lable at ScienceDirect

Vacuum 109 (2014) 61e67

Contents lists avai

Vacuum

journal homepage: www.elsevier .com/locate/vacuum

Measurement of LiCl removal behavior from porous solids by vacuumevaporation

Byung Heung Park a, *, Seung-Chul Oh b, Jin-Mok Hur b

a Department of Chemical and Biological Engineering, Korea National University of Transportation, 50 Daehak-ro, Chungju-si,Chungbuk 380-702, Republic of Koreab Korea Atomic Energy Research Institute, 1045 Daedeokdaero, Yuseong, Daejeon 305-353, Republic of Korea

a r t i c l e i n f o

Article history:Received 12 February 2014Received in revised form2 May 2014Accepted 12 June 2014Available online 27 June 2014

Keywords:EvaporationMolten saltPyroprocessingPorous solidLiCl

* Corresponding author.E-mail address: b.h.park@ut.ac.kr (B.H. Park).

http://dx.doi.org/10.1016/j.vacuum.2014.06.0110042-207X/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

Molten salt processes have been developed in various fields of engineering. In such a process, its effi-ciency and the quality of products would be enhanced when the used molten salt is effectively separatedfrom the product and recycled into the process. Vacuum evaporation has been applied to recover moltensalts due to the low vapor pressure and the high melting point. However, most of researches have beenfocused on the bulk salts evaporation. In this work, LiCl salt evaporation behavior from a porous solid wasinvestigated to develop a post-treatment process of an electrolytic reduction process which uses LiCl asan electrolyte and produces porous solid products. The electrolytic reduction process is one of the maincomponents of pyroprocessing to treat spent nuclear fuel and produce metallic uranium. Instead of usingradioactive material, we prepared porous MgO chips and rods to determine the conditions and measurethe behavior with different physical characteristics of the rods. The temperature and pressure were set to700 oC and 20 mTorr, respectively, and more than 70% of salt was removed within 5 h.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Molten salt or fused salt refers to a salt material in a liquid phasedue to elevated temperature, which is in a solid state at ambientcondition. It is an ionic and stable substance in a wide range oftemperature and exhibits a lot of interesting features such as anability to conduct heat and electricity. Molten salts have beenadopted as reacting or inert media in various technologies such asfuel cell, coal gasification, and pyroprocessing of non-ferrousmetals.

Recently, the molten salt has been used in pyroprocessing ofspent nuclear fuels, which has been developed as a promisingalternative to aqueous processes because of its compactness, radi-ation resistance, and enhanced proliferation resistance features[1e3]. Therefore, many countries including Japan, United States,and France have developed the pyroprocessing as a technical optionamong various nuclear fuel cycle technologies to accomplish theirown future nuclear energy plans as well as to cope with their spentnuclear fuel management issues [4].

In Korea, Korea Atomic Energy Research Institute (KAERI) hasled national R&Ds on fuel cycle technologies and established a basic

concept of pyroprocessing for the purpose of treating spent fuelsfrom light water reactors (LWRs) [5]. Based on electro-metallurgicalprinciples, three core unit processes were actively developed asconnected in series to convert the oxide fuels into metallic formand, then, collect uranium to be disposed of or recycled in a fuelcycle for the future fast reactors; they are referred to as electrolyticreduction, electrorefining, and electrowinning, respectively.

Pyroprocessing for spent fuels has focused on treating themetallic fuels [1] of which chemical components are readily ionizedinto their corresponding ionic species in a high temperaturemoltensalt medium and selectively recovered with respect to theirreduction potentials by an electrochemical means. This process isknown as the electrorefining. However, the commercial LWRs burnnuclear fuels in the form of oxide and, as a consequence, spent fuelsare discharged as oxides. The oxide spent fuels are stable in themolten salt medium of the electrorefining process and not allowedto be directly introduced into an electrorefining bath without apretreatment step. Converting the oxides into a metallic form couldresolve the technical difficulty on accepting the oxide spent fuels inthe molten-salt-based process.

