清洁家务垃圾的废物处理设施外文文献

Fusion Engineering and Design 87 2012 1040 1044 Contents lists available at SciVerse ScienceDirect Fusion Engineering and Design jo ur nal homep age Preliminary results from a detritiation facility dedicated to soft housekeeping waste X Lefebvrea b P Trabuca b K Ligera b C Perraisa b S Tostia c F Borgognonia c A Santuccia c aJET EFDA Culham Science Centre Abingdon OX14 3DB UK bCEA DEN Cadarache DTN STPA LIPC F 13108 Saint Paul lez Durance France cAssociazione ENEA Euratom sulla Fusione C R ENEA Frascati Via E Fermi 45 I 00044 Frascati RM Italy a r t i c l e i n f o Article history Received 7 September 2011 Received in revised form 15 February 2012 Accepted 16 February 2012 Available online 20 March 2012 Keywords Housekeeping Detritiation Tritium Membrane reactor a b s t r a c t Nuclear waste management has to be taken into account for fusion machine in tritium experimentations Soft housekeeping waste is produced during both operating and dismantling phases and is contaminated by tritium under reduced HT and oxidized HTO forms At CEA Cadarache a lab scaled facility has been built for soft housekeeping detritiation The tritiated gas exhausted from the process described above is foreseen to be treated by a tubular Pd Ag membrane reactor for gaseous tritium recovery Since this membrane reactor uses hydrogen as swamping gas the compatibility toward explosive hazard has to be taken into account Then this work presents a double objective A fi rst study is presented in order to identify the best conditions for the declassifi cation of soft housekeeping waste without tritium recovery Experiments carried out at 120 C are not effi cient enough and do not allow one to choose the most effi cient carrier gas Some other tests are being currently performed at higher temperatures 150 C Moreover due to safety issues the use of air has to be avoided during membrane reactor implementation phase Preliminary results obtained with hydrogen hazard free carrier gases are also presented 2012 Elsevier B V All rights reserved 1 Introduction Soft housekeeping waste is produced during both operating and dismantling phases of fusion reactor and is contaminated by tritium under reduced HT and oxidized HTO forms For exam ple concerning ITER project the production of 1600 3800 tons of housekeeping and process waste is foreseen and can be split as follows 1 20 very low level waste 75 short lived intermediate level waste 5 long lived intermediate level waste The higher the waste class the more expensive and stringent the storage 2 Moreover tritiated waste induces the outgassing particularity compared to usual nuclear waste Then the imple mentation of a detritiation facility could be interesting to declassify this kind of waste In this work the lab scaled detritiation facility designed by CEA Cadarache is presented as well as its future coupling with a mem brane reactor from ENEA Frascati 3 Corresponding author at CEA Cadarache DTN STPA LIPC B t 208 F 13108 Saint Paul lez Durance France Tel 33442256415 fax 33442257287 E mail address xavier lefebvre cea fr X Lefebvre Consequently this study consists of two main parts fi rst of all the technical feasibility study is performed with different carrier gases at different temperatures and fl ows in order to optimize the experimental conditions for detritiation effi ciency Moreover in order to avoid the explosive hazard due to the use of pure hydro gen as a swamping gas in the membrane reactor the study of the infl uence of the type of carrier gas has also been carried out Then the most appropriate carrier gas presenting the best compromise between safety and effi ciency has been determined 2 Presentation of the experimental facilities The housekeeping samples are milled nitrile gloves contami nated by tritium and potentially by beryllium They are placed in a glass reactor heated at a temperature lower than its melting point The carrier gas injected in the glass reactor removes the tritium that is fi nally trapped in liquid bubblers and measured via liquid scintillation see Fig 1 In a second step the tritiated gas exhausted from the process instead of going into the liquid traps is foreseen to be treated by a membrane reactor consisting of a tubular Pd Ag membrane fi lled with a Ni based catalyst as shown in Fig 2 4 12 In this way it is possible to recover tritium under HT form which can be further reused in the fusion fuel cycle The carrier gas containing tritium coming from the glass reactor is