Dynamic oscillatory shear properties of potentially melt processable high acrylonitrile terpolymers

Polyacrylonitrile (PAN) terpolymers consisting of acrylonitrile/methyl acrylate (AN/MA) systems with added amounts of a termonomer (acid based) were synthesized in an effort to generate melt spinnable carbon fiber precursors. The complex viscosity was measured as a function of time in the temperature range of 200-220 °C, and was found to be dependent on both, the amount, as well as the nature of the termonomer. For the different acid termonomers studied, the cyclization/crosslinking reactions occurring during stabilization as evidenced by the increase in viscosity over time of the PAN system ranged in the order, itaconic acid>acrylamide>methyl acrylic acid>acrylic acid (IA>AM>MAA>AA). Dynamic shear experiments indicated that an AN/MA system containing 3 mol% IA seemed well suited for melt spinning and stabilization. A chemorheological correlation was used to describe the viscosity variation in the PAN system as a function of IA content, temperature and time.     

Characteristics of a countercurrent reciprocating plate bubble column. I. Holdup,pressure drop and bubble diameter

The hydrodynamics of countercurrent air/water flow in a 5 cm diameter reciprocating plate bubble column have been studied; the plates contained 14 mm diameter perforations and had a fractional open area of 0.57. The ranges of superficial velocities of air and water were respectively 0-0.99 cm/s and 0-3.95 cm/s. The stroke was in most cases 4.5 cm and the reciprocation frequency was in the range 0–5 Hz. The pressure drops were measured in the absence of reciprocation for single phase and two phase flow conditions. Pressure fluctuations and time-averaged pressure drops were measured with plate reciprocation under single and two-phase conditions. The results were described in terms of the simple quasi-steady state model; the effective orifice coefficients of the perforations were within the range 0.4 to 0.97 depending on the reciprocation conditions. The Sauter mean diameters of the bubbles decreased with agitation; they were about twice the values predicted from an earlier correlation developed for liquid-liquid systems. The gas holdups were also substantially greater than predicted from correlations based on liquid-liquid systems. Both these effects were explained as due to the tendency for bubbles to cluster in the plate region.     

Axial mixing and scaleup of reciprocating plate columns

Axial mixing measurements in single phase (water) flow have been taken in open-type reciprocating plate columns of diameters 25.4 and 508 mm. In the case of the smaller column, two-phase axial mixing was measured, both in the dispersed phase (water dispersed in n-heptane) and the continuous phase (with n-heptane dispersed in water). Pulse injection of a tracer solution of ammonium chloride and methanol in water was used. Under single phase conditions, the axial dispersion coefficients were found to go through a minimum as the agitation level was increased from zero. The coefficients were nearly an order of magnitude higher in the 508 mm column than in the 25,4 mm column. In two phase flow in the 25,4 mm column, the continuous phase axial dispersion coefficients also went through a minimum as agitation was increased. The dispersed phase axial dispersion coefficients decreased monotonically as agitation was increased from zero. The results of this work and previous data are used in modelling the scale-up of reciprocating plate columns by means of Pratt's simplified technique. The existing empirical scale-up equation is consistent with an assumption that continuous phase mixing increased with column diameter but dispersed phase mixing remains unchanged.     

Holdup and flooding in reciprocating plate extraction columns

Measurements of dispersed phase holdup and flooding conditions have been carried out in a 5 cm diameter reciprocating plate extraction column. A simplified predictive model for holdup and flooding has been developed assuming uniform hydrodynamic conditions throughout the column. The model has been confirmed for several systems but may be limited in application by the following factors: (a) mass transfer, particularly from dispersed to continuous phase (coalescence effect), (b) wettability of plates by dispersed phase, (c) axial non-uniformity of holdup (noticed particularly for high interfacial tension systems), (d) the use of exceptionally high frequencies or low amplitudes of agitation.     

Miscibility and mechanical properties of poly(ether imide)/poly(ether ether ketone)/liquid crystalline polymer ternary blends

This paper is concerned with a novel ternary blend composed of poly(ether imide) (PEI), poly(ether ether ketone) (PEEK) and a liquid crystalline polymer (LCP; HX4000, Du Pont). Different compositions were prepared by extrusion and injection moulding. Dynamic mechanical thermal analysis and the observation of the fracture surfaces, before and after annealing, allowed determination of the cold crystallization temperatures and miscibility behaviour of these systems. PEEK/PEI blends are known from previous studies to be miscible at all compositions. In this case it was observed that the PEEK/HX4000 blend was miscible up to 50 wt% HX4000 but partially miscible above this value. The PEI/HX4000 blends were found to be partially miscible in the whole concentration range. As a result, some ternary blend compositions exhibited only one phase, while others exhibited two phases. The measurement of the tensile properties showed that ternary blends with high modulus can be obtained at high LCP loadings, while compositions with high ultimate tensile strength can be obtained with high loadings of PEI or PEEK.     

