ArticlePDF Available

Mechanical Properties of Carbon Nanotube Reinforced Cementitious Materials: Database and Statistical Analysis

May 2019
Magazine of Concrete Research 72(20):1-62

May 2019
72(20):1-62

DOI:10.1680/jmacr.19.00093

Authors:

Mahyar Ramezani

Florida International University

Young Hoon Kim

University of Louisville

Zhihui Sun

University of Louisville

This paper aims to provide guidelines for selecting the correct type of carbon nanotube (CNT) to improve the mechanical properties of cementitious materials. Previous researchers have discussed the effect of CNT characteristics on their dispersion quality. However, the effect of these characteristics on the mechanical properties of CNT-reinforced cementitious materials is not fully understood. To clarify this, the study reported in this paper was conducted in two phases. In the first phase, a database was established from the literature to study the influences of three different parameters associated with CNTs (length, diameter and concentration based on the weight percent of cement powder (c-wt%)) on compressive and flexural strengths. The analyses revealed that short and small-diameter CNTs could be beneficial for increasing compressive strength. Conversely, relatively long and large-diameter CNTs were more effective in increasing flexural strength. In general, an average CNT length of 10–20 μm and an average diameter of 20–32·5 nm resulted in the highest overall mechanical performance. The optimal upper limit concentrations for flexural and compressive strengths were found to be 0·15 and 0·20 c-wt%, respectively. In the second phase of this study, the statistical analyses were experimentally verified using the CNT optimum length, two diameters and three levels of concentrations.

Effect of CNT length and diameter at different concentrations on the pore size distribution: L, low concentration; H, high concentration; SL, short CNTs; LL, long CNTs; SD, small-diameter CNTs; LD, large-diameter CNTs; *, short and small-diameter CNTs, if clumped, divide big pores into smaller pores; **, long and large-diameter CNTs, if clumped, produce bigger pore sizes

…

Statistical data used to analyse the mechanical properties

…

Ancova results: p-values of interaction between potential and primary parameters

…

Figures - uploaded by Mahyar Ramezani

Content may be subject to copyright.

Content uploaded by Mahyar Ramezani

Content may be subject to copyright.

Mechanical properties of

carbon-nanotube-reinforced cementitious

materials: database and statistical analysis

Mahyar Ramezani

PhD candidate, Department of Civil and Environmental Engineering,

University of Louisville, Louisville, KY, USA (Orcid:0000-0002-3053-7736)

Young Hoon Kim

Associate Professor, Department of Civil and Environmental Engineering,

University of Louisville, Louisville, KY, USA (corresponding author:

young.kim@louisville.edu) (Orcid:0000-0001-9415-9001)

Zhihui Sun

Professor and Chair, Department of Civil and Environmental

Engineering, University of Louisville, Louisville, KY, USA

(Orcid:0000-0002-5905-1414)

This paper aims to provide guidelines for selecting the correct type of carbon nanotube (CNT) to improve the

mechanical properties of cementitious materials. Previous researchers have discussed the effect of CNT characteristics

on their dispersion quality. However, the effect of these characteristics on the mechanical properties of

CNT-reinforced cementitious materials is not fully understood. To clarify this, the study reported in this paper was

conducted in two phases. In the first phase, a database was established from the literature to study the influences of

three different parameters associated with CNTs (length, diameter and concentration based on the weight percent of

cement powder (c-wt%)) on compressive and flexural strengths. The analyses revealed that short and small-diameter

CNTs could be beneficial for increasing compressive strength. Conversely, relatively long and large-diameter

CNTs were more effective in increasing flexural strength. In general, an average CNT length of 10–20 μm and an

average diameter of 20–32·5 nm resulted in the highest overall mechanical performance. The optimal upper limit

concentrations for flexural and compressive strengths were found to be 0·15 and 0·20 c-wt%, respectively. In the

second phase of this study, the statistical analyses were experimentally verified using the CNT optimum length, two

diameters and three levels of concentrations.

Notation

dcarbon nanotube (CNT) average diameter

compressive strength

flexural strength

LCNT average length

obs

number of observations

Sstrength

Δf

percent change in compressive strength of

CNT-reinforced cementitious materials compared

with control (without CNTs)

Δf

percent change in flexural strength of

CNT-reinforced cementitious materials compared

with control (without CNTs)

ΔSchange in strength (Sˉ

κ2Sˉ

κ1)

ΔS=Δˉκdispersion quality indicator

Δˉ

κchange in CNT concentration (ˉ

κ2ˉ

κ1)

κaverage CNT concentration

Introduction

Cement-based materials exhibit low tensile strength and fracture

toughness, resulting in cracking (i.e. quasi-brittle behaviour).

Over the past few decades, fibres (macro to micro-sized) have

been incorporated to compensate for the low tensile strength

(Banthia and Nandakumar, 2003; Qian and Stroeven, 2000;

Song et al., 2005). Depending on the fibre type, the

contributions of fibres rely on the shape and size of the fibres,

their surface textures, the interfacial bond strength between the

fibres and the matrix, crack bridging ability and energy dissipa-

tion during crack propagation (Balaguru and Shah, 1992;

Bentur and Mindess, 2006). In this regard, smaller fibres are

effective at arresting comparably sized cracks at an earlier stage

of cracking. Meanwhile, because the nucleation of cracks starts

from the nanoscale (Konsta-Gdoutos et al., 2010a), even micro-

fibres cannot prevent the initiation of cracks (Tyson et al., 2011;

Zou et al., 2015). However, they are still capable of mitigating

crack propagation after the crack width reaches the microscopic

scale (Chuah et al., 2014; Lawler et al., 2003).

Being different from microfibres, nanofibres and nanotubes

including carbon nanotubes (CNTs) have been reported to

prevent or delay the nucleation of cracks at the nanoscale

(Konsta-Gdoutos et al., 2010b; Raki et al., 2010). There are

two main types of CNTs –single-walled carbon nanotubes

(SWCNTs) and multi-walled carbon nanotubes (MWCNTs) –

and both types have been used in cement-based materials.

SWCNTs typically possess diameters up to 2 nm, whereas

MWCNTs can have diameters ranging from 5 to about

100 nm (De Volder et al., 2013). The length of CNTs ranges

from less than several hundred nanometres up to even several

centimetres (Zheng et al., 2004; Zhu et al., 2002). Because of

the high aspect ratio (length-to-diameter ratio) of CNTs along

Cite this article

Ramezani M, Kim YH and Sun Z

Mechanical properties of carbon-nanotube-reinforced cementitious materials: database

and statistical analysis.

Magazine of Concrete Research,

https://doi.org/10.1680/jmacr.19.00093

Magazine of Concrete Research

Research Article

Paper 1900093

Received 12/02/2019; Revised 24/04/2019;

Accepted 14/05/2019

Keywords: composite materials/

compressive strength/tensile properties

with their nanoscale sizes, the distance between adjacent nano-

tubes can be reduced (Al-Rub et al., 2012a). This results in a

large number of CNTs at the crack plane, delaying the crack

propagation. Besides, CNTs possess a tensile strength and

Young’s modulus about 3000 and 20 times larger than those of

concrete, respectively (Demczyk et al., 2002; Yu et al., 2000).

The superior physical and mechanical properties of CNTs may

improve different aspects of concrete that exhibits quasi-brittle

behaviour. However, to the authors’best knowledge, there is

yet no consensus on the contribution of CNTs to mechanical

properties. Some researchers have reported significant improve-

ments in mechanical properties (Danoglidis et al., 2016a;

Konsta-Gdoutos et al., 2017; Nasibulina et al., 2012; Sun

et al., 2016) while others have reported negligible improvement

or even degradation in mechanical properties (Kumar et al.,

2012; Musso et al., 2009; Sobolkina et al., 2012; Stynoski

et al., 2015) (the mechanical properties investigated include

compressive and flexural (or tensile) strengths). These incon-

sistent results are mainly attributed to two issues: dispersion

and the interfacial bond strength between CNTs and the

cement matrix (Li et al., 2005; Stynoski et al., 2015; Tyson

et al., 2011).

The agglomeration of CNTs (i.e. poor dispersion) can produce

defects in the form of pores as well as creating unreacted

pockets, resulting in a reduction in strength (Fawaz and Mittal,

2014; Park and Bandaru, 2010). In addition, despite the

superior bond strength that CNTs with a higher aspect ratio

can provide (Wichmann et al., 2008), the higher aspect ratio is

also the primary reason for the adherence of individual CNTs

to one another (beyond certain concentrations of CNTs), thus

hindering their dispersion in different media. This may lead to

the premature debonding of CNTs from the cement matrix

when subjected to loading. Therefore, CNT characteristics

such as length, diameter and concentration are critical par-

ameters to determine both the dispersion quality and mech-

anical properties.

In previous studies, CNT characteristics have been tied to the

dispersion quality (Al-Rub et al., 2012a; Konsta-Gdoutos

et al., 2010a, 2010b), but how CNT characteristics could

differently affect mechanical properties has been overlooked.

Thus, it is worth knowing how each identified parameter of

CNT characteristics might influence the mechanical properties.

This led to the objective of this study –to investigate the

effects of CNT length, diameter and concentration on the

compressive and flexural strengths of cementitious materials.

Research significance

Despite the potential benefits of CNTs in improving the mech-

anical properties of cementitious materials, guidelines for

selecting the correct type of CNT are not yet established. To

address this, this paper provides a database from the available

literature and identifies the optimum ranges of CNT length,

diameter and concentration for improving compressive and

flexural strengths. The findings from this research provide prac-

tical guidelines for engineers and designers to select the type of

CNT to tailor mechanical properties for various structural

applications based on different strengthening mechanisms.

Mechanisms of CNTs affecting

mechanical properties

It has been found that the pore size distribution is affected

by CNT size and concentration (Hu et al., 2014; Kang et al.,

2015; Konsta-Gdoutos et al., 2010a; Li et al., 2005; Xu et al.,

2015). Thus, CNT characteristics such as length, diameter and

concentration are extremely important in terms of mechanical

properties. To date, most researchers have considered CNTs as

just very fine fibres, meaning that they can be distributed on

a much finer scale than other types of fibres in cementitious

materials (Al-Rub et al., 2012a). However, the mechanisms

of CNTs that affect mechanical properties are not yet fully

understood.

However, in other matrices (e.g. polymers, metals and cer-

amics), researchers have used CNT characteristics in different

models (e.g. the Kelly–Tyson model (Kelly and Tyson, 1965)

and the Halpin–Tsai model (Halpin and Kardos, 1976)) to

predict the flexural/tensile strength, modulus of elasticity and

interfacial shear strength between CNTs and various matrices

(Chen et al., 2016; Chen et al., 2017; Lachman et al., 2012;

Liao et al., 2010; Zare, 2015). Chen et al. (2017) observed

that the interfacial shear strength between CNTs and a metal

matrix increased when using longer CNTs, resulting in better

tensile strength. Conversely, the interfacial bond strength

between CNTs and the matrix might have a negligible

influence on the compressive strength. For example, Konsta-

Gdoutos et al. (2017) used two different types of MWCNTs to

increase the mechanical properties of cement mortars. Both

types of MWCNTs studied had similar lengths and diameters

but different surface conditions: pristine MWCNTs (type I)

and mechanically functionalised MWCNTs (type II) to

increase the interfacial bond strength between the CNTs and

the cement matrix. When using the type II MWCNTs, the

28 d flexural strength increased by 23% compared with the

type I MWCNTs. Nevertheless, the 28 d compressive strength

did not show further improvement when using either type I

or type II. Therefore, the compressive strength might be

increased by filling the internal pores of concrete using

CNTs. Consequently, smaller CNTs can fill bigger pores,

leading to smaller pore sizes. This improves the compactness

of the cement matrix, resulting in a significant improvement in

compressive strength.

Figure 1 shows schematic representations of these mechanisms.

Figures 1(a) and 1(b) show the influence of low (L) and high

(H) concentrations of the base CNTs with short length (SL)

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

and small diameter (SD). Figures 1(c) and 1(d) show the influ-

ence of an increase in CNT length (LL meaning long CNTs)

at low and high concentrations while using the same diameter

as the base CNTs. Figures 1(e) and 1( f) represent the influence

of an increase in the base CNT diameter at low and high con-

centrations (LD meaning large-diameter CNTs).

When using a low concentration of CNTs with various

physical characteristics (see Figures 1(a), 1(c) and 1(e)), the

well-dispersed CNTs reduce the total porosity and refine

the pore structure (Li et al., 2005; Xu et al., 2015). However,

utilising an excessive or high concentration of CNTs will

contribute differently to the total porosity and pore size

distribution. Agglomerated short and small-diameter CNTs

(see Figure 1(b)) might still be capable of filling big pores.

Conversely, the agglomeration of either long (see Figure 1(d))

or large-diameter (see Figure 1(f )) CNTs produces bigger

pores due to their larger sizes.

Compressive strength is closely related to pore structure:

smaller pore sizes increase the compressive strength while

bigger pores degrade it. Because of this case, the use of a low

concentration of short-length and small-diameter CNTs

(L-SL-SD, see Figure 1(a)) outperforms in terms of compres-

sive strength. In addition, the use of H-SL-SD could still con-

tribute to refining the pore structure due to the small CNT

size, which might be tolerable concerning compressive strength

(see Figure 1(b)). The agglomeration of long and large-diam-

eter CNTs (Figures 1(d) and 1(f)) might adversely affect the

compressive strength.

