Search

A combination of analytical and statistical methods is used to improve a tablet coating process guided by quality by design (QbD) principles. A solid dosage form product was found to intermittently exhibit bad taste. A suspected cause was the variability in coating thickness which could lead to the subject tasting the active ingredient in some tablets. A number of samples were analyzed using a laser-induced breakdown spectroscopy (LIBS)-based analytical method, and it was found that the main variability component was the tablet-to-tablet variability within a lot. Hence, it was inferred that the coating process (performed in a perforated rotating pan) required optimization. A set of designed experiments along with response surface modeling and kriging method were used to arrive at an optimal set of operating conditions. Effects of the amount of coating imparted, spray rate, pan rotation speed, and spray temperature were characterized. The results were quantified in terms of the relative standard deviation of tablet-averaged LIBS score and a coating variability index which was the ratio of the standard deviation of the tablet-averaged LIBS score and the weight gain of the tablets. The data-driven models developed based on the designed experiments predicted that the minimum value of this index would be obtained for a 6% weight gain for a pan operating at the highest speed at the maximum fill level while using the lowest spraying rate and temperature from the chosen parametric space. This systematic application of the QbD-based method resulted in an enhanced process understanding and reducing the coating variability by more than half.

A computational study using the discrete element method was performed to study the effect of pan speed, fill level and the design of the spray pattern on the coating variability of tablets coated in a rotating pan. The method simulates the movement of tablets in the pan and calculates the residence time of each tablet inside the spray zone, which is directly related to the amount of coating received by the tablet. The computational method was experimentally validated using a Laser Induced Breakdown Spectroscopy based analytical method. The simulations showed that the axial mixing was the most critical parameter affecting the coating variability. Although axial mixing was found to be better at higher pan speed, it did not affect the coating variability significantly. Lower variability was obtained when a 100% fill level was used as compared to 67% fill. Four spray patterns were used, two idealized (full surface spray and a symmetric band spray) and two realistic (5-ellipse and 5-circular spray guns). The full and band spray showed similar results while the ellipse and circular patterns were similar to each other (and much worse than the other two patterns) at all speeds and fill levels.

The application of computationally inexpensive modeling methods for a predictive study of powder mixing is discussed. A multidimensional population balance model is formulated to track the evolution of the distribution of a mixture of particle populations with respect to position and time. Integrating knowledge derived from a discrete element model, this method can be used to predict residence time distribution, mean and relative standard deviation of the API concentration in a continuous mixer. Low-order statisticalmodels, including responsesurfacemethods, kriging, and high-dimensional model representations are also presented. Their efficiency for design optimization and process design space identification with respect to operating and design variables is illustrated.

In recent years, high-pressure homogenization has been used with increasing frequency to reduce the size of pharmaceutical suspensions to the micron, submicron, and nano scales with the goal of increasing the dissolution rate of the drug, and consequently, the in vivo bioavailability. As particle suspensions become smaller in size, increased concentrations of surfactants and stabilizers are required to prevent particle agglomeration. Recently, relatively high concentrations of biocompatible polymers have been used to stabilize suspensions by imparting surface-active steric stability as well as kinetic stability through an increase in suspension viscosity and/or a transition to non-Newtonian rheological properties. While the benefits of stable suspensions are known, little is known about the effect of the high polymer concentration on the actual breakup of the particles in suspension. In this work, designed experiments were used to identify the key parameters that govern the final particle size of a drug suspension following homogenization. In particular, drug loading, operating pressure, and polymer concentration were all found to be statistically significant factors in determining the resulting size of particles in suspension. In addition, it was found that an optimal polymer concentration of 2% (w/w) was required to adequately stabilize the particle suspension while simultaneously avoiding a decrease in homogenizer performance. Concentrations higher than 2% (w/w) resulted in size reduction limitations due to the increased viscosity (>10 cP) and the transition to non-Newtonian behavior.

A laser-induced breakdown spectroscopy-based analytical methodology is developed to study tablet coating variability in pharmaceutical tablets. The method quantifies the amount of coating on a tablet by assigning an average coating thickness score to it. When tested using samples with different amounts of coating, the coating thickness scores showed direct correlation to the weight gain of the tablet, hence validating the analytical method. The relative significance of the processing parameters and the components of variability were computed using statistical techniques. The sampling frequency of laser shots was found to have no significant effect. The effect of the position of the laser pulse on the tablet surface was found to be significant; however, it was determined that this effect was due to the tablet curvature, which resulted in the laser optical path to intersect the coating diagonally. The variability between batches (lots) manufactured under the same processing conditions was not significant. The largest avoidable source of variability was the tablet-to-tablet component, possibly indicating the inadequate mixing performance of the coating device.

In many pharmaceutical process operations, first-principle models describing the behavior of the processed powders are difficult to derive or computationally very expensive and thus can be considered as “black-box” processes. Kriging is an efficient data-driven methodology that has been shown previously to predict steady-state operation of black-box processes while providing error estimates for each prediction. In this work, the ability of Kriging to capture the dynamic behavior of processes is evaluated and compared to the performance of dynamic Neural Network (NN) models. Design of Experiment (DoE) concepts are employed to collect the necessary dynamic input–output information that forms the initial multivariate database of the model. The most important feature of the proposed dynamic Kriging methodology is the ability to improve its predictive efficiency as the database is augmented by additional useful information while the process operates. An error tolerance is used in the adaptive procedure, so tha...

