Insights into the influence of waste ceramic tiles powder on cement hydration and microstructure from electrochemical impedance spectroscopy

To reduce carbon emissions from cement production and achieve the resource utilization of ceramic wastes, this work investigated the effects of waste ceramic tiles powder (WCTP), used as a mineral admixture at replacement levels of 0%–50% by mass, on the course of hydration, microstructural evolution, and compressive strength of cement-based materials. The research method centered on electrochemical impedance spectroscopy (EIS), and was complemented by uniaxial compression tests, mercury intrusion porosimetry analysis, and microstructural observation for cross-validation. The results show that the incorporation of WCTP delays the early-age hydration process, leading to a substantial decrease in 3-day compressive strength with increasing replacement rate; at a 50% replacement level, the 3-day strength is only 35% of that of the Portland cement reference sample. The addition of WCTP coarsens the pore structure of cement-based materials at early ages. Nevertheless, in the later stages of hydration (at 56 days), the pozzolanic effect of the WCTP gradually becomes prominent, resulting in a denser microstructure. Consequently, the compressive strength of specimen with a 20% replacement rate reaches 95% of the strength of the reference sample. EIS analysis reveals that during the early period of hydration (before 7 days), the incorporation of WCTP reduces both the bulk resistance and its growth rate. In the later stage, however, the bulk resistance growth rate surpasses that of Portland cement specimen. The pore solution resistance decreases with increasing WCTP content, particularly at early ages. At later ages, the differences in pore solution resistance among hardened pastes with 20%–50% WCTP replacement gradually narrow, indicating that the differences in porosity within the hardened pastes tend to decrease. Furthermore, EIS analysis shows that the charge transfer resistance is positively correlated with compressive strength. These findings confirm that bulk resistance, pore solution resistance, and charge transfer resistance are key electrochemical indicators suitable for the non-destructive monitoring of hydration processes and strength gain in cementitious materials incorporating WCTP.

1 Introduction

As the world’s predominant man-made building and construction material, concrete consumption averages about 1 cubic meter per person annually (Adesina, 2020). Portland cement is the principal cementitious component in concrete manufacturing, and its production is the main source of carbon emissions of concrete (Robalo et al., 2021). For every kilogram of cement produced, approximately 0.66–0.82 kg of carbon dioxide are emitted, and the concrete industry accounts for 33% of human-caused CO₂ emissions (Jyosyula et al., 2020). With the continuous growth in concrete demand, it is estimated that the consumption of Portland cement could reach a volume of 6 billion tons by 2060 (Raveendran and Vasugi, 2024). Therefore, the development of new concrete that is both environmentally friendly and sustainable has become a critical challenge that the construction industry and academia must tackle.

Meanwhile, a large amount of waste generated by the construction ceramics industry has caused serious environmental problems. In 2024, global ceramic tile production reached 14.95 billion square meters, with Asia contributing 10.9 billion square meters, accounting for 72.8% of the global output (Baraldi, 2025). Due to factors such as inappropriate raw material selection and poor kiln temperature control, the defect rate in the building ceramics industry exceeds 7%, while the defect rate for sanitary ceramics and daily-use ceramics is even higher. Coupled with the demolition of old ceramic products, the total amount of discarded ceramics is quite substantial. Currently, even in major ceramic-producing countries including Italy and Spain, the resource utilization of discarded ceramics remains a technical challenge.

Grinding waste ceramics into powder as a new type of mineral admixture is an effective way to achieve their resource utilization. The main chemical components of waste ceramics are silicon dioxide (SiO₂) and aluminum oxide (Al₂O₃). Preliminary tests indicate that when WCTP is directly mixed with water, even at elevated temperatures up to 60 °C, the paste fails to harden with no significant hydration reaction observed. Research by Chen et al. demonstrates that when water glass is used as an alkali activator to activate waste ceramic powder, the paste remains unhardened even after 24 h at room temperature (Chen et al., 2023). This further confirms the low reactivity of WCTP, which inherently cannot undergo hydration reactions. However, WCTP exhibits significant pozzolanic activity and can participate in pozzolanic reactions with calcium hydroxide generated by cement hydration, producing calcium silicate hydrate (C-S-H) gel (Ren et al., 2025). Additionally, ceramic powder with smaller fineness can fill the pore network within the hardened cement paste, exerting a microaggregate effect (Sun et al., 2021).

Existing studies on incorporating waste ceramics as a new mineral admixture into cement-based materials mainly fall into two categories: first, crushing waste ceramics into coarse or fine aggregates to replace natural aggregates in concrete, focusing on microstructure (Awoyera et al., 2017; Awoyera et al., 2021; Zareei et al., 2019), mechanical properties (Gonzalez-Corominas and Etxeberria, 2014; Bommisetty et al., 2019; Zareei et al., 2019), and sulfate attack resistance (Medina et al., 2016); second, grinding waste ceramics into powder as a mineral admixture, investigating the physical properties, workability (De Matos et al., 2020; De Matos et al., 2021; Elemam et al., 2023; Li et al., 2023), mechanical properties (Lasseuguette et al., 2019; Li et al., 2020; Zhao et al., 2020; Hoppe Filho et al., 2021), and durability (Samadi et al., 2020; Doleželová et al., 2022; Bayraktar et al., 2024; Tokareva and Waldmann, 2025) of cement composites incorporating ceramic powder.

Hydration is the core physicochemical process that governs the microstructure formation and macroscopic performance of cement-based materials (Zajac et al., 2020; Zajac et al., 2022; Liu et al., 2022; Yang et al., 2022). Traditional methods of material research such as mercury intrusion porosimetry and scanning electron microscopy are used to analyze the pore structure of cement-based materials (Zajac et al., 2020; Zajac et al., 2022; Yang et al., 2022) and to observe their microstructure (Liu et al., 2022; Yang et al., 2022; Zajac et al., 2022), respectively. Electrochemical impedance spectroscopy (EIS), a sensitive and efficient non-invasive method, can real-time monitor cement hydration and microstructure evolution (Chi et al., 2019; Fita et al., 2022). It has significant advantages in evaluating the microstructure development of cement composites (Chi et al., 2019; Li et al., 2024) and is widely applied to assess cement hydration, drying shrinkage, and carbonation behavior (Dong et al., 2016; Sun et al., 2017; Park et al., 2022; Li et al., 2024). By constructing a reasonable equivalent circuit (EC) model, electrochemical parameters related to the physicochemical state of the material can be extracted from impedance spectra, thereby deepening the understanding of hydration mechanisms.

Regarding the effect of ultrafine ceramic powder (with a maximum particle size not exceeding 10 microns) as a replacement for Portland cement on microstructural properties, overall, as the replacement ratio of ultrafine ceramic powder increases, the fraction of macropores shows a decreasing trend; the proportion of macropores (>50 nm) in the early hydration stage (7 days) is basically the same as that in the macropores of pure Portland cement paste specimens. In the late hydration stage (180 days), when the ceramic powder replacement level is 40%, the proportion of macropores can be reduced to one-third of that in pure Portland cement paste specimens. At replacement levels of 10%–40%, the proportion of macropores can be reduced to below 15% (Li et al., 2020). One major benefit of using waste ceramics as a substitute for Portland cement is the reduction of carbon emissions from cementitious materials. However, as the fineness of ceramic grinding increases, the energy required—and consequently the associated carbon emissions—also significantly rise. For ceramic powders with a particle size comparable to that of Portland cement, substituting cement results in a certain reduction in strength, although the pore structure of the paste has not yet been reported. Moreover, one of the fundamental questions in materials science concerns the relationship between microstructure and performance. The most quantifiable indicator of reaction microstructure is pore characteristics, while the most commonly assessed macroscopic property is strength. For porous materials, researchers have established several porosity-strength relationships, such as the empirical formulas by Balshin and Ryshkewitch. However, there is ongoing debate regarding the range of porosity values to be considered in these formulas. Are there other easily measurable, simple formulas that reflect both the pore characteristics and strength of materials? Regarding waste ceramic tiles, from a practical standpoint, if they are ground to a fineness comparable to cement and used to replace Portland cement at different substitution levels, how would their strength performance manifest?

Consequently, this study investigated the effects of varying ceramic powder dosage and hydration ages on the electrochemical response, microstructure, and macroscopic strength of waste ceramic powder-cement composite cementitious materials, with electrochemical impedance spectroscopy (EIS) as the primary research method. By integrating equivalent circuit modeling, compressive strength testing, MIP, and SEM analysis, this study is focused on uncovering the regulatory mechanism exerted by ceramic powder on the hydration and hardening processes of this system. Moreover, this work endeavors to establish intrinsic correlations between key electrochemical parameters and macroscopic material properties, thereby providing a scientific basis and practical reference for the engineering implementation of this massive low-carbon solid waste-based cementitious material.

