Drug Development and Industrial Pharmacy, 23(2), 193-202 (1997)

Drug Development and Industrial Pharmacy Downloaded from informahealthcare.com by University Of Wisconsin Madison on 07/19/13
For personal use only.

RESEARCH PAPER

Prospective Validation of High-Shear

Wet Granulation Process by Wet
Granule Sieving Method. I. Selection and
Characterization of Sieving Parameters
for Wet Granules
Hisatoshi Emori, Teruhisa Yoshizawa,
Toshiaki Nishihata, and Tadanori Mayumi2
Pharmacy Research, Upjohn Tsukuba Research Labs, Upjohn
Pharmaceuticals Ltd., 23 Wadai, Tsukuba, Ibaraki 300-42, Japan
2Faculty of Pharmaceutical Sciences, Osaka University, 1-6 Yamadaoka,
Suita, Osaka 565, Japan

ABSTRACT

To characterize the progression of high-shear wet granulation for various drugs

and formulations based on the particle size distribution of wet granules during
granulation, a general sieving method for wet granules was investigated. Wet
granulation was conducted in a 25-liter high-shear mixer using four model drugs
with different solubilities and particle sizes (ethenzamide, unmilled and milled
acetaminophen, and antipyrine). Because of its small size and eficient sifting
mechanism, a sonic sifter was used to determine the wet granulation particle size
distribution. From the good correlation of particle size distribution between wet
granules and dry-sized granules, an intensity of 80% of full-scale amplitude and
a sieving time of 3 min were selected as wet granule sieving parameters. 7% general sieving method showed good measurement precision as long as the determination was completed within 20 min after sampling, Further, the method was independent of sampling position within the mixer chamber.

*To whom correspondence should be addressed.

193
Copyright

1997 by Marcel Dekker, Inc.

Emori et al.

194

Drug Development and Industrial Pharmacy Downloaded from informahealthcare.com by University Of Wisconsin Madison on 07/19/13
For personal use only.

INTRODUCTION
A wet granulation method with a high-shear mixer is
widely used to produce granules for tablets in the pharmaceutical industry. The primary purposes of wet granulation are to improve the properties of powders, i.e.,
material handling, flowability, dissolution, and compression characteristics (1). Because the wet granulation
process is known to influence final tablet characteristics,
such as content uniformity. dissolution rate, disintegration time, hardness, and friability (2), this process is
often defined as a critical process for the final tablet
characteristics. The end point determination method in
this process influences the tablet performance.
The wet granulation process, especially granulation
end point, is often controlled by subjective operator
evaluation, by power consumption of the mixer motor
(3-12), or by torque of the mixer shaft (13-19). However, the operator evaluation technique has the inherent
intra- and interoperator variability. The power consumption and torque techniques are dependent on formulation, process, and equipment variables. Thus, these
techniques are not always adequate during formulation
development and scale-up, while they are very convenient methods to determine granulation end point in production site.
The method that directly characterizes the progression of wet granulation by monitoring the wet granule
particle size distribution (wet granule sieving method;
WGSM) has been briefly developed for a proprietary
formulation containing over 70 % insoluble micronized
drug in the granulation phase, in order to control the
granulation process and to determine the end point
(20,21). The in-process method is available on the spot,
simple, fast, independent of operator evaluation, and
requires a small sample amount. A sonic sifter has been
employed as wet granule particle sizing equipment because of its small size and efficient sifting mechanism.
Since this method is also independent of mixer design
and scale, it is very useful during scale-up. A correlation between the amount of wet granules larger than 2
mm and the degree of homogeneity of liquid distribution was reported by Holm et al. (22). However, they
did not focus this work on a granulation end point
method.
If the WGSM can be shown to be applicable to other
drugs and formulations, it will provide very useful information on in-process parameters in the high shear
wet granulation process. Also, the wet granule particle
size distribution should be determined by a characterized
sieving method, since it is expected that the WGSM is

capable of validating the wet granulation process prospectively. A general sieving method to determine the
wet granule particle size distributions of various drug
formulations was selected and then characterized in the
present study (Part I).
Four model drugs with different solubilities and particle sizes were used, and formulations containing high
percentages of active ingredients in the granulation portions were employed so that the drug properties were
well reflected in the wet granulation. The following experiments were carried out:
1.

