Accelerated Corrosion Test of Ceramic Recycled Gradient Concrete in Saline Soil Environment

Abstract

Given the problem that steel bar corrosion easily causes cracking of recycled concrete in saline soil areas of Western China, a kind of ceramic recycled gradient concrete is proposed. The polarization curves of steel bars were measured after cracking. The corrosion products of steel bars were analyzed by scanning electron microscopy (SEM) and X-ray energy spectrometry (XPS). The mass fraction of corrosion products was calculated. The volume expansion rate of corrosion products was calculated using the Eighted-Averaging method. The results show that N-60 group ceramic regenerated gradient concrete has the best resistance to corrosion and expansion. The main components of steel corrosion products are Fe₃O₄, FeOOH, and Fe₂O₃, which are consistent with the products under natural conditions, and the content of Fe₂O₃ in the products is the highest. Based on the polarization curves and corrosion expansion rate of steel bars after corrosion, it can be inferred that in a saline soil environment, corrosion products are mainly composed of Fe₃O₄, FeOOH, and Fe₂O₃. An electric accelerated corrosion steel bar can simulate the corrosion condition in a natural environment and has good applicability.

1. Introduction

There are many saline soils and salt lake areas in Western China, which seriously threaten local concrete’s durability. With an increase in construction waste, the application of recycled concrete is becoming more and more widespread. How to improve the service life of recycled concrete in western salt lakes and saline soil areas has become a new topic. As a major producer of ceramics, China generated approximately 9.7 million tons of ceramic waste in 2017 alone. Due to the high content of AL₂O₃, SiO₂, and other compounds in the main components of ceramics, these compounds can react with the hydration products of cement to produce volcanic ash, making concrete denser and improving its durability [1,2]. In the context of the national proposal for green and sustainable development, the treatment and recycling of waste ceramics and building aggregates can not only reduce environmental pollution and urban waste disposal but also achieve the recycling of industrial waste in China.

At present, many researchers at home and abroad have conducted research on recycled concrete and ceramic aggregate recycled concrete. Liu Jinlong et al. [3] summarized the properties of recycled concrete. Due to the high porosity of recycled aggregates, their durability is poor, especially in western high-altitude saline soil areas where their durability is even worse. A. M. Mustafa Al Bakri et al. [4] found that the addition of surface-treated waste ceramic tile aggregates to recycled concrete significantly improves its resistance to sulfate attack, reaching about 85% of that of ordinary concrete. Khuram Rashid et al. [5] studied the effect of waste ceramic tile aggregates on the mechanical properties of recycled concrete. When ceramic aggregates replace 20% of recycled fine aggregates, the frost and thaw resistance of recycled concrete is optimal. The concept of functionally graded materials was proposed by Japanese scholars in the 1990s to adapt to large temperature differences in aerospace equipment [6]. Due to the controllable structural properties of gradient materials, they have gradually been applied in the field of concrete in recent years. At present, there are many studies on the interfacial bonding of gradient concrete both domestically and internationally, but there is relatively little research on the durability of gradient concrete. David G.G. [7] used ordinary concrete as the internal material and 15% mineral powder admixture concrete as the external material to study its interfacial bonding strength. The study found that when the properties of the internal and external materials of gradient concrete are the same, its interfacial bonding ability is better. At present, the method of accelerating corrosion by electrification is widely used. Many scholars soak or semi-soak specimens to obtain corroded steel bars, and then only measure the electrochemical parameters of the steel bars. However, the soaking method reduces the contact surface between the specimens and oxygen, which is inconsistent with actual natural conditions [8,9], and does not quantitatively analyze and calculate the fundamental impact of the expansion rate of corroded materials on concrete cracking. As a widely used computational model, the Eighted-Averaging model has been rarely used to derive the corrosion expansion rate of steel bars in reinforced concrete specimens, and there is even no literature on the calculation of the corrosion expansion rate of steel bars in ceramic recycled gradient reinforced concrete, which is a new functional material, using the Eighted-Averaging model. The Eighted-Averaging model comes from the Model Averaging method, which is a statistical approach aimed at improving overall prediction accuracy by combining the prediction results of multiple models. This method calculates the weighted average of multiple models to obtain the final forecast, thus reducing the error of a single model [10].

