Recovery and recycling of uranium from rejected coated particles for compact high temperature reactors

Recovery and recycling of uranium from rejected coated particles for compact high temperature reactors

Journal of Nuclear Materials 473 (2016) 229e236 Contents lists available at ScienceDirect Journal of Nuclear Materials journal homepage: www.elsevie...

2MB Sizes 0 Downloads 4 Views

Journal of Nuclear Materials 473 (2016) 229e236

Contents lists available at ScienceDirect

Journal of Nuclear Materials journal homepage: www.elsevier.com/locate/jnucmat

Recovery and recycling of uranium from rejected coated particles for compact high temperature reactors Rajesh V. Pai a, *, P.K. Mollick b, Ashok Kumar a, J. Banerjee c, J. Radhakrishna a, J.K. Chakravartty b a b c

Fuel Chemistry Division, Bhabha Atomic Research Centre, Mumbai, India Powder Metallurgy Division, Bhabha Atomic Research Centre, Mumbai, India Radiometullurgy Division, Bhabha Atomic Research Centre, Mumbai, India

h i g h l i g h t s  The oxidation behaviour of coated particles was studied in air, O2 and moist O2.  It was observed that coated layers cannot be completely removed by mere oxidation.  Complete recovery of uranium from the rejected coated particles has been carried out using a combination of dry and wet recovery scheme.  A crushing step prior to oxidation is needed for full recovery of uranium from the coated particles.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 29 January 2016 Received in revised form 22 February 2016 Accepted 23 February 2016 Available online 27 February 2016

UO2 microspheres prepared by internal gelation technique were coated with pyrolytic carbon and silicon carbide using CVD technique. The particles which were not meeting the specifications were rejected. The rejected/failed UO2 based coated particles prepared by CVD technique was used for oxidation and recovery and recycling. The oxidation behaviour of sintered UO2 microspheres coated with different layers of carbon and SiC was studied by thermal techniques to develop a method for recycling and recovery of uranium from the failed/rejected coated particles. It was observed that the complete removal of outer carbon from the spheres is difficult. The crushing of microspheres enabled easier accessibility of oxygen and oxidation of carbon and uranium at 800e1000  C. With the optimized process of multiple crushing using die & plunger and sieving the broken coated layers, we could recycle around fifty percent of the UO2 microspheres which could be directly recoated. The rest of the particles were recycled using a wet recycling method. © 2016 Elsevier B.V. All rights reserved.

1. Introduction A prototype Compact High Temperature Reactor (CHTR) is being developed by the Bhabha Atomic Research Centre as part of the Indian programme to produce hydrogen as a substitute for fossil fuels [1e3]. TRISO (TRI structural ISOtropic) coated particles (schematically shown in Fig. 1) will be used as the fuel in these reactors. A TRISO coated fuel particle considered here composed of a heavy metal (e.g., uranium, plutonium, thorium, etc.) oxide or mixed oxide/carbide fuel kernel (~500 mm diameter) coated with four layers of three isotropic materials. The four layers are: a porous

* Corresponding author. E-mail address: [email protected] (R.V. Pai). http://dx.doi.org/10.1016/j.jnucmat.2016.02.030 0022-3115/© 2016 Elsevier B.V. All rights reserved.

carbon layer of around 90 mm thick (buffer), followed by a dense inner pyrolytic carbon layer (IPyC; 30 mm thick), a silicon carbide layer (30 mm thick), and a dense outer pyrolytic carbon layer (OPyC; 50 mm thick) in sequence. The buffer layer acts as a void volume for fission gas and CO. Its function is also to accommodate fuel kernel swelling and protects TRISO coating from fission recoil. The inner pyrolytic carbon layer retains the gaseous fission products and acts as a diffusion barrier for metallic fission products. It also acts as a mechanical substrate for SiC layer deposition. SiC layers are impervious to fission gases at normal operating temperatures and the SiC layer primarily acts as a structural material. The last layer, outer pyrolytic carbon layer retains gaseous fission products. This layer as well as IPyC layer, on irradiation, puts SiC layer into compression to limit stresses. Also it provides bonding layer with

230

R.V. Pai et al. / Journal of Nuclear Materials 473 (2016) 229e236

UO2 kernal Buffer layer IPyC SiC OPyC

Fig. 1. Schematic of cross section of TRISO coated particle.