2 Materials and methods

2.1 Raw materials

P.I. 42.5 Portland cement adopted in this experiment was manufactured by Liaoning Fushun Cement Co., Ltd. Waste ceramic powder, prepared by crushing and grinding defective tile wastes from the production line of Shijiazhuang Laoguo Tile Factory, served as the mineral admixture. According to the particle size distribution curves in Figure 1A, the ceramic powder particles possess a modal size of approximately 10 μm, while the cement particles are centered at 45 μm. The detailed chemical makeups of the waste ceramic powder and cement are listed in Table 1.

Figure 1

Figure 1. (A) Particle size distribution of cementitious materials and (B) Impedance testing flowchart.

Table 1

Table 1. Chemical compositions of cement and waste ceramic powder.

Municipal tap water from Shijiazhuang was utilized in this study. A polycarboxylate high-efficiency water reducer was adopted at a dosage of 2% relative to the total mass of cementitious materials.

2.2 Mix design and specimen preparation

In this investigation, the water-to-binder ratio was fixed at 0.25, and six experimental groups were designed, with the mass replacement rates of waste ceramic powder set at 0% (control group), 10%, 20%, 30%, 40%, and 50%. Table 2 lists the specific mix proportions used in this study.

Table 2

Table 2. Paste mix design.

After thoroughly mixing the raw materials in a mixer according to the specified proportions, they are cast into 40 mm × 40 mm × 160 mm steel molds and then placed in a standard curing chamber (temperature 20 °C ± 2 °C, relative humidity ≥95%) for 24 h. After demolding, the specimens continue to cure under the same conditions until the designated testing ages (3, 7, 14, 28, and 56 days).

2.3 Test methods

Cement-based materials are porous materials that contain numerous gel pores and capillary pores, which are filled with solutions containing various ions. These ions include hydroxide ions, calcium ions, sodium ions, sulfate ions, and others, forming a complex electrochemical system.

By placing electrodes on the surface or inside the specimen and applying low-amplitude, varying-frequency alternating current, the microstructure and material transport properties of the specimen can be studied.

The AC impedance method involves applying a small-amplitude sinusoidal AC signal to the system under test, which generates a corresponding response signal. The properties of the system are described by the ratio of the output response signal to the input perturbation signal. By using an input signal with frequency as a variable, as long as the signal is sufficiently small to satisfy the conditions of a linear dynamic system, the transfer function reflecting the system’s properties can be obtained through measurements of the output signal. The relevant formulas are shown in Equations 1–10.

$\text{Input\hspace{0.17em}sinusoidal\hspace{0.17em}voltage\hspace{0.17em}disturbance\hspace{0.17em}signal}:E=E_0\sin \left(\omega t\right)(1)$

$\text{Complex\hspace{0.17em}form}:E=E_0\exp \left(j\omega t\right)(2)$

$\text{Corresponding\hspace{0.17em}output\hspace{0.17em}response\hspace{0.17em}signal}:I=I_0\sin \left(\omega t+\theta \right)(3)$

$\text{Complex\hspace{0.17em}form}:I=I_0\exp \left[j\left(\omega t+\theta \right)\right](4)$

where $\omega$ is the frequency and $\theta$ is the phase angle.

The ratio of the output signal to the input signal, denoted as $Y\left(\omega \right)$, is called the admittance:

$Y\left(\omega \right)=\frac{I_0\sin \left(\omega t+\theta \right)}{E_0\sin \left(\omega t\right)}(5)$

$\text{Complex\hspace{0.17em}form}:Y\left(\omega \right)=\frac{I_0{e}^{j\omega t}{e}^{j\theta }}{E_0{e}^{j\omega t}}=\frac{I_0}{E_0}{e}^{j\theta }=\left|Y\right|e^{j\theta }(6)$

Using Euler’s formula $e^{j\theta }=\cos \theta +j\sin \theta$, the admittance can be converted to impedance.

Impedance Z:

$Z\left(\omega \right)=\frac{1}{Y\left(\omega \right)}=\frac{E}{I}{e}^{-j\theta }=\left|Z\right|e^{-j\theta }(7)$

Real and imaginary parts representation:

$Z^{\prime }\left(\omega \right)=\left|Z\right|\cos \theta (8)$

$Z^{″}\left(\omega \right)=\left|Z\right|\sin \theta (9)$

$Z\left(j\omega \right)=Z^{\prime }\left(\omega \right)-j{Z}^{″}\left(\omega \right)(10)$

The real part of the impedance is called resistance, and the imaginary part is called reactance.

Evaluation metrics: The charge transfer resistance $R_{ct}$ is very sensitive to changes in the microstructure and continuously increases as the hydration process progresses, which can indirectly reflect the degree of hydration of cement-based materials. The bulk resistance is closely related to porosity, and $R=R_{\mathrm{\Omega }}+R_{ct}$. Additionally, a logarithmic relationship exists between the compressive strength and flexural strength of NMK/FA cement mortar, its diffusion resistance coefficient, and its bulk resistance (Li Q. et al., 2025).

By fitting the measured impedance spectroscopy data with an equivalent circuit model, parameters reflecting the microstructure of cement-based materials and their mass transport properties can be obtained.

Electrochemical Impedance Spectroscopy (EIS) Testing: Cylindrical specimens with a diameter of 28 mm and a height of 20 mm were fabricated using the test mix ratio for EIS tests. Before specimen preparation, two 0.8 mm-thick stainless steel electrodes (30 mm × 20 mm) were pre-embedded in the mold. The electrodes were arranged in parallel and symmetrically with a spacing of 10 mm. After molding, the specimens were cured to the target ages (3 d, 7 d, 14 d, 28 d, 56 d) in a standard curing chamber. EIS tests were performed on the specimens of different ages using a HIOKI IM3570 precision impedance analyzer. Before testing, the specimens were subjected to vacuum water saturation treatment in a vacuum pump. The vacuum saturation solution was saturated calcium hydroxide solution, and the vacuum degree was maintained at approximately −90 kPa for 30 min. After standing for 4 h, the alternating current (AC) impedance spectroscopy tests were conducted. During the tests, the analyzer was connected to the two stainless steel electrodes, and after testing, the specimens were placed back into the saturated calcium hydroxide solution for continued curing. The frequency range was 4 Hz–5 MHz, with an applied AC voltage amplitude of 10 mV. The testing procedure is shown in Figure 1B. The collected impedance data were analyzed for Kramers-Kronig (K-K) relation verification and equivalent circuit (EC) fitting using ZView2 software.

Compressive Strength Test: According to the national standard Test method for strength of cement mortar (GB/T 17,671—2021), the compressive strengths of the specimens at 3 d, 7 d, 14 d, 28 d, and 56 d were measured using a YEW-2000 universal testing machine.

Mercury Intrusion Porosimetry (MIP): Specimens reaching the predetermined ages were crushed, and samples of approximately 3 mm in size were taken from their central parts. The hydration process was terminated by immersing the sample in anhydrous ethanol for 24 h, subsequently followed by oven-drying at 50 °C to achieve a constant weight. Porosity and pore size distribution measurements were conducted on the sample using an AutoPore V Mercury Porosimeter.

Scanning electron Microscopy (SEM) Observation: A sample of approximately 3 mm in size was obtained from the core region of the specimen that had reached the predetermined curing age. To avoid damaging the microstructure during the drying process, the sample was first subjected to solvent exchange in an acetone solution to remove pore water, followed by drying at 50 °C. The dried sample was subjected to gold sputtering treatment prior to microstructure observation using a Hitachi SU8010 Scanning electron microscope.

3 Results and discussion

3.1 Analysis of compressive strength test

Figure 2 presents the variation in compressive strength of specimens incorporating waste ceramic powder as a mineral admixture at various curing ages.

Figure 2

Figure 2. Compressive strength variation of specimens with age.

As depicted in Figure 2, both the pure cement specimens and the waste ceramic powder (WCTP)-blended cementitious paste specimens exhibit continuous strength development with increasing hydration age. The strength development behavior is significantly affected by the incorporation of WCTP. At each testing age, Higher WCTP replacement rates were found to negatively correlate with the compressive strength development of the cementitious paste.

The WCTP-incorporated specimens exhibited compressive strength at 28 d varying from 40.7 MPa to 73.7 MPa, and the WCTP-blended cement demonstrates lower strength than the pure cement specimen (the control). Higher WCTP replacement rates at 3 days reduce compressive strength. At 50% WCTP replacement, relative strength is 35% at 3 days, 55% at 28 days. This is due to a reduced cement clinker mass fraction in the paste, caused by WCTP. WCTP has a lower calcium oxide content and lower silica-alumina phase reactivity, resulting in a slower hydration rate. Consequently, the WCTP-blended cementitious materials exhibit insufficient internal pore filling, larger pore sizes, higher porosity, and thus lower early-stage strength.