Based on this, this article draws on traditional electrification acceleration tests and selects saline soil from a salt lake area in Qinghai Province as the electrolyte to detect the initial state of each specimen and the electrochemical parameters of the steel bars after specimen failure. The corrosion status of each group of steel bars is evaluated and analyzed, and the composition of the corrosion products is determined using an XPS energy spectrum analyzer and electron microscopy scanning. In addition, the Eighted-Averaging model was used to infer the corrosion expansion rate calculation model of each group of ceramic recycled gradient reinforced concrete structures, and the critical expansion rate of each group of specimens was calculated to reflect the ability of each group of concrete specimens to resist the expansion force of steel bars. Based on the natural conditions of the samples, the applicability of the electric acceleration test was comprehensively evaluated, and the model was used to predict and evaluate the service life of reinforced concrete structures in salt lake areas.

2. Raw Materials and Testing Plan

2.1. Raw Materials

(1) Cement: 42.5 ordinary Portland cement is used. All cement chemical composition tables are shown in

Table 1. Chemical composition of cement %.

Raw Materials	SiO2	Fe2O3	Al2O3	CaO	MgO	SO3	C3A
Cement	25.32	5.40	7.25	51.32	3.28	2.09	2.56

. (2) Recycled aggregate: All materials are made of self-made recycled aggregate, and the aggregate is well-proportioned. Recycled coarse aggregate’s void ratio and apparent density are 50.43% and 2486 kg/m³. All prepared recycled aggregates are shown in Figure 1. (3) Natural coarse aggregate: Natural coarse aggregate is crushed stone provided by a commercial concrete company in Lanzhou. The performance indicators of the natural aggregates used are shown in

Table 2. Performance indexes of natural coarse aggregate.

Void Ratio/%	Moisture Content/%	Water Absorption Rate/%	Apparent Density/kg/m3	Bulk Density/kg/m3
45.31	0.3	1.03	2010	1629

. (4) Water. (5) Waste ceramic particles: the performance indicators of ceramic particles and recycled fine aggregates are shown in

Table 3. Performance indexes of waste ceramic tiles and recycled fine aggregate.

Materials	Void Ratio/%	Water Absorption Rate/%	Apparent Density/kg/m3	Bulk Density/kg/m3	Size Distribution/mm
Ceramic particles	25.31	0.41	2680	1610	1–3 mm
Recycled fine aggregate	40.78	0.71	2257	1507	2–4 mm

. (6) Admixture: An SBTMJD-type high-efficiency water reducer was selected for the experiment, with a water reduction rate of 23% and a dosage of 2.5% of the total amount. The flowability of each group of specimens was good, and the slump was guaranteed to be within the range of 130 mm–180 mm.

2.2. Experimental Scheme Design

Select the same batch of steel bars; remove oil stains, impurities, etc., on the surface of the steel bars; choose a suitable size of recycled concrete isolator; place it in a 100 mm × 100 mm × 100 mm concrete test mold [11,12]; pour ceramic recycled concrete outside the isolator; and pour recycled concrete inside the isolator. During the vibration process, a 100 mm steel bar was inserted into the middle of the recycled concrete to ensure that the steel bar extended 10 mm beyond the specimen to connect the wires during the test. Then, while vibrating, the isolator was pulled away until there were no bubbles and bleeding on the surface of the specimen. All specimens were demolded after 24 h and then moved into a standard curing room (temperature 20 ± 1 °C, humidity above 95%) for 28 days of curing.

Table 4. Mix ratio of ceramic recycled concrete.

Cement	Natural Coarse Aggregate	Recycled Coarse Aggregate	Recycled Fine Aggregate	Ceramic Particles	Water	Substitution Rate of Ceramic Particles for Recycled Fine Aggregates%
1	1.61	1.61	1.568	0.392	0.5	20%

and

Table 5. Mix ratio of recycled concrete.

Cement	Natural Coarse Aggregate	Recycled Coarse Aggregate	Recycled Fine Aggregate	Water	Water Reducer%
1	1.61	1.61	1.96	0.5	1

show the mix proportions of recycled concrete and ceramic recycled concrete.

Four specimen groups are established for the test: the first group comprises internal recycled concrete 60 mm in size; the second group consists of internal recycled concrete 70 mm in size. Each specimen group maintains a perpendicular distance of 50 mm between the boundary of the external ceramic recycled concrete and the center of the inside recycled concrete circle. The third and fourth sets of specimens consist of standard recycled concrete examples. To mitigate the corrosion of exposed steel bars during the usual curing process, the specimens are demolded after 24 h of curing to eliminate the impact of corrosion on the test. Copper wires are then wound around the exposed steel rods and wax-sealed. The schematic diagram and physical figure of specimens are illustrated in Figure 2.

