外文翻译初稿

锅涂层过程的喷雾质量的测量

Wei Chen & Shih-Ying Chang & San Kiang &

William Early & Srinivasa Paruchuri & Divyakant Desai

Published online: 7 March 2008

# International Society for Pharmaceutical Engineering 2008

Abstract In this study, we describe the use of an innovative imaging system to measure and control the effect of nozzle operating parameters on the characteristics of a spray. These characteristics, including spray pattern, droplet size distribution, and droplet velocity, define the quality of the spray. They can have significant impact on the efficiency of the pan coating process and the quality of the coat. Suspensions of different composition were used in this study, and the authors demonstrated that the spray characteristics can be controlled with this approach. The main conclusions from this study were:

摘要：在这项研究中，我们描述了使用一种创新性的成像系统去测量和控制喷嘴操作参数对喷雾的特征的影响。这些特征，包括喷雾样式，液滴大小分布，液滴速度，对喷雾质量的定义。它们可以对锅涂层工艺的效率和涂层的质量产生重大影响。在这项研究中混悬剂使用不同成分组成，作者证明了这个喷雾特征可以用这种方法进行控制。本研究的主要结论是：

1. The AA/Spray (atomization air/spray rate) and AA/PA (atomization air/pattern air) mass flow ratios were the key parameters that affect spray characteristics. Although viscosity can impact the spray, there was minimal impact within the viscosity range tested in this study.

1.这个AA /喷涂(雾化空气/喷涂率)和AA / PA(雾化空气/空气模式)质量流率是影响喷雾特证的关键参数。虽然涂料粘度可以影响喷雾，本研究测试的是粘度范围内最小的影响。

2. With proper selection of the AA/Spray and AA/PA mass flow ratios, it was possible to generate sprays with consistent spatial distributions of volume flux with minimal variations of mean droplet size over the range of coating suspensions and spray rates studied.

2. 正确的选择AA /喷雾和AA / PA质量流量比率，可以研究导致喷涂与体积通量的立体分布

相一致的喷雾平均液滴大小的变化对涂料悬浮体的范围和喷雾率的影响。

Spray characterization can be a powerful tool for exploring and establishing the design space of nozzles operation in the pan coating process. When scaling or transferring a spray coating process, the focus should be on maintaining consistent spray qualities rather than limiting nozzle operating

parameters to a range. This approach embraces the FDA concept of process analytical technology (PAT) and design space (FDA, Guidance for industry PAT—A framework for innovative pharmaceutical development, manufacturing, and quality assurance, 2004) for science based operation flexibility.

在锅涂层过程中，喷雾特性可以成为一个强大的工具来探索和建立以及设计喷嘴操作的距离。当扩展或转移一个喷涂过程中，应该注重保持一致的喷雾质量而不是限制喷嘴操作参数范围。这种方法包含了FDA的概念过程分析技术(PAT)和设计距离(FDA，PAT产业的指导一在一个体系下创新药品开发，生产，质量保证，2004)具有科学基础的操作的灵活性。

Keywords Particle tracking velocimetry . Droplet size distribution . Spray pattern . Droplet velocity distribution . Spray coating . Process analytical technology (PAT) . Spray quality . Coating uniformity

关键词：粒子跟踪测速；液滴尺寸分布；喷雾样式；液滴速度分布；喷涂；过程分析技术(PAT)；喷雾质量；涂层均匀性

Introduction

Spray coating is a unit operation widely used in solid dosage manufacturing. Among different coating equipment, a rotating pan is commonly used for coating tablets [2]. When the coated film contains the active ingredient, tablet to tablet content uniformity of the coat becomes a critical quality attribute [3]. While other process factors such as spray rate and inlet air temperature [4] have been widely studied, the measurement of spray and tablet mixing in a coating pan have not drawn much attention in the literature. Our research has shown that USA these two qualities can have significant impact on the content uniformity of active coated tablets (unpublished results).

