. . I .
?.£.fi. 43.4w . .

V J .Tfifiwb Hunkymu
nh..s$.L...fl.:: “Wtwufi
. «MM .- g, 9

, . {.3

V . Au...
NW“ 7...

WW: .

A. Fun“.
3.”. V? fir...
( .
{VI
. V . .4 LL. V
Java. :2... flaw... R... V
x I .

. ,l A

., V V:
, in 8.3M:
.: 7.: 12:»...- .5.
a: ‘ 53.5 iii} I
2.3.. .. V...
. .11..

deE
H... , 13.35)}
I?

m?

V t3
(A? ma;
Y ~‘u s}

3464 .
z

‘3 ,3
£135}-

 

 

; g
.1... .dvfiafih.
nmmvfiwififi.

a) 21.

 

.1. 1.
61:3.1.
V a

1:! .-

 

 

:1...
a. 1 :qun

.731;

 

UP... .91.- .
cl. 1)! .v 1.11.

n 3;.an

.1111
.015... .
It...)- s 5.0

has

‘0‘)!
K 5:... I! 5...... 59.79.!

7.5: ..i$:.:

(3.33711. '.

 

 

’ .....-...__

LIBRARY
Michigan State
University

 

 

 

This is to certify that the
thesis entitled
FLOW LINE FORMATION OF INJECTION
MOLDED POLYPROPYLENE/BF RUBBER BLENDS
presented by

Paul Andrew Papworth

has been accepted towards fulfillment
of the requirements for

degree in Chemical Engineering

 

Date €‘(3t/O!

0-7 639 MS U 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

 

(‘ERLJZL 10??“

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 chIRC/DatoDuopfi-nts

 

FLOW LINE FORMATION OF INJECTION MOLDED POLYPROPYLENE/BF
RUBBER BLENDS
By

Paul Andrew Papworth

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

Department of Chemical Engineering

2001

ABSTRACT
FLOW LINE FORMATION OF INJECTION MOLDED POLYPROPYLENE/BF
RUBBER BLENDS
By

Paul Andrew Papworth

Flow line formation (typically referred to as tigerstriping) is a common manufacturing

problem of injection molded materials. This study investigates the morphology of two

“A” Lpolymflontaining polypropylene and one of two different grades of ethylene

-\‘~

propylene (EP) rubber. Image analysis of SEM rrricrographs from short shots taken at the
flow mark surface, out-of-flow mark surface, and cross-sectional core behind the flow
front regions reveals EP rubber particles with various degrees of stretching from the core
to the mold wall regions. From a detailed analysis of the micrographs, a mechanism for
flow line formation is proposed. As the mold is filled, an asymmetric melt front forms
and the melt head strays from the centerline creating a long and short span. If the EP
rubber particles flow along the short span, they simultaneously coalesce, become
stretched to their maximum, and become frozen at the mold wall to form the out-of—flow
mark region. If the particles flow along the long span, there is enough time for them to
simultaneously coalesce and stretch to their maximum, and then.partially retract. As the
partially retracted particles reach the mold wall, they freeze to form a flow mark region.
Both materials studied exhibited evidence of particle stretching, coalescing, and
retracting in the flow mark regions. A 25% reduction of the interfacial area of the

particles from the out-of-flow mark to flow mark regions was observed in both materials.

In Loving Memory of My Father, Charles Richard Papworth II
Who Did Not Live Long Enough to See His Son Graduate

iii

ACKNOWLEDGMENTS

This research was supported by Montell Polyolefins and I am please to acknowledge Dr.
Chi Chang Shu for his assistance in identifying materials and providing the SEM

micrographs for study.

I’d also like to thank Dr K. Jayraman for his insight and guidance to me during this
project. I’ve learned more from my conversations with him about engineering than I ever

have from a book.

Lastly, I would like to thank my mother, father, and three brothers Skip, Kyle, and Drew
for their support and encouragement over the years. I would also like to thank Guangda,
Darrel, Sang-Yoon, Jason, and Jon for making grad school fun. I’d especially like to
thank my two friends Frank and Kevin for helping me blow off some steam when I

needed too.

iv

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... vi
LIST OF FIGURES ........................................................................................................ vii
LIST OF SYMBOLS ........................................................................................................ ix
CHAPTER 1
INTRODUCTION ............................................................................................................. l
1.1 Flow Lines ....................................................................................................... 2
1.2 Previous Work ................................................................................................ 2
1.3 Proposed Mechanism ..................................................................................... 4
1.4 Problem Statement ......................................................................................... 6
CHAPTER 2
MORPHOLOGY OF POLYPROPYLENEIEP RUBBER BLENDS .......................... 7
2.1 Material Properties ........................................................................................ 7
2.2 SEM Analysis ....... 2' ............................. . ............................................................ 9
2.3 Results of TPO #1 Image Analysis .............................................................. 15
2.3.1 TPO #1 Particle Shapes ................................................................ 25
2.3.2 F low Mark vs. Out-of-flow Mark Comparison .............................. 32
2.3.3 Core Flow Front vs. Mold Wall Comparison ............................... 37
2.4 Results of TPO #2 Image Analysis .............................................................. 39
2.4.1 TPO #2 Particle Shapes ................................................................ 48
2.4.2 F low Mark vs. Out-of-flow Mark Comparison .............................. 55
2.4.3 Core Flow Front vs. Mold Wall Comparison ............................... 59
CHAPTER 3
CONCLUSIONS AND RECOMMENDATIONS ........................................................ 63
3.1 Conclusions .................................................................................................. 63
3.2 Recommendations for Further Work ........................................................ 65
APPENDICES
APPENDIX IMAGE ANALYSIS ................................................................................. 68
1. Determination of Aspect Ratio ...................................................................... 68
2. Determination of x, y, and 2 Dimensions ...................................................... 69
BIBLIOGRAPHY ........................................................................................................... 71

LIST OF TABLES

Table 2.1 Material properties of TPO #1 and #2 blends .................................................... 7
Table 2.2 Out-of-flow mark region comparison of particles less than 5 m2 vs.

greater than 5 m2. ............................................................................................................ 17
Table 2.3 Flow mark region comparison of particles less than 5 pm2 vs. greater

than 5 umz ......................................................................................................................... 21
Table 2.4 Measured values and calculated statistics for particles in TPO #1 .................. 28
Table 2.5 Geometric equations for various particle shapes ............................................. 29
Table 2.6 Variability of x, y, and z dimension lengths for TPO #1 ................................. 31
Table 2.7 Number average for the out-of-flow mark and flow mark regions ................. 32
Table 2.8 Average particle dimensions for the core flow front region ............................ 37

Table 2.9 TPO #2 Out-of-flow mark region comparison of particles less than

7 m2 vs. greater than 7 mm2 ............................................................................................. 40
Table 2.10 TPO #2 Flow mark region comparison of particles less than 7 umz vs.

greater than 7 umz. ............................................................................................................ 4-4
Table 2.11 Measured values and calculated statistics for particles in TPO #2 ................ 51
Table 2.12 Variability of x, y, and z dimension lengths for TPO #2 ............................... 54
Table 2.13 Number averaged statistics for the out-of-flow mark and flow mark

regions .......................................................................................................... . ..................... 55
Table 2.14 Average particle dimensions for the core flow front region .......................... 60
Table 3.1 Effects of changing the molding conditions on flow line formation ............... 65

vi

LIST OF FIGURES

Figure 1.1 Flow lines on the face and reverse sides of a rectangular mold ...................... 2

Figure 1.2 Cross-sectional melt front deformation showing the formation of the long and

short span ............................................................................................................................. 4
Figure 2.1 Tensile bar dimensions ..................................................................................... 8
Figure 2.2 Flow line patterns for (a) TPO #1 and (b) TPO #2 .......................................... 8
Figure 2.3 Micrograph locations. ...................................................................................... 9
1.0ut-of-flow mark on the plaque surface (in the xy plane) .......................... 9
2. Flow mark on the plaque surface (in the xy plane) ..................................... 9
3. Cross-section at the wall underneath the flow mark (in the xz plane) ........ 9

4. Cross-section at the wall underneath the out-of-flow mark (in the xz
plane) ........................................................................................................... 9
5. Cross-section behind the flow front (in the xz plane) ................................. 9
Figure 2.4 Out-of-flow mark surface morphology for TPO #1 ....................................... 10
(a) 0.44” short shot ................................................................................. 10
(b) 0.52” short shot ................................................................................. 10
(c) 0.58” short ........................................................................................ 10
Figure 2.5 Flow mark surface morphology for TPO #1 .................................................. 11
(a) 0.44” short shot ................................................................................. 11
(b) 0.52” short shot .................................................................................. 11
(c) 0.58” short ........................................................................................ 11
Figure 2.6 Cross sectional core flow front morphology for TPO #1 ............................... 12
(a) 0.44” short shot ................................................................................. 12
(b) 0.52” short shot .................................................................................. 12
(c) 0.58” short ........................................................................................ 12
Figure 2.7 Out-of-flow mark surface morphology for TPO #2 ....................................... 13
(a) 0.50” short shot ................................................................................. 13
(b) 0.53” short shot ................................................................................. 13
Figure 2.8 Flow mark surface morphology for TPO #2 .................................................. 13
(a) 0.50” short shot ................................................................................. 13
(b) 0.53” short shot .................................................................................. 13
Figure 2.9 Cross sectional core flow front morphology for TPO #1 ............................... 14
(a) 0.50” short shot ................................................................................. 14

vii

(b) 0.53” short shot .................................................................................. 14

