It’s Not Rocket Science #11: Keeping the Water Out

Back in February 2013 I wrote about the ancient Persian yakchal buildings for making ice in winter and storing it until summer.

Untitled

These buildings used a combination of night sky radiant cooing in conjunction with the thermal mass and insulating properties of mud brick. I wrote

Insulation: The walls of the dome were at least two metres thick at the base, and made of mud brick coated with a special waterproofing mortar composed of sand, clay, egg whites, lime, goat hair, and ash. This render had excellent insulating properties. I can’t find any information for how the optimum ingredients or mix for the mortar were discovered. I can imagine the goat hair may have functioned like the glass fibres do in fibreglass, but what properties do the egg whites add to the render? And how did anyone know they had those properties? 

Erica [Ritter] Wisner tracked me down and kindly explained what’s going on. It’s more fascinating than I’d imagined.

I suggest an answer in two parts:

Protein gels and starch gels are among the general categories of materials known to be compatible with clay-based plaster mixtures. Bodily fluids of all kinds have been used as paint fixatives and binders since the stone age; traces of blood, saliva, urine, egg, milk, etc. have been found in ancient petroglyphs and cave art. Vegetable materials like cactus juices/gels, wheat paste, and oils are also used.These binders are still used for natural paints, some more than others, and also can be added to finish plasters to make them more durable or more compatible with a given paint.  

For clay-based plaster work, these materials act as improved binders, adding hardness and resistance to erosion (and sometimes also for glossier finishes). Proteins and starch gels can be used in larger proportions than oils, without disrupting the clay-based bonds which makes an earthen plaster work. Clay swells with moisture to seal out further erosive moisture incursion; using too much oil (or mixing with too much Portland cement) reduces this self-sealing property, while the natural gel-type materials can work in tandem with the clay in a similar way. This wet-condition self-sealing prevents water erosion, yet allows breathability and release of moisture in dry conditions.

It’s not hard to see why the plaster for the ice domes should be extra erosion resistant.  Most desert climates are subject to the occasional storm deluge. The cold of the ice could also cause increased moisture condensation, and possibly (in combination with night-time cooling) lead to frost damage. The process you suggested of covering with straw would also call for a durable finish plaster.
Eggs are definitely in this category, one of the higher-performing additives for hardening natural paints and plasters. They are expensive but highly effective. 

But why egg whites particularly?

Egg whites are one of those things that get left over when someone is using egg yolks elsewhere. Egg yolk is an emulsifier, contains richer-tasting oils, and in general has more food value than egg whites. Egg yolks are used as an emulsifier and binder in natural paints, most famously the Italian frescos. Egg white is a protein gel. Its ability to hold a froth suggests relatively tough long-chain molecules such as are also found in glutinous flour and plastic polymers. 

Its very dearth of nutritive value might be a plus for structural uses, since it would not be as “tasty” to vermin as the whole egg. You would be in a better position than I am to verify whether the Persians used large quantities of egg yolk in paints, desserts, or other artisan or culinary uses, but I would not be surprised. There are just not that many natural emulsifiers that are as easily identified and produced. 

This I did. There’s a kind of Persian quiche that’s fairly popular but this next dish is also good candidate for producing a surfeit of egg whites.

chelo_kabab_kubideh

Chelow kabab or Chelo kabab (Persian: چلوکباب‎) is the national dish of Iran. The meal is simple, consisting of steamed, saffroned basmati or Persian rice (چلو chelow) and kabab, of which there are several distinct Persian varieties. It is an old north-western tradition that a raw egg yolk be placed on top of the rice.

Erica also provided a conjectural history for the origins of yakhchal.

You are an ancient Persian entrepreneur, working on a building design that basically makes its own weather. You run into plaster/mortar performance problems, and after a few experiments, you call on a local master plaster-and-paint artist for help with the formulation of a high-performing plaster. He might be recommended by your patrons because his family did the excellent and durable plaster work on a favorite folly, or an artist of good reputation who also does decorative fresco, faux-painting, mural, and sculptural plaster. As a master tradesman, one would expect him already to know a number of excellent plaster formulae for both indoor and outdoor work. He would know the value, use, and price of performance-enhancing additives. He might even be in the position to personally procure significant quantities of egg whites at the right price, after using the egg yolks in mural or fresco projects for other high-end clients.

I haven’t found any examples of ancient Persian frescoes but the frescoes in the Baptistry of the Dura-Europa Church in Syria are probably the oldest Christian paintings in existence, dating as they do from sometime between 233 and 256AD. Here’s Christ walking on water.

OLYMPUS DIGITAL CAMERA

The nearby Dura-Europaos Synagogue also dates from 250AD but its paintings were painted onto dry plaster and so technically aren’t frescoes.