carbonaceous fuel element matrix. The TRISO coating fuel system acts as a pressure vessel which accommodates the internal gas pressures generated during fission of the fuel kernel material [4,5]. The TRISO coated particles to be used in these reactors should withstand high temperature of ~1500e1600  C without any disintegration of the fuel particle [6]. The coated particle fuel is subjected to many forces that put stress on the TRISO coating. Potential failure mechanisms have been previously summarized [7]. These coatings of PyC and SiC on fuel kernel are formed by chemical vapour deposition (CVD) technique [8]. These coatings are obtained by pyrolytic decomposition of hydrocarbon gas or methyltrichlorosilane (CH3SiCl3) vapour in fluidised/spouted beds. A schematic of a spouted bed reactor is given in Fig. 2. Different coatings are prepared at different temperatures. All the layers are coated in an uninterrupted sequential process in the same fluidized bed reactor. The conditions under which layer deposition takes place are very important as these deposition parameters determine the properties and coating thickness of the coated particle formed. Parameters such as temperature, pressure, gas composition and gas ratios all play an important role in fixing the coated particle properties.

Since a very large number of TRISO particles (about 13.5 millions) [3] comprise the core, manufacturing of fuel kernels, deposition of multilayered coatings on the particles, and characterization of the coated as well as uncoated particles pose special challenges. It is very important to determine the fraction of defective coated fuel particle with respect to the specified coating thicknesses as the quality of the particles and, consequently, the acceptability of the fuel compacts and their irradiation performance is based on this parameter. The failure probability of a batch of fuel particles depends on statistical variations in the fuel design parameters as well as variation in the characteristic strengths of the coating layers in a batch. The fraction of defective or failed SiC layers is normally determined by burn-leach method and the gaseous acid leach method. In both methods the final step is extraction of kernel by leaching [9]. Recently, an integrated mechanistic coated particle fuel performance code known as “PARFUME” was developed at the Idaho National Laboratory which allows for statistical variations in the kernel diameter, the four layer thicknesses, the pyro carbon densities and several other parameters [10]. Since the fuel kernel contains 235U or 233U, recovery and recycling of these fissile materials from these failed/rejected coated particles are very much essential. The coated carbon layers should be removed by oxidation to recover the uranium from the failed/ rejected coated particles. Several attempts have been made in the past to remove the coatings layer by layer but considered difficult regardless of the method employed. This is mainly due to the chemical nature and strength of the coatings, in particular the silicon carbide layer, which is rigid and non reactive with most of the acids and bases [11]. It has been reported that removal of the layers by ion sputtering technique does not result in uniform removal of the layers [12]. Removal of the silicon carbide layer by molten salts at 800  C affects the distribution of fission products in the layer [13] for irradiated coated particle. In this paper, the oxidation behaviour of coated particles has been studied using Thermal Analyser and Temperature Programmed Desorption-Reduction-Oxidation (TPDRO) instrument. Based on these studies, a recovery procedure has been set up for the rejected coated particles (un irradiated). For these studies we have used sintered natural uranium oxide microspheres as a surrogate for 233UO2 prepared by Internal Gelation Technique which were coated using CVD technique. 2. Material and experimental methods 2.1. Preparation of UO2 microspheres The UO2 microspheres required for the coating experiments were prepared by Internal Gelation Process [14]. The dried UO3 microspheres prepared by this method were heated at 1050  C for 4 h in N2þ 8% H2 atmosphere. Further the spheres were sintered at vacuum/Ar þ8% H2 atmosphere at 1400  C for 3 h. The density of these microspheres was determined by liquid displacement method [15]. The size (diameter) of the sintered microspheres was determined by optical microscope and specific surface area of the microspheres was determined by BET method [16]. 2.2. Coating of carbon layers and SiC on UO2 microspheres

Fig. 2. Schematic of Spouted bed used for coating.

The coating experiment was carried out in a spouted bed CVD furnace on natural UO2 microspheres. Detailed reactor design had been reported elsewhere [17]. Spherical UO2 microspheres of size range 500e600 mm and density 1055 ± 20 kg/m3 were used for CVD coating. For coating porous buffer layer, acetylene (C2H2) gas was used. A mixture of acetylene and propylene (C2H2/C3H6) were used for coating IPyC and OPyC layers whereas, methyltrichlorosilane

R.V. Pai et al. / Journal of Nuclear Materials 473 (2016) 229e236

231

Fig. 3. (a) UO2 microspheres obtained from direct reduction and sintering of UO3 (b) SEM image.