Figure 2.10 (a) Out-of-flow mark surface micrograph .................................................. 16
(b) Detail of individual particles in the out-of-flow mark surface region. 16
(0) Flow mark surface micrograph ............................................................ 16
(d) Detail of individual particles in the flow mark surface region ............ 16
Figure 2.11 Aspect ratio distributions for the out-of-flow mark region of TPO #1 ........ 18
Figure 2.12 TPO #1 out-of-flow mark region particle lengths in the x and y directions
(measured from Figure 2.2(a)) .......................................................................................... 18
Figure 2.13 Cross-sectional morphology at the mold wall in the x, z plane for
TPO #1 ........................................................................................................ 19
Figure 2.14 Aspect ratio distributions for the flow mark region of TPO #1 ................... 22
Figure 2.15 TPO #1 flow mark region particle lengths in the x, y, and 2 dimensions
(measured from Figures 2.2(c) and 2.5(b)) ....................................................................... 22
Figure 2.16 Particle orientation distribution for the flow mark region of TPO #1 ......... 23
Figure 2.17 (a) Cross-sectional morphology at the core flow front for TPO #1 ............. 24
(b) Detail of individual particles in the core flow front region ................... 24
Figure 2.18 Aspect ratio distribution for the core flow front region of TPO #1 ............. 24
Figure 2.19 TPO #1 core flow front region particle lengths in the x and 2 dimensions
(measured from Figure 2.8). .............................................................................................. 25
Figure 2.20 Circular cylinder particle in the out-of-flow mark region ............................ 26
Figure 2.21 Elliptical cylinder particle in the flow mark region. .................................... 26
Figure 2.22 Ellipsoid particle in the core flow region ..................................................... 27
Figure 2.23 Particle size area distribution for TPO #1 .................................................... 29
Figure 2.24 Particle size distribution for TPO #1 ........................................................... 30
Figure 2.25 Aspect ratio distributions for TPO #1 .......................................................... 30
Figure 2.26 (a) Out-of-flow mark surface micrograph .................................................... 41
(b) Detail of individual particles in the out-of-flow mark surface region 41
(c) Flow mark surface micrograph .............................................................. 41

viii

(d) Detail of individual particles in the flow mark surface region ............... 41
Figure 2.27 Aspect ratio distributions for the out-of-flow mark region of TPO #2 ........ 42

Figure 2.28 TPO #2 Out-of-flow mark region particle lengths in the x, y, and
dimensions (measured from Figures 2.17(a) and 2.20(a)) ................................................ 42

Figure 2.29 Cross-sectional morphology at the mold wall in the x, 2 plane for

TPO #1 .............................................................................................................................. 43
Figure 2.30 Aspect ratio distributions for the flow mark region of TPO #2 ................... 45
Figure 2.31 TPO #2 flow mark region particle lengths in the x, y, and 2 directions
(measured from Figures 217(0) and 2.20(b)) ................................................................... 45
Figure 2.32 Flow mark particle orientation distribution for TPO #2 .............................. 46
Figure 2.33 (a) Cross-sectional morphology at the core flow front for TPO #2 ............. 47
(b) Detail of individual particles in the core flow front region .................... 47
Figure 2.34 Aspect ratio distribution for the core flow front region of TPO #2 ............. 47
Figure 2.35 TPO #2 core flow front region particle lengths in the x and 2 directions
(measured from Figure 2.33) ............................................................................................. 48
Figure 2.36 Elliptical cylinder particle in the out-of-flow mark region .......................... 49
Figure 2.37 Elliptical cylinder particle in the flow mark region ..................................... 49
Figure 2.38 Ellipsoid particle in the core flow region ..................................................... 49
Figure 2.39 Particle size area distributions for TPO #2 .................................................. 52
Figure 2.40 Particle size distributions for TPO #2 .......................................................... 52
Figure 2.41 Aspect ratio distributions for TPO #2 .......................................................... 53

ix

93‘“

TOTAL

are?

6964*

NFM

N 00m

PP

SEM
TEM
TPO

VT

XA WA

XNA

YA WA

YNA

ZA WA

ZNA

LIST OF SYMBOLS

Minor axis

Interfacial area

Total interfacial area

Particle area

Projected area in the xy plane
Projected area in the xz plane
Projected area in the yz plane
Major axis

Average length of b

y direction length

Diameter

Ethylene propylene

x direcion length

2 direction length

Molecular weight

Total number of particles in the Core Flow Front Region
Total number of particles in the Flow Mark Region

Total number of particles in the Out-Of-Flow Mark Region
Pi

Polypropylene

Radius

Reaction injection molded

Scanning electron microscope

Transmission electron microscope

Thermoplastic olepfrns

Total particle volume

Flow direction

x direction area weighted average length

x direction number averaged length
Tranverse flow direction

x direction area weighted average length
y direction number averaged length

Gap direction

x direction area weighted average length

2 direction number averaged length

CHAPTER 1

INTRODUCTION

With the increasing use of exotic plastics, polymer blends, and polymer composites in the
automotive manufacturing process, a common problem is the formation of surface defects
on injection—molded parts. Polymers used to make bumpers, specific body parts, and
interior components need a class A finish (within plain sight of the driver) and therefore
they must be free of any aesthetic defects. Such defects on injection-molded polymer
parts can take many different forms but this study will focus on flow lines (alternating
bands of gloss and dullness along the parts surface, commonly referred to as “tiger-
striping”) in thermoplastic olefins (TPO’s). Flow lines (transverse to the main flow) are
frequently formed on the surface of injection-molded thermoplastics. This effect can be
observed using neat polypropylene, if the injection molding conditions are pushed far
enough, but the problem is much more common when the injected material is a polymer
blend, a TPO, or a filled polymer (Han et al., 1996; Hobbs, 1996). The material studied
is a blend of polypropylene and ethylene propylene (EP) rubber that combines the
stiffness and structural characteristics of the polypropylene with the flexibility and
toughness of EP rubber. These TPO blends are replacing reaction injection molded
(RIM) polyurethanes because they provide better impact strength and paint durability

properties on the final molded parts.

1.1 Flow Lines

Flow lines are a common problem in the manufacturing of injection molded
thermoplastics. The surface has repeating alternating bands of gloss; the good area has a
high gloss finish while the flow lines have lower gloss and a dull appearance (Hamada,
1996). Comparing the face side and the back side of a rectangular molded part, the band
pattern is phase shifted 180 degrees, i.e., when one side has a high gloss area, the

opposite side has low gloss, even with symmetric mold cavities (Figure 1.1).

Flow Direction ——>

 

 

 

Face side

 

 

 

 

 

 

Back side

 

 

 

 

 

 

I
‘ 4,
C .t‘ Y
. _ ’0 ,
i I I ' ‘.
_ f7
3
4
. ’“ 3' -‘
s ‘i .1 7
O 1 _
- .

Flow line

Figure 1.1 Flow lines on the face and reverse sides of a rectangular mold.

1.2 Previous Work

The TPO’s examined were a two-phase immiscible blend of polypropylene and ethylene
propylene (EP) rubber. The EP rubber forms a disperse phase of deformable particles
suspended in the polypropylene matrix. The morphological properties of these blends

have been studied in order to predict a mechanism for flow line formation. It is believed

that as the mold is filled, the flow becomes unstable from the typical fountain flow
pattern, (Himont; 1990) and this unstable flow pattern leads to the formation of flow lines
(Chang, 1994; Hamada and Tsunasawa, 1996; Hobbs 1996). Flow near the advancing
front is affected by the presence of particles or drops which may deform the melt/air
interface significantly under some conditions, leading to surface bumps (Hoffman, 1985;
Stoos and Leal, 1989). These bumps, in turn, give rise to asymmetric instabilities in the
melt front, which leads to flow line formation. Himont (1990) observed and recorded the

melt front deformations using a high-speed camera.

For emulsions containing non-rigid, deformable particles or drops, the morphology can
be complicated further by the phenomena of drop break-up, coalescence, and stretching.
Bousmina and Muller (1998) studied the rheology and morphology of
polymethylmethacrylate rubber blends and found that the degree of stretch of the rubber
drops varied going from the center to the wall region. Flow plane micrographs of the
extrudate at various positions from the center of the circular conduit were obtained using
a transmission electron microscope (TEM). The particles were found to become more
stretched and aligned in the direction of the flow when going from the center to the wall.
The alignment in the skin region of the extrudate was found to result in a decrease in both
viscosity and post-extrusion swell of the blends. Bousmina and Muller did not, however,

report any radial concentration variations (i.e., no particle migration).

1.3 Proposed Mechanism

As the mold is filled, the melt front starts to deform and the stagnation point begins to
oscillate between the two mold walls (Yokoi 1994). The effect of this phenomenon on
flow lines is much more noticeable with filled polymers (Hobbs 1996) and polymer
blends (Hamada 1996; Han 1996). There are many possible reasons why the melt front
deforms in this manner, but that is not the subject of this research. The focus of this
study is the resulting morphology differences with polymer blends. The stagnation point
drifts away from the centerline as the melt front deforms and approaches one of the mold
walls (the top wall will be considered for illustrative purposes). During the oscillation, a

long span and short span are created that the EP rubber particles can follow (Figure 1.2).

l‘ Short Span —>

 

 

 

   

Center Line

   

 

 

 

 

 

Long Span

Figure 1.2 Cross-sectional melt front deformation showing the formation
of the long and short span.

There are two different cases that can be considered as the particles flow from the melt
front to the wall. The first case is that the flow mark region is formed after the particles
flow along the short span and an out-of-flow mark region is formed after they flow along
the long span. As the EP rubber particles flow along the short span, they undergo biaxial
elongational flow. The ellipsoidal core particles are stretched in the x and y directions to
form partially stretched drop structures. These incompletely stretched particles then
freeze at the mold wall in a random orientation to form a flow mark region. As the
particles flow along the long span, they have enough time to fully stretch and align
themselves parallel to the flow direction. These highly stretched and oriented particles

then freeze at the mold wall to form an out-of-flow mark region.

