{
  "id": 7887243,
  "title": "Chinese scientists achieve 10 billion writing cycles in major future memory endurance breakthrough that improves endurance by 100x",
  "url": "https://urgent.news/2026/09/16/chinese-scientists-achieve-10-billion-writing-cycles-in-major-future",
  "topic": "science",
  "section": "Science",
  "published": "2026-09-16T22:10:00.000Z",
  "source": {
    "name": "TechRadar",
    "slug": "techradar",
    "url": "https://www.techradar.com/pro/chinese-scientists-achieve-10-billion-writing-cycles-in-major-future-memory-endurance-breakthrough-that-improves-endurance-by-100x"
  },
  "original_language": "en",
  "account": "Scientists from Xidian University, City University of Hong Kong, and Fudan University have achieved a groundbreaking development in future memory technology, demonstrating the ability to complete 10 billion writing cycles – a significant improvement of 100 times over previous limits. This breakthrough relies on a structural design utilizing wurtzite ferroelectric materials, which offer the potential for faster, more durable memory chips while maintaining low energy requirements.\n\nWurtzite ferroelectrics, such as aluminum scandium nitride (AlScN), have been the focus of recent research due to their compatibility with semiconductor manufacturing processes and their ability to achieve low energy consumption while providing superior switching speed. However, previous attempts to utilize AlScN chips faced reliability issues, as they could only withstand around 100 million writing cycles before failing.\n\nResearchers discovered that the failure was due to nitrogen vacancies within the material, which, instead of remaining stationary, would move and group together during switching. This movement created \"pathways\" that allowed electricity to leak, ultimately leading to device failure. To address this issue, the team designed a structure to minimize the presence of nitrogen vacancies, thereby slowing down the deterioration of the material.\n\nIf successful, AlScN-based RAM and storage solutions could revolutionize future server farms, providing the foundation for greater adoption of large language models and artificial intelligence.",
  "summary": "Next-generation memory technology could be far more durable than existing modules following a breakthrough that reduces inconsistencies in the ferroelectric material.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}