Polarization curves of the internal reinforcement in the initial state of four groups of specimens that are standardly cured for 28 d are determined by using a CS350 electrochemical workstation before the test. A saturated KCl calomel electrode is selected as the reference electrode, and the reinforcement in the concrete specimens is used as the working electrode as shown in Figure 3. Subsequently, three sets of specimens with various sizes (N-60, N-70, and N-0) are placed in soil taken from a salt lake area in Qinghai Province, and then a set of recycled concrete specimens (N) is placed in the natural environment of the soil as a control group with N-0. The soil is tested to contain aggressive ions such as Cl⁻, SO4²⁻, Mg²⁺, and HCO⁻, with the maximum content of Cl⁻, and SO4²⁻, and the concentration of corrosive ions in the soil is about 10%. Three groups of ceramic regeneration gradient reinforced concrete specimens are used as anodes, and carbon rods with a diameter of 12 mm are taken as cathodes. A PS−3002d II DC power supply with a range of 5 V and 3 A is used for constant current energization. A constant current is set as 20 mA and a current density is set as 200 μA/cm². To accurately control the moment of cracking of the specimen, strain gauges are pasted on the two end surfaces perpendicular to the axial orientation of the reinforcement of the specimen, and the concrete strains around the reinforcement are controlled. The concrete strain around the rebar is controlled, and when the strain changes abruptly, the strain collector issues a command to cut off the power supply and halt the rusting of the rebar. For group N specimens, based on the group’s current research on recycled concrete, recycled concrete in salty soil field exposure conditions has an average of about 550 d to destruction [10]. Therefore, for this test, at the beginning of the 540 d, specimens were checked daily for crack detection to determine the specimen cracking time. The polarization curves of the internal reinforcement of each specimen at the time of concrete fracture are again determined using a CS350 electrochemical workstation. The measurement scan rate is 0.167 mV/s, and the scanning range is relative corrosion potential −0.1–0.1 V [12,13]. Referring to GB/T50344−2004 “Technical Standard for Building Structure Inspection” [14], the corrosion of steel bars is tested. The concrete specimens are destroyed; the corroded steel bars are removed; and the corroded materials are bagged for scanning electron microscopy (SEM) (Prusai Instrument Manufacturing Company, Beijing, China) and X-ray diffraction to determine the components of the corroded materials, and the expansion rate of the corroded materials of the reinforcement bars at the time of failure of each group of specimens is calculated by the Eighted-Averaging model.

3. Test Results and Analysis

3.1. Polarization Curve Analysis

During the test, group N specimens crack after 676 d under natural conditions, and N−0, N−60, and N−70 specimens produce cracks after 480 h, 624 h, and 576 h of energization, respectively. The polarization curves of each steel bar at the time of cracking of the specimens are detected and compared with their initial state as shown in Figure 4, and the polarization curve fitting parameters are shown in

Table 6. Polarization curves of reinforcing bars and their fitting electrochemical parameters for initial state and structural failure of specimens.

Specimen Number	Specimen Time	Ecorr/mV	Icorr/(10−3 mA.cm−2)
N (Natural conditions)	0 d	−378.15	1.04
	676 d	−405.21	3.25
N−0 (Energized)	0 h	−381.44	1.10
	480 h	−410.71	3.18
N−60 (Energized)	0 h	−361.50	0.98
	624 h	−407.81	3.67
N−70 (Energized)	0 h	−370.09	1.07
	576 h	−404.27	3.46

.

As can be seen from Figure 4 and Table 6, the corrosion potential of group N specimens moves negatively by 27.06 mV from the initial state to the time of cracking, which is the same as that of group N−0 specimens, indicating that when the concrete cracks, the corrosion degree of the reinforcement bars is the same in the two test conditions of natural conditions and electrified acceleration. When the specimen cracks, the corrosion potential of the N−60 group specimen moves negatively by 46.31 mV, suggesting that the corrosion degree of the steel reinforcement inside the specimen is the largest, and the expansion force caused by it is likewise the largest. The reason for this is that as the ceramic particles are mainly composed of compounds such as Al₂O₃ and SiO₂, these compounds react with hydrates in the concrete in a volcanic ash reaction (Al₂O₃ + Ca(OH)₂ + 3H₂O = Ca[Al(OH)₄]₂, Ca(OH)₂ + SiO₂ = CaSiO₃ + H₂O), filling the pores of the regenerated concrete and increasing specimen densification [15], resulting in the ultimate tensile strength of the specimens being greater than that of the remaining groups. Based on this macroscopic occurrence, it can be preliminarily deduced that when ceramic recycled concrete is utilized as an exterior material, it may considerably improve the expansion force of the concrete structure to resist corrosion and postpone fractures.