2.

3.

First experiment to select sieving parameters

(sieving time and intensity) for wet granules
Second experiment to determine the measurement precision for wet granule sieving using the
selected parameters and the effect of granule
standing time (time between sampling and analysis)
Third experiment to determine the effect of sampling position of wet granules within the highshear mixer bowl

MATERIALS
Ethenzamide (low solubility), acetaminophen (medium solubility), and antipyrine (high solubility) were
used. Further, two bulk drug lots with different particle
sizes were used for acetaminophen, i.e., unmilled (acetaminophen) and milled (acetaminophen-milled) lots.
Mean particle sizes were 1.1 pm for ethenzamide, 192.7
pm for acetaminophen, 36.1 pm for acetaminophenmilled, and 251.8 pm for antipyrine. Ethenzamide, acetaminophen, and acetaminophen-milled were of JP
grade and purchased from Yoshitomi Pharmaceutical
Co., Ltd. (Osaka, Japan); antipyrine JP was from Yashiro Pharmaceutical Co., Ltd. (Osaka, Japan). Lactose JP
(DMV 200M; DMV Japan, Tokyo, Japan) and hydroxypropyl cellulose JP (HPC-LE-P; Shin-Etsu Chemical Co., Tokyo, Japan) were obtained commercially.
Cornstarch (Cornstarch W; Nihon Shokuhin Kakou Co.,
Ltd., Tokyo, Japan) was of food grade.

METHODS
Wet Granulation
Wet granulation was conducted in a 25-liter highshear mixer (Model FM-VG-25; Powrex Co., Osaka,
Japan) using the core formulation shown in Table 1.
Powders were weighed and added to the bowl of the

Validation of High-Shear Wet Granulation. I

195

Table 1
Formulation Compositionfor Wet Granulation

Drug Development and Industrial Pharmacy Downloaded from informahealthcare.com by University Of Wisconsin Madison on 07/19/13
For personal use only.

Component

Amount per Batch (g)

D w
Cornstarch
Lactose
Hydroxypropyl cellulose
Total

3000
252
588
120
3960

high-shear mixer. After premixing the powders for 1

min, purified water was added to the powders using a
peristaltic pump and a spray nozzle while both main
(250 rpm) and chopper (3000 rpm) blades were activated. The powders were then mixed for a suitable time.
Further, the powders were kneaded longer or additional
purified water was added and mixed for an additional
time. The scheme of water addition and mixing was
repeated several times. In each experiment, one batch
per drug was wet granulated.

Wet Granule Particle Size Analysis

A sonic sifter (Model Gilsonic AutoSiever GA-1;
Gilson Company, Inc., Ohio, USA) was used as the wet

granule particle sizing equipment. A wide range (106 to

1700 pm) of screens were selected to cover wet granulations of various drugs-i.e., 7 screens (1700, 850,
425, 250, 180, 150, and 106 pm)-and a latex fines
collector to serve as a conventional pan screen. It was
found through experience that routinely, wet granule
samples contain a few unrepresentative lumps which
presumably are granule agglomerates formed by aggregating, and the large agglomerates cause variances in the
wet granule particle size determination. Thus, a presieving step was employed before the sieving step. A
2000-pm screen was used for presieving because it was
observed in a preliminary study that most of the granules except lumps passed through the screen easily. At
several time points during the granulation process, wet
granules were sampled from the high-shear mixer chamber. The cumulative amount of water added and the
cumulative mixing time at each sampling time point in
the first experiment are listed in Table 2. In the first and
second experiments, about 30 g of wet granules were
taken from the middle granulation bed of position B
within the high-shear mixer bowl (diagramed in Fig. 1).
In the third experiment, about 5 g of wet granules were
individually sampled from the middle granulation bed of
positions A, B, and C (positions AM, BM, and CM,
respectively) for ethenzamide and acetaminophen; and