Goat hair (or any animal hair) is a very common ingredient in both lime and clay-based plasters. The cheaper clay-based plasters often use straw or dung for the cellulose-type fiber, but animal hair lasts longer, is finer to work with, and makes it easier to get a solid, erosion-resistant, crack-free surface with plenty of fiber and binder. Any of the materials you mentioned would be stock-in-trade for a master craftsman in an era where natural plasters were state of the art.

Erica also mentioned anecdotal suggestions that the civilizations of the Middle East used blood to strengthen mortars (the Western sailors’ rumors being that it was slaves’ blood). There remains a high-end Spanish technique for earthen floor finishing using bulls’ blood, freshly slaughtered onto the floor.

I don’t have a Spanish example, but traces of blood were found in this clay floor from mid-18th century Montpelier. Here’s the floor, reconstructed.

clay_floor_dm_jpg_42010-60

In countries with a history of the ritual slaughter of animals, it’s easy to imagine how blood would be used to waterproof clay floors. As a general rule, if your dishwasher or washing machine struggles to remove it, then it’s probably a good candidate for a natural binder for a paint or plaster. 

So much for clay-based plasters and renders. Tadelakt (a transliteration of the Arabic “تدلاكت”, meaning “massaged”or “rubbed”) is a traditional Moroccan waterproof surface created by polishing a lime-based plaster with a stone and then rubbing it olive oil soap into it. The olive oil soap provides oleic acid which, I learn, is “a fatty acid that occurs naturally in various animal and vegetable fats and oils.” Making tadelakt is very labour intensive and time consuming.

Tadelakt-ouarzazate-morocco-1

This example is probably synthetic tadelakt as it looks rather high-end with those recessed halogens. We’ve know the look of tadelakt even if we’ve never seen the real thing.

Les Grands Bains du Monêtier, le hammam

Another reason for tadelakt substitutes is that the real thing is made from tadelakt lime washed downstream from the Marrakech Plateau. Over at realfinishes, Patrick Webb writes

The limestone is argillaceous, meaning it contains a relatively high percentage of clay. Also, there is a small infiltration of amorphous silica making Tadelakt lime slightly hydraulic. Combining its natural properties with traditional application methods, Tadelakt’s waterproofing qualities were subsequently put to decorative use in exterior façades, small drinking vessels and famously the “hammams” or public bath houses.

moroccan.plasterer writes of the medluk of Fez, quoting David Amster’s site A House in Fez,

“…the outer walls of houses were finished with medluk, made of extremely fine sand, lime (jeer), egg white, and sabon beldi (traditional soft soap made from olive by-products). Medluk develops a beautiful marbled effect over time. Simple geometric patterns are sometimes pressed or carved into the medluk. In Marrakech this mixture is called tadlakt, which is slightly finer and shinier due to the difference in the sand and lime from the two cities.

These fine regional distinctions and different names for the same substance or technique are typical of vernacular. Erica suggested there might be some similarly fortuitous geology upstream to account for the Yemeni people’s many ways with render.

11yemen-span-articleLarge-v3

I wonder about that waterproofing material derived from river sediment. Is it because of the size of particles, the sifting action of the river? Something the water deposits along with the minerals that makes it work better, like a hard water or algae coating on the silt particles?  I wouldn’t be surprised. I returned to Salma Samar Damluji’s The Architecture of Yemen – my only reference.

51IA7URwmvL._SY300_

The transliterations of the various names in Arabic below probably barely approximate the regional Yemeni Arabic but it’s not going to matter. All I want to show is the detail, the many classifications and the regionality of the thing named, and a glimpse of the general knowledge, built up over centuries, of the properties of local materials. The builders may not know why sand taken from a certain river at a certain time has those properties but they’ve learned how to make good use of them.

  • khulb: the general term for mud coating but the quality depends upon the type of turab (earth or clay) and the place of its extraction.
  • khulbah: a mixture of earth and water used for plastering the exterior of a flat roof. Inside, khulbah is mixed from soft or ‘light’ mud with water and used for plastering or finishing the walls.
  • tibil: the chopped straw, hay or chaff mixed with clay to make mud bricks
  • haddah: soft stalks of the tamarind tree mixed with clay to make mud bricks
  • mahdah: mud plastering carried out in the kitchen and living room during the month of Sha’ban.
    • In the area of Ghayl Ba Wazir, the mud is extracted from the saylah flood course silt that is fine and viscid.
    • In Yafi, a white-coloured earth from mountains called quri is known to give the best results. It is strengthened with wheat, chaff or animal dung.
  • qiddah: used for damp proofing. It is made by quarrying stone, cutting it into small pieces, firing it over wood and then burying it until it turns to a fine powder which is then mixed with water and small pebbles.
  • qatat: a grey-blue coloured clay extracted from the bed of Ghayl Habban in the area of Ghurayr and mixed with fine bullrushes to make a damp-proof course.
  • nays: sea sand
  • ruwaynah: fine red sea sand with soft grain
  • kafi nafsahu:  sea sand with medium-sized grains that doesn’t require mixing with any other kind of sand
  • nurah: a refined lime whitewash and plaster used as a damp-proof course. In some places, nurah is used internally and polished with a stone as with Moroccan tadelakt. After burning, nurah is pounded and left to soak and “ferment” before being beaten to a creamy paste.

nurah is generally preferred by master builders as it takes other plasters and renders better than cement, is malleable for longer, ages well without losing its shape and hardens over time. The downside is it takes much longer to dry. Small sections of wall have to be left for maybe as much as two months. This means a house rendered in nurah takes three years to build whereas one using cement render takes only one year. The general trend is for cement-based mortars and renders to replace traditional renders.

