» .43.. H5
“awn-w r ‘
M94792? .5
3533?: “L753: “mu.
” r» 2.1x:
“254?“

awn e-

‘C u
“WES“; ;

 

r‘
,3

rd- :.:._ :5;
y” fi;;_;;8y‘.

‘3. ‘ W ,3
e 3%.

 

.: fig-9,. ' ‘I
~ may ‘= «

x
O»
’7."

‘C

. I" f"
“th51.” .

.vl.

 

IVERS SITY LIB

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII| III

3 1293 01028 2840

IIIIIIIIIIIIII

 

 

 

 

 

This is to certify that the
thesis entitled
MODIFYING THE SORPTION CAPACITY OF POLYETHYLENE
FOR ORGANIC COMPOUNDS

presented by

Takashi Urata

has been accepted towards fulfillment
of the requirements for

Master degree in Packaging

 

 

 

 

I

Major professor

 

Date July 13, 1994

 

0-7639 MS U is an Aflirmative Action/Equal Opportunity Institution

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE II RETURN BOXtomnavothchockthyum

TO AVOID FINES Mom on or Mon duo duo.

DATE DUE DATE DUE DATE DUE

 

     

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU IIN'I Nflmntivo ActioniEqml Opportunity Im
W

 

m1

MODIFYING THE SORPTION CAPACITY OF POLYETHYLENE
FOR ORGANIC COMPOUNDS

By

Takashi.Urata

A THESIS

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

MASTER OF SCIENCE

School of Packaging

1994

ABSTRACT

MODIFYING THE SORPTION CAPACITY OF POLYETHYLENE
FOR.ORGANIC COMPOUNDS

By

Takashi‘Urata

Several concentrations of sorbents (activated carbon and
Tenax) blended polyethylene fiLms were made to observe their
sorption behavior against limonene, ethyl acetate, and
toluene. .A gravimetric method was used to determine the
sorption characteristics of the films. In addition, density
and selected mechanical properties of the sample were
measured. To obtain a close look of the distribution of
sorbents in the polymer's matrix, sample were observed under
an optical microscope.

Activated carbon blended polyethylene films sorbed
greater amounts of organic volatiles than polyethylene for
all sorbates. Tenax blended polyethylene films showed good
sorption behavior for ethyl acetate and toluene. However,
very little sorption was observed for limonene.

From the sorption study of several types of polyethylene
samples (granular, thin, and thick films), the sorption
characteristics seemed to be concentration dependent. But

that was not confirmed in this study.

Acknowledgements

I would like to express sincere thanks and appreciation
to my major advisor, Dr. R. Hernandez, for his inspiration,
counsel, and encouragement during the time spent at Michigan
State University.

Appreciation is also extended to Drs. J. Giacin, J.
Lucas, and P. Ng for serving as members of the guidance
committee and for their very helpful comments and suggestions
for the improvement of this thesis.

In addition, I would like to thank.Mr. Mike Rich for his
help and advice on extrusion process and optical observations
at Composite Materials and Structures Center, Michigan State
University.

Special thanks are expressed to Toppan Printing Co.
Ltd., who made this graduate work possible, for their
continued support.

Finally, I would like to express my appreciation to my

family and all my friends for their help and encouragement.

iii

Table of Contents

List of Tables ....................................... vi
List of Figures ..................................... vii
Nomenclature ......................................... ix
I. Introduction ....................................... 1
II. Literature review ................................. 4

1. Product/Package Interactions .................. 4

2. Sorption of Organic Compounds by Sorbents ..... 6
2.1 Activated Carbon ....................... 12
2.2 Tenax .................................. 13
2.3 Clay ................................... 14
2.4 Plastic Packaging Materials ............ 15

3. Analytical Method of Determining
Sorption Capacity .......................... 18

III. Materials and methods ........................... 20
1. Sorbent Blended with Polyethylene ............ 20
2. Determination of Density ..................... 21
3. Determination of Mechanical Properties ....... 22

4. Optical Microscope Observation
Of sample Film OIOOOOOOOOOOOOOOOOOOOO0.0... 23

5. Determination of Sorption Characteristics
by Gravimetric Method ...................... 24

IV. Results and Discussion ........................... 28
1. Density of Samples ........................... 28
2. Mechanical Properties ........................ 34
3. Optical Microscope Observation

Of sample Films oooooooooooooo ..... 000000000 40

iv

V

4. Sorption Characteristics by
Gravimetric Method ........................

v. sumaly 0.0...
VI. Future Studies

VII. Appendices

Appendix A : Blending Sorbents into Polyethylene
1. Determination of Sampling Time ..........
2. Blending Tenax and LDPE .................
3. Making Activated Carbon Blended.LDPE ....

Appendix B : Procedure for Making Film.Samples ..

Appendix C : Procedure for the Determination of

Sorbate Calibration Curve by Gas

Chromatography .....................

VIII. Bibliography

47
65
68

69
69
71
74

76

79
89

Table #
1

\IO‘U‘Ifi w

on

10
11

12

13

14

15

16
17
18

List of Tables

Title

Some structural characteristics of activated

carmnand Tenax ......OCOOOOOOOOIOOO

Samples used to determine density .............

Samples used to determine mechanical

properties and sorption characteristics .......

Density of activated carbon blended samples ...

Density of Tenax blended samples ..............

Mechanical properties of sample films

Concentration of each sorbate for each

vapor activity ................................

Results of sorption test for limonene

Results of sorption test for ethyl acetate ....

Results of sorption test for toluene

Results of sorption test of several types

of polyethylene samples for limonene

Maximum.calculated sorption capacity for

limnene ......OOOOOOOOOOOOOO......OOIOCOOOO...

Maximum.calculated sorption capacity for

ethYI acetate ......COCOOOOOOO......OOOOOOOO...

Maximum.calculated sorption capacity for

tOIuene 00.000.000.000.........OOOOIOOOIOOOOOOO

The number of molecules of organic compound

striking the surface of sorbents per 8

Data for calibration curve of limonene

econd ...

Data for calibration curve of ethyl acetate ...

Data for calibration curve of toluene

vi

10
20

21
31
31
35

47
48
49
50

51

52

53

53

84
84
85

List of Figures

Types of sorption separations ..................

carbon blended samples ........................

samples .......................................

Of activatedcarmn ..........OOOOOOOOOOOOOOOCC

Of Tenax 0.000.000.0000.......IOOICOCC0.00.0...

activated carbon and elastic modulus ..........

andelasticmdulus ......OOOIOOOOOOOOOO0......

and tensile strength ..........................

(SWt%' x200) ......OOOOIOOOOOOOOOOOO00.0.00...

(x400) ........................................

Figure #' Title

1

2 Shematic diagram.of electrobalance system. .....

.3 Calibration curve for density of activated

4 Calibration curve for density of Tenax blended

5 Relationship between density and concentration

6 Relationship between density and concentration

'7 Relationship between concentration of

8 Relationship between concentration of Tenax

9 Relationship between concentration of sorbent

10 Relationship between concentration of sorbent
and % elongation at break

11 .A particle of activated carbon (x200) .........

12 A particle of Tenax (x200)

13 Surface of activated carbon blended film

14 Surface of Tenax blended film (3.9 wt%, x100) .

15 Cross sectional view of polyethylene film

16 Cross sectional view'of activated carbon

blended film (5 wt%, x400)

vii

8
26

29

30

32

33

36

37

38

39
42
42

43
43

44

44

17

18

19

20

21

22

23

24

25

26
27

28
29
30
31

viii

Cross sectional view of Tenax blended

film (3.9 m%, x400) ........

Cross sectional view of polyethylene

film (x1000) ......O...’.00....OOIOOOOOO0......

Cross sectional view of activated carbon
blended film.(5 wt%, x1000) ...................

Cross sectional view of Tenax blended

film (3.9 wt%, x1000) .......

Relationship between concentrationof activated
carbon blended in a film and sorption

coefficient of limonene .....

Relationship between concentrationof activated
carbon blended in a film.and sorption

coefficient of ethyl acetate

Relationship between concentrationof activated
carbon blended in a film and sorption

coefficient of toluene ......

Relationship between sorption coefficient
of limonene and type of polyethylene ..........

Relationship between surface are per unit
mass and type of polyethylene sample ..........

Color gradient chart ..........................

Relationship between extruding time and

c010r gradient ...............O................

Shematic of the laboratory press ..............

Calibration curve for limonene

Calibration curve for ethyl acetate ...........

Calibration curve for toluene

45

45

46

46

55

56

57

58

59
72

73
77
86
87
88

B

C

[caqleq

[Cpleq

Nomenclature

activity of sorbent

experimental constant in reciprocal pressure units
which has limited practical application
(Equation 2)

penetrant driving force in units of concentration
or pressure (Equation 5)

experimental constant (Equation 3)

equilibrium.concentration of aroma compound in
aqueous solution (Equation 4)

equilibrium.concentration of aroma compound in
plastic film.(Equation 4)

concentration of sorbate in sample film
Boltzmann's constant
equilibrium partition coefficients

number of molecules striking each cm? of the

surface every second (molecules/cmzlsec)
(Equation 1)

pressure of sorbent at test temperature

saturated pressure of sorbent at test temperature
molecular weight (Equation 1)

weight of blended polyethylene (Equation 9)
weight of blended sorbent (Equation 9)

total amount (mass) of vapor absorbed by the

polymer at equilibrium for a given temperature
(Equation 5)

number of molecules of solvent (Equation 7)

ix

8x><z

X

number of molecules of polymer (Equation 7)
sorption coefficient (kg/kg-Pa)

entropy of mixing

maximum.calculated sorption capacity (Equation 9)
sorption coefficient of polyethylene (Equation 9)
sorption coefficient of sorbent (Equation 9)
absolute temperature (°K)

valves in Figure 2 (V1 ~ V9)

volume fraction of solvent (Equation 7)

volume fraction of polymer (Equation 7)

volume of gas adsorbed per unit mass of adsorbent
(Equation 2)

volume of gas adsorbed per unit of adsorbent with
a layer one molecule thick (Equation 2)

number of lattice in polymer matrix (Equation 6)
number of occupied sites in a lattice (Equation 7)
interaction parameter (Equation 8)

weight of the polymer sample (kg) (Equation 5)

Introduction

Nowadays, consumers want to have higher quality and
safer food products. Given that, attention to the
interaction between food products and.packages is becoming
more important. Keeping good environmental conditions
inside of a product's package provides longer shelf life and
better stability during storage of that product.

To modify the environment inside the package, several
active packages have been developed (Labuza and Breene,
1989). An active package is one which can interact with a
packed product or its environment to alter the in-package
environment (Sacharow, 1991). An example of an active
package is the sachet type oxygen absorber which has the
inherent property of absorbing oxygen. .Although this
packaging method requires high barrier packaging materials
and better seal integrity, it can extend the shelf life of
the product considerably.

