functional polymer foams from in-situ fibrillated …...ali rizvi doctor of philosophy department of...

Functional Polymer Foams from In-situ Fibrillated Polymer Blends

by

Ali Rizvi

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Department of Mechanical and Industrial Engineering University of Toronto

© Copyright by Ali Rizvi 2015

ii

Functional Polymer Foams from In-situ Fibrillated Polymer Blends

Ali Rizvi

Doctor of Philosophy

Department of Mechanical and Industrial Engineering

University of Toronto

2015

Abstract

Polymer foams are utilized in a range of applications such as absorption, mechanical cushioning,

thermal and sound insulation, and medical devices. In comparison with amorphous polymer

foams, foams of semicrystalline polymers are desirable in applications that require higher service

temperatures and chemical resistance, but preparing foams of semicrystalline polymers is

challenging due to their weak melt strength near processing temperatures. Blending of

immiscible polymers has emerged as a convenient route to tune the viscoelastic and

crystallization properties of polymers. Here we show that the in-situ generation of a fibrillar

morphology during physical blending of immiscible polymers can broaden the foam processing

window of semicrystalline polymers. These results formed the basis for a W.O. Patent. We

anticipate that these results will help in the development of alternative polymer formulations and

optimization of existing ones for the large-scale manufacture of semicrystalline polymer foams

with unique properties for advanced applications. For example, novel superhydrophobic and

oleophilic open-cell foams were obtained from polypropylene (PP) containing

polytetrafluoroethylene (PTFE) fibrillar dispersed phases in a fully scalable extrusion process.

These open-cell foams are technologically promising for applications such as oil-spill cleanup,

organic pollutant removal, and field water remediation.

iii

Acknowledgments

This thesis is a progression of the pioneering work by Prof. Masayuki Yamaguchi on structure-

rheology relationships of polymer blends with high aspect ratio dispersed phases. It is the

culmination of a fruitful collaboration for elucidating the effect of viscoelastic modification

through blending on foam processing, which had remained inadequately investigated since

experiments were few and restricted in the literature. I thank him for the most productive

collaboration I have ever had.

This document benefited from several insightful comments and discussion with my committee

members: Prof. Chul B. Park, Prof. Markus Bussmann, and Prof. Mohini Sain, who suggested

additional experiments, provided valuable feedback, identified errors in the scientific

presentation and generally facilitated making this document more convincing.

Takayama Nobuhisa from the Mitsubishi Rayon Co. Ltd., Japan Polypropylene Corporation,

SABIC, and Eastman Chemical Company donated materials used in the experimental

investigations. Funding for this research was provided by the members of the Consortium for

Cellular and Microcellular Plastics (CCMCP) and NSERC (Canada) Strategic and Discovery

Grants. Chandler Zhu, Lun Howe Mark, and Raymond Chu provided technical assistance and

training for several experiments presented in this document. Their support made this work

possible.

Brain-storming sessions in collaboration with the following people resulted in a few co-authored

publications: Alireza Tabatabaei assisted with laboratory-scale foam extrusion experiments,

Raymond Chu assisted with surface opening of the skin layer of open-cell foams using ultrasonic

irradiation, Seongso Bae and Pouria Vahidi assisted with pilot-scale foam extrusion experiments,

Jung Hyub (Sam) Lee assisted with contact angle measurements, Zamil K.M. Lee assisted with

fiber spinning experiments, Dr. Reza Barzegari assisted with rheological characterizations, and

Syed Hassan Mahmood assisted with solubility measurements of blowing agents.