fed into the membrane lumen 0920 3796 see front matter 2012 Elsevier B V All rights reserved doi 10 1016 j fusengdes 2012 02 076 X Lefebvre et al Fusion Engineering and Design 87 2012 1040 10441041 Fig 1 Schematic representation of the housekeeping detritiation facility Fig 2 Schematic representation of the catalytic membrane reactor while a hydrogen protium stream is injected in the shell side in counter current mode Since the Pd Ag membrane is permeable only to hydrogen isotopes the exchange reaction occurs over the Ni based catalyst HTO H2Ni 400 C HT H2O Tritiated gas HT formed by isotope swamping over the catalyst permeates the membrane to be fi nally recovered in the shell side of the membrane reactor Within the scope of the further study for analytic and waste management purposes the gaseous tritium will be oxidized in an oven containing a cartridge of copper oxide and the tritiated water formed is frozen in a liquid nitrogen cold trap The samples used in this study were milled nitrile gloves pro vided by JET fusion machine The initial mass activity of the samples has been assessed to 20 kBq g i e very close to the British ILW LLW threshold 12 kBq g 13 However due to the expected hetero geneity of the initial activity of the samples the residual activity has been always measured by liquid scintillation counting at the end of the experiment under the operating mode schematized in Fig 3 First of all a well known housekeeping quantity between 12 and 15 g was placed in the glass reactor see Fig 1 As the sam ples might contain some beryllium traces the sample preparation was made in a dedicated room for Be handling Then the air tight reactor was placed in the oven and a leak test was performed by reading the pressure variation The temperature of the oven was regulated and the carrier gas was injected in the system The demineralized water in the bubblers of a MARC 7000 device was changed every 1 or 2 h and a given quantity of this recovered water is analyzed by liquid scintillation counting At the end of the experiment the activity of the residue was also measured The residue housekeeping waste sample was burnt to ashes in a Hastelloy reactor placed in an oven heated up to 500 C under laboratory air for 3 h Fig 3 The tritium trapped in a MARC 7000 device was also measured via liquid scintillation counting The detritiation yield is calculated using the following equation Y t 100 n i 1ai N i 1ai ares with Y t the detritiation yield ai 3H removed activity as a func tion of time Bq ares 3H activity of the residue Bq n sampling number at t time dimensionless N sampling number at the end of the experiment dimensionless Fig 3 Schematic representation of the residue calcination facility 1042X Lefebvre et al Fusion Engineering and Design 87 2012 1040 1044 8 9 Ar 2 O2 Ar 5 H2 6 7 Ar 25 H2O Ar 40 H2O Air 3 4 5 Detritiation yield 0 1 2 0 5101520 Effective experimental time hours Fig 4 Soft housekeeping detritiation yield as a function of effective experimental time for different carrier gases T 120 C and Q 300 N mL min As the samples come from JET a residual activity of 12 kBq g of the waste has been considered for establishing the effectiveness of the detritiation process In fact such a value of waste activity is the threshold between intermediate level ILW and low level waste LLW according to the British regulation 13 3 Results and discussion The experiments whose results are presented here have been carried out during 3 days which correspond to 18 h of experiments 6 h being the effective working time in each day In Figs 4 6 the X axis represents the effective experimental time and the Y axis the detritiation yield The night gaps between two experimental days have not been considered in the graphs The results presented here only concern the detritiation part of the study using the process described in Fig 1 i e without the membrane reactor coupling 3 1 Infl uence of the nature of the carrier gas In order to identify the respective effi ciencies of the carrier gases several atmospheres have been tested 60 40 50 Ar 2 O2 120 C Ar 2 O2 150 C Ar 25 H2O 120 C Ar 25 H2O 150 C 20 30 Detritiation yield 10 0 50101520 Effective experimental time hours Fig 5 Detritiation yield as a function of effective experimental time with Ar 2 O2and Ar 25 H2O Q 300 N mL min Table 1 Housekeeping initial after 18 h experiment tritium activities and detritiation yield at 120 C and 300 N mL min gas fl ow Carrier gasai kBq g ares kBq g Y Ar 2 O223 0 21 3 7 6 