Sand varies a lot - it depends on the rock it weathered from originally. As detailed here, sand dunes create rounded grains that do not interlock in the cement hydrate matrix = weak concrete.
Lots more about it here.
https://en.wikipedia.org/wiki/Sand
Yes, land sand is no good for building directly due to grain shape, but I think it is fine as input to the glass stream. Most manufactured sand is from stone quarries today, but I think a process of making manufactured sand from recycled glass should be able to make something more consistent.
Recycled soda glass might have too little silica to make durable cement, but high silica glass (90%+) is a good way. recycled bottle glass - being mainly soda glass. A google finds this which shows that up to 15% of waste glass is OK, above that and properties decline. I believe they do use as much as they can as it is a disposal problem. Glass factories recycle their own waste into new glass. Used glass is removed from some recycling facilities, but with so many plastic bottles these days, separation is a labor cost problem.
Some references here in the bibliography.
https://iopscience.iop.org/article/10.1088/1757-899X/585/1/0...
Many are paywalled, so use https://sci-hub.se/ to get access for some of them.
Thanks, that article and it citations look very promising. I'm a little discouraged to see a first data point where clear is correlating with strength though since mixed color recycling waste was my first thought on sourcing. Still it gives me some ideas of areas to learn about with some specific goals in mind.
Well, clarity = lack of scattering centers, which are also points of stress concentration = dislocations start here and propagate. Now glass with color from metal ions, like gold = ruby red, does not produce weakness. This mean colored glass from dissolved ions is OK, colored with particulates(pigmented) carry weakness. There is a whole gamut of the ways pigments induce weakness.