In this work, carbon nanotube (CNT) based nanocomposites with high mass fraction are proposed by in-situ bridging carbon matrix into CNT paper through optimized chemical vapor infiltration (CVI). Nanoinfiltration behavior of CNTs is basically investigated under the CVI process. The contact between each CNT can be strengthened and the conductive pathways can be established, resulting in the better mechanical and electrical properties. Compared with the pristine CNT paper, the CNT/C composite after pyrolysis process confirms a remarkable advance in tensile strength (up to 310 ± 13 MPa) and Young’s modulus (up to 2.4 ± 0.1 GPa). Besides, a notable feature of electrical conductivity also shows an improvement up to 8.5 S/cm, which can be attributed to the mass fraction of CNT (41 wt%) breaking the limits of percolation thresholds and the efficient densification of this sample to establish the conductive pathways. This study has a broad application in the development of the multi-functional electrical and engineering materials.
Keywords: Carbon nanotube | Chemical vapor infiltration (CVI) | Nanocomposites

In this work, two simulations models have been developed to study the electrical percolation and the electrical conductivity of epoxy-based nanocomposite containing Multi-walled Carbon Nanotubes. The models are based on resistor-model and finite element analysis. The former was evaluated using MATLAB code and the finite element analysis using DIGIMAT software. The maximum tunneling distance and its influence on the percolation probability and final electrical conductivity were studied. Electrical measurements on the samples were conducted for numerical validation. The experimental data showed a percolation achievement around 2 wt%, which was confirmed in the numerical simulations. This study provides evidence of the effectiveness of the resistor model and finite element method approach to predict the electrical conductivity of nanocomposites.
Keywords: Polymer-matrix composites (PMCs) | Nanocomposites | Carbon nanotube | Electrical properties | Computational modelling

With rapid evolution of advanced microelectronic devices, thermally conductive polymeric materials with impressive through-plane thermal conductivity (κ⊥) and remarkable electrical insulating properties are urgently demanded for efficient thermal management. Assembly of boron nitride nanosheets (BNNS) into polymer matrix with vertically interconnected conformation was usually adopted. However, impeded by inferior oriented degree and insufficient overlapped interconnection of BNNS, polymer/BNNS nanocomposites still suffered from a limited enhancement of κ⊥. In this work, we firstly prepared regenerated cellulose (RC)/BNNS filaments with high thermal conductivity via wet-spinning method. Then, a tailored polydimethylsiloxane (PDMS)/(RC/BNNS) filaments nanocomposite with hierarchical assembled architectures is successfully fabricated with customizedmold vacuum impregnation. Numerous bunches of microscopical hybrid filaments penetrate through the nanocomposite from the bottom to the up, while vertically aligned microscopical BNNS with efficient filler interconnections confined within filaments provide many uniaxial thermal pathways. Attributed to tactfully engineered phonon pathways, the as-prepared material exhibits a remarkable κ⊥ up to 5.13 W/m K at 27.05 vol% BNNS loading, outperforming the characters in most literatures. Along with excellent electrical resistance and extraordinary thermal management performances, this type of nanocomposite with hierarchical structures provides an effective strategy to thickness-direction heat transfer issues of electronic devices.
Keywords: Through-plane thermal conductivity | Polymer | Boron nitride nanosheets | Vertical orientation

Rubber nanocomposites experiencing cyclic deformation undoubtedly exhibit Mullins effect whose underlining mechanisms are not yet clear. Herein this effect in nitrile butadiene rubber nanocomposites is systematically investigated for revealing the influences of pre-strain interval, loading and unloading velocities, temperature, filler type and content, as well as crosslinking agent. The results show that the recovery hysteresis energy and accumulative softening energy of the nanocomposites can be superposed onto master curves as a function of microscopic strain of the rubber phase, revealing that both involving the viscoelastic deformation of the rubber phase. Especially the recovery hysteresis highly depending on temperature and loading and unloading velocities is connected to the viscoelasticity of nonideally crosslinked rubber network in the nanocomposites. On the other hand, the accumulative softening energy loss comes from recovery retardation of rubber chains and is somewhat sensitive to the filler, temperature and crosslinking agent. The investigation would be instructive to clarify the physical origin of Mullins effect to produce low dissipation rubber nanocomposites.
Keywords: Mullins effect | Energy loss | Nanocomposites

In the study, the voltammetric electronic tongue based on three nanocomposites modified electrodes was applied for the identification of rice wines of different geographical origins. The nanocomposites were prepared by gold and copper nanoparticles in the presence of conducting polymers (polymer sulfanilic acid, polymer glutamic acid) and carboxylic multi - walled carbon nanotubes. The modified electrodes showed high sensitivity to guanosine - 5 - monophosphate disodium salt, tyrosine and gallic acid which have good correlation with the geographical origins of rice wines. Scanning electron microscopy was performed to display the surface morphologies of the nanocomposites, and cyclic voltammetry was applied to study the electrochemical behaviors of the taste substances on the electrode surfaces. Four types of electrochemical parameters (pH, scan rates, accumulation potentials and time) were optimized for getting a low limit of the detection of each taste substance. The geographical information of rice wines was obtained by the modified electrodes based on two types of multi - frequency large amplitude pulse voltammetry, and “area method” was applied for extracting the feature data from the original information obtained. Based on the area feature data, principal component analysis, locality preserving projection (LPP), and linear discriminant analysis were applied for the classification of the rice wines of different geographical origins, and LPP presented the best results; extreme learning machine (ELM) and alibrary for support vector machines were applied for predicting the geographical origins of rice wines, and ELM performed better.
Keywords: Nanocomposites modified electrodes | Conducting polymer | Multi - walled carbon nanotubes | Rice wine | Pattern recognition