Ar 5 H223 0 22 2 3 5 Ar 25 H2O 29 7 28 2 5 0 Ar 40 H2O 29 4 27 9 5 4 Air 10 2 9 4 8 1 Low oxidative Ar 2 O2 Low reductive Ar 5 H2 Intermediate oxidative air Humid gases Ar x H2O x 25 and 40 The gas feed fl ow has been fi xed at 300 N mL min which cor responds to the nominal workfl ow of the membrane reactor In order to avoid the housekeeping burning during the experiment the temperature has been fi xed at 120 C Fig 4 shows the behavior of the detritiation yield of soft house keeping as a function of the effective experimental time First of all is it possible to observe that none of the 5 plots shown in Fig 4 reaches a plateau and all obey a quasi linear regime which tends to demonstrate that the experimental time was not optimized for these experimental conditions The dark dots diamonds circles triangles and squares of the graph establishes that the detritiation yield did not exceed 8 what ever the carrier gases employed at 120 C during 18 h effective experimental time The amount of water in argon does not seem to have an infl uence on the detritiation yield see dark circles and triangles in Fig 4 Table 1 summarizes the experiments carried out at 120 C and 300 N mL min carrier gas fl ow First of all it is visible that the expected heterogeneity of initial activities of the samples is accu rate Indeed the initial activities vary from 10 to 30 kBq g It is also to notice that none of the Ar based carrier gases allows the declassifi cation to LLW after 18 h Regarding the sample detriti ated under air since its initial activity 10 kBq g was slightly lower than LLW ILW threshold it is not possible to conclude if that these experimental conditions are suffi cient to declassify such sample as LLW 16 18 20 450 NmL min Air 300 NmL min Air 10 12 14 4 6 8 Detritiation yield 0 2 50101520 Effective experimental time hours Fig 6 Detritiation yield as a function of effective experimental time with air T 120 C X Lefebvre et al Fusion Engineering and Design 87 2012 1040 10441043 Table 2 Housekeeping initial after 18 h experiment tritium activities and detritiation yields at 120 and 150 C 15 h for Ar 2 O2at 150 C Carrier gasT C ai kBq g ares kBq g Y Ar 2 O2120 23 0 21 3 7 6 Ar 2 O2150 17 3 7 6 56 1 Ar 25 H2O 120 29 7 28 2 5 0 Ar 25 H2O 150 18 8 14 3 23 6 3 2 Infl uence of the temperature The previous section has shown that the initial experimental conditions T 120 C and 300 N mL min gas fl ow were too mild to declassify the housekeeping waste Thus in order to optimize the operating conditions and anyway to avoid the combustion of the housekeeping sample experiments have been carried out at 150 C At this moment of the study we were only able to compare the infl uence of temperature with Ar 2 O2and Ar 25 H2O the experiments using the other carrier gases being currently carried out Fig 5 shows the comparison of the detritiation yields as a func tion of time for two different carrier gases at 120 C dark dots and 150 C light dots It is visible that a reasonable temperature increase of 30 C leads to a considerable rise of the detritiation yield Indeed for both Ar 2 O2 diamonds and Ar 25 H2O triangles the detritiation yield has increased nearly fi vefold 5 0 23 6 for Ar 25 H2O 7 6 56 1 for Ar 2 O2 It is also interesting to observe that a slope break occurs after 15 h for the 150 C Ar 25 H2O experiment enlightening the beginning of equilibrium between tritium removal and residual activity of the sample The residual activities listed in Table 2 show that 15 h have been enough to declassify to LLW the soft housekeeping treated by Ar 2 O2at 150 C On the contrary using Ar 25 H2O at 150 C does not allow to declassify the waste yet Nevertheless if the behavior of this experiment remained the same as its end yield increase of approx 0 75 h the declassifi cation to LLW might be reached after extra 13 h experimental time 3 3 Infl uence of the carrier gas fl ow This study was useful for two different reasons First of all deter mining the infl uence of the residence time on the detritiation yield allows one to assess the fl ow within an industrial detritiation facil ity and then the cost of the gas consumption especially coupled with a membrane reactor where air cannot be used for safety rea sons Moreover this study has also allowed one to determine the contamination mode of these housekeeping by tritium Indeed a detritiation yield increasing with carrier gas fl ow would have indi cated that the sample was contaminated by tritium on its surface while a constant