. a.
«44mm,

..;~
:
"..

. 1..
.444

“Ea“?
.o

imam»

1r...
& j
V #3:..." an

«we V, .45.

1:7

x
35:.

(a?
u... .

._..u.:.”.,nfi3n
» '0
fifififiufimfi?
at! W L A...
46 1 3 1 st
.4. i... W. x My. .
hung“: . 4 A ,

I...

u 4
‘ I 2.4! .1
3.9.2... 3.11".
12 {mg

3
.
I.

2;;
{fflf ’

Pat...“

quill,
. a...
«,4. a.

.435” V

. .4 .. .l$.:.J. £31.51... . SET»... ._‘.. ._ V . 4. .. 4,... z}.- 51... 1.... :.1..... .
.6... (4.. 42:: 4. . 2.. a... . . 4 H....a...fi.o. . .444f,...n£..$..\,2ffl?.. .mfjx Jummxfiw.._§

 

 

. i" “3’1 s

l

 

H? a UBRARY
6M957 Michigar. State
University

 

 

 

This is to certify that the
thesis entitled

ANALYSIS OF LINEAR POLYMER/HYPERBRANCHED
POLYMER EXTRUDATES

presented by

Joel K. Sutton

has been accepted towards fulfillment
of the requirements for the

Master’s degree in Chemical Enfleering

W/W

Major ProfessoW Signature

 

 

June 30, 2003

 

MSU is an Affirmative Action/Equal Opportunity Institution

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.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClRC/DateDuepes-p. 1 5

ANALYSIS or HYPERBRANCHED/LINEAR POLYMER EXTRUDATES
By

Joel Kellogg Sutton

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE

Department of Chemical Engineering and Material Science

2003

ABSTRACT
ANALYSIS OF HYPERBRANCHED/LINEAR POLYMER EXT RUDATES
By
Joel Kellogg Sutton

In this study the effect of flow through a capillary die on the surface
properties and morphology of a polymer blend was investigated. Previously, it
was shown that mixing a “hyperbranched polymer”, or HBP, with a linear polymer
resulted in an increase in the apparent shear rate for a given shear stress, when
compared to the virgin polymer. The postulate is that the HBP acts as a
processing aid, whose viability may be influenced by the capillary length.

The hypothesis is that the two polymers form an immiscible mixture and a
thin HBP layer near the fiber surface accounts for the increase in the apparent
shear rate. Thus, it is important to examine phase separation as a function of
capillary length to determine if a diffusive mechanism is present. Polymer
samples were extruded through a 30 mm length capillary die and a zero-length
capillary, or orifice plate; both the capillary and the orifice plate were 1.5 mm in
diameter. A 080 study was performed to determine the effect of temperature
and HBP concentration on phase separation. The extruded polymer samples
were viewed in TEM to visualize the layers.

The results show that the shear rate increase does exist and that the
HBPs and the virgin polymer do phase separate and that the HBP migrates to
the surface as hypothesized. Further, capillary length influences the morphology,

although the exact diffusive mechanism is unclear.

DEDICATION
I wish to dedicate this work to my parents, Bill and Sandra Sutton. All I have, I

owe to them.

iii

ACKNOWLEDGEMENTS

I wish to thank the following people. Without their help and inspiration, I
would never have made it.
1. Dr. Michael Mackay for helping me to learn more about myself as well as
teaching me the role and importance of research; without him, this project would
never have gotten off the ground
2. Drs. Richard Schalek and Alicia Pastor-Lecha for their instruction and helpful
advice in TEM sample preparation
3. Dr. Tien Thanh Dao for his technical advice in the experimental phase of this.
work and his thoughtful suggestions in proofing this manuscript
4. Dr. Jun Nogami for his helpful suggestions on the oral presentation of this
work

5. ACS-PRF 36388-AC7 for funding this work

iv

TABLE OF CONTENTS

LIST OF TABLES VII
LIST OF FIGURES VIII
KEY TO ABBREVIATIONS AND SYMBOLS X
1. INTRODUCTION 1
2. EXT RUSION RHEOLOGY FOR A 30 MM DIE 6
2.2. EXPERIMENTAL PROCEDURE 8
2.3. MATERIALS AND EQUIPMENT 9
2.4. DATA COLLECTION 9
2.5. RESULTS/DISCUSSION 1O
3. PHASE SEPARATION 17
3.1. INTRODUCTION 17
3.2. DSC EXPERIMENTAL PROCEDURE 17
3.3. DSC MATERIALS 18
3.4. DSC RESULTS AND DISCUSSION 18
3.5. TEM EXPERIMENTAL PROCEDURE 20
3.6. TEM MATERIALS 30
3.7. TEM RESULTS AND DISCUSSION 30
4. SURFACE ANALYSIS 41
4.1 . INTRODUCTION 41

4.2. MATERIALS 45

4.3.

4.4.

4.5.

4.6.

EXPERIMENTAL PROCEDURE

RESULTS

CONTACT ANGLE DISCUSSION

SURFACE ENERGY DISCUSSION

5. OVERALL CONCLUSIONS

6. SUGGESTED FUTURE WORK

WORKS CITED

vi

46

49

52

55

57

58

59

List of Tables

TABLE 1.1: HBP/LINEAR POLYMER SYSTEMS STUDIED ............................................... 5
TABLE 3.4.1: POLYMER TGS ................................................................................... 20
TABLE 3.5.1: OP FLow COMPARISON .................................................................... 28
TABLE 3.5.2: 30 MM CAPILLARY FLOW COMPARISON ................................................ 28
TABLE 3.7.1: COMPARISON OF BROWNIAN DIFFUSIVITIES FOR THE CAPILLARY ........... 40
TABLE 4.4.1: H20 CONTACT ANGLES FOR THE OP .................................................. 50
TABLE 4.4.2: CH2I2 CONTACT ANGLES FOR THE OP ................................................ 50
TABLE 4.4.3: SURFACE ENERGY RESULTS FOR OP ................................................. 51
TABLE 4.4.4: H20 CONTACT ANGLES FOR THE 30 MM CAPILLARY ............................. 51
TABLE 4.4.5: CH2I2 CONTACT ANGLES FOR THE 30 MM CAPILLARY ........................... 51
TABLE 4.4.6: SURFACE ENERGY RESULTS FOR THE 30 MM CAPILLARY ...................... 52
TABLE 4.5.1: COMPILED REFERENCE DATA FOR on. CONTACT ANGLES .................. 53
TABLE 4.6.1: LITERATURE SURFACE DATA40 ........................................................... 54

vii

List of Figures

FIGURE 1.1: CHEMICAL STRUCTURE OF A PS MONOMER1 .......................................... 1
FIGURE 1.2: CHEMICAL STRUCTURE OF A PE MONOMER1 .......................................... 1
FIGURE 1 .3: SCHEMATIC OF DIVERGENT GROWTH“ ................................................... 3
FIGURE 1 .4: SCHEMATIC OF CONVERGENT GROWTH5 ................................................ 3
FIGURE 1.5: SCHEMATIC OF A G3 HBP .................................................................... 4
FIGURE 2.1: CARTOON OF MELT FLOW ..................................................................... 6
FIGURE 2.5.1: PS AND PS/H50 MASTER CURVES, NORMALIZED TO 227 °C .............. 1 1

FIGURE 2.5.2: LLDPE AND LLDPE/H3200 MASTER CURVES, NORMALIZED TO 193 °C
.................................................................................................................... 12

FIGURE 2.5.3: LLDPE AND LLDPE/H40 MASTER CURVES, NORMALIZED TO 193 ’C. 13

FIGURE 3.5.1 : 3D IMAGE SHOWING AREA FOCUSED ON ............................................. 21
FIGURE 3.5.2: PS EXTRUDED THROUGH THE OP ..................................................... 22
FIGURE 3.5.3: PS/H50 EXTRUDED THROUGH THE OP; (L) 0.5 WT%; (R) 5.0 WT% ..... 22
FIGURE 3.5.4: PS EXTRUDED THROUGH THE 30 MM CAPILLARY ................................ 23

FIGURE 3.5.5: PS/H50 EXTRUDED THROUGH THE 30 MM CAPILLARY; (L) 0.5 WT%; (R)
5.0 WT% ....................................................................................................... 23

FIGURE 3.5.6: LLDPE EXTRUDED THROUGH THE OP ............................................... 24

FIGURE 3.5.7: LLDPE/H3200 EXTRUDED THROUGH THE OP; (L) 0.5 WT%; (R) 5.0

WTo/o ............................................................................................................. 25
FIGURE 3.5.8: LLDPE/H40 EXTRUDED THROUGH THE OP; (L) 0.5 WT%; (R) 5.0 WT%

.................................................................................................................... 25
FIGURE 3.5.9: LLDPE EXTRUDED THROUGH THE 30 MM CAPILLARY ......................... 26

FIGURE 3.5.10: LLDPE/H3200 EXTRUDED THROUGH THE 30 MM CAPILLARY; (L) 0.5
WT%; (R) 5.0 WT% ........................................................................................ 26

viii

FIGURE 3.5.11: LLDPE/H40 EXTRUDED THROUGH THE 30 MM CAPILLARY; (L) 0.5
WT%; (R) 5.0 WT% ........................................................................................ 27

FIGURE 3.5.12: FIBER SECTIONING DIAGRAM; FROM LEFT TO RIGHT: BEGINNING CROSS-
SECTION, “BULK” SECTION, AND “EDGE” SECTION. GRAYED OUT AREAS ARE THE

AREAS THAT WERE SECTIONED. ....................................................................... 28
FIGURE 3.7.1: PS/H50 30 MM CAPILLARY, BULK REGION, AT 27 KX ......................... 31
FIGURE 3.7.2: PS/H50 30 MM CAPILLARY, EDGE REGION, AT 14 KX ........................ 32
FIGURE 3.7.3: PS/H50 OP, BULK REGION, AT 2.7 KX ............................................. 32
FIGURE 3.7.4: PS/H50 OP, EDGE REGION, AT 10 KX .............................................. 33
FIGURE 3.7.5: LLDPE/H3200 30 MM CAPILLARY, BULK REGION, AT 27 KX .............. 34
FIGURE 3.7.6: LLDPE/H3200 30 MM CAPILLARY, EDGE REGION, AT 27 KX .............. 34
FIGURE 3.7.7: LLDPE/H3200 OP, BULK REGION, AT 14 KX .................................... 34
FIGURE 3.7.8: LLDPE/H3200 OP, EDGE REGION, AT 10 KX ................................... 35
FIGURE 3.7.9: LLDPE/H40 30 MM CAPILLARY, BULK REGION, AT 14 KX ................... 36
FIGURE 3.7.10: LLDPE/H40, 30 MM CAPILLARY, EDGE REGION, AT 14 KX ............... 36
FIGURE 3.7.11: LLDPE/H40 OP, BULK REGION, AT 14 KX ...................................... 37
FIGURE 3.7.12: LLDPE/H40 OP, EDGE REGION, AT 5 KX ........................................ 37
FIGURE 4.1.1: DIAGRAM OF THE WILHELMY PLATE TECHNIQUE30 .............................. 42
FIGURE 4.1.2: DIAGRAM OF THE CAPILLARY RISE TECHNIQUE30 ............................... 42
FIGURE 4.2.1: LIQUID PROPERTIES FOR THE GEOMETRIC AND HARMONIC MEAN

