A critical review of carbon materials engineered electrically conductive cement concrete and its potential applications

ABSTRACT Carbon materials engineered electrically conductive cement concrete (ECCC) is typically prepared by directly adding carbon-based conductive filler into the cement matrix and then mixing cement with aggregates. With such a strategy, ECCC possesses a high conductivity and strain/stress sensitivity and thus can be used for snow and ice melting, ohmic heating, cathodic protection system, electromagnetic shielding, structural health monitoring, and traffic detection. This paper aims to provide a systematic review on the development and applications of ECCC, especially the progress made in the past decade (from 2012 to 2022). The composition and manufacture of ECCC are first introduced. Then, the electrical performance of ECCC and its potential applications are reviewed. Finally, the remaining challenges for future work are discussed. GRAPHICAL ABSTRACT

Although carbon conductive fillers show great potential for developing ECCC, their practical applications are still limited due to the relatively poor dispersion quality and the technical bottlenecks between ECCC's mechanical and electrical properties [11,57,58]. In addition, the conductive mechanism of ECCC in practical applications is still unclear [58,59]. Therefore, there is an urgent need to review the current literature related to the carbon conductive fillers modified ECCC and to accelerate the practical applications of ECCC. It is believed that these gaps are mainly due to the failure to address the following issues: i) how to achieve high-quality dispersion of the carbon fillers in the concrete matrix and quantitative evaluation of their distribution; ii) the conductive mechanisms of conductive fillers inside the multi-scale, multi-interface concrete are unclear; and iii) how to cost-effectively and efficiently develop smart concrete that can be used in practical applications, rather than smart paste or mortar. Correspondingly, this review aims to provide a comprehensive summary of the development and applications of ECCC, especially the progress made in the past decade (from 2012 to 2022). The composition and manufacture of ECCC are first introduced. Then, the electrical performance of ECCC and potential applications are reviewed. Finally, the remaining challenges for future work are discussed. and mix ratio parameters can significantly affect the performance of concrete [47,[64][65][66]. This section systematically introduces the principles of ECCC, focusing on its composition, the dispersion of conductive fillers, and the preparation process.

Cement matrix
Generally, conductive fillers are directly mixed with cement, which is then mixed with aggregates to prepare ECCC [14,15,27,67]. In such a system, all kinds of cement can be adopted as the matrix for developing ECCC [19,21,54,68], such as Portland cement (PC), ordinary Portland cement (OPC), and alkali-activated slag cement. OPC is the most used binder, similar to developing ordinary cement concrete [20,55,69]. Unless otherwise stated, ECCC in this review is made of OPC.
Although cement matrix is close to an insulating material, some parameters, such as mixing ratios (e.g. water-cement ratio and cement type) and curing regimes, could influence the microstructure of concrete, affecting the migration of ions (i.e. ionic conductivity). The distribution of conductive fillers is crucial to the conductivity of concrete, which will be discussed in detail in the next section.

Carbon-based conductive fillers
The conductivity of concrete can be changed by directly adding conductive fillers (this review focuses on carbon-based fillers) [70][71][72]. The key to developing ECCC is constructing a conductive network within the concrete [73,74]. Commonly used carbon-based conductive fillers are available in different sizes, shapes, types, and conductivity. Among them, CF is favorable in fabricating ECCC because it can form conductive paths more efficiently [18,19]. However, the high cost and highly challenging dispersion process limit its practical applications.
CF can also enhance concrete's mechanical and durability properties by inhibiting the development and expansion of microcracks [19,75,76]. Considering that carbon is the most abundant element in nature and that carbon-based materials have achieved industrialization, which provides an excellent opportunity to use carbon-based materials to develop higher-performance and multifunctional concrete. According to the available literature [13,23,[77][78][79][80][81], the admixing carbon-based fillers can contribute to cement hydration and generate more calcium silicate hydrate (C-S-H) gels, improving the bonding between the cement matrix and aggregate. Additionally, they can also provide concrete with high conductivity, favorable EMI, and piezoresistivity.