Limitations to shelf life are not only caused by oxygen,
but also by many organic volatile substances. Such volatile
substances may come from.outside of the package. For
example, they may come from the ink on the carton in which
the product is packaged, or from.the secondary package such
as corrugated board container. The amount of these
impurities may be very small, but may impart an off-flavor to

1

2

the packaged product. ‘Very small amounts of volatile, low
molecular weight organic compounds can influence overall
product flavor (Parliment, 1987).

In order to control or minimize this type of
product/package interaction, at least two approaches are
possible. One is the use for packaging of a high barrier
polymer for which organic volatile substances have very low
diffusion coefficient values. The other is the use of a
polymer which has the barrier property of high capacity for
retention of organic substances.

There is a need for the development of a material that
has a higher sorption capacity than regular polymers. This
may be accomplished by modifying a commercial polymer with a
known compound which has high sorption properties such as
Tenax or clays (Bigg, 1992). .Additionally, these modified
polymers could be used to as a flavor releaser. Flavor
could be added to the sorbents before packing the product,
allowing the sorbents to release the flavor slowly during
storage. Therefore, even after opening, the product would

keep good flavor longer than in a typical package.

The objective of this thesis was to explore the
possibility of modifying a polymer to increase its sorption
characteristics by blending an active sorbent compound with a
polymer. Since the mechanical properties of the blended
polymer can be affected by the concentration and distribution
of an active sorbent compound, selected.mechanical properties

of the blended polymer were evaluated at several

3

concentrations of active sorbent compounds.

Literature Review

1. Product/Package Interactions

Flavor and aroma characteristics arise from.low
molecular weight, volatile organic substances. These
compounds are often present in food products at low
concentrations, which influence the flavor profile of the
foods (Parliament, 1987). Flavor causing compounds are
known to be sorbed into packaging materials during their
packaged period (Hotchkiss, 1988) and this is called
scalping. Packaging polymers can sorb enough aroma
compounds to be detected by human senses. Therefore, the
package material often alters the flavor characteristics of a
packaged food.

Other flavor changes can result from interactions with
heat, light, or oxygen in the head space, or from.ohemical
changes in the container (DeLassus et al., 1988).

Plastic containers have more complex flavor interactions
than glass or metal. In addition to the factors of heat and
light, plastic containers have three more concerns: (1)
Molecules from the environment can permeate the package wall
and enter the food. These molecules may be as simple as
oxygen or as complex as the floral essence from the laundry
products in the next aisle of the supermarket; (2) molecules

4

5

that may come from the plastic itself, from a coating, or
from.an adhesive might migrate to the food; and (3) flavor
molecules can leave the food by permeation into and through
the plastic walls of the package.

Therefore, organic solvents and flavor and aroma
molecules may contaminate food from.several external sources
such as printing inks or adhesives and cause flavor
degradation (Strandburg et al., 1991). To minimize flavor
losses many plastic food packages have a barrier layer in
their structures. This barrier layer minimizes the entrance
of flavor-degrading material from.outside of the package
(Hotchkiss, 1988). The barrier layer is made of high
barrier polymers such as polyvinyl chloride (PVDC)
copolymers, ethylene-vinyl alcohol (EVOH) copolymers,
amorphous nylon polymers, silica coated polymer films, and
high melt viscosity polyesters (Bigg, 1992). Flavor
degradation from external materials can be reduced or
eliminated by such barrier layers. Different techniques are
required to reduce flavor loss from.food into the packaging
material.

The loss of flavor compounds due to sorption by plastic
packaging materials can be considerable. The equilibrium.
uptake of volatile flavor compounds by plastic packaging
materials may be on the order of several weight percent of
the plastic. Because of sorption by the plastic packaging
materials, the polymer layer in contact with the food may
change the flavor of a food by selectively sorbing one or

more compounds which influence the overall aromatic

6

characteristic of that food. The specific compound which
would be sorbed by the polymer may be a key flavor component,
or one which combines with another to produce flavor of the
food. For example, polymers such as polyolefins have been
shown to selectively sorb flavor constituents (Foster, 1987).
Orange juice develops a “flat", non-fresh taste due to loss
of volatile compounds that were sorbed by polymeric packaging
materials (Marshall at al., 1985).

Migration of some component of food to the packaging
materials which raises possible health concerns. Food
processors generally claimuthat their suppliers provide
letters of certification that their materials will not
decrease quality of food as a result of unnecessary
migration, or meet certain Food and Drug.Administration (FDA)
regulations (Bentley, 1988; Heckman, 1991). Low'molecular
substances migrating from plastic packaging materials in
direct contact with the food are considered as indirect food

additives by FDA (21 CFR 177.101o-.2ooo, 1993).

2. Sorption of Organic Compounds by Sorbents

(/ The interaction at the interface between two phases
involving the transition of a molecule from one phase to
another and driven by different concentration values, can be
considered as the sorption of the molecule (Kovach, 1978).
Fromithe viewpoint of the distribution of the substances to

be sorbed, two types of sorption can happen: adsorption and

7

absorption. .Adsorption involves the accumulation or
concentration of substances at a surface or interface. The
process can occur at an interface between any two phases;
liquid-liquid, gas-liquid, gas-solid, or liquid-solid
interfaces. On the other hand, absorption is a process in
‘which the molecules or atoms of one phase interpenetrate
almost uniformly to the other phase to form.a solution-like
part with the second phase.)

The phenomenological differences between adsorption and
absorption are illustrated graphically in Figure 1 for
reactions of each type in which a substance moves from liquid
phase to solid phase (weber, 1972). For adsorption, curves
I and III indicate curvilinear dependence of the amount
concentrated at the solid surface on the amount remaining in
the solution phase for favorable and unfavorable separation
patterns, respectively. Curve II represents a linear
adsorption pattern. It also shows absorption which occurs
in direct proportion to concentration. These graphs show

the §g§9£EEiQQ,is described by a Langmuir type of isotherm,

 

 

and the absorptignflis represented by Eenry's Law. .As stated

 

above, absorption is a phenomena which the molecules or atoms
of one phase interpenetrate almost uniformly to the other
phase to form.a solution-like part with the second phase.
Therefore, the amount of substance moved across the interface
and concentration of substance moves from liquid phase to
solid phase ShOW’a linear relationship. .Adsorption involves
the accumulation or concentration of substances at a surface

or interface. So its curve indicates curvilinear dependence

 

 

 

 

 

Amount of substance moved across

the interface

Concentration of substance moves
from liquid phase to solid phase

Figure 1. Types of sorption separations
(Weber, 1972)

9

of the amount concentrated at the solid surface on the amount
remaining in the solution phase.

There are two methods by which organic compounds may be
adsorbed: physical and chemicalfladsorption (Cheremisinoff and
Morresi, 1978). When physical adsorption occurs as a result
of energy differences and/or electrical attractive forces, or
weak van der waals forces, the organic molecules become
physically fastened to the adsorbent molecules. This type
of adsorption may be multilayered: each molecular layer forms
on top of the previous layer with the number of layers being
proportional to the organic compound's concentration. .More
molecular layers form with higher concentrations of the
organic compound.

When.a chemical compound is produced by the reaction
between the adsorbed organic molecule and the adsorbent,
chemical adsorptignhas occurred. Unlike physical
adsorption, this process is one molecule layer and quasi-
irrggersible. Because energy is required to form.the new
chemical compound at the surface of the adsorbent, energy
would be necessary to reverse the process. The
reversibility of physical adsorption is dependent on the
strength of attractive forces between organic compound and
the adsorbent. If these forces are weak, desorption is
readily achieved.

Factors affecting adsorption include: (1) the physical
and chemical characteristics of the adsorbent, e.g.: surface
area, pore size, chemical composition, etc.; (2) the physical

and chemical characteristics of the organic compound, e.g.:

10

molecular size, molecular polarity, chemical composition,
etc.; (3) the concentration of the organic compound in the
phase in contact with the adsorbent; (4) the characteristics
of the phase in contact with the adsorbent, e.g.: pH,
temperature, and (5) the residence time of the system.

Therefore, the diameter of pores and the surface area of
adsorbents affect the capacity and characteristics of their
adsorption. Table 1 presents some structural

characteristics of activated carbon and Tenax.

Table 1. Some structural characteristics of
activated carbon and Tenax

 

Activated carbon Tenax
Surface area (mZ/g) 500~2000a 18.60
Average pore radius (A) 23~30b 720c
Density (g/cm3) 0.40--0.51b 0.37c

 

aKovach, 1978.
bDarco, Atlas Chemical Div., ICI America, Inc., 1965
cSakodynskii et al., 1974

In the case of the gas-phase adsorption, the number of
volatile molecules present at the surface of adsorbent is
dependent on the number that reach the surface and on the
residence time of these molecules on the surface (Kovach,
1978). The numbegygf volatile molecules that reach the
surface of adsorbent is directly related to the speed of the
molecules within the system. 'This number can be calculated

using Hansen's and. Boris-..Gar 911.9929? Pauatiorzsz-

11

n = 3.52x1022xfip (1)
where n = number of molecules striking each cm2 of the
surface every second
p = pressure of sorbent
M = molecular weight

absolute temperature

The Langmuir adsorption isotherm describes the
adsorption behavior between the adsorbent and the gas phase
of volatile compounds. The Equation of the Laggmpi;

adsorptigpflisothermLis shown as Equation 2.

V

V b
Vg = WE PP (2)
where Vg = volume of gas adsorbed per unit mass of adsorbent
Vm = volume of gas adsorbed per unit of adsorbent with

a layer one molecule thick
b = experimental constant in reciprocal pressure units

which.has limited practical application

The Langmuir isotherm only describes monolayer

adsorption of an adsorbent. For multilayer adsorption, the

 

__4.._.-q.p ...... .._..- cm H...

 

BET equation (Equation 3) can express the isotherm.at the gas

i

phase.

‘I/‘J

V C
m a (3)

v= (1 —a)[1+(C— 1)a]

12

where C experimental constants

a - activity of sorbent

.9.
Ps

a:

p8 = saturated pressure of absorbent at test

temperature

2.1 Activated Carbon

The effectiveness of activated carbon for the removal of
organic compounds by adsorption is enhanced by its large
surface area, a critical factor in the adsorption process
(Cheremisinoff and Morresi, 1978). This large surface area
is provided by its pore structure. Since adsorption is
possible only in those pores that can be entered by
molecules, the carbon adsorption process is dependent on the
physical characteristics of the activated.carbon and the
molecular size of the organic compound.

Of much less significance than its surface area is the
chemical nature of the carbon's surface. This chemical
nature or polarity varies with the carbon type and can
influence attractive forces between molecules. Silica gel,
an adsorptive media that is not a carbon compound, has a
polar surface which also shows an adsorptive preference for
unsaturated organic compounds as opposed to saturated
compounds. However, for the most part activated carbon

surfaces are nonpolar, making the adsorption of inorganic

13~
electrolytes difficult and the adsorption of organic
compounds easy.

