Nanofibrillated cellulose as nanoreinforcement in Portland cement: Thermal,mechanical and microstructural properties

Abstract

Nanofibrillated cellulose from eucalyptus pulp, produced by high-pressure homogenization, was used as cement partial replacement for cement paste at a content ranging from 0% to 0.5% by weight of cement. The effect of the content of nanofibrillated cellulose on porosity, thermal properties, compressive strength and degree of cement hydration was investigated. Results have shown an improvement in the compressive strength by more than 50% with 0.3 wt% of added nanofibrillated cellulose. The porosity was reduced by nanofibrillated cellulose addition, and the greatest result was achieved with mixture incorporating 0.3 wt% nanofibrillated cellulose. The coefficient of thermal expansion and the thermal conductivity measurements, relative to nanofibrillated cellulose-reinforced cement pastes, have pointed out the reinforcement effectiveness of nanofibrillated cellulose. The degree of cement hydration has increased with nanofibrillated cellulose content. This trend was confirmed by X-ray diffraction and Fourier Transform Infrared spectroscopy. These analyses have revealed that the presence of nanofibrillated cellulose promoted the hydration of cement, by producing more portlandite and calcium silicate gel, which is likely the main reason accounting for the strong enhancement in the compressive strength.

Keywords

Nanofibrillated cellulose Portland cement nanocomposite degree of cement hydration thermal properties mechanical properties microstructural properties

Introduction

Because of their unique mechanical properties, nanofibers, especially carbon nanotube (CNT) and carbon nanofibers, have recently been promising materials for cementitious materials.^1–3

Cellulose nanofibrils, also known as nanofibrillated cellulose (NFC), are composed of fibrils with width in the nanoscale and length in the micrometric range that form a network structure. This advanced biomaterial, made mainly from renewable forest and agricultural resources, has demonstrated exceptional performance in composites and nanocomposites.⁴

Cellulose nanofibers have several unique characteristics, such as their unique morphology, low density, very large surface to volume ratio, high surface area, good mechanical properties including a high Young's modulus, high tensile strength⁵ and low coefficient of thermal expansion (CTE).⁶ Taking advantage of these outstanding properties, there is an increasing interest in the development of cellulose nanofiber-based composites, including recently cement materials.^7–9

The most important potential application of nanocellulose is its incorporation as a reinforcing agent in polymer-based composite materials.⁷ It is now well accepted that the enhanced mechanical properties in polymer matrices, that can be displayed by these nanocomposites, are ascribed to the formation of a continuous network formed by the cellulosic nanoparticles.⁴ The formation of this rigid network, resulting from strong interactions between the nanoparticles, is assumed to be governed by a mechanical percolation mechanism. At a critical filler content which depends on the aspect ratio of the reinforcing phase, a huge improvement of the mechanical properties of the nanocomposite material was observed. The potential of these applications has been discussed in numerous publications.^10–12 For instance, Özgür and Oksman¹⁰ showed that the addition of 16.5 wt% of cellulose nanofibrils to polyurethane increased the strength nearly by 500% and the stiffness by 3000%. Shimazaki et al.¹¹ reported that reinforcing the conventional epoxy resin with cellulose nanofibers increased the thermal conductivity by 300%–500%, reduced three times the CTE value and improved the storage modulus of elasticity by 140%. Nakagaito and Yano¹² found that phenolic resin composites reinforced with microfibrillated cellulose exhibited a linear increase in young's modulus with nanofiber contents up to 40 wt%, and a rapid decrease of the thermal expansion under 60 wt% nanofiber content.

Given this great potential as reinforcement in polymer matrices, nanocellulose is expected to provide an improvement in the mechanical performance of cement-based composites. Recently, few studies have investigated the effect of the addition of nanocellulose as reinforcing materials in a cementitious matrix.^7–9 Onuaguluchi et al.⁷ found that the addition of 0.1 wt% of cellulose nanofibers improved the flexural strength and the absorption energy of the cement paste by approximately 106% and 184%, respectively. An increase in the cumulative heat of hydration and the degree of cement hydration of nanofiber reinforced mixtures was also reported, in comparison with the unreinforced paste.⁷ Recently, Cao et al.⁸ have shown that the addition of cellulose nanocrystals (CNCs) modified the performance of cement paste, more specifically a strong enhancement in the flexural strength of cement. It is hypothesized that this increase can be attributed to the increase of the degree of cement hydration of the cement pastes when CNCs are used. Two mechanisms are proposed to explain the increase of the degree of cement hydration. The first mechanism is the steric stabilization which is responsible for dispersing the cement particles. The second mechanism is that CNCs provide a channel for water transporting through the hydration products ring (i.e., high-density CSH) to the unhydrated cement particles and thereby improving hydration. Ardanuy et al.⁹ demonstrated that the cement mortar composites reinforced with NFC exhibited an improvement in the flexural modulus and strength compared with those reinforced with the conventional cellulose fibers.

