Read Ilass 2001 paper text version

Characterization of Spray Guns for the Tablet Coating Industry Rudolf J. Schick and Keith F. Knasiak Spraying Systems Co. Wheaton, IL, U.S.A ABSTRACT Automated spray guns are widely used in fully integrated pan-coating systems. Such pan-coating systems are used by the pharmaceutical/food industries to apply aqueous film, solvent and sugar coating for tablets, confectioneries, and a variety of other products. Typically, pan-coating systems are supplied from the equipment manufacturer fully configured for a desired application. The end user commonly makes necessary changes to the spray gun arrangement in order to accommodate a new application or a change in the coating formulation. Often, these changes do not produce the expected results due to a less than optimal spray gun configuration. Frequently an operator will change the gun operating conditions by simply varying the fluid and gas feeds. As a result, the spray pattern may deteriorate and drop size may increase beyond the desired size thus affecting the final product quality. A standard method for optimizing spray guns used in tablet coating applications does not exist at present. This paper proposes a method that includes an analysis of the critical elements of atomizer testing, consisting of layout and overlap optimization to reduce the errors commonly associated with spray manifold application, such as uneven spray distribution, heavy edges and poor overlap. Furthermore, this approach will also consider the effects of "normal" operating parameters such as fluid flow rate, atomization pressure and gun-to-target distance on spraypattern uniformity, drop size, and velocity. INTRODUCTION Aqueous film coating is one of the most important operations in the pharmaceutical industry. The process is very demanding and is not as forgiving as other coating applications such as solvent coating. A successful process greatly depends on controlling many of the variables with little or no allowance for significant variation. (conventional/side-vented), the air and tablet movement, and temperature. The dynamics of the tablet also affect the process; these include dimensional changes due to expansion and contraction in the moisture absorption and final drying stages. From a spray application standpoint, the problem can be analyzed by evaluating the steps in the coating process. As shown in Figure 1, these steps include the formation of drops, the contact of drops with the solid surface, droplet expansion, droplet coalescence, droplet penetration and finally droplet evaporation. Early research and field experiments indicate the need for a narrow span droplet size distribution with a limited number of small drops (or fines). Small drops have a tendency to drift off-target, therefore, the number of "fines" must be reduced in order to provide uniform coating. Tablet coating machines are equipped with automated spray guns specifically designed to deliver the coating solution. The general delivery method includes a manifold system that is fitted with multiple spray guns; the tablet coating machine control system controls these guns. Typically, these machines are designed and optimized for a specific process. As the process is modified, the operator alters the gun settings to accommodate the new process. However, frequently these changes do not produce the expected results due

Figure 1: Tablet Coating Process There are many governing parameters that relate to the processing conditions such as the type of tablet coater

As presented at the 15th Institute of Liquid Atomization and Spray Systems Conference, May 20-23, 2001 at The Dearborn Inn, Dearborn, Michigan