前言

喷涂是一个单元操作被广泛应用于固体剂量制造业中。在不同涂层设备中，一个旋转盘通常用于平板电脑的涂层[2]。涂膜包含活性成分时，平板电脑的平板含量均匀度成为质量的关键属性[3]。然而其他过程中的因素比如喷雾率和进气温度[4]已经被广泛研究，在文学中，在涂层容器里测量喷雾和平板混合没有引起关注。我们的研究表明在美国这两个特质对平板电脑的含量涂布均匀度有显著影响 (结果未发表)。

The quality of the spray affects not only the uniformity of the coat but can directly impact the microstructure of the coated surface, and therefore the dissolution profile of the coated tablet [5]. It also has process efficiency implications—a relatively high amount of fine droplets can cause a corresponding loss of materials due to spray drying. Process scaling and technology transfer should therefore focus on maintaining consistent spray characteristics. Such consistency is especially important with different nozzle types, spray rates, and suspension viscosities. Air volume control is preferred over air pressure control for maintaining atomization and pattern air control and in calculation of mass flow ratios.

喷雾的质量不仅影响涂层的均匀性，而且可以直接影响微观结构涂层的表面，因此分解平板电脑涂层的侧面[5]。它也有流程效率的含义-由于喷雾干燥的材料，一种相对较高的细水滴会造成相应的损失。过程扩展和技术转让应重点保持喷雾特性的一致性。不同的喷嘴类型，喷雾率和悬浮液体粘度的一致性尤为重要。在质量流率的计算下风量控制优先于为了维持雾化而进行空气压力控制和样式控制。

Several studies relating pneumatic atomization of aqueous liquids have been published, but the proposed correlations between nozzle setup and droplet sizes are not consistent with each other [6]. Such inconsistencies might have been due to the different nozzle designs, use of air pressure for atomization air/pattern air (AA/PA), or simply caused by measurement error. Available correlations usually apply to a particular type of nozzle operating within a specified range. In the following sections, we describe the use of an imaging system to characterize nozzle performance and correlate these characteristics to the operating parameters.

与气动雾化水液体有关的几项研究已经出版，但该喷嘴设置和液滴之间的相关性大小并不相互一致[6]。这种不一致可能是由于不同的喷嘴设计，使用空气压力雾化空气/空气模式(AA / PA),或者仅仅是由测量误差引起的。可利用的相关性通常适用于一种特定类型的喷嘴操作并且要在指定的范围内。在下面几节中，我们叙述使用成像系统去描述喷嘴的使用性能和与操作参数相关联的一些特征。

Principles of Liquid Atomization and its Impacts on Spray Coating

Atomization in a pneumatic nozzle involves impacting bulk liquid with a high velocity gas. The high

velocity gas creates frictional forces over the liquid surface and causes the liquid to disintegrate into droplets. The process is influenced by liquid properties such as viscosity, surface tension, and density. It is also affected by the gas velocity and density. Several investigators have studied such droplet correlations from specific nozzle designs [6, 7].

在一个气动喷嘴中雾化包括影响大量液体伴随速度快的气体。在液体表面，高速气体产生摩擦力，使液体分解成液滴。在这个过程中受液体性质如粘度、表面张力、密度的影响。它也受到气流速度和密度的影响。一些调查人员在具体的喷嘴设计中研究过这样的液滴的相关性 [6、7]。

Due to the large velocity difference between the droplets and surrounding air, the droplets will have relatively high heat and mass transfer rates across their air interface even at ambient temperature. The volatiles in relatively smaller droplets vaporize more quickly, making these droplets susceptible to spray drying. Such a mechanism of spray drying could lead to the loss of coating materials. Several researchers also suggest that the droplets’ size distribution, combined with the drying conditions, can affect surface roughness and the uniformity of coated tablets [8].

由于液滴和周围的空气速度有较大的差异，水滴将会有相对较高的热度和较大的转化速率当与他们的空气相连甚至是在室温下。挥发物在相对较小的液滴下蒸发更快，使这些液滴容易被喷雾干燥。这样一个喷雾干燥机制可能导致涂层材料的损失。几个研究还表明，液滴的粒径分布和结合干燥条件，会影响表面粗糙度和平板电脑涂层的均匀性 [8]。

To achieve a smooth uniform coat with high efficiency, the ideal spray should have uniformly distributed droplets, narrow droplet size distribution, and fairly fast velocity toward target tablets. Such spray characteristics should be the key criteria when selecting nozzle operating parameters [7, 9].