EQUATIONS31 ................................................................................................ 45
FIGURE 4.3.1: MICROBALANCE SET-UP ................................................................... 48
FIGURE 4.3.2: SAMPLE MICROBALANCE PLOT FOR PS/H50 AND WATER ..................... 49

ix

LLDPE
LDPE

PS

HBP
PDMS
G(Number)

H3200

H40

H50

OP

SAXS

DCA

ATR-FT I R

SANS

N/ R

Abbreviation

KEY TO ABBREVIATIONS AND SYMBOLS

Meaning

Linear Low Density Polyethylene
Low Density Polyethylene
Polystyrene

Hyperbranched Polymer
Polydimethylsiloxane

Generation (number) Dendrimer
Third Generation HBP, (10% -OH
groups, 90% -C20 groups)

Fourth Generation HBP, 100% -OH
groups

Fifth Generation HBP, 100% -OH
groups

Orifice Plate (zero-length capillary)
Small Angle X-Ray Scattering
Dynamic Contact Angle Analysis
Attenuated Total Reflection-Fourier
Transform Infrared Spectroscopy
Small Angle Neutron Scattering

Not Recorded

AP

O'w

xi

Meaning

Mass Flow Rate

Volumetric Flow Rate

Polymer Melt Density (also Liquid
Density); Definition depends on
Chapter

Apparent Shear Rate
Capillary Diameter
Capillary Length

Capillary Pressure Drop

Shear Stress

Depth of Immersion

Fiber Cross-sectional Area
Fiber Diameter

Gravitational Constant

Boltzmann’s Constant

Absolute Temperature

Particle Radius

1. Introduction

In the field of polymer science, engineering, and chemistry, much work
has been done on measuring the properties and behavior of linear polymers. A
linear polymer is a macromolecule with a primary straight chain and identical
repeating monomer units. Figures 1.1 and 1.2 depict two common linear

polymers: Polystyrene (PS) and Polyethylene (PE).

_CH2—-CH-—

Figure 1.1: Chemical Structure of a P3 Monomer1
——CH2—CH2—-
Figure 1.2: Chemical Structure of a PE Monomer1

A Simple literature search will reveal numerous papers and books on the
properties of linear polymers. Obviously, linear polymers’ properties may differ
drastically from each other.

In order to capitalize on these different properties, copolymers were
invented. A general definition of copolymers might be: a macromolecule
comprised of two or more different monomer units.2 Copolymers are subdivided
into classes according to the various ways in which the different monomers may
be organized in the molecule. Some examples of these classes are statistical,
alternating, block, and graft copolymers.2 Full definitions and descriptions of

these classes may be found in any general polymer text.

Copolymers Should not be confused with polymer mixtures or polymer
blends. A polymer blend is simply the mixture of two macromolecules, either in
solution or mechanically, whereas a copolymer is a macromolecule constructed
of two or more different monomers. Thus, it is possible to have a blend between
two different copolymers. Polymer blends also generally involve two non-
reactive and immiscible polymers, such as polystyrene and polyethylene.

Linear polymers differ radically from their cousins, the branched chain
polymers Since a branched chain polymer has either short chain or long chain
branching emanating from the linear Chain. A subgroup of the branched chain is
the dendrimer. Dendrimers are interesting molecules in that they are well
defined nanoparticles with a specific molecular weight and structure,3 with tree-
like branches emanating from a core.

The polymer chemistry used to create dendrimers can be quite involved.
Suffice it to say that it involves reacting an ABx structure repeatedly. Here, A and
B represent two different types of functional groups and x22. The two primary
methods proposed to date to create dendrimers are the “Divergent” and
“Convergent” Growth Methods. “Divergent growth” was first proposed by
Tomalia and coworkers.4 This method starts with a core and reacts the B)(

groups outward, as shown in Figure 1.3.

 

Figure 1.3: Schematic of Divergent Growth‘
The second method, first proposed by Hawker and Fréchet5 and used by

Mourey,6 begins with the desired end groups (8,) and reacts the A groups until
they reach a common point. These are called “dendrons”, which are then grafted
onto a core. This method is referred to as “convergent growth”. Figure 1.4, from

Hawker and Fréchet,5 demonstrates this process.

C
>—tp

S
S
c

S

 

Figure 1.4: Schematic of Convergent Growths

If the dendrimer synthesis is not strictly controlled, then a Hyperbranched
Polymer (HBP) is created. HBPS, therefore, exhibit similar properties to a
dendrimer, including low Viscosity, globular shape, and high functionality.7 The
imperfect structure is the source of some differences. In summary, HBPS are
essentially an intermediate between a dendrimer and a regular linear polymer.7
Some advantages of this imperfect structure are that HBPS are easier, cheaper,
and faster to produce than a dendrimer. For an industrial process, this should be
very attractive.

As a convention, both dendrimers and HBPS are given a generation
number. This generation number represents the number of layers from the core.
Each layer is a place where the Bx group has reacted. For example, a GB
dendrimer has three layers from the core. Figure 1.5 Shows an example of a

third generation HBP.

O

HO / ‘5 .\

 

 

Figure 1.5: Schematic of a Boltorn H30 (63) HBP Molecule

An interesting feature of both HBPS and dendrimers is the end group.
This “end group” corresponds to the Bx mentioned earlier. In traditional wet
chemistry, the outer groups of any molecule will dictate that molecule’s behavior
when it comes into contact with other molecules. The principle is the same for
polymers, regardless of the size or structure. An important question, then, might
be: what constitutes an outer group? In the case of dendrimers, the answer to
this question is still under discussion. A brief overview of this debate is given by
Fréchet.7 In the same paper, he concludes that in a dendritic polymer the end
groups will concentrate at the surface due to a dislike of the framework. It may
also be argued that due to their imperfect structure, HBPS are less constrained
sterically than dendrimers and thus, the “outer groups” are the “end groups”.

The HBPS used in this study were created using the divergent method.
The HBPS used also exhibit similarities and differences in their respective
generation numbers, end groups, and physical properties.

In this work, these HBPS were co-extruded with linear polymers and the
blends were analyzed through a variety of techniques. All of the polymers used
were commercial grade. Table 1.1 lists the basic information for all HBPS used,

as well as the linear polymers with which they were blended.

 

 

 

 

 

 

 

 

 

Generation End Linear System
HBP Name Number Group Polymer Name
Boltorn 10% -OH,
H3200 3 90 % ng LLDPE LLDPE/H3200
Boltorn H40 4 -OH LLDPE LLDPE/H40
Boltorn H50 5 -OH PS PS/HSO

 

 

Table 1.1: HBP/Linear polymer systems studied

2. Extrusion Rheology for a 30 mm Die

2.1 . Introduction

In order to properly understand the effect of the HBP on the extrusion

process, it is first necessary to examine the rheology of the linear polymers. In
Mackay and Henson,8 they concluded that in the flow of linear polymer melts that
there are two competing effects: a deposited layer which Slips along the capillary
wall and a decrease in the flow area due to this deposited layer. In their work,
they performed rheological tests on linear polystyrene, as well as LLDPE. Figure
2.1 shows a cartoon of this effect. The part labeled ‘Wall Layer” corresponds to

this deposited layer.

Extruder Wall

\

\

 

 

J

i I
/ wian Layer

Bulk Polymer Region

 

 

 

 

Figure 2.1: Cartoon of Melt Flow

While Mackay and Henson performed their experiments with a rheometer,
the polymer melt flow will be the same, regardless of flow geometry. There are
two key phenomena that are observed in any flow system: Shear stress and
shear rate. Shear stress may be defined as the effect a force has on a fluid,
while shear rate may be defined as the velocity gradient across a flow field. For

a Newtonian fluid, the relationship between these two values Should be linear.

That is to say that the fluid viscosity is independent of either of these two values.
Polymer melts, however, are Non-Newtonian fluids because their viscosities are
dependent on the shear stress and shear rate. Equations [2.1] and [2.2] give the

relations for the apparent shear rate and the shear stress in a capillary of finite

 

length.
' _32Q_ 32m [2.1]
7a - ”D3 - —'_3‘
,0er
0W = DAP [22] .
4L

The Apparent Viscosity is a ratio of the Shear Stress to the Apparent Shear Rate.

The other important boundary is an understanding of the HBP’s behavior.
Hsieh, et al.9 performed a series of rheometric studies on two of the three HBPS
used in this work: H40 and H50. Their main conclusion was that H40 and H50
behave as Newtonian fluids under both oscillatory and steady shear. Thus, the
viscosity remains constant when the shear rate is constant.

In her thesis,10 Carmezini reported that adding a HBP in a quantity as low
as 0.5 wt% to a linear polymer would result in a shift of the shear stress/shear
rate master curve to a lower apparent viscosity. Carmezini performed an
extensive study of the extrusion of several linear polymer/HBP mixtures and
observed this decrease for at least one system studied per linear polymer. Thus,
a necessary first step was independently corroborating this observation for three

systems, denoted PS/HSO, LLDPE/H3200, and LLDPE/H40.

2.2. Experimental Procedure
The following general procedure was followed for all three systems. Any
deviations will be noted at each step.
The first step was to extrude the pure polymer and plot the apparent shear
stress versus the apparent shear rate as a reference curve.
The next step was to extrude the 0.5 wt % polymer blends. In order to
introduce the HBP into the system, it was dissolved in Tetrahydrofuran (T HF). In

order to aid its solubility, 500 mg HBP was heated to 100 °C and kept there for
ten minutes. The HBP was then allowed to cool down to 60 °C. At this point, 50

mL THF was added and the solution was allowed to mix for a period of twenty-
four hours. The dissolved HBP solution was then poured over the bulk polymer
pellets, which were Spread out on two trays, lined with aluminum foil.

In the case of PS, it iS extremely important to note that the pellets must be
spread out as much as possible, otherwise, due to the high solubility of PS in
THF, the pellets could fuse together. After letting the pellets sit in the hood for a
couple of days, the PS/H50 pellets had to be cut apart. LLDPE is insoluble in
THF, and thus this cutting procedure is unnecessary.