Carbon fiber
Typically, CF has a carbon content of over 92 wt%, a conductivity of 10 −2 -10 −4 Ω·cm, a high Young's modulus of 300 GPa, and tensile strength of 200-3500 MPa [82,87,88]. Because of the outstanding electrical properties, excellent mechanical strengths, and high aspect ratio of CF (as shown in Figure 1a) [89,90], CF-modified concrete can achieve high conductivity and demonstrate self-sensing and self-heating abilities [91][92][93][94]. CF-reinforced concrete has desirable structural and electrical properties, making it structurally and functionally desirable in terms of acceptability. Polyacrylonitrile-based CF exhibits higher conductivity than bitumen-based CF, making it an ideal candidate for developing ECCC.

Carbon black
As a typical one-dimensional (0D) carbon nanomaterial, the admixing CB can decrease the resistivity of concrete [95,96]. Although CB has a conductivity of 0.1-2.3 Ω·cm, the low aspect ratio makes it less effective in increasing the conductivity of concrete compared to fibrous carbon-based fillers (e.g. CF, CNT, and CNF). However, CB-based ECCC exhibits higher signal stability than that fiber materials-based ECCC. Note that CB has an inert nature surface, and the interface between CB and the matrix is very weak (Figure 1b), which will seriously affect the mechanical properties of ECCC.

Graphite powder
GP is typically produced from graphite and has a resistivity of~10 −4 Ω·cm [97,98], which makes it widely used as a conductive filler for developing ECCC. Note that excellent conductivity of concrete can only be achieved with a high concentration of GP. Unfortunately, concrete's flowability and mechanical strengths would be severely damaged in this situation, mainly due to the high concentration of CB needing a large amount of water to wet their surface. Therefore, similar to CB, combining GP and CF is an ideal choice to develop ECCC.

Carbon nanotube
CNT is a tube structure made of single-layer graphene (see Figure 1c) [99,100], which has a high Young's modulus of 1,000 GPa and tensile strength of 200 GPa [84]. Typically, it has an electrical resistivity of~10 −2 Ω·cm, and thus it is regarded as an excellent filler for reducing the electrical resistance of concrete, and it is widely used to reinforce concrete due to its high availability and high reinforcement effect [101]. However, the CNT-based Figure 1. Carbon-based materials used for preparing ECCC: (a) a typical procedure for enhancing CF [82]; (b) the interface between CB and matrix [83]; (c) simulation of CNT structure [84]; (d) the microstructure of CNF [85]; and (e) the microstructure of GNP@CNT hybrid and the conductive network of composite [86].
concrete's performance is primarily related to the distribution of CNT in the concrete matrix. Therefore, future studies should be conducted to increase the dispersion quality of the admixing CNT in the matrix as much as possible to develop a high-performance ECCC.

Carbon nanofiber
CNF has a diameter of ~100 nm and a length of a few microns (see Figure 1d) [102]. It has a high electrical resistivity of 10 −4 -1 Ω·cm, a tensile strength of 3800 MPa, and a modulus of elasticity of ~ 230 GPa [85,103]. Thus, it is an attractive conductive filler with multi-functional properties and an excellent reinforcing effect. Similar to CNT, the large aspect ratio of the CNF makes it difficult to disperse uniformly in the matrix. More advanced dispersion techniques should be explored to ensure acceptable dispersion in the concrete matrix.

Graphene nanoplatelet
GNP is a two-dimensional (2D) nanomaterial consisting of graphene stacks [104,105], which is known for its low price, outstanding mechanical strengths, and excellent electrical properties [106][107][108]. It has a surface area of 13-450 m 2 /g and higher electric mobility than CNT [109][110][111], making it an ideal conductive filler to develop ECCC [112][113][114][115]. As suggested in Figure 1e, combining 1D CNT and 2D GNP can exploit both co-effects and is one of the candidates for developing high-performance ECCC.