The second case is that the out-of-flow mark region is formed after the particles flow
along the short span and the flow mark region is formed after the particles flow along the
long span. If an EP rubber particle follows the short span path, it will stretch to its
maximum, align itself parallel the to the flow field, and an out-of-flow mark region will
form as the particles freeze at the mold wall. If the particles flow along the long span, the
timescale that it takes to flow from the stagnation point to the mold wall is much greater
than the timescale for particles flowing along the short span. With this increased
timescale, three possible scenarios can occur that form flow lines. Regardless of which
scenario occurs, the flow mark particles will be highly oriented in the flow direction. The
scenarios are:

0 The particles become overstretched, start to break up and tailing occurs. As they

reach the bottom wall, the small broken particles freeze to form a flow line.

0 The particles are stretched to their maximum, begin to relax, partially retract back,
and form partially retracted drop structures that are frozen at the mold walls to
form flow lines.

0 The particles are stretched to their maximum, relax, fully retract back, and form

fully retracted drop structures that are frozen at the mold walls to form flow lines.

1.4 Problem Statement

The focus of this research is to take a detailed look at the dispersed phase morphology of
the EP rubber filler of two TPO blends and from the particle sizes, arrive at a conclusion
on the dominant mechanism for changes in the particle morphology from the core to the

different regions of the surface.

CHAPTER 2

MORPHOLOGY OF POLYPROPYLENEIEP RUBBER BLENDS

2.1 Material Properties

Both TPO blends were compounded in a reactor. Therefore, not all of the individual
component material properties are known. Table 2.1 does list the weight percentage of
EP rubber in each blend along with the molecular weights of the PP and EP rubber. The
solid density is of the PP is 0.908 g/cc. The solid density is of the EP is 0.86 g/cc. Each
blend was injection molded into a standard tensile bar mold, the dimensions of which are
shown in Figure 2.1. The mold wall temperature was 75°F, the melt temperature was
395°F, and the fill time was eight seconds. Short shots were molded for each material;
three for TPO #l at 0.44, 0.52, and 0.58 inches, and two for TPO #2 at 0.50 and 0.53

inches.

Table 2.1 Material properties of TPO #1 and #2 blends.

 

TPO Wt % EP Volume PP MW EP MW

 

 

Fraction
1 32 0.3 1 150,000 260,000
2 35 0.34 180,000 235,000

 

 

 

 

 

 

 

 

 

f 16cm J
r a C“
W/ A A £1

1 If N 4

3mm

 

 

 

 

Figure 2.1 Tensile bar dimensions.

The flow line patterns for TPO #1 and #2 are shown in Figure 2.2. For illustrative
purposes, the flow mark regions have been colored gray. For TPO #1, the flow line
frequency (distance between the centers of two flow lines) is 1.5 cm and the flow line
width is 6 mm. For TPO #2, the flow line frequency is 2.0 cm and the flow line width is

11 mm.

Flow —’

    

(a) TPO #1

 

(b) TPO #2

Figure 2.2 Flow line patterns for (a) TPO #1 and (b) TPO #2.
(The flow mark regions are shown in gray.)

2.2 SEM Analysis

Montell (Elton, Maryland) provided SEM micrographs for each short-shot. The short-
shots were cross-sectioned, microtomed, and the EP rubber was dissolved with
methylcyclohexane. The EP rubber particles left dark visible voids in the polypropylene

matrix. The micrographs were taken at the locations shown in Figure 2.3 for both TPO’s:

  

Flow ——> z, Gap Direction

y. Transverse
Flow Direction

x, Flow Direction

3 4 5

Figure 2.3 Micrograph locations.

I. Out-of-flow mark on the plaque surface (in the xy plane)

2. Flow mark on the plaque surface (in the xy plane)

3. Cross-section at the wall underneath the flow mark (in
the xz plane)

4. Cross-section at the wall underneath the out-of-flow
mark (in the xz plane)

5. Cross-section behind the flow front (in the xz plane)

The out-of-flow mark, flow mark, and core flow front region micrographs for each short

shot of both TPO’s are shown in Figures 2.4-2.9.

 

(a) (b)

31 i run

 

(C)

Figure 2.4 Out-of-flow mark surface morphology for TPO #l
(a) 0.44” short shot
(b) 0.52" short shot
(c) 0.58" short shot

Figure 2.5

 

Flow mark surface morphology for TPO #l
(a) 0.44" short shot

(d) 0.52” short shot

(e) 0.58” short shot

Figure 2.6

 

Cross sectional core flow front morphology for TPO #l
(a) 0.44" short shot
(b) 0.52" short shot

(C) 0.58" short shot

 

Figure 2.7 Out-of-flow mark surface morphology for TPO #2.
(a) 0.50" short shot
(b) 0.53" short shot

30 um

 

Figure 2.8 Flow mark surface morphology for TPO #2.
(a) 0.50" short shot
(b) 0.53" short shot

—
30 um

   

(b)

Figure 2.9 Cross sectional core flow front morphology for TPO #2.
(a) 0.50“ short shot
(b) 0.53” short shot

All of the micrographs were examined to ensure that the morphology particle shapes
observed were consistent in each short shot. Many measurements were made for all the
micrographs for each short, but a full set of clear, crisp micrographs showing the x, y, and
2 dimensions were only obtained for the 0.52 inch short-shot of TPO #1 and the 0.53 inch
short-shot of TPO #2. Thus, only these two sets were examined in great detail. The
micrographs for TPO #1 are shown in Figures 2.10, 2.13, and 2.17. The micrographs for
TPO #2 are shown in Figures 2.26, 2.29, and 2.33. The projected area and the
dimensions along the two orthogonal axes were measured for each rubber particle using
Jandel Scientifics’ Sigma Scan Pro 3.0 and Scion Corporation’s Scion Image version 3b
image analysis software packages. The out-of-flow mark and flow mark surface pictures

were used to determine the particles dimensions in the x and y directions (Figures 2.10

and 2.26). The cross-sectional wall micrographs in the xz plane yielded an average
particle thickness in the z direction (Figures 2.13 and 2.29). From these measurements
and observations, a three dimensional shape of what the particles at the mold wall might

look like was determined.

2.3 Results of TPO #1 Image Analysis

Figures 2.10, 2.13, and 2.17 show the micrographs for the out-of-flow mark, flow mark,
and core flow front regions. Examination of the particle size distributions, Figures 2.23-
2.25, indicates a distinct break at 5 pm2 in all regions between the presence of small and
large particles. Hereafter, small particles will be catorgized by having projected areas < 5

um2 and large particles will have projected areas > 5 umz.

For the out-of-flow mark surface region (Figure 2.10(a)), the micrographs reveal that the
particles are long, narrow, stretched parallel to the flow direction, and have an area
fraction of 0.168 (Table 2.2). The large (> 5 umz) particles have a high area weighted
average aspect ratio of 17.4 and take up 54% of the area fraction in the picture.
Examination of the detail, Figure 2.10(b), reveals the presence of many smaller, highly
stretched particles. These particles have a much smaller projected area (<< 5 umz) but
the area weighted average aspect ratio is still large at 12.6. The aspect ratio distribution
for all the particles (Figure 2.11) is very wide, ranging from 1-33, with an area weighted

average at 15.20.

15

 

 

 

(d) Detail

(c) Dull flow mark region

Figure 2.10 (a) Out-of—llow mark surface micrograph.
(b) Detail of individual particles in the out-of—flow mark
surface region.
(c) Flow mark surface micrograph.
((1) Detail of individual particles in the flow mark surface region.

Table 2.2 compares the small and large particles. It also contains the x and y dimensions
that were measured from the paricles in Figure 2.10(a). A value for the z dimension was
not obtained from Figure 2. 13(a) because not enough particles could be counted to
determine a representative thickness for the region. Figure 2.12 shows the x and y
dimension distributions for all of the out—of-flow mark particles. Typically, a distribution
this wide would indicate that the particle orientation was randomly distributed and that
the particles Ccounted are elliptical slices of particles at random angles. But examination
of Figure 10(a) shows that the particles are highly aligned in the flow direction. Even the
small particles shown in Figure 10(b) are aligned parallel to flow. Therefore, the aspect

ratio distribution is a true measurement of particle length and not an indication of

orientation.

Table 2.2 Out-of-flow mark region comparison of particles less than 5 ttm2 vs. greater than 5 m2.

 

 

 

 

 

 

 

 

 

 

 

 

 

Particleg< 5 ”ml Particles > 5 umz Au Particles

Projected Area Fraction of 0.077 0.091 0.168
Particles
Number of Particles 419 74 493
Number Averaged Aspect 10.34 18.23 1 1.53
Ratio
Area Weighted Aspect Ratio 12.61 17.37 15.20
Number Averaged Areas 1.50 10.13 2.80
trun’)
Weighted Average Areas 2.56 12.66 8.05

2

)
Iii-Dimension Area Weighted 6.01 15.25 11.03
AveIBged Length (um)
y-Dimension Area Weighted 0.55 1.11 0.85
Averaged Lenth (um)

 

17

 

0.025 P

" - All Particles
, \ — Particles < 5 um2
\ - - - Particles > 5 run2

0.02 I I

g 0.015

0.01

Are-F

0.005

 

 

 

Aepeetfletio

Figure 2.11 Aspect ratio distributions for the out-of—flow mark region of TPO

0.5 p

0.45 1"

 

x Dimension
0.4

- - - y Dimension

0.3 v

0.3
E 0.25

0.2

 

0.15

0.1

0.6

 

 

Wham

Figure 2.12 TPO #1 out-of-flow mark region particle lengths in the x and y
directions ( measured from Figure 2.10(a)).

18

For both the flow mark and out-of-flow mark xz plane pictures at the mold wall (Figure
2.5), only the particles at the mold wall surface were examined to determine an average
particle thickness in the z direction for each region. The particles in the xz plane cross
section underneath the surface were not counted. All of the particles both regions are
aligned parallel to the flow direction. The out-of-flow mark region particles are highly

stretched and the flow mark region particles are more globular.