为了实现高效率的、平稳的和统一的涂层，理想的液滴喷涂应均匀分布，液滴尺寸分布窄，以相对快的速度朝着目标平板电脑喷涂。这种喷雾特性的关键标准应该是选择喷嘴操作参数(7、9)。

Optical imaging is one of the most common approaches to spray characterization. Kim applied microscopic and sieve analysis for size determination [10]. Phase Doppler interferometry has also been used to measure both size and velocity in a calibrated range [11]. Chigier [12] has compiled a very comprehensive review of systems developed to characterize spray droplets.

光学成像是一种最常见的喷雾特性的方法。Kim运用微观和筛分析方法确定大小[10]。多普勒相位干涉法也被用于测量校准范围的大小和速度[11]。Chigier[12]已经完成了一个非常全面审查系统开发喷雾液滴的特点。

In this paper, we demonstrate the use of an innovative imaging system called SprayWatch (OSEIR Ltd. Tempere, Finland). The principle of measurement is based on fast imaging and a particle tracking velocimetry (PTV) algorithm. In the following sections, we will describe the experimental setup and tests that were performed to assess nozzle operating conditions. These conditions can affect spray cone angle, mean droplet size, velocity, and the spatial distributions of these properties. With an understanding of setup conditions, the authors controlled spray characteristics in a desired design space.

在本文中，我们演示了一个创新的成像系统称为SprayWatch OSEIR Ltd. Tempere, Finland)。测量原理是基于快速成像和粒子跟踪测速技术(PTV)算法。在下面几节中，我们将描述实验步骤和测试进行评估喷嘴操作条件的演示。这些条件会影响喷雾锥角，意味着影响液滴大小、速度和这些属性的空间分布。对步骤和条件的了解，作者控制了在所需的设计距离下的喷雾特性。 Experiments Equipment and Setup

The spray nozzle studied was a model 930/7-1 S35 from Dusen-Schlick GmBh (Untersiemau, Germany). Bore sizes of 1.2 and 0.8 mm were tested. The aqueous suspension used for spray coating

contained Opadry II white from Colorcon (West Point, PA). The solid content in the liquid suspension ranged from 4.2 to 13.2 wt%.

The imaging system, SprayWatch, from OSEIR, had been customized to measure the spray characteristics of typical suspensions used in the pharmaceutical industry. The detailed configuration of this system and the measurement principles are described in the following paragraphs.

Measurement Principles for Droplet Size and Velocity

The measuring system consists of two cameras. A CCD camera with microscope lens and a diode laser light source, shown in Fig. 1, were used to measure the droplet size and velocity distribution. The measurement area was 6.40×4.82 mm around the central line of the spray zone. This camera system was mounted on a metal frame with a stepper motor so as to move it along the spray zone with a measured increment. This arrangement allowed measurement of the spatial distributions of droplet size and velocity within the spray zone. The test nozzle was mounted so that its tip was 17 cm away from the measuring camera. This distance was selected to imitate a typical nozzle-to-tablet bed distance of 15 to 20 cm. A typical image captured with this setup is shown in Fig. 2.

Fig. 1 Setup of OSIER Spray-Watch system. It consists of a droplet camera for size and velocity measurements and a profile camera for measurement of pattern and geometry

Droplet velocities were derived from measurements using the PTV algorithm. The laser was triggered to generate three pulses at a fixed interval to capture triplet images corresponding to a single particle. A real-time image processing tool (LabVIEW, National Instrument, TX) was applied to enhance droplet images and then to identify those droplet tracks formed by the triplet images. The corresponding velocity of each track was calculated from measured displacement and the known pulse interval. The droplet size was derived from the circumference of the enhanced droplet images. There were usually more than a thousand droplets recorded in each image; distribution of velocities and sizes can therefore be derived. Based on the measured droplet size and velocity distribution, the droplet characteristics including volume flux (unit volume per unit area and time, nl/s mm2), number density (number/mm2),

volume average size (D90, D50, and D10), and standard deviation of the above were available. To minimize the interference from deposition of stray mist on the optical lens, there was an air knife or air curtain built into the light source and camera lens.We also recorded a new background occasionally to further mitigate the interference caused by droplets penetrating through the air curtain. The signal-to-noise ratio of recorded images remained high under this protective design. The system was configured to measure a droplet size range of 8–120 μm and velocities up to 60 m/s.