Financial support provided in the form of Mechanical and Industrial Engineering Fellowship,

Queen Elizabeth II-GSST Award: DuPont Canada Scholarship in Science, Society of Plastics

Engineers (SPE) PerkinElmer Award, and NSERC Alexander Graham Bell Canada Graduate

Scholarship (CGS-D2) bankrolled many years of graduate school.

iv

I thank all the members of the Microcellular Plastics Manufacturing Laboratory (MPML)

between the years 2010-2014 for the friendships that I predict, will outlast our fame. Most

importantly, I thank my advisor and mentor, Prof. Chul B. Park, for setting the direction of this

research.

v

Table of Contents

Acknowledgments .......................................................................................................................... iii

Table of Contents ............................................................................................................................ v

Introduction and Overview ............................................................................................. 1 Chapter 1

1.1 Introduction ......................................................................................................................... 1

1.2 Overview ............................................................................................................................. 4

1.3 Contributions ....................................................................................................................... 7

References ....................................................................................................................................... 9

Dispersed polypropylene fibrils improve the foaming ability of a polyethylene Chapter 2

matrix ....................................................................................................................................... 13

2.1 Abstract ............................................................................................................................. 13

2.2 Introduction ....................................................................................................................... 13

2.3 Experimental ..................................................................................................................... 14

2.3.1 Materials ............................................................................................................... 14

2.3.2 Sample Preparation ............................................................................................... 15

2.3.3 Measurements ....................................................................................................... 16

2.4 Results and Discussion ..................................................................................................... 17

2.4.1 Morphology of fibrils ............................................................................................ 17

2.4.2 Shear response in the linear viscoelastic regime ................................................... 19

2.4.3 Stress growth during elongation at a constant rate ............................................... 21

2.4.4 Isothermal crystallization behaviour ..................................................................... 24

2.4.5 Foam processing and cellular morphology ........................................................... 25

2.5 Conclusion ........................................................................................................................ 27

References ..................................................................................................................................... 29

vi

Enhanced foaming ability of poly(propylene-co-ethylene) random copolymer with Chapter 3

polyethylene terephthalate fibrils ............................................................................................ 35

3.1 Abstract ............................................................................................................................. 35

3.2 Introduction ....................................................................................................................... 35

3.3 Experimental ..................................................................................................................... 37

3.3.1 Materials ............................................................................................................... 37

3.3.2 Sample Preparation ............................................................................................... 37

3.3.3 Morphological Observations ................................................................................. 38

3.3.4 Rheological Characterization ................................................................................ 38

3.3.5 Investigation of Crystallization Kinetics .............................................................. 39

3.3.6 Investigation of Crystal Structures of PP in PP/PET ............................................ 39

3.3.7 Investigation of Foaming Behaviour .................................................................... 39


3.4.1 Morphology of Fibrils ........................................................................................... 40

3.4.2 Uniaxial Extensional Flow Response ................................................................... 42

3.4.3 Shear Response in the Linear Viscoelastic Regime .............................................. 43

3.4.4 Isothermal Crystallization Kinetics ....................................................................... 45

3.4.5 Isothermal Crystallization Kinetics in Presence of Dissolved CO2 ...................... 48

3.4.6 Effect of CO2 Pressure on Formation of γ-phase in PP/PET ................................ 51

3.4.7 Foaming Behaviour of PP/PET Fibrillar Composites ........................................... 52

3.5 Conclusion ........................................................................................................................ 55

References ..................................................................................................................................... 56

Preparing fibrillated polyethylene terephthalate in poly(propylene-co-ethylene) Chapter 4

through fiber spinning for foam processing ............................................................................ 62

4.1 Abstract ............................................................................................................................. 62

4.2 Introduction ....................................................................................................................... 63

4.3 Experimental ..................................................................................................................... 64

vii

4.3.1 Materials ............................................................................................................... 64

4.3.2 Blend preparation .................................................................................................. 65

4.3.3 Fiber spinning of blend ......................................................................................... 65

4.3.4 Uniaxial extensional viscosity measurements ...................................................... 67

4.3.5 Linear viscoelastic shear response ........................................................................ 67

4.3.6 Wide-angle X-ray scattering ................................................................................. 67

4.3.7 Foam extrusion procedure ..................................................................................... 68

4.3.8 Morphology characterization ................................................................................ 69


4.4.1 Effect of the draw ratio on the morphology of PP/PET ........................................ 69

4.4.2 Uniaxial extensional flow behaviour of PP/PET .................................................. 71

4.4.3 Linear viscoelastic characterization of PP/PET .................................................... 73

4.4.4 Effect of processing on PP crystal structure in PP/PET ....................................... 76

4.4.5 Continuous foam extrusion of PP/PET ................................................................. 77

4.5 Conclusion ........................................................................................................................ 80

