Kounadis A.N., Gdoutos E.E. (Eds.) Recent Advances in Mechanics

Подождите немного. Документ загружается.

Advanced Cement Based Nanocomposites 319

Fig. 4. Dispersant concentration effect on CNT dispersion: (a)-(c) represent a dispersant to

CNTs weight ratio of 0, 1.5, and 6.25 respectively

As expected, in samples where dispersion was achieved without the use of sur-

factant (Fig. 4(a)), CNTs appear poorly dispersed in cement paste, forming large

agglomerates and bundles. In the case where dispersion was achieved with a sur-

factant to CNT weight ratio of 1.5 (Fig. 4(b)), it is observed that CNTs mainly

remain as large agglomerates entangled in the cement paste hydration products

and only a small amount of CNTs was dispersed. Only individual CNTs were

identified on the fracture surfaces of samples where the surfactant to CNT weight

ratio lies within the range from 4.0 to 6.25 (Fig. 4(c)).

The mechanical performance of the CNTs nanocomposites was evaluated by

fracture mechanics tests. Notched specimens of 20×20×80 mm were tested at

the age of 3, 7 and 28 days, by three-point bending. The tests were performed

with a closed-loop testing machine with a 89 kN capacity. The feedback control

signal for running the test was the crack mouth opening displacement (CMOD)

at the notch, which was advanced at a rate of 0.12 mm/min. The load and the

CMOD were recorded during the test. A typical load-CMOD curve is shown in

Fig. 5.

28d CP w/c=0.3

100

150

200

250

300

350

400

450

0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05

CMOD (mm)

Load (N)

Plain Cement Paste (CP)

CP+CNTs (0.08wt%)

Fig. 5. CMOD-Fracture load of 28 day w/c=0.5 cement paste reinforced with CNTs

(0.08wt% by weight of cement)

(c)

(b)

(a)

1 μm

320 S.P. Shah, M.S. Konsta-Gdoutos, and Z.S. Metax

The results of the 28-day maximum load for the cementitious nanocomposites

with different surfactant to CNTs weight ratio are plotted in Fig. 6. It is observed

that samples treated with different amounts of surfactant exhibit higher fracture

load than plain cement paste. Samples where dispersion was achieved without the

use of surfactant exhibit lower fracture load. This decrease in strength indicates

that CNTs are poorly dispersed in cement matrix and that in order to achieve proper

dispersion the use of a surfactant is absolutely required. The samples with surfac-

tant to CNT weight ratio of 4.0 give a higher average fracture load increase at all

ages. Specimens where dispersion was achieved at surfactant to CNTs weight

ratios either lower or higher than 4.0 exhibit lower fracture load. A possible expla-

nation could be that at lower surfactant to CNT weight ratios, less surfactant mole-

cules are absorbed to the carbon surface and the protection from agglomeration is

reduced. At higher surfactant to CNT weight ratios, bridging flocculation can oc-

cur between the surfactant molecules. Too large amount of surfactant in the aque-

ous solution is causing the reduction of the electrostatic repulsion forces between

the CNTs. Based on those results and the SEM imaging of the fractured surfaces it

was concluded that for effective dispersion there exist an optimum weight ratio of

surfactant to CNTs close to 4.0.

170

190

210

230

250

01.53 4 56.258

Fracture Load (N)

Surfactant/CNTs weight ratio

CP w/c=0.5

Fig. 6. Fracture load of 28 days w/c=0.5 cement paste reinforced with CNTs (0.08wt% by

weight of cement)

2.2 Effect of CNTs/CNFs Type and Concentration

Current research is focused on exploring different processing methods (e.g. coat-

ing fibers with additives, varying the rheological properties of the matrix, mixing

procedure, etc) and on the optimization of the amount of the CNTs and CNFs used

to make the product cost effective. Moreover, the influence of CNT type (short

versus long) and the effect of the concentration of CNTs on the fracture properties

of nanocomposite samples were studied for a constant weight ratio of surfactant to

CNTs.

The fracture mechanics test results of the average Young’s modulus of the na-

nocomposites which demonstrate the best mechanical performance for water to

Advanced Cement Based Nanocomposites 321

cement ratios of 0.5 and 0.3 are illustrated in Figs 7(a) and 7(b), respectively.

