Nikil Kapur

A parametric study of direct gravure coating

An experimental investigation of direct gravure coating, focusing on roll-to-web fluid transfer over a wide range of conditions is presented. The results highlight that the pickout of fluid from the gravure cells (i.e. the fraction of the cell volume transferred to the web) is affected significantly by the ratio of the web-to-roll speed, the fluid properties and the shape and size of the gravure cells. The stability of the operability window is also explored. At high web-to-roll speed ratios the pickout from the gravure cells approaches one and an instability, caused by starvation, results in streaking on the web.

Displacement of liquid droplets on a surface by a shearing air flow.

The motion of droplets on surfaces is crucial to the performance of a wide range of processes; this study examines the initiation of droplet motion through a shearing mechanism generated here by a controlled air flow. Systematic experiments are carried out for a range of fluids and well defined surfaces. A model is postulated that balances surface tension forces at the contact line and the drag force due to the air motion. Experiments reveal that the critical velocity at which droplet motion is initiated depends on the contact angle and the droplet size. Visualizations highlight three modes of motion: (I) the droplet retains a footprint similar to that at the point of motion; (II) a tail exists at the rear of the droplet; (III) a trail remains behind the droplet (that can shed smaller droplets). The predictions of droplet initiation velocity are good for type I motion, in accordance with the assumptions inherent within the model. This model confirms the dominant physics associated with the initiation of droplet motion and provides a useful predictive expression.

Optimized implementation of the Lattice Boltzmann Method on a graphics processing unit towards real-time fluid simulation

Real-time fluid simulation is an active field of research in computer graphics, but they usually focus on visual impact rather than physical accuracy. However, by combining a lattice Boltzmann model with the parallel computing power of a graphics processing unit, both real-time compute capability and satisfactory physical accuracy are now achievable. The implementation of an optimized 3D real-time thermal and turbulent fluid flow solver with a performance of half a billion lattice node updates per second is described in detail. The effects of the hardware error checking code and the competition between appropriate boundary conditions and performance capabilities are discussed.

Bead-break instability

Flow is considered in a “fluid bead” located in the nip between two contra-rotating rolls, and bounded by two curved menisci. Such a flow arises in meniscus roll coating where fluid is transferred from the lower applicator roll to a substrate, in contact and moving with the upper metering roll, by means of a transfer jet (snake). Equilibrium of the bead is maintained through a balance of hydrodynamic and capillary stresses, the stability of which is considered experimentally by increasing the speed of the metering roll while that of the applicator roll remains constant. At a critical speed ratio, the upstream meniscus becomes unstable; the bead contracts as the meniscus accelerates forward and merges with its downstream counterpart—giving rise to “bead-break.” A mathematical model, based on lubrication theory, exhibits multiple solutions and a limit point for the existence of steady solutions. A linear stability analysis identifies the stable solution and shows that the flow becomes unstable at the limit ...

A theoretical and experimental investigation of reservoir-fed, rigid-roll coating

An industrially-important variant of reverse roll coating is studied in which the metering gap sits beneath a large, liquid reservoir, the influence of which is investigated via complimentary experimental, analytical (lubrication) and computational (finite element) methods, and for which gravitational effects are shown to be influential. Experimental measurements for both the flow rate and wetting line position are given over a wide range of roll speed ratio and capillary number and it is shown that, provided the wetting line is sufficiently far from the nip, the flow rate depends linearly on the reservoir level. A key feature of the mathematical models is that, unlike previous reverse roll coating studies, the variation of dynamic contact angle with metering roll speed has been accounted for. The lubrication model also uses boundary conditions which incorporate free surface, surface tension and wetting line effects and the predictions from both models are found to be in generally good agreement with experiment. Finally, streamlines obtained from Finite Element solutions of the flow in the reservoir and wetting line regions are found to compare well with corresponding experimental flow visualisations. The flow in the reservoir is recirculating in nature, the size and number of recirculations depending on the reservoir geometry.

Inducing protein aggregation by extensional flow

Significance Proteins are inherently sensitive to environmental factors that include hydrodynamic flow. Flow-induced protein remodeling is used in vivo and can also trigger the aggregation of therapeutic proteins during manufacture. Currently, the relative importance of shear and extensional hydrodynamic flow fields to aggregation remains unclear. Here we develop a flow device that subjects proteins to a defined and quantified flow field that is dominated by extensional flow. We show that extensional flow is crucial to induce the aggregation of globular proteins and that flow-induced aggregation is dependent on both protein structure and sequence. These observations rationalize the diverse effects of hydrodynamic flow on protein structure and aggregation propensity seen in both Nature and in protein manufacture. Relative to other extrinsic factors, the effects of hydrodynamic flow fields on protein stability and conformation remain poorly understood. Flow-induced protein remodeling and/or aggregation is observed both in Nature and during the large-scale industrial manufacture of proteins. Despite its ubiquity, the relationships between the type and magnitude of hydrodynamic flow, a protein’s structure and stability, and the resultant aggregation propensity are unclear. Here, we assess the effects of a defined and quantified flow field dominated by extensional flow on the aggregation of BSA, β2-microglobulin (β2m), granulocyte colony stimulating factor (G-CSF), and three monoclonal antibodies (mAbs). We show that the device induces protein aggregation after exposure to an extensional flow field for 0.36–1.8 ms, at concentrations as low as 0.5 mg mL−1. In addition, we reveal that the extent of aggregation depends on the applied strain rate and the concentration, structural scaffold, and sequence of the protein. Finally we demonstrate the in situ labeling of a buried cysteine residue in BSA during extensional stress. Together, these data indicate that an extensional flow readily unfolds thermodynamically and kinetically stable proteins, exposing previously sequestered sequences whose aggregation propensity determines the probability and extent of aggregation.

