High-Volume Natural Pozzolan Concrete for Structural Applications

INTRODUCTION

Concrete, the most widely-used construction material, has some advantages over other well-known structural materials, such as low cost, ease of production, and durability under aggressive conditions. However, the adverse environmental impact of the concrete industry, associated with the portland cement production process, is a problem that needs resolution. The most important impact of the cement industry on the environment is CO2 emission resulting from the calcination of raw materials and the fossil fuels used during clinker production. It is well accepted that CO2 is one of the major gases causing the greenhouse effect that is responsible for global warming. Approximately 1 tonne (1.1 tons) of portland cement production results in the release of approximately 1 tonne (1.1 tons) of CO2 into the atmosphere. The cement industry alone is responsible for approximately 7% of the total CO2 emissions in the world.1 The question is whether it is possible to reduce the portland cement clinker production without decreasing the cement supply needed by the building industry.

The use of mineral admixtures to reduce the clinker content in portland cement results not only in an environmentallyfriendly product but also provides many benefits to the properties of fresh and hardened concrete, such as improvement in workability, reduction in the heat of hydration, low permeability, high ultimate strength, and control of alkalisilica expansions, when compared with plain portland cement concrete. Therefore, high-volume use of mineral admixtures in structural concrete is desirable to obtain proper workability, adequate early-age strength, high strength, and high durability. This is also a useful approach to reduce the environmental impact of the cement and concrete industries. A great achievement in this regard is the development and use of a high-volume fly ash system containing 50% or more fly ash by mass of the cementitious components. This type of cementitious system has good mechanical properties and superior durability characteristics when compared with plain portland cement concrete.2-4 However, it requires a good quality fly ash with high fineness and low carbon content. In the past, the highvolume fly ash concrete mixtures generally did not perform well with respect to strength development and drying shrinkage and durability, because fly ashes produced by old thermal power plants were coarser and usually contained high carbon.5

Uzal and Turanli6,7 previously investigated the properties of blended cements containing high volumes of natural pozzolans (55% by mass) as an alternative environmentally-friendly product when natural pozzolans of desirable properties are available and good quality fly ash is not available. The results of these studies have shown that when compared with ordinary portland cement, it is possible to obtain high-volume natural pozzolan blended cements with comparable mechanical properties in late ages and very good durability characteristics.

This study was carried out to develop structural concrete mixture containing 50% natural pozzolan by mass of the total cementitious material. Three different kind of natural pozzolans from Turkish deposits, that is, natural zeolite, perlite (a glassy pozzolan), and volcanic tuff were used to prepare the high-volume natural pozzolan (HVNP) concrete mixtures with 680 lb/yd^sup 3^ (400 kg/m^sup 3^) total cementitious material content and 0.45 water-cementitious material ratio (w/cm). For comparison purposes, two other concrete mixtures, one with a low-calcium fly ash and the other with a granulated blast-furnace slag, were also prepared with the same mixture proportions. In addition, a conventional portland cement concrete mixture with 0.61 water-cement ratio (w/c) was prepared as a reference mixture. The concrete mixtures were tested for air content, setting time, compressive strength, splitting-tensile strength, and resistance to chlorideion penetration (ASTM C 1202).

RESEARCH SIGNIFICANCE

Most pozzolanic materials, especially natural pozzolans, tend to increase the mixing water requirement for concrete and lower the rate of strength development. Therefore, for structural applications, their proportion in cementitious components of concrete mixtures is generally limited to 30% or less. Published literature contains no reports on concrete mixtures containing a high volume of natural pozzolans for structural purposes. One of the objectives of this study is to fulfill this need.

