Evaluation of the durability and shielding properties of high-strength concrete incorporating locally available materials and carbon additives

Abstract

High-strength concrete (HSC) has been extensively studied for major applications, particularly in high-rise buildings and bridges. Furthermore, conductive concrete exhibits enhanced electrical properties, enabling its use in specialized applications such as heating, electromagnetic shielding, and antistatic systems. This study aims to combine these two types to produce conductive high-strength concrete (CHSC) using locally available materials. The proposed mixes’ performance measurements include compressive strength, flexural strength, modulus of elasticity, long-term effects (like creep and shrinkage), and electromagnetic shielding. HSC with compressive strength up to 100 MPa and modulus of rupture of 8.96 MPa was successfully produced using locally available materials. The mix containing low dune sand achieved the best mechanical performance. The inclusion of steel fibers (FLDUNE) increased 28-day compressive and flexural strengths by 3.5 and 22%, respectively, and reduced shrinkage and creep by 25 and 10%. However, the addition of carbon materials along with steel fibers (FCLDUNE) decreased these strengths by 19.6 and 15%, and increased shrinkage and creep by 18 and 9%, respectively. The measured electrical properties of the proposed concrete mixes show resistivity of 33.3 Ω-m for FLDUNE and 25.7 Ω-m for FCLDUNE mixes and measured relative complex permittivity of 22.57-j2.22 with relative complex permittivity of 22.57-j2.22. Electromagnetic shielding showed an attenuation of -70 dBm using steel fiber, compared to −28.3 dBm for the control mix. Adding carbon material with steel fiber did not provide significant added value for shielding applications.

Introduction

The rapid growth of technological advancements and global development accelerates, increasing the demand for novel, multifunctional structural materials. Therefore, the new focus started to shift away from conventional concrete to high-strength concrete (HSC)¹. HSC can be utilized in challenging structures such as tall buildings and bridges due to smaller cross-sections and less dead load². Furthermore, conductive concrete is an emerging material that became popular in the late 1990s. Conductive concrete is a mixture of conventional concrete and conductive materials such as metal powder, graphite, steel fibers, steel shavings, carbon fiber, etc.³. It can be used as a strain/stress sensor⁴, an electromagnetic radiation reflector for shielding against electromagnetic interference (EMI)^5–7, a self-sensing material for tracking the structural health^8–10, and a resistance material in heated pavement^11,12.

Several studies in the literature have investigated the effects of different construction materials on the mechanical strength, time-dependent deformation, and electrical conductivity of concrete:

Effect of fibers

The use of fiber in concrete has a significant effect on its mechanical behavior, shrinkage, creep, and electrical conductivity¹³. The mechanical properties of concrete are significantly enhanced with the addition of fibers, including tensile strength, flexural strength, and impact resistance. Fibers improve post-peak behavior by increasing ductility and toughness, which prevents sudden brittle failure, although they have little influence on compressive strength¹⁴. The overall impact of fibers depends on their type, dosage, and distribution. Jang and Yun¹⁵ examined the effects of different volumetric ratios of hooked-end steel fibers on the properties of HSC. They concluded that adding 1.5% steel fibers by volume significantly increased the flexural strength and post-peak behavior of HSC. Afroughsabet and Ozbakkaloglu¹⁶ conducted a similar experiment, concluding that a 1% dosage of hooked-end steel fibers would significantly enhance the flexural and tensile properties of concrete.

Shrinkage in concrete often causes crack formation due to restricted volume changes. This effect is more critical in HSC and ultra-high-strength concrete (UHSC) due to the low water content, which results in autogenous shrinkage^17–19. Autogenous shrinkage becomes significant in HSC due to the internal water consumption during hydration, which leads to self-desiccation and volumetric contraction²⁰. According to Mohebbi et al.²¹, the shrinkage of UHSC exceeds the shrinkage levels commonly observed in conventional concrete. Wu et al.²² reported that autogenous shrinkage in HSC can be reduced using low heat cement, fly ash, shrinkage reducing agents, lightweight aggregates, and fibers. Fibers can mitigate shrinkage-induced cracking by bridging microcracks and distributing tensile stresses¹⁴. The strong bond of fibers with the concrete matrix and its high tensile strength improves internal confinement, limiting shrinkage strains and enhancing durability²³. However, concrete creep is defined as the deformation of concrete over a period of time due to sustained loading²⁴. According to Su et al.²⁵, concrete creep is specifically related to the increase in deformation of concrete that occurs under constant or sustained loads, distinct from the initial or instantaneous deformation that happens during the loading process. Concrete creep leads to prestressing losses, which can affect the integrity of concrete infrastructures and can lead to expensive maintenance and rehabilitation of structures. The addition of fibers to concrete has a considerable impact on its creep behavior. Fibers reduce long-term deformations under sustained loading by enhancing load transfer within the concrete matrix²⁶.

On the other hand, fibers have been the subject of ongoing research because of their conductivity, which is highly beneficial for enhancing the electrical properties of concrete. Yehia et al.²⁷ successfully developed a conductive concrete mix that was able to attenuate electromagnetic signals by 50 dB using steel fiber as a conductive material. Steel wool and steel fibers were employed by Liu et al.²⁸ as conductive ingredients in asphalt concrete. By using conductive materials of various lengths and widths, they demonstrated that while long fibers with small diameters improve electrical conductivity, they also reduce concrete strength. A conductive concrete mixture containing steel fibers and shavings was created by Christopher et al.²⁹ for the purpose of deicing bridge decks. The mixture exhibited a compressive strength of 31 MPa, a heating rate of 0.14 °C/min, and thermal power density of 590 W/m². According to Sassani et al.³⁰, the conductivity of concrete is enhanced by using carbon fibers with an aspect ratio of 833 and a dosage of 0.75% by volume. Furthermore, Shishegaran et al.³¹ used a combination of steel powder and steel wire rope with optimal percentages of 5.8% and 3% by volume, respectively, to improve conductivity.

