home home journal journal
listserv listserv photos photos knowledge knowledge
home faq
earth • ship - a dwelling designed to bring people into a more intelligent and sustainable relationship with their environment
Earthship Journal prologue previous journal entry next journal entry journal index
02.19.05   The Plaster Experiments

At some point after the workshed was finished, I decided to change things up a little and experiment with so-called natural cement. I was interested in natural cement for reasons which included appearance, breathability, and environmental issues. I was eager to try just about everything I read about including a few mixes of my own, and found the results to be very interesting, even if my methodology wasn't pristine.
General Notes:
All plaster mixes which utilized sand were made with regular plaster sand. No special sifting of materials were used to get a consistently sized aggregate. The samples are designated by the ratios of:
An F designates wether fibers were included. Either fiberglass or chopped straw fibers were used. There are many examples where no fibers were used at all in order to observe the behavior of the raw material. The photos in the first set at the right were taken about 2 weeks after the initial samples were spread. They were protected from the elements while they dried. The second set shows the examples several months later, having been exposed to the elements in a southerly direction. The pigment had been added closer to the time the photos were taken however. Not all samples in the second set are represented. Where they are missing, the photos have not changed, and links to detail photos have not been provided.
Lime, Soil, and Sand:
By far the most appealing results came from mixing various portions of Lime, Soil, and Sand. The straight Lime/Sand mixes (1:4) were very impressive, but tended to be a little grainy. When a little soil was added, the plasticity and workability of the mix seemed to improve, and provided a visibly less grainy finish. Over time the added earth did not appear to be a liability for the strength and water-resistance of the mix. Retained strength was most evident in samples where earth was added very sparingly, such as less than 1/4 that of sand.
When pigment was later added to the surface of a few of the cured samples, the samples with earth added took the pigment more evenly and resulted in richer color. In the samples where only lime and sand was used, the pigment tended to roll off the surface of the grains of sand, giving it a paler tint. Of course, ideally you'd want to add pigment to the whole mix, not just to the surface, so that exposure to the elements doesn't produce white spots where damage may occur. It would be interesting to see if similar benefits could be derived by using an extra fine aggregate.
Another benefit of the Lime/Sand mixes is the color luminosity of pigment derived from the light to white base of the lime. The compressive and surface strength of these mixes has really stood the test of time. After the photos in the second set were taken, the palette was laid on the gound, pointing the samples straight up. Almost a year later, only mild erosion is evident. The lime/sand samples are even still holding some of the surface pigment, and it has been unusually wet in Taos this winter. I am impressed enough with these results that I plan to use a lime plaster in place of cement wherever appropriate.
Some of the examples included caliche, since it is so abundant in my area. The caliche gave the mixes a very tough, mealy texture, almost like auto body putty. It contracted a great deal when dry, and had a brittle appearance. The caliche mixes that included fibers did much better, but even they showed slight signs of cracking. The best performing caliche mix included some earth to help it relax. I think it would be worth some further experiments with caliche based on these tests.
One other note, the variations in aggregate size in the caliche mixes made spreading dicey, so I'd recommend sifting the caliche before using it. In fact, if I did these experiments over again I'd likely pre-sift all the earth-ingredients before mixing. Caliche tends to give the plaster a delicate reddish tone.
Fly Ash:
The samples that included fly ash produced the most cement-like material when dried. It possessed a hard tough surface that was also somewhat brittle in nature. It sounded like ultralite concrete when a piece was dragged against itself. Only the fly ash samples which included fibers resisted cracking.
One big detriment to the fly ash samples was their color. Not only was the natural color unappealing, it took pigment very poorly as you will see, producing a very chalky, dull and somewhat lumpy finish.
The Kaolin clay was by far the biggest dissapointment. I was hoping it would help induce some sort of cementitious reaction, but it did not. It was soft and frail even after several weeks of drying.
I set out to produce a cementitious hydraulic reaction from these mixes, and on that level I believe the experiments failed completely. I'm not entirely sure wether I would even recognize a genuine hydraulic reaction in the case of these materials, since the lime will carbonate by air alone. To really give hydraulic setting a go I'd need to really set up a situation where drying occurred more slowly. On the other hand, I am completely happy with the results of the lime/sand mixes and will likely use a lime plaster in place of cement wherever appropriate.
previous journal entry next journal entry
related images (click to view photos in full resolution)
Plaster Samples Set 1
Plaster Samples Set 1 (click for details)
Plaster Samples Set 2
Plaster Samples Set 2 (click for details)
more photos
 other areas of