Note that it is important to keep the PS/H50 and LLDPE/H40 out of the
open air, as both H50 and H40 are highly hydroscopic, and any moisture or
solvent in the system will cause foaming in the extrusion process. Finally, the
pellets were allowed to sit under vacuum at room temperature for a week, after

which extrusion was performed.

When performing the extrusion experiments, the pellets were fed into the
hopper and the system was allowed to run until the system reached a pseudo-
Steady state. At this point the extrudate was cutoff and a stopwatch was started
simultaneously. After about 30 seconds, the stopwatch was stopped and the
extrudate cutoff at the same time. The time was recorded and the extrudate
weighed and saved. Three runs were performed for each motor setting from 15
to 85 at increments of 10. These motor settings roughly correspond to 10 to 70
RPMS, respectively. Optical micrographs of the extrudates may be found in
Section 3.5.

2.3. Materials and Equipment

The PS and LLDPE used were industrial grade Dow products. The PS
was Dow Styron 67BC-W, Lot #OH10017P08; the LLDPE was Dow Dowlex
2045A, Lot #NH1109210. The HBPS were products of Perstorp, Inc. The THF
was a product of Fischer Scientific. All products were used as delivered and
without modification, except as noted above. The extruder used was a home
built device and was the same extruder used in Carmezini’s experiments.

2.4. Data Collection

The data collected for this section included the melt pressure, amperage
used, motor RPMS, melt temperature, die temperature, barrel temperature, time
of extrusion, and extrudate mass. This data was then used to generate the
extrusion master curves. Equations 2.1 and 2.2 were used to calculate the

apparent Shear stress and shear rate.

2.5. Results/Discussion
Figures 2.5.1 through 2.5.3 are the master curves for shear stress/Shear
rate created following the procedure outlined above. While not identical to the
previous work, they are extremely close. Thus, the observed behavior10 was
independently corroborated, and it must be concluded that the phenomena
exists.
In Figure 2.5.1, a comparison of the PS curve and the PS/H50 flow curve

is made. These flow curves were conducted Over a range of 178 °C to 227 °C for

the apparent Shear rate/shear stress as well as the apparent shear rate/apparent
viscosity master curve. Adding H50 to PS resulted in a significant melt pressure
drop for higher apparent shear rates. For example, a mixture of PS and H50 at

an operating temperature of 227 °C and running at 43 RPMS has a melt pressure

of 310 psi; PS, running at Similar conditions, has a melt pressure of 460 psi.
Thus, adding H50 to PS grants a 33% decrease in the melt pressure. In order to
get PS to run at 340 psi, the extruder motor would have to be set at about 26
RPMS. In addition, adding H50 to PS yielded a substantial decrease in the
apparent viscosity for higher apparent shear rates. The reason for this decrease

is due to the reduction in melt pressure.

10

 

 

 

 

 

 

1.0E+03 ,_ X 11 1.805
s 4‘“ A
‘ ,
r “3? Q “0‘ ”X
A - X
”‘10“ 1.E+O4
’3
C
U
i s
3' o o » 1.E+03 c.
'3 I’. ~ 0% 8
.0
§ 1.0E+02 ,. an?“ 0 g
— ‘ ' '
> I o. (n
g 'I 1.E+02 E
E ’ 0 “’
< O
I 1.E+01
1.0E+01 1.E+00
1.0E+01 1.0E+02 , 1.0E+03 1.0E+04

Apparent Shear Rate (8")

 

"‘o’PfSVTsIE: BAPSH50 Visc. A PS Shear x PSHso Shear;

 

Figure 2.5.1: PS and PSIHSO Master Curves, Normalized to 227 °C
Figure 2.5.2 depicts the LLDPE/H3200 master curves in comparison to the

LLDPE master curves, over a temperature range of 174 °C to 225 °C. In the
higher apparent shear rate regimes, adding H3200 resulted in a melt pressure
drop of about 19%. For example, at an operating temperature of 193 °C and
about 44 RPMS, the melt pressure for LLDPE/H3200 is approximately 1640 psi,
whereas at similar conditions, the melt pressure for plain LLDPE is approximately
2030 psi. In order to achieve a melt pressure of 1640 psi for the LLDPE system,
the extruder motor would have to be slowed to 25 RPMS. Once again, adding
H3200 resulted in a decrease in the apparent Viscosity; this decrease also

corresponds to a decrease in the melt pressure.

11

 

 

 

 

1.0E+04 1.0E+06
O
O
2 4‘
,3 9.03.
r. ‘2‘)
a ’ e 01‘“ 1.0E+05 1T
"' 0 AA
Q 0' if
g 0. g
> e 9
E t 0 g
2 . 9 . m
i n O. 1.0E+04
< " °
I}. Q ‘
U‘ ~
'5 3.
I ° 0.
B o
'I I, 3
1 .0E+03 "I 1.0E+03
1.0E+01 1.0E+02 1.0E+03

Apparent Shear Rate (s4)

 

fiIIDPE YE ijLoPEHazoo Visc. e LLDPE Shear A LLDPEH3200 Site;

 

 

Figure 2.5.2: LLDPE and LLDPE/H3200 Master Curves, Normalized to 193 '0

Figure 2.5.3 illustrates the LLDPE and LLDPE/H40 master curves over a
range of 171 °C to 227 °C. At higher apparent shear rates, adding H40 to LLDPE
also resulted in a reduction in the melt pressure of about 44%. This value comes
from comparing LLDPE/H40 operating at 193 °C and 44 RPMS, with a melt
pressure of 1410 psi to LLDPE operating at similar conditions, with a melt
pressure of 2030 psi. In order to extrude LLDPE at 1410 psi, the motor would
need to run at about 20 RPMS. Here, again, the apparent viscosity is greatly

reduced. Once again, this effect is due to the drop in the melt pressure.

12

 

 

 

 

 

A A
M ‘ ‘ . O.
“‘.‘3Af;,*:.o ’71.0E+05
5‘ .0 .
00
1? . ° 9 o
.P . 9 g + 10E+04
e" ’g ‘ ' 0‘ . ?
9.. e
E it a. v
a J .- ° s 3
§ “' ° ' 2
_ 1.0E+03 I a 9+ 1.0E+03 iii
>
E I '. I 3
9 II
8, 1.0E+02 m
a.
4
«t 1.0E+01
1 .0E+02 1 .0E+OO
1.0E+01 1.0E+02 1.0E+03

Apparent Shear Rate (8")

 

.. LLDPEVIsE— nLLD_P_E:l40 VIEJA LLDPE Shear e LLDPEH40 Shea—Ir“

 

 

Figure 2.5.3: LLDPE and LLDPE/H40 Master Curves, Normalized to 193 “C
From the specific cases cited earlier, it is readily apparent that H50 and

H40 increased the apparent shear rate for a given shear stress to a greater
degree than H3200. The individual effects of H40 and H50 on the apparent
shear rate are incomparable due the innate differences in the base polymers for
these systems. However, H4O does affect a LLDPE blend master curve to a
greater degree than does H3200 (44% vs. 19%). H50 also substantially affects
the PS master curve (33%) quite a bit. These two large values (44% and 33%)
suggest that the hydroxyl end groups are largely responsible for the shifts in the
master curves. In support of this statement, all three HBPS studied have
hydroxyl end groups, albeit varying amounts. H3200 has the least; it is a H30

HBP that has been altered such that 90% of its end groups are —Czo groups,

13

instead of the usual —OH groups. H40 and H50 have 100% hydroxyl end groups,
and they caused greater shifts in the master curves than did H3200.

Another Obvious observation is that the increase in the apparent shear
rate for both PS and LLDPE blends was greatest in the higher shear regions.
Previous work by Hong‘“ ‘2 has shown that a HBP may be used as a processing
aid.

This idea has recently been corroborated by Kil, et al.13 who extruded a
reactive blend with a co-rotating twin-screw extruder. Their sample preparation
differs from that stated in this work, but they observed the same decrease in
viscosity as reported in this work.

Sendijarevic, et al.14 showed similar behavior for LLDPE/Alkyl terminated
HBP systems. They Showed, as was found here, that at higher operating
temperatures that the benefit of the HBP is lost. One major difference between
their work and this work is that they used a slit die for their extrusions, while a
capillary die was used here.

Mulkem and Tan15 reported a decrease in the viscosity/shear rate curve
for a PS/H40 system and also a PS/SMA copolymer blended with H40 (SMA
stands for styrene maleic anhydride). This result concurs with these
observations on a PS/HSO system, as well as Carmezini’s results.‘0

All of this work was done with therrnoplastics, i.e. LLDPE, PS, etc.
Mezzenga, et al.” showed that the phase separation and modified rheology was

present in thermosets as well. In this work, they examined the effect of four

14

different HBPS, similar to those used in the present work, on DGEBA (Diglycidyl
Ether Bisphenol A).

Thus, adding a HBP will affect the rheology of a wide range of base
polymers, and it may be said that these HBPS are acting as processing aids. A
processing aid is a component added to a polymer flow system to enhance
system performance while reducing wear on a processing machine, such as an
extruder, mixer, etc. Some examples of processing aids are fluoroelastomers, a
second polymer, or some other compound.

One of the first recorded instances of “processing aid” being used was by
Zelinger, et al. in 1976.17 In this work, they added methylmethacrylate,
polystyrene, and polyvinylacetate to polyvinylchloride (PVC) in varying
concentrations up to 5 wt%. They found that the addition of these compounds
increased the polymer output without Significantly affecting the overall physical
properties of the bulk polymer.

Continuing with this work, Kanu and Shaw18 examined the effects of
different capillary entrance geometries in an Ethylene-Propylene-Diene
(EPDM)Niton system. EPDM is a common copolymer; Viton is the trade name
for a fluoroelastomer. They found that the Viton caused Slip, but it did not form a
true layer. Instead, they proposed that the Viton acted more like ball bearings, in
which the bearings assisted the polymer melt flow through the capillary.

Building on this work, Akay19 investigated the rheology of a glass or
calcium carbonate filled polypropylene and a glass or calcium carbonate filled

Nylon 6,6 polymer system. In this work, Akay discovered that the glass and

15

calcium carbonate aligned in the polymer melt as it entered and passed through
a capillary. In addition, he found that as the polymer melt exited the capillary that
an aid-free region formed in the center of the polymer melt. He hypothesized
that the reason for this separation was particle exclusion through vortex flow. He
suggested that the aligned region is continuous throughout the entire length of
the capillary.

In all of these cases, the processing aid phase separated from the bulk

polymer to some extent.

16

3. Phase Separation
3.1 . Introduction

In her thesis, Carmezini” hypothesized that the decrease in the master
curve was due to phase separation in the barrel, thus causing slippage on the
screw. In order to test Carrnezini’s assertion of phase separation, two tests were
performed on the extrudates. The first test was to create a phase diagram for the
PS/H50 system in order to see if temperature or concentration played any role in
the hypothesized phase separation. The secOnd test involved using
Transmission Electron Microscopy (T EM) to see the different phases. Thus, if
Caremezini’s assertion is correct, then there Should be a larger fraction of HBP in
the Wall Layer than the Bulk, as shown in Figure 2.1.

