There is a rescue mission to save Scihub, you can help by seeding the torrents or contributing to related open source efforts https://www.reddit.com/r/DataHoarder/comments/nc27fv/rescue_...

there are some discussions on using IPFS for decentralized storage and indexing of existing 85M Scihub papers http://freeread.org/ipfs/