Evaluation of the shrinkage and creep of medium strength self compacting concrete

The difference between self compacting concrete (SCC) and conventional concrete (CC) is in fresh state, is the high fluidity at first and the need for vibration at second, but in hardened state, both concretes must comply with the resistance specified, in addition to securing the safety and functionality for which it was designed. This article describes the tests and results for shrinkage and creep at some medium strength Self Compacting Concrete with added sand (SCC-MSs) and two types of cement. The research was conducted at the Laboratorio de Tecnología de Estructuras (LTE) of the Universitat Politécnica de Catalunya (UPC), in dosages of 200 liters; with the idea of evaluating the effectiveness of implementation of these new concretes at elements designed with conventional concrete (CCs).


Experimental details
Two (2) batches were done for each type of SCC-MS (with CEM II 32.5 R CEM I 42.5 R and 120 liters each), in the forced action mixer LTE capacity of 250 liters.
Each SCC-MS were performed two (2) of characterization tests in fresh, as are trials Slump and Extension Bars with ring (De La Cruz, 2006), (Bravo, 2004), to verify self-compactability of concrete; because they are the two (2) that define processes quickly and reliably the behavior in fresh SCC. Three (3) specimens of each type of control SCC-MS were also developed for evaluating the compressive strength after curing in a humid chamber to specification UNE 83-301 (Gettu et al, 2000), at a temperature of 23 ° C and humidity on 90%, tested at 28 days according to the UNE 83-304 (Gettu et al, 2000). Dosages and results of characterization tests are shown in Table 1. In the same batches 14 cylindrical samples, 150 x 300 mm were prepared; corresponding three (3) of each type of concrete for testing shrinkage and four (4) of each type of concrete for fluence. These concretes were compared with a conventional concrete (CC) of the same strength. To evaluate deformations, the molds were provided with strain gauges (120 Q) of length 10 cm, arranged in the center of the vertical axis as shown in Figure 1. This is to capture the axial strain in the central third of the specimen. Processed specimens for the tests were demolded at 24 hours of being emptied, they are polished and immersed in water in a tank located in the climate chamber with relative humidity (RH) of 50% at a temperature of 19 ° C, for 28 days.

Shrinkage tests
Drying shrinkage is determined for three (3) specimens of each type of SCC-MS. Once the samples were immersed in water they were connected to data acquisition equipment. This, with the idea of tracking the processes of deformation of the SCC-MS during curing. In Figure 2, the process sequence once the specimens manufactured for trials presented. After 28 days of curing in water, each specimen was adhered external measurement points (DEMEC) located each on its side surface 120° as shown in Figure  3. . Deformation monitoring with electronic readings during the curing time, for both concrete showed an expansion (deformation negative values), which was expected given the curing conditions to which the specimens were undergoing. So, given the performance of the test conditions (curing in water), endogenous contribution is negligible and, therefore, the measured strains are mainly to drying shrinkage. As shown in the graphs below, the securities electronic gauges in the early hours of drying, have a rapid growth of the strain, which tends to increase more gradually to 24 hours after the two SCC-MSs. Hence, the behavior is very similar during retraction measured for both SCCs-MS. In Table 2, the results of deformation are summarized shrinkage both methods (points DEMEC and average of electronic values embedded gauge), which shows that the specific values (old) are approximately equal from 19 days. The trend and behavior of each individual specimen and their average age (three (3) specimens for concrete) compared to the average electronic readings from the two (2) days of curing be removed. The differences in average readings deformation of the two (2) methods. For each type of SCC-MS at 28 and 42 days it is approximately 18% with CEM II 32.5 R and 25% with CEM I 42.5 R (De La Cruz, 2006).
It seems that at the age of 19 days of trial, the concrete specimens manage both internal and external balance. So the data obtained with both methods are becoming very similar, if not the same after this age. The differences are not comparatively high if one considers that the strain readings with DEMEC points, were performed manually with higher possibility of error (attached implement points-and strain measurements-sample), and electronic values are taken directly from the data acquisition equipment (embedded gauges). So, given the performance of the test conditions (curing in water), endogenous contribution is negligible and, therefore, the measured strains are mainly to drying shrinkage.
In Table 3   with the embedded gauges, the results have not large percentage differences, suggesting verify the effectiveness of the measurements with both methods, being either I them suitable for determining shrinkage SCC-MSs.
One of the conclusions of De La Cruz (2006), regarding the shrinkage behavior of the SCC-MSs with different types of cement; for the same strength, they are similar. And also it notes that trends in the evolution of the shrinkage after 28 days of curing (considering zero (0) to read just after curing) for the two SCC-MSs are very similar during the 42 days of drying. In Figure 6, the evolution of the shrinkage occurring up to 530 days, the specimens SCC-MS with CEM II 32.5 R Reaching a maximum strain of 331 microstrain.