3.2. DSC Experimental Procedure

A Differential Scanning Calorimeter (DSC) is an instrument that is
designed to measure the heat flow required to increase the temperature at a
constant rate. At some point a phase transition will occur and the heat flow will
change to maintain the constant rate. The heat flow versus temperature graph
can be used to observe the phase transition.

The DSC experimental procedure was fairly straight forward. The DSC
used in this study was a TA Instruments Q-1000. Part one entailed obtaining the
glass transition temperature (T g) for the HBPS. The sample preparation
procedure followed was specified by the DSC documentation. Essentially, this
involved weighing out 3-5 mg of sample into a special pan and placing the pan

into the instrument. The first run was set up to heat the sample up to 100 'C and

17

keep it there for 10 minutes in order to drive off any bound water. Then, two
isothermal modulated runs were performed on the sample, in 10 'C increments.20
From the second run, the reversible heat flow curve was plotted and this plot was
used to obtain the T9 for all of the HBPS tested.

Part two of this experiment was to create a phase diagram for the PS/H50
system. Because LLDPE is highly insoluble in any solvent, an attempt at such a
phase diagram would be very difficult. PS, on the other hand, will easily dissolve
in many solvents. Ideally, then, this phase diagram would demonstrate the
concentration and temperature effects on the HBP phase separation.

The set-up was fairly simple. First, the HBP was dissolved in glass vials
with THF as described in Section 2.2 in solutions of 5, 10, 15, 20, 30, and 50
weight percent. The Vials were then sealed with polypropylene caps and allowed
to mix overnight on a stir plate. The following day, the linear PS was measured
into the Vials in the appropriate concentrations and allowed to mix overnight.
Finally, cyclohexane was used as a precipitating agent. This solution was
allowed to sit in the hood for a week, after which it was placed into a vacuum
oven for another week to remove any remaining solvent. The mixtures were then

measured into the DSC pans as described above. Then, DSC runs were

performed on the samples, starting at 100°C. The temperature was then
increased by 10°C, and the test re-run. This process was repeated until phase

separation was observed.

18

3.3. DSC Materials
Materials studied here include Boltorn H20, H30, H40, H50, as well as JE3
and JE4. All Boltorn HBPS have hydroxyl end groups; however, JE3 and JE4 are
modified Boltorn H30 and H40.21 JE3 and JE4 differ from their base polymers in
that per-fluorinated benzene groups have been substituted for the hydroxyl
groups?"1 PS and LLDPE are the same as that listed for the previous section.
The solvents, THF and cyclohexane, were products of Fischer-Scientific.
3.4. DSC Results and Discussion
Table 3.4.1 shows the Tgs of all HBPS and linear polymers tested. H3200
is not listed because no Tg could be found; only a melt temperature transition
was observed at 57 °C. The Tgs for H30 through H50 came from MalmstrOm, et
al.;22 these values were measured on a Perkin-Elmer DSC-7, also at 10 '0 per
minute. No value was reported in this paper for H20. Mulkem and Tan15 report a
value of 32 °C for H40, which agrees very well with the value measured here.
The Tgs for JE3 and JE4 come from Englund’s thesis.21 In his work, he used
both a Perkin-Elmer DSC as well as a Mettler DSC; the data reported here are
for the Perkin-Elmer.
Generally, the rate of temperature increase for a DSC run is about 5 “C

per minute. Runs were originally done at 5 °C per minute, but the results were
inconclusive. Therefore, the heating rate was increased to 10 °C per minute

since Kim and Beckerbauer20 found that this higher rate gave better results for

their substituted hyperbranched polyphenylenes by avoiding the crystallization

19

present at lower heating rates. This higher heating rate provided more accurate

results for the HBPS studied.

 

 

 

T9 (DSC) T9 (Literature)

Polymer [°C] [°C]21’ 2’ % Error
JE3 23. 27.7 14.1
JE4 31 . 321 5.2
H20 2. N/A N/A
H30 1 8. 34. 47.1
H40 ' 30. 37. 17.0
H50 34. 41 . 1 5.9
PS 9 ~100 ~1

LLDPE IN/A ~-2o N/A

Table 3.4.1: Polymer T,s
The Q-1000 DSC data, on the whole, corresponded within reason for all

but H30. Some possible reasons for these discrepancies in the data might be a
difference in the instruments, sample size, difference in calibrations, or the type
of sample pan used.

As was mentioned above, these Tgs would have been used to obtain the
phase diagram for PS/H50. However, the precipitation step provided an
unexpected result. Instead of forming a nice, even mixture, the HBP precipitated
onto the sides of the vial as the cyclohexane/T HF solution evaporated. This
precipitation demonstrates that these linear and hyperbranched polymers are
immiscible. This result agrees with the results that Massa, et al.23 found for their
linear polymer/HBP blends. Therefore, one of the driving forces in this system
must be inherent immiscibility.

3.5.TEM Experimental Procedure
In performing these tests, it was desirable to determine whether the phase

separation was a result of flow in the capillary. Therefore, two conditions were

20

set for each system. Condition one involved extrusion through the 30 mm
capillary discussed earlier, while condition two involved extrusion through a zero
length capillary, which iS referred to as an orifice plate (OP).

Previous work11 has shown that adding a HBP to a linear polymer will
affect the surface appearance. AS such, it was important to take optical
micrographs of the surface in order to see how the surface was altered. In taking
these pictures, a JAl, Inc. CV-S3200 digital camera was used with ViSIlog
software. Because the fibers studied were cylindrical, it was Impossible to get an
image of the full surface. Therefore, only the highest point was focused on, as .
shown in Figure 3.5.1. The extrusion conditions for all optical micrographs are

analogous to those shown in Tables 3.5.1 and 3.5.2.

Area of Focus

 

 

 

Figure 3.5.1: 30 image showing area focused on

Figure 3.5.2 Shows the virgin PS, while Figure 3.5.3 Shows the effects of
0.5 and 5.0 wt% H50 on PS through the OP. Figures 3.5.4 and 3.5.5 Show the
same systems, except extruded through the 30 mm capillary. All images are at a
lens magnification of 10X and a camera magnification of 26X. The addition of

H50 to PS results in a smoother surface for both the OP and capillary.

21

 

Figure 3.5.3: PSIHSO extruded through the OP; (L) 0.5 wt%; (R) 5.0 wt%

22

 

Figure 3.5.5: PSIHSO extruded through the 30 mm Capillary; (L) 0.5 wt%; (R) 5.0 wt%

23

Figure 3.5.6 shows a sample LLDPE extrudate through the OP. Figure
3.5.7 shows a sample LLDPE/H3200 extrudate of both 0.5 wt% and 5.0 wt%,
also through the OP. Figure 3.5.8 shows the difference between 0.5 wt% and
5.0 wt% for the LLDPE/H40 system. Figures 3.5.8 through 3.5.11 Show these
same systems, except that they have been extruded through the 30 mm
capillary, instead of the OP. These images differ from those shown for PS in that
the surface now appears smoother for both the OP and the Capillary. This
increased smoothness has been shown in preVious work for a LLDPE/H3200

system.1 1

 

Figure 3.5.6: LLDPE extruded through the OP

24

  

Figure 3.5.7:

 

 

 

 

 

Figure 3.5.8: LLDPE/H40 extruded through the OP; (L) 0.5 wt%; (R) 5.0 wt%

25

 

Flgure 3.5.10: LLDPE/H3200 extruded through the 30 mm Capillary; (L) 0.5 wt%; (R) 5.0
wt%

26

 

Figure 3.5.11: LLDPE/H40 extruded through the 30 mm Capillary; (L) 0.5 wt%; (R) 5.0 wt%
Originally, it was intended to obtain TEM images of the 0.5 wt% blends

extruded in the process of constructing the master flow curve, but this proved
impractical due to the low HBP concentration. That is, even at high
magnification, it was impossible to see the HBP. Thus, the HBP concentration
was increased to a 5 wt% solution. According to previous work,” a 5 wt% blend
will also result in a master curve shift.

The pellets for the 5 wt% blends were created following the same
procedure outlined in Section 2.2, with the obvious exception that more was HBP
dissolved into solution.

After the pellets were created, they were extruded at operating
temperatures of approximately 200 °C for PS/H50 and 190 °C for LLDPE/H3200
and LLDPE/H40. Additionally, all samples were extruded at a motor setting of

55, which equates approximately to 43 RPMS. These extrusion settings were

27

selected on the basis of the master curves presented earlier (Figures 2.5.1 and
2.5.2); on these curves, the chosen temperatures and motor speed correspond to
the reduced master curve regime. Thus, these two temperatures should yield
TEM images analogous to ones expected for a lower weight percent. Tables
3.5.1 and 3.5.2 Show the comparison of the flow rates at the selected motor

settings and temperatures for both the OP and 30 mm capillary.

 

 

 

 

 

 

 

 

OP, 43 For LLDPE, Tme|t=190 C;
RPMs For PS, Tmen=200 C
Mass Apparent Shear Apparent

mass time Rate Amps Pm... Shear Stress Viscosity
Material (9) (s) J8) ijL (psi Rate (s4) JPa) (Pa's)
PS 1.2 35.53 0.0328 350 30 101 N/R N/R
PS/H50
5.0 wt% 0.9 46.5 0.0086 240 40 60 N/R N/R
LLDPE 1.2 37.62 0.0306 660 130 95 N/R N/R
LLDPE/
H3200
5.0 wt% 0.9 27.47 0.0335 270 130 104 N/R N/R
LLDPE/
H40
5.0 wt% 0.9 37.43 0.0230 370 90 71 N/R N/R

 

 

 

 

 

 

 

 

 

 

Table 3.5.1: OP Flow Comparison

 

 

 

 

 

 

 

 

30 mm
Cap” 43 FOI‘ LLDPE, Tme|t=190 C;
“PMS FOI' PS, Tme|t=200 C
Mass Apparent Shear Apparent

mass time Rate Amps Pm... Shear Stress Viscosity
Material is) (s) M) (mAL (psi Rate (3") (Pa) (Pa‘s)
PS 1.1 31.69 0.0344 370 450 106 38000 360
PS/H50
5.0 wt% 1.2 45 0.0267 360 260 82 22000 270
LLDPE 0.9 30.85 0.0299 650 2030 120 173000 1440
LLDPE/
H3200
5.0 wt% 0.5 20.87 0.0255 320 620 102 53000 520
LLDPE!
H40
5.0 wt% 0.7 37.06 0.0200 380 520 80 44000 550

 

 

 

 

 

 

 

 

 

 

Table 3.5.2: 30 mm Capillary flow comparison

28

These tables Show that the extrudates studied will fall within the desired
apparent shear rate range, due to the increased mass flow rate and the
decreased pressure.

