Inglês (pdf)
Artigo em XML
Referências do artigo
Tradução automática
Enviar este artigo por email
Citado por SciELO 
Citado por Google 
Similares em
SciELO 
Similares em Google 
Permalink
INTRODUCTION
Ultra-high Performance Concrete (UHPC) is a composite material constituted by an improved gradation of components, a proportion between water and cementitious materials lower than 0.25, and a relevant quantity of reinforcement. Among its mechanical properties, it is worth highlighting a compressive strength greater than 150 MPa, as well as a tensile strength greater than 5 MPa [1].
In the tensile behavior of UHPC, the strength provided by the fiber reinforcement is greater than the cementitious matrix cracking strength [10]. There is however a huge disparity between compressive and tensile UHPC's strengths. In particular, High performance concrete (HPC) has interested engineers during last decades due to its unique properties allowing high durability and high tensile and compressive strengths [11], [12]. The compressive and tensile strengths of HPC are less unbalanced. Early versions of HPC, whose compressive strength exceeds 50 MPa [1], were based on the concept of homogeneity of the concrete matrix taken from Densified Small Particles Concrete (DSP) and Reactive Powder Concrete (RPC) studies [9]. The use of ultrafine aggregates (e.g. silica fume, fly ash, etc.) to optimize the packing density [11], [13], [14] or the use of coarse aggregates to decrease manufacturing costs is usual [15]. In fact, high compressive strength and low porosity of the concrete are the result of two main principles: packing optimization of the material particles [16]-[18], and low water to binder ratios (w/b) [2], [11]. Otherwise, specialized manufacturing and curing methods to produce HPC are frequently required [11], [15]. HPC is a concrete suitable for highway infrastructure repair and replacement [10], the fabrication of precast and prestressed elements for bridges [1], [3] and urban furniture [19], [20]. In the future, HPC precast elements will play an important role in the field of civil engineering [21], [22]. Although it is a material with high performance, its manufacturing is not straightforward because of its high cost, as well as the complexity of the production method and curing. HpC mixing will require equipment that provides more energy than for conventional concrete because the particles in HpC mixes are smaller and the water to cement ratios are low [23]. Insufficient mixing energy can result in inappropriate distribution throughout the mix when ultrafine particles are used, which is detrimental to the HPC performance. It is important to clarify that a particle is defined as fine if its diameter is shorter than 125 μm, while it is ultrafine if it is less than 5 μm. Ultrafine particles (metakaolin, rice husk ash, copper slag, colloidal nanosilica, pulverized fly ash, eggshell powder, tire rubber powder, pulverized granulated blast furnace slag, silica fume, etc.) are normally used in HPC to increase mechanical properties. Among the materials employed to produce HPC, silica fume is the frequently used. However, it is characterized by high surface, like other ultrafine materials, and absorb more water, leading to a decrease the workability of HPC [24]. The latter also increases HPC manufacturing costs, as higher mixing speeds and longer mixing times would be required. In effect, manufacturing HPC may require special mixing techniques and the application of heat, steam, or pressure treatment before and during setting [12], [15], [23], [25]. In addition, these methods require expensive facilities and is sometimes difficult to apply to full size elements like beams or slabs [16], [19]. Therefore, large-scale HPC production is not attractive to industry, limiting its use [26], [27]. The production method is a crucial step in the preparation of any type of concrete, but it is still more relevant when the mechanical properties must be enhanced, such as in HPC. It is worth mentioning here that sometimes when producing and analyzing small quantities of HPC in the laboratory for research purposes, equipment with high speed and long mixing time is used. These conditions clearly differ from the production of an industrial concrete. The objective of the manufacturing method should be thus twofold: ensuring high quality standards and employing simple fabrication steps, without the requirement of specialized and costly materials and curing regimes [23], [28]. Several authors, such as Mohd Zahid et al. [15], [25], Shi et al. [26] and Sobuz et al. [23], indicated that the economic analysis of fabrication of HPC has been little studied so far, highlighting the importance of the study presented here.
For these reasons, it is desirable to consider a simple production method of HPC. Therefore, this work focuses on the manufacture of HPC with techniques and tools commonly applied to standard concrete. Here, 3 HPC mix designs were analyzed, and 23 cylindric concrete samples (100 mm x 200 mm) were tested for each of them to study the correlation between the main parameters of interest (strengths, water to binder proportion, and matrix packing density).