detritiation yield would have shown that the most part of tritium had penetrated the waste At this time of the study only experiments with air have been carried out More experiments performed with other carrier gases are foreseen to confi rm the following conclusions which can be only considered up to now as tendencies Fig 6 shows the behavior of the detritiation yield as a function of time for two different air fl ows It is visible that the beginning of the curves is quasi superposed for the 6 fi rst hours which corresponds to the fi rst day of experimentation It is also remarkable to observe that the following black diamond shown a jump probably due to a tritium degassing during the night This phenomenon occurred again during the second night stop see the next to last black dia mond in Fig 6 This is only visible on the 450 N mL min experiment because that one has been carried out before the 300 N mL min one Table 3 Housekeeping initial after 18 h experiment tritium activities and detritiation yield at 120 C for different gas fl ows Carrier gasFeed fl ow N mL min ai kBq g ares kBq g Y Air 300 10 2 9 4 8 1 Air450 13 1 11 4 13 5 and the natural tritium outgassing occurs much more easily with fresh samples Nevertheless these two points excluded the slope is higher for higher feed fl ows 0 32 h for 300 N mL min 0 46 h for 450 N mL min Despite that Fig 6 establishes that the higher the feed fl ow the higher the detritiation yield Furthermore the gap between the detritiation yields obtained using these two feed fl ows tends to demonstrate that the tritium contamination is mainly on the waste surface Table 3 sums up the results obtained for the detritiation of soft housekeepings under air at different gas fl ows As already mentioned in Section 3 1 due to the mild initial activities it is impossible to conclude about the possibility to declassify the waste treated at 300 N mL min This conclusion is also valid for a 450 N mL min gas fl ow Indeed the initial activity of the sample is clearly too close to the ILW LLW threshold 3 4 Safety considerations Considering that the membrane reactor which will be coupled with the existing detritiation facility uses hydrogen protium as a swamping gas and that the effi ciency of the isotopic exchange depends on the temperature the optimal temperature has been determined at around 400 C it appears essential to assess the explosive hazard between hydrogen and Ar 2 O2which is the most effi cient carrier gases among those listed above Fig 7 shows a detail of the fl ammability limits of a ternary nitro gen oxygen hydrogen mixture as a function of the temperature The fl ammability regions have been determined by the CAST the ory 14 ambient temperature 70 F excepted In this case up to 400 C 752 F corresponding to the working temperature of the membrane the fl ammable region is never crossed Fig 7 Detail of fl ammable region limits of a ternary O2 H2 N2mixture as a function of the temperature F 14 1044X Lefebvre et al Fusion Engineering and Design 87 2012 1040 1044 4 Conclusion and prospects In this study the detritiation effi ciencies of low activity soft housekeeping by different carrier gases at different temperatures and fl ows have been assessed In the future it appears essential to perform the same study on much higher sample amounts and much higher tritium contaminated housekeeping waste as expected dur ing the ITER project The initial goal of this study was to demonstrate the tech nical feasibility of the detritiation of soft housekeepings under mild conditions i e without housekeeping combustion After 18 h experiments the 120 C initial reference temperature is too mild to declassify soft housekeeping waste weakly contaminated but also to establish a hierarchy concerning the most effi cient carrier gases even if the oxidative gases give higher detritiation yields Thus the reference temperature was to be slightly increased up to 150 C The fi rst results of experiments carried out at this new refer ence temperature reveal that this temperature increasing leads to a fi vefold detritiation yield increasing leading to the possibility to declassify waste treated with Ar 2 O2 Further experiments currently carried out may confi rm the sur face contamination of the waste by tritium as well as the infl uence of the feed fl ow on the detritiation yield The most appropriate carrier gas to be employed for housekeep ing detritiation and gaseous tritium recovery by the membrane reactor designed by ENEA is