Dispersion of carbon fibers and conductivity of carbon fiber-reinforced cement-based composites

Ceramics International 43 (2017) 15122–15132

Contents lists available at ScienceDirect

Ceramics International journal homepage: www.elsevier.com/locate/ceramint

Dispersion of carbon fibers and conductivity of carbon fiber-reinforced cement-based composites

MARK

⁎

Wang Chuanga, , Jiao Geng-shengb, Li Bing-lianga, Peng Leia, Feng Yinga, Gao Nia, Li Ke-zhic a b c

Shaanxi Province Engineering Laboratory of High Performance Concrete, Shaanxi Railway Institute, Weinan 714099, PR China School of Materials Science and Chemical Engineering, Weinan Teachers’ University, 714000, PR China School of Materials Science and Engineering, Northwestern Polytechnical University, Xi’an 710072, PR China

A R T I C L E I N F O

A BS T RAC T

Keywords: Carbon fibers Cement Dispersion CFRC Conductivity

Dispersion of carbon fibers in the cement matrix remains a hot topic in the preparation of carbon fiberreinforced cement-based composites (CFRC) because it affects greatly both the mechanical and electrical properties of the composites. In this work, a new dispersant hydroxyethyl cellulose was used with the aids of pre-dispersion by ultrasonic wave to realize the uniform distribution of chopped carbon fibers in the cement matrix. The fracture surface of the prepared CFRC was observed by scanning electron microscopy, the elemental distribution was investigated by energy dispersive spectroscopy, and the components was analyzed by X-ray diffraction. Influences of carbon fiber lengths and contents, water/cement weight ratio, molding process, curing time, and silica fume content over the conductivity of the CFRC composites were studied. The mechanism of conductivity was discussed. Results shown that the electrical resistivity intended to decrease with the increasing of carbon fiber contents. The mass fraction 0.6% of carbon fibers was a turning point. The concentration of hydroxyethyl cellulose between 1.66% and 1.86% was mostly beneficial for the dispersion of carbon fibers. The resistivity was increased first and decreased then with the increase of water/cement ratio. When the CFRC sample was prepared by the vibrating pressing method, the resistivity of the sample was reduced far greatly than that of the sample by the vibrating method. The incorporation of silica fume into the CFRC composites exerted not only a good effect on the dispersion of carbon fibers, but also increased the density of the composites to further influence the conductivity of the CFRC.

1. Introduction Cement is a construction material commonly used in engineering constructions due to its rich resources, good environmental adaptability, low cost and high compressive strength [1–6]. It self is an electrical insulating material. Carbon fibers exhibit a series of outstanding properties of high strength, high modulus, high temperature resistance, corrosion resistance, fatigue resistance, creep resistance, light weight, and electric conduction [1,3,7–9]. It is convenient to produce carbon fiber-reinforced cement-based composites (CFRC) [6,10–12] by adding short carbon fibers into cement matrix. In the CFRC composites, carbon fibers are advantageous in their superior ability to increase the tensile strength of cement. They can improve both mechanical and electrical behaviors of the material as well as the electromechanical and electromagnetic behaviors [1,13–15]. They are also advantageous in the relative chemical inertness. In relation to most functional properties, carbon fibers are exceptional compared to other fiber types [1,3]. ⁎

Corresponding author. E-mail address: [email protected] (W. Chuang).

http://dx.doi.org/10.1016/j.ceramint.2017.08.041 Received 3 July 2017; Received in revised form 29 July 2017; Accepted 5 August 2017 Available online 06 August 2017 0272-8842/ © 2017 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

CFRC is of great technological interest owing to the combination of good structural properties and exceptional electrical properties. It can be used both as structural materials and as functional materials. The electrical resistivity of CFRC has been widely studied because of its utility as multifunctional materials [16,17]. Its applications, particularly in military use, have attracted more and more researchers' attention and become a hot topic for the cementitious materials [1,6,9]. The incorporation of an appropriate amount of carbon fibers into cement matrix can not only enhance the tensile ductility, flexural strength, and toughness to reduce the dry shrinkage and to improve the bond strength of CFRC, but can also adjust the electrical conductivity in a large range [18,19]. Meanwhile, CFRC is more attractive for its multifunctional behavior, which derives from its electrical property, including strain sensing, temperature sensing, piezoresistivity effect, thermoelectric effect, and joule heating [1,20–22]. It special features involves pressure sensitivity, thermoelectric effect, Joule effect, low drying shrinkage, high specific heat, low thermal conductivity, high electrical conductivity, high corrosion resistance and weak thermo-

Ceramics International 43 (2017) 15122–15132

W. Chuang et al.

electric behavior, and temperature resistance effect [23,24]. Therefore, it is a kind of promising materials in the engineering fields of civil engineering, industrial anti-static, health monitoring, non-metal heating elements and buildings against electromagnetic wave shielding [19,25,26]. The effect of carbon fiber addition on the properties of cement increases with the fiber content, unless the content is so high that the air void content becomes excessively high. The air void content increases with the fiber content and the air voids tend to have a negative effect on many properties, such as the compressive strength. In addition, the workability of the mix decreases with the fiber content. Effective use of carbon fibers in cement requires their homogeneous dispersion in the cement matrix [6,27,28]. There are lots of factors affecting the properties of CFRC composites. Among them, the dispersion of carbon fibers in the cement matrix directly affects the mechanical properties and electrical properties. In order to make carbon fibers uniformly dispersed in the cement matrix, an appropriate amount of suitable dispersants and additives [29,30] should be added in the preparation of CFRC composites. The dispersion is also enhanced by using silica fume (a fine particulate) as an admixture. A typical silica fume content is usually 15% by weight of cement [31,32]. Silica fume is typically used along with a small amount (0.4% by weight of cement) of methylcellulose for helping the dispersion of fibers and the workability of the mixture [33,34]. The improved structural properties rendered by carbon fiber addition pertain to the increased tensile and flexible strengths, the increased tensile ductility and flexural toughness, the enhanced impact resistance, the reduced drying shrinkage and the improved freeze-thaw durability [35,36]. The fiber dispersion greatly affects the air void content, which in turn exerts influence on the mechanical and electrical performances of the CFRC composites [34,37,38]. Many works have been done on the electrical properties of CFRC as of today [1,16,39–41]. However, the relationship between carbon fiber dispersion and conductivity of CFRC has been little reported yet. With this in mind, in the present work, carbon fibers were firstly dispersed in the aqueous solution with the aids of a good dispersant hydroxyethyl cellulose. The dispersed system was then incorporated into the cement matrix under moderate stirring to ensure carbon fibers were uniformly dispersed to have achieved homogeneous CFRC samples. The flexural strength and the compressive strength of the samples were tested. The fracture surface of the samples were observed by scanning electron microscopy. The distribution of major elements in the CFRC was analyzed by energy dispersive spectroscopy (EDS). The composition of CFRC was analyzed by X-ray diffraction (XRD) at different amounts of carbon fibers. The resistance was measured to further calculate the resistivity. Influences of carbon fiber contents and lengths, water/cement ratio, curing age, molding process, and content of silica fume on the conductive properties of CFRC composites were investigated. The experimental results and theoretical analysis may provide helpful references for CFRC used as functional materials both for civilian and military purposes. 2. Raw materials and experimental methods 2.1. Major raw materials Short carbon fibers used were polyacrylonitrile-based 5–7 mm in length provided by Jiyan Carbon Co., Ltd. (Jilin, China). Their major parameters are shown in Table 1. The matrix was 32.5R Portland