Cookson (1978) reported that aromatic rings of aromatic
compounds act as electron acceptors, and carbonyl oxygens on
the carbons are electron donors; this implies a donor-
acceptor relationship. For non-polar aliphatic compounds,
the oxides on the surface of activated carbon prohibit
adsorption, and the metals on the activated carbon's surface
allOW'it. To permit the adsorption of polar compounds,
acidic oxides provide polar characteristics to activated

carbon.

2 . 2 Tenax

Tenax is a 2,6-dipheny1-p—phenylene oxide based porous
polymer column packing material (Van‘Wijk, 1970). It is
suitable for the separation of high boiling polar compounds.
This material is also used in adsorbent traps to collect
organic compounds from air or gas (Patton et al., 1992; Sharp
et al., 1992). For this use, the collection efficiencies of
Tenax for low'molecular weight, volatile compounds less than
or equal to C6 are often unacceptable for quantitative
purposes (Heavner, 1992). Therefore, to collect volatile
compounds in the air, “multisorbent” cartridges containing
Tenax and other solid adsorbents are usually used.

The polarity of Tenax is relatively low, and its

specific surface area is small (Daemen et al., 1975). These

14

characteristics provide short retention times at low
temperature for effective separation.

Vejrosta et a1. (1988) reported the Langmuir adsorption
isotherm was fit for benzene sorption on Tenax. This
relationship was also applied to the sorption behavior of n-

alkanes (Vejrosta et al., 1989).

2.3 Clay

The interaction between.clay minerals and organic
compounds has been a subject of research because this
interaction raises the expectation that the adsorption of the
organic fraction in the soil on the mineral particles would
contribute to the physical stability of soil (Frissel, 1961).

The interactions between.clays and organic substances,
especially organic cation, have been studied extensively
(Dixon and weed, 1977; Mortland, 1970, 1986). The following
factors may affect clay-organic cation interactions: (1) the
properties of minerals, including cation exchange capacities,
surface charge density, the cations on the exchange sites,
the origin of charge, etc.; (2) the characteristics of
organic cations, including molecular charge, lone pair
electrons, polarity, molecular weight, etc.; and (3) the
environments, including pH buffering, solution ionic
strength, temperature, concentration of the organic cation,
the presence of other cations, etc.

Many mechanisms for clay-organic interactions are

15

discussed (Mortland, 1970, 1986), such as: (1) ion exchange:
substitution of organic cations for hydrated metal cations on
the mineral surface; (2) protonation: proton-donating and
proton-accepting; (3) coordination and ion.dipole: organic
cations acting as ligands; (4) hydrogen bonding: relatively
weak interactions between.proton and electronegative
elements; and (5) van der waals forces: relatively weak

physical interactions.

2.4 Plastic Packaging Materials

Many studies are done for sorption characteristics of
organic materials by plastic packaging materials (Hirose et
al., 1988; Imai, 1988; Schroeder, 1989).

Equilibrium partition coefficients (Re) for sorption of
aroma compounds by plastics packaging materials in an aqueous
system were calculated by following equation (Kwapong and

Hotchkiss, 1987).

_ [C9199

K _ _______
e [C aqieq

J" (4)

where [C P199 : the equilibrium concentration of aroma

compound in the plastic film

[C aq] eq : the equilibrium concentration of aroma

compound in the aqueous solution

16

The resulting Ke values were indicative of an aroma
compound's strong affinity for plastic films, such as low
density polyethylene, and ionomers. Although Equation 4 is
based on an aqueous system, the partition coefficient would
be able to apply between polymers and sorbents.

A sorption coefficient can be calculated from.sorption

experiments using the following equation:

a
8
<

U)
II

(5)

8
m

where S : sorption coefficient
1h» : total amount (mass) of vapor sorbed by the
polymer at equilibrium for a given temperature
: weight of the polymer sample

(0
B : penetrant driving force in units of concentration

or pressure

Equation 5 can be applied to organic vapors exhibiting
non-ideal diffusion, as well as non—interacting penetrants
(Hernandez et al., 1986).

To explain the mixing of polymer solution, a two-
dimensional lattice structure is used to represent non-
polymer liquid (Billmeyer, 1984). Although the molecules of
a pure component can be arranged in only one way on such a
lattice, assuming that they cannot be distinguished from one
another, the molecules of a.mixture of two components can be

arranged on a lattice in a large but calculable number of

17

ways: W. By the Boltzmann's principle, the entropy of

mixing is expressed as following:

AS = k ln(W) (6)

where AS = entropy of mixing

k = Boltzmann's constant

It is assumed that the polymer molecules consist of a
large number of chain segments of equal length, flexibly
joined together. Each link occupies one lattice site. The
solution is assumed to be concentrated enough that the
occupied lattice sites are distributed at random rather than
lying in.well-separated regions with X occupied sites each.
There are fewer ways in which the same number of lattice
sites can be occupied by polymer segments: fixing one segment
at a lattice point severely limits the number of sites
available for the adjacent segment. . To calculate W for such

a model, the Flory-Huggins'the9§y_of polymer solutions are

 

used (Flory, 1953). The equation is as follows:

AS = -k(N11n(V1) + N21n(V2)) (7)

number of molecules of solvent

g
(D
H
ID
2
H
ll

2
A)
II

number of molecules of polymer

V1 and V2 = volume fractions defined as:
_ N1

‘71— N1 +52

XN2

V2=N1+XN2

18

‘where X = number of occupied sites in a lattice

The entropy of mixing of each component can be
represented by the volume fraction of the other component in
diluted solutions (Huggins, 1942). For high molecular
weight polymers, the Flory-Huggins expression is shown as

Equation 8.

ln (a) = 1n (V1) + V2 4- XV22 (8)

where x = interaction parameter

3. Analytical Method of Determining Sorption Capacity

The Electrobalance Sorption/Desorption Apparatus (Cahn
Instruments Inc., Cerritos, CA) has been used in sorption
studies of volatile flavor components (MohneyLM1988; Roland

and Hotchkiss, 1991). In this apparatus, a test sample film

——. ...—tn»)... ... —.—->'-r"‘

is hung in the hangdown tube of an electrobalance and a gas-
phase vapor generating/dilution system is used to get
constant vapor concentration. .A.constant concentration of
penetrant vapor stream.is produced by bubbling nitrogen
through liquid penetrant and dilution with untreated
nitrogen, then passing it over the polymer. The polymer is
continuously weighed by the electrobalance.

This method can be used to accurately measure flavor

sorption by plastic materials which have been surrounded with

19

a constant concentration of organic vapor. .A limitation is
that it can analyze only one component or set of components
at a time. The minimum weight to detected by the

electrobalance is 1/10 pg.

Materials and methods

1. Sorbent Blended with Polyethylene

Table 2 shows samples which were used to determine the
density of them. The procedure to make these samples are

described in appendix A.

Table 2. Samples used to determine density

 

Type of resin concentration of sorbent
(wt is)
polyethylene 0
activated carbon
blended polyethylene 0.135
1
5
Tenax blended polyethylene 1.6
2.8
3.9

 

Table 3 shows samples which were used to determine the
mechanical properties and sorption characteristics of them.

The procedure to make these samples are described in appendix

B.

20

21

Table 3. Samples used to determine mechanical
properties and sorption characteristics

 

Type of film concentration of sorbent
(wt %)
polyethylene 0
activated carbon
blended polyethylene 0.135
1
5
Tenax blended polyethylene 1 .6
2.8
3.9

 

\/ 2. Determination of Density

Several polyethylene samples were blended with the
indicated sorbents. The blended concentration was different
for every sample. Therefore the density of these samples
varied with type of blended sorbent and its concentration.

To understand the relationship between density and sorbent,
the density of each sample was measured.

The procedure was based on ASTM D 1505-85.

A density gradient column (Cole-Parmer Instrument Co. ,

Chicago, IL) was prepared using isopropanol and water. The
density range of this system is from 0.79 to 1.00 g/cm3.

Distilled water and 2-propanol (100 %, J.T.Baker Inc. ,
Phillipsburg, NJ) was used.

22
Three standard glass beads (0.9282, 0.9342, and 0.9516

g/cm3) were used to make the calibration curve for density of

samples.
.After the column was filled with the gradient mixture,

the three standard glass beads were put into the column.
After their sinking stabilized, the height of each of the
standard floats from the bottom of the column was measured by
a ruler. From these data, density versus height was
plotted. The plot was used as a calibration curve to
determine the density of samples. The calibration curve was
made separately for Tenax and activated carbon.

All the samples were put into the same column as the
standard glass beads, and their height from the bottom of the
column were measured. Using the calibration curve, the

density of each sample was determined with an approximation
of 10.001 g/cm3 for activated carbon and $0.011 g/cm3 for

Tenax, respectively.

3. Determination of. Mechanical Properties

Several polyethylene samples were blended with the
indicated sorbents. The blended concentration was different
for every sample. Therefore the mechanical properties of
these samples varied with type of blended sorbent and its
concentration. To understand the relationship between

mechanical properties and sorbent, tensile strength, %

23
elongation of at break, and elastic modulus of each sample
was measured.

The procedure was based on ASTM D 882.

Sample films were cut by a JDC Precision sample cutter
(model JDC25, Thwing-Albert Instrument Co., Philadelphia, PA)
to get proper size for testing. The thickness of all
specimens was measured by a micrometer (model 549, Testing
Machines Inc., Amityville, NY) before the test. The
specimen was placed in the grip of the Instron machine (model
4201, Instron Corporation, Liviona, MI), and the machine was
started.

The testing conditions were as follows:

Width of specimen: 1 inch
Grip separation: 2 inches
Crosshead speed: 20 in/min.
The result was charted, and stress, strain, and modulus

of elasticity were calculated for each specimen.

4. Optical Microscope Observation of Sample Films

It is extremely difficult to observe the distribution of
sorbent in a film by the naked eye. In particular,
observation of the cross sectional distribution of sorbent in
a film is almost impossible. To know the distribution
condition of sorbent is important to understand the sorption
characteristics of the sorbent blended sample films. Such

observations may help to explain the interactions between

24

polyethylene and sorbent. Optical observation can provide
magnified images of film.samples, so it would be helpful in
understanding the interactions between polyethylene and
sorbent.

Each sample film was cut about 1 cm.x 3 cm. To cut
samples, a razor blade was used to obtain sharp edges. A
small piece of sample film was put on a slide glass and set
on the optical microscope (model BHS, Olympus Optical Co.,
Ltd., Tokyo, Japan). First, the sample film was focused at
a lOW'magnification (x100), then the magnification was
changed to the desired setting. .An automatic
photomicrographic system (model PM-lOADS, Olympus Optical
Co., Ltd., Tokyo, Japan) was attached to the microscope and
magnified images of the sample film were taken at the desired
magnification. The exposure time was automatically
determined by this system.