Dispersion agents
Carbon-based materials used to prepare ECCC often have large specific surface areas, large aspect ratios, and van der Waals forces, making their dispersion very challenging [116][117][118]. Especially for conductive fibrous fillers, such as CF, CNT, and CNF, the high aspect ratio and flexibility increase the difficulty of their dispersion in the matrix [119]. Currently, surfactants are usually used to disperse the conductive fillers, and the dispersion agents mainly include two types [120][121][122]: surfactant and mineral admixture. The former is achieved by wetting, electrostatic repulsion or potential spatial resistance effect, and the latter by adsorption or separation effects [58,59]. As shown in Table 1, using surfactant pre-treat conductive fillers is the most adopted method with the benefits of easy operability and simple process. More importantly, it is adaptable and can disperse almost any conductive filler. But the effective dispersion of fibrous conductive additives (e.g. CNT and CNF) is still challenging and has not been adequately explored and investigated. In addition, acid treatment ( Figure 2a) and pre-treatment with  Figure 2b) can also improve the dispersion quality of the admixing carbon-based materials.

Dispersion of carbon-based conductive fillers in the concrete matrix
It is particularly challenging to disperse conductive fillers in water and concrete matrix because of the extremely high van der Waals force among the fillers and the hydrophobic nature [126][127][128]. According to the previous reports [129], three types of techniques are usually used to disperse carbon-based fillers in the concrete matrix, including: 1) mixing the carbon conductive filler with dry cement powder by mechanical shear mixing or mechanical ball milling, which however may damage the microstructure of the filler and the dispersion effect is very limited (Figure 3a [130]; 2) the combined usage of surfactants and mechanical methods to achieve uniform dispersion of filler in solution, and then the filler dispersion is mixed with cement. As indicated by Du et al. [131], they combined chemical (polycarboxylate) and mechanical (power output of 300 W) methods to obtain a better dispersion of GNP ( Figure  3b; and Figure 3) Optimal dispersion of carbon-based conductive filler in solution is achieved by using surfactants, mechanical stirring, and ultrasonic treatment, followed by mixing the uniform filler dispersion with cement [132], as presented in Figure 3c. In general, the latter two methods, which use mechanical and chemical dispersion techniques, are more effective in improving the dispersion quality of carbon-based fillers. However, it is worth noting that the pre-treatment of carbon-based materials suspension is extremely tedious and time-consuming, which should be further optimized for the practical application of ECCC.

Preparation of ECCC
Similar to the manufacture of conventional concrete, the preparation of ECCC consists of three steps [26,108,133]: dispersing/mixing, molding, and curing. After casting the fresh mixture, additional vibration is needed to decrease the air bubble. A homogeneous conductive filler dispersion is typically prepared before mixing with cement particles. As such, the pre-dispersion of conductive filler is a key step for preparing ECCC, which will significantly affect the distribution of the admixing conductive fillers, thus affecting the homogeneity and conductivity of ECCC. According to previous studies [93,[134][135][136], the mixing process includes three methods ( Figure 4): first admixing, synchronous admixing, and latter admixing. Based on previous studies and our experience [137][138][139][140], the first admixing method is recommended for preparing ECCC because it is highly manipulable and can be easily extended to practical applications. Table 2 summarizes suitable mixing methods for different fillers. For hybrid fillers, joint mixing processes are recommended to be adopted.

Conductive mechanisms
This section will discuss in detail the conductive mechanisms of ECCC, which involves the percolation threshold of the conductive fillers, contact conductivity, tunneling effect, and field emission.

