{"data":{"id":1340,"lang":"en","title":"USC Engineers Build Memory Chip That Works at 700°C, Hotter Than Molten Lava","summary":"USC engineers created a memristor chip from tungsten, hafnium oxide, and graphene that operates at 700°C and performs AI computations 92% more efficiently.","category":"technology","category_label":"Technology","published_at":"2026-04-07T18:02:17Z","updated_at":"2026-04-07T18:02:17Z","url":"https://techtools.cz/tools/news/article/1340-usc-engineers-build-memory-chip-that-works-at-700-c-hotter-than-molten-lava","api_url":"https://techtools.cz/news-api/v1/articles/1340?lang=en","markdown_url":"https://techtools.cz/news-api/v1/articles/1340.md?lang=en","image_url":"https://viterbischool.usc.edu/wp-content/uploads/2026/03/chips-on-Venus.png","word_count":137,"reading_time_minutes":1,"sources":[{"url":"https://viterbischool.usc.edu/news/2026/03/usc-scientists-build-a-memory-chip-that-survives-temperatures-hotter-than-lava/","domain":"viterbischool.usc.edu"},{"url":"https://www.sciencedaily.com/releases/2026/04/260406192904.htm","domain":"sciencedaily.com"}],"translation":{"lang":"cs","id":1398,"title":"Inženýři z USC vytvořili paměťový čip fungující při 700 °C, tedy více než roztavená láva","url":"https://techtools.cz/tools/news/clanek/1398-inzenyri-z-usc-vytvorili-pametovy-cip-fungujici-pri-700-c-tedy-vice-nez","api_url":"https://techtools.cz/news-api/v1/articles/1398?lang=cs"},"content":"Engineers at the University of Southern California have created a memristor chip that operates at 700°C (1,300°F)—temperatures exceeding molten lava—far surpassing any existing electronic memory device. The chip, built from tungsten, hafnium oxide ceramic, and graphene, held data for over 50 hours at peak temperature without refresh, survived more than one billion switching cycles, and ran on just 1.5 volts.\n\nLead researcher Joshua Yang noted that 700°C was simply the maximum their equipment could test, suggesting the device may tolerate even higher temperatures. Beyond extreme-environment applications like Venus surface missions, geothermal drilling, and nuclear reactor systems, the memristor also performs AI matrix multiplication 92% more efficiently than conventional chips. Yang has co-founded a startup, TetraMem, to commercialize the technology.\n\nThe research was published in the journal Science on March 26, 2026."}}