This optical microscope system is available at the
Composite Materials and Structures Center, Michigan State

University.

5. Determination of Sorption Characteristics by

Gravimetric Method

Theoretically, given sorbent sorbs sorbate up to its
capacity. Therefore, it can be expected that the more
sorbate sorbed into a sample film, the greater the weight of

the tested sample. The gravimetric method is based on this

25

expectation. The degree of sorption of each sorbate was
determined by the weight change of the sample. This system
requires a very sensitive balance to detect the weight change
of samples

The schematic diagram of the electrobalance system.(Cahn
Instruments Electrobalance model 2000, Cahn Instruments,
Cerritos, CA) is shown as Figure 2.

(R)-(+)-limonene, toluene, and ethyl acetate were
studied as sorbates in this study.

Nitrogen gas was used as the carrier gas for the
Electrobalance system.

The degree of absorption of each sorbate was determined
by the weight change of the sample. To measure the
absorption of the samples, the following techniques were

applied:

1. The system was purged by nitrogen gas overnight to
eliminate sorbate residues. In this phase, valve No.7 (V7)
‘was closed, three-way valve No.9 (V9) was turned to waste,

and valve No.1 (V1) and valve No.8 (V8) were opened.

2. For a film.sample, the film was cut about 1 cm.x 2 cm
and weighed precisely by a.Mettler.AE-160 balance (Mettler
Instrument, Hightstown, NJ). .After weighing of sample, it
was hung in the hangdown tube of the electrobalance and left
until the balance stabilized.

For a granular sample, the sample was put on a small

aluminum pan and weighed using the same method used for a

26

oases

Emumhn cocoaonoupomao mo Ecumowo owuoaocom .N enemas

vuom ucaHmEum

h>

 

 

 

 

 

 

m>
32.: AIIIVATIII

v>

 

 

 

 

VAT

 

), o>

 

 

 

avast AI

<I
ab

oocdaonouvooan
,IIIIIIJ

 

 

 

 

 

«z

 

 

 

 

 

27

film.sample. Then the aluminum.pan was hung in the hangdown
tube of the electrobalance with a stirrup and left until the

balance stabilized.

3. During the stabilizing of the balance, valves No.2 (V2) ,
No.3 (V3) and No.4 (V4) were opened, and valve No.5 (V5) was
closed to adjust the concentration of sorbate. 500 pl of
sample used to determine the concentration of sorbate was
obtained from the sampling port using a Gastight® #1750
syringe (Hamilton Co. , NV), and directly injected into the
gas chromatograph (Hewlett Packard HP5890A with flame
ionization detector, Hewlett-Packard, Avondale, PA). The
analyzing conditions of the gas chromatograph were set as
shown in appendix C.

The concentration of sorbate was determined by the

responded area units and calibration curve of each sorbate.

4. After stabilization of the sample, V6 and V8 were

closed, V7 was opened, and V9 was turned to electrobalance to
allow the carrier gas with sorbate to pass into the hangdown
tube of the Electrobalance. The weight change of the sample

was recorded by a recorder.

5. When the weight change reached equilibrium, the recorder
was stopped; V7 was closed, V9 was turned to waste, V6 and V8

were opened to purge the inside of the hangdown tube.

Results and Discussion

1. Density of Samples

The calibration curves of density for activated carbon
and Tenax are presented as Figure 3 and 4, respectively.
The density of both samples was calculated using these
calibration curves, and are shown in Table 4 and 5. From
both Table 4 and 5, the concentration of sorbent versus
density for each sorbent were plotted in Figures 5 and 6,
respectively.

In Figures 5 and 6, both activated carbon and Tenax show
a linear relationship between concentration of sorbent and
density. Comparing these two results, Tenax showed a larger
decrease in density than activated carbon; even the blended
amount of Tenax is lower than that of activated carbon.
This is caused by the density difference of these two
sorbents. Table 1 shows that Tenax is lighter than

activated carbon.because of their relative densities
(activated carbon: 0.40-0.51 g/cm3; Tenax: 0.37 g/cm3).

Therefore, it required less Tenax than activated carbon to

affect a given density of blended polymer.

28

Density (g/cn?)

29

0.955

 

0.9504

0.945“

0.940-

0.935-

0.930-

 

 

 

0.925 I I I
20 25 30 35 40

Distance from bottom of column (cm)

Y = -1.248x10'3x + 9.779x10'1
r2 = 0.989

Figure 3. Calibration curve for density
of activated carbon blended
samples

30

0.955

 

0.950‘
0.945q
0.940“

0.935-

Density (glen?)

0.9304

 

 

0.925 r l T T7 I
15 20 25 30 35 40 45

 

Distance from bottom of column (cm)

Y = -1.05x10'3x + 9.721(10-1
r2 = 0.989

Figure 4. Calibration curve for density
of Tenax samples

31

Table 4. Density of activated carbon blended samples

 

Concentration of Density*
activated carbon (%) (g/cm3)
0 0.9230
0.135 0.9210
1 0.9210
5 0.9174

 

* The values were averages of replicated measurements.
Approximation: 10.001 g/cm3

Table 5. Density of Tenax blended samples

 

Concentration of Density*
Tenax (%) (g/cm3)
0 0.9266
1.6 0.9242
2.8 0.9161
3.9 0.9008

 

* .
The values were averages of replicated measurements.
Approximation: 10.011 g/cm3

32

0.925

 

0.920-

Density (g/cm’)
0

 

 

0.915 I I I

Concentration of activated
carbon (wt%)

Y = -9.36x10'4x + 9.22::10-1
r2 = 0.892

Figure 5. Relationship between density
and concentration of
activated carbon

 

33

0.930 7

0.925-

0.920-

(9/0"?)

0.915-

0.910-

Density

0.905“

 

0.900 1

l I

0 1 2

Concentration of Tenax (wt%)

y = -6.39x10'3x + 9.30::10'1
r2 = 0.840

Figure 6. Relationship between density
and concentration of Tenax

 

 

34

2. Mechanical Properties

Results of thickness, tensile strength, % elongation at
break, and elastic modulus of each sample film are presented
on Table 6. The values are averages of replicated
measurements.

From.those results, the elastic modulus versus the
concentration of sorbents for both sample films is plotted in
Figures 7 and 8, respectively. In addition, tensile
strength and % elongation at break versus concentration of
sorbents is presented in Figure 9 and 10, respectively.

It was observed that the greater the concentration of
blended sorbents, the worse the mechanical properties of both
materials. However, the degree of decrease in those
properties varied.

It appears that a linear relationship between
concentration of Tenax and elastic modulus is observed over
the sorbent concentration levels investigated (Figure 8).

On the other hand, a logarithmic relationship is shown for
activated carbon over the blending concentration levels
studied (Figure 7). The activated carbon drastically
decreases the flexibility of polyethylene film, even at a
loading level of 0.135 wt%. The degree of decrease
approached equilibrium when 5 wt% of activated carbon was
added. By contrast, Tenax reduced flexibility relatively

slowly, but lost flexibility did not reach equilibrium even

35

after 3.9 wt% of Tenax was blended. This phenomena may be
related to differences in the size of the sorbent particles.
From.the optical microscope observation (Figures 11 and 12),
the particle size of activated carbon is smaller than that of
Tenax (activated carbon: 2.2 mm; Tenax: 3.1 mm). Therefore,
Tenax blended polyethylene could lose its flexibility easier
than activated carbon blended samples. This tendency can be
seen graphically in Figures 9 and 10. As shown in Figure
10, the slope of losing elongation was sharper for Tenax than
for activated carbon. The same effect was observed with

decreasing elastic modulus.

Table 6. Mechanical properties of sample films

 

thickness tensile % elongation elastic
(mil) strength at break modulus
(PSI) (M (981)
LDPE 3.64 1795.23 123.18 27738.03
LDPE* 1200~4550 100-650 25000~41000

Carbon blended LDPE

0.135 % 4.38 1655.53 123.15 19821.55
1 % 4.17 1631.05 122.50 18211.68
5 % 4.21 1558.40 121.33 14646.70

Tenax blended LDPE

1.6 % 3.59 1758.43 120.30 23238.08
2.8 % 3.60 1692.38 119.75 17977.65
3.9 % 3.71 1543.82 115.40 17508.47

 

*Data from Modern Plastic Encyclopedia (1990)

36

 

 

 

 

30000
II
'3.“
e 25000"
In
S
H I
'3 20000‘
8
O D
-H
.p
m
'3 15000'I 0
II:
10000 I I I I I
0 1 3 5 6

Concentration of activated carbon (%)

Figure 7. Relationship between concentration
of activated carbon and elastic
modulus

37

 

30000

27500

25000-

22500-

20000-

Elastic modulus (psi)

 

 

 

17500

 

Concentration of Tenax (%)

Figure 8. Relationship between concentration
of Tenax and elastic modulus

Tensile strength (psi)

38

 

1800i—

1750‘

1700‘

1650-D

1600-

1550-

CI Activated carbon

0 Tenax

 

 

1500

 

-

p-a/

Concentration of sorbents (%)

Figure 9. Relationship between concentration

of sorbent and tensile strength

% elongation at break (%)

 

39

 

CI Activated carbon

0 Tenax

 

 

 

Concentration of sorbents (%)

Figure 10. Relationship between concentration
of sorbent and % elongation at
break

40

3. Optical Microscope Observation of Sample Films

The results of optical observation are shown as Figures
11 ~ 20.

Figures 11 and 12 show a particle of activated carbon
and Tenax, respectively. For these two figures, the
magnification of microscope was x200. In these figures, it
can be seen the different diameter and surface structure
between the two sorbents.

Figure 13 shows the distribution of activated carbon
particles in a 5 wt% concentration of sample film. The
distribution and concentration of activated carbon appear to
be random.

Figures 14 exhibits the distribution of Tenax particles
in a 3.9 wt% concentration of sample film. .As with
activated carbon film.sample, it suggests random.distribution
of Tenax. In addition, Tenax particle size is bigger than
that of activated carbon. It mirrors the results shown in
Figures 11 and 12.

The cross sectional view of polyethylene film.is shown
in Figure 15, which can be compared to Figures 16 and 17
presenting a cross sectional view of each blended sorbent.
These figures present the condition of both sorbents in the
thickness of the sample films. .As shown in Figure 17, some
particles of sorbent are completely surrounded in

polyethylene.

41

To confirm.these results, cross sectional pictures at
the highest magnification (x1000) were taken as shown in
Figures 18 ~ 20. They also show most of sorbent particles
are surrounded in polyethylene.

