Strength optimization of “tailor-made cement” with limestone filler and blast furnace slag

https://doi.org/10.1016/j.cemconres.2004.09.023Get rights and content

Abstract

The use of cements made with portland clinker and two or three additions has grown because they present several advantages over binary cements. Production of composite cements has produced a necessary shift in the manufacture process used in the cement industry. Now, it is known that the separate grinding and mixing technology is more convenient in order to produce these cements, called market-oriented or tailor-made cements. However, their optimum formulations require the help of methods of experimental design to obtain an appropriate performance for a given property with the least experimental effort.

In this study, the interaction between limestone filler (LF) and blast-furnace slag (BFS) is analyzed in mortars in which portland cement (PC) was replaced by up to 22% LF and BFS. For this proposition, a two-level factorial design was used permitting to draw the isoresponse curves. Results show that compressive and flexural strength evaluated at 2, 7, 14, 28, 90 and 360 days are affected in different ways by the presence of mineral additions.

Introduction

Blended cement is an old component of concrete mixtures. Pozzolanic cement was standardized in Italy in 1929, blast-furnace slag cements were produced in Germany, France, Luxembourg and Belgium for more than a century and cements containing fly ash appeared in France in 1950. However, the importance of mineral additions has notoriously rose in the last decades due to cement industry requirements, as well as the need of longer service life for concrete structures.

The use of mineral additions in cement production implies a reduction in consumption of fossil fuel, mineral natural resources and the gas emissions that contribute to the green-house effect. Consequently, partial replacement of portland clinker is a feasible solution from an economic, ecological and technical point of view. Since the 1970s, many efforts have been done to find appropriate mineral additions from natural resources (pozzolans, limestone filler), heat-activated additions (clay, metakaolin) or industrial by-products (fly ash, blast-furnace slag, silica fume, rice husk ash) [1]. Around the world, several combinations of mineral additions have been used to formulate blended cements depending on the available resources that can be found in each country.

During the 1990s, the use of cements made with portland clinker and two additions, called ternary or composite cements, has grown because they present some advantages over the binary cements [2]. Nowadays, composite cements containing combinations of fly ash and silica fume or slag and silica fume are commonly used and many studies have been carried out on this topic [3], [4]. Ternary cements could contribute to achieving the needed balance between the industry's quest for high-performance products and the restrictive environmental regulations. In the ternary cement, the synergistic effects could allow individual component ingredients in such blends to compensate for their mutual shortcomings. They present an opportunity to produce environmentally friendly concrete with tailor-made properties for the requirements of market, without expansive investment cost [5].

In Europe, the EN 197-1 standard identifies two types of portland composite cements according to the replacement level of active mineral additions (slag, silica fume, natural pozzolan, fly ash): type II/A-M containing 6–20% and type II/B-M containing 21–35%. These cements, as other types defined by this standard, can contain a proportion up to 5% of minor additional constituents and they are classified for different strength classes according to the strength gain until 28 days. A similar trend is observed in Latin American countries. Brazilian standard (EB-2138/91) defines the types CPII-E (BFS<34%+LF<10%) and CPII-Z (Pozzolanic material<14%+LF<10%) cements. Composite cements also were standardized in Argentina (IRAM 50000-00) and Mexico (NMX C-414-0/99) admitting the incorporation of two or more mineral additions with a maximum content of 35%. In USA, ASTM C 1157 standard has introduced performance-based hydraulic cements that do not limit the type and the amount of mineral additions that can be blended with portland cement.