Percolation threshold
According to the statistical percolation theory [12], the relationship between ECCC's conductivity and the conductive filler concentration is developed and analyzed, determining the optimal amount of conductive filler inside ECCC. As shown in Figure 5a, the conductive filler's concentration is below the percolation threshold (Zone A), and it fails to construct a continuous conductive pathway. In this situation, the resistivity of ECCC slowly decreases with the increasing concentration of the conductive filler. As the concentration of conductive filler increases to the percolation threshold region (Zones B), the adjacent conductive fillers come into direct contact and form stable conductive pathways (Figure 5d), resulting in a dramatic decrease in the resistivity of ECCC by several orders of magnitude. As the concentration of conductive filler exceeds the percolation threshold, there are small fluctuations in resistivity (Zone C). In this case, SSCC has a more stable but less sensitive sensing performance. In summary, the percolation threshold is the basic parameter for developing and optimizing the conductivity of ECCC, and the appropriate concentration of conductive filler should be selected according to this theory before developing ECCC.

Contact conductivity
When directly adding conductive fillers into cement and then mixing it with aggregates and other components, the admixing conductive fillers would contact with each other in   [91,142] the continuous cement matrix phase (Figure 6a). The conductive channels constructed by the conductive fillers in direct contact allow electric currents to pass through the cement concrete, thus making it electrically conductive.

Tunneling effect and field emission effect
Some conductive fillers are dispersed inside the matrix as isolated particles. Thermal vibrations and electron transitions activate electrons, and as a thin layer of hydrates between these isolated fillers, a tunneling effect occurs where electrons can jump over this thin hydration layer and enter the adjacent conductive particles (Figure 6b). Especially when a strong internal electric field exists between conducting particles, an electric field emission current is formed, allowing electrons to pass through an electron barrier. Note that tunneling conduction is related to the distance between two adjacent conductive fillers, and the local tunneling resistance R as a function of the interparticle distance d can be described as follows: where m and e are the mass and electric charge of an electron; h is Planck's constant, λ is the potential barrier height; and A is the contact area of two conductive fillers. Note that field emission-induced tunneling conduction would contribute to some conductive filler with nanoscale tips such as CNT.
Typically, the three zones discussed above demonstrated different conduction mechanisms: ionic conduction for Zone A, tunneling conduction and/or field emission conduction for Zone B, and contact conduction for Zone C.

Electrical resistivity
In general, the resistivity decreases with increasing concentration of the conductive phase [35,147]. However, as mentioned earlier, the type of conductive phase and the doping amount is only the dominant factors affecting the resistivity and not ultimately determining the resistivity due to the differences in the preparation methods of ECCC, which lead to large fluctuations in their resistivity.
Graphite powder is a 0D conductive particle, and it is very challenging to construct a conductive path in the cement matrix with graphite powder alone compared to fiber-based carbon materials [62]. Therefore, a large amount of graphite powder needs to be incorporated to achieve high electrical conductivity. However, the introduction of overdosage of GP increases the porosity of the cement matrix, adversely affecting the mechanical strength of ECCC. In general, the CB percolation threshold is 12-20 wt% (corresponding to 7.22-11.39 vol%) [69,71], and it is important to note that the percolation threshold is not constant and may be altered by changes in the mix design. Therefore, 0D materials are usually combined with fibrous materials (CF, CNT or CNF, etc.) to improve cementitious composites' conductive network and mechanical properties through the bridging.
For 1D nanocarbon materials, adding CNT can significantly reduce the resistivity of concrete, and the CNT dosage is basically linearly related to the resistivity. Numerous studies have shown that the percolation threshold of CNT is 0.1-1.0 wt% [91,92], which mainly depends on the cementitious material mix ratio and CNT physical parameters (parameters such as surface area and aspect ratio). The main problem of CNF on the resistivity of cementitious materials is also the difficulty in ensuring high-quality dispersion, mainly because CNF with a high aspect ratio is easy to agglomerate, which can reduce the probability of its direct contact in concrete. Numerous studies have found that the percolation threshold of CNF is~0.5 wt% [92,148,149]. However, the enhancement efficiency of CNF in the compressive strength of concrete is weaker. 2D GNP nanosheet is one of the potential options for preparing ECCC due to its high conductivity [150,151]. The percolation threshold of GNP is ~ 2 vol% [142], when the introduced GNP can form a stable conductive channel inside the ECCC.
In summary, the current strategy for preparing conductive cementitious materials is to introduce the conductive phase (specifically carbon-based nanomaterials in this study) directly into the cement matrix and then mix it with the aggregate phase. However, the lower resistivity is usually obtained at the expense of workability and mechanical strength (see Table 3). Therefore, it is necessary to develop a high-performance (e.g. workability, mechanical, and durability properties are satisfied) ECCC to solve the current technical bottleneck and dilemma and to promote the development and application of conductive cement matrix composites. Some micromechanics approaches have been shown quite effective at predicting the conductivity of ECCC doped with fiber-like carbon inclusions (e. g. CNT) [156][157][158]. For instance, a simplified micromechanical model has been developed by Kim et al. [158] to predict the electrical performance and percolation threshold of cement composites. García-Macías et al. [153] have proposed a micromechanical model that considers CNT's wavy and nonuniform spatial distribution to predict the overall conductivity of CNT-modified ECCC, and both mechanisms governing the conductivity: electron hopping and conducting networks were incorporated into the developed model. However, the heterogeneity nature of cement composites makes it difficult to accurately predict their electrical conductivity. Therefore, more investigations are needed to obtain the parameters used in the developed models.