The effect of feed frames, a device used in rotary tablet presses to drive the powders into compression dies, on the properties of the powders entering the dies is systematically described in this article. The work focuses on the effect of blend composition, feed frame parameters (blade speed and residence time), and rotary die disc parameters (die disc speed and die diameter) on powder overlubrication, the flow properties of the blends, and the resulting tablet properties (tablet hardness and dissolution). For lubricated blends, the feed frame had a large impact on powder hydrophobicity and powder flow properties. Hydrophobicity results demonstrated that the shear applied by the feed frame caused overlubrication of the powders, which improved the powder flow properties but affected adversely the tablet hardness and dissolution. © 2011 American Institute of Chemical Engineers AIChE J, 2012

Previous experimental and computational work has demonstrated that the geometry of the impeller has a significant effect on the mixing performance of continuous powder mixers. In particular, different blade patterns using either all blades pushing the powder forward (‘forward pattern’) are less effective than patterns where some of the blades push the powder backwards (‘alternate pattern’). In this article, we use Discrete Element Method to investigate this issue, as well as to examine whether batch mixers can be used to estimate the cross-sectional mixing rate of continuous mixing process. Mixing and flow of particles are examined in simple geometries consisting of a cylindrical cross-section agitated by impellers. In these geometries, blades are designed ‘forward’ and ‘alternate’ to study different blade patterns. Periodic boundary conditions are used to approximate an idealised cross-section of a continuous mixer. In addition, to examine whether an experimentally realisable batch system could be used to validate these cross-sectional simulations, we examine mixing in geometries with solid end walls. Performance in these elemental systems is compared qualitatively in terms of radially averaged velocity vectors, and quantitatively in terms of the mixing rate and the fill level profile. Results show that end walls and symmetric blade pairs of the ‘alternate’ blade patterns lead to faster mixing. Non-symmetric elemental systems display both increased fill level uniformity and similar mixing performance for both blade patterns. Results demonstrate the need for caution when attempting to use batch mixing information to estimate the cross-sectional mixing rate of continuous powder mixers.

A method comparison study was undertaken to characterize the effects of the formulation composition on the flow and shear properties of pharmaceutically relevant powders. Fourteen blends with various concentrations of an active ingredient, magnesium stearate as a lubricant and silica as a glidant were prepared. These blends were characterized with two very different techniques: the gravitational displacement rheometer (GDR), and a rotational shear cell. The values of GDR flow index were compared to the values of principal stresses and cohesion obtained with the shear cell. These measurements are different in that the GDR operates in the gravity-driven flow regime while the shear cell utilizes an imposed normal force, which leads to a much more pronounced powder consolidation. In spite of these differences, the study demonstrated a significant correlation between the two methods, although some discrepancies were observed due to differences in the consolidation state for each technique. This observation was confirmed by measuring compressibility of these formulations with the FT4 Powder Rheometer; the presence of a cohesive component influenced the values of the shear cell principal stresses (and therefore, the shear cell flow factor) in a non-linear manner, contributing to the discrepancies in the correlation between the flow factor and the GDR flow index. Conversely, the correlation between the shear cell cohesion parameter and the GDR flow index was significantly better, as both indices test the materials at similar degree of consolidation. The study examined the limits and the ranges of applicability of each technique, offered recommendations on applications, where the use of each method was more reliable, as well as provided reasons for the methods' failure.

Computational methods using mechanistic modeling with a specific application in the area of continuous blending are presented. These methods complement experimental designs and aim to reduce the amount of time, effort, and material required to characterize a device or a process. The discrete element methodis applied to the specific case of continuous mixing using two approaches. The first approach models an entire blender and studies the impact of changing impeller speed on residence time distribution (RTD), number of blade passes experienced by the powder, and mean centered variance of the particle residence time. The mean centered variance and the number of blade passes exhibit a maximum with increasing impeller speed, indicating optimal mixing behavior at an intermediate speed range. The second DEM approach utilizes a periodic slice of the full blender in order to explore in detail the effect of speed, fill level, and cohesion on mixing performance. The results show that the transverse mixing rates were generally higher than the corresponding axial mixing rates. Transverse diffusion was highest for large impeller speeds at larger fills, and axial diffusion was highest for large impeller speeds at smaller fills. Variations in particle‐ particle cohesion within the investigated range do not affect diffusion values significantly. The axial diffusion coefficient can be reasonably predicted from the residence time distribution using the 1D advection-diffusion model.