cement with the execution standard GB175-1999 from Qinling Cement Plant, Shaanxi, China. The water/cement ratio was from 0.3 to 0.5. The mass fractions of carbon fibers by weight of cement were 0.2%, 0.4%, 0.6%, 0.8%, and 1.0%, respectively. No aggregate (fine or coarse) was used. The dispersants used were hydroxyethyl cellulose made in Shandong Yiteng Chemical Co., Ltd with the viscosity of 30000 Pa.S. The mass fraction of hydroxyethyl cellulose was in the amount of 0.6% by mass of cement. Silica fume (300 mesh, purity 99.7%, Kaihua Yuantong Silicon Industrial Co., Ltd, Zhejiang) was used in the amount of 5%, 10%, and 15% by mass of cement. The diameter of silica fume particles ranges between 0.01 and 0.1 µm. Naphthalenesulfonate formal condensate, a water-reducing agent, purchased from Wuhan Iron and Steel Corporation Admixture Factory was used in the amount of 0.8% by mass of cement. The defoamer was liquid tributyl phosphate (Henghao Science and Technology Co., Ltd, Tianjin, China), which was used in the amount of 0.02% by mass of cement. A DDG-A high efficiency electric contact conductive paste was made in Wuhan Changjiang Mechanical and Electrical Equipment Industrial Co., Ltd. 2.2. Preparation of CFRC samples To prepare homogenized CFRC samples, the dispersion of short carbon fibers are divided into two steps. In the first step, they are dispersed into the aqueous solution, and in the second step they are distributed in the solid mixture. Carbon fibers were first placed in a 500 mL glass beaker. About three-fifth of the entire water to be used in the subsequent test was added to make sure that carbon fibers were immersed fully into water. The beaker was vibrated by ultrasonic wave for 10 min for predispersion. The temperature of the water was kept between 38 and 44 °C. The power of the ultrasonic wave was 250 W. Hydroxyethyl cellulose, a dispersant, was dissolved in water and stirred by hand for 2 min. Two drops of tributyl phosphate were added to eliminate air bubbles. The beaker was continuously vibrated by ultrasonic wave for another 10 min. The dispersion process can be seen clearly through the transparent glass beaker. In the first step, the mass fraction of hydroxyethyl cellulose in the aqueous solution was controlled between 1.66% and 1.86%. A glass rod was used to stir carbon fibers intermittently to ensure their uniform dispersion in the transparent, sticky solution. No breakage of carbon fibers occurred during the ultrasonic vibration. This is the first step for the dispersion of carbon fibers. Now comes to the second step. Silica fume, cement, standard sand, naphthalenesulfonate formal condensate, and other additives were mixed in a J-160A rotary mixer with a flat beater for 2 min. Then, the prepared liquid dispersive system in the first step was poured into the mixer. The left two-fifth of water was added and mixed for another 2 min. The mixture was stirred for 1 min quickly. By these two steps, carbon fibers are well dispersed in the prepared hardened CFRC samples, though the mixing and later molding operations might change the dispersion degree more or less. After pouring the prepared mixture above into the rectangular oiled moulds of 160 × 40 × 40 mm and the cube oiled moulds of 40 × 40 × 40 mm, respectively, an external vibrator was used to facilitate compaction and decrease the amount of air bubbles. For the rectangular beam, three samples of each mix were prepared for flexural strength and compressive strength test. For the cube block, thin sheets

Table 1 Main parameters of short carbon fibers. Diameter/μm

Density/g/cm3

Tensile strength/MPa

Shear strength/MPa

Elastic modulus/GPa

Elongation/ %

Electrical resistivity/Ω·cm

7 ± 0.2

1.76–1.78

2500–3000

80

200–220

1.25–1.5

3.1 × 10−3

15123

Ceramics International 43 (2017) 15122–15132

W. Chuang et al.

Fig. 1. Mechanical properties of CFRC varies with carbon fiber contents (a) Relationship between the bending strength and the mass fractions of carbon fibers; (b) The compressive strength varies with the mass fractions of carbon fibers.

of coppers (50 × 10 mm) were embedded in the opposite end of the cube moulds, which are used as electrodes. The samples were placed in a curing box inside which the temperature was kept around 22 °C and the relative humidity was ≥ 96%. They were demolded after 24 h and then allowed to cure in a curing box for 28 days. Prior to test, the samples were placed in a constant temperature drying box for 24 h. 2.3. Test of resistance of CFRC Alternating current (AC) is often used in the measurement of electrical conductivity to avoid the problem of polarization because of charge carriers moving back and forth as the voltage polarity varies under AC condition [40,42,43]. Comparatively, direct current (DC) is simpler and less expensive than AC. In this work, a two electrode method was adopted to measure the resistance of CFRC samples. The copper electrodes in the CFRC samples were sanded to be shone. The essence of this method is to directly measure the resistance with an ohmmeter. A pair of electrodes were shared for measuring the current and the voltage. Three samples of each composition were tested. A FLUCK89 precision digital resistance tester was directly connected with the electrodes at either ends of the test piece, the resistance being tested, and the resistivity was then calculated according to formula (1) [39,44,45]:

ρ = RS / L

(1)

where ρ refers to the resistivity of the sample (Ω·cm), R stands for the resistance (Ω), S signifies the cross-sectional area (cm2), and L is the distance between the electrodes (cm). 2.4. SEM observation and XRD analysis The morphology and microstructure of the fractures of CFRC samples was observed and analyzed by JEOL JSM-6460 scanning electron microscopy. Half of the broken specimen was used for the compressive strength test. The composition of hardened CFRC was analyzed by an X-ray diffractometer (XRD, D/max-2200PC, Rigaku, Japan). 3. Mechanism of conductivity of CFRC 3.1. Uniform dispersion and mechanical properties of CFRC In order to make carbon fibers show continuously good dispersion in the cement matrix, ultra-fine particles of silica fume was added in