Factors affect the conductivity of ECCC
The main factors influencing ECCC's resistivity include the concentration of conductive filler, type of conductive filler (i.e. diameter, thickness, and aspect ratio), dispersion quality of conductive filler, and some parameters of the concrete matrix, such as water to cement ratio, sand to cement ratio, preparation process, and curing regime [127,134,138,139,159].
The concentration of conductive filler is the most critical factor for concrete to affect resistivity, as discussed in Section 3.1. Note that the huge specific surface area of carbonbased conductive fillers tends to agglomerate. This will affect the workability and mechanical properties of the mixes. Therefore, the selection of the proper conductive filler concentration is crucial for the development of conductive concrete. Fibrous conductive fillers are generally more difficult to disperse than granular ones. The dispersion process also affects the filler distribution in the concrete matrix. Therefore, the matching dispersion process should be carefully selected according to the characteristics of the conductive filler. Note that multichannel biphasic measurement methods eliminate polarization issues while keeping the multichannel measurement feature, which is not the case when performing alternating current measurements. A high water-cement ratio is beneficial to achieve good dispersion of conductive fillers in ECCC and improve the mixtures' workability. However, a higher water-cement ratio means more water in the matrix, which leads to an increase in the proportion of the ionic conductive contribution of ECCC, resulting in a less stable conductive signal. Furthermore, the moisture content greatly affects ECCC's resistivity, with an increase of nearly 1-3 orders of magnitude in the resistivity of the completely dried specimens compared to those in the saturated state. Therefore, moisture interference should be minimized in using ECCC.

Self-sensing performance of ECCC and its applications in SHM and traffic detection
The well-dispersed and distributed conductive fillers can construct a continuous and stable conductive pathway inside the concrete matrix, enabling concrete to demonstrate excellent conductivity. Under external loading, the conductive network of the concrete will change, which in turn leads to changes in its electrical properties (i.e. electrical resistance). Based on this principle, stress (or force), strain (or deformation), cracks, and damage of concrete structures can be detected under static or dynamic conditions. As illustrated in Figure 7, a Figure 7. Self-sensing of ECCC and its applications in SHM and traffic detection [143]. conductive network constructed by conductive fillers acts as a 'nervous system' to transmit electrical signals from internal and external stimuli to a computational center, namely, a 'brain' mimicking human behavior. As such, The ECCC is prepared in the form of a piezoresistive sensor, which can be applied to SHM for concrete infrastructures (e.g. buildings, bridges, tunnels, dams, etc.) and traffic detection (e.g. traffic volume monitoring and dynamic weighing of roads, etc.), as presented in Figure 7.

