Abstract
A radiation shield consists of borated polyethylene (7 % wt of boron) and lead was used for the HPGe detector in Isfahan MNSR's PGNAA facility in order to remove scattered neutrons and gamma rays, respectively. Therefore, the boron peak in gamma spectroscopy is related to both resources in the sample and in the shield, and it has caused a challenge in determination of boron concentration in the samples. In this research, various methods have been investigated to remove the effect of boron in the shield of the detector. For this purpose, Monte Carlo simulations of radiation transportation have been used. The results showed that by covering the detector shield using a cadmium sheet can greatly (99.32 %) reduce the height of the boron peak in prompt gamma spectrum recorded in the detector that come from the shield. The experiments have been done to validate the results obtained from the simulations.
Keywords: PGNAA, MNSR, Boron concentration, Detector shield, Monte Carlo
1. Introduction
Measurement of boron concentration is important in material analysis like detergents and bleaches, glass, ceramics, insecticides, textiles, and semiconductors [1]. Despite the NAA that had a delay time between the neutron irradiation and gamma spectroscopy, neutron irradiation of sample and acquisition of prompt gamma rays must be at the same time in PGNAA [2]. Since the detection of these gamma rays is usually done using HPGe, the detector needs an appropriate radiation shield in order to prevent neutron damage and decrease the background level of the recorded spectrum [3]. The detector shield in the PGNAA facilities consists of two main parts, one to absorb background gamma rays and the second to absorb neutrons scattered from the sample. Therefore, absorbers such as lead, tungsten, and steel are usually used to reduce the gamma background, and absorbers such as boron, lithium and cadmium are used to remove neutrons [4].
Robinson conducted a study on the optimization of the Oregon Research Reactor PGNAA facility in the United States in 2009. The detector shield used in this facility includes lead to absorb background gamma rays and borated polyethylene to remove neutrons [5]. Zhang and his colleagues used the lead with a thickness of 10 cm and Li2CO3 (in which 6Li was enriched up to 95 %) with a thickness of 5 mm in the detector shield of a PGNAA system related to In-Hospital Neutron Irradiator (IHNI) in Beijing [6]. Raja performed a study in 2020 on the analysis of solid samples by PGNAA and Particle-induced gamma emission (PIGE) methods and the detector shield consisted of lead bricks with a thickness of 15 cm and a thin layer of cadmium on them [7]. In addition, a hole of 1 cm in diameter was embedded in the lead bricks in front of the detector so that the prompt gamma rays of the sample enter the detector.
One of the problems of detector shields, where boron is used in their structure, is in measurement of boron concentration. The boron peak in gamma spectroscopy is caused by both resources in the sample and in the shield. Therefore, it causes disturbances and errors in measurement of the boron concentration in samples.
Recently a PGNAA facility has been installed in Isfahan MNSR [[8], [9], [10], [11], [12]]. A radiation shield consists of borated polyethylene (7 % wt of boron) and lead was used for the HPGe detector in Isfahan MNSRs’ PGNAA facility [12]. In this research, various methods have been investigated without physically changing this shield to remove the effect of boron peak caused by the shield and accurately determine the boron concentration in samples. Therefore, Monte Carlo simulations of radiation transportation have been used. The experimental test also examined the chosen method to validate the obtained results from the simulations.
2. Materials and methods
2.1. Isfahan MNSR and PGNAA facility
Isfahan MNSR is a low-power (30 kW) and tank-in-pool research reactor. This research reactor is used for education and training, research, neutron radiography, and elemental analysis by NAA method. Light water is used as moderator, coolant, and biological shield with natural convection flow in the core. 343 fuel rods containing high enriched uranium in ten concentric circles, they form the core of the reactor and the core is surrounded by an annular beryllium reflector. This reactor has 5 inner irradiation sites (IIS) inside the annular beryllium and 5 outer irradiation sites (OIS) outside the beryllium. The thermal neutron fluxes at 30 kW reactor power are 1 × 1012 n/cm2s at the IIS and 5 × 1011 n/cm2s at the OIS [[13], [14], [15], [16]]. Fig. 1 shows the schematic view of the Isfahan MNSR.
