Nuclear forensics training — it’s a dirty job

Although the likelihood of a terrorist nuclear attack is extremely low, a lot of work is required to prepare for such an unthinkable event. That’s why a response team assembled by the National Nuclear Security Administration (NNSA) recently trained in eastern Idaho’s desert on ways to collect and analyze simulated debris from a nuclear detonation. Nuclear forensics—the science of determining the origin of nuclear material—is an essential element of the United States’ strategy to prevent nuclear terrorism.

DOE Announces $70 Million to Improve Supercomputer Model of Earth’s Climate System

The U.S. Department of Energy (DOE) today announced $70 million in funding for seven projects that will improve climate prediction and aid in the fight against climate change. The research will be used to accelerate development of DOE’s Energy Exascale Earth System Model (E3SM), enabling scientific discovery through collaborations between climate scientists, computer scientists, and applied mathematicians. Data from this model will enhance scientists’ understanding of climate change, which will be crucial to furthering President Biden’s commitment to tackling the climate crisis at home and abroad.

Scientists Recruit New Atomic Heavyweights in Targeted Fight Against Cancer

Researchers from Berkeley Lab and Los Alamos National Laboratory have developed new methods for the large-scale production, purification, and use of the radioisotope cerium-134, which could serve as a PET imaging radiotracer for a highly targeted cancer treatment known as alpha-particle therapy.

Lack of Damage After Secondary Impacts Surprises Researchers

When a material is subjected to a shock or blast wave, damage often forms internally through spall fracture, and research is needed to know how these damaged materials respond to subsequent shock waves. Recent experimentation on spall fracture in metals found that, in certain cases, there was an almost complete lack of damage with only a thin band of altered microstructure observed. In the Journal of Applied Physics, researchers narrowed down exactly why the expected damage was missing.

Supercomputers Pave the Way for New Machine Learning Approach

Researchers have developed a machine learning approach called transfer learning that lets them model novel materials by learning from data collected about millions of other compounds. The new approach can be applied to new molecules in milliseconds, enabling research into a far greater number of compounds over much longer timescales.