 

 

 

I ‘ LIBRARY
2 002* Michigan State
University

 

This is to certify that the
thesis entitled

EXPERIMENTS FOR CLIMATE ZONE IV HYPOTHESIS

presented by
XIN Ll

has been accepted towards fulﬁllment
of the requirements for the

MS. degree in PACKAGING

MM 9. Jam/1%

Mﬁbr Prpfessor’s Signature

Sag] 4153 Q 1; £007

Date

 

MSU is an affinnative-action, equal-opportunity employer

 

- -.-.-.-.-.-.-.-.- -n-a-s-IQr-Iq-IIQ-J'Q

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DAIEDUE

DATEDUE

DAIEDUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:lClRC/DaleDue.mdd-p.1

EXPERIMENTS FOR CLIMATE ZONE 1v HYPOTHESIS
By

Xin Li

ATHESIS

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

MASTER OF SCIENCE

Department of Packaging

2007

ABSTRACT

Experiments for Climate Zone IV Hypothesis
By
Xin Li
Stability tests are required for all drug products. The purpose of the stability test
is to find the shelf life of the products; the test should be performed under the
climatic conditions of the destination marketplace. In order to make the test
widely applicable, the world is divided into four climate zones with different test
conditions specific to each zone. Climate Zone IV represents the hot and humid
zone. The conditions 30°C/65°/o RH were speciﬁed by WHO (World Health
Organization) as the test condition for this area. However, the Association of
Southeast Asian Nations (ASEAN) proposed to WHO in 2004 to change this
condition to 30°C/7 5% because of the severe environmental condition of ASEAN
countries (27°C/79%). Based on a theoretical calculation, 30°C/65°/o RH could be
used as the stability test condition for ASEAN’s countries.'The purpose of this
study was to support the result of theoretical calculation using experimental data.
HDPE bottle, one of the most popular packages for pharmaceutical products,
was selected as the test sample. The WVTR for the bottles was determined
under 5 different conditions, 30°C/59.6°/o RH, 30°C/60.9% RH, 30°C/64.5% RH,
30°C/74.5% RH and 27°C/79°/o RH, using a gravimetric method. More water
vapor permeated into the bottle at 30°C/65°/o RH than it did at 27°C/79% RH.

This result supports the theoretical calculation.

ACKNOWLEDGEMENTS

There are many people to thank for their contributions toward the completion of
my master degree during my time at MSU. First, I want to thank my advisor, Dr.
Hugh E. Lockhart, for his support and guidance. I also want to express my thanks

to Dr Dennis Gilliland and Dr Maria Rubino who serve on my committee.

Thanks to my friends for their support and friendship. To my parents, parents-in-
law, and brother, thanks for the encouragements and the sacrifices they have

made.
Most importantly, I would like to express my deepest gratitude to my husband,

Hui, for all of the love and support she has given me over the years. I would not

have been able to accomplish any of these without you.

iii

TABLE OF CONTENTS

LIST OF TABLES ................................................... _ .............................. v

LIST OF FIGURES ............................................................................... vi

Chapter 1. Introduction ............................................... ' ......................... 1
Chapter 2. Literature Review ................................................................ 4
Chapter 3. Materials and Methods ........................................................ 27
Chapter 4. Results and Discussion ....................................................... 33
Chapter 5. Conclusion ........................................................................ 54
References ......................................................................................... 55

iv

Table 1.1.
Table 2.1.
Table 3.1.
Table 3.2.
Table 3.3.
Table 3.4.
Table 4.1.

Table 4.2.

LIST OF TABLES

Definitions and stability test conditions of climatic zones .............. 2

Definitions, criteria and stability test conditions of climatic zones...8

Thickness of the bottle walls ................................................. 28
Crystallinity of the bottles ....................... » .............................. 28
Salt solutions used to provide speciﬁed RH for bottle test ........... 29
The balances used in the test for the bottles ............................ 31
WVTR value for each bottle .......................... i ........................ 4 1

t-test for WVTR values gotten from 30°C/64.5%RH and
27°C/79%RH ..................................................................... 45

Figure 2.1.

Figure 2.2.

Figure 3.1.

Figure 4.1.

Figure 4.2.

Figure 4.3.

Figure 4.4.

Figure 4.5.

Figure 4.6.

Figure 4.7.

Figure 4.8.

Figure 4.9.

Figure 4.10.

LIST OF FIGURES

Schematic drawing of ASTM E 96 test method: Desiccant Method
and Water Method .............................................................. 21

Schematic drawing of ASTM 1249 test method ........................ 22

Schematic drawing of exposure bucket for WVTR test of

bottles .............................................................................. 31
The temperatures recorded during the test of the
bottles .............................................................................. 33
The RH recorded during the test of the bottles .......................... 34

The weight gains of the bottles stored at condition 30°C/59.6%RH
from day 0 to day 63 ........................................................... 36

The weight gains of the bottles stored at condition 30°C/59. 6%RH
from day 64 to day 119 ....................................................... 36

The weight gains of the bottles stored at condition 30°C/60.9%RH
from day 0 to day 63 ........................................................... 37

The weight gains of the bottles stored at condition 30°C/60.9%RH
from day 64 to day 119 ........................................................ 37

The weight gains of the bottles stored at condition 30°C/64.5%RH
from day 0 to day 63 ........................................................... 38

The weight gains of the bottles stored at condition 30°C/64.5%RH
from day 64 to day 119 ........................................................ 38

The weight gains of the bottles stored at condition 30°C/74.5%RH
from day 0 to day 63 ........................................................... 39

The weight gains of the bottles stored at condition 30°C/74.5%RH
from day 64 to day 119 ........................................................ 39

vi

Figure 4.11.

Figure 4.12.

Figure 4.13.

Figure 4.14.

Figure 4.15.

Figure 4.16.

Figure 4.17.

Figure 4.18.

Figure 4.19.

Figure 4.20.

Figure 4.21.

Figure 4.22

Figure 4.23

Figure 4.24

Figure 4.25

The weight gains of the bottles stored at condition 27°C/79%RH
from day 0 to day 63 ........................................................... 40

The weight gains of the bottles stored at condition 27°C/79%RH
from day 64 to day 119 ........................................................ 4O

Scatter plot of WVTR from two different periods ........................ 43

Comparison of all the WVTR values of the bottles (day 0 ~ day
63) .................................................................................. 44

Comparison of all the WVTR values of thebottles (day 64 ~ day '
119) ................................................................................. 44

The weight gains of the control bottles stored in condition
30°C/59.6%RH from day 0 to day 63 ...................................... 47

The weight gains of the control bottles stored in condition
30°C/59.6%RH from day 64 to day 119 ................................... 47

The weight gains of the control bottles stored in condition
30°C/60.9%RH from day 0 to day 63 ...................................... 48

The weight gains of the control bottles stored in condition
30°C/60.9%RH from day 64 to day 119 ................................... 48

The weight gains of the control bottles stored in condition
30°C/64.5%RH from day 0 to day 63 ...................................... 49

The weight gains of the control bottles stored in condition
30°C/64.5%RH from day 64 to day 119.... .............................. 49

The weight gains of the control bottles stored in condition
30°C/74.5%RH from day 0 to day 63 ...................................... 50

The weight gains of the control bottles stored in condition
30°C/74.5%RH from day 64 to day119...........' ........................ 50

The weight gains of the control bottles stored in condition
27°C/79°/oRH from day 0 to day 63 ......................................... 51

The weight gains of the control bottles stored in condition
27°C/79%RH from day 64 to day 119 ..................................... 51

vii

Chapter 1. Introduction

In United States Pharmacopeia, “stability is deﬁned as the extent to which a
product retains, within speciﬁed limits, and throughout its period of storage and
use (i.e., its shelf life), the same properties and characteristics that it possessed
at the time of its manufacture.” There are generally ﬁve types of stability:
chemical, physical, microbiological, therapeutic and toxicological. These ﬁve
factors are evaluated through stability testing (USP 1 2004). The purpose of
stability testing is to determine proper storage conditions and expiration dates of
an active substance or pharmaceutical product. It is required by United States
food and drug regulations for all drug products. In the regulation for Current Good
Manufacturing Practice of Finished Pharmaceuticals, at 21CFR 211.166, the
Food and Drug Administration (FDA) says: "There shall be a written testing

program designed to assess the stability characteristics of drug products. “

The stability of ﬁnished pharmaceutical products depends on several factors.
Basically, those factors can be divided into two parts: one is environmental
factors, such as ambient temperature, humidity and light; the other is product
related factors, such as chemical and physical properties of the active substance,
excipients, dosage form, manufacturing process, container closure system and
packaging material (WHO, 1). The shelf life should be established based on the

climatic conditions of the destination marketplace.

Since the result of stability testing is affected by the environmental conditions,
different testing conditions are adopted for different areas. In order to be able to
reduce the amount of stability testing, the number of different long term testing
conditions must be reduced to a sufﬁcient extent. Based on the proposals to
World Health Organization (Schumacher, 1972; Grimm, 1986 and Grimm, 1998),
the world is divided into four different climatic zones (CZ), and different testing

conditions are used for each of them.