Fig. 1.
The schematic view of Isfahan MNSR shows the reactor component [10].
MNSR has low-power and originally designed for education and training, neutron activation analysis, and production of short-live radioisotopes. In order to expand the utilization of reactor, neutron beam tubes were added to the reactor [13,14,[17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31]].
Another thermal neutron beam was installed inside the tank of reactor. In addition, the feasibility study of performing PGNAA experiments has been done using an HPGe detector and appropriate radiation shield [[8], [9], [10], [11], [12]].
The radiation shield of HPGe detector consist of two main parts; lead and borated polyethylene (BPE). The lead part is a cylinder with height of 15 cm and outer diameter of 19 cm. There are different hollows inside this cylinder to place the detector and collimator. The BPE part is a 20 x 30 × 30 cm3 cubic consists of different inside hollows to place the lead part and collimator. This radiation shield has a dual function to absorb both the scattered neutrons and prevent to avoid hitting the detector's crystal and absorb the background gamma radiation to reduce the noise of gamma spectrum [8,12]. The collimator that directs the gamma rays of the sample to the detector is machined inside these two parts (lead and BPE) and includes holes with different diameters 3, 3.5, 4, 6, 8 and 13 cm. Fig. 2a shows the position of beamline, reactor tank, and detector shield and Fig. 2b show the detail structure of detector shield. As can be seen, there is no lead in the first 5 cm of the entrance hole of collimator in the shield. The diameter of the inlet hole of the collimator is 3 cm. The detector shield is placed 90 cm higher than the reactor hall floor and 160 cm from the pool water level. The distance of the sample position to the inlet hole of the collimator and the surface of the detector are 10 cm and 15 cm, respectively.
Fig. 2.
a) The schematic view of the Isfahan MNSR's PGNAA facility of and b) the assembled detector shield of detector [12].
2.2. Correction approaches
The boron content in the BPE part of the detector shield causes an interference in determination of boron concentration of samples. Here, different methods to remove the effect of boron present in the detector shield have been investigated. The MCNP as a Monte Carlo (MC) radiation transport code has been used in this regard. It makes it possible to simulate and calculate photon and neutron transport in different media and geometries [[32], [33], [34]]. The number of histories was considered so that the statistical error of calculations remain below 5 %.
In addition, an experimental setup was provided to determine the impact of boron content in the BPE part of detector shield on the measurement of boron concentration. For this purpose, two gamma spectra corresponding to the background and 20 ml of pure water were recorded for 1000 s. The reactor was 15 kW. All geometrical arrangements of the sample - detector shield - detector in the simulations is similar to the experiments.
2.2.1. Background correction method
Thermal neutrons absorb by boron and emit alpha particle in 10B(n, α)7Li reaction and after that emits 477.6 keV prompt gamma rays. The boron peak in the gamma spectrum at this energy is due to these gamma rays. By irradiating the boron containing sample with neutrons, some neutrons are absorbed in the boron nuclides in the sample and produce 477.6 keV gamma rays. Some other neutrons are also scattered from the nuclei in the sample, including hydrogen, and go towards the detector shield and absorbed in the boron nuclides of the shield and produce another 477.6 keV gamma rays. Therefore, the surface area under the boron peak will be due to boron inside the sample and boron inside the shield. To accurate determination of the boron concentration inside the sample, the contribution caused by the boron inside the shield must be subtracted from the under peak area. One of the ways to improve the measurement of boron concentration can be to measure and correct the gamma rays reaching the detector due to boron present in the shield from the main spectrum of the samples. If the spectrum is taken from a sample containing pure water, in this case all the 477.6 keV gamma rays recorded in the spectrum are due to the boron in the shield. In this case, the under the boron peak area is considered as the background. Now, if there is some boron in the sample in addition to the pure water, the boron peak in the spectrum taken from the sample also includes the gamma rays emitted from the boron content of the shield and the boron in the sample. The background correction method is based on calculating the under boron peak area in the spectrum of pure water and subtracting it from the corresponding peak in the spectrum of samples containing boron.