3.2. Microanalysis of the Composition of Corrosion Products

When cracks are generated in the specimen, the energization is halted; the corroded bars are removed; and the corrosion products on the surface of the bars are collected. The composition of the sample taken is evaluated by employing electron microscope scanning and XPS energy spectroscopy (ULVCA–PHI, New York, NY, USA), along with XPS energy spectroscopy analysis data of our group’s corrosion products. The macroscopic view of the corroded steel bar and the microscopic picture of the corrosion products and XPS energy spectra are displayed in Figure 5.

As can be seen in Figure 5, there are reddish-brown loose corrosion products on the surface of the corroded steel bar, and its microscopic morphology is an irregular flake structure with a rough surface; after removing the red corrosion products on the surface, there is a layer of relatively dense black products close to the surface of the steel bar. From the XPS spectra of corrosion products, it can be found that the corrosion products are predominantly made of Fe, O, and C. Combined with XPS spectral analysis of the corrosion products of steel bars by this group, the Fe in the corrosion products mainly exists in the form of divalent (Fe²⁺) and trivalent (Fe³⁺) ions, and the O exists primarily in the form of C−O bonds and Fe−O bonds, and it can be deduced that the composition of corrosion products is as follows: FeOOH, Fe₃O₄, and Fe₂O₃. Through comparison with the literature [16], the corrosion products produced by the energization test and the composition of the corrosion products of the rebar in the natural environment are basically the same, which can reflect that the energization accelerated test can be a better response to the corrosion of the rebar in natural conditions.

3.3. Calculation of the Volumetric Expansion Rate of Corroded Material

Using Jade 6.0 software to examine XRD patterns, the diffraction peak intensity I and diffraction intensity ratio R of these two corrosion products can be obtained, and the mass fraction w may be computed using [17,18].

$$w_Z={\displaystyle \frac{I_i}{(R_A^N\underset{i=1}{\overset{N}{{\displaystyle \mathrm{\Sigma }}}}\frac{I_i}{R_A^i})}}$$

(1)

In the above equation the following applies: In

Table 7. Composition and phase mass fraction of rust.

Specimen Number	Components of Rust	I	R	w/%
N−0	FeOOH	29	1.187	31.62
	Fe3O4	37	1.542	16.33
	Fe2O3	59	4.997	52.05
N−60	FeOOH	31	1.121	28.15
	Fe3O4	40	1.499	18.21
	Fe2O3	64	5.201	53.64
N−70	FeOOH	27	1.099	29.69
	Fe3O4	36	1.552	17.87
	Fe2O3	61	5.026	52.44
N	FeOOH	25	1.190	11.77
	Fe3O4	36	1.542	17.22
	Fe2O3	60	4.811	71.01

, the results of the calculations for each physical phase are presented. Z is a type of physical phase; N is the total number of physical phases in the corrosion products; A is a type of physical phase that is chosen as the internal standard phase; $w_Z$ is the mass fraction of the physical phase Z; I_i is the diffraction peak intensity of the physical phase I; and $R_A^i$ is the relative diffraction peak intensity ratio of the physical phase i when the physical phase A is the internal standard.

This research employs the Eighted-Averaging approach to ascertain the corrosion expansion of the expansion [17]:

$$V=\underset{i=1}{\overset{n}{{\displaystyle \mathrm{\Sigma }}}}{V}_i$$

(2)

$$m=\underset{i=1}{\overset{n}{{\displaystyle \mathrm{\Sigma }}}}{m}_i$$

(3)

$$w_i={\displaystyle \frac{m_i}{m}}$$

(4)

