Richard Turton

Modeling weight variability in a pan coating process using Monte Carlo simulations

The primary objective of the current study was to investigate process variables affecting weight gain mass coating variability (CVm) in pan coating devices using novel video-imaging techniques and Monte Carlo simulations. Experimental information such as the tablet location, circulation time distribution, velocity distribution, projected surface area, and spray dynamics was the main input to the simulations. The data on the dynamics of tablet movement were obtained using novel video-imaging methods. The effects of pan speed, pan loading, tablet size, coating time, spray flux distribution, and spray area and shape were investigated. CVm was found to be inversely proportional to the square root of coating time. The spray shape was not found to affect the CVm of the process significantly, but an increase in the spray area led to lower CVms. Coating experiments were conducted to verify the predictions from the Monte Carlo simulations, and the trends predicted from the model were in good agreement. It was observed that the Monte Carlo simulations underpredicted CVms in comparison to the experiments. The model developed can provide a basis for adjustments in process parameters required during scale-up operations and can be useful in predicting the process changes that are needed to achieve the same CVm when a variable is altered.

Movement of Different-Shaped Particles in a Pan-Coating Device Using Novel Video-Imaging Techniques

The purpose of this study was to investigate the effects of particle shape on the movement of particles in a pan-coating device using novel video-imaging techniques. An area scan CCD camera was installed inside a 24-in pan coater at the same location as that of a spray nozzle, and the movement of particle was tracked using machine vision. A white tracer particle was introduced inside a bed of black-coated particles. The effects of pan loading, pan speed, and particle shape on the movement of particles was studied. The response variables were circulation time, surface time, projected area of particle per pass, dynamic angle of repose, cascading velocity, and dispersion coefficient. Experiments were conducted at 3 different pan speeds, 6, 9, and 12 rpm, and 2 fill levels (ratio of bed depth to pan diameter), one eighth and one quarter, and data were collected over a 30-minute time period. The differences in circulation times of spheres and tablets, with similar volume equivalent diameter as that of the sphere, were found to be insignificant at the 95% confidence interval. The circulation time ranged from 2.8 to 10.8 seconds depending on the operating condition and increased with increasing pan load and decreasing pan speed. The distributions of circulation time, surface time, and projected surface area were found to be nonnormal. The dynamic angle of repose for tablets was higher than for spheres. Also, the bed surface for spheres was much flatter in comparison with tablets where the bed shape attained a “wave-like” form. The average velocity of tablets in the cascading layer was found to be significantly higher than spheres. A linear model (R2>0.98) best described the variation of velocity as a function of pan speed for all of the operating conditions.

Scale-up of a pan-coating process

The purpose of this work was to develop a practical scale-up model for a solvent-based pan-coating process. Practical scale-up rules to determine the key parameters (pan load, pan speed, spray rate, air flow) required to control the process are proposed. The proposed scale-up rules are based on a macroscopic evaluation of the coating process. Implementation of these rules does not require complex experimentation or prediction of model parameters. The proposed scale-up rules were tested by conducting coating scale-up and scale-down experiments on 24-inch and 52-inch Vector Hi-coaters. The data demonstrate that using these rules led to similar cumulative drug release profiles (f2≫50; and P Analysis of Variance [PANOVA]≫0.05 for cumulative percentage of drug released after 12 hours [Cum 12] from tablets made at 24- and 52-inch scales. Membrane characteristics such as opacity and roughness were also similar across the 2 scales. The effects of the key process variables on coat weight uniformity and membrane characteristics were also studied. Pan speed was found to be the most significant factor related to coating uniformity. Spray droplet size was found to affect the membrane roughness significantly, whereas opacity was affected by the drying capacity.

Nonuniformity of particle coating on a size distribution of particles in a fluidized bed coater

Abstract The focus of the current work is to evaluate the effect that particle size has on the uniformity of coating received during a batch fluidized bed coating process. Batches of near-spherical Nu-Pareil™ sucrose particles comprising a wide range of sizes (548–1205 μm) were formulated so as to have the same mass average particle size but with different size distributions. These batches were then coated under identical coating conditions using a coating medium of FD&C (Food, Drug, and Cosmetic) blue dye no. 1 dissolved in a suitable aqueous-based polymeric binder. At the end of each run the particles were sieved back to the different size fractions. From each size fraction 30 particles were chosen at random and analyzed for their blue dye content. The results showed that the amount of blue dye deposited on a given size of particle was proportional to the particle diameter to the power of 3.4 (dp3.4). The variance of blue dye content for different size cuts was also measured. It was found that the variance could also be correlated with particle diameter but that the dependence was lower than that for the mean (dp3.0). The experimental procedures and possible mechanistic explanations for these results are presented.

Data interpretation techniques for inferring bubble size distribution from probe signals in fluidized systems

Abstract Bubble behavior in fluidized beds may have a profound effect on other aspects of bed hydrodynamics. Submersible probes are often used to study the local bubble properties in these systems. Generally, the axes of the pierced bubbles need not be aligned with the vertical axis of the bed. This matter is further complicated by the fact that there is a distribution of bubble size. Therefore, data interpretation of probe signals should be performed carefully to yield bubble size distribution information. In this paper an analytic data interpretation technique in closed form has been presented based on the assumption that the bubble rise velocity depends on bubble size (usually the velocity increases as the square root of the linear size). A method employing a non-parametric approach to infer the bubble size distribution from the probe time signals (pierced time durations) without need for information about the nature of the pierced time duration distribution has also been described. A statistical data transformation approach for obtaining the average bubble size and the standard deviation of the local bubble size distribution has been established. Monte-Carlo simulations have been conducted to demonstrate the correctness of these techniques.

