Materials and Design 54 (2014) 864–868
Contents lists available at ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/matdes
Mechanical characterization of multiwalled carbon nanotubes-polycarbonate composites Prashant Jindal a, Meenakshi Goyal b, Navin Kumar c,⇑ a
University Institute of Engineering & Technology, Panjab University, Chandigarh 160014, India University Institute of Chemical Engineering & Technology, Panjab University, Chandigarh 160014, India c Indian Institute of Technology Ropar, Punjab 140001, India b
a r t i c l e
i n f o
Article history: Received 9 May 2013 Accepted 30 August 2013 Available online 7 September 2013
a b s t r a c t Present study is aimed to examine through experiments the mechanical properties of multi-walled carbon nanotubes-polycarbonate (MWCNT-PC) composite. Composites of MWCNT-PC were prepared by a two-step method of solution blending followed by compression molding. Multiwall carbon nano-tubes (MWCNTs) compositions in polycarbonate (PC) were varied by weight from 0.5% to 10%. Nano-indentation techniques were used to evaluate mechanical properties like elastic modulus and hardness. A marked increase in the elastic modulus (up to 95%) and hardness (up to 150%) has been observed even with the addition of small quantity (up to 2 wt%) of MWCNTs. The increase in mechanical properties is found to supplement earlier experimental investigation of these composites using dynamical impact through Split Hopkinson Pressure Bar (SHPB) instrument. Better load transfer property of MWCNTs, larger surface area and interaction between reinforcement with base matrix are the suggested reasons for this increase in mechanical properties. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Ever since the synthesis of carbon nanotubes (CNTs)  and study that followed exploring mechanical (100 times stronger than steel) and structural properties of CNTs [2–4], there has been wide ranging interests in scientiﬁc and engineering communities to exploit these for varying applications. The unusual mechanical strength of the carbon nanotubes has motivated scientists to fabricate and modify other useful materials which are cheaply available in bulk form, by combining them as composites with carbon nanotubes. Polycarbonate (PC) is a light weight polymer which is available in bulk form and is widely used for several engineering applications due it moldability. For taking advantage of the useful properties of polymers in combination with unique structural properties of carbon nanotubes, multi-walled carbon nanotubespolymer composites are being researched and fabricated over the past few years . In order to exploit the usefulness of these composites for speciﬁc mechanical engineering applications, their static and dynamic mechanical properties need to be evaluated. Among the static properties, hardness and elastic modulus of the specimen are very important. In order to evaluate static mechanical properties low weight testing becomes an important study for all materials that need to be deployed for any stress or pressure related application. Nano-indentation is one of the processes that has ⇑ Corresponding author. Tel.: +91 1881 223395. E-mail address: [email protected]
(N. Kumar). 0261-3069/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.matdes.2013.08.100
been used to evaluate these parameters for different type of materials in the past few decades. Nano-indentation is conveniently used to ﬁnd various static mechanical properties of various materials using an indenter tip under various set of loads for a speciﬁc period of time. The nanoindentation technique was developed in the mid-1970s to measure the hardness of small volumes of material [6,7]. In a traditional indentation test (macro or microindentation), a hard tip is used whose mechanical properties are known. It is pressed into a sample whose properties are unknown. Based on the formulations given by Oliver and Pharr [8,9] an applied load is preset at a ﬁxed value on the indenter tip for a given loading time during which the tip penetrates/presses into the specimen until it reaches preset ﬁxed load value. During this process, the maximum penetration depth (hmax) is calculated. After reaching the ﬁxed load the indenter starts unloading and moves away from the sample towards its initial position. The indent in the sample starts recovering and ﬁnally reaches a depth (hf). Fig. 1 represents the general process of indentation on any specimen. During this process, various shapes and materials of indenter tips can be used as per the type of specimen [10,11]. A three sided pyramid shaped diamond indenter tip, commonly known as Berkovich tip is widely used for this purpose. Recently, results have also been reported pertaining to static load tests on composites comprising of MWCNTs. Lot of work has been published related to tensile testing of MWCNT-PC composites. These tests have reported that minor compositions (up to
P. Jindal et al. / Materials and Design 54 (2014) 864–868
Fig. 1. General loading and unloading ﬂow during Nano-indentation process.