$\underset{i=1}{\overset{n}{{\displaystyle \mathrm{\Sigma }}}}{w}_i=1$. Assuming that the corroded material’s overall density is $\rho$, the computation can be achieved by substituting Equations (2)–(4):

$$\rho ={\displaystyle \frac{m}{V}}={\displaystyle \frac{\underset{i=1}{\overset{n}{\mathrm{\Sigma }}}{m}_i}{\underset{i=1}{\overset{n}{\mathrm{\Sigma }}}{V}_i}}={\displaystyle \frac{1}{\underset{i=1}{\overset{n}{\mathrm{\Sigma }}}\frac{w_i}{{\rho }_i}}}$$

(5)

$$V_{Fe}=\underset{i=1}{\overset{n}{{\displaystyle \mathrm{\Sigma }}}}{V}_{Fei}$$

(6)

$${\alpha }_i={\displaystyle \frac{V_i}{V_{F\mathrm{e}i}}}$$

(7)

This can be determined by substituting Equations (5) and (7) into Equation (6):

$$V_{Fe}={\displaystyle \sum _{i=1}^n\frac{V_i}{{\alpha }_i}}=m{\displaystyle \sum _{i=1}^n\frac{w_i}{{\rho }_i{\alpha }_i}}$$

(8)

By substituting Equation (8) into Equation (7), it is deduced that the expansion rate of the corroded material is $\alpha$:

$$\alpha ={\displaystyle \frac{V}{V_{Fe}}}={\displaystyle \frac{m}{\rho V_{Fe}}}={\displaystyle \frac{{\displaystyle \sum _{i=1}^n\frac{w_i}{{\rho }_i}}}{{\displaystyle \sum _{i=1}^n\frac{w_i}{{\rho }_i{\alpha }_i}}}}$$

(9)

In the above equation, V_i is the volume of the ith phases; V_Fe is the volume of rust-producing iron; m_i is the mass of the ith phases; w_i is the mass fraction of the ith phases out of the total rust; ${\rho }_{\mathrm{i}}$ is the density of the ith phases; and ${\alpha }_{\mathrm{i}}$ is the expansion rate of the ith phases.

Through Formula (5), it can be found, from the definition of density deduction, the density of the corrosion material only by the density and mass of the corrosion components contained in the decision. For this kind of composition of less corrosion material, it only needs to be measured through the density of the corrosion material through a densitometer and can be calculated through Formula (5). The mass fraction of each component, through Formula (9), can directly find out the expansion rate.

The densities of FeOOH, Fe₃O₄, and Fe₂O₃ are 4.31, 5.18, and 4.87 g/cm³, respectively, and the volumetric expansions are 2.87, 2.09, and 2.31 [19], respectively. By checking the reference [20], substituting the mass fractions w of the different rustications for each group of specimens in Table 5 into Equation (9), the rustications of N, N−0, N−60, and N−70 are determined, respectively, by the expansion rate.

Table 8. Volume expansion rate of each specimen under different working conditions.

Specimen	Volume Expansion%
N−0 (Energized)	2.99
N−60 (Energized)	3.46
N−70 (Energized)	3.37
N (Natural conditions)	2.97

reports the specific results.

Multiple references have shown that the expansion percent of corroded steel bars in reinforced concrete structures is closely related to the service life of the structure [21,22]. As can be observed from Table 8, when all groups of specimens reach the cracking stage, the N−60 group has the largest expansion rate of the corroded reinforcement, followed by N−70, but both expansion rates are larger than those of the N−0, N group. This suggests that the ultimate tensile capacity that ceramic recycled concrete can bear is substantially stronger than recycled concrete. As for the ceramic gradient recycled concrete, the specimens of group N−60 have the strongest ability to resist the expansion of steel corrosion products, and cracks are harder to produce, which is consistent with the macroscopic performance; moreover, when cracks are produced, the expansion rate of the reinforcement of the specimens of group N under a natural environment is basically the same as that of the reinforcement of group N−0 under the condition of an energized accelerated test, which reflects that the energized accelerated test can be a better reaction to reinforcement corrosion under natural conditions.

4. Conclusions

Analyzing the expansion rate of reinforcement and polarization curves of reinforcement in group N and group N−0 specimens, it can be seen that under the conditions of the energized accelerated test, the corrosion of reinforcement is basically the same as that in natural conditions, indicating that the energized accelerated test can better simulate the corrosion effect of reinforcement in natural conditions. According to the analysis of each group of specimens, it is known that, compared with recycled concrete, ceramic recycled gradient concrete prepared under the concept of functional gradient material can effectively resist the expansion force of steel reinforcement and delay crack opening. The specimens in the N−60 group have the best ability to withstand the expansion of corroded steel reinforcement.