A metallothermic technology was adopted as a reductionmethod from 1990s [6e8]. It utilized a reducing ability of Li metaland produced Li2O with the reduced metal product in a molten LiClsalt. The Li reduction process required an additional process forcollecting Li metal from the used molten salt after the reduction

B.H. Park et al. / Vacuum 109 (2014) 61e6762

process so that Li metal could be recycled within the processes.Recently, the two steps in the reduction process were combinedinto a single step under an electrochemical concept and referred toas an electrolytic reduction process [9e12]. The electrolyticreduction adopting LiCl could be summarized by the followingequation.

MxOy þ 2yLi in LiClð Þ/xMþ yLi2O in LiClð Þ (1)

In the electrolytic reduction cell, oxide (MxOy) to be reduced isused as an electrode where oxygen elements in the solid oxide areelectrochemically dissolved into a molten salt. The shape of theloaded oxide (powder, pellet and so on) is maintained throughout areduction operation and the reduced metal becomes sparse due tothe loss of oxygen elements. The molten salt gradually fills thepores of the reducedmetal product, which are formed by the loss ofoxygen as the reaction proceeds. Therefore, the entangled LiCl saltis pulled out with the porous metal product after the oxidereduction. The percolated salt should be removed from the metalbefore being transferred to the following electrorefiner, otherwisethe accompanied LiCl salt will shift the salt composition of theelectrorefiner where LiCleKCl eutectic salt is used, resulting in thealteration of the physico-chemical properties of the salt medium.

Vacuum evaporation method has been applied to pyro-processes as a post-treatment step for separating a molten salt[13] or a liquid metal [14,15] from metal products or precipitates[16e18]. In principle, the method could be readily applied to LiClsalt removal from the metal products obtained after the electrolyticreduction process because the effective removal of LiCl saltincluded in the porous metal product of the electrolytic reductionprocess can be achieved by applying the high vacuum evaporationtechnique due to the low vapor pressures even at high tempera-tures. Experimental data are required in designing a large-scaleevaporator and calculating material balances throughout thepyro-processes. However, the evaporation condition and thebehavior of LiCl salt evaporation from a porous medium under highvacuum have not been investigated. In this work, we addressedthermodynamic considerations on the LiCl evaporation, examinedthe LiCl removal conditions by changing process variables such astemperature and pressure, and measured the evaporation ratesfrom different porous solids to investigate the effect of porosity andshape of the solid. The porosity and shape of uranium oxide to beelectro-reduced can be controlled by adjusting the conditions ofthe voloxidation process [5] which feeds materials to the electro-lytic reduction process. Therefore, the study of the effect of poroussolid characteristics on the evaporation rate is of significance tointegrating the two processes.

The electrolytic reduction process using spent fuels producesradioactive metals which should not be handled in a typicalchemical laboratory. Thus, in this work, a porous inert solid mate-rial was adopted to observe the behaviors of the LiCl evaporationwith respect to solid characteristics prior to the use of the radio-active materials.

Fig. 1. Pressureetemperature diagram of LiCl and temperature limits of LiClevaporation.

2. Thermodynamic considerations

Thermodynamic analysis on a considered process is of essencein determining the limit of the process and setting the processconditions. As for the vacuum evaporation of LiCl, solid LiCl wouldbe evaporated after being fused to a liquid phase when an imposedtemperature is higher than its triple point. Otherwise, solid LiClwould be changed directly into a vapor phase. The triple point ofLiCl is not known. On a pressureetemperature (PeT) diagram of apure substance, a sublimation line meets a melting line at the triplepoint. Therefore, the triple point can be estimated by PeT behavior

on the solideliquid phase change line expressed as the followingClapeyron equation.

dPsat

dT¼ DHsl

TDVsl(2)

In general, the volume change by the solideliquid phase tran-sition DVsl is nearly negligible for a number of substances, whichimplies that the solideliquid transition temperature is nearly in-dependent of the saturated pressure and the triple point is veryclose to themelting point. Therefore, a higher temperature than themelting point of LiCl (610 �C) on a system would evaporate LiClfrom a liquid phase even under an extremely reduced pressure.