An experimental investigation was carried out to study the mixing performance and flow behavior in a continuous powder mixer for a typical pharmaceutical mixture. Blender performance, characterized by the relative standard deviation (RSD) of composition of blend samples taken at the blender discharge and by the variance reduction ratio (VRR) of the blender, was measured as a function of impeller rotation rate, flow rate and blade configuration. The flow behavior in the continuous mixer was characterized using the residence time distribution (RTD) and powder hold-up measurements. To quantify the strain applied to the powder in the blender, the number of blade passes experienced by the powder in the blender was calculated using the residence time measurements. The relationship between different experimental parameters and mean residence time and mean centered variance was examined. The mixing performance was largely dominated by the material properties of the mixture, which had a larger effect than the ingredient flow rate variability contributed by the feeders. Holdup was strongly dependent on impeller rotation rate; as impeller rotation rate increased, holdup (and therefore, residence time) decreased sharply. As a result, intermediate rotation rates showed the best mixing performance. Blade configuration affected performance as well; blade patterns where some of the blades push the powder backwards improved the mixing performance.

Continuous powder mixing is an efficient alternative in high volume manufacturing of powder-based products. A new method is communicated in this paper to determine the effects of different operating conditions and mixer configurations on mixing performance. The main idea of the proposed methodology is the utilization of experimental residence time distribution (RTD) measurements to (a) determine the contributions of feeding variability, powder segregation and RTD variability on output composition variance and (b) develop a predictive model of the output variance of a continuous mixer. The method is illustrated using experimental data for mixing acetaminophen and lactose in convective continuous mixer.

In this paper, we examine the movement of particles within a continuous powder mixer using PEPT (Positron Emission Particle Tracking). The benefit of the approach is that the particle movement along the vessel can be measured non-invasively. The effect of impeller rotation rate, powder flow rate, and powder cohesion on the particle trajectory, dispersive axial transport coefficient, and residence time is examined. Increase in the impeller rotation rate decreased the residence time, increased the axial dispersion coefficient, and resulted in longer total path length. Effect of flow rate was different at two different rotation rates. At lower rotation rate, increase in flow rate increased the residence time, decreased the axial dispersion, and resulted in longer total path length. At higher rotation rate, increase in flow rate decreased the residence time, increased the total path length and showed a complex dependence on the axial dispersion coefficient. Increasing cohesion (measured using the flow index, dilation, and the Hausner ratio) did not affect the axial dispersion coefficient significantly, but had significant effects on the total particle path length traveled and the residence time. These results, relevant to pharmaceutical powders, provide better physical understanding of the influence of operating parameters on the flow behavior in the continuous mixer. In addition, one of the main obstacles of modeling continuous mixing of particles is to know the appropriate values for the modeling parameters as well as validate modeling approaches. One example is the dispersion coefficient which leads to an analytical solution for the axial dispersion model of a continuous blending process.

A non-destructive NIR spectroscopic method was used to acquire on-line spectra of a continuous mixing process, and evaluate the performance of this novel system. Partial least squares (PLS) calibration models were developed and used for real-time determination of active ingredient concentration on the blends produced with a continuous mixer. The NIR method was developed for concentrations ranging from 0 to 15% (w/w) of acetaminophen (APAP), the active pharmaceutical ingredient used in the experiments. The calibration model's overall accuracy was 0.41% (w/w), and estimated through the root mean square error of cross validation (RMSECV) for samples predicted using leave-class-out cross validation. In this cross validation, each concentration was defined as a class, and when a sample of a specific concentration was predicted all samples of that concentration were left out of the calibration model. The precision of the calibration model was also estimated at various concentration levels, to facilitate the differentiation between the variation in drug concentration due to the analytical method's measurement uncertainty and the variation in the drug distribution throughout the powder blend. The results obtained are very promising since in three of the five powder mixes, the variation in the drug concentration in the powder blends was similar to that of the analytical method indicating a high degree of blend homogeneity

Objective: The purpose of this study was to investigate the effect of mechanical shear on hydrophobicity of pharmaceutical powder blends as a function of composition and particle size, and to determine the impact on drug release from tablets. Methods: Four powder formulations were subjected to three different shear strain conditions (40 rev, 160 rev, and 640 rev) in a controlled shear environment operating at a shear rate of 80 rpm. A total of 12 blends were tested for hydrophobicity. Subsequently, sheared blends were compressed into tablets at 8 kN and 12 kN in a rotary tablet press. During tablet compression, powder samples were collected after the feed frame and their hydrophobicity was again measured. Results: Results indicated that increase in shear strain could significantly increase hydrophobicity, predominantly as an interacting function of blend composition. Blends with both colloidal silica and magnesium stearate (MgSt) were found to show higher hydrophobicity with shear than other blends. Additional shear applied by the tablet press feed frame was found to change the powder hydrophobicity only in the absence of MgSt. Conclusions: Studies showed that the drug release rates dropped with shear more for the blends with both colloidal silica and MgSt than the other blends. Furthermore, the rate of drug release dropped with a decrease in particle size of the main excipient. Surprisingly, the relationship between the relative increase in hydrophobicity and a corresponding drop in the drug release rate was not found when either MgSt or colloidal silica was mixed alone in the blends.

Introduction The identification and graphical representation of process design space are critical in locating not only feasible but also optimum operating variable ranges and design configurations. In this work, the mapping of the design space of pharmaceutical processes is achieved using the ideas of process operability and flexibility under uncertainty.

