OK, I did this originally many months ago, then got caught up in things, and now it’s a lot later than I expected. That’s good in a way, as long term slow reactions have had time to operate, but bad in that I’ve not taken any measurements / observations in the middle.
As this was intended as a recreation of ‘my guess as what was likely to happen by accident’, it isn’t that big a deal in any case. It mostly just shows what is likely a bad way to do things, that might have stimulated the interest of some Egyptian Kid playing in the dross next to a shop and got somebody thinking about better ways.
Right now I’m packing things up (and have been for a couple of days) as I’m “between jobs” and that means I can’t carry the costs of a ‘bi-coastal’ lifestyle, so the Florida Rental has to go (end of the month I move in with a friend in their Florida house on a lake… so not exactly having a reduction in life style, just costs ;-)
At any rate, “the junk must go”, and part of it is this plate of samples that have been asking me to write a posting about them. Each is a paper cup with about 2 to 4 ounces (volume – heaping spoons, actually…) of stuff in them. Call it 125 cc or so. Some as low as 50 cc. They are all a light gray or white, so showing pictures of bland dirty white hockey pucks is unlikely to be informative. Instead, you will get a short description of what each was made from, and what it has become.
I will be saying one of the ingredients is Natron. It is commercial Sodium Carbonate (Arm and Hammer brand washing soda) but as natron was just natural Sodium Carbonate, I think that’s appropriate as it is close to the original material. I also found a source of “lime” that was a bit of a challenge. The garden shop stuff isn’t CaO, but a mix of some CaO and some Calcium Hydroxide (‘slaked lime’) as that is what was sold at the local stores. Most likely, fully “burnt lime” would be better, but as a first approximation of ‘lime on a crap pile’ having the mixed stuff is likely not too bad. But instead I used some more pure lime for this round. I found what looks to be real ‘burnt lime’ in the pickle and canning section of the grocery store as “Pickling Lime”, and that’s what I used for these small experiments. Clay was from a standard bag of clay based Cat Litter. Using specific kinds of clay would likely be better. Again, though, as a ‘quick test of stuff from a trash heap in old Egypt’, this is a reasonably common crummy clay. I also used some Diatomacious Earth (D.E. from here on out) as a silicon dioxide sample. I’d have rather used some quartz sand, but figured this was OK for a first look.
More detail on chemistry and methods than you might want from the original posting about this batch here:
https://chiefio.wordpress.com/2014/02/12/some-notes-on-geopolymer-cement-clay-bricks-unfired-and-diy/ Hard to believe it was a year ago, but it is.
All units are given in ratios. So, for a 1:1:2 lime:natron:clay you would use one “unit of volume of your choice” of the first two, and two such scoops of the last one. Eventually it would be better to use mass ratios, but I don’t have a scale right now. Typically I added just enough water to make a stiff paste, but some mixes had excess water and were a bit more soupy / slurry like (usually causes weaker cement). Once the basic evaluation is done, then a strength test (perhaps to destruction) by standing on each (about 2 inch) diameter puck will be done. Followed by a resistance to water test – i.e. does it crumble when soaked.
In no particular order, the mixes and my sensory evaluation:
2-3 Clay (by eye is not exact…)
Hard white lump. Surface rubs off as a white powder (might be excess alkali that turned to carbonate on the surface). Not a particularly dense stone, but usable (most likely for ‘light duty’). Does not fracture on light tapping on a hard surface. Broke in half when I stood on it, and then falls apart when placed in water. Broken face shows many clay inclusions of gray color. Clearly not enough alkali soaks into the clay grains to convert them outside the surface (so likely more clay dust and mixing would be better). When wet, the mass has a gray color like clay.
Enough to get the idea into someone’s head that “this stuff sets up”. Dumping some lime and natron on a natural clay surface or clay dust pile, that gets wet for just a while, would be enough to get someone started looking at ‘making stones’, even if fairly weak ones. From that, it’s time, experiment, and observation.