The vapor pressure of the salt can be calculated using a corre-lated equation for phase changes of pure salt vapor. The equationand the temperature parameters for the vapor pressures of liquidLiCl were found from tables of a metallurgical thermochemistrybook [19]. The equation including coefficients for LiCl is expressedas follows with units of mmHg and K for pressure and temperature,respectively.

log p ¼ �10760=T � 4:02 log T þ 22:3 (3)

If we reduce the pressure on an evaporation system, the tem-perature difference between the boiling and the melting point getscloser and the liquid region narrows as expected from Fig. 1 whichis drawn by using Eq. (3). The temperature range of the stable liquidphase is not so significant for the evaporation. However, it becomesan important factor to be considered when recovering the evapo-rated salt for recycling. If the liquid phase range is wide, theevaporated salt would be condensed as a liquid phase in a receiverand then solidified to be a crystalline formwhich is tightly adheredon the receiver. Meanwhile, an instant cooling rapidly crossing theliquid range could change the salt vapor directly into a solid. In sucha case the powder could be formed due to the reduced pressure andthe sparsity of salt molecules in the vapor phase, which could bereadily collected and recovered from a receiver. Consequently, thehigh vacuum should be imposed on the evaporation system takingthe salt recovery into consideration along with the rate ofevaporation.

Chemical stability of the substances in the evaporation system isalso an important aspect in deciding the temperature condition.After the electrolytic reduction, the reduced metal of which majorcomponent is uranium contains the LiCl salt including Li2O which is

Fig. 2. MgO chips before wetting.

B.H. Park et al. / Vacuum 109 (2014) 61e67 63

produced as a result of the metal oxide reduction. The substancesreadily react when exposed to a high temperature condition andthe undesirable reaction between the uranium metal and Li2Owould take place as follows.

Uþ 2Li2O/UO2 þ 4Li ð>909 �CÞ (4)

It means that the post-treatment stepwould degrade the qualityof the metal product by the above reaction if the temperature ex-ceeds 909 �C.

Another material to be considered is a perforated or wovenbasket containing the metal product during the electrolyticreduction process. Due to the high temperature operation (650 �C)of the reduction, usually an iron-based material such as a stainlesssteel (STS) is adopted as a cathode basket containing initially themetal oxide and finally themetal product. The cathode is wet by themolten salt during the reduction and the separation of the metalproduct cannot be attained without the removal of the consoli-dated salt holding the metal product and the basket tightly.Therefore, the metal product is inevitably carried into a salt evap-oration equipment as contained in the basket. Iron element of thebasket material could react with the uranium at high temperatures.A phase diagram of UeFe reveals [20] that intermetallic compoundsof Fe2U and FeU6 exist throughout a wide range of temperature.Moreover, the binary system exhibits a eutectic melting at 725 �C.The eutectic melting is commonly found for uranium with com-ponents of stainless steel at 740 and 859 �C with nickel and chro-mium, respectively. Accordingly, as the temperature of theevaporation process is increased, the potential for the formation ofthe intermetallic compounds and/or the low-melting eutectics isenhanced.

The temperature limits of the LiCl evaporation are presented onthe PeT diagram of LiCl as Fig. 1. As shown by Fig. 1, the uppertemperature and pressure limits for the LiCl evaporation systemshould be 725 �C and 0.29 Torr, respectively.

3. Evaporation condition determination

As explained in the above section, the salt evaporation at therelatively low temperature e slightly higher than the melting pointof LiCl e is important to restrict the unwanted chemical reactions.In this work, experiments to prove the applicability of the evapo-ration and to determine the conditions for the salt removal werecarried out as the first step.