Variance reduction ratio (VRR) is generally considered as an important index in characterizing the continuous powder mixing. Although the capacity of the mixer to smooth out feeder fluctuations can be expressed by the VRR, few studies are performed quantitatively in this area. The feeder effects are investigated on the solid mixer through the Fourier series analysis. The VRR is deduced quantitatively using the Fourier series of the feed rate variability and the residence time distribution (RTD), which facilitate the explicit decomposition of VRR into intensities of different frequency components. It provides a novel model to determine whether the integrated feeder-mixer system satisfies specific solid mixing performance criteria, and provides guidelines of system improvement. © 2010 American Institute of Chemical Engineers AIChE J, 2011

This paper studies systematically the effect of a feed frames, a device used in rotary tablet presses to drive the powders into compression dies, on the properties of the powders entering the tablet press dies. The work focused on the effect of blend composition, feed frame parameters (blade speed, residence time), and rotary die disc parameters (die disc speed, die diameter) on the flow pattern, uniformity of die filling, applied shear, and the flow properties of pharmaceutical blends. The flow pattern suggests a stratified filling of the dies and therefore, non-uniform properties of the tablets. The amount of powder entering the dies depended on blend flow properties, feed frame speed, and dies disc speed. In addition, blend properties changed significantly as the powder flowed through the feed frame. The flowability of lubricated blends improved significantly as the number of feed frame blade passes increased, decreasing in turn the RSD of the die filling weight.

This communication empirically correlates flow in two systems; an instrumented rotating drum (GDR) and a set of bench scale hoppers. A flow index obtained from measurements in the GDR is directly correlated to the flow through hoppers, providing a predictive method for hopper design and a convenient experimental test for screening materials and determining their suitability for specific hopper systems. Simulations were performed to understand the dynamics of flow in hoppers by using the same flow parameters in hoppers and rotating cylinders. Simulations showed that as cohesion increased it becomes harder for the particles to flow through the hoppers, in good agreement with the experiments. The effect of hopper angle also yields similar findings to experiments for Avicel, K =60, where the powder does not flow through the 45° hopper but flows well in a 75° hopper. Simulations were also used to calculate the normal forces on the walls of the hopper and the wall pressure distributions in both hoppers. As depth increases, the wall pressure increases for all cases. Finally, the simulations also helped understand the different flow behaviors (funnel and mass flow) that take place in a hopper. The simulated dynamics of flow in the rotating drum and in the hopper correlate very closely to experiments, indicating that the model cohesion parameters are, as desirable, material-specific but independent of geometry.

Electrostatic forces can significantly alter flow properties of granular materials and can adversely affect many industrial particulate processes in unpredictable ways. We investigate here the effect of higher order dielectric electrostatic forces, which are created by non-uniform electric fields, on the agglomeration, adhesion and flow of several granular materials, including pharmaceutical powders. We find that materials can adhere consistently and reproducibly to a metallic rod in a sufficiently strong electric field, which can be produced by either a DC source or tribocharged surfaces. These results provide a simple way to characterize material susceptibility to electrostatic agglomeration. The effect of applied non-uniform fields on the flow of grains falling from a cylindrical hopper is studied and found to significantly reduce the particle flow rate. The effects of humidity, particle size, coatings, and the grounding of equipment are also tested. Finally, contrary to common intuition, we find that grounding a metallic surface can actually exacerbate particle adhesion and agglomeration.

Lack of knowledge of the first principles, that describe the behavior of processed particulate mixtures, has created significant attention to data-driven models for characterizing the performance of pharmaceutical processes—which are often treated as black—box operations. Uncertainty contained in the experimental data sets, however, can decrease the quality of the produced predictive models. In this work, the effect of missing and noisy data on the predictive capability of surrogate modeling methodologies such as Kriging and Response Surface Method (RSM) is evaluated. The key areas that affect the final error of prediction and the computational efficiency of the algorithm were found to be: (a) the method used to assign initial estimate values to the missing elements and (b) the iterative procedure used to further improve these initial estimates. The proposed approach includes the combination of the most appropriate initialization technique and the Expectation Maximization Principal Component Analysis algorithm to impute missing elements and minimize noise. Comparative analysis of the use of different initial imputation techniques such as mean, matching procedure, and a Kriging-based approach proves that the two former used approaches give more accurate, “warm-start” estimates of the missing data points that can significantly reduce computational time requirements. Experimental data from two case studies of different unit operations of the pharmaceutical powder tablet production process (feeding and mixing) are used as examples to illustrate the performance of the proposed methodology. Results show that by introducing an extra imputation step, the pseudo complete data sets created, produce very accurate predictive responses, whereas discarding incomplete observations leads to loss of valuable information and distortion of the predictive response. Results are also given for different percentages of missing data and different missing patterns. © 2010 American Institute of Chemical Engineers AIChE J, 2010