It is very interesting to note the similarity in the values of the apparent
Shear rate for both the capillary and orifice plate. There are no reported values
for the Shear Stress and Apparent Viscosity for the OP because the shear stress
calculation requires dividing by finite length, and an OP has zero length (see
Equation 2.2), and the apparent viscosity is dependent on the Shear stress.

To facilitate TEM sample preparation, the fibers were pulled Slightly as the
polymer exited the extruder. These fibers were also handled with gloves in order
to preserve the surface.

The extruded fibers were next carved into blocks, prior to sectioning.
Figure 3.5.1 Shows how the blocks were prepared from a cross-sectional
perspective. It is important to note that the “Edge” region may also include a
small section of “Bulk”, but the goal was to get a section within the first few
hundred nanometers from the edge. In this diagram, straight lines represent

cutting lines and the grayed out areas are the places on the blocks that were

sectioned. Sizes are exaggerated for illustration purposes.

 

Figure 3.5.12: Fiber sectioning diagram; from left to right: beginning cross-section, “Bulk”
Section, and “Edge” Section. Grayed out areas are the areas that were sectioned.

29

After making the blocks, an Edgemaster II diamond knife in a RMC MT-7
ultramicrotome, with a cryo unit attachment, was used to section the 5 wt%
extruded fibers into 80 nm sections. The cutting temperature used for all blends
was -60 °C. This temperature was selected in order to ensure that the samples
were below the T9 for both polymers in the system. The thin polymer sections
were then placed onto 300 nm copper grids and vapor stained with OsO4. In
both previous works,” 24 RUO4 was used as the staining agent, but 0804 is very
similar in most respects to RuO4. From page 106 of Flegler, et al.,25 “Its [RUO4’s]
fixative properties are closely related to those of OsO4...but is a stronger
oxidizing agent” (brackets and ellipses added).

These samples were then examined with a JEOL .JEM-1OOCX ll TEM
using the 100 kV setting.

3.6.TEM Materials

The HBPS used were Boltorn H3200, H40, and H50 from Perstorp, Inc.
The linear polymers were the same PS and LLDPE used in Section 2. 0804 was
the staining agent and was obtained as a powder from Electron Microscopy
Sciences and prepared by Dr. Alicia Pastor-Lecha in the Center for Advanced
Microscopy at Michigan State University. The copper grids were from Ted Pella,
Inc.

3.7.TEM Results and Discussion
The following pages show the TEM images recorded for the different

systems. The systems will be presented in the following order: PS/HSO,

30

LLDPE/H3200, and LLDPE/H40, with the 30 mm capillary images first, followed
by the orifice plate images.

In Figures 3.7.1 through 3.7.4 the results of the PS/H50 system are
compared. All images have a gray area, as well as darker spots. These darker
spots represent the HBP, while the surrounding gray area represents the bulk
polymer. These images correspond very well with the results observed by Kim
and Webster.24 Kim and Webster examined a 5 wt% PS/Hyperbranched
Polyphenylene system.

It is interesting to note that in both the OP and capillary cases that there
are more spots in the edge region than in the bulk region. The spots have been

circled for easier identification.

 

Figure 3.7.1: PS/H50 30 mm Capillary, Bulk Region, at 27 RX

31

 

Figure 3.7.2: PSIHSO 30 mm Capillary, Edge Region, at 14 kX

 

Figure 3.7.3: PSIHSO OP, Bulk Region, at 2.7 RX

32

  

1300' Ea

Figure 3.7.4: PSIHSO OP, Edge Region, at 10 kX .
In Figures 3.7.5 through 3.7.8 the results of the LLDPE/H3200 system are

compared. Once again, it is obvious that there are some dark areas and some
lighter areas. The darker areas again represent the HBP, while the lighter areas
represent the bulk polymer. This interpretation is in disagreement with Hong et
al.,11 but the final results are the same in that they show a large amount of HBP
in the edge region.

In the case of LLDPE/H3200, there is a higher concentration of the HBP in
the edge region of the 30 mm capillary fibers than in the bulk region. In addition,
the OP results also show a larger amount of HBP in the edge region than in the

bulk region. The HBP has once again been circled for easier identification.

33

 

Figure 3.7.5: LLDPE/H3200 30 mm Capillary, Bulk Region, at 27 kX

     
  

 

we

Figure 3.7.6: LLDPE/H3200 30 mm Capillary, Edge Region, at 27 kX

 

Figure 3.7.7: LLDPE/H3200 OP, Bulk Region, at 14 RX

34

  

' «his?
Figure 3.7.8: LLDPE/H3200 OP, Edge Region, at 10 kX

Figures 3.7.9 through 3.7.12 illustrate the LLDPE/H40 system. It is
interesting to note the similarities between Figure 3.7.8 and Figure 3.7.12. This
system shows the same phase separation as the other two. The large, white
circles in Figure 3.7.12 are holes in the sample. These holes are probably the
results of water vapor interacting with the H40 during the staining process. A
similar phenomenon was observed in the PS/H50 system. As before, the darker
areas generally represent the stained HBP. Some exceptions are grid bars,

curled polymer parts, and Shadowing.

35

 

Figure 3.7.9: LLDPE/H40 30 mm Capillary, 8qu region, at 14 kX

  

a»

Figure 3.7.10: LLDPE/H40, 30 mm Capillary, Edge region, at 14 RX

36

 

(w M. .. no

Figure 3.7.11: LLDPE/H40 OP, Bulk Region, at 14 kX

       

Figure 3.7.12: LLDPE/H0 P, Edge rlon 5 kX
The most important conclusion that may be drawn from these images is
that the two polymers form two separate phases, even in the presence of
mechanical mixing. In addition, the HBP migrates to the surface of the polymer
melt regardless of whether or not the capillary die is present.
Other workze' 27 has shown strong microphase separation in HBP/linear

'26

polymer diblock copolymer systems. Mackay, eta attached a GS polybenzyl

ether to a linear polystyrene chain. They observed that dendron/Iinear copolymer

37

system would phase separate. This separation moves from an ordered cylinder
to an ordered lameller to a disordered Iamellar as the linear mass fraction
decreases. Roman, et al.27 observed a Similar phenomenon for chemically
dissimilar hybrid block copolymer system. These phase separations were
visualised by TEM and SAXS. Therefore, phase separation is a common
occurrence between linear and hyperbranched polymers, even when they are
physically attached to each other.

Lee and Archer28 showed through Dynamic Contact Angle Analysis (DCA)
and Attenuated Total Reflection-Fourier Transform Infrared Spectroscopy (ATR-
FTIR) that for a PS/PDMS copolymer that the PDMS would concentrate at the
surface of the blend, even when mixed in at a 2 wt% concentration. Despite the
fact that the PDMS is linear, as was the PS, this work is valuable in that it shows
that copolymer phase separation also exists for linear/linear systems as well as
linear/branched systems.

Sendijarevic, et al.14 claimed that HBPS with C18 end groups are miscible,
and that HBPS with smaller Size and groups are immiscible. The TEM images
presented here of the LLDPE/H3200 system, along with Hong’s image of the
same system,11 Show that this is not the case, and that a C20 terminated HBP
phase separates. It is possible that the 10% -OH end groups are sufficient to
cause the observed phase separation, but is unlikely due to steric hindrance.

Mulkem and Tan15 observed phase separation through Scanning Electron
Microscopy (SEM) for a PS/H4O system. Ideally, they also would have used

TEM, but it is important to note the complimentary result.

38

Bauer,‘et al.29 showed that mechanical mixing of a fatty acid dendrimer
and a polyolefin will result in an immiscible mixture through Small Angle Neutron
Scattering (SANS). Their materials were analogous to the ones used here.

Therefore, a highly branched polymer and a linear polymer will refuse to
blend together either when mixed in solution or mechanically dispersed in the
extruder at any concentration, and that the HBP will migrate to the surface layers
in both the capillary and the OP systems.

If this phase separation is independent of temperature, concentration, and
capillary presence, why is the HBP to rapidly diffusing to the surface?

One possibility is Brownian diffusion. Brownian diffusion may be defined
as the continuous fluctuation of a particle on a molecular scale. Equation [3.1] is
the equation for the approximate diffusivity for Brownian motion.

D z kBT [3.1]
6717770
Using the temperatures and the apparent shear rates from Table 3.5.2 for the
capillary, along with the approximate diameters measured for HBP Spots in
Figures 3.7.2, 3.7.5, and 3.7.10, the approximate Brownian diffusivities may be
obtained. In comparison, a molecule is generally on the order of 1 nm in
diameter. Table 3.7.1 compares the Brownian diffusivities for both the 1 nm

particle, as well as the HBP particles.

39

 

 

 

 

 

 

30 mm Cap.,
43 RPMs For LLDPE, Tmfl=190 C; For PS TM=200 C

Apparent D (cm’ls)

Viscosity [d=previous D (cm’ls)

Material (Pa's) d (nmL column] [d=1 rim]

PS/H50 5.0
wt% 268 400 6.46E-14 2.58E-1 1
LLDPE/H3200
5.0 wt% 516 170 7.91 E14 1.32E-11
LLDPE/H40
5.0 wt% 550 200 6.30E-14 1.23E-11

 

 

 

 

 

 

Table 3.7.1: Comparison of Brownian Diffusivlties for the Capillary

A quick examination of Table 3.7.1 shows the obvious fact that the
diffusivity for a 1 nm particle is greater than the diffusivity for a particle greater
than 100 nm. The self-diffusivity of PS is approximately between 10‘8 and 10'12
cm2/s. Assuming that the self-diffusivity of LLDPE is of the same magnitude as
the self-diffusivity of PS, it is obvious that the HBP diffusivity in the base polymer
is at least two orders of magnitude smaller than this self-diffusivity. Thus,
Brownian diffusion cannot explain the observed rapid diffusion, and the observed

rapid diffusion must be flow induced.

40

4. Surface Analysis
4.1 . Introduction
What effect, then, might a HBP have on a polymer surface?
One important way of determining the effects on surfaces is through
advancing and receding contact angles. The advancing contact angle represents

the dynamic change of a drop of liquid on a surface” 3" 32

as the drop
“advances” on the surface. In other words, it is a measure of a surface’s
amphophobicity. The receding contact angle represents the dynamic change as
a droplet “recedes” on the surface. That is, the receding contact angle is a
measure of a surface’s amphophilicity.