Composite cements can be formulated allowing the compensation of mineral additions shortcomings by the synergistic effect produced in ternary blends. For these cements, the basis is usually clinker plus BFS and a third addition such as pozzolan, metakaolin, fly ash, silica fume or limestone filler. They can be produced by intergrinding process or by the grinding and mixing technology. The intergrinding process is easier, technologically simple and the homogenization is made in the grinder device. However, the particle size distribution (PSD) of composite cements obtained by this process depends primarily on the grindability of each component. The second technology consists in grinding and storing separately the components (clinker+gypsum, slag, limestone) and finally mixing them in previously determined proportions to obtain a composite cement according to user requirements. This process has several advantages such as control of particle size distribution of each component, choice the PSD parameter according the role of each component, choice the appropriate technology or process for milling hard or soft component and it is an economic way to produce small amounts of different type of cements according to the market requirement [6], commonly called tailor-made cement[5] or market-oriented[6].

In general, fillers are incorporated in order to complete the granulometric distribution of cement decreasing the water demand, to enhance its granular packing factor and to block up capillary pores. Moreover, filler particles accelerate the hydration of silicate and alumina phases of clinker grains acting as nucleation centers for CH crystals and, as a consequence of this chemical and physical interaction, increasing the early strength of cement [7], [8].

Pozzolanic and cementitious materials, when mixed with portland clinker and water, produce C–S–H similar to that generated from the hydration of calcium silicate of clinker [9]. This reaction is slow compared with that of the portland cement, leading to a lower strength at early ages and similar or higher values at later ages [10]. Also, investigations and practical experiences show that fly ash, natural pozzolan and BFS could increase the water demand [6].

For the optimum formulation of these cements, methods of experimental design can be used to determine the influence of mineral addition proportions on the development of strength and other properties of cement analyzing its multifactorial dependencies. These methods highlight the significance of the effect of experimental variables and their interactions and they present a predictive capability for the response of other experimental points located within the experimental domain. They were successfully applied to find out the optimum proportion of mixtures containing mineral additions for several properties of paste, mortar or concrete [11], [12], [13].

According to the literature review, the combination of LF and BFS in composite cement can help to formulate a cement with an adequate development of strength, because LF contributes to the early strength and BFS increases the long-term strength. On the other hand, experimental design can lead to find the best combination of these mineral additions to obtain an appropriate response for a given property with the least experimental effort.

The objective of this paper is to provide information for the optimization of the compressive and flexural strength in composite cements containing limestone filler and blast-furnace slag.

Section snippets

Materials

A portland cement (PC), limestone filler (LF) and blast-furnace slag (BFS) were used in this investigation. The chemical composition obtained by X-ray fluorescence spectrometry, and the physical characteristics of these materials are reported in Table 1. The mineralogical composition (Bogue) of portland cement was C3S=59.2%, C2S=17.5%, C3A=4.5% and C4AF=9.6%. According to XRD analysis, LF contains 87% of CaCO3 in calcite form and the main crystalline impurity is quartz. As results of chemical

Results and discussion

Experimental results of compressive and flexural strength for each mixture are reported in Table 3. The β0β5 coefficients obtained for the quadratic model for compressive and flexural strength at each test age are given in Table 4, where it can be observed that the coefficient of determination (R2) was always equal or higher than 0.80, indicating a good correlation between calculated and experimental results. The maximum difference between the experimental and calculated compressive strength

Conclusion

For the LF–BFS–PC system studied, containing up to 22% of LF and BFS, the following conclusions can be drawn:

  • At all ages, there is a ternary blend of LF, BFS and PC that present an optimum strength, better than binary LF or BFS cement and plain portland cement. It is attributable to the complementary behavior of both admixtures: LF improves early strength while BFS improves later strength by its cementing reaction.

  • The isoresponse method highlights the significance of the effect of the mineral

Acknowledgements

This research was supported by the Secretaría de Ciencia y Técnica de la Universidad Nacional del Centro de la Provincia de Buenos Aires.

References (24)

  • S. Mindess et al.

    Concrete

    (1981)
  • K. Kovler et al.

    Influence of mixture proportions and curing conditions on compressive strength of high-performance concrete

    ACI Mater. J.

    (2000)
  • Cited by (95)

    View all citing articles on Scopus
    View full text