Heat transport through flowing particulate materials is an essential component of modern technologies such as heterogeneous catalytic reactors, high performance cryogenic insulation, construction material, and powder metallurgy. In catalyst manufacturing, heat transfer through granular media occurs in the drying and calcination stages. In this paper, we describe the use of experiments and discrete element methods to examine flow, mixing, and mass and heat transport in rotary calciners. Alumina powder (200 μm) and cylindrical silica pellets (2 mm diameter and 3 mm long), which are common support materials for catalysts, are used in our experiments. A parametric study was conducted by varying the material properties of granular material, and rotational speed of the calciner. We use the discrete element model to simulate flow, mixing, and heat transport in granular flow systems in rotary calciners. Simulations and experiments show that the rotation speed has minimal impact on heat transfer. As expected, the material with higher thermal conductivity (alumina) warms up faster in experiments and simulations. Various baffle configurations (rectangular and L-shaped flights) in the calciner and their effect on the flow and heat transfer of granular material are simulated. The average wall-particles heat transfer coefficient of the granular system is also estimated from the experimental findings.

This paper discusses how the hydrophobicity of lubricated pharmaceutical formulations is affected by process variables such as shear rate and strain. Hydrophobicity is a critical property that affects the dissolution of powder formulations, tablets and capsules as well as the performance of tablet coating and granulation operations. In this paper, hydrophobicity is measured using a modified Washburn method. Results show that, in the absence of lubricant, the hydrophobicity of powders does not change substantially as a function of shear rate or strain. However, when magnesium stearate is present (concentrations studied here range between 0.5% and 2%), hydrophobicity increases as a function of strain, shear rate and lubricant concentration. Observed changes range over several orders of magnitude, readily explaining common “overlubrication” observations of delayed drug dissolution.

The problem of the Andersen cascade impactor data inversion is studied using the maximum entropy method to recover the particle size distribution function from a set of measured masses collected on the impactor stages. The classical entropy maximization and its modification to accommodate data errors are reviewed using the techniques of convex optimization. Numerical experiments were performed to test the method on several unimodal and multimodal distributions commonly observed in practice. The Maximum Entropy Method for data inversion produced excellent agreement between the recovered and test Particle Size Distribution Functions without the need for a priori assumptions.

Powder feeding is a fundamental unit operation in the pharmaceutical industry. For the cases in which first-principle process models are unknown, such as when new powder mixture feeding operations are being evaluated, or no longer accurately describe current operating behavior, surrogate model-based approaches can be employed in order to quantify input–output behavior. In this work, two such metamodeling techniques—kriging and response surface methods—are used to predict a loss-in-weight feeder unit’s flow variability in terms of unit flowability and feed rate. Based on a comparison of predicted with experimental values, an iteratively constructed kriging model is found to more accurately capture the feeder system behavior compared with the response surface methodology. Although feeders are used as a case study in this paper, the kriging methodology is general to address other processes where first-principle models are not available.

The dynamics involved in powder mixing remains a topic of interest for many researchers; however the theory still remains underdeveloped. Most of the mixers are still designed and scaled up on empirical basis. In many industries, including pharmaceutical, the majority of blending is performed using “tumbling mixers”. Tumbling mixers are hollow containers which are partially loaded with materials and rotated for some number of revolutions. Some common examples include horizontal drum mixers, v-blenders, double cone blenders and bin blenders. In all these mixers while homogenization in the direction of rotation is fast, mediated by a convective mixing process, mixing in the horizontal (axial) direction, driven by a dispersive process, is often much slower. In this paper, we experimentally investigate a new tumbling mixer that rotates with respect to two axes: a horizontal axis (tumbling motion), and a central symmetry axis (spinning motion). A detailed study is conducted on mixing performance of powders and the effect of critical fundamental parameters including blender geometry, speed, fill level, presence of baffles, loading pattern, and axis of rotation. In this work Acetaminophen is used as the active pharmaceutical ingredient and the formulation contains commonly used excipients such as Avicel and Lactose. The mixing efficiency is characterized by extracting samples after pre-determined number of revolutions, and analyzing them using Near Infrared Spectroscopy to determine compositional distribution. Results show the importance of process variables including the axis of rotation on homogeneity of powder blends.

In this paper we examine the effect of rotation rate, mixing angle, and cohesion on the powder residence time and the content uniformity of the blend exiting from two continuous powder mixers. In addition, differences in mixing performance between the two blenders are examined. Analysis of variance is used to determine significance of main effects and their interactions. The results show that the effect of powder cohesion is scale-dependent, having a significant effect in the larger mixer. The overall rotation rate was the least influential parameter in terms of content uniformity. The residence time is significantly affected by both rotation rate and mixing angle.

A single jet model is described and used to analyze the performance of an 8-stage Andersen cascade impactor (ACI). The two-dimensional axisymmetric jet flow field was calculated numerically by solving the Navier-Stokes equations and particle trajectories were then analyzed to obtain the collection efficiency curves at different flow rates. The resulting efficiency curves and corresponding cutoff diameters were compared and found to be in good agreement with the available experimental data. We also examined the effect of gravity on impactor performance and discuss the limitations of the single jet model.