When compared to the OPC paste the samples reinforced with CNTs clearly ex-

hibit improved mechanical performance. It is also noticed that, independently

from the water to cement ratio, specimens reinforced with either short CNTs at an

amount of 0.08 wt% or long CNTs at an amount of 0.025 wt% exhibit the same

level of mechanical performance. In general, it can be concluded that the opti-

mum concentration of CNTs depends on the aspect ratio of CNTs.

When CNTs with a low aspect ratio are used (short CNTs), a higher amount

close to 0.08wt% by weight of cement is needed to achieve effective reinforce-

ment. When CNTs with higher aspect ratio (long CNTs) are used, amounts less

3d 7d 28d

Young's Modulus (GPa)

Age (Days)

CP w/c=0.5

Cement Pa ste (CP)

CP+0.025wt% Long CNTs

CP+0.08wt% Short CNTs

Plain

0.025wt%

Long

0.08wt%

Short

(a)

3d 7d 28d

Young's Modulus (GPa)

Age (Days)

CP w/c=0.3

Cement Paste (CP)

CP+0.025wt% Long CNTs

CP+0.08wt% Short CNTs

Plain

0.08wt% Short

0.025wt%

Long

(b)

Fig. 7. Fracture mechanics test results of the Young’s modulus of cement paste nanocom-

posites with water to cement ratio of (a) w/c=0.5 and (b) w/c=0.3

322 S.P. Shah, M.S. Konsta-Gdoutos, and Z.S. Metax

than 0.048 wt% are needed to achieve a similar level of mechanical performance.

These differences are attributed to the degree of dispersion of the CNTs. Compar-

ing similar amounts of CNTs in the mixes, long CNTs exhibit a lower degree of

dispersion due to their higher aspect ratio. Consequently, adequate dispersion can

be achieved at lower amounts. Short CNTs exhibit a higher degree of dispersion

however, because they are shorter, a higher concentration in cement paste matrix

is needed to reduce the fiber free area and arrest the nanocracks.

2.3 Nanomechanical Properties

For the characterization at the nanoscale and the determination of local mechani-

cal properties, a nanoindenter with unique advantage of in-situ AFM like imaging

that allows pre and post-test observation of the cementitious samples, is being

used [16]. In any indentation technique, one material of known properties is used

to indent the material with unknown mechanical properties such as elastic

modulus and hardness. This technique has its origins in Mohs hardness scale de-

veloped in 1822, in which one material is considered to be harder if it can leave a

permanent scratch on another material. In nanoindentation, a small indenter is

pushed into a sample. Load applied by the indenter is plotted continuously with

the displacement of the indenter into the sample. This kind of plot is commonly

known as load-indentation or simply p-h plot (Fig. 8). The data obtained is then

analyzed to estimate elastic modulus, E and hardness, H of the sample.

The nanomechanical properties of the CNTs nanocomposites were investigated

using a Hysitron Triboindenter (Fig. 9) following the technique described in [17].

Before testing, thin slides of approximately 5 mm were cut out of the specimens.

The surfaces were polished with silicon carbide paper discs and diamond lapping

films in order to obtain a very smooth and flat surface. Nanoindentation was per-

formed in a 12×12 grid (10 μm between adjacent grid points). This procedure was

repeated in at least two different areas on each sample.

100

200

300

400

500

600

700

800

0 50 100 150 200

Load (micro N)

Displacement (nm)

Fig. 8. Force-displacement plot from nanoindentation data

Advanced Cement Based Nanocomposites 323

Fig. 9. Hysitron Triboindenter

0.1

0.2

0.3

0.4

0 - 5

5 - 10

10 - 15

15 - 20

20 - 25

25 - 30

30 - 35

35 - 40

40 - 45

45 - 50

Probability

Young's Modulus (GPa)

CP w/c=0.5

CP+0.08wt% Short CNTs

High Stiffness

C-S-H

(Ca(OH)

)

PorousPhase

Low Stiffness

C-S-H

Fig. 10. Probability plots of the Young’s modulus of 28 day cement paste and cement paste

reinforced with 0.08wt% short CNTs with w/c=0.5

The structure of cement paste at the nanoscale is dominated by the calcium-

silicate-hydrate (C-S-H) phase. Fundamental properties such as strength, fracture

behavior, shrinkage and durability are basically controlled by the properties of the

C-S-H and the porosity.

