- Email: [email protected]
Experimental research on the radioactive dust in the primary loop of HTR-10
Nuclear Engineering and Design 324 (2017) 372–378
Contents lists available at ScienceDirect
Nuclear Engineering and Design journal homepage: www.elsevier.com/locate/nucengdes
Experimental research on the radioactive dust in the primary loop of HTR10
MARK
⁎
Feng Xiea, Jianzhu Caoa, Xiaogui Fenga, Xuegang Liua, , Jiejuan Tonga, Haitao Wanga, Yujie Donga, Zuoyi Zhanga, S.K. Loyalkab a Institute of Nuclear and New Energy Technology, Collaborative Innovation Center of Advanced Nuclear Energy Technology, Key Laboratory of Advanced Reactor Engineering and Safety of Ministry of Education, Tsinghua University, Beijing 100084, China b Nuclear Science and Engineering Institute, University of Missouri, Columbia, MO 65211, United States
A R T I C L E I N F O
A B S T R A C T
Keywords: HTR-10 Radioactive dust Fission and activation products Dust concentration Particle size distribution
Presence of radioactive dust, and also potential for its release, are considered important issues as regards the safety of high temperature gas cooled reactors, especially the pebble bed types. To obtain data and insights into the quantity and characteristics of such dust, a sampling loop (with stainless steel powder filter elements to collect the dust) has been built in the helium purification system of HTR-10 enabling sampling of radioactive dust circulating in the primary loop of HTR-10. Two sampling experiments were conducted during the operation of HTR-10 in 2015. The concentration of the radioactive dust in the primary loop was deduced from the dust deposited on the filters, and was estimated to be 5.57(17) μg/m3STP and 1.97(8) μg/m3STP, respectively. The nuclides present in the dust have been identified by their γ spectra. After the removal of dust from filter elements, the particle size distribution was measured with an optical microscope. In the first experiment, the average particle diameters of the radioactive dust were determined as 43.0 μm and 13.6 μm captured respectively, by the filter elements of 80 μm and 50 μm pore size. In the second experiment, the average particle diameters were 62.9 μm and 11.3 μm as captured respectively, by similar filter elements. The data show existence of large particles in the primary loop of HTR-10, generated most likely by abrasion.
1. Introduction With the development of the high temperature gas-cooled reactors (HTGRs), the very high temperature reactor (VHTR) has been considered as a candidate for the six proposed Generation IV concept reactors by the Generation IV International Forum (GIF) (NERAC and GIF, 2002). With the use of fuel elements embedded with tristructural-isotropic (TRISO) coated particles, the inherent safety performance of the HTGRs has attracted wide attention. The high thermal efficiency and the high output temperature which can be used for hydrogen production and process heat, bring large commercial space for the future development of HTGRs (Zhang and Yu, 2002). There are, however, several safety issues that have been identified previously through operation of research reactors. These include the local high temperature in the core leading to the failure of TRISO coated particles, the release of the radioactive dust in the depressurization accident, the contamination due to the radioactive dust in the primary loop, etc. (Bäumer et al., 1990; Moormann, 2008a,b). It has been noted also that dust related problems are more important in the pebble bed type
⁎
reactors than in the prismatic type reactors (Humrickhouse, 2011). In the pebble-bed HTGRs, the dust is thought to be generated by several mechanisms, including the abrasion among fuel elements, and friction between fuel elements and other graphite structures or pipelines when the fuel elements cycled (Luo, 2004; Moormann, 2008b; Kissane, 2009). Several papers have been published regarding the behavior of dust in HTGRs, including the generation, characterization, transport, coagulation, aggregation, deposition, resuspension, etc. Humrickhouse (2011) summarized the dust safety issues in HTGRs and indicated that the dust related research was urgently needed for the development of pebble bed HTGRs on topics such as the dust distribution under normal operation, dust generation, dust-fission product interaction, etc. Cogliati et al. (2011) surveyed the available literature on graphite dust production and measurements in pebble bed reactors and concluded that there was significant uncertainty on the severity of the dust production and its consequences in pebble bed reactors. Troy et al. (2012, 2015) and Shen et al. (2016) investigated the characteristics of graphite dust particles experimentally. Simones et al. (2011), Simones and Loyalka (2015) measured the charge-size distributions of
Corresponding author at: Room 320, Energy Science Building A, Institute of Nuclear and New Energy Technology, Tsinghua University, Beijing 100084, China. E-mail address: [email protected] (X. Liu).
http://dx.doi.org/10.1016/j.nucengdes.2017.09.015 Received 21 February 2017; Received in revised form 21 August 2017; Accepted 19 September 2017 0029-5493/ © 2017 Elsevier B.V. All rights reserved.
Nuclear Engineering and Design 324 (2017) 372–378
F. Xie et al.
was bulky and difficult to dismantle, an experimental system for accurately measuring the radioactive dust in the primary loop of HTR-10 was designed and built. Though details of the radioactive dust sampling loop of HTR-10 have been presented in Xie et al. (2015a,b), it is necessary to give a brief description here. As shown in Fig. 1, the main sampling loop includes two electric check valves (KBE03 AA001 and KBE03 AA002), one electric control valve (KBE03 AA003), two thermometers (KBE03 CT001 and KBE03 CT002), two pressure gauges (KBE03 CP001 and KBE03 CP003), one flowmeter (KBE03 CF001), one differential pressure transmitter (KBE03 CP002), and five manual globe valves (KBE03 AA004 ∼ KBE03 AA008). Primary features for this dust sampling loop are as follows:
graphite aerosol particles and coagulation of charged aerosols pertinent to HTGRs. Moormann (2008a,b), Kissane (2009), and Lind et al. (2010) discussed the transport of the radioactive dust which can be an important source term in pebble bed reactors. Gutti and Loyalka (2009), Boddu et al. (2011), Barth et al. (2013), and Peng et al. (2016) experimentally and/or numerically investigated the deposition behavior of dust in HTGRs. Kazuhiro et al. (1992), Stempniewicz et al. (2008), and Kissane et al. (2012) studied the resuspension behavior of dust in HTGRs. There have been, however, very few experimental investigations regarding the radioactive dust directly in pebble bed reactors. The only available data on such dust are from Arbeitsgemeinschaft Versuchsreaktor (AVR) and Thorium Hochtemperatur Reaktor (THTR) (Bäumer et al., 1990; Moormann, 2008b; Fachinger et al., 2008). There is a need to verify conclusions from theoretical calculations and nonradioactive dust experiments against data from the radioactive dust experiments in actual reactors. Recently, several experiments related to the source terms of the 10 MW high temperature gas-cooled reactor (HTR-10) have been conducted. These include investigations of the radioactive dust, H-3, and C-14 in the primary loop, post irradiation graphite spheres from the reactor core, etc. (Xie et al., 2015a,b; Xu et al., 2017; Wang et al., 2014; Liu et al., 2017; Li et al., 2017). Previous measurements also indicated the existence of radioactive dust in the primary loop of HTR-10 (Xie et al., 2013, 2015a,b). Accordingly, an experimental sampling loop has been designed and built in the helium purification system of HTR-10. In the present study, the radioactive dust in the primary loop of HTR-10 was investigated with use of the above sampling loop when HTR-10 was restarted in 2015. The radioactive dust has been collected with a sampling filter. The concentration of the dust in the primary loop was estimated with use of the coolant flow data. The types of solid fission and activation products which were absorbed or present in the dust were determined with a γ spectrometer (GC3018, High-purity Germanium Detectors, from CANBERRA Industries Inc.) in the radiochemistry lab in the Institute of Nuclear and New Energy Technology (INET), Tsinghua University. The counting rates of typical nuclides, including Co-60, Cs-137, I-131, etc., were measured and compared with each other. The particle size distributions of the dust from the filter elements were obtained through imaging with an optical microscope. The data reported can shed light on the behavior of fission/activation products and radioactive dust in HTGRs, and would aid in modeling and safety studies.
(1) A small and easily disassemblable sampling filter was designed, as shown in Fig. 2. It can contain as many as 8 filter elements with different pore sizes, from 1 μm, 3 μm, 5 μm, 10 μm, 20 μm, 50 μm, and 80 μm, and can separate the radioactive graphite dust according to the particle diameter. (2) Two electric check valves (KBE03 AA001 and KBE03 AA002) can isolate the experimental system from the helium purification system effectively. The electric control valve (KBE03 AA003) before the dust filter can adjust the flow ratio between the experimental loop and the main circuit. (3) Two thermometers (KBE03 CT001 and KBE03 CT002), two pressure gauges (KBE03 CP001 and KBE03 CP003), one flowmeter (KBE03 CF001), and one differential pressure transmitter (KBE03 CP002), can record the temperature, the absolute pressure, the flow capacity, and the pressure drop across the sampling filter, respectively. The operation for the radioactive dust sampling loop can be divided into preparation, sampling, and shutdown phases. In the preparation phase, the electric check valves (KBE03 AA001 and KBE03 AA002) at the entrance and exit of the sampling loop are closed, and the manual globe valves in the sampling loop are open. The whole sampling loop is pumped to remove the air inside from KTT system whose function is to pump the air out to the system and to relieve the high pressure in the pipe, if necessary. The preparation phase is completed when the pressure of the system is below 100 Pa and the manual globe valve is closed. In the sampling phase, the radioactive dust sampling loop will be connected to the helium purification system by opening of the electric check valves (KBE03 AA001 and KBE03 AA002). The electric control valve (KBE03 AA003) can be adjusted to the predetermined flow capacity, and the sampling filter would collect the dust particles for several days or weeks. In the shutdown phase, the sampling loop would be separated from the main circuit by closing of the electric check valves (KBE03 AA001 and KBE03 AA002). After the pressure in the sampling loop drops to normal conditions due to opening of the manual globe valve (KBE03 AA008) and the temperature falls to the room temperature, the sampling filter can be dismantled and sent to the radiochemistry lab for further analysis.