On the other hand, the interfacial bond strength between

CNTs and the cement matrix is an important parameter that

affects flexural strength. If the CNTs lack proper dispersion

(see Figures 1(b), 1(d) and 1(f )), frictional forces become mini-

mal within the agglomerated CNTs and they easily debond

from the matrix (Bakshi and Agarwal, 2011; Tyson, 2010),

degrading the flexural strength. In the case of good CNT dis-

persion (see Figures 1(a), 1(c) and 1(e)), the higher aspect ratio

CNTs (i.e. longer and smaller diameter CNTs, see Figure 1(c))

outperform in terms of flexural strength.

Dispersion quality is thus the prerequisite condition to utilise

the advantages of CNTs for improving the properties of har-

dened concrete (e.g. mechanical and durability). Durability

properties are not the focus of the current study, but previous

studies have reported on the influence of CNT characteristics

on the durability of cementitious materials such as freeze–thaw

resistance (Li et al., 2015; Yakovlev et al., 2013), corrosion of

steel reinforcement bars (del Carmen Camacho et al., 2014;

Shah et al., 2016) and permeability (Han et al., 2013; Lu

et al., 2016).

Effect of dispersion techniques on

mechanical properties

According to the literature survey, six representative dis-

persion techniques have been used to exploit the superior phys-

ical and mechanical properties of CNTs within cementitious

materials

&ultrasonication process with surfactant (US + S)

(Al-Rub et al., 2012a; Konsta-Gdoutos et al., 2010b;

Ramezani et al., 2016)

&ultrasonication process without surfactant (US)

(Kumar et al., 2012; Sun et al., 2016)

(a) L-SL/SD (b) H-SL/SD

Increase in CNT length Increase in CNT diameter

Matrix

CNTs

Pore

Figure 1. Effect of CNT length and diameter at different concentrations on the pore size distribution: L, low concentration; H, high

concentration; SL, short CNTs; LL, long CNTs; SD, small-diameter CNTs; LD, large-diameter CNTs; *, short and small-diameter CNTs,

if clumped, divide big pores into smaller pores; **, long and large-diameter CNTs, if clumped, produce bigger pore sizes

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

&pre-dispersion of CNTs in cement using an ultrasonication

process (US + CC) (Hunashyal et al., 2011; Makar et al.,

2005)

&dry mixing CNTs and cement (DM) (Kim et al., 2014;

Tamimi et al., 2016)

&ball milling (BM) (Li et al., 2015; Lu et al., 2016)

&direct synthesis of CNTs onto the surface of cement or

mineral admixtures (DS) (Mudimela et al., 2009;

Nasibulin et al., 2013).

Figure 2 shows the boxplots of the percentage change in com-

pressive and flexural strengths of cement-based materials con-

taining CNTs compared with the respective control specimens

(i.e. without CNTs) for the different dispersion techniques

obtained from the literature (see the Appendix). The boxplots

demonstrate median, first and third quartiles, minimum and

maximum values, and outliers. A data point is considered to

be an outlier if it lies 1·5 times the interquartile range below

the first quartile or above the third quartile. The median values

of the percentage change in compressive and flexural strengths

using the various dispersion techniques are all positive, except

for the DS method, which shows median values of −21%

and −9% for compressive strength and flexural strength,

respectively.

Although the boxplots did not show a noticeable difference

between dispersion methods, analysis of variance (Anova) was

used to statistically confirm the observed trend. Anova, which

determines whether there are significant differences between

the means of different groups, was used to investigate the influ-

ence of various dispersion techniques on the compressive and

flexural strengths. The probability value ( p-value) is used to

determine the significance of the analysis: if the p-value ≤0·05,

there is a significant difference between the groups. Table 1

compares the p-values of the different dispersion techniques

for compressive ( f

) and flexural ( f

) strengths. For example,

for compressive strength, the p-value of US + S compared with

the US dispersion technique was 0·316, indicating no signifi-

cant difference between these techniques in terms of increasing

the compressive strength. Comparing the DS method, most of

the p-values were close to 0·050 for the flexural strength. This

might be explained by the direct influence of several factors

(e.g. substrate material, inert gas, flow speed rate, catalyst and

applied temperature) resulting in high variations in CNT

300

250

200

150

100

–50

–100

Change in compressive strength: %

300

250

200

150

100

–50

–100

Change in flexural strength: %

US + S US DM

Dispersion technique

DS BM US + S US + CCUS DM

Dispersion technique

(a) (b)

DS BM

–30%

30%

–30%

30%

Outlier

Maximum

Third quartile

Median

First quartile

Minimum

Figure 2. Influence of different dispersion techniques on (a) compressive strength and (b) flexural strength

Table 1. Anova results: p-values of different dispersion techniques for compressive and flexural strengths (—indicates that the evalu-

ation was not valid and * indicates data were not available to perform the Anova)

Dispersion

technique

US + S US US + CC DM BM DS

US + S ——0·316 0·618 * 0·744 0·944 0·348 0·834 0·956 0·596 0·068

US ——* 0·576 0·552 0·173 0·817 0·820 0·945 0·076

US + CC —* 0·848 * 0·735 * 0·189

DM ——0·860 0·355 0·750 0·012

BM ——0·912 0·041

DS ——

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

concentration from 0·3 c-wt% (weight percent of cement

powder) (Ludvig et al., 2011) to 20 c-wt% (Cwirzen et al.,

2009; Nasibulin et al., 2009). This is the primary reason for

the high variations in compressive and flexural strengths using

the DS dispersion method (see Figure 2).

The ANOVA results suggest that the dispersion technique

does not significantly contribute to the mechanical properties.

However, to achieve better dispersion, each particular method

does have certain steps (or critical factors) that need to be fol-

lowed. For example, when using the US+ S method, the ultra-

sonication process (energy and amplitude) and surfactant

dosage are two crucial factors that would directly contribute

to CNT dispersion (Grossiord et al., 2005; Strano et al., 2003;

Yu et al., 2007). Assuming that the best efforts were given to

achieving a good dispersion of CNTs in the reported studies,

this paper now focuses on the influences of the intrinsic

characteristics of CNTs on concrete properties.

Identification of critical parameters and

data distribution

Identification of critical parameters

To identify the possible critical parameters, a thorough litera-

ture review was conducted. As shown by the dashed lines in

Figure 2, data of above 30% increase and below 30% decrease

in compressive and flexural strengths were further investigated.

These ranges were close to the maximum and minimum

strength levels of most of the dispersion techniques. Therefore,

other potential parameters could be identified regardless of the

dispersion method used. Through analysing the experimental

results reported in the literature, the observed trends demon-

strated the need for discussion about the effects of CNT

characteristics on the mechanical properties of cementitious

materials. Therefore, this study investigated the influences of

three parameters associated with CNTs –their average length

LÞ, average diameter ð

dÞand concentration (ˉ

κ). The optimum

ranges of CNT characteristics were identified to improve mech-

anical properties with minimal adverse effects of other

parameters.

Data distribution

A total of 42 studies (1359 data points) were collected from

the literature for data analysis (including the authors’data).

Table 2 shows the number of available data points and the

number of sources (independent studies) concerning the three

different properties of CNTs studied (

dand ˉκ). Table 3

shows the minimum, maximum, average, median and standard

deviation of the studied properties of CNTs obtained from the

literature.

Land

dof the CNTs were determined according to

the physical characteristics of the CNTs reported in each study.

For example, if the length and diameter of CNTs were reported

to be 10–30 μm and 20–40 nm, respectively, then

L=20 μm

and

d= 30 nm were used. The authors have collected all the

information from the reported CNT intrinsic properties (length

and diameter) in each individual study. It must be acknowl-

edged that such data are provided by CNT manufacturers. In

the early stage of this study, a CNT manufacturer claimed that

when the CNT length is reported to be 10–30 μm, 80 wt% of

CNTs will be 20 μm long. Therefore, the average length and

diameter of the CNTs are assumed to be the representative

characteristics of CNTs in data analysis.

Approach

To analyse the influence of CNT characteristics on the

mechanical properties of cementitious materials, this study was

conducted in two phases. In the first phase, using the data col-

lected from the literature (see the Appendix), the relationships

between CNT characteristics and compressive and flexural

strengths were evaluated to find the optimum ranges of par-

ameters for superior mechanical properties. In the second

phase, an experimental study was conducted to verify the find-

ings from the literature data analysis.

Due to a lack of available data in the literature, this study

investigated the influence of each investigated parameter with-

out taking into consideration the influence that other potential

parameters might have on the mechanical properties

(e.g. cement matrix composition, testing age, curing condition,

type of CNT etc.). However, an analysis of covariance

(Ancova) was used to evaluate the effect of other parameters

that were not primary factors on the response (i.e. mechanical

properties). The Ancova was performed to evaluate whether

other potential parameters (water-to-cement (w/c) ratio, sand-

to-cement (s/c) ratio, testing age) interacted with the critical

Table 2. Statistical data used to analyse the mechanical properties

Mechanical property

Number of data points/number

of studies

L¯

d¯κ

Compressive strength 216/20 255/26 266/29

Flexural strength 198/26 206/28 212/30

Table 3. Ranges of physical characteristics of CNTs used to

analyse the mechanical properties

μm

κ:

c-wt%

Compressive

strength

Minimum 3·00 1·5 0·01

Maximum 25·25 80·00 1·00

Average 14·80 33·28 0·33

Median 15·00 30·00 0·20

Standard deviation 5·82 23·86 0·32

Flexural strength Minimum 1·50 1·50 0·01

Maximum 55·00 80·00 2·00

Average 16·48 28·06 0·24

Median 17·50 30·00 0·10

Standard deviation 13·74 23·23 0·30

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

parameters (

dand ˉ

κ) on the mechanical properties.

When analysing the influence of each critical parameter on the

mechanical properties, the Ancova confirmed that other poten-

tial parameters had negligible influence (see Table 4;

p-value > 0·05).

Data analysis (phase I)

Effect of CNT characteristics on mechanical properties

This work aimed to identify the optimum ranges of each para-

meter for superior compressive and flexural strengths. Through

analysis of extensive data in the literature, three different

ranges were assigned to each CNT characteristic. Table 5 lists

the different ranges of CNT characteristics for compressive and

flexural strengths. The detailed analysis of identifying these

ranges will be presented elsewhere.

Effect of CNT length

Figure 3 shows the influence of three different CNT length

ranges on the mechanical properties. Figure 3(a) shows

boxplots for the change in compressive strength ( f

)

compared with a control (i.e. without CNTs) with respect

to the three different ranges of CNT length (Table 5). The per-

centage change in compressive strength Δf

was calculated

using

ΔfCS ð%Þ¼ fCSðCNTsÞfCSðcontrolÞ

fCSðcontrolÞ

100

Figure 3(b) shows the effect of different ranges of

Lof CNTs

on the percentage flexural strength change, Δf

, defined as

ΔfFS ð%Þ¼ fFSðCNTsÞfFSðcontrolÞ

fFSðcontrolÞ

100

Table 4. Ancova results: p-values of interaction between potential

and primary parameters

Potential parameter

Primary parameter

L¯

d¯

κ¯

L¯

d¯

Compressive

strength Flexural strength

w/c ratio 0·78 0·47 0·58 0·45 0·62 0·64

s/c ratio 0·26 0·63 0·25 0·66 0·11 0·11

Age 0·66 0·57 0·33 0·52 0·45 0·89

300

250

200

150

100

–50

–100

ΔfCS: %

300

250

200

150

100

–50

–100

ΔfFS: %

Range 1 Range 2 Range 3

L: μm

Range 1 Range 2

(a)

(b)

Range 3

L: μm

Range p-value

1 vs 2 0·0125

2 vs 3 0·1668

1 vs 3 0·0203

Range p-value

1 vs 2 0·0039

2 vs 3 0·0015

1 vs 3 0·4532

Figure 3. Effect of CNT length on (a) compressive strength and (b) flexural strength

Table 5. Identified ranges of critical parameters

Critical parameter

Range 1 Range 2 Range 3

L:μm <15 <10 15 ≤¯

L≤20 10 ≤¯

L≤20 >20 >20

d: nm <20 <15 20 ≤¯

d≤45 15 ≤¯

d≤32·5 ≥50 ≥50

κ: c-wt% ≤0·2 ≤0·15 0·2 < ¯

C≤0·5 0·15 < ¯

C≤0·5 >0·5 >0·5

Note: 32·5 < ¯

d< 50 nm not available for flexural strength

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

To statistically confirm the differences between different ranges

of each critical parameter, the Anova test results ( p-values) are

also included in Figure 3.

Figure 3(a) shows that as

Lincreased from range 1 to 3, the

median, first quartile and minimum values gradually

decreased. In contrast, the third quartile and maximum values

indicate the limited contribution of CNT length to higher

compressive strength. This can be statistically confirmed by

the p-values smaller than 0·05 between range 1 and other

ranges. Therefore, range 1 is identified as the optimum range

to increase the compressive strength, as highlighted in

Figure 3(a).

Figure 3(b) shows that the minimum, first quartile, median

and third quartile values increased as

Lincreased from range 1

to range 2. Δf

degraded beyond range 2, most probably due

to dispersion issues. The p-value of range 2 was lower than

0·05 when compared with length ranges of 1 and 3, indicating

that range 2 exhibited a higher mean flexural strength than the

other ranges.

Effect of CNT diameter

Figures 4(a) and 4(b) show boxplots of Δf

and Δf

, respect-

ively, with respect to the various ranges of

d(Table 5).

Figure 4(a) shows that as

dincreased from range 1 to 2, the mini-

mum and first quartile values increased. In contrast, the

maximum value of Δf

decreased from 50% to 35·5% and the

third quartiles remained with minimal changes (15·9–17·1%).

Therefore, CNTs within range 2 exhibited more cases of posi-

tive Δf

. However, their contribution is limited. This trend

suggests that CNTs of smaller diameter are beneficial for an

increase in compressive strength. However, due to difficulties

in achieving proper dispersion of small-diameter CNTs

(range 1), the likelihood of achieving Δf

< 0 is also high.