We study the relation between flow and electrical conductivity of multiple formulations of pharmaceutical powder blends. Ten formulations were tested, consisting of two excipient sets, two active preparations, and a variety of food-grade additives including magnesium stearate (MgSt), and ionic and conductive materials such as ascorbic acid, talc, sodium carbonate, colloidal silica and TiO2. Electrical impedance, flow index and dilation were independently measured for all of the blends, and a strong correlation was found between every pair of these three properties. The relation between flow and dilation has been observed before; we find for the first time that there is an exponential relationship between flow index or dilation and impedance. This indicates that cohesive powder behavior depends on powder electrical properties, raising the questions of whether additives such as MgSt affect friction and conductivity per se and what mechanism and phenomenon links cohesion and conductivity.

In this paper we focus on the effect of cohesion and compression speed on the outcome of the compression process for both monodisperse and polydisperse granular systems. A three dimensional discrete element model (DEM) which incorporates static and dynamic friction is used in this study to simulate die filling, and the compaction and decompaction of cohesive granular system in a confined cylindrical die similar to those used in a commercial tablet press driving the pre-compressive stage. The magnitude of the cohesive force is represented in terms of a parameter K = Fcohes/mg, where K is called the bond number and is the measure of cohesiveness. Force displacement curves are used to characterize the compression and deformation properties of the materials and are obtained by measuring the force on the upper punch and the corresponding displacements in the die. Results show that a considerable more energy is needed to compress the cohesive material as compared to free flowing materials. It is found that the time required to fill the die strongly depends on the cohesion of the material. The energy for the tableting process is directly proportional to the upper punch speed.

This paper examines the effect of three protocols with several units and blender parameters on the mitigation of active pharmaceutical ingredient (API) agglomeration in solid formulations. The three protocols either preblend API with a portion of excipients in a high shear unit followed by dilution in a large blender or prepare the entire blend in a single blender followed by milling. In general, the three protocols yield blends with statistically similar API concentration variance and deagglomeration. The scale-up of the three protocols leads to more extensive API deagglomeration, which suggests that blender parameters still influence the degree of API deagglomeration, even when high shear units are present in the protocol. Lower blender fill levels and larger blenders lead to blends with fewer API agglomerates. Regarding the use of blender internals, results show that baffles have no substantial effect on API agglomeration. The inclusion of a moving internal (i.e., impeller) in a bin blender may not alw...

In this study, a quality by design approach was applied to the design and scale-up of a batch mixing process. Mixtures of acetaminophen and lactose were sampled at different mixing times using a groove sampler. Samples were subsequently analyzed using NIR reflection spectroscopy. The effects of four processing parameters on the empirical mixing rate in a bin blender were examined. Blender rotation rate (two levels), powder fill level (two levels), powder cohesion (two levels), and blender size (three levels) represent the four parameters studied. Blender geometry and blender loading method were treated as constant parameters. Statistical analysis was used to assess the impact each parameter had on the mixing rate. Blender size (p = 0.02), powder cohesion (p = 0.05), and rotation rate (p = 0.07) all significantly affected the mixing rate. The least significant parameter was the vessel fill level (p = 0.18), indicating mixing performance is not strongly affected by fill level, given the range studied.

This paper investigates the effects of processing and equipment parameters of a ribbon blender (i.e. loading method of lubricant, fill level, blade speed and blade design) on magnesium stearate homogeneity. A core sampling technique is used to obtain at least one hundred samples per sampling event, which are extracted throughout the blender and yield a thorough characterization of the entire bed. The results presented here can be used as a guideline to develop appropriate blending processes and characterization protocols for ribbon blenders.

In this paper, the development of a compartment model to simulate mixing within a continuous blender is reported. The main benefit of the method is that it can generate extensive modeling predictions in very short computational time. The model can also be used to explore the effect of sampling parameters on estimated mixing performance, a topic that has been central to pharmaceutical manufacturing for the past 15 years and that remains a central issue in the PAT initiative. However, this method requires more input than conventional particle dynamics methods. Thus, we investigate the effects of modeling parameters on mixing performance to develop general guidance needed to adapt this modeling framework to any continuous process. An experimental technique based on longitudinal sampling is used to examine the content uniformity of the blend along the continuous mixer. The model compares favorably with continuous mixing experiments, capture the effects of feeding rate variability, active product ingredient concentration, and blender processing angle, while effectively capturing and making explicit the effect of sampling parameters such as number of samples and sample size. The modeling approach provides a convenient tool for process design.

Laminar flow structure and mixing patterns in stirred tanks with eccentrically located impellers is examined using tracer visualization techniques and laser induced fluorescence (LIF). The displacement of the shaft from the centerline of the tank has a remarkable effect on the manifold structure of the flow: segregated regions of regular motion observed in concentric systems are destroyed, impeller mid-plane separatrices are eliminated, and the axial circulation is greatly improved even in systems agitated even by radial impellers. On examine par des techniques de visualisation par traceurs et la fluorescence laser (LIF) la structure de l'ecoulement laminaire et les modeles de melange dans des reservoirs agites munis de turbines decentrees. Le deplacement de l'axe de l'agitateur par rapport a l'axe central du reservoir a un effet notable sur la structure d'ensemble de l'ecoulement : les regions isolees a mouvement regulier observees dans les systemes concentriques sont detruites, les separatrices au plan median de la turbine sont supprimees et la circulation axiale est grandement amelioree meme dans des systemes agites par des turbines radiales.

