Fe-N-C interfacial engineering on commercial Pt/C catalysts, via unique 5d-3d/2p orbital hybridization, enables a breakthrough in fuel cell durability and performance. The engineered PtFe/C@Fe-N-C catalyst maintains 97.3% of its activity after 30,000 accelerated stress cycles. This exceptional stability and its high mass activity decisively surpass both the performance of pristine Pt/C and the U.S. Department of Energy's 2025 technical targets.

A new study shows that hydrochar, a carbon-rich material made from wet biomass, can improve soil structure and help soils store more carbon more effectively than several common organic amendments.

A new study published in Biochar reports a sustainable membrane technology that converts microalgae-derived biochar into an advanced material for municipal wastewater treatment, offering a promising route to cleaner water and waste valorization.

Researchers led by Illinois Grainger Engineering professor Qing Cao have demonstrated a scalable way to directly and sequentially stack high-performance silicon circuits. This advance marks a critical step toward realizing the full potential of three-dimensional chips that could carry computing beyond the limits of traditional scaling.

In an environment of radiation, extreme temperatures and razor-sharp dust, researchers are designing how humans will build, and ultimately survive, on the moon.

Researchers have developed a fully light-controlled artificial synapse that could improve how computers process visual information. The device uses a rare-earth-doped crystal to mimic how biological synapses strengthen and weaken connections based on past activity, but operates entirely with optical signals rather than electricity. By modeling and validating how trapped charge carriers govern this behavior, the team demonstrated both signal enhancement and suppression, key features of neural function. Integrated with an imaging sensor, the system can reduce noise and enhance contrast directly at the point of capture. When applied to a neural network, the approach achieved nearly 96% accuracy on handwritten digit recognition, highlighting its potential for low-power, efficient neuromorphic vision systems.