Effect of fine materials

The incorporation of supplementary cementitious materials (SCMs), such as ground granulated blast furnace slag (GGBS), silica fume (SF), and fly ash (FA), in addition to fine aggregate, in concrete significantly affects its mechanical performance, conductivity, and time-dependent properties. The concrete compressive strength and durability are enhanced as a result of using the fine materials, which fill voids, lowering porosity, and improving the microstructure. For example, the utilization of GGBS in concrete improves the long-term compressive and flexural strengths and slightly promotes the elastic modulus. In addition, the use of GGBS can enhance the concrete durability by reducing drying shrinkage and creep, reducing water, gas, improving abrasion resistance and chloride ion permeability, and significantly improving sulfate and alkali silica reactions^32,33. Lukowski and Salih³⁴ reported that the inclusion of GGBS in concrete can reduce the porosity due to its greater surface area. Additionally, through the development of conductive networks inside the concrete, the inclusion of GGBS, which contains iron oxide, can improve electrical conductivity³⁵. Also, SF improves the mechanical properties of concrete as a result of its pozzolanic reaction with calcium hydroxide, which forms extra calcium silicate hydrate (C-S-H) gel³⁶. In terms of conductivity, SF decreases porosity and refines the pore structure, creating a denser microstructure that restricts ion and electron movement and lowers electrical conductivity³⁷. On the other hand, because the pozzolanic reaction of FA is usually a slow process, the enhancement it gives to the strength and microstructure of concrete is mainly seen at later ages³⁸. According to Saha³⁹, the dry shrinkage decreases with the increase in FA content in concrete. The properties and content of fine aggregate, on the other hand, affect concrete’s shrinkage, creep, and overall durability. By decreasing voids and improving aggregate interlock, a well-graded sand helps create a denser microstructure, which reduces drying shrinkage and creep. Conversely, higher shrinkage and long-term deformation might result from sands with a high fine content or poor grading, which raises water consumption⁴⁰. The type of fine aggregate used in mixtures can also impact concrete conductivity. The use of conductive materials, such as magnetite sand, in the concrete mix has been shown to enhance electrical conductivity. Ren et al.⁴¹ revealed that adding nanographite and magnetite sand to concrete can significantly increase the conductivity of concrete.

Effect of carbon-based fillers

The incorporation of carbon-based materials into concrete increases its conductivity. The study by Khalid et al.⁴² showed that the conductive concrete enhanced with graphite powder, carbon black, and steel fibers achieves superior electromagnetic shielding effectiveness. Electromagnetic shielding receives high importance in modern society due to the rapid growth of wireless communication systems, high-power radar, electric vehicles, and sensitive electronic infrastructure. Uncontrolled EMI can disrupt critical equipment, compromise data security, affect human health, and even damage electronics through electromagnetic pulses. Electromagnetic shielding techniques are used to reduce or block the penetration of electromagnetic fields into a certain region by using a conductive or magnetic material barrier. Shielding is achieved via three mechanisms: reflection, absorption, and multiple internal reflections of wave energy within a conductive material. Common techniques include the use of highly conductive metals (copper, aluminum), magnetic alloys, conductive coatings, paints, and carbon- or metal-fiber-loaded polymers. Multifunctional materials such as conductive concrete reinforced with carbon black, graphite, or steel fibers have emerged in applications in electromagnetic shielding. Pyramidal samples outperform carbon-laced polyurethane, which performs 50 dB better, with approximately 65 dB shielding across 1–5.5 GHz. The resistivity of graphite filling asphalt concrete drops from 3.7 × 10¹¹ Ω m to 1.4 × 10³ Ω m with graphite content increasing from 8% to 12.7% according to Wu et al.⁴³. El-Dieb et al.⁴⁴ studied the effects of three different conductive materials, which are steel shavings, carbon powder, and graphite powder. The carbon powder is the best performer because it has the least negative effect on the compressive strength and the durability of conductive concrete. Also, Li et al.⁴⁵ studied the effect of three types of conductive materials (GGBS, steel slag, and graphite powder) on the mechanical and electrical properties of concrete by substituting those materials as percentages of the total cementitious materials by weight. All conductive fillers were found to improve the conductivity of concrete, with graphite powder showing the highest enhancement effect, followed by steel slag and GGBS. The optimized mixture design was found to be 4% graphite powder, 15% GGBS, and 20% steel slag, which yielded a compressive strength of 36 MPa, a flexural strength of 3.4 MPa, and an excellent resistivity of 77.79 Ω-m. On the contrary, the use of carbon materials in concrete resulted in a significant decrease in concrete mechanical performance^46,47. Chaturvedy et al.^48–51 revealed that the incorporation of graphene oxide (GO) in HSC significantly improves its mechanical and durability properties. Small amounts of GO enhance compressive, tensile, and flexural strength, refine the pore structure, and improve workability. These effects are observed even in sustainable concrete mixtures containing recycled rubber. Optimization and modeling studies further confirm that GO can effectively enhance both the structural performance and long-term durability of high-strength concrete.

Many researchers have studied HSC and conductive concrete. It was mentioned that sometimes there are limitations on the use of conductive concrete due to strength requirements. Also, there is little research in the literature when it comes to combining HSC properties and conductive concrete properties to create conductive high-strength concrete (CHSC). This highlights the need for further research to explore the potential of CHSC in various applications.

Research significance

This research aims to develop CHSC for applications where traditional conductive concrete proves impractical. By bridging the gap between HSC and conductive concrete, CHSC can leverage the extensive knowledge of HSC to enhance feasibility and broaden the applications of conductive concrete, potentially leading to a significant breakthrough in the field. Furthermore, since concrete is susceptible to various forms of shrinkage and creep, this study investigates how CHSC performs under these conditions.