to a lack of standardized procedures that can be used to optimize the guns' settings. This paper evaluates a method that will include an analysis of the critical elements of atomizer testing. This method consists of layout and overlap optimization to reduce the errors commonly associated with spray manifold application, such as uneven spray distribution, heavy edges and poor overlap. This approach will document a baseline condition for an existing setup in a tablet coater and will proceed in optimizing the flux/density distribution to obtain the optimal spray setup and overlaps. Optimization will be accomplished by using a patternation technique that involves collecting liquid flux density data at various gun spacings and gun-to-target distances. Data will be collected at various feed rates that are typically used in tablet coating applications. The patternation data is then correlated with drop size and velocity data in order to determine the effects of "normal" operating parameters such as liquid flow rate, atomization air pressure, and gun-to-target distance on spray-pattern uniformity, drop size, and velocity. Figure 3: Schematic of PDPA set up. Drop size data is collected using an Aerometrics Inc. Phase Doppler Particle Analyzer (PDPA). The sampling method consists of collecting data at various radial locations throughout the spray. Information such as volume flux and velocity are correlated with the D32 measured at these locations. Results will demonstrate the variation in drop size as a function of spray distance and the sampling location within the spray. APPROACH Experiment. The spray patternation data was obtained using a 4.55m linear patternator with 2.54 cm on-center collection tubes, as shown in Figure 2. Drop size and velocity data were collected using a PDPA system as shown in Figure 3. A Spraying Systems Co. 7310-1/4JAU+SUE15B spray gun (SPG1) was used in this test. This is an externalmix two fluid atomizer. This atomizer features independent air and liquid inlets. The liquid is directed through a metering orifice to a nozzle exit. The liquid exits the nozzle orifice and makes contact with the coaxial atomizing air initiating primary breakup. Droplets are then spread into a 45° sheet by the fan air. The fan-air is generated by two impinging air jets, these jets are also responsible for secondary droplet breakup. Both nozzle and body assemblies are constructed of 303 Stainless Steel. A schematic of this atomizer is shown in Figure 4. The tests were conducted under laboratory conditions. The atomizers were mounted on a 2-axis traverse. A clamp assembly was used to hold the atomizers in the test positions. Liquid flow to the atomizer was delivered using a 40-liter pressure tank and was monitored using a Micro Motion D6 flow meter. The MicroMotion flow meter is a Coriolis Mass flow meter that measures the density of the fluid to determine the volume flow rate. The meter is accurate to ±0.4% of reading. Liquid pressure was monitored immediately upstream and downstream of the atomizer manifold, using 0-10 bar, class 3A analog pressure gauges. The test matrix for this experiment is shown in Table 1. Spray gun Gun spacing Gun-to-target distance Flow rate (4) 7310-1/4JAU+SUE15B 75-102 mm 275 mm, typical range 196 393 mm 137 - 300 mL/min

Figure 2: Schematic of linear patternator

Table 1: Test Matrix

As presented at the 15th Institute of Liquid Atomization and Spray Systems Conference, May 20-23, 2001 at The Dearborn Inn, Dearborn, Michigan

200-mm lens was used for the transmitter and 500-mm lens was used for the receiver units. For all testing, a size range of 1.1-150.0µm was used to ensure capturing the full range of droplet sizes. The velocity range was set with a range of 0-30m/s, which was more than ample to capture the full range of droplet velocities. For each test point, a total of 10,000 samples or a 30-second collection time were used. A schematic of the test setup is shown in Figure 3. PDPA theory and sizing accuracy has been fully described in many references. For accurate determination of the mean and median droplet sizes, along with accurate measurements of volume flux and number density, all droplet size classes must have an equal probability of detection. Without correction, the PDPA exhibits sample volume bias due to the Gaussian intensity profile of the laser beams. The PDPA software performs a sample volume correction based upon an empirical determination of the maximum sample volume size for each size class in order to give all size classes equal probability of detection. This correction is imperative for accurate flux and number density determination, which can be within ± 10% with proper high voltage setting. PMT gain was set via the intensity function. For all test conditions, a gain of 350-400 volts was adequate. The system was operated with transit based probe volume correction and number density. The intensity validation routine was active at all times, thus reducing the possible error due to beam attenuation and improper sizing due to reflection of multiple particles throughout the varying spray conditions. The 50:1 dynamic size range was set to encompass the maximum particle size, while the velocity range was set to include the maximum particle velocity present. RESULTS AND DISCUSSION The success of most coating applications will depend on evenly distributing the spray's volume flux over the target area. As with most flat fan atomizers a typical volume flux distribution can be described by having a heavy concentration of volume at the center of the spray and a light concentration of volume towards the edges of the spray. For reference, a volume flux distribution from a single (SPG1) is shown in Figure 6. When a series of atomizers are installed on a manifold, the atomizer spacing should be optimized via overlap analysis in order to obtain a uniform coating of the target. This optimization involves a careful balance of the gun-to-gun spacing and gun-to-target distance along with an optimization of the atomizers operating conditions. The overlap analysis is important primarily for the prediction and determination of the uniformity of

Figure 4: 7310-1/4JAU+SUE15B Spray Gun (SPG1) INSTRUMENTATION AND PROCEDURE The volume flux density (patternation data) was collected using a specially designed lineal patternator, as shown in Figure 2. This device allows for the collection of volume flux in the x-plane at the various spray heights and gun spacings listed in the test matrix (Table 1). This patternator is ideal for analyzing the spray performance of flat fan atomizers. The patternator was fitted with a manifold containing four (4) (SPG1) atomizers. The manifold allowed for adjustment of atomizer spacing and gun-to-target distance in order to simulate the layout in a tabletcoating machine. A schematic of the test manifold is shown in Figure 5.