Effect of comonomers on melt processability of polyacrylonitrile

This work is concerned with the feasibility of achieving melt‐processable polyacrylonitrile copolymer systems as precursors for the manufacture of carbon fibers. Identification of suitable comonomers and achieving the optimum content of the comonomer were some of the main objectives of this investigation. It was seen that methyl acrylate (MA) was found to be suitable for enabling the melt processing of polyacrylonitrile (PAN), in terms of melt viscosity, time stability, and char yield. Below 10 mol % of MA, the PAN copolymer exhibited no flowability, even at very low molecular weights (～ 20,000). The long‐range order inherently present in PAN is speculated to be broken down at a critical value of about 10% MA at 220°C in the PAN‐based system, enabling its melt flowability. It was also seen that the incorporation of MA was seen to improve the temperature and molecular weight window of processability of the PAN system. The molecular weight cutoff for the 90/10 mol % acrylonitrile/methyl acrylate (AN/MA) system was about 50,000 at 220°C, whereas it was increased to about 100,000 in the presence of 15 mol % MA comonomer. Feasibility studies on the use of other comonomers such as higher acrylates and acrylamides were also conducted. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 85: 69–83, 2002     

Hypertext — a Bibliometric Briefing

Students' opinions about the degree of impact, status, and socio-cognitive viability with the utilization of the emerging three-dimensional (3D) computer-generated technologies may vary. Indisputably, 3D technology-enhanced environments have provided considerable benefits and affordances to the contemporary e-Education. In these circumstances, virtual worlds (VWs) like second life (SL) have generally intensified with an extensive perpetuation and penetration of innovative performances that encapsulated or enacted from the vast majority of higher education fields. At the same time, there is growing widespread recognition of reasons affecting the high or low degree of students' engagement in online and blended course delivery methods held in 3D VWs. Notwithstanding that most notable studies have disclosed SL functional capabilities from a plethora of pilot case studies, however, it is still lacking an experiential-based research approach to determine the degree of students' engagement in blended and online courses at university level through SL. The present comparative study explores students' engagement overall as a multidimensional construct consisting of emotional, behavioral, and cognitive factors. One hundred and thirty-five undergraduate and postgraduate students in almost identical blended and online instructional conditions held in SL took part in this project. Preliminary results have decoded students' satisfaction for both methods, despite the fact that the voluntary sample composed of different educational disciplines. The quantitative analysis showed that postgraduate students of the online course had more positive results and the degree of engagement significantly increased compared to those who enrolled with the blended course delivery method. The instructional affordances from the utilization of SL were the collaborative climate between users (instructor and students) who eliminated various intractable boundaries which were predominantly observed by several conventional methods. Specifically it was revealed that the online course delivery method engaged more students with the collaborative activities. Educational implications and recommendations for future research are also included.     

Introduction

We describe an SQL relational database schema for representing the objects in HyperCard, along with a technique for automatically populating this schema from a HyperCard stack using the facilities in HyperTalk with calls to the database manager. The standard relational database query language SQL can then be used to perform more general hypertext searches than are possible with the string search feature found in most hypertext browsing environments. Semiautomatic updates of the content of a hypertext are also possible using SQL updates on the object representations in the database to trigger corresponding HyperCard updates on the objects themselves. We describe a prototype implementation and present several example queries and updates to motivate this approach. These techniques, although demonstrated here specifically using HyperCard and Oracle for Macintosh, are generally applicable to a wide range of hypertext systems and relational databases.     

Analysis of selected trace organics in advanced wastewater treatment systems

Effluents through four different pilot tertiary wastewater treatment systems were monitored for selected trace organic compounds. The effects of using ozone, free and combined chlorine residuals in these systems were also studied. Advanced treatment of secondary effluent using various combinations of flocculation (alum and polymer), dual media filtration, and carbon adsorption were evaluated for production/removal of volatile halogenated organics, polynuclear aromatics, chlorinated pesticides, and polychlorinated biphenyls. Gas chromatographic methods were used for the analysis of these different classes of compounds: specific techniques and analytical parameters are described. Salient results included: drastic increases in trihalomethane production using free chlorine residuals: disinfection with combined chlorine species does not produce significant levels of trihalomethanes: approximately 90% reduction in trihalomethane levels by carbon adsorption: absence of detectable quantities of polynuclear aromatics: significant decreases in pesticide and PCB levels by carbon adsorption and chlorination. Statistical dependence of trihalomethane production on soluble COD, suspended solids and chloramine levels was evident from multiple linear regression calculations.     

Using localizing learning to improve supervised learning algorithms

Slow learning of neural-network function approximators can frequently be attributed to interference, which occurs when learning in one area of the input space causes unlearning in another area. To mitigate the effect of unlearning, this paper develops an algorithm that adjusts the weights of an arbitrary, nonlinearly parameterized network such that the potential for future interference during learning is reduced. This is accomplished by the reduction of a biobjective cost function that combines the approximation error and a term that measures interference. An analysis of the algorithm's convergence properties shows that learning with this algorithm reduces future unlearning. The algorithm can be used either during online learning or can be used to condition a network to have immunity from interference during a future learning stage. A simple example demonstrates how interference manifests itself in a network and how less interference can lead to more efficient learning. Simulations demonstrate how this new learning algorithm speeds up the training in various situations due to the extra cost function term.