References ..................................................................................................................................... 81

In-situ fibrillation of CO2-philic polymers: sustainable route to polymer foams in a Chapter 5

continuous process .................................................................................................................. 86

5.1 Abstract ............................................................................................................................. 86

5.2 Introduction ....................................................................................................................... 86

5.3 Experimental ..................................................................................................................... 88

5.3.1 Materials ............................................................................................................... 88

5.3.2 Sample preparation ............................................................................................... 88

5.3.3 Measurements ....................................................................................................... 88

5.3.4 Continuous Foam Processing ................................................................................ 89


viii

5.4.1 Morphology of PP/PTFE fibrillar blends .............................................................. 91

5.4.2 Strain-induced hardening in uniaxial extensional flow ........................................ 93

5.4.3 Enhancement of CO2 sorption in the PP/PTFE fibrillar blends ............................ 95

5.4.4 Continuous foam processing of PP and PP/PTFE fibrillar blends ........................ 97

5.4.5 Minimum admissible temperature for PP and PP/PTFE fibrillar bends ............. 102

5.5 Conclusions ..................................................................................................................... 103

Crystallization-induced structural heterogeneities for open-cell foam extrusion of Chapter 6

polypropylene/polytetrafluoroethylene fibrillar blend .......................................................... 109

6.1 Abstract ........................................................................................................................... 109

6.2 Introduction ..................................................................................................................... 110

6.3 Experimental ................................................................................................................... 112

6.3.1 Materials ............................................................................................................. 112

6.3.2 Blend preparation ................................................................................................ 112

6.3.3 Morphology characterization .............................................................................. 112

6.3.4 Differential Scanning Calorimetry (DSC) characterization ................................ 113

6.3.5 Mass uptake of CO2 ............................................................................................ 114

6.3.6 Open-cell content determination ......................................................................... 114

6.3.7 Foam extrusion procedure ................................................................................... 114

6.4 Results and Discussion ................................................................................................... 115

6.4.1 Preparation of open-cell foams of PP/PTFE fibrillar blends .............................. 115

6.4.2 Free energy difference for PP transcrystallization on PTFE fibrils and

spherulitic crystallization in the bulk .................................................................. 117

6.4.3 Isothermal crystallization kinetics of PP/PTFE fibrillar blends ......................... 122

6.4.4 Isothermal crystallization kinetics of PP/PTFE fibrillar blends with dissolved

CO2 ...................................................................................................................... 125

6.4.5 CO2-philicity of PTFE in PP/PTFE fibrillar blends ............................................ 128

6.4.6 Characterization of open-cell foams ................................................................... 130

ix

6.5 Conclusion ...................................................................................................................... 132

Superhydrophobic and oleophilic open-cell foams from fibrillar blends of Chapter 7

polypropylene and polytetrafluoroethylene .......................................................................... 138

7.1 Abstract ........................................................................................................................... 138

7.2 Introduction ..................................................................................................................... 139

7.3 Experimental ................................................................................................................... 141

7.3.1 Materials ............................................................................................................. 141

7.3.2 Blend preparation ................................................................................................ 141

7.3.3 Foam extrusion procedure ................................................................................... 142

7.3.4 Morphology characterization .............................................................................. 142

7.3.5 Oil absorption test ............................................................................................... 143

7.3.6 Mechanical properties of the foam ..................................................................... 143

7.3.7 Contact angle determination ............................................................................... 144

7.3.8 Ultrasound treatment ........................................................................................... 144

7.4 Results and Discussion ................................................................................................... 145

7.4.1 Preparation of open-cell foams of PP/PTFE fibrillar blends .............................. 145

7.4.2 Superhydrophobicity of the open-cell foams ...................................................... 147

7.4.3 Oil uptake study .................................................................................................. 149

7.4.4 Effect of mechanical squeezing for oil recovery ................................................ 152

7.4.5 Kinetics of oil uptake .......................................................................................... 155