Table 1.1 Deﬁnitions and stability test conditions of climatic zones (WHO 1)

 

 

 

 

 

CZ Deﬁnition Long term testing conditions
I Temperate climate 21°C/45% RH
II Subtropical and Mediterranean 25°C/60% RH
climate
III Hot and dry climate 30°C/35% RH
IV Hot an humid climate 30°C/65% RH

 

 

 

 

 

At the “Consultation to Discuss Stability Studies in a Global Environment” on 13-
14 December 2004 at WHO Headquarters in Geneva, the Association of South
East Asian Nations (ASEAN) proposed to WHO to change the stability testing
condition for Climate Zone IV from 30°C/65% RH to 30°C/75% RH. Three
recommendation were made after the meeting: A) revert to 30°C/70% RH as the
long-term stability testing condition for Zone IV; 8) change to 30°C/75% RH as

the long-temI stability testing condition for Zone IV; C) add a new Climate Zone

IVb to accommodate hot and very humid areas (30°C/7 5%), and keep the current
Zone IV as Climate Zone IVa (WHO, 2). This proposal was made just based on
the ASEAN climate zone average condition which is 27°C, 79% RH. The effect of
barrier packaging on control of the humidity inside the package was not
considered. Actually, it is the storage temperature, partial pressure for water
vapor and moisture permeability of the package that affect the result of stability
testing. Based on the theoretical calculation (Lockhart, 2006), the current Climate
Zone IV stability condition 30°C/65°/o RH should be adequate to represent the
ASEAN average condition 27°C/79°/o RH. The hypothesis was made: the stability
test condition 30°C/65% RH should be severe enough to represent the ASEAN
average condition 27°C/79%. This theoretical conclusion needs to be tested by

direct experimentation.

The objective of this research was:
Use HDPE bottles as test samples to demonstrate that the stability test condition

30°C/65% RH should be severe enough to represent the ASEAN average

condition 27°C/79%.

Chapter 2. Literature Review

Association of South East Asian Nations

ASEAN was founded on 8 August 1967 in Bangkok. The original member
countries included: Indonesia, Malaysia, Philippines, Singapore, and Thailand.
The ASEAN Declaration was made on the same day by all the members. It
states that the aims and purposes of the Association are: “(1) to accelerate
economic growth, social progress and cultural development in the region, (2)'to
promote regional peace and stability through abiding respect for justice and the
rule of law in the relationship among countries in the region and adherence to the
principles of the United Nations Charter" (ASEAN, 1). As the development of the
ASEAN continued and the collaboration increased, Brunei Darussalam, Vietnam,
Lao People’s Democratic Republic, Myanmar and Cambodia joined the ASEAN
later. Now, the ASEAN region has a population of about 500 million, a total area
of 4.5 million square kilometers, a combined gross domestic product of almost

US$ 700 billion, and a total trade of about US$ 850 billion.

In order to facilitate and liberalize trade and investment in the region, ASEAN
Consultative Committee for Standards and Quality (ACCSQ) was set up in 1992.
The purpose of the organization is to facilitate and liberalize trade and investment
in the region. It has endeavored to harmonize national standards with
international standards and implement mutual recognition arrangements on
conformity assessment to achieve its end—goal of “One Standard, One Test,

Accepted Everywhere”. All Member Countries have accomplished the

harmonization of standards for 20 priority products and 81 standards for Safety
and Electromagnetic Compatibility (EMC). New areas for harmonization are
currently being identiﬁed. Priority for harmonization will be given to those

standards used in technical regulations in Member Countries (ASEAN 2).

At the 13th Meeting of the ACCSQ held in March 1999 in Manila, the members
agreed to set up a Pharmaceutical Product Working Group (PPWG), with
Malaysia as the lead country. Accordingly, ACCSQ-PPWG was launched in
September 1999 in Kuala Lumpur, Malaysia. The objective of the ACCSQ-PPWG
is to develop harmonization schemes of pharmaceuticals' regulations of the
ASEAN member countries to complement and facilitate the objective of ASEAN
Free Trade Area (AFTA), particularly the elimination of technical barriers to trade
posed by these regulations, without compromising on drug quality, safety and

efﬁcacy (ASEAN 3).

Stability Testing

“The purpose of stability testing is to provide evidence on how the quality of a
drug substance or drug product varies with time under the inﬂuence of a variety
of environmental factors, such as temperature, humidity, and light, and to
establish a retest period for the drug substance or a shelf life for the drug product
and recommended storage conditions” (WHO, 1). In the stability testing, the
marketed container-closure systems for the products must be involved. Therefore,

the environment that the products are exposed to may not be as severe as the

storage environment because of the protection of the package. This is especially

true when a barrier package limits the ingress of water in vapor form.

Stability testing should be performed for both drug substances and drug products.
In manufacturing, the stability of the drug substance should be decided at ﬁrst.
And based on the information of the behavior and properties of the active
substance, stability test for drug product is performed. Generally, at least three
primary batches of product should be used in the stability study, and the product
should be packaged in the container closure system that is the same as the
packaging proposed for storage and distribution. The testing should cover the
physical, chemical, biological, and microbiological attributes, preservative content
and functionality test. About the storage condition used in the testing, generally, a
drug product should be evaluated under storage condition that test its thermal
stability and, if applicable, its sensitivity to moisture or potential for solvent loss.
The storage conditions and the lengths of studies chosen should be sufﬁcient to

cover storage, shipment, and subsequent use (WHO 1).

Based on the results from the stability testing, a storage statement should be
established for the labeling. Speciﬁc instructions should be provided if applicable.

An expiration date should be displayed on the container label.

Climate Zone (CZ)

The shelf life should be established with regard to the environmental condition in
which the product is to be marketed. For certain preparations, the shelf life can
be guaranteed only if speciﬁc storage instructions are complied with. So, different
testing conditions are supposed to be used for different climate zones. In order to
reduce the amount of stability testing required, the number of different long term
testing conditions must be reduced to a sufﬁcient extent. Therefore, the concept
of climate zone was proposed by Schumacher (Schumacher, 1972). In the
proposal, the world was divided into four zones based on the temperature and
humidity, and the storage conditions for stability testing were recommended for
each zone. The mean annual temperature and the mean annual partial pressure
of water vapor are the main criteria used for the Inclusion of a region in a
particular set of climatic conditions (Bott, 2007). The mean annual partial

pressure of water vapor can be calculated using equation (1)

P = P" x RH (1)
Where P is the partial water vapor pressure in the environment, P' is the
saturated water vapor pressure at the same temperature and RH is relative

humidity of the environment.

Table 2.1 Deﬁnitions, criteria and stability test conditions of climatic zones

 

CZ Deﬁnition Criteria Long term
Mean annual testing

temperature in open air/ conditions

mean annual partial

water vapor pressure

 

 

 

 

I Temperate climate s 15°C I s11hPa' 21°C/45% RH

II Subtropical and >15 to 22°C / > 11 to 18 25°C/60% RH
Mediterranean climate hPa

I” Hot and dry climate > 22°C / s15hPa 30°C/35% RH

IV Hot an humid climate > 22°C / >15hPa 30°C/70% RH

 

 

 

 

 

 

* 1 hPa = 0.75 mmHg

This concept is described in WHO’s regulatory standards (WHO, 1996) and The
United States Pharmacopeia and became an established standard in developing

pharmaceutical products.

In fact, the storage conditions for the stability testing in a region are determined
based on the mean kinetic temperature (MKT) and the average of the relative
humidity. The MKT is a single calculated temperature at which the total amount
of degradation over a particular period is equal to the sum of the individual

degradations that would occur at various cycles of higher and lower temperature

(WHO, 2004). It expresses the cumulative thermal stress undergone by a product
at varying temperatures during storage and distribution. It can also be considered
as an isothermal storage temperature that simulates the non-isothermal effects of
storage temperature variations.

The temperature dependence for the degradation of drug products is based on
principles of chemical kinetics, especially the Arrhenius equation. The
relationship between the reaction rate constant and the temperature can be

expressed as (PMA’s, 1991 ):

AE

where K is the degradation rate, A is a constant dependent on the molecule of
interest, AE is the activation energy. R is the universal gas constant and T is the

absolute temperature.

Based on Equation 2, MKT can be calculated (Bailey,1993):

T _ — AE/R
MKT '- _AE/RT, 45/er —AE/RT.. (3)
lnie + e + + e ]

n

 

 

Where TMKT is MKT, n is the number of temperatures collected and T is absolute
temperature. AE is generally considered as 20 kcal/mol, since the activation

energies of most reactions of organic molecules are about 20 kcal/mol.

Using Equation 3, the temperature of stability testing for each climate zone can
be obtained.

Relative humidity can be calculated based on Equation 1. But the relative
humidity speciﬁed for the stability testing for each climate zone was selected
based on the average condition of each zone. And there are not generally
accepted methods to build a relationship between RH and time or temperature.
That’s also why the argument about stability testing condition always focuses on

the relative humidity.