Experimental program

Materials

The HSC used in this research was prepared with the following constituents: cement, GGBS, SF, dune sand, crushed aggregate, water, superplasticizer, steel fibers, and carbon materials. Type I ordinary Portland cement (OPC) conforming to BS EN 197-1:2011⁵² was used in all mixes. The properties of the fine and powder materials are summarized in Table 1. The following volumetric ratios were considered during the optimization process: 15–22% of the cementitious materials, 5–10% of silica fume, 20–40% of fine aggregate, 0.5-2% of steel fiber and 0–30% of carbon and/or graphite⁵³. To create a denser matrix and enhance the strength of the concrete, coarse aggregate was omitted from the mixes^54–56.

Table 1.

Material properties.

Material	Specific gravity	Absorption (%)	Moisture content (%)
OPC	3.15	-	-
GGBS	2.9	-	-
SF	2.2	-	-
Crushed dolomite	2.54	1.5	1.14
Dune sand	2.51	1.1	0.57

Hybrid steel fibers with aspect ratios of 40 and 87, along with five different types of carbon products, were utilized in this study. These five types of carbon were used in equal percentages as partial replacements for the fine aggregate. Additionally, superplasticizer Type F, which complies with ASTM C494⁵⁷ and has a specific gravity of 1.15, was employed.

Mix proportions and specimens’ Preparation

Initially, a mix was adopted based on the literature review as well as the previous experience of the researchers. This mix was defined as very high dune content (VHDUNE). In addition, the mix was checked against Fuller’s model, which provides a theoretical optimal packing density for concrete⁵⁸. However, a big gap was observed between Fuller’s model and the VHDUNE mix, as shown in Fig. 1. This might be attributed to the fact that the mixes in this study do not contain coarse aggregate. Another model, which is the Modified Andreasen and Andersen model (Modified A&A model) for particle packing, which is more applicable to fine mixes, was checked⁵⁹. A smaller gap in comparison was found between the VHDUNE mix and the A&A model, as presented in Fig. 1. Nevertheless, an attempt was made to bring the gradation curve closer to the Modified A&A model. Changing the aggregate-to-binder ratio did not induce a significant change to the gradation curve. However, changing the absolute volume of the crushed aggregate and dune sand while keeping the total fine aggregate volume constant changed the shape of the gradation curve. Four additional mixes to VHDUNE were introduced and were plotted against the modified A&A model. VHDUNE had the biggest gap, followed by HDUNE, MDUNE, LDUNE, and VLDUNE with respect to the Modified A&A as shown in Fig. 1. The quantities are referred to as very low (VL), low (L), medium (M), high (H), and very high (VH).

Fig. 1.

Gradation of mixes against theoretical models.

Figure 2 provides a comprehensive summary of the experimental work conducted in this investigation. The research was divided into two distinct phases:

Fig. 2.

Flow chart of experimental work.

Phase I: initial pilot mixing and evaluation

This stage focused on the development of the HSC mix, which can later be turned into CHSC by incorporating conductive materials in the mix. Mainly, five mixes were developed and evaluated, each with a different quantity of dune sand (VL, L, M, H, and VH). The volume of fine aggregate (dune sand and crushed sand) was kept constant for all mixes. Then the compressive and flexural strength were tested according to BS EN 12390-3:2019⁶⁰ and ASTM C78M-21 standard⁶¹; respectively, to evaluate and choose the optimal one. It is to be noted that the mix yielding the best performance will be selected for further investigation in the second phase of the experimental program.

Phase II: development of conductive concrete mixes with comprehensive evaluation

In this stage, the mix that achieved the highest compressive and flexural strength was selected from the first phase and replicated for verification. Two other mixes were conducted, one mix containing steel fibers with a volume fraction of 1.5%, and the other containing steel fibers with the addition of carbon products. The 1.5% steel fiber content was chosen based on previous studies as an optimal dosage for improved mechanical performance and adequate workability. The addition of fibers was expected to decrease workability. Also, carbon materials tend to absorb the mixing water, indicating a dry mixture, which requires careful monitoring of the mixing process to eliminate excessive use of fluids to avoid segregation. Therefore, during the mixing process, the addition of water and superplasticizer was critical and was conducted at a reasonable pace. It is important to note that only the mixing water is used, and the desired workability was achieved by the addition of the superplasticizer. For each mix, several mechanical, time-dependent, and electrical properties are investigated. Compressive strength, flexure strength, modulus of elasticity, shrinkage, creep, resistivity, and electromagnetic signal attenuation were evaluated. Then, the results are compared to recommend an optimal CHSC mix, which is prepared using local materials. Table 2 provides details on the testing conducted in this phase.

Table 2.

Extensive evaluation of experimental program.

Measured property	Sample (cm)	Quantity	Testing events (day)	Testing standard
Compressive strength	10 × 10 × 10 cube	9	3, 7, and 28	BS EN 12,390- 3:201960
Modulus of rupture	10 × 10 × 50 Prism	3	28	ASTM C78M-22 for plain61, ASTM C1609/C1609M-2462
Modulus of elasticity	30 × 15 Cylinder	3	28	ASTM C469M-2263
Shrinkage	30 × 15 Cylinder	3	Continuous Non-Destructive	-
Creep	30 × 15 Cylinder	3	Continuous Non-Destructive	ASTM C512M-2464
Electromagnetic signal attenuation	30 × 30 × 1.5 30 × 30 × 1.0 Slab–plane and corrugated	4	28 (Non-Destructive)	-
Resistivity of concrete	30 × 30 × 10 Slab– plane	4	28 days (Non-Destructive)	ASTM C1876-1965