Therefore, the spectroscopy process of 10 cc of samples containing 0 %–0.85 % mass fraction of boron was simulated. The weight fraction of each element nuclei in the samples is given in Table 1. However, the correctness of this method requires the stability of the background boron peak for samples containing different amount of boron. Since the effect of boron in the shield on the spectrum is directly related to the scattered neutrons from the sample and reaching the shield, the number of scattered neutrons from different samples and reaching the detector shield must be calculated first. For this purpose, the number of neutron particles crossing the shield surface (tally F1) was calculated. The number of histories was considered so that the statistical error of calculations remain below 5 %.
Table 1.
The weight fraction (WF) of each element nuclei in the samples containing different boron mass fractions from 0 % to 0.85 %.
| Mass fraction of natural boron | WF of | WF of | WF of | WF of |
|---|---|---|---|---|
| 0.00 % | 11.18983 % | 88.81017 % | 0 % | 0 % |
| 0.005 % | 11.18803 % | 88.80697 % | 0.00092 % | 0.00408 % |
| 0.01 % | 11.18623 % | 88.80377 % | 0.00184 % | 0.00816 % |
| 0.02 % | 11.18263 % | 88.79737 % | 0.00368 % | 0.01632 % |
| 0.05 % | 11.17182 % | 88.77818 % | 0.0092 % | 0.0408 % |
| 0.10 % | 11.15296 % | 88.74704 % | 0.01843 % | 0.08157 % |
| 0.25 % | 11.09976 % | 88.65024 % | 0.04608 % | 0.20392 % |
| 0.50 % | 11.00969 % | 88.49031 % | 0.09216 % | 0.40784 % |
| 0.75 % | 10.91961 % | 88.33039 % | 0.13823 % | 0.61177 % |
| 0.85 % | 10.88358 % | 88.26642 % | 0.15667 % | 0.69333 % |
2.2.2. Gamma removal method
Another way to the improvement of boron concentration measurement can be to remove the gamma rays produced in the shield and prevent them from reaching the detector. This method has been implemented based on placing sheets of lead in the inner wall corresponding to the inlet hole of the collimator. The available space for placing the lead layer is limited because the diameter of the inlet of the collimator is only 3 cm. Simulations were performed with different thicknesses of the lead layer. The gamma spectra were obtained from 20 ml of pure water. The energy spectrum of the prompt gamma rays in the position of the HPGe has been calculated by the pulse height tally (F8). Also, in order to make more precise calculations and considering peak broadening characteristics observed in a real spectrum, a Gaussian Energy Broadening card (GEB) has been incorporated in simulations. The simulated geometry of this lead layer in the inlet hole of the detector shield is shown in Fig. 3.
Fig. 3.
The simulated geometry of the additional lead layer in the inlet hole of the collimator.
2.2.3. Neutron removal method
Preventing scattered neutrons from the sample to reach the detector and its shield can be another way. This method has also been implemented by placing a neutron absorber layer. Cadmium is a material with good availability, economical price, no need for enrichment, and high thermal neutron absorption cross-section. There are many peaks in the prompt gamma neutron activation spectrum of cadmium, but only one of them is in the range of the boron peak [2]. So here cadmium has been used to remove the scattered neutrons.
As shown in Fig. 4, a cadmium sheet has been used as a cover. The thickness of this sheet is 0.5 mm. The effectiveness of cadmium in correcting the impact of boron content in the shield has been investigated by recording the gamma spectrum from 20 ml of pure water in simulations. In addition, an experimental setup based on this neutron removal configuration was implemented for validating the Monte Carlo simulations.