It is a similar story with paints. The Buqshun Palance is the major building in Khalyah. It was overpainted with oil-based “emulsion” rather than rendered with time-consuming nurah and its inherently softer pastels

This is the fate of the vernacular. The same process that led people to use one type of sand taken from a particular river at a particular time to produce a building more suited to immediate but largely unchanging circumstances, is the same process that leads people to eschew it in favour of another product with obvious advantages for circumstances newly immediate. We do it all the time and call it progress but, if cement render, oil-based paint and, for that matter, parametric rainscreens were to disappear from the face of the earth tomorrow, there’d still be some people who could make good use of what’s at hand to keep the water out.

[cite]

6 thoughts on “It’s Not Rocket Science #11: Keeping the Water Out

  1. abdu_a.m.

    As a ‘Shibami’ I admit I was pleasantly surprised to see my family’s hometown up there in the banner!
    Very informative post. The amount of wisdom unpretentiously embedded within the details of vernacular traditions is beyond amazing.
    Your last paragraph got me thinking quite a bit, though. I’ve always wondered whether there, genuinely, were any valid criteria by which we can decide if it was a good idea to use a certain technology once it became feasible and practical. Obviously, not screwing up the earth on the long run might be a valid one. Maybe resilience and acknowledging limitations of scale are others. I guess what I mean to say is maybe there could be another way by which we can deal with the thing that we call progress other than a one-dimensional view of efficiency. Hassan Fathy and other “misfits” spoke a lot about resilience, empowerment and that process matters as important drives behind the way we should go about building and architecture.
    It’s too scattered an argument I guess but I wonder if there could be something in that.

    1. Graham McKay

      It’s nice to know someone from Shibham. That’s a good question and I think you were close to answering yourself when you said “unpretentiously embedded”. Imagine some place with a vernacular tradition that they’ve never felt the need to update. However, one day, some builder decides to use a cement based render instead of a ‘mud’n’blood’ one because it’s quicker and with maybe 90% performance. I’d see his decision to use that as a judgment based on a comprehensive evaluation of all the factors involved – and therefore a correct call. HOWEVER, if he made that decision because cement renders are “modern” then it becomes pretentious and wrong. Look at it the other way. If we use cement-based renders because we have a history of using them then there’s nothing wrong with that. Cement was good enough for the Romans, etc. HOWEVER, if we choose to use a clay-based render because it is somehow “authentic” and “natural” (despite being more labour-intensive and possibly more expensive) then it is us who are now being pretentious in our choice.

      Remember Hassan Fathy’s Gourna project? Mud brick is a good material but the problem for the inhabitants was that it wasn’t pretentious (“modern”) enough. My next post will deal with various pretensions of the last century of architecture. This is to set the scene for the post after that, the plan of which is to set out “some valid criteria by which we can decide” if architecture is going in a good direction or not. It’s half written now, but it will involve concepts of “pretentious” and “unpretentious”.

  2. Erica Wisner

    So flattering to be included! Lovely to see the specificity of the terms. This is detail that is very hard to get when discussing trades across international borders.

    Just after reading this, I stumbled into a fascinating conversation about fermented plasters with a natural builder from Montana USA. Apparently some people are mixing natural plaster materials (clay, lime, sand) with chopped hay (instead of straw or hair fiber). They add fermented apple starter culture, and allowing the mixture to ripen before using. He says it is incredibly hard and weather-resistant. I wonder if the biological ‘matrix’ in the plaster might even continue to grow for a while after application, turning the wall coating into a sort of living felt that might toughen and integrate the material after application. There might be amazing benefits to the longer waiting period of the ‘nurah’ plasters.

    1. Graham McKay

      You’re welcome Erica! I did come across the word “ferment” quite often as a translation of some Arabic term for the production of lime-based plasters. I thought it might be in the sense of “prove” as in pastry, but it probably really does mean ferment. From what I remember from the book, the master plasterers appreciated that concrete render hardened more quickly whereas the hardness of nurah continued to increase over time – the implication being that concrete render had inferior long-term performance.

  3. Jonathan Strauss

    excellent
    we are seeing a return to using hydraulic limes and soured limes in mortar mixes as they perform better

Comments are closed.