7.5 Conclusion ...................................................................................................................... 157

Conclusions and Recommendations ........................................................................... 165 Chapter 8

8.1 Conclusions ..................................................................................................................... 165

8.2 Recommendations ........................................................................................................... 167

Appendix A ................................................................................................................................. 169

Appendix B ................................................................................................................................. 171

1

Chapter 1

Introduction and Overview

1.1 Introduction

Polymer foams are commercially and technologically important materials characterized by

properties such as light weight, excellent strength/weight ratio, thermal and sound insulation, and

energy absorption performance. 1

These properties make foams advantageous structures for

various applications in the construction, packaging, automotive, and medical industries.

The foaming of polymers typically involves the following steps: 1) dissolution of a blowing

agent into a polymer; 2) generation of bubbles or cells by phase separation of the blowing agent

from the polymer; and 3) stabilization of the porous or cellular structure.

The primary mode of deformation during cell growth in polymer foaming is an extensional

deformation. Consequently, the viscoelastic properties of a polymer melt and its ability to

crystallize are important in foam processing. During the foaming process, the melt viscosity must

increase, either by cooling the melt or through strain hardening in extensional flow, to a degree

suitable for stabilizing the growing cells and preventing them from rupturing.

The use of semicrystalline polymers for foaming applications is restricted because these

polymers generally exhibit inadequate rheological properties at processing temperatures. Unlike

amorphous polymers, semicrystalline polymers show a rapid change in viscosity and melt

strength around the melting transition temperature. Therefore, accessing the processing

temperature window for generating stable foams, where the polymer is stiff enough to prevent

cells from rupturing but also soft enough to deform under relatively small stresses during bubble

growth, is challenging. 2

Methods have been proposed to broaden the foam processing window of semicrystalline

polymers. Crosslinking of semicrystalline polymers is an effective strategy for increasing their

melt strength and strain hardening response in extensional flows but the approach is undesirable

as it renders the polymer non-recyclable. 3, 4

Furthermore, the degree of crosslinking must be

carefully controlled because an excessive degree of crosslinking can restrict cell growth. Long-

chain branching is also effective in increasing the melt strength and strain hardening behaviour

2

of the semicrystalline polymers but the cost of such resins are several folds higher than their

linear counterparts. For example, long chain branched polypropylene (PP) is effective in

producing low density foams of PP, however, commercially available high melt strength PP is at

least twice as expensive as linear PP. 5, 6, 7, 8

Melt viscosity modification in semicrystalline

polymers such as poly(lactic acid) (PLA) has been accomplished using chain extenders,

however there is a potential of gel formation which dramatically reduces the material flowability

characteristics. 9 Exfoliation of nanoparticles such as organically modified layered nanoclay,

carbon nanofibers, carbon nanotubes etc. in semicrystalline polymers has been effective in

improving the melt viscosity and elasticity, 10, 11, 12, 13, 14

however, exposure to such nanoparticles

poses significant health hazard. Furthermore, exfoliation of nanoparticles in polymers is

challenging and requires fine tuning of the nanoparticle surface chemistry as well as their

synthesis and processing conditions. 15, 16, 17

Despite the availability of a number of methods to compensate for the poor melt strength and

weak strain-induced hardening of semicrystalline polymers, it still remains a challenge to

improve their rheological properties for foam processing applications. The deficiencies of these

existing strategies call for a need to develop methods to improve the foaming ability of

semicrystalline polymers in an environmentally-sustainable, inexpensive, and completely

scalable process.

Blending of immiscible polymers has emerged as an effective tool to control the rheological

properties of polymers. The properties of immiscible polymer blends are closely related to their

morphology. Morphological control of polymer blends arise from: a) the complexity of thermal

and flow fields experienced by the melt during physical blending, b) the competition between

droplet break-up and coalescence of the dispersed phase domains; and c) the viscoelastic

properties of the phases. 18, 19

It is believed that dispersed phases break up into smaller domains

by stretching into sheets. 20, 21

Once the domains reach the micrometer length scale, they may be

stretched into filaments and then break into droplets. 22

At this length scale, the interfacial tension

becomes important and causes the edges of the stretched domains to be pulled in, yielding

filament-type or cylindrical morphologies referred to as fibrils.

