115
International RILEM Conference on Materials, Systems and Structures in Civil Engineering
Conference segment on Service Life of Cement-Based Materials and Structures
22-24 August 2016, Technical University of Denmark, Lyngby, Denmark
THE IMPORTANCE OF MULTIPHYSICS AND MULTISCALE
MODELLING OF CONCRETE TO UNDERSTAND ITS COMPLEX
MACROSCOPIC PROPERTIES
Jörg F. Unger
(1)
, Vitaliy Kindrachuk
(1)
, Volker Hirthammer
(1)
, Thomas Titscher
(1)
,
Christoph Pohl
(1)
(1) Federal Institute for Materials and Structures, Berlin, Germany
Abstract
Concrete is a complex material. Its properties evolve over time, especially at early age, and
are dependent on environmental conditions, i.e. temperature and moisture conditions, as well
as the composition of the material. This leads to a variety of macroscopic phenomena such as
hydration/solidification/hardening, creep and shrinkage, thermal strains, damage and inelastic
deformations. Most of these phenomena are characterized by specific set of model
assumptions and often an additive decomposition of strains into elastic, plastic, shrinkage and
creep components is performed. Each of these phenomena are investigated separately and a
number of respective independent models have been designed. The interactions are then
accounted for by adding appropriate correction factors or additional models for the particular
interaction. This paper discusses the importance of reconsider even in the experimental phase
the model assumptions required to generalize the experimental data into models used in
design codes. It is especially underlined that the complex macroscopic behaviour of concrete
is strongly influenced by its multiscale and multiphyscis nature and two examples (shrinkage
and fatigue) of interacting phenomena are discussed.
1. Introduction
The understanding and the prediction of the macroscopic behaviour of concrete is very
important to ensure a safe design of structural components. In the design phase, it is often
very difficult and certainly not appropriate to use complex models with a multiscale and
multiphysics approach. The focus is in many cases a safe design for the ultimate limite state.
However, a profound understanding of the material behaviour with a realistic modelling of
hardening and the interacting phenomena is required to understand the behaviour under
service conditions or the evaluation of a structure after extraordinary loading situations. This
ranges from the detection of microcracking, crack spacing and crack opening, creep and
shrinkage deformations up to the influence of high temperatures or fatigue loading. In this
116
International RILEM Conference on Materials, Systems and Structures in Civil Engineering
Conference segment on Service Life of Cement-Based Materials and Structures
22-24 August 2016, Technical University of Denmark, Lyngby, Denmark
case, the experimental investigations of interacting phenomena are usually performed with a
specific modelling assumption, e.g. a linear decomposition of the strain into elastic, plastic,
shrinkage and creep components. Interactions are then taken into account by additional
interaction terms.
An example is concrete under combined thermal and mechanical loads, where often an
additive decomposition into mechanical and thermal strains is done. The mechanical model is
calibrated for constant room temperature, whereas the thermal model is adjusted for zero
mechanical load. The (nonlinear) interaction is then characterized by so called load induced
thermal strains [1].
Even though this approach seems to be straightforward in general, it has several drawbacks
these interaction terms have to be calibrated separately and only model the real
physics in a homogenized, macroscopic way;
it is often difficult to design experiments, where only a single parameter is changed
and calibrated, and the distribution of this parameter is
often not fully homogeneous
within the specimen;
the predictive capacities of models with interaction factors are often limited to the set
of training/validation data, since the interaction model is purely phenomenological and
does not capture the real physics;
the consideration of additional inputs requires a recalibration of the model, i.e. the
heating of concrete does not only change the temperature, but also induces drying
which itself leads to shrinkage phenomena. So depending on whether shrinkage is
considered as additional (additive) strain, the interaction term varies;
the number of parameters for those models is often large and particularly the
interaction terms often do not have a clear physical meaning;
the thermodynamic consistency of the model is sometimes difficult to ensure, e.g.
when modelling a series of loading scenarios with a parallel temporal evolution of
macroscopic properties such as strength, Young’s modulus or fracture energy, which
is especially important at early age.
As a consequence, the choice of the right model assumptions is very important, especially
when dealing with multi-physics problems. Two models might give very similar results, e.g. a
damage or softening plasticity model for monotonic loading. However, both models will give
significantly different results for general loading sequences including unloading. A similar
example is the modelling of shrinkage in concrete based either on a strain or stress based
approach. Pure shrinkage deformations can be modelled appropriately with both approaches,
but it will be demonstrated that combined loading (mechanical, thermal, moisture content)
might lead to significantly different results.
A second reason for the macroscopic complex behaviour of concrete is the importance of its
heterogeneous structure. This comprises very different scales. Some (non exhaustive)
examples are
atomistic models
the influence of moisture on macroscopic creep behaviour is often explained by
the disjoining pressure due to very thin layers of water;
microscale models
micro-cracks lead to a macroscopic softening with a quasi-brittle post-peak