There is less destruction of physical structure of Tenax
than expected because the average diameter of Tenax is almost
twice that of the thickness of the sample film. Therefore,
the destruction of the porous structure of sorbent by making
sample film would not affect the sorption characteristic of
the sample film. The surface conditions of the sorbent
which is buried in polyethylene cannot be confirmed from
these pictures. Consequently, it would be difficult to say
whether or not polyethylene surrounding sorbents interrupts
sorption from these observations.

However, it appears that there is a good contact between
the surface of the sorbent and polyethylene. This means no
air gap appears to exist between the discontinuous and matrix
phases. This is important because the uptake of the sorbate
by the sorbent has to be made through the diffusion of the

sorbate though the polyethylene matrix.

42

 

L I
lOOpm

Figure 11. A.particle of activated carbon (x200)

 

100nm

Figure 12. A.particle of Tenax (x200)

43

Figure 13. Surface of activated carbon blended film
(5 wt%, x200)

0‘ V .5sz

fl. "“‘ :"

('(f'v‘ ‘ a: ‘ ‘ \
. ‘ {gs-‘5

“‘ . .1

Figure 14. Surface of Tenax blended film
(3. 9 wt%, x100)

44

 

Figure 15. Cross sectional View of polyethylene
film (x400)

 

Figure 16. Cross sectional view of activated carbon
blended film (5 wt%, x400)

45

 

Figure 17. Cross sectional View of Tenax blended
film (3.9 wt%, x400)

 

 

Figure 18. Cross sectional view of polyethylene
film (x1000)

46

 

Figure 19. Cross sectional View of activated carbon
blended film (5 wt%, x1000)

 

Figure 20. Cross sectional View of Tenax blended
film (3.9 wt%, x1000)

47

4. Sorption Characteristics by Gravimetric Method

For this sorption study, the activity of sorbent was
adjusted to 0.02 and 0.05. Table 7 shows the concentration
in ppm (v/v) of each organic vapor at a = 0.02 and 0.05.

The saturated vapor pressure was calculated based on the data

reported by Perry et a1. (1984).

Table 7. Concentration in ppm (v/v) of
each sorbate for each vapor

 

activity
vapor activity 0.02 0.05
limonene 0.35 0.70
ethyl acetate 9.59 19.19
toluene 2.71 5.41

 

Tables 8 ~ 10 ShOW'the results of sorption tests for
each sorbent. In addition, the sorption characteristic of
limonene was observed for granular, thin film.and thick film
polyethylene. The results are presented as Table 11. In
these tables, CS represents sorbate concentration (wt%) in

the sample film,.A/g stands for surface area of sample per
unit mass (mmZ/g), and S is the sorption coefficient (kg/kg-

Pa). It can be seen from these tables that for polyethylene
the solubility of limonene is almost three order of magnitude

larger than for ethyl acetate and two order larger than

48

ucmHOHuumoo cosumuom «m
Ease mHQEMm ecu ca cowumuuemocoo mumoHom «no

 

 

wcmcoEHH HOH vmmu cowumHOm no measmmm .m canoe

mnoexo.e Hee.o am.~H me~m.o ommo.o e.m
euoexe.m meo.o NH.m moHH.o emmo.c e.m «m.m
NMCOB
mnoexH.H smo.o me.- omme.c eemo.c ~.e
arcane.» eHo.o mm.m smme.o mseo.o ~.e em
muonH.H mmo.o ~H.MH emee.o mmmo.o ~.e
euoexm.e Hmo.o em.e Heoe.o Ammo.o ~.e we
mueHxH.H meo.o Hm.mH eeee.c mmmo.e e.e
euofixo.e mHo.o He.m eoo~.o memo.o e.e mmme.o
GOQHMO UOHMbflHOfl
muesxo.e seo.o Ho.eH Neee.o smmo.o e.m
euoexe.m -o.o me.m mome.o meeo.o e.m memeseemsaoa
mmagficm_saaw
mioexe.e mmo.o se.ee mmmo.o ~moo.o
muoex~.m -o.o oe.m. omso.o omoo.o seams
Nucexo.~ emo.c ee.mH~ emmm.H Heoo.o
Nicene.e mHe.o oe.HH~ smmm.fl meoo.e eonueo omee>auoa
eucaxm.e mmo.o ~H.¢ memo.o msoo.o
euoexe.m eeo.e ee.m eHMH.e emme.o memeseumsfioe
mmameom Hmasccum
Aaeuox\mxv sua>auoe .m\me. Amsvpeoems .ovueaa03_ lease
m Homm> mu ommcmuocH HchHcH mmmchdne mmamfimm no seas

49

ucmHoHuuooo :oaumuom «m

Edam mamemm on» ca cowumuucmocoo mucouom «no

 

 

sIon~.m mec.o mH.o omoo.o mmmo.o n.m
hioflxn.e mmc.o HH.c omoo.o mmmo.o >.m wm.m
mecca
sloaxm.e meo.o -.o omoo.o mmmo.c ~.e
bloaxm.m emo.o no.9 mmoc.o mmmc.o ~.¢ wm
hIonH.¢ omo.o -.o whoo.o memc.o ~.e
Floaxw.m mmo.o oa.o emoo.o ommc.o ~.¢ wH
snofixm.m meo.o mo.o mooo.o mmoo.o e.e
nIonH.m emo.o mo.o ~moo.o ~mno.o e.e amma.o
:onumo omuo>auom
sioaxm.m meo.o No.0 mooo.o meeo.o m.m
hioaxm.~ Hmo.o mo.c omoo.o emmo.o e.m mcwahcumaaom
moamecm Edam
mioaxo.e neo.o mm.~ ooHo.o meoo.o
muofixm.h -o.o mm.H omoo.o meoo.o seems
mIon~.e meo.o mo.- thH.c mmoo.o
muoaxs.n emo.o eH.om mmHH.o mmoo.o cooumo odom>auom
mIonm.m omo.o eo.o eooo.o amoo.o
hioaxm.fi -o.o eo.o eooo.o Hmoo.o ocmaacumaaom
mmHQEMm umaocmum
Ammlmx\oxv >ua>fluom .m\oav Amsvucmwm3 .mvucmflm3 Aaflsv
m Homm> mo ommmmuocH HospfloH mmmchflca mwamecm so some

mpmuoom amend Ho“ and» coauQHOm no mpaommm .m manna

ucmaowuumoo coaumuom um

Ease wHQEMm on» as coflumuucmocoo mucouom «mo

 

50

 

moosaou HOH ummp coaeQHOm no measmmm .oH magma

oioflxm.m ~mo.o eo.H mneo.o ommo.o m
miofixa.m amo.o mm.o oomo.o eHmo.o m wm.m
xmcwa
enoaxm.m ~eo.o mm.H ommo.o memo.o
enoaxm.s mmo.o ~m.o mefio.o emmo.o mm
eIonH.m meo.o oe.H omeo.o mmmo.o
wuoflxm.s mmo.o om.o omao.o mmac.o we
mnoflxm.m meo.o mm.H «Heo.o oomo.o
mioaxm.> omo.o em.o memo.o ~mmo.o wmmH.o
conumo oopm>fluom
ouoaxh.m meo.o mm.H emmo.o mmmo.o
onofixm.s mmo.o om.o mmao.o mHNo.o mcoaanuwaaom
moameom Edam
mioaxe.~ evo.o hw.m ommo.o Hmoo.o
mIoHNo.m NNo.o mH.~ mwoo.o Heoo.o NMGOB
euoflxm.a eeo.o me.m~ omm~.o moHo.o
euoaxm.m Hmo.o ms.- mmmo.o ~eoo.c :onumo ompm>fluom
mloax~.> meo.o vo.H omoo.o whoo.o
suede... Hmo.o me.o mooo.e aeoo.o mamaseeosaoa
mdeEMm Hoascmum
Aomnox\oxv >ua>fluom Am\mE. .msvpnmflmz onucoflws
m Homm> mo oommmuocH HmeHcH mmmcx0flca endgame no mm>a

1
5

ucmflOfluumoo :oapmuom “m
mums use: Hem mamecm no menu mocunom «axe

 

 

mIonH.H e.onm No.0 wmme.o mmmo.o m.m
mIono.H m.mmmnfi mo.o ~m~¢.o smmo.o e.m
eueexe.m o.eemee ~e.e mome.o meeo.o e.m memeseeesaoa
mmamecm EHHH
eIonm.s e.mmee mo.o mmmo.o msco.o
euofixn.m s.~mm> mo.o mHmH.o mmmo.o momamcuoaaom
mmHQEMm unannoum
Acmlmx\oxv Am\meev >9a>apom Amevpcmamz Amvvcmwmz AHdE.
m m\4 Homm> ommmmuocH HMHuHcH mmwcxoaca meQEMm no wows

mochEHH no“ mmHmEMm mcoamcumaaom no momma Hmum>wm mo pump cowumHOm mo muaommm .HH manna

52

toluene. .A similar relation exists for Tenax and activated
carbon. Limonene, ethyl acetate and toluene sorbed better
by activated carbon alone than by Tenax alone.

A maximum calculated sorption.capacity (Smax) is

described by Equation 9:

Smax = MPESPE + M553 (9)
where MPE = weight of blended polyethylene

SPE = sorption coefficient of polyethylene

3
to
II

weight of blended sorbent

sorption coefficient of sorbent

U)
I!)
II

Tables 12, 13, and 14 show'maximum.calculated sorption
capacity for limonene, ethyl acetate, and toluene,
respectively.

Table 12. Maximum.calculated sorption capacity
for limonene

 

vapor activity 0.02 0.05
polyethylene 5.7x10'4 7.3x10’4
activated carbon blended

0.135% 6.3x10-4 7.6x10-4

1% 1.0x10-3 9.2x10-4
5% 2.9x10-3 1.7x10'3

Tenax blended
3.9% 6.7x10-4 7.7x10-4

 

53

Table 13. Maximum calculated sorption capacity
for ethyl acetate

 

vapor activity 0.02 0.05
polyethylene 1.9x10-7 8.3x10‘8
activated carbon blended

0.135% 2.9::10-7 1.4x10'7

1% 9.6x10'7 5.01110-7
5% 4.0x10'6 2.2x10‘6

Tenax.blended
3.9% 4.9x10'7 2.6x10'7

 

Table 14. Maximum.calculated sorption capacity
for toluene

 

vapor activity 0.02 0.05
polyethylene 1.7::10-6 7.2::10-6
activated carbon blended

0.135% 2.1::10-6 7.4x10‘6

1% 5.0::10"6 8.7x10"6
5% 1.8x10'5 1.5x10'5

Tenax.blended
3.9% 2.8x10‘6 7.9x10'6

 

54

The values of CS versus concentration of blended
activated carbon in film.samples (wt%) and Smax are plotted
in Figures 21, 22 and 23. Figures 24 and 25 show
relationship between the sorption coefficient of limonene in
polyethylene samples having different surface area per unit
mass, and the values of the area per unit of mass for each
sample, respectively.