Fig. 4.
a) The simulated geometry of the additional cadmium cover. b) Experimental setup of the neutron removal configuration at the Isfahan MNSR's PGNAA facility.
Fig. 4a shows the simulated geometry of the additional cadmium cover and Fig. 4b shows a photo of experimental setup of the neutron removal configuration at the Isfahan MNSR's PGNAA facility.
3. Results and discussions
As shown in the two gamma spectra in Fig. 5, one of which was recorded in the presence of 20 ml of pure water and the other without the presence of the sample, only related to the background in the Isfahan MNSR's PGNNA facility, the boron peak can be seen at the energy of 477.6 keV. Among the other most prominent peaks close to this energy is the annihilation peak of 511 keV and the germanium peak of 595.9 keV (caused by the (n, γ) reactions in the detector).
Fig. 5.
The recorded gamma spectrum corresponding to the background and 20 ml of pure water in the Isfahan MNSR's PGNNA facility.
As shown in Fig. 5, the boron peak is wider than other peaks, this broadening is due to the Doppler phenomenon. The reason for the occurrence of the Doppler phenomenon is that the prompt gamma rays obtained from the absorption of neutrons in boron are actually emitted from the excited state of the recoiling ∗7Li nucleus [7].
Since the recording time of both spectra was 1000 s, the higher spectrum counts related to the presence of the sample are due to the increased amount of neutrons scattered around due to the collision with hydrogen nuclei in pure water. The counting of the area under the boron and germanium peaks for these two spectra is given in Table 2.The comparison of the two corresponding peaks of boron and germanium shows that the counts of gammas from the pure water sample are almost 8.5 and 15 times that of the case without sample, respectively.
Table 2.
Counting of the area under the boron and germanium peaks for two recorded spectra in the Isfahan MNSR's PGNNA facility.
| Case | Counts of boron peak | Counts of germanium peak |
|---|---|---|
| Without sample | 31620 ± 264 | 3459 ± 118 |
| Pure water sample | 280352 ± 691 | 52154 ± 362 |
In the investigation of the background correction method, the calculation result of the neutrons scattered current from different boron-containing samples and reaching the detector shield is shown in Fig. 6It can be seen that the current of scattered neutrons from the samples that reach the detector and its shield decreases significantly with increasing the concentration of boron. So that the scattered neutrons current from sample with the highest boron mass fraction is about 33.7 % less than the pure water sample. Therefore, the impact of boron in the shield on the under boron peak area in the spectrum taken from that sample will be different for each sample.
Fig. 6.
The current of neutrons scattered from different boron-containing samples reaching the detector shield.
Therefore, it is not possible to measure the boron peak background in pure water and subtract it from other spectra, and the background correction method does not have a suitable result. A control element can be used to obtain the amount of changes in the neutron current scattered from each sample and calculate a background correction factor for it. However, an element that can have suitable characteristics as a background measurement indicator is not present in the sample materials, shield and detector.
In the investigation of the method of gamma removal, the calculation results of counting corresponding the boron peak in the recorded gamma spectrum of 20 ml pure water for the lead layer with different thicknesses in the entrance hole of the detector shield is summarized in Table 3. Due to the limitation in the diameter of the entrance hole, it is practically impossible to use lead thicknesses greater than 0.6 cm. Consequently, the attenuation of 477.6 keV gamma rays produced in the borated polyethylene of the shield will not be done completely, and it is not possible to measure the amount of the remaining gamma rays that pass through the lead.
Table 3.