Table 2
Amount of Water Added and Mixing Time for Wet Granulation
Sampling
Time Point
1st
2nd
3rd
4th
5th
1st
2nd
3rd
4th
1st
2nd
3rd
4th
1st
2nd
3rd
4th

Ethenzamide

Acetaminophen

Acetaminophen-milled

Antipyrine

-~
~

~~~

~~~~

aBased on the total solid weight in the wet granulation.

Cumulative
Water Amount
(%)a

15.0
17.1
18.2
19.3
19.3
8.0
9.1
10.2
12.4
10.0
12.1
14.3
16.6
5.0
5.0
6.1
8.4

Cumulative
Mixing Time
(rnin)
8.8
11.0
12.7
14.3
15.3
4.0
5.4
6.8
11.0
6.5
8.7
10.8
13.1
3.8
6.8
9.3
13.4

Emori et al.

196

Drying

7 P
Main Blade

Drug Development and Industrial Pharmacy Downloaded from informahealthcare.com by University Of Wisconsin Madison on 07/19/13
For personal use only.

Blade
-

\\

In the first experiment, about 100 g of wet granules

were additionally sampled at the same time points for
the wet granule particle size determination. The wet
granules were immediately dried in a fluid-bed dryer
(Model Pulvis Mini-Bed GA22; Yamato Scientific Co.,
Ltd., Tokyo, Japan) using an inlet temperature of 80C.
Drying proceeded until a product temperature of 45C
was reached. Loss on drying (LOD) of the dry granules
was measured using an infrared moisture meter (Model
Kett FD-220; Kett Kagaku, Tokyo, Japan) at 80C. All
LOD results were less than 3 % . Dry granules were then
hand-screened through a 2000-pm screen.

Dry-Sized Granule Particle Size Analysis

Figure 1. Sampling positions of wet granules within highshear mixer bowl.

from the upper, middle, and lower beds of position B

(positions BU, BM, and BL, respectively) for acetaminophen-milled and antipyrine. The wet granule sample
was placed in a sealed container immediately after sampling.
Three grams of wet granules were placed on the top
of a nest screens (2000-pm screen, 7 screens, and latex fines collector), and passed through the 2000-pm
screen mildly and quickly by loosening the granule agglomerates with a spatula. The 2000-pm screen was
replaced with the top screen holder and the assemblys
diaphragm, and the whole assembly immediately placed
in the sifter unit. The sieving test was then run. In the
first experiment the sieving parameters were selected;
i.e., sieving times of 0.5, 1, and 3 min and intensities
of 20 % , 40 % , and 80% of full-scale amplitude were
compared. For the other two experiments, the selected
sieving time and intensity were used. Bottom and side
tapping, 0.2-min ramp-up time, and 0.2-min ramp-down
time were employed for all experiments.
For each granule sample in the first experiment, nine
measurements at different sieving conditions were completed within 60 min after sampling. To determine the
measurement precision, the particle size distribution was
measured in triplicate within 20 min after sampling. An
additional determination was conducted 60 min after
sampling in order to investigate the granule standing
time effect. Three granule samples taken from different
positions at each time point were also measured within
20 min after sampling.

The particle size distribution of dry-sized granules

was determined using the same method as the wet granule particle size analysis except that only 50% of fullscale amplitude was used for the sieving intensity. This
sieving condition was selected in a preliminary study
from the result that there was no difference in the particle size distribution of dry-sized granules among the
sieving intensities of 20%. 50%. and 80% of full-scale
amplitude.