In this study, the effect of the addition of NFCs up to 0.5 wt% on thermal, mechanical and microstructural properties of the cement paste nanocomposites was investigated. The nanocomposites were characterized using thermal conductivity instrument (TCI), thermal mechanical analysis (TMA), X-ray diffraction (XRD) and scanning electron microscopy (SEM).

Materials and test methods

In this research, the used cementis CEM I 32.5 N, produced in Bizert in a Tunisian plant, conforming to the Tunisian standard NT 47-01:1983¹³ and the European standard EN 197-1:2000.¹⁴

The chemical composition and the physical characteristics of this cement are shown in Table 1.

Table 1.

Chemical composition and physical characteristics of Portland cement (wt%).

Material	SiO₂	Al₂O₃	Fe₂O₃	CaO	MgO	SO₃	Na₂O	K₂O	Loss on ignition
Cement	19.4	4.8	3.6	63.7	1.9	2.7	0.2	0.8	2.4

Specific gravity: 3.26 g/cm³.

Specific surface area (blaine): 3870 cm²/g.

Commercial bleached eucalyptus pulp (Eucalyptus globulus) was used as starting material for the preparation of NFC. Prior to the homogenization process, the fibers were first submitted to a TEMPO-mediated oxidation pre-treatment to bring the carboxyl content up to 500 µmol/g and to facilitate the defibrillation process. The oxidation pretreatment aims to generate carboxylic groups whose ionization facilitates the fibrillation and the break-down of the cell wall of the fibers. Details of the TEMPO ((2,2,6,6-tetramrtylpiperidin-1-yl)oxyl) mediated oxidation were reported in our previous work.¹⁵

The fiber suspension was dispersed in water at a fiber content of 1.5 wt% and then homogenized at different passes through a high-pressure homogenizer (NS1001L PANDA 2K-GEA) operating at pressure ranging from 100 to 1000 bar and at a temperature operating between 60 and 70℃ until obtaining a translucent gel. The homogenization was conducted in two steps. First, the fiber suspension at a concentration of 1.5 wt% was passed several times through thin slits at a pressure of 300 bar (4350 Psi) until the suspension turned to a gel. Then, the fibrillation was pursued by further passes at a pressure of 600 bar (8700 psi).¹⁵

Operation conditions were set at 600 bar pressure and 60℃–70℃. This operation was carried on five to six times until a transparent gel-like product was produced (Figure 1(a)). Based on FE–SEM observation (Figure 1(b)), the NFC consisted in nano-sized fibrils with a width within the range of 5–10 nm and length within the micron scale. The distribution of the width is fairly narrow without any trace of visible fragment of fibers.

A water-to-cement (w/c) ratio of 0.26 was applied for all mixtures according to the Tunisian standard NT 47.05.¹⁶ The cement pastes were mixed in a specified mixer according to NT 47.07.¹⁷

Six paste mixtures were prepared, incorporating NFC at 0%, 0.01%, 0.05%, 0.1%, 0.2%, 0.3% and 0.5% by weight of cement, and are identified as reference, NFC0, NFC0.01, NFC0.05, NFC0.1, NFC0.2, NFC0.3 and NFC0.5, respectively.

To ensure the NFC dispersion, the NFC gel was pre-mixed with the mixing water for 5 min using a hand mixer, and then the nanofiber-mixing water suspension was agitated for 10 min in an ultrasonic bath.

After the mixing procedure, cubic specimens with 40 mm edge length were prepared for each mixture. After 24 h, the samples were demolded and cured at a 95% relative humidity and at 23.0℃ ± 2.0℃ prior to test days.