 

---i- ' “en-“us smug.» ”A. "‘~.'.:“" ‘ ”refer-w. V . .._ , . "\"“'r"‘- ~ K... ; ..; no... ~ ..-.-- .
, W ,.. " , v“, . . Au. . ,..- , «. ~ , ..,..... 3'”;

cm A‘ " I {w - I - ”m . ll..-‘ ‘ . ,_ «gym-‘47 ~¢~¢ ‘ .9...
~ 'm- ‘ 1 . We , fi - - , 4 h
.- .

.. - V.
K o... - ~--1a%' . ve-W*M¢w --- ,.. .~--1'n' “ .» -uw-- .“ 4. ‘ ~ W ,- ' .‘9' ' . w’
'. :_-7 ‘v .. i .
. .
... -

 

 

a) Out-of-flow mark region

2, Gap Direction

 

y, Transverse Flow
Direction

 

 

x, Flow Direction

 

 

 

 

b) Flow mark region

 

Figure 2.13 Cross-sectional morphology at the mold wall in the x, 2 plane for TPO

The particles in the flow mark region are elliptical in shape, aligned parallel to the flow
direction, and have an area fraction of0.115 (Table 2.3). The large (> 5 umz) particles

have a small area weighted aspect ratio of 2.01. The small (< 5 umz) particles also have a

 

small area weighted aspect ratio of 2.13. The aspect ratio distribution (Figure 2.14) for
all of the particles is narrow, ranging from 1-5, with an overall area weighted aspect ratio
of 2.04. Figure 2.8 shows the particle orientation distribution for all of the flow mark
particles. Ther orientations were determined by measuring the angle difference between
the paricle’s major axis to the flow direction. The Hermann’s planar orientation

parameter ([5,) for TPO #1 is 0.807. It was calculated using:

fl, = 2<cos2 6) —l

The small and large particles for the flow mark region are compared in Table 2.3. It also
contains the x and y dimensions that were measured from the particles in Figure 2.10(c).
The 2 dimension number averaged length was determined by measuring the thickness of
the particles at the mold wall (Figure 2.13). This average thickness was taken as the z
dimension of all of the particles, regardless of size. Figure 2.15 shows the x, y, and z

dimension distributions for all of the flow mark particles.

20

Table 2.3 Flow mark region comparison of particles less than 5 1.1m2 vs. greater than 5 umz.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Particles < 5 pm2 Particles > 5 run2 All Particles
Projected Area Fraction of 0.028 0.087 0.115
Particles
Number of Particles 166 64 228
Number Averaged Aspect 2.18 2.04 2.14
Ratio
Area Weighted Aspect Ratio 2.13 2.01 2.04
Number Averaged Areas 1.41 l 1.59 4.18
(run’)
Weighted Average Areas 2.66 16.65 13.20
(um’)
x-Dimension Area Weighted 2.70 6.69 5.71
Averaged Length (um)
y-Dimension Number 0.92 2.99 1.48
Averaged Length (pm)
z-Dimension Number 0.77 0.77 0.77
Averaged Leflth (um)

 

21

 

0.05.
o'm‘ ’n\ — " All Particles
0.041 I \ --- Particles < 5 1.1m2
0.0351 I {‘3‘ ------ Particles > 5 urn2

0.03 1 I '5 \
0.025 1 I ".‘1
g 0.02 . l "'1‘
0.0151 .-°"‘\
0.01 1 I H .\
0.0051 . .

 

 

 

Figure 2.14 Aspect ratio distributions for the flow mark region of TPO #1.

 

x Dimension
- - - y Dimension
— -z Dimention

  
  

 

Length in um

Figure 2.15 TPO #1 flow mark region particle lengths in the x, y, and 2 dimensions
(measured from Figures 2.10(c) and 2.13(b)).

22

 

 

F0

2111,

par

(< 5
flou
Ham
dime
not b

ifansr

0.35 P

0.3 I

0.25 ll

 

 

 

0.13 I
o A
~90 -75

Degrees

Figure 2.8 Particle orientation distribution for the flow mark region of TPO #1.

For the core flow front region, the micrograph reveals that the particles are elliptical,
aligned parallel to the flow direction, and have an area fraction of 0.208 (Table 2.4). The
particle size distribution (Figure 2.23) indicates that almost all of the particles are small
(< 5 umz) so there was no reason to do a small and large particle comparison like the
flow mark and out-of—flow mark regions. The aspect ratio distribution (Figure 2.18) is
narrow with a range of 1-5 and area weighted average at 1.91. Only the x and 2
dimensions were measured for the core flow front (Figure 2.19). The y dimension could
not be determined because none of the micrographs taken show the cross sectional

transverse xy plane.

23

  

0.09

0.08

  

:'

(b) Detail

 

(a) Core flow front region

Figure 2.17 (a) Cross-sectional morphology at the core flow front for TPO #1.
(b) Detail of individual particles in the core flow front region.

 

 

I
- 5' ‘-, ------ Core Flow Front
D 1
I
\-
L .
l A A .: I . ~ I ' ' e A - --------- A
o 1 2 3 4 5 6
Aspect Ratio

Figure 2.18 Aspect ratio distribution for the core flow front region of TPO #1.

24

 

0.4 p

0.35 I I \
x Dimension

 

\
— -z Dimension

Frequency

 

 

 

 

Length in um

Figure 2.19 TPO #1 core flow front region particle lengths in the x and 2 dimensions
(measured from Figure 2.17).

2.3.1 TPO#I Particle Shapes

The shape of the particles in each region can be determined from the aspect ratios and the
area weighted averaged lengths in the x, y, and 2 directions (Table 2.4). For the out-of-
flow mark region, Figures 2.10(a) and 2.13(a) indicate that the particles are strand-like
cylindrical cylinders with their lengths oriented in the x direction. The particles at the
surface are highly stretched, with an area weighted aspect ratio of 15.2 between the x and
y dimensions. The broad particle size and area weighted aspect distributions (Figures
2.33-2.25) indicate the presence of both small and large highly stretched particles. Figure

2.70 shows a representative particle and its’ orientation.

25

I‘— Lx —_.l 2, Gap Direction

{7 y, Transverse Flow Direction

x, Flow Direction

Figure 2.20 Circular cylinder particle in the out-of—flow mark region.

The flow mark region particles shown in Figures 2.10(c) and 2.13(b) are elliptical
cylinders whose length is oriented in the z direction. The area weighted aspect ratio is
2.04, its distribution (Figure 2.25) is very narrow, indicating that particles are elliptical
shaped in the xy plane. In comparison to the area weighted x and y dimensions, the z
dimension listed in Table 4 is small. Therefore, the particles resemble thin discs that are
slightly stretched along their x axis. The particle size distributions (Figures 2.23 and
2.24) are broad, indicating a mixture of small and large particles. Figure 2.21 shows a

typical particle and its orientation in the flow mark region.

2. Gap Direction

(1

b fl // ‘,1L

7

y, Transverse Flow Direction

2. x, Flow Direction

Figure 2.21 Elliptical cylinder particle in the flow mark region.

In the core flow front region, (Figure 2.4) the particles are ellipsoids. The area weighted

x and 2 dimensions are listed in Table 2.4, but since there are no micrographs that show

26

the transverse yz plane, the y dimension could not be measured. The y dimension should
be roughly equivalent to the z dimension. The particles are aligned parallel to the flow
direction and the area weighted x dimension is almost double that of the z dimension,
which corresponds to the area weighted aspect ratio of 1.91. The narrow particle size
distributions (Figures 2.23 and 2.24) confirm the presence of a large number of small (< S
umz) particles. Figure 2.22 shows a typical ellipsoid particle and it’s orientation in the

core flow front region.

2, Cap Direction

y, Transverse Flow Direction

x, Flow Direction

 

Figure 2.13 Ellipsoid particle in the core flow region.

27

Table 2.4 Measured values and calculated statistics for particles in TPO #1.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Core Flow Out-Of-Flow Mark Flow Mark
Front Surface Surface

Projected Area Fraction of 0.208 0.168 0.115
Particles
Bulk Volume Fraction 0.31 0.31 0.31
Total Micrograph Area 8244.6 8229.4 8280.1
tum‘)
Total Particle Area (umz) 1714.5 1379.1 953.3
Total Number of Particles 1373 493 228
Number Averaged Particle 1.249 2.797 4.181
Areas (umz)
Weighted Average Particle 3.77 8.05 13.20
Areas (ymz)
Number Averaged Aspect 1.84 1 1.53 2.14
Ratio
Area Weighted Aspect Ratio 1.91 15.20 2.04
x-Dimension Area Weighted 2.77 11.03 5.71
Averaged Lenggi (um)
y-Dimension Area Weighted - 0.85 3.00
Averaged Length (um)
z-Dimension Area Weighted 1.53 - 1.31
Averaged Length (um)
Number Averaged Particle 0.35 1.34 3.23
Volume (um’)
Total Interfacial Area (umz) 6495 5802 3606

 

 

 

 

The equations for volume, projected area in the xy, xz, and yz planes, and interfacial area

for the different representative particle shapes are listed in Table 2.5.

28

 

Table 2.5 Geometric equations for various particle shapes.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Elliptical Cylinder Elliptical Cylinder Circular Ellipsoid
(Length in x (Length in 2 Cylinder
Direction) Direction)
Volume nabL, zrabLz 7:021.z nabc
4 4 4 6
xy Plane
Projected aLr fl DLZ Eb—c
Area (A,,) 4 4
xz Plane nab
Projected 12L] bLz DLz -4—
A"; (An)
yz ane b 2
Projected ”i aL: ”D ”LC
Area (An) 4 4 4
Interfacial 2 2 7m}, 2 2 nab ”DZ
Area (Al) 27(14", 20 + 2’) + *2— Zfl'L: 2a + 21) + T 7'(,'DLz + —2-— ”a;
0.18 n n
0.16 » ’I 2 ------ Core Flow Front
— —Out of Flow Mark Surface
0.14 I - —Flow Mark Surface
0.12 I .'
g 0 1 I 2
1L 5
g 0.03 D:
0.06 I;
: r s
0.04 t' I \\
1! x . ._
0.02 1, ‘ -33: "
o A A 7 '1 - n A
0 5 10 15 20 25 30

Prolectod Partlclo Area (“m“)

Figure 2.23 Particle size area distribution for TPO #l.