the second step. Because the fume particle size was less than that of cement particles and the diameter of carbon fibers, it exerted a good filling effect to eventually separate carbon fibers. Only when carbon fibers are dispersed uniformly in the cement matrix, can homogeneous CFRC composite samples be prepared. In this case, it is meaningful, either mechanically or electrically, to further investigate the properties of CFRC. Therefore, hydroxyethyl cellulose was added into the aqueous solution in the first step of preparation, the mass fraction of which was controlled between 1.66% and 1.86%. It increased the density and viscosity of the liquid phase, which reduced the tendency of bundling of carbon fibers. Meanwhile, it was attached to the surface of carbon fibers, forming a layer of solvent water film, which increased the wettability of carbon fibers and prevented them from assembling and bundling. As a result, they tended to disperse in monofilament. At the same time, in the mixing process, a moderate stirring speed should be selected to not damage carbon fibers but to disperse them well. The operation speed should be fast and slow alternately. The mechanical properties of the prepared CFRC samples were first investigated. According to the Q/GB95-92 criterion, the samples were used for the three-point bending test at the age of 28 days. The flexural strength and compressive strength tests were conducted using an Instron 1195-typed electronic universal testing machine (Instron Company, US). The span was 100 mm and the load speed was 0.5 mm/min. Half of the sample after the flexural test was used to test the compressive strength. Fig. 1(a) shows the flexural strength as a function of fiber mass fractions when carbon fibers are well dispersed in the cement matrix. Obviously, the flexural strength increased linearly with the increasing of mass fractions of carbon fibers. This is because carbon fibers have the function of strengthening and toughening, namely, reinforcement, which can prevent the formation and propagation of micro cracks in the cement matrix. Fig. 2(b) demonstrates the compressive strength varies with the mass fractions when carbon fibers are dispersed well in the cement matrix. It increased slowly before the percentage of 0.6% and increased by 20% at this percentage. However, the strength doesn’t always keep increasing as the mass fraction rises. When carbon fibers are used in an excessive amount, they are difficult to disperse evenly in the cement matrix and are likely to fasciculate, leading to the reduction of strength. Meanwhile, a large number of air bubbles would be introduced owing to the mix several times, which might produce many pores in the matrix. This will severely affects the compressive strength [6,46,47]. Consequently, it decreased dramatically after 0.6% owing to the holes and fascicule inside the composites. The fracture surface of the CFRC samples after mechanical test was observed under a scanning electron microscope (JSM-6460) as shown

15124

Ceramics International 43 (2017) 15122–15132

W. Chuang et al.

Fig. 2. Morphology of the fracture surface of the CFRC samples: (a) good dispersion of carbon fibers in the cement matrix; (b) poor dispersion of carbon fibers.

in Fig. 2. Clearly, carbon fibers were well distributed in the cement matrix as seen in Fig. 2(a), while they were not well dispersed as seen in Fig. 2(b). In the condition of poor dispersion in Fig. 2(b), carbon fibers were separated by a thin layer of cement wall. In this case, it is very difficult for them to contact each other to form a conductive network. Meanwhile, electron transition were forbidden or in other words the tunneling effect of electrons among carbon fibers were difficult to occur [23,48]. Good dispersion of carbon fibers is not only beneficial to the mechanical properties but also to the electrical performances owing to the possible contact of carbon fibers one another, whereas the poor dispersion may exert negative influences on the mechanical and electrical properties of the composites [49,50]. It is known that the rupture of the composites is closely related with the development of micro-cracks inside it [24,26]. Adding carbon fibers into cement can stop the free expansion of cracks, which not only enhances the compressive strength but also improve the toughness of CFRC [8,16]. So, it is essential for uniform dispersion of carbon fibers to achieve homogenized CFRC composites. 3.2. Analysis of the conduction mechanism of CFRC As mentioned above, when carbon fibers are uniformly dispersed in the cement matrix, they can contact or overlap each other to form a conductive network. To show clear overlap of carbon fibers in the cement matrix, two other SEM images of CFRC samples are provided in Fig. 3. When carbon fibers in the cement matrix are in a bundle state as shown in Fig. 2(b), they are separated by cement barriers and can’t contact each other. In this case, it is difficult to form a conductive network in the CFRC composites. When the distance between carbon

fibers are far apart from each other, tunnel effect can’t take place, either [16,18]. Generally, the electrical conduction in CFRC involves ions, electrons and/or holes and comprises three types of conduction ways [13,51]. In the case of ions, there is only one type of conduction, which is associated with the motion of ions of Ca2+, Na+, K+, OH− and SO42− in pore solution. Ionic conductivity changes in a particularly wide range when cement contains a substantial amount of free water. Under dry conditions, the cement matrix is almost an insulating material due to the high resistivity (105–107 Ω·m), so the ionic conduction through pore solution is negligible [1,49]. By analyzing the chemical compositions, it is known that cement clinkers are C3S, C2S, C3A and C4AF. After hydration reaction, hardened cement pastes are composed of unhydrated clinker particles, hydration products and pore network which is occupied by water and a small amount of air, forming a porous solid-liquid-gas system. The hydration products include hydrated calcium silicate gel of poor crystallization, calcium hydroxide, and crystal bodies of hydrated calcium aluminate. Pores are composed of gel holes (15–30A) inside the gel, capillary pores (2000A) filled with water, and a small amount of water in the hole as well as aqueous electrolyte ions such as Ca2+, Na +, K+, OH− and SO42−. They all give a certain conductive ability to the hardened cement paste [7,52,53]. In the case of electrons and/or holes, there can be two types of conduction: one is associated with the motion of free electrons and/or holes through the conductive paths formed by a large amount of contacting carbon fibers, and the other is associated with the transmission conduction of electrons between adjacent disconnected carbon fibers. The transmission conduction is dominated by the tunneling effect. These types of conduction modes coexist in CFRC, but they are

Fig. 3. Formation of a conductive network between mutual lap of carbon fibers (Mass fraction of carbon fibers 0.8%).

15125

Ceramics International 43 (2017) 15122–15132

W. Chuang et al.

not independent. The conductivity of CFRC by electrons can also be explained from the perspective of chemical structure. Chemically, there are aromatic rings in carbon atoms. Delocalized π bonds can be formed in the rings. They are actually a large conjugated system in which π electrons move freely, which gives conductivity to the CFRC composites. This is the major reason why the composites are of conductivity. Two theories are usually employed to interpret the conduction mechanism of CFRC: one is percolation theory and the other is tunneling effect theory. An idealized model of percolation is that adjacent fibers physically contact and form a geometrically connected phase when the fiber content is near or above the percolation threshold. However, the tunneling effect also contributes to the conduction for the case when two adjacent fibers are not physically in contact, even when the fiber content is below the percolation threshold. In general, a low content of fibers (usually near the percolation threshold) is preferred in practice not only since lower cost, good workability and higher compressive strength (low air void content) can be obtained, but also a decrease of fiber addition close to the percolation threshold will increase the sensitivity of CFRC as a strain sensor [11,34]. According to the percolation theory，when the fiber content is higher than the percolation threshold, the moving of electrons and/or holes through continuous fibers, or designated as Ohmic contacting conduction mode, dominates. In this case, the percolation networks are formed by the short-cut randomly distributed carbon fibers, and the conductivity for the composites is increased. In contrast, the electrons’ transmission plays an important role when the carbon fiber content is below the percolation threshold or falls within the percolation transition zone [23,45]. From analysis above, it is known that, to form a conductive network, the amount of carbon fibers must be appropriate. It can neither be too much nor too little. Usually, the mass fractions of carbon fibers lie in between 0.6% and 0.8% during which carbon fibers can contact and overlap each other, forming a conductive network. When the mass fraction is more than 1.0%, a phenomenon of fasciculus of carbon fibers often takes place and it is very difficult for carbon fibers to disperse [16,23,54]. 3.3. Analysis by XRD and energy dispersive spectroscopy (EDS) 3.3.1. XRD analysis To find out the chemical compositions of the prepared CFRC composites, X-ray diffraction patterns are carried out for different mass fractions of carbon fibers of 0.2%, 0.4%, 0.6%, and 0.8% as shown in Fig. 4. The sharpest diffraction peak in Fig. 4 is marked by “a”, standing for the compound SiO2 plus 3CaO·Al2O3·3CaSO4·32H2O (Ettringite, or AFt) and the symbol “b” stands for the compound

Fig. 4. XRD patterns of CFRC with different mass fractions of carbon fibers (0.2%, 0.4%, 0.6%, 0.8%).