Fig. 2 A typical image of three-pulse droplets for velocity and size measurements, with backlight

Measurement Principle for Spray Pattern and Geometry

The spray profiles were recorded by a second cameramounted at the proximity of the nozzle tip. This camera was illuminated by a matrix of LED at near-infrared wavelength and focused through a wide-angle lens (shown in Fig. 1). The cone angle was derived from the measured angle between the cone

borderlines. The borderlines were detected through software by differentiating the intensities between spray and dry air. Figure 3 is a typical image from the profile camera.

1. The range of spray rate studied was from 60 to 180 g/min for the nozzle with 1.2 mm bore size. One spray rate at 25 g/min with a bore size of 0.8 mm was also tested. These spray rates encompass operating ranges for commercial production and laboratory batches.

Based on the vendor’s (Schlick) operating chart for spray nozzle S35, an atomization air (AA) of 180 standard liter per minute (SLPM) was selected as a starting point for the 150 g/min water spray rate test. The experiments covered AA/Spray mass ratios between 1.03 and 2.06. The AA was converted to a mass flow rate by multiplying the volumetric flow rate by the density of air. 2. The scaling spray study was performed with the Schlick model S35 nozzle with a 0.8-mm insert using 20–30 g/min spray rates and a Schlick model S35 with 1.2-mm insert and 100–180 g/min spray rates.

3. (1.29 mg/cm3) [13]. For the 120 and 180 g/min spray rates, the corresponding AA was applied to keep the same mass ratio as the 150 g/min condition.

4. The cone angle and spray-covered area were controlled by the ratio of the AA to PA. There were five AA/PA ratios tested in this study including 1.1, 1.3, 1.5, 1.7, and 1.9.

5. Solid content levels of 4.2% w/w, 8.1% w/w, and 13.2% w/w of Opadry II white were investigated with a 1.2-mm bore nozzle. The viscosities of these suspensions were measured in the range of 2 to 25 cP at 25°C with an inline viscometer (Galvanic Scientific Inc., MA).

The operating variables and recorded nozzle response variables are listed in Table 1. All the response variables in Table 1 were measured around the center of the spray zone. The spatial distribution of the response variables across the spray zone were recorded in a separate set of measurements.

Fig. 3 A typical spray image from the profile camera with geometry analysis

Results and Discussions Variables Affecting Droplet Size

As suggested by Kim and Marshall [10], the mean droplet size varies with the mass flow ratio (AA/Spray) to the power of 0.36. The size also changes with the air velocity to the power of 1.14. The curve fitting function in Excel was used to fit the measurement data for the exponent parameters of 0.2 and 0.36 in Eq

In Eq. 1, l represents the initial liquid sheet (m), ug is the air velocity at the point where the air jet impinges on the target (m/s), ρl is the density of liquid (kg/m3), σl is the surface tension of liquid (kg/m2), c is a constant representing the

Table 1 Tested nozzle operating and response variables Operating variables Tested ranges

Operating variables Tested ranges

Atomization air (SLPM) 52–288 Pattern air (SLPM) 47–261.8 Solid content (%wt) 4.2–13.2

Suspension viscosity (cP) 2–25 Spray rate (g/min) 25–180 Nozzle to ―bed‖ distancea 17 cm Response variables Measurement units and range Droplet size D10, D50 and D90 (microns);

8–100 μm

Droplet speed Mean and STD of velocity

(m/s); 0–100 m/s

Volume flux Spray volume in unit area unit time

(nl mm−2 s−1)

Droplet number density Droplet density in measured area

(no./mm2)

Cone angle Spray span (degree) Ratio of D90 to D10 Polydispersity of size distribution

(D90/D10)

SLPM standard liter per minute

a Distance from the nozzle to spray target to reflect actual distance in coating pan