The fibrillar morphology of polymer blends is particularly important. Owing to their large aspect

ratio, the fibrils are able to bend substantially in response to interfibrillar interactions during

3

flow. 23

At a concentration greater than the random close packing of fibers this bending results in

the formation of a disordered physical network characterized by superior mechanical properties.

24, 25, 26, 27, 28, 29 The presence of a fibrillar network defined by topological (entangled) interactions

is expected to create additional and large contributions to the viscoelasticity of the polymer

matrix. Studies demonstrate that the presence of a fibrillar network in a polymer-host results in a

marked enhancement of strain hardening behaviour in extensional flow. The mechanism of strain

hardening in extensional flow for the fibrillar network is analogous to that of long chain

branched polymers, where strain hardening originates from restricted stretching of the backbone

between branching points. 30

Thus, a network structure that exhibits strong resistance against

stretching should show strain hardening. It seems that the origins of strain hardening in a fibrillar

networks is generally related to the inability of the fibrils to disentangle quickly enough with the

strain and follow the deformation.

From both an experimental and theoretical standpoint, the morphology evolution of polymer

blends has been described and excellent reviews exist. 31, 32

Droplet deformation during flow is

controlled by the rheological properties of the components and by the ratio between the applied

hydrodynamic stress and the interfacial stress, , where is the interfacial tension and

is the unperturbed droplet radius. In the case of a steady shear field applied to Newtonian

phases, the capillary number ( ) is 33

(1.1)

The shear stress is defined as , with the viscosity of the continuous phase, and

the shear rate and is the interfacial tension between two polymers. The second dimensionless

number controlling droplet deformation is simply the viscosity ratio , where is the

viscosity of the dispersed phase. Taylor showed that for , the shape of a droplet in a steady

shear field becomes unstable for a critical value , of the capillary number. 34

When the

magnitude of is lower than , the interfacial stress overrules the hydrodynamic stress and

droplets do not break but undergo only slight deformation from equilibrium shape. Upon

exceeding the hydrodynamic stress dominates the interfacial stress, and droplets continue

to stretch until they break due to growing disturbances at the interface. Finally, when

, where is about 2 for simple shear flow and 5 for elongational flow 35

the interfacial

4

stress becomes negligible with respect to the hydrodynamic stress and the droplet undergoes

affine deformation with respect to the applied macroscopic strain i.e., the droplet acts as a

material element, and it is stretched into an extended filamentous structure. Upon flow cessation,

the filamentous structure recoils back into spherical shape by the interfacial tension in the molten

state. However, a prompt solidification process prohibits this deformation upon flow cessation

and results in a fibril dispersion system. 18, 36

The quenched, meta-stable fibrillar structure of the

dispersed domains is ultimately responsible for the properties observed.

Modification of crystallization and rheological characteristics of a host-polymer through

microstructure control of the dispersed phase can have substantial effects on the foam processing

of the matrix. This thesis shows the relation between polymer blend morphology and foam

processability of semicrystalline polymers, with specific focus on the fibrillar morphology. The

aim is to better understand the role of fibrillar domains on the viscoelastic properties and

crystallization kinetics of the matrix polymer so that superior polymer formulations can be

developed for large-scale foam processing.

1.2 Overview

This work is organized as follows:

In Chapter 2, in-situ fibrillation of polypropylene (PP) in a metallocene-catalyzed polyethylene

(mPE) matrix is performed on a small-scale twin-screw compounder. The foaming ability of neat

mPE is compared with mPE containing fibrillated PP domains (mPE/fibrillated-PP) and mPE

containing spherical PP domains (mPE/spherical-PP). The mPE/fibrillated-PP shows the best

foam morphology with the highest number of bubbles per unit volume. This enhancement in

foaming ability is explained through rheological and crystallization studies. Winter-Chambon‘s

analysis is used to identify that the fibrils form a percolated network at a fibril content of 4.5

wt%. The network of entangled fibrils results in strain hardening in uniaxial extension. Shear

thickening responses are observed in shear flow. Such responses are not observed for mPE or

mPE/spherical-PP. Oscillatory shear flow investigation of the isothermal melt crystallization

reveals a two decade decrease in the time for the onset of crystallization in the mPE/fibrillated-