RESULTS AND DISCUSSION

Selection of Sieving Parameters for Wet Granules
The extent of wet granulation is defined practically
on the basis of characteristics of the dry or dry-sized
granules. For selection of sieving conditions (sieving
time and intensity of a sonic sifter), the particle size
distribution of wet granules was compared with that of
dry-sized granules. In all cases, the wet granules were
larger than the dry-sized granules, due to breaking down
of granule agglomerate and/or granule itself in the fluidbed dryer and during the dry-sizing process. Since it has
been reported that the wet granulation progression involves a decrease in small particle size fraction and an
increase in large particle size fraction (l), the correlation between the wet and dry-sized granules was determined for the increasing fraction (up-fraction) and decreasing fraction (down-fraction), instead of comparing
wet and dry-sized granules for each individual particle
size fraction.
Figure 2(a) shows the changes in individual dry-sized
granule particle size fractions of ethenzamide with the

Validation of High-Shear Wet Granulation. I

197

70 r

50

.-

a
5
Drug Development and Industrial Pharmacy Downloaded from informahealthcare.com by University Of Wisconsin Madison on 07/19/13
For personal use only.

01st sampling

_i

EBB@
! 2nd
B@B@ 3rd

30

4th

0
Pan

106

150

180

250

425

850

1700

Pan

106

Screen Size (pm)

70

r
!

-0-

60

40
50

Over850pm

+ Under 180pm

.=.

50

3
E

a,

.r,

3rd

4th

5th

Sampling Time Point

----

250

425

1700

850

(b)

---.
'r,

30

Over 1700pm
-C- Under425pm
-0-

20

' 0O

2nd

180

Screen Size (pm)

LL

1st

150

1st

2nd

3rd

4th

Sampling Time Point

Figure 2. Changes in particle size distributions of dry-sized

granules of ethenzamide with wet granulation progression: (a)
changes in individual particle size fractions; (b) changes in upfraction and down-fraction.

up-fraction and down-fraction.

wet granulation progression. Fractions larger than 850

pm increased and fractions smaller than 180 pm decreased continuously as the sampling time point of wet
granules went. Therefore, fractions over 850 pm and
under 180 pm were defined as up-fraction and downfraction for the ethenzamide formulation, respectively.
Changes in the up-fraction and down-fraction are illustrated in Fig. 2(b), showing the granulation progression
that the small particle size fraction decreased continuously and disappeared by the fourth sampling, and that
the large fraction increased drastically after the third
sampling.
Results for dry-sized granules for the acetaminophen,
acetaminophen-milled, and antipyrine formulations are
shown in Figs. 3 to 5 . Fractions over 1700 pm and
under 425 pm, fractions over 425 pm and under 250

pm, and fractions over 850 pm and under 425 pm were

defined as up-fraction and down-fraction for acetaminophen, acetaminophen-milled, and antipyrine, respectively, because of their continuous increase and decrease
with the wet granulation progression [Figs. 3(a), 4(a),
and 5(a)]. Because the bulk drug particle sizes of acetaminophen (192.7 pm) and antipyrine (251.8 pm)
were large, there was little fraction smaller than 180 pm
in the dry-sized granule particle size distribution. Figures 3(b), 4(b), and 5(b) indicate that the granulation
process involved a continuous granule growth.
Similarly, the wet granule particle size data obtained
under various sieving conditions were analyzed. Next,
percent changes in up-fraction and down-fraction of drysized granules were compared with those of wet granules for the various sieving conditions. Correlations

Figure 3. Changes in particle size distributions of dry-sized

granules of acetaminophen with wet granulation progression:
(a) changes in individual particle size fractions; (b) changes in

Emori et al.

198
70

70

Drug Development and Industrial Pharmacy Downloaded from informahealthcare.com by University Of Wisconsin Madison on 07/19/13
For personal use only.