The measurement of instantaneous heat transfer coefficients around the circumference of a tube immersed in a high temperature fluidized bed

Abstract Instantaneous and time-averaged local heat transfer coefficients are obtained for an immersed heat transfer tube in a high temperature (900 K) fluidized bed. Surface temperature measurements are made by very fine ribbon thermocouples circumferentially located 120° around the surface of the immersed tube. Surface temperature measurements are converted to local instantaneous heat transfer coefficients by solving the conduction equation for the tube wall. Results from this work are compared to previous work in fluidized beds.

The application of modeling techniques to film-coating processes

Background: The modeling of the processes for film coating of solid dosage forms can aid in the prediction and troubleshooting of coating operations. Aim: A review of the existing approaches to modeling the coating phenomena was undertaken with the aim of identifying key assumptions and limitations of each approach. Method: Models are categorized into macrolevel and microlevel approaches. Macrolevel models can predict mass coating uniformity for a batch of solids. On the other hand, microlevel models consider phenomena at the scale of the liquid droplets sprayed into the process. Macrolevel models consist either of phenomenological approaches, whereby the inherent variabilities in the coating process are described by probability distributions. Alternatively, first-principle models based on rigorous descriptions of solid and gas-phase motion can be used to describe the multiphase flow behavior in coating equipment. Result: The advantages of the rigorous macrolevel approach are that extrapolation of results for different operating conditions is more reliable than the phenomenological approach and much more information is available about the flow of both phases. However, these first-principle models are computationally intensive and cannot be used for real-time predictions. On the other hand, microlevel models describe chemical and physical phenomena occurring when droplets of coating solution impact the surfaces of the solid cores and are useful in determining the cause of morphological variations in coating properties. Conclusions: Both micro- and macro-level approaches can be used to predict and troubleshoot problems in coating processes. The use of such models in the control of coating operations is not widespread but offers significant potential to improve quality control.

The application of computer-based imaging to the measurements of particle velocity and voidage profiles in a fluidized bed

Abstract Computer-based video imaging techniques were utilized in order to measure the axial and radial components of particle velocity and voidage profiles in the draft tube region of a semi-circular spouted fluid bed coating device. Computer images were obtained using standard RS-170 video signals (30 frames/s) from an image processing board and a CCD variable shutter speed camera. When measuring particle velocities, low shutter speeds of 1–2 ms were utilized. Due to rapid motion of particles, the images obtained were blurred streaks several millimeters in length. The length and direction of streaks were proportional to the magnitude and direction of the projected velocity vector. By using dispersed back lighting, the particle velocity images appeared as light streaks on a dark background. Custom software was written in order to automatically search for and to identify bright streaks and calculate particle velocities. The velocity measurements obtained from the software were compared with data taken with a high speed Kodak video system (1000 frames/s with a 20-μs strobe). The results were in close agreement. The video images for analysis of voidage measurements were obtained using the same techniques as for the velocity measurements except with a shutter speed of 0.1 ms. The images thus obtained showed the particles to have a dark outer ring with a light center. The search algorithm for identifying particles involved establishing the gray level for nine different locations: eight compass points and a center point. By searching the frame for the presence of particles, and then comparing the gray levels at the circumference and the center of the particle image it was possible to distinguish between in-focus and out-of-focus particles. A comparison of the results using the software with those of standard models made with glass beads, and by visually counting the particles, was in close agreement.

Interpreting probe signals from fluidized beds

Abstract Bubble size distribution in fluidized beds may have a profound effect on reactor mass transfer and hydrodynamics. Although several different types of probes are available to monitor bubble events in fluidized beds, the probe signals must be processed carefully to yield bubble size information, because a probe will not always intersect a bubble centrally and because bubble rise velocity will depend on bubble size (usually velocity increases as the square root of linear size). Using a truncated ellipsoid as a model of bubble shape, the relationship between probe signals and bubble size has been explored using probability density functions. A back-transform technique has been presented to yield the bubble size distribution for a given distribution of time intervals when the probe ‘cuts’ a bubble. This back-transform has been demonstrated for two different bubble size distributions. When too few bubble measurements are taken or too many intervals in the discrete bubble size distribution are used, instabilities in the back-transform may result.

PARTICLE VELOCITY AND VOIDAGE PROFILES IN A DRAFT TUBE EQUIPPED SPOUTED-FLUIDIZED BED COATING DEVICE

Particle motion in the spray region of a semicircular spouted-fluidized bed coating device was investigated. Radial and axial particle velocity and voidage profiles were obtained using computer-based video imaging techniques and custom software. In order to investigate the effect of liquid droplets on fluidized particles in the spray region, measurements were carried out both with and without an atomized water spray fed to the bed. In the draft tube region, radial profiles of particle velocities varied from parabolic to nearly flat. At the spray rates used in this work there was no difference in velocity and voidage measurements with or without spray.