PP and a decade decrease in mPE/spherical-PP relative to neat mPE. The observed enhancement

in foaming ability of mPE/fibrillated-PP is attributed to the concurrent increase in strain-induced

hardening response and improved crystallization kinetics.

5

To demonstrate that the strategy of fibrillating dispersed phases in polymer blends to improve the

foaming ability is general and can be extended to other semicrystalline polymers, in Chapter 3,

in-situ fibrillation of polyethylene terephthalate (PET) in a PP random copolymer matrix is

performed on a small-scale twin-screw compounder. The fibrillar blend exhibits strain hardening

behaviour in uniaxial extensional flow. The strain hardening response is attributed to the

formation of a disordered physical network of PET fibrils, confirmed by applying Winter-

Chambon‘s analysis on the linear viscoelastic oscillatory shear data. Differential Scanning

Calorimetry (DSC) at ambient pressures and at elevated CO2 pressures reveals that the PET

fibrils enhance the kinetics of isothermal crystallization of PP random copolymer under both

ambient and elevated CO2 pressures. Wide-angle X-ray scattering (WAXS) study shows that

isothermal crystallization of the PP/PET fibrillar composite in the presence of pressurized CO2

facilitates the formation of γ-phase crystals in the PP matrix. Foaming of the PP/PET fibrillar

composite shows that the presence of fibrils lead to marked enhancements in the foaming ability

of the PP random copolymer. The improvements in foaming are explained based on the

rheological and crystallization results.

While preparation of fibrillar blends in a small-scale twin-screw extruder is effective for research

and development of material formulations, the results do not necessary scale-up to larger-volume

twin-screw extruders because the equipment design attributes and processing conditions are

fundamentally different. Consequently, in Chapter 4, a fully-scalable fiber spinning process is

used to fibrillate PET in PP random copolymer. Macroscopic extensional flows applied during

fiber spinning cause both the PP matrix and the PET domains to stretch and orient in the flow

direction. When these blend fibers are isothermally treated at a temperature between the melting

temperatures of the two components, PP relaxes while the PET fibrils maintain their

morphology. These fibrils influence the viscoelastic behaviour of the PP matrix. The uniaxial

extensional viscosity measurements show strain hardening behaviour for the PP/fibrillated-PET,

not observed for the neat PP matrix or for PP/PET blends when the PET domains are spherical.

This strain hardening behaviour is attributed to the formation of a disordered physical network of

flexible PET fibrils. The oscillatory shear behaviour in the linear viscoelastic region is studied to

understand the percolation properties of PET fibrils in the PP matrix. When the PET domains are

fibrillated, the storage (G‘) and loss (G‘‘) moduli show an increase and their slopes decrease at

low frequencies compared to neat PP or when the PET domains are spherical, indicating that the

6

fibril network responds elastically over long timescales. The WAXS data shows the presence of

γ-polymorph crystals of PP in both PP and PP/PET after the fiber spinning process. Foam

extrusion is used as a model polymer process to study the effect of the PET fibrils on the

processability of PP. Results reveal that the presence of PET fibrils in PP yields foams that

exhibit up to two orders of magnitude higher number of bubbles per unit volume and up to a

five-fold increase in the expansion ratio relative to the neat PP. Enhancing the foaming ability of

polymer blends by fibrillating the dispersed phase using fiber spinning is technologically

promising.

