Elsevier

Polymer Testing

Volume 85, May 2020, 106455
Polymer Testing

Effect of microstructure uncertainty and testing frequency on storage and loss moduli of injection molded MWCNT reinforced polyamide 66 nanocomposites

https://doi.org/10.1016/j.polymertesting.2020.106455Get rights and content

Highlights

  • Injection molding is an effective approach for manufacturing well dispersed nanocomposites.

  • PA 66/CNT disk samples with five different CNT contents ranging from 3 wt % to 15 wt % were manufactured.

  • Nanocomposites with CNT content as high as 15 wt % shows improved storage modulus without saturation.

  • The variability in loss modulus is much higher than the variability in storage modulus.

Abstract

The viscoelastic behavior of multiwall carbon nanotube (MWCNT) reinforced polyamide 66 (PA 66) was evaluated to investigate the effect of CNT content and loading frequency on dynamic moduli (i.e. storage modulus E′ and loss modulus E″) and damping factor tanδ. PA 66/CNT disk samples with five different CNT contents ranging from 3 wt % to 15 wt % were manufactured by injection molding. Testing was performed over the frequency range of 0.1–100 Hz at room temperature. Dynamic mechanical analysis results show that the mechanical properties are highly functions of tested frequency and the improvement on loss and storage modulus of nanocomposites with the addition of CNT is highly dependent on tested frequencies. The variability in loss modulus is significantly higher than the variability in the storage modulus indicating the correlation of loss modulus with uncertainties present in nanocomposite microstructure while storage modulus is essentially independent of microstructure for a given reinforcement content.

Introduction

Nanocomposites have received extensive attention due to their superior thermal and electrical behavior compared to the neat polymer resin. While the addition of carbon nanotubes (CNT) improve thermal [[1], [2], [3]] and electrical [[3], [4], [5], [6], [7]] conductivity, their effect on mechanical properties must be considered as well. The outstanding mechanical properties of CNT, such as high strength-to-weight ratio and high stiffness does not directly translate to the resulted nanocomposite. Experimental as well as simulation results show that addition of CNT could even degrade the mechanical properties of polymer base [[8], [9], [10]]. Dotchev et al. [9] showed that the addition of 3% CNT to PA 66 reduced the elongation by 180%, without any noticeable change in strength and a slight increase in stiffness. Therefore, one should be cognizant of the effect of CNT on mechanical properties even if the objective is to improve electrical and thermal properties.

In addition to elastic properties, viscous behavior of nanocomposite must be evaluated. Dynamic mechanical analysis (DMA) is a powerful technique to decompose elastic and viscous behavior of materials. The DMA applies sinusoidal stress with a set frequency to the sample and the corresponding deformation is measured. For a viscoelastic material, the phase lag (δ) between stress and strain profile enables separating the elastic versus viscous component of the modulus. Therefore, the storage modulus (E′) associated with energy storage and loss modulus (E″) corresponding to energy dissipation of material can be determined. Tan δ is obtained as the ratio of E/E′ also referred to as mechanical damping factor. Tan δ can be interpreted as a measure of material energy loss due to molecular rearrangement and internal friction [11]. DMA has the advantage of quantifying mechanical properties at a range of frequencies and temperatures, which serves as a great tool for locating the dynamic properties as well as the glass transition temperature.

DMA has extensively been used in composite testing to evaluate the effect of micro-fiber composite microstructure on viscoelastic and damping characteristics. The viscoelastic behavior of polymer/fiber microcomposites has been well documented in a variety of thermoplastic materials including polystyrene and polyester reinforced with different fibers ranging from sisal fibers, banana fibers, and carbon fibers [[12], [13], [14], [15], [16]]. It is found that with increasing fiber content, the storage modulus is usually increased below Tg. Studies by Akay [12] examined the temperature at which the maximum values for E′, E″, and tanδ is observed in the glass transition region for carbon-fiber/epoxy composite laminates. Akay used DMA over a range of temperatures for different fiber alignments. They observed that tanδ peak always occurs at a higher temperature than E″ and that variations in temperature interval between peaks depend on the orientation of fiber with respect to the applied load. Manikandan et al. [13] studied the effect of reinforcing polystyrene with various wt. % of treated and untreated sisal fiber of different lengths on mechanical properties. They used DMA to measure dynamic moduli and tan δ over a range of temperature at set frequencies. They found a negative relationship between E′ and temperature. They further noted that the glass transition temperature of the composite was lower than that of a neat polystyrene. The negative relationship between E′ and the temperature was also observed in studies by Ray et al. [14]. They conducted dynamic mechanical thermal analysis (DMTA) at a set frequency on raw jute/vinyl ester composites of different jute volume percentages and alkali treated jute fiber/vinyl ester composites. They reported that increasing fiber volume fraction yielded higher E′. Such studies confirm the significant change in viscoelastic behavior of polymer with addition of reinforcement.