The compressive strength tests were carried out after 1, 7 and 28 days of curing according to the NT 47-30 standard,¹⁸ using a universal testing machine. The compressive strength results are the average of the three test values.

The thermal conductivity of the control mixture and the nanocomposites was measured after 28 days of curing, using a TCI. Specimens 100 × 100 × 20 mm cured for 28 days were used for this measurement.

The degree of cement hydration was estimated based on the amount of the non-evaporable water in the cement pastes. Cement paste powders obtained from crushed test prisms were treated with acetone to stop hydration before they were dried to constant mass at 140℃. Thereafter, samples were heated and kept at 1100℃ for 3 h. Samples were then cooled to room temperature in a desiccator. The non-evaporable water content of samples was calculated as the mass loss between 140℃ and 1100℃, normalized by the ignited mass.^7,8

The porosity of pastes was measured using the methanol displacement method.¹⁹ High purity methanol, which can penetrate into the microstructure without damage, was used.^20,21

After 28 days of curing, specimens were crushed into small particles. The crushed particles were then vacuum-dried until attaining a constant weight W₀. After that, the dried particles were immersed in methanol for 24 h, and the weight of sample dipped in the methanol was noted W₁. Afterward, particles were brought out of the methanol and were surface dried, and the measured weight was noted W₂. The porosity of cement pastes was calculated as

P = (W_{2} - W_{0}) / (W_{2} - W_{1})

(1)

The thermal expansion of the NFC0, NFC0.1 and NFC0.3 mixtures, cured for 28 days, was measured using TMA Setaram model Setsys Evolution. The test was carried out between 30℃ and 400℃, at a heating speed of 5℃/min, and under nitrogen atmosphere. A compression load of 1 g was applied to ensure good contact between the specimen and probe surfaces.

The XRD analysis was performed in a Philips PW 3050/60 θ/2θ goniometer and a PW 3373/00 copper cathode, at a scanning speed of 2°/min.

The microstructure and morphology of the prepared nanocomposites were analyzed using a Zeiss DSM 960A scanning electron microscope, with voltage that varies between 0.49 and 30 KV, and microscope resolution of 25 to 4 nm.

A Weiss SEM was used to obtain images by capturing secondary electrons emitted from the surface of a NFC sample, prepared from a drop of the NFC suspension (with a solid content about 0.05 wt%) deposited on the surface of a silicon wafer and coated with a thin carbon layer, applied by ion sputtering with a thickness limited to around 1 nm. To ensure a good image resolution without any damage to the samples during the analysis, the acceleration voltage was maintained at a relatively low range (1–3 kV).

Results and discussion

Setting time

Figure 2 shows that both the initial and the final setting times have decreased as NFC contents mixtures decrease, with the NFC0.5 mixture achieving the fastest setting times. The increase of the initial and final setting times of NFC-cement nanocomposites ranged from 6.1% to 38% and from 1.8% to 23.38%, respectively. Accordingly, NFC can act as a set accelerator inducing a faster hydration process.

Degree of hydration

The degree of cement hydration was measured based on the Pane and Hansen's method.^8,19 This method states that the weight loss between 140℃ and 1100℃ is considered as the amount of the chemically bound water (CBW) in the hardened cement pastes. The thermogravimetric analysis profile of the unhydrated cement indicated some mass loss within a temperature range of 600℃–780℃ associated to the decomposition of calcium carbonate. This mass loss is about 1% of the initial weight. Such a correction can be done by subtracting the mass loss within a temperature range of 600℃–780℃ from the total measured CBW.²² The degree of cement hydration is calculated by dividing the amount of CBW by the final weight of the material to obtain the mass of CBW per unit gram of unhydrated cement. According to the analysis by Copeland et al.,²³ it was postulated that the CBW is 0.23 g per unit gram of cement when fully hydrated. This value is typical for type I Portland cement. Thus, the degree of cement hydration can be easily obtained by dividing the mass of CBW per unit gram of unhydrated cement with 0.23 g. The degree of cement hydration at different curing ages, as function of NFC content, is presented in Figure 3. For all curing ages, the degree of cement hydration increases as the NFC content of cement pastes increases. The greatest degree of cement hydration value is achieved with the mixture containing 0.5 wt% NFC (NFC0.5 mixture). At this addition level, the degree of cement hydration is increased by 102%, 23% and 28%, relative to the control mixture, after 1, 7 and 28 days of hydration, respectively. Similar trend is observed with cement pasts incorporating CNCs. However, the degree of cement hydration increase is limited to 8.3% and 10% after 7 and 28 days, respectively, with the mixture containing 0.5 wt% CNCs.⁸