Made a thin puck of about 2 cm height. On handling while still young, it broke into 1/2 and 2 x 1/4 chunks. Now, it is white and weak. Simple finger squeeze caused one of the 1/4 chunks to fly apart into crumbs. Not going to try the ‘stand on it’ test. While this clay was fully divided into more dust like particles, this mix has no strong alkali in it. Clearly neither clay nor D.E. binds much (at all?) in the weak base of Natron. You need more alkaline catalyst like lime or lye.
But even a negative result tells you things… Look to lye and lime (CaO) as the catalytic agents, or perhaps a bit of KOH from wood ashes.
90% Clay slurry
10% Natron / D.E. mix
This gives a puck that looks like it is mostly gray clay with a white layer on top (where the Natron collected as the water evaporated?). A fairly strong feeling puck (rather like an unfired brick) but friable to strong finger pressure. IMHO, just a clay brick with some D.E. scattered in it and some Natron floated to the top / mixed in. It simply turns to mush in water.
But this confirms that Natron alone is not enough, and that the ratio with clay is not having much effect between these two mixes. Clay makes nice bricks, but to harden it takes fire, or something stronger than natron.
This was a small thin puck, made with excess water. It is a purer white, but very friable with low strength and crumbled getting it out of the cup and handling it. Lime is not enough to set up the D.E. into a solid. It does fall apart in water, but starts out so prone to crumble that it’s hard to say if the water makes it go much faster. Lime alone is not cutting it as a catalyst.
3 Shells (ground up sea shells for feeding birds)
Makes an irregular puck. Surface prone to spalling and flaking, but some parts fairly hard. This was to test a low ‘catalyst’ level with lots of clay and carbonates. It might give the idea of a stone, but it is just not anywhere near a usable thing, really. I suspect that a very strong lye is needed to make silicates “go”, while the carbonates would work better as a classical Roman Cement kind of mixture. At any rate, a bit of natron and some lime on a pile of dirt (clay, limestone or shells midden, some D.E. or related sea muds) isn’t going to cut it. Though it might get someone started looking at the hard lumps and ‘trying stuff’.
It doesn’t turn to mush in water as fast as the others (I think the surface is a bit less permeable) but does soften and can be ‘mushed’ easily (showing a drier under layer). I could see folks making mud bricks finding that a bit of lime and sand might make them stronger and a bit more water resistant. A possible path for discovery to start.
Finally, the last one:
Just checking to see if a large excess of ‘catalyst’ might make the silicon dioxide of D.E. do anything. This makes a very white, very solid puck, with a flake layer on the top (that tastes rather alkaline, and that when wet – flakes in hand not on the puck – didn’t break up easily but did feel ‘soapy’ so alkaline). About 1 inch thick. Once the top flakes are off, the puck itself is solid and not friable to the fingers.
Standing on it, the ‘rim’ where it was raised from wetting the paper cup, crumbled, but the puck itself held up. Some bits of the side surface show cracking. Water wetting gives a soapy feeling wash water, but the puck does not fall apart. It does weaken a bit, so I can crack off chunks with finger pressure ( I have strong hands, so YMMV). I’m doing a longer ‘soak test’ for longer term strength when wet.
My opinion of this is that there is significant excess natron / lime in the mix and that is what is washing out in the water, leaving voids and a weaker structure. The bulk of the D.E. has had some degree of ‘binding’ happen, but not enough for high strength. It would be quite reasonable to use silicon containing sands and a lower ratio of lime (perhaps with lye added or instead) and see if an approximation of sandstone could be made.
Clearly in a ‘strong enough’ alkaline environment, silicon dioxide can be made to have some degree of binding that would start one down the path to cements and liquid stones.
A strong alkali or lye is needed, natron alone is not enough.
Silicon Dioxide and Clay both react to some degree in Lime solutions. Exploring lye, and lime / lye mixes (both quantities and ratios), ought to be fruitful. In particular, using ‘wood ash’ as a lye containing material that would be dumped in quantities near places making clay bricks and / or on silicon bearing sands would be an interesting area to explore.
Clay crumbles tend to remain clay crumbles if only lightly mixed in modest water level slurries. While this is ‘good enough’ to cause someone to notice an effect, it makes for a weak product, and would be a defect in any final method. D.E. is an easy to find silicon dioxide source for experiments, but each shell is a hollow thin thing. LOTS of surface area for surface chemistry to be demonstrated, but not very good for ‘strong bulk’. Using a good grade of sand would be the next step to take.