Shm
As summarized in Table 4, ECCC can be used in different forms of structures for SHM [12,141,143,160,161]. Among them, the last four forms of ECCC structure have the advantages of low cost and high efficiency compared to the first form. Gratifyingly, the negative effects of the conductivity of the ECCC on human safety and the corrosion of precast reinforcement can be avoided when these forms are applied.
Overall, the reliable loading-resistance variation of ECCC incorporating carbon-based conductive fillers, the excellent compatibility with the parent concrete, and the low cost enable ECCC to demonstrate great potential for SHM in infrastructures. Future research should promote the practical applications of ECCC on a large scale and long-term monitoring, and the collected data should be compared with conventional embedded sensors to confirm the effectiveness and reliability of ECCC-based sensors.

Traffic detection
The pavement or bridge integrated with ECCC-sensors can detect essential traffic data, such as traffic flow rate, vehicle speed and density, and implement weighing in motion [12,64,67]. For example, Han et al. [67] proposed a CNT-based ECCC pavement system for traffic detection (Figure 8a), and they found that the ECCC pavement system can accurately detect the passage of different vehicles at different speeds, which means that using such an approach to achieve real-time vehicle flow detection is possible. They also developed another kind of CNT-based ECCC pavement [64], as shown in Figure 8b, and found that the CNT-based ECCC showed sensitive and stable responses to repetitive compressive loads, which implies that CNT-ECCC has great potential for traffic flow detection and vehicle speed detection. Recently, Ding et al. [30] developed a novel ECCC using CNT grown on the cement surface (CNT@Cem), which achieved a maximum stress sensitivity of 2.87%/MPa and exhibited excellent repeatability and stability. Thus, the self-sensing circuit integrated with ECCC is robust to polarization changes within ECCC structures and the external environment.
Compared with other commercial embedded sensors used in SHM, self-sensing cement concrete made of ECCC has a natural compatibility with the parent concrete and has the same lifespan due to its cementitious characteristics. In addition, the Table 4. Different forms of ECCC structures used for SHM [12,70,129,162].

Category
Form of Existence Bulk ECCC Coating form The surface is covered with an ECCC layer Sandwich form The top and bottom surfaces are made of ECCC layers Bonded form The ECCC-based sensor was attached to the concrete Embedded form ECCC-based sensor admixing carbon-based fillers can reinforce concrete and thus exhibiting outstanding mechanical and durability properties. Overall, the tunability and scalability of this manufacturing process enable its great potential to mediate sensing concrete by controlling and adjusting the composition, dimension, and performance for various practical applications. As a result, ECCC can be used to manufacture smart infrastructure with sensing and health monitoring capabilities, thereby enhancing the suitability, safety, reliability, and durability of the infrastructures.

Deicing/Snow thawing
On many occasions in winter, snow and ice accumulation often lead to huge economic losses, safety hazards for infrastructures, and even tragic loss of life [163,164]. Although traditional snow and ice removal methods such as mechanical tools, chimerical salts or other heating candidates have been in use for many years, some shortcomings, such as lower heating efficiency, environmental pollution, and concrete/rebar erosion, have limited their application [163]. Delightfully, using ECCC as a heating element can be used for snow and ice melting on pavements, bridges, highways and airport sites, and this strategy offers a practical and environmentally friendly approach. As shown in (Figure (9a, b), [165] developed a graphene-based concrete road snow melting system and reported that graphene-based concrete has excellent temperature rise rates, with energy utilization rates of 43.41%, 41.36%, and 41.27% for road modules with heat fluxes of 1500 W/m 2 , 2000 W/m 2 , and 2500 W/m 2 when the thickness of snow is 100 mm. Besides, Li et al. [166] developed a CNF-based self-deicing road system, as shown in  [67]; vehicular loading experiment with CNT-based ECCC sensor [64]; and ECCC-engineered track slab for high-speed rail monitoring [30]. Figure (9c, 9d)). The resistivity of the CNF-based ECCC exhibited piecewise linear temperature-dependent characteristics within a certain temperature range of 0-280°C, demonstrating that this CNF-based ECCC has the potential to apply in pavement deicing/snow thawing.