The Process Analytical Technology (PAT) initiative has encouraged the development of new technology to improve upon the current manufacturing paradigm. As a result substantial attention has recently focused on continuous processing due to the ability to control disturbances online, avoiding the loss of processing materials and enabling effective process scale-up. In this paper, a pharmaceutical formulation is blended using a continuous flow “high shear” mixer utilizing different operating and design parameters. The mixing efficiency is characterized by extracting samples at the discharge of the blender, and analyzing them using Near Infrared Spectroscopy to determine compositional distribution. Operational conditions such as the inclination angle of the mixer and impeller rotation rate were investigated and showed to affect the mean residence time. The effects of mixer angle, agitation speed, number of blades, blade angle, number of passes through the mixer on the mixing performance of a powder continuous convective mixer are also examined and shown to affect mixing performance whereas the cohesive properties of the material did not significantly affect the mixing operation.

Accurate prediction of the time required to heat up granular materials to a target temperature is crucial for several processes. However, we do not have quantitative models to predict the average temperature or the temperature distribution of the particles. Here, we computationally investigate the scaling of heat transfer in granular flows in rotating drums. Based on our simulations, which include a wide range of system and material properties, we identify the appropriate characteristic time that is used to derive equations that predict the particles' average temperature and the particles' temperature distribution.

This paper communicates a methodology that uses experimental data and a new statistical method that uses a simulation to reconstruct the composition distribution of a powder blend containing drug agglomerates. The reconstructed distribution can be used subsequently to optimize sampling protocols, compute operating characteristic curves, and estimate process capability for blends containing agglomerates. A blend containing a cohesive API and an excipient is prepared in a blender. Although the API is sieved (mesh 100 μm) prior to its addition to the blender, API agglomerates are found in samples after blending. These agglomerates have either survived the mixing operation, or been created by the operation. The population of sample concentrations is used to estimate the parameters of a statistical model that combines multiple distribution functions to describe the binary mixture with the minor component partially agglomerated. The local concentration for the non-agglomerated portion of the minor component throughout the entire blend is estimated using a Gaussian distribution with parameters $ \overline{x} $ (mode) and s (standard deviation) that are typically well established. The function and the parameters that describe the population of agglomerates of the minor component are estimated using (typically sparse) experimental data. The optimization of the parameters describing the agglomerate population is performed using a simulation of the powder bed in an iterative manner. The simulated blend, built using the statistical model, is extensively examined and a distribution of RSD estimates is obtained. The iterative process for seeking the parameters that describe the agglomerate population converges when the mode of the distribution of RSD estimates matches the experimental RSD value, which is the best available estimate of the blend homogeneity. This methodology is applied to populations of samples for different binary formulations prepared in different blenders. In all cases, after iterating to adjust the parameters of the model, there is an agreement between the experimental and the simulated population of samples with a confidence of, at least, 95%. The methodology described here is an effective tool to create an accurate representation of agglomerated blends based on limited sampling results.

This communication correlates two empirical methods; flow in a rotating drum with flow in bench scale hoppers. The flow characteristics of 13 cohesive granular materials were studied in the gravitational displacement Rheometer (GDR) and compared to flow in hoppers of varying angle and discharge diameter at fixed temperature and moisture conditions. The GDR proved to be an effective and convenient tool for examining flow properties of pharmaceutical materials, both pure and mixtures. A flow Index obtained from GDR measurements is directly correlated to the flow through hoppers, providing a predictive method for hopper design and a convenient experimental test for screening materials and determining their suitability for specific hopper systems. Analysis of industrial scale hoppers will be addressed in future communication.

Statistical methods to assess similarity of dissolution profiles are introduced. Sixteen groups of dissolution profiles from a full factorial design were used to demonstrate implementation details. Variables in the design include drug strength, tablet stability time, and dissolution testing condition. The 16 groups were considered similar when compared using the similarity factor f2 (f2 > 50). However, multivariate ANOVA (MANOVA) repeated measures suggested statistical differences. A modified principal component analysis (PCA) was used to describe the dissolution curves in terms of level and shape. The advantage of the modified PCA approach is that the calculated shape principal components will not be confounded by level effect. Effect size test using omega-squared was also used for dissolution comparisons. Effects indicated by omega-squared are independent of sample size and are a necessary supplement to p value reported from the MANOVA table. Methods to compare multiple groups show that product strength and dissolution testing condition had significant effects on both level and shape. For pairwise analysis, a post-hoc analysis using Tukey's method categorized three similar groups, and was consistent with level-shape analysis. All these methods provide valuable information that is missed using f2 method alone to compare average profiles. The improved statistical analysis approach introduced here enables one to better ascertain both statistical significance and clinical relevance, supporting more objective regulatory decisions.