With the prolongation of hydration age, the strength development advantage of WCTP-blended cementitious materials becomes increasingly evident. When the WCTP replacement rate is below 20%, the 56-day compressive strength of the modified paste reaches 95% of that of the Portland cement paste specimens. Even at a 50% replacement rate, the 56-day compressive strength still reaches 65% of that demonstrated by the pure cement paste specimens. The 3-day strength of the pure cement paste represents 71% and 65% of its 28-day and 56-day strength values, respectively. For the WCTP-blended cementitious pastes with replacement rates of 10%, 20%, 30%, 40%, and 50%, their 3-day strengths account for 63% and 57%, 61% and 49%, 51% and 46%, 51% and 39%, and 44% and 34% of their respective 28-day and 56-day strengths, respectively.

These data indicate that an increasing WCTP content progressively retards the early-stage hydration process; however, it conversely accelerates the strength evolution of the WCTP-incorporated paste during the late hydration age. This phenomenon can be ascribed to the secondary hydration reactions occurring between the alumina (Al₂O₃) and silica (SiO₂) in WCTP and the calcium hydroxide generated during cement hydration. The hydration products formed in this process lead to a finer pore structure and a denser matrix, thus promoting the development of late-stage strength.

In the study by Heidari et al. (ceramic powder <75 μm), at 7 and 28 days of age, the compressive strength of cement-based specimens under a 40% ceramic powder substitution rate decreased to 0.51 and 0.59 of that of the ordinary Portland cement specimens (Heidari and Tavakoli, 2013), both lower than the strength ratios of 0.59 and 0.64 in this study. In contrast, compared with the study by Li et al. (ceramic powder 1–10 μm), at 7 days of age, the compressive strength with a 40% ceramic powder substitution rate also dropped to about 0.59 of the ordinary Portland cement specimens; however, at 28 days of age, in Li et al.'s study, the compressive strength of cement-based specimens with a 40% ceramic powder substitution rate decreased to 0.71 of that of ordinary Portland cement specimens (Li L. et al., 2025), higher than the strength ratio of 0.64 in this study. This indicates that the finer the particle size, the smaller the reduction in strength at the same substitution rate.

Pitarch et al. studied the effect of different dosages and three types of raw ceramic powders on the hydration and hardening performance of cement-based materials, where the ceramic powders were derived from ceramic tiles, sintered clay bricks, and sanitary ceramics, and all had an average diameter close to 20 μm, with 90 vol% of particles below 56 μm and 10 wt% under 1.62 μm, slightly finer than cement. The results showed that the incorporation of all three types of raw ceramic powders reduced the compressive strength of the cement matrix (Pitarch et al., 2021). This indicates that when ceramic powder is ground to a fineness comparable to or slightly finer than cement, its addition still has a reducing effect on strength. Additionally, Parashar et al. and Alani et al. used waste nano-ceramic and nano brick powders as Portland cement alternatives, respectively, and at 7 and 28 days of age, with a 10% ceramic powder substitution rate, the compressive strength of cement-based specimens even increased compared to ordinary Portland cement specimens (Parashar et al., 2022; Alani et al., 2025). In Parashar et al.'s study, it was 1.14 and 1.15 times that of ordinary Portland cement, and in Alani et al.'s study, it was 1.29 and 1.26 times that of ordinary Portland cement.

The comparison of compressive strength results under different particle size distributions indicates that the finer the ceramic powder, the higher the chemical reactivity, the more hydration products are generated, the denser the structure, and the higher the compressive strength. At the same time, this also suggests that when ceramic powder is ground to a certain fineness, its incorporation can achieve strength enhancement, albeit with increased carbon emissions and energy consumption.

3.2 Pore structure analysis

The pore size distributions curves for the hardened WCTP-blended cementitious paste specimens at the 3-day and 56-day hydration points are illustrated in Figures 3A,B.

Figure 3

Figure 3. Pore size distribution of hardened paste: [(A) 3 days of hydration; (B) 56 days of hydration], (C) Porosity of hardened paste specimens with different waste ceramic powder (WCTP) replacement rates at various hydration ages and Pore size statistics of hardened paste specimens with different waste ceramic powder (WCTP) replacement rates at various hydration ages: [(D) 3 days; (E) 56 days].

As presented in Figures 3A,B, each curve exhibits a distinct peak corresponding to the most probable pore size. Figure 3A indicates that the most probable pore size of the hardened pure cement paste is smaller than that of the waste ceramic powder (WCTP)-blended cement-based materials. With the increase in WCTP replacement rate, the most probable pore size of each curve progressively increases, with a peak concentration in the 70 nm–100 nm range.

The quantity of hydration products increases progressively with advancing curing age. In the alkaline solution produced by cement hydration, WCTP starts to participate in chemical reactions, consuming calcium hydroxide and producing calcium-silicate-hydrate gel to fill the interior of the hardening paste, continuously refining the pore size. After 56 days of hydration, the WCTP-blended cement-based materials are characterized by a larger most probable pore size compared to the pure cement hardened paste. Nevertheless, the most probable pore size of all samples decreased significantly, shifting from approximately 100 nm to the range of 40 nm–60 nm.

3.2.1 Porosity and pore size

From Figure 3C, it can be observed that at the 3-day hydration age, the WCTP-modified pastes were more porous than the pure cement paste. Furthermore, a positive correlation was observed, with porosity increasing markedly as the WCTP content rose. At low dosages (≤20%), the porosity is 1.05 and 1.14 times that of pure cement hardened paste; at high dosages (>20%), the porosity reaches 1.42, 1.46, and 1.55 times that of pure cement hardened paste.

This is because the reactivity of cement is much higher than that of ceramic powder—cement dominates the hydration reaction in the early stage. When the ceramic powder dosage is excessively high, although it can increase the local water-to-binder ratio (w/b) to facilitate hydration, it is still insufficient to compensate for the decrease in cementitious phases due to the shortage of hydration raw materials. The effect of filling and refining pores inside the samples is weaker, leading to higher porosity in the hardened paste of ceramic powder-cement-based materials.

By 56 days of curing, the hydration reaction has proceeded relatively sufficiently, producing a large amount of hydration products that fill the pore structure, giving rise to a significant reduction in porosity When the WCTP content is low (≤20%), the porosity of the hardened composite cementitious material is 1.03 times and 1.02 times that of the hardened Portland cement paste, respectively; when the WCTP content is high (>20%), the porosity is 1.09 times, 1.17 times, and 1.24 times that of the hardened Portland cement paste, respectively. The increase in porosity caused by high WCTP content is significantly improved. This may be a consequence of the pore-filling effect of the products generated by the secondary hydration of WCTP, which optimizes the pore structure.

Pores can be divided into four types on the basis of their size: gel pores (<4.5 nm), small capillary pores (4.5–50 nm), medium capillary pores (50–100 nm), and large capillary pores (>100 nm). The volume of each pore type can be calculated from the mercury intrusion volume under different pressures during the MIP test, as shown in Figures 3D,E.

As illustrated in Figures 3D,E, the porosity of the hardened sample gradually increases with the rise in ceramic powder dosage, while it decreases significantly with prolonged curing age. At 3 d, the pores in the ceramic powder-cement-based hardened paste are dominated by medium and large capillary pores, accounting for over 50% of the total pore volume—even reaching 75% at a 50% replacement rate. At this stage, ceramic powder barely participates in the hydration reaction and only plays a minor physical filling role. By 56 d, the pore volume of the hardened paste decreases remarkably for all replacement rates: a large number of large capillary pores are converted into small and medium capillary pores, which together account for over 90% of the overall pore volume. The distribution of the micro-pore structure tends to be optimized, and the structure becomes increasingly dense. This phenomenon stems from the progressive filling of larger pores by the secondary hydration products of WCTP.

3.3 Microscopic morphology analysis

The microstructures of cement-based samples with different WCTP contents at 3 d and 56 d hydration ages, obtained via scanning electron microscopy, are presented in Figure 4. Figure 4A displays the microstructure of the Portland cement specimen at 3 d. The specimen surface is generally uneven, with inherent defects such as pores and microcracks. A large number of acicular-ettringite (AFt) crystals, flaky Ca(OH)₂ crystals, and stacked flocculent hydrated C-S-H gel are observed around the pores and cracks, with partial interconnection between some C-S-H gel and Ca(OH)₂ crystals.

Figure 4

Figure 4. Micromorphology of cement paste at different ceramic powder replacement rates (3 days of curing). [(A) 0% WCTP; (B) 10% WCTP; (C) 20% WCTP; (D) 30% WCTP; (E) 40% WCTP; (F) 50% WCTP].

The hydration products are closely bonded, and a small quantity of unreacted cement particles are present. This indicates that in the early ages, the plain Portland cement paste contains sufficient reactants for hydration, leading to a high hydration rate and considerable formation of hydration products. For this reason, the cement paste achieves a certain level of strength at an early age.