EXPERIMENTAL

Materials

Ordinary portland cement and supplementary cementing materials-An ordinary portland cement (PC) was used in the study. Its chemical composition and physical properties are given in Table 1. Three different natural pozzolans, a low-calcium fly ash, and a blast furnace slag were used in the study as supplementary cementing materials (SCMs). Natural pozzolans and slag were ground by a ball-mill to have approximately 80% passing through a 45 µm sieve, whereas the fly ash was used as received with 84% particles passing through a 45 µm sieve. The chemical composition and physical properties of SCMs, along with their notations, are given in Table 2. The grinding times applied for natural pozzolans and slag to obtain powders having 45 µm passing value of 80% are also given in Table 2 for comparison purposes. Mineral composition and particle size distribution curves of the SCMs are given in Table 3 and Fig. 1, respectively.

Particle size distribution curves indicated that NZ and VT are finer than the other mineral admixtures for the particles smaller than 5 µm, although all the SCMs have similar percent passing 45 µm values.

High-range water-reducing admixture-A high-range water-reducing admixture of sulfonated melamine condensate type in dry powder form was used in combination with mineral admixtures to obtain approximately 100 mm (4 in.) slump.

Aggregates-A crushed quartzite-type aggregate was used in the study with nominal maximum aggregate size of 25 mm (1 in.). Three different size groups of aggregates were combined to obtain proper gradation. The gradation curve for the combined batch is shown in Fig. 2.

Proportions of concrete mixtures

One concrete mixture for each type of SCM and also a conventional portland-cement concrete mixture were made. The proportions of the concrete mixtures are given in Table 4. A conventional mixture was designed to achieve 4.35 ksi (30 MPa) compressive strength at 28 days, 100 to 150 mm (3.9 to 5.9 in.) slump with 0.61 w/c and 400 kg/m^sup 3^ (680 lb/yd^sup 3^) cement content. On the other hand, high-volume SCM mixtures were made with 50% replacement of portland cement by mass and 0.45 w/cm. The slump of the high-volume mixtures was adjusted to 90 to 100 mm (3.5 to 3.9 in.) with the help of the high-range water-reducing admixture. High-volume mixtures were made with 0.45 w/cm because the goal was to achieve a 28-day compressive strength similar to that of the conventional mixture.

Test methods

The slump, unit weight, and air content of the fresh concretes were determined immediately after mixing. A cubic specimen was cast for each mixture to determine the initial and final setting time of concrete in accordance with ASTM C 403. Cylindrical specimens, 10 x 20 cm (4 x 8 in.), were cast to determine the compressive and splitting tensile strength at various ages and the chloride-ion penetration of concretes. The cylinders were demolded after 24 hours and cured in lime-saturated water until the test ages.

Compressive and splitting tensile strengths of the hardened concrete samples were determined for 3, 7, 28, 91, and 180 days of age in accordance with ASTM C 39 and ASTM C 496, respectively. The resistance of the concrete specimens to the chloride-ion penetration was determined at 28 and 91 days according to ASTM C 1202.

RESULTS AND DISCUSSION

Properties of fresh concrete

The slump, air content, unit weight, and setting time of the fresh mixtures are given in Table 5. The dosage of the highrange water-reducing admixture used for high-volume SCM mixtures to provide a slump of approximately 100 mm (4 in.) varied with the type of SCM. HVNP mixtures required 9 kg/m^sup 3^ (15.3 lb/yd^sup 3^), 2.6 kg/m^sup 3^ (4.4 lb/yd^sup 3^), and 6 kg/m^sup 3^ (10.2 lb/yd^sup 3^) high-range water-reducing admixture dosage for NZ, P, and VT, respectively. Due to relatively high specific surfaces of NZ and VT which are indicated by Blaine fineness values (Table 2), the high-range water-reducing admixture requirement of the mixtures was significantly higher than that of the mixture with P. For the mixtures with FA and BFS, 2 kg/m^sup 3^ (3.4 lb/yd^sup 3^) dosage of high-range water-reducing admixture was enough to provide the specified slump.