Cubes and prisms were used to evaluate compressive strength and flexural strength, respectively. Cylinders were prepared to test the modulus of elasticity, shrinkage, and creep. The elasticity modulus was obtained by placing strain gauges vertically at the side face of the cylinder and testing under compressive axial loading. Also, the shrinkage test had strain gauges attached to the surface as well, and continuous readings were taken to measure the deformation. The creep test, as shown in Fig. 3, was conducted on cylinders placed in a creep testing rig. Two magnetic bases were placed on top of the bottom plate of the rig and near the top of the cylinder head to hold the strain dial in place. The dial gauge is used to measure the deformation of the concrete specimens during the test. The readings were taken at different time intervals. Mainly hourly, daily, weekly, and then monthly, following the ASTMC512⁶⁴ recommendations. It is to be noted that sulphur capping (Fig. 4) was done on cylinders that were tested in compression to ensure smooth surface contact during testing. Before being exposed to drying conditions or loading, the shrinkage and creep specimens were water-cured for 28 days under standard conditions (20 ± 2 °C and 95 ± 5% relative humidity). The creep tests were initiated at 28 days, consistent with standard practice, to represent long-term structural behavior after most early-age effects have diminished. About 40% of the concrete’s 28-day compressive strength was represented by the continuous load applied during the creep test. In compliance with applicable standards, this stress level was selected to prevent cracking or nonlinear deformation while remaining within the elastic range. The same environmental conditions were applied to each specimen for the duration of the long-term monitoring process. This was carried out in order to preserve consistency and guarantee accurate comparisons between the two tests. The specimens were kept in a controlled indoor environment with a relative humidity of 60 ± 5% and a temperature of 20 ± 2 °C.

Fig. 3.

Creep test on cylinder.

Fig. 4.

Cylinder Sulphur capping.

Figure 5a shows the resistivity measurement setup for the conductive concrete samples. Two metallic electrodes are embedded in the concrete sample, separated by a distance of 25 cm. The resistivity measurement involves sourcing a voltage (V), and measuring the current (I) flow between the electrodes spaced by a distance (L)⁶⁵. The resistivity is calculated using the cross-sectional area A = w×t of the sample as follows in Eq. (1)⁶⁵:

1

Fig. 5.

(a) The resistivity testing of concrete, (b) Electromagnetic signal attenuation test.

The electromagnetic shielding effectiveness (SE) of the proposed mixes is investigated using the measurement setup shown in Fig. 5b. The SE measures the amount of power attenuation propagated through the slab. The measurement setup consists of two rectangular horn antennas, connected to both the transmitter and receiver, with a sample positioned between them. To reduce wave dispersion, the antennas are positioned 20 cm apart. The transmitter (Agilent 83752 A synthesized sweeper 0.01–20 GHz) generates RF signals covering frequencies from 1 to 10 GHz at power levels of 10 dBm (absolute power relative to 1 milliwatt). The receiving antenna is connected to a spectrum analyzer (Rohde & Schwarz FSL Spectrum Analyzer, 9 kHz-18 GHz) to measure the power received through the sample. The samples included in the study were panels, measuring 30 × 30 cm² with thicknesses of 1 cm and 1.5 cm, which were cast using two types of molds: one with a flat surface and the other with a corrugated surface, as illustrated in (Fig. 6). The SE is given by Eq. (2)²⁷:

2

Fig. 6.

Corrugated panel (a) wood molds with required configuration (b) panels after casting.

where P_in is the power of the incident RF signal on the sample and P_out is the power of the transmitted signal after passing through the shielded sample.

Results and discussions

Phase I results

Figures 7 and 8 show the compressive strength and the modulus of rupture test results, respectively. The results show that the mix with low content of dune sand (LDUNE) had the best performance in terms of compressive strength (99.9 MPa) and modulus of rupture (8.96 MPa). This could be attributed to the dense matrix, which was achieved by adding low dune sand content to achieve the optimum proportion and better performance. The fine particles of dune sand can disturb the normal particle size distribution when utilized excessively, resulting in increased voids and a reduction in concrete density, which negatively affects the concrete’s mechanical properties⁶⁶. Lee et al.⁶⁷ reported that using dune sand up to 10% achieved the highest compressive strength, while the higher content led to a reduction in mechanical characteristics. The results were comparable to what was predicted by the modified A&A model since a better gradation correlates with a higher strength. LDUNE had the best particle distribution according to the modified A&A model, and its strength turned out to be the highest when compared with the other mixes. Except for VLDUNE, the compressive strength decreased as the content of dune sand increased, which was expected by the modified A&A model. In Fig. 1, it was shown that as the content of dune sand increased, the gradation started to sway away from the ideal modified A&A model curve, and as the gradation gets poorer, the strength decreases. The 28-day compressive strength of MDUNE, HDUNE, and VHDUNE was around 10.1, 13.8, and 25.7% less compared to that of LDUNE, respectively. However, the 28-day compressive strength of VLDUNE was 9.1% less than that of the LDUNE, which might be attributed to the bigger gap that was introduced beyond 600 microns on the gradation graph. The modulus of rupture showed a similar trend since it is a function of . The LDUNE exhibits the highest modulus of rupture (8.96 MPa) at age 28 days. In comparison, the VLDUNE, MDUNE, HDUNE, and VHDUNE exhibit a reduction of 8.1, 11.4, 24.8, and 33.8%, respectively, compared to the LDUNE. Figure 8 compares the experimental modulus of rupture results of the mixes with the modulus of rupture predicted by the ACI equation Eq. (3). Figure 8 presents a comparison between the experimental modulus of rupture results and the predicted values derived from Eq. (3), as specified in ACI 318 − 19⁶⁸. It is illustrated that, except for VHDUNE, all mixes had a modulus of rupture that exceeded what was predicted by Eq. (3), and the maximum difference was 28.5% for LDUNE⁶⁸.

3

Fig. 7.

Compressive strength results of phase I.

Fig. 8.

Modulus of rupture results of phase I Vs. ACI equation.

