Civil Engineering Reference
In-Depth Information
2.2.2 Theoretical nano/micro-mechanics
Atomistic simulations
The exponential growth of computer power over the past decades has
springboarded the development of computational methods and code inter-
faces that provide the potential of material simulations for a variety of
scientifi c problems. The community of concrete science and engineering was
reluctant to employ these techniques primarily because the important char-
acteristics of C-S-H were poorly understood.
A realistic molecular model of hydrating cement has recently been pre-
sented in Pellenq
et al.
(2009). Starting from a monoclinic periodic compu-
tational cell of dry Tobermorite and subtracting SiO
2
groups, they created
a defected silicate chain that achieve C/S ratios close to the experimentally
obtained values from NMR (Cong and Kirkpatrick, 1996). The resulting
molecular model was then enriched, using grand canonical Monte Carlo
simulations, with water molecules to reach a crystal structure of
(CaO)
1.65
(SiO
2
)(H
2
O)
1.75
and a density of 2.56 g/cm
3
which are in close
agreement with the experimental values reported through SANS/SAXS by
Allen
et al.
(2007) of (CaO)
1.7
(SiO
2
)(H
2
O)
1.8
and 2.6 g/cm
3
, respectively. The
model was then contrasted to experimental data on hydrated cement,
namely fi ne structures X-ray absorption spectroscopy signals, X-ray diffrac-
tion intensities, nanoindentation results, and vibrational density from infra-
red spectroscopy with very good agreement.
The studies that have already been presented in the literature include the
calculation of elastic and chemical bonding properties of the most common
cement analoges, Tobermorite and Jennite (Churakov, 2008, 2009; Shahsa-
vari
et al.
, 2009), the atomistic modeling of the major hydration products of
cement (Manzano
et al.
, 2006, 2007, 2009), the development of an empirical
force fi eld (CSH-FF) for calcio-silicate hydrates (Shahsavari
et al.
, 2011),
the impact of chemical impurities on the hydration/structure/mechanical
performance of alite (Ca
3
SiO
5
or in short C
3
S) and belite (Ca
2
SiO
4
or in
short C
2
S) clinker phases (Manzano
et al.
, 2011), and the chloride binding
on the various hydration products (Kalinichev and Kirkpatrick, 2002), to
name a few. The development of a robust computational model at the
nanoscale will allow scientists and engineers to fi ne-tune and optimize the
nanostructure of material through cost-effective virtual simulations. Proper
benchmarking against detailed nanoscale experiments will be an essential
part of the development process.
Continuum micromechanical modeling
There have been signifi cant developments in the fi eld of micromechanics
over the last 50 years (i.e., Torquato, 2001; Dormieux
et al.
, 2006). As the
Search WWH ::
Custom Search