MATERIALS AND METHODS
Three mix designs of HPC were considered and at least 23 cylindrical concrete specimens (100 mm x 200 mm) were manufactured for each mix (see Table 1). Most researchers suggest using a specialized Type II/III cementitious material [15], [29], [30]. In contrast, to fabricate HPC mixes in this research, a standardized general purpose cement or Type I with a density of 3150 kgf/m3 (ASTM C1157 [4]) was used. At present, cement Type I could cost 25 percent less than cement Type III. To produce the HPC mixes, fine aggregate was also used, standardized under ASTM C136 [5], with a maximum particle size of 1.18 mm, a fineness modulus of 1.9 and a density of 2600 kgf/m3 (see Figure 1); a micro silica powder with a density of 2300 kgf/m3 was also used (see Figure 2); a High Range Powder Water Reducer (HRWR) was included (0.3 % by weight of the cement) and spherical particles of steel with a diameter smaller than 2.36 mm with a density of 7000 kgf/m3 were used (see Figure 3). For comparison purposes, polypropylene microfibres with a length of 19 mm, a density of 910 kgf/m3 and a tensile strength of 300 MPa were used (see Figure 4). The target slump of HPC mixtures is 100 mm. No coarse aggregates were used in any mix.
The conventional economic production process used in this research and summarized below is to ensure fluidisation of dry constituents with a mixture of water and HRWR to prepare HPC with better mechanical behavior. First, 50 % of all dry materials, fine aggregate, micro silica powder and cement, are placed in a low speed mixer (drum or planetary mixer) for about two minutes. Then, the water and HRWR are blended, gradually adding the 50 % of it to the dry constituents and stirring at once for about four minutes. Premixing the water and HRWR aids in more uniformly distributing the HRWR. Next, the other 50 % of all dry materials are gradually added and mixed for about four minutes. Finally, the remaining water HRWR mixture and eventually the spherical particles of steel and polypropylene microfibres are combined for about two minutes until the constituents are thoroughly merged and wetted.
Cylindrical specimens were mould for each mix. For each mix design, compressive strength, tensile strength, permeability, and density were determined.
TABLE 1 HPC MIXTURES
Source: The Authors.
Mixture HPC1, with just cement and fine aggregate (quartz sand), was intended to attain a compressive strength greater than 50 MPa. Evidently, the amount of cement used in this sample is considerable. In HPC2 and HPC3, the cement and water are reduced, and the quantity of fine aggregate is optimized to reach a compressive strength greater than 70 MPa. Steel particles were added to increase the compressive strength while leaving constant the amount of cement. For comparative purposes, polypropylene microfibres were added in HPC3 mixture. The specimens were removed from curing tank, dried, and tested. Permeability tests were performed evaluating the end faces of the concrete specimens, according to SIA 262/1 [31] (see Figure 5).
Thirty cylindrical concrete specimens of each mixture were analyzed and tested after 28 days of curing to estimate their compressive strength (see Figure 6), and the ASTM C39 [6] Standard was fulfilled.
Ten cylindrical concrete specimens of each mixture were tested according to ASTM C496 [7] (see Figure 7).
The density of hardened concrete according to ASTM C642 [8] was determined for all cylindrical concrete specimens.
RESULTS AND DISCUSSION
The average permeability coefficient (kT) of the samples was estimated (see Table 2). This coefficient allows measuring the ease with which small particles (e.g. chloride ions) can penetrate the matrix of the HPC. All tested HPC specimens have low permeability compared to conventional concretes whose permeability coefficients ranging from 0.70 to 3 [32]. This is due to the low water to cement ratio and optimized packing system of the specimens manufactured in this research.