3.1. Material and experiment

LiCl of 99% purity was purchased from Alfa Aesar and usedwithout further treatment. The main purpose of this work was toprove the applicability of the salt evaporation under vacuum as apost-treatment process of the electrolytic reduction of uraniumoxide and to measure the rate of the salt removal. Therefore, it isnot significant to use the reduced uranium as a salt-occludedporous solid matrix since the uranium itself is inert during thesalt evaporation process under the considered conditions. Instead,various types of porous magnesia (MgO) were prepared to absorbthe molten salt. The magnesia is stable against a reaction with LiCleven at high temperatures, and the porosity and the shape of it areadjustable with ease in manufacturing. Porous magnesia baskethad already been used as a container for U3O8 in an early typeelectroreducer [21,22] (recently, it was replaced by a porous steelbasket) and exhibited a chemical stability in a high temperatureLiCleLi2O molten salt.

Porous rectangle magnesia chips (W1.5 cm � H5.0 cm �T0.3 cm) of which porosity is about 20% were prepared and

submerged into a molten LiCl bath in a glove box under an Ar at-mosphere as hung up on a metal frame as shown in Fig. 2. After12 h of the LiCl wetting in the chips, the metal frame was liftedabove the melt surface and the salt-occluded porous chips werecooled down. The amounts of the uptake LiCl in the chips wereestimated by measuring weight differences of the chips before andafter the wetting.

Four different conditions were examined for the evaporation ofthe salt using the prepared salt-occluded porous chips. At the firstrun (EC-01) a chip was exposed to conditions at 650 �C under anatmospheric pressure for 12 h in a closed cylindrical vessel of2.4 dm3 volume. The purpose of the salt evaporation from thereduced product includes the recovery of the salt to recycle it intothe electroreducer and one of the possible approaches for the saltcollection was to drain down the salt as liquid to a collector. In EC-01, molten LiCl salt would be dripped from a chip by a gravitationalforce.

The second chip was placed in a small vessel with1.4 dm3 volume as hung up on a metal frame. The inside pressurewas reduced to about 20 mTorr absolute pressure which was lowerthan the vapor pressure of LiCl 53 mTorr at 650 �C and then thecrucible vessel was tightly closed. The cruciblewas heated to 650 �Cand the temperature was maintained for 12 h in a heating chamber(EC-02 run).

The third and the fourth chips were placed in a vacuum heatingchamber without a vessel to utilizing the relatively large volume ofthe chamber inside (>12 dm3). In the third run (EC-03), the pres-sure was maintained at 20 mTorr and the temperature was held at650 �C for 12 h. The temperature and the durationwere changed to700 �C and 6 h for the fourth run (EC-04) under the same physicalconditions with EC-03.

After each run, the weight changes were measured to calculatethe evaporated fractions. The experimental conditions and the re-sults were summarized in Table 1.

3.2. Results and discussion for the evaporation conditiondetermination

As shown in Table 1, the weights of the porous MgO chips wereless than 7 g and they absorbed about 1 g of molten LiCl before theexperiments. The results obtained from EC-01 to EC-03 were notsatisfactory for the purpose of the research while the EC-04 runachieved a satisfactory result of 95.9% salt removal. Based on theexperiment results, we concluded that the temperature of 700 �C

Table 1Experimental conditions and results of EC runs.

Id. Conditions Porous MgO weight (g) Salt removal(%)

Vessel volume (dm3)

P (Torr) T (�C) Time (h) Prepared Salt occluded(Salt%)

Salt removed(Salt%)

EC-01 760 650 12 6.43 7.36(12.6%)

7.10(9.4%)

28.0 2.4

EC-02 2.0E-2 650 12 6.52 7.46(12.6%)

7.31(10.8%)

16.0 1.4

EC-03 2.0E-2 650 12 6.80 7.75(12.3%)

7.18(5.3%)

40.0 >12

EC-04 2.0E-2 700 6 6.12 7.13(14.2%)

6.16(0.6%)

95.9 >12

B.H. Park et al. / Vacuum 109 (2014) 61e6764

which is close to but lower than the eutectic melting temperatureof a uranium and iron binary system should maintained under highvacuum for the effective removal of the salt.