It can be observed that the addition of NFC to cement paste has accelerated the early hydration. In fact, at one day of curing the percentage increase of the degree of cement hydration (DOH) is about 1.6%, 19.3%, 32.6%, 66.%, 87.3% and 102% for the mixtures NFC0.01, NFC0.05, NFC0.1, NFC0.2, NFC0.3 and NFC0.5, respectively. The improvement of the degree of cement hydration was expected as NFCs can have the same steric stabilization effect of water reducing admixtures, which supply a uniform distribution of cement particles during the hydration process.^8,24 Also, the NFCs can act as a nuclei to promote the formation of the hydrated products, and consequently, to improve the early degree of cement hydration.

Mineralogical analysis of the hydration products by X-Ray diffraction

To investigate how the addition of NFCs is likely to affect the phase composition of cement, XRD of cement with different contents of NFCs was performed. The corresponding diffraction patterns, at an early age of curing (one curing day), are shown in Figure 4.

The XRD patterns of the three samples showed the expected hydration products, including portlandite, ettringite and unreacted calcium silicate phases (C₃S and C₂S). However, as the calcium silicate hydrates (C–S–H gel) are poorly crystallized, the corresponding diffraction peaks cannot be clearly identified in the patterns. Therefore, the evolution of C–S–H gel is evaluated referring to the unreacted anhydrous cement phases.

The intensity of calcium silicate main peaks decreases with NFC addition, as a greater amount of anhydrous cement phases reacts in the presence of NFCs (C₃S and C₂S are transformed into C–S–H). Besides, higher amounts of portlandite and ettringite are formed in NFC0.1 and NFC0.3 samples, compared to the control paste NFC0. It can be observed that the intensity of peak changes continues to increase as a function of NFC amount.

Compared to the neat cement sample, it can be seen that the presence of NFC promotes the early hydration of cement by producing more Portlandite, Ettringite and C–S–H gel. As C–S–H is one of the major hydration products and the main binding phases in Portland cement controlling cement mechanical properties, the higher content in C–S–H phase is likely the main reason accounting for the strong enhancement in the compressive strength of the cement matrix.

Hydration products analysis by Fourier transform infrared spectroscopy

Fourier transform infrared (FTIR) spectra of the control cement paste and cement nanocomposites containing different NFCs contents, hydrated after one curing day, are shown in Figure 5. The results show a small peak at about 3650 cm⁻¹ associated with O–H stretching vibrations of portlandite (Ca(OH)₂),²⁵ which seems to be intensified with increasing the NFC content.

The broadbands observed at 3440 and 1638 cm⁻¹ are attributed to the stretching and bending vibrations of water lattice in calcium silicate, calcium aluminate and calcium aluminosilicate hydrates, and the characteristic band of calcium silicate gel appeared at about 984 cm⁻¹.²⁶ Their intensities increase with NFC amount.

The band appearing in the range between 1112 and 1129 is due to ${SO}_{4}^{2 -}$ associating with ettringite formation.²⁵ Its intensity increases with NFC addition.

The results of FTIR spectra have shown one more time that the NFC addition has promoted the early age hydration, as higher amount of hydrated products (portlandite, ettrigite and calcium silicate gel) are formed at an early age of curing, and the degree of the hydration improvement increases with the % of NFC addition.

Thermal conductivity

Figure 6 shows the measured thermal conductivities of the paste mixtures as function of NFC content. As expected, the NFC addition clearly shows a systematic trend of increasing the thermal conductivity. The maximum of thermal conductivity was achieved by adding 0.3 wt% NFC (NFC0.3 mixture). These results demonstrate that the use of NFC as reinforcement enhances the thermal conductivity of the cement nanocomposites. In fact, a small amount of NFC, as low as 0.3 wt% FNC, was sufficient to increase the thermal conductivity by 20%. Similar observations were reported by Shimazaki et al.¹¹ for nanocomposites based on NFC and a polymer matrix. These authors have pointed an increase in the thermal conductivity of the nanocomposite by more than five times, with respect to the neat matrix, when NFCs were incorporated in the polymer at a content of 58 wt%.