The effects of 0.08wt% short, 0.025wt% and 0.048wt% long CNTs on the na-

nomechanical properties of cement paste with w/c=0.3 and 0.5 were investigated.

Similar results were obtained for both water to cement ratios. Fig. 10 shows the

probability plot of the 28 days Young’s modulus of plain cement paste and cement

paste reinforced with 0.08wt% short CNTs for w/c of 0.5. The probability plots

are in good agreement with results from the literature [16, 18-19]. Values of the

Young’s modulus less than 50 GPa represent four different phases of cement paste

corresponding to the porous phase, low stiffness C-S-H, high stiffness C-S-H and

calcium hydroxide phase, while values greater than 50 GPa are attributed to na-

noindentation on unhydrated particles (clinker phases) presented in the material

[16, 18-19]. The different phases have been found to exhibit properties that

324 S.P. Shah, M.S. Konsta-Gdoutos, and Z.S. Metax

considered as distinct material properties and are independent of the mix propor-

tions. As expected, the peak of the probability plots of plain cement paste falls

in the area of the low stiffness C-S-H, which is the dominant phase of cement

nanostructure. On the other hand, the peaks of the probability plots of the nano-

composites are in the area of 20 to 25 GPa which corresponds to the high stiffness

C-S-H, suggesting that the addition of CNTs results to a stronger material with

increased amount of high stiffness C-S-H. Moreover, it is observed that the prob-

ability of Young’s modulus below 10 GPa is significantly reduced for the samples

with CNTs for both water to cement ratios. The nanoindentation results provide

an indirect method of estimating the volume fraction of the capillary pores [18]

indicating that the CNTs reduce the amount of fine pores by filling the area

between the C-S-H gel.

Fig. 11. Typical 60 × 60 µm AFM images of cement paste with (a) w/c=0.5 and (b)

w/c=0.3 reinforced with CNTs (0.08 wt% short) showing indentation modulus in GPa

Nanoindentation of the phases of Portland cement (C3S and alite, C2S and be-

lite, C3A, C4AF) have shown that the Young’s modulus of the clinker phases

range between 125 and 145 GPa [20]. In addition, in a previous study [19] using

the same type of cement and testing procedure, values of the Young’s modulus of

cement unhydrated particles close to 110 GPa (16,000 ksi) have been reported.

Fig. 11 shows typical 60×60 µm AFM images, captured using the nanoindenter

tip, of the nanostructure of cement paste with w/c=0.5 and w/c=0.3 reinforced

with 0.08 wt% short MWCNTs prior to nanoindentation. The values of Young’s

modulus calculated from the grid indentation are reported on the images in GPa at

the respective indent locations. It is observed that the stiffness varies reaching

values as high as 185 GPa which are attributed to the presence of unhydrated par-

ticles. The high values obtained from the nanocomposites indentation can be pos-

sibly attributed to the presence of CNTs in the vicinity of the unhydrated particles.

Please note that the nanoindentation tip is in the order of 100 nm and indentation

on individual MWCNTs can not be distinguished. These results suggest that

Advanced Cement Based Nanocomposites 325

MWCNTs appear to alter the nanostructure of the cement matrix resulting in a

stiffer and stronger material.

2.4 Autogenous Shrinkage

Typically, changes in the nanostructure affect the transport properties (properties

that are related with the movement of the water in the pores). Recently, it has

been increasingly recognized that high strength and high performance concrete is

sensitive to the microcracking that occurs at early ages, as a result of the volumet-

ric changes due to the development of high autogenous shrinkage stresses.

The autogenous shrinkage of cement nanocomposites was studied using a

modified version of ASTM C 341 and ASTM C 490. Cement paste specimens of

20x20x80 mm were cast following the procedure described above. Immediately

after setting (~6 hours after casting) specimens were demolded and sealed using

plastic wrap. Stainless steel gage studs were glued directly to the surface of the

specimens maintaining a 50.8 mm gage length, using a five minutes epoxy resin.

