Part 2 – Nature and origins
P. J. Withers and H. K. D. H. Bhadeshia
Residual stress is that which remains in a body that is stationary and at equilibrium with its surroundings. It can be
detrimental when it reduces the tolerance of the material to an externally applied force, as is the case with welded
joints. On the other hand, it can be exploited to design materials orcomponents which are resistant to damage,
toughened glass being a good example. This paper, the second part of a two part overview, the ﬁrst part having been
devoted to measurement techniques, examines the nature and origins of residual stresses across a range of scales.
This extends from the long range residual stress ﬁelds in engineering components and welded structures, through the
interphasestresses present in composites and coatings, to the microscale interactions of phase transformations with
Published by Maney Publishing (c) IOM Communications Ltd
Professor Withers is in the Manchester Materials Science Centre, University of Manchester and UMIST, Grosvenor Street,
Manchester M1 7HS, UK (firstname.lastname@example.org). Professor Bhadeshia is in the Departmentof Materials Science
and Metallurgy, University of Cambridge, Pembroke Street, Cambridge CB2 3QZ, UK (email@example.com). Manuscript
received 3 March 2000; accepted 6 December 2000.
# 2001 IoM Communications Ltd.
Macro residual stresses in engineering
As the design of engineering components becomes less
conservative, there is increasing interest in how residual
stressaffects mechanical properties. This is because
structural failure can be caused by the combined effect of
residual and applied stresses. In practice, it is not likely that
any manufactured component would be entirely free from
residual stresses introduced during processing. Furthermore, in natural or artiﬁcial multiphase materials, residual
stresses can arise from differences in thermalexpansivity,
yield stress, or stiffness.
As discussed in part 1,1 residual stresses arise from misﬁts
in the natural shape between different regions (as in shot
peening), different parts (such as the stresses around a rivet
in a plate), or different phases (as is the case for composites).
The terminology for describing these is given in detail in
part 1. In essence, the stresses can be discussed in termsof
their characteristic length,2 l0, which is the length over
which the stresses equilibrate. Long range stresses (type I)
equilibrate over macroscopic dimensions (l0,I#the scale of
the structure). Such stresses can be estimated using
continuum models which ignore the polycrystalline or
multiphase nature of the material, often calculated using
ﬁnite elements. Type II residual stresses equilibrateover a
number of grain dimensions (l0,II#3 – 106grain size). An
example of these is the interphase thermal stresses in a metal
matrix composite. Type III stresses, on the other hand, exist
over atomic dimensions and balance within a grain
(l0,IIIvgrain size), for example, those caused by dislocations
and point defects.
The materials technologist has a plethora of techniques of
differing capabilitywith which to characterise residual
stress.1 The challenge is to use information gained with
these techniques in the optimisation and management of the
residual stress state with the goal of improved processing
and component design. In this paper, part 2 of the overview,
the nature and origins of residual stress across a range of
scales are examined, from the short range stresses typical ofcomposite materials and phase transformations to longer
range stresses in thin ﬁlms, welds, and engineering
366 Materials Science and Technology
April 2001 Vol. 17
There are at least four ways in which macro residual stresses
can arise in engineering components: through the interaction between misﬁtting parts within an assembly, and
through the generation of chemical, thermal, and...