CaCO3. Apparently, the two compounds are main compositions. The diffraction peaks for different carbon fiber contents are much similar, which indicates that the content of the two compounds keeps stable, though different additives were added to adjust the workability in the preparation stage of CFRC. The most dominant compound SiO2 is attributed to the excess amount of silica fume during the preparation process. Ettringite (3CaO·Al2O3·3CaSO4·32H2O) is one of the most important early hydration products of hardened cement paste, accounting for about 7% of the Portland cement hydration product. It may reach 25% in the expansive cement. It can improve the early strength of concrete or cementitious materials, almost the main contributor to the early hardening properties of cementitious materials. Micro expansion will occur during the formation of ettringite phase. This expansion can compensate the shrinkage of cement in the early stage. However, if a large amount of ettringite is formed after hardening, the volume expansion will cause the cracking in the cement matrix, resulting in the loss of strength. It is conducive to inhibition of later expansion induced by ettringite by reducing the water cement ratio, adding fly ash and slag or other mineral admixtures. Ettringite is a relatively unstable component. Its formation and stability are not only related closely to alkalinity of the solution in the cement pore and concentrations of Ca2+, SO42−, and AlO2− ions, but also related to the alkali-silicic acid intergrowth, environmental temperature and humidity. XRD analysis showed that when carbon fiber contents varied, the amount of additives, especially that of the dispersants, changed accordingly. But the main chemical compositions of the mixture did not change obviously. During the formation of ettringite, the sample was already dried for a certain period of time. Under such circumstances, there were almost no free moving ions. In other words, there was little impact on the conductive properties of CFRC. 3.3.2. EDS graphics of a standard cement mortar sample and CFRC sample As previously discussed, it is essential that carbon fibers be homogenized in the cement matrix to ensure the prepared CFRC samples possess good mechanical and electrical properties [28,30]. To compare the variations of major elements in the cement matrix, EDS of a standard cement mortar sample and a CFRC sample are shown in Figs. 5 and 6, respectively. Fig. 5 shows the SEM image of the fracture surface of a standard cement mortar sample without any carbon fibers and the corresponding EDS graphics. The ratio of cement/sand/water is 1:3:0.5. It can be seen from the SEM image that the structure of cement is relatively compact. So, it can be inferred that the material may exhibit a good compressive strength. From the corresponding EDS graphics in the right side, it can be seen that major elements of O, Ca, Fe, C, and Si exhibit relatively stronger peaks. Meanwhile, the spectrum also shows the relative contents of elements of S, K, Al, and Sn. To describe quantitatively the contents of different elements in the cement mortar, the element analysis is given in Table 2. The weight percentage of major elements such as O, Ca, C, and Si is 54.03%, 31.03%, 9.20%, and 2.42%, respectively. Of course, there are other elements such as Al, S, K, Fe and Sn existing in little amount in the cement matrix due to introduction of impurities in cement production. Fig. 6 shows the SEM image of the fracture surface of a CFRC sample with different additives and the corresponding EDS graphics at mass fraction 0.8% of carbon fibers. When carbon fibers were incorporated into the cement matrix to prepare CFRC samples, different admixtures were added to ensure the good dispersion of carbon fibers. Therefore, the distribution status of major elements O, Ca, C, and Si varied. As shown in the EDS graphics in Fig. 6, the spectrum dot was chosen in the cement matrix not in some dot in carbon fibers. Clearly, the peaks of Ca, O, C, Si, and Fe are relatively apparent.

15126

Ceramics International 43 (2017) 15122–15132

W. Chuang et al.

Fig. 5. SEM image and EDS graphics of a standard cement mortar sample.

Compared with the distribution of major elements in the standard cement mortar sample, the weight percentages of major elements of O, Ca, C, and Si in the CFRC composites are 52.49%, 23.84%, 9.79%, and 7.82%, respectively, as shown in Table 3. To clearly compare the variations of O, Ca, C, and Si before and after additives were added, the percentage variations of weight and atomic were calculated as listed in Table 4. From Table 4, it can be seen that elements of O, Ca, and C varied greatly (variation value over 90% is referred), the weight percentages of which are 103%, 130%, and 94%. The corresponding atomic percentages were 104%, 132%, and 95%. It should be noted here that the element Sn varied remarkably, 142% in weight percentage and 143% in atomic percentage, which may be caused by the instrument itself in the measurement of EDS. Analyzing the distribution of major elements such as O, Ca, C, and Si is helpful for selecting appropriate additives and moderate amount of dispersants. Meanwhile, by referencing the variation of major elements, the water/ cement ratio can be altered by adjusting different additives during the preparation of CFRC to keep the moderate workability and achieve the most ideal dispersion of carbon fibers in the cement matrix.

Table 2 Elemental analysis of a standard cement mortar sample (Cement: Sand:Water=1:3:0.5). No.

Element

Weight %

Atomic %

1 2 3 4 5 6 7 8 9

OK Ca K CK Si K Sn L SK Al K KK Fe K Totals

54.03 31.03 9.20 2.42 1.18 0.96 0.51 0.38 0.30 100.00

66.51 15.25 15.09 1.69 0.20 0.59 0.37 0.19 0.11

Remarks

Table 3 Elemental analysis of CFRC sample (mass fraction of carbon fibers 0.8%).

4. Results and discussion 4.1. Influences of carbon fiber contents and lengths over the conductivity of CFRC Fig. 7 shows the resistivity of CFRC specimens varied with the amount and length of carbon fibers when the length was 4, 7 and 10 mm, respectively. Clearly, when the carbon fiber content lay in between 0 and 0.2%, the resistivity dropped sharply; when it fell in between 0.2% and 0.3%, the resistivity decreased slowly, and in this

No.

Element

Weight %

Atomic %

1 2 3 4 5 6 7 8 9

OK Ca K CK Si K Al K KK Sn K Fe K Na K Totals

52.49 23.84 9.79 7.82 2.16 1.80 0.83 0.81 0.47 100.00

63.87 11.58 15.87 5.42 1.56 0.90 0.14 0.28 0.39

Fig. 6. SEM image and EDS graphics of a CFRC sample (Mass fraction of carbon fibers 0.8%).