Figure 5: Schematic of test manifold Drop size and velocity data were collected using a onecomponent Aerometrics PDPA instrument. The transmitter and receiver were mounted on a rail assembly with rotary plates; a 30-degree forward scatter collection angle was used. For this particular test, a

As presented at the 15th Institute of Liquid Atomization and Spray Systems Conference, May 20-23, 2001 at The Dearborn Inn, Dearborn, Michigan

the composite spray across a linear dimension defined by the spray manifold. Uniformity is judged visually and statistically by analyzing the composite spray distribution produced by the manifold.

Single Nozzle Spray Distribution

approach the condition shown in Figure 7(c). This is generally a swift process and the solution is normally approached very quickly.

20

Tube Volume (ml)

15

10

5

0

8

6

4

2 20 .3

40

32

24

16

0. 00

5. 08

.0

10 .1

15 .2

-2

-2

-1

-1

Tube Position (cm)

Figure 6 The first step in providing parameters for determining a quality overlap is a determination of some process specific parameters. These usually include: allowable gun-to-target distance, pressure ranges, quantity of product required, etc. From these an atomizer can be selected that provides an appropriate capacity and spray angle. Selection of the atomizer will also allow for determination of the expected operating pressure. As a starting point, an arbitrary gun-to-target distance (h) can be selected that falls in the application's allowable range. Using the atomizers' spray angle and some simple trigonometry the distance (D) between spray atomizers can be set such that the spray patterns of two adjacent atomizers will intersect at approximately h/2. This is assuming that a triangular spray pattern is generated by the atomizers. With a known h/D ratio a patternation test can now be run at a previously determined pressure. Figure 7 below shows some possible outcomes of such a test. In the first case, Figure 7(a), we note that the peaks in the spray distribution fall at the overlap point of two atomizers. This indicates that the value of the h/d ratio is too large. Decreasing the atomizer spray height or increasing the distance between the atomizers will lower the value of the h/D ratio. As shown in Figure 7(b) the atomizers can be set such that the h/D ratio is also too small. In the case of a small h/D ratio, the peaks have moved to a position where they are symmetrical about the atomizer centerline. A remedy would include raising the atomizer spray height or reducing the atomizer-toatomizer distance. Finally, Figure 7(c) shows a correct h/D ratio. In general an iterative process is used to

25 .4

5.

0.

5.

0.

-5

0

Figure 7: Overlap Configurations Once a suitable h/D ratio is found, all spray parameters must be held constant. Otherwise variations in the atomizer's flux density distribution will occur and will generally provide a net negative effect on the overall overlap distribution. In the case where varying air or liquid pressures must be used, an h/D ratio should be found such that these pressure changes will bound a median set of pressure settings on both the upper and lower sides. The preferred method to determine the effect of these changes on the spray overlap is to statistically analyze the composite spray distribution produced by the manifold. The basis for this analysis is discrete spray distribution data obtained from a lineal patternator. The individual volume collected in the tubes of the patternator is used to calculate Cv (coefficient of variation). Cv is defined in this case as the ratio of the sample-weighted standard deviation of the collected spray volume to the average volume collected.

Cv =

1

__

n x 2 - ( x ) n(n - 1)

2

X

As presented at the 15th Institute of Liquid Atomization and Spray Systems Conference, May 20-23, 2001 at The Dearborn Inn, Dearborn, Michigan