In Figure 21, it appears to reach the equilibrium at 5
‘wt% of concentration of activated carbon at lower vapor
activity. On the other hand, there is very little
difference of sorption coefficient among 3 blended
concentrations at higher vapor activity. This result
suggests that limonene may saturate the 0.135 wt% of
activated carbon. This shows that activated carbon is
effective at low concentration of limonene by a factor of
almost 2. To compare experimental data and Smax data of
0.135 wt% blended concentration, experimental data shows
higher values than Smax° Most of other data are below the
estimated Smax values. These results would suggest all
activated carbon did not sorb limonene.

When the sorbate is ethyl acetate, the graphs for both
the lower and higher vapor activity levels showed the
equilibrium.at 5 wt% of blended activated carbon, as shown in
Figure 22. It means both lower and higher concentration of
sorbate would be sorbed by 5 wt% of blended activated carbon.

To compare experimental data and Smaxr almost same tendency
as limonene is observed. These results would suggest all

activated carbon did not sorb ethyl acetate.

55

 

30

25‘

20‘

 

 

Sorption coefficient
(x10’4 kg/kg-Pa)

 

 

 

Concentration of activated
carbon (wt%)

Figure 21. Relationship
of activated
and sorption

a=0.02
a=0.05
Smax at a = 0.02
Smax at a = 0.05

between concentration
carbon blended in a film
coefficient of limonene

56

 

 

 

 

 

 

50
p -
a 40 .
g A
' (U
U
a ‘I‘ ..
u. 3‘ 30-
Q) \ .'
O ox
0 x
G5 .' ,’
o —l '0’ ”
"'1 lo 20 '0’ ”
4., H a ’1
Q: N ,’
H ~v ,’
o '- ,I
m 10‘ ox ,”
fiFF;=flE:: <iJ
0‘f’ I’ I I I I
5

Concentration of activated
carbon (wt%)

E]

<>

a=0.02

a=0.05

0.02

0.05

Figure 22. Relationship between concentration
of activated carbon blended in a film
and sorption coefficient of ethyl acetate

57

 

20

...:
U1
1
\

H
0
ml
J
.1‘
. \
\
. \

Sorption coefficient
(x10'6 kg/kg-Pa)
U1
1
\
\

 

 

 

El a=0.02

0 a=0.05
.......... smax at a = 0.02
..... Smax at a = 0.05

Figure 23. Relationship between concentration
of activated carbon blended in a film
and sorption coefficient of toluene

Sorption coefficient
(x10‘4 kg/kg-Pa)

58

 

 

 

 

12
11- E]
10- BK
El a=0.02
9—i
HE a=0.05
8..
3K
7—
6—.
E] E]
5 j l I
granular thin thick

Type of polyethylene sample

Figure 24. Relationship between sorption
coefficient of limonene and
type of polyethylene sample

59

 

 

 

 

20000
El
3 36
m
E
:3 15000-
a
g C] 0 02
A a: O
8.13“ 1
3K a=0.05
8% 10000 D
s . n
H
8 2‘
..I
a g 5000 X
H
a 4i
:0 o
0 I I I
granular thin thick

Type of polyethylene sample

Figure 25. Relationship between surface area
per unit mass and type of polyethylene
sample

60

For toluene, Figure 23 shows very little sorption
coefficient difference among 3 concentrations of blended
activated carbon. It means 0.135 wt% of activated carbon
can sorb such low concentration of toluene. .As contrasted
with this result, the higher concentration of toluene
(a=0.05) may need.more than 5 wt% of activated carbon to
sorb. To compare experimental data and Smax data, almost
same tendency as other sorbates is observed: for 0.135 wt%
blended concentration, experimental data shows higher values
than Smax' Most of other data are below the estimated Smax
values. These results would suggest all activated carbon
did not sorb toluene.

It seems that all these results show the curvilinear
sorption behavior in these blended concentration of this
sorbate. The sorption of polyethylene would also happen on
these samples, however, activated carbon may sorb more
amounts of sorbate than polyethylene. But such an analysis
is inconclusive given the small data set.

Tenax blended film sorbed very little limonene.
However, it sorbed more ethyl acetate and toluene than the
activated carbon blended.filuh On the other hand, activated
carbon blended film sorbed greater sorbate amounts overall
than polyethylene film. So, it can be said activated carbon
has more capacity but no selective sorption characteristics
for these three organic substances.

Tenax alone sorbed two times or more limonene than pure
polyethylene. This may suggest some hindrance phenomena or

the inacceptibity of Tenax in polyethylene to reach limonene.

61

From.the optical observations, some particles of Tenax are
completely surrounded by polyethylene, that may support the
latter presumption. .As shown in Table 1, Tenax has
relatively large pores on its surface. Therefore,
polyethylene would be packed more easily than activated
carbon, and interrupt sorption of large molecules such as
limonene.

From Equation 1, the number of volatile molecules
striking the surface every second, n, was calculated for each
organic compound and vapor activity at test temperature.
They are shown in Table 12.

Table 15 shows limonene is able to be sorbed about half
the number of molecules as the other two sorbates. This is

due to its large molecular weight.

Table 15. The number of molecules of organic
compound striking the surface of

sorbents per second (molecules/cmZ/sec)

 

vapor activity 0.02 0.05

limonene 4.74x1017 1.19x1o18
ethyl acetate 9.11x1017 2.28x1018
toluene 8.52x1017 2.13x1018

 

The uptake by the sorbent is so low or below expected
for both two sorbents. Two reasons may be considered to
explain this question. First, the sorption process needs

very long time to reach equilibrium. Therefore, the blended

62

sample films haven't reached the equilibrium. Secondly,
there may be some problem at the interface between sorbents
and polyethylene. If there is any interaction between them,
it may interrupt the sorption behavior. These have not
satisfactory answered.yet.

From the results of comparison of sorption coefficient
of several different polyethylene samples (Table 11, and
Figures 24 and 25), two thin polyethylene films show
completely different sorption coefficients even though their
values of surface area per unit mass are very close. This
appears to be dependent upon the concentration of sorbate,
but such an analysis is inconclusive given the small data

set .

Possible sources of experimental errors are listed as

follows:

1. Sensitivity of electrobalance.

Since the electrobalance is very sensitive, it must be
placed where external vibrations are minimized. External
vibrations may cause fluctuation of the equipment

sensitivity.

2. Gas flow regulator.

Due to the sensitivity of the electrobalance and the
tubing system, vapor concentration cannot be easily checked
during the course of the experiment. Sometimes the gas flOW'

regulator affected the results with noise generation. In

63

addition, a small change of vapor concentration can result in
a large measured.weight increase for the sample film.
Consequently, the flow regulators used must be accurate and
precise to generate an identical flow rate during the test
and between subsequent runs. Also, the nitrogen gas used to
generate proper vapor concentration had to be checked to
maintain constant pressure throughout the experiment. No

action was taken during this study to minimize these effects.

3. Distribution of sorbents.

When sample films were made by a laboratory press, both
activated carbon and Tenax tended to spread to the edge of
the film with the flow of melted polyethylene. Therefore,
there were some parts of the sample film which had different
sorbent concentrations. For this sorption study, the test
films were visually checked to ensure sorbent concentrations

were as equal as possible.

4. Temperature of testing environment.

Sorption is temperature dependent. Even though the
laboratory is air conditioned, room.temperature changes
easily due to weather, number of people in the room, and so
on. Hence, the electrobalance system should be in a well-

controlled, temperature-stabilized.chamber.

5. Static electricity in the hangdown tube.
If static charge develops inside the hangdown tube, it

attracts the wire of the electrobalance. If the wire

64

touches the wall of hangdown tube, the detected weight will
differ greatly from actual weight. Therefore, it is
important to minimize static electric buildup in the

electrobalance system.

Summary

(1) Tenax showed a larger decrease in density than
activated carbon when both were blended with polyethylene.
Even the weight of blended Tenax film.samples is lower than
that of blended activated carbon. This is caused by the

density difference of these sorbents.

(2) It was observed that the greater the concentration
of blended sorbents, the poorer the mechanical properties of
both materials. However, a linear relationship between
concentration of Tenax and elastic modulus is observed. On
the other hand, a logarithmic relationship is shown for
activated carbon. This phenomena may be caused by the
difference of size of sorbent particles. The particle size
of activated carbon is much smaller than that of Tenax.
Therefore, Tenax blended polyethylene could lose its

flexibility easier than activated carbon blended samples.

(3) It is observed by optical observation that some
particles of sorbent are completely surrounded by
polyethylene. There is less destruction of physical
structure of Tenax than expected, even though the thickness
of the sample film is almost half of the average diameter of

Tenax .

65

66

(4) Tenax blended film.sorbed very little limonene.
However, it showed more sorption than activated carbon
blended film.for ethyl acetate (1.2 times) and toluene (1.1
times). On the other hand, activated carbon blended film
sorbed greater amounts of all other sorbates than
polyethylene (limonene: 1.1~1.4 times; ethyl acetate: 1.0~1.7
times; toluene: 1.0~1.1 times). Tenax can sorb limonene
‘when it is granular. This selective sorption characteristic
may be caused by polyethylene which can be packed into the
porous structure of Tenax. This is not confirmed on this

study.

(5) From the result of comparison of sorption
coefficients of several types of polyethylene samples, two
thin polyethylene films show completely different sorption
characteristics, even though their values of surface area per
unit mass are very close. This appears to be concentration
dependent, but that is difficult to conclude from.these data.
A.more extensive database would be required to resolve that

issue.

(6) The uptake by the sorbent is so low or below
expected for both two sorbents. Two reasons may be
considered to explain this question. First, the sorption
process needs very long time to reach equilibrium.
Therefore, the blended sample films haven't reached the
equilibrium. Secondly, there may be some problem.at the

interface between sorbents and polyethylene. If there is

67

any interaction between them, it may interrupt the sorption

behavior. These have not satisfactory answered yet.

Future Studies

(1) To confirm the sorption characteristics of sorbent
blended samples, electron microscopic observation of the
specific case of sorbent which is completely enveloped in
polyethylene would be helpful. This method would provide a
visual representation of the surface of the sorbent, so the
effect of polyethylene on the sorbent's porous structure

would be clarified.

(2) To give better sorption.characteristics, the
blending of multiple sorbents into polyethylene could be
considered. Multiple sorbents may provide more uniform.
sorbent coverage of organic volatiles, by closing the gaps in

coverage left by individual sorbents.

68

Appendix A

Blending Sorbents into Polyethylene

mils—A.