Phase II results

Based on the initial results, LDUNE was chosen as the optimal mix, which served as a basic mix to which conductive materials have been added in the second stage of the experimental program. Two new derivative mixes have been created from LDUNE. The first mix, labelled as FLDUNE, contains 1.5% steel fiber by volume in addition to the regular proportions of LDUNE. The second mix, labelled as FCLDUNE which contained 1.5% steel fibers and 10% of carbon by volume. The mix is created by substituting 10% of the aggregates present in LDUNE with a blend of carbon powder. As discussed in the experimental program, those three mixes (LDUNE, FLDUNE, and FCLDUNE) have been tested for mechanical, time-dependent, and electrical properties, and the results are as follows:

Mechanical properties

Table 3 illustrates the effect of steel fibers and carbon products on the mechanical properties of the LDUNE mix. The FLDUNE achieved a 28-day compressive strength of 103.4 MPa, which is a 3.5% increase compared with LDUNE. This is attributed to the addition of the steel fiber, which improves crack control and allows the concrete to absorb more energy, leading to a slightly higher compressive strength. On the other hand, FCLDUNE achieved a 28-day compressive strength of 80.3 MPa, which is a 19.6% decrease compared with LDUNE. These results agreed with Luo and Wang⁴⁶ and He et al.⁴⁷, which illustrated that the incorporation of carbon materials into cementitious composites can lead to a reduction in compressive strength, which may be attributed to the increased water demand and the poor adhesion and low interlocking with cement paste. Nevertheless, the decrease in strength is reasonable and can be accepted if the increase in the conductive properties of concrete is substantial.

Table 3.

Phase II results.

Mix	’c 3−day (MPa)	’c 7−day (MPa)	’c 28−day (MPa)	r 28−day (MPa)	F600 (MPa)	F150 (MPa)	T150 (J)	Fe150 (MPa)	Rt150 (%)	Ec 28−day (GPa)
LDUNE	54.3	67.1	101.6	7.31	-	-	-	-	-	34.8
FLDUNE	57.6	73.8	103.4	8.90	4.80	2.20	26	3.90	43.8	36.1
FCLDUNE	44.1	54.2	80.3	6.23	3.90	2.05	22	3.30	53.0	30.8

F₆₀₀, F₁₅₀: Residual strengths corresponding to deflections of and , respectively.

T₁₅₀: Toughness, representing the energy absorbed by the concrete specimen up to a net deflection of .

Fe₁₅₀: Average stress value in the concrete specimen up to a deflection of .

Rt₁₅₀: Equivalent flexural strength ratio (Fe_{150/ r 28−day)} x 100.

Since FLDUNE and FCLDUNE contain steel fibers, the standard third-point loading test in accordance with ASTM C78M⁶¹ could not be performed. Instead, ASTM C1609M⁶² was used to assess the flexural strength of FLDUNE and FCLDUNE. ASTM C78M produces only one parameter representing the modulus of rupture, which is the stress in the tension zone (bottom layer) of concrete at the time of failure. However, measuring the flexural strength of fiber-reinforced concrete can be complex due to the fact that the sample continues to deflect even after cracking. Therefore, several parameters can be measured using ASTM C1609M. _r values are summarized in (Table 3). For LDUNE, _r represents the modulus of rupture, while for FLDUNE and FCLDUNE, _r represents the peak strength on the stress-deflection curve. FLDUNE exhibited a pre-cracking flexural strength of 8.90 MPa, compared to 7.31 MPa for LDUNE, showing a notable increase of 22%. This enhancement is attributed to the incorporation of steel fibers, which significantly strengthen the concrete matrix. The fibers increase the material’s overall pre-cracking strength by enhancing its capacity to absorb energy and distribute stresses, which delays the beginning of cracking. In contrast, FCLDUNE achieved a pre-cracking flexural strength of 6.23 MPa, a 15% decrease compared to LDUNE. The reduction could be attributed to the addition of carbon-based components, which could weaken the matrix by causing inadequate bonding with the cementitious phase or poor particle dispersion. The reduced flexural strength may come from the less even distribution of stress caused by the poorly distributed carbon particles inside the cement matrix. This result aligns with the findings of Jayashree and Ganapathy⁶⁹, who reported that the addition of carbon-based materials to concrete negatively impacted its mechanical properties, including flexural strength, due to similar dispersion and bonding issues.

The parameters such as F₆₀₀, F₁₅₀, T₁₅₀, Fe₁₅₀, and Rt₁₅₀ listed in Table 3 are essential for comprehending the material’s post-cracking behavior and serviceability performance in the analysis of FRC. F₆₀₀ and F₁₅₀, which represent residual tensile stresses at deflections of L/600 and L/150, respectively, indicate the concrete’s ability to absorb loads after cracking. In order to evaluate the concrete’s ability to sustain strength under flexural loads during service conditions, a crucial component of long-term durability, these characteristics are necessary. T₁₅₀ provides insights about the toughness of the material by measuring the energy absorbed by the concrete until a net deflection of L/150. High toughness indicates a material’s ability to absorb significant energy before failure, which contributes to improved structural resilience under cyclic loading or impact. Similarly, Rt₁₅₀, the equivalent flexural strength ratio, which assesses the material’s post-crack behavior, is determined using Fe₁₅₀, the average stress value from the start of loading until the net deflection of L/150. Better performance in terms of crack resistance and ductility is indicated by a high Rt₁₅₀ value, which shows that the concrete maintains a reasonably high stress after achieving its peak strength. On the other hand, a lower Rt₁₅₀ highlights brittle post-cracking behavior by showing a sharp decrease in stress following peak strength. Table 3 shows that FCLDUNE had a slightly higher Rt₁₅₀ value than FLDUNE, while FLDUNE had a higher peak flexural strength. This indicates that FCLDUNE has greater energy dissipation capabilities and more stable post-cracking behavior. These results agreed with Shah and Ribakov⁷⁰ highlighted that the incorporation of steel fibers improves post-cracking behavior by enhancing energy absorption and reducing stress drops, resulting in improved toughness and more stable stress profiles under loading. The stress-deflection curves, shown in Fig. 9, further illustrate the differences between the two mixes, with FCLDUNE showing a more gradual decrease in stress after peak strength, which indicates a more ductile post-crack response. The overall analysis of these parameters reinforces the importance of both peak strength and post-cracking behavior in assessing the performance of FRC, particularly in applications where serviceability and durability are critical.