Fig. 4 Effect of atomization air to liquid mass flow ratio (AA/Spray) on mean droplet size

efficiency of the atomization process, and AA/Spray is the mass flow ratio of atomization air to

sprayed liquid. The predicted mean sizes from Eq. 1 were overlaid with measurements shown in Fig. 4. The parameters for predictions were: l =0:0012m; c = 15; ρl=1000kg/m3 σ=0.075j/m2

Effect of Atomization Air–Liquid Mass flow Ratio (AA/Spray)

The atomization air–liquid mass flow ratio (AA/Spray) is one of the most important variables affecting the mean droplet size. An increase in the ratio leads to smaller droplet sizes as shown in Fig. 4. Also

plotted in the same figure was the predicted droplet size from Eq. 1 with parameters fitted at AA/PA=1.5. The effect of AA/PA ratio was not included in Eq. 1 and therefore may explain the deviations from measured data. At a constant liquid flow rate, the air flow rate can be increased, resulting in the increase of the mass flow ratio. However, the mean droplet size can only be decreased to a limited value when the mass flow ratio becomes larger than a threshold [9]. Droplets in this condition can be considered fully atomized. Further increases of the mass flow ratio will not result in a significant change of the droplet size. The decreasing size observed in the mass flow ratio between 1 and 2 (Fig. 4) indicates that the spray was not in the fully atomized region. However, the measured mean droplet size were in the 20 and 30 μm range, which was in the working range of15–50 μm suggested by the nozzle vendor [9].

Fig. 5 Effect of AA/PA ratio on mean droplet size

Fig. 6 The effect of suspension viscosity on droplet size, affected by suspension solids content

Effect of Atomization–Pattern Air Ratio (AA/PA)

The effect of the AA/PA ratio on mean droplet size is shown in Fig. 5. In addition to its impact on the shape of the spray cone, the AA/PA ratio also affects the size distribution. Note that the droplet size was smaller at lower AA/PA ratios or higher pattern air. The possible mechanism for pattern air

contributing to the liquid break down involves increased and more directed energy to the atomization of the liquid.

Effect of Viscosity or Surface Tension

粘度和表面张力的影响

Atomization is a process generating large amounts of surface area in a bulk liquid by transforming the liquid into droplets. Accordingly, higher surface tension will increase the energy barrier of atomization. Suspensions with higher content of Opadry II have higher viscosity and higher surface tension. Figure 6 implies that larger droplet sizes will arise from suspensions with higher solids content.

雾化是这样的一个过程在表面积上产生大量的液体，并且液体转化成水滴。因此，高表面张力会增加雾化的能量屏障。悬浮液中有较高含量的欧巴代II具有高粘度和高表面张力。图6意味着更大的液滴大小来自含有更高的固体含量的悬浮液。

Effect of Spray Rate

喷率的影响

Studies have indicated that mean droplet size increase with increasing spray rate [1]. However, in this work, there was no observable variation of mean droplet sizes at different spray rates when the AA/Spray ratio was fixed (Fig. 7). This observation supports AA/Spray ratio as one of the most important variables affecting mean droplet size.

研究表明平均液滴大小随着喷射率增加会增加 [1]。然而，在这项工作中，没有可观察到的变化意味着在不同的喷射率下的液滴大小固定当AA /喷雾比率固定时(图7)。这个发现支持AA /喷雾作为一个最重要的变量影响着液滴的大小。

Fig. 7 Effect of spray rate, at constant AA/Spray mass flow ratio, on droplet sizes

Variables Affecting Droplet Spatial Distribution

To obtain a uniform coat, it was desirable to generate a spray with a narrow droplet size distribution and with droplets uniformly distributed within the spray. In addition to the spatial distribution of droplets, the width of spray wetted area (spray cone angle) is also affected by the AA/PA ratio. With the SprayWatch system, the spatial distribution of droplets within the spray was measured by scanning across the entire spray zone.

Effect of AA/PA Ratio on the Spray Cone Angle

The spray shape and cone angle were mainly affected by the AA/PA ratio. Increasing the pattern air will increase the cone angle. The effect from varying atomization air was relatively small, as

demonstrated in Fig. 8. Equation 2 is empirically fitted from the measured data and shows very good correlation, as shown in Fig. 8.