The initial and final setting times of the HVNP mixtures varied depending on the type of natural pozzolan used in the mixture. The HVNP mixtures demonstrated shorter initial and final setting times when compared with the high-volume mixtures with FA and BFS. It should be noted that the HVNP mixture with NZ exhibited the shortest initial and final setting times when compared with the other mixtures, which can be attributed to the high specific surface of NZ and a possible chemical interaction between NZ and the highrange water-reducing admixture.

Compressive strength and splitting-tensile strength

The compressive and splitting-tensile strength of the mixtures are given in Tables 6 and 7, respectively. The HVNP concrete mixtures exhibited lower 3-day compressive strengths and lower or similar 7-day compressive strengths when compared with the conventional portland-cement concrete mixture. The compressive strength of the HVNP mixtures with NZ and VT exceeded the strength of the conventional mixture at 28 days, whereas the compressive strength of the HVNP mixture with P was still slightly lower than that of the conventional mixture. Comparing the HVNP mixtures to those made with FA, it can be concluded that the 3-, 7-, and 28-day compressive strengths of HVNP mixtures were comparable to HVFA concrete mixture; however, after 28 days, the mixtures with FA showed higher compressive strengths. This behavior was attributed to slower pozzolanic activity of FA resulting from its relatively lower specific surface (Table 2) and coarser particle size distribution (Fig. 1) when compared with the natural pozzolans. Concrete mixtures containing BFS showed high strength even at 3 days. At 91 days, the HVFA concrete gave the highest strength.

Compared with the Control Mixture No. 1, all cements containing SCMs exhibited higher splitting-tensile strength at the ages of 91 and 180 days, which is attributable to enhanced strength of the interfacial transition zone between aggregate and cement paste as a result of the lower w/cm and pozzolanic reaction. The pozzolanic reaction is slow but it has the effect of filling up the capillary voids with additional C-S-H.

Resistance to chloride ion penetration

The data on the resistance of the concrete mixtures to chloride-ion penetration determined in accordance with ASTM C 1202 for the ages of 28 and 91 days are shown in Table 8. Again both ages, the concretes containing SCMs exhibited much higher resistance to the chloride-ion penetration (lower charge passed) when compared with Control Mixture No. 1, due primarily to the lower w/cm and pozzolanic reaction. According to the permeability rating system suggested by ASTM C 1202, at 91 days, Concrete Mixtures No. 2 to 6 exhibited very low permeability, whereas Control Mixture No. 1 showed high permeability. This is attributable partially to the lower w/cm of pozzolan concretes and partially to the fact that pozzolanic reaction reduces the ionic concentration of the pore fluid.

CONCLUSIONS

Based on the experimental results, the following conclusions are drawn concerning high-volume natural pozzolan concrete mixtures suitable for structural applications;

1. Due to their high fineness, the high-volume natural pozzolan concretes with natural zeolite and volcanic tuff required a significantly higher high-range water-reducing admixture dosage than the concrete mixtures with perlite, fly ash, and ground granulated blast-furnace slag. For the same reason, concrete mixtures with natural zeolite and volcanic tuff exhibited shorter initial setting times when compared with other concrete mixtures of this study;

2. Again, due to their higher fineness, concrete mixtures with natural zeolite and volcanic tuff showed higher compressive strengths at 28 days when compared with the conventional mixture, whereas concrete mixtures containing perlite and slag showed slightly lower strength. At 91 days and beyond, however, the compressive strengths of all highvolume pozzolan concrete mixtures, except those containing perlite, exceeded the strength of Control Mixture No. 1. A similar trend is shown by the splitting-tensile strengths data;

3. At 28 and 91 days, the concrete mixtures with a high volume of natural pozzolans, fly ash, or slag showed considerably higher resistance to chloride-ion penetration when compared with the conventional mixture. At 91 days, all high-volume pozzolanic concrete mixtures showed very low permeability rating according to ASTM C 1202 when compared with the control concrete mixture which gave a high permeability rating; and

4. Based on the results, it appears that the high-volume natural pozzolan concretes in this study are similar in properties to high-volume fly ash and high-volume slag concretes and, therefore, are suitable for structural use.