According to the nuclear power plant (NPP) decommissioning plan, the Kori Unit 1, one of the pressurized water reactors (PWR) in Korea, is expected to perform chemical decontamination of the reactor coolant system after permanent shutdown . Decontamination of the reactor coolant system for decommissioning has been carried out at many overseas nuclear power plants [2-5], but none at all in Korea. Therefore, the development of the decontamination process technology for decommissioning the Kori Unit 1 is urgently needed . In the case of nuclear power plants, as the number of years of operation increases, radioactive materials generated from the reactor are deposited in the form of a corrosion oxide layer on the surface of the coolant system, thereby increasing the level of radiation. Accordingly, it is necessary to reduce the radiation level by removing the corrosion oxide layer through decontamination before the decommissioning operation, since it causes heavy radiation exposure to workers during the decommission operation.
In general, the oxide layers deposited on the inner surface of structures of the coolant system of the PWR-type NPP are composed of metal oxides of iron, nickel, and chromium. Chemical decontamination by oxidative and reductive dissolutions has so far been recognized as the most effective method. The three key reactions for dissolving metal oxides consist of protonation to decompose oxygen bonds, reduction of Fe3+ ions to Fe2+ ions, and oxidation of Cr3+ ions to Cr6+ ions. Oxidative dissolution of chromium(III) oxide is carried out mainly with potassium permanganate or permanganic acid. Iron and nickel oxides are reduced and dissolved using aqueous solutions of acids like oxalic acid, citric acid, ethylelenediaminetetraacetic acid, and mixtures thereof. Thus, it is necessary to properly combine the oxidation and reduction processes when performing the decontamination of the coolant system . Several permanganate processes using nitric permanganate (NP), alkaline permanganate (AP), and permanganic acid (HP) have been developed mainly for the oxidative dissolution. The reductive decontamination processes (CANDEREM, CITROX, LOMI, and CORD) have been developed to dissolve the iron oxides .
In this study, multi-component chemical equilibrium calculations were performed for heterogeneous aqueous systems that dissolve corrosion oxides using oxalic acid (H2C2O4) as a reducing agent. Magnetite (Fe3O4) is considered as a target corrosion oxide in this work. By performing the equilibrium calculations using a Pitzer-based aqueous solution model that can be applied to aqueous electrolytes containing ions, the effect of the input amount of Fe3O4 and H2C2O4 raw materials on the equilibrium amounts and activity coefficients of chemical species and ions involved in the dissolution reactions was investigated at various temperatures. The pH changes and the formation of ferrous/ferric oxalates were studied as well.
The scope of the work in this manuscript is to clarify the proposed dissolution systems in the thermodynamic perspective. From many experimental studies we have been aware of dissolution kinetics of the corrosion oxides in the reductive dissolution systems. The kinetic variation is actually not correlated with the thermodynamic consideration. Since the reductive dissolution reactions are greatly influenced by kinetics, it is revealed that the interpretation of the reactions through the equilibrium calculations has limitations.
2. Reaction Mechanism and Method of Calculations
The reductive dissolution using a reducing agent in the NPP chemical decontamination is to dissolve transition metal ions from the corrosion oxides deposited on the structure surfaces in the coolant system. In this work, only oxalic acid is used as a strong reducing agent which transfer electrons to Fe3+ ions to be reduced to Fe2+ ions. The reductive decontamination of magnetite with oxalic acid is a reaction mechanism in which three reaction steps occur simultaneously: (a) acidic dissolution of magnetite, (b) reduction of dissolved ferric ions by oxalic acid, and (c) complex formation of the dissolved ferrous ions with the oxalate. The reactions for the acidic dissolution of magnetite are as follows:
Hydrogen ions (H+) are provided by the dissociation of oxalic acid:
Fe3+ ions are reduced to Fe2+ ions by electrons generated by the decomposition of oxalic acid:
Combining the reactions (3) and (7) gives
From reactions (3) and (8), the acidic and reductive dissolution of magnetite by oxalic acid becomes:
Plausible reaction schemes for the formation of complexes of ferric/ferrous ions with oxalate are as follows:
Finally, from reactions (10) and (11), the dissolution reaction of magnetite by oxalic acid can be summarized as follows
The calculations of multicomponent chemical equilibria in an aqueous system uses the Gibbs free energy minimization principle [9-11]. The theories and methods of the chemical equilibrium calculations are described in detail in the author’s previous publication . The expressions of the activity coefficients in electrolyte solutions and the model equations are given as well [13- 16]. In this study, a semi-experimental model developed by Pitzer  and later modified by Harvie et al.  was used as a thermodynamic model suitable for calculating the activity coefficients of species and ions in a multicomponent aqueous solution. The calculations for the aqueous systems covered in this study are based on 1 molality (= 55.5082 mole of H2O).
The chemical equilibrium for the aqueous system of the reductive dissolution of magnetite using oxalic acid was calculated by using the GEM module in the HSC Chemistry ® version 9 . The GIBBS solver and the AQUA module were used to calculate the amounts and activity coefficients of neutrals and ions in the dissolution reactions at equilibrium by varying the concentration of raw materials and the temperature. Additionally, the change of the pH values of the aqueous solution was observed.