Figures 4B–F illustrate the microstructures of ceramic powder-cement-based paste specimens at 3 d. It can be seen that the hydration products of the ceramic powder-incorporated pastes are almost identical to those of plain Portland cement, consisting of acicular AFt crystals, flaky Ca(OH)₂ crystals, and flocculent C-S-H gel. However, as the ceramic powder dosage increases, the quantity of hydration products decreases significantly, and the interlocking between hydration products becomes poor.

Comparing Figure 4B with Figure 4F, the hydration products and unhydrated ceramic powder particles are intertwined at the 10% replacement rate. In contrast, the hydration products at the 50% replacement rate are mostly individual flocculent C-S-H gel, with no obvious connection between them. The hydration rate of ceramic powder-cement-based pastes is significantly lower than that of plain Portland cement paste. With the increase in ceramic powder dosage, the number of unhydrated ceramic powder particles increases noticeably—at this stage, ceramic powder only plays a physical filling role in the system.

From a microstructural perspective, the increased content of non-reactive ceramic powder particles, coupled with insufficient early-stage hydration products and loose interlocking between hydration products, leads to a relatively porous microstructure of the hardened paste. Simultaneously, pores and cracks in the paste begin to proliferate and expand, with an increased formation of AFt around the pore openings. Although some hydration products are also formed within the pores, they are insufficient to fill these pores, as depicted in Figure 4E. The results of the compressive strength and MIP tests align with the microstructural evidence, which demonstrates that the ceramic powder cementitious matrix paste develops lower strength and higher porosity at early ages than the ordinary Portland cement paste.

Figure 5 presents the microstructural morphology of ordinary Portland cement paste and waste ceramic powder (WCTP)-blended cementitious paste specimens at the 56-day curing age. Compared with the specimens at the 3-day age, the hydration process of cement clinker is largely completed, and a large quantity of hydration products are formed in the system. The microstructure of the cement paste exhibits a cohesive texture.

Figure 5

Figure 5. Micromorphology of cement paste at different ceramic powder replacement rates (56 days of curing). [(A) 0% WCTP; (B) 10% WCTP; (C) 20% WCTP; (D) 30% WCTP; (E) 40% WCTP; (F) 50% WCTP].

Figure 5A illustrates the microstructure of Portland cement paste at 56 days, where the system has undergone relatively complete hydration. At this stage, C-S-H gel and crystalline Ca(OH)₂ are tightly interlocking. In the microstructure, the acicular ettringite (AFt) crystals around pore openings have decreased, and the internal pores have been filled with hydration products. Compared with the 3 d age, the pores are significantly smaller, and the microstructure is much smoother and denser. This is because with the extension of hydration age, both the surface of cement particles and the ions dissolved in the solution participate in the hydration reaction, generating a large amount of reaction products. These reaction products fill and refine the inherent defects in the system, resulting in a denser microstructure observed at 56 d.

Figures 5B–F show the microstructures of ceramic powder-cement-based pastes at 56 d. Compared with the ceramic powder-cement-based pastes at 3 d, the hydration-generated C-S-H gel forms an integrated matrix. For pastes with high ceramic powder dosages (>20%), the individual flocculent C-S-H gel disappears, and the hydration products are tightly bonded. In contrast to the plain Portland cement paste at 56 d, the crystalline Ca(OH)₂ in the ceramic powder-incorporated system is significantly reduced.

This stems from the fact that cement dominates the hydration reaction in the early stage of ceramic powder-cement-based paste hydration, producing Ca(OH)₂ and rendering the hydration environment alkaline. In the middle and late hydration stages, ceramic powder undergoes secondary hydration with Ca(OH)₂ under alkaline conditions, consuming Ca(OH)₂ to generate a large amount of C-S-H gel with a low calcium-silicon ratio. This partially compensates for the low formation of hydration products at an early age.

Compared with the microstructure of ceramic powder-cement-based pastes at 3 d, the specimens with the same ceramic powder dosage exhibit a transformation from a rough, porous, and crack-prone surface to a uniform, flat, and dense surface. This is the result of specimen hardening: the inherent surface defects are gradually filled by new hydration products, forming an integrated structure.

The mechanism underlying the formation of dense structures differs slightly between ordinary Portland cement paste and ceramic powder-modified cementitious materials. For ordinary Portland cement paste, abundant raw hydration materials (cement clinker minerals) contribute to the generation of ample hydration products, which directly densify the microstructure. For ceramic powder cement-based materials, the C-S-H gel formed during the mid-to-late stages of secondary hydration, characterized by a low silica-to-calcium ratio, serves to refine and densify the microstructure of the system.

3.4 EIS test results and analysis

3.4.1 Nyquist plot analysis

To explain the impedance characteristics of hydrating cementitious materials via EIS, the Nyquist plot represents the most widely adopted analytical approach in the existing research. In this plot, the complex impedance values, which are measured across a frequency spectrum, are represented by plotting their real and imaginary components on a single complex plane.

Figure 6A presents a typical Nyquist plot of the electrochemical system of cement-based materials, where the X-axis denotes the real component and the Y-axis represents the imaginary component of the system impedance. This plot is composed of a straight line and a semicircular arc, distributed from the low-frequency to high-frequency regions.

Figure 6

Figure 6. (A) Typical Nyquist plot for cementitious materials; Nyquist plots of hardened paste at different ages. [(B) 3 days; (C) 7 days; (D) 14 days; (E) 28 days; (F) 56 days]; (G) Photograph of Cracks in Specimen at 56 days of Age.

On the Nyquist plot, the low-frequency line corresponds to the electrode polarization effect, which is controlled by mass diffusion processes. The high-frequency semicircle is related to the bulk properties of the material, involving pore structure, porosity, and the transport of ions in the material’s pore solution.

The main characteristic parameters of the high-frequency semicircle in the Nyquist plot contain rich microstructural information of porous materials. Some important parameters are as follows:

High-frequency limiting impedance ($R_{\mathrm{\Omega }}$): Represents the real-axis value at the intersection of the high-frequency semicircle and the real axis, corresponding to the maximum frequency point on the Nyquist plot. Song (2000) ascribe this value to the resistance of the pore solution in the material’s pore system.

Diameter of the high-frequency semicircular arc ($R_{ct}$): It exhibits high sensitivity to microstructural changes and can indirectly reflect the hydration degree of cementitious materials. With the progression of the hydration process, $R_{ct}$ increases gradually.

Bulk resistance ($R$): This parameter is determined from the junction of the high-frequency semicircle and the low-frequency linear segment, and its value follows the relationship: $R=R_{\mathrm{\Omega }}+R_{ct}$.

Diffusion impedance coefficient ($\sigma$): Under quasi-Randles conditions, the low-frequency segment of the Nyquist plot presents a straight line with a slope of 1. This straight segment intersects the real axis at the point corresponding to $R_{\mathrm{\Omega }}+R_{Ct}-2{\sigma }^2{C}_d$, and the intercept at this intersection can be used to derive the value of $\sigma$. In practical operation, a tangent line with a slope of 1 can be drawn along the low-frequency segment, and the intercept of this tangent line with the real axis enables the determination of $\sigma$.

Figures 6B–F illustrates the Nyquist plots of hardened paste at various ages of curing (3d, 7d, 14d, 28d, 56d).

As illustrated in Figure 6B, at the curing age of 3 days, the bulk resistance of Portland cement paste is higher than that of all ceramic powder-modified cementitious materials. A higher ceramic powder replacement rate leads to a progressive reduction in bulk resistance. To illustrate, the bulk resistance of the control Portland cement specimen is approximately eightfold that of the specimen incorporating 50% ceramic powder. The relatively high water absorption potential of ceramic powder can partially account for this phenomenon. The decrease in early-age hydration product formation, observed under a fixed water-to-binder ratio, is attributed to the reduced content of free water in the system.

Additionally, the constituent makeup of the raw materials available for hydration is changed by the incorporation of ceramic powder. Portland cement contains a notably greater proportion of calcium oxide (CaO) than ceramic powder. CaO is a major component in the early-age hydration products of cement paste, it can generate calcium hydroxide by reacting with water. Subsequently, active components such as aluminum oxide (Al₂O₃) and silicon dioxide (SiO₂) react with Ca(OH)₂ to form calcium-silicate-hydrate gel that fills the pores among hydration products. For the cementitious system containing ceramic powder, the total CaO content is reduced, which inhibits the formation of hydration products. Eventually, these effects are reflected in a gradual decrease in bulk resistance as the ceramic powder content increases. The specific reaction equations are presented as Equations 11–13:

$\text{CaO}+{\mathrm{H}}_2\mathrm{O}=\text{Ca}{\left(\text{OH}\right)}_2(11)$

$x\text{Ca}{\left(\text{OH}\right)}_2+{\text{SiO}}_2+m_2{\mathrm{H}}_2\mathrm{O}\to x\text{CaO}·{\text{SiO}}_2·m_2{\mathrm{H}}_2\mathrm{O}(12)$

$y\text{Ca}{\left(\text{OH}\right)}_2+{\text{Al}}_2{\mathrm{O}}_3+m_2{\mathrm{H}}_2\mathrm{O}\to y\text{CaO\hspace{0.17em}}{\text{Al}}_2{\mathrm{O}}_3{\mathrm{H}}_2\mathrm{O}(13)$

A comparison of Figures 6C,D E, indicates that from 7 to 28 days, the high-frequency semicircular arc of the ceramic powder-modified cementitious paste gradually converges toward that of the Portland cement paste, demonstrating a higher volume resistivity growth rate than the latter. The early-stage hydration of Portland cement paste is largely completed within 7 days, and the cement hydration process enters a stable phase after 28 days.