Argument about Stability Testing Condition for Zone IV

Climate Zone IV represents a hot and humid climate. It corresponds to central
Africa and south Paciﬁc, such as Zaire, Kenya, Indonesia, and Philippines etc.
The stability testing condition for Zone IV has always been in dispute since the
beginning of the concept. Testing at 30°C/70% RH used to be the condition used
in industry, and it was also recommended by WHO In 1996. In 2000, WHO
changed the testing condition from 30°C/70% RH to 30°C/60% RH based on a
suggestion from the International Conference on Harmonization (ICH) Q1 Stability
Expert Working Group. But, African countries objected to it later. They believed
the condition could not represent the high humidity of some coastal African
countries. After a series of meetings, in 2001, WHO made the ﬁnal decision: they
changed the condition from 30°C/70% RH to 30°C/65%. Consequently, ICH
Stability Guideline 01F recommended the same condition (Singh, 2006). In Dec.

2004, ASEAN proposed to WHO to change the stability testing condition for

10

Climate Zone N from 30°C/65% RH to 30°C/75% RH because the average
environmental condition for ASEAN countries is 27°C/79% RH. After gathering
the responses from manufacturers and regulators, in Oct. 2005, the 40th WHO
Expert Committee on Speciﬁcations for Pharmaceutical Preparations decided to
split the current Climatic Zone IV (hot and humid countries) into CZ IVA, for
which 30°C/65% RH will remain the standard long-term testing condition, and CZ
IVB, for which 30°C/75% RH will become the testing condition (WHO 2). The
Expert Committee also agreed that each individual Member State within the
former Zone IV would need to indicate which of these conditions (Zone lVa or lVb)

would be applicable in its territory.

Moisture Permeation Mechanism of Container-Closure System

A container-closure system can be classiﬁed as either permeable or
impermeable system. Generally, most of those packages used in pharmaceutical
industry are permeable. Only ﬂame-sealed glass ampuls and aluminum blisters
can be considered impermeable systems. Commonly used moisture permeable
container-closure systems in pharmaceutical industry are: high—density
polyethylene (HDPE) containers with any metal or plastic screw closures, low-
density polyethylene (LDPE) containers with any metal or plastic screw closures,
glass containers with any metal or plastic closure, polyvinyl chloride (PVC)
blisters with lidding, PVC/polyvinylidene chloride (PVdC) blisters with lidding, and

PVC/Aclar (polychIorotriﬂuoroethylene) blister with lidding. Aluminum blisters are

11

a special case. The permeation rate is so low that it is difﬁcult to measure with

current methods, but there is permeation.

For the glass container-closure system, moisture permeates into the container
mainly through the cap liner which is part of the closure and is used as a
moisture seal. The permeability of glass itself is considered zero. For plastic
containers, most of them are semi-crystallized materials. So, permeation is the
result of ingress of water vapor through the wall of the container and the ingress
of water vapor through the liner in the cap. If a hermetic wax seal is made for the
closure, the permeation of the bottle alone can be measured. For aluminum
blisters, a very low rate of ingress can occur through the seal between the

backing and the blister.

Theory of Permeation
This discussion of theory is taken from course pack of School of Packaging,

Michigan State University, for class PKG 815. (Rubino, 2005).

Permeability is a measure of how easily a permeant compound transports
through a solid medium. The permeation process takes place when a permeable
wall separates two ﬂuid phases (air or liquid) having different concentration
values in each phase. For water vapor, there are two mechanisms of permeation:
1) convective mass transfer through channels, 2) diffusion-controlled molecular

exchange through semi-permeable materials such as plastic. The ﬁrst

12

mechanism is generally not taken into account for the permeation of the moisture
into a package used for pharmaceutical product. This is because it only happens
when the packages have signiﬁcant defects like pinholes or cracks. Diffusion-
controlled permeation can be described as a three-step event: 1) the permeant
molecule from the contacting ﬂuid phase permeates the polymer surface, 2) the
permeant diffuses within the polymer material, 3) the diffused permeant desorbs
from the polymer interface into the adjacent continuous phase which has low
concentration. The fundamental driving force that prompts a molecular specie to
transfer within the polymer, or between a polymer and a surrounding phase is,
according to the solution theory, the tendency to equilibrate the species’ chemical
potential across the thickness of the polymer.

Fick’s laws of diffusion can be used to describe the permeation process in
packaging systems. It includes two parts. The ﬁrst Fick’s law is used for steady
state diffusion. It describes the rate of transfer of diffusing substance through unit
area of a section as being propotional to the concentration gradient aclax. the
equilibrium unidirectional diffusion along the x-axis can be expressed as,

F = -—D x a (4)

where F is the rate of transfer of diffusing substance per unit of area, c is the
permeant concentration, x is the spatial coordinate in the direction of transfer,
and D is the molecular diffusion coefﬁcient which is assumed to be independent

of penetrant concentration.

13

Fick’s second law is used to describe non-steady state, one-dimensional

diffusion in an isotropic phase. It can be described as,

2
6_C=Dx6c
at 6x2

 

(5)

where t is time.

In order to use the previous two equations efﬁciently, the concentration of the
penetrant both in the packaging material phase and in the contacting ﬂuid phase

must be known. Henry’s law can be used to calculate the concentration:

C = Sp (6)

Where 0 is the concentration of the diffusing substance in the solid or liquid
phase, S is the Henry’s law constant, which is the solubility coefﬁcient, and p is
partial pressure of permeant at the contacting phase interface. Henry’s law only

applies for the non-interactive permeant/polymer system.

Based on equations (4), (5), (6), and assuming that the D and S are independent
of concentration, which is true for water vapor permeating through polymers that

do not strongly interact with water, the following equation can be obtained:

P = DS = ql (7)
AtAp

 

Where P is the permeability coefﬁcient, D is the diffusion coefﬁcient which

describes how fast the permeant molecules move through the material, S is the

14

solubility coefﬁcient which describes how many permeant molecules are
absorbed by the material, q is the amount of permeant transferred through the
material during time t, A is the area available for permeation, I is the thickness of
the material being permeated, and Ap is the difference in partial pressure of the
permeating substance in the gas phase in contact with the two surfaces of the

material.

Equation (7) is valid for both steady-state transfer and nonsteady-state transfer. If
the mass transfer is not at steady state, the permeability coefﬁcient can be

calculated for different time t and integrated over time.

When water vapor is permeating into a package that is maintained at a constant
temperature and relative humidity and the package contains a desiccant, steady
state permeation can be assumed because the RH inside the package is
maintained at a low level (around zero) by the desiccant. The time to reach
steady state depends on the thickness of the material, its initial condition, and its
barrier characteristics but generally is short in relation to desired shelf life. If one
of the three conditions is not met (constant temperature, RH and desiccant inside
package), steady state cannot be assumed. When water vapor is permeating into
a package that is maintained at a constant temperature and relative humidity but
contains a product other than desiccant, the water vapor partial pressure inside
the package changes with time. In such cases, the moisture sorption isotherm for

the product can be used to relate the moisture content of the product to the

15

partial pressure inside the package. It can also tell a great deal about the
physical nature of the association between water and product. This assumes that
moisture in the product and the package headspace equilibrate relatively rapidly
in comparison to the rate of transfer of moisture through the package and that

chemical consumption of water is negligible.

When the permeability coefﬁcient of a package is calculated, generally equation

(8) is used:

 

q
p =
tAp (8)

In the equation, P represents the permeability coefﬁcient of the whole package
for the speciﬁc package because the thickness of the package wall varies for
different areas. Equation (6) still can be used for differential areas of the material
and can be integrated over the entire area. As a practical matter this is a tedious
process. For this reason, permeation for a package is usually measured on a per

package basis.

Water Vapor Transmission Rate (WVTR) is another parameter to evaluate the
moisture barrier of a speciﬁc material. It is the amount of water vapor that
permeates per unit time through a ﬁlm or a package at speciﬁed conditions of
temperature and relative humidity. Values are generally expressed in g/100
in2/day for ﬁlm, or g/package/day for package. For permeation of barrier

packages for drugs, it is accepted practice to calculate and report WVTR in

16

mg/day/package. Sometimes it is also useful to report the permeation of water
vapor into a package on the basis of unit partial pressure differential. This is
called permeance. The relationship between WVTR and permeance for a speciﬁc

ﬁlm or package can be expressed using equation (9).

WVTR
P = (9)
A p

 

Where P is the permeance of the package or material, Ap is the difference in
water vapor partial pressure on the two sides of the material, and the WVTR is

the water vapor transmission rate for a whole package.

Factors Affecting the Permeability

This discussion is taken from course pack of School of Packaging, Michigan
State University, for class PKG 815 (Rubino, 2005).