15127

Remarks

Ceramics International 43 (2017) 15122–15132

W. Chuang et al.

Table 4 Comparison of percentage of major elements in standard cement mortar and CFRC. Element

O

Ca

C

Si

Al

K

Fe

Na

Sn

Weight% Variation % Atomic% Variation %

54.03 /52.49 103 66.51/63.87 104

31.03/23.84 130 15.25 /11.58 132

9.20 /9.79 94 15.09/15.87 95

2.42/7.82 31 1.69/5.42 31

0.51/2.16 24 0.37/1.56 24

0.38/1.80 21 0.19/0.90 21

0.30/0.81 37 0.11/0.28 39

0/0.47 – 0/0.39 –

1.18/0.83 142 0.20/0.14 143

Fig. 7. The electrical resistivity of CFRC varies with carbon fiber length and content.

Fig. 8. Variation of the electrical resistivity of CFRC with the water/cement ratio.

case the resistivity remained almost unchanged for the fiber length of the sample of 7 mm. When the fiber content was 0.3–0.6%, the resistivity increased sharply with the increase of the fiber content, in other words, there existed a seepage phenomenon [22,35]. At this moment, the fiber content corresponding to the inflection point was approximately 0.4%. When the fiber content was more than 0.6%, the resistivity changed gradually with the increase of carbon fiber content. The electrical conductivity of CFRC materials depends mainly on the tunnel effect which is both formed between overlapped carbon fibers that are randomly dispersed in the cement matrix and between adjacent carbon fibers that are separated by a very thin layer of cement matrix. In CFRC composites, there are a large number of interconnected capillary channels, inside which there is pore water existing. When the carbon fiber content is low, the conductivity is mainly carried out by two ways: hydrated ions moving freely in the internal pore water in the cement matrix and current carriers moving in the conductive carbon fibers through the tunnel effect. Hydrated ions like OH−, K+, Na + and Ca2+ migrate towards the interface between the channel and the cement matrix, eventually forming a continuous conductive water film inside the channel. It takes some time for the completion of this process, so the stage appears in which the resistivity declined slowly. With the increase of the carbon fiber content, the conductions by electrons, by current carriers through the tunnel effect, and by overlapping of carbon fibers become the main conducting mode. At the same time, the continuous conductive water film has been formed, lowering the barrier between carbon fibers to increase the tunneling probability of π electrons. As a result, it is easier for electrons to transfer freely between carbon fibers, so the resistivity decreases greatly. However, with the increasing of the carbon fiber content, more carbon fibers are overlapped in the cement matrix, on which the conductivity of the system depends mainly. When the carbon fiber content exceeds the percolation threshold [27,39], in other words, a conductive network is fully formed inside the material, the resistivity varies hardly with the increasing of carbon fiber contents. There is only a slight variation. The resistivity tends to be a certain limit. It is also found that the resistivity of the length 10 mm is smaller than that of the length 4 mm and 7 mm when carbon fiber contents are

the same. Moreover, the carbon fiber length exerts greater influences over the resistivity when the amount of carbon fiber content is smaller. When the amount of fiber content is over 0.6%, the influence of fiber length on the resistivity decreases gradually. The effect of carbon fiber length on the conductivity of CFRC depends mainly on the overlap degree of carbon fibers in the composite. When the carbon fiber is longer, it is easier to form the conductive network by overlapping. The conductivity along the network is stronger. Therefore, the resistivity is smaller. 4.2. Influences of water/cement ratio on the conductivity of CFRC The water/cement ratio exerts influences over the conductivity of CFRC. Fig. 8 shows the resistivity varies with the water/cement ratio changing from 0.23 to 0.6 with fiber length of 7 mm and mass fraction of 0.6%. The sample was prepared by the vibration method and cured for 28 days. Under certain conditions, the increase of the water/cement ratio can improve the workability of the mixture, enhance the dispersion of cement particles and carbon fibers, and improve the uniformity of the composite system. When the water/cement ratio changed from 0.23 to 0.30, the resistivity decreased suddenly. This is because that this period lay in the initial stage of hydration, during which the number of free ions in the liquid phase of hardened cement paste is relatively large, so the conductivity was stronger. When the water/cement ratio rose up from 0.30 to 0.5, the resistivity increased gradually. The reason is that the hydration reaction increased gradually and the number of free ions decreased continuously. When the water/cement ratio exceeded 0.5, the hydration reaction was basically completed. In this case, the number of free ions remained almost constant, so, the resistivity kept basically unchanged. Obviously, the water/cement ratio 0.30 is a turning point in the curve. In addition, the change of the water/cement ratio will lead to the variation of the porosity in the hardened samples, and further the size of the porosity exerts a direct impact on the test result of the resistivity [17,29,34].

15128

Ceramics International 43 (2017) 15122–15132

W. Chuang et al.

Fig. 9. Influence of different molding methods on the electrical resistivity of CFRC (a): Carbon fiber content 0.2 wt%; (b): Carbon fiber content 0.4 wt%.

4.3. Influences of molding technology on the resistivity The densification process also has an effect on the resistivity of the CFRC sample. When a certain amount of carbon fibers were added, the resistivity of the samples which are prepared by the vibrating pressing method was much smaller than that of the samples prepared by the vibrating method as shown in Fig. 9(a) and (b). Fig. 9(a) shows the resistivity curve with different molding methods when the carbon fiber content was 0.2%, whereas Fig. 9(b) shows the resistivity curve when the carbon fiber content was 0.4%. The curves above in Fig. 9(a) and (b) are the resistivity variation obtained by the vibration molding method, and those below are the resistivity variation achieved by the vibration compaction method. In the vibration compaction process, the water in the mixture was squeezed under pressure and the formed hardened cement paste was much denser. The distance between carbon fibers became smaller, the overlapping degree increased, and the porosity decreased, so, the electrical conductivity was good and the resistivity was lower [26,28,32]. At the same time, the contact resistance between carbon fibers and cement paste was reduced under pressure. In addition, the distance between carbon fibers was shortened, leading to the change of conducting modes from the free ions in the cement paste to the electrons in carbon fibers. As a result, the conductivity of CFRC was greatly enhanced. 4.4. Influences of curing age on the conductivity of CFRC The hydration of cement is a long process. With the increase of the curing period, the resistivity of the material is bound to change. To

demonstrate the resistivity of the CFRC composites varies with the increase of the curing period, 6 sets of samples were prepared to test the resistivity at different curing periods. The results are shown in Fig. 10. It can be seen from Fig. 10 that when the carbon fiber content lay in between 0% and 0.6%, the resistivity before 28 days increased sharply. After 28 days, the resistivity still grew, but the growth rate was slow. After 84 days, the resistivity as a whole was relatively stable. When the carbon fiber content ranged from 0.8% to 1%, the resistivity before 7 days increased rapidly, after which it still grew, but the growth rate was slow. After 84 days, the resistivity was generally relatively stable. It is analyzed that, when the carbon fiber content was between 0% and 0.6%, the water in the free state in CFRC was gradually reduced with the continuous hydration of cement clinker. The free water was converted into adsorption water or gel water of larger dielectric constant. The interface between cement and fiber transformed gradually into solid-solid contact. The interface layer on the fiber surface became thicker. The conductive pathway in the composites was blocked. The conductivity by way of ions was weaker. The absolute barrier between fibers became larger. Under such a circumstance, it was difficult or even impossible for the electrons to jump by tunneling [36,55]. When the hydration reaction was fully accomplished (after 28 days), the condensation body structure in the internal of the composites was basically formed and the concentration of the free ions was relatively stable. In this case, the conductivity of the composites was stable, in other words, the resistivity changed little along with the prolonging of the curing age. At higher percentage of carbon fibers (over 0.6%), the conductivity of CFRC was slightly affected by the curing age. This is because when the carbon fiber content reached a certain value the distance between fibers was narrowed. So, it is easier for the electrons to jump through tunneling. Meanwhile, some fibers were closely overlapped together through which electrons transferred by carbon fibers to carry out conductivity. Thus, the conductivity of CFRC was hardly influenced by the curing age. Therefore, the resistivity varied slightly. 4.5. Influences of silica fume content on the resistivity of CFRC samples