29

 

Freq uency
.0
‘

0.09 v

0.00 I

0.00 i
0.05 L
0.04 -

0.03 t

 

‘ 5 " ------ Core Flow Front
— -—Out 01 Flow Mark Surtece
-—Flow Mertt Suflece

 

 

 

—_—_
~—_
I —_-

0 5 10 15 20 25 30

Projected Particle Aree (um')

Figure 2.24 Particle size distribution for TPO #1.

------ Core Flow Front

Flow Mark
— -Out of Flow Mark

 

I
e
r
u
r
r
e
-
I
-
r
r
e
r
I
r
r
t
e

 

Aepect Ratio

Figure 2.25 Aspect ratio distributions for TPO #l.

30

 

Table 2.6 shows the standard deviations for the number averaged aspect ratios and

number averaged x, y, and z dimension lengths for TPO #1. For the out—of-flow mark

region, the aspect ratio distribution (Figure 2.11) and x-dimension length distribution

(Figure 2.12) are non-gaussian. Because of this, the standard deviation cannot be

accurately calculated for them. Therefore, the most probable range where the particles lie

 

 

 

 

 

 

 

 

 

 

 

 

 

 

has been listed.
Table 2.6 Variability of x, y, and z dimension lengths for TPO #1.
Number Standard Most Probable
Averaged Value Deviation Range
Aspect Ratio 1 1.53 5-27
x-Dimension 5.59 1-7
Out-Of-Flow Lergth (pm)
Mark Region y-Dimension 0.61 $0.32
Ltflgth (um)
z-Dimension - -
Lengg (11m)
Aspect Ratio 2.14 $0.84
x-Dimension 2.9 12. 1
Flow Mark Leflth (um)
Region y-Dimension l .48 $0.46
Length (um)
z-Dimension 0.77 :0.48
Length (pm)
Aspect Ratio 1.84 10.71
x-Dimension 1.37 :1 1
Core Flow Length (um)
Front Region y-Dimension - -
Length (um)
z-Dimension 0.76 :0.58
Length (um)

 

 

 

 

 

31

 

2.3.2 F low Mark vs. Out-of-flow Mark Comparison
The particle volume is conserved between the flow mark and out-of-flow mark regions
and there is negligible particle migration difference between the two. For the two

regions, Table 2.7 shows the number averaged areas and the total number of particles.

Table 2.7 Number average for the out-of-flow mark and flow mark regions.

 

 

 

 

 

 

 

 

Out-Of-Flow Mark Flow Mark Surface
Surface

Total Number of Particles 493 228
Number Averaged 2.8 4.2

Particle Area (pm’)

y-Dimension Number 0.6] 1.48
Averafid Length M)

z-Dimension Number - 0.77
Averaggd Length (um)

 

To obtain the total particle volume Win the flow mark region, the xy plane projected area
is substituted into the volume equation for an elliptical cylinder whose length is oriented
in the z dimension (Table 2.5) and is then multiplied by My (the total number of
particles).

VT : A“. L, N W (2'1)
Substituting in the flow mark surface values from Table 2.7 obtains:

VT 2 737ttm3

 

Therefore, the average particle volume [ VT ] in the flow mark surface is 3.23 um3.
FM

32

 

The interfacial area was calculated using the equation for an elliptical cylinder whose

length is oriented in the z direction (Table 2.5). The total interfacial area (Ah )was

0“"

calculated by summing the interfacial area for every particle in the flow mark region.

The value used for L2 was the number averaged length in the z direction.

703

,4,an =:[2an\/2a2 +21;2 +3533] (2-2)
‘ ' i=1 -

l

A, = 3606 pm2

T0 TA 1.

The ratio of total interfacial area to total particle volume is

AITOTAI. : 489 #m-I
7.

The shape with the minimum interfacial area is a sphere, therefore, as a comparison for
retraction, the radius for a sphere of equal volume to a single particle in the flow mark
region is calculated. The weighted average particle area and z dimension area weighed

average length (Table 2.4) were used for A... and L.

R = 3.44 am

3 -1
— = 0.872 m
R H

Thus the actual interfacial area in the flow mark region is 5.6 times greater than the fully

retracted particle interfacial area.

33

To obtain the total particle volume Vrin the out-of—flow mark region, the xy plane
projected area is substituted into the volume equation for a circular cylinder (Table 2.5)
and is then multiplied by Noam (the total number of particles).

7: DA . . N
V : .1) OOFM

T 4 (2-4)

 

Substituting in the out-of-flow mark surface values from Table 2.6, using the y dimension
number averaged length for 0, yields:

v, =66lpm"

 

Therefore, the average particle volume ( VT J in the out-of—flow mark region is 1.34
OOFM

um3, which is 2.4 times smaller than the average particles volume for the flow mark

region.

The interfacial area was calculated using the equation for a circular cylinder given in

Table 2.5. The total interfacial area (A,mm )for the out-of-flow mark region was

calculated by summing the interfacial area for every particle in the out-of-flow mark
region.

493

2
A, _ =2(JIDL, +7”) J (25)
TOTAL i=| ” 2 i

 

A, = 5802 pm2

T0 TA 1.

The ratio of total interfacial area to total volume is

AlTUT/ll. :878 #m-l
T

34

To calculate the radius for a sphere with equal volume to a single particle in the out-of-
flow mark region, the weighted average particle area and y dimension area weighed
average length (Table 2.4) were used for A... and D.

3

mt“ D
_ 4

(2-6)

 

firm
3

R = 2.55 pm
3

— = 1.78 m‘I
R #

Thus the actual interfacial area in the out-of-flow mark region is 7.44 times greater than

the fully retracted particle interfacial area.

The particles in the core flow front, flow to the melt front and become stretched by
biaxial elongational flow. As they are being stretched, they flow along the short span of

the melt front to the out-of-flow mark region and freeze in their stretched state at the

mold wall.

Since the orientation distribution of the flow mark region shows that the particles are
aligned parallel to the flow direction rather than randomly distributed, it can be concluded
that the dominant mechanism is particle retraction as the particles flow along the long
span from the stagnation point to the mold wall. The particles flow from the core flow
front and become stretched by the biaxial elongational flow at the melt front in the same
way as the out-of-flow mark particles. But due to the increased timescale as they flow
along the long span from the stagnation point to the mold wall, the strands have enough

time to retract into the observed disc shaped particles before they freeze at the wall

35

forming a flow mark region. Normally, if a free-floating particle retracts in a polymer
matrix, it will form a sphere, but this does not happen for the EP rubber particles. They
form discs instead. There are two reasons to explain this behavior. First, the mold wall
constrains the particles and does not allow the spheres to form. Second, there is a large
amount of shear at the wall, which hinders expansion in the z direction. The particles are
constrained by these two factors and so they can only retract in the xy plane, which
results in the formation of the observed discs. Even though they cannot form spheres, the
particles still undergo a significant amount of retraction when compared to the fully
stretched particles. Comparing the total interfacial area, there is 38% more area in the
out-of-flow mark region than the flow mark region. Also, as the particles retract, they
coalesce into larger particles that have 2.4 times the volume of the out-of-flow mark

particles. Keeping the volume difference in mind, the ratio of the flow mark particle’s
calculated interfacial area to an equivalent sphere’s £— value is 5.60. That value is 1.3

times smaller than the out-of-flow mark region’s ratio of 7.44.

To account for the reduction of the number of particles between regions and the increased
average volume per particle in the flow mark region, coalescence must be taking place.
As the stretched particles retract, they could be coalescing with each other to form larger
particles. The other option is that as two of the stretched strand-like particles retract to
form the disc shaped particles of the flow mark region, one slides underneath the other

and they stack like coins.

36

2.3.3 Core F low Front vs. Mold Wall Comparison
The volume fraction of EP rubber (total particle volume) is the same in the core flow
front region, flow mark region, and out-of-flow mark regions. For the core flow front

region, Table 2.8 shows the number averaged areas and the x and 2 dimensions.

Table 2.8 Average particle dimensions for the core flow front region.

 

Core Flow Front

 

 

 

Total Number of Particles 1373
Number Averaged 1.25
Particle Area (umz)

x-Dimension Number 1.37
Averaged Length (um)

 

y-Dimension Number -
Averaged Length (um)
z-Dimension Number 0.76
Averaged Length (um)

 

 

 

 

 

To obtain the total particle volume VT equation for the core flow front region, the xz plane
projected area can be substituted into the volume equation for an ellipsoid (Table 2.5) and

then be multiplied by NC (the total number of particles).

V.

I‘

o
: i-C'Au N ,. (2‘7)

An average value for c can then be obtained by substituting in 699 um3 for the volume
(the average volume between the flow mark and out-of-flow mark regions) and the core
flow front values from Table 2.8.

€20.61 um

37

Now that an average value for c has been obtained, the interfacial area for a single

particle can be calculated. From Table 5, the interfacial area formula for an ellipsoid is:

A, = no]? (2-8)
I 2 2
where b = _2b_§-_2_c_

Substituting in the number averaged values for b and 0 yields:

1) =2 0.69 p 111

To obtain the total interfacial area, all of the areas must be summed.

I373

AITUMI = 2 (Ital?) (2'9)

1:1 r'

A] TorAI. : 4038 ,u m-
Dividing the total volume by the total number of particles yields an average particle
volume of 0.51 1.1m3 . If the assumption that volume is conserved between all of the

regions is valid, then coalescence must be taking place.

Considering the case of the core flow front to the out-of—flow mark surface, the only time
that coalescence can take place is when the particles approach the melt front and become
stretched by elongational flow. On average, 2.8 of the core particles would have to

coalesce to form one stretched particle.

In the flow mark re gion. the particles can coalesce twice. The first time would be as the

particles approach the melt front and become stretched by elongational flow in the same

38

manner as the out-of-flow mark surface case. After being stretched, they can coalesce

again as the particles retract into discs.