Fig. 4. (a) Sintered UO2 microspheres obtained through modified heating scheme (b) SEM image.

(CH3SiCl3) was used for the SiC layer coating. In all the pyrolytic carbon layers, argon was used as a carrier gas, while hydrogen was used for the SiC layer. According to the simulated process parameters, operating parameters were set through LABVIEW programming to control all precursor and carrier gases flow through the reactor inlet nozzle.

at a heating rate of 10  C min1 for studying the oxidation behaviour of the coated particles. The crushing strength of the coated particles was determined by a crush strength apparatus. The crushed particles were also subjected to TPO to study their oxidation behaviour. 2.5. Recovery and recycling of uranium from the coated particles

2.3. Evaluation of microstructure by scanning electron microscopy For micro structural examination under a Scanning Electron Microscope, a double sided thin layer of gold (~100 Å) was coated on the microspheres/coated particles by thermal evaporation in a vacuum coating unit. A vacuum of 3.5  109MPa was maintained using a diffusion pump having a rotary backing. A measured quantity of gold wire (99.99% pure) was wrapped on a tungsten wire which in turn was heated to evaporate the gold wire. The coated samples were analysed by a Scanning Electron Microscope for evaluation of texture and morphology of sintered UO2 microspheres and coated UO2 particles.

20e30 g of coated particles were heated in a furnace at 1000  C in air, O2 and moist O2 atmospheres at different durations in separate experiments. The heated microspheres were weighed after cooling to room temperature and observed under a microscope for monitoring the carbon loss. Alternatively, around 15 g of coated particles were taken in 20 mm die plunger and pressed for 2 min (100 MPa). The crushed coated particles were separated from the uncrushed particles by sieving and tabling. The crumbled UO2 microspheres were oxidized in flowing O2 at 1000  C for 5 h. The fine powder obtained was dissolved in concentrated HNO3. The obtained solution was analysed for carbon impurity and recycled for UO2 microspheres preparation.

2.4. Thermal and oxidation behaviour of coated particles 3. Results and discussion The thermal stability of the coated particles was studied using a simultaneous TG-DTA analyser. For this, about 40 mg of coated particles were loaded in the thermal balance. The microspheres were heated up to 1100  C in flowing O2 atmosphere at a heating rate of 4 C/min. Temperature Programme Oxidation (TPO) studies of the crushed coated microspheres was carried out using a TPDRO1100 analyser (Thermo Quest, Italy) under the flow of O2 (5%) þHe at a gas flow rate of 10 ml min1, in temperature range 25e850  C,

The UO2 microspheres required for coating were prepared by reducing UO3 microspheres in N2 þ 8% H2 atmosphere. These reduced UO2 microspheres were sintered in Ar þ 8% H2 at 1400  C for 4 h. These microspheres observed under a microscope looked glossy and smooth. Fig. 3(a) and (b) show the microsphere surface observed through a microscope and SEM respectively. The deposition of porous buffer layer is carried out using spouted bed reactor

232

R.V. Pai et al. / Journal of Nuclear Materials 473 (2016) 229e236

Fig. 5. TRISO coated UO2 kernel. Fig. 7. Room temperature XRD pattern of (a) un-coated UO2 kernel and (b) coated particle oxidized in O2 at 1000  C.

0.4 0.2

-0.2 -0.4 4

-0.6 -0.8

Heat flow (mV)

8

0.0

Weight change (mg)