However, the pattern presented in Figure 6F deviates from this general trend. The measured bulk resistance of the 56-day cement paste showed a significant increase. Upon analysis, there are two reasons for this. First, a separation occurred between the electrode and the specimen, resulting in an increased measured impedance. Second, cracks appeared in the tested cement paste specimen, which also contributed to the increase in impedance (As shown in Figure 6G).

Figure 7 presents the Nyquist plots of cement pastes with different ceramic powder replacement rates (0%, 10%, 20%, 30%, 40%, 50%). It can be observed that all cement pastes at 3 d already exhibit distinct high-frequency semicircles, but these are much smaller compared to the impedance spectra at 14 d and 28 d. With the extension of curing age, the high-frequency semicircles of both types of cement pastes continue to develop and shift to the right along the X-axis. Since bulk resistance can be understood as DC resistance to a certain extent, the continuous increase in bulk resistance indicates that the microstructure of the hardened paste becomes progressively dense, resulting in reduced electrical conductivity of the paste.

Figure 7

Figure 7. Nyquist plots of hardened paste at various replacement rates. [(A) 0% WCTP; (B) 10% WCTP; (C) 20% WCTP; (D) 30% WCTP; (E) 40% WCTP; (F) 50% WCTP].

The 7-day bulk resistance of Portland cement accounts for 82% and 77% of the 14-day and 28-day bulk resistance, respectively. This indicates that the early-stage hydration process of Portland cement is essentially completed within 7 days, and the system then enters the mid-hydration stage. Ceramic powder-cement-based materials, in contrast, demonstrate slightly delayed early hydration kinetics, and this delay becomes more prominent with the increase in ceramic powder replacement ratio.

At a 10% replacement rate, the 7-day bulk resistance of the ceramic powder-modified cementitious paste reaches 75% and 60% of the 14-day and 28-day bulk resistance, respectively. For the paste with a 50% ceramic powder replacement rate, the 7-day bulk resistance is 68% and 52% of the 14-day and 28-day bulk resistance, respectively. After 7 days, the shape of the impedance spectrum remains largely unchanged, characterized by a semicircular arc in the high-frequency region and nearly identical linear slopes in the low-frequency region. This phenomenon suggests that by approximately 7 days of hydration, the internal microstructures of both types of cement pastes have essentially stabilized. At this stage, the primary structural transformation involves the conversion of interconnected pores into isolated pores.

3.4.2 Bulk resistance analysis

To improve the comprehension of the variation pattern of bulk resistance with curing age during cement hydration, bulk resistance values measured from 3 to 28 days were extracted. Figure 8A illustrates the changes in bulk resistance at different curing ages for cementitious materials with various ceramic powder replacement rates. The curves corresponding to all replacement rates exhibit a continuous increasing trend, and a higher replacement rate is associated with a lower bulk resistance. Additionally, the growth patterns of bulk resistance with curing age are consistent across different ceramic powder replacement rates.

Figure 8

Figure 8. (A) Variation of bulk resistance with curing age and (B) Trend in bulk resistance with curing age at 0% and 20% ceramic powder replacement rates.

Two sets of data corresponding to ceramic powder-modified cementitious pastes with 0% and 20% replacement rates were selected for analysis. Linear fitting was conducted on the data obtained from 3 to 7 days and 7–28 days, respectively. The fitting curves are shown in Figure 8B.

As can be seen in Figure 8B that the increasing trend of bulk resistance in the first 7 days of cement hydration is significantly more pronounced than that after 7 days. For the fitting line of the 0% replacement rate, the slopes before and after 7 days are 164 and 14.9, respectively. For the 20% replacement rate, the slopes of the fitting line are 102 and 33.6 before and after 7 days. This indicates that the variation law of bulk resistance with curing age can be divided into two stages, which correspond to the early and middle stages of cement hydration. The slope of the regression line for the 20% replacement rate is lower than that of the 0% replacement rate before 7 days, but higher after 7 days. This phenomenon reflects the secondary hydration effect of ceramic powder during the hydration process of the entire system. Therefore, measuring the bulk resistance of cement paste can be used as a method to evaluate the degree of cement hydration.

3.5 Validation of the Kramers-Kronig relations for impedance data

Three fundamental assumptions must be satisfied during impedance testing: the tested system must comply with the conditions of causality, linearity, and stability. In 1945, Bode applied the dispersion relation derived by Kramers and Kronig based on the causality principle to linear circuits, and this relation was named the Kramers-Kronig relation (commonly abbreviated as the K-K relation).

Mathematically, the K-K relation can be defined as follows: The analyticity of a complex function in the upper half-plane implies that its real part is interrelated with the imaginary part through a specific mathematical formalism. When the tested system meets the conditions of causality, linearity, and stability, the real and imaginary components of its complex-valued function can be linked through the Hilbert transform. The complex impedance is given by: $Z\left(\omega \right)=Z^{\prime }\left(\omega \right)+i{Z}^{″}\left(\omega \right)$, where $Z^{\prime }\left(\omega \right)$ is the real axis component of the impedance data and $Z^{″}\left(\omega \right)$ is the imaginary axis component. The expression for the Kramers-Kronig (K-K) relations is as Equations 14–17:

$Z^{\prime }\left(w\right)-Z^{\prime }\left(\infty \right)=\left(\frac{2}{\pi }\right){\int }_0^{\infty }\frac{x{Z}^{″}\left(x\right)-\omega Z^{″}\left(\omega \right)}{x^2-{\omega }^2}dx(14)$

$Z^{\prime }\left(\omega \right)-Z^{\prime }\left(0\right)=\left(\frac{2\omega }{\pi }\right){\int }_0^{\infty }\left[\frac{\omega }{x}{Z}^{″}\left(x\right)-Z^{″}\left(\omega \right)\right]\frac{1}{x^2-{\omega }^2}dx(15)$

$Z^{″}\left(\omega \right)=-\left(\frac{2\omega }{\pi }\right){\int }_0^{\infty }\frac{Z^{\prime }\left(x\right)-Z^{\prime }\left(\omega \right)}{x^2-{\omega }^2}dx(16)$

$\theta \left(\omega \right)=\left(\frac{2\omega }{\pi }\right){\int }_0^{\infty }\frac{\ln \left|Z\left(x\right)\right|}{x^2-{\omega }^2}dx(17)$

To verify the validity of the impedance data obtained from the EIS tests, all data were subjected to Kramers-Kronig (K-K) validation using ZView software. Figure 9 shows the K-K validation results for six sets of impedance data, where “M” represents the measured data and “K” represents the K-K analyzed data.

Figure 9

Figure 9. K-K analysis of measured impedance data. [(A) 3 d, 0% WCTP; (B) 3 d, 20% WCTP; (C) 7 d, 40% WCTP; (D) 7 d, 50% WCTP; (E) 28 d, 10% WCTP; (F) 28 d, 30% WCTP].

It can be observed that the data derived from K-K verification are in excellent agreement with the measured data. Meanwhile, residual errors were introduced to quantitatively evaluate the discrepancy between the K-K verified data and the measured data. The formula for residual error is given in Equation 18, where Z_{r, exp} and Z_{i, exp} are the real and imaginary parts of the measured impedance data, respectively, and Z_exp and Z_K-Ktran represent the measured data and the corresponding K-K verified data. Figure 10 shows the residual error plots of the aforementioned six groups of K-K verification: the horizontal axis represents the test frequency, the vertical axis denotes the residual error between the measured data and the regression data, $ReZ$ stands for the residual error of the real part of impedance, and $ImZ$ denotes the residual error of the imaginary part.

$\Delta Z\left(\omega \right)=\frac{Z_{\mathit{exp}}\left(\omega \right)-Z_{K-Ktran}\left(\omega \right)}{\sqrt{Z_{r,\mathit{exp}}{\left(\omega \right)}^2+Z_{i,\mathit{exp}}{\left(\omega \right)}^2}}(18)$

Figure 10

Figure 10. Residual values from Kramers-Kronig verification for impedance tests. [(A) 3 d, 0% WCTP; (B) 3 d, 20% WCTP; (C) 7 d, 40% WCTP; (D) 7 d, 50% WCTP; (E) 28 d, 10% WCTP; (F) 28 d, 30% WCTP].