2. Experimental setup The HTR-10 is a helium cooled, graphite-moderated, and thermal neutron spectrum test reactor, which was designed and built by INET, Tsinghua University, China (Wu et al., 2002). It reached its criticality in December 2000, ran up to full power operation in January 2003, and was shut down in July 2007. It was restarted at the end of November 2014 (Wei et al., 2016). During the shutdown stage of HTR-10, an experimental sampling loop in the primary system aiming to study the behavior of radioactive dust was designed. It was approved by the State Bureau of Nuclear Safety of China in June 2012. In December 2013, the experimental loop with a sampling filter was built at the entrance of the helium purification system of HTR-10, as shown in Fig. 1. The helium purification system contains two normal purification lines and one dehumidification line. Each normal line contained a dust filter, a copper oxide bed, a molecular sieve adsorber, and a low temperature adsorber. This system is very important for the reactor operation since it can purify a bypass helium from the primary coolant system and maintain the quality of the primary helium by continuously reducing the quantity of chemical impurities and removing the dust, fission products and activation products. Parallel to the dust filter in the helium purification system, which
3. Results After a long term shutdown from 2007 to 2014, HTR-10 was restarted at the end of November 2014, and was operated in a power stage in March 2015. During June 1st, 2015, to August 6th, 2015, two sampling experiments about the radioactive dust in the primary loop were conducted. The operational thermal power of HTR-10 was about 2.9 MW compared to the rated thermal power of 10 MW. The primary helium temperature at the reactor inlet and outlet were respectively, ∼185 °C and ∼575 °C. The primary helium pressure was about 2.1 MPa and the flow rate was about 1.39 t/h. Note that the rated values for the primary helium temperature at the reactor inlet and outlet, primary helium pressure, and primary helium flow rate at the full power operation of HTR-10 were 250 °C, 750 °C, 3.0 MPa, and 4.32 t/h, respectively. The concentration of the radioactive dust, the types of 373
Nuclear Engineering and Design 324 (2017) 372–378
F. Xie et al.
Fig. 1. Flowchart of the radioactive dust sampling loop with the schematic diagram of the primary loop of HTR-10.
Fig. 2. Sampling filter containing four filter elements (the left one) and comparison of the filter element after the sampling process (the middle one) and after the decontamination process (the last one).
flow was 2911.2 m3STP totally. Fig. 2 shows a filter element after the sampling which clearly identified the radioactive dust on the surface. In order to obtain the mass of the radioactive dust, several steps needed to be implemented. First, the filter element was immersed into the ethanol, and the dust particle separated from the filter element by the ultrasonic separation. Second, the above process was repeated many times and all suspensions were mixed together. Finally, ethanol was evaporated by heating so the mass of the dust could be obtained if it was present. After removing the radioactive dust into the liquid, the weight and particle size distribution of the radioactive dust could be measured. However, only the weight of radioactive dust in the filter element with 80 μm pore size in the first sampling experiment was measurable, and it was found to be 0.0162(5) g. Pi et al. (2014) studied filtration characteristics of sintered stainless steel powder filter element experimentally, and indicated that the filter element with 80 μm pore size can reach a high filtration efficiency ∼95% even with the particle diameter around 1 μm. Considering the overwhelming majority of radioactive dust with bulk mass was retained in the filter element with 80 μm pore size, the concentration of the radioactive dust in the primary loop of HTR-10 can be deduced as 5.57(17) μg/m3STP. This is very close to the average dust concentration of AVR, ∼5 μg/m3STP. In the second sampling experiment, three filter elements with the pore size of 80 μm, 50 μm, and 20 μm were put into the sampling filter in sequence. With the same process as that in the first sampling experiment, only the mass of the radioactive dust in the filter element with 80 μm pore size was measurable and was found to be 0.0254(10) g. With the accumulated helium flow data which was 12,926.2 m3STP in the second sampling experiment, the concentration of the radioactive dust in the primary loop of HTR-10 was deduced to be 1.97(8) μg/m3STP.
solid fission and activation nuclides adsorbed in the dust, and the particle size distribution of the dust in the filter element are determined from the experimental measurement. For convenience, the filter elements used in two sampling experiments are denoted as follows: The four filter elements used in the first sampling experiment are denoted as 1#1, 1#2, 1#3 and 1#4, corresponding to the filter element with pore sizes of 80 μm, 50 μm, 20 μm, and 5 μm, respectively. The three filter elements used in the second sampling experiment are denoted as 2#1, 2#2 and 2#3, corresponding to the filter element with pore sizes of 80 μm, 50 μm, and 20 μm, respectively. The purpose of removing the filter element of 5 μm pore size in the second experiment was to reduce the resistance pressure of the sampling loop. With a portable γ dose rate spectrometer, the recording value from the filter element 1#4 is close to the background. 3.1. Concentration of the radioactive dust in the primary loop of HTR-10 Table 1 lists the typical values of main operational parameters of the radioactive dust sampling loop of HTR-10 during its operation. In the first sampling experiment, four filter elements with the pore sizes of 80 μm, 50 μm, 20 μm, and 5 μm were put into the sampling filter in sequence, which means the radioactive dust in the primary helium would go through the filter element with biggest pore size and then to the smaller one. Though several breaks due to the preplanning or unexpected events occurred during the first sampling process, the accumulated helium Table 1 Main operational parameters of the radioactive dust sampling in HTR-10. Item
Unit
1st Experiment
2nd Experiment
Helium pressure Sampling loop inlet temperature Sampling loop outlet temperature Flow rate
MPa °C °C m3/h STP
2.1 ∼100 ∼50 ∼15
2.1 ∼107 ∼70 ∼33
3.2. Types of solid fission and activation nuclides in the dust Fig. 3 shows four sintered stainless steel powder filter elements used in the first sampling experiment and the corresponding γ spectra from the direct measurement of these filter elements with the γ spectrometer. 374
Nuclear Engineering and Design 324 (2017) 372–378
F. Xie et al.
Fig. 3. Four sintered stainless steel powder filter elements in the first experiment and the corresponding γ spectra.
Fig. 4. Assignment of nuclides from the γ spectra with the radioactive dust from 2#1 filter element.
Along with the direction of the helium flow, the first sintered stainless steel powder filter element 1#1 clearly captured remarkable amount of dust on the surface. While there was no visible dust on other filter elements, the γ spectra clearly showed the presence of fission and activation nuclides which are considered to be adsorbed in the dust. Fig. 4 shows the nuclides identified in the radioactive dust of HTR-10. Table 2 lists the nuclides with the corresponding half lives and peaks. This is the first time, that the solid fission products, including I-124, I-131, Cs-137, Ba-140, La-140, Eu-152 and Hf-181, and the solid activation products, including Cr-51, Mn-54, Fe-59, Co-57, Co-58, Co-60 and Se-75, including short lived nuclides and long lived nuclides, have been experimentally shown to exist in the primary loop of HTR-10. Further, the unexpected nuclide Hg-203 also was detected in the γ spectra. The Hg203 can be produced by a neutron induced n-α reaction from Pb-206 and neutron capture from Hg-202. In the structure graphite materials of HTR-10, Hg and Pb elements were determined experimentally. Thus,
Hg-203 appears to be an activation product. Table 3 and Table 4 list counting rates for typical nuclides from the measurement of γ spectra of filter elements in the first and the second experiment, respectively. For short lived nuclides, like I-131 and Hg203, the measurement time affected the counting rates significantly. In Table 3, the counting rate of I-131 and Hg-203 exhibited the similar decreasing behavior along with the cascade sequence filter elements since the measurement time were quite close. For the relatively long lived nuclides, like Co-60 and Cs-137, the measurement time had little influence on counting rates. Compared with the I-131 and Hg-203, the counting rates of the Co-60 and Cs-137 decreased much faster with the cascade placement of filter elements, which implied the transport behavior were very different. However, the counting rates of I-131 and Hg-203 in 1#4 filter element in Table 2 were rather large even considering the effect of the spontaneous decay. A possible reason may be that the I-131 and Hg-203 could be transported not only by the dust but 375
Nuclear Engineering and Design 324 (2017) 372–378
F. Xie et al.
shows the dust particles in the suspension from 1#1, 1#2, 2#1, and 2#2 filter elements with an optical microscope. With a simple statistical method, the average particle diameters of the radioactive dust are determined as 43.0 μm, 13.6 μm, 62.9 μm and 11.3 μm in the suspensions from 1#1, 1#2, 2#1 and 2#2 filter elements, respectively. Due to too little and relatively small sized dust in 1#3, 1#4 an 2#3 filter elements, the measurement of the average particle diameter of the radioactive dust was not successful. The particle size distribution in the 1#2 filter element has been deduced based on about 5000 particles observed in the suspension, as seen in Fig. 6. It appears that the particles with diameter between 5 μm and 10 μm are the dominant part in the suspension of the 1#2 filter element. Since much fewer particles (< 1000) were recorded in the observation range of suspensions of 1#1, 2#1 and 2#2 filter elements, we do not report the particle size distribution in these filter elements.
