LIQUID-LIQUID EXTRACTION AND MEMBRANE SEPARATION

Abstract

The removal of strontium from highly alkaline solutions is a challenging task in the nuclear wastewater treatment. In this thesis, the possible approach for strontium removal by solvent extraction was investigated. D2EHP A [di-(2-ethylhexyl) phosphoric acid] was used as an extractant. An equilibrium study on solvent extraction under highly alkaline (pH = 14) and a high sodium concentration ([Na] = 4M) conditions was conducted. It was observed that even under such conditions, strontium still could be extracted by D2EHP A. The corresponding distribution coefficient was near 0.3. The influence of OH on the distribution coefficient was investigated, and the complexing reaction between OH⁻ and Sr²+ was incorporated into the mechanism. Due to the high concentration of sodium in nuclear wastewater, solvent extraction efficiency is usually reduced by formation of a third phase. This is because the low solubility of sodium salt of D2EHP A To solve this problem, we used TBP (tributyl phosphate) as diluent. The corresponding stoichiometric study was implemented and the synergistic effect of TBP was addressed. Usually, alkaline nuclear waste storage tanks also have organic compounds, such as EDTA, glycolic acid, oxalic acid etc. in solution. Their influence on the strontium extraction was investigated. EDTA was found to hinder an extraction. We also proposed a possible method to minimize the hindrance effect. In order to remove strontium directly from the storage tank, an extraction- reextraction process was designed and tested in model experiments. Up-hill mass transfer of strontium from a donor phase into a stripping phase was obtained. At the end of the testing run, 98% of total strontium was removed out of the donor solution. In the second part of this thesis, the kinetics of the mass transfer from the organic phase into the aqueous phase was studied in detail. The role of different components m different phases on the mass transfer resistance was addressed experimentally. This study is helpful to understand the mechanism of mass transfer in liquid membrane separation process. In order to obtain the effective diffusion coefficient, transient and steady state mass transfer data were analyzed with corresponding equations. At the steady state, the flux was used to calculate the diffusion coefficient, while in the transient state, time lag was employed to estimate the diffusion coefficient. A discrepancy between the diffusion coefficients obtained from different states was observed (2.4x10⁻⁶ cm²/s in transient state and 1.5xl0⁻⁶ cm²/s for steady state, all inn-octane). This maybe due to the formation of reversed micelles in the membrane during the mass transfer. The transient state of diffusion through multiple layers is traditionally expressed by Barrer's Equations, which is useful for a time lag calculation. The limitation of these equations lies in the equilibrium assumption at the layer-layer interface. We derived new equations that incorporated a non-equilibrium interface into the traditional equations. These general equations could find their applications in the description of liquid-liquid mass transfer process.