Fig. 9.

Stress vs. deflection for FCLDUNE and FLDUNE mixes.

The modulus of elasticity (Ec) was also evaluated in this phase, and the results are shown in Table 3. According to the results, FLDUNE had the highest modulus of elasticity, exceeding LDUNE by about 4%. Yehia et al.⁷¹ reported that fibers achieved a small increase in modulus of elasticity in the range of 6–13% compared to the control mix. This could be attributed to the steel fiber’s reinforcing effect. Steel fibers act as internal reinforcements, bridging microcracks and contributing to a more cohesive matrix, resulting in a slight increase in stiffness and overall structural performance⁷². FCLDUNE, on the other hand, performed 11% less than LDUNE. The high surface area of fine carbon particles may raise the water demand during mixing, which could result in less workability. Furthermore, carbon particles may not be able to bridge microcracks effectively due to their small size and distribution, which would restrict their ability to increase strength and stiffness⁷³. Figure 10 presents a comparison between the experimental modulus of elasticity results and the predicted values derived from Eq. (4), as specified in ACI 363 for HSC⁷⁴. The LDUNE, FLDUNE, and FCLDUNE mixes had a modulus of elasticity that is less than the predicted value by 13.7, 11.2, and 16%, respectively. This difference could be due to the absence of coarse aggregates in the mix, which is not accounted for in Eq. (4)⁷⁴.

4

Fig. 10.

Experimental and predicted modulus of elasticity.

Electrical properties and shielding performance

In this study, concrete mixes LDUNE, FLDUNE, and FCLDUNE are referred to as M1, M2, and M3, respectively. The resistivity measurements are investigated on 4 samples of the different mixes; each has a size of 30 × 30 × 1.0 cm³ and two electrodes separated by 25 cm. The average measured resistivity is 33.3 Ω-m for M2 and 25.7 Ω-m for M3, respectively. The way electromagnetic waves interact with a shielding material depends on the material’s electrical properties, specifically its complex permittivity (ε_r) or complex conductivity (σ). These properties govern how deeply the waves can penetrate the shield and the extent of energy loss as the waves pass through. Complex permittivity measures a material’s ability to store and dissipate electric energy, while complex conductivity indicates how well it conducts electric currents. Together, they determine the shield’s effectiveness in blocking or attenuating the waves, with higher losses and shallower penetration indicating better shielding performance. In⁷⁵, the authors developed a multiphase dielectric mixing model for accurately estimating the relative complex permittivity of the proposed concrete mixes. This model uses scattering parameter measurements obtained via a rectangular waveguide in the C-band frequency range, as shown in (Fig. 11). The proposed model accounts for the volume fractions and dielectric properties of individual constituents (e.g., cement, aggregates, and water) in both conventional and conductive concrete mixtures. A comparison between the calculated and measured relative complex permittivity of both conventional and conductive concrete mixtures using the proposed dielectric mixing model and different techniques in the literature is listed in (Table 4). The calculated relative complex permittivity using the proposed model is closely matches the measured values, with a minor discrepancy. More details about the multiphase dielectric mixing model can be found in⁷⁵.

Fig. 11.

The relative complex permittivity measurement setup with a concrete mix block placed inside the waveguide.

Table 4.

Comparison between relative permittivity calculation methods with the measured data⁷⁵.

Method	Concrete type	Calculated (εr)	Measured (εr)	Error in (εr)
Maxwell Garnett method	Normal concrete	5.1756-j0.7595	5.45-j0.3091	0.2744 + j0.4504
	Conductive concrete	7.2565-j1.5839	22.57-j2.22	15.3135-j0.6361
Sihvola and Kong method	Normal concrete	5.1587-j0.9450	5.45-j0.3091	0.2913 + j0.6359
	Conductive concrete	6.9918-j1.6531	22.57-j2.22	15.5782-j0.5669
Proposed dielectric mixing model	Normal concrete	7.7303-j0.3191	5.45-j0.3091	2.2803 + j0.01
	Conductive concrete	20.2310-j3.1522	22.57-j2.22	2.339 + j1.2922

Several key observations regarding electromagnetic shielding were made from the measured data. The test measured the received signal power in dBm, which is not a direct measure of attenuation, but it indicates the achieved attenuation level. The power measurements are represented in dBm scale, where more negative values indicate higher attenuation. For each mix, three types of specimens were tested: a plain 1.5 cm plate, a corrugated 1 cm plate, and a corrugated 1.5 cm plate. Figure 12 compares the results for free space (minimal shielding), a steel conducting plate (maximum shielding), and the three concrete mixes with 1.5 cm thickness when illuminated with the same signal power at different frequencies. Free space acts as a reference medium, demonstrating signal attenuation due to wave propagation between the transmitter and receiver without interference from test samples. The steel plate, being fully conductive, is expected to exhibit the greatest attenuation compared to concrete plates.

Fig. 12.

The attenuation measurements for 1.5 cm thick plates.

As shown in Fig. 12, the received power decreases as the frequency increases. In free space, the power was nearly 0 dBm at 1 GHz, due to minimal wave spreading, but it dropped to -26 dBm at 10 GHz as spreading increased. For the steel plate, the received power started at -20.5 dBm and decreased to -68 dBm at 10 GHz.