Ohmic heating
Studies have shown that increasing the curing temperature is beneficial to reduce the capillary absorption and carbonation depth of concrete, which positively affects the mechanical and durability performance of concrete [84,167]. Ohmic heating is a novel thermal treatment method for cement concrete, and it is vital for the hardened performance of ohmic heating-cured concrete due to its importance for the maintenance costs and service life of the concrete structure and should be given attention.
For example, [167] developed a CF-based ECCC and investigated its ohmic heating curing at ultra-low temperatures. When subjected to freezing, the ohmically heated CF-based ECCC exhibited high strength recovery and compressive strength with values of 97.13% and 64.3 MPa, showing great potential to promote early-age strength development for cement concrete. [47] used CF and CB to develop an ECCC and studied the effect of curing conditions on the self-heating properties of concrete (Figure 10c). They found that the self-heating properties of ECCC were improved due to dry curing, as confirmed by the higher maximum surface temperature of ECCC. [168] found that the thermal images of the CNT-based ECCC's surface temperature can reach 101.5°C, demonstrating great potential for heating curing. [169] indicated that the maximum average surface temperature of ECCC reached~77°C when Figure 9. Self-heating performance of ECCC and its applications in pavement deicing/snow thawing: (a) melting process of snow [165]; (b) isotherm diagram of the ECCC module at different depths [165]; (c) preparation of CNF-ECCC [166]; and (d) the setup for the deicing experiment [166].
the CB content was 0.8 vol% for a given CF content. [170] suggested that the compressive strength of CF-based ECCC cured by ohmic heating curing at − 20°C for 48 h reached 51.94 MPa, higher than the control group samples cured at room temperature for seven days, indicating that ohmic heating is a promising way for curing CF-reinforced ECCC under ultra-low temperature conditions.

Electromagnetic interference shielding
Nowadays, more and more electronic devices and some electromagnetic waves often interfere with digital equipment and even human beings [53,87]. Protection against EMI is increasingly required in our daily life [96,171], as shown in Figure 11a. Figure 10. Self-heating performance of ECCC and its applications in ohmic heating: (a) surface temperature measurement for ECCC [47]; (b) thermal images of CNT-modified ECCC after 28 days [168]; (c) ECCC connected to power supplies and the thermal image [169]; and (d) the preparation of CF-based ECCC and its self-heating properties [170]. Figure 11. ECCC and its applications in EMI: (a) schematic diagram of a person surrounded by electromagnetic pollution [96]; (b) EMI mechanisms of GP-ECCC [172]; (c) setup for testing EMI shielding effectiveness [171]; and (d) EMI-SE testing procedure [52].
As compared to conventional concrete, ECCC-containing carbon-based fillers can effectively shield and absorb electromagnetic waves. For instance, [172] synthesized silica-coated CNT using a sol-gel method and found that a high EMI effectiveness of 24.2 dB was achieved for the ECCC containing 5 vol% CNT (Figure 11b). [52] used 0-2.0 wt % CNT to develop ECCC, and they found the addition of CNT substantially improves the conductivity and leads to shielding effectiveness up to the percolation threshold ( Figure 11d). Overall, developing ECCC based on carbon-based conductive fillers is an effective strategy for EMI.