using CVD technique. The force which is responsible for the adherence of the buffer layer to the kernelis physical in nature. Because of this, the physical adherence of buffer carbon layer is expected to be better on a rough surface. For our experimental conditions of coating, it was experienced that smooth shiny kernels were non conducive for coating of carbon layer. The carbon layer coated found to be peeling off the kernel surface. Further, a calcination step at 850  C in air atmosphere was introduced prior to reduction and pre-sintering. Further, sintering of these microspheres was carried out at 1400  C in vacuum. The microspheres were sieved using standard ASTM sieves for obtaining required size fraction of 550 ± 50 mm. The smaller and bigger sizes were rejected. The suitable size fraction was tabled using a smooth tray several times in order to separate any cracked microspheres. The smoothly flowing microspheres were separated from the cracked microspheres using a suction device. The density of these microspheres was found to be 97% of theoretical density of UO2 as determined by liquid displacement method. The morphology of these microspheres was observed under an optical microscope and SEM. It can be visualized that the surface roughness of UO2 microspheres using this modified heat treatment scheme was more compared to glossy UO2 microspheres obtained from direct reduction of UO3 microspheres followed by sintering. The size of these microspheres observed under microscope also showed that they are in the required size range. The aspect ratio of these microspheres was determined by optical microscope to be 1.01 ± 0.01. The specific surface area of these microspheres determined by BET method was ranging from 0.20 to 0. 56 m2/g. These modifications in heat treatment scheme resulted in UO2 microspheres suitable for coating carbon layer. Fig. 4(a) & 4 (b) show the optical photograph of UO2 microspheres and their SEM image prior to coating experiments. From these figures it can be observed, that the surface of

-1.0 0

-1.2 -1.4 -1.6

0

200

400

600

800

1000

o

Temperature ( C) Fig. 8. TG-DTA pattern of coated particle.

these microspheres is rough enough for adhering the first pyrolytic carbon layer.

3.1. Thermal and oxidation studies The coated particles used for the thermal and oxidation studies & recovery and recycling studies were rejected batches resulted due to lesser coating thickness of C and SiC layers than the specified

Fig. 6. Coated particles heated in static air (a) and flowing O2 atmosphere (b) at 1000  C for 5 h.

R.V. Pai et al. / Journal of Nuclear Materials 473 (2016) 229e236 1600 1 1400 1200

-1

1000

-2

800

-3

600

-4 -5

400

-6

200

-7

0

-2000

0

2000

4000

6000

8000

10000

12000

Temperatuture (°C)

Wt. change (%)

0

14000

Time (sec) Fig. 9. Thermo gram of coated particles in moist O2 atmosphere.

233

not undergo oxidation which suggested that the SiC layer was intact. The room temperature XRD pattern of starting UO2 kernel was matched with that of oxidized coated particle. From this, it was concluded that the calcination in O2 at 1000  C could not oxidize UO2 to its higher oxides as the SiC layer withstood the oxidation step at this temperature. This can also be seen from Fig. 6 (b). The silverish texture could be seen (SiC layer) due to the oxidation of OPyC layer. The XRD also showed the presence of graphitic carbon which suggests that removal of pyrolytic carbon layers is a sluggish process and not complete during oxidation. Also we did not observe any presence of SiO2 in the XRD pattern that would have formed due to the oxidation of SiC. For the detailed investigation, thermal stability of coated particles was evaluated by performing a simultaneous TG-DTA analysis. Fig. 8 shows the thermogram of coated particles performed at 1100  C in flowing O2 atmosphere. The thermogram shows a sharp weight loss in the range 550e750  C. This weight loss was due to the oxidation of OPyC layer. Beyond 750  C, a weight loss was very sluggish showing that removal of carbon is difficult beyond this temperature. This shows the integrity of the carbon coating to the SiC surface. A corresponding heat release due to the oxidation of carbon could also be seen at this temperature range. It has been reported that moist O2 oxidizes carbon better than dry O2 [18]. The effect of moisture in the oxidation of carbon layers and SiC in O2 atmosphere was studied. For this, a few milligrams of coated particles were loaded in a thermo balance. The sample was heated up to 1400  C in moist O2 atmosphere. The moist O2 atmosphere was created by bubbling O2 at a flow rate of 100 mL/min through a water column. Fig. 9 shows the thermogram of coated particles heated in moist O2 atmosphere. It can be seen from this thermogram, that there is around 6 wt% weight loss in the temperature range 720e950  C. This weight loss is corresponding to the loss of OPyC layer present in the coated particles. Beyond 950  C, aslope change is observed in the thermogram indicating the loss of carbon from just above the surface of SiC is somewhat sluggish. The oxidation of SiC at high temperature has been reported by many workers [19,20]. They have reported that formation of SiO2 is the rate controlling step in the oxidation of SiC. The formation of SiO2 should result predominantly from the reaction

SiC þ 2O2 /SiO2 þ CO2 Fig. 10. Dry and wet combination scheme for recovery of uranium from rejected coated particles.