This might be attributed to the larger surface area of smaller

diameter CNTs, which causes them to agglomerate if not prop-

erly dispersed (Manzur et al., 2014). This can be statistically

confirmed by p-value smaller than 0·05 between ranges 2

and 3 ( p-value = 0·0033; see Figure 4(a)), while there was no

significant difference between the mean of Δf

by showing a

p-value of 0·3490 between ranges 1 and 2. In range 3, most

cases had a compressive strength lower than that of the control.

Therefore, range 2 was identified as the optimum range for

superior compressive strength, as highlighted in Figure 4(a).

Figure 4(b) shows that, despite the acceptable performance of

CNTs with small

din terms of compressive strength (see

range 1 in Figure 4(a)), CNTs with diameters within range 1

did not show a clear positive impact on achieving a higher

Δf

. This is confirmed by the higher minimum, first quartile,

median and third quartile values of CNTs with

dwithin

range 2 compared with range 1. In range 3, the likelihood of

obtaining a positive Δf

was significantly reduced compared

with range 2. Therefore, range 2 can be considered the opti-

mum range for superior flexural strength. This was statistically

confirmed by the p-values smaller than 0·05 between range 2

and the other ranges.

Effect of CNT concentration

Figures 5(a) and 5(b) show the effect of different ranges of

CNT ˉ

κon Δf

and Δf

, respectively. Figure 5(a) shows that

the dominant data for Δf

fell in the positive zone for CNT

concentration within range 1 (see the positive median and first

quartile values of range 1 in Figure 5(a)). Although the

p-value between ranges 1 and 2 showed no significant differ-

ence ( p-value = 0·2610 > 0·05), the compressive strength of

range 2 was degraded compared to range 1. The first quartile

300

250

200

150

100

–50

–100

ΔfCS: %

300

250

200

150

100

–50

–100

ΔfFS: %

Range 1 Range 2 Range 3 Range 1 Range 2

(a)

(b)

Range 3

d: nm

Range p-value

1 vs 2 0·3490

2 vs 3 0·0033

1 vs 3 0·1578

Range p-value

1 vs 2 <0·0001

2 vs 3 <0·0001

1 vs 3 0·8129

Figure 4. Effect of CNT diameter on (a) compressive strength and (b) flexural strength

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

value of range 2 was negative, while the median value was still

positive. Therefore, range 1 is considered the optimum range

for superior compressive strength. This will be confirmed in

the following section. In range 3, most of the data fell in the

negative zone. Therefore, this range of CNT concentration is

not recommended. The p-values smaller than 0·05 between

range 3 and the other ranges confirm this.

A similar trend can be observed in Figure 5(b) where the

first quartile, median and third quartile values continuously

decreased from range 1 to 3. Therefore, range 1 is considered

the optimum range for superior flexural strength, as high-

lighted in Figure 5(b).

Figures 3–5 demonstrate that, within the identified optimum

ranges, (a) lower variations in both Δf

and Δf

were

observed compared with other ranges and (b) the majority of

data were greater than zero (i.e. Δf

> 0 and Δf

> 0). This

further confirms the superior performance of CNTs if selected

from the optimum ranges. However, Figures 3–5 yield no infor-

mation on the CNT dispersion quality and its influence

on strength properties. This will be discussed in detail in the

following section.

Effect of CNT dispersion quality on

mechanical properties

This section discusses how the quality of dispersion affects the

strength of cementitious materials with respect to

dand ˉ

κ.

A good-quality dispersion leads to an increase in mechanical

properties as the CNT concentration increases. However, when

the quality of dispersion is poor, the mechanical properties are

degraded by the addition of CNTs (Jang et al., 2016; Kim

et al., 2014; Konsta-Gdoutos et al., 2010b; Wang et al.,

2013a). This study thus indirectly evaluated the quality of

dispersion using the relationship between the increment in

CNT concentration (Δˉ

κ¼ˉ

κ2ˉ

κ1;ˉ

κ2.ˉ

κ1) and the difference

between the measured strengths between two levels of CNT

concentration (ΔS; compressive strength ðΔSCS ¼fCSðˉ

κ2Þ

fCSðˉκ1ÞÞand flexural strength ðΔSFS ¼fFSðˉκ2ÞfFSðˉκ1ÞÞ), in each

individual study. Similarly, Konsta-Gdoutos et al. (2010b)

indirectly evaluated the dispersion quality using the measured

fracture load of cement pastes containing MWCNTs. Note

that, by comparing ΔS=Δˉ

κfor both compressive and flexural

strengths, it is possible to study how a degradation in dis-

persion quality affects compressive and flexural strengths. For

example, when short CNTs were used, many cases that showed

poor dispersion quality still had higher compressive strength

than the control sample. However, in the case of the flexural

strength, most cases that showed poor dispersion quality exhib-

ited lower flexural strength than the control. This indicates the

different contributions of CNT length to compressive and flex-

ural strengths.

With this in mind, the data obtained from the literature were

categorised into two groups –uniform dispersion and poor dis-

persion. This categorisation was merely done by taking into

account the ratio of strength change ðΔS¼Sˉ

κ2Sˉ

κ1Þto the

increment in CNT concentration (Δˉ

κ) for each individual study.

When the value of ΔS=Δˉ

κis positive, the strength is increased

by adding additional CNTs. This indicates possible uniform

dispersion. When the value of ΔS=Δˉ

κis negative or zero, the

compressive or flexural strength degrades by adding a higher con-

centration of CNTs. This indicates the possible poor dispersion

of CNTs within the cement matrix. Note that Δˉ

κis always posi-

tive; however, ΔScan be either positive (i.e. increase in strength

when adding additional CNTs) or negative (i.e. reduction in

strength when adding additional CNTs). Also, a negative value

of ΔS=Δˉ

κdoes not mean lower strength than the control.

300

250

200

150

100

–50

–100

ΔfCS: %

300

250

200

150

100

–50

–100

ΔfFS: %

Range 1 Range 2 Range 3 Range 1 Range 2

(a)

(b)

Range 3

κ: c-wt% κ: c-wt%

Range p-value

1 vs 2 0·2610

2 vs 3 0·0200

1 vs 3 <0·0001

Range p-value

1 vs 2 0·5743

2 vs 3 0·0477

1 vs 3 0·0001

Figure 5. Effect of CNT concentration on (a) compressive strength and (b) flexural strength

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

Figures 3–5 provide no information on the number of obser-

vations (N

obs

) for the quality of dispersion (ΔS=Δˉ

κ) and its

impact on strength changes within the proposed ranges against

the control (Δf

or Δf

). When counting either positive or

negative values of ΔS=Δˉ

κand Δf(Δf

or Δf

) with respect to

different parameters, it can be useful to assess the condition of

dispersion quality and the strength level of each range

simultaneously.

Effect of CNT length

Figure 6 shows the number of observations (N

obs

)ofeitherposi-

tive or negative value of ΔS=Δˉ

κand Δffor the given CNT length

ranges (Table 5). Figures 6(a) and 6(c) show the N

obs

of positive

and negative ΔS=Δˉ

κfor compressive strength ðΔSCS =Δˉ

κÞand

flexural strength ðΔSFS=Δˉ

κÞ, respectively. Figures 6(b) and 6(d)

respectively show the N

obs

of positive and negative Δf

and Δf

with respect to the different length ranges.

Figure 6(a) shows the N

obs

of positive and negative values of

the difference in the compressive strength between fCSðˉκ2Þand

fCSðˉ

κ1Þin terms of the increment in CNT concentration ðΔˉ

κ¼

κ2ˉ

κ1Þ.InthecaseofΔSCS=Δˉ

κ≤0 (i.e. poor dispersion),

shorter CNTs exhibited a higher N

obs

of positive Δf

than

longer CNTs, as shown in Figures 6(a) and 6(b). For example,

CNTs with

Lin range 1 showed 38 N

obs

of negative dispersion

indication (ΔSCS=Δˉ

κ≤0), while there were only 22 N

obs

negative Δf

(42% of poor dispersion still showed positive

Δf

). Increasing the CNT

Lto range 2 and range 3 led to only

17% and 7% Δf

> 0 in the case of ΔSCS=Δˉ

κ≤0, respectively.

This could be attributed to the effect of CNT length on the

pore size distribution (see Figure 1). The effect of pore size dis-

tribution on the strength has also been observed by other

researchers (Hu et al., 2014; Kang et al., 2015; Li et al., 2005).

For example, Wang et al. (2013b) incorporated a high concen-

tration of CNTs with

L=10 μm. Compared with plain cement

paste, they observed an increase in total porosity by 1·5%,

120

100

Nobs

Range 1 Range 2 Range 3

(a)

120

100

0Range 1 Range 2 Range 3

(b)

120

100

Nobs

Range 1 Range 2 Range 3

L: μm

(c)

120

100

0Range 1 Range 2 Range 3

L: μm

(d)

ΔSCS/Δκ > 0 ΔfCS > 0

ΔfCS ≤ 0

ΔSCS/Δκ ≤ 0

ΔSFS/Δκ > 0 ΔfFS > 0

ΔfFS ≤ 0

ΔSFS/Δκ ≤ 0

10 15

11 14

25 28 28 25

18 19 21

Figure 6. Effect of average length of CNTs ( ¯

L) on quality of dispersion (ΔS=Δ¯

κ) and strength gain/loss (Δf) for compressive and flexural

strengths

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

whereas the median volume pore diameter decreased by 26%.

Kang et al. (2015) also observed an increase in the compressive

strength of cement pastes containing CNTs with

L=12·5 μm,

resulting from a substantial decrease in average pore diameter

even though the total porosity was increased.

Despite the compressive strength, Figure 6(c) shows a higher

obs

of negative ΔSFS =Δˉ

κthan positive ΔSFS=Δˉ

κfor

Lin

range 1. In this range, the N

obs

of positive and negative Δf

were 28 and 25, respectively. This observation is in good agree-

ment with the findings in Figure 3(b), which showed short

CNTs (range 1) had a limited contribution to the increase of

Δf

. This might be explained by the ineffectiveness of short

CNTs in bridging cracks. With an increase in CNT length

to range 2, there were 35 N

obs

of ΔSFS=Δˉ

κ> 0 and 18 N

obs

ΔSFS=Δˉ

κ≤0. Also, 83% of the total N

obs

were correlated

withΔf

> 0 (44 positive out of 53 total N

obs

). This shows that

longer CNTs outperformed more in flexural strength than in

compressive strength, as highlighted in Figures 6(c) and 6(d).

However, an increase in CNT length to range 3 degraded the

quality of dispersion, adversely affecting the flexural strength

due to the creation of bigger pore sizes and premature debond-

ing of CNTs from the cement matrix.

In summary, shorter CNTs exhibited better performance in

terms of compressive strength, but this trend was not observed

for flexural strength. In general, CNTs with

L=10–20 μm

showed higher strength gain compared with the control, for

both compressive and flexural strengths. This effect was more

pronounced for flexural strength, which might be explained by

the better anchorage of longer CNTs in the hydration products,

which can bridge ongoing cracks.

Effect of CNT diameter

Figure 7 shows the N

obs

of positive or negative values of

ΔS=Δˉ

κand Δffor both compressive (Figures 7(a) and 7(b))

100

Nobs

Range 1 Range 2 Range 3

(a)

100

0Range 1 Range 2 Range 3

(b)

100

Nobs

Range 1 Range 2 Range 3

(c)

100

0Range 1 Range 2 Range 3

(d)

ΔfCS > 0

ΔfCS ≤ 0

ΔSCS/Δκ > 0

ΔSCS/Δκ ≤ 0

ΔfFS > 0

ΔfFS ≤ 0

ΔSFS/Δκ > 0

ΔSFS/Δκ ≤ 0

36 37

38 35 38

38 35

42 39 37

18 19

d: nm d: nm

Figure 7. Effect of average diameter of CNTs ( ¯

d) on quality of dispersion (ΔS=Δ¯

κ) and strength gain/loss (Δf) for compressive and

flexural strengths (Note 32·5 < ¯

d< 50 nm not available for flexural strength)

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

and flexural (Figures 7(c) and 7(d)) strengths for the given

ranges of CNT diameter.

As shown in Figure 7(a), the N

obs

of ΔSCS=Δˉ

κfor CNTs with

din range 1 was almost identical for poor (i.e. ΔSCS =Δˉ

κ0)

and uniform (i.e. ΔSCS=Δˉ

κ.0) dispersions. Nevertheless, 22%

of ΔSCS=Δˉ

κ≤0 still exhibited higher compressive strength

than the control (Δf

> 0; see Figure 7(b)). In the case of flex-

ural strength, only 12% of ΔSFS=Δˉ

κ≤0 for CNTs with

in range 1 exhibited Δf

> 0 (see Figure 7(d); five cases out of

42 total N

obs

). Thus, CNTs with diameters in range 1 tend to

perform better for compressive strength than for flexural

strength.

CNTs with 20 nm ≤

d≤45 nm for compressive strength and

15 nm ≤

d≤32·5 nm for flexural strength (range 2) dram-

atically reduced the number of cases that exhibited lower

strength than the control in Δf

and Δf

, respectively (see the

highlighted ranges in Figures 7(b) and 7(d)). For example, 51%

and 91% of the total N

obs

of Δf

were positive for range 1

and range 2, respectively. When considering the dispersion

quality, within the optimum range of

d(range 2), although

38 cases showed poor dispersion quality (see Figure 7(a)),

only 17 cases exhibited lower compressive strength than

the control (see Figure 7(b)). CNTs with

din range 3 degraded

both compressive and flexural strength gains. This might be

explained by the effect of CNT diameter on the pore

structure of the cement matrix (see Figure 1). Pore sizes

greater than 50 nm (referred to as macropores) have been

reported to have adverse effects on strength (Mehta and

Monteiro, 2006).