Blending Sorbents into Polyethylene
Materials:

Low density polyethylene

(LDPE, Dow Chemical U.S.A., Freeport, TX)
Tenax-TA®

(80/100 mesh, Alltech Associates, Inc., Deerfield, IL)
Activated carbon

(Norit® A, 100 mesh, Fisher Scientific, Fair Lawn, NJ)
Baker-Perkins MP2030 Compounder

(Baker-Perkins , Saginaw, MI)
Extruder model KLB-100

(Killion Extruders Inc., Cedar Grove, NJ)
Concentrations of 1.6, 2.8 and 3.9 % (w/w) Tenax, 0.135,
1 and 5 % (w/w) activated carbon in LDPE, respectively, were
made for the sorption tests.

1. Determination of Sampling Time

The determination of the residence time distribution in

69

70

an extruder was important to control the concentration of
blended sorbents because the concentration of sorbents in
LDPE increased gradually after putting them.into an extruder.
Therefore, to determine the sampling time of blended.LDPE the
residence time distribution was measured.

Butler (1990) reported a study of the residence time
distribution for processing because most polymers are heat
sensitive. Butler's procedure was applied for the

determination of sampling time.

1. Turn on the extruder (Baker-Perkins, Saginaw, MI) to
preheat. The temperature and processing conditions of the
extruder'were:
Temperatures:
section 1: 150 °C
section 2: 150 °C
section 3: 150 °C
die: 150 °C
Screw speed: 70 rpm
Automatic LDPE feed rate: 4 %

2. The extruder was purged about 30 minutes using LDPE

which was same grade as test samples.

3. Set the screw speed and confirulthe rate of extrusion.
To confirm.the rate of extrusion, the weight of extruded LDPE

per minute was measured.

71

4. After enough purging and stabilized extruding, 0.02
grams of activated carbon were added from the hopper every
two minutes for a total of eight times during the extrusion
of LDPE. This activated carbon worked as a pigment and was

used as a controlling variable.

5. The extruded LDPE were sampled every one minute, and
their colors were determined by comparing to a color gradient

chart. The color gradient chart is shown as Figure 43.

6. The color gradient vs. extruding time was plotted to
determine time when the stable concentration of sample was
obtained. This plot is shown as Figure 44. In Figure 44,

the times carbon were added are represented as symbol V7.

From Figure 44, a sampling time from the extruder of 17

to 26 minutes was selected after the sorbent was admitted.

2. Blending Tenax and LDPE

Tenax blended LDPE pellets were made using a Baker-

Perkins MP2030 Compounder. The procedure was as follows:

1. Turn on the extruder to preheat. The temperature and
processing conditions of the extruder were:
Temperatures:

section 1: 150 °C

72

 

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Figure 26. Color gradient chart

73

ucwflocum HoHoo
one was“ mcfloouuxm :wm3umn mflcmcofluoamm .hm whomflm

Amocoommv mafia

coma coca com o
p _ C

DDDDD >%

IoH

 

 

ImH

10103

74

section 2: 150 °C
section 3: 150 °C
die: 150 °C
Screw speed: 70 rpm
Automatic LDPE feed rate: 4 %

2. The extruder was purged about 30 minutes using LDPE

which was of the same grade as the test samples.

3. Set the screw speed and confirm the rate of extrusion.
To confirm the rate of extrusion, the weight of extruded LDPE

per minute was measured.

4. After sufficient purging and stabilized extruding,
certain amounts of Tenax were added from the hopper while
extruding LDPE. The amount of Tenax was determined by the
rate of extrusion.

For example, when the rate of extrusion was 6 grams per

minute, Tenax was added 0.3 grams per minutes for 5 % (w/W).

5. Using a pair of nippers, the blended polymer was

pelletized. The pelletized blended resins were stored far

from any organic substances to avoid contamination.

3. Making Activated Carbon Blended LDPE

Activated carbon blended LDPE pellets were made using an

75

extruder (model KLB-lOO, Killion Extruders, Inc., Cedar

Grove, NJ). The procedure was as follows:

1. Turn on the extruder to preheat. The temperature and
processing conditions of the extruder were:
Temperatures:
section 1: 270 °F (132 °C)
section 2: 350 °F (177 °C)
section 3: 365 °F (185 °C)
die: 320 °F (160 °C)
Screw speed: 10 rpm

2. The extruder was purged about 30 minutes using LDPE

which was the same grade as the test samples.

3. Set the screw speed and confirm.the rate of extrusion.

To confirm.the rate of extrusion, the weight of extruded LDPE

per minute was measured.

4. After sufficient purging and stabilization of extrusion,

LDPE resins mixed with certain amounts of activated carbon

were added from.the hopper. The amounts of activated carbon

were determined by weight basis.

For example, 0.5 grams of activated carbon were mixed to

9.5 grams of LDPE to make 5 % (w/W) concentration.
5. Using a pair of nippers, the blended polymer was
pelletized. The pelletized blended resins were stored far

from.any organic substances to avoid contamination.

end' B

Procedure for Making Film Samples

 

Appendix B

Procedure for Making Film Samples
Materials:

Low density polyethylene (LDPE, DuPont Chemical Co.)
Tenax-TA® blended LDPE pellets

Activated carbon blended LDPE pellets
Carver Laboratory Press (Model M, Fred 5. Carver Inc.)
PET sheets

Procedure:

All concentrations of Tenax and carbon blended LDPE

pellets were processed to film for sorption tests.

1. The laboratory press was preheated prior to resin set

up. The temperature condition for both platens was 150 W2.

2. When 150 °C was reached, the sample resins were set on
the bottom.platen and covered with two sheets of PET as shown

as Figure 45.

3. 30000 psi of pressure was applied 10 minutes to make a

76

77

 

 

 

 

 

«Upper Platen

 

 

 

 

 

I
l-'.- .2-‘.:-'.1-i1-'.1-'.1-'.°.-'.;- .l- 33* 1? PET f :1. 1m
I

 

 

\ ( Lower Platen

 

 

 

\

Polymer resins

 

 

 

 

 

 

(——Oil pressured jack

 

 

Figure 28. Schematic of laboratory press

78

thin layer of blended polymers.

4. After 10 minutes of applied pressure, these platens were
cooled to 70 °C by water. During this cooling period, 30000

psi of pressure was applied.

5. When the temperature reached 70 °C, the pressure was
released from the platens and the sample film was collected.
All sample films were stored far from any organic substances

to avoid contamination .

Anaemia—2

Procedure for the Determination of Absorbate

Calibration Curve by Gas Chromatography

 

Passed—134

Procedure for the Determination of Sorbate

Calibration Curve by Gas Chromatography

Materials:

Four 25 m0 volumetric flasks with stoppers and two 50 ml
volumetric flasks with stoppers were used to prepare standard
solutions.

A 10 pl liquid sampling syringe was used to inject
samples into the gas chromatograph.

A 10 ml pipette and four 1 ml pipettes were used to
adjust the concentrations in the preparation of standard
solutions.

A Hewlett Packard HP5890A gas chromatograph with flame
ionization detector (FID) was used to quantify the sorbates
in the standard solutions (Hewlett-Packard, Avondale, PA).

(R)-(+)-limonene (97%) was obtained from Aldrich
Chemical Company (Milwaukee, WI). Toluene (99.9%) was
obtained from Mallinckrodt Inc. (Paris, KY), and ethyl
acetate (100%) was secured from J. T. Baker Chemical Company
(Phillipsburg, NJ). They were used as sorbates in this
study.

Dichloromethane (DCM, 99.5%) and.o-dichlorobenzene (DCB,
98%) were obtained from EM Science (Gibbstown, NJ). DCM.was

79

80

used as a solvent for limonene and toluene, and DCB was used

as a solvent for ethyl acetate.

Concentrations of 5, 10, 20, 50, and 100 ppm (v/v) of
sorbate in solvent were prepared to make the calibration

01113738 .

Procedure:

The standard curves of the injected sorbate quantities
vs. responded area units for all the sorbates were
constructed using standard solutions of known concentrations.

The standard solutions were prepared by dissolution of known
quantities of both limonene and toluene in dichloromethane
and ethyl acetate in dichlorobenzene. To prepare the

calibration curves, the following steps were applied:

1. The flasks and the syringe were heated overnight in an
oven at 80 °C to remove or reduce any residual organic
compounds that could be adsorbed on the flasks"walls prior
to use. The flasks were then left covered at room

temperature prior to utilization.

2. The purity of the solvents, DCM and DCB, were tested on
the gas chromatograph to verify the existence of interfering
peaks near the solvents' retention times. 1 pl of pure
solvent was injected into each of the solvents, and no

interfering peak was observed on either solvent.

81

3. The standard solutions were prepared by the following

dilution steps:

a)

b)

One 50 ml volumetric flask was partially filled with
solvent using a 10 ml pipette.
0.5 ml of sorbate was added into the 50 ml flask.

c) A stopper was set and the flask was slightly swirled

d)

to mix.
The flask was filled to volumetric line with the

solvent.

e) A stopper was set and the flask's contents were mixed

completely.

The procedure shown above provided the 10000 ppm stock

solution. From this stock solution, the other

concentrations of standard solutions were obtained by

following dilution steps:

a)

b)

C)

One 50 ml volumetric flask was partially filled*with
solvent using a 10 ml pipette.

The flask of stock solution was swirled to ensure
proper mixing.

0.5 ma of stock solution was added into the partially
filled flask.

d).A stopper was set and the flask was slightly swirled

e)

to mix.
The flask was filled to volumetric line with the

solvent.

f) A stopper was set and the flask's contents were mixed

82

completely.

The procedure shown above provided the 100 ppm standard
solution. The other concentrations were obtained by similar

procedures.

4. The analyzing conditions of the gas chromatograph were

set as shown below.

Column:
SUPELCOWAX' 10
Fused Silica Capillary Column
60 m, 0.25 mm I.D., 0.25 pm film thickness

Analysis Conditions:

Initial temperature 75 °C
Initial time 8.0 min.
Rate 4.0 °C/min.
Final temperature 200 °C
Final time 4.0 min.
Injection temperature 200 °C
Detector temperature 250 °C
Helium.gas flow 30 ml/min.

5. A 1 pl sample was injected directly into the gas
chromatograph and the corresponding area units were recorded

for each concentration of sorbates.

83

6. The corresponding area units (A.U.) of the gas
chromatograph versus the quantities of injected sorbates
were plotted and a linear relationship was observed for each
sorbate. The slope of this curve equals the calibration
factor. The data and standard curves were shown in Table 11
~ 13 and Figure 40 ~ 42 for all sorbates, respectively. In
addition, the calibration factors (C.F.) were determined from

the slope of calibration curves.

3.47 x 10'12 g/A.U.

C.F. for limonene

1.19 x 10‘11 g/A.U.

C.F. for ethyl acetate

C.F. for toluene = 2.53 x 10‘12 g/A.U.

84

Table 16. Data for calibration curve of limonene

 

Concentration Volume Area Unit*
(ppm) injected(9)
5 4.20x10"9 942
10 8.40x10‘9 2399
20 1.68x10‘8 3895
50 4.20::10"8 13644
100 8.40x10'8 23299

 

* retention time = 11.6 min.