DMA testing on nanocomposites have revealed similar but not identical results compared to conventional composites. Almagableh et al. [17] compared dynamic moduli and creep behavior of graphite platelet, nanoclay/vinyl ester, and MWCNT/PA 66 composites. They conducted DMA tests on samples over a range of temperature at a single frequency. They reported that E′ increased as reinforcement increased. Bao et al. [18] found a similar trend for the relationship between CNT wt. % and storage modulus up to 1 wt % CNT focusing on MWCNT/polypropylene (PP) nanocomposites. They reported that a small amount of CNT remarkably increased storage and Young's moduli compared to pure polypropylene. This finding was also reported in Piatt's work [19] which examines the mechanical properties of MWCNT/epoxy composites. Additionally, Piatt explored the effect of CNT length on damping and mechanical properties of MWCNT/epoxy and found that long CNT/epoxy has slightly higher damping than short CNT/epoxy. Dalmas et al. [20] also recorded the positive relationship between CNT content and E′ for MWCNT/poly (styrene-co-butyl acrylate) [P(S-BuA)] nanocomposites. They noted that with the highest CNT content (3 vol %), MWCNT/P(S-BuA) nanocomposite retained their high ultimate strength. Zhao et al. [21] studied 3D undoped-MWCNT sponge reinforced polydimethylsiloxane (PDMS) and 3D boron-doped MWCNT reinforced PDMS. They found that with a low CNT content (1.5 wt %), the apparent modulus of the composite exceptionally increased. They also showed that with regards to E′ and loss factor, undoped-MWCNT/PDMS composite outweighs the boron –doped MWCNT/PDMS composite.

The present study evaluates the effects of load frequency on PA 66/CNT nanocomposites. The impact of loading frequency is important for PA 66/CNT high-end engineering applications such as automotive, aerospace, and sporting good applications. Studying dynamic properties of nanocomposite over a range of two decades of frequency with change in CNT wt. % revealed that E′ and E″ show different behavior at different frequencies. Degree of entanglement and reinforcement efficiency factor were determined to evaluate the dispersion and effectiveness of CNTs, respectively. In addition to the average properties, due to uncertainty present in nanocomposite microstructures, the variability was also quantified. This paper is the first attempt in evaluating frequency dependent properties of nanocomposites with high CNT content (15 wt %) using injection molding process.

Section snippets

Material

Pre-compounded PA 66 with MWCNT in the form of pellets of 15 wt % CNT (Plasticyl PA1501) was obtained in masterbatch form NanoCyl (Sambreville, Belgium). The master batch was diluted with the same commercial grade neat PA 66 to achieve lower CNT content nanocomposites through the injection molding process.

Sample preparation

The Plasticyl PA1501 and PA 66 were both dried at 80 °C for six hours, then the temperature was dropped down to 60 °C for 10 h to prevent moisture absorption overnight. The masterbatch was

Results and discussions

DMA testing was performed to quantify E′, E″, and tan δ over a range of frequencies. Nanocomposites with five different CNT wt. %, as well as a neat sample were tested. For each CNT wt. %, 8 samples were manufactured to capture the variability in the obtained properties. Fig. 3, Fig. 4, Fig. 5 use the averaged values of 8 samples at each CNT loading level. Fig. 3 compares E′, over a range of frequencies from 0.1 to 100 Hz with increasing CNT wt. %. There is a direct correlation between the CNT

Conclusions

This paper characterizes the effect of CNT wt. % and load frequency on dynamic mechanical properties of nanocomposites. Testing was performed over a range of frequencies from 0.1 Hz to 100 Hz at room temperature for five different CNT wt. %.

  • 1.

    High content of CNT (15 wt %) still changes the mechanical properties and saturation was not observed.

  • 2.

    Tan δ is primarily reliant on frequency rather than the CNT wt. %. Highest dampening is observed at the lowest frequency.

  • 3.

    Storage and loss moduli show

Declaration of competing interest

The authors confirm that there is no conflict of interest in preparation of this paper. All the sources have been properly cited.

References (35)

  • S.H. Yetgin

    Effect of multi walled carbon nanotube on mechanical, thermal and rheological properties of polypropylene

    J. Mater. Res. Technol.

    (2019)
  • S.H.R. Sanei et al.

    Uncorrelated volume element for stochastic modeling of microstructures based on local fiber volume fraction variation

    Compos. Sci. Technol.

    (2015)
  • S.H.R. Sanei et al.

    Length-scale dependence of variability in epoxy modulus extracted from composite prepreg

    Polym. Test.

    (2016)
  • D. Savvas et al.

    Effect of waviness and orientation of carbon nanotubes on random apparent material properties and RVE size of CNT reinforced composites

    Compos. Struct.

    (2016)
  • R.A. Ilyas et al.

    Development and characterization of sugar palm nanocrystalline cellulose reinforced sugar palm starch bionanocomposites

    Carbohydr. Polym.

    (2018)
  • B. Arash et al.

    Mechanical properties of carbon nanotube/polymer composites

    Sci. Rep.

    (2014)
  • C.E. Corcione et al.

    Characterization of nanocomposites by thermal analysis

    Materials (Basel)

    (2012)
  • Cited by (11)

    • The effect of uncertainty sources on the dynamic instability of CNT-reinforced porous cylindrical shells integrated with piezoelectric layers under electro-mechanical loadings

      2021, Composite Structures
      Citation Excerpt :

      Based on the structure of CNT-reinforced materials in nanoscale, such as direction, location, and geometry, the mechanical and physical properties of these materials may be associated with some level of uncertainty. This issue has previously been mentioned by various researchers [71–74]. In addition to the uncertainties related to the characteristics of the central core and the piezoelectric layers, the external sources of uncertainty such as those related to support and loading conditions exist in reality.

    • Water absorption and hydrothermal ageing of epoxy adhesives reinforced with amino-functionalized graphene oxide nanoparticles

      2021, Polymer Degradation and Stability
      Citation Excerpt :

      As seen from Figs. 7 and 8 the reinforcement efficiency is highly influenced by test conditions (frequency and ageing state) and the method or parameter used. This fact introduces some uncertainties in data interpretation [57], however, could also be used for quantitative monitoring of the reinforcement efficiency under different conditions. Similar considerations on the activation energies of epoxy/graphene nanocomposites were analysed in [65], where ΔE = 670 kJ/mol was found to be independent on the nanofiller loadings of 1 and 5 wt.

    View all citing articles on Scopus
    View full text