It was found that CNTs with smaller diameters had a more

positive influence on compressive strength than on flexural

strength. This might be explained by the different mechanisms

of CNTs in terms of compressive and flexural strength. In case

of compressive strength, CNTs with small diameters, even

when clumped together, can still act as fillers and contribute

to the refinement of pore structures. However, premature

debonding of agglomerated CNTs adversely affects the

load transfer mechanism between the cement matrix and

CNTs, thus degrading flexural strength. Generally, CNTs with

d=20–32·5 nm may be considered as the best range for

increasing both compressive and flexural strength.

Effect of CNT concentration

Figure 8 shows the effect of the concentration of CNTs on the

obs

of ΔS=Δˉ

κand Δffor both compressive and flexural

strengths. As shown in Figure 8(a), in range 1 ( ˉ

κ≤0·2 c-wt%),

the N

obs

of uniform dispersion (98 N

obs

for ΔSCS=Δˉ

κ>0) was

2·2 times more than that of poor dispersion (44 N

obs

for

ΔSCS=Δˉ

κ≤0). A similar trend was also observed for the flex-

ural strength, with the N

obs

of uniform and poor dispersion for

range 1 (ˉ

κ≤0·15 c-wt%) being 91 and 50, respectively (see

Figure 8(c)). Figures 8(b) and 8(d) show that there were many

cases of higher strength than the control in range 1. This indi-

cates that dispersion quality is highly correlated to strength

gain/loss.

Beyond range 1, increasing the CNT concentration con-

siderably increased the probability of obtaining lower strength

than the control. For example, in range 2, 35% of the overall

obs

exhibited lower compressive strength than the control.

The situation is even worse for CNTs with ˉ

κin range 3, which

showed 29 N

obs

of Δf

≤0, while there were only 18 N

obs

Δf

> 0 (Figure 8(b)). A similar trend was observed for flex-

ural strength (Figure 8(d)).

The results show that there are different thresholds of CNT

concentration for compressive and flexural strengths. As shown

in Figures 5 and 8, the threshold concentration for flexural

strength (ˉ

κ= 0·15 c-wt%) is slightly lower than that for com-

pressive strength (ˉ

κ= 0·2 c-wt%). When the CNT concentration

exceeds the threshold, both compressive and flexural strengths

decreased with an increase in CNT concentration. The differ-

ent mechanisms of CNTs with respect to compressive and flex-

ural strength could explain the different thresholds. In the

threshold of 0·2 c-wt%, agglomerated CNTs can degrade the

bond between the cement matrix and CNTs, resulting in a

reduction in flexural strength, while the agglomerated CNTs

can still be tolerable with respect to compressive strength. For

example, Danoglidis et al. (2016a) incorporated different con-

centrations of CNTs (in the range 0·08–0·50 c-wt%) to increase

the mechanical properties of cement mortars. Their results

indicated that, beyond ˉ

κ= 0·1 c-wt%, the flexural strength

degraded with an increase in CNT concentration, whereas

the compressive strength was not reduced with a higher

CNT concentration (this study was not included in the data

analysis).

Authors’experimental study (phase II)

To further confirm the proposed optimum ranges, experimen-

tal data obtained by the authors were assessed. Two different

types of MWCNTs were used; their specifications are listed in

Table 6. Both types of MWCNTs had the same length, which

was within the optimum range of CNT average length found

using the statistical analysis (

L=20 μm; the influence of length

was not experimentally investigated in this phase of the study).

The diameter of the type I MWCNTs was within the optimum

range of average diameters found using the statistical analysis

(

d= 25 nm). However, the type II MWCNTs were of diameter

not within the proposed optimum range. Three levels of CNT

concentrations (0·05, 0·10 and 0·30 c-wt%) were examined.

Four replications of 25 25 285 mm (1 111·25 in.)

beam specimens and three replications of 50 mm (2 in.) cube

specimens were cast for measurements of the 28 d flexural

strength and compressive strength, respectively. The w/c and s/c

ratios used were 0·35 and 2, respectively.

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

Figure 9 shows the influence of CNT diameter and con-

centration on the 28 d compressive strength (Figure 9(a)) and

flexural strength (Figure 9(b)), with the other experimental

variables (e.g. testing age, CNT length, CNT dispersion pro-

cedure etc.) fixed. When the MWCNT with the optimum

diameter was used (type I), both the compressive and flexural

strengths increased by increasing the CNT concentration from

0·05 to 0·10 c-wt%, which was within the optimal upper limit

for CNT concentration found using the statistical analysis

(as shown by the vertical dashed lines in Figure 9). However,

beyond the threshold concentrations, the mechanical properties

degraded by increasing the CNT dosage to κ= 0·30 c-wt%.

This was also confirmed by the p-values of <0·05 when com-

paring κ= 0·10 c-wt% with the other concentration levels

(κ= 0·05 and 0·30 c-wt%) for both compressive and flexural

strengths.

When the type II MWCNT was used, the mechanical pro-

perties decreased with an increase in CNT concentration.

However, the p-values suggest that, within the threshold limits

of concentration, the change in the mechanical properties from

0·05 to 0·10 c-wt% was not significant ( p-values of 0·12609

120

100

Nobs

Range 1 Range 2 Range 3

(a)

120

100

0Range 1 Range 2 Range 3

(b)

120

100

Nobs

Range 1 Range 2 Range 3

(c)

120

100

0Range 1 Range 2 Range 3

(d)

ΔfCS > 0

ΔfCS ≤ 0

ΔSCS/Δκ > 0

ΔSCS/Δκ ≤ 0

ΔfFS > 0

ΔfFS ≤ 0

ΔSFS/Δκ > 0

ΔSFS/Δκ ≤ 0

44 39 38

117

109

3231

κ: c-wt% κ: c-wt%

Figure 8. Effect of average CNT concentration ( ¯

κ) on quality of dispersion (ΔS=Δ¯

κ) and strength gain/loss (Δf) for compressive and

flexural strengths

Table 6. Properties of MWCNTs (type II MWCNT is COOH-functionalised (3·67–4·05 wt%))

Outside diameter: nm Inside diameter: nm Length: μm Aspect ratio Surface area: m

/g Purity: %

Type I 20–30 5–10 10–30 800 >500 >95

Type II <8 2–510–30 2500 >500 >95

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

and 0·35316 for compressive and flexural strength, respect-

ively). Beyond the threshold concentrations, the mechanical

properties degraded with the use of additional CNTs.

Within the threshold concentrations, the smaller diameter

CNTs (type II) were beneficial in increasing the compressive

strength, while the larger diameter CNTs (type I) benefited

the flexural strength. For example, when using κ= 0·10 c-wt%,

utilising either type I or II MWCNTs had no significant

influence on the compressive strength ( p-value > 0·05). On the

other hand, the type I MWCNTs contributed significantly to

higher flexural strength compared with the type II MWCNTs

(p-value = 0·0039 < 0·05).

Conclusions

The optimum ranges of CNT length, diameter and con-

centration for improving compressive and flexural strengths

were identified. To identify the optimum ranges, in the first

phase, experimental data from 42 studies (1359 data points) in

the literature were analysed. In the second phase, the authors’

experimental data were used to further verify the proposed

optimum ranges associated with CNTs. The analysis revealed

that the physical parameters of CNTs contribute differently to

the mechanical properties of cementitious materials. The fol-

lowing conclusions were drawn from this work.

(a) The length of CNTs was found to have a minimal

influence on compressive strength, but longer CNTs were

found to outperform in terms of flexural strength.

(b) CNTs with small diameters benefited compressive

strength but adversely affected flexural strength.

range 10–20 μm and 20–32·5 nm, respectively,

significantly contributed to the achievement of higher

mechanical performance.

(d) Concerning CNT concentration, optimal upper limits of

0·15 c-wt% and 0·20 c-wt% were obtained for flexural

and compressive strengths, respectively.

In addition to the proposed optimum ranges, there are other

important parameters to be considered to make stronger con-

crete, such as testing age, cement matrix composition, curing

condition, type of CNT and so on. Further research is thus

needed to investigate the possible contributions of these par-

ameters to the mechanical properties of cementitious materials.

Acknowledgements

The authors wish to express their gratitude and sincere appreci-

ation for the partial financial support from the Intramural

Research Incentive Grant and the Civil and Environmental

Engineering Department of the University of Louisville.

–10

Change in compressive strength: %

–10

Change in flexural strength: %

0 0·05 0·10 0·15 0·20 0·25 0·30 0·35

(a)

κ: c-wt%

κ =0·2 c-wt%

κ =0·15 c-wt%

0 0·05 0·10 0·15 0·20 0·25 0·30 0·35

(b)

κ: c-wt%

Type I

Type II

Threshold limit

(statistical analysis)

Threshold limit

(statistical analysis)

Figure 9. Influence of CNT diameter and concentration on (a) compressive strength and (b) flexural strength

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

Appendix

CNT characteristic

w/c ratio Dispersion method Age: d

Change in mechanical property: %

Type ¯

κ: c-wt% ¯

L:μm¯

d:nm CS DC FS DF

Lu et al. (2016)

MWCNTs 0·03 10 30 0·2 Nanosand mill + PVP + TP 7 4·6 G 5·1 G

0·05 28 5·6 G 7·5 G

0·1 2·9 B 3·3 B

0·15 0·7 B −0·7 B

0·03 4·2 G ——

0·05 4·6 G ——

0·1 2·1 B ——

0·15 −2B ——

Kim et al. (2014)

MWCNTs 0·15 10 26 0·25 DIY mix + SP + 0% SF 14 2 G ——

0·15 10% SF 32 G ——

0·15 20% SF 15 G ——

0·15 30% SF 16 G ——

0·3 0% SF −7B ——

0·3 10% SF 12 B ——

0·3 20% SF 15 B ——

0·3 30% SF −6B ——

Tamimi et al. (2016)

MWCNTs 0·15 20 30 0·3 DIY mix + 0% SF 14 6·5 G ——

C-MWCNTs 0% SF 8·7 G ——

O-MWCNTs 0% SF 8·7 G ——

H-MWCNTs 0% SF 13 G ——

MWCNTs 15% SF 3·6 G ——

C-MWCNTs 15% SF 3·6 G ——

O-MWCNTs 15% SF 7·1 G ——

H-MWCNTs 15% SF 12·5 G ——

MWCNTs 30% SF −5·1 B 33·8 G

C-MWCNTs 30% SF 20·3 G 50·6 G

O-MWCNTs 30% SF 15·2 G 47·4 G

H-MWCNTs 30% SF −16·94 B 30·4 G

Wang et al. (2014)

MWCNTs 0·02 10 30 0·3 US + GA 28 2·7 G 25·5 G

0·05 6·8 G 29·4 G

0·08 17·8 G 39·2 G

0·1 23·3 G 33·3 B

0·12 16·4 B 15·7 B

0·15 13·7 B 7·8 B

Fakhim et al. (2015)

MWCNTs 0·1 10 50 0·4 US + SP 28 ——50·4

0·3 ——69·2

0·5 ——7·4

1——−15·5

(continued on next page)

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

CNT characteristic

w/c ratio Dispersion method Age: d

Change in mechanical property: %

Type ¯

κ: c-wt% ¯

L:μm¯

d:nm CS DC FS DF

1·5 ——−36·3

2——−47

Li et al. (2015)

C-MWCNTs 0·1 20 15 0·45 Ball milling + SP 28 18·7 G 18·2 G

0·3 22·9 G 21·2 G

0·5 12·5 B 12·1 B

Li et al. (2005)

A-MWCNTs 0·5 0·5–500 20 0·45 Dry mix 28 18·9 G 25·1 G

Danoglidis et al. (2016b), Gdoutos et al. (2016)

MWCNTs 0·1 >10 32·5 0·485 US + SP 3 ——73·2 G

7——53·8 G

28 ——86·7 G

Morsy et al. (2011)

MWCNTs 0·005 —5·5 0·5 Dry mix +

nanometakaolin

28 5·5 G ——

0·02 11·1 G ——

0·05 3·7 B ——

0·1 −1·8 B ——

Elkashef et al. (2015)

A-MWCNTs 0·2 —35 0·5 US 28 32 G ——

Nochaiya et al. (2008)

MWCNTs 0·5 ——0·5 US 7 24·5 G 24 G

1 7 34·1 G 10·8 B

0·5 28 10·3 G 31 G

1 28 16·5 G 21·7 B

Chaipanich et al. (2010)

MWCNTs 0·5 ——0·5 US + FA 7 9·5 G ——

1 7 9·15 B ——

0·5 28 8 G ——

1 28 9·7 G ——

0·5 60 8·2 G ——

1 60 8·9 G ——

Xu et al. (2015)

MWCNTs 0·025 10 60 0·33 US+ TNWDIS 7 9·4 G ——

0·05 7 18·3 G ——

0·1 7 21·8 G ——

0·025 28 6·2 G 7·5 G

0·05 28 12·7 G 15 G

0·1 28 14·6 G 30 G

0·2 28 ——40 G

Wang et al. (2013b)

MWCNTs 0·05 10 30 0·35 US + GA 28 7 G 6 G

0·08 7·3 G 43·4 G

0·1 9·1 G 36·1 B

0·12 4·3 B 19·5 B

0·15 −5·7 B −4·6 B

(continued on next page)

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

CNT characteristic

w/c ratio Dispersion method Age: d

Change in mechanical property: %

Type ¯

κ: c-wt% ¯

L:μm¯

d:nm CS DC FS DF

Amin et al. (2015)