Table 17. Data for calibration curve of ethyl acetate

 

Concentration Volume Area Unit*
(ppm) injected(g)
5 4.47x10’9 354
10 8.93x10‘9 751
20 1.79::10‘8 1485
50 4.47::10-8 3728
100 8.93x10-3 7472

 

* retention time = 4.8 min.

85

Table 18. Data for calibration curve of toluene

 

Concentration Volume .Area Unit*
(PPm) injected ( 9 I
5 4.34x10'9 2285
10 8.67x10'9 3811
20 1.73::10‘8 8142
50 4.34::10'8 19456
100 8.67x10‘8 34033

 

* retention time = 6.8 min.

86

 

 

 

 

100

3

a. 75—

b

H

31

8

u 50-

8

oh D

G

H

0)

E 25-

PI

0

:>
0"" I I I l
0 0.5 1 1.5 2 2.5

Area Unit (x104)

Volume Injected = 3.47x10'12xA.U.+3.68x10'10
r2 = 0.990

Figure 29. Calibration Curve for Limonene

87

 

 

 

100
3
m 75"
bi
H
1‘.
'U
3 50-
U
0)
'I"
G
H
(I)
E
'3 25‘
O
>
.1. , .

 

j,
0.2 0.4 0.6 0.8

0

Area Unit (x101)

Volume Injected = 1.19x1011xAJU,+3,52x10-11
r2 = 1.000

Figure 30. Calibration Curve for Ethyl Acetate

88

 

 

 

 

100
3
m 75-
bi
H
x
o
3 50—
8
..., [II
c
H
g 25—
.-I
o
>
0",- I I T
0 1 3 4

Area Unit (x104)

Volume Injected = 2.53x1012xA.U.-1.77x10-9
r2 = 0.995

Figure 31. Calibration Curve for Toluene

Bibliography

Bibliography

Baner, A.L., R.J.Hernandez, R.Jayaraman, and J.R.Giacin.
1986. Current technology in flexible packaging. ASTM STP
912:49

Bentley, D.J.Jr., 1988. A.guide to FDA regulation of
adhesives and coatings for flexible packaging. Tappi Journal,
September, p 125

Bigg, D.M. 1992. The newest developments in polymeric
packaging materials. IoPP Technical Journal, Fall, p 24

Billmeyer, F.W.Jr. 1984. Textbook of Polymer Science. John
Wiley & Sons, Inc., New York

Butler, T.I. 1990. The influence of extruder residence time
distribution on polymer degradation. Journal of Plastic film
& Sheeting, 6:247

Cheremisinoff, P.N., and.A.C.Morresi. 1978. Carbon adsorption
applications. Ch. 1. In Carbon adsorption handbook, P.N.
Cheremisinoff, F. Ellerbusch (Eds.), p 1. Ann Arbor Science,
.Ann.Arbor, MI.

Cookson, J.T.Jr. 1978. Adsorption.mechanisms: The chemistry
of organic adsorption on activated carbon. Ch. 7. In Carbon
adsorption handbook, P.N. Cheremisinoff, F. Ellerbusch
(Eds.), p 241. Ann Arbor Science, Ann Arbor, MI.

Daemen, J.M., W.Dankelman, and M.E.Hendriks. 1975. Properties
and applications of Tenax GC as a column packing material in
gas chromatography. Journal of Chromatographic Science, 13:79

Darco, Atlas Chemical Div., ICI America, Inc., 1965. Darco
powdered and granular activated carbon. Wilmington, DE.

89

90

DeLassus, P.T., J.C.Tou, M.A.Babinec, D.C.Rulf, B.K.Karp, and
B-A.Howell. 1988. Transport of apple aromas in polymer films.
Ch. 2. In Food and Packaging Interactions, J.H. Hotchkiss
(Ed.), p 11. ACS Symposium Series 365. American Chemical
Society, Washington, DC.

Dixon, J.B., and S.B.Weed. 1977. Minerals in Soil
Environments, Soil Sci. Soc. Amer., Inc. Madison, WI.

Flory, P.J. 1953. Principles of Polymer Chemistry. Cornell
University Press, Ithaca, NY.

Food and Drug Administration, Department of Health and Human
Services. 1993. Title 21 of the Code of Federal Regulations,
Part 177, section 1010-2000. U.S.Government Printing Office,
washington, DC.

Foster, R.H. 1987. Ethylene vinyl alcohol copolymer resins
for better solvent, aroma and flavor barrier. Future-Pac, 5th
International Conference on Packaging Innovations, Atlanta
GA, Ryder Associates.

Frissel, M.J. 1961. The Adsorption of Some Organic Compounds,
Especially Herbicides, on Clay Minerals,‘Wageningen, Centrum.
voor Landbouwpublikaties en Landbouwdocumentatie.

Beaver, D.L., M.W.Ogden, and P.R.Neison. 1992. Multisorbent
thermal desorption/gas chromatography/mass selective
detection method for the determination of target volatile
organic compounds in indoor air. Environ. Sci. Technol.,
26(9):1737

Heckman, J.H. 1991. Packaging industries and the food-
additives amendment. Ch. 8. In Food and Packaging
Interactions II, J.H. Hotchkiss (Ed.), p 88..ACS Symposium
Series 473. American Chemical Society, Washington, DC.

Hernandez, R.J., J.R.Giacin, and A.L.Baner. 1986. Measuring
the aroma barrier properties of polymeric packaging
materials. Pkg. Technol., 16(4):12

91

Hirose, K., B.R.Harte, J.R.Giacin, and J.Miltz. 1988.
Sorption of d-limonene by sealant films and effect on
mechanical properties. Ch. 3. In Food and Packaging
Interactions, J.H. Hotchkiss (Ed.), p 28. ACS Symposium
Series 365. American Chemical Society, Washington, DC.

Hotchkiss, J.H. 1988..An overview of food and food packaging
interactions. Ch. 1. In Food and Packaging Interactions, J.H.
Hotchkiss (Ed.), p 28. ACS Symposium.Series 365. American
Chemical Society,‘Washington, DC.

Huggins, M.L. 1942. Thermodynamic properties of solutuion of
long-chain compounds. Annals of the New York.Academy of
Science, 43:1

Imai, T. 1988. Absorption of organic volatiles by sealant
films. M.S. thesis, Michigan St. Univ., East Lansing

Konczal, J.B., B.R.Harte, P.Hoojjat, and J.R.Giacin. 1992.
Apple juice flavor compound sorption by sealant films.
Journal of Food Science, 57(4):967

Kovach, J.L. 1978. Gas-phase adsorption and air purification.
Ch. 9. In Carbon adsorption handbook, P.N. Cheremisinoff, F.
Ellerbusch (Eds.), p 331. Ann Arbor Science, Ann Arbor, MI.

Kwapong, O.Y., and J.H.Hotchkiss. 1987. Comparative sorption
of aroma compounds by polyethylene and ionomer food-contact
plastics. Journal of Food Science, 52(3):761

Labuza, T.P., and'W.M.Breene. 1989. Applications of “Active
packaging" for improvement of shelf-life and nutritional
quality of fresh and extended shelf-life foods. Journal of
Food Processing & Preservation, 13:1

Marshall, M.R., J.P.Adams, and J.W.Williams. 1985. Flavor
absorption by aseptic packaging materials. Proceedings
ASEPTIPAK '85. Third International Conference and Exhibition
on Aseptic Packaging, Princeton, NJ.

Modern.Plastics Encyclopedia ‘91. 1990. Mid-October 1990
issue, 67:11

92

Mohney, S.M., R.J.Hernandez, J.R.Giacin, B.R.Harte, and
J.Miltz. 1988. Permeability and solubility of d-limonene
vapor in cereal package liners. Journal of Food Science,
53(1):253

Mortland, M.M. 1970. Clay-organic complexes and interactions.
Adv. Agron., 22:75

Mortland, M.M. 1986. Mechanisms of adsorption of nonhumic
organic species by clays. Interaction of Soil Minerals with
Natural Organics and Microbes. Soil Sci. Soc. Amer., Madison,
WI.

Parliament, T.H. 1987. Sample analysis in flavor and
fragrance research. American.Lab., 19(1):51

Patton, G.W., L.L.McConnell, M.T.Zaranski, and.T.F.Bidleman.
1992. Laboratory evaluation.of polyurethane foamegranular
adsorbant sandwich cartridges for collecting chlorophenols
form.air. Anal. Chem. 64:2858

Perry, R.H., D.W.Green, and J.O.Maloney. 1984. Perry's
Chemical engineers' handbook,.McGraw-Hill chemical
engineering series, McGraw-Hill, Inc., New'York

Roland, A.M., and J.H.Hotchkiss. 1991. Determination of
flavor-polymer interactions.by vacuumpmicrogravimetric
method. Ch. 13. In Food and Packaging Interactions II, J.H.
Hotchkiss (Ed.), p 149. ACS Symposium Series 473. American
Chemical Society, Washington, DC.

Sacharow, S. 1991. Packaging meets 1990s needs through active
technology. Paper, Film.& Foil Converter, July, p 52

Sakodynskii, K., L.Panina, and N.Klinskaya. 1974..A study of
some properties of Tenax, a porous polymer sorbent.
Chromatographia, 7(7):339

Schroeder, M.A. 1989. Effect of sorption of flavor volatiles
on the adhesive and cohesive bond strength of multilayer
laminations. M.S. thesis, Michigan St. Univ., East Lansing

93

Sharp, G.J., Y.Yokouchi, and H.Akimoto. 1992. Trace analysis
of organobromine compounds in air by adsorbent trapping and
capillary gas chromatography/mass spectroscopy. Environ. Sci.
Technol. 26:815

Strandburg, G., P.T.DeLassus, and B.A.Howell. 1991.
Thermodynamics of permeation of flavors in polymers. Ch. 12.
In Food and Packaging Interactions II, J.H. Hotchkiss (Ed.),
p 133. ACS Symposium Series 473. American Chemical Society,
washington, DC.

Van Wijk, R. 1970. The use of poly-para-2,6-diphenyl-
phenylene oxide as a porous polymer in gas chromatography.
Journal of Chromatographic Science, 8:418

Vejrosta, J., M.Mikesova, A.Ansorgova, and J.Drozd. 1988.
Sorption of benzene on Tenax. Journal of Chromatography,
447:170

Vejrosta, J., M.Mikesova, and J.Drozd. 1989. Sorption of n-
alkanes on Tenax. Journal of Chromatography, 464:394

‘Wagner, B., and N.Vaylen. 1990. The packaging activists.
Prepared Foods, 159:172

Weber,‘W.J.Jr., 1972. Physicochemical processes for water
quality control. John Wiley & Sons, New York

 

MICHIGAN STATE UNIV. LIBRARIES
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
31293010282840