As illustrated in Figure 10, both the real and imaginary part errors from the K-K verification are extremely small, almost all controlled within 3%. This indicates that the impedance data measured in this experiment are accurate and valid, and can reflect the changes in impedance characteristics of ceramic powder-cement-based materials. Only impedance data that satisfy the Kramers-Kronig (K-K) relations are considered suitable for equivalent circuit fitting.

3.6 Selection of equivalent circuit

Generally, there are three conduction paths within cementitious materials: continuous paths formed by interconnected pores or microcracks; discontinuous paths composed of isolated pores partially blocked by cementitious materials and hydration products; and “insulating” paths formed by the bulk matrix of the specimen (Song, 2000). The equivalent circuit constructed by paralleling these three paths is presented in Figure 11A below.

Figure 11

Figure 11. (A) Equivalent Circuit of AC conduction paths in concrete and (B) Equivalent Circuit adopted in this study.

According to basic circuit theory, each resistance-capacitance parallel combination in a circuit corresponds to a semicircular arc in the Nyquist plot. However, in the actual impedance testing of cementitious materials, the Nyquist plot does not exhibit two distinct semicircular arcs. To account for the diffusion resistance of the porous medium, the equivalent circuit depicted in Figure 11B was adopted in this study, which is an extension of the aforementioned model. In this model, $W_0$ represents the diffusion resistance of the hardened paste, $R_s$ denotes the electrolyte resistance of the pore solution, $CP{E}_P$ is the constant phase element describing the polarization effect inside the hardened paste, and $R_{ct}$ represents the charge transfer resistance of free charges in the paste.

3.7 Analysis of equivalent circuit model fitting results

The internal microstructural changes of ceramic powder-modified cementitious pastes are analyzed using two key parameters, namely $R_s$ and $R_{ct}$. $R_s$ corresponds to the electrolyte resistance of the pore solution in hardened paste, and negatively correlates with the porosity and ion concentration within the pore solution (Song et al., 2018). $R_{ct}$ has a dual significance: it represents the charge transfer resistance of free charges while simultaneously reflecting the progression of the hardening reaction in hardened paste (Dong et al., 2017).

Data fitting for the aforementioned equivalent circuit model was conducted using the software ZView. The fitting results are presented in Figure 12, where “M” represents the measured experimental data and “C” denotes the values fitted by the software.

Figure 12

Figure 12. Fitted Nyquist plots for different curing age groups. [(A) 3 days; (B) 7 days; (C) 14 days; (D) 28 days; (E) 56 days].

3.7.1 Pore solution electrolyte resistance (${\mathit{R}}_{\mathit{s}}$)

The development of the hardened paste’s pore solution electrolyte resistance ($R_s$) over various curing ages is illustrated in Figure 13A.

Figure 13

Figure 13. Variation with curing age at different ceramic powder replacement rates [(A) pore solution electrolyte resistance ($R_s$); (B) charge transfer resistance ($R_{ct}$)] and (C) Fitting results of charge transfer resistance ($R_{ct}$) and compressive strength.

The experimental results indicate that the $R_s$ value of hardened paste grows with increasing curing age. This phenomenon is the result of the combined effects of porosity and pore solution resistivity. Increasing the amount of ceramic powder will increase the porosity, as shown in Figure 3C. The incorporation of ceramic powder lowers the pH of the porous solution. The literature (Chen et al., 2021) shows that the pH value of the pore solution of silicate cement hardening paste is about 13, that is, [OH⁻] is 0.10 mol/L at 90 days of age. When the substitution rate of ceramic powder was 30%, the pH value of the composite cementitious material hardened paste pore solution was about 12.6, that is, [OH⁻] is 0.04 mol/L. The main anion in the pore solution is OH⁻, followed by SO₄^2- (its concentration is much lower than OH⁻ because sulfur is present in the form of calcium sulfoaluminate hydrate in the hardened paste, not SO₄^2-). From the perspective of anion and cation balance, the concentration of cations will also decrease accordingly. As the ion concentration decreases, the mobility of the ions increases. Considering the formula u = q/(6πηr) for calculating ion mobility, q is the charge carried by the ion, η is the viscosity of the solution, and r is the radius of the ion, that is, the ion mobility is inversely proportional to the viscosity of the solution. Since the pore solution of cement-based materials is a dilute solution, the decrease in ion concentration will not cause a large increase in mobility. In other words, the incorporation of ceramic powder reduces the conductivity of the pore solution and thus increases the resistivity.

As illustrated in Figure 13A, the $R_s$ value increases significantly with curing age. For hardened cement paste at 3 d, the $R_s$ value of all samples is comparatively small. This is because the Portland cement paste is in the early stage of hardening, with numerous large-diameter pores existing inside the entire system. Subsequently, between 3 d and 14 d, the $R_s$ value of all hardened cement pastes increases rapidly. After 14 d, as hydration proceeds, the $R_s$ value enters a stable and slow growth stage. This suggests that the chemical processes within the hardened cement paste proceed continuously, reacting with ions in the pore solution to produce hydration products The connected pores within the system are filled, the connectivity paths of the pores become more tortuous, and the porosity decreases.

For hardened cement pastes of the same age, the $R_s$ value decreases with the increase in ceramic powder dosage, and this trend is particularly pronounced at 3 d. The $R_s$ value of plain cement paste is three times that of hardened cement paste with 50% ceramic powder replacement, indicating that the porosity of the high-dosage ceramic powder-cement-based paste is significantly higher than that of Portland cement paste at this stage. After 56 d of curing, the $R_s$ value of all hardened cement pastes has increased significantly, and the gap in $R_s$ values among specimens with different replacement rates has gradually narrowed. This indicates that the porosity inside all hardened cement pastes has decreased, which is consistent with the porosity variation law obtained from mercury intrusion porosimetry tests.

When considering the 56-day period as the entire hydration process of the Portland cement paste, the $R_s$ growth of the pure cement paste exhibits a distinct inflection point at 7 days, with the $R_s$ value at 7 days already exceeding 50% of the total $R_s$ increase during the entire hydration process. For the ceramic powder-modified cementitious paste, the inflection point of $R_s$ growth appears at 14 days. This is mainly due to the minimal participation of ceramic powder in chemical reactions during the early hydration stage, leading to insufficient formation of hydration products.

3.7.2 Charge transfer resistance (${\mathit{R}}_{\mathit{c}\mathit{t}}$)

Figure 13B illustrates the charge transfer resistance ($R_{ct}$) of the hardened specimens, which contain varying ceramic powder contents, at different curing ages.

The charge transfer resistance of hardened paste develops with increasing curing time. For specimens with a given curing age, the charge transfer resistance decreases as the ceramic powder content increases. This phenomenon arises from the fact that a higher ceramic powder content reduces the cement concentration in the paste. Although an increased effective water-to-binder ratio is conducive to cement hydration, the overall reduction in cement proportion results in a decrease in early-stage hydration products. Consequently, the charge transfer resistance is reduced at higher WCTP dosages.

This reveals that a continuous capillary pore structure has formed inside the hardened paste, but the charge transfer resistance is still relatively low. Between hydration ages of 3 and 7 days, the charge transfer resistance of the hardened cement paste increases rapidly. This is a consequence of, as hydration progresses, the amount of hydration products inside the hardened paste gradually increases, filling the pores, reducing pore size and porosity, and forming a denser microstructure. As a result, the charge transfer paths become more tortuous, leading to an increase in resistance. From 7 to 14 days, the growth rate of the charge transfer resistance of the WCTP-cement paste is faster than that of ordinary cement paste. The WCTP undergoes secondary hydration, generating additional hydration products that further fill the micropores and enhance the density of the paste.

The charge transfer resistance is very sensitive to the microstructural changes of the hardened paste and can reflect its compactness. As the curing time increases, the hardening reaction of the WCTP -cement paste continues. Hydration products are continuously formed and fill the internal pores, optimizing the pore structure, improving the specimen’s density, and enhancing compressive strength. In EIS tests, this is manifested as a more complex pathway for free charge transfer within the paste, leading to increased resistance, which ultimately results in an increase in charge transfer resistance with age, showing a similar trend between Figures 13B, 2.

Comparing $R_{ct}$ with compressive strength shows a clear linear correlation between the two, as shown in Figure 13C. This indicates that both $R_{ct}$ and compressive strength are closely related to the development of the microstructure during hardening process, confirming an intrinsic correlation between them.

