Papers
Study of the relation between hydrated Portland cement composition and leaching resistance

https://doi.org/10.1016/S0008-8846(98)00061-1Get rights and content

Abstract

The present paper addresses cement compositions that have an optimal resistance against acid attack and hence, low leaching rates and optimal waste containment. To this end a shrinking core leaching model is used that describes the leaching of metals from a cement sample. This process is directly related to the calcium hydroxide removal from the sample by the acidified leachant. Effective diffusion coefficients in this so-called leached shell were calculated using the equations derived from a cement hydration model. This results in equations in which leaching rates were dependent on cement composition, especially the calcium hydroxide fraction. Optimizing the calcium hydroxide fraction yields cement compositions possessing the optimal leaching resistance as a function of the water porosity or as a function of the hydration degree and water to cement ratio used. The results were also used for the determination of optimal amounts of silica fume and fly ash. A comparison with experimental and practical data reported in literature yields good agreement.

Introduction

Portland cement can be combined with wastes in order to prevent leaching of species such as heavy metals into the environment. This technique is referred to as immobilization or stabilization/solidification (S/S). The ability of cement with respect to fixing hazardous components and prevention of leaching lies in both the physical and chemical properties of the matrix. In leaching tests, cement specimens are subjected to acidified water for prescribed time periods after which metal concentrations in the leachant are measured. In the U.S., the TCLP (1) and ANS/ANSI 16.1 (2) prescribe the execution of such tests, whereas in The Netherlands recently, NEN 7343 (3) and NEN 7345 (4) have been introduced as regulatory tests.

In the past, efforts have been made in order to model the leaching from immobilisates and predict leaching test results. Godbee and Joy (5) used a semi-infinite medium diffusion model and obtained an expression for the leaching from monoliths. Based on the same bulk diffusion model, Brouwers (6) derived an expression for the leaching of granular materials. The bulk diffusion model however does not account for the leachant pH nor the observed matrix dissolution taking place in the cement during exposure to an acidic environment 7, 8. During this process the portlandite (or Ca(OH)2, in cement chemistry notation: CH) present in the cement matrix dissolves. This results in the presence of an unaltered shrinking core and a moving leached shell in which the CH is removed 9, 10. Cheng and Bishop (11) found that in the leached shell stabilized metals were removed while metal concentrations in the core were still unchanged. Hinsenveld and Bishop (12) presented a shrinking core model which describes the transport of species by diffusion through this leached shell.

The presence of CH in the cement matrix is of major importance for the leaching. First, this component acts as acid buffer for the acidified water that enters the matrix and releases the contaminants. On the other hand, after dissolution the removed CH generates extra porosity facilitating transport of species by diffusion through the leached shell and resulting in an increasing progress of the dissolution front in the matrix.

Equations for the computation of cement-gel fraction (in cement chemistry notation: CSH), water fraction (or porosity), and CH fractions as function of the water/cement ratio were derived from the cement hydration model developed by Bentz and Garboczi (13). Furthermore, they simulated CH leaching using their model and derived an expression for the effective diffusion coefficient in a cement matrix during the leaching process 14, 15. This effective diffusion coefficient mainly depends on the porosity of the matrix, and was in accordance with experimental data (16). Their objective however, was limited to reducing the porosity below the critical value where porosity is not connected anymore. However, the positive effect of the CH phase when it acts as buffering barrier against the acid attack was not considered.

In this paper, the results of the mentioned cement hydration model will be used to predict leaching rates as described by the model of Hinsenveld and Bishop (12). Combining the analytical expressions yield an optimal composition of hydrated cement and w/c ratio that minimizes the leaching rate of the sample. It is believed that this information is of major importance in predicting regulatory test results and creating cement matrices that are effective in containing hazardous contaminants. Subsequently, the analytical predictions are compared with experimental results provided by the literature. Finally, the positive effect of adding pozzolanic admixtures such as silica fume and fly ash is analyzed in some detail.

Section snippets

Implementation of matrix composition into leaching model

Following the shrinking core model, the cumulative amount leached per unit exposed surface area can be calculated as follows 12, 17: M(t)=2·De·C02·fmo2·CHβ·t where M(t) is the cumulative amount leached contaminant per unit exposed surface area [mol/m2], De is the effective diffusion coefficient [m2/s], C0 is the initial metal concentration in sample [mol/m3], fmo is the mobile fraction of metal, CH is the H+ concentration in leachant [mol/m3], β is the acid neutralization capacity (ANC) [mol/m3

Calculation of optimal w/c ratio

In the previous section, ϕCH and ϕw have been treated as independent variables. These properties can, however, be related using the equation for C3S hydration used by Bentz (23), which was given in Eq. 10 and their typical specific gravity of 3.2 kg/dm3 for cement. Using this information from their model, the ϕCH and ϕw can be described as a function of the degree of hydration (α) and water/cement ratio (w/c) as follows: ϕCH=0.191αw/c+0.313ϕw=w/c−0.410αw/c+0.313 and hence, ϕtCHw=w/c−0.219α

Comparison with experiments

In the previous section, a theoretical model has been presented for the leaching from hydrated cement matrices. In this section, the theoretical predictions are compared with some experimental data obtained from the literature.

