Neutronic Aspect Of Samop Reactor

Table of Contents International Conference on Advances in Nuclear Science and Engineering in Conjunction with LKSTN 2007 (23-26)

Neutronic aspect of Subcritical Assemby for Mo-99 Production (SAMOP) Reactor Topan Setiadipura* , Elfrida Saragi Computational Division PPIN-BATAN, Serpong, Indonesia2Affiliation, City, Country *E-mail:[email protected]

Abstract NEUTRONIC ASPECT OF SUBCRITICAL ASSEMBLY FOR Mo-99 PRODUCTION (SAMOP) REACTOR. Design of a subcritical assembly for Mo-99 production (SAMOP) is in progress at National Nuclear Energy Agency. The main purpose of the project is to b able to produce Mo-99, which is a parent of Tc-99m an important nuclide for nuclear medicine application. The major source of Mo-99 is from fission of U-235. The conventional technique is by forming the uranium into targets and irradiated by neutrons from research or test reactors, then the irradiated targets are dissolved and the Mo-99 is extracted from the solution. Another technique to produce Mo-99 from U-235 is by using homogeneous reactor fueled with uranyl nitrate. This method introduced by Ball in 1992 and some advantages compared with the conventional method. On the SAMOP design, the low enriched uranyl nitrate is placed in the stainless steel container and irradiated by neutron generator. In this paper, the neutronic aspect of the design will be reported. Including the criticality analysis to secure the subcriticality of the design and the neutron flux distribution analysis, also the effect of the graphite reflector. The neutronic analysis was using the general monte carlo code, MCNP. Keywords: Mo-99, uranyl nitrate, subcritical, neutronic, monte carlo

1. Introduction 99

Mo is used as a parent isotope of the widely used medical radiotracer 99mTc. It is estimated that 99mTc is used in over 85%of all nuclear medicine clinical studies in the world. The strong demand for 99mTc has stimulated a search for reliable supplies of 99Mo [1]. Basically, there are two process or reaction to produce 99mTc, fission from U-235 or capture reaction of 98Mo as shown in the picture 1. The fission yield of 99Mo is about 6.1%. The fissionproduced 99Mo has a high specific activity (~104 Ci 99 Mo /g Mo) which makes it the most important source of 99Mo in the world. The fission technique, however, requires considerable capital investment and produces large quantities of radioactive waste. The cross section of the 98Mo (n,γ) 99Mo reaction is small (σth ~ 0.14 barns) and only a small portion of the 98Mo is converted to 99Mo. The resulting specific activity (~Ci 99Mo /g Mo) is much lower than that of the fission-produced 99Mo. Advanced generator technologies are required to produce high quality 99mTc generators from the capture-produced 99Mo. For the fissioning process, the conventional way is to form the U-235 into a target which then irradiated by neutrons from research or test reactor. These irradiated targets are dissolved and the fission

product, 99Mo, is extracted from the solutions. In 1992, Ball [2] introduced a method of "targetless" production of fission product Mo-99 using an aqueous homogeneous reactor fueled with uranyl nitrate. The design anticipated that the uranium salt could be made with low enriched uraniurn [3,4].

Figure 1. Two different reaction to produce 99 Tc from 99Mo. Subcritical Assembly for Mo-99 Production (SAMOP) is designed at the PTAPB-BATAN based on the Ball’s method. The core of the SAMOP is the Uranyl Nitrate solution in the SS-304 tank which irradiated by the neutrons from a D-T neutron

23

International Conference on Advances in Nuclear Science and Engineering in Conjunction with LKSTN 2007 (23-26)

generator. Geometrical data of the SAMOP design is given in the table below Table 1. Samop Geometrical Data Parameter Core tank (Inner tank) Diameter Height Uranyl Nitrate Solution Height SS-304 thick Coolant Tank (Outer tank) Diameter Height Distance Inner Tank – Reflector Distance Inner Tank – Outer tank based

Value 15.35cm 35cm 30.7cm 0.3cm 80cm 190cm 1cm 40cm

2. Methodology The neutronic calculation is done using MCNP [6], a general purpose monte carlo code. To do the calculation on the MCNP we modeled the neutron source term from the neutron generator, the geometry and material of the SAMOP reactor. The geometrical model of the SAMOP reactor on the Visual Editor of MCNP is shown in the picture below:

θ, degree

Intensity

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180

1,059 1,048 1,046 1,043 1,038 1,031 1,024 1,017 1,008 1,000 0,992 0,984 0,976 0,969 0,964 0,959 0,956 0,954 0,953

Energy (MeV) 14.962 14.948 14.906 14.839 14.748 14.637 14.509 14.368 14.220 14.069 13.919 13.776 13.642 13.523 13.421 13.339 13.278 13.241 13.229

Uranyl Nitrate used in this calculation is enriched to 20%, and the Uranium concentration is 300 g/L. The material composition of Uranyl Nitrate is given in the table below Table 3. Material composition of the Uranyl Nitrate Nuclide

Figure 2. Geometrial model of the SAMOP reactor, from the side of the reactor (left) and from the top of the reactor. (right). Energy and intensity of the source neutron as a function of the angle is modeled using several probability distribution. The intensity distribution is modeled using Source Information (SI) card for the cosines of the angle where the intensity is known and the Source Probability (SP) card for the related intensity. Dependent Source (DS) cards is used for the energy distribution because the energy is depends on the direction distribution. The intensity (normalized for the value of angle 90o degree) and energy of the neutron source is given in the table below [5] Table 2. Neutron source characteristic