1st sampling
2nd
3rd
4th

a,

a
Pan

106

150

180

250

10

850 1700

425

Pan

106

Screen Size (pm)

loo

a,

2
a

1;

11

180

250

425

850

1700

Screen Size (pm)

I
8o
+Under425pm
Over850ym
60
+

loo

LL

+OUnder250ym
v e r

.
.-a

1st

150

2nd

3rd

4th

Sampling Time Point

1st

2nd

3 rd

4th

Sampling Time Point

Figure 4. Changes in particle size distributions of dry-sized

granules of acetaminophen-milled with wet granulation progression: (a) changes in individual particle size fractions; (b)
changes in up-fraction and down-fraction.

Figure 5. Changes in particle size distributions of dry-sized

granules of antipyrine with wet granulation progression: (a)
changes in individual particle size fractions; (b) changes in upfraction and down-fraction.

between the dry-sized and the wet granules of ethenzamide are shown in Fig. 6. All results of the four
granulations are also listed in Table 3. For the up-fraction of ethenzamide, stronger sieving conditions (40%
for 1 and 3 min; and 80% for 0.5, 1, and 3 min) provided very high correlation coefficients ( r = 0.99530.9991), and also for the down-fraction all conditions
showed a good correlation (r = 0.9019-0.9616).
Regarding acetaminophen-milled, only three stronger
sieving conditions (intensities of 20%, 40%, and 80%
of full-scale amplitude for 3 min) produced up- and
down-fractions of wet granules, whereas the clear
changes in up- and down-fractions of the dry-sized granules were observed. Continuous decrease in small particle size fraction and increase in large particle size fraction of wet granules with the wet granulation

progression were not observed when sieving times of

0.5 and 1 min were used. This suggests that the granulation of acetaminophen-milled proceeded to formation
of strong granules and strong granule agglomerates.
Two (20% and 80% of full-scale amplitude for 3 min)
of the three conditions gave good correlations between
the dry-sized and wet granules ( r = 0.9035-0.9899) for
both up- and down-fractions, although the reason for the
low correlation coefficient obtained at 40% intensity for
3 min is not known.
The up- and down-fractions of acetaminophen drysized granules were well correlated with those of wet
granules under all sieving conditions except the weakest condition evaluated (20% for 0.5 min), i.e., r =
0.9187-0.9902 for the up-fraction and r = 0.92050.9999 for the down-fraction. From these results, the

Validation of High-Shear Wet Granulation. I

100

80
60
40

Drug Development and Industrial Pharmacy Downloaded from informahealthcare.com by University Of Wisconsin Madison on 07/19/13
For personal use only.

v)

Q)
2

20

0
50

40

5 30
.-c
E 20
0

9
: o
3
c

Q)

10

50

40

30

b/
20%

40%

20

A
20%

15

10

Q)

0 25

20

15

40%

.o

10

20

199

20

80%

15

10

10

0 20 40 60 80

20

40

60

Percent Down-Fraction
Percent UpFraction
of Dry-Sized Granules
of Dry-Sized Granules
0 / 0 0.5 rnin, 0 /
1 rnin,
a / A 3 rnin.
Figure 6 . Correlations between dry-sized granules and wet

Therefore, correlations of geometric mean diameter

(Dso)of wet granules obtained at 20%, 40% or 80% of
full-scale amplitude for 3 rnin to D,,of dry-sized granules were determined. The correlation coefficients for
20%, 40% and 80% intensities for 3 min were 0.9710,
0.9977, and 0.9941 for ethenzamide; 0.9464, 0.9383,
and 0.9859 for acetaminophen; 0.7894, 0.9016, and
0.9586 for acetaminophen-milled; and 0.9918, 0.991 1,
and 0.9852 for antipyrine, respectively. For all of the
four granulations, the three wet sieving conditions, especially 80% intensity for 3 min. resulted in good correlations.
From the above results, an intensity of 80% of fullscale amplitude and a sieving time of 3 rnin were selected as general sieving parameters for the Gilsonic
AutoSiever to obtain wet granule particle size data reflecting the granulation progression of various drugs and
formulations. A sieving time as short as possible IS desirable to minimize the time the granulation process is
interrupted. For this purpose, the 0.5-min sieving time
is more useful than 3 min. However, it is more important for the sieving time to reflect granulation progression than to minimize interruption time. Further, the
3.4-min total sieving time (0.2-min ramp-up, 3-min
sieving, and 0.2-min ramp-down) is not long enough
from the practical standpoint during formulation development and scale-up.