The counting of the boron peak of 20 ml pure water for the lead layer with different thicknesses in the entrance hole of the detector shield.
| Thickness of lead layer(cm) | Counts of boron peak | Entrance area of the shield opening hole (cm2) |
|---|---|---|
| 0 | 15751 ± 131 | 7.069 |
| 0.2 | 6448 ± 87 | 5.309 |
| 0.4 | 2492 ± 56 | 3.801 |
| 0.6 | 1066 ± 38 | 2.545 |
Fig. 7 depicts the simulated and recorded gamma spectra with and without of cadmium cover for 20 ml of pure water. The spectra are just given in the energy range of 400–700 keV because of the location of the boron peak. Despite the measured spectrum, the prompt gamma peaks of cadmium and germanium are not seen in the calculated gamma spectrum using Monte Carlo simulations due to the inability of this method and deficiency in the ENDF-VII cross section library [33].The vertical axis is shown as normalized until both the experimental and simulated spectra without cadmium cover are brought together.
Fig. 7.
The gamma spectra simulated and experimentally recorded with and without cadmium cover for 20 ml of pure water.
The effect of cadmium cover in removing the scattered neutrons reaching the detector shield is clearly evident with the fading of the boron peak in the spectrum. The experimental spectrum after placing the cadmium cover has several other peaks in this energy range, including energies of 558 and 651 keV. These peaks are not included in the simulated gamma spectrum due to the mentioned cross section library deficiency. In addition, for this reason, the simulated spectrum with coverage has fewer counts than the spectrum without coverage.
Boron peak counts are determined by measuring the under peak area and are summarized in Table 4.The reduction of the under boron peak area occurs while, unlike the gamma removal method, the area of the entrance opening of the shield does not decrease, and there is no need to increase the spectroscopic time.
Table 4.
The counting of boron peak with and without cadmium cover for the 20 ml of pure water sample.
| Case | Experiment | Simulation |
|---|---|---|
| With cadmium cover | 18895 ± 619 | 212 ± 20 |
| Without cadmium cover | 280352 ± 691 | 31013 ± 188 |
| Relative difference (%) | 93.26 % | 99.32 % |
The results of this research have been successfully used to overcome the challenge of presenting the boron in the detector shield and boron concentration has been determined in liquid samples [35].
4. Conclusion
The use of the detector shield, which includes borated polyethylene, in the Isfahan MNSR's PGNAA facility leads to the presence of the boron peak in the recorded gamma spectrum. Therefore, it causes disturbances and errors in the precise determination of boron concentration in samples. Several methods were investigated to remove this peak from the spectrum. Monte Carlo calculations showed that designing a cadmium cover can greatly (99.32 %) reduce the height of the boron peak in the recorded gamma spectrum. According to this, the experimental arrangement was implemented, and its data showed a good conformity with the simulation results. However, the presence of cadmium increases the background of the spectrum and the lower limit of detection of other elements.
CRediT authorship contribution statement
M. Jafari: Writing – original draft, Visualization, Validation, Investigation, Data curation, Conceptualization. H. Jafari: Writing – review & editing, Visualization, Validation, Supervision, Investigation, Data curation. M. H. Choopan Dastjerdi: Writing – review & editing, Visualization, Validation, Supervision, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. J. Mokhtari: Writing – review & editing, Visualization, Investigation, Data curation.