MWCNTs 0·02 7·5 25 0·3 US + SP 1 15·8 G ——

0·05 1 27·6 G ——

0·1 1 35·5 G ——

0·2 1 7·9 B ——

0·02 3 5·4 G ——

0·05 3 12·7 G ——

0·1 320 G ——

0·2 3 14·5 B ——

0·02 7 5·1 G ——

0·05 7 10·3 G ——

0·1 7 15·4 G ——

0·2 7 8·1 B ——

0·02 28 6·5 G ——

0·05 28 10 G ——

0·1 28 12·5 G ——

0·2 28 2·5 B ——

0·02 90 10 G ——

0·05 90 23 G ——

0·1 90 28·2 G ——

0·2 90 15·9 B ——

Metaxa et al. (2012)

MWCNTs 0·08 20 30 0·3 US + SP 3 ——43·7 G

7——34·4 G

28 ——37·6 G

Konsta-Gdoutos et al. (2010a)

MWCNTs 0·048 20 30 0·3 US + SP 3 ——22·5 G

0·08 20 3 ——45 G

0·1 20 3——31·2 B

0·025 10–100 3 ——28·7 G

0·048 10–100 3 ——33·7 G

0·08 10–100 3 ——15 B

0·048 20 7 ——18·9 G

0·08 20 7 ——34·4 G

0·1 20 7 ——25 B

0·025 10–100 7 ——22·2 G

0·048 10–100 7 ——25·5 G

0·08 10–100 7 —— 8·9 B

0·048 20 28 ——17·2 G

0·08 20 28 ——36·5 G

0·1 20 28 ——23·6 B

0·025 10–100 28 ——22·6 G

0·048 10–100 28 ——23·6 G

0·08 10–100 28 —— 9·7 B

Sun et al. (2016)

A-MWCNTs 0·05 >1 13 0·3 US 28 ——99·4 G

(continued on next page)

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

CNT characteristic

w/c ratio Dispersion method Age: d

Change in mechanical property: %

Type ¯

κ: c-wt% ¯

L:μm¯

d:nm CS DC FS DF

Luo et al. (2009)

MWCNTs 0·2 10 30 0·4 US + SDBS 28 13·8 G 15 G

US + NaDC 29·5 G 35 G

US + GA 2·5 G 4·7 G

US + SDBS + TX10 21 G 28·7 G

Hunashyal et al. (2011)

MWCNTs 0·25 1·5 20 0·4 US in cement and ethanol 28 —— 3·7 G

0·5 ——25·9 G

0·65 ——85·2 G

0·75 ——−48·1 B

Kang et al. (2015)

A-MWCNTs 0·15 13 20 0·4 Dry mix + SP + SF 14 57·8 G 25·6

MWCNTs SP + SF 17·8 G 11·6

A-MWCNTs SF 68·9 G 46·5

Kumar et al. (2012)

MWCNTs 0·5 25·25 80 0·4 US 7 21·4 G 17·8

0·75 7 16·7 B −1·9

1 7 −16·7 B −41·2

0·5 28 14·7 G 36·3

0·75 28 −2·9 B 24·1

1 28 −29·4 B 5·4

0·5 60 7·9 G 9·8

0·75 60 −2·4 B 0

1 60 −27 B −10

0·5 90 11·6 G 9·8

0·75 90 −2·1 B −13

1 90 −28·6 B −43·3

0·5 180 12·7 G 9·8

0·75 180 −1·6 B 0

1 180 −27 B −20

Al-Rub et al. (2012b)

MWCNTs 0·1 1·5 9·5 0·4 US + SP 7 ——−29·2 B

MWCNTs 0·2 7 ——−67·7 B

A-MWCNTs 0·1 7 ——63·1 G

A-MWCNTs 0·2 7 ——50·8 B

MWCNTs 0·1 14 ——−69·7 B

MWCNTs 0·2 14 ——−53·9 B

A-MWCNTs 0·1 14 ——−60·5 B

A-MWCNTs 0·2 14 ——−59·2 B

MWCNTs 0·1 28 ——−47·1 B

MWCNTs 0·2 28 ——37·9 G

A-MWCNTs 0·1 28 ——−51·7 B

A-MWCNTs 0·2 28 ——−73 B

Tyson et al. (2011)

MWCNTs 0·1 1·5 9·5 0·4 US + SP 7 ——−21·4 B

0·2 7 ——−64·3 B

0·1 14 ——−73·3 B

(continued on next page)

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

CNT characteristic

w/c ratio Dispersion method Age: d

Change in mechanical property: %

Type ¯

κ: c-wt% ¯

L:μm¯

d:nm CS DC FS DF

0·2 14 ——−60 B

0·1 28 ——−52·9 B

0·2 28 ——41·2 G

Al-Rub et al. (2012a)

MWCNTs 0·04 1·5 9·5 0·4 US +SP 7 ——76·7 G

0·1 1·5 9·5 7 ——20·9 B

0·2 1·5 9·5 7 ——−34·9 B

0·04 20 <8 7 ——27·9 G

0·1 20 <8 7 ——39·5 G

0·04 1·5 9·5 14 ——−22·2 B

0·1 1·5 9·5 14 ——−46·7 B

0·2 1·5 9·5 14 ——−28·9 B

0·04 20 <8 14 ——26·7 G

0·1 20 <8 14 ——−33·3 B

0·04 1·5 9·5 28 ——18·2 G

0·1 1·5 9·5 28 ——33·3 G

0·2 1·5 9·5 28 ——260·6 G

0·04 20 <8 28 ——51·5 G

0·1 20 <8 28 ——60·6 G

Sobolkina et al. (2012)

Mixed CNTs 0·05 20 <8 0·4 US + SDS 28 44·8 G 6·7

0·25 SDS 20·7 B 6·7

0·05 Brij 35 15 G −30

0·25 Brij 35 −12·5 B −15

Lelusz (2015)

MWCNTs 0·06 >2 11·5 0·45 US + SP 7 −9·4 B ——

0·12 7 −16·3 B ——

0·06 28 29·7 G ——

0·12 28 4·6 B ——

Nasibulina et al. (2012)

A-MWCNTs 0·02 —7 0·4 US 28 83·3 G ——

0·03 US 97·2 G ——

0·05 US 80·5 B ——

0·09 US 63·9 B ——

—US + SDS −65 B ——

Musso et al. (2009)

MWCNTs 0·5 400–1000 60 0·4 SP + VMA 28 10·6 G 34·7 G

C-MWCNTs 5·1 15 0·4 −85·1 B −61·3 B

C-MWCNTs 5·1 15 0·56 −26·3 B −3·2 B

Hu et al. (2014)

MWCNTs 0·05 20 15 0·2 US + surfactants 28 0·5 G ——

MWCNTs 0·1 15 −2·3 B ——

C-MWCNTs 0·05 <8 5·3 G ——

C-MWCNTs 0·1 <8 5B ——

Choi et al. (2015)

MWCNTs 1 15 5 0·4 US + surfactant 7 34·6 G ——

0·4 14 50 G ——

0·4 28 34 G ——

(continued on next page)

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

CNT characteristic

w/c ratio Dispersion method Age: d

Change in mechanical property: %

Type ¯

κ: c-wt% ¯

L:μm¯

d:nm CS DC FS DF

0·6 7 −27·8 B ——

0·6 14 9·5 G ——

0·6 28 4 G ——

0·65 7 −35·3 B ——

0·7 7 −43·7 B ——

0·35 7 22·2 G ——

0·45 7 12 G ——

Manzur and Yazdani (2010)

MWCNTs 0·1 15 50 0·6 US + —7 6·2 G ——

0·2 15 50 0·6 —7 7·8 G ——

0·3 15 50 0·485 —7−10·6 B ——

0·3 15 50 0·6 —7 6·2 B ——

0·3 15 50 0·485 SP 7 3·6 G ——

0·3 15 50 0·485 SP 7 5·3 G ——

0·3 15 50 0·485 SP 7 2 G ——

0·5 15 50 0·485 —7−18·7 B ——

0·5 15 50 0·6 —77G ——

0·5 15 50 0·485 SP 7 4·5 G ——

0·5 15 50 0·485 SP 7 6·9 G ——

0·5 15 50 0·485 SP 7 −17·9 B ——

0·8 15 50 0·485 —7−29·3 B ——

0·8 15 50 0·6 —7−4·1 B ——

0·8 15 50 0·485 SP 7 −10·6 B ——

0·8 15 50 0·485 SP 7 −3·6 B ——

0·3 20 25 0·485 —713 G ——

0·3 20 25 0·6 —7 28·5 G ——

0·3 20 25 0·485 SP 7 17·9 G ——

0·5 20 25 0·6 —7 1·6 B ——

0·8 20 25 0·6 —7 0·4 B ——

0·1 15 50 0·6 —28 10·2 G ——

0·2 15 50 0·6 —28 10·8 G ——

0·3 15 50 0·485 —28 −0·5 B ——

0·3 15 50 0·6 —28 16·5 G ——

0·3 15 50 0·485 SP 28 9·1 G ——

0·3 15 50 0·485 SP 28 8 G ——

0·3 15 50 0·485 SP 28 10·3 G ——

0·5 15 50 0·485 —28 −25·9 B ——

0·5 15 50 0·6 —28 8·4 B ——

0·5 15 50 0·485 SP 28 −12·4 B ——

0·5 15 50 0·485 SP 28 −16·5 B ——

0·5 15 50 0·485 SP 28 −24·2 B ——

0·8 15 50 0·485 —28 −30·1 B ——

0·8 15 50 0·6 —28 −1·5 B ——

0·8 15 50 0·485 SP 28 −16·5 B ——

0·8 15 50 0·485 SP 28 −7·4 B ——

0·3 20 25 0·485 —28 3·83 G ——

0·3 20 25 0·6 —28 24·8 G ——

0·3 20 25 0·485 SP 28 −0·6 B ——

(continued on next page)

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

CNT characteristic

w/c ratio Dispersion method Age: d

Change in mechanical property: %

Type ¯

κ: c-wt% ¯

L:μm¯

d:nm CS DC FS DF

0·5 20 25 0·6 —28 3·6 B ——

0·8 20 25 0·6 —28 −1·2 B ——

Jang et al. (2016)

MWCNTs 0·05 12·5 20 0·5 US + SP 28 2·2 G 6·1

0·1 US + SP 6·5 G 30·6

0·25 US + SP 11·9 G 36·7

0·5 US + SP 9·8 B 14·3

0·05 US −1·1 B 10·9

0·1 US −3·3 B 26·1

0·25 US −11 B 32·6

0·5 US −12·1 B 8·7

Konsta-Gdoutos et al. (2010b)

MWCNTs 0·048 20 30 0·5 US + SP 3 ——13·1 G

0·08 20 3 ——34·2 G

0·048 10–100 3 ——26·3 G

0·08 10–100 3 ——15·8 B

0·048 20 7 ——14·6 G

0·08 20 7 ——22·9 G

0·048 10–100 7 ——20·8 G

0·08 10–100 7 ——18·8 B

0·048 20 28 ——12·9 G

0·08 20 28 ——20·4 G

0·048 10–100 28 ——25·9 G

0·08 10–100 28 ——22·2 B

Collins et al. (2012)

MWCNTs 0·5 10 80 0·5 —28 −7·8 B ——

20 —−6·5 B ——

15 —−11·2 B ——

80 US 200·4 G ——

80 Air entrainer −10·1 B ——

80 US + air entrainer −13·8 B ——

80 Lignosulfonate −28·5 B ——

80 US + lignosulfonate −35·4 B ——

80 SP 25·8 G ——

80 US + SP 22·9 G ——

Ludvig et al. (2011)

CNTs/CNFs 0·3 ——0·4 Direct synthesis +

lignosulfonate

7——24·2 G

28 88·3 G 14·1 G

Parveen et al. (2015)

MWCNTs 0·1 20 <8 0·5 US (bath) + surfactants 28 −42·7 B −27·1 B

MWCNTs 0·1 17·5 1·5 28 −15·5 B ——

MWCNTs 0·1 42 −8·9 B ——

MWCNTs 0·08 56 −1·7 B ——

C-MWCNTs 0·1 28 −10·1 B ——

C-MWCNTs 0·1 42 −0·6 B ——

MWCNTs 0·1 56 ——−11·2 B

C-MWCNTs 0·1 28 ——−0·8 B

(continued on next page)

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

CNT characteristic

w/c ratio Dispersion method Age: d

Change in mechanical property: %

Type ¯

κ: c-wt% ¯

L:μm¯

d:nm CS DC FS DF

MWCNTs 0·1 ——−16 B

C-MWCNTs 0·1 ——−8B

MWCNTs 0·08 —— 3·1 B

C-MWCNTs 0·08 ——−6·1 B

SWCNTs 0·1 —— 6·2 G

C-SWCNTs 0·1 ——−0·8 B

SWCNTs 0·1 19·1 G 6·7 G

SWCNTs 0·1 9·8 G ——

C-SWCNTs 0·1 −8·9 B −6·4 B

SWCNTs 0·08 15·4 G 5·6 G

SWCNTs 0·08 12·4 G −18·9 B

SWCNTs 0·08 12·36 G 13·3 G

C-SWCNTs 0·08 −12·6 B −10·2 B

C-SWCNTs 0·08 1·12 G 11·9 G

C-SWCNTs 0·08 20·8 G 17·5 G

C-SWCNTs 0·1 1·1 ———

Zou et al. (2015)

C-MWCNTs 0·038 1·5 9·5 0·4 US + SP 28 ——10·7 G

0·038 ——25·3 G

0·038 ——24·8 G

0·038 ——19·5 G

0·038 ——17 G

0·075 —— 7·8 G

0·075 ——32·1 G

0·075 ——49·9 G

0·075 ——48·7 G

0·075 ——38·2 G

Cwirzen et al. (2008)