2 wt%) of MWCNT in PC enhance the modulus and tensile strength from 10% to even 70%. Choi et al.  used styrene and acrylonitrile (SAN) grafted MWCNTs with PC instead of pristine MWCNTs and observed that when SAN-grafted MWCNTs (1 wt%) were used with PC, both tensile strength and modulus increased by nearly 5% and 10% respectively in comparison to pristine MWCNT-PC composites. Liu et al.  observed that at 3 wt% MWCNTs in PC, the composites exhibited nearly 40% higher tensile strength in comparison to pure PC. However, for 5 wt% MWCNT composition the strength reduced drastically. Surface hardness for composites increased by 10–18% as MWCNT composition increased. There are also contrary results  which report no improvement in the presence of MWCNTs in the polymer poly(methyl methacrylate) (PMMA) for any static mechanical property. Even with change in compositions from 1% to 5% of MWCNTs the elastic modulus and hardness remained almost same to that pure PMMA. However, if the MWCNTs were coated with silica, the composite showed remarkable results upon nano-indentation. With only 4% MWCNT-Silica in PMMA, the hardness measured about two times to that of pure PMMA and modulus measured about three times to that of pure PMMA. Reinforcement on polyvinyl alcohol (PVA) and PMMA with few-layer graphene (FG) was also tested using nanoindenter by Das et al. . Low compositions of FG (0.6%) in PVA increased the modulus by about 20% and hardness by about 50%. For PMMA also there was 50% increase in hardness by reinforcing only 0.6% FG. Vivekchand et al.  have explained the use of inorganic nanowires (NW) as reinforcement in PVA to be as effective than MWCNTs. Elastic modulus increased by almost 2 times with 0.8% (by volume) reinforcement of inorganic NW. However, MWCNTs have a very smooth surface due to which the strength imparted by reinforcing MWCNTs is lesser than NW. Kim et al.  used a compatiblizer as Two poly-g-polycaprolactones (P3HT-g-PCLs) with bisphenol-A-PC-MWCNT composite. When PC-MWCNT composite was combined with P3HT-g-PCL then there was an increase of nearly 22% in the Young’s modulus and 30% in the tensile strength in comparison to pure PC. For small concentrations (0.1–0.5 wt%) of MWCNTs this increase was found to be consistent However, when MWCNTs concentration was further increased to 1 wt%, then it was observed that both Young’s modulus and tensile strength reduced considerably. Eitan et al.  used bisphenol-A-PC with MWCNTs as composite for mechanical characterization. Tensile tests were performed using a Universal Testing Machine and it was found that for composites with surface modiﬁed MWCNTs (5 wt%) the modulus improved by 95% in comparison to pure PC. Even for composites using pristine MWCNTs (5 wt%) the modulus rise was nearly 70% in comparison to pure PC. Ayatollahi et al.  have used epoxy-MWCNT composite under shear and bending load using a Santan Universal Testing Machine. They have also found that there is a gradual increase in elastic modulus and tensile strength as MWCNT composition
increased in epoxy. Compositions of 0.1%, 0.5% and 1.0% MWCNTs in Epoxy were fabricated and as composition of MWCNTs increased, both elastic modulus and tensile strength increased by 10%. Montazeri et al.  used a Hounsﬁeld machine and also evaluated visco-elastic behavior of epoxy-MWCNT composite. They have presented that with further increase in MWCNT composition of 2%, the elastic modulus increased about 20% as compared to pure Epoxy samples. On the basis of available experimental study, it has been observed that consistent and exhaustive study on the composites of pristine MWCNT with PC alone is lacking and the results reported do not elaborate the behavior of mechanical properties of the composites of PC by systematic enhancement in composition of MWCNT. Further, there seems to be reasonable variation in observations with composites comprising of MWCNTs with various other types of polymers although most of these experiments indicate a gradual increase in mechanical properties with increase in MWCNTs composition but study on MWCNT-PC composites alone has not been adequately reported. The results reported using MWCNT composites also need to be supplemented by nanoindenter study as an alternate widely accepted measuring technique and evaluation of modulus under compressive loading. Therefore, the present study is undertaken to investigate changes in the static mechanical properties under the inﬂuence of varying compositions of pristine MWCNTs in PC by employing nano indentation techniques without any additional component or any surface modiﬁcation in the composite. The properties like elastic modulus and hardness have been studied to supplement the investigations of a dynamical impact study on these composites already made . These properties will help us in mechanical characterization of PC based composites and the usefulness of MWCNTs as reinforcement from a comprehensive applications point of view.