Creep test
For the creep tests (ASTM C 512-87), they were prepared four (4) samples of each type of SCC-MS. The specimens were embedded strain gauges as shown in Figure 7, once demolded spot faced and were immersed in water for 28 days. As for the shrinkage test, once the SCC-MS (120 liters for each SCC-MS) obtained the procedure shown in Figures 1 and 2. After the curing is followed, a voltage was applied uniaxial compression equivalent to 40% of its compressive strength at the age of 28 days; that is for SCC-MS with CEM I 42.5 R of 17 MPa and with the SCC-MS CEM II 32.5 R of 12 MPa; oleopneumatic using racks of charge as shown in Figure 7.
In Table 4, results are summarized creep test deformation with both methods (DEMEC points and average values of the embedded electronic gauges).  In this paper partial data and the evolution of creep deformation (under constant pressure) of the SCC-MSs occur during the 629 days of being subjected to load in the racks (I report given by the LTE (records sent to Colombia by the LTE (UPC) / 2008)). This evolution occurs both with electronic readings of gauges, as with DEMEC points.
To determine the creep deformation, it must be subtracted from the total deformation (£ total ) (Eq. φ (t-t0): Creep coefficient whose expression involves φo basic creep coefficient.
Then the evolution of creep for SCC-MSs occurs after 530 days of experiment. In Figures 8 and 9, the evolution of creep deformation occurs in SCC-MSs with cements CEM I 42.5 R and CEM II 32.5 R (Barcelona-Spain). (2) With recent records deformation (microstrain) of the specimens prepared for determining shrinkage, creep coefficient ($) and creep of the SCC-MSs, it is shown in Figure 10. Next, the evolution of creep coefficients ($) for both concretes.
ϕvalue, for SCC-MS with CEM II 32.5 R 629 days to prepare specimens, is 1.67, as shown in Figure  11. Similarly, the $ results are presented for SCC-MSs concretes with CEM I 42.5 R, up to 629 days: As shown in the previous figure, ϕ values are lower for concrete with CEM I 42.5 R 50% at 28 days and 61% at 42 days, compared to the other concrete CEM II 32.5 R.
According to the results of De La Cruz (2006), which are presented in Table 5, in the same specimens and 169 days, ϕ it had a value of 1.15 for SCC-MS with CEM I 42.5 R . Establishing compared to CC and SCC (Roncero et al, 2001) ϕ has a value of 0.80 and 0.81 respectively; which according to the experimental data to 629 days, it is 1.24 (see Figure 11).  In Table 6 below, the values of ϕ for concrete with the same mechanical properties similar environmental conditions while testing are presented, as proposed calculation Article 39 (EHE (2003) (Gettu et al, 2000)), Roncero et al. (2001) and Mari and Cladera, 2003 ("RH = 60%"); and it is observed that the value of ϕ with CEM I 42.5 R is less than for concrete with CEM II 32.5 R, by 75% and 62% respectively.  Table 5.). Another conclusion of De La Cruz (2006), regarding the creep coefficient ($) of the SCC-MSs with different types of cement, and according to estimates suggesting CCs Article 39 is applicable to SCC-MSs.

Conclusions
• As the percentage comparisons with the SCC-MS CEM II 32.5 R, it has a deformation similar to a CC yield the same mechanical properties. This situation differs in a high percentage to the other concrete CEM I 42.5 R. This does not preclude in any way that the latter has a strain similar to that of a CC creep, as it has been compared to concrete characteristic strength of 35 MPa, and its compressive strength at 28 days is 43 MPa. • If the methodologies and approaches used only with SCC-MS with CEM II 32.5 compared, one can say that the results have not large percentage differences, suggesting verify the effectiveness of the measurements with both methods, whichever one it appropriate to determine the deformation creep of the SCC-MSs. • It seems that the calculations suggesting the Article 39 to determine the evolution of the creep coefficient is applicable to SCC-MSs. • The strain values obtained to 629 days, the specimens for shrinkage and creep tests show the same trend until 169 days (De La Cruz, 2006), but are increasing. Higher for SCC-MS with CEM II 32.5 R, which for SCC-MS with CEM I 42.5 R. • As in the case of determining the retraction of this type of concrete, it is not excluded in any way, do a more thorough study of creep deformation and evolution of $ for SCC-MSs, continuing with measurements and calculations to help identify reliably, the deferred behavior of this type of SCCs.