4 Conclusion

This work explored the effects of using waste ceramic tiles powder as a mineral admixture on the strength development and structural formation in cementitious systems. Through cross-validation with multi-scale experimental methods, the following conclusions were drawn:

1. The addition of WCTP reduces the compressive strength of cement-based materials. At an age of 3 days, the higher the WCTP dosage, the more significant the decrease in the compressive strength. As the curing time increases, the strength development of WCTP-incorporated cement-based materials becomes remarkably noticeable. At WCTP contents up to 20%, the 56-day compressive strength can reach 95% of that observed in ordinary Portland cement (OPC) specimens. From the perspective of strength performance, a 20% replacement rate is considered the optimal WCTP content.

2. Both the most probable pore size and the porosity of the cement-based materials are increased by WCTP dosage, with this effect being more significant in the early stage. At the 56-day curing age, the WCTP-modified cement-based materials showed substantial pore structure refinement, as indicated by the significant reduction in most probable pore size and porosity. As the hydration reaction advances, substantial macropores within the hardened paste are gradually transformed into medium and small capillary pores, resulting in an optimized pore structure distribution.

3. Microscopic test results show that in the early stages, WCTP reduces the hydration rate, and the WCTP remains largely inert and functions primarily as a physical filler. In the subsequent middle and late stages, it undergoes secondary hydration reactions. The resulting hydration products fill the internal pore structure, leading to a refined microstructure of the cement paste.

4. The hydration progress of cement-based materials can be evaluated via the bulk resistance of the hardened paste, and the kinetics of the hydration reaction are reflected in the growth rate of this resistance. The 7-day bulk resistance of OPC accounts for 82% and 77% of the 14-day and 28-day bulk resistance, respectively, indicating that the early hydration process of OPC is basically completed within 7 days, and then enters the middle hydration stage. In contrast, the early stage of hydration of WCTP-modified cement-based materials is slightly delayed, and this delay effect is amplified at higher WCTP replacement rates.

5. There is a good linear correlation between the charge transfer resistance ($R_{ct}$) and compressive strength, and ($R_{ct}$) can effectively reflect the compressive strength development of WCTP-modified cement-based pastes.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

Author contributions

SW: Methodology, Resources, Writing – original draft. QH: Writing – review and editing, Investigation. YF: Resources, Writing – review and editing, Data curation, Investigation. LW: Visualization, Formal Analysis, Project administration, Writing – review and editing, Investigation, Funding acquisition, Conceptualization.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This research was funded by the financial support from the Guiding Funds of Central Government for Supporting the Development of the Local Science and Technology (Grant No. 236Z3810G) and the Natural Science Foundation of Hebei Province-Beijing-Tianjin-Hebei Special Project (Grant No. E2021210136).

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Adesina, A. (2020). Recent advances in the concrete industry to reduce its carbon dioxide emissions. Environ. Challenges 1, 100004. doi:10.1016/j.envc.2020.100004

CrossRef Full Text | Google Scholar

Alani, A. A., Hama, S. M., and Jahami, A. (2025). Influence of a nano-dispersed additive from waste brick powder on the structure formation of concrete and heat dissipation characteristic. Constr. Build. Mater. 481, 141561. doi:10.1016/j.conbuildmat.2025.141561

CrossRef Full Text | Google Scholar

Awoyera, P. O., Dawson, A. R., Thom, N. H., and Akinmusuru, J. O. (2017). Suitability of mortars produced using laterite and ceramic wastes: mechanical and microscale analysis. Constr. Build. Mater. 148, 195–203. doi:10.1016/j.conbuildmat.2017.05.031

CrossRef Full Text | Google Scholar

Awoyera, P. O., Olalusi, O. B., and Babagbale, D. P. (2021). Production of lightweight mortar using recycled waste papers and pulverized ceramics: mechanical and microscale properties. J. Build. Eng. 39, 102233. doi:10.1016/j.jobe.2021.102233

CrossRef Full Text | Google Scholar

Baraldi, L. (2025). World production and consumption of ceramic tiles. Ceram. World Rev. 35, 40–54.

Google Scholar

Bayraktar, O. Y., Tunçtan, M., Benli, A., Türkel, İ., Kızılay, G., and Kaplan, G. (2024). A study on sustainable foam concrete with waste polyester and ceramic powder: properties and durability. J. Build. Eng. 95, 110253. doi:10.1016/j.jobe.2024.110253

CrossRef Full Text | Google Scholar

Bommisetty, J., Keertan, T. S., Ravitheja, A., and Mahendra, K. (2019). Effect of waste ceramic tiles as a partial replacement of aggregates in concrete. 1st Int. Conf. Manuf. Mater. Sci. Eng. 19, 875–877. doi:10.1016/j.matpr.2019.08.230

CrossRef Full Text | Google Scholar

Chen, M., Zhang, R., Zhao, W., and Xiao, J. (2021). Effect of curing temperature on hydration of ceramic powder-cement composite colloid. J. East China Jiaot. Universtty 38, 48–55. doi:10.16749/j.cnki.jecjtu.20211026.004

CrossRef Full Text | Google Scholar

Chen, R., Zhang, Z., and Shi, C. (2023). Effects of blending waste ceramic powder in Alkali activated slag on early reaction, strength of hardened products and shrinkage property. Mater. Rep. 37, 22030249. doi:10.11896/cldb.22030249

CrossRef Full Text | Google Scholar

Chi, L., Wang, Z., Lu, S., Zhao, D., and Yao, Y. (2019). Development of mathematical models for predicting the compressive strength and hydration process using the EIS impedance of cementitious materials. Constr. Build. Mater. 208, 659–668. doi:10.1016/j.conbuildmat.2019.03.056

CrossRef Full Text | Google Scholar

de Matos, P. R., Jiao, D., Roberti, F., Pelisser, F., and Gleize, P. J. P. (2020). Rheological and hydration behaviour of cement pastes containing porcelain polishing residue and different water-reducing admixtures. Constr. Build. Mater. 262, 120850. doi:10.1016/j.conbuildmat.2020.120850

CrossRef Full Text | Google Scholar

de Matos, P. R., Sakata, R. D., Onghero, L., Uliano, V. G., de Brito, J., Campos, C. E. M., et al. (2021). Utilization of ceramic tile demolition waste as supplementary cementitious material: an early-age investigation. J. Build. Eng. 38, 102187. doi:10.1016/j.jobe.2021.102187

CrossRef Full Text | Google Scholar

Doleželová, M., Krejsová, J., Scheinherrová, L., Keppert, M., and Vimmrová, A. (2022). Investigation of environmentally friendly gypsum based composites with improved water resistance. J. Clean. Prod. 370, 133278. doi:10.1016/j.jclepro.2022.133278

CrossRef Full Text | Google Scholar

Dong, B., Qiu, Q., Gu, Z., Xiang, J., Huang, C., Fang, Y., et al. (2016). Characterization of carbonation behavior of fly ash blended cement materials by the electrochemical impedance spectroscopy method. Cem. Concr. Compos. 65, 118–127. doi:10.1016/j.cemconcomp.2015.10.006

CrossRef Full Text | Google Scholar

Dong, B., Li, G., Zhang, J., Liu, Y., Xing, F., and Hong, S. (2017). Non-destructive tracing on hydration feature of slag blended cement with electrochemical method. Constr. Build. Mater. 149, 467–473. doi:10.1016/j.conbuildmat.2017.05.042

CrossRef Full Text | Google Scholar

Elemam, W. E., Agwa, I. S., and Tahwia, A. M. (2023). Reusing ceramic waste as a fine aggregate and supplemental cementitious material in the manufacture of sustainable concrete. Buildings 13, 2726. doi:10.3390/buildings13112726

CrossRef Full Text | Google Scholar

Fita, I. C., Cruz, J. M., Bouzón, N., Borrachero, M. V., and Payá, J. (2022). Monitoring the pozzolanic effect of fly ash in blended OPC mortars by electrical impedance spectroscopy. Constr. Build. Mater. 314, 125632. doi:10.1016/j.conbuildmat.2021.125632

CrossRef Full Text | Google Scholar

Gonzalez-Corominas, A., and Etxeberria, M. (2014). Properties of high performance concrete made with recycled fine ceramic and coarse mixed aggregates. Constr. Build. Mater. 68, 618–626. doi:10.1016/j.conbuildmat.2014.07.016

CrossRef Full Text | Google Scholar

Heidari, A., and Tavakoli, D. (2013). A study of the mechanical properties of ground ceramic powder concrete incorporating nano-SiO2 particles. Constr. Build. Mater. 38, 255–264. doi:10.1016/j.conbuildmat.2012.07.110

CrossRef Full Text | Google Scholar

Hoppe Filho, J., Pires, C. A. O., Leite, O. D., Garcez, M. R., and Medeiros, M. H. F. (2021). Red ceramic waste as supplementary cementitious material: microstructure and mechanical properties. Constr. Build. Mater. 296, 123653. doi:10.1016/j.conbuildmat.2021.123653

