1. Introduction
For several decades, concrete has been widely used in the field of civil and architectural engineering all over the world, because of its excellent mechanical strength, durability, and economic efficiency. However, because of some drawbacks of concrete associated with low tensile strength, low ductility, and low strength-to-weight ratio, its application to thin-plate structures, e.g., thin walls and long-span bridge decks, has been limited thus far. In order to overcome these drawbacks, reactive powder concrete (RPC), which is the forerunner of the currently used ultra-high-performance concrete (UHPC), was developed by Richard and Cheyrezy [
1] in the 1990s. Since UHPC is made from a granular mixture optimized by the packing density theory with a low water-to-binder ratio (W/B) and since it includes a high volume content of steel fibers, it exhibits both superior strength and ductility with a unique strain-hardening response.
However, because of the low W/B and the inclusion of high-fineness admixtures in UHPC, the evaporation rate of water from the surface is normally too large to be replenished by bleeding. Because of this, the surface exposed to the atmosphere dries and condenses very quickly, even while most of the interior mortar is still fresh. This causes plastic shrinkage cracks at the surface, as well as durability and aesthetic problems. In addition, much quicker initial and final setting times of UHPC are obtained when these are measured as per ASTM C403 [
2], because the needle penetrates through the condensed surface. Therefore, there is a pressing need for new ideas of appropriate methods to prevent the rapid drying and setting of the UHPC surface.
According to a previous study [
3], both free and restrained shrinkage of UHPC increase steeply at a very early age, leading to a high possibility of shrinkage cracking. In order to predict the cracking potential precisely, the tensile strength development needs to be investigated along with residual stress resulting from the restraint of shrinkage. For this reason, Yoo
et al. [
4] developed a direct tensile test apparatus to measure the tensile strength before the final set, based on a previous study [
5]. However, it is time-consuming to measure the tensile strengths at certain times (especially at a very early age); thus, the field application is limited. An alternative nondestructive method for determining the strength development of concrete has been adopted by several researchers [
6,
7,
8,
9,
10]. Because of its many advantages (e.g., continuous measurement of microstructural change in concrete, strong relationship between ultrasonic pulse velocity (UPV) and cement hydration,
etc.), the nondestructive method has attracted much attention from engineers for field applications. However, to the best of the authors’ knowledge, only very limited study [
4] is currently available for predicting the tensile strength development of UHPC at a very early age using a nondestructive method.
In order to improve the shrinkage cracking resistance of UHPC, a number of restrained shrinkage tests have been conducted by many researchers [
3,
11,
12,
13,
14,
15]. In particular, Yoo
et al. [
3] reported that the cracking potential can be mitigated by selecting a lower reinforcement ratio and by using a reinforcing bar with a lower stiffness such as a glass fiber-reinforced polymer (GFRP) bar. Park
et al. [
14] employed a ring test to investigate the effect of using shrinkage-reducing admixture (SRA) and expansive admixture (EA) on the restrained shrinkage behavior of UHPC. Based on the test results, they concluded that the combined use of 1% SRA and 7.5% EA exhibits the best performance compared to other mixtures such as a UHPC mixture without SRA and EA or a UHPC mixture with SRA or EA alone. Therefore, the UHPC mixture with 1% SRA and 7.5% EA has been adopted for several structures built in Korea [
16]. The effectiveness of the combined use of 1% SRA and 7.5% EA in reducing the shrinkage crack width of UHPC slabs has also been reported by Yoo
et al. [
15].
Accordingly, the present study experimentally investigated the setting, tensile strength, and UPV evolutions of two types of UHPC (with and without SRA and EA) at a very early age (before 24 h). In order to determine the optimum surface treatment method for preventing rapid surface drying, four different methods using plastic sheet, curing cover, membrane-forming compound, and paraffin oil were adopted. In addition, the very early age tensile strength was evaluated by a newly developed tensile test apparatus, and a simple power function relationship was developed to predict tensile strength on the basis of UPV.
4. Conclusions
This study investigated the effects of the surface treatment method on the setting properties and UPV evolution of two types of UHPC. In addition, very early age tensile strength development (before 24 h) was measured by a newly developed tensile test apparatus and correlated with the UPV. From the above discussions, the following conclusions can be drawn:
(1) The use of a plastic sheet caused an overestimation of the penetration resistance, whereas the addition of a membrane-forming compound led to an underestimation. The use of either paraffin oil or a curing cover resulted in intermediate values with behaviors that were similar to each other. However, the paraffin oil was chosen to be most appropriate for measuring the penetration resistance of UHPC because of its lower density (0.88 g/cm3), its immiscibility with water, and the fact that it did not interfere with hydration.
(2) The shape of UPV development with age was not affected by the surface treatment method. In contrast, the use of a curing cover or membrane-forming compound resulted in slightly lower UPV values at certain ages, compared to the results when paraffin oil was used. Attaching a plastic sheet to the exposed surface was effective at preventing water evaporation, and thus, it is considered to be a simple method for preventing the rapid surface drying of UHPC elements.
(3) The specimen UH-A (with 1% SRA and 7.5% EA) exhibited more rapid development of penetration resistance, leading to quicker initial and final setting times, and an earlier starting time for the steep increase in the UPV evolution, compared to those of UH-N (without SRA and EA). This is mainly caused by the accelerated stiffness development in the UHPC mortar at an early age due to the large quantities of ettringite that result from the addition of EA.
(4) The S-shaped tensile strength development of UHPC at a very early age (before 24 h) was successfully measured in this study. The tensile strength increased monotonically with the UPV values and could be accurately predicted by using a simple power function.