Simply put, this is the standard deviation of the volumes collected in the tubes divided by the average value and expressed in percent. In practice, a minimum Cv can be found while using constant spray input conditions. This should be treated as a pivot point when making allowances for variable spray settings. Overlap uniformity is simply judged by the Cv value obtained. In standard industrial applications, a Cv15% can be considered a good overlap distribution. In processes where the spray target is being mixed, this constraint can be loosened if it is expected that the material is also going to be spread through a mixing process as well. In the initial setup for proper overlap, mid-range values should be used so that optimal settings are arrived at in a minimal number of iterations that will best serve the entire range of spray parameters expected to be used. PATTERN OPTIMIZATION PROCESS To show the spray overlap optimization experimentally, four (SPG1) atomizers were setup on a manifold. An arbitrary h/D ratio was selected so that the sprays overlapped at the very edges of their spray pattern. This is shown below in Figure 8(a). Four distinct peaks are shown in this plot. These peaks coincided with the center of each gun. Though this is an extreme case it does show that if the h/D ratio is too small that the peaks will reside under the centerline of the atomizers. The Cv for this case is extremely high, 49.25%. This is much too high of a Cv value generated due to a lack of spray overlap. Figure 8(b) shows the first iteration of trying to improve the spray overlap. The volume change from the peaks and valleys in the spray has been decreased. This was done by increasing the h/D ratio by decreasing the atomizer-to-atomizer distance (D). Though there is improvement here, the Cv is still at a high value of 33.77%. This is much too large, leading to a second iteration. The second iteration is shown in Figure 8(c). There is marked improvement at this point. The variability of the spray overlap has been decreased significantly as shown by a Cv= 24.79%. Though this is a marked improvement, it is still a fairly large value and can be further reduced. Comparing Figure 8(c) to Figure 8(a) above it, it is obvious that the h/D ratio has been increased too far. This is demonstrated by the peaks in the spray appearing to have switched from being under

the centerlines of the atomizers to being in the overlap region of the sprays. A third and final iteration was performed to try to reach a smaller Cv. This was done by decreasing the h/D ratio, again by increasing the atomizer-to atomizer distance (D). In this case, shown in Figure 8(d), the Cv has been decreased significantly to 6.81%. This is an extremely high quality overlap and should provide a relatively stable starting point if the material being sprayed is changed or the liquid and/or air pressures need to be changed. Figure 8(e) and (f) show the effects of variation in flow rate and air pressures while maintaining the h/D ratio shown in Figure 8(d). In the case shown in Figure 8(e), the flow rate was dropped from around 200ml/min to 137ml/min. The Cv value increases slightly from case (d) to 9.68% for case (e), but this still provides a good overlap distribution. This was done by decreasing atomization air so that the proportion of fan air to liquid flow remained fairly close to the ratio used in the 200ml/min case. The case shown in Figure 8(f) shows an overlap distribution for 137ml but the gun-to-target distance (h) has been changed to 25.4cm as compared to 16.5cm. To compensate for the change in gun-to-target distance, (D) must be increased such that the h/D ratio remains the same. It appears that the distribution has deteriorated a small amount and this is evidenced by the increase in Cv to 13.82%. This is still a relatively good overlap and demonstrates that changing system parameters will effect the overlap distribution. However, as shown in case (d) it seems a good central point has been found for this atomizer operating under these conditions. Table 2 shows the gun spacing parameters, flow rates and Cv for all test conditions shown in Figure 8. Case a b c d e f Q (ml/min) 200 200 200 200 137 137 h (cm) 17.78 25.40 16.51 15.24 16.51 25.40 D (cm) 15.24 12.38 7.62 7.62 7.62 10.16 h/D 1.167 2.051 2.167 2.000 2.167 2.500 Cv (%) 49.25 33.77 24.79 6.81 9.68 13.82

Table 2

As presented at the 15th Institute of Liquid Atomization and Spray Systems Conference, May 20-23, 2001 at The Dearborn Inn, Dearborn, Michigan

Overlap Distribution

35

(a)

30

(b)

Tube Volume (ml)

25

20

15

10

5

0

35

(c)

30

(d)

Tube Volume (ml)

25

20

15

10

5

0

35

(e)

30

(f)

Tube Volume (ml)

25

20

15

10

5

0

10 .1 6

15 .2 4

20 .3 2

25 .4 0

10 .1 6

15 .2 4

20 .3 2

-2 5. 40

-2 0. 32

-1 5. 24

-1 0. 16

-2 5. 40

-2 0. 32

-1 5. 24

Tube Position (cm)

-1 0. 16

Tube Position (cm)

Figure 8

As presented at the 15th Institute of Liquid Atomization and Spray Systems Conference, May 20-23, 2001 at The Dearborn Inn, Dearborn, Michigan

25 .4 0

0. 00

5. 08

0. 00

-5 .0 8

-5 .0 8

5. 08

Percent (%)

Figure 9 shows a comparison of the baseline condition and the three overlap improvement iterations and then compares these to cases (e) and (f). This figure shows that a good spray overlap uniformity can be arrived at quickly. Once this point is known for certain operating conditions, tests that demonstrate the effects of operating parameter changes can be made to determine the effect on the overlap distribution.