The 28-day average compressive strength of the cylindrical HPC specimens was determined (see Table 2). The different HPC mixtures developed in this investigation obtained strengths above 50 MPa. These values are comparable with those obtained in other high performance concretes containing ultrafine particles [11], [13], [14] or whose aggregates were partially substituted by other types of materials [33], [34] (see Table 3). For all different mixes in this investigation, substantial improvements in compressive strength were obtained, since the packing system was optimized by using only fine aggregates of small grain size. The virtual packing density (VPD) represents the maximum potential packing density of a mixture (De Larrard [16]). This model predicts the properties of a mixture using the size distribution and packing density of each monogranular constituent in the concrete mix. The equation describing the VPD of a granular mix containing n particles classes is presented by Eq. (1) (De Larrard [16]):
βi: Residual packing density
di: Diameter of ith dominant class
aij, bij: Factors representing the loosening and wall effect exerted by class j
yj: volume fraction of the ith class
TABLE 2 PERMEABILITY, DENSITY, VIRTUAL PACKING DENSITY AND COMPRESSIVE AND TENSILE STRENGTHS
| Mixture | kT (1016 m2) | Permeability | CS (MPa) | STS (MPa) | Density (kgf/m3) | VPD |
|---|---|---|---|---|---|---|
| HPC1 | 0.09 | Low | 50 | 5.6 | 2183 | 0.70 |
| HPC2 | 0.05 | Low | 70.9 | 4.4 | 2335 | 0.86 |
| HPC3 | 0.08 | Low | 70.4 | 4.9 | 2345 | 0.86 |
CS Compressive strength, STS Split tensile strength.
Source: The Authors.
The 28-day average split tensile strength of cylindrical HPC specimens was determined (see Table 2). The specimens developed in this research have performed identical to high performance concretes found in the literature (see Table 3).
TABLE 3 HPC CHARACTERISTICS
| Aggregates and ultrafine particles | CS (MPa) | STS (MPa) | Cement (kgf/m3) | Mixing technique | Reference |
|---|---|---|---|---|---|
| HPC with crushed stone coarse aggregate, river sand, coarse, and fine recycled concrete aggregate, silica fume | 71 - 85.3 | 3.6 - 4.6 | 500 | High-mix energy | Vinay Kumar et al. [34] |
| HPC with coarse aggregate, metakaolin, rice husk ash | 52.3 - 71.4 | 4.3 - 6.3 | 420-560 | Low-mix energy, drum mixer | Swaminathen et al. [33] |
| HPC with crushed granite coarse aggregate, copper slag, colloidal nano-silica solution | 64 - 70 | 4 - 5 | 500 | High-mix energy, pan mixer | Chithra et al. [13] |
| HPC with crushed stone coarse aggregate, eggshell powder, tire rubber powder | 31.6 - 64.1 | - | 510-600 | Low-mix energy, drum mixer | Othman et al. [35] |
CS Compressive strength, STS Split tensile strength.
Source: The Authors.
By using fine aggregates with smaller sizes than usual, the packing system was optimized (see Figure 8 and Figure 9), demonstrated by the fact that adequate VPD associated with high compressive strengths and low permeability coefficients were achieved (see Table 2). While it is true that smaller particles increase the fluidity of the mixes, they do not necessarily reduce the final strength of the concrete. The optimum mixture is HPC2 because it has a high compressive strength and an adequate density. HPC1 shows the highest tensile strength, although it has a very high amount of cement, which makes the concrete element expensive.
CONCLUSIONS
This research studies the fabrication of HPC, without coarse aggregate, with easily available resources, with simple fabrication and curing procedures, achieving indirect tensile and compressive strengths similar to the HPC produced at present time.
HPC mixes are fabricated using a particle packing design approach to improve the matrix density and reduce the porosity.
It would be believed that the durability of the HPC manufactured in this research is superior to that of conventional concrete, since the permeability coefficients are considerably lower. At the same time, it should increase its lifespan and reduce maintenance costs.
The use of a general purpose or Type I cement and a very small amount of HRWR, which improves the workability of the mix, reduces the production costs of this HPC. This is a significant result since the cementitious materials and superplasticizers used in currently commercialized high-performance concretes are high-priced.
The manufacturing techniques and instruments used here are the same as those used in the fabrication of conventional concretes. A common low speed mixer was used. According to the experience gained in this work, it is possible to state that the production of HP mix can be scaled up and applied in the construction industry; it does not require special curing regimes at high temperatures, nor high speed concrete mixing machines, nor wet chambers, as is required for currently commercialized high performance concretes.
In future research, we will try to increase the tensile strength of HPC to have a material with much more balanced mechanical properties. The long-term performance of this HPC will be investigated in a future study too.