Conclusions from a published study [3] suggested that broader spray distribution with a large cone angle could improve coating uniformity. The vendor also suggested using a cone angle up to 60° in their operation manual. From the measured correlation in Fig. 8, an upper limit of cone angle (approximately 60°) can be achieved at the AA/PA ratio of approximately 1:1 for the tested suspensions.

Fig. 8 Effect of AA/PA ratio on the cone angle of spray

Effect of AA/PA Ratio on the Spatial Distribution of Volume Flux

The distribution curves shown in Fig. 9 exhibit the spatial distribution of droplet volume flux (liquid volume per unit area per unit time) at different AA/PA ratios. Note that the smaller the AA/PA ratio, the flatter the flux spatial distribution was. This trend was comparable to the effect of the larger cone angle from smaller AA/PA ratio.

Fig. 9 Effect of AA/PA ratio on the spatial distribution of spray volume flux

Conditions Affecting the Spatial Distribution of Droplet Size (Mean Size and Width)

The ratio of D90/D10 can be a good indication of the droplet polydispersity. A ratio of D90/D10 equal to 1 indicates a uniform droplet size distribution. A large ratio may indicate wide distribution, tailing on both large and small sizes. Figure 10 shows the spatial distribution of D90/D10 ratio at various AA/PA values. Lower AA/PA ratios generate a narrow distribution of droplet size, while higher AA/PA ratios generate a wider size distribution. Figure 11 illustrates the impact of solids content on the spatial distributions of various size statistics, including the D90, D10, and mean droplet size. Because D90 was affected by suspension viscosity, this observed trend implies the following two phenomena: 1. suspensions with high solids content tend to increase the proportion of large droplets 2. large droplets tend to travel on the edge of spray zone

Fig. 10 Effect of AA/PA ratio to the spatial distribution of size polydispersity

Fig. 11 Spatial distribution of droplet size from different suspensions

Variables Affecting Droplet Velocity and its Distribution

Due to the relatively high heat and mass transfer across the air–liquid interface, the droplet velocity may impact coating efficiency and the surface roughness of the coated tablets. Small droplets traveling

in the center of a spray zone tend to have higher velocity and will be dried faster. The dried solid particles are therefore lost in effluent air more easily than larger droplets [6]. The air dynamics within the spray cone may provide one possible explanation of why large droplets travel slowly and on the edges. When the air plume spreads and slows, the droplets also travel more slowly and tend to recombine more frequently due to the turbulent vortex on the edges of the spray zone.

The measured spatial distribution of droplet velocity, shown in Fig. 12, was a function of the velocity profile across the spray zone. The maximum velocity was observed in the center of the spray zone and is less than 30 m/s at 17-cm distance from nozzle tip for most of the tested AA/PA ratios.

Fig. 12 Spatial distribution ofdroplet velocities at various AA/PA ratios

Effect of Mass Flow Rate Ratio (AA/Spray)

A higher mass flow rate ratio (AA/Spray) results in a higher velocity, as shown in Fig. 13. At a fixed spray rate, increasing atomization air will increase the corresponding droplet velocity at all levels of AA/PA ratios.

Effect of AA/PA Ratio

The effect of AA/PA ratio on droplet velocity is shown in Fig. 14 where the atomization air was the dominant factor affecting the droplet velocity. Note that increasing pattern air resulted in lower droplet velocity at all tested AA/Spray ratios. Equation 3 describes the empirically fitted correlation between droplet velocity and AA/PA ratio.

Scaling Spray Processes with Different Nozzle Inserts

Characterizing spray quality can greatly facilitate scaling because the spray process can be better controlled based on the spray quality instead of nozzle setup parameters. The authors compared droplet velocity, mean size, and volume flux from different nozzles at different spray rates.

The curves shown in Fig. 15 are the velocity profilesfrom the Schlick-S35 nozzle with different inserts and spray rates. Velocity ranged from 5 to 25 m/s with the two inserts. The spatial distributions of velocities from the two nozzle inserts were consistent.