Chemical structure

The chemical structure of material of package and permeant is the primary factor
in determining the permeability. The basic rule for each pair of material/permeant
is “like dissolves like”. For example, a polar material can be more or less
hydrophilic. And the permeability can be changed because of the presence of
water. Similarly, non-polar materials are not affected by the presence of water.
HDPE and LDPE are used as good water vapor barrier because there is no polar

group in the structures, so they are non-polar.

17

Crystallinity and Orientation

Crytallinity of the material can affect the permeability. Generally, permeation
occurs in the amorphous regions. Therefore, increasing the crystallinity will
decrease permeability. Orientation itself does not increase crystallinity. But after
orientation, molecules are packed tightly; free volumes inside the structure are

decreased. Therefore, permeability is decreased.

Temperature

Based on equation (7), the permeability coefﬁcient is the product of diffusion and
sorption coefﬁcients. Both of the two coefﬁcients are affected by temperature.
The diffusion coefﬁcient follows the Arrhenius relationship:

-Ed

D=D0xeRT (1o)

 

Where Do is a constant, Ed is the activation energy of diffusion which is usually a
linear function of temperature. R and T stand for the universal gas constant and
absolute temperature, respectively (Van Krevelen, 1997). Generally, the diffusion

coefﬁcient increases with temperature for both gases and vapors.

For sorption coefﬁcient, the temperature dependence of the solubility coefﬁcient
over relatively small ranges of temperature can also be expressed by an
Arrhenius-type relation (Van Krevelen, 1997),

—AHS
S = SO xe RT

 

(11)

18

Where So is a constant, AHs is the heat consumed by dissolving a mole of
permeant in the polymer. For volatile substances, it is common for solubility to
decreases as temperature increases, but for solids and liquids, solubility

generally increases as temperature increases.

Generally, over reasonably small temperature ranges, the change in permeability
coefﬁcient is given by the following equation (Van Krevelen, 1997),

—Ea
P = P0 X 6 RT (12)

 

Where E, is the activation energy. R is the gas constant, P0 is a pre-exponential
term and T is temperature in degree Kelvin.
Since P0 is a constant, we can rewrite the equation as:

aaa,
R T T
Bzﬁxe '2 am

And this equation can be used to calculate the permeability P2 at temperature at

T2 when the permeability P1 at temperature T1 is known.

Measuring Permeability

Many experimental techniques have been developed for measuring the
permeability of gases and vapors through polymeric materials. Some techniques
have been adapted as standard methods but many of them have remained non-

standardized procedures and are used for development and research purposes.

19

Most techniques for measuring permeability share a number of basic elements:
permeant source, permeation unit (cell or device to support the test sample),
permeant detector and temperature controlled environment. These techniques
can be grouped in one of the three basic methods based on the way the
permeated compound is collected: pressure-variable, volume-variable, and
isostatic or isobatic.

When a pressure-variable method is used, a two-chamber cell is separated by
the test ﬁlm. The high-pressure chamber is ﬁlled with the permeant at relatively
high pressure and the permeated gas diffuses to the other side of the ﬁlm. The
permeant is collected in a constant volume chamber, and the pressure increase
in the low-pressure chamber is measured by a pressure gauge as a function of
time. This method was used to measure the water vapor permeability through

packaging polymer (Doty, 1946).

The volume-variable method is also uses a two-chamber cell separated by the
test ﬁlm. But the permeant is allowed to expand against a constant atmospheric
pressure. And the change in volume as a function of time is measured in a
capillary tube by displacement of a liquid. This method offers some advantages
over the pressure-variable by the fact that the test proceeds under constant
pressure differential and during the steady state the gas transmission rate,

measured as change in volume, is also constant with time.

20

In the isostatic method, the partial pressure differential across the test ﬁlm for
permeating gas remains constant during the total permeation process. While the
high pressure side remains constant at a certain value, the low pressure side is
maintained in most cases, at zero partial pressure by either sweeping a gas
stream or using a trapping agent. The isostatic technique itself can be divided
into continuous ﬂow method, cup method, and dish method. It is one of the most
popular techniques used to test the permeability of water vapor. Two test
methods described in ASTM International standards for water vapor transmission
are set up based on the isostatic technique. One is ASTM E 96 — Standard Test
Methods for Water Vapor Transmission of Materials. The other is ASTM F 1249 -
Standard Test Method for Water Vapor Transmission Rate through Plastic Film
and Sheeting Using a Modulated Infrared Sensor, using continuous technique.
ASTM E 96 includes two basic methods, the Desiccant Method and the Water
Method. Testing is performed by sealing a specimen to the open mouth of a test
dish containing either desiccant or water and placing the assembly into a
controlled atmosphere. The test unit is weighed periodically and the weight is
plotted as a function of time. Water vapor transmission is taken as the slope of
the curve (in the linear region) divided by the area of the dish opening (ASTM E
96)

21

Test
specimen

 
 
 
 

  
  

Controlled Test

atmosphere specimen —" Controlled

atmosphere

Desiccant Water

 

Figure 2.1 Schematic drawing of ASTM E 96 test method: Desiccant Method and Water Method

In the method of ASTM F 1249, the test specimen is held such that it separates
two sides of a test chamber. The "wet side” of the specimen is exposed to a high
relative humidity atmosphere, while the "dry side" is subjected to a zero relative
humidity atmosphere. Infrared sensors on the "dry side" detect the amount of
water vapor present. Testing is complete when the concentration of water vapor

in the dry side atmosphere is constant (ASTM F 1249).

 

 

 

 

 

 

 

SQIISOI'

 

 

Figure 2.2 Schematic drawing of ASTM 1249 test method

22

For packages used for drug products, the United States Pharmacopeia (USP)
also provides the guidelines for a water vapor permeation test. They are
contained in USP <671> - Permeation. There are two sections in USP Chapter
<671>. One is Multiple-Unit Containers for Capsules and Tablets, which
describes the test applied to multiple-unit containers. The other is Single-Unit
Containers and Unit-Dose Containers for Capsules and Tablets, which describes
the test applied to single-unit and unit-dose containers. In this chapter, it is
speciﬁed that the term “container’ refers to the entire container-closure system
comprising, usually, the container itself, the cap and the cap liner in the case of
multiple-unit containers (bottles). In the case of single-unit and unit-dose
containers (blisters or pouches), the container is the Iidding sealed to the blisters
or a sealed pouch. The method described in USP is similar to the Desiccant
Method in ASTM E 96. For its application in USP, it provides a means for
classifying containers as tight, or well-closed with respect to water vapor
transmission.

Multiple-Unit Containers for Capsules and Tablets (USP 2, 2004)

12 containers (bottle and cap) are used to perform the test. 10 of the bottles,
designated test containers, are ﬁlled with desiccant. Then, the screw — capped
containers are closed with adequate torque and stored at 75 :I: 3% relative
humidity and a temperature of 23 :I: 2°C for 336 :1 hours (14 days). The other 2
bottles, designated controls, are ﬁlled with glass beads and stored in the same

condition for 14 days. The weight of the glass beads is similar to the weight of the

23

desiccant. The bottles are weighed before and after the storage. The rate of

moisture permeability is calculated using equation 14.

1 OOO

WX[(Tr‘TI)’(Cf—Cl)i (14)

in which V is the volume, in mL, of the container, (Tf — Ti) is the difference, in mg,
between the ﬁnal and initial weights of each test container, and (Cf - Ci) is the
difference, in mg, between the average ﬁnal and average initial weights of the 2
controls. Based on the calculation result, the containers are labeled as either
tight containers or well-closed containers. For the tight containers, no more than
1 of the 10 test containers exceeds 100 mg per day per liter in moisture
permeability, and none exceeds 200 mg per day per litter. For the well-closed
containers, no more than 1 of the 10 test containers exceeds 2000 mg per day

per liter in moisture permeability, and none exceeds 3000 mg per day per litter.

Single-Unit Containers and Unit-Dose Containers for Capsules and Tablets (USP 2 2004)

There are two test methods described in the section.

Method l

10 unit-dose containers with 1 pellet desiccant in each are designated as test
containers. 10 empty unit-dose containers are used to provide the controls. All
the containers are stored at 75 :I: 3% relative humidity and at a temperature of 23
:t 2°C. Weigh all the containers every 24 hours. After weighing, the containers
are returned to the chamber. And the rate of moisture permeation of each

container is calculated using equation 15.

24

TL-XKWf-WI-i-(CI-Ciii (15)

in which N is the number of days expired in the test period; (Wf — W) is the
difference, in mg, between the ﬁnal and initial weights of each test container, and
(Cf - Ci) is the difference, in mg, between the average ﬁnal and average initial
weights of the controls.

Method 2

This method is used for packs that incorporate a number of separately sealed
unit-dose containers or blisters. Four or more packs and a total of not less than
10 unit-dose containers or blisters ﬁlled with 1 desiccant pellet in each unit are
used as test containers. The same amounts of empty packs are used as the
controls. All the containers are stored at 75 :I: 3% relative humidity and at a
temperature of 23 1: 2°C. Weigh all the containers every 24 hours. After weighing,
the containers are returned to the chamber. And the rate of moisture permeation

of each container is calculated using equation 16.

ﬁgXKWI-WII- (Cf-0.11 (16)

in which N is the number of days expired in the test period; X is the number of
separately sealed unit per pack; (WI - W.) is the difference, in mg, between the
ﬁnal and initial weights of each test container, and (Cf — Ci) is the difference, in
mg, between the average ﬁnal and average initial weights of the controls.

For the two methods, the containers are classiﬁed as Class A, Class B, Class C

or Class D according to the calculation result in both of the two methods. In

method I, the individual unit-dose containers are designated Class A if not more

25

than 1 of 10 containers tested exceeds 0.5 mg per day in moisture permeation
rate and none exceeds 1 mg per day; they are designated Class B if not more
than 1 of 10 containers tested exceeds 5 mg per day in moisture permeation rate
and none exceeds 10 mg per day; they are designated Class C if not more than
1 of 10 containers tested exceeds 20 mg per day and none exceeds 40 mg per
day, and they are designated Class D if the containers tested meet none of the
requirements above. In method II, the packs are designed Class A if no pack
tested exceeds 0.5 mg per day in average blister moisture permeation rate; they
are designed Class B if no pack tested exceeds 5 mg per day in average blister
moisture permeation rate; they are designed Class C if no pack tested exceeds
20 mg per day in average blister moisture permeation rate; and they are
designed Class D if the packs tested meet none of the above average blister
moisture permeation rate. For both of the methods, the test intervals are
determined based on the classiﬁcation: 24 hours for Class D; 48 hours for Class

C; 7 days for Class B; and not less than 28 days for Class A.

26

Chapter 3. Materials and Method

3.1 Materials and Test Conditions

3.1.1 Material

Plastic bottle is one the most popular packages used in the pharmaceutical
industry. It can provide enough protection for many drugs although it is
permeable. High-density polyethylene (HDPE) bottle was used as the package
which the WVTR would be tested at ﬁve different conditions. The bottles were
provided by Eli Lilly and Company. For each bottle, the overall capacity is 50 cc,
and the thickness of 5 randomly selected bottles was measured. The results
were shown in table 3.1. The crystalllinity of the bottle was measured using TA

Instruments DSC 0100. The temperature of the test ranged from -10° C to 160° C.

5 bottles with 5 samples from each bottle were tested. For each bottle, 4 of the 5
samples were gotten from the side wall of the bottle. The other one was the
bottom of the bottle. The results of the crystallinity were shown in table 3.2. The
28/400 white child resistant closures came with the bottles. The closures
included two parts: Polypropylene (PP) caps and induction heat seal liners which

were made of aluminum foil and paper board.

27

Table 3.1 Thickness of the bottle walls

 

 

 

 

 

 

 

 

 

 

Bottle Thickness (in)
Sidewall Bottom
1 0.0402 0.0385 0.0421 0.0401 0.0611
2 0.0427 0.0399 0.0431 0.0406 0.0634
3 0.0426 0.041 0.044 0.0396 0.0601
4 0.0419 0.039 0.0434 0.0399 0.0567
5 0.0439 0.0396 0.0429 0.0409 0.0547
Table 3.2 Crystallinity of the bottles
Bottle Crystallinity %
Sidewall Bottom
1 42.85 44.41 44.66 43.67 45.96
2 41 .26 44.39 43.45 43.27 46.57
3 43.78 42.82 42.95 44.78 43.92
4 41.88 44.76 43.72 44.51 44.26
5 41.28 43.82 43.49 44.58 42.95

 

 

 

 

 

3.1.2 Test conditions

28

The tests for the bottles were performed under 5 different conditions: 27°C/79%
RH, 30°C/60% RH, 30°C/65% RH, 30°C/70% RH and 30°C/75% RH. Among
them, 27°C/79% RH is the average environmental condition of the ASEAN
countries. 30°C/60% RH, 30°C/65% RH and 30°C/70% RH have been used as

the stability test condition for Climate Zone IV. 30°C/65% RH was the condition

 

used for Zone IV stability test when ASEAN proposed to WHO to change the

stability test condition to 30°C/75% RH.

3.2 Test Method - Gravimetric Method

The gravimetric method was also used to detect the WVTR of the HDPE bottles.
USP Chapter <671> was used as the reference. Some changes were adopted.
Five different conditions were prepared using ﬁve different salt solutions
combined with two temperature-controlled chambers. The salt solutions were
made following the description about Salt-Water System in ASTM E 104 (ASTM
E 104). Based on the references (Wexler, 1954 and O’Brien, 1948), the ﬁve salts
selected were:

Table 3.3 Salt solutions used to provide speciﬁed RH for bottle test

 

 

 

 

 

 

Salt Target RH @ 30°C Real RH 30°C"
(average during the
test)
Cobalt Chloride (COCIz) 62% 59.5%.
Sodium Nitrite (NaNOz) 65% 61 %'
Potassium Iodide (KI) 68% 64.5%T
Sodium Chloride (NaCl) 75% 74.5%'
Ammonium Sulfate 80% 79%. @ 27°C"
(NH4)2SO4

 

 

 

 

 

' Humidity variances: :I_-2%; " Temperature variances: i2°C

29

Two controlled chambers were used to provide the temperatures needed. But the
relative humidities of the chambers were also set based on the condition needed.
The purpose of this is to decease the inﬂuence from the outside environment
once the lid of the bucket is opened. One of the chambers was set at 30°C, 60%
RH, the other was set at 27°C, 79% RH. the relative humidity of the two
chambers were set based on the The salt solutions were stored in the two
chambers for stabilization for 2 weeks before the test. Salt solution CoCIz ,
NaNOz, KI and NaCl were stored in the chamber set at 30°C. While, (NH4)ZSO4
solution was placed in another chamber which was set at 27°C. Five randomly
selected bottles ﬁlled with calcium sulfate desiccant (about 50 gram) were tested
for each condition. Two bottles ﬁlled with the same amount of glass beads were
used as controls (blanks) for each condition. Enough torques were applied to all
the bottles once the desiccants or glass beads were added. Then, the bottles
were induction heat sealed and retorqued. Each set of 7 bottles was placed in a

bucket with humidity sensor.

30

 

H um idity

Samples sensor Connect to the humidity
+controls : collector

 

 

 

 

Bucket

Figure 3.1 Schematic drawing of exposure bucket for WVTR test of bottles
The samples were weighed every 7 days at the beginning. After 63 days, the
bottles were weighed every 14 days till the end. Two balances were used
because the ﬁrst one was broken in the middle of the test. The information about

the two balances is listed in table 3.4

Table 3.4 The balances used in the test for the bottles

 

Description Maximum Readability Test duration

 

Sartorious 210 g 0.1 mg Beginning ~ day 63

 

Ohaus 110 g 0.1 mg Day 64 ~ end

 

 

 

 

 

 

31

The temperatures for both of the chambers were recorded continuously using
analog recorders. The relative humidity of each bucket was measured using
humidity sensor which was placed in the bucket. For the two chambers, the
temperature variances are -_l-2°C. For all the humidity sensors inside the buckets,

the humidity variances are i2%. The temperature and the relative humidity of

each bucket were monitored before each weighing. Average WVTR value for

each bottle was calculated based on the weight gain of the sample.

When the test was ﬁnished, visual check was made for all the bottles. The seal

and the body of the bottle were checked.

32

Chapter 4. Results and Discussion

In the test, the temperature of each storage chamber and the relative humidity of
each bucket were checked every time before weighing. The temperatures
recorded were the digital reading of the chambers. The ﬁgure 4.1 and 4.2 show

the temperature and the relative humidity during the test.

 

 

 

32 I

30 90000.... O O O .—

 

 

28 .. -------——
III..IIIIIIIIII II I II I

 

 

‘ 0 Chamber 30°C
26 i

 

I Chamber 27°C
24 L_._- - . -_____-. I

 

Temperature °C

22 -- - 4H ,m i

 

 

 

 

 

0 50 100 1 50

 

 

Figure 4.1 The temperatures recorded during the test of the bottles

33

 

.__ -_#- _—.-—_— _.-._ ._._ _- .-_ _ ---» -u r A?

 

 

 

 

 

 

 

 

 

 

 

I 90 1
i 80 x x i ‘i x X x x 1* ﬁ‘“ 7 “i a:
$ x x x x x x x x x x x x x
70
o\° AAAAAAAAA A A A A OBucket1
3. 60 JJHJIm—Je—J—l—U _____,
.‘9 I Bucket 2
E 5° ‘“ “’ “’* ‘—‘
:3 A Bucket 3
5 40 P _ _r _ _ __ __ ‘ Li M.—
a; x Bucket4
E 30 _.., ﬂ _ FT ‘_—_ 'T x Bucket 5
3:) 20 4* a 4 —— — - I
10 _____. _E__#_E _-__ —— - —————— I
I 0 . . i
i o 20 4o 60 so 100 120 140 i
I I

 

 

 

 

 

 

 

Figure 4.2 The RH recorded during the test of the bottles

From the two ﬁgures, it can be seen that both temperature and the relative
humidity were very stable during the test. The records of temperature gotten from
the analog recorder were also checked each time. They also demonstrated the
stability of the temperature during the test. The effect of opening the lids of the
buckets (either to- take the samples out of the buckets or to put the samples into
the buckets) was also evaluated. Once the lid was opened, the reading of RH
sensor was monitored hourly till the RH went back to the original relative humidity.
It was found the RH inside the bucket decreased when the lids were opened, and
the RH would come back to the original value in about 6 hours once the lid is

closed.

34

4.1 Result of the test bottles

The weight gains vs time for each bottle was plotted in Fig 4.3 to 4.12. The
weight gain was considered as the amount of water vapor transmitted into the
bottle. When the ﬁrst balance used to weigh the samples was broken after 63
days, another balance was used as the substitute. So, for each condition, two
ﬁgures were plotted, one for each balance/time period. The Water Vapor
Transmission Rate (VVVT R) was calculated from linear regression of the weight
gained of the test bottle versus time. The WVTR of each bottle and the average

of ﬁve for each condition are shown in table 4.1.

35

 

i Bottles stored in 30°C. 595% R“ Chambe' i

 

 

 

 

 

 

 

 

 

 

40 I
23 3o — —— —— —— —V——— ——))I((—;~~— —— "T'BottIe—lﬁ
B 20 4— ——— -- —— —f——%- j . Bottle 3
:5, 15 if . . x Bottle 4
g 10 ~—-—,, if . e - x actual:

0
0 20 40 60 80
Day

 

 

 

 

Figure 4.3 The weight gains of the bottles stored in condition 30°C/59.6%RH

from day 0 to day 63

 

i Bottles stored in 30°C, 59.6%RH chamber

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

120

110 - — R+*_____ﬁw
’cE?) 100 -- 0 Bottle 1
E 90 mum -- x x “a I Bottle 2
‘3, 80 — —--———— ——y ———-- A Bottle 3
E, 70 »—‘~.——++e—------—~—i—! ‘ xBottIe4 i
g so -— -_ 3532sz

50 — —— — —— —— a

40 *

0 50 100 150

 

 

 

 

Figure 4.4 The weight gains of the bottles stored in condition 30°C/59.6%RH

from day 64 to day 119

36

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

60 I . _________
is?) 50 ~———~— —— oBottIe1
E, 40 __ _ - V A r' IBottle2
'ITI __ ABottle3
m 30 i R — — ' B ttl 4
vo-v ‘ X 0 e
J: __ ___I § _, _

.3 2° x i! 3' ﬁx Bottle5
; 10 ‘
o 1
. 0 20 40 60 80
. Day

 

 

Figure 4.5 The weight gains of the bottles stored in condition 30°C/60.9%RH

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

from day 0 to day 63
[—____ ”A W, .._-_.a
Bottles stored in 30°C, 60.9%RH chamber
100
,\ 95 if
E 90 ——-— - ---—~ ~ I I r 0 Bottle 1 7'
E 85 E“ I Bottle 2
3: 80 as — f—M—--——— A Bottle 3
E, 75 LL - hr" ‘x Bottle 4
g 70 ‘— ’———+%+ — x Bottle 5
65 ——m~--- -- - -
60
0 50 100 150
Day

 

 

Figure 4.6 The weight gains of the bottles stored in condition 30°C/60.9%RH

from day 64 to day 119

37

A-‘NNww-b-h
0100100!

Weight gain (mg
O 01 O 01 O

 

 

Bottles stored in 30°C, 64.5%RI-l chamber

 

 

 

 

 

 

 

 

 

 

 

 

—— F - ‘ — —~—-+A——~——
;:fi:m;:m____ ‘ i _c___..
20 40 60 80
Day

"M’I

I
I

 

iBottle 1

I Bottle 2
A Bottle 3
x Bottle 4
a: Bottle 5

 

 

 

 

 

 

 

Figure 4.7 The weight gains of the bottles stored in condition 30°C/64.5%RH

I
j 110
I

90
80
70
60
50

Weight gain (mg)

 

from day 0 to day 63

 

Bottles stored In 30°C, 64.5%RH chamber

 

100 ~

 

 

 

 

 

 

100 150

Day

 

”WRRRI

i786tt|e 1 I
I Bottle 2
A Bottle 3
x Bottle 4

 

1: Bottle 5) I

 

Figure 4.8 The weight gains of the bottles stored in condition 30°C/64.5%RH

from day 64 to day 119

38

 

Bottles stored in 30°C, 74.5%RH chamber

 

 

 

W“ W W I
I

 

 

 

 

 

 

 

 

 

5" . I
3’40 — ——-———— +I——- oBottle1I
E 30 H z __ w Th! _, I Bottle 2
3 A Bottle 3
E, 20 ”_i ‘ ‘W ‘ W W W W x Bottle 4
£10 J . —- ——— ———— xBottIeSJ-

‘ __ -
I 0 1 .
f o 20 4o 60 so I
I Day I
I , __ _. a- Z J

Figure 4.9 The weight gains of the bottles stored in condition 30°C/74.5%RH

from day 0 to day 63

 

W I

j Bottles stored in 30°C, 74.5%RH chamber I

 

I 120 I

 

 

 

 

 

 

 

 

 

 

 

 

 

I A 110 —~-- A ﬂ __g f ,I
g 100 F‘. ii _ A L ____ 0 Bottle 1 i
c A 8 - Bottle 2 I
a) 80 ‘ f—i ,. _ __ A Bottle 3
f9, 70 I x Bottle 4 I
é’ W WW W R WT”_ a: Bottle 5:
60 J» _-»__ _ __ __,
50 ' 1
0 50 100 150
I Day

 

Figure 4.10 The weight gains of the bottles stored in condition 30°C/74.5%RH

from day 64 to day 119

39

Figure 4.11 The weight gains of the bottles stored in condition 27°C/79%RH from

 

Figure 4.12 The weight gains of the bottles stored in condition 27°C/79%RH from

Weight gain (mg)
-t N on A
o O O O O

100 —

90

80

70

Weight gain (mg)

60

50

 

Bottles stored in 27°C, 79%RH chamber

 

 

 

 

 

 

 

I I
______ T+ML _
”*fn‘ ’ z c —
_4_l__..___ _ _ __ ____ .
20 40 60 80

day 0 to day 63

 

Bottles stored in 27°C, 79%RH chamber

 

 

 

 

 

 

 

 

L ,L-_-
‘ I

___, _ cc ‘ 1 __

_. ”i _E__ _-___--

50 100 150

Day

 

day 64 to day119

40

,, _ﬁ-___.,7 TI

I
I
SERIES]
I Bottle 2 I
A Bottle 3 I

x Bottle 4}
{Bottle 5:

L.

 

 

 

I

 

W

I Bottle 2 i
A Bottle 3 I
x Bottle 4 l
.*@%&I

 

$ Bottle 1‘

 

Table 4.1 WVTR value for each bottle

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Storage Bottl Balance 1/Day1 ~day 63 Balance 2/Day64 ~day
condition* e 119
Linear regression R2 Linear regression R2
(WVTR mg/d/bt) (WVTR g/d/bt)
30°C/59.6%RH 1 0.4138 0.975 0.4171 0.995
2 0.3986 0.972 0.4071 0.996
3 0.3952 0.970 0.4093 0.996
4 0.4579 0.980 0.4721 0.998
5 0.4064 0.973 0.4186 0.995
Average 0.4144 0.4248
30°C/60.9%RH 1 0.4500 0.977 0.4293 0.990
2 N/A N/A N/A N/A
3 N/A N/A N/A N/A
4 0.4788 0.980 N/A N/A
5 0.4299 0.994 0.4264 0.984
Average 0.4529 0.4279
30°C/64.5%RH 1 0.4798 0.981 0.4879 0.997
2 0.5240 0.984 0.5479 0.998
3 0.5498 0.986 0.5471 0.999
4 0.4781 0.981 0.4964 0.998
5 0.4840 0.981 0.4943 0.999
Average 0.5031 0.5147
30°C/74.5%RH 1 0.5243 0.989 0.5093 0.999
2 0.5526 0.987 0.5621 0.987
3 0.6607 0.991 0.6393 0.999
4 0.5264 0.982 0.5236 0.998
5 0.5095 0.984 0.5136 0.999
Average 0.5547 0.5496
27°C/79%RH 1 0.4398 0.975 0.4214 0.998
2 0.4267 0.974 0.4114 0.998
3 0.4929 0.979 0.4743 0.998
4 0.5238 0.984 0.5121 0.998
5 0.4840 0.978 0.4714 0.998
Average 0.4743 0.4581

 

 

 

 

 

 

*Humidity variances: i2%; Temperature variances

41

: i2°C

 

In ﬁgure 4.5 and 4.6, it can be found that bottle 2 and 3 reacted abnormally at
day 56, 63 and 77 comparing with other bottles. The weight gains for these two
bottles were much higher than other bottles at the three measurements. Linear
regression could not be made for them. So, in table 4.1, they are listed as WA.
For bottle 4, WVTR from day 64 to day 119 was also listed as N/A because the
R2 value is less than 0.9 when the linear regression was made. When the test
was ﬁnished, visual check was made for the three bottles. No defect was found.

The seal was intact and the body of the bottle was undamaged.

In order to make sure that there is no difference for WVTR values which were
calculated based on the data from two different balances, a scatter plot Fig 4.13
was made of x = WVTR for period 0 - 63 and y = WVTR for period 77 - 119 for
the n = 22 bottles (25 bottles minus the 3 for which there were no slopes

computed for at least one of the periods).

42

 

c.__-____ M-)

Scatterplot of Slopes
y = 0.9437x + 0.0264

 

 

 

 

 

 

 

 

 

 

 

 

 

I .
I

I 0 7 R2 = 0.9504 i

. I

' I

I §_>‘ I

5 I

(’6‘ I

e I

0‘ I

E i I

, I I

I 0.35 I , L . ‘

I 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 |

I WVTR (Day 0 - 63) I

L- ___. ______ _______ _ _.- _J

 

Figure 4.13 Scatter plot of WVTR from two different periods

Linear regression was made for all the data. The plot shows a consistency
between the two periods in regard to WVTR determinations. The regression line
is essentially the 45-degree line. This tells us that there is no difference between

the WVTR for the two different periods.

From table 4.1, it can be seen that the average WVTR value of the bottles at
30°C, 65.4%RH is higher than the average value at 27°C, 79%RH. The WVTR
values for all the bottles stored in 5 different conditions were compared in the two
ﬁgures, Figure 4.14 and 4.15. It also can be found that the range of the WVTR
values of the bottles at 30°C, 64.5%RH are higher than the range of the values at

27°C, 79%RH.

43

 

m—,__' w.]

 

 

 

 

 

 

 

 

 

I

I

I 0.7

i ’5‘ 0.65 ——-—- ———— x '* ocondition1 ,

I E 06 A ”ﬂ #2. ___, ﬂ _ __ E___ 30/59.6 f
g ' I condition 2

I 3 0.55 ~ — # —————— —A# —><— ——~— — —~~ 30/60.9

I 2 0 5 b. __ _ _ __ _____‘ § X J Acondition 3 ,

. g ' u A if 30/64.5 I

I v 0.45 e ——1 - E —- § — 2“ ~——-——~ xcondition 4 I

I m 0 4 4 __ __:_ _§ * J +____'_____ 30/745 I

i E ” ‘ ' xcondition5

I 0.35 -—— —— 27/79 ’

0.3 I
0 2 4 6
I Condition ,

 

Figure 4.14 Comparison of all the WVTR values of the bottles (day 0 ~ day 63)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I" *T “T— WW WI
I WVTR Comparison (day 64 ~ day 119) i
I ,6. 0-7 0 condition 1 I
a 0.65 x -* '~— 30/59.6
g 0 6 a?-—— --——A I condition 2
,5 0 55 4. _-* _ . x 30/69-9
2 0 5 __.__~__ _ it A condition 3
3’ - . r 30/64.5
V 0.45 ~~——~— I x condition 4
o: 0.4 , __,_ ____ ,____._,___ __ 30/74.5
0 35 %_ a: condition 5
- 27/79 I
0.3 I I
I 0 2 4 6 I
I condition

 

Figure 4.15 Comparison of all the WVTR values of the bottles (day 64 ~ day 119)

44

One tail, two-sample t-tests were performed for the WVTR values gotten from the
two different conditions 30°C, 64.5%RH and 27°C, 79%RH. Separate analyses
were done for the two periods where different balances were used. In each test,
the null hypothesis is that there is no difference in means and the alternative is
that mean WVTR values gotten from 30°C, 64.5%RH exceeds that of WVTR

values gotten from 27°C, 79%RH.

The p-values are reported in Table 4.3. Using the 0.05 level as indicating
signiﬁcance, the difference in average WVTR values 0.5031 — 0.4743 = 0.0297
mg/day for Days 0 - 63 is not signiﬁcant whereas the difference in average

WVTR values 0.5147 - 0.4581 = 0.0566 mg/day for Days 64 -- 119 is signiﬁcant

Table 4.3 P-values for t-tests of T-test for WVTR values gotten from

30°C/64.5%RH and 27°C, 79%RH

 

 

Result from Result from
balance 1 balance 2
(Day 0~63) (Day 64~119)
P-value 0.12 0.02

 

 

 

 

 

The P-values in both cases are small and the null hypothesis can be rejected for

the second period. These results support the hypothesis that the current WHO

45

long test condition is severe enough to represent the condition of ASEAN

countries for purpose of measuring WVTR of package

4.2 Result of the control bottles

The weight gain of every control bottle vs time was also plotted, and shown in Fig
4.16 to 4.25. Same balances were used for weighing the controls as were used
for weighing the test bottles. Two ﬁgures were plotted for the bottles stored in

each condition.

46

 

 

TC _Iwﬂ_m_ _I

I Control bottles stored in 30°C, 59.6%RH chamber

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I
I
6 I
As~w——w 4L - I

c» I___.’_,. _*_ ____#__H
s§_4 - ”1- _« I
% 2 __._£ ‘7 “-4 '“’I --_----r_ _ ocontrol bt1 I
3’31::f‘Tfiﬁ::i;:;f;f::;:::i '“m_m“m2I
a -1 _..__.._-_ﬁo _ _ _. I
-3 . . __. 4w.» __ O. __ a I

.4

o 20 4o 60 80 I
I Day I
I

 

Figure 4.16 The weight gains of the control bottles stored in condition

30°C/59.6%RH from day 0 to day 63

 

 

.5
O

 

 

 

 

 

 

 

 

 

 

 

 

 

I
I
A 38 —~ I
°’ I
g I I
.E 36 - r ' - --—' ~ I
g l 0 control bt 1 II
E 34 #H F IIcontrol bt2I
U) W .
.5 I
3 32 —-- a --—--. 9 a I
I . I
I 30 I
I o 50 100 150 I
I

Day

I. ____ ______.__*I,- _ _ .. _ I

 

Figure 4.17 The weight gains of the control bottles stored in condition

30°C/59.6%RH from day 64 to day 119

47

 

Control bottles stored in 30°C, 60.9%RH chamber

 

 

 

 

 

 

30 I
I
€20 - - ~~ —~—— —--ﬁ m--___I
% 15 ' Iocontrol bt1 II
D) *‘ ”i 'f a «— _ _____~_ I
E ’ LEMPQII
.2>1o —mv~ ____. --m__I II‘ #_ __ﬂ
é, o e e I
o
5 TTI‘ﬁ“”‘t I 1 ~-- I
I
I 0 I I .
I 0 20 4O 60 80 I
I Day I
I
I

 

Figure 4.18 The weight gains of the control bottles stored in condition

30°C/60.9%RH from day 0 to day 63

 

_ ,,, i;A,___.____._,

Control bottles stored in 30°C, 60.9%RH chamber I

 

 

 

,_I V.—

’ _ __ _. Iocontrolbt’1-II

. . e Icontrolbt2I

 

 

 

 

 

I I

 

I
I

 

 

 

o 50 100 150 I

I Day
L______________ "MAI

 

 

Figure 4.19 The weight gains of the control bottles stored in condition

30°C/60.9%RH from day 64 to day 119

48

 

 

I _ -__ _
Control bottles stored in 30°C, 64.5% RH chamber

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I 12
310 —— -— —— ——-——— -—— ~ ——
E . °
Z 3 — ”* r,“ 1 . ‘ —— _____ ~—
'66 . . ’ ° . ocontrolbt1I
O) 6 ’- —— _'I_« .-I #~.——~———I
E - . - control bt2 I
.9 4 ~— ‘ "W
on
3 2 -w __ #-_- ___# *_.
I 0 T ‘ I
I o 20 4o 60 80 I
I
I Day

- __________ ._ _ _ _ ____ ._ _____ _I

Figure 4.20 The weight gains of the control bottles stored in condition

30°C/64.5%RH from day 0 to day 63

 

I, ,,___

I Control bottles stored in 30°C, 64.5%RH chamber

 

 

 

 

 

 

 

 

 

 

 

 

50
A 48 —
8’
v46 . f u r ————— few _s ______,W_
'§ 9 control bt 1I=
a: 44 F , 7 I
E I control bt 2 I
g 42 —— o . e
o I
3 4o _-_______-_,_, ._ ' u ' - __ I
I
38 I
I o 50 100 150 I
I Day I

I ,I________W___,_.____ M- __ I

 

Figure 4.21 The weight gains of the control bottles stored in condition

30°C/64.5%RH from day 64 to day 119

49

 

I,_.__‘,._._

Control bottles stored in 30°C, 74.5%RH chamber

 

 

 

 

 

 

 

 

 

 

20 I
A 15 __.-, w—w +-———————e---——r-»-L—i~ .

I a) .

