Radiation shielding design, verification and dose distribution calculations for industrial and insect irradiation facilities

Abstract

This study focused on analysis of the new radiation shield design, verification and dose distribution calculations for three irradiation facilities: The analysis of the new radiation design concrete shield for the Citrusdal facility, and shield verification calculations (radiological safety assessment) for the already constructed Stellenbosch and HEPRO irradiation facilities. Additionally, as a subordinate objective, the study reports on dose distributions in the irradiated insect containers of the Citrusdal irradiation facility, and sample product boxes of the HEPRO irradiation facility. In 2006 a private company, Citrus Research International (CRI), began the engineering design of a new insect sterilization facility in Citrusdal in the Western Cape region of South Africa, which was then constructed from ordinary concrete density (3 ¸ 2.35 g/cm$ ) in the year 2007. This facility employs a very successful ionising radiation based method, the Sterile Insect Technique (SIT), for drastically reducing or controlling insect population, which poses a serious pest problem in the Citrusdal agricultural area. The SIT uses the radionuclide 60Co as the source of ionising radiation. The design-base activity of the 60Co source for the Citrusdal facility was 740 TBq (20 kCi). Therefore, the new radiation shield design was done for this source strength. Analysis of the shield design has been carried-out using the integrated point-kernel (QAD-CGGP and the MathCAD worksheet) and confirmed with a Monte Carlo (MCNPX) radiation transport computer codes. This was done by considering the concrete shielding walls, the roof as well as the entrance/exit labyrinth at critical points for radiological protection purposes. However, only MCNPX was used to ascertain whether the labyrinth entrance/exit was radiologically safe because the point-kernel uses a semi-analytical, approximate methodology that is not able to account for radiation streaming through labyrinth. Satisfactory agreement was found between the three computational methods. Stellenbosch irradiation facility is being operated by the Agricultural Research Council as insect sterilization facility at Nietvoorbij, that is situated approximately 70 km from Cape Town. This facility uses a radiation based Sterile Insect technique; using 60Co as the source of ionising radiation. The facility is designed chiefly for sterilization of fruit flies which presents a major problem in the agricultural area in South Africa. The facility was designed for a 5 kCi (0.185 PBq) 60Co source, but is now operated with a 10 kCi (0.37 PBq) 60Co source (reference source strength in the year 1999). The source activity was 3.32 kCi (0.12 PBq) in October 2007, i.e., shield verification calculations was performed for this activity. This was carried out using QAD-CGGP and verified with MCNPX radiation transport computer codes, and where possible, validated with measurements. Dose rates were measured with a calibrated gamma-ray monitor, GRAETZ X5C. The spatial distribution of dose rates around the shielding concrete walls, on top of the roof and in various positions along the maze were considered. The dose rates at the closest distance (at 10 cm) from the exterior of the concrete shielding walls were slightly higher. However, good agreement was found between measured and simulated (calculated) values. HEPRO is a commercial irradiation facility situated in Ferrule Avenue, Montagu Gardens, Milnerton, City of Cape Town, Western Cape, South Africa. The HEPRO facility is mainly used for food irradiation and sterilization of medical instruments. This facility was originally designed to operate with a maximum '!Co source strength of 1 MCi (37 PBq). However, to increase the product throughput rate, the source strength had to be increased to a nominal 1.5 MCi (55.5 PBq) in 2005/6. Nevertheless a radiological safety assessment was performed for a source strength of 2 MCi (2 37 PBq 74 PBq) 60Co, because any radiation safety system such as radiation shielding must have a “safety factor”. Again the point-kernel method was used for radiological safety assessment and verified with MCNPX radiation transport computer code. Observation of critical spots and where possible the spatial distribution of dose rates were measured using a calibrated gamma-ray monitor, GRAETZ X5C, to provide additional confirmation of results. The spatial distribution of dose rates around the shielding concrete along the walls, roof, cable penetration inside the hut on roof, inside irradiation vault, above water pool, hotspot on wall, pneumatic pipe penetrations, sliding plug-door, and ducts of the hoistcable situated on the roof of the locked hut were considered. The comparison of dose rate values obtained using the aforementioned methods showed a slight deviation such that they were considered satisfactory. The use of Monte Carlo computational tools such as MCNPX in support of the prediction/assessment of the absorbed dose distributions in the irradiated product of the irradiation facilities can prove to be economically effective, representing savings in the utilization of dose meters, among other benefits. The absorbed dose distributions within the insect containers were calculated with the model developed in MCNPX for one hour of exposure. The modelling was carried out for a 60Co source with a nominal activity of 16.8 kCi, i.e., 622 TBq. The calculated values were compared to an average value from the preliminary experimental (sterility test) measurements. The sterility test for the insects, False Codling Moths, was done during the onset of the SIT programme at Citrusdal irradiation facility. The variability of the dose distribution inside the insect canisters were within acceptable dose values for adequate sterilising False Codling Moths. A systematic computational calculation methods was used to model/simulate the sample product undergoing irradiation in a 60Co source for HEPRO industrial irradiator facility. This was carried-out using the Monte Carlo code—MCNPX. The calculation was done for a source-rack containing 4 25 100 60Co pins, having a total activity of about 1 MCi (37 PBq). Four boxes were stacked around about 50 cm away from the source centerline. Each square box of sample product was rotated by 90° around the horizontal axis (perpendicular to the source rack), i.e., three times so that each of the four side of the box faced the source rack once during an irradiation session. A specific Fortran-90 code uses an algorithm to calculate the absorbed dose distribution for equal-time irradiation was therefore applied. The Fortran-90 code uses the dose values calculated by MCNPX and sums them in a suitable way in order to reproduce the mentioned rotations. The calculations revealed highest absorbed dose at the edges of the box, and even slightly higher at the four corners. The lowest absorbed dose was noticed at the centre of the sample product.