Walter,
The national Machine tool manufactuer's association was taken over by The Association For Manufacturing Technology. Their URL is www.amtonline.org
I've put in a request to their general information e-mail for a source of the publication. I also looked it up on www.worldcat.com and found that purdue university is likely the nearest library to you with a copy. . . Georgia tech is the closest one to me. . .
That looks like a good paper.
I've been looking at your samples and musing about the air bubbles and some other factors. I need to redo the calculation for how much stress I think your sample is under at the breaking point so that I can compute the Griffith (critical) flaw size. The Griffith Flaw size is the size of flaw at a given stress level which will cause a crack to propagate through the material leading to failure. From some back of the enveloping, I think the critical flaw size could be as small as 2mm right now.
It also looks like your samples failed with the aggregate breaking. This is a sign that the failure is not necessarily the matrix's fault. A possibly cost impractical test would be to redo the experiment with alumina, zirconia, or agate as they have much much higher fracture toughness values than quartz.
In that the sample that utilized the wide range of aggregates performed better than the one with the large aggregates, I think it supports the generally accepted theory result that the fracture toughness of small aggregates is better than large aggregates. I believe the fracture toughness of the aggregate is right up there with the overall density of the material as a critical factor.
As a note, several items I've read lately indicate that water vapor supports the creation of critical flaws in glass fiber which is chemically very similar to quartz.
Your use of heat and waiting for the material to settle are consistent with all of the epoxy manufacturer best practices sheets I read lately. Kudos for the good work on your technique!! It would also be good to leave the samples overnight in the oven at about 200F for absolute maximum epoxy crosslinking.
I'm still of the belief that the air bubbles have to go to get truly strong material: especially with what I've been looking into about Griffith flaws.
I haven't started experimenting yet and I goofed today and fell asleep when I should have been bidding for this vacuum pump on ebay Larry warned me about. . . Worse yet, I have to go back to work tomorrow Sanity should be limited this week as my Bridgeport and lathe will be here and I need to put up the rollup door in the shop.
While the epoxy we have may not be perfect, it does look at least decent. I still think silyated silica fume or nanosand could bring us the order of magnitude improvement based on the papers I've read. DAK3333 Are you out there?
--Cameron