2. Sample fabrication MWCNTs having diameter about 10–30 nm and length 1– 10 lm were procured from Nanoshel Intelligent Materials Pvt. Ltd., USA. These were characterized using Renishaw inVia Raman microscope at a wavelength of 514 nm. The spectrum is shown in Fig. 2 and the peaks at 1338/cm and 1563/cm are indicative of the D and G peaks respectively which are characteristics for MWCNTs. Pure PC beads were also procured from the same supplier. After the characterization process, MWCNTs were ﬁrst ultrasonically dispersed in Chloroform to obtain a stable suspension of carbon nanotubes as an initial step for the fabrication of MWCNT-PC composites . These suspensions were then mixed with similar dispersed solutions of PC in Chloroform to obtain a series of compositions of MWCNT/PC containing different weight percent (wt%) of MWCNT as 0.5, 0.75, 2, 5.0 and 10 wt% in pure PC. The compositions were then ultra-sonicated to obtain a uniform dispersion of MWCNTs in PC. Thin ﬁlms were casted from this solution by pouring the solution into a glass petri dish and allowing the solvent to evaporate. This resulted in formation of dried thin ﬁlms of varied compositions. Blank PC ﬁlms were also cast by the same technique. These composite ﬁlms were removed from the petridish and broken into pieces and pressed in a die mold to fabricate disc shaped composites with 10 mm diameter and 5 mm thickness. These discs were then used as specimen for nano-indention testing. Before indenting the specimens, various specimens were characterized using Raman spectroscopy. For composites with 5% and 10% MWCNT compositions, a Raman Spectra (Fig. 3) has been obtained for analyzing the synthesized composites. It can be observed that the intensity for D and G peaks was different for both composites. For 5% MWCNT composition the intensities for D (1343/cm) and G
P. Jindal et al. / Materials and Design 54 (2014) 864–868
Raman Shift (cm )
Fig. 4a. SEM image for 5% MWCNT-PC composite indicating dispersion.
Fig. 2. Raman spectra for MWCNTs with D (1338/cm) peak, G (1563/cm) and G0 (2678/s) peaks.
(1562/cm) peaks are nearly half in comparison to those for 10% MWCNT composition, indicating the synthesized samples have appropriate quantity of MWCNTs in the composites. Scanning Electron Microscopy (SEM) images of some of the samples thus prepared after being subjected to static and dynamic impact are shown in Figs. 4a and 4b to observe the extent of dispersion of MWCNTs in PC It can be seen that no aggregation of MWCNT appears in the ﬁeld of the images which have been presented at high concentrations of MWCNT from 5% to 10%.
3. Nano indentation Nano-indentation tests were performed using ISO 1457:2002 standards  on the above samples using Hysitron T1 950 TriboIndentor. Different compositions of MWCNT-PC composites were used as specimen for indentation and the results were then compared to the indentation results of pure PC samples. The disc shaped specimens were polished to a surface roughness of 0.2 lm. Berkovich tip was used for nano-indentation. As a standard practice, the tip area function was calibrated using a standard quartz sample. Load range for PC based specimen is very small, so the calibration was done for low depth ranges. Nano-indentation tests were done on single crystal Aluminum to verify the standard values of harness and modulus. Both values were found to be
10% MWCNT 5% MWCNT
Fig. 4b. SEM image for 10% MWCNT-PC composite indicating dispersion.