The behavior of the thermal conductivity can be related to the contribution of NFC addition to the densification of cement matrix, since its expected effect is to decrease the porosity and to increase the bulk density. As a result, the solid phase becomes more continuous,¹⁶ and the phonon-transporting in the cementitious matrix is improved. In fact, as shown in Figure 7, the bulk densities of NFC-cement nanocomposites evolve linearly with the NFC content, from 1.776 g/cm³ of the control cement paste, towards 1.817 g/cm³ of NFC0.3 mixture. Besides, Figure 3 shows that the NFC addition has clearly reduced the porosity of samples. The lower porosity was achieved with the mixture containing 0.3 wt% FNC (NFC0.3 mixture). Compared to the control sample, the porosity of NFC.3 mixture is reduced by 36%.

The changes in bulk density and porosity as NFC is added can be directly attributed to the increase in the degree of cement hydration. In fact, the NFCs behave not only as a filler to improve the microstructure but also as an activator to promote hydration reaction. They also act as nucleating sites to form more accumulation and precipitation of hydrated products in the open pores originally filled with water, leading to the formation of more homogeneous, dense and compact microstructure than the mixture without NFC addition.

Nevertheless, a 2% increase in the density seems not able to induce a 20% increase in the thermal conductivity. However, it was found that in this case (NFC0.3 mixture), the porosity was reduced by 36%, which can justify the thermal conductivity change. In Yuzuru et al.²⁷ research, it was reported that the thermal conductivity of NFC/epoxy resin nanocomposite was three to five times higher than that of the conventional resins. This increase was attributed to the crystalline nature of the cellulose nanofibers that provided excellent phonon pathways through the nanocomposite. It is also known^28–36 that not only the porosity but also the pore's characteristics, such as the size, shape, distribution, orientation, structure and the emissivity of the pores surfaces, have some effects on thermal conductivity. It was found that,^37,38 with equal density, a material with spherical air holes has twice the thermal conductivity of a material in which the solid constituents consist of spherical particles. This is because the number of contact points of the solids influences the conductivity.

However, beyond 0.3 wt% FNC addition level (NFC0.5 mixture), NFC affected negatively the porosity and density enhancement of the nanocomposites, as result of fiber agglomeration.

In Figure 8, the latter parameters are correlated with the measured values of the thermal conductivity to examine the relation that each makes with the thermal conductivity of the studied nanocomposites. The results agree with the general trend that the thermal conductivity of a material increases as density increases and porosity decreases.³⁸ This trend can be explained as following: Since the gaps in the solid matrix are occupied by non-reacted water (λ_water = 0.6 W/mK) and air (λ_air = 0.025 W/mK), they have lower thermal conductivities compared to the cementitious materials. Consequently, the higher is the porosity, the higher is moisture and air entrapped within the material and the lower is the thermal conductivity.

Thermal expansion

In Figure 9, the calculated CTE of control, NFC0.1 and NFC0.3 mixtures are plotted. The thermal expansion behaviors of cement nanocomposites and the control specimen are presented in Figure 10. As it is shown, the thermal expansion behaviors and the CET increase with increasing the NFC content of mixtures. The CTEs were increased by about 0.8% and 7% in presence of 0.1 and 0.3 wt% NFC, respectively.

Based on the general conclusion of Shui et al.³⁹ that “… more porous materials have lower CTE, because the void could accommodate the internal thermal expansion of material,” it is expected that the increase of the CTE is attributed to the potential of NFCs to decrease the porosity. This was proved in the previous section.

Strength analysis of cement composites

The influence of the NFCs as reinforcement on the mechanical performance of the cement past composites, at different curing ages, is shown in Figure 11. Irrespective of the curing time, it can be seen that incorporation of NFCs leads to a steady enhancement in the compression strength up to a content of 0.3 wt%. For example, the compression strength grows from 30 MPa for the neat reference cement, to about 35 and 43 MPa in presence of 0.1 and 0.3 wt% of NFC, respectively, which corresponds to about 16% and 43% enhancement. However, at 0.5 wt% NFC, the strength dramatically falls down below that of the neat cement.