A length comparator was used to measure the distance between the stubs from the

time of final setting up to 96 hours after casting.

Fig. 12 shows the autogenous shrinkage results of plain cement paste and ce-

ment paste reinforced with 0.025% and 0.048wt% long CNTs. It is observed that

the samples reinforced with CNTs exhibit lower shrinkage than the plain cement

paste. Also, the samples reinforced with a higher amount of CNTs demonstrate

lower autogenous shrinkage. The shrinkage development is known to be propor-

tional to the amount of the fine pores (pores with diameter < 20 nm) in the binder

at early ages [21-22]. Higher percentage of the volume fraction of small pores in a

cementitious system at early ages leads to the increase of autogenous shrinkage.

Due to their small diameters (20 to 40 nm) CNTs appear to reduce the amount of

fine pores and also reinforce the nanostructure of cement paste, which leads to the

reduction of the capillary stresses, resulting in lower autogenous strains.

0.05

0.1

0.15

0.2

0 24487296

Autogenous Shrinkage (%)

Hours after casting (h)

CP w/c=0.3

CP+Long CNTs 0.025wt%

CP+Long CNTs 0.048wt%

Time of

setting

Fig. 12. Autogenous shrinkage of cement paste (w/c=0.3) and cement paste reinforced with

0.025wt% and 0.048wt% long CNTs

326 S.P. Shah, M.S. Konsta-Gdoutos, and Z.S. Metax

3 A New Generation of Cement Based Materials-Conclusions

The development of high-performance cementitious nanocomposites reinforced

with multiwall carbon nanotubes was studied. Effective dispersion was achieved

by applying ultrasonic energy and with the use of a surfactant. Results have

shown that for proper dispersion, the application of ultrasonic energy is required.

The combination of the SEM and fracture mechanics test results suggests that for

effective dispersion, a weight ratio of surfactant to CNTs close to 4.0 is required.

The fracture mechanics test results indicate that the fracture properties of cement

matrix increased through proper dispersion of small amounts of CNTs (0.025wt%

and 0.08wt%). In particular, higher concentrations of short CNTs are required to

achieve effective reinforcement, while lower amounts of longer CNTs are needed

to achieve the same level of mechanical performance. The nanoindentation results

suggest that CNTs can modify and reinforce the cement paste matrix at the nano-

scale by increasing the amount of high stiffness C-S-H and decreasing the poros-

ity, which leads to the reduction of the autogenous shrinkage. The autogenous

shrinkage results indicate that CNTs, except for the reinforcement effect, can also

have beneficial effect on other properties such as the transport properties of ce-

mentitious materials.

Acknowledgements

The authors would like to acknowledge the financial support from the ITI Institute

at Northwestern University under Grant DTRT06-G-0015/M1. The SEM experi-

ments were carried out in the EPIC facility of NUANCE center at Northwestern

University. The nanoindentation experiments were carried out in the NIFTI facil-

ity of NUANCE center at Northwestern University.

References

1. Lawler, J.S.: Hybrid fiber-reinforced in mortar and concrete. PhD thesis, Northwestern

University (2001)

2. Akkaya, Y., Shah, S.P., Ghandehari, M.: Influence of fiber dispersion on the perform-

ance of microfiber reinforced cement composites. ACI Special Publications 216: Inno-

vations in Fiber-Reinforced Concrete for Value, SP-216-1, 1-18 (2003)

3. Salvetat, J.P., Bonard, J.M., Thomson, N.H., Kulik, A.J., Forro, L., Benoit, W., et al.:

Mechanical properties of carbon nanotubes. Appl. Phys. 69, 225–260 (1999)

4. Belytschko, T., Xiao, S.P., Schatz, G.C., Ruoff, R.: Atomistic Simulations of Nano-

tubes Fracture. Phys. Rev. B 65, 235430–235437 (2002)

5. Lawrence, J.G., Berhan, L.M., Nadarajah, A.: Elastic Properties and Morphology of

Individual Carbon Nanofibers. ACS Nano 2, 1230–1236 (2008)

6. Cao, J., Wang, Q., Dai, H.: Electromechanical Properties of Metallic, Quasimetallic,

and Semiconducting Carbon Nanotubes under Stretching. Phys. Rev. Lett. 90,

157601–157604 (2003)

