No abstract

Analyses of fluid flow over a wavy wall are of interest because of their applications to the physical problems mentioned in § 1. The authors have therefore devoted their attention to the effect of waviness of one of the walls on the flow and heat-transfer characteristics of an incompressible viscous fluid confined between two long vertical walls and set in motion by a difference in the wall temperatures. The equations governing the fluid flow and heat transfer have been solved subject to the relevant boundary conditions by assuming that the solution consists of two parts: a mean part and a disturbance or perturbed part. To obtain the perturbed part of the solution use has been made of the long-wave approximation. The mean (zeroth-order) part of the solution has been found to be in good agreement with that of Ostrach (1952) after certain modifications resulting from the different non-dimensionalizations employed by Ostrach and the present authors respectively. The perturbed part of the solution is the contribution from the waviness of the wall. The zeroth-order, the first-order and the total solution of the problem have been evaluated numerically for several sets of values of the various parameters entering the problem. Certain qualitatively interesting properties of the flow and heat transfer, along with the changes in the fluid pressure on the wavy and flat wall, are recorded in §§ 5 and 6.

Iron was coordinately linked to the hydroxyl-terminated polybutadiene (HTPB) backbone using iron carbonyl via a ligand displacement reaction. The modified HTPB thus obtained was reddish brown in color and was characterized by GPC, FTIR, NMR, thermal, and propellant studies. No significant changes in the rheology, molecular weight, and molecular weight distribution were seen in the modified resin when the bonded Fe content was Ͼ0.8.0 wt %. However, the hydroxyl value of the resin decreased by 3-7 irrespective of the weight percent of the bonded Fe, and this was more likely due to the Fe-catalyzed oxidation of the -CH 2 OH moiety, mostly to the -CHO group. Apparently, this has not affected the cure characteristics of the binder, as demonstrated by the good mechanical properties of the gum stock and the propellant. The catalytic efficiency of the bonded Fe on the burn rate of the propellant was more efficient than was the free Fe added to the propellant. The aging characteristics of the resin for the bound iron content of Յ0.8 wt % was apparently good, as its viscosity and molecular weight did not undergo any drastic changes even after 18 months' storage under ambient conditions.

Crosslink density (CLD) is an important characteristic for elastomeric polymer networks. The mechanical and viscoelastic properties of the elastomers are critically dependant on the CLD. Several methods have been adopted for its determination, but swelling and stress-strain methods continue to be more popular because of the convenience associated with these techniques. In this article, the determination of CLD of allophanate-urethane networks based on hydroxyl-terminated polybutadiene and toluene diisocyanate with swelling and stress-strain methods is reported. The Flory-Rhener relationship was applied to calculate CLD from the swelling data. CLDs were also calculated from the initial slope of the stress-strain curve (Young's modulus), Mooney-Rivlin plots, equilibrium relaxation moduli, and dynamic mechanical properties. A comparison was drawn among the values obtained with the various methods. Although the CLD values obtained from Mooney-Rivlin plots were slightly lower than those obtained from swelling data, the values obtained with Young's modulus and storage modulus were considerably higher. The values obtained with swelling and equilibrium relaxation moduli data were very close to each other.