Four possible reasons have been proposed for the huge improvement in the mechanical properties with the addition of NFCs within cement.^7–9 First, as a result of their hydrophilic character, the hydration of cement is accelerated, and larger volumes of hydration products are formed. Second, the high-specific surface area of the NFC improves the nanofiber–matrix interface, which assures the improvement of the stress transfer between the matrix and the nanofibrils. Third, the high surface area to volume ratio of NFC promoted the nucleation of cement phases namely CSH, providing excellent chemical reactivity to promote cement hydration. Finally, the application of the TEMPO-mediated oxidation method, as a chemical pre-treatment for modifying the surface of native cellulose, also contributed to the strength enhancement: nanofibrillated cellulose (NFC) displays some drawbacks, which are associated with its intrinsic physical properties. One of them, the high number of hydroxyl groups, which leads to strong hydrogen interactions between two nanofibrils. The chemical surface modification of NFC is the most feasible solution to reduce the interactions of hydroxyl groups. TEMPO-mediated oxidation is the more commonly used chemical pre-treatment that modifies selectively the surface of the native cellulose. By using this method, the C6 primary hydroxyl groups of cellulose are selectively converted to carboxylate groups. Accordingly, the repulsive forces of the ionized carboxylate groups over whelm the hydrogen bonds, holding the nanofibrils together.⁴⁰ This ameliorates the NFCs dispersion within the cement matrix. The carboxylic groups on the surface of NFC react with the calcium silicate hydrate and result in a strong embedding strength between the nanofiber and matrix.⁴¹ This reaction can provide higher stress when the crack goes straight toward the NFC reinforced zone area and prevent propagation of cracks at high speed.^42,43 Also, when the nanofibers are chemically reacted with hydration products, they can behave as nucleation sites and form interfacial mechanical bonds.^44–46

Over 0.3 wt% NFC, a possible reason for the drop in the strength might be due to the aggregation of NFC, leading to an inhomogeneous dispersion of the nanofibrils within the cement matrix. This agglomeration is attributed to the localization of a high content of NFCs and the formation of networks. Such agglomerated NFCs create weak zones in the form of pores. After curing process, these aggregated particles will act as stress concentrators in the cementitious matrix and promoted premature cracking. In fact, the SEM image of the sample containing 0.5 wt% NFC shows the presence of cracks (Figure 12). Previous studies have shown that excessive nanofiber content does not improve strength, and sometimes even deteriorates the material properties dramatically. In the study of Onuaguluchi et al.,⁷ it was found that beyond 0.1% cellulose nanofiber addition level, fiber agglomeration affected strength enhancement negatively, especially at 0.4% (at this addition level, the strength decreased even below that of the control sample). Also, Cao et al.⁸ have shown that the strength reaches a peak at around 0.2% of CNC, and then decreases.

In order to confirm the hypothesis that the cement strengthening can be mainly attributed to the NFC potential to improve the degree of cement hydration, the compressive strengths against the degrees of cement hydration are plotted in Figure 13. Figure 13 shows the relationship between the compressive strength and the degree of cement hydration at the ages of 1, 7 and 28 days. As it can be seen, the compressive strength increases almost linearly with increasing the degree of cement hydration. This trend is not followed by the last point of each age series, that correspond to mixtures containing 0.5 wt% NFC, cured for 1, 7 and 28 days. As previously observed in Figure 10, NFC-cement nanocomposites, with such addition level of NFC, exhibited a drawback in strength caused by NFC agglomeration.

Composite morphology

Figure 14 shows SEM micrographs of NFC0 (a) and NFC0.3 (b) specimens cured for seven days.

The addition of low-volume fractions (0.3 wt%) of NFCs into the cementitious matrix leads to a denser matrix with lower porosity. These observations confirm the acceleration of the hydration process at an early age of curing (seven days), and the strengthening of the cementitious composite as a result of NFC addition.

Conclusion

In this work, the NFC potential, as nanoreinforcement for a cementitious matrix, was investigated. Due to its hydrophilic potential, high reactivity and high-specific surface area, the NFC addition has shown an improvement in the thermal, mechanical and microstructural properties of the new NFC–Portland cement (PC) nanocomposite.