Characterization of Sieving Parameters for Wet

Granules

granules of ethenzamide.
granule formation of acetaminophen appeared to be
weak, contrary to the granulation of acetaminophenmilled.
There were no remarkable differences in correlation
coefficient for the up-fraction of antipyrine among the
sieving conditions ( r = 0.8082-0.8962), and all except
weaker conditions (20% and 40% for 0.5 min) showed
good correlation coefficients for the down-fraction ( r =
0.9252-0.9747).
The evaluation results of wet sieving conditions are
summarized in Table 4. Each formulation had its own
ideal sieving conditions. In all of the four different wet
granulations, however, the strongest sieving condition
evaluated in this study (intensity of 80% of full-scale
amplitude and sieving time of 3 min) showed good correlations between wet and dry-sized granules for both
up- and down-fractions.
It is also hypothesized that wet granulation involves
an increase in mean particle size of component particles.

To characterize the sieving method consisting of the

selected parameters (intensity of 80% of full-scale amplitude and sieving time of 3 rnin), the measurement
precision was determined. Since there was a possibility
that the wet granule particle size distribution changed as
the granules were drying, the influence of granule standing time between sampling and sieving was also investigated. Further, the effect of sampling position of wet
granules within the mixer bowl was investigated.
Figures 7(a) to 7(d) show the reproducibility of wet
granule sieving using 80% intensity and 3-min sieving
time for ethenzamide, acetaminophen, acetaminophenmilled, and antipyrine, respectively. For each of two
wet granule samples at an early time point and at a late
time point in each granulation, triplicate measurements
were completed within 20 min after sampling. The particle size distribution was very reproducible for all
samples. Standard deviations of individual particle size
fractions were less than 5 % . Therefore, it was verified
that the measurement precision of wet granule sieving

Emori et al.

200

Table 3
Correlations Between Dly-Sized and Wet Granules for Up- and Down-Fractions
UP

Drug Development and Industrial Pharmacy Downloaded from informahealthcare.com by University Of Wisconsin Madison on 07/19/13
For personal use only.

Wet Sieving
Intensity and Time
Ethenzamidea
20% 0.5 min
1 min
3 min
40% 0.5 rnin
1 rnin
3 min
80% 0.5 min
1 min
3 min
Acetaminophenb
20% 0.5 min
1 min
3 min
40% 0.5 rnin
1 min
3 min
80% 0.5 rnin
1 min
3 min
Acetaminophen-milledC
20% 3 min
40% 3 rnin
80% 3 min
Antipyrined
20% 0.5 min
1 min
3 min
40% 0.5 rnin
1 min
3 rnin
80% 0.5 min
1 min
3 min

Down

Wet Fraction
(Over)

Slope

Correlation
Coefficient (r)

Wet Fraction
(Under)

Slope

Correlation
Coefficient (r)