MWCNTs 0·0007 <10 10 0·3 US + PAP 28 −14·6 B −40·3 B

MWCNTs 0·03 <10 0·3 PAP −39 B −22·8 B

MWCNTs 0·042 <10 0·3 PAP −6·7 B 17·5 G

MWCNTs 0·03 <10 0·3 GA 5·3 B 2·2 B

MWCNTs 0·0007 <10 0·3 GA 13·2 G 11·1 G

MWCNTs 0·042 <10 0·3 GA −21·1 B −24·4 B

C-MWCNTs 0·06 3 0·4 —−0·5 B 13·3 G

C-MWCNTs 0·05 3 0·4 PAP 6·06 G 6·7 G

C-MWCNTs 0·15 3 0·3 PAP 1·92 G 15·9 G

C-MWCNTs 0·045 3 0·3 PAP 63·16 G −0·5 B

Petrunin et al. (2013)

MWCNTs 0·01 —45 0·3 US + SP 1 −0·5 B ——

MWCNTs 0·02 1 6·8 G ——

MWCNTs 0·05 1 22·7 G ——

A-MWCNTs 0·05 1 29·5 G ——

MWCNTs 0·13 1 36·4 G ——

MWCNTs 0·25 1 −2·3 B ——

MWCNTs 0·01 7 2·1 G ——

(continued on next page)

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

CNT characteristic

w/c ratio Dispersion method Age: d

Change in mechanical property: %

Type ¯

κ: c-wt% ¯

L:μm¯

d:nm CS DC FS DF

MWCNTs 0·02 7 2·1 G ——

MWCNTs 0·05 7 6·4 G ——

A-MWCNTs 0·05 7 10·6 G ——

MWCNTs 0·13 7 10·6 B ——

MWCNTs 0·25 7 8·5 B ——

MWCNTs 0·01 28 3·9 G ——

MWCNTs 0·02 28 11·8 G ——

MWCNTs 0·05 28 21·6 G ——

A-MWCNTs 0·05 28 11·7 G ——

MWCNTs 0·13 28 25·5 G ——

MWCNTs 0·25 28 2 B ——

Stynoski et al. (2015)

MWCNTs 0·125 21·25 30 0·485 US + SP + —7—— 5·1 G

MWCNTs SF 7 —— 4·6 G

S-MWCNTs —7——−1B

S-MWCNTs —7—— 2G

MWCNTs —28 —— 4·4 G

MWCNTs SF 28 —— 4·6 G

S-MWCNTs —28 —— 9·2 G

S-MWCNTs —28 —— 7·2 G

Direct tension test

Splitting tensile strength

AR, average aspect ratio; CS, compressive strength; DC, CNT dispersion quality for compressive strength; FS, flexural strength; DF, CNT dispersion quality for flexural strength.

C-MWCNTs, COOH-functionalised MWCNTs; O-MWCNTs, OH-functionalised MWCNTs; A-MWCNTs, acid-treated MWCNTs; H-MWCNTs, commercial pre-dispersed MWCNTs in water; S-MWCNTs, silica-

functionalised MWCNTs

Brij 35, polyoxyethylene (23) lauryl ether; NaDC, sodium deoxycholate; SDBS, sodium dodecyl benzene sulfonate; SDS, sodium dodecyl sulfate; TNWDIS, CNTs-based water dispersant, Chengdu Organic

Chemicals Co; TX10, Triton X-100

US; ultrasonication; SP, superplasticiser; SF, silica fume; FA, fly ash; PVP, polyvinyl pyrrolidone; TP, tributyl phosphate; GA, gum arabic; VMA, viscosity-modifying agent; PAP, polyacrylic acid polymer

B, bad; G, good

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

REFERENCES

Al-Rub RKA, Ashour A and Tyson BM (2012a) On the aspect ratio effect

of multi-walled carbon nanotube reinforcements on the

mechanical properties of cementitious nanocomposites.

Construction and Building Materials 35: 647–655.

Al-Rub RKA, Tyson BM, Yazdanbakhsh A and Grasley Z (2012b)

Mechanical properties of nanocomposite cement incorporating

surface-treated and untreated carbon nanotubes and carbon

nanofibers. Journal of Nanomechanics and Micromechanics 2(1):

1–6.

Amin MS, El-Gamal SMA and Hashem FS (2015) Fire resistance and

mechanical properties of carbon nanotubes–clay bricks wastes

(Homra) composites cement. Construction and Building Materials

98: 237–249.

Bakshi SR and Agarwal A (2011) An analysis of the factors affecting

strengthening in carbon nanotube reinforced aluminum

composites. Carbon 49(2): 533–544.

Balaguru PN and Shah SP (1992) Fiber-Reinforced Cement Composites.

McGraw-Hill, New York, NY, USA.

Banthia N and Nandakumar N (2003) Crack growth resistance of hybrid

fiber reinforced cement composites. Cement & Concrete

Composites 25(1):3–9.

Bentur A and Mindess S (2006) Fibre Reinforced Cementitious

Composites. CRC Press, Boca Raton, FL, USA.

Chaipanich A, Nochaiya T, Wongkeo W and Torkittikul P (2010)

Compressive strength and microstructure of carbon nanotubes–fly

ash cement composites. Materials Science and Engineering: A

527(4): 1063–1067.

Chen B, Kondoh K, Imai H, Umeda J and Takahashi M (2016)

Simultaneously enhancing strength and ductility of carbon

nanotube/aluminum composites by improving bonding conditions.

Scripta Materialia 113: 158–162.

Chen B, Shen J, Ye X et al. (2017) Length effect of carbon nanotubes on

the strengthening mechanisms in metal matrix composites. Acta

Materialia 140: 317–325.

Choi H, Kang D, Seo G and Chung W (2015) Effect of some parameters

on the compressive strength of MWCNT-cement composites.

Advances in Materials Science and Engineering 2015:1–8.

Chuah S, Pan Z, Sanjayan JG, Wang CM and Duan WH (2014) Nano

reinforced cement and concrete composites and new perspective

from graphene oxide. Construction and Building Materials 73:

113–124.

Collins F, Lambert J and Duan WH (2012) The influences of admixtures

on the dispersion, workability, and strength of carbon

nanotube–OPC paste mixtures. Cement & Concrete Composites

34(2): 201–207.

Cwirzen A, Habermehl-Cwirzen K and Penttala V (2008) Surface

decoration of carbon nanotubes and mechanical properties of

cement/carbon nanotube composites. Advances in Cement

Research 20(2):65–73, https://doi.org/10.1680/adcr.2008.20.2.65.

Cwirzen A, Habermehl-Cwirzen K, Nasibulina LI et al. (2009) CHH

cement composite. In Nanotechnology in Construction: Proceedings

of NICOM3 (Bittnar Z, Bartos PJM, Nemecek J, Smilauer V and

Zeman J (eds)). Springer, Berlin, Germany, pp. 181–185.

Danoglidis PA, Konsta-Gdoutos MS, Gdoutos EE and Shah SP (2016a)

Strength, energy absorption capability and self-sensing properties

of multifunctional carbon nanotube reinforced mortars.

Construction and Building Materials 120: 265–274.

Danoglidis PA, Falara MG, Katotriotou MK, Konsta-Gdoutos MS and

Gdoutos EE (2016b) MWCNT and CNF cementitious

nanocomposites for enhanced strength and toughness. Mechanics

of Composite and Multi-functional Materials 7: 241–246.

del Carmen Camacho M, Galao O, Baeza FJ, Zornoza E and Garcés P

(2014) Mechanical properties and durability of CNT cement

composites. Materials 7(3): 1640–1651.

Demczyk BG, Wang YM, Cumings J et al. (2002) Direct mechanical

measurement of the tensile strength and elastic modulus of

multiwalled carbon nanotubes. Materials Science and Engineering:

A334(1): 173–178.

De Volder MFL, Tawfick SH, Baughman RH and Hart AJ (2013) Carbon

nanotubes: present and future commercial applications. Science

339(6119): 535–539.

Elkashef M, Wang K and Abou-Zeid MN (2015) Acid-treated carbon

nanotubes and their effects on mortar strength. Frontiers of

Structural and Civil Engineering 2:1–9.

Fakhim B, Hassani A, Rashidi A and Ghodousi P (2015) Preparation and

microstructural properties study on cement composites reinforced

with multi-walled carbon nanotubes. Journal of Composite

Materials 49(1):85–98.

Fawaz J and Mittal V (2014) Polymer nanotube nanocomposites: a

review of synthesis methods, properties and applications.

In Polymer Nanotube Nanocomposites: Synthesis, Properties,

and Applications (Mittal V (ed.)). Wiley, Hoboken, NJ, USA,

pp. 1–44.

Gdoutos EE, Konsta-Gdoutos MS, Danoglidis PA and Shah SP (2016)

Advanced cement based nanocomposites reinforced with

MWCNTs and CNFs. Frontiers of Structural and Civil Engineering

10(2): 142–149.

Grossiord N, Regev O, Loos J, Meuldijk J and Koning CE (2005)

Time-dependent study of the exfoliation process of carbon

nanotubes in aqueous dispersions by using UV-visible

spectroscopy. Analytical Chemistry 77(16): 5135–5139.

Halpin J and Kardos J (1976) The Halpin–Tsai equations: a review.

Polymer Engineering & Science 16(5): 344–352.

Han B, Yang Z, Shi X and Yu X (2013) Transport properties of

carbon-nanotube/cement composites. Journal of Materials

Engineering and Performance 22(1): 184–189.

Hu Y, Luo D, Li P, Li Q and Sun G (2014) Fracture toughness

enhancement of cement paste with multi-walled carbon nanotubes.

Construction and Building Materials 70: 332–338.

Hunashyal A, Sundeep GV, Quadri SS and Banapurmath NR (2011)

Experimental investigations to study the effect of carbon

nanotubes reinforced in cement-based matrix composite beams.

Proceedings of the Institution of Mechanical Engineers, Part N:

Journal of Nanoengineering and Nanosystems 225(1)17–22.

Jang SH, Kawashima S and Yin H (2016) Influence of carbon nanotube

clustering on mechanical and electrical properties of cement

pastes. Materials 9(4):1–11.

Kang ST, Seo JY and Park SH (2015) The characteristics of CNT/cement

composites with acid-treated MWCNTs. Advances in Materials

Science and Engineering 2015:1–9.

Kelly A and Tyson AW (1965) Tensile properties of fibre-reinforced

metals: copper/tungsten and copper/molybdenum. Journal of the

Mechanics and Physics of Solids 13(6): 329–350.

Kim HK, Nam IW and Lee HK (2014) Enhanced effect of carbon

nanotube on mechanical and electrical properties of cement

composites by incorporation of silica fume. Composite Structures

107:60–69.

Konsta-Gdoutos MS, Metaxa ZS and Shah SP (2010a) Multi-scale

mechanical and fracture characteristics and early-age strain

capacity of high performance carbon nanotube/cement

nanocomposites. Cement & Concrete Composites 32(2): 110–115.

Konsta-Gdoutos MS, Metaxa ZS and Shah SP (2010b) Highly dispersed

carbon nanotube reinforced cement based materials. Cement and

Concrete Research 40(7): 1052–1059.

Konsta-Gdoutos MS, Danoglidis PA, Falara MG and Nitodas SF (2017)

Fresh and mechanical properties, and strain sensing of

nanomodified cement mortars: the effects of MWCNT aspect

ratio, density and functionalization. Cement & Concrete

Composites 82: 137–151.

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

Kumar S, Kolay P, Malla S and Mishra S (2012) Effect of multiwalled

carbon nanotubes on mechanical strength of cement paste. Journal

of Materials in Civil Engineering 24(1):84–91.

Lachman N, Wiesel E, de Villoria RG, Wardle BL and Wagner HD (2012)

Interfacial load transfer in carbon nanotube/ceramic microfiber

hybrid polymer composites. Composites Science and Technology

72(12): 1416–1422.

Lawler JS, Wilhelm T, Zampini D and Shah SP (2003) Fracture processes

of hybrid fiber-reinforced mortar. Materials and Structures 36(3):

197–208.

Lelusz M (2015) Carbon nanotubes influence on the compressive

strength of cement composites. In Czasopismo Techniczne

(Gawlick J (ed.)). Wydawnictwo Politechniki Krakowskiej,

Kraków, Poland, pp. 5–11.

Li GY, Wang PM and Zhao X (2005) Mechanical behavior and

microstructure of cement composites incorporating surface-treated

multi-walled carbon nanotubes. Carbon 43(6): 1239–1245.

Li WW, Ji WM, Wang YC et al. (2015) Investigation on the mechanical

properties of a cement-based material containing carbon nanotube

under drying and freeze-thaw conditions. Materials 8(12):

8780–8792.

Liao JZ, Tan MJ and Sridhar I (2010) Spark plasma sintered multi-wall

carbon nanotube reinforced aluminum matrix composites.

Materials & Design 31: S96–S100.

Lu L, Ouyang D and Xu W (2016) Mechanical properties and durability

of ultra high strength concrete incorporating multi-walled carbon

nanotubes. Materials 9(6):1–11.

Ludvig P, Calixto JM, Ladeira LO and Gaspar ICP (2011) Using converter

dust to produce low cost cementitious composites by in situ

carbon nanotube and nanofiber synthesis. Materials 4(3): 575–584.

Luo J, Duan Z and Li H (2009) The influence of surfactants on the

processing of multi-walled carbon nanotubes in reinforced cement

matrix composites. Physica Status Solidi a 206(12): 2783–2790.