The experimental results have shown that the incorporation of NFC has greatly enhanced the compressive strength. The highest strength property was observed by adding 0.3 wt% of NFC. However, beyond this level of addition, a problem of strength deficiency was confronted.

NFC addition has increased both the thermal conductivity and the CTE of cement pastes, which is mainly attributed to the NFC potential to reduce the porosity and to improve the microstructure of cement matrix.

It is also found that, in the presence of NFC, more hydration products were formed at an early age of curing, inducing the composite strengthening and confirming the NFC role as hydration's nuclei.

Footnotes

Acknowledgements

The authors would like to express gratitude for the help of Pere Bellvehi Casadellà (Construction Engineering Department, Girona University), Hichem Eloussifi (Laboratory of Research in Materials and Thermodynamic, Girona University), Joan Lopez (Laboratory of Research in Materials and Thermodynamic, Girona University) and Yassine Slama (LEPOLE secondary school).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Tyson

Abu Al-Rub

Yazdanbakhsh

. Carbon nanotubes and carbon nanofibers for enhancing the mechanical properties of nanocomposite cementitious materials. J Mater Civil Eng 2011; 23: 1028–1035.

Nur

Vinoth

. Carbon nano-tube and nano-fiber in cement mortar: effect of dosage rate and water-cement ratio. IJMSCI 2014; 4: 45–52.

Zoi

Maria

Surendra

. Carbon nanofiber-reinforced cement-based materials. J Transport Res Board 2010; 2142: 114–118.

Boufi

Kaddami

Dufresne

. Mechanical performance and transparency of nanocellulose reinforced polymer nanocomposites. Macromol Mater Eng 2014; 299: 560–568.

Takagi

Asano

. Characterization of “green” composites reinforced by cellulose nanofibers. Key Eng Mat 2007; 334–335: 389–392.

Nishino

Matsuda

Hirao

. All-cellulose composite. Macromolecules 2004; 37: 7683–7687.

Onuaguluchi

Panesar

Sain

. Properties of nanofibre reinforced cement composites. Constr Build Mater 2014; 63: 119–124.

Cao

Zavaterri

Youngblood

. The influence of cellulose nanocrystal additions on the performance of cement paste. Cement Concrete Compo 2015; 56: 73–83.

Ardanuy

Claramunt

Arévalo

. Nanofibrillated cellulose as a potential reinforcement for high performance cement mortar. Composites 2012; 7: 3883–3894.

10.

Özgür Seydibeyoğlu

Oksman

. Novel nanocomposites based on polyurethane and micro fibrillated cellulose. Compos Sci Technol 2008; 68: 908–914.

11.

Shimazaki

Miyazaki

Takezawa

. Excellent thermal conductivity of transparent cellulose nanofiber/epoxy resin nanocomposites. Biomacromolecules 2007; 8: 2976–2978.

12.

Antonio

Hiroyuki

. The effect of fiber content on the mechanical and thermal expansion properties of biocomposites based on microfibrillated cellulose. Cellulose 2008; 15: 555–559.

13.

NT 47-01:1983. Tunisian standard specification of ordinary Portland.

14.

EN 197-1:2000. European standard specification of ordinary Portland.

15.

Besbes

Alila

Boufi

. Nanofibrillated cellulose from TEMPO-oxidized eucalyptus fibres: effect of the carboxyl content. Carbohyd Polym 2011; 84: 975–983.

16.

NT 47.05:1983. Liants hydroliques – Technique des essais: Pate Normale.

17.

NT 47.07:1983. Liants hydroliques – Technique des essais: Malaxeur.

18.

NT 47.30:1991. Méthodes d'essais des ciments – Détermination des résistances mécaniques.

19.

Mingxun Y and Hooton RD. Influence of slag, fly ash and silica fume on porosity of hardened cement paste. J Chin Ceram Soc 1990; 18: 179–183.

20.

Guimer T. Contribution de l'addition minérale sur les caractéristiques physico-mécaniques des liants et mortiers. Master thesis, University ofMohammed Kheider, Biskra, 2007.

21.

Mellas

Mezghiche

Ash

. Estimation of the porosity of Portland cement pastes using back scattered electron image. Courrier du Savoir 2003; 4: 47–51.

22.

Pane

Hansen

. Investigation of blended cement hydration by isothermal calorimetry and thermal analysis. Cement Concrete Res 2005; 35: 1155–1164.

