Rock Salt - Rock Mechanics Based on an Anisotropic Jointed Rock Model - page 115

5 Rock Salt

5.1 Introduction

Rock salt exhibits an exceptional rock mechanical behavior because its pronounced
time-dependent stress-strain behavior, even at low stress levels, deviates from the hydro-
static stress state

σ 3 also referred to as the “petrostatic stress state” or “litho-
static stress state”. Because of this particular constitutive behavior, which shows some
similarity with a viscous liquid, the specifi c fi eld of “salt mechanics” was founded,
which can be considered as a subdomain of rock mechanics.

σ 1 = σ 2 =

5.2 Stress-Strain Behavior

Rock salt almost exclusively consists of halite, which chemically corresponds to sodium
chloride (NaCl). The mechanisms that control time-dependent behavior of rock salt
are believed to have their origin in the movement of dislocations in the crystal lattice
of halite, which form potential sliding paths (dislocation glide) and subgrains that can
migrate further (dislocation climb or polygonization). Both processes are strongly de-
pendent on stress and temperature (Weertmann 1955, Carter & Heard 1970, Munson &
Dawson 1979, Langer 1984). It has been found that clean rock salt (halite) has a more
pronounced time-dependent stress-strain behavior - also referred to as “creep” - than
rock salt, which contains impurities of other minerals (Hampel et al. 1998).
Figure 5.1 schematically illustrates the shape of strain versus time curves obtained dur-
ing constant uniaxial loading of rock salt specimens. Under this unconfi ned condition
even very low stress levels lead to an elastic strain

el not depending on time and also to

ε

c increasing with time:

a creep strain ε

el + c (t).

ε (t) = ε ε (5.1)

If the applied stress σ f , the so-called “uniaxial yield stress”, the

increase in creep strain with time, i.e. the strain rate , is largest immediately after the
stress is applied and then converges to a constant value. The creep strain

σ is smaller than a stress

c in this case

ε

can be subdivided into two components:

c (t) = p (t) + s (t) if

ε ε ε σ < σ f . (5.2)

p is the so-called “primary creep” component. In the course of deformation a hard-
ening takes place which leads to a reduction of creep rate with time. Therefore the

ε