For specimens with a thickness of 1.5 cm, the M1 corrugated plate demonstrated the highest measured power of -2.8 dBm at 1 GHz and − 28.3 dBm at 10 GHz, indicating the lowest attenuation. The low attenuation of M1 is attributed to the absence of conductive materials in the mix, which allows electromagnetic waves to pass through. The plain plate exhibited lower power (better attenuation) than the corrugated one because the corrugations reduced the plate’s effective thickness.

For M2, the plain plate showed a lower power of -24.9 dBm at 1 GHz and − 70 dBm at 10 GHz, compared to the corrugated plate with a power of -23.8 dBm at 1 GHz and − 66 dBm at 10 GHz. The inclusion of steel fibers in M2 enhances its conductivity compared to M1. Despite the presence of carbon powder in M3, M2, and M3 exhibited similar attenuation performance. Although M3 outperformed M2 at some frequencies and vice versa, the overall difference was negligible, indicating that the addition of carbon powder added minimal value to attenuation. This limited impact likely stems from the fine particle size and poor dispersion of carbon powder, which fails to form a continuous conductive network within the concrete matrix. Studies have shown that coarse conductive fillers (e.g., carbon fibers or graphite) are more effective in creating such networks, whereas finer powders tend to remain as isolated particles, reducing their efficiency in influencing physical properties such as electrical resistivity or wave attenuation⁷⁶.

For the 1 cm thick plates shown in Fig. 13, the corrugated M1 sample exhibited the lowest performance, which is consistent with its lack of conductive material. The M2 and M3 samples followed trends similar to those observed in the 1.5 cm plates, without leading to new conclusions. Data from Yehia et al.²⁷ confirmed that the developed mixes (M2 and M3) achieved better attenuation compared to previous studies. Figure 14 illustrates the effect of time on the SE of the different mixes, with a thickness of 1.5 cm, after two years of preparation. A reduction in SE is noticeable across all mixes due to the hydration process over time. The SE of M1 has slightly degraded for the plane configuration at most frequencies, whereas a general enhancement of SE for the corrugated configuration is observed at most frequencies. Over the two years, both configurations of M2 and M3 experienced a decrease in SE, with a reduction of approximately 10–20 dBm at low frequencies and slightly less at higher frequencies. These results suggest that the corrugated configuration may provide better long-term shielding performance than the plane configuration across different frequencies.

Fig. 13.

The attenuation measurements for 1 cm thick plates.

Fig. 14.

The SE of different mixes with a thickness of 1.5 cm over time.

Time-dependent properties

The shrinkage and creep strains were recorded over 6 months. As shown in Fig. 15, all three mixes underwent significant shrinkage during the first three weeks of testing. This behavior is consistent with the common pattern of early-age concrete shrinkage, which is largely caused by internal reactions and rapid moisture loss. However, as time passed, the gradient of the curve began to approach zero, indicating a stabilization phase in shrinkage behavior. The results demonstrated that, during the first 21 days of testing, LDUNE, FLDUNE, and FCLDUNE exhibited 62, 70, and 57% of their total 6-month shrinkage strain, respectively. This notable early-age shrinkage is attributed to the high moisture content maintained during the curing process. Capillary tension effects and the evaporation of free water originally led to increased rates of shrinkage by facilitating enhanced hydration of cement particles⁷⁷.

Fig. 15.

Shrinkage results.

As the concrete hardened, the rate of shrinkage gradually decreased. Over the 6 months, the maximum shrinkage strain values recorded were 0.000286 for LDUNE, 0.000215 for FLDUNE, and 0.000338 for FCLDUNE. The results illustrate that FCLDUNE exhibited the highest shrinkage strain, which is attributed to the presence of extremely fine carbon particles. These particles accelerated water loss, which in turn increased drying shrinkage by decreasing the pore structure’s capacity to retain water. This result is consistent with the findings of Du et al.⁷⁸, who examined the impact of particle size on moisture dynamics and early-age shrinkage. In contrast, FLDUNE experienced 25% less shrinkage strain compared to LDUNE. This behavior can be attributed to the incorporation of steel fibers, which provide resistance to tensile stresses and decrease deformations caused by moisture evaporation. Yousefieh et al.⁷⁹ have obtained similar outcomes, demonstrating the effectiveness of fibers in reducing cracking caused by shrinkage. As expected, the control mix (LDUNE) showed shrinkage values between those of FLDUNE and FCLDUNE. Crucially, the shrinkage values of all three mixes were below normal ranges. According to the Federal Highway Administration (FHWA), shrinkage strains should not exceed 0.0004 at 56 days to reduce the chance of cracking⁸⁰. The results illustrate that the shrinkage strains of the mixes are within acceptable ranges, ensuring durability and minimizing cracking risks. Figure 15 also compares the experimental and predicted values of shrinkage obtained from the AASHTO LRFD Bridge Design Specification 9th Edition, Eq. (5)⁸¹. For the first 14 days of the experiment, all three mixes had actual shrinkage performance very close to the performance predicted by the AASHTO equation. After that, the measured shrinkage strain values became lower than the predicted values, which was more pronounced for FLDUNE. This suggests that all three mixes had better long-term performance compared to theoretical predictions. This might be attributed to the AASHTO equation being slightly conservative, as it is used in design applications, or due to the fact that the AASHTO equation does not account for the absence of coarse aggregates or the presence of steel fibers⁸¹.

5

Where: k_s = 1.45–0.13(V/S) ≥ 1.0, V = volume of sample, S = surface area of sample;

k_hs = (2.00–0.014 H), H= average relative humidity;

k_f =, = concrete strength at time of loading or 0.8;

k_td = , t = days between end of curing and time of analysis.