23.

Copeland LE, Kantro DL and Verbeck GJ. Chemistry of hydration of Portland cement. In: Proceedings of the symposium on the chemistry of cement, cement and concrete association, Portland Cement Association, Washington, DC, 1960, pp. 429–468.

24.

Nawa

. Effect of chemical structure on steric stabilization of polycarboxylate-based superplasticizer. J Adv Concrete Tech 2006; 4: 225–232.

25.

Kontoleontos

Tsakiridis

Marinos

. Dry-grinded ultrafine cements hydration. physicochemical and microstructural characterization. Mater Res 2013; 16: 404–416.

26.

Delgado

Paroli

Beaudoin

. Comparison of IR techniques for the characterization of construction cement minerals and hydrated products. Appl Spectrosc 1996; 50: 970–976.

27.

Yuzuru

Yasuo

Yoshitaka

. Excellent thermal conductivity of transparent cellulose nanofiber/epoxy resin nanocomposites. Biomacromolecules 2007; 8: 2976–2978.

28.

Euchen A. Allgemeine Gesetzmäßigkeiten für dasWärmeleitvermögen verschiedener Stoffarten und Aggregatzustände. Forsch Gebiete Ingenieur-W 1940; 11: 353.

29.

Russell

. Principles of heat flow in porous insulators. J Am Ceram Soc 1935; 1: 1–5.

30.

Andreas Ö, Graeme EM and Marcelo JS (Eds). Cellular and Porous Materials: Thermal Properties Simulation and Prediction. Weinheim, 2008, p. 89.

31.

Waddanis AL. Transactions and Communications. J Soc ChemInd 1944; 63: 357.

32.

Francl J and Kingery WD. Fundamental study of clay: xiii, drying behavior and plastic properties. J Am Ceram Soc 1954; 37: 596–602.

33.

Loeb

. A theory of thermal conductivity of porous materials. J Am Ceram Soc 1954; 37: 96–99.

34.

Krischer O and Esdorn H. Heat transfer in damp porous substances of different structures. Forsch Gebiete Ingenieur-W 1956; 22: 1–8.

35.

Toshio N, Qian G and Christopher CB. Effects of pores on mechanical properties of plasma-sprayed ceramic coatings. J Am Ceram Soc 2000; 83: 578–84.

36.

Sugawara

Yoshizawa

. An investigation on the thermal conductivity of porous materials and its application to porous rock. Aust J Phys 1961; 14: 469–480.

37.

Gunst EV and Zuilen DV. The influence of moisture content on the thermal conductivity of building materials. BuildResCongr 1951; Division 3: 94–98.

38.

Ganjian

. The relationship between porosity and thermal conductivity of concrete 2012PhD Thesis, University of Leeds, Leeds.

39.

Shui

Z-H

Zhang

Chen

. Effects of mineral admixtures on the thermal expansion properties of hardened cement paste. Constr Build Mater 2010; 24: 1761–1767.

40.

Eichhorn

Dufresne

Aranguren . Review: current international research into cellulose nanofibres and nanocomposites. J Mater Sci 2010; 45: 1–33.

41.

Wang

Zhao

. Mechanical behavior and microstructure of cement composites incorporating surface-treated multi-walled carbon nanotubes. Carbon 2005; 43: 1239–1245.

42.

Sobolkina

Mechtcherine

Khavrus

. Dispersion of carbon nanotubes and its influence on the mechanical properties of the cement matrix. Cem Concr Compos 2012; 34: 1104–1113.

43.

W-W

W-M

Wang

Y-C

. Investigation on the mechanical properties of a cement-based material containing carbon nanotube under drying and freeze-thaw conditions. Materials 2015; 8: 8780–8792.

44.

Larisa

Ilya

Albert

. Effect of carbon nanotube aqueous dispersion quality on mechanical properties of cement composite. J Nanomater 2012. 2012: Article ID 169262.

45.

Wang

Zhao

. Mechanical behavior and microstructure of cement composites incorporating surface treated multi-walled carbon nanotubes. Carbon 2005; 43: 1239–1245.

46.

Missoum

Belgacem

Bras

. Nanofibrillated cellulose surface modification: a review. Materials 2013; 6: 1745–1766.