Safeguards monitoring ensures that nuclear technologies can be employed for peaceful uses without fear of nuclear materials being diverted to nefarious purposes. Nuclear forensics techniques are used to discriminate the origins of nuclear material found outside of regulatory control. Both of these fields depend upon extremely accurate estimations of concentrations of actinides and fission products to provide the best assessments of the materials in question.
“As the global security environment continues to rapidly change,” explained Seray Cerezo, “Nuclear forensics and nuclear safeguards measurements need to be accurate, and the sensitivities known to ensure timely detection.”
Cerezo is a nuclear engineering graduate student who worked with Dr. Sunil Chirayath, the former director of the Center for Nuclear Security Science and Policy Initiatives (NSSPI), to improve the accuracy of estimates that are useful in nuclear safeguards monitoring and nuclear forensics. Her master’s thesis work involved applying Systematic Uncertainty Quantification to MCNP fuel burnup simulations. MCNP, or the Monte Carlo N-Particle transport code, is often used to simulate nuclear fuel burnup and depletion. It estimates the concentrations of actinides and fission products, and these estimates can be compared to actual material measurements.
“During fuel burnup simulations,” said Cerezo, “The uncertainties in the predicted nuclide concentration due to the uncertainty in the nuclear data of the MCNP methodology are not propagated.”
In MCNP, the nuclide concentration is calculated by the embedded CINDER 90 isotope generation and depletion module. CINDER90 uses the neutron reaction rates and flux values provided by MCNP. The reaction rates can be broken down into flux, number density, and microscopic cross-section terms. The microscopic cross sections contain a systematic uncertainty, but this systematic uncertainty is not propagated in the MCNP fuel burnup simulation through each time step. To propagate the effects of systematic uncertainty in microscopic cross sections, Cerezo utilized a Backward Euler numerical scheme that allows for the reporting of the systematic relative error in the predicted nuclide concentrations. This methodology was executed through Python scripting, and a program was developed to estimate the overall systematic relative error in concentrations of the user-desired nuclides predicted in an MCNP fuel burnup simulation.
As a NSSPI student, Cerezo completed a summer internship at Los Alamos National Laboratory and gained experience in nondestructive assay through a training course offered at the lab. She was a member of the Texas A&M Institute for Nuclear Materials Management (INMM) Student Chapter, which allowed her to attend presentations by various experts. She also had the opportunity to present her research at the 2023 INMM/ESARDA Joint Annual Meeting in Vienna, Austria, and to participate in the International Nuclear Facilities Experience to the European Union, which she cited as a highlight of her NSSPI student experience.
“Not only was it enjoyable to experience other cultures,” she said, “but it was exciting to learn more about international safeguards and nonproliferation.”
In May, Cerezo earned a Master of Science in nuclear engineering with a nuclear nonproliferation specialization. She previously graduated with a Bachelor of Science in nuclear engineering from Texas A&M in May 2022. She plans to continue as a Texas A&M graduate student and pursue a Ph.D.