My daughter Samantha, who is fascinated by lava, asked if it was possible to make lava at home (or in the lab). Since lava is essentially molten rock, I figured it should be possible, given a suitable furnace. If you can melt metal, you should be able to melt rocks, assuming the temperatures are reasonable.
First, you can go by color. Lava is orange yellow, which implies a temperature around 1000C, just on the correlation between color and temperature. A little research on the web turned up a very useful site: Volcano World (http://volcano.und.nodak.edu/vw.html), particularly "ask a volcanologist", which has been discontinued, but the info is in the FAQ's. We turned up the following useful data. The conversions are approximate
| Rock type | Temperature (C) | Temperature (F) |
|---|---|---|
| Rhyolite | 700-900 | 1300-1700 |
| Dacite | 800-1100 | 1500-2000 |
| Andesite | 950-1200 | 1750-2200 |
| Basalt | 1000-1250 | 1850-2300 |
Note that the lighter the color of the rock, the lower the melting point. More generally, lava is typically between 900 and 1200C (1600F to 2200F).. say about the temperature of molten iron (1200C) , but not as hot as steel (1400C)... but a lot hotter than brass (600C) or aluminum (658C).
And, just as a guideline, glass is essentially molten silica, which is rock, so, if you can melt glass, you can melt rock.
In terms of absolute energy required to melt a kg of rock, I approximated it by using sulfur (38 kJ/kg) for the heat of fusion, and assuming the specific heat is that of glass (.84 kJ/kg K). Heat to 1000C requires 840 kJ, and melting requires another 40, so you need to put about 900 kJ into that kilo of rock to melt it.
Since I had other projects for which I wanted a furnace, I decided to make a sort of generic crucible furnace. I've always wanted to try casting metal, making high temperature superconductors (you need a furnace that goes to 1000C) , and a variety of other high temperature projects. The choice is between gas fired and electric fired. Certainly, electrically operated furnaces are cleaner and quieter, but, there is something satisfying about a big gas burner roaring away. And, gas fired furnaces are cheaper to operate (not that I expect to spend anything more than a nominal amount on either approach). Finally, I figured it would be easier to get a lot of heat quickly with a gas burner (I hate waiting, and it's easy to get 50-100 kW out of a gas flame (propane is 46 MJ/kg, and a standard 5 gal/20lb propane tank (like for a barbecue) will burn for quite a while at that rate (about 5-10 lb/hr))
Mind you, a lot of the heat goes into heating up the furnace, the air, and so forth, otherwise it would only take 10 seconds to pump 900 kJ out.
If you go electric, and figure that 2.5 kW is about the maximum electrical power you want to put in (=20 Amps on a 240 V service), it will take a bit longer to put the heat in (375 seconds or just over 6 minutes).
I built a pair of propane burners based on the forge burner designs on Ron Reil's page: http://www.reil1.net/design1.shtml specifically, the EZ burner, which has a minimum of machining, and is made from a bunch of standard black iron plumbing fittings. I made the burner using 3/4" pipe.
My initial test was running the burner at 20 psi through a #57 orifice.
For my first lash-up (which may actually be the last), I bought a couple dozen firebricks (at $0.98/each) from Carlson Building Materials. They get their firebricks from Pacific Clay ( http://www.pacificclay.com/ ) in Lake Elsinore, California. I called Pacific Clay to find out the specs, and they sent me a fax saying that they aren't real refractory bricks, but that they should work to 2300F.
Companies like BNZ Materials (http://www.bnzmaterials.com/) make the real thing, with bricks at various temperature ratings from 2300 up to 3000F. This assumes the traditional furnace approach with refractory bricks with relatively low thermal conductivity, but still quite massive. You wind up heating the bricks mostly. A potentially better approach is to use high temperature ceramic fiber insulators like Kaowool from Thermal Ceramics (http://www.thermalceramics.com/ ) and various insulating/reflective/refractory coatings from International Technical Ceramics (http://users.southeast.net/~itc/) which gives you much better thermal insulation, requiring less total heat input to get the oven up to temperature. Of course, if you aren't energy cost constrained, then the traditional approach probably works as well.
lava.htm - 23 Feb 2000 - Jim
Lux
Jim's home page