Makar JM, Margeson JC and Luh J (2005) Carbon nanotube/cement

composites-early results and potential applications. Proceedings of

the 3rd International Conference on Construction Materials:

Performance, Innovations and Structural Implications, Vancouver,

BC, Canada (Banthia N, Uomoto T, Bentur A and Shah SP

(eds)). UBC Press, Vancouver, Canada, pp. 1–10.

Manzur T and Yazdani N (2010) Strength enhancement of cement

mortar with carbon nanotubes: early results and potential.

Transportation Research Record 2142: 102–108.

Manzur T, Yazdani N and Emon MAB (2014) Effect of carbon nanotube

size on compressive strengths of nanotube reinforced cementitious

composites. Journal of Materials 2014:1–8.

Mehta K and Monteiro P (2006) Concrete: Microstructure, Properties,

and Materials. McGraw-Hill, New York, NY, USA.

Metaxa ZS, Seo JWT, Konsta-Gdoutos MS, Hersam MC and Shah SP (2012)

Highly concentrated carbon nanotube admixture for nano-fiber

reinforced cementitious materials. Cement & Concrete Composites

34(5): 612–617.

Morsy MS, Alsayed SH and Aqel M (2011) Hybrid effect of carbon

nanotube and nano-clay on physico-mechanical properties of

cement mortar. Construction and Building Materials 25(1):

145–149.

Mudimela PR, Nasibulina LI, Nasibulin AG et al. (2009) Synthesis of

carbon nanotubes and nanofibers on silica and cement matrix

materials. Journal of Nanomaterials 2009:1–4.

Musso S, Tulliani JM, Ferro G and Tagliaferro A (2009) Influence of

carbon nanotubes structure on the mechanical behavior of cement

composites. Composites Science and Technology 69(11):

1985–1990.

Nasibulin AG, Shandakov SD, Nasibulina LI et al. (2009) A novel

cement-based hybrid material. New Journal of Physics 11(2):

1–10.

Nasibulin AG, Koltsova T, Nasibulina LI et al. (2013) A novel approach to

composite preparation by direct synthesis of carbon nanomaterial

on matrix or filler particles. Acta Materialia 61(6): 1862–1871.

Nasibulina LI, Anoshkin IV, Nasibulin AG et al. (2012) Effect of carbon

nanotube aqueous dispersion quality on mechanical properties of

cement composite. Journal of Nanomaterials 2012:1–6.

Nochaiya T, Tolkidtikul P, Singjai P and Chaipanich A (2008)

Microstructure and characterizations of Portland-carbon

nanotubes pastes. Advanced Materials Research 55: 549–552.

Park SH and Bandaru PR (2010) Improved mechanical properties of

carbon nanotube/polymer composites through the use of

carboxyl-epoxide functional group linkages. Polymer 51(22):

5071–5077.

Parveen S, Rana S, Fangueiro R and Paiva MC (2015) Microstructure and

mechanical properties of carbon nanotube reinforced cementitious

composites developed using a novel dispersion technique. Cement

and Concrete Research 73: 215–227.

Petrunin S, Vaganov V and Sobolev K (2013) The effect of functionalized

carbon nanotubes on the performance of cement composites.

In Proceedings of the 5th International Conference NANOCON,

Brno, Czech Republic. Tanger Ltd, Ostrava, Czech Republic,

pp. 1–5.

Qian CX and Stroeven P (2000) Development of hybrid

polypropylene-steel fibre-reinforced concrete. Cement and

Concrete Research 30(1):63–69.

Raki L, Beaudoin JJ, Alizadeh R, Makar JM and Sato T (2010) Cement

and concrete nanoscience and nanotechnology. Materials 3(2):

918–942.

Ramezani M, Kim YH, Hasanzadeh B and Sun Z (2016) Influence of

carbon nanotubes on SCC flowability. Proceedings of 8th

International Rilem Symposium on Self-Compacting Concrete,

Washington DC, USA (Khayat KH (ed.)). Rilem Publications,

Bagneux, France, pp. 397–406.

Shah SP, Hou P and Konsta-Gdoutos MS (2016) Nano-modification of

cementitious material: toward a stronger and durable concrete.

Journal of Sustainable Cement-Based Materials 5(1-2):1–22.

Sobolkina A, Mechtcherine V, Khavrus V et al. (2012) Dispersion of

carbon nanotubes and its influence on the mechanical properties

of the cement matrix. Cement & Concrete Composites 34(10):

1104–1113.

Song PS, Hwang S and Sheu BC (2005) Strength properties of nylon-and

polypropylene-fiber-reinforced concretes. Cement and Concrete

Research 35(8): 1546–1550.

Strano MS, Moore VC, Miller MK et al. (2003) The role of surfactant

adsorption during ultrasonication in the dispersion of

single-walled carbon nanotubes. Journal of Nanoscience and

Nanotechnology 3(1-2):81–86.

Stynoski P, Mondal P and Marsh C (2015) Effects of silica additives on

fracture properties of carbon nanotube and carbon fiber reinforced

Portland cement mortar. Cement & Concrete Composites 55:

232–240.

Sun G, Liang R, Lu Z, Zhang J and Li Z (2016) Mechanism of

cement/carbon nanotube composites with enhanced mechanical

properties achieved by interfacial strengthening. Construction and

Building Materials 115:87–92.

Tamimi A, Hassan NM, Fattah K and Talachi A (2016) Performance of

cementitious materials produced by incorporating surface treated

multiwall carbon nanotubes and silica fume. Construction and

Building Materials 114: 934–945.

Tyson BM (2010) Carbon Nanotube and Nanofiber Reinforcement

for Improving the Flexural Strength and Fracture Toughness of

Portland Cement Paste. Texas A&M University, College Station,

TX, USA.

Tyson BM, Al-Rub RKA, Yazdanbakhsh A and Grasley Z (2011) Carbon

nanotubes and carbon nanofibers for enhancing the mechanical

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

properties of nanocomposite cementitious materials. Journal of

Materials in Civil Engineering 23(7): 1028–1035.

Wang B, Han Y and Liu S (2013a) Effect of highly dispersed carbon

nanotubes on the flexural toughness of cement-based composites.

Construction and Building Materials 46:8–12.

Wang B, Han Y, Pan B and Zhang T (2013b) Mechanical and

morphological properties of highly dispersed carbon nanotubes

reinforced cement based materials. Journal of Wuhan University of

Technology –Materials Science Edition 28(1): 82–87.

Wang B, Han Y and Zhang T (2014) Reinforcement of surface-modified

multi-walled carbon nanotubes on cement-based composites.

Advances in Cement Research 26(2):77–84, https://doi.

org/10.1680/adcr.12.00074

Wichmann MHG, Schulte K and Wagner HD (2008) On nanocomposite

toughness. Composites Science and Technology 68(1):

329–331.

Xu S, Liu J and Li Q (2015) Mechanical properties and microstructure of

multi-walled carbon nanotube-reinforced cement paste.

Construction and Building Materials 76:16–23.

Yakovlev G, Pervushin G, Maeva I et al. (2013) Modification of

construction materials with multi-walled carbon nanotubes.

Procedia Engineering 57: 407–413.

Yu J, Grossiord N, Koning CE and Loos J (2007) Controlling the

dispersion of multi-wall carbon nanotubes in aqueous surfactant

solution. Carbon 45(3): 618–623.

Yu MF, Lourie O, Dyer MJ et al. (2000) Strength and breaking

mechanism of multiwalled carbon nanotubes under tensile load.

Science 287(5453): 637–640.

Zare Y (2015) Effects of interphase on tensile strength of polymer/CNT

nanocomposites by Kelly–Tyson theory. Mechanics of Materials

85:1–6.

Zheng LX, O’Connell MJ, Doorn SK et al. (2004) Ultralong single-wall

carbon nanotubes. Nature Materials 3(10): 673–676.

Zhu HW, Xu CL, Wu DH et al. (2002) Direct synthesis of long single-

walled carbon nanotube strands. Science 296(5569): 884–886.

Zou B, Chen SJ, Korayem AH et al. (2015) Effect of ultrasonication

energy on engineering properties of carbon nanotube reinforced

cement pastes. Carbon 85: 212–220.

How can you contribute?

To discuss this paper, please submit up to 500 words to the

editor at journals@ice.org.uk. Your contribution will be

forwarded to the author(s) for a reply and, if considered

appropriate by the editorial board, it will be published as a

discussion in a future issue of the journal.

Magazine of Concrete Research Mechanical properties of

carbon-nanotube-reinforced

cementitious materials: database

and statistical analysis

Ramezani, Kim and Sun

Optimization Design of Mix Proportion for Fly Ash–Silica Fume–Basalt Fiber–Polypropylene Fiber Concrete under Steam Curing Condition

Article

Full-text available

Apr 2024

To enhance the physical and mechanical characteristics of steam-cured concrete, an orthogonal experimental design was utilized to examine the effects of varying contents of fly ash (0 wt%, 10 wt%, 15 wt%, 20 wt%), silica fume (0 wt%, 5 wt%, 10 wt%, 15 wt%), basalt fiber (0 vol%, 0.05 vol%, 0.1 vol%, 0.2 vol%), and polypropylene fiber (0 vol%, 0.05 vol%, 0.1 vol%, 0.2 vol%) on its mechanical properties. Utilizing range and variance analyses, this study identified four preliminary optimized compositions of concrete incorporating fly ash, silica fume, basalt fiber, and polypropylene fiber. On this basis, in order to determine the optimal mix proportion, the mechanical performances, the pore characteristics, and the microstructure of four optimized mix proportions were analyzed. According to the results of macroscopic, fine, and microscopic multi-scale tests, the addition of 15 wt% fly ash, 10 wt% silica ash, 0.2 vol% basalt fiber, and 0.1 vol% polypropylene fiber to the steamed concrete is the best to improve the performance of the steamed concrete. Compared to ordinary concrete, the compressive strength increases by 28%, the tensile strength increases by 40%, and the porosity decreases by 47.2%.

Cementitious composites modified by nanocarbon fillers with cooperation effect possessing excellent self-sensing properties

Article

Full-text available

Mar 2024

The safety and durability of concrete structures are prone to damage and result in significant harm to human society. Hence, monitoring and estimating the states of concrete structures is of great significance to protecting human safety. Graphene nanoplatelets (GNPs) and carbon nanotubes (CNTs) are promising candidates to endow cementitious composites with piezoresistivity for the health monitoring of concrete structures. Therefore, the aim of this article is to explore the effect of the hybrid GNPs and CNTs on behavior of cementitious composites with cooperation effects. The cementitious composites containing the hybrid GNPs and CNTs with simplified manufacturing methods are developed first. And then, the mechanical behaviors, electrical conductivities, and piezoresistive performances of the composites are investigated systematically. Finally, the modification mechanisms are also discussed. The results proved that the electrical resistivity of the composites with the hybrid GNPs and CNT concentration of 5.0 wt% is reduced by three orders of magnitudes, and the FCR and sensitivities are reached at 36.0% and 1.1% MPa⁻¹/177.9, respectively. Its compressive strength/elastic modulus is achieved at 73.3 MPa/16.9 GPa. Therefore, the hybrid GNPs and CNTs modified cementitious composite present great potential application in monitoring and evaluating service states of civil infrastructures.

Electrostatic self-assembly effect of Fe3O4 nanoparticles on performance of carbon nanotubes in cement-based materials

Article

Full-text available

Feb 2024

The beneficial effect of carbon nanotubes (CNTs) to enhance the electrical conductivity and piezoresistivity of cement-based materials was highly contingent upon its dispersion. To achieve an appropriate dispersion of CNTs, ultrasonication, high-speed stirring, and chemical dispersion were commonly used, which raises the risk of structural damage of CNTs caused by the excessive energy. In this study, electrostatic self-assembly of Fe3O4 nanoparticles on CNTs was employed to efficiently disperse CNTs. To optimize the dispersion effect of conductive fillers in cement paste, the mix proportions including sodium dodecyl sulfate (SDS) concentration, CNTs concentration, and Fe3O4/CNTs ratios were adjusted. The dispersion degree and electrical property were evaluated by UV–vis absorption and zeta potential. In addition, the effect of self-assembled conductive filler dosage on the electrically conductive property of cement pastes was examined. The results show that the occurrence of electrostatic self-assembly was proved by the change of zeta potential, and the grape-bunch structure was observed by transmission electron microscopy. Further, the optimal proportions of self-assembled conductive fillers were 0.20 wt% SDS concentration, 0.05 wt% CNTs concentration, and 1:1 Fe3O4/CNTs ratio. The self-assembled conductive filler dosage between 0.02 and 0.10 wt% can effectively improve the electrical conductivity of cement paste with up to 68% reduction of resistivity.

Experimental Studies and Application of Fiber-Reinforced Polymers (FRPs) in Civil Infrastructure Systems: A State-of-the-Art Review

Article

Full-text available

Jan 2024

The application of FRPs in civil infrastructure has increased, particularly in the last 20 years. FRPs have gained importance because of their resistance to harsh environments, high strength-to-weight ratio, and good corrosion resistance, and they are faster and easier to apply than other traditional methods. The rehabilitation of structures is the main area in which FRPs have been developed, because they have allowed for compliance with architectural restraints in historic structures. This review is a compilation of the research conducted on the laboratory and field applications of FRPs, highlighting the different applied methods, installation difficulties, and failure modes of FRPs. Moreover, this review compares studies on the types of fibers such as CFRPs, GFRPs, and AFRPs, and their effects would affect the mechanical properties of civil infrastructure and the durability characteristics of civil infrastructure in challenging environmental conditions. In addition, this review focuses on the modification of the mechanical properties of structural elements using different methods of installing FRPs, including externally bonded reinforcement (EBR), and their main problem: debonding failure before the ultimate load.

