Elsevier

Cement and Concrete Research

Volume 53, November 2013, Pages 18-24
Cement and Concrete Research

Evaluation of internal bleeding in concrete using a self-weight bleeding test

https://doi.org/10.1016/j.cemconres.2013.05.015Get rights and content

Highlights

  • A linear poroelastic model for bleeding in cement-based materials is proposed.

  • The amount of internal bleeding is then evaluated with a self-consistent scheme.

  • The model confirms the use of round gravel decreases the internal bleeding.

Abstract

Bleeding occurs when freshly mixed concrete consolidates in a form. After the setting of concrete, formation of microstructure in hydrating concrete initiates on the consolidated frame. External bleeding water is excluded and the space occupied by internal bleeding becomes pore in hardened concrete. Understanding and furthermore estimation of bleeding are therefore important to assure the quality of fresh concrete. This paper provides a comprehensive model and advanced experimental setup for analyzing both external and internal bleeding.

Introduction

Bleeding is a phenomenon occurring in freshly mixed concrete and consists in the draining of water out to the surface due to settlement of solid particles after placing in a form [1]. A measure of bleeding, according to ASTM C232 or ASTM C243 [2], [3], is represented by the volume of bleeding water per surface area or the proportional ratio of bleeding water and the net mixing water, in percentage.

Bleeding of a cement-based material mainly depends on the mix proportion and the dimensions of the concrete member. The characteristics of constituents, including particle size, shape, density, and specific surface of cement and aggregates, also influence bleeding [1]. In addition, after a fresh mixture is placed, other conditions affect the amount of bleeding water. If the rate of evaporation is high, the bleeding water on the sample surface decreases. Self-desiccation causes reabsorption of bleed water [4].

Several models have been proposed based on empirical or theoretical approaches to predict the bleeding parameters such as the total amount of bleeding water and the bleeding rate [5], [6], [7], [8], [9], [10], [11]. Self-weight consolidation theory has been widely used to analyze the bleeding phenomenon [1], [6], [7], [9], [10], [11], [12], [13], [14], [15], [16]. Some models considered large strain on the consolidation; however, under a relatively small amount of loading (self-weight), the assumption of small strain is still valid for cement-based materials [17].

This paper proposes a comprehensive model based on small-strain theory of linear poroelasticity, in order to understand and predict the bleeding phenomenon, including internal bleeding (of which quantitative investigation is scarcely reported in the literature). External manifestation of bleeding, which is measured in the proposed experiment, involves only some of the bleeding water. A large amount of bleeding water is trapped within the mixture, a phenomenon known as internal bleeding. A major portion of internal bleeding water is reportedly captured in the vicinity of coarse aggregates and reinforcing bars. Space occupied by excessive internal bleeding can permanently remain as porosity in a material, even though the space is partially filled by cement hydration. The excessive porosity due to internal bleeding weakens the interfacial transition zone around aggregates and the bonding strength of reinforcing bars. Excessive internal bleeding thus causes deterioration of durability and strength performance [5], [18], [19].

While internal bleeding obviously initiates sparse microstructure, the amount of external bleeding is related to the seeding density of cement particles. According to the cement hydration model [20], [21], [22] such as HYMOSTRUC and CEMHYD, a shell of hydration products (for example C–S–H) forms around the cement grains in pore space and the initial pore space, generally controlled by the water-to-cement ratio (the seeding density), affects the formation of microstructure in hardening concrete. The initial pore space should exclude the external bleeding and count on the internal bleeding. The model and experimental measurement on the initial bleeding are important in the purpose, and the duration of the initial bleeding would be less than 30 min after placement.

Different from the previous study [5], [6], [7], [8], [9], [10], [11], this study evaluate the initial bleeding with an advanced experimental setup and the proposed comprehensive model. The initial plastic settlement of freshly mixed concrete was measured with a noncontact laser displacement sensor. Comparing the measured settlement with the model prediction based on the theory of micromechanics gave the material properties to explain the internal bleeding. Finally, the amount of initial (both external and internal) bleeding of concrete could be evaluated.

Section snippets

Theory of linear poroelasticity

Assuming a freshly mixed cement-based material as a fluid-infiltrated solid allows us to adopt the theory of linear poroelasticity. The theory under the small-strain assumption considers the coupling between linear diffusion of fluid mass, water in the problem, and the deformation of a linear elastic porous solid [12], [23].

Self-weight of a sample causes a certain amount of pore pressure, and consequently bleeding will occur. Fig. 1 shows a porous material placed in a hard and impervious

Self-weight bleeding test

Fig. 2 shows the self-weight bleeding test setup using two noncontact laser displacement sensors (KL4-50NV, KAIS Co.). This setup is modified from a plastic shrinkage measurement apparatus [11]. The signal from the sensors is converted to voltage within ± 10 V and recorded via a data logger at 1 point per minute. As shown in Fig. 2(C), two sensors are arranged to detect the vertical movement of two probes, and measure the distance up to each probe using the basic principles of optics. One of the

Results and discussion

All of the results showed no change of the water level, although a small experimental error of 0.1 mm to 0.2 mm was found. The finding corresponds to the theoretical prediction represented by Eq. (13), which supports that the assumption of incompressible ingredients (Ks, Kp, Kf  K) is plausible in the problem of self-weight bleeding. If the Skempton coefficient B > 1, the water level should go down (even with the Biot coefficient ζ = 1  K / Ks < 1).

The measured settlement thus solely account for the

Conclusion

Bleeding is a consequence of a gravity-induced migration phenomenon. After a freshly mixed concrete is placed, solid sediments are separated from the bleeding water. The amount of bleeding water depends on the mix proportion, the characteristics of the ingredients, and their weight. Some of the bleeding water is trapped within the mixtures, referred to as internal bleeding. An excessive amount of internal bleeding decreases the strength and durability of hardened concrete. This paper proposes a

Acknowledgments

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (Award 2012R1A1A1005402).

References (24)

  • G. Sant et al.

    Measurement of volume change in cementitious materials at early ages: review of testing protocols and interpretation of results

    Trans. Res. Board Rec.

    (2006)
  • A. Radocea

    A new method for studying bleeding of cement paste

    Cem. Concr. Res.

    (1990)
  • Cited by (41)

    • Absorption-desorption process of internal curing water in ultra-high performance concrete (UHPC) incorporating pumice: From relaxation theory to dynamic migration model

      2022, Cement and Concrete Composites
      Citation Excerpt :

      As a result, it becomes imperative to cure the UHPC matrix to mitigate the drop of its internal relative humidity, which is a major concern. At present, maintenance technology commonly used in engineering mainly includes internal curing and external curing [15–17]. External curing is to delay or prevent concrete surface moisture loss, which is isolating the concrete from its surroundings [17–20].

    View all citing articles on Scopus
    View full text