Table 2 Nuclides determined from the γ spectra with corresponding half lives and peaks. Nuclides
Half lives
Peaks (keV)
Cr-51 Mn-54 Fe-59 Co-57 Co-58 Co-60 Se-75 I-124 I-131 Cs-137 Ba-140 La-140 Eu-152 Hf-181 Hg-203
27.704 d 312.7 d 44.63 d 270.9 d 70.8 d 5.271 a 119.78 d 4.18 d 8.04 d 30.17 a 12.789 d 40.22 d 13.6 a 42.39 d 46.6 d
320.0 834.9 142.7, 192.3, 1099.2, 1291.4 121.9, 136.2 511.1, 810.6 1173.2, 1332.6 121.9, 136.2, 264.3, 279.4 511.1, 602.2 284.5, 364.5, 636.9 661.2 162.7, 304.7, 537.1 328.5, 486.9, 815.5 121.9, 344.3, 1408.0 133.2, 344.3, 482.0 279.4
4. Discussion As stated earlier there have been very few experimental data available about the radioactive dust in the pebble bed reactors. Bäumer et al. (1990) reported that the average concentration of dust in the primary loop of AVR was 5 μg/m3STP, but it could be as high as 2000 μg/ m3STP at the startup or shutdown stage of the reactor. In HTR-10, this was the first attempt to investigate the radioactive dust in the primary loop. In the first sampling experiment, though only the dust in the first filter element 1#1 was measureable, we believe that the most of the mass of the dust was obtained as the sintered stainless steel powder filter element possessed rather high filtration efficiency (Pi et al., 2014). During the first sampling experiment, the sampling loop experienced the startup and shutdown stages of the HTR-10. Besides, when the experimental loop was carried on the trial experiment, the dust in the primary loop was also gathered but the flow of the helium was not cumulatively added in the dust concentration calculation. Thus, the dust concentration value of 5.57(17) μg/m3STP in the primary loop of HTR-10 is an integrated and conservative result. In the second sampling experiment, the accumulated helium flow was 12,926.2 m3STP, which was much higher than that in the first experiment and the concentration of the radioactive dust in the primary loop of HTR-10 can be deduced as 1.97(8) μg/m3STP. Based on our measurements, the average dust concentration in the primary loop of HTR-10 is slightly smaller than the average value of dust concentration in AVR. Given the same filtration process, the counting rate of typical nuclides in each filter element exhibits individual features. Co-60 is thought to have a strong correlation with the graphite dust since it was produced from the activation reactions of impurities of Co-59 and/or Ni-60 in dust materials. The experimental average values for the Co and Ni mass fractions in the graphite spheres of HTR-10 have been investigated (Li et al., 2017). Cs-137, as a fission product, could diffuse from the fuel element and then adsorb into the dust. I-131 is also a fission product but it exhibits a different transport behavior. It can absorb into the dust and then circulate in the primary loop. Another possible way is that it can transport with the helium gas in the atomic/ molecular form. Hg-203 is a special nuclide, which behaves in a unique way. A possible explanation may that Hg-203 can transport in the atomic/molecular form even at not high temperature and is not easy to condense in the filter element as I-131. Another possibility may be the interaction between Hg-203 and the fine particle is much stronger than with the large particle. In any case, the transport behavior of Hg-203 needs to be studied further. From the γ spectra, Co-60 plays an important role in the specific activity of the dust. With a standard source of Co-60, the specific activities of Co-60 on the dust in HTR-10 are conservatively estimated as about 3.3 × 104 Bq/g and 6.5 × 103 Bq/g from these two experiments respectively. These values are lower than dust experimental results of AVR, 2.0 × 105–8 × 106 Bq/g for Co-60 (Gottaut and Krüger, 1990). The specific activities of Cs-137 on the dust in HTR-10 are
Table 3 Counting rates of typical nuclides from γ spectra of filter elements in the first experiment. Measurement Time
2015/7/6
2015/7/2
2015/7/3
2015/6/26
Nuclides
Peak (keV)
1#1 (CPS)
1#2 (CPS)
1#3 (CPS)
1#4 (CPS)
Co-60 Co-60 Cs-137 I-131 Hg-203
1332.5 1173.2 661.2 364.5 279.4
2.2521(52) 2.4874(54) 0.0149(4) 0.0333(6) 0.0197(5)
0.1366(12) 0.1522(13) 0.0018(1) 0.0135(4) 0.0789(9)
0.0234(3) 0.0252(3) 0.0013(1) 0.0028(1) 0.0134(2)
0.0208(3) 0.0224(3) 0.0015(1) 0.0122(2) 0.0222(3)
Table 4 Counting rates of typical nuclides from γ spectra of filter elements in the second experiment. Measurement Time
2015/8/18
2015/8/19
2015/8/19
Nuclides
Peak (keV)
1#1 (CPS)
1#2 (CPS)
1#3 (CPS)
Co-60 Co-60 Cs-137 I-131 Hg-203
1332.5 1173.2 661.2 364.5 279.4
0.6899(34) 0.7660(35) 0.0163(5) 0.2127(19) 0.0502(9)
0.0638(19) 0.0692(20) 0.0022(3) 0.1487(29) 0.0259(12)
0.0071(3) 0.0084(4) 0.0016(2) 0.0331(7) 0.0058(3)
also by the helium in molecular or atomic form. Since the 1#4 filter element had the smallest pore size of 5 μm implying a small flow rate of the helium, the free I-131 and Hg-203 could condense in the filter element much more easily, especially for Hg-203 whose normal state is liquid even at the room temperature. The counting rates of most radioactive nuclides decreased along with the cascade sequence filter elements since they had a high quality of filtration efficiency, except for I-131 and Hg-203. From Table 3 and Table 4, it seems that the counting rate of Co-60 decreases a little faster than that of Cs-137, which indicates that the interaction of the dust with each of these two nuclides may be different. A surprising finding is that of Hg-203, which exhibits a strong peak in the 1#2, 1#3, and 1#4 filter elements but rather weak peak in the 1#1 filter element. However, Hg-203 appeared in all three filter elements in the second sampling experiment with a regular decreasing behavior as I-131. The production mechanism and transport behavior of Hg-203 need to be investigated further. 3.3. Particle size distribution of the dust on the filter element As indicated in Fig. 3, the radioactive dust on the filter element is collected with ethanol liquid by ultrasonic oscillation. The suspension was used to investigate the particle size distribution of the dust. Fig. 5 376
Nuclear Engineering and Design 324 (2017) 372–378
F. Xie et al.
Fig. 5. Particle image of the dust in the suspension from the 1#1 (a), 1#2 (b), 2#1 (c), and 2#2 (d) filter elements.
Percentage of the number of the particles
50%
diameters were detected with an optical microscope. However, the experimental observation doesn’t contradict the theoretical analysis, which predicts the diameter of the dominant dust particle in HTR-10 is less than 1 μm (Luo, 2004). The fine particles with diameter less than 1 μm need to be detected with an electron microscope and filter element with smaller pore size, which was beyond the sampling and detection methods of this work. We note that this is the first time that the radioactive dust sampling experiment was conducted while the HTR-10 was in operation. The radioactive dust has been sampled and the dust concentration has been deduced from these samples. The solid fission and activation products including short lived and long lived nuclides have been identified. The particles with large diameter have been observed with an optical microscope. The present work shows that more experiments on the dust in the primary loop are necessary to delineate the interactions between individual nuclides and the dust in an actual reactor.
45% 40% 35% 30% 25% 20% 15% 10% 5% 0% 0-5
5-10
10-15
15-20
20-25
25-50
>50
Particle Diameter ( m) Fig. 6. Particle size distribution of the dust from 1#2 filter element.
5. Conclusion A loop for sampling the radioactive dust in the primary loop of HTR10 has been designed, built, and used. For the first time, since the start of the HTR-10, radioactive dust was collected and characterized in two successful experiments. The types of solid fission and activation products present in the dust have been determined, including short lived nuclides, Co-58, I-131, etc. and long lived nuclides, Co-60, Cs-137, etc. The interaction between fission or activation products and dust particles exhibits distinct characteristics. Most nuclides, like Co-60, show a similar behavior in that the activity decreases with the cascade filter elements along the helium flow. I-131 and Hg-203 appear as unique in that the activity decreases much more slowly with the cascade filter elements along the helium flow. The reason may be that I-131 and Hg203 are in molecular or atomic form in the helium or the dust particle of small sizes have a much stronger adsorption ability for I-131 and Hg203.
conservatively deduced as about 1.5 × 102 Bq/g and 1.0 × 102 Bq/g, which are much smaller than dust experimental results of AVR, 2 × 106–9.6 × 107 Bq/g for Cs-137 (Gottaut and Krüger, 1990). All these experimental results indicate that specific activities of the typical fission and activation products in the dust in the primary loop of HTR10 are smaller than those of AVR. An accurate deduction of specific activities of other nuclides on the dust needs to consider the geometrical configuration of the filter and the detector with a Monte Carlo method. This is presently under investigation. The measured particle size distribution of the dust depends on the sampling method. In the current experiment, the sintered stainless steel powder filter element was adopted. The average particle diameters of dust in the 1#1, 1#2, 2#1 and 2#2 filter elements were 43.0 μm, 13.6 μm, 62.9 μm and 11.3 μm, respectively. The particles with large 377
Nuclear Engineering and Design 324 (2017) 372–378
F. Xie et al.
121, 143–153. Gutti, V.R., Loyalka, S.K., 2009. Thermophoretic deposition in a cylindrical tube: computations and comparison with experiment. Nucl. Technol. 166 (2), 121–133. Humrickhouse, W., 2011. HTGR dust safety issues and needs for research and development. In: INL/EXT-11-21097. Idaho National Laboratory, U.S.. Kazuhiro, S., Takeshi, N., Yoshiaki, M., 1992. Experimental study of dust behavior during depressurization. J. Nucl. Sci. Technol. 29 (10), 1018–1025. Kissane, M.P., 2009. A review of radionuclide behavior in the primary system of a veryhigh-temperature reactor. Nucl. Eng. Des. 239, 3076–3091. Kissane, M.P., Zhang, F., Reeks, M.W., 2012. Dust in HTRs: its nature and improving prediction of its resuspension. Nucl. Eng. Des. 251, 301–305. Li, H., Liu, X., Xie, F., Jia, F., 2017. Experimental study on the content and distribution of key nuclides in an irradiated graphite sphere of HTR-10. Nucl. Eng. Des. 323, 39–45. Lind, T., Güntay, S., Dehbi, A., Liao, Y., Rycroft, C.H., 2010. PSI project on HTR dust generation and transport. In: Proceedings of HTR-2010, October 18–20, 2010, Prague, Czech Republic. Liu, X., Huang, X., Xie, F., Jia, F., Feng, X., Li, H., 2017. Source term analysis of irradiated graphite in the core of HTR-10. Sci. Technol. Nucl. Install Submitted 2017, 2614890 (1-6). Luo, X., 2004. Research on Graphite Dust Behavior in 10MW High Temperature Gascooled Reactor (Ph.D. dissertation). Tsinghua Univ., Beijing. Moormann, R., 2008a. A safety re-evaluation of the AVR pebble bed reactor operation and its consequences for future HTR concepts. Berichte des Rorschungszentrums Jülich 4275, ISSN 0944-2952. Moormann, R., 2008b. Fission product transport and source terms in HTRs: experience from AVR pebble bed reactor. Sci. Technol. Nucl. Install 597491 (1–14). Peng, W., Zhang, T., Sun, X., Yu, S., 2016. Thermophoretic and turbulent deposition of graphite dust in HTGT steam generators. Nucl. Eng. Des. 300, 610–619. Pi, Y., Xie, F., Cao, J., Chadwick, Ch., 2014. Analysis on the efficiency of filter elements in the radioactive graphite dust sampling loop of HTR-10. At. Energy Sci. Technol. 48 (Suppl), 511–516 (in Chinese). Shen, K., Peng, W., Liu, B., Kang, F., Yang, X., Li, W., Yu, S., 2016. Characterization of graphite dust produced by pneumatic lift. Nucl. Eng. Des. 305, 104–109. Simones, M.P., Gutti, V.R., Meyer, R.M., Loyalka, S.K., 2011. Measurements of aerosol charge and size distribution for graphite, gold, palladium, and silver nanoparticles. Nucl. Technol. 176 (2), 211–226. Simones, M.P., Loyalka, S.K., 2015. Measurements of charged aerosol coagulation. Nucl. Technol. 189, 45–62. Stempniewicz, M.M., Komen, E.M.J., de With, A., 2008. Model of particle resuspension in turbulent flows. Nucl. Eng. Des. 238, 2943–2959. Troy, R.S., Tompson, R.V., Ghosh, T.K., Loyalka, S.K., 2012. Generation of graphite particles by rotational/spinning abrasion and their characterization. Nucl. Technol. 178 (3), 241–257. Troy, R.S., Tompson, R.V., Ghosh, T.K., Loyalka, S.K., Gallego, N.C., 2015. Generation of graphite particles by sliding abrasion and their characterization. Nucl. Technol. 189 (3), 241–257. U.S. DOE Nuclear Energy Research Advisory Committee (NERAC) and the Generation IV International Forum (GIF), 2002. A Technology Roadmap for Generation IV Nuclear Energy Systems. GIF-002-00. Wang, S., Pi, Y., Xie, F., Li, H., Cao, J., 2014. Study on 14C content in post-irradiation graphite spheres of HTR-10. At. Energy Sci. Technol. 48 (Suppl), 506–510 (in Chinese). Wei, L., Xie, F., Chen, X., Ma, T., Tong, J., Li, F., 2016. Summary and experience feedback on the restart and power operation after a long term shutdown of HTR-10. In: Proceedings of HTR-2016, November 5–10, 2016, Las Vegas, NV, USA. Wu, Z., Lin, D., Zhong, D., 2002. The design features of the HTR-10. Nucl. Eng. Des. 218, 25–32. Xie, F., Cao, J., Chen, Z., Dong, Y., 2015a. The design and study of the radioactive graphite dust experimental system in the primary loop of the HTR-10. At. Energy Sci. Technol. 49 (4), 744–749 (in Chinese). Xie, F., Cao, J., Tong, J., Wei, L., Liu, X., Dong, Y., Yu, S., 2015b. The research progress on the radioactive graphite dust in HTR-10. In: Proceedings of ICONE23, May 17–21, 2015, Chiba, Japan. Xie, F., Li, H., Cao, J., Yu, S., Zhang, L., Li, W., Fang, S., 2013. A reform in the helium purification system of the HTR-10: γ dose rate measurement and suggestions for decommissioning. In: Proceedings of ICEM 2013, September 8–12, 2013, Brussels, Belgium. Xu, Y., Li, H., Xie, F., Cao, J., Tong, J., 2017. Source term analysis of tritium in HTR-10. Fusion Sci. Technol. 71, 671–678. Zhang, Z., Yu, S., 2002. Future HTGR developments in China after the criticality of the HTR-10. Nucl. Eng. Des. 218, 249–257.
The concentrations of the radioactive dust in the primary loop of HTR-10 deduced from the collected data are 5.57(17) μg/m3STP and 1.97(8) μg/m3STP. The average particle diameters, as obtained with collection by the filter elements of 80 μm pore size and 50 μm pore size are respectively, 43.0 μm and 13.6 μm in the first experiment and 62.9 μm and 11.3 μm in the second experiment. This indicates that dust particles of diameter larger than 10 μm, make up a significant portion of particle mass in the primary loop of HTR-10. The presence of fine particles needs to be investigated with the filter elements of smaller pore size and electronic microscopy. The specific activities in the dust of HTR-10 are conservatively estimated as 3.3 × 104 Bq/g and 6.5 × 103 Bq/g for Co-60, and 1.5 × 102 Bq/g and 1.0 × 102 Bq/g for Cs-137. Direct comparisons of the present data and results with the measurements in AVR are difficult, because of many factors (graphites, operational characteristics, measurement techniques, experimental uncertainties, etc.). Nevertheless it should be noted that both the amount of the dust and the activities associated with the HTR-10 dust are lower than those reported for AVR. The methods, data and analysis reported here are of basic significance for not only the radiation safety evaluation of HTR-10, but also other pebble bed reactors as regards the study of behavior of fission and activation products and characteristics of radioactive dust. Acknowledgements This work was supported by the National Natural Science Foundation of China (No. 11575099), the Beijing Natural Science Foundation (No. 2163051), the Chinese National Significant Science and Technology Project (No. ZX06901), and the Tsinghua University Initiative Scientific Research Program (No. 20151080375). Fruitful discussions with Prof. Dazhi Xue, Prof. Meisheng Yao, Prof. Fu Li, Mr. Liqiang Wei and Mr. Ling Liu at INET, Tsinghua University, Beijing, China, and Dr. Terttaliisa Lind at Paul Scherrer Institute, PSI, Switzerland, are gratefully acknowledged. We also thank one of the anonymous reviewer for pointing out the production mechanism of Hg203. References Barth, T., Kulenkampff, J., Bras, S., Gründig, M., Lippmann-Pipke, J., Hampel, U., 2013. Positron emission tomography in pebble beds. Part 2: graphite particle deposition and resuspension. Nucl. Eng. Des. 267, 227–237. Bäumer, R., Barnert, H., Baust, E., Bergerfurth, A., Bonnenberg, H., Bülling, H., Burger, St., Von der Decken, C.-B., Delle, W., Gerwin, H., Hackstein, K.-G., Hantke, H.-J., Hohn, H., Ivens, G., Kirch, N., Kirjushin, A.I., Kröger, W., Krüger, K., Kuzavkov, N.G., Lange, G., Marnet, C., Nickel, H., Pohl, P., Scherer, W., Schöning, J., Schulten, R., Singh, J., Steinwarz, W., Theymann, E., 1990. AVR: Experimental High-Temperature Reactor: 21 Years of Successful Operation for a Future Energy Technology. Association of German Engineers (VDI). The Society for Energy Technologies. Boddu, S.R., Gutti, V.R., Meyer, R.M., Ghosh, T.K., Tompson, R.V., Loyalka, S.K., 2011. Carbon nanoparticle generation, collection, and characterization using a spark generator and a thermophoretic deposition cell. Nucl. Technol. 173, 318–326. Cogliati, J.J., Ougouag, A.M., Ortensi, J., 2011. Survey of dust production in pebble bed reactor cores. Nucl. Eng. Des. 241, 2364–2369. Fachinger, J., Barnert, H., Kummer, A., Caspary, G., Seubert, M., Koster, A., Makumbe, M., Naicker, L., 2008. Examination of dust in AVR pipe components. In: Proceedings of HTR 2008, 28 September-1 October 2008, Washington DC., USA. Gottaut, H., Krüger, K., 1990. Results of experiments at the AVR reactor. Nucl. Eng. Des.