1700 pm
1700 pm
850 pm
1700 pm
1700 pm
1700 p m
1700 pm
1700 pm
1700 pm

0.6493
0.6850
1.2192
0.5922
0.6309
0.5961
0.5848
0.5380
0.5275

0.8628
0.8867
0.8352
0.9382
0.9956
0.9991
0.9953
0.9982
0.9990

250 pm
250 pm
180 pm
250 pm
250 pm
180 pm
250 pm
250 pm
180 pm

0.1193
0.2045
0.3098
0.1272
0.2107
0.3502
0.1455
0.2361
0.2735

0.9343
0.9616
0.9019
0.9309
0.9524
0.9090
0.9522
0.9529
0.9482

1700 pm
1700 pm
1700 pm
1700 pm
1700 pm
1700 p m
1700 pm
1700 pm
1700 pm

2.1314
4.0278
4.358 1
5.4895
5.7928
6.2327
4.7072
5 4842
5.7838

0.7005
0.9187
0.9706
0.9796
0.9864
0.9902
0.9642
0.9593
0.9795

1700 pm
1700 pm
1700 pm
1700 pm
1700 pm
1700 p m
1700 pm
1700 pm
1700 pm

0.4500
0.8835
0.9276
1.1860
1.2445
1.3325
0.9507
1.1234
1.1858

0.6985
0.9506
0.9704
0.9985
0.9999
0.9990
0.9205
0.9285
0.9488

850 pm
850 pm
425 pm

0.4522
0.0785
0.3455

0.9899
0.7210
0.9035

425 pm
425 pm
425 pm

0.4719
0.2085
0.3634

0.9600
0.5719
0.9828

1700 pm
1700 pm
1700 pm
1700 pm
1700 p m
1700 pm
1700 pm
1700 prn
1700 pm

0.5369
0.5094
0.5098
0.4734
0.4614
0.4432
0.3930
0.3941
0.3798

0.8950
0.8927
0.8962
0.8172
0.8174
0.8243
0.8082
0.8193
0.8189

850
850
850
850
850
850
850
850
850

0.2193
0.3032
0.5934
0.1191
0.2684
0.4239
0.1896
0.3192
0.4155

0.8212
0.9400
0,9747
0.8279
0.9264
0.9400
0.9371
0.9294
0.9252

pm
pm
pm
pm
pm
pm
pm

pm
pm

aUp-fractionand down-fraction of dry-sized granules are over 850 pm and under 180 pm, respectively.
hUp-fractionand down-fraction of dry-sized granules are over 1700 pm and under 425 pm, respectively.
CUp-fractionand down-fraction of dry-sized granules are over 425 pm and under 250 pm, respectively.
dUp-fraction and down-fraction of dry-sized granules are over 850 pm and under 425 pm, respectively.

Validation of High-Shear Wet Granulation. I

20 1
Table 4
Evaluation of Wet Sieving Conditions

Drug Development and Industrial Pharmacy Downloaded from informahealthcare.com by University Of Wisconsin Madison on 07/19/13
For personal use only.

Wet Sieving
Intensity and Time
20%

40%

80%

Ethenzamide

Evaluationa
AcetaminophenAcetaminophen
Milled

Antipyrine

0.5 rnin
1 min
3 min
0.5 rnin
1 min
3 min
0.5 min
1 min
3 min

aA: both correlation coefficients between dry-sized and wet granules for up- and down-fractions are more than 0.9; B: one coefficient is more than 0.9 and another 0.8 to 0.9; C: both coefficients are 0.8 to 0.9; D: one or both coefficients are less than 0.8;
E: Up- and/or down-fraction is not produced.

A\

600

1200 1800 0

600

1200

1800

500 1000 1500 2000 0

Particle Size (m)

Particle Size (pn)

-0- Early Point (meanksd of 3 measurements),
-0- Late Point (meanltsd of 3 measurements),
-C-Early Point (60 min after sampling),

-tLate Point (60 min after sampling).

Figure 7. Effect of granule standing time on wet granule
particle size distribution: (a) ethenzamide, (b) acetaminophen,
(c) acetaminophen-milled, and (d) antipyrine.

500 1000 1500 2000

-0-

/+

AM,+

I+

B M , d /+

/+
BU,
/+
B M , U 1-0Open; 1st sampling, Close; 2nd sampling.

-G-

CM,
BL.

Figure 8. Effect of sampling position on wet granule particle

size distribution: (a) ethenzamide; (b) acetaminophen, (c) acetaminophen-milled, and (d) antipyrine.

Emori et al.

Drug Development and Industrial Pharmacy Downloaded from informahealthcare.com by University Of Wisconsin Madison on 07/19/13
For personal use only.

