Artificial neural networks (ANN) have been applicable in almost every field, especially in basic sciences and engineering, in recent years. Nuclear physics, one of the fastest-growing fields of basic sciences, is often used in interpreting various experimental data or predicting the results of nuclear events. It has also been applied to Giant Dipole Resonance (GDR), which has been known since the 1930s and is a common feature for all nuclei in the periodic table. GDR is essential in understanding the internal structure of the nucleus, modeling nuclear reactions, and discovering the internal dynamics of atomic nuclei. Accurate estimation of GDR energies and peak widths is critical for interpreting the results of experimental studies in nuclear physics and validating theoretical models. GDR peak energies have been estimated using ANN for many even-even (even-N and even-Z) and odd-mass (odd-N, even-Z or odd-Z, even-N) nuclei in agreement with experiments. However, many odd-A nuclei exist, especially in the rare earth region, where ANN has not yet been applied. It is essential to complete the missing pieces of the puzzle to understand the behavior of GDR in this region.
The project aims to accurately estimate the GDR peak energies (E1 ve E2) of odd-N 167-179Lu and odd-Z 167-179Yb nuclei in the rare earth region where ANN has not been applied. By taking advantage of the strong learning capabilities of ANN, a large dataset will be created from the peak energies obtained in experimental studies within the project's scope, and estimations will be made using these datasets. The results will be compared with experimental data and theoretical calculation (TGI-QPNM available in the literature) to evaluate the ability of the ANN to predict GDR peak energies and widths in the odd-mass 167-179Lu and 167-179Yb nuclei.
This study aims to reveal the gamma shielding properties of the Fe3Cu1C compound obtained by doping pure Iron (Fe) with Copper (Cu) and Carbon (C) elements by exposing it to gamma radiation at different energies (1173 keV and 1332 keV) emitted from Co-60 radioactive source for the first time. In the study, linear attenuation coefficients (LAC) of Fe3Cu1C and pure Fe will be measured with a 3"×3" NaI(Tl) scintillation detector, and with this data, mass attenuation coefficient (MAC), half value layer (HVL), tenth value layer (TVL), and mean free path (MFP) characterizing gamma shielding properties will be determined. The same procedures will be performed for pure Fe to understand the effects of copper and carbon on the gamma radiation permeability of iron. The experimentally obtained data will be estimated and verified with artificial neural networks (ANN), popular modeling tools in recent years.
ANN is a method frequently used to estimate the gamma permeabilities of different materials. Still, it will be used for the first time in determining the gamma shielding properties of the newly produced Fe3Cu1C compound discussed in this study. For this purpose, a large data set will be created with the MAC of the Fe3Cu1C material at different energies from the XCOM database. The shielding properties of the Fe3Cu1C compound will be estimated in the 1173 keV and 1332 keV energies by utilizing the powerful learning capabilities of the ANN.
The subject of the project is to investigate the effect of the spin-spin interaction parameter (κ), which is responsible for the formation of M1 excitations in odd-A deformed nuclei, on the discrepancies between the experimental and Rotational Invariant Quasiparticle Phonon Nuclear Model (RI-QPNM) results. Until now, the spin-spin strength parameter used in the M1 calculation of odd-A nuclei was determined by comparing the theoretical and experimental magnetic moment values of odd-A deformed nuclei. However, using the interaction parameter obtained from the magnetic moment calculation, which characterizes the static properties of the nucleus, in the description of M1 excitations, which is one of its dynamic properties, is not a valid approach. Therefore, determining the appropriate κ value is important to achieve satisfactory agreement between experiment and theory. The effect of the change in the κ parameter on the M1 excitation properties in odd-A deformed nuclei has never been investigated.