Cathodic protection system
Cathodic protection is recognized as a suitable technique for protecting reinforced concrete (RC) structures damaged by chloride ion-induced corrosion, such as marine structures exposed to the action of deicing salts or pavement structures [50,51]. It has been generally accepted that the pulsed cathodic current method is an effective corrosion prevention method for contaminated RC elements. Previous studies show that several anode systems are being developed to address this issue [46,47], including metallized zinc and conductive organic coatings. Unfortunately, these anode systems suffer from high prices, incompatibility with the parent concrete, and poor durability.
Developing an anode system made of ECCC as a secondary covering anode is an ideal candidate for protecting RC structures. For example, [51] confirmed that an ECCCembedded CF could be used as an anode for the cathodic protection of steel-RC structures, and they found the cement-based conductive coatings for cathodic protection show acceptable behavior after two years of application of current densities up to 20 mA/m 2 (Figure 12a) [50] developed a type of CF-modified ECCC for the cathodic protection of RC structures (Figure 12b) , and it has many advantages over most of the other anode materials. [49] also studied the current distribution in a designed three-layer RC cathodic protection system with CF-reinforced ECCC as the anode (Figure 12c) , and it was found that higher electrical resistivity does not deteriorate the current distribution. This study provides a theoretical basis and strong evidence for the cathodic protection of RC structures in practical applications.

Summary and conclusions
Although ECCC has been studied and developed for almost 20 years, its composition design, performance optimization, and practical applications are still in progress. More efforts are needed to facilitate its development and application. The following challenges and insights are critical to the further development of ECCC.
(1) Carbon-based conductive fillers usually possess large specific surface area, aspect ratio, and van der Waals forces, leading to difficulties in solution and cement matrix dispersion. However, it is worth noting that the perfect dispersion of carbon fillers in solution does not mean they can be well-dispersed in the cement matrix. Moreover, semi-quantitative or quantitative assessments of the dispersion quality and distribution of the fillers in the concrete matrix are not available yet, which limits its broad application in the construction industry. More convenient, less costly, and operational strategies should be explored to disperse carbon fillers in the concrete matrix. (2) ECCC is a multi-phase material whose conductivity is greatly influenced by conductive filler type, mix design, and fabrication method. Among these factors, the type and concentration of the carbon-based filler are the keys to constructing the conductive pathway inside the concrete. To obtain high conductivity of the concrete, a high concentration of filler is desired, which may result in difficult dispersion and poor workability, thus affecting the mechanical and durability properties of ECCC. In order to realize the practical applications of the ECCC, it is necessary to simplify the manufacturing process and decrease the relevant cost. (3) Although directly introducing carbon-based fillers to modify cement matrix is a simple and scalable strategy to develop ECCC, the high cost of admixing conductive fillers (its concentration should be over the percolation threshold) and the complex dispersion process both limit its practical applications. Developing a novel conductive aggregate for improving electrical properties while reducing the consumption of carbon-based nanofillers can be a promising way to develop the next generation of smart conductive concrete. (4) The performance of ECCC is highly affected by the service environment. For instance, moisture, temperature, and erosion mediums may significantly affect the electrical signals. As such, it is urgent to develop a measurement circuit design and new electrodes to monitor the electrical signals of ECCC accurately. (5) The manufacture of ECCC promotes the development of smart infrastructures. ECCC possesses high conductivity and strain/stress sensitivity, enabling many potential applications, such as snow and ice melting, a cathodic protection system, EMI, SHM, and traffic detection.

Prospects and future work
Various reasonable explanations for the conductive mechanisms of the ECCC have been proposed through extensive experimental and theoretical analyses. However, since the conductive mechanisms of ECCC are quite complex in practical applications, the current explanations are still unclear or need further validation to drive the further design and optimization of ECCC. Currently, the practical applications of the ECCC are very limited, and the corresponding supporting technology remains immature. Future studies should aim to establish standard methods and guidelines for the ECCC design and optimization for practical applications. Although some studies have attempted to explore 3D printing of smart cement composites [173][174][175], laboratory applications of bulk smart concrete for crack identification for instance using resistor mesh models [161,176], and vibration strain measurements using concrete-based sensors [177][178][179], more research is needed to promote their practical applications.