A New Dawn, page-54

yes once you get past 10% the ion in flow is too dramatic causing cell rupture- so progressively worse cycle life until it totally collapses.Thats why graphite is used because it’s very stable. see table.
The only way if you want to add more silicon is to buffer this swelling caused by the silicon expanding by over 300%.
There are lab trials with lining on carbon tubes etc but it’s very tricky, costly so inevitably unscable for mass production into battery cells.
NVX in their patents have already described and demonstrated that they can get high silicon load due to the abilities of the DPMG process to multiple layer within other materials to buffer the silicon expansion.This is scalable, not costly and reliable.
Though it’s unknown what exact silicon load they got. I’d expect some results when the Samsung/Sanyo agreements get to the nxt stage.