**1. Introduction**

Reinforced concrete (RC) is one of the most widely used materials for bridge structures. The deteriorated durability of RC structures causes structural safety problems and expensive maintenance costs. Besides, in the process of bridge reinforcement, environmental problems including noise pollution, dust pollution and consumption of natural resources will occur. Meanwhile, the corrosion of rebar in concrete is crucial to the durability of RC structures, especially in chloride aggressive environments [1,2]. Rebar in concrete are not prone to corrosion within a short time in sound concrete due to the high and stable alkalinity of concrete pore solution. However, cracks will be formed due to material shrinkage, thermal gradients and repeated mechanical loading causing a higher osmotic pressure, which often leads to a higher chloride diffusion coefficient in concrete [3,4]. Besides, cracks provide a convenient channel for water ingression, harmful ions and oxygen towards the internal concrete and the surface of rebar, which could result in accelerating corrosion of rebar. Therefore, it is necessary to investigate corrosion behavior of rebar under the effect of crack in chloride aggressive environment [5–8].

A large number of laboratorial studies have been conducted focusing on durability of concrete and rebar corrosion under the actions of chloride penetration and crack. Wang et al. [9] introduced feedback controlled splitting tests to generate crack width-controlled concrete specimens and studied the relationship between crack characteristics and concrete permeability. Ye et al. [10] established a model of chloride penetration into cracked concrete subject to drying-wetting cycles based on Fick's second law. Marsavina et al. [11] made artificial cracks on concrete specimens by a thin copper to perform a concrete chloride penetration test. They demonstrated that the chloride diffusivity in cracked concrete is much stronger than sound concrete. Du et al. [12] and Liu et al. [13] simulated chloride diffusivity in cracked concrete using multi-component ionic transport models and found that the geometry of crack affects chloride transport. Papakonstantinou and Shinozuka [14] established a probabilistic model for rebar corrosion in reinforced concrete structures of large dimensions considering crack effects to simulate the complex phenomena involved in a detailed and simple way, appropriate for implementation on large-scale, real structures. Cao et al. [15] found that oxygen is a crucial factor influencing rebar corrosion propagation process during the full RC structures' service life. Zhu et al. [16] investigated the influence of load-induced cracking behavior on the process of chloride penetration into concrete. Pacheco et al. [17] analyzed bending cracks in RC samples by measuring the electrical resistance across the crack and illustrated that cracks in concrete represent fast routes for chloride penetration, which can result in rebar corrosion. Besides, Šavija and Schlangen [18] claimed that chloride induced corrosion of rebar is one of the most important mechanisms causing deterioration of RC structures and the requirements for their premature repair or replacement. Pedrosa and Andrade [19] indicated that rebar corrosion causes several damage types that influence the structure service life. They analyzed the effect of a range of distinct corrosion rates on the crack growth rate, and they established an empirical model to describe the relation between crack width and the corrosion rate applied for the first stage of cracking. Based on the existed researches, there are different types of cracks caused by temperature, loading and rebar corrosion deteriorating the durability of RC structures, and each crack type has its own distribution inside concrete cover. Thus, performing a laboratorial study on comprehensive effects of crack parameters (such as crack number, width and spacing) on corrosion behavior of rebar could better illustrate the effect of crack on the durability of RC structures.

Nowadays, electrochemical measurements are popular to analyze the corrosion behavior of rebar. The reason is that rebar corrosion is an electrochemical process. An oxidation reaction occurs (Fe → Fe2+ + 2e−) at the anodic area of rebar. The electrons given by anodic area are consumed at the cathodic area (O2 + 2H2O + 4e− → 4OH−). Pore solution as a conducting medium for the transportation of electrons and ions ensure the corrosion process to proceed [20]. The electrochemical nature of corrosion means that electrochemical techniques can be used to monitor the corrosion behavior such as corrosion rate or corrosion current density of rebar in concrete. The commonly used electrochemical techniques include liner polarization (LP), Tafel potentiodynamic polarization (TPP) and electrochemical impedance spectroscopy (EIS) measurements [21]. These techniques are widely used in the studies of carbon steel and alloy steel corrosion [22]. There are a large volume of published researches describing the corrosion behavior of rebar by immersing them into chloride solution [23] or cement extract solution [24,25]. Meanwhile, some researches focus on the corrosion behavior of rebar in real concrete material. Andrade et al. [26] analyzed influence of environmental factors and cement chemistry on corrosion behavior of rebars in concrete by using EIS measurement, and they indicated that redox activity caused by harmful ions in the rebar's oxides layer greatly influences the electrochemical behavior of rebars in the passivity potential domain. In addition, Andrade et al. [27] indicated that different geometrical dispositions of the electrodes used in EIS measurement may affect the test results much. Wang et al. [21] analyzed corrosion rate of rebar in concrete under cyclic freeze-thaw and chloride salt action using LP and TPP measurements. They illustrated that TPP measurement is rapid and easy to operate, read corrosion current directly and provides sufficiently accurate results. Gerengi et al. [28] used EIS measurements to investigate corrosion behavior of rebar in reinforced concrete exposed to sulphuric acid, and they claimed that EIS measurement is one of the most widely used techniques in recent years. Andrade and Alonso [29] applied a non-destructive electrochemical test method for the estimation in large size concrete structures of the instantaneous corrosion current density and discussed the accuracy and applicability of this measurement used in real RC material. In summary, most of the literatures applied one of the electrochemical measurements. However, combining multiple electrochemical measurements may show a more accurate way to undertake the study on rebar corrosion.

In this paper, corrosion current density (*icorr*) of rebar was treated as the index for estimating the effect of crack on durability of RC material. Firstly, *icorr* values of rebar were tested by TPP, LP and EIS measurements. Subsequently, a more reasonable electrochemical testing method was recommended for rebar in RC material. Finally, the effect of crack width, number and spacing on durability of RC material was analyzed by statistical analysis methods.

#### **2. Materials and Methods**

#### *2.1. Materials and Mixture*

Q235 rebars (equivalent to SS400 and A36, i.e., yield strength is 235 MPa) with diameter 7 mm were cut into 110 mm long. Rebars were polished with 340# to 2000# grit silicon carbide emery paper in order to remove the passivation layers and guarantee no pit corrosion on their surfaces. Subsequently, according to a national standard [30], ethanol and acetone treatments were used to degrease surfaces of rebar, so the electrochemical properties of each rebar were sensitive to the effect of chloride ions. One end of each rebar was welded to a copper wire.

PO 42.5 type ordinary Portland cement was used in this study conforming to the requirements of the national standard [31]. Crushed stones with diameters ranging from 2.36 mm to 20.00 mm and natural sands with fineness modulus of 2.7 were adopted as coarse and fine aggregates, respectively. The mixture proportions of concrete used in this study are listed in Table 1. Slump of the mixture was tested to be 40 mm, which means the concrete mixture has favorable cohesiveness and meets well with the requirement of the national standard [32].


**Table 1.** Mixture proportions of concrete.