C v vs . T es t C as e s a -f

Drop size distribution at Q: 30 and 60ml/min 2 h: 314mm - Pa: 2kg/cm

100 90 80 70 60 50 40 30

30ml 60ml

C o e fficie n t o f V a rian ce : C v (% )

50

a

20 10

40

0 1 10 100

b

30

Drop Size (µm)

Figure 10: Drop Size Distribution

c

Velocity distribution at Q: 30 and 60ml/min 2 h: 314mm - Pa: 2kg/cm

20

f

10

1000 900 800 700

e d

30ml 60ml

0

Figure 9

Count

600 500 400 300 200 100 0 0 5 10 15 20 25

Drop Size Tests The tests were conducted on a single atomizer at a gunto-target distance (h): 236, 314 and 393mm, atomizing air pressure (Pa): 0.3 ­ 4.5kg/cm2, and flow rate (Q): 30 and 60ml/min. The 30 and 60ml/min flow rate is the range of flow rates through a single atomizer on the manifold. This data was collected using an Aerometrics Inc. Phase Doppler Particle Analyzer (PDPA). The sampling method consists of collecting data at various radial locations throughout the spray. Information such as volume flux and velocity are correlated with the D32 values. Effect of liquid flow rates. The effects of liquid flow rate on D32 and average velocity Vavg were evaluated at flow rates of 30 and 60ml/min at a constant h=314mm and a constant Pa=2kg/cm2. The Sauter Mean Diameter (D32) increased from 24.3µm to 30.1µm as the liquid flow rate was increased from Q=30ml/min to Q=60ml/min. The Drop size distribution, shown in Figure 10 illustrates the increase in size with a slight shift in the slope of the distribution for the 60ml/min condition. This shift illustrates the wider span of sizes, at 30ml/min the span was 4.66-60.79µm, where as the drop size span at 60ml/min was 4.66-75.66µm. The Vavg remained virtually unchanged from 10.64m/s at Q=30mil/min to 10.42m/s at Q=60ml/min. The velocity distribution is shown in Figure 11.

Velocity (m/s)

Figure 11: Velocity Distribution Effect of atomizing air pressure. The effects of atomizing air pressure on D32 and Vavg were also evaluated at the liquid flow rates of 30 and 60ml/min at a constant h=314mm. The atomizing air pressure range was set at 0.3 ­ 4.5kg/cm2. At 30ml/min, D32 decreased from 68.1µm to 18µm as the air pressure was increased from 0.3 ­ 4.5kg/cm2. No significant decrease in D32 was realized beyond 2.05kg/cm2. At the same atomizing air pressure range, Vavg increased from 7.4m/s to 15.8m/s. Unlike the D32, Vavg continued to increase with an increase in air pressure. This data suggests that a relatively constant D32 can be achieved at a wide velocity range. The D32 and Vavg data for the Q=30ml/min test condition are shown in Figure 12. The same trend was observed at Q=60ml/min. D32 decreased from 91.8µm to 24.7µm as the air pressure was increased from 0.3 ­ 4.5kg/cm2. In this case also,

As presented at the 15th Institute of Liquid Atomization and Spray Systems Conference, May 20-23, 2001 at The Dearborn Inn, Dearborn, Michigan

no significant decrease in D32 was realized beyond 2.05kg/cm2. At the same atomizing air pressure range, Vavg increased from 6.63 m/s to 13.49 m/s. This data suggests that at an atomizing air pressure in excess of 2.05kg/cm2, this atomizer can be operated at 30 and 60 ml/min and yield similar D32 and Vavg values. The D32 and Vavg data for the Q=60ml/min test condition are shown in Figure 13.