The aim of the project designed to eliminate this deficiency is to determine the effects of the change in the spin-spin interaction strength parameter (κ) on the energy distributions of M1 transitions and the reduced transition probabilities of B(M1) in odd-A deformed nuclei and to select the appropriate κ value. The objectives to be achieved in the project in question are to determine for the first time the effects of the change in the κ parameter on the M1 properties in 151,153Eu, 155Gd, 157Gd, 159Tb, 161,163Dy, 165Ho, 167Er, 169Tm, 175Lu, 231,233Th, 233Pa, 235-239U, 239Pu deformed nuclei with low energy M1 excitations. Furthermore, this project aims to determine the optimum value of the κ parameter for the first time, which will provide an agreement between experiment and theory for the mass region investigated (actinide and rare earth).
Giant Dipole Resonance (GDR) is one of the most important collective excitation modes in atomic nuclei and is an important topic of interest in nuclear physics. Since the 1950s, GDR has been the focus of theoretical and experimental research and has led to extensive studies in the field of nuclear physics. GDR spectra have been measured for numerous nuclei in different mass regions of the periodic table (from 4He to 239Pu) using photonuclear cross-section experiments and have helped to determine the fundamental parameters of GDR (such as resonance energy distribution, resonance width, and photonuclear cross-sections).
This project aims to estimate the resonance energies and widths of even-even and odd-A nuclei in the actinides region for the first time using the ANFIS. In addition, the GDR parameters of odd-A 237,239U, 229-235Th, and 238Np nuclei, for which no experimental data are available, will be estimated using the ANFIS module in the MATLAB program, and these estimates constitute the original value of the project.
The GDR data set obtained from the RIPL (https://www-nds.iaea.org/RIPL-3/) reference input parameter libraries for nuclei with 219 different mass numbers will be used in the project. Mass number (A) and atomic number (Z) values will be used in the input layer of the ANFIS architecture, and resonance energies and widths will be used separately in an output layer. ANFIS calculations in the project will be performed with randomly selected data using the relevant toolbox in the MATLAB (MATLAB R2023a) software language and then divided into two groups training (80%) and testing (20%). The performance of the ANFIS model will be evaluated by root mean square error (RMSE) and correlation coefficient (R2).
People are exposed to radiation daily through soil, air, water, food consumed, etc. This radiation exposure can be natural or human-induced. Natural 40K,232Th, 238U, and human-induced 137Cs radionuclides are important sources of radiation in the soil. The main factors affecting the activities of these radionuclides in the soil are the geological structure of the land and fertilization activities. Determining the activation of radionuclides in agricultural lands' soil is essential to predict the effects of radiation transferred to people and agricultural products of farm activities. From this perspective, it is thought that the radionuclide rate in the soil where tea, the second most preferred beverage after water, is grown should be determined and monitored. One of the indispensable aplications of tea farming is fertilization. There are many different types of fertilizers, natural or synthetic, in tea farming. Natural fertilizers and artificial fertilizers are also very diverse. The effect of these fertilizer types on the radionuclide activation rates of the soil constitutes the research problem of this study.
For this purpose, the Fındıklı district, located in the east of Rize province, which is the city where tea cultivation is most commonly carried out in our country, was selected as the study area. Soil samples were collected from the study area, including the entire district and all types of fertilizers used in the region. Samples were also collected from tea-grown soils with no fertilizer to determine the effect of the region's geological structure on the soil's radionuclide activation rate. Thus, radionuclide activities were compared in soil samples where fertilizer was used and not used. Furthermore, soil samples were taken twice at 3-month intervals from the determined points where fertilized soil samples were taken.
The soils were brought to Sakarya University, Faculty of Science, Department of Physics, Nuclear Physics Research Laboratory. Before the radionuclide activation measurements, the soil samples were subjected to a preparation stage, which included cleaning, drying, beating, and sieving. After this stage, the soil samples were placed in airtight, special 250 ml containers and made ready for measurement. The soil samples were measured using an ORTEC brand 3"x3" crystal NaI(Tl) scintillation detector, and the analyses were carried out using MAESTRO software. The measurements were repeated 3 times, thus ensuring the reliability of the measurements.
This project differs from the studies conducted so far regarding the number of samples collected. In addition, no study has yet been conducted in the region comparing the radionuclide activations of fertilizer varieties.
