ELSEVIER EngineeringGeology 40 (1995) 137-138 EN£~INEERIING GEOLOGY Opinion section The adobe reaction and the use of loess mud in construction C.D.F. Rogers, I.J. Smalley Civil and Building Engineering Department, Loughborough University of Technology, Loughborough LE11 3TU, UK Received 31 July 1995; accepted 14 September 1995 Abstract Adobe brick and wall constructions depend for their structural usefulness on the adobe reaction. This is similar to the pozzolanic reaction produced by the addition of siliceous material to Portland cement concrete. The adobe of southwestern USA is similar to loess. Fringe loess in arid regions (North Africa, southern Spain) gave rise to the term adobe and the constructional applications. The adobe reaction may be similar to the reactions involved in lime stabilisation of engineering soils. Adobe soil makes adobe mud bricks and adobe mud walls. It is a constructional material which has been used for many thousands of years. The word came to the English language from the Spanish, but it came to Spanish from Arabic (At-tub =mud brick). This suggests that the mate- rial was used in North Africa and in southern Spain before its utilisation in its best known region, the southwest of the USA and in Central America. Holmes (1978, p. 486) equated adobe and loess by stating: "In the Mid-west of the United States there are deposits of loess, locally called adobe, that correspond in all essentials to those of Europe and Asia." He was obviously misinformed about loess; loess in the Mid-west is called loess, and is recognised as such. It was not perhaps, recognised fully in Idaho and this lack of recognition helped to bring down the Teton Dam (Rogers et al., 1994). The questions to be asked concern the adobe in the southwestern states; is that also loess and does its recognition as such have any conse- quences for soil engineering and construction? 0013-7952/96/$9.50© 1996ElsevierScience B.V. All rights reserved SSDI 0013-7952(95)00064-X The mechanism of mud brick formation has not been fully investigated. Presumably there is some low order cementing action that develops in the wet mass, which gives the dried artefact some strength, and a modest resistance to further wet- ting. It is possible to envisage a form of pozzolanic action, very like the original form of pozzolanic action in Roman cement. Today's pozzolans oper- ate by reaction with the by-product lime which is released when Portland cement hydrates, and tend to be largely composed of finely divided silica. Loess is, of course, the classic natural occurrence of finely divided silica (mode size 30 I~m -- as quartz). The quartz co-exists with clays and car- bonates. Roman cement was produced by adding siliceous material (often volcanic ash) to lime mortar, which produced some simple calcium sili- cates that hydrated to give added strength to the set lime mortar. So in loess (or adobe), the wetting process can initiate a series of very low order chemical reactions: