- Illllll'lllllilllllllllfl'lllllllllflllll

3 1293 01093 7005

_ a-..

W
4’,"
3’." e ' ' r .1; '0‘?“ 7
THESIS , . '-
! '1 ..¢ " ‘-”‘" fir; .
p/ ('1 1‘ 3 ' ‘
. .~ , V I, e Mm- "‘ t :s .fi
war ‘ .; L'»..:.‘ (I'u 5CD
»- Iv iv we V’h' fit!

"7".“

Jr

éLV.
I

€3.35. ' ,4 V . ‘1 “1"?
“MW- ' wet ww-imy

‘\__.____.l_‘_______

5. 3" W I"
‘ ‘ a.
. t .1 . « q. ‘~)“ ‘ y” "'1. s l

L“;En~..~.-_

/"

 

This is to certify that the
thesis entitled
Test Methods for Evaluating
Moisture Permeation of Flexible

Unit Dose Packages

presented by

Michael D . Kentala

has been accepted towards fulfillment
of the requirements for

Master of degree in Packaging

Science

1% f. atoll/cw

Major professor

Date '7 //533

MS U is an anm've Action/Equal Opportunity Institution

04639

 

 

 

MSU

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from

 

 

 

LlBRARlES
.—__. your record. FINES will
be charged if book is
returned after the date
stamped below.
.4;
a ‘ B
v an
(m5 g '9‘ L

fififil 53:5 C33?”

 

/4//.._ Poesy

TEST METHODS FOR EVALUATING
MOISTURE PERMEATION OF FLEXIBLE
UNIT DOSE PACKAGES

By
Michael D. Kentala

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

School of Packaging
1983

ABSTRACT
TEST METHODS FOR EVALUATING

MOISTURE PERMEATION OF FLEXIBLE
UNIT DOSE PACKAGES

By
Michael D. Kentala

Many hospital pharmacies are involved with the
repackaging of drug products. A common type of packaging
machine for this purpose is the form-fill-seal. These
machines are used to repackage oral solid drugs which are
moisture sensitive. A problem that arises is how to
insure that the packages formed are good moisture barriers.

This study includes an evaluation of 2 commonly used
packages. The effects of external environment, seal
discontinuity, and defective material on the permeability
of the package were tested. Testing was done according
to the United States Pharmacopeial Convention test
procedures for permeation of single-unit containers.

In addition, a test method is introduced by which an
estimation of permeation, USP class, and package integrity
can be achieved.

Results indicate that seal disantinuities, pinholes
in materials, and increased relative humidity in the storage
area all cause increased package permeability. None of these
potential problems are readily discernable by the operator
without the use of tests. The test procedure developed

can detect these conditions before problems arise.

ACKNOWLEDGMENTS

Special thanks to my major professor, Dr. Hugh
Lockhart without whose guidance this thesis would not
have been possible. His countless hours of hard work,
genuine concern, and unselfishness will always be
remembered.

Thanks also to my committee members, Dr. Jack
Giacin and Dr. Robert Zustiak for their help and
direction.

Finally, thank you to Mr. Robert Fassezke and the
Ingham Medical Center Pharmacy Staff for the use of
their machinery and materials as well as their friendly

advice.

ii

TABLE OF CONTENTS

List of Tables

List of Figures

Introduction

The Repackaging Operation
Package Formation

Materials and Machinery

Test Methods

Discussion of Results
Conclusion and Recommendations
Package Test Method

Method of USP Classification

iii

iv

<2

H m) «a .p H

28
29
31

LIST OF TABLES

USP Classification of Packages
Storage Conditions for Package WVTR Measurements

USP Classification of Packages
Storage Conditions: 72°F, 75% RH

USP Classification of Packages
Storage Conditions: 72°F, 75% RH

Water Vapor Transmission Rates
Package A/Machine A 72°F, 50% RH

Water Vapor Transmission Rates
Package A/Machine A 72°F, 75% RH

Water Vapor Transmission Rates
Package A/Machine A 100°F, 85% RH

Permeation Rates at Various Storage Conditions

USP Classification of Packages with Seal
Defects 72°F, 75% RH

Moisture Gain in Package A with Pinhole
72°F, 50% RH

Estimation of USP Classification Based on
Desiccant Color Change

iv

15

17

18

19

19

20
20

22

23

31

LIST OF FIGURES

Moisture Gain at Various Storage
Conditions (Package A/Machine A)

Weight Gain of Defective and
Intact Packages

21

24

INTRODUCTION

The repackaging of oral solid drugs in single units
is widespread in hospitals across the United States (5).

One reason for this is that the institutional pharmacist
must repackage the drugs obtained in bulk containers into
single units to facilitate the use of a unit dose distri-
bution system (7). This process has its shortcomings,
however. There have been reports that pharmacists need more
information about the permeation characteristics of

packages produced in the pharmacy. Reamer and Grady state
that not only must moisture permeation of unit dose packages
be measured, but a separate verification of the integrity
of the heat seal must be done (11).

These problems were presented to the directors of three
institutional pharmacies in the Lansing, Michigan area to
obtain their views. Again, a concern was expressed by them
about the permeation characteristics of the repackaging
materials and the shelf life of the product after repackag-
ing. In addition, these pharmacists were also uncertain
about the integrity of the finished package. Presently all
the information regarding permeation is obtained from the
material supplier. Often this data is given for materials
only and not for fabricated packages.

The United States Pharmacopeia (USP) (14) provides a

2
method by which packages can be classified based on the
amount of moisture gained by a packaged desiccant. Packages
are classified from A through D as indicated in Table 1.
TABLE 1

USP Classification of Packages

 

 

Class Moisture Permeation (ngday)
not more than none
1 of 10 exceeds: exceeds:
A 0.5 1.0
B 5.0 10.0
C 20.0 #0.0
D Above criteria not met

This method provides a general guide to permeation rates and
a package supplier may use this classification when describ-
ing a package system. The ranges are quite large within each
class as well as from one class to another.

The moisture permeation of the package can also be
expressed as a Water Vapor Transmission Rate (WVTR). The
procedure for determining WVTR is similiar to the USP method
but it allows for measurements of moisture gain at several se-
quential time intervals. The moisture gain is then plotted
versus time and a statistical tool such as linear regression
may be applied to calculate the rate of permeation. In com-
parison, the USP method calls for a measurement of moisture
gain at only one time interval. This does not present a
good picture of what is occurring before and after that
point in time.

There still remains the Lansing pharmacists' concern

3
about package integrity. Even though the supplier provides
information on material permeation and package USP classi-
fication, how does one know if a good package is being
formed? Can this be monitored? If so, how?

This thesis will address the issues concerning the
pharmaceutical repackaging operation. Moisture permeation
and USP classification of selected packages will be determin-
ed. The integrity of the package will be altered to
determine the effects on WVTR and USP classification.
Finally, a test method will be presented by which the .
pharmacist can monitor the package system for moisture

permeation and package integrity.

THE REPACKAGING OPERATION
Because of the widespread repackaging of drug products,
many different machines and materials have become commer-
cially available (10,11). With this variety comes the neces—
sity for the pharmacist to make a decision about which
packaging process and material to select to efficiently
carry out the repackaging operation. A general overview of
a few packaging concepts will help the pharmacist get
started in the right direction.
Whenever a package selection process is initiated
there are 4 major areas to consider:
1. Package
2. Product
3. Environment
4. Machinery
PACKAGE
When selecting the package, one must have some knowledge
about both the product and the environment the product will
face after packaging. Once the decision has been made as
to which environmental factors the product must be protected
from, the process of selecting the proper package can
proceed.
PRODUCT
A determination must be made as to whether or not the
product needs a package. In the case of an oral solid drug
product, it may be necessary to protect the product from cer -
tain elements of the environment which may be detrimental

to that product. This is where proper packaging and prOper

L;

5
package selection comes into play. Drugs may be sensitive to
such factors in the environment as oxygen, water vapor, or
light or they may lose volatile components (1).

ENVIRONMENT

This study will concentrate on water vapor only since
this is the environmental factor causing most problems in
oral solid drug shelf life.

Many oral solid drugs are sensitive to water vapor.
There is a wide variation in water vapor present in storage
areas. In some pharmacies humidity and water vapor are
fairly well controlled at 30-50% relative humidity through-
out the year. In other pharmacies relative humidity may
vary from 15% in winter to 70% in spring or summer.

MACHINERY

In the repackaging operation, selection of material
and machinery go hand in hand. Decisions most likely will
be based on such factors as cost of machinery, cost of
materials, complexity of machinery, and degree of training
necessary for the operators. Repackaging in single units
may be done with different types of packages, such as strip
packages, blister packs, or skin packs (10,11). Strip
packages may be formed on horizontal or vertical form-
fill-seal machinery. All other things being equal, the
aesthetics of the package may be the deciding factor. However,
the performance of the package is the key issue that must
not be overlooked.

Bacon (4) has stated that there are definite advantages

6
to using form-fill-seal operations. Among them are: low
capital investment, low tooling costs, and minimum complex-
ity of equipment. In addition to cost effectiveness, the
form-fill-seal package is reduced in size to dimensions
which are only slightly larger than the product itself
resulting in material savings as well as decreased storage
space requirements. This is preferable to using pre-formed
pouches or pre—formed blisters. If transparent material is
used, product visibility is enhanced. This is important
because 85% of hospital nursing supervisors claim that product
visibility makes transparent packages preferable (4). Also,
specific product information can be imprinted on the package
while it is still on the packaging machine (4). The heat
sealing operation used to seal the package can be controlled
allowing for good package integrity. Thus, the three major
advantages of form-fill-seal are cost effectiveness, space
efficiency, and package integrity (4).

With the preceeding in mind, the study includes only
one type of package: the strip package. The strip package
was selected because:

1. it is typical.

2. it is in use in the Lansing area hospitals.

3. local pharmacists eXpressed concern about
whether or not the strip package is adequate
as a moisture barrier.

4. these pharmacists also eXpressed concern about
' seal integrity and how it could be monitored.

5. the Lansing area hospitals are'the basis of
this study because they comprise a coopera-
ting group with the School of Packaging for
study of these and other hospital packaging
questions.

PACKAGE FORMATION
Two methods of package formation will be discussed:
horizontal and vertical form-fill-seal strip packages.
The methods have some similarities. Among them are:

a) packages are formed from 2 separate webs of mater-
ial into a strip of 4-side-seal pouches.

b) packages are sealed by a hot bar coming in contact
with one side of the package and applying a force
to help construct the heat seal bond.

c) one tablet or capsule is placed in each package.

d) temperature of the hot bar is easily controlled
by means of a thermostat.

Differences also exist and they will be apparent as each
method of package formation is separately studied.

The vertical form-fill-seal, as the name implies,
involves a process whereby the material is fed into the
machine and the package is formed in a vertical orientation.
The tablets or capsules are dropped into each package by
gravity, usually being fed into the package by means of a
slotted, rotating wheel located at the top of the machine (3).
The sealing operation consists of 2 steps. The first step
seals one end of the package. The tablet then drops into the
pouch. In the second step the material is advanced and the
other end of the package is sealed, along with the sides and
end of the new package. Since one sealing step seals the ends
of 2 packages, the tablets will not fall into a previously
filled package as the drop. As the packages exit the machine
they all may be in one strip with perforations between each
package or they may be cut into separate pouches.

The horizontal form-fill-seal package is very similar

7

8
in formation. The webs of material are fed in and the package
is formed horizontally. Since the tablets cannot be gravity
fed, an operator will place one tablet or capsule in each
package. 0n the machine used in the study, a press made
an indentation in one web so that the tablet would have
a place to nest (4). The sealing Operation and formation
of the strips exiting the machine are the same as described

for the vertical form-fill-seal.

MATERIALS AND MACHINERY
Testing for water vapor transmission rates was
performed on two package types made of similar material,
fabricated on two different packaging machines. For
purposes of simplicity in the presentation and discussion
of results, the following designations will be used.
Package A: Materials supplied by Odessa Packaging
Services, Inc. Odessa, DE
Machine A: Mercury U-Pack
G-L Industries
Mercury Packaging Machine Corp.

Philadelphia, PA
Sold by Odessa Packaging Services Inc.

Type: Vertical F/F/S

Package B: Materials supplied by In—Pac Systems Inc.
Mount Vernon, OH

Machine B: In-Pac Machine, System III

In-Pac Systems, Inc.
Mbunt Vernon, OH

Type: Horizontal F/F/S

The materials used in the studies are laminations.
Both packages are constructed from two different materials.
Each package has a transparent material heat sealed to a
non-transparent material containing foil and paper lamin-
ated to polyethylene. Information regarding the composi-
tion of the materials was obtained from the respective
suppliers and is as follows:

Package A:

Transparent material: 250 gauge Saran coated
cellophane / 1.5 mil LDPE

Non-transparent material: 25 lb. paper / 0.5 mil LDPE /
.35 mil foil / 1 mil LDPE

9

10
Package B:

Transparent material: 195 gauge Saran coated cello-
phane / 1.5 mil LDPE

Non-transparent material: 35 lb. paper / 0.67 mil
LDPE / .35 mil foil / 1 mil LDPE

For the tests, packages were fabricated from each pair
of materials on that machine sold by the supplier of that
material. That is, Package A (Odessa) formed on Machine A
(Odessa), and Package B (In-Pac) on Machine B (In-Pac).
Then, to measure machine effect, each pair of materials was
run on the other suppliers' machine to see if any differ-
ences resulted. The combinations were Package A (0dessa)/
Machine B (In-Pac) and Package B (In-Pac)/ Machine A (Odessa).

This approach was taken because each of the suppliers
offers a machine/material system of his own selection.
The function and geometries of the machines were enough

different that we felt the crossover test chould be made.

TEST METHODS

The usual test methods of interest for hospital
repackaging are those for moisture permeation and for seal
integrity. A descussion of possible methods of measuring
these follows:

NDISTURE PERMEATION

Testing of moisture permeation through packages and
packaging materials is a very important step to insure that
good packages are produced which will protect moisture
sensitive products. Procedures are available to test both
materials and packages for their ability to serve as moisture
barriers. These tests usually involve the use of a desiccant
and a gravimetric method for measuring weight gain of the
desiccant. The procedure commonly used to measure the rate
of transmission of water vapor through materials is ASTM
procedure E-96 (Appendix A). Procedures for measuring
water vapor transmission into or out of containers include
ASTM D 3199 (Appendix A) and USP (671) Containers -
Permeation (Appendix A) (1.14).

When water vapor is being transmitted through a given
material or package there may be 2 mechanisms involved:
leakage and permeation.~ Leakage can occur when the vapor
passes through some discontinuity in a material or between
surfaces of materials in contact. Sources of these leaks
include pinholes in material or faulty seals. Both are
of concern when dealing with strip packages for drugs that
are moisture sensitive. Permeation, on the other hand,

occurs when the vapor passes through the material itself.
11

12
When dealing with moisture sensitive products, the goal is
to eliminate leakage and decrease permeation. Some ways to
minimize leakage are to (a) decrease the surface area of
the seal so the number and size of seal discontinuities
are decreased, (b) use a lamination which will help elimi-
nate pinholes since it is very unlikely that pinholes in
2 or more materials will coincide, (c) optimize the sealing
conditions of temperature, pressure, and dwell time to gain
maximum seal continuity and (d) avoid wrinkles or product
particles in the seal area (2).

Permeation rates can be reduced by (a) using thicker
material, (b) using a different material with superior
barrier properties, or (c) decreasing the surface of the
package (2).

SEAL INTEGRITY

Seal integrity can be measured as the strength of the
seal, or as the degree of continuity of the seal. In this
study, the second of these is treated as the property to
be measured.

Various methods of testing seals are cited in the lit-
erature. Miltz (8) describes a peel test using a tensile
testing machine. Reamer and Grady (12) discuss a method for
testing seals using thermal analysis techniques, and
Auslander (3) refers to a pressurized ammonia vapor test
for evaluating the integrity of package seals. All of these
test procedures require special equipment and would be
useful in the initial development of optimum machine settings.

They could also be used to confirm that Optimum sealing

13
conditions have been achieved. These procedures are time
consuming as well as costly in terms of machinery and labor
and may not be feasible in the institutional pharmacy, since
the pharmacy staff is normally totally occupied with dispen-
sing medications and monitoring medication regimens.
Furthermore, manufacturers of heat sealing equipment
provide reliable information concerning the machine settings
necessary to achieve prOper seals when using specified
material. For this particular application in the pharmacy,
seal strength alone is not necessarily a good measure of
package performance. The concern is how much moisture
enters the packages. A seal that tests as being "strong" by
a peel test may indeed leak. This can happen when there is
an open channel or discontinuity running through a wide
seal which is very strong. The effect of channels will be
demonstrated later. A "weak" seal may be continuous and be
a good moisture barrier. There should be a simple test by
which the pharmacist can monitor the seal integrity with
respect to moisture transmission of packages produced on
in-house machinery,thus avoiding the costly and painstaking
process of conducting a test specifically for seal

integrity of either kind.

PERMEATION

The water vapor barrier properties of the two package
types were determined using the USP method and a gravi-
metric method. In both tests the weight gain of a desiccant
pellet* was used to measure the moisture permeation of the
packages. The pellets were obtained from:

Hunt Sales Co.

13700 Bardon Rd.

Pheonix, MD 21131
These pellets were chosen because of the reference to them
in the USP (14). Other desiccants are available, but the
one selected was especially well suited to this testing
because of its size and shape. A unique feature of the
pellets is a change in color from blue to pink when they have
absorbed 10% of their weight in moisture. Since the pellets
range in weight from 250 to 300 milligrams, the color change
occurs when the pellets have gained 25 to 30 milligrams
of moisture.

In the determination of moisture permeation, blanks
were run with each test to account for any moisture gain by
the package.

Packages were tested under three storage conditions.

given in Table 2.

* Grade 944 Indicating Pellets, P/N 944-0%-X1746 from
Davison Chemical Division of W.R. Grace

14

15
Table 2

Storage Conditions for Package WVTR Measurements

Temperature Relative Humidity
72°F 50%

72°F * 75% *

100°F 85%

* USP Test Conditions

Conditions were chosen to represent ambient and acceler-
ated storage environments as well as the USP test condi-
tions. The USP classification of both Package A and B
appears in Table 3.

Next the materials were run on the other supplier's
machine. The combinations were Package A (Odessa)/Machine B
(In-Pac) and Package B (In-Pac)/Machine A (Odessa). These
results are presented in Table 4.

To compare the effects of different external environ-
ments on the moisture permeation of the package, Package A
was tested at the 3 storage conditions listed in Table 2.
These results are summarized in Tables 5,6, and 7 and
presented graphically in Figure 1.

LEAKAGE

Occasionally a package may be formed which is less than
acceptable. That is, the material may be defective or the
seal may be non-continuous. This should result in a
higher WVTR. To verify this assumption, packages were
intentionally damaged.

First, packages were formed having non-continuous seals

Itill'wfi .|.r

ll]! 10’ ulii .|

 

 

 

16
by inserting a common paper clip wire ( .032 inches in
diameter) into the seal area while the seal jaw closed.
The result was a channel through the seal area. The
packages were then tested by the USP method. The data
appears in Table 9. Both materials were run on Machine A
due to difficulty in creating the seal defect on machine B.
Because of the geometry of Machine B, there were physical
obstructions making it impossible to create the defect.
Also, Package A was tested with two channels to determine
what degree of seal discontinuity was needed to produce a
Class 0 package.

Secondly, to simulate defects in material, a hole was
made in the transparent side of Package A. The packages
were then stored at 720E, 50% RH and weighed periodically.
The results are in Table 10.

For ease of comparison between intact packages and

those with defects, refer to Figure 2.

TABLE 3

USP Classification of Packages
(Values represent moisture gain in mg/day)
Storage Conditions: 72°F, 75% RH

Package A/Machine A Package B/Machine B
0.7 3-3
0.7 4.1
0.8 6.3
1.3 3.1
0.7 4.4
5-8* 3.0
1.0 3.2
0.7 2.9
0.7 3.0
0-9 3-3
Average 0.8 3-7
Std Dev 0.2 1.1
USP

Class B . B

* bad seal - value not included in average

17

TABLE 4

USP Classification of Packages
(Values represent moisture gain in mg/day)
Storage Conditions: 720F, 75% RH

Package A/Machine B Package B/Machine A
0.4 3.6
0.7 3.6
0.4 3.4
0.4 3.4
0.4 15.2*
0.6 3-5
0.4 3.6
0.4 3.6
0.3 3.8
0.4 3.4

Average 0.4 3-5

Std Dev 0.1 0.1

USP

Class B B

* bad seal - value not included in average

18

TABLE 5

Water Vapor Transmission Rates (mg/pkg)
Package A/Machine A
72°F 50% RH

Time (days)
1 3 5 7 12

Average
weight gain* 0.04 1.4 1.9 2.6 4.6

Std Dev 0.07 0.1 0.2 0.2 0.2

Desiccant

color: Blue Blue Blue Blue Blue
TABLE 6

Water Vapor Transmission Rates (mg/pkg)
Package A/Machine A
72°F 75% RH

Time (days)
1 2 3 4 6

Average
weight gain* 1.7 1.0 2.4 3.0 4.2

Std Dev 0.2 0.4 0.7 0.9 1.5
Desiccant
color: Blue Blue Blue Blue Blue

* average of 10 samples

19

17

6.2

0.2

Blue

13

9-9

2.9

Blue

TABLE 7

Water Vapor Transmission Rates (mg/pkg)
Package A/Machine A
100°F 85% RH

Time (days)

1 2 3 4
Average
weight gain* 3.2 8.9 12.4 19.2
Std Dev 0.4 0.4 0.3 0.4
Desiccant
color: Blue Blue Blue Pink

* average of 10 samples

TABLE 8

Permeation Rates at Various Storage Conditions

Storage Condition Permeation Rate USP Class I”
(mg/day)
72°F 50% RH 0.4 B 0.040
72°F 75% RH 0.7 B 0.046
100°F 85% RH 5.2 B 0.124

* P'is the permeability constant expressed in the units:

mg H20 x pkg.

 

day x mmHg

20

Figure 1

Moisture Gain at Various
Storage Conditions

Package A/Machine A

 

 

:c:
a:
EL.
0 SR
010
bAn
:1:
tr
EL.
0 Si
an
54\
:1:
lit-«II:
0
OR
Otn
«up
I J _1 l l
l l l j l
o \o N a) .3
N H H

15 18 21

12
Time (days)

TABLE

9

USP Classification of Packages with Seal Defects
(Values represent moisture gain in mg/day)

72°F

Package A/Machine A

(one channel)

Average

Std Dev

USP
Class

.2

A) k» c— tr k0 -¢ ¢- tr k0 -¢
(DKRO\H\OONO\3

3-9
0.5

B

Package A/Machine A

75% RH

(two channels)

22

6.0
5.0
11.4
12.2
13.0
6.3
12.5
6.7
13.0
14.2

10.1
3-7

Package B/Machine A
(one channel)

5.4
6.1
8.5
6.7
11.8
7.6
6.8
6.8
8.9
7.6

7.6
1.8

Average
weight gain*
Std Dev

Desiccant
color:

TABLE 10

Moisture Gain (mg) in Package A
with Pinhole
72°F 50% RH

Time (days)

1 2

3.6 7.4
0.3 0.8
Blue Blue

* average of 10 samples

23

L.

13.2

Blue

15.7

Blue

19.2

Pink

Figure 2

Weight Gain of Defective
and Intact Packages
Package A/Machine A

ma

 

xmsmm
:msom
:msmm
:msom

emu
emu
ea.
emu

Amzmuv mafia

ma NH m

mommxomm poMPCH o
wmwmxomm pomysH .
manom CH maoccmco 4

Hmflampms CH moH052flm x

 

NH

we

3N

on

DD
8

BQHWSP WNHC

24

DISCUSSION OF RESULTS

USP CLASSIFICATION VS WVTR

According to data in Table 3, and the classification
in Table 1, packages A and B have the same USP classifi-
cation in spite of the fact that Package B has four and
one half times greater rate of gain than Package A. This
difference is not explained by differences in material
unless Package A has a thicker coating of Saran on the 250
cellophane. The hydroscopic nature of cellophane is such
as to indicate that thicker cellophane alone should not
account for this large difference. Note that packages A
and B have the same USP classification, but Package A
will provide a shelf life perhaps 4 times longer than
Package B.

In order to gain an insight into what moisture per-
meation occurs due to the environment, Package A was
stored at the three conditions presented in Table 2.
The results (Table 8) show that as external relative
humidity increases, so does the permeation rate. At the
constant temperature of 72°F, the increase in relative
humidity from 50% to 75% causes a small rise in permeation
rate from 0.4 to 0.7 mg/day/package. Taking into account
that the balance used can measure only to the nearest 0.1
mg, these results are very close. However, when the package
is stored in an extremely severe environment, 100°F and 85%
RH, the permeation rate increases more than five-fold. This
demonstrates the importance of monitoring or controlling

storage conditions.

25

CROSSOVER TEST OF MATERIAL ON OPPOSITE MACHINE

From Table 4, Package B has about the same WVTR on
either machine. Package A has an average WVTR about one
half the value obtained on Machine A. However, if the
standard deviation is taken into account, it must be recog-
nized that Package A is probably about the same when pro-
duced on either machine.

LEAKAGE INDUCED BY DEFECT IN PACKAGE 0R MATERIAL

While package permeability will vary based on the
external environment, defects in materials or seal areas
will also cause changes in permeability. The data in Table
9 helps to illustrate this statement. Package A has a
permeability of 0.8 mg/day/package under USP test conditions.
With the introduction of one channel in the seal area the
permeability increases by 3.1 mg/day, or nearly four-fold.
Two channels caused an increase of 9.3 mg/day (twelve-fold)
and the package is no longer a Class B but a Class C package.
Meanwhile, Package B with one channel has an increase in
permeability of 3.9 mg/day and also becomes a Class C
paflfige. Material defects cause the same effect. Package
A has an increase of 2.2 mg/day when one hole is introduced
into the material and the package is stored at ambient
conditions (72°F, 50% RH). Package A has a permeability of
2.6 mg/day as determined by linear regression of the data
in Table 10, whereas from Table 8 one can see that the
same intact package has a permeation rate of 0.4 mg/day.

The interesting aspect of the previous discussion is

that these defects are not readily visible. Even upon

26

27
careful inspection of each package the defect is difficult
if not impossible to spot. Since the machines used in
this study operate at 50-60 packages per minute, the
Operator could not possibly inspect each package. Even if
100% inspection were desirable it wouldn't be practical.
Thus, the need is established for a way to identify

defects and consequently prevent potential problems.

CONCLUSION AND RECOMMENDATIONS

Many problems may arise when attempting to package a
moisture sensitive product. The problems could result in
a more than desirable amount of moisture entering the pack-
age and affecting the product. The pharmacist has guide-
lines available to assist in the repackaging operation
(9.13). These guidelines do not provide a means for
evaluating the integrity of packages produced in the pharma-
cy. The pharmacist can also obtain information from his
machine and material supplier about machine settings and
material USP classification or WVTR. However, these do not
provide a means to monitor the integrity of packages as
they are produced.

Based on data taken, and observations made during
testing, two methods have been devised for the purpose of
evaluating the integrity of packages as produced in the
pharmacy. Furthermore, they provide a means for estimating
the shelf life (or monitoring the approach of the end of
shelf life) of products packaged in the hospital pharmacy.
These methods have the advantage that they do not require
the pharmacist to apply time to learn and to use skills
which are not a part of his primary duty. These methods,
properly applied, should allow the monitoring of package
integrity and shelf life on a routine basis, using only

a minimum of time.

28

PACKAGE TEST METHOD

With the possibility of problems due to changing
environment, non-continuous seals, or defective material,
it is necessary to have a way to determine what kind of
moisture protection the package will provide. There is no
replacement for the standard water vapor transmission
testing of packages in determining moisture barrier proper-
ties. In the hospital pharmacy, however, this process can be
time consuming and costly, especially if done on a frequent
basis. A simple test to give an estimation of WVTR and/or
package USP classification would be beneficial.

The desiccant pellets used in these studies would be
ideal for such a test. Since the pellets turn from blue to
pink as moisture is absorbed, a simple visual test could be
used to monitor package permeability. Such a test has been
devised using the results of this study. This test provides
an on—line check without extensive use of weighing devices
on a regular basis, thus saving time and money.

The test procedure is as follows:

1. Place one desiccant pellet in a package at the
beginning, middle, and end of each run.

2. Identify each of these three packages and record
its weight to the nearest 0.1 mg (1.0 mg is accept-
able but 0.1 is preferred.)

3. Identify and weigh a blank package without a pellet.

4. Store the packaged desiccant and blank in the same
area where packaged drug products are stored.

5. Daily observe the pellets until a pink color is
apparent. NOTE: Controls must also be set up:
that is, one pink and one blue pellet to help
distinguish minor color changes.

29

30
6. Weigh the package and blank.
Once these weights have been determined, the following

calculation will estimate the moisture permeation.

PI = Initial weight of package and desiccant (mg)
PF = Final weight of package and desiccant (mg)
BI = Initial weight of blank (mg)

BF = Final weight of blank (mg)

D = Weight gain of desiccant (mg)

T = Time (days) from initial to final weighing

( PF - BF ) - ( PI - BI ) = D

D

T = moisture permeation (mg/day)

This is an estimate of permeation since only an
initial and final weighing are used. Ideally, several
weighings should be done and moisture gain versus time
plotted. Linear regression of these points would provide
a more realistic WVTR ( slope of the line ) (1).

An example can be taken from Table 10. The pellets
stored at 72°F and 50% RH turned pink after 7 days. The
calculations yield a moisture gain of 19.2 mg over this
period. Dividing 19.2 by 7 gives a permeation rate of
about 2.7 mg/day/package. “The same package with no pin-
holes (Table 8) had a permeation rate of 0.4 mg/day/package.
This would alert the pharmacist to a possible problem for
the drugs packaged the same day as the pellets. Obviously
some initial studies must be done to establish baselines
for non-defective packages. Once the time has been es-

tablished for the desiccant in a good package to turn pink,

31

the weighing may be eliminated or reduced.

METHOD FOR USP CLASSIFICATION

A simple test method can be used to get a rough
estimate of USP classification. By noting the time re-
quired for a packaged desiccant pellet to turn pink, a
chart such as Table 11 can be used to estimate the USP
classification.

TABLE 11

Estimation of USP Classification Based
on Desiccant Color Change

7201“ 75% RH
Time (days) USP Class
40 or more A
4 - 40 B
1 - 4 C
Less than 1 D

The assumptions made in constructing Table 11 are:
1. The desiccant pellet weighs close to 200 mg.

2. The desiccant turns from blue to pink when it
gains 20 mg of moisture (10% by weight). This
is true for the desiccant used in this study.

3. Any testing to verify USP classification or WVTR
will be conducted according to the specific
procedure. ,

APPENDIX A

q r, Designation: D 3199 - 73'

Standard Test Method for

WATER VAPOR TRANSMISSION
THROUGH SCREW-CAP CLOSURE LINERSl

This Standard is issued under the fixed designation D 3W9: the number immediately following the designation indicates the
year of original adoption or. in the case of tension. the year of last I'CVISIOII. A number in parentheses indicates the year of last

reapproval.

 

'Non—The title was changed editorially in January 1977.

l.Scope

L] This method covers the measurement of
the barrier efficiency of screw- cap liners against
water vapor transmission either into or outward
from a container capped under standard condi—
tions.

2. Summary of Method

2.] The method involves periodic weighing
of containers sealed with screw caps containing
the liners to be evaluated to determine moisture
gain or loss under controlled conditions of
temperature and humidity.

Significance

3.1 This test may be used to compare or
screen new or existing closure liner materials by
their performance in a standard test.

3.2 it also may be used to indicate whether
the liner material will be satisfactory for use
with moist products. dry products. or both
types of products. When this method is used to
assist in the choice of liner material for a
specific use. the choice of containers and clo-
sures should simulate the conditions of use as
closely as practicable.

3.3 The method may be used to establish
performance specifications.

4. Apparatus

4.! Bottles. l:oz (approximately l5 cm“)
with a 28 mm Glass Container Manufacturers
Institute (GCMI) 400 or 405 finish. Select test
bottles for uniformity of flatness of finish.
width of lip sealing surface. and freedom from
splits. chips. or cracks.

4.2 Screw Caps. 28 mm continuous thread
(CT) of either metal or plastic. ,

4.3 Torque Tester’. to determine application
torque. accurate to within a: ' 2 lbf-in. (0.06
N -m).

4.4 Humidity Cabinet. capable of control to
t I°F (:0.6°C) and :2‘}: relative humidity at
IOO°F (38°C). Ranges required are 25 and 75%
relative humidity.

4.5 Analqi'tical Balance. with an accuracy to
at least 0.001 g.

4.6 Pipete or Graduate. capable ofdelivering
IO ml of water with an accuracy of $0.1 ml.

5. Reagent

5.! Activated silica gel (Grade H. Type 46.
I6 mesh) anhydrous calcium chloride. or other
suitable desiccant material.

6. Test Specimens

6.! The closure-liner test specimen should be
die-cut by either standard manufacturers'
screw-cap lining equipment or by a single
punch. using care that the liner is made uniform
and without distortion of the liner facing.

6.2 No torn liners or wrinkled liners should
be used. White. Virgin pulp hacking should be
used for all tests. Prior to use. the liner backing
combination must be laminated together and

‘This method is under the iurisdictmn of ASTM Committee
D-Itl tin Packaging

('urrent edition approx-cu -\pti| 27. W73. Published May
'97]

‘Oucns-Illinois torque tester. .i\.iil.ihle from 0‘6!“-
lllin-vis. Inc, (ILIss (_ untamer Du . P 0 Bot I015. Toledo.
()h... 41min. or equualcnt. has been Iound satisfactory for
Ibis method.

634

 

iilb

adhered to the cap. if plastic: or forced into the
cap. if metal. At least ten test specimens and
ten control specimens of each liner to be
evaluated should be used for this test.

7. Procedure

7.1 Water Vapor Transmission Outward
From 0 Container:

7.1.1 Tare weigh each '2-oz (28-mm) glass
container together with its respective lined
closure. Use five containers for this portion of
the test.

7.1.2 Place l0.0 ml of distilled water in each
tared glass container.

7.l.3 Apply the screw caps carrying the test
liners with an application of 15 lbf-in. (approxi-
mately 1.7 N.'m).

7.1.4 Assemble with the caps as in accord-
ance with 7.1.1 and 7.1.3 and weigh five empty
bottles for use as controls. Correct the test
results by subtracting the average mass loss
observed in the control containers from the
average mass loss in the test containers.

7.1.5 Place the containers upright in the
humidity chamber at 100:I°F (38:0.6°C)
and 25 12% relative humidity.

7.1.6 Seven days after starting the test. begin
weighing each test container to the nearest
0.001 g at weekly intervals for 4 weeks. Allow
the containers to come to room temperature
before weighing.

7.1.7 Replace the test containers in the hu-
midity cabinet immediately after each weigh-
ing.

7.2 Water Vapor Transmission Into 0
Container:

7.2.1 Tare weigh each '2-02 (28-min) glass
container together with its respective lined
closure. Use five containers for this portion of
the test.

7.2.2 Place 10.0 g of activated silica gel or
other suitable desiccant in each container.

D 3199

7.2.3 Apply the screw caps carrying the test
liners with an application torque of 15 Ibf-in.
(approximately 1.7 N-m).

7.2.4 Assemble with the caps in accordance
with 7.2.1 and 7.2.3 and weigh five empty
bottles for use as controls. Correct the test
results by subtracting the average mass gain
observed in control containers from the average
mass gain in the test containers.

7.2.5 Place the containers upright in the
humidity chamber at 100 a; I°F (38 a: 06°C)
and 75 :29} relative humidity.

7.2.6 Seven days after starting the test. begin
weighing each test container at weekly intervals
for 4 weeks. Allow the packages to come to
room temperature before weighing.

7.2.7 Replace‘the test containers in the hu-
midity cabinet immediately after each weigh-
inc.

8. Report

8.1 The report shall include the following:
8.1.1 Material of construction of the liner
and whether adhered to or only forced into the

l Material of construction of the cap.
1. ame of the desiccant used.

'1 umber of replicates and controls.
.1 5 Time interval or intervals of the test.
.lih Temperature and humidity ofthe test.

8.1.7 Mass losses and gains in tabular form.
and their averages.

8.1.8 Average mass gain and loss and the
range. in milligrams corrected for the average
mass gain and loss of the controls.

8.1.9 Calculation of daily gain or loss (op-
tionall.

8.1.10 Graph on arithmetical paper of mass
gain or loss versus time (optional).

8.1.11 Evaluation of comparative results if
more than one liner construction was tested.

8.1.12 Statement that this test was done in
accordance with ASTM Method D 3199.

N
N

I
o

The American Society for Testing and Materials takes no position respectth the i-aliditv ofunv patent rights asserted
in connection with any item mentioned in this standard. Users of this standard are expressly advised that determination
of the validity of any such patent rights. and the risk of infringement Ufsut'h rights. is entirely their own responsibility.

This standard is subject to revision or any time by the responsible technical committee and must be reviewed ercrv five
years and if not revised. either reopproved or withdrawn. Your commcnn are invited either for revision of this standard or
for additional standards and should be addressed to AS TM Headquarters. )nur comments will receive careful consideration
at a meeting of the responsible technical committee. which you mai- attend. If mum: that your comments have not received
a fair hearing you should make your views known to the ASTM Committee on Standards. Ivln Race St. Philadelphia. Pa.
thli. which will schedule a further hearing regarding your comments. I'uiling solo/union there. you mm- appcul to the

A 5 TM Board of Directors.

635

.-1
. t] r) Designation: E 96 - 66 (Reepproved 1972)

Standard Test Methods for

WATER VAPOR TRANSMISSION OF MATERIALS

1N SHEET FORM1

'iis Standard is issued under the fixed designation E96: the number immediately following the designation indicates the
..'.:r of original adoption or. in the case of revision. the year of last I'CVhlun. A number tn parentheses indicates the year of

st reapproval.

5. Scope

1.1 These methods cover determination of
fine rate of water vapor transmission (WVT)
'-l' materials in sheet form. The methods are
applicable to materials such as paper. plastic
:ilms. and sheet materials in general. The
:nethods are most suitable for specimens ’l.
"'1. (3.18 mm) or less in thickness. but may be
used with caution for somewhat thicker speci-
mens. For specimens of materials used in
building construction and similar materials
greater than 'I. in. in thickness. see ASTM
.‘vlethods C 355. Test for Water Vapor Trans-
mission of Thick Materials.2 Six procedures
are provided for the measurement of trans-
mission under different test conditions. as
follows:

l.l.l Procedure A—For use when the ma-
terials to be tested are employed in the low
range of humidities.

l.l.2 Procedure B—For use when the ma-
terials to be tested are employed in the high
range of humidities. but will not normally be
“cited.

l.l.3 Procedure BW—-For use when mate-
rials to be tested may in service be wetted on
one surface but under conditions where the
hydraulic head is relatively unimportant and
moisture transfer is governed by capillary and
water vapor diffusion forces.

1.1.4 Procedure C—Conducted at an ele-
vated temperature for use with materials em-
ployed in the low range of humidities. and
intended to shorten the time of testing of
highly impermeable materials. This procedure
eliminates. in most cases. the need for refrig-
eration.

1.1.5 Procedure D~Conducted at an ele-

vated temperature for use with materials em-
ployed in the high range of humidities. but not
normally wetted. and intended to shorten the
time of testing of highly impermeable materi-
als employed in this range. This procedure
also eliminates. in most cases. the need for
refrigeration.

1.1.6 Procedure E—For use in measuring
the WVT at an elevated temperature. with a
very low humidity on one side of the sheet and
a high humidity on the other side.

NOthl--The values stated in US. customary
units are to be regarded as the standard. The metric
equivalents of U.S. customary units may be approx-
imate.

2. Summary of Methods

2.1 The material to be tested is fastened
over the mouth of a dish. which contains ei-
ther a desiccant or water. The assembly is
placed in an atmosphere of constant tempera-
ture and humidity. and the weight gain or loss
of the assembly is used to calculate the rate of
water vapor movement through the sheet
material under the conditions prescribed in
Table 1.

3. Definitions

3.1 rate of water vapor transmission. W VT
—of a body between two specified parallel sur-
faces- the time rate of water vapor flow nor-
mal to the surfaces. under steady conditions.
through unit area. under the conditions of

' These methods are under the jurisdiction of ASTM
Committee (3'3 on Durability tit .Nonmetalltc Materials.

Current edition etlecme Sept. 2“. l9hh. Originally is-
sued I953. RL'Dltth's h 9h - 1‘3.

" Annual Iimik o].-1.87.\I Standards. Part 18.

739

vmvuwfi

. - . —-——-

15119

test. An accepted unit of WVT is l g/24 h - m.2
The test conditions must be stated.

3.2 water vapor permeance—of a body
between two specified parallel surfaces—the
ratio of its WVT to the vapor pressure differ-
ence between the two surfaces. An accepted
unit of permeance is a metric perm. or 1
g/24 h-m2 -mm Hg. Since the permeance of a
specimen is generally a function of relative
humidity and to a lesser extent temperature.
the test conditions must be stated.

3.3 water vapor permeability—the product
of its permeance and thickness. This property
may vary with conditions of exposure. An
accepted unit of permeability is a metric
perm—centimeter. or 1 g/24 h -m2 -mm
Hg-cm of thickness. Since the permeability
of most materials is a function .of relative
humidity and to a lesser extent temperature.
the test conditions must be stated.

Non 2—The definition does not imply that
permeance is inversely proportional to thickness
when the material is not fully homogeneous.

3.4 The following US. customary units are
also widely used:

For WVT: l grain/h-ft".

For permeance: 1 perm. or 1 grain/hit2
in. Hg.

For permeability: l perm-inch. or 1 grain/
h - ft2 in. Hg - in. of thickness.

Metric-US. customary conversion factors
are stated in Table 2.

4. Significance

4.1 The purpose of these tests is to obtain.
by means of simple apparatus. reliable values
for the rate of water vapor transmission of
sheet materials. expressed in suitable units.
Correlation of test values with any given use
must be determined by experience.

4.2 In any application. the vapor transmis-
sion of a sheet is determined by its permeance
under the given conditions. With many mate-
rials the permeance of a sheet increases
markedly when the relative humidity on ei-
ther or both sides is raised. In contrast. the
effect of a moderate change of temperature is
quite small. Consequently. Procedures A and
C. each with an average relative humidity of
25 percent. yield about the same permeance.
Procedures 8 and D. with 75 percent average
relative humidity. yield a different pair of

740

596

similar values. The broad range of relative
humidity provided by these two pairs of pro-
cedures (sometimes referred to as the desic-
cant and water methods. respectively) indi-
cates in a practical way the permeance of a
given sheet in many applications.

4.3 Procedure E provides the highest WVT.
but the permeance generally falls between the
values obtained by the desiccant and water
methods. Although it has special applications.
it is generally not as revealing as the water
method for severe relative humidity expo-
sures.

4.4 In Procedure BW. the water cup is in-
verted to wet the specimen. but otherwise it is
the same as Procedure B. For some speci-
mens. capillary transfer will be increased. It is.
intended specifically for‘those applications
where wetting of one surface over extended
periods is anticipated.

4.5 The report of a test result as a perme-
ance value (rather than WVT) largely elimi-
nates a fault in the test space conditions al-
lowed in the tolerances stated in 5.2. provided
the actual vapor pressure in the space is prop-
erly determined and used in the calculation
as required in 15.2. Also. the permeance value
of a sheet is a rational basis for evaluating its
performance and comparing various sheets for
a given applicati it. Therefore. values of vapor
permeance both in specifications and in re-
porting test results are to be encouraged.

4.6 The calculation of permeability
(permeance X thicknefi) is also logical when
the sheet is unquestionably homogeneous.
with smooth surfaces and uniform thickness.
The value so obtained can be used for calcu-
lating the permeance of a different thickness
of like material in the same exposure. In Ta-
ble 2 permeability units and metric-US. cus-
tomary conversion factors are shown. but no
report of the permeability is required.

5. Apparatus

5.1 Test Dishes—-Open-mouth dishes shall
be of such size or shape that they can be ac-
commodated readily on the pan of an analyt-
ical balance. The dishes shall be constructed
of a noncorroding. nonpcrmeable material.
and shall be as light as is consistent with the
necessary rigidity. The area of the Opening
shall be as large as practical and at least 30

all)»

cm". The area of the desiccant or water in the
dish shall be at least equal to the exposed area
of the specimen (see Procedures A. B. BW, C,
l). and E. Sections 9 to 14). There shall be no
obstructions within the dish that would re-
strict the flow of water vapor between the
specimen area and the water or desiccant in
the dish. For examples of suitable dishes and
methods of sealing the specimen to the test
dishes. see the Appendix.
.2 Test Chambers:

5.2.1 The room or cabinet where the as-
sembled test dishes are to be placed shall have
controlled temperature and humidity which
are measured frequently or preferably re-
corded continuously. The air conditions at any
test location shall be one of those shown in
Table l and shall be held constant within
temperature limits of $0.6C (1 F) and rela-
tive humidity limits of :2 percemt.

1}.

NOTE 3—These tolerances permit a range of va-
“or pressure up to 11 percent of nominal (tn 'Proce-
sures A and B) when temperature and relative hu-
midity peaks occur together. This variation can be
accepted only when control cycles are short and the
"tean vapor pressure is measured with due regard
:or such variation.

5.2.2 Air shall be continuously circulated
throughout the chamber with a velocity suffi-
Cient to maintain uniform conditions at all
:est locations. lts velocity over the specimen
in meters per minute shall be (numerically)
101 less than five times the permeance of the
~pecimen. in metric perms (fpm = 10 times
:‘ermSL but not over 150 m (500 ft)/min. Suit-
.iale racks shall be provided on which to place
1.1: test dishes within the test chamber.

\nrt 4—1f the dishes are to be placed on the

.exs in an inverted position. care must be taken
it the. specified air velocity over the surfaces of
' .,- specmtens is maintained.

5.3 Balance and Weights—The sensitivity
‘: the balance shall be not less than 1 percent
3 the weight change during the period when a
'Tz‘udy state is considered to exist. The
*‘thhts used shall be accurate to 1 percent of
'5; “eight change during the steady-state pe-
'~)d. A light wire sling may be substituted for
7.: usual pan to accommodate a larger and
J‘mer load.

-‘--1 Weighing Carers—Weighing covers for
J test dishes shall be provided if the dishes

Mi

have to be removed from the test room or cab.
inet for weighing (see the Appendix).

5.5 Template—A template may be used for
defining the test area and effecting the wax
seal (see the Appendix).

6. Materials

6.1 Sealant—The sealant used for attach-
ing the specimen to the dish. in order to be
suitable for this purpose. must be highly re-
sistant to the passage of water vapor (and
water). It must not lose weight to. or gain
weight from. the atmosphere in an amount.
over the required period of time. that would
affect the test result by more than 2 per-
cent. It must not affect the vapor pressure
in a water-filled dish. Molten wax or asphalt
is required for specimens having a permeance
of less than 2.5 metric perms. The sealant
shall be so applied that the test area of the
specimen is well-defined and the same on both
sides. and that no leakage of water vapor can
occur at or through the edges of the specimen.

6.2 Desiccant (for Procedures A. C. and E)
-—A desiccant having a high affinity for water
vapor and a high drying efficiency. that is. giv-
ing a low water vapor pressure after absorbing
a‘ large amount of water. The desiccant shall
remain essentially unchanged in physical con-

. dition and exert no chemical or physical ac-

tion. other than dehydration effects. on test
specimens with which it is in contact. A suit-
able desiccant is anhydrous calcium chloride
in the form of small lumps that will pass a No.
8 (2.36-mm) sieve. and free of lines that will
pass a No. 30 (600mm) sieve.’ It shall be
dried at 200C (392 F) before use and the
moisture gain during the test shall be limited
to 10 percent of its starting weight. If CaCl:
will react chemically on the specimen. an ab-
sorbing desiccant such as silica-gel. activated
at 200C. may be used but the moisture gain
by this desiccant during the test shall be lim-
ited to 4 percent.

‘7. Sampling
7.1 The material shall be sampled in ac-

Detailed requtrements for these sieves are given in
.-\ST \1 Specification 1', 11. for \VII’C'CIOIh Sieves for Test-
ing Purposes. which appears in the Annual Book of ASTM
Standards. Parts 14. 125. 2b. 30. and 41.

481p

cordance with standard methods of sampling
applicable to the material under test. The
sample shall be selected so as to represent the
material to be tested. and shall be of uniform
thickness. ‘If the material is of nonsymmetri-
cal construction. the two faces shall be desig-
nated by distinguishing marks (for example.
on a one-side-coated sample. “1“ for the
coated side and "11“ for the uncoated side).

8. Test Specimens

8.1 Test specimens shall be representative
of the sample. When the faces of a specimen
are indistinguishable. at least three specimens
shall be tested by the same method. When the
material is designed for use in one position
only. at least three specimens shall be tested
by the same method with the vapor flow in the
designated direction. When the material is in
ready-to-use form and may be used with ei-
ther face toward the vapor source. at least
foar specimens shall be tested by the same
method. two being tested with the vapor flow
in each direction and so reported. Great care
shall be taken not to contaminate the test area
of the specimen.

NOTE S—When testing a low-permeance material
that may be expected to low or gain weight
throughout the test (because of its evaporation or
oxidation). it may be advisable to provide an addi-
tional specimen or “dummy" tested exactly like the
other except that no desiccant or water is put in the
dish.

8.2 The thickness of each specimen shall be
measured at the center of each quadrant to
the nearest 0.025 mm (0.001 in.) and the re-
sults averaged.

9. Procedure A. Desiccant Method at 23C
(73.4 F)

9.1 Fill the dish with desiccant to within 6
mm of the specimen. Leave enough space so
that shaking of the dish. which must be done
at each weighing with the dish in the upright
position. will mix the desiccant. The depth of
the desiccant shall be at least 12 mm. Seal the
specimen to the opening of the dish in such a
manner that leakage of water vapor at and
through the edges is prevented. (See the Ap-
pendix for examples of suitable methods of
sealing.)

9.2 Weigh the assembly on the analytical
balance. Then place the assembly on a rack in

a test chamber. as described in 5.2. and locate
so that the conditioned air circulates over the
exposed surface of the specimen with the
specified velocity. The specimen may be
placed in either the upright position or in-
verted so that the desiccant is in direct contact
with the specimen. This latter position is pre-
ferred for specimens having a high rate of
water vapor transmission. but care must be
taken to ensure that the seal is not broken or
the surface of the specimen damaged.

9.3 Make successive weighings of the as-
sembly at suitable intervals until a constant
rate of gain is attained. Weighing should be
accomplished without removal of the test
dishes from the controlled atmosphere: if
removal is necessary. the specimen must be
tightly covered. and the weighing made imme-
diately after removal with the assembly re-
turned to the test chamber immediately after
each weighing. Plot the results as prescribed
in 15.1. Terminate the test or change the des-
iccant before the water added to the desiCo
cant exceeds the limits specified in 6.2.

10. Procedure 8. Water Method at 23 C

(73.4 I")

10.1 In this method it is intended to pro.
vide 100 percent relative humidity in the dish
but to avoid direct wetting of the surface fac-
ing the water. This requires that the test be
carried out with the dish in the upright posi-
tion and that a certain distance be maintained
between water and specimen. As a result.
there will exist a layer of air between the sur-
face of the water and the under surface of the
specimen through which water vapor will dif-
fuse at a rate that cannot be measured con-
veniently. The influence of this air layer will
become apparent only on specimens of a high
rate of water vapor transmission: in this case
the relative humidity on the under side of the
specimen will be less than 100 percent: and
the measured WVT will be slightly too low.

10.2 Place distilled water in the dish'to
bring the level of the water to such a height
that the distance between the surface of the
water and the under surface of the specimen is
20 :1; 5 mm. A depth of 5 mm of water will
adequately take care of losses by evaporation
through the most permeable specimens. The
depth shall be not less than 3 mm to ensure

742

 

(Elli

coverage of the dish bottom throughout the
test. The possibility of splashing may be re-
duced by the use of a noncorroding grid in the
dish to break the water surface. This grid
shall be at least 6 mm below the specimen and
it shall not reduce the water surface by more
than 10 percent.

NOTE b—For any of the water methods. baking
the empty dish and promptly coating its mouth with
sealant before assembly is recommended. The water
may be added most conveniently after the specimen
is attached. through a small scalable hole in the dish
above the water line.

10.2 Attach the Specimen to the dish. and
proceed with the weighing as prescribed for
Procedure A in 9.2 and 9.3. except place the
assembly on the test racks in the upright posi-
tion. Small variations in the temperature to
which the dish and specimen might be ex-
posed can result in condensation of water va-
por on the underside of the specimen. Such
condensation can significantly affect the meas
ured water vapor transfer rate of some mate-
rials and is to be avoided. For this reason
weighing should be carried out within the test
chamber. 1f removal is necessary. the speci-
men must be tightly covered. The dishes shall
not be exposed to a temperature that differs
more than 3C from the cabinet temperature
to minimize condensation on the specimen.
Plot results as prescribed in 15.1.

11. Procedure BW. Inverted Water Method
at23C (73.4 F)

11.1 This is similar to Procedure 8 except
that the dish is inverted. so that the water is
in contact with the specimen surface at all
times during the test.

11.2 Place distilled water in the dish to a
depth of not less than 5 mm and not greater
than 10 mm.

11.3 Attach the specimen to the dish and
proceed with the weighing as prescribed for
Procedure A in 9.2 and 9.3. placing the dish
on the test racks in an inverted position. The
dish must be sufficiently level so that water
covers the inner surface of the specimen de-
spite any distortion of the specimen due to the
“eight of the water. With highly permeant
\pecimens it is especially important to locate
the test dish so that air circulates over the
exposed surface at the specified velocity. The

743

test dishes may be placed on the balance in
the upright position for weighing. but the pe-
riod during which the wetted surface of the
specimen is not covered with water must be
kept to a minimum.

12. Procedure C. Desiccant Method at 32.2 C
(90 F)

12.1 Follow Procedure A (Section 9). ex-
cept place the assembly in a chamber main-
tained at 32.2 C (90 F) as prescribed in 5.2.2.

13. Procedure D. Water Method at 32.2 C
(90 f 1
13.1 Follow Procedure B (Section 10). ex-

cept place the assembly in a chamber main-
tained at 32.2 C (90 F) as prescribed in 5.2.2.

14. Procedure E. Desiccant Method at 37.8C
(100 F)

14.1 Follow Procedure A except place the
assembly in a chamber maintained at 37.8C
(100 F) and 90 percent relative humidity. as
prescribed in 5.2.2.

Nort-z 7— The use of weighing covers is recom-
mended. particularly with this procedure. since
weighings will have to be made in a room whose

humidity'and temperature will differ from the con-
ditions i1 the cabinet.

15. Plot'ting and Calculation

15.1 Plot the results of successive weighing
against elapsed' time and draw a smooth curve
through the plotted points. Judgment is re
quired here and therefore numerous points are
helpful. When a straight line adequately fits
the plot of four properly-spaced. successive
points (with due allowance for weighing er-
rors). a nominally steady state exists. and the
slope of the straight line is the rate of vapor
transmission for the test area.

NOTES The time required to reach the steady
state will depend upon the permeance. thickness.
hygrosc0pieity. and initial moisture content of the
specimen. for specimens of very low permeance.
buoyancy effects due to changes in barometric pres-
sure may C‘JLhC 'J gl’L'JlL'I’ change ”'1 measured
weight between weighings than that due to meisture
transfer. Long periods of test are required to obtain
significant results in these instances.

15.2 Calculate the water vapor transmis-
sion (WVT) and permeance as follows:

Water vapor transmission (WVT)
= (e x 24i/ti x a)

45117

where:

g = weight gain or loss. g.

r = time. during which gain or loss. 2.
was observed. h.

a = exposed area of specimen. in". and

WVT = rate of water vapor transmission.

g/mJ -24 h.
Permeance = WVT/AP = WVT/[$(R. -— Ryyl

where:

AP = vapor pressure difference. mm
Hg.

5 -- saturation vapor pressure. at
test temperature. mm Hg.

R1 = relative humidity at the source.

R: = relative humidity at the sink. and

Permeance = WVT/mm Hg expressed in

metric perms.

1n the controlled chamber the relative humid-
ity and temperature are the average values
actually measured during the test and (unless
continuously recorded) these measurements
shall be made as frequently as the weight
measurements. In the dish the relative humid-
ity is nominally 0 percent for the desiccant
and 100 percent for the water. These values
are usually within 3 percent relative humidity
of the actual relative humidity for specimens
below 2.5 metric perms when the required
conditions are maintained (no more than 10
percent moisture in CaCl: and no more than
2.5-mm air space above water).

15.3 When the test specimen is homoge-
neous (not laminated) and thick relative to
pore dimensions. the average water vapor
permeability. in metric perm-centimeters.
may be calculated as follows:

Average permeability = permeance
(metric perms) X thickness (centimeters)

For thin specimens the accuracy of the re-

E96

sults will depend greatly on the precision with
which the average thickness is determined.

15.4 Other units and conversion factors are
given in Table 2.

16. Report

16.1 The report shall include the following:

16.1.1 A description of the sheet tested. in-
cluding identification of side 1 and side 11. If
indistinguishable. so state.

16.1.2 Thickness of the specimen tested and
exposed area.

16.1.3 The following actual conditions of the
test:

16.1.3.1 Procedure. whether A. 8. BW. C.
D. or E.

16.1.3.2 Measured temperature.

16.1.3.3 Measured relative humidity in the
test spaces. and

16.1.3.4 State side of sheet. 1 or 11. exposed
to the test space.

16.1.4 WVT of each specimen in stated units
as tested.

16.1.5 Average WVT of all specimens tested
in each direction of vapor flow.

16.1.6 Permeance of each specimen in
stated units under the test conditions.

16.1.7 The average permeance of all speci-
mens tested in each direction of vapor flow.

17. Precision

17.1 Results obtained by any one pro-
cedure on several specimens from the same
sample may differ as much as 10 percent
from their average. Two significant figures in
water vapor transmission or permeance will.
therefore. Usually sullice to characterize the
sample. Nevertheless. careful attention to all
aspects of the procedure is required in order
to obtain test results of acceptable precision.

744

REFERENCES

10.

11.

REFERENCES

Amini, M.A. Lecture: "Testing Permeation and Leakage
Rates of Containers for Pharmaceuticals." Pharm Tech
Conference, Sheraton Centre Hotel, New York City,
Sept. 22-24, 1981.

Amini, Mary A. and Darrell R. Morrow. "Leakage and
Permeation: Theory and Practical Applications,"
Package Development and Systems, May/June 1979, pp.
20-27.

Auslander, D.E. "Hermetic Packaging of Drugs: Opti-
mized Sealing of Foil Pouches," Package Development
and Systems, September/October 1978, pp. 20-22.

Barcan, Donald S. "Answering the Question: 'Is
F/F/S the Best Solution For My Products?'," Packaging
Technology, Dec. 1981, p. 10.

Das Gupta, V. and others. "Stability of Oral Solid
Drugs after Repackaging in Single-unit Containers,"
American Journal 2; Hospital Pharmacy, 37: 165-169.
Feb. 1980.

Giacin, J.R., H.E. Lockhart, Robert Adams and Michael
Kentala. "Storage Stability of Moisture Sensitive Oral
Solid Drugs After Unit Dose Repackaging." Lecture:
Pharm Tech Conference, Sheraton Centre Hotel, New York
City, Sept. 22-24, 1981.

"Guidelines for Repackaging Oral Solids and Liquids
in Single Unit and Unit Dose Packages," American

Journal 2: Hospital Pharmacy, 34:1355. Dec. 1977.

Miltz, Joseph. "Effect of Structure on Heat Sealing
Properties, Seal Strength of LDPE'S," Package Develop-
ment and Systems, March/April 1980, pp. 21-25.

"Minimum Standards for-Pharmacies in Institutions,"
Amerigan Journal 9: Hospital Pharmacy. 34: 1356-1358.
Dec. 1977.

Reamer, Jeanne T. and Lee T. Grady. "Moisture Perme-
ation of Newer Unit Dose Repackaging Materials,"
American Journal g: Hospital Pharmagy, 35: 787-793.

IJuly 1978.

Reamer, Jeanne T. and others. "Moisture Permeation of
Typical Unit Dose Repackaging Materials," American

Journal 2: Hospital Pharmacy, 34: 35-42. Jan. 1977.

12.

13.

14.

Reamer, Jeanne T. and Lee T. Grady. "Testing of Heat
Sealing by Thermal Analysis," Journal of Pharmaceu-
tical Sciences, 65: 628-630. April 1973.

 

"Unit Dose Repackaging," American Journal 2§_Hospital
Pharmacy, 34: 27. January 1977.

United States Pharmacopeia, The. Twentieth Revision,
Washington D.C.: The United States Pharmacopeial
Convention, 1980. P. 955.

nICHIan STQTE UNIV. LIBRARIES
1|I‘I‘lllllm‘IIHIIHWINNHIHWW"WWW“
31293010937005