D32 & Vavg vs. Atomizing Air Pressure Q=30ml/min -- h=314mm

not significant and indicate minimal effect on the drop size at Pa=2.0kg/cm2. The same trend was observed at 60ml/min. The recorded D32 values were 29.7, 30.1 and 30.8µm at h= 236, 314 and 393mm respectively. At 30ml/min, the recorded Vavg values were 13.7, 10.64 and 8.63m/s at h= 236, 314 and 393mm. The velocity data indicates a noticeable effect of the gun-target distance on Vavg. The decrease in Vavg is due to aerodynamic drag effects that resulted in a reduction in the spray velocity. The same trend was observed at the 60ml/min condition. The recorded Vavg values were 12.2, 10.42 and 8.7m/s at h= 236, 314 and 393mm. The D32 and Vavg data for the 30 and 60ml/min conditions is shown in Figure 14.

Distance Effects on D 32 and Vavg

32 30 28

393 236 314 393 314 236

70 60

D32 Vavg

D32 (µm), Vavg (m/s)

50 40 30 20 10 0 0 1 2 3

2

32 30 28 26 24 22 20 18 16 14

236

26 24 22 20 18 16 14 12 10 8 6 4 2 0

314 393 236

D32 (µm), Vavg (m/s)

4

5

Atomizing Air Pressure (kg/cm )

Figure 12 Effect of gun-to-target distance. The effects of gun-totarget distance on D32 and Vavg were also evaluated at the liquid flow rates of 30 and 60ml/min. The atomizing air pressure was set at 2.0kg/cm2 and the gun-to-target distance was set at 236, 314 and 393mm.

D32 & Vavg vs. Atomizing Air Pressure Q=60ml/min -- h=314mm

D32 (µm), Vavg (m/s)

12 10 8 6 4 2 0

314 393

Q=30ml/min

Q=60ml/min

D32

Vavg

D32

Vavg

(Note: Numbers next to data point indicate gun-to-target distance)

Figure 14 CONCLUSIONS A method for optimizing spray guns used in tablet coating applications was presented in this work. The approach consists of optimizing the spray flux density and spray overlap and evaluating the effects of "normal" operating parameters such as fluid flow rate, atomization pressure, and gun-to-target distance on spray-pattern uniformity, drop size, and velocity. All tests conducted in this study were performed using water. Through an iterative spray pattern optimization process a substantial reduction in the coefficient of variation Cv was possible. At the base line condition, the Cv value was extremely high at 49.25%, Figure 8(a). Through subsequent iterations (Figures 8(b), 8(c) and 8(d)), the Cv values were reduced to 33.77%, 24.79% and 6.81%. A Cv of 6.81% represents an extremely high quality spray overlap uniformity and could be considered optimal. At the optimal h/D condition variations in

90 80 70

D32 Vavg

D32 (µm), Vavg (m/s)

60 50 40 30 20 10 0 0 1 2 3

2

4

5

Atomizing Air Pressure (kg/cm )

Figure 13 At 30ml/min, the recorded D32 values were 27, 24.3 and 28.4µm at h= 236, 314 and 393mm respectively. The variation in D32 at the indicated gun-to-target distance is

As presented at the 15th Institute of Liquid Atomization and Spray Systems Conference, May 20-23, 2001 at The Dearborn Inn, Dearborn, Michigan

flow rate and air pressures increased the Cv values to 9.68% and 13.82% (Figures 8(e) and 8 (f)). This increase in Cv is not considered significant but demonstrates that changing a system's parameters will effect to the overlap distribution. Figure 9 shows a comparison of the baseline condition and the three overlap improvement iterations and then compares these to cases (e) and (f). An increase in liquid flow rate from 30 to 60ml/min (at h=314mm and Pa=2kg/cm2 ) increased the D32 by 23% and increased the drop size span as shown in Figure 10. At same test conditions the Vavg remained virtually unchanged. This data confirm a direct effect of an increase in flow rate on drop size and a minimal to no effect of an increase in flow rate on the average velocity. An increase in atomizing air pressure from 0.3 to 4.5kg/cm2 (at h=314mm and Q=30 ml/min) reduced the D32 from 68.1µm to 18µm, however no significant decrease in D32 was realized beyond 2.05kg/cm2. At the same test conditions, Vavg increased from 7.4m/s to 15.8m/s and continued to increase with an increase in air pressure as shown in Figure12. The same trend was observed at Q=60ml/min, this data is shown in Figure 13. The test data suggests that at an atomizing air pressure in excess of 2.05kg/cm2, this atomizer can be operated at 30 and 60 ml/min and yield similar D32 and Vavg values.