Advanced Cement Based Nanocomposites 327

7. Tombler, T.W., Zhou, C., Alexseyev, L., Kong, H., Dai, H., Liu, L., Jayanthi, C.S.,

Tang, M., Wu, S.Y.: Reversible electromechanical characteristics of carbon nanotubes

under local-probe manipulation. Nature 405, 769–772 (2000)

8. Makar, J.M., Beaudoin, J.J.: Carbon Nanotubes and Their Applications in the Con-

struction Industry. In: Bartos, P.J.M., Hughes, J.J., Trtik, P., Zhu, W. (eds.) in

Nanotechnology in Construction Proceedings of the 1st International Symposium on

Nanotechnology in Construction, pp. 331–341. Royal Society of Chemistry (2004)

9. Makar, J.M., Margeson, J., Luh, J.: Carbon nanotube/cement composites – Early re-

sults and potential applications. In: Proceedings of the 3rd International Conference on

Construction Materials: Performance, Innovations and Structural Implication, Vancou-

ver, B.C., Canada, pp. 1–10 (2005)

10. Li, G.Y., Wang, P.M., Zhao, X.: Mechanical behavior and microstructure of cement

com-posites incorporating surface-treated multi-walled carbon nanotubes. Carbon 43,

1239–1245 (2005)

11. Li, G.Y., Wang, P.M., Zhao, X.: Pressure-sensitive and microstructure of carbon nano-

tube reinforced cement composites. Cem. Concr. Comp. 29, 377–382 (2007)

12. Konsta-Gdoutos, M.S., Metaxa, Z.S., Shah, S.P.: Nanoimaging of highly dispersed

carbon nanotube reinforced cement based materials. In: Gettu, R. (ed.) Proceedings of

the Seventh International RILEM Symposium on Fiber Reinforced Concrete: Design

and Applications, pp. 125–131. RILEM Publications S.A.R.L. (2008)

13. Junrong, Y., Grossiord, N., Koning, C.E., Loos, J.: Controlling the dispersion o multi-

wall carbon nanotubes in aqueous surfactant solution. Carbon 45, 618–623 (2007)

14. Hielscher, T.: Ultrasonic production of nano-size dispersions and emulsions. In: Pro-

ceedings of 1st Workshop on NanoTechnology Transfer in Europe, TIMA Editions,

Grenoble, France, pp. 138–143 (2006)

15. Yang, Y., Grulke, A., Zhang, G.Z., Wu, G.: Thermal and rheological properties of car-

bon nanotube-in-oil dispersions. J. Appl. Phys. 99, 114307 (2006)

16. Mondal, P., Shah, S.P., Marks, L.D.: Nanoscale characterization of cementitious mate-

rials. ACI Mater. J. 105, 174–179 (2008)

17. Konsta-Gdoutos, M.S., Metaxa, Z.S., Shah, S.P.: Multi-scale mechanical and fracture

characteristics and early-age strain capacity of high performance carbon nano-

tube/cement nanocomposites. Cem. Con. Com. 32, 110–115 (2010)

18. Constantinides, G., Ulm, F.J.: The nanogranular nature of C-S-H. J. Mech. Phys. Sol-

ids 55, 64–90 (2007)

19. Mondal, P.: Nanomechanical properties of cementitious materials. PhD thesis, Evans-

ton, Northwestern University (2008)

20. Velez, K., Maximilien, S., Damidot, D., Fantozzi, G., Sorrentino, F.: Determination by

nanoindentation of elastic modulus and hardness of pure constituents of Portland ce-

ment clinker. Cem. Concr. Res. 31, 555–561 (2001)

21. Tazawa, E. (ed.): Autogenous shrinkage of concrete. E&FN Spon, London (1999)

22. Konsta-Gdoutos, M.S., Shah, S.P., Dattatraya, D.J.: Relationships between engineering

characteristics and material properties of high strength-high performance concrete. In:

Dhir, R.V., Newlands, M.D., Csetenvi, L.J. (eds.) Proceedings of International Sympo-

sia Celebrating Concrete: People and Practice, in Role of Cement Science in Sustain-

able Development, pp. 37–46. Tomas Telford Limited (2003)