within 5% of the values prescribed by the manufacture, which validates the tip calibration process. Indents at different points on the surface were done at a ﬁxed load value of 1000 lN for a loading and unloading time period of about 24 s each. There was no holding period of the indenter at peak load. Fig. 5 depicts the complete loading and unloading process for various samples. The ﬁgure indicates that both hf, hmax (as discussed in Fig. 1) for different compositions of MWCNTs are much lower than that of pure PCs. Lower penetration depths at same peak loads is an indicator that higher amounts of MWCNTs increase the hardness of PCs which does not allow the indenter to penetrate deeper into the sample. Fig. 6 presents a real time image of the indent formation on one of the specimen to ascertain the indentation process. Table-1 gives the values of hardness and elastic modulus for all these compositions when the peak applied load is 1000 lN. Fig. 7 presents the hardness data and a polynomial least square curve ﬁt for variation of hardness with MWCNTs composition in PC.
4. Results and discussion 0 0
Raman Shift (cm ) Fig. 3. Raman spectra for composite with 5% MWCNT and 10% MWCNT composition in PC.
It is evident from Table 1 and Fig. 7 that static mechanical properties increase consistently as MWCNT composition increases in PC. The error bar in Fig. 7 is indicative of the difference in hardness values at different points on the surface where nano-indentation was done. When the MWCNTs composition is increased to only 2%, the mechanical properties are considerably enhanced indicating that
P. Jindal et al. / Materials and Design 54 (2014) 864–868
Fig. 5. Loading and unloading curves for various compositions under applied load of 1000 lN.
Fig. 6. Real time image of the indent formed on the specimen.
Measured points Polynomial fit
MWCNT % in PC Fig. 7. Hardness as measured using nano-indenter of various compositions of MWCNT-PC composites. The line is least square ﬁt to the data. The error bars are deviations from mean measurements on different points on the surface.
the MWCNTs bind strongly with the PC molecules forming a strong material in most sections as observed in the indention process. The values given in Table-1 for elastic modulus and hardness for these compositions indicate this consistent rise, however the growth of this rise in properties for higher composition is reducing. It can be
observed that hardness for 10% composition is 382.77 MPa, which in comparison to 5% composition (352 MPa) is only about 10% higher, but for minor compositions like 0.5% to 0.75% this rise is nearly 40%. Fig. 6 also conﬁrms that the increase in hardness number is very consistent up to 5% of MWCNTs. A further increase in concentration of MWCNTs tends to saturate the hardness effects. Addition of MWCNTs enhances the load transfer  capability in polymers which results in the enhancement of their mechanical properties. During the indenting process, indentation was done at several points on the complete surface. The overall average effect of these indents is considered to evaluate the properties of the base material so that the effect of non-homogeneity of the composite maybe minimized. Previous literature work relating to MWCNT composites with PC, PMMA etc. emphasizes the improvement in mechanical properties of these composites with increase in composition of MWCNTs. Along with MWCNTs, add on components were also used to either modify the surface of MWCNTs or to enhance interaction [12,17,18] between MWCNTs and PC which further increase the strength of these composites. The improvement in mechanical properties was found for minor compositions of MWCNTs in PC and they further increased as the composition of MWCNTs was increased. However, beyond a certain limit of composition  this pattern was not followed. It has been argued that proper dispersion at lower compositions, stronger p–p interaction  between MWCNTs and base matrix, combination of large aspect ratio and high surface to volume ratio of MWCNTs  and improved load transfer capability  of MWCNTs assisted in improvement of properties but for higher compositions aggregates [12,13,17,18] of MWCNTs were formed and the curvy and slippery  nature of MWCNTs did not assist in further improvement of mechanical properties. In the present study, there is an increase in hardness which tends to saturate as MWCNTs composition increases from 5 wt% to 10 wt%. The mechanism of increase in hardness is possibly due to stronger short range interactions between PC molecules mediated by MWCNTs and lasts up to some critical