The Geohazards Research Group at the Nottingham Trent University outlines the background to Loughborough Loess - an idealised universal benchmark for collapsible soils
Some soils exhibit the potential to collapse as a result of bond weakening under conditions of wetting and/or loading and as a result are a major hazard in geotechnical engineering. Although these soils are widely variable in their nature and location, their essential behaviour is consistent throughout the world. A considerable body of research has been - and continues to be - undertaken to investigate the engineering behaviour of collapsible soils. However, much of the knowledge gained from this work tends to be lost due to the use of formats or literature which engineers find unfamiliar or difficult to access.
A wide variety of research materials has been produced, reflecting the range of disciplines actively involved in collapsible soils research, which has in turn led to disparate literature. As economic growth demands further development, engineers are increasingly being asked to work with collapsible soils. These range from natural deposits (eg loess, quick- clays and tropical soils) to man-made soils such as PFA and non-engineered fill. Such deposits are one of the main global problem soils which engineers today must face.
The cost of poor engineering design and construction associated with these soils runs into billions of pounds and loss of life, for example the Teton Dam failure, Idaho, US in 1976 and the St Jean Vianney landslide, Canada in1971. Other geotechnical problems commonly arising on collapsible soils include foundation collapse, slope instability and collapse of underground space, although the use of loess has been widely advocated for siting engineering structures including as a host for radioactive waste.
The reality is that the wealth of research that exists is still largely unavailable to the engineering community. This is leads to many engineers being unable to either recognise a collapsible soil or be able to deal effectively with engineering works using them when they are encountered. Furthermore, much of the existing literature suffers from being rather technically specific, both in terms of geographical applicability and the terminology used to write up the work. This is further hampered by the range of criteria used to define what is actually meant by a collapsible soil and whether the collapsible character is significant enough to be likely to cause problems.
The International Association of Engineering Geologists has recently established a commission, Collapsible Soils (No 18), whose membership not only represents the main geographical regions of the world but also includes individuals from each of the key disciplines involved with collapsible soils research and related work. This is building on the work of the NATO advanced workshop on the 'Genesis and properties of collapsible soils', which was held at Loughborough in April 1994.
As a result, 'Loughborough Loess' has been proposed as an alliterative material in a similar sense to Cam Clay and Granta Gravel: a structural and lithological ideal, rather than a rheological one. This ideal silty material has been produced in the laboratory, allowing full control over its material constituents. Thus for the first time the effects of factors like silt size, shape and nature, together with the mineralogy of the bonding clay minerals can be fully examined.
Loughborough Loess samples are produced in such a way as to simulate air-fall deposition, a key characteristic of natural loess soils. Thus the metastable structures, which are the source of loess collapsibility, can be reproduced and examined in detail. Work at both Loughborough and Nottingham Trent Universities has illustrated the importance of clay mineral content and mineralogy on the collapsible nature of loess soils, with the maximum collapse occurring, for example, at a kaolinite content of 20% to 25%.
Through extensive testing of Loughborough Loess, actual loess behaviour has been reproduced, offering a new window to understanding and coping with this problematic soil.
These and wider issues will be discussed at conferences over the next four years, starting with the 34th conference of the Engineering Group of the Geological Society in Nottingham this September. A benchmark volume, Silts: Problems and engineering solutions, to be published in spring 2000,will provide a base for pooling results pertinent to geotechnical engineers and combine know- ledge relevant to the needs of the end user.
Mike Rosenbaum, Ian Jefferson and Ian Smalley are members of the Geohazards Research Group at Nottingham Trent University.