U-234 U-235 U-238 N H O

Atomic Density (atom/barn cm) 1.2977E-06 1.5374E-04 6.0591E-04 1.5219E-03 5.5457E-02 3.3816E-02

KCODE card is used to calculate the effective multiplication factor, k-eff, of the SAMOP core. And the F4 tally is used to calculate the flux distribution of the SAMOP core. Using the F4 tally, the average neutron flux is estimated by summing the neutrons track length in the cells. The track length estimator is generally quite reliable to because there are frequently many tracks in the cell (compared to the number of the collisions), leading to many contribution to this tally[6]. To calculate the flux of different area of the core, F4 tally is applied to a small spherical cells with different position that represented the area of the core. To simplified the tallying, the height of the cells position is divided into three level, and in each level there are nine cells 24

International Conference on Advances in Nuclear Science and Engineering in Conjunction with LKSTN 2007 (23-26)

represented the center and the pheripery of the core at that level. The detail cells configuration and numbering is shown in the figure below,

Effect of diferent Graphit Reflector Width 0.99 0.98 0.97 0.96

k-eff

0.95 0.94 0.93 0.92 0.91 0.9 0.89 0

5 10 15 20 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 tebal reflektor (cm)

Figure 5. Effect of Diferent Graphite Reflector Thicknes

Figure 3. Cell configuration for detail flux calculation. The above figure picturing the cells of the upper level, with the center of the cell height is 27.55cm (from the base of the core tank), cells 17-25 for the middle level, and cells 26-34 for the lower level with the height of the cell’s center is 15.35 and 3.15 respectively.

From the calculation, the k-eff of the core without the reflector is 0.92319 which is to low for the application. This k-eff can be increased by additional (radial) reflector. The calculation shows that the increase of the k-eff is more significant by using berrylium reflector than using graphite reflector. The results also shows that using any reflector material, there is a limitation of the k-eff, it can be higher even the reflector thickness is increase. The flux distribution inside the SAMOP core for different reflector width is given in the figure Flux Distribution

3. Results and Discussion

Effect of diferent Be Reflector Width 1.04

0.04 0.035 0.03

sam_f3

0.025

sam_f4

0.02

sam_g6

0.015 0.01

1.02

0.005

1

8 10 12 14 16 18 20 22 24 26 28 30 32 34

0

0.98

k-eff

0.05 0.045

N o rm a liz e d F lu x

The k-eff that represented the criticality condition of the SAMOP reactor is calculated for different width of the reflector. In this calculation we calculate for two reflector material berrylium and graphit, each with different width starting from zero to 40cm.

Cell Number

0.96

below Figure 6. Flux distribution of the SAMOP core for different reflector width.

0.94 0.92 0.9 0.88 0.86 0

5 10 15 20 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 tebal reflektor

Figure 4. Effect of Diferent Be Reflector Thicknes

The above figure picturing the distribution for reflector thickness 10cm, 5 cm and without reflector. As the reflector getting thicker the flux is getting higher as expected. The results shows that the higher flux is in the center of the core. The average flux in the core for different reflector thickness are given in the table below 25

International Conference on Advances in Nuclear Science and Engineering in Conjunction with LKSTN 2007 (23-26)

Table 4. Flux average of the SAMOP core for different reflector thickness Reflector thickness 0 (non reflector) 5cm 10cm 16cm

Average flux (n/cm2-s) 4.21e+9 6.85e+9 9.27e+9 1.16e+10

BATAN Yogyakarta for the opportunity to involve in the project.

References 1. 2.

4. Conclusion The neutronic aspect of the SAMOP reactor design is already done including the criticality of the core and the neutron flux distribution on the core. Regarding the criticality, it is confirmed that the core is subcritic and the level of the subcriticality, the k-eff, can be increase close to critical at least by using thicker reflector. And it is investigated that the Be reflector is much more effective to increase the k-eff of the core. The average flux in the core is about 1.16E+10 for the graphite reflector thickness 16cm where the k-eff value is between 0.97 – 0.98.

Acknowledgment

3.

4. 5. 6.

S.C.Mo (1993), Production of 99Mo Using LEU and Molybdenum Targets, RERTR Meeting 1993. Ball,R.M.(1992), Testimony Before the Congressional Committee on U.S. Resources on the Production of Mo-99 with Aqueous Homogeneous Reactors, Mike Synar, Chairman. Ball, R.M. (1994), Use of LEU in the Aqueous Homogeneuos Medical Isotope Production Reacto, RERTR Meeting, Williamsburg, Virginia 1994. Ball,R.M.(1995), The Mo-99 Solution, Nuclear Engineering International. Slamet Santoso (2007), Neutron Yield and Energy Calculation of thr Neutron Generator for SAMOP, not published. X-5 Monte Carlo Team (2003), “MCNP-A General Monte Carlo N-Particle Transport Code, Version 5 “ Vol.1, LA-UR-03-1987

The Authors would like to aknowledge Prof. Sarip and the SAMOP Team at the PTAPB-

Table of Contents 26