But volume may not be conserved between the core flow front and the wall regions
because the average y dimension calculated for the ellipsoid particles in the core was 0.61
um. For an ellipsoid aligned parallel to the flow direction, it is assumed that the average
y dimension should be at least equal to or greater than the z dimension. But since this is
not the case, the core could contain much more volume than the wall regions. For
example, if b and c were equal (prolate spheroid geometry), the total volume would be
1748 um3 and the interfacial area would be 6858 m2. To explain the missing volume,
there would have to be some sort of particle migration or break-up occurring. Particle
break-up does not seem very likely because there would be an increased number of small
particles seen in the flow front and out-of-flow front regions. Just the opposite is
occurring in these regions, the particles are becoming larger and being reduced in
number. Therefore, some combination of particle migration and coalescence is

occurring, but the extent of which is unknown.

2.4 Results of TPO #2 Image Analysis

The micrographs for the three regions of TPO #2 are shown in Figures 2.26, 2.29, and
2.33. As with TPO #1, the particle size distributions, Figures 2.39 and 2.40, indicate the
pressence of small particles. Hereafter, the small particles will be catorgized by having

projected areas < 7 ttm2 and large particles will have projected areas > 7 umz.

39

The particles in the out-of-flow mark surface region (Figure 2.26(a)) are long, narrow,

stretched parallel to the flow direction, and have an area fraction of 0.20 (Table 2.10).

The large (> 7 umz) particles have a high area weighted average aspect ratio of 17.1 and

take up 46% of the area fraction in the picture. The small particles (< 7 m2) also have a

large area weighted average aspect ratio of 14.3. The aspect ratio distribution for all the

particles (Figure 2.30) is very wide, ranging from 1-50, with an area weighted average at

15.60.

Table 2.9 shows the comparison to the small and large particles. It also contains the x, y,

and 2 dimensions that were measured from the paricles in Figures 2.26 and 2.33. Figure

2.31 plots the x, y, and z dimension distributions for all of the out-of—flow mark particles.

Table 2.9 TPO #2 Out-of-flow mark region comparison of particles less than 7 pm2 vs. greater than 7 umz.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Particles < 7 pm2 Particles > 7 pm2 All Particles
Projected Area Fraction of 0.108 0.092 0.200
Particles
Number of Particles 380 53 433
Number Averaged Aspect 11.87 17.06 12.51
Ratio
Area Weighted Aspect Ratio 14.29 17.13 15.60
Number Averaged Areas 2.33 14.26 3.79
(111112)
Weighted Average Areas 3.53 19.63 10.94
(11m?)
x-Dimension Number 5.71 17.99 7.22
Averaged Length (gm)
y-Dimension Number 0.54 1.30 0.64
Averaged Length (um)
z-Dimension Number 0.56 0.56 0.56
Averaggd Length (um)

 

40

 

 

    

D tail
(a) Glossy out-of—flow mark (b) 6

Flow —§

 

   

(d) Detail
(c) Dull flow mark region

Figure 2.26 (a) Out-of-flow mark surface micrograph.
(b) Detail of individual particles in the out-of-flow mark surface region.
(c) Flow mark surface micrograph.
(d) Detail of individual particles in the flow mark surface region.

41

— — All Particles

0.045 -
0.04 - I"
I \ . 2
, \ — Particles < 7 pm
0.035 . \ ...... . 2
I \ Particles > 7 pm

 

o 5 10 15 20 25
Aspect Ratio

Figure 2.27 Aspect ratio distributions for the out-of-flow mark region of TPO #2.

 

x Dirnenston
- - - y Dimension
— -z Dimension

 

0.6 f

0.5

 

14 1O

12

I 10

2 4
Lengthtnpm

 

Figure 2.28 TPO #2 Out-of-flow mark region particle lengths in the x, y, and
dimensions (measured from Figures 2.17(a) and 2.20(a)).

42

For both the flow mark and out-of-flow mark xz plane pictures (Figure 2.29), only the
particles at the mold wall were examined to determine an average particle thickness in the
z direction for each region. The particles in the xz plane cross-section underneath the
surface were not counted. All of the particles in both regions are highly stretched and

aligned parallel to the flow direction.

  
 

z, Gap Direction

 

y, Transverse Flow
Direction

x, Flow Direction

 

 

 

 

Flow mark region

 

   

 

Figure 2.29 Cross-sectional morphology at the mold wall in the x, 2 plane for TPO #1.

The particles in the flow mark region are moderately stretched, aligned parallel to the
flow direction, and have an area fraction of 0.1.26 (Table 2.10). The large (> 7 umz)
particles have an area weighted aspect ratio of 9.54. The area weighted aspect ratio of the
small (< 7 umz) particles slightly smaller at 9.16. The aspect ratio distribution (Figure

2.30) for all of the particles wide, ranging from 1-40, with an overall area weighted

43

aspect ratio of 9.27. Figure 2.32 shows the particle orientation distribution for all of the
flow mark particles. Ther orientations were determined by measuring the angle
difference between the paricle’s major axis to the flow direction. The Hermann’s planar
orientation parameter (fp) for TPO #2 is 0.949. It was calculated using:

fp = 2<cos2 6) -1
The small and large particles for the flow mark region are compared in Table 2.10. It
also contains the x and y dimensions that were measured from the particles in Figure
2.26(c). The 2 dimension number averaged length was determined by measuring the
thickness of the particles at the mold wall (Figure 2.29(b)). This average thickness was
taken as the z dimension of all of the particles, regardless of size. A plot of the x, y, and

z dimension distributions for all of the flow mark particles is shown in Figure 2.31.

Table 2.10 TPO #2 Flow mark region comparison of particles less than 7 111112 vs. greater than 7 umz.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Particles < 7 um2 Particles > 7 um2 All Stretched
Particles

Projected Area Fraction of 0.090 0.036 0.126
Particles
Number of Particles 327 23 350
Number Averaged Aspect 8.07 9.40 8.16
Ratio
Area Weighted Aspect Ratio 9.16 9.54 . 9.27
Number Averaged Areas 2.28 12.99 2.99
mm”)
Weighted Average Areas 3.47 15.57 6.93
(umz)
x-Dimension Number 4.88 13.31 5.43
Avergged Length (um)
y-Dimension Number 0.69 1.92 0.77
Avergged Lergth (um)
z-Dimension Number 0.87 0.87 0.87
Avegged Length (tun)

 

 

Area Fraction

0.6

0.5

0.03 I-

— — All Particles
0.025 1- rs — Particles < 7 urn2
\ ------ Particles > 7 am2

0.02 I I

0.015

0.01

0.005

 

 

 

Figure 2.30 Aspect ratio distributions for the flow mark region of TPO #2.

 

1. x Dimension
I :: ' " ' y DlIIIOIWO"
{r1 — -z Dimension

 

 

 

Lengthin um

Figure 2.31 TPO #2 flow mark region particle lengths in the x, y, and 2
directions (measured from Figures 2.26(c) and 2.29(b)).

45

(135 r
0.3 I

0.25 I

 

3015 1-
u.
0.1 I

0.05-

 

Degrees
Figure 2.32 Flow mark particle orientation distribution for TPO #2.

For the core flow front region, the micrograph reveals that the particles are generally
elliptical and have an area fraction of 0.217 (Table 2.9). Examination of the particle size
distributions (Figures 2.39 and 2.40) indicates that almost all of the particles are small (<
7 umz). The aspect ratio distribution (Figure 2.34) has a range of 1-9 and area weighted
average at 4.22. Figure 2.35 plots the x and 2 plane dimensions for the core flow front.
The y plane dimension could not be determined because none of the micrographs show

the cross sectional transverse xy plane.

46

   

(b) Detail
(a) Core flow front region

Figure 2.33 (a) Cross-sectional morphology at the core flow front for TPO #2.
(b) Detail of individual particles in the core flow front region.

0.‘ I
------ Core Flow Front

 

 

0.04 I '. .
0.02 I ."' .
' a . u .. . -
0 ll . I I I l I I l l l
0 1 2 3 4 5 6 7 8 9 10

Aspect Ratio

Figure 2.24 Aspect ratio distribution for the core flow front region of TPO #2.

47

 

 

 

 

 

 

0.6 I
0. 5 _ I‘\ x Dirnenslon
I \ - -z Dimension
I
l
l
l
i
l
i
I
l
l
4 fi"==—
0 1 2 3 4 5 6

Length In um

Figure 2.35 TPO #2 core flow front region particle lengths in the x and 2 directions
(measured from Figure 2.33).

2.4.1 TPO #2 Particle Shapes

A representative particle shape for each region can be determined from the aspect ratios
and the area weighted averaged lengths in the x, y, and 2 directions (Table 2.11). The
out-of-flow mark region particles shown in Figure 2.26(a) and 2.29(b) are elliptical
cylinders with their length oriented in the x direction. As the area weighted aspect ratio
of 15.60 suggests, the particles are highly stretched. Comparing the y dimension area
weighted averaged length to the z dimension number averaged length (not enough
particles were counted to obtain a true weighted averaged length), it is shown that the
particle width in the y direction is twice its’ thickness in the z direction. Figure 2.36

shows the representative particle shape and its’ orientation.

48

 

L .| z, Gap Direction
I

 

4.. ...- j -u‘va.._ .

y, Transverse Flow Direction

x, Flow Direction

 

Figure 2.36 Elliptical cylinder particle in the out-of-flow mark region.

The flow mark region particles shown in Figure 223(0) and 2.26(d) are also elliptical
cylinders with their length oriented in the x direction. These particles are moderately
stretched when compared to the out-of-flow mark particles, with an area weighted aspect
ratio of 9.27. Comparing the y dimension area weighted averaged length to the z
dimension number averaged length (not enough particles were counted to obtain a true
weighted averaged length), it is shown that the particle width in the y direction is roughly
1.4 times its’ thickness in the z direction. Therefore, the particles are shorter, fatter, and
thicker than the out-of-flow mark particles. Figure 2.37 shows the representative particle

shape and its’ orientation.

r— L ___.| z, Gap Direction
x

y, Transverse Flow Direction

x, Flow Direction

 

Figure 2.37 Elliptical cylinder particle in the flow mark region.