concentration of MWCNTs. The highly elastic nature of MWCNTs takes over at higher concentration causing reduction in hardness. It appears that for higher concentrations the elastic deformation becomes very difﬁcult to achieve but once it is reached then the composite will easily transit to plastic deformation. A competition between the two is covered by the saturation region from 5 wt% to 10 wt%. Most literature reports the characterization of PC based materials by using universal testing machine. However, in this study a nano-indenter has been used which is also used for characterization of various other types of materials. The composite materials used in this study as specimen comprise of nano sized ﬁllers in the form of MWCNTs. Hence it becomes imperative to study the properties at nano or microscale owing to the size of the ﬁller under observation and properties at this level are complimentary to the properties at macro level. Moreover, the study at speciﬁc points becomes important considering that initiation of any defect, fracture, void or non-homogeneity needs to be detected at such scales. Ultimately, the observations can be used to evaluate properties based on both speciﬁc locations and complete cross section, where as a universal testing machine can only evaluate properties based on the cross section. In an earlier work , dynamic load tests on similar samples using Split Hopkinson Pressure Bar (SHPB) were reported but the ﬁndings were somewhat different. It was found that for low concentrations of 0.5% MWCNTs, the increase in dynamic strength was most signiﬁcant and remained almost same up to 2% of MWCNTs, while for higher concentrations of 5% MWCNTs, this increase was negligible. The method used to determine the data based on SHPB was not dependent on a single point measurement
P. Jindal et al. / Materials and Design 54 (2014) 864–868
Table 1 Comparison of elastic modulus and hardness for different compositions. The data does not indicate ﬂuctuations due to different points on the composite surface. MWCNT (%) Average elastic modulus, E (GPa) Average hardness, H (MPa)
0.0 2.96 80.50
0.5 3.70 136.00
– rather a total effect on the whole cross section and the extent of the applied load was much higher which caused crushing of the specimen. For samples with 0.5–2% MWCNTs the impact on the surface got better absorbed. But in the case of 5% MWCNT samples impact strength of the composite was not enhanced as a whole. From the results of this paper, hardness number which gives a measure of local performance was found to rise to a certain limit with increase in MWCNTs. Considering a combined effect of both studies, it is safe to state that 2% MWCNT compositions in pure PC is the most suitable composition to enhance both static and dynamic mechanical properties to sufﬁcient extent for any mechanical load and pressure related application. Plastic deformation in amorphous polymers occurs due to nucleation and propagation of shear bands. In the unreinforced polymer matrix, shear bands propagate unchecked as there are no barriers for their movement. On the other hand, the presence of MWCNTs in the composites could offer resistance for the propagation of shear bands. The reasons for the enhancement of hardness and elastic modulus in polymer MWCNTs nano-composites are good mechanical interlocking and the presence of obstacles to the motion of shear bands. 5. Conclusions This paper presents results of static tests for hardness and elastic modulus obtained by a nano-indenter on a MWCNT based PC composite. The results indicate unambiguously that static mechanical properties of pure PC are greatly enhanced by composing these with less than 10% of MWCNT. The hardness increase by more than a factor of four is a recommendation for industrial use of PC in composite with MWCNT by a simple process of composite formation where strength is a weak point of PC. On the basis of the ﬁndings of this paper on static properties of MWCNT-PC composites, coupled with the ﬁndings of an earlier paper on dynamic impact studies, it is safe to conclude that a concentration around 2% of MWCNT is a good compromise to form composites suitable for enhancing both static as well as dynamic properties. Acknowledgements Prashant Jindal gratefully acknowledges ﬁnancial support from the Defence Research Organization (DRDO) for a research project (No. ARMREB/DSW/2011/129). The authors also wish to thank Dr. R.B. Mathur and his group from National Physical Laboratory, New Delhi for their assistance in sample fabrication. References  Ijima S, Endo M. Nanotube. Carbon 1995;33(7):869–1019.