Fig. 10. Variation of the resistivity of CFRC samples with the curing days.

The addition of silica fume in the CFRC composites exerts not only a good effect on the dispersion of carbon fibers in the cement matrix, but can also increase the density of cement-based materials, which further influence the conductivity of the CFRC. Table 5 shows the samples with different amount of silica fume. The samples are numbered with A1, A2, A3, A4, B1, B2, B3, B4, C1, C2, C3 and C4. 15129

Ceramics International 43 (2017) 15122–15132

W. Chuang et al.

Table 5 Mix ratio of CFRC samples with different amount of silica fume. Test no.

Cement/g

Carbonfiber/%

Water/g

Standardsand/g

Hydroxyethyl cellulose /% (Aqueous solution)

Silica fume /%

Water-reducing agent/ %

Defoaming agent /%

A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 C3 C4

450 450 450 450 450 450 450 450 450 450 450 450

0 0 0 0 0.2 0.2 0.2 0.2 0.6 0.6 0.6 0.6

225 225 225 225 225 225 225 225 225 225 225 225

1350 1350 1350 1350 1350 1350 1350 1350 1350 1350 1350 1350

0 0 0 0 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4

0 5 10 15 0 5 10 15 0 5 10 15

0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8

0 0 0 0 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01

Table 6 Variation of resistivity of CFRC samples with the silica fume content. Test no.

Carbon fiber/ %

Silica fume /%

Resistivity /Ω·m

A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 C3 C4

0 0 0 0 0.2 0.2 0.2 0.2 0.6 0.6 0.6 0.6

0 5 10 15 0 5 10 15 0 5 10 15

265 273 296 326 195 201 178 168 22 26 21 20

Variation range /%

3.0 11.7 23.0 3.1 −8.7 −13.8 18.2 −4.5 −9.1

Cement, water, standard sand, and water-reducing agent used in all samples were fixed and they were 450 g, 225 g, 1350 g, and 0.8%, respectively. The content of carbon fibers in A1…A4 was 0%, in B1…B4 was 0.2%, and in C1…C4 was 0.6%. Hydroxyethyl cellulose in A1…A4 was 0%, in B1…B4 and C1…C4 was 0.4%. Silica fume in A1…A4, B1… B4, and C1…C4 was successively 0%, 5%, 10%, and 15%, respectively. None of the defoaming agent was added in A1…A4, but was used in 0.01% in B1…B4 and C1…C4. After the CFRC samples were prepared, they were placed in water for curing until the 28th day. The resistance was measured, according to which the resistivity was calculated. The results are shown in Table 6. It can be seen from Table 6 that the resistivity of the standard cement mortar sample without any carbon fibers increased gradually with the increase of the content of silica fume, namely, 265, 273, 296, and 326 Ω.m. When the content of silica fume was 15%, the resistivity increased by 23%. When the carbon fiber content was 0.2%, the resistivity of the CFRC composites was increased firstly and then decreased. When the content of silica fume was 5%, the resistivity reached the maximum value 201 Ω.m or it was increased by 3.1%. It is interesting that, when the content of silica fume was over 5%, the resistivity was decreased with the increase of the content of silica fume. When the content of silica fume rose up to 15%, the resistivity was decreased by 13.8%, compared to the resistivity without any silica fume. When the carbon fiber content was 0.6%, the resistivity of CFRC composites was increased first and then decreased. When the content of silica fume was still 5%, the resistivity reached the maximum value 26 Ω.m, which was increased by 18.2% compared to that without any silica fume. When the content of silica fume was more than 5%, the resistivity started to decrease: at the content of silica fume 10%, the resistivity was 21 Ω•m; at the content of 15%, it was 20 Ω•m. It can be

seen here that the resistivity was decreased with the increase of the content of silica fume, but reduced very slightly. The reason may be ascribed to the conductive ways. When there were none of carbon fibers in the cement mortar, the conductivity was carried out through ions migration by pores in the gel matrix. When silica fume was incorporated into the cement mortar, the number of pores was reduced and the density of the composite was increased. As a result, it is more difficult for ions to migrate, leading to the increase of the resistance of the composite, so, the resistivity was increased as well. When the carbon fiber content was 0.2%, the incorporation of silica fume into cement mortar improved the dispersion of carbon fibers, which make the resistivity of the CFRC composites decreased. However, when the carbon fiber content rose up to 0.6%, the resistivity changed hardly with the content of silica fume. This is because a dense overlapped conductive network of carbon fibers had been already formed in the CFRC composites, the resistivity tended to be a stable value. Under such a circumstance, the incorporation of silica fume can improve the dispersion of carbon fibers, but exerts little effect on the resistivity [43,47]. 5. Conclusions The dispersion of carbon fibers in the cement matrix greatly influenced the conductivity of CFRC composites. Only when carbon fibers were well dispersed, it was meaningful to investigate the conductivity. Major conclusions were drawn as follows: (1) The dispersion of carbon fibers in achieving homogenized CFRC composites fell into two steps: the first step is that they were dispersed evenly in the aqueous solution; the second step is that the achieved disperse system was uniformly distributed in the cement matrix. In the first stage, the mass fraction of hydroxyethyl cellulose lay in between 1.66% and 1.86%. (2) The mechanical performances of CFRC increased with the increasing of carbon fiber contents before the percentage 0.6%. The compressive strength increased by 20% at 0.6%, the modulus increased by 1.29%, whereas the flexural stress decreased by 0.89%. (3) Carbon fibers were randomly distributed in the cement matrix in a three-dimensional ways. They contacted and overlapped each other to form a conductive network, giving CFRC composites good conductivity. When the water/cement ratio, carbon fiber content and molding process were fixed, the conductivity of CFRC increased with the increase of carbon fiber length. Generally, the length is 4–8 mm, better not exceeding 10 mm. (4) The resistivity of CFRC composites decreased with the increase of carbon fiber content. It tended to be constant when the carbon fiber content was more than 0.6%, which is the optimum value. Selecting an appropriate water/cement ratio and pressing forming