In the core flow front region, (Figure 2.33) the particles are ellipsoids. The area weighted
x and 2 dimensions are listed in Table 2.11, but since there are no micrographs that show

the transverse yz plane, the y dimension could not be measured. The y dimension should

49

be roughly equivalent to the z dimension. The particles are aligned parallel to the flow
direction and the area weighted x dimension is over four times that of the z dimension,
which corresponds to the area weighted aspect ratio of 4.22. The narrow particle size
distributions (Figures 2.39 and 2.40) confirm the presence of a large number of small (< 7
umz) particles. Figure 2.38 shows a typical ellipsoid particle and its orientation in the
core flow front region.

2, Gap Direction

y, Transverse Flow Direction

 

 

x, Flow Direction

Figure 2.38 Ellipsoid particle in the core flow region.

50

Table 2.11 Measured values and calculated statistics for particles in TPO #2.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Core Flow Front Out-Of-Flow Flow Mark
Mark Surface Surface

Particle Area Fraction 0.217 0.200 0.126
Bulk Volume Fraction 0.34 0.34 0.34
Total Micrograph Area 8244.7 8229.4 8280.1
(um’)
Total Particle Area (umz) 1790.5 1642.8 1045.8
Total Number of Particles 2068 433 350
Number Averaged Areas 0.87 3.79 2.99
(111112)
Area Weighted Averages 3.69 10.94 6.93
om‘)
Number Averaged Aspect 2.60 12.51 8.16
Ratio
area Weighted Aspect 4.22 15.60 9.27
Ratio
x-Dimension Area 2.57 13.92 8.85
Weighted Averaged
Length (Jim)
y-Dimension Area — 1.04 1.19
Weighted Averaged
Length (um)
z-Dimension Number 0.50 0.56 0.87
Averaged Lergth (pm)
z-Dimension Area 0.59 - -
Weighted Averaged
Length (pm)
Number Averaged 0.35 1.67 2.04
Particle Volume (umj)
Total Interfacial Area 6495 27962 22247
(um’)

 

51

 

   

------ Core Flow Front
— -Out of Flow Mark Surface
—Flow Mark Surface

 

 

 

o 5 10 15 20 25 30 35 4o 45 50
Projected Particle Area (umz)

Figure 2.39 Particle size area distributions for TPO #2

 

 

 

 

 

 

0.8

it.
0.7 -.": ------ Core Flow Front

g g - —om of Flow Mark Surface

3 ': —Flow Mark Surface
0.6 ._ .

f . "——-—-—==____" .-
5 10 15 20 25 30
Projected Particle Area (um’)

Figure 2.40 Particle size distributions for TPO #2.

52

0.12 I

0.1 r ,
: g - -Out Of Flow Mark Surface
0.08 - ;' ': -—F|ow Mark Surface
% t ------ Core Flow Front
E 0.05 i :-
g s '1.
< :' "

 

 

 

 

Aspect Ratio

Figure 2.41 Aspect ratio distributions for TPO #2.

Table 2.12 shows the standard deviations for the number averaged aspect ratios and
number averaged x, y, and z dimension lengths for TPO #2. The x-dimension length
distributions for the out-of-flow mark region (Figure 2.28) and the core flow front (Figure
2.35) are non-gaussian. Because of this, the standard deviation cannot be accurately
calculated for them. Therefore, the most probable range where the particles lie has been

listed.

53

Table 2.12 Variability of x, y, and z dimension lengths for TPO #2.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Number Standard Most Probable
Averaged Value Deviation Range

Aspect Ratio 12.5 28.5

x-Dimension 7.22 2—10
Out-Of-Flow Length (pm)
Mark Region y-Dimension 0.63 10.41

Length (um)

z-Dimension 0.56 :0. 14

Length (gag

Aspect Ratio 8.16 15.39

x-Dimension 5.43 :3.7
Flow Mark Length (um)
Region y-Dimension 0.76 $0.5

Lengg (11m)

z-Dimension 0.87 4.0.4

Length (um)

Aspect Ratio 2.6 21.67

x-Dimension 1.45 0.5-2
Core Flow Length 0"“)
Front Region y-Dimension - -

Length (um)

z-Dimension 0.5 10.3

Lengt_h (pm)

 

54

 

2.4.2 Flow Mark vs. Out-of-flow Mark Comparison
The particle volume is conserved between the flow mark and out-of-flow mark regions
and there is negligible particle migration difference between the two. For the two

regions, Table 2.13 shows the number averaged areas and the total number of particles.

Table 2.13 Number averaged statistics for the out-of-flow mark and flow mark regions.

 

 

 

 

 

 

 

 

 

 

Out-Of-Flow Mark Flow Mark Surface
Surface

Total Number of Particles 433 350
Number Averaged 3.79 2.99
Particle Area (umz)

x-Dimension Number 7.22 5.43
Averaged Length (pm)

y-Dimension Number 0.63 0.77
Averaged Length (pm)

z-Dimension Number 0.56 0.87
Averaged Length (um)

 

To obtain the total particle volume Vrin the flow mark region, the xy plane projected area
is substituted into the volume equation for an elliptical cylinder whose length is oriented
in the x dimension (Table 2.5) and is then multiplied by Nm (the total number of
particles).

V = 7:11,,be m

r 4 (MO)

Substituting in the flow mark surface values from Table 2.11 obtains:

V, = 715er3

55

 

Therefore, the average particle volume [ VT ] in the flow mark surface is 2.04 um3.
FM

The interfacial area was calculated using the equation for an elliptical cylinder whose

length is oriented in the x direction (Table 2.5). The total interfacial area (Ah )was

calculated by summing the interfacial area for every particle in the flow mark region.

The value used for a was the number averaged length in the z direction.

350
A, = 2( 27:1,, 2a2 + 2122 +23) (211)
TOTAL i=1 2 i

A, = 22,247 pm2

TOTAL

The ratio of total interfacial area to total volume is

’ TOTAL

=3l.1;rm’l

T
The shape with the minimum interfacial area is a sphere, therefore, as a comparison for
retraction, the radius for a sphere of equal volume to a single particle in the flow mark
region is calculated. The weighted average particle area and y dimension area weighed

average length (Table 2.11) were used for Ax, and b.

ItA b
gay =__48 (212)
R = 1.16 pm
3
—= 2.59 m"
R #

Thus the actual interfacial area in the flow mark region is 12.0 times greater than the fully

retracted particle interfacial area.

56

To obtain the total particle volume Win the out-of-flow mark region, the xy plane
projected area is substituted into the volume equation for an elliptical cylinder (Table 2.5)

and is then multiplied by Noam (the total number of particles).

V : ”bAryNoom

r 4 (2-13)

 

Substituting in the out-of-flow mark surface values from Table 2.11, using the y

dimension number averaged length for b, yields:

V, = 722 pm3

Vr

 

Therefore, the average particle volume [ J in the out-of-flow mark region is 1.67

00PM
um3, which is 1.22 times smaller than the average particles volume for the flow mark

region.

The interfacial area was calculated using the equation for an elliptical cylinder given in

Table 2.5. The total interfacial area (A,mm ) for the out-of-flow mark region was

calculated by summing the interfacial area for every particle in the out-of-flow mark

region.

I

433
Aime = 2(27zL,\/2a2 +21;2 +3533] (214)
i=1

A, = 27,962 ,umz

TOTAL

The ratio of total interfacial area to total volume is

57

—A'T0“L = 38.7 ,um"

r
To calculate the radius for a sphere with equal volume to a single particle in the out-of-
flow mark region, the weighted average particle area and y dimension area weighed
average length (Table 2.10) were used for Ax, and b.

nA b
g-IIR3: 4” (2-15)

 

R = 1.29 pm
3

— = 2.33 m"
R lu

Thus the actual interfacial area in the out-of-flow mark region is 16.6 times greater than

the fully retracted particle interfacial area.

The particles in the core flow front flow to the melt front and become stretched by
elongational flow. As they are being stretched, they flow along the short span of the melt

front to the out-of-flow mark region and freeze in their stretched state at the mold wall.

Since the orientation distribution of the flow mark region shows that the particles are
highly aligned parallel to the flow direction rather than randomly distributed, it can be
concluded that the dominant mechanism is partial particle retraction as the particles flow
along the long span from the stagnation point to the mold wall. The particles flow from
the core flow front and become stretched by the biaxial elongational flow at the melt
front in the same way as the out-of-flow mark particles. But due to the increased

timescale as they flow along the long span from the stagnation point to the mold wall, the

58

strands partially retract into the observed shorter, thicker particles before they freeze at
the wall. Normally, if a free-floating particle retracts in a polymer matrix, it will form a
sphere. Under the high shear conditions at the mold wall, the particles form discs, such
as the particles in TPO #1. In this case though, the EP rubber particles in TPO #2 only
have enough time to partially retract. Even though they cannot form spheres, the
particles still undergo a significant amount of retraction when compared to the fully
stretched particles. Comparing the total interfacial area, there is 21% more area in the
out-of-flow mark region than the flow mark region. Also, as the particles retract, they
coalesce into larger particles that have 1.22 times the volume of the out-of-flow mark

particles. Keeping the volume difference in mind, the ratio of the flow mark particle’s
calculated interfacial area to an equivalent sphere’s {:— value is 12.0. That value is 1.38

times smaller than the out-of-flow mark region’s ratio of 16.6.

To account for the reduction of the number of particles between regions and the increased
average volume per particle in the flow mark region, coalescence must be taking place.
As the stretched particles retract, they are coalescing with each other to form larger
particles. On average, 1.2 stretched particles in the out—of-flow mark region coalesce to

form a partially retracted flow mark particle.