378
Contents lists available at ScienceDirect
Nuclear Engineering and Design journal homepage: www.elsevier.com/locate/nucengdes
Experimental research on the radioactive dust in the primary loop of HTR10
MARK
⁎
Feng Xiea, Jianzhu Caoa, Xiaogui Fenga, Xuegang Liua, , Jiejuan Tonga, Haitao Wanga, Yujie Donga, Zuoyi Zhanga, S.K. Loyalkab a Institute of Nuclear and New Energy Technology, Collaborative Innovation Center of Advanced Nuclear Energy Technology, Key Laboratory of Advanced Reactor Engineering and Safety of Ministry of Education, Tsinghua University, Beijing 100084, China b Nuclear Science and Engineering Institute, University of Missouri, Columbia, MO 65211, United States
A R T I C L E I N F O
A B S T R A C T
Keywords: HTR-10 Radioactive dust Fission and activation products Dust concentration Particle size distribution
Presence of radioactive dust, and also potential for its release, are considered important issues as regards the safety of high temperature gas cooled reactors, especially the pebble bed types. To obtain data and insights into the quantity and characteristics of such dust, a sampling loop (with stainless steel powder filter elements to collect the dust) has been built in the helium purification system of HTR-10 enabling sampling of radioactive dust circulating in the primary loop of HTR-10. Two sampling experiments were conducted during the operation of HTR-10 in 2015. The concentration of the radioactive dust in the primary loop was deduced from the dust deposited on the filters, and was estimated to be 5.57(17) μg/m3STP and 1.97(8) μg/m3STP, respectively. The nuclides present in the dust have been identified by their γ spectra. After the removal of dust from filter elements, the particle size distribution was measured with an optical microscope. In the first experiment, the average particle diameters of the radioactive dust were determined as 43.0 μm and 13.6 μm captured respectively, by the filter elements of 80 μm and 50 μm pore size. In the second experiment, the average particle diameters were 62.9 μm and 11.3 μm as captured respectively, by similar filter elements. The data show existence of large particles in the primary loop of HTR-10, generated most likely by abrasion.
1. Introduction With the development of the high temperature gas-cooled reactors (HTGRs), the very high temperature reactor (VHTR) has been considered as a candidate for the six proposed Generation IV concept reactors by the Generation IV International Forum (GIF) (NERAC and GIF, 2002). With the use of fuel elements embedded with tristructural-isotropic (TRISO) coated particles, the inherent safety performance of the HTGRs has attracted wide attention. The high thermal efficiency and the high output temperature which can be used for hydrogen production and process heat, bring large commercial space for the future development of HTGRs (Zhang and Yu, 2002). There are, however, several safety issues that have been identified previously through operation of research reactors. These include the local high temperature in the core leading to the failure of TRISO coated particles, the release of the radioactive dust in the depressurization accident, the contamination due to the radioactive dust in the primary loop, etc. (Bäumer et al., 1990; Moormann, 2008a,b). It has been noted also that dust related problems are more important in the pebble bed type
⁎
reactors than in the prismatic type reactors (Humrickhouse, 2011). In the pebble-bed HTGRs, the dust is thought to be generated by several mechanisms, including the abrasion among fuel elements, and friction between fuel elements and other graphite structures or pipelines when the fuel elements cycled (Luo, 2004; Moormann, 2008b; Kissane, 2009). Several papers have been published regarding the behavior of dust in HTGRs, including the generation, characterization, transport, coagulation, aggregation, deposition, resuspension, etc. Humrickhouse (2011) summarized the dust safety issues in HTGRs and indicated that the dust related research was urgently needed for the development of pebble bed HTGRs on topics such as the dust distribution under normal operation, dust generation, dust-fission product interaction, etc. Cogliati et al. (2011) surveyed the available literature on graphite dust production and measurements in pebble bed reactors and concluded that there was significant uncertainty on the severity of the dust production and its consequences in pebble bed reactors. Troy et al. (2012, 2015) and Shen et al. (2016) investigated the characteristics of graphite dust particles experimentally. Simones et al. (2011), Simones and Loyalka (2015) measured the charge-size distributions of
Corresponding author at: Room 320, Energy Science Building A, Institute of Nuclear and New Energy Technology, Tsinghua University, Beijing 100084, China. E-mail address: [email protected] (X. Liu).
http://dx.doi.org/10.1016/j.nucengdes.2017.09.015 Received 21 February 2017; Received in revised form 21 August 2017; Accepted 19 September 2017 0029-5493/ © 2017 Elsevier B.V. All rights reserved.
Nuclear Engineering and Design 324 (2017) 372–378
F. Xie et al.
was bulky and difficult to dismantle, an experimental system for accurately measuring the radioactive dust in the primary loop of HTR-10 was designed and built. Though details of the radioactive dust sampling loop of HTR-10 have been presented in Xie et al. (2015a,b), it is necessary to give a brief description here. As shown in Fig. 1, the main sampling loop includes two electric check valves (KBE03 AA001 and KBE03 AA002), one electric control valve (KBE03 AA003), two thermometers (KBE03 CT001 and KBE03 CT002), two pressure gauges (KBE03 CP001 and KBE03 CP003), one flowmeter (KBE03 CF001), one differential pressure transmitter (KBE03 CP002), and five manual globe valves (KBE03 AA004 ∼ KBE03 AA008). Primary features for this dust sampling loop are as follows:
graphite aerosol particles and coagulation of charged aerosols pertinent to HTGRs. Moormann (2008a,b), Kissane (2009), and Lind et al. (2010) discussed the transport of the radioactive dust which can be an important source term in pebble bed reactors. Gutti and Loyalka (2009), Boddu et al. (2011), Barth et al. (2013), and Peng et al. (2016) experimentally and/or numerically investigated the deposition behavior of dust in HTGRs. Kazuhiro et al. (1992), Stempniewicz et al. (2008), and Kissane et al. (2012) studied the resuspension behavior of dust in HTGRs. There have been, however, very few experimental investigations regarding the radioactive dust directly in pebble bed reactors. The only available data on such dust are from Arbeitsgemeinschaft Versuchsreaktor (AVR) and Thorium Hochtemperatur Reaktor (THTR) (Bäumer et al., 1990; Moormann, 2008b; Fachinger et al., 2008). There is a need to verify conclusions from theoretical calculations and nonradioactive dust experiments against data from the radioactive dust experiments in actual reactors. Recently, several experiments related to the source terms of the 10 MW high temperature gas-cooled reactor (HTR-10) have been conducted. These include investigations of the radioactive dust, H-3, and C-14 in the primary loop, post irradiation graphite spheres from the reactor core, etc. (Xie et al., 2015a,b; Xu et al., 2017; Wang et al., 2014; Liu et al., 2017; Li et al., 2017). Previous measurements also indicated the existence of radioactive dust in the primary loop of HTR-10 (Xie et al., 2013, 2015a,b). Accordingly, an experimental sampling loop has been designed and built in the helium purification system of HTR-10. In the present study, the radioactive dust in the primary loop of HTR-10 was investigated with use of the above sampling loop when HTR-10 was restarted in 2015. The radioactive dust has been collected with a sampling filter. The concentration of the dust in the primary loop was estimated with use of the coolant flow data. The types of solid fission and activation products which were absorbed or present in the dust were determined with a γ spectrometer (GC3018, High-purity Germanium Detectors, from CANBERRA Industries Inc.) in the radiochemistry lab in the Institute of Nuclear and New Energy Technology (INET), Tsinghua University. The counting rates of typical nuclides, including Co-60, Cs-137, I-131, etc., were measured and compared with each other. The particle size distributions of the dust from the filter elements were obtained through imaging with an optical microscope. The data reported can shed light on the behavior of fission/activation products and radioactive dust in HTGRs, and would aid in modeling and safety studies.
(1) A small and easily disassemblable sampling filter was designed, as shown in Fig. 2. It can contain as many as 8 filter elements with different pore sizes, from 1 μm, 3 μm, 5 μm, 10 μm, 20 μm, 50 μm, and 80 μm, and can separate the radioactive graphite dust according to the particle diameter. (2) Two electric check valves (KBE03 AA001 and KBE03 AA002) can isolate the experimental system from the helium purification system effectively. The electric control valve (KBE03 AA003) before the dust filter can adjust the flow ratio between the experimental loop and the main circuit. (3) Two thermometers (KBE03 CT001 and KBE03 CT002), two pressure gauges (KBE03 CP001 and KBE03 CP003), one flowmeter (KBE03 CF001), and one differential pressure transmitter (KBE03 CP002), can record the temperature, the absolute pressure, the flow capacity, and the pressure drop across the sampling filter, respectively. The operation for the radioactive dust sampling loop can be divided into preparation, sampling, and shutdown phases. In the preparation phase, the electric check valves (KBE03 AA001 and KBE03 AA002) at the entrance and exit of the sampling loop are closed, and the manual globe valves in the sampling loop are open. The whole sampling loop is pumped to remove the air inside from KTT system whose function is to pump the air out to the system and to relieve the high pressure in the pipe, if necessary. The preparation phase is completed when the pressure of the system is below 100 Pa and the manual globe valve is closed. In the sampling phase, the radioactive dust sampling loop will be connected to the helium purification system by opening of the electric check valves (KBE03 AA001 and KBE03 AA002). The electric control valve (KBE03 AA003) can be adjusted to the predetermined flow capacity, and the sampling filter would collect the dust particles for several days or weeks. In the shutdown phase, the sampling loop would be separated from the main circuit by closing of the electric check valves (KBE03 AA001 and KBE03 AA002). After the pressure in the sampling loop drops to normal conditions due to opening of the manual globe valve (KBE03 AA008) and the temperature falls to the room temperature, the sampling filter can be dismantled and sent to the radiochemistry lab for further analysis.
