Introduction
Concrete is the most widely used man-made material in the world: it is estimated that roughly 25 billion tons of concrete are manufactured globally each year [
1]. However, concrete is a quasi-brittle material which, depending on its constituents, can be very strong in compression (>200 MPa ultimate strength), but presents generally a low resistance in tension and consequently a limited bending strength. It is also characterized by relatively low fracture toughness. Studies show that cement, aggregate and mineral admixture can improve the toughness of concrete, but the effect is not obvious. Therefore, fibers have been added to concrete for almost 50 years to increase its tensile and flexural strength [
2]. In the last years, carbon nanotubes (CNTs) [
3,
4], cellulose fibers [
5] and CaCO
3 whiskers [
6] have been also investigated to this aim. The observed micromechanical mechanisms were based on CNTs/fibers/whiskers pull-out and bridging, crack deflection, as well as whiskers breakage. Recently, carbonized nano/microparticles have been proposed to increase cement composites toughness too [
7–
9].
Besides many advantages, modern developments in wireless and communication systems are contaminating our surroundings with electromagnetic waves (EMWs) pollution and cement matrix possess poor shielding against EMWs. Therefore, several researchers have worked to enhance shielding effectiveness of cement composites by conducting element additions [
10–
14]. Although CNTs and graphene possess ideal properties in terms of high specific surface area, low density, high aspect ratio and very high electrical conductivity [
10,
15], there are several major concerns that normally restrict their utilization on large scales. One of them is their high production cost and another one is their effective dispersion in cement matrix. Moreover, acid functionalization of CNTs can lead to an increase of CNTs toxicity [
16].
Thus, the current study is focused on the utilization of sub-micronized, carbonized agricultural residues, specifically peanut and hazelnut shells to enhance the mechanical properties (strength and fracture toughness), as well as, the electromagnetic interference shielding effectiveness of cement composites.
Experimental program
The raw peanut shell (PS) and hazelnut shell (HS) were washed with tap water and dried in oven for 48 h at 105±5°C. All the washed and dried raw materials were then pyrolyzed in a quartz reactor at 850°C for 1 h under inert atmosphere and a constant flow of argon under a 0.2 bar pressure. Finally, the carbonized materials were first manually ground in an agate mortar with an agate pestle, and then in ethanol to sub-micron scale by ball milling for 24 h, followed by 2 h of attrition milling. After grinding, the powders were dried in an oven at 50°C and stored in airtight containers until use.
Raman analyses were carried out on carbonized particles by means of a micro-Raman spectrometer (Olympus BH-2 UMA, Japan) equipped with a CCD detector. All the measurements were performed at room temperature. Field emission scanning electron microscopy (FESEM, Hitachi S4000) equipped with elemental microanalysis (Kevex) was used to study the morphology and particle size distribution of the carbonized particles, as well as to determine their chemical composition. The amount of carbonized nano/microparticles was 0.025, 0.05, 0.08, 0.2, 0.5 and 1 wt% with respect to cement. On the basis of preliminary tests, the water to cement ratio (
w/
c) was kept to 0.35, while the superplasticizer content was always fixed to 1.5 wt% (Mapei Dynamon SP1, Italy; modified acrylic based polymer) with respect to cement (ordinary Portland cement provided by Buzzi Unicem, Italy, Type-1, grade 52.5R). Four samples were prepared for each composition. The required quantity of carbonized particles and admixture were added into the measured quantity of distilled water and the mixture was sonicated for 15 min in an ultrasonic bath. After sonication, the mixture was transferred into the mixing bowl (Janke and Kunkel homogenizer) and mixing speed was set at 440 rpm while the cement was added gradually in the initial 60 s. The mixing speed was kept constant for further 60 s. Then, the speed was increased to 630 rpm and maintained for 120 s, making the total mixing time of four minutes. After mixing, the cement paste was poured into the acrylic molds (20 × 20 × 75 mm
3), which were then kept in closed plastic containers having 90% relative humidity for 24 h. Finally, the prism shaped samples were removed from the molds, weighed, labeled and cured in water for 28 days at room temperature. The dimensions of the samples were chosen based on literature [
4,
17]. Prism shaped cement composite specimens flexural strength was evaluated with a single column displacement controlled flexural testing machine (Zwick Line-Z010) with a load cell capacity of 1 kN. The crack mouth opening displacement (CMOD) mode was used and a 0.003 mm/min rate was adopted. Prior to the flexural testing, 6 mm deep U shaped notches were made in the specimens with Remet type TR100S abrasive cutter having 2 mm thick diamond cut-off wheel.
Complex permittivity measurements of cement composites containing carbonized particles were performed in the frequency range 0.20–10 GHz with a commercial dielectric sensor (85070D) and a network analyzer (E8361A). The system was calibrated for water and air before starting measurements. In all measurements, the perfect contact between sensor probe and specimen’s surface was ensured. Measurements were taken at six different sites of each single specimen to ensure homogeneity and to reduce error margin. The reported results are the average of the six measurements on samples having 6.5 cm in diameter and 1.0 cm in height.
Results and discussion
From FESEM observations, carbonized peanut and hazelnut shells have an average diameter of 600 and 750 nm, respectively. However, few particles up to 5–6 microns in size were sometimes observed. FESEM observations also revealed that the carbonized particles dispersed well in the cement matrix (Fig. 1). Moreover, the rather limited presence of cement paste residue onto carbonized particles seems to indicate a rather weak interaction between the reinforcement particles and the cementitious matrix. Microanalysis evidenced that in the synthesized materials the carbon content was equal to 87.7 wt% and 93.8 wt%, respectively for hazelnut and peanut shells. The main impurities were aluminum and calcium.