I g 10 __._ _ Q : L_._.._ _.__ __ I
% 5 _ _ f __ A_ 2 w _ ocontrolbth
3:3 I control bt 22I;
.% O I——— ii 74. __.___L_ _._.___ l____ I I
3 '5 I _ _l—_“Li—"“ I

10 I I I I I I

I ' I

I 0 20 40 60 80

Day I

 

Figure 4.22 The weight gains of the control bottles stored in condition
30°C/74.5%RH from day 0 to day 63

Control bottles stored in 30°C, 74.5%RH chamber !

 

 

 

 

(A)
o
I
I
I
I
I

 

50

O o O . I

I A45-~—f Jm—w—~n_~w— I

I a, I

E 40 ——‘ I w_r______ I m ____._ I

E omMmWHI

‘31 35 — I I

.. Icontrol bt2I.

c l
.9
(D
3

 

N
01
I
I
I

 

 

 

 

I

' o 50 100 150 I

I Day I

Figure 4.23 The weight gains of the control bottles stored in condition

30°C/74.5%RH from day 64 to day 119

50

 

____.___~ _—.— gur—

 

L, a ___ H

 

Control bottles stored in 27°C, 79% RH chamber

 

_I
m

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A 16 Lee TTT‘T ﬂeece—es
C» __ ___-.. M“ _ __ “ ___
EEDWW lbw: -Iwg *
E10 _,- ' ‘ . .D ocontrolbt1I.
E 8 M__ T“ ”a“ ~ m— I-contnolthI
a, 6 ___ .‘______ e__ ., _.e__ _._ _ % ,_
g 4 ___.. ._. ____ -_~___ W_W __*v__~_ ‘_I
2 —— — ——_—_ _ _ __ __ __ _
o
o 20 4o 60 so
Day
“J

 

 

 

Figure 4.24 The weight gains of the control bottles stored in condition

27°C/79%RH from day 0 to day 63

 

 

 

Control bottles stored in 27°C, 79%RH chamber

 

 

I
I
I
I
I
I

 

 

~——~ +~ww»~~-m__ :ammmq
LogntmLtﬂII

 

 

 

 

 

 

Weight gain (mg)
8 s t 8 3 8 a; 2 8 8
I
I
I
I

 

Day

I
I
I
e I I
I
I
I
i
I

 

 

Figure 4.25 The weight gains of the control bottles stored in condition

27°C/79%RH from day 64 to day 119

51

 

Theoretically, the weight of the control bottle should increase ﬁrst, then become
stable once the steady state is reached and the control bottle is at equilibrium.
The weight gain of the control bottle mainly comes from the sorption of the
package system and the permeation of water vapor. Once the stable state is
reached, which means the RH inside the control bottle is equal to the RH of the
storage environment, the weight of the bottle shouldn’t change any more. If linear
regression is performed between the weight gain of the control bottle and time
under steady state, the slope should be zero. From the ﬁgures above, it could be
found the control bottles displayed erratic behaviors. Some bottles lost weight
instead of gaining weight. And stable state could not be found for any control
bottle. In this test, the control bottles served as the reference of the system error
produced in measuring and handling. The result was expected to be used in data
analysis. However, from the outcome gotten from the control bottles, it’s hard to
relate it to the results of the test bottles. And WVTR, which is the rate of weight
gain, should be a constant when the steady state is reached. Therefore, it ‘3

possible to ignore the change of the control bottles.

In this test, there are a few possible reasons that could affect the results of the

test and control bottles:
1. The variance of the bottles. Although the bottles are from the same batch,
there are still some differences in thickness, crystallinity and etc which

affect the WVTR of the bottle.

52

. When the bottles were induction sealed, there might be differences of the
current used for each bottle. If the current is strong enough, micro hole
could be produced on the aluminum layer of the liner. That could affect the
WVTR of the bottles.

. During the test, the temperature and the relative humidity were stable.
However, there were some small variance that could affect the result.

. System error was produced by handling and measuring.

. Sample size was not big enough.

53

Chapter 5 Conclusion

The water vapor transmission rates of HDPE bottles in ﬁve different conditions
were tested using gravimetric method, and statistical analysis was performed for
all the data. The hypothesis that the stability test condition 30°C/65% RH should
be severe enough to represent the ASEAN average condition 27°C/79% was
veriﬁed. It means that 30°C/65% RH could be used as the stability test condition
for all the countries in Climate Zone IV. And it’s not necessary to use 30°C/75%

RH as the test condition, although it’s helpful to make the drug products safer.

The results also proved that the relative humidity of the environment alone is not
the only issue that affects the result of stability testing. Actually, it is the storage
temperature, partial pressure for water vapor and moisture permeability of the
package that affect the result of stability testing, because they all affect the
humidity inside the package. It is the humidity inside the package that affects the

product.

There is some further investigation which could lead to an increased
understanding of this research.
1. Monitor and record the change of storage condition continuously. In this
case, more information could be gathered and used in data analysis.
2. Enlarge the sample size.
3. Investigate the behavior of the control bottles, such as sorption of the

bottle and liner.

54

REFERENCES

ASEAN 1, “Overview Association of Southeast Asian Nations”,
http://www.aseansec.org_/64.htm

ASEAN 2, “ASEAN Cooperation on Standards and Conformance to Facilitate
Trade in The Region”, http://www.aseansec.org/6667.htm)

ASEAN 3, “ACCSQ Pharmaceutical Product Working Group”,
http://www.aseansec.org/14903mm.

ASTM E 96, “Standard Test Methods for Water Vapor Transmission of Materials”,
ASTM lntemational

ASTM F 1249, “Standard Test Method for Water Vapor Transmission Rate
through Plastic Film and Sheeting Using a Modulated Infrared Sensor", ASTM
lntemational

ASTM E 104, “Standard Practice for Maintaining Constant Relative Humidity by
Means of Aqueous Solutions”, ASTM lntemational

Bailey, L.C.; Medwick, T. 1993, “Mean Kinetic Temperature — A Concept for
Storage of Pharmaceuticals”, Phannacopeial Forum, 19(5), 6163-6166

Bott, R. F.; Oliveira, W. P. 2007, “Storage Conditions for Stability Testing of
Pharmaceuticals in Hot and Humid Regions”, Drug Development and Industrial
Pharmacy, 33, 393-401

Doty, P.M.; Aiken, W.H.; Mark, H. 1946, “Temperature Dependency of Water
Vapor Permeability”, Ind & Eng. Chem, 38(8), 788-791

Grimm, W. 1986, “ Storage conditions for Stability Testing (Part 2)”, Drug made
in Germany, 29, 39-47

55

Grimm, W. 1998, “Extension of the lntemational Conference on Harmonization
Tripartite Guideline for Stability Testing of New Drug Substances and Products to
Countries of Climatic Zones Ill and IV”, Drug Development and Industrial
Pharmacy, 24, 313-325

Lockhart, H.; Selke, 8.; Yoon, S. 2006, “The Role of Container-Closure Systems
in Stability Testing For Climate Zone IV”, Phannacopeial Forum, 32(2), 688-694

O’Brien, F.E.M. 1948, “The control of Humidity by Saturated Salt Solutions — A
Compilation of Data”, J. Sci. lnstr., 25, 73

PMA’s Joint QC-PDS Stability Committee 1991, “Room-Temperature Stability
Studies: Storage Conditions”, Pharmaceutical Tech., Sep., 93-96

Rubino, M. 2005, “Permeability and Shelf Life” Course pack for School of
Packaging, MSU PKG 815, 2.1-3.30

Schumacher, P. 1972, “The Impact of Climate Classiﬁcation on The Stability of
Medicines”, Pharm Ind, 34, 481-483

Singh, S.; Kumar, V. 2006, “Recent Developments on Long — Term Stability Test
Conditions”, The Pharma Review, Dec, 61-68

USP 1 2004, “Stability Considerations in Dispensing Practice”, United States
Pharmacopeia, 27, 2605

USP 2 2004, “Containers — Permeation” United States Pharmacopeia, 27, 2296
- 2297

Van Krevelen, D.W. 1997, Properties of Polymers. Their come/ation with
chemical structure: their numerical estimation and prediction from additive group
contributions. 3’“ edition, Elsevier, Amsterdam

Wexler, A.; Hasegawa, S. 1954, “ Relative Humidity-Temperature Relationship of

Some Saturated Salt Solutions in the Temperature Range 0° to 50°C”, J.
Research NBS, 53, 19

56

WHO 2004, “Aspects of the Quality Assurance”, WHO Drug lnfonnation, 18(2),
113-116

WHO 1, “Stability Testing of Active Substances and Pharmaceutical Products”,
WHO, Working Document OAS/06.179

WHO 2, “Stability Studies in a Global Environment” WHO, Working Document
OAS/05.146

57

   

II IIIIIIIIIIIIIIIIIIIII