2. Experimental setup The HTR-10 is a helium cooled, graphite-moderated, and thermal neutron spectrum test reactor, which was designed and built by INET, Tsinghua University, China (Wu et al., 2002). It reached its criticality in December 2000, ran up to full power operation in January 2003, and was shut down in July 2007. It was restarted at the end of November 2014 (Wei et al., 2016). During the shutdown stage of HTR-10, an experimental sampling loop in the primary system aiming to study the behavior of radioactive dust was designed. It was approved by the State Bureau of Nuclear Safety of China in June 2012. In December 2013, the experimental loop with a sampling filter was built at the entrance of the helium purification system of HTR-10, as shown in Fig. 1. The helium purification system contains two normal purification lines and one dehumidification line. Each normal line contained a dust filter, a copper oxide bed, a molecular sieve adsorber, and a low temperature adsorber. This system is very important for the reactor operation since it can purify a bypass helium from the primary coolant system and maintain the quality of the primary helium by continuously reducing the quantity of chemical impurities and removing the dust, fission products and activation products. Parallel to the dust filter in the helium purification system, which
3. Results After a long term shutdown from 2007 to 2014, HTR-10 was restarted at the end of November 2014, and was operated in a power stage in March 2015. During June 1st, 2015, to August 6th, 2015, two sampling experiments about the radioactive dust in the primary loop were conducted. The operational thermal power of HTR-10 was about 2.9 MW compared to the rated thermal power of 10 MW. The primary helium temperature at the reactor inlet and outlet were respectively, ∼185 °C and ∼575 °C. The primary helium pressure was about 2.1 MPa and the flow rate was about 1.39 t/h. Note that the rated values for the primary helium temperature at the reactor inlet and outlet, primary helium pressure, and primary helium flow rate at the full power operation of HTR-10 were 250 °C, 750 °C, 3.0 MPa, and 4.32 t/h, respectively. The concentration of the radioactive dust, the types of 373
Nuclear Engineering and Design 324 (2017) 372–378
F. Xie et al.
Fig. 1. Flowchart of the radioactive dust sampling loop with the schematic diagram of the primary loop of HTR-10.
Fig. 2. Sampling filter containing four filter elements (the left one) and comparison of the filter element after the sampling process (the middle one) and after the decontamination process (the last one).
flow was 2911.2 m3STP totally. Fig. 2 shows a filter element after the sampling which clearly identified the radioactive dust on the surface. In order to obtain the mass of the radioactive dust, several steps needed to be implemented. First, the filter element was immersed into the ethanol, and the dust particle separated from the filter element by the ultrasonic separation. Second, the above process was repeated many times and all suspensions were mixed together. Finally, ethanol was evaporated by heating so the mass of the dust could be obtained if it was present. After removing the radioactive dust into the liquid, the weight and particle size distribution of the radioactive dust could be measured. However, only the weight of radioactive dust in the filter element with 80 μm pore size in the first sampling experiment was measurable, and it was found to be 0.0162(5) g. Pi et al. (2014) studied filtration characteristics of sintered stainless steel powder filter element experimentally, and indicated that the filter element with 80 μm pore size can reach a high filtration efficiency ∼95% even with the particle diameter around 1 μm. Considering the overwhelming majority of radioactive dust with bulk mass was retained in the filter element with 80 μm pore size, the concentration of the radioactive dust in the primary loop of HTR-10 can be deduced as 5.57(17) μg/m3STP. This is very close to the average dust concentration of AVR, ∼5 μg/m3STP. In the second sampling experiment, three filter elements with the pore size of 80 μm, 50 μm, and 20 μm were put into the sampling filter in sequence. With the same process as that in the first sampling experiment, only the mass of the radioactive dust in the filter element with 80 μm pore size was measurable and was found to be 0.0254(10) g. With the accumulated helium flow data which was 12,926.2 m3STP in the second sampling experiment, the concentration of the radioactive dust in the primary loop of HTR-10 was deduced to be 1.97(8) μg/m3STP.
solid fission and activation nuclides adsorbed in the dust, and the particle size distribution of the dust in the filter element are determined from the experimental measurement. For convenience, the filter elements used in two sampling experiments are denoted as follows: The four filter elements used in the first sampling experiment are denoted as 1#1, 1#2, 1#3 and 1#4, corresponding to the filter element with pore sizes of 80 μm, 50 μm, 20 μm, and 5 μm, respectively. The three filter elements used in the second sampling experiment are denoted as 2#1, 2#2 and 2#3, corresponding to the filter element with pore sizes of 80 μm, 50 μm, and 20 μm, respectively. The purpose of removing the filter element of 5 μm pore size in the second experiment was to reduce the resistance pressure of the sampling loop. With a portable γ dose rate spectrometer, the recording value from the filter element 1#4 is close to the background. 3.1. Concentration of the radioactive dust in the primary loop of HTR-10 Table 1 lists the typical values of main operational parameters of the radioactive dust sampling loop of HTR-10 during its operation. In the first sampling experiment, four filter elements with the pore sizes of 80 μm, 50 μm, 20 μm, and 5 μm were put into the sampling filter in sequence, which means the radioactive dust in the primary helium would go through the filter element with biggest pore size and then to the smaller one. Though several breaks due to the preplanning or unexpected events occurred during the first sampling process, the accumulated helium Table 1 Main operational parameters of the radioactive dust sampling in HTR-10. Item
Unit
1st Experiment
2nd Experiment
Helium pressure Sampling loop inlet temperature Sampling loop outlet temperature Flow rate
MPa °C °C m3/h STP
2.1 ∼100 ∼50 ∼15
2.1 ∼107 ∼70 ∼33
3.2. Types of solid fission and activation nuclides in the dust Fig. 3 shows four sintered stainless steel powder filter elements used in the first sampling experiment and the corresponding γ spectra from the direct measurement of these filter elements with the γ spectrometer. 374
Nuclear Engineering and Design 324 (2017) 372–378
F. Xie et al.
Fig. 3. Four sintered stainless steel powder filter elements in the first experiment and the corresponding γ spectra.
Fig. 4. Assignment of nuclides from the γ spectra with the radioactive dust from 2#1 filter element.
Along with the direction of the helium flow, the first sintered stainless steel powder filter element 1#1 clearly captured remarkable amount of dust on the surface. While there was no visible dust on other filter elements, the γ spectra clearly showed the presence of fission and activation nuclides which are considered to be adsorbed in the dust. Fig. 4 shows the nuclides identified in the radioactive dust of HTR-10. Table 2 lists the nuclides with the corresponding half lives and peaks. This is the first time, that the solid fission products, including I-124, I-131, Cs-137, Ba-140, La-140, Eu-152 and Hf-181, and the solid activation products, including Cr-51, Mn-54, Fe-59, Co-57, Co-58, Co-60 and Se-75, including short lived nuclides and long lived nuclides, have been experimentally shown to exist in the primary loop of HTR-10. Further, the unexpected nuclide Hg-203 also was detected in the γ spectra. The Hg203 can be produced by a neutron induced n-α reaction from Pb-206 and neutron capture from Hg-202. In the structure graphite materials of HTR-10, Hg and Pb elements were determined experimentally. Thus,
Hg-203 appears to be an activation product. Table 3 and Table 4 list counting rates for typical nuclides from the measurement of γ spectra of filter elements in the first and the second experiment, respectively. For short lived nuclides, like I-131 and Hg203, the measurement time affected the counting rates significantly. In Table 3, the counting rate of I-131 and Hg-203 exhibited the similar decreasing behavior along with the cascade sequence filter elements since the measurement time were quite close. For the relatively long lived nuclides, like Co-60 and Cs-137, the measurement time had little influence on counting rates. Compared with the I-131 and Hg-203, the counting rates of the Co-60 and Cs-137 decreased much faster with the cascade placement of filter elements, which implied the transport behavior were very different. However, the counting rates of I-131 and Hg-203 in 1#4 filter element in Table 2 were rather large even considering the effect of the spontaneous decay. A possible reason may be that the I-131 and Hg-203 could be transported not only by the dust but 375
Nuclear Engineering and Design 324 (2017) 372–378
F. Xie et al.
shows the dust particles in the suspension from 1#1, 1#2, 2#1, and 2#2 filter elements with an optical microscope. With a simple statistical method, the average particle diameters of the radioactive dust are determined as 43.0 μm, 13.6 μm, 62.9 μm and 11.3 μm in the suspensions from 1#1, 1#2, 2#1 and 2#2 filter elements, respectively. Due to too little and relatively small sized dust in 1#3, 1#4 an 2#3 filter elements, the measurement of the average particle diameter of the radioactive dust was not successful. The particle size distribution in the 1#2 filter element has been deduced based on about 5000 particles observed in the suspension, as seen in Fig. 6. It appears that the particles with diameter between 5 μm and 10 μm are the dominant part in the suspension of the 1#2 filter element. Since much fewer particles (< 1000) were recorded in the observation range of suspensions of 1#1, 2#1 and 2#2 filter elements, we do not report the particle size distribution in these filter elements.