thickness. When these coated particles were sectioned and observed through optical microscope and SEM, it was revealed that the thickness of SiC layer which is acting as a pressure vessel in these types of fuel materials was 10 mm compared to a specified 30 mm SiC thickness. Typical micrograph is shown in Fig. 5. In order to remove the OPyC and other coating layers, coated particles were subjected to calcination in an air furnace from temperatures 800e1000  C for different durations batch wise. During the oxidation of coated particles in static air it was observed that the oxidation of OPyC layer was incomplete which suggested that air atmosphere was not able to oxidize the carbon layer completely even at 1000  C. Alternatively, the coated particles were heated in a furnace under flowing O2 atmosphere at 1000  C for 5 h. The surface of the microspheres observed under microscope suggested that though small patches of carbon layers were still present in the microspheres, the oxidation of carbon layers was more severe. Fig. 6 shows the coated particles heated in (a) static air and (b) flowing O2 atmosphere at 1000  C for 5 h. Fig. 7 shows the room temperature XRD pattern of coated particle oxidized at 1000  C in O2. The XRD shows that UO2 kernel did

It has been reported that [21] SiO2 forms during passive oxidation which gets deposited over the surface of SiC leading to a net increase in the mass. They observed that SiO2 formed at lower temperature develops a layer over the surface of the SiC.A small weight gain observed in Fig. 9 in temperature range 1050  Ce1400  C could be due to the oxidation of SiC to SiO2. Though the room temperature XRD pattern of this heated coated particles indicated presence of SiO2, any higher oxides of uranium were absent indicating that moist O2 could not remove the carbon layers and SiC completely to oxidize uranium oxide. Further, a gradual weight loss could be observed due to the removal of IPyC in the isothermal region. It was observed that even though, a soak of 2 h at 1400  C in moist O2 could not remove the layers of carbon layers fully. 3.2. Recovery and recycling of uranium from coated particles For recovery and recycling of uranium from the coated particles, a combination of dry and wet recycling scheme was employed. The schematic of the combined scheme is given in Fig. 10. 3.2.1. Dry recovery scheme The coated carbon layers and SiC layer need to be removed

234

R.V. Pai et al. / Journal of Nuclear Materials 473 (2016) 229e236

Fig. 11. SEM of a partially crushed coated particle.

Fig. 12. Photograph of coated particles crushed with a die plunger.

before uranium in the particle could be recycled. As a simple oxidation step was unable to remove the coated layers completely, these layers have to be removed by careful crushing of coated particles. The crushing behaviour of coated particle studied using a crush strength apparatus showed that the average crushing strength of these particles for the outer pyrolytic carbon was 15e20 N/particle and for crushing of all the layers down to UO2 kernel was 45e50 N/particle. SEM image of a partially crushed coated particle is presented in Fig. 11. In order to remove the coating layers, around 20 g rejected coated particles were pressed using a 15 mm diameter die and plunger. The pressure was varied from 50 to 100 MPa depending up on the amount of material used for a particular crushing. The pressure was suitably optimized for crushing the carbon layers and removal of SiC layer. It was observed that although the coated layers got crumbled the majority of inner sintered UO2 microspheres were still intact. A photograph of the crushed sample recorded using an optical microscope is shown in Fig. 12. These intact microspheres were separated from the crushed carbon powder and SiC shells using multiple sieving and tabling. These were then subjected to an air jet of ~0.41 MPa pressure to remove carbon residue adhering on the surface of UO2 microspheres if any. Smooth UO2 microspheres were obtained without any carbon residue sticking on the surface which was also confirmed by observing through an optical microscope. These recovered UO2 kernels were directly recycled for further coating experiments. 3.2.2. Wet recovery scheme Although, some of the UO2 kernels were recycled directly for coating the carbon and SiC layers, a large fraction of microspheres were found not suitable for coating experiments due to surface erosion and some got crumbled during the pressing of using a die and plunger. This may be due to the non-uniform compaction pressure throughout the coated particle bed. The SEM image of

Fig. 13. SEM image of coated particles crushed with a die plunger.

R.V. Pai et al. / Journal of Nuclear Materials 473 (2016) 229e236

Crushed, sintered UO2 microspheres

600

Detector voltage (mV)

235

300

Crushed coated particles Fig. 16. Crushed, oxidized coated particles.

