Abstract
The geology of chalk and conditions of special engineering significance which characterise chalk areas are described and discussed. Methods of investigating, sampling and testing chalk, both in the field and laboratory, are considered at length; the importance of in-situ observation and examination is emphasised. Following a short historical review of existing published information, the results of an extremely wide ranging collection of laboratory tests are presented and discussed. Simple index and classification tests are used to illustrate why chalk can be such a sensitive engineering material, but the deformation and strength properties of chalk in terms of total and effective stresses, receive most attention. It is apparent that for chalk, the value of laboratory testing is strictly limited, and great importance is attached to establishing the mass properties of chalk. This is best accomplished by field testing, the scale of which must be commensurate with the dimensions of the fabric of the rock mass. Plate bearing tests are shown to provide the most reliable means of determining the strength and deformation properties of chalk but, properly executed the simple dynamic standard penetration test can be used for correlation purposes. Analysis of spread foundation design and performance indicates that bearing pressures between 4 and 6 tons/ft for Grades IV and II chalk, increasing to a maximum of 20 tons/ft2 for massive Grade I chalk, can be applied and that settlement can be predicted reliably from plate bearing tests. For completely weathered or recompacted chalks it is shown that bearing pressures of 1 to 2 tons/ft can be applied and the settlement can be predicted from conventional laboratory oedometer tests. The carrying capacity of piles is shown to be derived almost entirely from shaft adhesion, the magnitude of which reduces with penetrations into chalk exceeding 10 to 15 pile diameters. Under-reaming appears ineffective. Pile length or depth of penetration into chalk does not appear to significantly influence settlement which is generally very small and can be predicted reliably from plate bearing tests. Driven steel piles appear to be the least efficient pile in chalk.