Many different methods exist for finding these contact angles. Some of
the most common methods are the Wilhelmy Plate, Capillary Rise, and the
Sessile Drop.3°' 3" 32 The Wilhelmy method consists of inserting a fiber into a
liquid and measuring the angle that the liquid forms with the surface. The
Capillary Rise method involves inserting a thin tube into a liquid and measuring
the angle the meniscus form with the inside of the tube. The Sessile Drop
method is Simply the placement of a drop on a flat surface and measuring the
angle that the edge of the drop makes with the surface. Figure 4.1.1 and Figure

4.1.2, from page 253 of Hiemenz and Rajagopalan,30 show the Capillary Rise

and the Wilhelmy Plate methods.

41

 

 

 

. 0-
. Liquid

Figure 4.1.1: Diagram of the Wilhelmy Plate Technique3°

——L RC . ‘—

Figure 4.1.2: Diagram of the Capillary Rise Technique30
In this study, the Wilhelmy Plate method will be used. Equation [4.1]

 

 

 

Shows the relationship between the surface tension of a liquid (y), the fiber

diameter (d), the contact angle (61, and the wetting force for a Wilhelmy probe
(gAm).

2
pgd z

[4.1]

gAm = 7Idycos 6—

However, for a very thin fiber, the second term is approximately zero, since dis
very small, and therefore the term is negligible. Thus, Equation [4.2] reduces to
the simplified form of the Wilhelmy probe equation. Using a liquid with known

properties, thiS equation may be used to compute the contact angle.

42

gAm = 72217 cos 6? [421

The advancing contact angle is further useful in that it provides a measure
of the polymer surface properties. Over the last fifty years, Good33 and Zisman34
have both published excellent reviews tracing the history of contact angle usage
from Aristotle and Archimedes to Young and Dupré to their respective present
times (1992 and 1964). In addition, several good texts may be referenced on this
subject, including Wu,31 Heimenz and Flajagopalan,30 and Adamson and Gast.32
It is not the purpose of the present work to recount in detail the history or

derivation before Young’s Equation, shown as Equation [4.3], which relates the

energy of the liquid, vapor, and solid interfaces.

7Lv COS 6 = 75v + 73L [4'3]

Here, L, S, and Vrepresent the Liquid, Solid, and Vapor portions. Generally
speaking, though, one may use L, S, and Vto mean phases 1, 2, and 3.

The next step, proposed by Girifalco and Good, applied the Geometric
Mean Equation to Young’s Equation and yields the Girifalco-Good Equation,
shown as Equation [4.4]. Alternatively, the Harmonic Mean Equation is

sometimes applied to Young’s Equation; the result is shown as Equation [4.5].

43

(1+ COSgadquq = 2KW )+ ( W )1 [4.4]

<1+cosa...>y.. =4 ”W + ”‘47” ”-51

_ 7/11'1q + 7501 ”I: + 7.5:)!

 

 

 

 

Equation [4.4] will be used here as most of the advancing contact angles

will be greater than 90° and the Harmonic Mean Equation fails in those cases.

The main discrepancy is in how to apply this equation. Van Oss, et al.,

cited in Adamson and Gast,32

page 376, suggested using a term that the overall
surface tension was composed of three parts: the Liftshitz-Van der Waals
(dispersion) forces, and the electron donor/electron acceptor interactions for the

solid and liquid as well as the liquid and the solid. Equation [4.6] shows the

commonly accepted form, from Van Oss, et al.35

“”0086” = 2(x/73LW716W +\/7:7; Wyn/Z) [4.61

Another method was proposed by Fowkes.36' 37 Fowkes separated the
overall surface tension into dispersive and polar parts, as shown in Equation
[4.4]. Further, he proposed that for hydrocarbons only the dispersive form was

crucial, thus leading to the following form, labeled Equation [4.7].

(1.1-€086)?qu : 2 / 7627/3101] [4_7]

Both of these methods assume that the equilibrium spreading pressure is
approximately zero for a hydrocarbon. Note, too, that Equation [4.7] and
Equation [4.4] are identical if the solid acts as only a proton donor, and the liquid
acts only as a proton acceptor, or vice versa. Good38 stated that for most
hydrocarbons that the overall equation (Equation [4.4]) would hold, except for a
fluorinated hydrocarbon. As no fluorinated hydrocarbons were used in this study,
Equation [4.4] will be sufficient to compute the values of the surface energy.

4.2. Materials

Polymer materials used are the same as those listed in previous sections.
The methylene iodide was a product of Spectrum, Inc. The deionized (DI) water
was purchased from Culligan by the Composite and Materials Center at Michigan
State University. Table 4.2.1 shows the surface tension data for methylene
iodide and DI water at 20 “C from page 179 of Wu31 for both the Geometric Mean
and the Harmonic Mean Equations (Equations [4.4] and [4.5]).

Harmonic Equation Data

d p
I] 7: ll 1: II VI
Liquid (mN/m) (mN/m) (mN/m)
Methylene Iodide 43.5 6.6 50.1
DI Water 21.9 50.1 72.0
Geometric Equation Data

 

 

 

Y." 11” TI

Liquid " (mN/m " (mN/m) ll (mN/m)

Methylene Iodide 47.7 2.3 50.0
DI Water 21.6 50.4 72.0

Figure 4.2.1: Liquid Properties for the Geometric and Harmonic Mean Equations31

 

 

 

 

 

 

There is a fair amount of discussion as to what the actual values are for
methylene iodide;31 the ones used in these computations were measured using
the interfacial tension between water and methylene iodide. These values were

chosen in order to give allowance for any polarity in the systems studied.

45

4.3. Experimental Procedure

In their work, Sauer and DiPaoIo39 give a more extensive explanation of
the experimental method followed here. However, the steps will be outlined here
briefly. It is imperative to remember that this method is intended to obtain the
parameters needed to solve either Equation [4.1] or Equation [4.2]. This same
procedure was performed on both the virgin polymers as well as the polymer
blends.

The first step, naturally, is to create the probe by stretching the polymer
fiber as it is extruded. The fibers were extruded at the same conditions as the
fibers created for the TEM analysis (see Tables 3.5.1 and 3.5.2). Fibers were
extruded from the OP as well as the 30 mm capillary in order to be able to
compare and contrast the surface energy of the melt inside the barrel with the
melt exiting the capillary. The fibers were handled with gloves to prevent
contaminants from adhering to the surface.

In order to attach this fiber to the Cahn 322 Microbalance at the High
Sensitivity location, one end of the polymer fiber was bent into a “U” shape and
held briefly against the extruder barrel in order to melt the fiber into a hook. The
fiber was then trimmed with ceramic scissors as needed. All fibers were used
within a day of extrusion.

The second step was to obtain the probe diameter. According to Sauer

and DiPaolo,39 the fiber should be 200-500 pm. For large diameter fibers, a set

46

of microcalipers may be used to measure the diameter. Because these fibers
were relatively large, the calipers were used to measure the diameter.

Sauer and DiPaolo39 further state that even for fibers of 200-500 pm that
the buoyancy correction should be negligible and that Equation [4.2] may be
used. However, as these fiber diameters seem rather large, the buoyancy will
not be neglected, and Equation [4.1] will be used. Normally, measuring the
buoyancy force is difficult due to the uncertainty of the parameters involved,
primarily the depth of immersion. The microbalance used, though, allowed the
operator to fix the approximate depth of immersion, and thus, this value may be
reasonably evaluated.

Equation [4.1] was used, and through the computer’s autoanalyze function
together with the buoyancy correction option, the advancing and receding contact
angles were determined. The Cahn 322 microbalance was programmed to lower
the fiber into the liquid at a speed of 80 um/s. The fiber was inserted into the
liquid to a maximum depth of 1 mm, and then permitted to equilibrate there for
about 15 seconds. This maximum depth corresponds to z, the depth of
immersion. The fiber was then pulled out of the liquid at a speed of 80 um/s.
This same procedure was used throughout the experiment. Figure 4.3.1 shows a

cartoon of the experimental set-up.

47

 

 

 

 

 

 

 

 

Cahn 322 M
Microbalance -
Polymer V \

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fiber
Computer
”3:“
Motorized I N Beaker
Sta e e—
g Baffle

 

 

 

Figure 4.3.1: Microbalance Set-up

The computer was used to record and save the results. The fiber insertion
yielded the advancing contact angle, while the fiber retraction yielded the
receding contact angle. This experiment was repeated at least three times with
both methylene iodide and DI water for all polymer systems. The resulting
angles were then averaged. Figure 4.3.2 shows a sample plot generated from

the data for PS/H50 and water.

48

 

 

 

Am (Pa)

 

 

 

 

 

 

 

-2 :i 37.—Ir v TVHLE’ IT“ MEI VFxm 117"-“ ET F anE E" H ”F. [1' I'imrlm .11. [[73 [W
°8NQQVSQ'QQQTPEBSQSSEKR
'- '-
time(s)

Figure 4.3.2: Sample microbalance plot for PSIHSO and water

The advancing contact angle as well as the known surface tension data,
found in Table 4.2.1, was inserted into Equation [4.4]. This yielded two equations
with two unknowns, one equation for each liquid. These two equations were then
solved simultaneously to get the two parts of the surface energy. The total
surface energy was then simply the sum of these two parts.

4.4. Results

In addition to Sauer and DiPaolo,39 several other studiesa1'40’ 4‘ have been
devoted to using the polar-dispersive equation (Equation [4.4]) to compute the
surface tension. Table 4.4.1 and Table 4.4.2 show the contact angles measured
by the microbalance for the OP extrudates. The standard deviation was also

computed.

49

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

OP H20

eadv erec
Material (deg) (deg) Hysteresis (deg)
PS 99 1 3 85 14 1 3
PS/H50 5.0 wt% 95 1 9 88 7 1 9
LLDPE 97 1 5 83 14 1 5
LLDPE/H3200
5.0 wt% 91 11 74 18 1 1
LLDPE/H40 5.0 wt% 95 1 2 54 41 1 2

Table 4.4.1: H20 Contact Angles for the OP

OP CH- l2

9adv 9rec
Material (deg) (deg) Hysteresis (degL
PS 50 1 11 N/R N/R
PS/H50 5.0 wt% 47 1 4 N/R N/R
LLDPE 61 1 7 36 25 1 7
LLDPE/H3200
5.0 wt% 75 1 3 48 27 1 3
LLDPE/H40 5.0 wt% 52 1 8 28 24 1 8

 

 

Table 4.4.2: CH2I2 Contact Angles for the OP

The receding angles for PS and PS/H50 were not able to be measured.

The advancing angles were then inserted into Equation [4.4] and solved to

 

 

 

 

 

 

 

 

 

 

 

generate Table 4.4.3.