2.4.3 Core Flow Front vs. Mold Wall Comparison
The volume fraction of EP rubber (total particle volume) is the same in the core flow
front region, flow mark region, and out-of-flow mark regions. For the core flow front

region, Table 2.12 shows the number averaged areas and the x and 2 dimensions.

59

Table 2.14 Average particle dimensions for the core flow front region.

 

 

 

 

 

 

 

 

 

Core Flow Front
Total Number of Particles 2068
Number Averaged 0.87
Particle Area (timz)
x-Dimension Number 1.45
Averaged Length (pm)
y-Dimension Number -
Averaged Length (um)
z-Dimension Number 0.50
Average; Lergth (um)

 

To obtain the total particle volume Vrequation for the core flow front region, the xz plane
projected area can be substituted into the volume equation for an ellipsoid (Table 2.5) and
then be multiplied by NC (the total number of particles).

VT = %CszNC (2-16)

An average value for c can then be obtained by substituting in 719 um3 for the volume
(the average volume between the flow mark and out-of-flow mark regions) and the core
flow front values from Table 2.12.

c = 0.60 am
Now that an average value for c has been obtained, the interfacial area for a single

particle can be calculated. From Table 2.5, the interfacial area formula for an ellipsoid is:

A, = flab (2'17)

2 2
where b = M

60

Substituting in the number averaged values for b and c yields:
5 = 0.55 ,um

To obtain the total interfacial area, all of the areas must be summed.

2068

ATM = H (not?) (2-18)

A, = 6495 ,um2

TOTAL

Dividing the total volume by the total number of particles yields an average particle
volume of 0.35 um3. If the assumption that volume is conserved between all of the

regions is valid, then coalescence must be taking place.

Considering the case of the core flow front to the out—of—flow mark surface, the only time
that coalescence can take place is when the particles approach the melt front and become
stretched by elongational flow. On average, 4.8 of the core particles would have to

coalesce to form one stretched particle.

In the flow mark region, the particles can coalesce twice. The first time would be as the
particles approach the melt front and become stretched by elongational flow in the same
manner as the out-of-flow mark surface case. After being stretched, they can coalesce

again as the stretched particles retract into the shorter, thicker particles.

The assumption that volume is conserved between the three regions is much more likely

to be valid with TPO #2 than TPO #1 because of the geometry of the ellipsoid particles

61

(the average y dimension of 0.60 is greater than the average z dimension of 0.50).
Therefore, particle migration between the core and wall regions is not likely and the

dominant mechanism between the core flow front and wall regions is coalescence.

62

CHAPTER 3

CONCLUSIONS AND RECOMMENDATIONS
3.1 Conclusions
The dominant mechanism of flowlmark formation for TPO #1 is a combination of
particle coalescence and retraction. As the ellipsoidal particles flow along the melt front,
they coalesce and stretch into circular cylindrical particles, with their maximum area
weighted aspect ratio reaching 11.5. As they flow from the stagnation point to the long
span of the melt front to the wall, they have enough time to retract back into disc shaped
particles aligned parallel to the flow direction, with area weighted aspect ratios of 2.14.
As the particles retract, they can coalesce again or the discs can slide underneath each
other to form stacks. A gauge for the amount of retraction that occurs in the system is the
reduction of the total interfacial area of the particles in the flow mark compared to the out
of flow mark region. For TPO #1, the reduction is 25% between the two regions.
Comparing the core flow front region to the wall regions, particle migration might be
occurring, but this is unconfirmed because no SEM micrographs show the yz plane of the
core flow front region and therefore, a true estimate of the total volume of the core flow

front particles cannot be determined.

The dominant mechanism of flow mark formation for TPO #2 is also a combination of
particle coalescence and retraction. As the ellipsoidal particles flow along the melt front,
they coalesce and stretch into elliptical cylindrical particles, with their maximum area

weighted aspect ratio reaching 15.6. As they flow from the stagnation point to the flow

63

mark region, they start to retract and form shorter, wider, and thicker elliptical cylindrical
particles oriented parallel to the flow direction with an area weighted aspect ratio of 9.27.
The particles freeze in this partially retracted state once they reach the mold wall. The

reduction of the interfacial area between the two regions is 28%.

The key to reducing the amount of flow marks in the injection molding process may be in
the rheology between the matrix and the EP rubber particles. In TPO #1 there is a
difference in molecular weight of 110,000 between the two components. Naturally, one
would also expect a large viscosity difference as well. Because of this viscosity
difference, the relaxation time for the EP rubber particles is very small and the particles
retract very quickly. This would explain why there is such a gross difference in the
morphology between the out of flow mark and flow mark regions. For TPO #2, the
molecular weight difference between the PP and EP rubber is 55,000, and therefore the
viscosities between the two are more comparable. Because of this, the EP rubber
particles are stretched further and the relaxation time is longer because the similar
viscosities hinder the particle retraction process. For this reason, the flow mark particles
in TPO #2 are still in a partially stretched state as they freeze at the mold wall. If the
rheology of the matrix and EP rubber particles could be tailored to matCh each other more
closely, the relaxation time could be large enough that if the particles flow along the long
span of the melt front, they can still reach the mold wall in a highly stretched state and

form an out of flow mark region instead of creating flow marks.

Table 3.1 shows the effects on flow line formation by changing the processing

conditions, mold geometry, and polymer rheology of the injection molding system.

Table 3.1 Effects of changing the molding conditions on flow line formation.

 

Condition Changed Effect

 

Heat mold +

 

Processing Conditions Increase injection rate

 

 

 

 

 

 

 

Decrease injection rate +
Increase part thickness +
Decrease part thickness _
Mold Geometry
Simplify mold geometry 0
Complicate mold geometry 0
Decrease the viscosity ratio between matrix/filler _
Polymer Rheology

 

Increase the viscosity ratio between matrix/filler +

 

 

 

+ reduces the formation of flow lines

— increases the formation of flow lines

0 no effect

3.2 Recommendations for Further Work

The goal of this research was to determine the dominant mechanism for flow mark
formation during the injection molding process and provide insight on how to reduce the
occurrence of their formation. Now that an explanation for the morphology changes
between the regions has been determined, future work can be done in two areas:

0 Additional micrographs in the yz plane could be taken and examined to determine

if there is significant particle migration from the wall to the core regions.

65

 

A new polymer blend of PP and EP rubber could be made that has its’ rheology
tailored to lengthen the EP rubber particle relaxation time. This blend could be
molded and tested to see if there is a reduction in the occurrence of flow mark
formation.

Using Table 3.1 as a guide, quantitative studies could be performed for each
variation of the molding conditions to see its effect on the occurrence of flow

lines for both TPO’s.

66

APPENDIX

67

APPENDIX
IMAGE ANALYSIS

1. Determination of Aspect Ratio

The SEM micrographs were scanned into TIF image files and analyzed using Sigma Scan
3.0 from Jandel Scientific. The border of a particle was outlined using the software and
then the major and minor axis lengths were calculated. The major axis is defined by
searching all of the border pixels of the particle and choosing the two pixels that are the
farthest apart. Once a major axis is defined, the minor axis is calculated. The minor axis
is drawn between the two pixels defining the longest line perpendicular to the major axis.

The aspect ratio for the particle is then calculated by

g __ major axis length
‘ minor axis length I.

The number averaged aspect ratio is then determined by

n [
[NAzz '

i=1 [TOTAL

 

The area weighted aspect ratio is determined by dividing the sum of the aspect ratio

multiplied by the particle area by the sum of the particle area

68

2. Determination of x, y, and 2 Dimensions

The x, y, and 2 dimensions of the particles were determined by measuring the major and
minor axes of the particles. The major and minor axes were defined as the x, y, or z
dimension depending on what plane the micrograph showed. For example, for a
micrograph showing the xz plane of the core flow front, the major axis would be the x
dimension and the minor axis would be the z dimension. The number averaged x, y, and

z dimension lengths are calculated with the following equations

 

n xi
xNA =
i=1 xTOTAL
_ " yl
yr. - 2—
1:1 yTOTAL

n

ZNA = 2 Zi

i=1 ZTOTAL

 

The area weighted x, y, and 2 dimensions are determined by dividing the sum of the

dimension multiplied by the particle area by the sum of the particle area

xAWA ,.
4.
i=1
. (Ii-y.)
yAWA = 1:1,.
i=1
. (44)
ZAWA = Fl"
4.

BIBLIOGRAPHY

70

Bibliography

Chan, C.-M. and J. Feng, J. Rheol., 41 (2), 319 (1997).

Chang, M.C.O., SPE Tech Papers, 40, p.36 (1994).

Choi, S.J., Schowalter, W.R., Physics of Fluids, 18 (4), 420-427, (1975).
Hamada, H. and H. Tsunasawa, J.Appl. Polym. Sci., 60 (3), 353-362 (1996).
Han, C.Y., M.D. Wolkowicz, D. Dong, and S.S. Dagli, SPE Tech Papers, 42, (1996).
Hoffman, R.L., J. Rheol., 29(5), 579-604 (1985).

Hobbs, S.Y., Polym. Eng. Sci., 36 (1 1), 1489-94 (1996).

Koh, C.J., Hookham, P., and LG. Leal, J. Fluid Mech., 266, 1-32, (1994).
Minale, M., Mewis, J., and P. Molenaers, AIChE J., 44 (4), 943-950, (1998).
Newman, 8. and Q. Trementozzi, J. Appl. Polym. Sci., 9, 3071-3089 (1965).

Riemann, R.-E., Cantow, H.-J., and C. Friedrich, Macromolecules, 30, 5476-5484,
(1997).

Yokoi, H., S. Nagami, A. Kawsak, Y. Murata, SPE ANT ECH, 1994, 368-372 (1994).

Yokoi, H., S. Nagami, A. Kawsak, Y. Murata, SPE ANT ECH, 1994, 829-832 ( 1994).

71