The influences of admixtures on the dispersion, workability, and strength of carbon nanotube–OPC paste mixtures

https://doi.org/10.1016/j.cemconcomp.2011.09.013Get rights and content

Abstract

Carbon nanotubes (CNT) have excellent mechanical properties and have the potential, if combined with Ordinary Portland Cement (OPC), traditionally a brittle material in tension, to become a nano-composite with superlative mechanical properties. However, highly attractive van der Waals forces between CNTs create coherent agglomerates that prove difficult to disperse within the cement matrix and reduce the fluidity of the fresh mixture. Good dispersion of CNTs, while maintaining good workability of fresh OPC–CNT paste mixtures, is a prequalification before CNT-cement nanocomposites can be considered as a future building and construction material with enhanced mechanical properties. This paper reports the results of investigations of the dispersion, workability, and strength of CNT aqueous and CNT–OPC paste mixtures, with and without several generically different dispersants/surfactants that are compatible as admixtures in the manufacture of concrete. These include an air entrainer, styrene butadiene rubber, polycarboxylates, calcium naphthalene sulfonate, and lignosulfonate formulations. Aqueous mixtures were initially assessed for dispersion of CNTs, followed by workability testing of selected OPC–CNT-dispersant/surfactant paste mixtures. A broad range of workability responses were measured and the CNT dispersion within hardened pastes was qualitatively assessed by SEM analysis.

Introduction

Carbon nanotube (CNT) fibers have superlative mechanical properties and have future promise when combined with Ordinary Portland Cement (OPC) as a nanocomposite. The first evidence of tubular shaped carbon nanotubes was observed under transmission electron microscope by Radushkevich and Lukyanovich [1] and Monthioux and Kuznetsov [2]. CNTs are macromolecules of carbon atoms in a periodic hexagonal arrangement with a cylindrical shell shape and categorized as single-walled nanotubes (SWNTs) and multi-walled nanotubes (MWNTs). The structure of a SWNT can be conceptualized by wrapping a graphene sheet into a seamless cylinder while MWNT consists of multiple graphene sheets rolled in on themselves to form a tube shape. CNTs can have diameters ranging from 1–100 nm and lengths up to millimetres. However, because of high attractive van der Waals forces between CNT particles [3], considerable surface area, and high length/diameter ratio, with lengths of tens of microns and diameters from 2 to 100 nm [4], it can be difficult to ensure uniform dispersion of CNTs within the OPC paste. Insufficient dispersion of CNTs is cited a key diminishing factor in performance of composites incorporating CNTs [5], [6]. Poor dispersion of CNTs leads to the formation of many defect sites in the nanocomposite and limits the efficiency of the CNTs in the matrix.

Many past studies have focused on dispersion of CNTs within CNT–polymer composites, however there are few published studies on dispersion of CNTs within OPC pastes. Many of the surfactants that can effectively disperse CNTs in polymer liquids are incompatible with cement hydration; they can retard or prevent hydration, entrap substantial air in the paste, or undergo reactions with water-reducing admixtures, resulting in reagglomeration [7].

When mixed in water, OPC–CNT paste mixtures are highly viscous and difficult to place, even when only 2% CNT (weight of cement) has been admixed within cement paste prepared with a 0.8 w/c ratio [8]. When embedded as fillers, CNTs with poor dispersion act as crack initiators, thereby impairing the elastic properties of the resultant composite [9].

Kowald [10] incorporated 0.5% MWNTs (% cement weight) and measured 12% increase in compressive strength, however the consistency of the paste (i.e. the ability of freshly mixed cement paste to flow) was substantially reduced and pressures up to 125 MPa were applied for more than 45 min for adequate compaction.

Methods for dispersing CNTs have included mechanical ultrasonication to disperse CNTs [11], or changing the surface tension of the wetting liquid [12], or functionalization of the CNT surface [9], [13], [14] to change the surface energy of the nanotubes. However, chemical functionalization of the CNT surface has the potential for damage to the CNTs following this treatment [11], [15]. Although previous dispersion investigations have been undertaken on OPC–CNT-surfactant paste mixtures [16], [17], [18], [19], [20], none link the simultaneous need for aqueous dispersion of both CNTs and OPC particles; a key factor when considering future applications that involve future placement and fabrication of CNT–OPC mixtures.

This paper reports the results of preliminary investigations of the dispersion and rheology of CNT–OPC paste mixtures, with and without dispersants/surfactants. The work was conducted over the following stages:

  • (i)

    Initial screening of aqueous solutions containing CNTs and several generically different types of admixtures that were designed to disperse OPC particles.

  • (ii)

    Based on the results of (i), a short-list of OPC–CNT-admixture combinations were selected for consistency and compressive strength testing.

  • (iii)

    Hardened samples from (ii) were fractured and the fracture surface was assessed by SEM for distribution of CNTs within the cement paste.

Section snippets

Materials

Ordinary Portland Cement (OPC), conforming to the requirements of Australian Standard AS3972 as a Type GP cement, was used as the binder material [21]. The chemical composition and properties of the binder is summarized in Table 1. The multiwalled CNTs (MWCNTs) used in the experiments were sourced from Nanotech Port Corporation (Shenzhen, China). The physical properties of the MWCNTs used in the investigation are summarized in Table 2, and are distinguished by diameters within the ranges of

Sedimentation in aqueous solutions

Although there are methods to quantify dispersion (e.g. image analysis and UV–vis spectroscopy), the purpose of the visual observations of aqueous solutions containing carbon nanotubes, with and without dispersants, was to identify those mixtures that showed very clear contrasts, for the purpose of selecting mixtures for subsequent cement paste mixes. There was little ambiguity of the opaque black CNT agglomerations that separated from the mixture, settling underneath a very clear and

Conclusions

This paper reports the results of investigations of the dispersion and consistency and strength of CNT-aqueous and CNT–OPC paste mixtures, with and without several generically different dispersants/surfactants for OPC-based mixtures, including air entrainer, styrene butadiene rubber, polycarboxylate, calcium naphthalene sulfonate, and lignosulfonate. The outcomes of the work were:

  • (i)

    As per previous studies, CNTs in aqueous solutions agglomerate despite mechanical agitation by magnetic stirring and

Acknowledgements

The authors wish to acknowledge the financial support provided to the project by a Monash University Engineering Small Grant. The support and efforts provided by Tim Williams from the Monash Centre for Electron Microscopy, Jeff Doddrell and the Department of Civil Engineering concrete laboratory technical staff, is greatly appreciated.

References (28)

  • F. Ko et al.

    Electrospinning of continuous carbon nanotube-filled nanofiber yarns

    Adv Mater

    (2003)
  • J.N. Coleman et al.

    Small but strong: a review of the mechanical properties of carbon nanotube–polymer composites

    Carbon

    (2006)
  • A. Yazdanbakhsh et al.

    Carbon nanofibers and nanotubes in cementitious materials: some issues on dispersion and interfacial bond

    ACI Special Publication

    (2009)
  • J. Makar et al.

    Carbon nanotube/cement composites – early results and potential applications

  • Cited by (391)

    View all citing articles on Scopus
    View full text