D32 vs. Atomizing Air Pressure Q: 30 ml/min

In total the effects of flow rate, atomizing air pressure and gun-to-target distance have a combined effect on D32 and Vavg. Figures 15 and 16 present this data in its totality for both the Q=30ml/min and Q=60ml/min respectively. These plots provide the tablet coater operator a calibration curve that can be used to optimize the drop size and velocity based on the desired process conditions. The extent of this effect and its relevance to the tablet coating application are determined by other factors such as type of tablet coater, the air and tablet movement, and temperature. These factors were not within this scope of this study. More work is needed in order to investigate these factors and their effect on the spray application in the tablet coating applications.

D32 vs. Atomizing Air Pressure Q: 60 ml/min

100 90 80 70

Spray Height 236mm 314mm 393mm

D32 (µm)

60 50 40 30 20 10 0 0 1 2 3

2

4

5

Atomizing Air Pressure (kg/cm )

Figure 16

70 60 50

Spray Height 236mm 314mm 393mm

D32 (µm)

40 30 20 10 0 0 1 2 3

2

ACKNOWLEDGMENTS The authors would like to acknowledge Mr. Jim Nightingale and Mr. Dylan Mason from Bend Research Inc. and Mr. Jim Wlodarczyk from Thomas Engineering Inc. for their assistance with this project. REFERENCES 1. T Yao, M Yamada, H Yamahara and M Yoshida, "Tableting of coated particles. II. Influence of particle size of pharmaceutical additives on protection of coating membrane from mechanical damage during compression process." Chemical & Pharmaceutical Bulletin (Tokyo), Vol 46, No 5, pg. 826-830, 1998. 2. Jyrki Heinamaki, Mirja Ruotsalainen, Veli-Matti Lehtola, Osmo Antikainena and Jouko Yliruusi, "Optimization of aqueous-based film coating of tablets performed by a side-vented pan-coating system," Pharm. Dev. Technol, Vol 2, No 4, pg. 357-364, 1997

4

5

Atomizing Air Pressure (kg/cm )

Figure 15 The gun-to-target distance had minimal effect on drop size at both flow rate conditions (Q=30 ml/min and Q=60 ml/min). In contrast the effect of gun-to-target on Vavg was noticeable and can be attributed to aerodynamic drag effects that resulted in a reduction in the spray velocity, this data is shown in Figure 14.

As presented at the 15th Institute of Liquid Atomization and Spray Systems Conference, May 20-23, 2001 at The Dearborn Inn, Dearborn, Michigan

3.

4. 5.

6.

7.

Atomization and Spray Drying. W. R. Marshall. Department of Chemical Engineering. University of Wisconsin Madison, June 1954. Curt Henry Appelgren, "A method for coating granules, pills, and tablets," SE, September 24, 1986 R. J. Schick, "A Guide to Drop Size for Engineers," Spraying Systems Co. Bulletin 459. W. D. Bachalo and M. J. Houser, "Phase Doppler Spray Analyzer for Simultaneous Measurement of Drop Size and Velocity Distribution", Optical Engineering, Vol. 23, No. 4, pp. 583, 1984. E799-92 (Re-approved 1998): Standard Practice for Determining Data Criteria and Processing for Liquid Drop Size Analysis. "2000 Annual Book of

ASTM Standards, General Methods and Instrumentation, Volume 14.02 pp. 343-347". 8. E1296-92: Standard Terminology relating to liquid particle statistics. "1996 Annual Book of ASTM Standards, General Methods and Instrumentation, Volume 14.02 pp. 810-812". 9. C. Signorino "Trouble Shooting 18 Common Problems Associated with the Coating of Tablets", Proc. COATING TECHNOLOGY '94, Atlantic City, NJ, August 1994 10. Eric Forster and Kenneth Lawrence Morrow, "Improvements in tablet coating apparatus and method" GB, September 24, 1996

As presented at the 15th Institute of Liquid Atomization and Spray Systems Conference, May 20-23, 2001 at The Dearborn Inn, Dearborn, Michigan

Information

Ilass 2001 paper

10 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

686843


You might also be interested in

BETA
Ilass 2001 paper
Microsoft Word - WP005_Tablet Coating.doc