0

0

200

400

600

800

ο

Temperature ( C)

Intensity (counts)

Fig. 14. TPO profile of crushed sintered UO2 microspheres and coated particles.

600

Coated particles oxidised

300

0

Intensity (counts)

30

60

1200

UO2 microspheres oxidised

600

0 30

60

2θ Fig. 15. Room temperature XRD pattern of sintered UO2 particles and coated particles after oxidation.

crushed coated particles shows residue of UO2 particles in the back scattering mode (Fig. 13). It can be noted that in the back scattering mode of SEM, high Z element like U will appear brighter compared to low Z element like Si and C. These crushed UO2 kernels need to be recovered and recycled. Oxidation studies of these crushed coated particles were carried out using Temperature Programmed Oxidation (TPO). TPO studies of the crushed coated microspheres were carried out using a TPDRO-1100 analyser (Thermo Quest, Italy) under the flow of O2 (5%) þHe at a gas flow rate of 10 ml min1, in the temperature range 25e850  C, at a heating rate of 10 oC min1 for studying the oxidation behaviour of the coated particles. Fig. 14 shows the TPO of the crushed coated particles. The TPO profile recorded for sintered UO2 microspheres crushed in a similar way can also be seen from Fig. 14. The profile of coated particles shows two peaks corresponding to the oxidation of carbon as well as the oxidation of UO2 to U3O8. The first peak appeared at around 550  C which shows the UO2 oxidation to U3O8. The second peak which is almost merged with

the first peak can be seen at 580  C which is due to the oxidation of carbon layers. This can be ensured from the TPO profile of sintered UO2 microspheres which shows only one oxidation peak corresponding to UO2 to U3O8. The room temperature XRD pattern of TPO residue of UO2 microspheres matched with that of U3O8 which is presented in Fig. 15. It can also be noted from the TPO profiles that the oxidation of UO2 to U3O8 in coated particles takes place at higher temperature compared to uncoated UO2 microspheres. This may be due to the easier accessibility of the oxygen to the surface of UO2 microspheres compared to crushed coated particles as some coated layers could be seen on the surface of the UO2 microspheres when observed through microscope (Fig. 16). The sieved powder in the dry recovery scheme was first oxidized in O2 atmosphere at 1000  C for 10 h to remove the carbon completely. The carbon content from these powder samples after oxidation was found to be 70 ppm. The U3O8 powder along with SiC was dissolved in 1 M HNO3 solution. Sanyal et al. [22] reported that there is no action of hot concentrated HNO3 on SiC. We have also observed that 1 M HNO3 used for dissolution of U3O8 powder was not able to dissolve any SiC residue. The residue was mainly accounted for SiC. The solution obtained was filtered through a Whatman filter paper 542 to remove any undissolved residues (SiC etc.) and washed several times with distilled water. This filtered supernatant solution was analysed by Atomic Emission Spectroscopic Technique (AES) for trace metal analysis. Table 1 shows the trace elements present in the U3O8 powder recovered from coated particles. The uranium content was analysed by Davies and Grey method [23]. The Si content present was 120 ± 23 ppm and carbon content was 10 ± 1 ppm. And also, all other metallic impurities present in the powder were within the specified limit [24,25]. The concentration of uranium in the solution was adjusted for using this recovered uranyl nitrate solution for further UO2 microsphere preparation. Good quality UO2 microspheres could be prepared from this recovered uranyl nitrate solution.

4. Conclusions The study showed that direct calcination of coated particles in air, O2 and moist O2 could not remove all the coated layers completely to recover the uranium even if the coated particles were heated up to 1400  C. It was concluded that a crushing step is essential prior to calcination step and a combination of dry and wet recycling would be ideal irrespective of extent of failure in the coating process for recovering the uranium from the coated particles. By this method, the impurity level with respect to Si and C in the recovered uranyl nitrate solution was within the specified limit and could be recycled successfully without any compromise in the quality of microsphere preparation.

236

R.V. Pai et al. / Journal of Nuclear Materials 473 (2016) 229e236

Table 1 Trace elements present in U3O8 powder recovered from the coated particles analysed by AES. Sr. No.