15130

Ceramics International 43 (2017) 15122–15132

W. Chuang et al.

process can reduce the resistivity of CFRC. When the carbon fiber content was 0.6% and the water/cement ratio was 0.3, the minimum resistivity 3.6 kΩ.cm occurred. When the carbon fiber content was 0.2% and 0.4% and the pressing forming process was adopted, the resistivity was reduced by 50% and 90%, respectively. (5) By comparing the percentages of major elements O, Ca, and C in the standard cement mortar with those in CFRC, it was found that they varied greatly, the variation values being all over 90%. The percentages of weight variation were 103%, 130%, and 94%, and the corresponding atomic percentages were 104%, 132%, and 95%. Acknowledgments This work was financially supported by Shaanxi Province Engineering Laboratory of High Performance Concrete with the grant No. G2015-03 and No. G2015-05 as well as by Weinan Basic Research Innovation Talent Project with the grant No. 2015KYJ-3-3. It was also partially supported by the National Natural Science Foundation of China under Grant No. 51472202 and 51521061. The authors would also like to thank Prof. Li Kezhi of Northwestern Polytechnical University for his experimental support. References [1] D.D.L. Chung, Cement reinforced with short carbon fibers: a multifunctional material, Composites: B 31 (6–7) (2000) 511–526. [2] P. Garcés, J. Fraile, E. Vilaplana-Ortego, D. Cazorla-Amorós, E.G. Alcocel, L.G. Andión, Effect of carbon fibres on the mechanical properties and corrosion levels of reinforced portland cement mortars, Cem. Concr. Res. 35 (2) (2005) 324–331. [3] Farhad Reza, Jerry A. Yamamuro, Gordon B. Batson, Electrical resistance change in compact tension specimens of carbon fiber cement composites, Cem. Concr. Compos. 26 (7) (2004) 873–881. [4] Victor C. Li, Karthikeyan Hobla, Effect of fiber length variation on tensile properties of carbon fiber cement composite, Compos. Eng. 4 (9) (1994) 947–964. [5] Dragos-Marian Bontea, D.D.L. Chung, G.C. Lee, Damage in carbon fiber-reinforced concrete, monitored by electrical resistance measurement, Cem. Concr. Res. 30 (4) (2000) 651–659. [6] C. Wang, K.-Z. Li, H.-J. Li, G.-S. Jiao, J. Lu, D.-S. Hou, Effect of carbon fiber dispersion on the mechanical properties of carbon fiber-reinforced cement-based composites, Mater. Sci. Eng: A 487 (1–2) (2008) 52–57. [7] Binmeng Chen, Bo Li, Yan Gao, Tung-Chai Ling, Zeyu Lu, Zongjin Li, Investigation on electrically conductive aggregates produced by incorporating carbon fiber and carbon black, Constr. Build. Mater. 144 (7) (2017) 106–114. [8] Jian Wei, Qian Zhang, Lili Zhao, Lei Hao, Chunli Yang, Enhanced thermoelectric properties of carbon fiber reinforced cement composites[J], Ceram. Int. 42 (10) (2016) 11568–11573. [9] Bing Chen, Keru Wu, Wu Yao, Conductivity of carbon fiber reinforced cementbased composites, Cem. Concr. Compos. 26 (4) (2004) 291–297. [10] Ming-qing Sun, Jun Li, Ying-jun Wang, Xiao-yu Zhang, Preparation of carbon fiber reinforced cement-based composites using self-made carbon fiber mat, Constr. Build. Mater. 79 (3) (2015) 283–289. [11] Min Chen, Peiwei Gao, Fei Geng, Lifang Zhang, Hongwei Liu, Mechanical and smart properties of carbon fiber and graphite conductive concrete for internal damage monitoring of structure, Constr. Build. Mater. 142 (7) (2017) 320–327. [12] R.K. Graham, B. Huang, X. Shu, E.G. Burdette, Laboratory evaluation of tensile strength and energy absorbing properties of cement mortar reinforced with microand meso-sized carbon fibers, Constr. Build. Mater. 44 (2013) 751–756. [13] F. Azhari, N. Banthia, Cement-based sensors with carbon fibers and carbon nanotubes for piezoresistive sensing, Cem. Concr. Compos. 34 (7) (2012) 866–873. [14] O. Galao, F.J. Baeza, E. Zornoza, P. Garcés, Strain and damage sensing properties on multifunctional cement composites with CNF admixture, Cem. Concr. Compos. 46 (2) (2014) 90–98. [15] Enrique García-Macías, Antonella D'Alessandro, Rafael Castro-Triguero, Domingo Pérez-Mira, Filippo Ubertini, Micromechanics modeling of the electrical conductivity of carbon nanotube cement-matrix composites, Compos. B: Eng. 108 (1) (2017) 451–469. [16] Manuela Chiarello, Raffaele Zinno, Electrical conductivity of self-monitoring CFRC, Cem. Concr. Compos. 27 (4) (2005) 463–469. [17] J. Xu, W.H. Zhong, W. Yao, Modeling of conductivity in carbon fiber-reinforced cement-based composite, J. Mater. Sci. 45 (13) (2010) 3538–3546. [18] Sung-Hwan Jang, Daniel Peter Hochstein, Shiho Kawashima, Huiming Yin, Experiments and micromechanical modeling of electrical conductivity of carbon nanotube/cement composites with moisture, Cem. Concr. Compos. 77 (3) (2017) 49–59. [19] Ming Jin, Linhua Jiang, Mengting Lu, Shuya Bai, Monitoring chloride ion penetration in concrete structure based on the conductivity of graphene/cement