OP Polymer
Material 7,,“ (mN/m) 75° (mN/m) 75‘ (mN/m)

P8 35.0 1 5.3 3.9 1 3.2 38.9 1 3.8
PS/H50
5.0 wt% 36.5 1 2.0 6.5 1 6.4 42.9 1 6.3
LLDPE 28.6 1 3.5 8.0 1 3.8 36.6 1 3.9
LLDPE/H3200
5.0 wt% 20.5 1 1.4 16.1 1 1.1 36.6 11.2
LLDPE/H40
5.0 wt% 33.6 1 4.1 7.1 1 2.3 40.7 1 2.8

 

Table 4.4.3: Surface Energy Results for OP

50

 

Table 4.4.4 and Table 4.4.5 show the contact angles measured for the 30

 

 

 

 

 

 

 

mm capillary.

Capillary H20
eadv erec

Material (deg) (degL Hysteresis (deg)
PS 95 1 8 76 20 1 8
PS/H50 5.0
wt% 99 1 4 83 17 1 4
LLDPE 92 1 4 86 6 1 4
LLDPE/H3200
5.0 wt% 94 1 1 91 4 1 1
LLDPE/H40
5.0 wt% 90 1 3 31 59 1 3

 

 

 

 

 

 

Table 4.4.4: H20 Contact Angles for the 30 mm Capillary

 

 

 

 

 

 

 

 

 

 

Capillary CHz'g

eadv 9rec
Material (deg) (deg) Hysteresis (deg)
PS 54 1 5 N/R N/R
PS/H50 5.0
wt% 53 1 4 10 42 1 4
LLDPE 56 1 5 21 35 1 5
LLDPE/H3200
5.0wt°/o 7313 59 1413
LLDPE/H40
5.0 wt% 59 1 3 20 38 1 3

 

 

 

Table 4.4.5: CH-J, Contact Angles for the 30 mm Capillary
No receding contact angle was able to be measured for PS. Table 4.4.6

shows the results of the surface energy calculations for the 30 mm capillary.

51

 

 

 

 

 

 

 

Capillary Polymer

Material ygmN/m) y,” (mN/m) y; (m N/m)
PS 32.6 1 2.4 7.7 1 5.5 40.3 1 5.4
PS/HSO 5.0
wt% 33.6 1 1.8 4.0 1 2.9 37.6 1 2.9
LLDPE 31.7 1 2.5 10.9 1 3.3 42.6 1 3.4
LLDPE/H3200
5.0 wt% 21.4 1 1.6 13.2 1 0.8 34.6 1 1.0
LLDPE/H40
5.0 wt% 29.9 1 1.4 13.1 1 2.1 42.9 1 2.1

 

 

 

 

 

 

Table 4.4.6: Surface Energy Results for the 30 mm Capillary
4.5. Contact Angle Discussion

Wu,31 as well as others,4°' 4" 42' 43 have reported empirical literature values
for both the contact angles and the surface energy values. There are two slight
problems with this empirical work. First, in order to accurately measure the
surface properties, one needs the advancing contact angles, not the equilibrium
contact angles. Second, the empirical values apparently were measured using
the Sessile Drop Method, whereas the work done here was with the Wilhelmy
Method. Nevertheless, it is useful as a guide.

A brief attempt44 has been made to provide theoretical correlation to these
empirical values. Unfortunately, this attempt is highly idealized.

Volpe,43 tested the advancing contact angle as a function of the speed of
immersion of the Wilhelmy Plate. He found that the importance of the probe’s
insertion velocity depended greatly on the particular material being tested.

Table 4.5.1 shows a comparison between the equilibrium values used by

4
1,0

Dala the advancing angles also reported in Wu,31 the advancing angles

reported by Owens and Wendt,41 the advancing angles by Volpef’3 and the

52

Cont. Ang.: H20

theoretical values by El Ghzaoui.‘14 The numbers across the top represent these

reference numbers. Contact angles, if reported, are in degrees.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source By Reference Number)
Material 31 40 42 43 44
PS 91 91 91 96 N/A
LDPE 1 02 96 1 04 N/A N/A
Cont. Ang.: C_H;!2

Source (By Reference Number)
Material 31 40 42 43 44
PS N/A 35 35 N/A 35.9
LDPE N/A 53 53 N/A 52.8

 

 

 

Table 4.5.1: Compiled Reference Data for Adv. Contact Angles
Once again, note that the values reported here, except for Reference 43,

are for LDPE, not LLDPE, and were measured with the Sessile Drop Method
under controlled conditions. In addition, the actual value of water on LDPE is still

somewhat debatable. Fowkes37 cites several authors, including Tardros,

Adamson, et al.,45 who reported a value of 88° for the system in question.

According to Fowkes,37 the polymer studied by Tardros, et al."5 were not purified
of all impurities, as was the case in other studies. The values measured in this
work fitted within the boundaries cited.

As the fibers in this work were extruded and are industrial quality, it is
possible that a small amount of contaminant is present in the system.
Nevertheless, the contact angle values measured here for the base polymers are
still very close to these literature values, with the exception of PS/methylene

iodide (35° versus 54°), but this discrepancy could be explained by the presence

of contaminants.

53

According to Mackay and Carmezini,46 the advancing contact angle for
both water and methylene iodide on the HBPs studied (H30 and H3200), was
less than the angle measured for the two bulk polymers listed above. Thus, one
might expect that blending the HBP with a linear polymer would reduce the
advancing contact angle. This, however, is not the case. The value obtained
here for the water advancing contact angle on LLDPE/H3200 agrees very well

with previously published results.43 The value measured in this study was 94 1

1° for the capillary and 91 1 1° for the OP. .These values agree with Mackay and
Carmezini,46 who measured a value of 96° for a LLDPE/H3200 0.5 wt% film.

However, there is a difference in the measured advancing contact angles for

methylene iodide (OP: 75 1 3° and Capillary: 73 1 3° versus 57°, from Mackay

and Carmezini).46 The reason for this discrepancy is twofold. First, there is a
difference in method; i.e., a fiber versus a film. Second, the fibers measured
here have a slightly higher HBP weight percent. Shafrin and Zisman42 reported a

maximum contact angle of 77° for a methylene rich, single crystal hydrocarbon.

In this light, the values measured here make sense since LLDPE has a large
number of methyl groups and H3200 has almost pure carbon (90% C20 end
groups). Thus, the measured value is believable. It is interesting to note that
within the margin of error that there is no difference between the base polymer
and the other polymer blends.

Sendijarevic, et al.14 measured contact angles on their LLDPE films and
found that altering the end group changed the advancing contact angle. There

are two problems with their results. First, they used a manual goniometer on a

54

polymer film. These instruments are difficult to use because the change from
advancing contact angle to equilibrium contact angle is very fast, and no margin
of error is reported in their work. Second, extruding a polymer film is not easy.
Great care must be taken to keep the film flat and straight. In addition, surface
defects are much more common with polymer films than with polymer fibers.
Attempts were made to measure contact angles using tapes as was done by
Sendijarevic, et al.,14 but were abandoned in favor of Sauer’s and DiPaolo’s
method39 that utilizes fibers and the Wilhelmy method due to the above stated
problems.
4.6. Surface Energy Discussion

Despite any differences in the contact angle measurements, the surface
energy values for the base polymers agree with each other within the stated
margin of error. Table 4.6.1 lists literature values for the components of the

'40

surface energy from Dala for methylene iodide and water. Remember, these

values were computed using equilibrium contact angles.

 

d p t

YB 76 75
Material (mN/m) (mN/m) (mN/m)
PS 42.0 0.5 42.5
LDPE 33.7 0.0 33.7

Table 4.6.1: Literature Surface Data“

 

 

 

 

 

 

 

 

Now, comparing those results with these from page 184 of Wu31 for LDPE,
ys‘ = 35.7 mN/m, and for PS, 75‘ = 40.7 mN/m, it is obvious that the difference in
angle is important, but not critical. It is also obvious that the experimental results
for the base polymers easily fall between the two boundaries within their

respective margins of error. Thus, these results are acceptable.

55

The next obvious question is: how good are the values for the blends?
Mackay and Carmezini46 have published some results for a 0.5 wt%
LLDPE/H3200 blend using Fowkes’ approximation (Equation [4.7]). Despite the
fact that they concentrated only on the dispersive part, their results agree
reasonably well with those reported in this work. Thus, the results for the blends
are also reasonable.

The next question is: what trends exist in the data? Previously in this
work, it was hypothesized that the capillary would affect the phase separation,
and it does. The question then asked pertained to how the capillary affected the
surface properties. The answer, of course, is that these two questions are one
and the same. Looking at the data, it is readily apparent that there is no
significant overall effect of the HBP on the surface energy of the blend within the
margins of error. In addition, within the margins of error, there is no difference
between the OP and the capillary results for any of the systems studied.

However, it is interesting to note that there are some changes in the
makeup of the overall tension. Mackay and Carmezini46 reached the same
conclusion in their brief study. The main conclusion to be drawn from these
results is that adding a HBP does not significantly alter the surface energy for
either the OP or the capillary. This fits since similar results were observed in the

TEM images.

56

5. Overall Conclusions

What, then, has been learned from this work? First, adding a HBP to a
linear polymer does cause a reduction in the shear stress for a given shear rate.
This reduction indicates that the HBP does act as a processing aid, provided it
phase separates at the surface.

Second, TEM and DSC tests show that the polymers do phase separate.
While the TEM images do not show any HBP directly on the surface of the OP or
capillary fibers, they do show a higher concentration of the HBP in the edge
layers than in the bulk region. Some possible reasons for the absence of the
HBP directly at the interface are: 1. they were lost during the microtoming
process, 2. the concentration in the edge region is sufficient to cause the
observed flow increase, 3. the HBP layer was left attached to the interior of the
capillary, as shown by Barone, et al.47 for fluorinated LLDPE with Confocal
Microscopy and Duschene, et al.,48 for a fluoroelastomer/Polyethylene
GlycoVLLDPE blend, 4. the HBP acts as a “ball-bearing”,18 or 5. some
combination of the above. Whatever the true reason, the phenomenon exists
and it is general for an immiscible linear polymer/HBP system. An immiscible
system will consist of two highly dissimilar molecules. This dissimilarity may
arise from chemical constituents or structure. The phase separation is
independent of temperature, concentration, and capillary use. The rapid diffusion
of the HBP to the surface layers is driven by the flow.

Third, adding a HBP leaves the overall surface energy relatively unaltered,

which may be an attractive feature. Optical microscopy shows that both LLDPE

57

blends and PS blends have smoother surfaces than the virgin polymers’
sudaces.

Therefore, adding an HBP to a linear polymer is doubly useful in that it
acts as a processing aid while leaving the overall surface energy unchanged and
providing a potentially useful surface for a wide range of applications.