After the EC-04 run, the distillated chip was horizontally cut atmiddle. The top surfaces of the upper and the lower fragments wereexamined by SEM-EDS method and elements of Mg, O, and Cl wereanalyzed to compare the extents of the salt removal with respect tothe height. Fig. 3(a) and (b) show the mapping images and analysisresults on the top surfaces of the two fragments. At both ends thechlorine compositions were similar and low which indicated thesalt evaporation uniformly took place along the height of the chip.

The formation of the eutectic alloy is not only governed by athermodynamic aspect but also by reaction kinetics. The investi-gation on the eutectic melts between uranium and STS componentssuch as Fe, Ni, and Cr revealed that uraniumwould not dissolve intoan STS 304 domain even at 750 �C and 800 �C for 6 and 2 h,respectively [23]. Recently, 2 kg of bulk LiCl salt was completelyevaporated and recovered within 5 h at 900 �C [24] which is muchhigher than the eutectic melting point. However, in this study, theevaporation temperature was determined as 700 �C consideringthe repeated usage of the basket materials.

Fig. 3. SEM-EDS analysis of MgO chip surfaces at (a

4. Evaporation rate measurement

4.1. Experiment system

The rates of salt distillation were measured by a thermo-gravimetric method. A column-type salt vacuum distillationapparatus was installed as shown in Fig. 4. It was composed of avertical alumina tube embraced by an electric furnace, a load cell,and a digital pressure sensor. High purity Ar gas was introducedfrom the top to sweep the evaporated salt into a receiver at thebottom. A salt-occluded porous MgO rod was loaded in a cage-typebasket and inserted into the heating zone of the column from thetop as hung up to the load cell tungsten wire. The experimentaltemperature was controlled by monitoring temperature from athermocouplewhichwas inserted from the top and placed adjacentto a sample. A vacuum pump connected at the bottom continuouslyaspirated the column atmosphere making a downward flow andthe pressure was adjusted by a throttle valve at an outlet line.

Porous MgO rods were prepared with respect to diameter,porosity, and length. As explained in the evaporation conditiondetermination experiments, they were submerged into a molten

) top and (b) sectioned middle after EC-04 run.

Fig. 4. Evaporation rate measurement apparatus: (a) photo and (b) schematic diagram.

Table 2

B.H. Park et al. / Vacuum 109 (2014) 61e67 65

LiCl bath in a glove box to absorb LiCl salt to be removed. As a salt-occluded porous MgO rod was loaded, the weight change wasrecorded with a temperature and a pressure throughout an evap-oration run.

4.2. Results and discussion for the evaporation rate measurement

A typical result of an evaporation behavior was presented byFig. 5. It was obtained by using a salt-occluded porous MgO of15 mm diameter (D), 10 cm length (L), and 0.28 porosity (f). As thetemperature increased with time, the weight was slightlydecreased and the evaporation of the salt was rapidly took placewhen the temperaturewas stabilized at the experimental conditionof 700 �C. The temperature was controlled to be increased fromambient temperature to 700 �C within 2.5 h and then kept at theexperimental temperature for 3 h. The initial weight measured bythe load cell included a cage containing a sample. Pressure was

Fig. 5. Typical evaporation behavior of a salt-occluded porous MgO rod. (ER-06).

reduced at the beginning of an experiment and stably maintainedat 20 mTorr throughout a run as shown in Fig. 5. Due to a down-ward Ar flow the weight change responses included some noises.

Experiments were carried out by varying the physical charac-teristics of the porous MgO rods. Different types of rods were usedas summarized in Table 2 which also presents the amounts ofloaded salt and the removed salt fractions. The same experimentalconditions (T ¼ 700 �C and P ¼ ~20 mTorr) were applied in theseevaporation rate (ER) runs. The porosity (f) of a rod was definedand calculated by the following equation,

f ¼ 1� ra=rr (5)

where, ra and rr are the apparent density of a porous MgO rod andthe real density of solid MgO (3.58 g/cm3), respectively.