Excess natron and / or lime solution does not rapidly react away, and may weaken the final product. Exploration of the lower bound and upper bound effects would likely point to where the optimum can be found.
I expect that using Water Glass (Sodium silicate or metasilicate) with sand would work well, and be an easy bypass of the difficulty I’ve had getting lye. From the wiki:
They are used in cements, passive fire protection, textile and lumber processing, refractories, and automobiles. Sodium carbonate and silicon dioxide react when molten to form sodium silicate and carbon dioxide:
Water Glass was defined in Von Wagner’s Manual of Chemical Technology (1892 translation) as any of the soluble alkaline silicates, first observed by Jean Baptist van Helmont circa 1640 as a fluid substance made by melting sand with excess alkali. Glauber made what he termed “liquor silicum” in 1646 from potash and silica. Johann Nepomuk von Fuchs, in 1818, obtained what is now known as water glass by treating silicic acid with an alkali, the result being soluble in water, “but not affected by atmospheric changes”. Von Wagner distinguished soda, potash, double (soda and potash), and fixing (i.e., stabilizing) as types of water glass. The fixing type was “a mixture of silica well saturated with potash water glass and a sodium silicate” used to stabilize inorganic water color pigments on cement work for outdoor signs and murals.
So this is a reasonably well traveled path…
Prepare Sodium Silicate
Wear proper safety gear, which includes gloves.
Heat 4-8 grams of sodium hydroxide in 10 ml of water.
Once the sodium hydroxide is dissolved, slowly add the 6 grams of crushed silica gel beads. Heat the solution between additions. If the crushed beads won’t dissolve, add a little more water to the solution.
That’s it! You now have sodium silicate or water glass. NurdRage has a YouTube video of this procedure, if you’re interested in seeing how it’s done.
Silica gel being a synthetic form of silicon dioxide that is porous. But this shows that lye reacts with silicon dioxide to make sodium silicate. We can skip a long slow process with lye and sand in a bucket at warm Egyptian temps (but not heated on a burner…) but just mixing ‘water glass’ with sand and evaluating the product for strength and characteristics. If that works well enough, finding a path to that product via other forms of natural silicates and ‘Egyptian Desert Warm’ lye could be followed over a longer time.
It’s pretty clear that using water glass will work:
Concrete and general masonry treatment
Concrete treated with a sodium silicate solution helps to significantly reduce porosity in most masonry products such as concrete, stucco, plasters. A chemical reaction occurs with the excess Ca(OH)2 (portlandite) present in the concrete that permanently binds the silicates with the surface making them far more durable and water repellent. It is generally advised to apply this treatment only after the initial cure has taken place (7 days or so depending on conditions). These coatings are known as silicate mineral paint.
Water glass is a useful binder of solids, such as vermiculite and perlite. When blended with the aforementioned lightweight aggregates, water glass can be used to make hard, high-temperature insulation boards used for refractories, passive fire protection and high temperature insulations, such as moulded pipe insulation applications. When mixed with finely divided mineral powders, such as vermiculite dust (which is common scrap from the exfoliation process), one can produce high temperature adhesives. The intumescence disappears in the presence of finely divided mineral dust, whereby the waterglass becomes a mere matrix. Waterglass is inexpensive and abundantly available, which makes its use popular in many refractory applications.
which strongly implies that some kind of ‘in situ’ creation of sodium silicate from sand and lye in a mix of minerals ought to be usable for a kind of ‘liquid stone’ pathway.
And that concludes this round.
In large part just showing that the most likely path (using lye) is not the only path, but that the lime / natron pathways are at best weak, and show a direction, but not a method.
Going forward, trying water glass as a quick path to products, then working backward to ‘in situ’ sodium silicate formation / reaction is likely a good path of investigation.
Finding nice strong lye solutions available without a load of trouble is next on my list of ‘things to do’ on this topic (and thanks to all the folks who have given suggestions of where to find it…) but only just after I order a $20 jug of Water Glass from the hardware store (Lowe’s says they can order it).
Probably in another year ;-)