Data and code availability
The authors do not have permission to share data.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References
- 1.Bakirdere S., Orenay S., Korkmaz M. Effect of boron on human health. Open Miner. Process. J. 2010;3(1) [Google Scholar]
- 2.Molnar G., editor. vol. 1. Springer Science & Business Media; 2004. (Handbook of Prompt Gamma Activation Analysis: with Neutron Beams). [Google Scholar]
- 3.Advanced Measurement Technology, Inc Neutron radiation damage in HPGe detectors. 2007. https://www.ortec-online.com/service-and-support/library/tech-notes/neutron-damage Retrieved October Wednesday, 2008, from ortec-online:
- 4.Alfassi Z.B., Chung C. 1995. Prompt Gamma Neutron Activation Analysis. [Google Scholar]
- 5.Robinson J., Hartman M., Reese S. Design, construction and characterization of a prompt gamma activation analysis facility at the Oregon State University TRIGA® reactor. J. Radioanal. Nucl. Chem. 2010;283(2):359–369. . [Google Scholar]
- 6.Zhang Z., Chong Y., Chen X., Jin C., Yang L., Liu T. PGNAA system preliminary design and measurement of In-Hospital Neutron Irradiator for boron concentration measurement. Appl. Radiat. Isot. 2015;106:161–165. doi: 10.1016/j.apradiso.2015.07.049. . [DOI] [PubMed] [Google Scholar]
- 7.Raja S.W., Samanta S.K., Sharma V., Acharya R., Pujari P.K. Application of PGNAA utilizing thermal neutron beam for quantification of boron concentrations in ceramic and refractory neutron absorbers. J. Radioanal. Nucl. Chem. 2020;325:933–940. . [Google Scholar]
- 8.Dastjerdi M.C., Mokhtari J., Toghyani M., Soleimani B. Feasibility study on PGNAA experiments using a prototype neutron beam at Isfahan MNSR. J. Instrum. 2023;18 [Google Scholar]
- 9.Ghasemi Z., Hassanvand M., Dastjerdi M.C., Mokhtari J. Design of a radiation shield for an HPGe detector for a reactor-based PGNAA facility. Radiat. Phys. Chem. 2023;210 [Google Scholar]
- 10.Vatani M., Hassanvand M., Mokhtari J., Dastjerdi M.C. Design of an in-tank thermal neutron beam for PGNAA application at Isfahan MNSR. Nucl. Eng. Des. 2023;412 [Google Scholar]
- 11.Rahmati E., Barough M.S., Dastjerdi M.C., Mokhtari J. Design of an out-of-tank neutron beam line for prompt gamma neutron activation analysis at low-power Miniature Neutron Source Reactor. J. Instrum. 2023;18(9) [Google Scholar]
- 12.Dastjerdi M.C., Mokhtari J., Toghyani M. Design, construction, and characterization of a prompt gamma neutron activation analysis (PGNAA) system at isfahan MNSR. Nucl. Eng. Technol. 2023 [Google Scholar]
- 13.Mokhtari J., Choopan Dastjerdi M.H. Development and characterization of a large thermal neutron beam for neutron radiography at Isfahan MNSR. Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip. 2023;1051 . [Google Scholar]
- 14.Dastjerdi M.C., Mokhtari J., Asgari A., Ghahremani E. A neutron radiography beamline relying on the isfahan miniature neutron source reactor. Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip. 2019;928:20–25. . [Google Scholar]
- 15.Mokhtari J., Faghihi F., Dastjerdi M.C., Khorsandi J. Neutronic feasibility study of using a multipurpose MNSR for BNCT, NR, and NAA. Appl. Radiat. Isot. 2020;161 doi: 10.1016/j.apradiso.2020.109147. [DOI] [PubMed] [Google Scholar]
- 16.Moslehi A., Dastjerdi M.C., Torkzadeh F., Mokhtari J. Feasibility study of Isfahan MNSR as a calibration thermal neutron source. Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip. 2022;1024 [Google Scholar]
- 17.Pourimani R., Choopan Dastjerdi M.H., Abdi M. Determination of trace elements in different pistachio variety by instrumental neutron activation analysis technique. Iranian Journal of Science. 2023;47(2):581–588. [Google Scholar]
- 18.Tabas E.J., Mokhtari J., Abbassi Y. Application of artificial neural network for prediction of operational performance of MNSR. Nucl. Eng. Des. 2024;419 [Google Scholar]