A similar trend to shrinkage was observed in the creep strain results. The creep strain reached maximum values of 0.000655, 0.000587, and 0.000713 for LDUNE, FLDUNE, and FCLDUNE, respectively, as shown in Fig. 16. Among the mixtures, FLDUNE had a 10% lower creep strain than LDUNE. This is because FLDUNE contains steel fibers, which improve tensile resistance and minimize creep deformation, creating a reinforcing effect. By enhancing the concrete matrix’s load transfer mechanisms, the fibers likely contributed to the reduction of microstructural deformation under sustained loads⁸². On the other hand, FCLDUNE showed a creep strain that was 9% higher than LDUNE. The inclusion of fine carbon particles is likely the main cause of the increase in creep deformation for FCLDUNE. These particles may have altered the pore structure and increased matrix flexibility under sustained stress, leading to higher creep deformation over time. Overall, FLDUNE’s superior shrinkage performance corresponds to its superior creep strain reduction performance compared to the other two mixes. Steel fibers improved the concrete’s long-term load-bearing capacity by lowering creep strain and minimizing shrinkage. Figure 16 compares the creep strain results with predicted values from the AASHTO LRFD Bridge Design Specification 9th Edition, Eq. (6)⁸¹. For the first 21 days, both LDUNE and FLDUNE had performance similar to the AASHTO equation’s prediction. After 21 days, the actual creep performance exceeded the expectations set by the AASHTO equation. In contrast, FCLDUNE exhibited higher actual creep strain values than those predicted by the AASHTO equation for the entire duration of the experiment. At 180 days, LDUNE and FLDUNE outperformed the theoretical predictions by 16.6% and 26.6%, respectively. The steel fibers in FLDUNE successfully resisted creep, a factor not considered in the AASHTO equation. Consequently, FLDUNE outperformed LDUNE and showed superior performance compared to the predictions. Meanwhile, at 180 days, FCLDUNE showed a creep strain 9.7% greater than the AASHTO equation’s prediction. The inclusion of carbon-based materials adversely affected creep behavior, increasing long-term deformation under sustained load. This negative effect, combined with the AASHTO equation’s inability to account for the influence of carbon-based products, resulted in significant differences between the predicted and actual creep strain values.

Fig. 16.

Creep results.

Furthermore, Managat and Azari⁸³ developed an equation, Eq. (7), to predict the creep strain of steel fiber reinforced concrete based on the creep of the control mix. Applying this equation to FLDUNE, using the 180-day creep strain value of LDUNE as a reference, yielded a predicted value of 0.000566, while the experimentally measured value was 0.000586. With a percentage difference of only 3.5%, this demonstrates that Eq. (7) offers highly accurate creep predictions for FRC^81,83.

6

Where: Ψ = creep coefficient;

k_hc= 1.56–0.008 H, H = average relative humidity;

t_i = age of concrete at time of load application (days)

7

Where: = creep of steel fiber reinforced concrete, = creep of control plain concrete mix, = volume of fiber, = coefficient of friction, = fiber aspect ratio.

Conclusions

CHSC is a newly developed innovation that involves mixing two different types of concrete (HSC and conductive concrete). The effect of incorporating conductive materials (steel fiber and carbon materials) on the properties of CHSC was studied. The following provides an overview of the current study’s findings:

1. HSC with compressive strength and modulus of rupture up to 100 MPa and 8.96 MPa, respectively, could be produced using locally available materials at low content of dune sand.

2. The incorporation of steel fiber in HSC (FLDUNE) resulted in an increase in 28-day compressive and flexural strengths by 3.5 and 22%, respectively, compared with LDUNE. However, the addition of carbon materials along with steel fiber (FCLDUNE) resulted in a decrease in 28-day compressive and flexural strengths by 19.6 and 15%, respectively, compared with LDUNE.

3. The steel fibers demonstrated effective crack control, contributing to the sustained modulus of rupture across different deflection levels.

4. CHSC with significant attenuation performance was obtained when using steel fiber in the production of concrete mix, while the use of carbon material along with steel fiber did not lead to a significant added value for the shielding application.

5. The two-wire technique measured resistivities of 33.3 Ω-m for FLDUNE and 25.7 Ω-m for FCLDUNE mixes, while the waveguide measurement technique determined a relative complex permittivity of 22.57-j2.22 for the CC mix. An accurate multiphase dielectric mixing model is developed for estimating the relative complex permittivity of the proposed concrete mixes, incorporating the proportions of various aggregates.

6. The M2 mix, incorporating steel fibers, demonstrated significantly higher attenuation (-24.9 dBm at 1 GHz, -70 dBm at 10 GHz for the plain plate) compared to M1, attributed to enhanced conductivity. The M3 mix, with added carbon powder, showed similar attenuation to M2 (-23.8 dBm at 1 GHz, -66 dBm at 10 GHz for the corrugated plate).

7. The plain configuration showed a slight SE reduction, while the corrugated configuration exhibited improved SE at most frequencies.

8. The presence of steel fiber resulted in a significant decrease in shrinkage and creep by 25 and 10%; respectively, while the use of carbon along with steel fiber increased the shrinkage and creep by 18 and 9%; respectively, compared to the control mix.

Author contributions

Obaida Othman: Conceptualization, data curation, formal analysis, investigation, methodology, validation, visualization, writing—original draft preparation, and writing—review and editing.Sherif Yehia: Conceptualization, methodology, software, validation, formal analysis, investigation, resources, data curation, writing—original draft preparation, writing—review and editing, visualization, supervision, project administration, and funding acquisition.Mohamed Elchalakani: Conceptualization, investigation, methodology, validation, and writing—review and editing.Nasser Qaddoumi: Conceptualization, methodology, validation, formal analysis, investigation, resources, data curation, writing—review and editing, visualization, supervision.Hend Malhat: Conceptualization, methodology, software, validation, formal analysis, investigation, data curation, writing—original draft preparation, and writing—review and editing.Walid E. Elemam: Conceptualization, methodology, software, validation, formal analysis, investigation, data curation, writing—original draft preparation, writing—review and editing, and visualization.

Data availability

The original contributions presented in this study are included in the article.

Declarations

Competing interests

The authors declare no competing interests.