**1. Introduction**

Concrete is a key material in the construction industry. It is in fact the most used element both in building and infrastructures construction such as bridges, hydraulic works, pavements, etc. Its attributes include easy and affordable manufacturing [1,2], good durability if properly executed [3,4], and remarkable structural capabilities [5–7], which make this material the most widely consumed globally. Concrete materials are designed solely to carry compressive loads, although their continuos presence in structures may induce to seek for additional properties which could make concrete a more exploitable material. However, the very nature of cement based materials imposes limitations to develop highly customizable features with added functionalities.

One of the properties that cement microstructure prevents from developing is the electrical conductivity. Wet concrete behaves as a semiconductor, with resistivity in the range of 10<sup>5</sup> Ω/mm. However, dry concrete has resistivity in the range of 1012 Ω/mm, which makes the material be considered as an insulator. The variation in the measured electrical resistivity in wet and dry concrete can be interpreted to find that concrete electrical conductivity is an effect of the evaporable water present in the material [8].

The volume of evaporable water found in saturated concrete varies from 60% at the time of mixing to 40% when cement is fully hydrated. This water contains ions whose concentration varies over time, directly affecting concrete conductivity. When the concentration of these ions is very high, ionic association begins, giving rise to C-S-H gel formation and ettringite, which in turns generates an electrical insulation layer in the cement grains. This leads to a decrease in ions mobility, therefore increasing the resistivity. The increase in resistivity with time is also due to less porosity and more tortuosity [9]. After this abrupt rise, resistivity continues to increase at a much slower rate, due to the

**Citation:** Cerro-Prada, E.; Pacheco-Torres, R.; Varela, F. Effect of Multi-Walled Carbon Nanotubes on Strength and Electrical Properties of Cement Mortar. *Materials* **2021**, *14*, 79. https://dx.doi.org/10.3390/ ma14010079

Received: 29 November 2020 Accepted: 22 December 2020 Published: 26 December 2020

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/ licenses/by/4.0/).

decrease in hydration reactions. In a porous medium such as concrete, resistivity reflects the ability to transport electric charge throughout the ions dissolved in the aqueous phase of a certain volume, assuming that aggregates are electrically inert since their resistivity turns out to be several orders of magnitude higher than that of the porous solution.

This physico-chemical approach to analyze the capacity to carry electrical charge will not only make it possible to develop a concrete with conductive properties-which will surely lead to very interesting applications-but also, it will provide information on the mechanical performance of the material. Upon mixing cement with water, a suspension is obtained whose resistivity is very low, however, as the cement hydrates and the concrete sets and hardens, the resistivity increases. The evolution of resistivity is therefore parallel to the evolution of strength. In this way, resistivity allows to predicting performance and can act as an indicator of the "age factor", which is essential for certain durability models.

At this point, we incorporate into our study carbon nanotubes (CNTs) within the cementitious material microstructure. CNTs are emerging nanomaterials that have captured the interest of researchers in recent years due to their attractive physical and chemical properties. CNTs mechanical properties, such as strength, stiffness and toughness, have been broadly studied and reported. In particular, Young's modulus of CNTs is in the range of 1–1.2 TPa, which indicates elastic behavior, with a tensile strength in the order of 36 GPa, denoting yield strains of up to 10% [10–14]. These outstanding mechanical characteristics provide a promising horizon in the achievement of CNT-composites with very high performance strength.

As nanoadditions incorporated into the cement microstructure, CNTs have been widely studied expecting that, due to CNTs superb mechanical performance, an impressive reinforcement in cement-based composites will be achieved. There are indeed a significant number of researchers that have demonstrated great enhancements in flexural and compressive strengths in cement composites by adding low concentration of CNTs. In particular, considering 28 days curing, Li et al. [15] achieved up to 19% and 25% increases in compressive and flexural strength respectively, by adding CNTs 0.5 wt.% to Portland cement paste. Hawreen and Bogas [16] incorporated CNTs additions varying between 0.05% and 0.5% by cement weight to concrete, obtaining improvement in the compressive strength of concrete up to 21%. CNTs were added in ultra high strength concrete by Lu et al. [17], achieving 4.63% increase in compressive strength with 0.05 wt.% CNTs loading. More recently, Hu et al. [18] achieved 2.4% and 9.6% improvements in compressive and flexural strength respectively with the addition of 0.05% CNTs in cement mortar at 28 days curing. However, these researchers reported detrimental effects in compressive and flexural strength with the addition of 0.5% CNTs.