This study describes how the strain on formulation components affects dissolution and how near infrared spectroscopy can be used to predict dissolution. Strain (exposure to shear stress) applied during powder mixing affects the interaction between formulation components. Particles experience shear strain when they move relative to each other in a process affecting the properties of the final product. This stress affects the dissolution of oral solid dosages forms. However, dissolution testing destroys the entire tablet, making it impossible to further evaluate tablet properties when an out of specification result is obtained. Thus, a nondestructive technique such as near infrared spectroscopy is desirable to predict dissolution. The aim of this study was to predict dissolution on tablets with different levels of strain (shear) using near infrared spectroscopy in combination with multivariate data analysis. Shear was induced using a modified Couette cell on the powder mixture and tablets from these mixtures were produced using a tablet press emulator. Tablets produced with different strain levels were measured using near infrared spectroscopy. Spectra were obtained in diffuse reflectance mode and pretreated with baseline correction to maintain the physical and chemical information of the tablets. Dissolution profiles were obtained using USP Apparatus 2 as a reference method. Principal component analysis was used to study the sources of variation in the spectra obtained. Partial least squares 2 was used to predict dissolution on tablets with different levels of strain.

Mammalian cells are extremely sensitive to mechanical stress. Ideally, a bioreactor design for mammalian cell culture should assure adequate mixing at low mechanical stress. This paper focuses on the mixing characterization of a novel stirred tank bioreactor configuration, proposed for the culture of mammalian cells, based on the principle of displacing the agitation shaft to an eccentric position and replacing the impellers normally used for mammalian cell culture with a disc impeller with no blades. Experiments in a 1.0 L prototype are conducted to study flow patterns using UV light visualization techniques. Three different impeller shaft positions are tested E = 0.0 , 0.21 , and 0.42. For the purposes of this work, eccentricity ( E ) is defined as the distance between the shaft and the vertical centerline of the tank/tank radius. The mixing performance of two different impeller disc diameters (3.0 and 5.0 cm) are compared. Experimental results show that adequate mixing conditions are achieved at very low Re numbers for some of the eccentric cases considered. Computations are used to illustrate mixing improvement caused by eccentricity, and to validate the existence of globally chaotic conditions for the eccentric cases tested experimentally.

This paper describes a new method to quantitatively measure the flow characteristics of unconfined cohesive powders in a rotating drum. Cohesion plays an important role, affecting flow properties/characteristics, mixing rates, and segregation tendencies. The method relies on measuring the change in center of mass of the powder bed as it avalanches in the vessel, using a load cell that is sampled continuously. Filtering and analysis of the signal is done using Fast-Fourier transform into the frequency domain, where noise is eliminated using signal processing methods. The filtered data is transformed back to the time domain by using an inverse Fast-Fourier transform to give quantitative information on the powder flow characteristics. In order to understand the nature of the forces controlling powder flow behavior, a computational model was developed to estimate the relationship between inter-particle cohesive strength and experimental measurements. A "flow index" generated by the method correlates well with the degree of bed expansion (dynamic dilation) of the cohesive powders. The flow index also predicts the dynamics of flow through hoppers. As the flow index increases it becomes increasingly difficult for the powder to flow through the hoppers.

Heat transfer in particulate materials affects a wide variety of applications ranging from multi-phase reactors to kilns and calciners. In catalyst manufacturing, heat transfer through granular media (catalyst) occurs in the impregnation and calcinations stages. We use the discrete element model to simulate flow, mixing, and heat transport in granular flow systems in rotary calciners and impregnators. Granular flow and heat transport properties are taken into account in order to develop a fundamental understanding of their effect on dryer and calcination performance. Simulations have shown that as rotation speed decreases, both heat transfer and temperature uniformity of the granular bed for both calciner and impregnator increase. Depending on baffle size, baffles can either increase or decrease heat transfer in double cone impregnators. Granular cohesion does not affect heat transfer in the range of values examined.

Powder mixing has been the subject of substantial research due to its importance in a variety of industrial sectors, including pharmaceuticals, food, and polymer manufacturing. Although a number of different models have been proposed in the literature, most of them are either empirical or require computationally intensive calculations that make them difficult to implement for realistic systems. The aim of this paper is to develop a simplified framework, based on compartment modeling that efficiently and accurately captures the system behavior. Using the V-blender as a model system, the compartment modeling approach was used to illustrate the effects of vessel loading on mixing as well as the impact of sampling methods on the accuracy of mixing characterization.

Rotary drums are used as mixers, dryers, kilns and granulators. In all of these systems, powder cohesion deeply affects mixing and segregation, and it is critical in process scale up. In this paper, we focus on the effect of cohesion in mixing and size segregation of binary mixtures of uniform and non-uniform sizes in a partially filled rotating drum. The cohesive force between particle is simulated using a square-well potential and the numerical model is used to characterize flow and mixing properties. The model is validated by comparison to experimental images. Results show a time-dependent spatial distribution of cohesive powder that depends on the magnitude of cohesion and friction. In uniform binary systems, as cohesion increases, the rate of mixing first increases and then decreases, however for the case of non-uniform binary systems, we observe different mixing patterns depending on the relative magnitude of forces acting between particles of same/different sizes. Unlike free flowing material, for cohesive mixtures, a higher rotation speed is found to enhance mixing performance.