202

using the 80%-3 min condition is good as long as the

determination is completed within 20 min.
Additionally, samples taken for determination of
measurement precision were measured 60 rnin after
sampling. For all of the wet granule samples, the wet
granule particle size distribution determined 60 rnin
from sampling was similar to those determined within
20 min after sampling (Fig. 7). It thus appears that the
granule standing time effect is not a problem for the wet
granule sieving as long as the sieving is completed
within 60 rnin after sampling.
In each granulation, wet granules were sampled at
two time points and from three different positions within
the mixer chamber at each time point. As shown in Fig.
8, in all cases there was no difference in particle size
distribution among three sampling positions. Standard
deviations of individual particle size fractions were less
than 5 % and were similar to those for the reproducibility of wet granule sieving described above. Thus, it is
believed that the wet granule particle size distribution is
homogeneous within the mixer bowl. and that the sampling position does not influence the particle size determination.

REFERENCES
1.

2.

3.

H. G. Kristensen and T. Schaefer. Drug Dev. Ind.

Pharm., 13, 803 (1987).
D. E. Former, N. R. Anderson, and G . S. Banker, in
Pharmaceutical Dosage Forms: Tablets, Vol. 2 (H. A.
Lieberman and L. Lachman, eds.). Dekker. New York,
1981, Chap. 5.
B. M. Hunter and D. Ganderton, J. Pharm. Pharmacol.,
25S, 71P (1973).

4.
5.
6.
7.
8.
9.
10.
11.
12.
13.

14.

15.
16.
17.
18.
19.
20.

21.

22.

C. J. Flesch, Pharm. Ind., 40, 757 (1978).

H . Leuenberger, H. P. Bier, and H . Sucker, Pharm.
Tech., 3, 60 (1979).
H. P. Bier, H. Leuenberger, and H. Sucker, Pharm.
Ind., 41, 375 (1979).
H. Leuenberger, Pharm. Acta Helv., 57, 72 (1982).
H. Leuenberger and M. Durrenberger, Chem.
Rundschau, 35, l(1982).
H. Leuenberger, Acta Pharm. T e c h . , 29, 274 (1983).
H. Leuenberger, Manuf. Chemist, 5 , 67 (1984); 6, 53
(1984).
H. Leuenberger and G. Imanidis, Pharm. Tech., 10, 56
(1986).
J. Werani, Acta Pharm. Suec., 25, 247 (1988).
N. 0. Lindberg, L. Leander, H. Helgesen, B.
Reenstierna, and L. Wenngren, Acta Pharm. Suec., 11,
603 (1974).
D. N. Travers, A. G. Rogerson, and T. M. Jones, J.
Pharm. Pharmacol., 27s. 3P (1975).
N. 0. Lindberg, P. G. Nilsson, B. Reenstiema, and L.
Leander, Acta Pharm. Seuc., 14, 191 (1977).
N. 0. Lindberg, L. Leander, and B. Reenstiema, Drug
Dev. Ind. Pharm., 8, 775 (1982).
N. 0. Lindberg, Drug Dev. Lnd. Pharm., 10, 45 (1984).
S. R. Ghanta. R. Srinivas, and C. T. Rhodes, Drug
Dev. Ind. Pharm., 10, 305 (1984).
M. Kopcha. E. Roland, G . Bubb, and W. A. Vadino,
Drug Dev. Ind. Pharm.. 18, 1945 (1992).
K. Yamamoto, R. J. Wald, H . Emori, S. Narita, T.
Yoshizawa, T. Nishihata, and G. E. Amidon. Pharm.
Res.. 12, S159 (1995).
R. J. Wald, G. E. Amidon, D. J. Reits, F. J. McGill.
K. Yamamoto, H. Emori, and T. Nishihata, Pharm.
Res., 12, S160 (1995).
P. Holm, 0. Jungersen, T. Schaefer, and H. G.
Kristensen. Pharm. Ind., 45, 806 (1983).