CrossRef Full Text | Google Scholar

Jyosyula, S. K. R., Surana, S., and Raju, S. (2020). Role of lightweight materials of construction on carbon dioxide emission of a reinforced concrete building. First Int. Conf. Adv. Lightweight Mater. Struct. 27, 984–990. doi:10.1016/j.matpr.2020.01.294

CrossRef Full Text | Google Scholar

Lasseuguette, E., Burns, S., Simmons, D., Francis, E., Chai, H. K., Koutsos, V., et al. (2019). Chemical, microstructural and mechanical properties of ceramic waste blended cementitious systems. J. Clean. Prod. 211, 1228–1238. doi:10.1016/j.jclepro.2018.11.240

CrossRef Full Text | Google Scholar

Li, L., Liu, W., You, Q., Chen, M., and Zeng, Q. (2020). Waste ceramic powder as a pozzolanic supplementary filler of cement for developing sustainable building materials. J. Clean. Prod. 259, 120853. doi:10.1016/j.jclepro.2020.120853

CrossRef Full Text | Google Scholar

Li, X., Tian, B., Lv, Y., Zhang, C., Jiang, D., Xu, J., et al. (2023). Ultra-high performance concrete prepared with ceramic polishing waste: hydration, microstructure and mechanical property. Powder Technol. 424, 118567. doi:10.1016/j.powtec.2023.118567

CrossRef Full Text | Google Scholar

Li, Q., Fan, Y., Qi, Y., Zhang, S., and Shah, S. P. (2024). Effect of nano-metakaolin on the chloride diffusion resistance of cement mortar with addition of fly ash. J. Build. Eng. 88, 109171. doi:10.1016/j.jobe.2024.109171

CrossRef Full Text | Google Scholar

Li, L., Liu, W., Zeng, Q., and Zhou, C. (2025a). Sustainable reuse of household ceramic wastes as a substitute of cement and natural sand in construction materials: mechanical, transport, and microstructural properties. Case Stud. Constr. Mater. 22, e04708. doi:10.1016/j.cscm.2025.e04708

CrossRef Full Text | Google Scholar

Li, Q., Fan, Y., Zhao, Y., Chen, H., and Shah, S. P. (2025b). AC impedance spectroscopy interpretation of the hydration behavior for cement mortar containing fly ash and nano-metakaolin. Constr. Build. Mater. 468, 140436. doi:10.1016/j.conbuildmat.2025.140436

CrossRef Full Text | Google Scholar

Liu, X., Liu, L., Lyu, K., Li, T., Zhao, P., Liu, R., et al. (2022). Enhanced early hydration and mechanical properties of cement-based materials with recycled concrete powder modified by nano-silica. J. Build. Eng. 50, 104175. doi:10.1016/j.jobe.2022.104175

CrossRef Full Text | Google Scholar

Medina, C., Sánchez de Rojas, M. I., Thomas, C., Polanco, J. A., and Frías, M. (2016). Durability of recycled concrete made with recycled ceramic sanitary ware aggregate. Inter-indicator relationships. Constr. Build. Mater. 105, 480–486. doi:10.1016/j.conbuildmat.2015.12.176

CrossRef Full Text | Google Scholar

Parashar, A. K., Sharma, P., and Sharma, N. (2022). An investigation on properties of concrete with the adding of waste of ceramic and micro silica. Mater. Today Proc. 62, 4036–4040. doi:10.1016/j.matpr.2022.04.603

CrossRef Full Text | Google Scholar

Park, J., Song, H., and Choi, H. (2022). Estimation of water-to-cement ratio in cementitious materials using electrochemical impedance spectroscopy and artificial neural networks. Constr. Build. Mater. 350, 128843. doi:10.1016/j.conbuildmat.2022.128843

CrossRef Full Text | Google Scholar

Pitarch, A. M., Reig, L., Tomás, A. E., Forcada, G., Soriano, L., Borrachero, M. V., et al. (2021). Pozzolanic activity of tiles, bricks and ceramic sanitary-ware in eco-friendly portland blended cements. J. Clean. Prod. 279, 123713. doi:10.1016/j.jclepro.2020.123713

CrossRef Full Text | Google Scholar

Raveendran, N., and Vasugi, K. (2024). Synergistic effect of nano silica and metakaolin on mechanical and microstructural properties of concrete: an approach of response surface methodology. Case Stud. Constr. Mater. 20, e03196. doi:10.1016/j.cscm.2024.e03196

CrossRef Full Text | Google Scholar

Ren, Y., Tomoyose, A., and Maruyama, I. (2025). Long-term hydration and pozzolanic reactivity of volcanic glass powder in blended cement systems: microstructural evolution and compressive strength development. Constr. Build. Mater. 494, 143331. doi:10.1016/j.conbuildmat.2025.143331

CrossRef Full Text | Google Scholar

Robalo, K., Costa, H., do Carmo, R., and Júlio, E. (2021). Experimental development of low cement content and recycled construction and demolition waste aggregates concrete. Constr. Build. Mater. 273, 121680. doi:10.1016/j.conbuildmat.2020.121680

CrossRef Full Text | Google Scholar

Samadi, M., Huseien, G. F., Mohammadhosseini, H., Lee, H. S., Abdul Shukor Lim, N. H., Tahir, M. M., et al. (2020). Waste ceramic as low cost and eco-friendly materials in the production of sustainable mortars. J. Clean. Prod. 266, 121825. doi:10.1016/j.jclepro.2020.121825

CrossRef Full Text | Google Scholar

Song, G. (2000). Equivalent circuit model for AC electrochemical impedance spectroscopy of concrete. Cem. Concr. Res. 30, 1723–1730. doi:10.1016/S0008-8846(00)00400-2

CrossRef Full Text | Google Scholar

Song, S., Jiang, L., Jiang, S., Yan, X., and Xu, N. (2018). The mechanical properties and electrochemical behavior of cement paste containing nano-MgO at different curing temperature. Constr. Build. Mater. 164, 663–671. doi:10.1016/j.conbuildmat.2018.01.011

CrossRef Full Text | Google Scholar

Sun, H., Ren, Z., Memon, S. A., Zhao, D., Zhang, X., Li, D., et al. (2017). Investigating drying behavior of cement mortar through electrochemical impedance spectroscopy analysis. Constr. Build. Mater. 135, 361–368. doi:10.1016/j.conbuildmat.2016.12.196

CrossRef Full Text | Google Scholar

Sun, Y., Wang, K. Q., and Lee, H. S. (2021). Prediction of compressive strength development for blended cement mortar considering fly ash fineness and replacement ratio. Constr. Build. Mater. 271, 121532. doi:10.1016/j.conbuildmat.2020.121532

CrossRef Full Text | Google Scholar

Tokareva, A., and Waldmann, D. (2025). Durability of cement mortars containing fine demolition wastes as supplementary cementitious materials. Constr. Build. Mater. 477, 141316. doi:10.1016/j.conbuildmat.2025.141316

CrossRef Full Text | Google Scholar

Yang, H. M., Zhang, S. M., Wang, L., Chen, P., Shao, D. K., Tang, S. W., et al. (2022). High-ferrite portland cement with slag: hydration, microstructure, and resistance to sulfate attack at elevated temperature. Cem. Concr. Compos. 130, 104560. doi:10.1016/j.cemconcomp.2022.104560

CrossRef Full Text | Google Scholar

Zajac, M., Skocek, J., Durdzinski, P., Bullerjahn, F., Skibsted, J., and Ben Haha, M. (2020). Effect of carbonated cement paste on composite cement hydration and performance. Cem. Concr. Res. 134, 106090. doi:10.1016/j.cemconres.2020.106090

CrossRef Full Text | Google Scholar

Zajac, M., Irbe, L., Bullerjahn, F., Hilbig, H., and Ben Haha, M. (2022). Mechanisms of carbonation hydration hardening in portland cements. Cem. Concr. Res. 152, 106687. doi:10.1016/j.cemconres.2021.106687

CrossRef Full Text | Google Scholar

Zareei, S. A., Ameri, F., Bahrami, N., Shoaei, P., Musaeei, H. R., and Nurian, F. (2019). Green high strength concrete containing recycled waste ceramic aggregates and waste carpet fibers: mechanical, durability, and microstructural properties. J. Build. Eng. 26, 100914. doi:10.1016/j.jobe.2019.100914

CrossRef Full Text | Google Scholar

Zhao, J., Yan, C., Liu, S., Zhang, J., Li, S., and Yan, Y. (2020). Effect of solid waste ceramic on uniaxial tensile properties and thin plate bending properties of polyvinyl alcohol engineered cementitious composite. J. Clean. Prod. 268, 122329. doi:10.1016/j.jclepro.2020.122329

CrossRef Full Text | Google Scholar