3. Results and Discussion
Table 1 shows the conditions for calculating the chemical equilibrium for the reductive dissolution system of magnetite (Fe3O4) using oxalic acid (H2C2O4). The concentration of oxalic acid concentration used for the NPP chemical decontamination is low and typically set to be 20 mM·L–1. Equilibrium calculations were performed while varying the raw amounts of Fe3O4 and H2C2O4 to observe the effect of the input amount of raw materials on the chemical equilibrium. The temperature dependence on the equilibrium was studied by performing the equilibrium calculations while changing the temperature of the system in the range of 25℃ to 125℃ at a given condition of the raw amounts. Note that the raw amount of H2O is fixed at 55.5082 mole (1 liter) as our calculations are based on the aqueous solution of 1 molality. The pressure was kept constant at 1.0 bar for all calculations. Table 2 shows the chemical species and ions present in the solid, aqueous, and gas phases of the dissolution reactions of magnetite with oxalic acid.
Equilibrium calculations were repeatedly carried out at various input amounts of raw H2C2O4 of 0.020 mol, 0.010 mol, 0.005 mol, and 0.004 mol. The results of calculations at two representative input amounts of raw H2C2O4 (0.020 mol and 0.004 mol) are shown in Tables 3 and 4, respectively. Note that the input amount of raw Fe3O4 fed to the aqueous phase are set to be 0.001 mol for all the cases. The amounts and activity coefficients at equilibrium of species and ions participating in the reactions as well as the amounts of raw materials fed are shown at 10℃ intervals in the temperature range from 25℃ to 95℃. The pH values of aqueous solutions at equilibrium are also given as a function of temperature.
For four different amounts of H2C2O4 fed to the aqueous phase, the changes in the equilibrium amounts of main chemical species and ions [CO2(g), Fe2+, Fe3O4(s) undissolved, H2C2O4 unreacted, FeC2O4, HC2O4−, C2O42−, H+] are shown in Fig. 1 as a function of temperature. As shown in Fig. 1, when the input amount of the H2C2O4 reducing agent was from 0.020 mol to 0.005 mol, the undissolved magnetite did not remain in the aqueous solution. In other words, when the concentration of the added H2C2O4 reducing agent was 0.020~0.005 mol∙L–1, it can be seen that 0.001 mol of Fe3O4 was completely dissolved. Fe3+ ions were almost completely reduced to Fe2+ ions by oxalic acid. As can be seen in Table 4 and Fig. 1(d), the Fe3O4 undissolved was observed, when the input amount of raw H2C2O4 was 0.004 mol. This indicates that 0.001 mol of Fe3O4 could not be completely dissolved when the concentration of the added H2C2O4 reducing agent was 0.004 mol·L–1. On the other hand, when the concentration of H2C2O4 was 0.005 mol·L–1, the Fe3O4 undissolved was not observed, which is shown in Fig. 1(c). Therefore, in order to completely reduce and dissolve 0.001 mol of Fe3O4, the concentration of the oxalic acid should be kept higher than 0.004 mol·L–1.
Table 4 and Fig. 2 show the results of the equilibrium calculations when the input amount of raw Fe3O4 was increased 5 times to 0.005 mol while the input amount of H2C2O4 was maintained at 0.020 mol. Fe3O4 was not completely dissolved at all temperatures. Therefore, it is suggested that 0.020 mol of H2C2O4 is insufficient to completely dissolve 0.005 mol of Fe3O4.
The equilibrium amounts of ferrous/ferric oxalates (FeC2O4, FeC2O4+, FeHC2O42+), which can be generated by the complex formation of ferrous/ferric ions and HC2O4– or C2O42− ions, was calculated. In particular, the results for the equilibrium amount of FeC2O4 are schematically shown in Fig. 3. It can be seen that a very small amount (10−8~10−5 mol of order of magnitude) of FeC2O4 was produced, and the amount of FeC2O4 produced decreased as the input amount of H2C2O4 added increased. As can be seen in Tables 3 to 5, FeC2O4+ and FeHC2O42+ were hardly produced.
Fig. 4 shows the change in the pH of aqueous solution as a function of temperature for the Fe3O4-H2C2O4-H2O system at various amounts of raw H2C2O4 added into the aqueous solution. Not surprisingly, when compared at the same temperature, the pH of the solution increased markedly as the raw amount of H2C2O4 decreased. In addition, in the case of the raw amount of H2C2O4 in the range of 0.020 to 0.005 mol, the pH of the solution gradually increased as the temperature increased. However, when the raw amount of H2C2O4 was 0.004 mol, the pH of the solution decreased as the temperature increased, which seems to be due to the fact that Fe3O4 was not completely dissolved.
The chemical equilibrium of the aqueous system for the reductive dissolution of magnetite with oxalic acid (Fe3O4-H2C2O4-H2O) was calculated at high temperatures up to 125℃, by using the HSC Chemistry® version 9 software. This study focused on the effect of the amounts of raw materials fed into the aqueous phase on the equilibrium amounts of species and ions involved in the reductive dissolution reactions and the pH of the aqueous solution. In particular, we observed the conditions in which Fe3O4 was completely dissolved as well as the equilibrium amount of FeC2O4 complex produced.
For the raw amount of Fe3O4 of 0.001 mol, when the raw amount of H2C2O4 was from 0.020 mol to 0.005 mol, no Fe3O4 undissolved was present in the solution and Fe3+ ions were almost completely reduced to Fe2+ ions by H2C2O4. On the other hand, at the raw amount of H2C2O4 as low as 0.004 mol, Fe3O4 was not completely dissolved. The pH of the solution increased with increasing the solution temperature. The equilibrium amount of FeC2O4 produced decreased as the raw amount of H2C2O4 increased. The calculation results obtained from this work provide information for the input amounts of the raw materials and the pH values of the aqueous solution in which Fe3O4 is completely dissolved at high temperatures.