Elements

Amount present (ppm)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Al Be C Ca Cd Co Cr Cu Fe Mg Mn Mo Na Ni Pb Si Sn W

35 0.3 10 10 0.1 5 5 4 13 5 4 12 50 8 5 120 1 60

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

3 0.02 1 2 0.01 1 2 1 5 1 1 2 6 2 1 23 0.2 8

Acknowledgements The authors are thankful to Dr S. Kannan, Head, Fuel Chemistry Division for his keen interest and support in the work. The authors sincerely thank Dr. S.K. Sali, FCD, BARC for his help in Thermo gravimetric analysis. References [1] R.K. Sinha, S. Banerjee, in: International Conference on Roadmap to Hydrogen

Energy Organized by INAE, Hyderabad, March 4e5, 2005. [2] D. Saha, R.K. Sinha, in: Sixteenth Annual Conference of Indian Nuclear Society, INSAC-2005, Mumbai, November 15e18, 2005. [3] I.V. Dulera, R.K. Sinha, J. Nucl. Mat. 383 (2008) 183e188. [4] L. Wolf, G. Ballensiefen, W. Froehling, Nucl. Eng. Des. 34 (1975) 93e108. [5] T.D. Gulden, H. Nickel, Nucl. Technol. 35 (1977) 206e213. [6] I.V. Dulera, A. Basak, P.P. Kelkar, R.K. Sinha, Barc. News Lett. 273 (2006) 172e179. [7] D.A. Petti, J. Buongiorno, J.T. Maki, R.R. Hobbins, G.K. Miller, Nucl. Eng. Des. 222 (2003) 281e297. [8] J. Bernard Wood, Angel Sanjurjo, T. Gilbert Tong, E. Stacy Swider, Surf. Coatings Technol. 49 (1991) 228e232. [9] B.F. Myers, ORNL, DOE-HTGR 88260, ORNL-TM 10805. [10] F. William Skerjanc, T. John Maki, P. Blaise Collin, A. David Petti, J. Nucl. Mat. 469 (2016) 99e105. [11] D.H. van Dorp, J.L. Weyher, J.J. Kelly, J. Micromech Micro Eng. 17 (2007) S50eS55. €ck, H. Wotke, J. Nucl. Mater. 38 (1971) [12] G. Betz, R. Dobrozemsky, F.P. Viehbo 77e82. [13] K. Fukuda, K. Iwamoto, Mikrochim. Acta 2 (1976) 99e110. [14] V.N. Vaidya, S.K. Mukerjee, J.K. Joshi, R.V. Kamat, D.D. Sood, J. Nucl. Mater. 148 (1987) 324e331. [15] P. Grabar, S. Nikitine, J. Chim. Phys. 33 (1936) 721e741. [16] S. Brunauer, P. Emmett, E. Teller, J. Am. Chem. Soc. 60 (1938) 309e319. [17] P.K. Mollick, D. Sathiyamoorthy, P.T. Rao, V.G. Rao, Ind. Eng. Chem. Res. 50 (2011) 13313e13321. [18] G.A. Hope, I.M. Ritchie, L.A. Bruce, Corros. Sci. 16 (1976) 177e181. [19] A.C. Lea, J. Soc. Glass Technol. 33 (1949) 27e50. [20] G. Ervin Jr., J. Am. Ceram. Soc. 44 (1958) 347e352. [21] J. Roy, S. Chandra, S. Das, S. Maitra, Rev. Adv. Mater. Sci. 38 (2014) 29e39. [22] K. Sanyal, S. Dhara, S. Sanjay Kumar, N.L. Misra, P.K. Mollick, P.T. Rao, R. Venugopalan, Rajesh V. Pai, N. Kumar, S.K. Mukerjee, J.K. Chakravartty, S.K. Aggarwal, J. Radioanal. Nucl. Chem. 302 (2014) 1357e1361. [23] W. Davies, W. Gray, Talanta 11 (1964) 1203e1211. [24] R.B. Yadav, B. Gopalan, Chemical characterization of power reactor fuels at Nuclear fuel complex, in: C. Ganguli, R.N. Jayaraj (Eds.), Characterization and Quality Control of Nuclear Fuels, Allied Publishers, New Delhi, 2002, pp. 455e471. [25] S.V. Godbole, Trace metal assay of nuclear materials, chemical characterization of nuclear fuel, IANCAS Bull. 7 (2008) 169e176.