Summary of ER runs.

Id. MgO rodcharacteristics

Porous MgO weight (g) Saltremoval(%)

D(mm)

L(cm)

f Prepared Salt occluded(Salt%)

Salt removed(Salt%)

ER-01 15 10 0.18 51.63 57.77(10.6%)

51.63(0.0%)

100

ER-02 10 10 0.18 23.15 26.02(11.0%)

23.79(2.7%)

77.7

ER-03 5 10 0.19 5.71 6.39(10.6%)

5.71(0.0%)

100

ER-04 15 8 0.20 40.35 44.31(8.9%)

41.21(2.1%)

78.3

ER-05 15 6 0.18 31.16 34.13(8.7%)

32.01(2.7%)

71.4

ER-06 15 10 0.20 50.91 56.91(10.5%)

52.57(3.2%)

72.3

ER-07 15 10 0.28 45.72 53.80(15.0%)

46.82(2.3%)

86.4

Fig. 6. Comparison of evaporation behavior with MgO rod diameter. Fig. 8. Comparison of evaporation behavior with MgO rod porosity.

B.H. Park et al. / Vacuum 109 (2014) 61e6766

Taken as a reference, the result of ER-01 was compared withother series of runs as given by Figs. 6e8. The first series of ex-periments were devised to understand the evaporation behaviorwith the diameter of porous MgO rod. As shown in Fig. 6, the initialsalt removal fraction is greater and the removal completes earlierwhen the diameter is smaller. The data acquisition for ER-02 wasfinished at 3.8 h and that would be reflected the relatively low saltremoval fraction compared with ER-01 and ER-03.

The effects of the length and the porosity on the salt removalbehavior were presented in Figs. 7 and 8, respectively. In Fig. 7, theoverall behaviors of the salt removal were seen similar withdifferent length of the rod. However, it was found that the saltevaporation completed earlier when the length was decreased. Theporosity change within 18e28% seems to exert no significant effecton the overall salt removal as shown in Fig. 8.

The final salt removal fractions of ER runs summarized in Table 2ranged from around 70 to 100%. The main reason of the differencewas supposed to be due to the pore characteristics of rods. Weprepared the porous rods focused on the apparent characteristicssuch as radius, length, and porosity. However, the micro charac-teristics such as pore size distribution and tortuosity of poreswouldbecome an important aspect as salt evaporation proceeds tocompletion.

Fig. 7. Comparison of evaporation behavior with MgO rod length.

Most of evaporated salt was condensed in the salt receiver.However, the salt was tightly attached as crystalline form notstacked as powder. The relatively long cooling zone below theheater might give enough time for salt crystal to grow. Thus,increasing temperature gradient between the evaporation zoneand the receiver with a rapid cooling on the zone is required torecover the salt as power.

5. Conclusions

Pyroprocessing to treat spent nuclear fuel adopts molten salts aselectrochemical reaction media. The high melting points and thelow vapor pressures of the molten salt require high vacuum con-ditions not to increase temperature when a separation of the saltfrom the reaction product is necessary for conditioning the productin the intermediate and/or final steps.

In this work, porous MgO was prepared to investigate theremoval behavior of molten LiCl under high vacuum conditions,which was used in an electrolytic reduction process of pyropro-cessing. As the first step of determining the effective separationcondition, porous MgO chips were exposed to various pressure andtemperature conditions in a batch system. Considering the tem-perature limits by undesirable chemical reactions, 700 �C and 20mTorr were set and it was found that pressure should be main-tained during the evaporation of the salt by increasing the vesselvolume in a batch operation or by flowing an inert gas whichcarries the evaporated salt in a continuous operation.

Continuous evaporation experiments were performed toinvestigate the removal behavior under the set conditions.Different kinds of MgO rod were prepared by changing physicalcharacteristics of radius, length, and porosity. The salt included inporous MgO rod was gradually removed as the temperature in-creases and evaporated at the set temperature. Compared with theexperimental results from the different radius and length, it wasfound that the evaporation dominantly took place in a radial di-rection. However, the difference in the porosity within 18e28% hadno significant influence on the overall evaporation behavior. Inaddition, it seemed that the micro structure of the pores wouldaffect the final removal fraction.