General discussions from these and other reported studies conclude that low loadings of carbon nanotubes are very effective in improving the performance of cement-based materials. However, the incorporation of higher amounts of 0.5% CNTs seems to lead to lower compressive strengths. Furthermore, even the addition of 0.5 wt.% appears to produce contradictory strength results in different cement-based materials, as indicated in the previous paragraph. It seems necessary, therefore, to study in depth the incorporation of CNTs in cementitious materials in order to refine the amounts of CNTs to be added without producing detrimental effects.

However, researchers are clearly in agreement on the existence of a major downside to incorporating CNTs into cementitious composites. CNTs are atomic crystals and therefore do not dissolve in water. Furthermore, the large surface area of nanotubes induces strong attractive forces between the CNTs themselves. These facts result in the formation of clusters and agglomerations of CNTs when mixed with water. These clusters remain even after their insertion into the clinker, due to the van der Waals forces between carbon nanotubes. Furthermore, as CNTs are chemically inert, they do not participate in the hydration process. Consequently, agglomeration of CNTs around the cement grains hinders clinker hydration promoting defects formation in cement composites during the development of their microstructure [19]. All this results in deficiencies in the mechanical properties of the

cement-base material, since optimal resistance is developed during the correct hydration of the cement grains. In short, good CNTs dispersion is crucial and helps to obtain maximum mechanical performance by filling the pores and increasing adhesion with cementitious and hydration products [20–23].

Consequently, some authors have promoted surface modification processes to ensure an adequate dispersion of CNT and a homogeneous inclusion in the cement-water mixture [23–25]. Another interesting approach consists of using sonication to achieve a homogeneous dispersion of CNTs in the cementitious matrix. Many authors combine sonication with other shear mixing methods, such as mechanical, magnetic, or/and hand-stirring mixing methods [26–28]. On the other hand, cement type influences CNTs dispersion within the matrix, mainly due to the clinker particle size. In fact, most of the tests are carried out with cement of the CEM I 52.5 R type corresponding to Portland cement with a finer particle size and therefore more similar to the size scale of the nanomaterials, as they can be worked together in a more appropriate way. In any case, the hydrophobility of these nanostructures is maintained in the absence of electric field, since the water molecules cannot enter into the graphene layer spontaneously [29]. Consequently, due to their hydrophobility, we can expect that the presence of CNTs in cement microstructure will give rise to movements of evaporable water in the porous structure, and therefore variations in the intrinsic electrical resistivity of the cement composite.

Apart from ionic current due to water presence, electrical conductivity can also be developed in cement microstructure by means of CNTs connectivity, as this nanomaterial exhibits electrical conductivity similar to those of metallic materials [30]. However, it is clear that uniform CNTs distribution is essential to produce a continuous electrically conductive network. In fact, Kim et al. [31] reported that CNTs agglomerations induce damage in the electrically conductive pathways that CNTs should otherwise create. Remarkable improvement in electrical conductivity was achieved by Kim at al. [32], obtaining more than 1000 times reduction in electrical resistivity of well dispersed CNTs-cement paste compared to that of plain cement paste at 28 days curing, whereas poorly-dispersed CNTscement paste only achieves electrical resistivity approximately 2 times lower than that of control.

Very large number of CNT modified cement-based materials studies in the current literature aims at assessing the optimal CNT dosage to produce remarkable improvements in strength performance. Despite their importance, these studies use small-scale specimens that are not likely to reflect the actual mechanical behavior of large-scale structures. Taking into account that the electrical resistivity of cement-based materials can be used in quality control or for service life prediction of full size elements, this paper aims at providing more insights at important features that may need to be captured in the ongoing development of standard test methods to be used for CNT modified cement-based materials.

In this work, we present an experimental study on the effect of multi-walled carbon nanotubes (MWCNTs) on the microstructure of cement mortars, in terms of mechanical and electrical properties. For this, standard mortar prismatic samples will be modified with different proportions of MWCNTs, i.e., 0.00 wt.%, 0.01 wt.%, 0.015 wt.%, and 0.02 wt.% by the mass of cement. The nanomaterials will be introduced into the mortar material by priorly prepared water-MWCNT nanofluids following a combined mechanical and sonication dispersion procedure. The consistency, density, setting time and compressive and flexural strength of mixes will be tested and analyzed at 28 and 90 days curing time, in order to account for possible effects due to MWCNT presence in fully hydrated mortar. Finally, this paper will also investigate the electrical resistivity of the different MWCNT-mortar composites and the influence of temperature on the samples resistivity, to confirm the correlation of mechanical and electrical properties of CNT modified cementbased materials.
