No abstract

Polyamic acid, the precursor of polyimide, was used for the preparation of polyimide/multiwalled carbon nanotubes (MWCNTs) nanocomposite films by solvent casting technique. In order to enhance the chemical compatibility between polyimide matrix and MWCNTs, the latter was surface modified by incorporating acidic and amide groups by chemical treatment with nitric acid and octadecylamine (C18H39N), respectively. While the amide-MWCNT/polyimide composite shows higher mechanical properties at low loadings (<3 wt%), the acid-MWCNT/polyimide composites perform better at higher loadings (5 wt%). The tensile strength (TS) and the Young’s modulus (YM) values of the acid-MWCNT/polyimide composites at 5 wt% MWCNT loadings was 151 and 3360 MPa, respectively, an improvement of 54% in TS and 35% in YM over the neat polyimide film (TS = 98 MPa; YM = 2492 MPa). These MWCNT-reinforced composites show remarkable improvement in terms of thermal stability as compared to that for pure polyimide film. The electrical conductivity of 5 wt% acid modified MWCNTs/polyimide nanocomposites improved to 0.94 S cm−1(6.67 × 10−18 S cm−1for pure polyimide) the maximum achieved so far for MWCNT-polyimide composites.

No abstract

A comprehensive approach for modelling dependent radiative heat transfer in beds of large (geometric range) spherical particles is presented. Such a system of large spheres lies in the dependent range even for large porosities. We show that the dependent properties for a bed of opaque spheres can be obtained from their independent properties by scaling the optical thickness while leaving the albedo and the phase function unchanged. The scaling factor is found to depend mainly on the porosity and is almost independent of the emissivity. We show that such a simple scaling for non-opaque particles is not feasible. The transparent and semi-transparent particles are treated by allowing for the displacement across an optical thickness (because of transmission through a particle) while solving the equation of radiative transfer. When combined with the scaling approach, this results in a powerful method of solution called the dependence included discrete ordinates method (DIDOM). The results obtained from the DIDOM give good agreement with the results obtained from the Monte Carlo method.