A Combined Experimental and Computational Fluid Dynamics Analysis of the Dynamics of Drop Formation

Abstract This article presents a complementary experimental and computational investigation of the effect of viscosity and flowrate on the dynamics of drop formation in the dripping mode. In contrast to previous studies, numerical simulations are performed with two popular commercial computational fluid dynamics (CFD) packages, CFX and FLOW-3D, both of which employ the volume of fluid (VOF) method. Comparison with previously published experimental and computational data and new experimental results reported here highlight the capabilities and limitations of the aforementioned packages.

Stirring and transport enhancement in a continuously modulated free-surface flow

The transport of fluid from a recirculation region adjacent to a free surface is studied using a numerical method validated with experimental flow visualization. The flow is an example of a liquid film coating process, and consists of two counter-rotating rolls placed side-by-side and half-submerged in a bath of fluid. In the gap between the rolls a recirculation zone exists just below the free surface, around which the flow splits into two films. Fluid recirculating for long periods has been identified as a source of coating defects, so this paper considers a possible method of inducing stirring. The flow is modulated by driving one of the rolls through a Hookes joint, which delivers a well-characterized periodic perturbation to the roll speed. In response to this speed modulation, the free surface undergoes a periodic change in position and shape which drives an exchange of fluid between the recirculation region and the surrounding flow. The amplitude of the free-surface motion is strongly dependent on modulation frequency. The dynamics of the free surface preclude a quasi-steady approach, even in the small-frequency limit, and so a fully time-dependent analysis based on the finite element method is employed. Trigonometric temporal interpolation of the finite element data is used to make passive tracer advection calculations more efficient, and excellent agreement is seen between simulation and experiment. Computations of the stable and unstable invariant manifolds associated with periodic points on the free surface reveal that the exchange of fluid is governed by a self-intersecting turnstile mechanism, by which most fluid entrained during a modulation cycle is ejected later in the same cycle. Transport over several cycles is explored by observation of the evacuation of passive tracers initially distributed uniformly in the recirculation zone. Results demonstrate the persistence of unmixed cores whose size is dependent on the modulation frequency. By considering the percentage of tracers remaining after a fixed number of cycles, contours in frequency-amplitude space show that for each modulation amplitude there is a frequency which produces the most effective transport, with up to 80 % of tracers removed by a modulation which produces only a 5 % change in film thickness. Finally it is shown how modulation of both rolls at slightly different phases can reduce the film thickness variation to about 1 % while maintaining the level of transport.

Flow Cell Design for Effective Biosensing

The efficiency of three different biosensor flow cells is reported. All three flow cells featured a central channel that expands in the vicinity of the sensing element to provide the same diameter active region, but the rate of channel expansion and contraction varied between the designs. For each cell the rate at which the analyte concentration in the sensor chamber responds to a change in the influent analyte concentration was determined numerically using a finite element model and experimentally using a flow-fluorescence technique. Reduced flow cell efficiency with increasing flow rates was observed for all three designs and was related to the increased importance of diffusion relative to advection, with efficiency being limited by the development of regions of recirculating flow (eddies). However, the onset of eddy development occurred at higher flow rates for the design with the most gradual channel expansion, producing a considerably more efficient flow cell across the range of flow rates considered in this study. It is recommended that biosensor flow cells be designed to minimize the tendency towards, and be operated under conditions that prevent the development of flow recirculation.

Assessing cement injection behaviour in cancellous bone : An in vitro study using flow models

Understanding the cement injection behaviour during vertebroplasty and accurately predicting the cement placement within the vertebral body is extremely challenging. As there is no standardized methodology, we propose a novel method using reproducible and pathologically representative flow models to study the influence of cement properties on injection behaviour. The models, confined between an upper glass window and a lower aluminium plate, were filled with bone marrow substitute and then injected (4, 6 and 8 min after cement mixing) with commercially available bone cements (SimplexP, Opacity+, OsteopalV and Parallax) at a constant flow rate (3 mL/min). A load cell was used to measure the force applied on the syringe plunger and calculate the peak pressure. A camera was used to monitor the cement flow during injection and calculate the following parameters when the cement had reached the boundary of the models: the time to reach the boundary, the filled area and the roundness. The peak pressure was comparable to that reported during clinical vertebroplasty and showed a similar increase with injection time. The study highlighted the influence of cement formulations and model structure on the injection behaviour and showed that cements with similar composition/particle size had similar flow behaviour, while the introduction of defects reduced the time to reach the boundary, the filled area and the roundness. The proposed method provides a novel tool for quick, robust differentiation between various cement formulations through the visualization and quantitative analysis of the cement spreading at various time intervals.