Table 2 Nuclides determined from the γ spectra with corresponding half lives and peaks. Nuclides
Half lives
Peaks (keV)
Cr-51 Mn-54 Fe-59 Co-57 Co-58 Co-60 Se-75 I-124 I-131 Cs-137 Ba-140 La-140 Eu-152 Hf-181 Hg-203
27.704 d 312.7 d 44.63 d 270.9 d 70.8 d 5.271 a 119.78 d 4.18 d 8.04 d 30.17 a 12.789 d 40.22 d 13.6 a 42.39 d 46.6 d
320.0 834.9 142.7, 192.3, 1099.2, 1291.4 121.9, 136.2 511.1, 810.6 1173.2, 1332.6 121.9, 136.2, 264.3, 279.4 511.1, 602.2 284.5, 364.5, 636.9 661.2 162.7, 304.7, 537.1 328.5, 486.9, 815.5 121.9, 344.3, 1408.0 133.2, 344.3, 482.0 279.4
4. Discussion As stated earlier there have been very few experimental data available about the radioactive dust in the pebble bed reactors. Bäumer et al. (1990) reported that the average concentration of dust in the primary loop of AVR was 5 μg/m3STP, but it could be as high as 2000 μg/ m3STP at the startup or shutdown stage of the reactor. In HTR-10, this was the first attempt to investigate the radioactive dust in the primary loop. In the first sampling experiment, though only the dust in the first filter element 1#1 was measureable, we believe that the most of the mass of the dust was obtained as the sintered stainless steel powder filter element possessed rather high filtration efficiency (Pi et al., 2014). During the first sampling experiment, the sampling loop experienced the startup and shutdown stages of the HTR-10. Besides, when the experimental loop was carried on the trial experiment, the dust in the primary loop was also gathered but the flow of the helium was not cumulatively added in the dust concentration calculation. Thus, the dust concentration value of 5.57(17) μg/m3STP in the primary loop of HTR-10 is an integrated and conservative result. In the second sampling experiment, the accumulated helium flow was 12,926.2 m3STP, which was much higher than that in the first experiment and the concentration of the radioactive dust in the primary loop of HTR-10 can be deduced as 1.97(8) μg/m3STP. Based on our measurements, the average dust concentration in the primary loop of HTR-10 is slightly smaller than the average value of dust concentration in AVR. Given the same filtration process, the counting rate of typical nuclides in each filter element exhibits individual features. Co-60 is thought to have a strong correlation with the graphite dust since it was produced from the activation reactions of impurities of Co-59 and/or Ni-60 in dust materials. The experimental average values for the Co and Ni mass fractions in the graphite spheres of HTR-10 have been investigated (Li et al., 2017). Cs-137, as a fission product, could diffuse from the fuel element and then adsorb into the dust. I-131 is also a fission product but it exhibits a different transport behavior. It can absorb into the dust and then circulate in the primary loop. Another possible way is that it can transport with the helium gas in the atomic/ molecular form. Hg-203 is a special nuclide, which behaves in a unique way. A possible explanation may that Hg-203 can transport in the atomic/molecular form even at not high temperature and is not easy to condense in the filter element as I-131. Another possibility may be the interaction between Hg-203 and the fine particle is much stronger than with the large particle. In any case, the transport behavior of Hg-203 needs to be studied further. From the γ spectra, Co-60 plays an important role in the specific activity of the dust. With a standard source of Co-60, the specific activities of Co-60 on the dust in HTR-10 are conservatively estimated as about 3.3 × 104 Bq/g and 6.5 × 103 Bq/g from these two experiments respectively. These values are lower than dust experimental results of AVR, 2.0 × 105–8 × 106 Bq/g for Co-60 (Gottaut and Krüger, 1990). The specific activities of Cs-137 on the dust in HTR-10 are
Table 3 Counting rates of typical nuclides from γ spectra of filter elements in the first experiment. Measurement Time
2015/7/6
2015/7/2
2015/7/3
2015/6/26
Nuclides
Peak (keV)
1#1 (CPS)
1#2 (CPS)
1#3 (CPS)
1#4 (CPS)
Co-60 Co-60 Cs-137 I-131 Hg-203
1332.5 1173.2 661.2 364.5 279.4
2.2521(52) 2.4874(54) 0.0149(4) 0.0333(6) 0.0197(5)
0.1366(12) 0.1522(13) 0.0018(1) 0.0135(4) 0.0789(9)
0.0234(3) 0.0252(3) 0.0013(1) 0.0028(1) 0.0134(2)
0.0208(3) 0.0224(3) 0.0015(1) 0.0122(2) 0.0222(3)
Table 4 Counting rates of typical nuclides from γ spectra of filter elements in the second experiment. Measurement Time
2015/8/18
2015/8/19
2015/8/19
Nuclides
Peak (keV)
1#1 (CPS)
1#2 (CPS)
1#3 (CPS)
Co-60 Co-60 Cs-137 I-131 Hg-203
1332.5 1173.2 661.2 364.5 279.4
0.6899(34) 0.7660(35) 0.0163(5) 0.2127(19) 0.0502(9)
0.0638(19) 0.0692(20) 0.0022(3) 0.1487(29) 0.0259(12)
0.0071(3) 0.0084(4) 0.0016(2) 0.0331(7) 0.0058(3)
also by the helium in molecular or atomic form. Since the 1#4 filter element had the smallest pore size of 5 μm implying a small flow rate of the helium, the free I-131 and Hg-203 could condense in the filter element much more easily, especially for Hg-203 whose normal state is liquid even at the room temperature. The counting rates of most radioactive nuclides decreased along with the cascade sequence filter elements since they had a high quality of filtration efficiency, except for I-131 and Hg-203. From Table 3 and Table 4, it seems that the counting rate of Co-60 decreases a little faster than that of Cs-137, which indicates that the interaction of the dust with each of these two nuclides may be different. A surprising finding is that of Hg-203, which exhibits a strong peak in the 1#2, 1#3, and 1#4 filter elements but rather weak peak in the 1#1 filter element. However, Hg-203 appeared in all three filter elements in the second sampling experiment with a regular decreasing behavior as I-131. The production mechanism and transport behavior of Hg-203 need to be investigated further. 3.3. Particle size distribution of the dust on the filter element As indicated in Fig. 3, the radioactive dust on the filter element is collected with ethanol liquid by ultrasonic oscillation. The suspension was used to investigate the particle size distribution of the dust. Fig. 5 376
Nuclear Engineering and Design 324 (2017) 372–378
F. Xie et al.
Fig. 5. Particle image of the dust in the suspension from the 1#1 (a), 1#2 (b), 2#1 (c), and 2#2 (d) filter elements.
Percentage of the number of the particles
50%
diameters were detected with an optical microscope. However, the experimental observation doesn’t contradict the theoretical analysis, which predicts the diameter of the dominant dust particle in HTR-10 is less than 1 μm (Luo, 2004). The fine particles with diameter less than 1 μm need to be detected with an electron microscope and filter element with smaller pore size, which was beyond the sampling and detection methods of this work. We note that this is the first time that the radioactive dust sampling experiment was conducted while the HTR-10 was in operation. The radioactive dust has been sampled and the dust concentration has been deduced from these samples. The solid fission and activation products including short lived and long lived nuclides have been identified. The particles with large diameter have been observed with an optical microscope. The present work shows that more experiments on the dust in the primary loop are necessary to delineate the interactions between individual nuclides and the dust in an actual reactor.
45% 40% 35% 30% 25% 20% 15% 10% 5% 0% 0-5
5-10
10-15
15-20
20-25
25-50
>50
Particle Diameter ( m) Fig. 6. Particle size distribution of the dust from 1#2 filter element.
5. Conclusion A loop for sampling the radioactive dust in the primary loop of HTR10 has been designed, built, and used. For the first time, since the start of the HTR-10, radioactive dust was collected and characterized in two successful experiments. The types of solid fission and activation products present in the dust have been determined, including short lived nuclides, Co-58, I-131, etc. and long lived nuclides, Co-60, Cs-137, etc. The interaction between fission or activation products and dust particles exhibits distinct characteristics. Most nuclides, like Co-60, show a similar behavior in that the activity decreases with the cascade filter elements along the helium flow. I-131 and Hg-203 appear as unique in that the activity decreases much more slowly with the cascade filter elements along the helium flow. The reason may be that I-131 and Hg203 are in molecular or atomic form in the helium or the dust particle of small sizes have a much stronger adsorption ability for I-131 and Hg203.