The mean droplet size profiles are shown in Fig. 16 where the 1.2-mm insert provided larger droplets. The impact of droplet sizes on coated film quality may not be significant but should be determined experimentally. The volume flux is an indication of the deposition rate of liquid volume per unit of spray area. As demonstrated in Fig. 17, with proper nozzle adjustment, consistent volume flux may be achieved from unique inserts at different spray rates.

Fig. 13 Effect of mass flow ratio on mean droplet velocity

Fig. 14 Effect of AA/PA ratio on the mean droplet velocity

Fig. 15 Comparison of droplet velocity profiles from two nozzles with different bore sizes and spray rates

Fig. 16 Comparison of droplet size profiles from two nozzles with different bore sizes and spray rates

Fig. 17 Comparison of droplet volume fluxes from two nozzles with different bore sizes and spray rates

Conclusions 结论

1. The AA/Spray (atomization air/spray rate) and AA/PA (atomization air/pattern air) mass flow ratios are the operating parameters that affect the droplet characteristics. While suspension viscosity is known to affect the spray quality, the viscosities tested in this study were low and did not have observable impact.

这个AA /喷雾(雾化空气/喷雾率)和AA / PA(雾化空气/空气模式)质量流率是影响液滴特征的操作参数。虽然人们知道悬浮液粘度影响喷涂质量，本研究中液体粘度测试比较低并且没有可观察到的影响。

2. With proper selection of the AA/Spray and AA/PA mass flow ratios, it is possible to generate sprays with consistent spatial distributions of volume flux across the spray wetted area. Also, minimal variation of mean droplet size is achievable from different coating suspensions at different spray rates. 选择正确的AA /喷雾和AA / PA质量流量比率，在整个喷雾地区可以生成与喷雾体积通量一致的空间分布。同时，最小的变化意味着液滴大小在不同的喷率下和在不同的喷雾涂层悬浮液是可以实现的。

Acknowledgments Jussi Larjo and Ismo Linden from OSEIR are acknowledged for customizing the commercial system for the spray characterization. Howard Stamato, Charles Van Kirk, Sanjeev Kothari, Howard Miller, and Megan Schroeder are acknowledged for the suggestion of process conditions and support of process equipment. Daniel Hsieh is

acknowledged for the help and discussion of nozzle setup and calibration.

References

1. FDA. Guidance for industry PAT—A framework for innovative pharmaceutical development, manufacturing, and quality assurance; 2004 (September).

2. Turton R, Cheng XX. The scale-up of spray coating processes for granular solids and tablets. Powder Technol 2005;150:78–85.

3. Muller R, Kleinebudde P. Scale-down experiments in a new type of pan coater. Pharm Ind. 2005;67(8):950–7. 4. Rege BD, Gawel J, Kou JH. Identification of critical process variables for coating actives onto tablets via statistically designed experiments. Int J Pharm. 2002;237:87–94.

5. Ruotsalainen M. Studies on aqueous film coating of tablets performed in a side-vented pan coater. Dissertation, Department of Pharmacy, University of Helsinki; 2003.

6. Lavernia EJ, Wu Y. Spray atomization and deposition. 3rd ed. Chichester: Wiley; 1997.

7. Scattergood LK, Cunnungham CR, Vesey CF, Fefely KA. Optimization of spray-nozzle performance for aqueous filmcoating process, Colorcon Technical Report (www.colorcon.com/ pharma/post_articles/opt_spray.pdf). 8. Shi D, Li M, et al. Rule-based modeling of coating microstructure. Ind Eng Chem Res. 2004;43:3653–65. 9. Operating Instruction for Schlick spray nozzle model 930/7-1 S35. Orthos Liquid Systems, Inc.; 2006. 10. Kim KY, Marshall WR Jr. Drop-size distributions from pneumatic atomizers. AIChE J. 1971;17(3):575–84. 11. Zhou Y, Lee S, McDonell VG, Samuelsen S, Kozarek RL, Lavernia EJ. Influence of operating variables on average droplet size during linear atomization. At Sprays. 1997;7:339–58.

12. Chigier N. Optical imaging of sprays. Prog Energy Combust Sci. 1991;17:211–62.

13. Perry RH, editor. Perry’s chemical engineer’s handbook. 7th ed. New York: McGraw-Hill; 1997.