Zamorani and Serrini (26) performed leaching experiments on Cs+-ions immobilized in cement samples at different w/c values. Ordinary Portland cement was used and all samples were cured at 60°C, 98% R.H. for 11 days. The samples were leached in water at a surface-to-liquid

Addition of pozzolanic admixtures

Silica fume, a highly reactive amorphous silica material, can be used as an admixture. It reacts with the CH released during hydration and forms CSH. A typical density of 2.2 kg/dm3 for silica will be used. Two situations should be considered when silica (in cement chemistry notation: S) is present:

  • 1.

    All CH produced during hydration of the cement is consumed by the initial amount of S, forming CSH. This implies that some unreacted silica remains in the hydrated sample, while ϕCH = 0. The

Conclusions

Leaching of metals in a solidified cement sample as a result of an acid attack can be described by a shrinking core leaching model. Both the effective diffusion coefficient and the acid neutralization capacity can be described in terms of cement phase fractions. Using the leaching model and a cement hydration model, it is possible to describe leaching rates as a function of cement paste composition. When the water porosity fraction is known a CH fraction can be calculated at which leaching

References (28)

  • H.J.H. Brouwers

    J. Hazard. Mat.

    (1997)
  • N.I. Fattuhi et al.

    Cem. Concr. Res.

    (1988)
  • K.Y. Cheng et al.

    J. Hazard. Mat.

    (1992)
  • P.G. Baker et al.

    J. Hazard. Mat.

    (1997)
  • C. Carde et al.

    Cem. Concr. Res.

    (1996)
  • P. Faucon et al.

    Cem. Concr. Res.

    (1996)
  • A.A. Kyi et al.

    Cem. Concr. Res.

    (1994)
  • F.P. Glasser

    Waste Management

    (1996)
  • U.S. EPA, TCLP, Solid waste procedure manual, SW-924, Cincinnati...
  • ANSI/ANS 16.1, Measurement of the leachability of solidified low-level radioactive wastes by a short term test...
  • NEN 7343, Leaching characteristics of solid earthy and stony building and waste materials, determination of the...
  • NEN 7345, Leaching characteristics of solid earthy and stony building and waste materials, Determination of the...
  • H.W. Godbee and D.S. Joy, Assessment of the Loss of Radioactive Isotopes from Waste Solids to the Environment, Part 1,...
  • K.Y. Cheng, Controlling mechanisms of metal release from cement-based waste forms in acetic acid solution, PhD Thesis,...
  • Cited by (38)

    • Numerical analysis the calcium leaching effect in concrete and concrete-rock interface on concrete gravity dam anti-sliding stability

      2022, Case Studies in Construction Materials
      Citation Excerpt :

      Gawin [24] adopted Ulm’s rate equation and proposed a hydro-chemo-mechanical model of cementitious materials exposed to contact with deionized water. Many modeling studies focus on the predictions of the degraded depth of the leached materials(Kamali et al. [14]), characterize its mechanical behaviors(Nguyen et al. [4,5], Kuhl et al. [25,26], Stora et al. [27,28]), and microstructure changes(Phung et al. [23], Garboczi and Bentz [29], van Eijk and Brouwers [30], Bentz et al. [2,3]). Only a few studies discuss the Influence of calcium leaching on concrete projects.

    • The influence of disposed fly ash on Ca<sup>2+</sup> leaching and physico-mechanical properties of mortars

      2019, Journal of Cleaner Production
      Citation Excerpt :

      Leaching behaviors of PC pastes containing some powder materials were investigated by different researchers. Some researchers reported that silica fume increased the leaching resistance of concrete against the aggressive environment, having low pH values (Mehta, 1985; Kawai et al., 2008; Jain and Neithalath, 2009; Matte and Moranville, 1999; Lin et al., 2011; van Eijk and Brouwers, 1998; Bentz and Garboczi, 1992; Rozière et al., 2009) but Monteny et al. (2003) reported a decrease in leaching resistance. Some researchers argued that the class F fly ash effectively increased the leaching resistance in the PC system (van Eijk and Brouwers, 1998; Jain and Neithalath, 2009; Kawai et al., 2008; Novais et al., 2016).

    View all citing articles on Scopus
    View full text