Study on the Properties of Multi-Walled Carbon Nanotubes (MWCNTs)/Polypropylene Fiber (PP Fiber) Cement-Based Materials

Article

Full-text available

Dec 2023

In order to improve the mechanical properties and durability of cement-based materials, a certain amount of multi-walled carbon nanotubes (MWCNTs) and polypropylene fiber (PP fiber) were incorporated into cement-based materials. The mechanical properties of the multi-walled carbon nanotubes/polypropylene fiber cement-based materials were evaluated using flexural strength tests, compressive strength tests, and splitting tensile tests. The effects of multi-walled carbon nanotubes and polypropylene fiber on the durability of cement-based materials were studied using drying shrinkage tests and freeze–thaw cycle tests. The effects of the multi-walled carbon nanotubes and polypropylene fibers on the microstructure and pore structure of the cement-based materials were compared and analyzed using scanning electron microscopy and mercury intrusion tests. The results showed that the mechanical properties and durability of cement-based materials can be significantly improved when the content of multi-walled carbon nanotubes is 0.1–0.15%. The compressive strength can be increased by 9.5% and the mass loss rate is reduced by 27.9%. Polypropylene fiber has little effect on the compressive strength of the cement-based materials, but it significantly enhances the toughness of the cement-based materials. When its content is 0.2–0.3%, it has the best effect on improving the mechanical properties and durability of the cement-based materials. The flexural strength is increased by 19.1%, and the dry shrinkage rate and water loss rate are reduced by 14.3% and 16.1%, respectively. The three-dimensional network structure formed by the polypropylene fiber in the composite material plays a role in toughening and cracking resistance, but it has a certain negative impact on the pore structure of the composite material. The incorporation of multi-walled carbon nanotubes can improve the bonding performance of the polypropylene fiber and cement matrix, make up for the internal defects caused by the polypropylene fiber, and reduce the number of harmful holes and multiple harmful holes so that the cement-based composite material not only has a significant increase in toughness but also has a denser internal structure.

Mechanical characterization of high-strength and ultra-high-performance engineered cementitious composites reinforced with polyvinyl alcohol and polyethylene fibers subjected to monotonic and cyclic loading

Article

Apr 2024
CEMENT CONCRETE COMP

Optimizing building insulation with silica nanoparticles: a study on lime-based roof and screed pastes for enhanced compressive strength and thermogravimetric analysis

Article

Mar 2024

Research into leveraging nanomaterials to enhance building material performance has shown significant progress in recent decades. Silica nanoparticles (SiO2), a prevalent chemical element in ceramics, glass, sand, cement, and binders, have been a focus. This study delved into the impact of varying nanosilica (NS) percentages (ranging from 0 to 2.5% of LRP weight) on lime-based pastes for roofs and screeds (LRP), commonly used in building insulation. Specimens underwent demolding after 48 h, with 75% treated at 40, 60, and 80 °C for 7 h and then cured at room temperature for 3, 7, and 28 days. Analysis revealed that SiO2 and CaO in LRP transformed into calcium silicate and calcium carbonate during hydration. The addition of NS notably enhanced compressive strength by facilitating calcium silicate formation and reducing pore structures, with the best results at 40 °C. Thermogravimetric analysis indicated a significant weight loss reduction (2.3 times less) in samples with 2.5% NS content compared to those without. Optimal NS content percentages of 1.5% and 2.5%, replacing LRP at water/binder ratios of 0.6 and 0.75, respectively, were identified based on tensile and compressive strengths and desirable flowability.

A study of the effects of MWCNTs on the fresh and hardened state properties of 3D printable concrete

Article

Jul 2024

Effect of Modified Cow Dung Fibers on Strength and Autogenous Shrinkage of Alkali-Activated Slag Mortar

Article

Full-text available

Oct 2023

Alkali-activated slag (AAS) presents a promising alternative to ordinary Portland cement due to its cost effectiveness, environmental friendliness, and satisfactory durability characteristics. In this paper, cow dung waste was recycled as a renewable natural cellulose fiber, modified with alkali, and then added to AAS mortar. The physico-chemical characteristics of raw and modified cow dung fibers were determined through Fourier transform infrared (FTIR), X-ray diffraction (XRD), and Scanning electron microscope (SEM). Investigations were conducted on the dispersion of cow dung fibers in the AAS matrix, as well as the flowability, strength, and autogenous shrinkage of AAS mortar with varying cow dung fiber contents. The results indicated that modified fiber has higher crystallinity and surface roughness. The ultrasonic method showed superior effectiveness compared to pre-mixing and after-mixing methods. Compared with raw cow dung fibers, modified fibers led to an increase of 11.3% and 36.3% of the 28 d flexural strength and compressive strength of the AAS mortar, respectively. The modified cow dung fibers had a more significant inhibition on autogenous shrinkage, and the addition of 2 wt% cow dung fibers reduced the 7 d autogenous shrinkage of the AAS paste by 52.8% due to the “internal curing effect.” This study provides an alternative value-added recycling option for cow dung fibers as a potential environmentally friendly and sustainable reinforcing raw material for cementitious materials, which can be used to develop low autogenous shrinkage green composites.

Synergic effect of triethanolamine and C-S-H seeding on early hydration of the limestone calcined clay cement in UHPC

Article

Oct 2023
CONSTR BUILD MATER

Influence of Carbon Nanotubes on SCC Flowability

Conference Paper

Full-text available

May 2016

Carbon Nanotubes (CNTs) are considered as one of the most promising nano particles for strengthening concrete structures. Most researches are focused on its contribution to the mechanical properties of concrete, e.g. tensile strength and toughness. Very few studies have been conducted to investigate its influence on the flowability of concrete, especially the relationships among water-to-cement (w/c) ratio, superplasticizer (SP) dosage, and CNTs concentration. This paper studies the effect of Multi Walled Carbon Nanotubes (MWCNTs) addition on the workability of Self Consolidating Concrete (SCC) pastes and mortars. Samples with three w/c ratios of 0.35, 0.45 and 0.6 have been studied. And for each specific w/c ratio, different dosages of superplasticizer were used. The flowability of the SCC pastes and mortars are represented by viscosity, yield stress, and flow diameter obtained from the rheological and flow table tests. The research found that the addition of CNTs in pastes had significant impacts on the pastes’ rheological properties. And the SP/CNTs ratio also had a remarkable influence on the shear thickening behavior of the material. However, being different from the SCC pastes, the SCC mortars’ flowability was less sensitive to the variations of CNTs content and SP/CNTs ratio.

Mechanical Properties and Durability of Ultra High Strength Concrete Incorporating Multi-Walled Carbon Nanotubes

Article

Full-text available

May 2016

In this work, the effect of the addition of multi-walled carbon nanotubes (MWCNTs) on the mechanical properties and durability of ultra high strength concrete (UHSC) is reported. First, the MWCNTs were dispersed by a nano sand-mill in the presence of a surfactant in water. The UHSC specimens were prepared with various amounts of MWCNTs, ranging from 0% to 0.15% by weight of cement (bwoc). Results indicated that use of an optimal percentage of MWCNTs (0.05% bwoc) caused a 4.63% increase in compressive strength and a 24.0% decrease in chloride diffusion coefficient of UHSC at 28 days curing. Moreover, the addition of MWCNTs also improved the flexural strength and deformation ability. Furthermore, a field-emission scanning electron microscopy (FE-SEM) was used to observe the dispersion of MWCNTs in the cement matrix and morphology of the hardened cement paste containing MWCNTs. FE-SEM observation revealed that MWCNTs were well dispersed in the matrix and no agglomerate was found and the reinforcing effect of MWCNTs on UHSC was thought to be pulling out and microcrack bridging of MWCNTs, which transferred the load in tension.

Performance of cementitious materials produced by incorporating surface treated multiwall carbon nanotubes and silica fume

Article

Full-text available

Jul 2016
CONSTR BUILD MATER

The outstanding mechanical properties of carbon nanotubes (CNTs) highlight them as potential candidates for cementitious material reinforcement. However, their low surface friction and the Van der Waals forces of attraction between them, cause the CNTs to aggregate with each other rather than bind with the cement matrix. A number of methods have been investigated by researchers to reduce the aggregation, improve dispersion and activate the graphite surface to enhance its interfacial interaction. These methods involve surface functionalization and coating, optimal physical blending, use of surfactant and other admixtures. This research investigates the use of silica fumes (an admixture), surface functionalized CNTs and cement paste to overcome those obstacles. CNTs with polar impurities end groups OH and COOH were examined. Mortar samples with non-functionalized CNTs dispersed in water solution, another with non-dispersed, non-functionalized CNTs, and a third batch with no CNTs (as control) was used also studied. Silica fumes volume fraction was varied from 0 to 30% to determine its effect. Compressive and flexural strengths of the different mixes were measured and compared. Qualitative analysis using Scanning Electron Microscope (SEM) and Energy-Dispersive Spectroscopy (EDS) were carried out to study the morphology of each mix. Results reveal a much higher enhancement in strength both compressive and flexural strengths for the functionalized CNTs with 30% silica fumes over the other samples.

Fibre Reinforced Cementitious Composites

Book

Nov 2006

Length Effect of Carbon Nanotubes on the Strengthening Mechanisms in Metal Matrix Composites

Article

Sep 2017
ACTA MATER

In the present work, we studied the effect of the aspect ratio of carbon nanotubes (CNTs) on strengthening aluminum metal matrix composites (Al MMCs). To this end, Al samples reinforced with CNTs of various aspect ratios were produced via three different powder metallurgy methods. Microstructural examination revealed that the CNTs were uniformly dispersed in the materials with a range of aspect ratios from 6.5 to 55. The tensile results showed that the CNTs exhibited a strong strengthening effect in the composites regardless of their aspect ratios. However, the post-loading examination and quantitative analysis indicated that there was a strengthening mechanism transition for CNTs, which was closely associated with the aspect ratio or length of CNTs. The origin of such transition was explored from the viewpoint of dislocation-CNTs interaction under loading. The findings may provide a new insight in understanding the strengthening behaviors of CNT-reinforced MMCs.

Fresh and mechanical properties, and strain sensing of nanomodified cement mortars: The effects of MWCNT aspect ratio, density and functionalization

Article

May 2017
CEMENT CONCRETE COMP

A comprehensive analysis on the effect of aspect ratio, bulk density and functionalization of multi walled carbon nanotubes (MWCNTs) in the development of nanomodified mortars, reinforced with different types of MWCNTs is presented herein. A structural characterization of the pristine and functionalized carbon nanotubes was carried out with scanning electron microscopy (SEM), transmission electron microscopy (TEM), and thermogravimetric analysis (TGA). A simple one step dispersion method, involving the application of ultrasonic energy and the use of a superplasticizer (SP) was utilized for the preparation of uniformly dispersed MWCNT suspensions. The experimental determination of the fresh and 28d mechanical properties of mortars with w/c = 0.5 and s/c = 3.0, using four different types of well dispersed pristine and functionalized MWCNTs at an amount of 0.1 wt% of cement took place through: (i) flow and time of setting tests; (ii) three point bending experiments on 4 × 4x16 cm specimens; and (iii) uniaxial compression on the half prisms of the flexural test specimens (4 × 4x8cm). The piezoresistive behavior of the mortars reinforced with the pristine MWCNTs was experimentally determined using the 4-pole method, and compared with the strain sensing ability of the mortars reinforced with the functionalized MWCNTs. All MWCNT reinforced mortars exhibit a remarkable enhancement in the mechanical properties. However, the 28d flexural strength, young's modulus and energy absorption capability of the mortars reinforced with the mechanically functionalized MWCNTs at an amount of 0.1 wt% increased by 120%, 124%, and 103% respectively. Finally, depending on the procedure of the functionalization, chemical or mechanical, a different effect on the intrinsic properties of MWCNTs was observed. The carboxylic groups attached to the surface of the chemically functionalized MWCNTs indeed provided them with the ability of a uniform and effective dispersion, without the need of a sonication procedure. On the other hand, it was found that functionalized MWCNTs do not always retain the electrical properties of pristine MWCNTs.

Fiber Reinforced Cement Composites

Article

Jan 1992

Strength, energy absorption capability and self-sensing properties of multifunctional carbon nanotube reinforced mortars

Article

Sep 2016
CONSTR BUILD MATER

Mechanism of cement/carbon nanotube composites with enhanced mechanical properties achieved by interfacial strengthening

Article

Apr 2016
CONSTR BUILD MATER

Advanced cement based nanocomposites reinforced with MWCNTs and CNFs

Article

Apr 2016

Cementitious materials reinforced with well dispersed multiwall carbon nanotubes (MWCNTs) and carbon nanofibers (CNFs) at the nanoscale were fabricated and tested. The MWCNTs and CNFs were dispersed by the application of ultrasonic energy and the use of a superplasticizer. Mechanical and fracture properties including flexural strength, Young’s modulus, flexural and fracture toughness were measured and compared with similarly processed reference cement based mixes without the nano-reinforcement. The MWCNTs and CNFs reinforced mortars exhibited superior properties demonstrated by a significant improvement in flexural strength (106%), Young’s modulus (95%), flexural toughness (105%), effective crack length (30%) and fracture toughness (120%).

Mechanical Properties of Carbon Nanotube Reinforced Cementitious Materials: Database and Statistical Analysis

Abstract and Figures

Recommended publications

Production and mechanical properties of aligned multi-walled carbon nanotubes-M140 composites

Modeling the mechanical properties of cementitious materials containing CNTs

Design and predicting performance of carbon nanotube reinforced cementitious materials mechanical pr...

Design and predicting performance of carbon nanotube reinforced cementitious materials: mechanical p...

Probabilistic Model for Flexural Strength of Carbon Nanotube Reinforced Cement-based Materials