composite, Constr. Build. Mater. 136 (4) (2017) 394–404. [20] D.D.L. Chung Sihai Wen, Self-sensing of flexural damage and strain in carbon fiber reinforced cement and effect of embedded steel reinforcing bars, Carbon 44 (8) (2006) 1496–1502. [21] F.J. Baeza, O. Galao, E. Zornoza, P. Garcés, Effect of aspect ratio on strain sensing capacity of carbon fiber reinforced cement composites, Mater. Des. 51 (10) (2013) 1085–1094. [22] Panjasil Payakaniti, Supree Pinitsoontorn, Prasit Thongbai, Vittaya Amornkitbamrung, Prinya Chindaprasirt, Electrical conductivity and compressive strength of carbon fiber reinforced fly ash geopolymeric composites, Constr. Build. Mater. 135 (3) (2017) 164–176. [23] Baoguo Han, Liqing Zhang, Chenyu Zhang, Yunyang Wang, Xun Yu, Jinping Ou, Reinforcement effect and mechanism of carbon fibers to mechanical and electrically conductive properties of cement-based materials, Constr. Build. Mater. 125 (10) (2016) 479–489. [24] Egemen Teomete, Measurement of crack length sensitivity and strain gage factor of carbon fiber reinforced cement matrix composites, Measurement 74 (10) (2015) 21–30. [25] Faezeh Azhari, Nemkumar Banthia, Cement-based sensors with carbon fibers and carbon nanotubes for piezoresistive sensing, Cem. Concr. Compos. 34 (7) (2012) 866–873. [26] Panagiota T. Dalla, Konstantinos G. Dassios, Ilias K. Tragazikis, Dimitrios A. Exarchos, Theodore, E. Matikas, Carbon nanotubes and nanofibers as strain and damage sensors for smart cement, Mater. Today Commun. 8 (9) (2016) 196–204. [27] Jie Gao, Zhenjun Wang, Ting Zhang, Liang Zhou, Dispersion of carbon fibers in cement-based composites with different mixing methods, Constr. Build. Mater. 134 (3) (2017) 220–227. [28] Jiang Cuixiang, Zhong Dongwang, hen Hao, Interface resistance between carbon fiber and cement, Procedia Eng. 15 (2011) 5333–5337. [29] Zhenjun Wang, Jie Gao, Tao Ai, Wei Jiang, Peng Zhao, Quantitative evaluation of carbon fiber dispersion in cement based composites, Constr. Build. Mater. 68 (10) (2014) 26–30. [30] Ke-zhi Li, Chuang Wang, He-jun Li, Xin-tao Li, Hai-bo Ouyang, Jian Wei, Effect of chemical vapor deposition treatment of carbon fibers on the reflectivity of carbon fiber-reinforced cement-based composites, Compos. Sci. Technol. 68 (5) (2008) 1105–1114. [31] V. Afroughsabet Mahmoud Nili, Combined effect of silica fume and steel fibers on the impact resistance and mechanical properties of concrete, Int. J. Impact Eng. 37 (8) (2010) 879–886. [32] V.T. Giner, F.J. Baeza, S. Ivorra, E. Zornoza, O. Galao, Effect of steel and carbon fiber additions on the dynamic properties of concrete containing silica fume, Mater. Des. 34 (2) (2012) 332–339. [33] F. Koksal, F. Altun, I. Yigit, Y. Sahin, Combined effect of silica fume and steel fiber on the mechanical properties of high strength concretes, Constr. Build. Mater. 22 (2008) 1874–1880. [34] Hongyu Shao, Binmeng Chen, Bo Li, Shengwen Tang, Zongjin Li, Influence of dispersants on the properties of CNTs reinforced cement-based materials, Constr. Build. Mater. 131 (1) (2017) 186–194. [35] Jian Wei, Qian Zhang, Lili Zhao, Lei Hao, Chunli Yang, Enhanced thermoelectric properties of carbon fiber reinforced cement composites [J], Ceram. Int. 42 (1) (2016) 11568–11573. [36] Wei-wen Li, Wei-ming Ji, Guo-hao Fang, Yu-qing Liu, Feng Xing, Yu-kai Liu, Biqin Dong, Electrochemical impedance interpretation for the fracture toughness of carbon nanotube/cement composites, Constr. Build. Mater. 114 (7) (2016) 499–505. [37] P.F.G. Banfill, G. Starrs, G. Derruau, W.J. McCarter, T.M. Chrisp, Rheology of low carbon fibre content reinforced cement mortar, Cem. Concr. Compos. 28 (9) (2006) 773–780. [38] Inkyu Rhee, Jun Seok Lee, Yoong Ahm Kim, Jin Hee Kim, Ji. Hoon Kim, Electrically conductive cement mortar: incorporating rice husk-derived high-surface-area graphene, Constr. Build. Mater. 125 (10) (2016) 632–642. [39] Farhad Reza, Jerry A. Yamamuro, Gordon B. Batson, Electrical resistance change in compact tension specimens of carbon fiber cement composites, Cem. Concr. Compos. 26 (7) (2004) 873–881. [40] D.D.L. Chung Jingyao Cao, Electric polarization and depolarization in cementbased materials, studied by apparent electrical resistance measurement, Cem. Concr. Res. 34 (3) (2004) 481–485. [41] Luo Lailong, Tunnel current in carbon fiber reinforced concrete (in Chinese), Acta Phys. Sin. 54 (6) (2005) 2540–2544. [42] Sihai Wen, D.D.L. Chung, Double percolation in the electrical conduction in carbon fiber reinforced cement-based materials, Carbon 45 (3) (2007) 263–267. [43] A.O. Monteiro, P.B. Cachim, P.M.F.J. Costa, Electrical properties of cement-based composites containing carbon black particles, Mater. Today: Proc. 2 (1) (2015) 193–199. [44] H.K. Kim, I.W. Nam, H.K. Lee, Enhanced effect of carbon nanotube on mechanical and electrical properties of cement composites by incorporation of silica fume, Compos. Struct. 107 (1) (2014) 60–69. [45] Nataliya Kalashnyk, Eric Faulques, Jan Schjødt-Thomsen, Lars R. Jensen, Jens Chr M. Rauhe, Ryszard Pyrz, Monitoring self-sensing damage of multiple carbon fiber composites using piezoresistivity, Synth. Met. 224 (2) (2017) 56–62. [46] Jing Xu, Wu Yao, Ruiqing Wang, Nonlinear conduction in carbon fiber reinforced cement mortar, Cem. Concr. Compos. 33 (3) (2011) 444–448. [47] M.ònica Ardanuy, Josep Claramunt, Romildo Dias Toledo Filho, Cellulosic fiber reinforced cement-based composites: a review of recent research, Constr. Build. Mater. 79 (3) (2015) 115–128.

15131

Ceramics International 43 (2017) 15122–15132

W. Chuang et al. [48] Baomin Wang, Zhiqiang Guo, Yu Han, Tingting Zhang, Electromagnetic wave absorbing properties of multi-walled carbon nanotube/cement composites, Constr. Build. Mater. 46 (2013) 98–103. [49] D.D.L. Chung Sihai Wen, The role of electronic and ionic conduction in the electrical conductivity of carbon fiber reinforced cement, Carbon 44 (11) (2006) 2130–2138. [50] Josef Foldyna, Vladimír Foldyna, Michal Zeleňák, Dispersion of carbon nanotubes for application in cement composites, Procedia Eng. 149 (2016) 94–99. [51] Manuel Hambach, Hendrik Möller, Thomas Neumann, Dirk Volkmer, Carbon fibre reinforced cement-based composites as smart floor heating materials, Compos. B: Eng. 90 (4) (2016) 465–470. [52] R.D. Verástegui-Flores, G. Di Emidio, Impact of sulfate attack on mechanical

properties and hydraulic conductivity of a cement-admixed clay, Appl. Clay Sci. 101 (9) (2014) 490–496. [53] M. Pokharel, M. Fall, Combined influence of sulphate and temperature on the saturated hydraulic conductivity of hardened cemented paste backfill, Cem. Concr. Compos. 38 (4) (2013) 21–28. [54] Dawid Janas, Stefanie K. Kreft, Krzysztof K.K. Koziol, Printing of highly conductive carbon nanotubes fibres from aqueous dispersion, Mater. Des. 116 (2) (2017) 16–20. [55] S. Nasri, M. Megdiche, M. Gargouri, DC conductivity and study of AC electrical conduction mechanisms by non-overlapping small polaron tunneling model in LiFeP2O7 ceramic, Ceram. Int. A 42 (1) (2016) 943–951.

15132