Shrinkage-induced warping of UHPC overlay cast on hardened NSC substrate under various conditions

https://doi.org/10.1016/j.cemconcomp.2022.104772Get rights and content

Abstract

Ultra-high performance concrete (UHPC) shows very different properties such as high level of autogenous shrinkage and low creep coefficient compared to conventional concrete. One of the problems posed by this is that when UHPC is used as a thin, bonded overlay for concrete bridge decks or pavements, there is limited information regarding the shrinkage-induced warping of the overlay and optimization designs to minimize it. This study conducted a hygro-mechanical finite element analysis to investigate the warping behavior of UHPC overlay cast on mature substrate of normal strength concrete (NSC). The simulation results were validated with the observations from outdoor overlay tests. It is shown that incorporating the shrinkage and creep of UHPC and the bonding behavior of the overlay-to-substrate interface can predict the UHPC overlay's warping deformation in an accurate way. Based on the model, a parametric study was carried out to investigate the influences of sealed curing period, the roughness degree (the bond strength) of overlay/substrate interface, the overlay thickness, the rebar reinforcement inside the overlay, and the amount of total (free) shrinkage of UHPC on the magnitude of the overlay's warping. The findings obtained in this study include the following: (1) The UHPC's shrinkage and the bond strength between UHPC overlay and substrate are two essential factors affecting the overlay's warping behavior, (2) With an increase in UHPC overlay thickness, the calculated overlay's warping deformation is slightly increased, and (3) Internal rebar reinforcement has a limited contribution to reduce the overlay's warping.

Introduction

The superior properties of UHPC, such as excellent compressive and tensile strengths, ductility, ultra-low permeability, and long-term durability make UHPC attractive for various applications [1]. One emerging utilization of UHPC is as a thin and bonded overlay for concrete bridge deck construction or durable rehabilitation [[2], [3], [4], [5]], given the benefits of strengthening the bridge deck against cracking and deterioration due to fatigue loads and environmental erosions. In Europe and North America, bonded UHPC overlays have been deployed on more than 30 bridges since the early 2000s. Typically, the thickness of the UHPC overlay ranges from 25-mm to 55-mm [6], which is much thinner than the traditional concrete overlays and therefore the dead loads on the bridge can be significantly reduced. As expected, the applications of thin, bonded UHPC overlay can greatly improve the strength and stiffness of the bridge deck system [7]. Moreover, the wear resistance of UHPC overlay is demonstrated to be excellent regardless of its thickness [8].

Meanwhile, one of the most critical issues associated with bonded UHPC overlays applied in concrete pavements or bridge decks is the compatibility between the two contact layers made of different materials. The restraint provided by concrete substrate to the shrinkage deformation of UHPC overlay and the dismatch of coefficient of thermal expansion (CTE) between the two materials are the main reasons for the concern. Different modes of failure may arise therefrom, including debonding, warping, and fine cracking, etc. Importantly, the scattered bond strength of UHPC to concrete substrate renders the debonding failure of UHPC overlay a challenge for practical projects. It has been pointed out in a previous study [9] that the failure of conventional overlay resulting from restrained overlay shrinkage is either debonding or transverse cracking, which is mainly dependent on the magnitudes of shear stresses at the overlay/substrate interface and the tensile stresses developed within the overlay. But as is well known, UHPC features an ultra-high tensile strength (10–30 MPa at 28 d), therefore, the loss of bond of UHPC overlay to existing concrete substrate should be the major concern to researchers and engineers. Multiple types of bond assessment tests have been designed for this, among which the slant shear test and the direct shear test are used to assess the shear bond strength of UHPC to NSC, while the direct tension test [10,11], the indirect tension test [12], and the pull-off test [13] are commonly utilized to obtain the tensile bond strength. Sarkar has evaluated the slant shear bond strength between UHPC and NSC substrate made with three different types of surface textures (smooth, low-rough, and high-rough) [11]. It is reported that the slant shear bond strength is sensitive to the roughness of the substrate surface. On the other hand, the splitting tensile strength measured from UHPC-concrete composite specimen was found to be not very dependent on the interface roughness [14]. Zhu et al. [15] carried out an experimental study on shrinkage-induced stresses in bonded UHPC overlay under normal curing and steam curing. It is shown that the substrate surface treatments and curing conditions can produce evident impacts on restrained shrinkage and the warping stress in UHPC.

While a lot of studies have been carried out to quantify the bonding of UHPC material to conventional concrete, limited research has focused on the warping or curling characteristic of bonded UHPC overlay. It is known that compared to conventional concrete, UHPC exhibits a higher level of autogenous shrinkage (especially at the early ages) but a lower drying shrinkage. From this perspective, numerical studies [16,17] simply assumed that the shrinkage strain over the thickness of a thin UHPC overlay is uniform, but the validity of this is questionable. Despite the dominance of autogenous shrinkage deformation in UHPC, drying shrinkage should be paid attention either, especially under the condition of inadequate curing [18], or in some cases that large volume mineral additives are used in UHPC mixture [19]. Excessive cracking rising from UHPC's shrinkage has been found in Refs. [18,20].