0.75 4.35 187.00
2.0 4.43 207.00
5.0 5.74 355.00
10.0 6.99 382.77
 Ruoff Rodney S, Qian Dong, Liu Wing Kam. Mechanical properties of carbon nanotubes: theoretical predictions and experimental measurements. C R Physique 2003;4:993–1008.  Sears A, Batra RC. Macroscopic properties of carbon nanotubes from molecular-mechanics simulations. Phys Rev B 2004;69:235406.  Dresselhaus MS, Dresselhaus G, Eklund PC. Science of fullerenes and carbon nanotubes. San Diego: Academic Press; 1996.  Moniruzzaman Mohammad, Winey Karen I. Polymer nanocomposites containing carbon nanotubes. Macromolecules 2006;39:5194–205.  Bulychev SI, Alekhin VP, Shorshorov MK, Ternovskii AP, Shnyrev GD. Determining Young’s modulus from the indentor penetration diagram. Zavodskaya Laboratoriya 1975;41(9):1137–40.  Poon B, Rittel D, Ravichandran G. An analysis of nanoindentation in linearly elastic solids. Int J Solids Struct 2008;45(24):6018–33.  Oliver WC, Pharr GM. An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation techniques. J Mater Res 1992;7:1564–83.  Oliver WC, Pharr GM. Measurement of hardness and elastic modulus by instrumented indentation: advances in understanding and reﬁnements to methodology. J Mater Res 2004;19(1):3–20.  Bhushan Bharat, Li Xiaodong. Nanomechanical characterisation of solid surfaces and thin ﬁlms. Int Mater Rev 2003;48(3):125–64.  Pharr GM. Measurement of mechanical properties by ultra-low load indentation. Mater Sci Eng 1998;A253:151–9.  Choi Won Seok, Ryu Sung Hun. Improvement of interfacial interaction via ATRP in polycarbonate/carbon nanotube nanocomposites. Colloids Surf A: Physicochem Eng Aspects 2011;375(1–3):55–60.  Liu Sung-Po, Hwang Shyh-shin, Yeh Jui-Ming, Pan Kuan-Wei. Enhancement of surface and bulk mechanical properties of polycarbonate through the incorporation of raw MWNTs—using the twin-screw extruder mixed technique. Int Commun Heat Mass Transfer 2010;37(7):809–14.  Olek M, Kempa K, Jurga S, Giersig M. Nanomechanical properties of silicacoated multiwall carbon nanotubespoly(methyl methacrylate) composites. Langmuir 2005;21(7):3146–52.  Das Barun, Eswar Prasad K, Ramamurty U, Rao CNR. Nano-indentation studies on polymer matrix composites reinforced by few-layer graphene. Nanotechnology 2009;20:125705.  Vivekchand SRC, Ramamurty U, Rao CNR. Mechanical properties of inorganic nanowire reinforced polymer–matrix composites. Nanotechnology 2006;17:S344–50.  Kim Ki Hong, Jo Won Ho. A strategy for enhancement of mechanical and electrical properties of polycarbonate/multi-walled carbon nanotube composites. Carbon 2009;47(4):1126–34.  Eitan A, Fisher FT, Andrews R, Brinson LC, Schadler LS. Reinforcement mechanisms in MWCNT-ﬁlled polycarbonate. Compos Sci Technol July 2006;66(9):1162–73.  Ayatollahi MR, Shadlou S, Shokrieh MM. Fracture toughness of epoxy/multiwalled carbon nanotube nano-composites under bending and shear loading conditions. Mater Des 2011;32:2115–24.  Montazeri Arash, Montazeri Nasser. Viscoelastic and mechanical properties of multi walled carbon nanotube/epoxy composites with different nanotube content. Mater Des 2011;32:2301–7.  Jindal Prashant, Pande Shailaja, Sharma Prince, Mangla Vikas, Chaudhury Anisha, Patel Deepak, Singh Bhanu Pratap, Mathur Rakesh Behari, Goyal Meenakshi. High strain rate behavior of multi-walled carbon nanotubes– polycarbonate composites. Compos B: Eng 2013;45(1):417–22.  Pande Shailaja, Mathur RB, Singh BP, Dhami TL. Synthesis and characterization of multiwalled carbon nanotubes-polymethyl methacrylate composites prepared by in situ polymerization method. Polym Compos 2008;30(9):1312–7.  ISO 1457. Metallic materials – Instrumented indentation test for hardness and materials parameters’’. ISO Central Secretariat, 1 Rue de Varembé, 1211 Geneva 20, Switzerland; 2002.