In this work, the applicability of high vacuum evaporation ofmolten salt from a porous solid was demonstrated especially for theconditioning the reaction products from electrochemical processusing molten salt. As a post-treatment step of an electrolytic

B.H. Park et al. / Vacuum 109 (2014) 61e67 67

reduction process, the experimental results would be used for thedevelopment of a scaled-up equipment.

Acknowledgment

This work was supported by a grant from the Nuclear Researchand Development Program of National Research Foundation (NRF)funded by the Ministry of Science, ICT & Future Planning (MSIP) ofRepublic of Korea.

References

[1] Laidler JJ, Battles JE, Miller WE, Ackerman JP, Carls EL. Prog Nucl Energ1997;31:131e40.

[2] Nawada HP, Fukuda K. J Phys Chem Solids 2005;66:647e51.[3] Inoue T, Koch L. Nucl Eng Technol 2008;40:183e90.[4] OECD/NEA. Pyrochemical separations in nuclear applications. OECD Report;

2004.[5] Lee H, Park G-I, Kang K-H, Hur J-M, Kim J-G, Ahn D-H, et al. Nucl Eng Technol

2011;43:317e28.[6] McPheeters CC, Pierce RD, Mulcahey TP. Prog Nucl Energ 1997;31:175e86.[7] Usami T, Kurata M, Inoue T, Sims HE, Beetham SA, Jenkins JA. J Nucl Mater

2002;300:15e26.

[8] Usami T, Kato T, Kurata M, Inoue T, Sims HE, Beetham SA, et al. J Nucl Mater2002;304:50e5.

[9] Kurata M, Inoue T, Serp J, Ougier M, Glatz J-P. J Nucl Mater 2004;328:97e102.[10] Herrmann SD, Li SX, Simpson MF, Phongikaroon S. Sep Sci Technol 2006;41:

1965e83.[11] Jeong SM, Shin H-S, Hong S-S, Hur J-M, Do JB, Lee HS. Electrochim Acta

2010;55:1749e55.[12] Choi E-Y, Lee JW, Park JJ, Hur J-M, Kim J-K, Jung KY, et al. Chem Eng J

2012;207e208:514e20.[13] Bourges G, Faure S, Fiers B, Saintignon S, Lemoine O, Cardona-Barrau D, et al.

Procedia Chem 2012;7:731e9.[14] Kato T, Iizuka M, Inoue T, Iwai T, Arai Y. J Nucl Mater 2005;340:259e65.[15] Westphal BR, Price JC, Vaden D, Benedict RW. J Alloy Comp 2007;444e445:

561e4.[16] Eun HC, Yang HC, Cho YZ, Lee HS, Kim IT. J Hazard Mater 2008;160:634e7.[17] Eun HC, Yang HC, Lee HS, Kim IT. J Nucl Mater 2009;395:58e61.[18] Yang H-C, Eun H-C, Kim I-T. Vacuum 2010;84:751e5.[19] Kubaschewski O, Alcock CB. Metallurgical thermochemistry. 5th ed. Pergamon

Press; 1979.[20] Factsage, http://www.factsage.com [accessed 30.01.14].[21] Seo CS, Park SB, Park BH, Jung KJ, Park SW, Kim SH. J Nucl Sci Technol

2006;43:587e95.[22] Park SB, Park BH, Jeong SM, Hur JM, Seo CS, Choi S-H, et al. J Radioanal Nucl Ch

2006;268:489e95.[23] Park SB, Cho DW, Hwang SC, Kang YH, Park KM, Jun WG, et al. J Kor Radioact

Waste Soc 2010;8:41e8.[24] Kim IS, Oh SC, Im HS, Hur JM, Lee HS. J Radioanal Nucl Ch 2013;295:1413e7.