6. Suggested Future Work

Examine the surface effects of using a fluorinated polymer. Perform
Scanning Electron Microscopy (SEM) or Atomic Force Microscopy (AFM) and
measure the surface roughness. Dissolve the HBP out in solution and observe

the effects. Test these blends in a twin screw extruder and observe the effects.

58

 

Works Cited

. Kissin, Y.V. “Principles of Polymerizations with Ziegler-Nutta Catalysts”.
Handbook of Polymer Science: Vol. 1: Synthesis and Properties. Marcel
Dekker: New York, 1989.

. Brandrup, J.; Immergut, E.H., Eds. Polymer Handbook: 3'" Ed. John
Wiley and Sons: New York, 1989.

. Newkome, G.R.; Moorefield, C.N. Dendritic Molecules: Concepts
Synthesis Perspective, 15’ Ed. VCH Publishers, Inc. : New York, 1996.

. Tomalia, D.A.; Baker, H.; Dewald, J.; Hall, M.; Kallos, G.; Martin, S.;
Roeck, J.; Ryder, J.; Smith, P. “Dendritic Molecules: Synthesis of
Starburst Dendrimers”. Macromolecules 1986, 19, 2466.

. Hawker, C..J.; Fréchet, J.M.J. “Preparation of Polymers With Controlled
Molecular Architecture. A New Convergent Approach to Dendritic
Macromolecules”. Journal of the American Chemical Society 1990, 112,
7638.

. Mourey, T.H.; Turner, S.R.; Rubinstein, M.; Fréchet, J.M.J.; Hawker, C..J.;
Wooley, K.L. “Unique Behavior of Dendritic Molecules: Intrinsic Viscosity
of Polyester Dendrimers”. Macromolecules 1992, 25, 2401.

. Fréchet, J.M.J.; Hawker, C.J.; Gitsov, l.; Leon, .J.W. “Dendrimers and
Hyperbranched Polymers: Two Families of Three-Dimensional
Macromolecules With Similar But Clearly Distinct Properties”. Journal of
Macromolecular Science 1996, A33, 1399.

. Mackay, M.E.; Henson, DJ. "The Effect of Molecular Mass and
Temperature on the Slip of Polystyrene Melts at Low Stress Levels”.
Journal of Rheology 1998, 42, 1505.

. Hsieh, T.-T. ;Tiu, C.; Simon, G.P. “Melt Rheology of Aliphatic
Hyperbranched Polyesters With Various Molecular Weights”. Polymer
2001, 42, 1931.

10. Carmezini, G. (2001) MS Thesis. Stevens Institute of Technology.

11.Hong, Y.; Cooper-White, J.J.; Mackay, M.E; Hawker, C.J.; Malmstrém, E.;

Rehnberg, R. “A Novel Processing Aid for Polymer Extrusion: Rheology
and Processing of Polyethylene and Polymer Blends”. Joumal of
Rheology 1999, 43(3), 781 .

59

12.Hong, Y.; Coombs, S.J.; Cooper-White, J.J.; Mackay, M.E.; Hawker, C.J.;
Malmstrbm, E.; Rehnberg, R. “Film Blowing of Linear Low Density
Polyethylene Blended With a Novel Processing Aid”. Polymer 2000, 41,
7705.

13.Kil, S.B.; Augros, Y.; Leterrier, Y.; Manson, J.A.E. “Rheological Properties
of Hyperbranched Polymer/Poly(Ethylene Terphthalate) Reactive Blends”.
Polymer Engineering and Science 2003, 329.

14.Sendijarevic, l.; McHugh, A..J. “The Influence of Alkyl End Groups on the
Miscibility of Hyperbranched Polymers with Polyolefins”. Polymer
Engineering and Science 2002, 42, 2393.

15. Mulkem, T.J.; Tan, N.C.B. “Processing and Characterization of Reactive
Polystyrene/Hyperbranched Polymer Blends”. Polymer 2000, 41, 3193.

16.Mezzenga, R.; Luciani, A.; Manson, J.A.E. “Phase Separation and
Gelation of Epoxy Resin/Hyperbranched Polymer Blends”. Polymer
Engineering and Science 2002, 42, 249.

17.Zelinger, .J.; Volfova, E.; Zahradnfkova, H. “PVC Mixtures With
Processing Aids”. lntemational Joumal of Polyermic Materials 1976, 5,
99.

18. Kanu, R.C.; Shaw, M.T. “Rheology of Polymer Blends: Simultaneous
Slippage and Entrance Pressure Loss in the Ethylene-Propylene-Diene
(E PDM)Niton System”. Polymer Engineering and Science 1982, 22, 507.

19.Akay, G. “Unstable Capillary Flow of Reinforced Polymer Melts”. Joumal
of Non-Newtonian Fluid Mechanics 1 983, 13, 309.

20. Kim, Y.H.; Beckerbauer, R. “Role of End Groups on the Glass Transition
of Hyperbranched Polyphenylene and Triphenylene Derivatives”.
Macromolecules 1 994, 27, 1968.

21 . Englund, J.E. (2000) MS Thesis. Royal Institute of Technology.

22.Malstr6m, E.; Johansson, M.; Hult, A. “Hyperbranched Aliphatic
Polyesters”. Macromolecules 1995, 28, 1698.

23. Massa, D.J.; Shriner, K.A.; Turner, S.R.; Voit, B.I. “Novel Blends of
Hyperbranched Polyesters and Linear Polymers”. Macromolecules 1995,
28, 3214.

24. Kim, Y.H., Webster, O.W. “Hyperbranched Polyphenylenes”.
Macromolecules 1992, 25, 5561.

60

25. Flegler, S.L.; Heckman, Jr.; J.W., Klomparens, K.L. Scanning and
Transmission Electron Microscopy: An Introduction. Oxford University
Press: New York, 1993.

26.Mackay, M.E.; Hong, Y. ; Jeong, M.; Tande, B.M.; Wagner, N.J.; Hong, S.;
Gido, S.P.; Vestberg, R.; Hawker, C.J. “Self Assembly of Hybrid
Dendrimer-Linear Block Copolymers into Ordered Structures”.
Macromolecules 2002, 35, 8391.

27. Roman, C.; Fischer, H.R.; Meijer, E.W. “Microphase Separation of
Diblock Copolymers Consisting of Polystyrene and Acid-Functionalized
Poly(propylene imine) Dendrimers”. Macromolecules 1999, 32, 5525.

28. Lee, H.; Archer, L.A. “Functionalizing Polymer Surfaces By Field Induced
Migration of Copolymer Additives. 1. Role of Surface Energy Gradients.”
Macromolecules 2001, 34, 4572.

29.Bauer, B.J.; Ramzi, A.; Liu, D.W.; Scherrenberg, R.L.; Froehling, P.;
Joosten, J. “Blends of Fatty-Acid-Modified Dendrimers with Polyolefins”.
Journal of Polymer Science: Part B: Polymer Physics 2000, 38, 95.

30. Hiemenz, P.C. ; Rajagopalan, R. Principles of Colloid and Surface
Chemistry. Dekker: New York, 1997.

31 .Wu, 8. Polymer Interface and Adhesion. Dekker: New York, 1982.

32.Adamson, A.W.; Gast, A.P. Physical Chemistry of Surfaces: 6‘" Ed. John
Wiley and Sons: New York, 1997.

33.Good, R.J. “Contact Angle, Wetting, and Adhesion: a Critical Review”.
Journal of Adhesion Science and Technology 1992, 6, 1269.

34.Zisman, W.A. “Relation of the Equilibrium Contact Angle to Liquid and
Solid Constitution”. Advances in Chemistry Series 1964, 43, 1.

35.Van Oss, C.J.; Giese, R.F.; Li, 2.; Murphy, K.; Norris, J.; Chaudhury, M.K.;
Good, R.J. “Determination of Contact Angles and Pore Sizes of Porous
Media by Column and Thin Layer Wicking”. Journal of Adhesion Science
and Technology1992, 6, 413.

36. Fowkes, F.M.; McCarthy, D.C; Mostafa, MA. “Contact Angles and the

Equilibrium Spreading Pressures of Liquids on Amohophobic Solids”.
Journal of Colloid and Interface Science 1980, 78, 200.

61

37. Fowkes, F.M. “Dispersion Force Contributions to Surface and lnterfacial
Tensions, Contact Angles, and Heats of lmmersion”. Advances in
Chemistry Series 1964, 43, 99.

38.Good, R.J. “Theory for the Estimation of Surface and lnterfacial
Energies”. Advances in Chemistry Series 1964, 43, 74.

39.Sauer, B.B.; DiPaolo, N.V. “Surface Tension and Dynamic Wetting of
Polymers Using the Wilhelmy Method: Applications to High Molecular
Weights and Elevated Temperatures”. Journal of Colloid and Interface
Science 1991, 144, 527.

40. Dalal, E.N. “Calculation of Solid Surface Tensions”. Langmuir1987, 3,
1009.

41 .Owens, D.K.; Wendt, R.C. “Estimation of the Surface Energy of
Polymers”. Joumal of Applied Polymer Science 1969, 13, 1741.

42. Shafrin, E.G.; Zisman, W.A. “Upper Limits to Contact Angles on Solids”.
Advances in Chemistry Series 1964, 43, 145.

43. Volpe, C.D.; Cassinelli, C.; Morra, M. “Wihelmy Plate Measurements on
Poly(N-isopropylacrylimide)-Grafted Surfaces”. Langmuir1998, 14, 4650.

44. EI-Ghzaoui, A. “Determination of Contact Angles: Consistency Between
Experiment and Theory”. Journal of Applied Physics 1 999, 86, 2920.

45.Tardros, M.E.; Hu, P.; Adamson, A.W. “Adsorption and Contact Angle
Studies: I. Water on Smooth Carbon, Linear Polyethylene, and Stearic
Acid-Coated Copper". Journal of Colloid and Interface Science 1974, 49,
605.

46. Mackay, M.E.; Carmezini, G. “On the Surface Properties of
Hyperbranched Polymers”. Langmuir 2001, 17, 1708.

47. Barone, J.R.; Wang, S.Q.; Farinha, J.P.S; Winnik, M.A. “Polyethylene
Melt Adsorption and Desorption During Flow on High-Energy Surfaces:
Characterization of Post-Extrusion Die Wall by Laser Scanning Confocal
Fluorescence Microscopy". Langmuir 2000, 16, 7038.

48.Duschene, D.; Blacklock, J.E.; Johnson, B.V.; Blong, T.J. “Improved
Processability of Linear Low Density Polyethylenes Through the Use of
Fluorocarbon Elastomers/Polyethylene Glycol Blends”. In Society of
Plastic Engineers (US) Technical Conference (47”7 ANTEC 89): Plastics
Create a World of Difference. ANTEC: New York, 1989, pp. 1343-1347.

62

AAAAAAAAA

' IIIIIIIIIIIIIIIII‘IIIIIIIII III