UHPC overlays deployed in outdoor environments experience a considerable amount of total shrinkage that can lead to distresses such as warping and debonding. Critically, interfacial debonding may be caused by autogenous shrinkage alone and then facilitates overlay warping. The aim of the present study is to evaluate the warping behavior of UHPC as a bonded overlay on NSC substrate. To this end, a sequential numerical analysis was performed to simulate the relative humidity (RH) profile of UHPC overlay at first, and then the consequent shrinkage and restrained stresses are calculated based on the RH results. That is, the numerical procedure includes a moisture model that could capture the RH development of UHPC and a mechanical model to compute strains and stresses. A practical interface interaction model is also crucial so as to identify overlay debonding when large stresses at the interface of the UHPC overlay and substrate are produced by shrinkage. The creep of UHPC has apparent influences on the stress results, which is properly considered in the mechanical model. Outdoor bonded UHPC overlay tests were performed to validate the numerical results, and to identify the restrained shrinkage strains of the bonded UHPC layer. From the outdoor tests, the effects of sealed-curing period and steel rebar reinforcement were investigated in respect to the UHPC overlay's warping and the restrained shrinkage. And at the same time, laboratory tests assessing the (free) autogenous shrinkage and the CTE of UHPC were carried out. The contents are organized as follows: Section 2 contains the numerical modeling methodology and the identification of the relevant parameters. Analytical cases are presented, with each group having a single factor varied to numerically investigate the effect of sealed-curing period, the roughness (the bond strength) of the overlay-substrate interface, the overlay thickness, the rebar reinforcement, and the amount of total free shrinkage of UHPC on the magnitude of the overlay's warping. Section 3 describes the outdoor bonded UHPC overlay tests. The strain measurement and temperature correction are introduced as well. Section 4 discusses the simulated results of the UHPC overlay's warping, the detailed testing results, and the validation of the numerical model to the outdoor overlay tests in Section 3. Finally, the main conclusions are drawn in Section 5.

Section snippets

Hygro-mechanical finite element analysis

In this paper, the simulation was undertaken in the framework of ABAQUS 6.14 finite element package. The sequential hygro-mechanical analysis is a two-step process as shown in Fig. 1. In the first step, a user-defined material subroutine (UMATHT) for moisture diffusion analysis is applied to simulate the RH distribution in the UHPC overlay. The DC3D8 element was used to mesh the overlay. In this step, only the RH variation of the UHPC layer is considered because the internal RH and shrinkage

Materials and properties

Two concretes were used for the outdoor bonded overlay test, that is, a laboratory made UHPC and a commercial normal strength concrete with a target compressive strength of 40 MPa at 28 d. The designed mixture proportions are shown in Table 3. The UHPC was mixed using a mixer with a capacity of 60 L. The total mixing time exceeded 8 min to achieve a uniform distribution of all mix components.

The compressive strength of UHPC was measured on cubes of 100 mm following standard GB/T 31387-2015 [41

Measured strain development from outdoor overlay test

The restrained shrinkage strains in three overlays were measured under the outdoor temperature condition (5°C–28 °C). Strain correction for CTE mismatch between the strain gauge and the UHPC material was made according to Eq. (6). The corrected strain results obtained in different locations (depicted in Fig. 9) of each overlay are shown in Fig. 11. The three bonded overlays all exhibited initial expansion in the first 2 days after casting. This expansion may be attributed to the continuous

Conclusions

UHPC overlays deployed in outdoor environment experience a considerable shrinkage that can lead to warping and debonding. The scattered bond strength of UHPC to NSC substrate renders the warping of UHPC overlay a challenge for practical projects. This research conducted FE modeling to simulate shrinkage-induced warping of UHPC overlay casting on a mature NSC substrate. A parametric analysis was made to numerically investigate the influences of several factors on overlay's warping behavior.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors thank the support of the National Natural Science Foundation of China under Grant No. 52078273 and the Technology Innovation and Demonstration Project of the Department of Transport of Yunnan Province (2021) under Grant No. 25.

References (44)

  • M. Chilwesa et al.

    Shrinkage induced edge curling and debonding in slab elements reinforced with bonded overlays: influence of fibers and SRA

    Cement Concr. Compos.

    (2019)
  • Ç. Yalçınkaya et al.

    Effects of ambient temperature and relative humidity on early-age shrinkage of UHPC with high-volume mineral admixtures

    Construct. Build. Mater.

    (2017)
  • D.Y. Lei et al.

    Micro-mechanical model for ultra-high strength and ultra-high ductility cementitious composites (UHS-UHDCC)

    Construct. Build. Mater.

    (2021)
  • Y. Wei et al.

    Stress prediction in very early-age concrete subject to restraint under varying temperature histories

    Cement Concr. Compos.

    (2017)
  • L. Østergaard et al.

    Tensile basic creep of early-age concrete under constant load

    Cement Concr. Res.

    (2001)
  • M. Jirásek et al.

    Microprestress–solidification theory of concrete creep: reformulation and improvement

    Cement Concr. Res.

    (2014)
  • Y. Wei et al.

    Microprestress-solidification theory-based tensile creep modeling of early-age concrete: considering temperature and relative humidity effects

    Construct. Build. Mater.

    (2016)
  • Y. Liu et al.

    Effect of calcined bauxite powder or aggregate on the shrinkage properties of UHPC

    Cement Concr. Compos.

    (2021)
  • V. Baroghel-Bouny et al.

    Characterization and identification of equilibrium and transfer moisture properties for ordinary and high-performance cementitious materials

    Cement Concr. Res.

    (1999)
  • T. Rougelot et al.

    Water desorption and shrinkage in mortars and cement pastes: experimental study and poromechanical model

    Cement Concr. Res.

    (2009)
  • Y. Zhang et al.

    Interfacial bond properties between normal strength concrete substrate and ultra-high performance concrete as a repair material

    Construct. Build. Mater.

    (2020)
  • S. Feng et al.

    Experimental study on bonding behaviour of interface between UHPC and concrete substrate

    Construct. Build. Mater.

    (2021)
  • Cited by (5)

    View full text