- 19.Pour Imani R., Choopan Dastjerdi M.H., Ghahani S., Mashayekhi M. Iranian Journal of Applied Physics; Iran: 2024. Determining Trace Elements in Agricultural Products of Chogha Village of Arak. [Google Scholar]
- 20.Pourimani R., Dastjerdi M.H.C., Farahani E.H., Siavashani M.R.A. Measurements of concentrations of some elements in the meat and skin of farmed and marine fish of the Persian Gulf using the neutron activation analysis technique. The European Physical Journal Plus. 2024;139(6):552. [Google Scholar]
- 21.Asgari A., Hosseini S.A., Choopan Dastjerdi M.H., Mokhtari J. Power calibration of MNSR using solid state track detector. Journal of Radiation Safety and Measurement. 2023;12(3):145–153. [Google Scholar]
- 22.Pourimani R., Choopan Dastjerdi M.H., Shekari M., Feyzi M. Analysis of trace elements in various types of Iranian and imported rice using the neutron activation method. Radiation Physics and Engineering. 2024 [Google Scholar]
- 23.Gholamzadeh Z., Ebrahimzadeh R., Choopan Dastjerdi M.H., Mokhtari J. Radiation Physics and Engineering. 2024. Assessment of shielding performance of NBR through simulation and experiments at MNSR beam line. [Google Scholar]
- 24.Asgari A., Hosseini S.A., Dastjerdi M.H.C., Mokhtari J. Determination of the neutron scattering contribution in a neutron calibration field based on MNSR. Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip. 2024;1061 [Google Scholar]
- 25.Abbassi Y., Mokhtari J., Mirvakili S.M., Dastjerdi M.H.C. Adoption of a fast dynamic model to simulate transient operations of MNSR. Nucl. Eng. Des. 2024;421 [Google Scholar]
- 26.Gholamzadeh Z., Bavarnegin E., Ebrahimzadeh R., Mokhtari J., Jafari M., Dastjerdi M.C. Experimental evaluation of transition rate of sapphire crystal for thermal and fast neutrons using MNSR vertical neutron beam line. Heliyon. 2024;10(2) doi: 10.1016/j.heliyon.2024.e24160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Asgari A., Dastjerdi M.H.C., Mokhtari J. A computational-experimental model of reactor kinetic for investigating the linearity response of in-core neutron detectors of a low power research reactor. Nucl. Eng. Des. 2024;424 [Google Scholar]
- 28.Bagherzadeh M., Karimi M., Choopan Dastjerdi M.H., Asadi Asadabad M., Mokhtari J., Babanejhad A. Long-time irradiation effect on corrosion behavior of aluminum alloy in pool water of low-power research reactor. Sci. Rep. 2023;13(1) doi: 10.1038/s41598-023-44287-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Bagherzadeh M., Dastjerdi M.C., Mokhtari J. Feasibility study of nanomaterials synthesis at MNSR research reactor through design and construction of a gamma irradiation cell. Nucl. Eng. Des. 2023;414 [Google Scholar]
- 30.Asgari A., Hosseini S.A., Dastjerdi M.H.C., Mokhtari J. Determination of the linear behavior of FC detectors in Isfahan MNSR using ex-core offline and online experiments. Nucl. Eng. Des. 2023;415 [Google Scholar]
- 31.Teimoory E., Allaf M.A., Mokhtari J., Dastjerdi M.C. Development and characterization of fission chamber neutron detectors in Isfahan miniature neutron source reactor. Radiat. Phys. Chem. 2024;215 [Google Scholar]
- 32.Ringuest J.L. Springer Science & Business Media; : 2012. Multiobjective Optimization: Behavioral and Computational Considerations. [Google Scholar]
- 33.Pelowitz D.B., Durkee J.W., Elson J.S., Fensin M.L., Hendricks J.S., James M.R.…Wilcox T.A. Los Alamos National Laboratory; : 2011. MCNPX 2.7. 0 Extensions. [Google Scholar]
- 34.Kargaran H., Jafari H., Minuchehr A. A novel approach based on improved combinatorial geometry (ICG) model for photon transport in Monte Carlo codes. Prog. Nucl. Energy. 2021;131 . [Google Scholar]
- 35.Jafari M., Jafari H., Dastjerdi M.C., Mokhtari J. Designing a system of boron concentration measurement in solution samples by the PGNAA facility of the Isfahan MNSR reactor. Nucl. Eng. Des. 2024;416 [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The authors do not have permission to share data.