The Raman results indicate that the two materials are equivalent, as the ratio of I
D to I
G band is similar in the two synthesized products (about 0.85). This ratio also means that the materials contain a limited amount of amorphous carbon or of defective graphite crystals [
18].
Table 1 reports the average MOR of the investigated compositions: the MOR value of neat cement pastes was rather low, but in line with literature data [
19].
Generally, the flexural strength of HS-cement samples was higher respect to PS composites (Figs. 2, 3; Table 1). This fact may be due to a higher density (determined by means of a pycnometer) of HS particles (2.35 g/cm
3), respect to PS ones (2.20 g/cm
3). The observed scattering in the results of the samples having 0.08 wt% PS and 0.2 wt% of HS and PS is probably due to a lower homogeneity of these samples. However, no correction has been applied to the data, as suggested in the ASTM C348 standard, for example. The area under the stress-separation curve represents the total energy dissipated by fracture per unit area of the crack plane (G
F, J/m
2) [
20]. After carbonized particles addition, fracture energy increased too with respect to plain cement samples (Fig. 4).
In a material, cracks open or propagate when local stresses provide the energy necessary for rupture of material bonds and for the creation of fracture surfaces. In composite materials, when a crack meets aggregates, pores, voids and fibers, it is stopped and an increase of energy by increasing the load is needed for its further propagation [
21]. Then, the crack may pass across obstacles or may contour them to follow another path along a weaker region [
21]. In that case, a crack may be subdivided into several finer cracks, leading to crack branching because of an obstacle. Additional energy is also needed for breaking it or for destroying the bond strength in the interface around it. Therefore, all kinds of inhomogeneities, which produce crack deviation, branching or multiplication increase the total area of fracture surface, which becomes several times the area of the cross-section of the element [
21]. To this aim, the addition of fine aggregates to cement paste matrix increases the fracture toughness because of the occurrence of the above described mechanism of microcrack shielding and crack deflection in the mortar [
20]. For increasing the toughness, it is important to understand how cracks intercept aggregate zones and this is influenced by the nature of the interfacial bond between the aggregate and the matrix [
22]. When a crack encounters an angular aggregate which is not bonded to the matrix, the aggregate is able to act as a bridge because the crack follows the flat fracture path and irregular shape leads to pinning [
22]. The dependence of the aggregate shape on toughening is critical and angular grains are needed to produce effective aggregate bridging [
22]. In our case, as the carbonized particles are the result of an extensive milling, they possess angular shapes favorable to aggregate bridging. Finally, it is known too that the surface roughness of the paste near the aggregates is greater than that of the paste far from aggregates. This fact implies that the Interfacial Transition Zone (ITZ) associated with them toughens the paste in concrete [
23].
The relative complex permittivity (
er =
e' -
je'') was measured using dielectric probe sensor in 0.2-10 GHz frequency range. The real part refers to the content of energy stored by a material when exposed to external electric field while imaginary part is the measure of dissipated or lost energy. An appreciable increment was observed in the values of real and imaginary permittivity on induction of the composite samples. Total shielding effectiveness (SE) depends upon three major losses occurring while propagation of electromagnetic waves (EMWs) through the medium: the reflection loss, which occurs due to reflection of EMWs at the material’s interfaces, the absorption loss of the wave as it proceeds through the barrier, and the re-reflections and transmissions phenomena of EMWs (multiple reflection loss). Total SE along with the three major contributors was numerically evaluated from measured values of dielectric constant (
e') and dielectric loss (
e''). More details on these measurements are reported in Ref. [
24].
For a better evaluation of results, total SE of all five formulations was compared at four different frequencies of 0.94 GHz (GSM mobiles), 1.56 GHz (GPS communication devices), 2.46 GHz (microwave ovens), and 10.0 GHz (radio communication devices). With respect to plain cement, a maximum increase of 353% in shielding effect was observed at a frequency of 0.9 GHz after adding 0.5 wt% of carbonized peanut shells. This percentage reduces to 223%, 126% and 83% when the frequency values increase (1.56, 2.46 and 10 GHz, respectively). In case of 0.5 wt% hazelnut shells addition, a similar increase of 335%, 214%, 122% and 76% was achieved at the four specified frequencies, in comparison to plain cement samples. The addition of 0.5 wt% HS leads not only to the maximum in electromagnetic SE but also to the maximum in fracture energy (Fig. 5). For PS, the maximum in fracture energy is reached for a 0.08 wt% addition, but a higher amount (0.5 wt%) gives rise to a significant increase in fracture energy too, with respect to plain cement. To conclude, the total increase in SE when adding 0.5 wt% of PS or HS in cement composites is much higher in comparison to the ones reported by other researchers using CNTs in the same content [
10,
25], at the four above mentioned frequencies.
Conclusions
Nano/microsized carbonized particles were successfully synthesized from hazelnut and peanut shells and further used in the production of high performance cement composites. In the case of PS addition, the amount of particles which ensure a maximum shielding effectiveness gives rise also to a maximum value of the fracture energy, making possible to prepare cementitious materials optimized both from a mechanical and an EMI shielding point of view.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(517KB)