conservatively deduced as about 1.5 × 102 Bq/g and 1.0 × 102 Bq/g, which are much smaller than dust experimental results of AVR, 2 × 106–9.6 × 107 Bq/g for Cs-137 (Gottaut and Krüger, 1990). All these experimental results indicate that specific activities of the typical fission and activation products in the dust in the primary loop of HTR10 are smaller than those of AVR. An accurate deduction of specific activities of other nuclides on the dust needs to consider the geometrical configuration of the filter and the detector with a Monte Carlo method. This is presently under investigation. The measured particle size distribution of the dust depends on the sampling method. In the current experiment, the sintered stainless steel powder filter element was adopted. The average particle diameters of dust in the 1#1, 1#2, 2#1 and 2#2 filter elements were 43.0 μm, 13.6 μm, 62.9 μm and 11.3 μm, respectively. The particles with large 377
Nuclear Engineering and Design 324 (2017) 372–378
F. Xie et al.
121, 143–153. Gutti, V.R., Loyalka, S.K., 2009. Thermophoretic deposition in a cylindrical tube: computations and comparison with experiment. Nucl. Technol. 166 (2), 121–133. Humrickhouse, W., 2011. HTGR dust safety issues and needs for research and development. In: INL/EXT-11-21097. Idaho National Laboratory, U.S.. Kazuhiro, S., Takeshi, N., Yoshiaki, M., 1992. Experimental study of dust behavior during depressurization. J. Nucl. Sci. Technol. 29 (10), 1018–1025. Kissane, M.P., 2009. A review of radionuclide behavior in the primary system of a veryhigh-temperature reactor. Nucl. Eng. Des. 239, 3076–3091. Kissane, M.P., Zhang, F., Reeks, M.W., 2012. Dust in HTRs: its nature and improving prediction of its resuspension. Nucl. Eng. Des. 251, 301–305. Li, H., Liu, X., Xie, F., Jia, F., 2017. Experimental study on the content and distribution of key nuclides in an irradiated graphite sphere of HTR-10. Nucl. Eng. Des. 323, 39–45. Lind, T., Güntay, S., Dehbi, A., Liao, Y., Rycroft, C.H., 2010. PSI project on HTR dust generation and transport. In: Proceedings of HTR-2010, October 18–20, 2010, Prague, Czech Republic. Liu, X., Huang, X., Xie, F., Jia, F., Feng, X., Li, H., 2017. Source term analysis of irradiated graphite in the core of HTR-10. Sci. Technol. Nucl. Install Submitted 2017, 2614890 (1-6). Luo, X., 2004. Research on Graphite Dust Behavior in 10MW High Temperature Gascooled Reactor (Ph.D. dissertation). Tsinghua Univ., Beijing. Moormann, R., 2008a. A safety re-evaluation of the AVR pebble bed reactor operation and its consequences for future HTR concepts. Berichte des Rorschungszentrums Jülich 4275, ISSN 0944-2952. Moormann, R., 2008b. Fission product transport and source terms in HTRs: experience from AVR pebble bed reactor. Sci. Technol. Nucl. Install 597491 (1–14). Peng, W., Zhang, T., Sun, X., Yu, S., 2016. Thermophoretic and turbulent deposition of graphite dust in HTGT steam generators. Nucl. Eng. Des. 300, 610–619. Pi, Y., Xie, F., Cao, J., Chadwick, Ch., 2014. Analysis on the efficiency of filter elements in the radioactive graphite dust sampling loop of HTR-10. At. Energy Sci. Technol. 48 (Suppl), 511–516 (in Chinese). Shen, K., Peng, W., Liu, B., Kang, F., Yang, X., Li, W., Yu, S., 2016. Characterization of graphite dust produced by pneumatic lift. Nucl. Eng. Des. 305, 104–109. Simones, M.P., Gutti, V.R., Meyer, R.M., Loyalka, S.K., 2011. Measurements of aerosol charge and size distribution for graphite, gold, palladium, and silver nanoparticles. Nucl. Technol. 176 (2), 211–226. Simones, M.P., Loyalka, S.K., 2015. Measurements of charged aerosol coagulation. Nucl. Technol. 189, 45–62. Stempniewicz, M.M., Komen, E.M.J., de With, A., 2008. Model of particle resuspension in turbulent flows. Nucl. Eng. Des. 238, 2943–2959. Troy, R.S., Tompson, R.V., Ghosh, T.K., Loyalka, S.K., 2012. Generation of graphite particles by rotational/spinning abrasion and their characterization. Nucl. Technol. 178 (3), 241–257. Troy, R.S., Tompson, R.V., Ghosh, T.K., Loyalka, S.K., Gallego, N.C., 2015. Generation of graphite particles by sliding abrasion and their characterization. Nucl. Technol. 189 (3), 241–257. U.S. DOE Nuclear Energy Research Advisory Committee (NERAC) and the Generation IV International Forum (GIF), 2002. A Technology Roadmap for Generation IV Nuclear Energy Systems. GIF-002-00. Wang, S., Pi, Y., Xie, F., Li, H., Cao, J., 2014. Study on 14C content in post-irradiation graphite spheres of HTR-10. At. Energy Sci. Technol. 48 (Suppl), 506–510 (in Chinese). Wei, L., Xie, F., Chen, X., Ma, T., Tong, J., Li, F., 2016. Summary and experience feedback on the restart and power operation after a long term shutdown of HTR-10. In: Proceedings of HTR-2016, November 5–10, 2016, Las Vegas, NV, USA. Wu, Z., Lin, D., Zhong, D., 2002. The design features of the HTR-10. Nucl. Eng. Des. 218, 25–32. Xie, F., Cao, J., Chen, Z., Dong, Y., 2015a. The design and study of the radioactive graphite dust experimental system in the primary loop of the HTR-10. At. Energy Sci. Technol. 49 (4), 744–749 (in Chinese). Xie, F., Cao, J., Tong, J., Wei, L., Liu, X., Dong, Y., Yu, S., 2015b. The research progress on the radioactive graphite dust in HTR-10. In: Proceedings of ICONE23, May 17–21, 2015, Chiba, Japan. Xie, F., Li, H., Cao, J., Yu, S., Zhang, L., Li, W., Fang, S., 2013. A reform in the helium purification system of the HTR-10: γ dose rate measurement and suggestions for decommissioning. In: Proceedings of ICEM 2013, September 8–12, 2013, Brussels, Belgium. Xu, Y., Li, H., Xie, F., Cao, J., Tong, J., 2017. Source term analysis of tritium in HTR-10. Fusion Sci. Technol. 71, 671–678. Zhang, Z., Yu, S., 2002. Future HTGR developments in China after the criticality of the HTR-10. Nucl. Eng. Des. 218, 249–257.
The concentrations of the radioactive dust in the primary loop of HTR-10 deduced from the collected data are 5.57(17) μg/m3STP and 1.97(8) μg/m3STP. The average particle diameters, as obtained with collection by the filter elements of 80 μm pore size and 50 μm pore size are respectively, 43.0 μm and 13.6 μm in the first experiment and 62.9 μm and 11.3 μm in the second experiment. This indicates that dust particles of diameter larger than 10 μm, make up a significant portion of particle mass in the primary loop of HTR-10. The presence of fine particles needs to be investigated with the filter elements of smaller pore size and electronic microscopy. The specific activities in the dust of HTR-10 are conservatively estimated as 3.3 × 104 Bq/g and 6.5 × 103 Bq/g for Co-60, and 1.5 × 102 Bq/g and 1.0 × 102 Bq/g for Cs-137. Direct comparisons of the present data and results with the measurements in AVR are difficult, because of many factors (graphites, operational characteristics, measurement techniques, experimental uncertainties, etc.). Nevertheless it should be noted that both the amount of the dust and the activities associated with the HTR-10 dust are lower than those reported for AVR. The methods, data and analysis reported here are of basic significance for not only the radiation safety evaluation of HTR-10, but also other pebble bed reactors as regards the study of behavior of fission and activation products and characteristics of radioactive dust. Acknowledgements This work was supported by the National Natural Science Foundation of China (No. 11575099), the Beijing Natural Science Foundation (No. 2163051), the Chinese National Significant Science and Technology Project (No. ZX06901), and the Tsinghua University Initiative Scientific Research Program (No. 20151080375). Fruitful discussions with Prof. Dazhi Xue, Prof. Meisheng Yao, Prof. Fu Li, Mr. Liqiang Wei and Mr. Ling Liu at INET, Tsinghua University, Beijing, China, and Dr. Terttaliisa Lind at Paul Scherrer Institute, PSI, Switzerland, are gratefully acknowledged. We also thank one of the anonymous reviewer for pointing out the production mechanism of Hg203. References Barth, T., Kulenkampff, J., Bras, S., Gründig, M., Lippmann-Pipke, J., Hampel, U., 2013. Positron emission tomography in pebble beds. Part 2: graphite particle deposition and resuspension. Nucl. Eng. Des. 267, 227–237. Bäumer, R., Barnert, H., Baust, E., Bergerfurth, A., Bonnenberg, H., Bülling, H., Burger, St., Von der Decken, C.-B., Delle, W., Gerwin, H., Hackstein, K.-G., Hantke, H.-J., Hohn, H., Ivens, G., Kirch, N., Kirjushin, A.I., Kröger, W., Krüger, K., Kuzavkov, N.G., Lange, G., Marnet, C., Nickel, H., Pohl, P., Scherer, W., Schöning, J., Schulten, R., Singh, J., Steinwarz, W., Theymann, E., 1990. AVR: Experimental High-Temperature Reactor: 21 Years of Successful Operation for a Future Energy Technology. Association of German Engineers (VDI). The Society for Energy Technologies. Boddu, S.R., Gutti, V.R., Meyer, R.M., Ghosh, T.K., Tompson, R.V., Loyalka, S.K., 2011. Carbon nanoparticle generation, collection, and characterization using a spark generator and a thermophoretic deposition cell. Nucl. Technol. 173, 318–326. Cogliati, J.J., Ougouag, A.M., Ortensi, J., 2011. Survey of dust production in pebble bed reactor cores. Nucl. Eng. Des. 241, 2364–2369. Fachinger, J., Barnert, H., Kummer, A., Caspary, G., Seubert, M., Koster, A., Makumbe, M., Naicker, L., 2008. Examination of dust in AVR pipe components. In: Proceedings of HTR 2008, 28 September-1 October 2008, Washington DC., USA. Gottaut, H., Krüger, K., 1990. Results of experiments at the AVR reactor. Nucl. Eng. Des.
378