Proceedings of the International rilem conference Materials, Systems and Structures in Civil Engineering 2016

62

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 

 

also be taken into account (total heat release, concrete thermal conductivity,… [8]). Besides, 

the initial concrete temperature has a significant effect on the maximal temperature but to 

avoid this third parameter, we used a relationship between the external temperature T

ext

 and 

the initial concrete one T

ini

, proposed by Torrenti and Buffo-Laccarrière [8]. 

 

 

 

Figure 2: Temperature difference between the core and the surface in a massive wall 

(thickness 1.2m) with respect to the wind velocity and of the external temperature. 

 

The results of this study are resumed on figure 1 and 2. One can see on the figure 2 a slight 

effect of wind velocity on maximal temperature. This can be explained by the fact that, as the 

thermal concrete conductivity is rather low, the thermal exchange conditions at the core are 

close to adiabatic ones. On the contrary, the evolution of the maximal temperature due to the 

external temperature is quasi-linear. This can be explained by the fact that the initial 

temperature is assumed to be a linear function of the external temperature. On contrary to the 

previous results, the wind velocity has a high effect on "temperature gradient" (Figure 2), 

whereas the external temperature has a slight one. This indicates that with regards to these 

two phenomena, there are no optimal casting conditions (autogenous and drying shrinkage are 

not considered). 

 

2.1.2 Dissymmetry of creep: effect on stresses 

There is no consensus on the tensile creep of concrete at early-age and at long term. In most 

cases, only compressive creep tests are performed, since they are easier to perform. Besides, 

stresses (compression/tension) may change during service life of concrete structures. In order 

to show the impact of this dissymmetrical behavior, material parameters were identified on 

experimental results for creep of Briffaut et al. [9] (see the results in Figure 3 at left). Then, 

numerical simulations were performed on a 1.2 m thick concrete structure [3]. The results 

shown in Figure 3 (right) highlight a difference which is greater than the difference between 

the results obtained with and without the taking into account or not of creep (after 10 days). 

However, the difference occurs only after 4 days. 

 

---

63

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 

 

 

Figure 3: Comparison between simulated (thick lines) and experimental evolutions of strains 

(left). Stress and temperature evolutions versus time in a 1.2 m thick massive wall [3] (right). 

 

2.1.3 Simulation of RG8 specimen: influence of creep and creep/cracking coupling 

Numerical simulations are performed on a large beam specimen realized for ConCrack 

international Benchmark (for Control of Cracking in Reinforced Concrete Structures, [4]). 

After casting, the structure is thermally isolated and protected from drying during 48 hours. 

Then, the isolation and the formwork are removed and the structure is kept during two months 

outside. During all the test, longitudinal strains of the structure are globally restrained by two 

metallic struts. 

 

Numerical simulations on an active thermal ring test [9] and on RG8 beam [4] show that the 

coupling between creep and cracking should be taken into account in order to retrieve 

experimental simulations (i.e. the occurrence of 1 crack at least). Indeed, as displayed in 

Figure 4 and Table 1, no cracking is predicted with an approach where creep is not coupled 

with damage (3 cracks appear during the experiment at different times). Taking into account 

creep reduces drastically the crack opening. However, it should be emphasized that only 2 

(considering the symmetry) cracks occur, as 3 cracks (the first one in the center of the 

specimen) have been reporting during the experiments. 

 

 

Figure 4: Damage field on the RG8 beam for different creep approach: 1 = without creep, 2 = 

Creep without coupling with cracking, 3&4 = creep coupled with cracking for 2 different 

meshes (3 is not displayed but it is very similar to 4). 

 

---

64

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 

 

Table 1: Predicted crack opening for different models  

Model 

Without creep 

(1) 

Creep without 

coupling with 

cracking (2) 

Creep coupled 

with cracking, 

mesh 1 (3) 

Creep coupled 

with cracking, 

mesh 2 (4) 

Crack opening 

[μm] 

359 0  158 152 

 

 

2.2. Effect of drying at long term 

The effect of drying shrinkage is illustrated on a RC-tie specimen (w/c = 0.48, section of 

10x10 cm², 1 m long reinforced with a steel bar of 12 mm diameter, see [10]). Ninety days of 

drying are taken into account (only a quarter is meshed). Figure 5 shows the damage and axial 

stress fields after 90 days. Along the interface, it can be observed a slight concrete 

degradation (D around 0.3) due to the strain incompatibility between concrete and the couple 

rebar-interface. However, considering a uniform drying shrinkage strain, the rebar restraint 

induces significant tensile quasi-uniform axial stresses in the specimen. The axial stresses are 

equal to about 0.75 MPa (=1/4 of the tensile strength). The mechanical response of the tension 

test will be therefore influenced by the ‘‘initial’’ state of stress in the specimen. 

 

 

Figure 5: RC tie: experimental set-up; stress and damage fields in concrete, after 90 days in 

drying conditions (quarter of structure) – the steel rebar and the interface are not represented. 

 

The tension test is then numerically carried out. The mechanical response is plotted in Figure 

6. The response is greatly impacted by the drying shrinkage strain in the specimens. The first 

cracking force decreases by about 30%, in accordance with the initial internal stresses before 

loading. The first crack establishes the course of the further crack initiations. It is also 

observed a slight loss of initial stiffness if drying. That is explained by concrete degradation at 

the steel–concrete interface before loading (Fig. 5). 

 

---