Polytetrafluoroethylene (PTFE) has a tendency to undergo plastic deformation upon application

of a weak flow field because the chains packed in PTFE crystals are characterized by weak

cohesive forces. In Chapter 5, large-scale production of PP/PTFE fibrillar blends is

accomplished in a twin-screw extruder. Unlike the method to prepare fibrillar blends of PP/PET

which required a second step using a fiber spinning line to elongate the PET disperse phase

domains (explained in Chapter 4), the shear and extensional stresses applied during twin-screw

extrusion of PTFE with PP are sufficient to elongate the PTFE domains. Dynamic oscillatory

shear experiments confirm that the PTFE elongates into fibrils during blending and forms a

physical network of entanglements in the melt which results in a low frequency plateau in the

elastic modulus. Uniaxial extensional flow experiments show strain-induced hardening

behaviour. CO2 solvency in the PP/PTFE fibrillar blend is enhanced due to the CO2-philic

character of PTFE. Remarkably, adding only 0.3 wt% of PTFE is sufficient to markedly enhance

the CO2 sorption capacity of the matrix. Continuous foam extrusion of the in-situ fibrillar blend

reveals a three orders of magnitude increase in bubble density, a ten-fold increase in volume

expansion ratio, and a marked broadening of the foaming window with respect to neat PP. These

improvements are attributed to the simultaneous enhancement in CO2 solvency and strain

hardening behaviour of the melt in the in-situ fibrillar blend.

Within a narrow processing window, open-cell foams of PP/PTFE can be obtained in extrusion

with extremely high open-cell contents. In Chapter 6, we report a novel approach to produce

low-density open-cell foams of PP/PTFE in continuous foam extrusion. The foams exhibit open-

cell contents as high as 97.7% and mass densities as low as 0.07 g/cm3. Crystallization of the PP

matrix around the PTFE fibrils creates the structural heterogeneity required for inducing cell

opening. The thermodynamics and kinetics of transcrystallization of PP on PTFE fibrils is

7

investigated. The growth rate of the transcrystalline layer on PTFE fibrils remains the same as

the growth rate of spherulites in the bulk and the application of Lauritzen-Hoffman‘s growth

theory for polymer crystals yields similar results. However, the induction time for

transcrystallization is shorter than the induction time for bulk crystallization. Ishida‘s approach is

used to quantify the free energy difference for transcrystallization and for spherulitic

crystallization in the bulk and the magnitudes are calculated to be 7.8×10-4

J∙m-2

and 1.3×10-3

J∙m-2

, respectively. The lower free energy difference seen for transcrystallization suggests it is a

more thermodynamically favorable crystallization process than bulk crystallization. The

increased uptake of CO2 by PP/PTFE increases the plasticization of the sample, and facilitates

cell wall opening during the bubble growth stage of the foaming process.

These open-cell foams of PP/PTFE show excellent oil uptake capacity. In Chapter 7, the

superhydrophobic and oleophilic behaviour of the PP/PTFE open-cell foams are studied. The oil

uptake capacity of the open-cell foams is characterized using various petroleum products such as

octane, gasoline, diesel, kerosene, light crude oil and heavy crude oil from water. Ultrasonic

irradiation is used to increase the surface porosity of the thin, impervious, foam ‗skin‘ layer to

enhance the uptake kinetics of the open-cell foam. The reusability of the open-cell foams after oil

absorption is investigated by subjecting the foams to a cyclic-compression stress-strain test as a

controlled analogue for mechanical ―squeezing‖ required for extracting the oil from the open-cell

foam. To the best of our knowledge, these PP-based open-cell foams outperform PP-based

absorbents conventionally used for oil-spill cleanup applications such as nonwoven PP fibers or

melt-blown PP pads.

Finally, in Chapter 8, the conclusions of this thesis are drawn and suggestions for future work

are made.

1.3 Contributions

1. Demonstration of the broadening of the foaming window of semicrystalline polymers by

controlling the morphology of the dispersed phase domains in immiscible polymer

blends.

2. Fabrication of superhydrophobic and oleophilic PP-based open-cell foams in a

completely scalable foam extrusion process. To the best of our knowledge, the resultant

8

open-cell foams exhibit oil uptake capacities that exceed those of PP-based absorbents in

literature.

9

References

1. Ashby, M. F.; Gibson, L. J. In Cellular Solids; Cambridge University Press: Cambridge, UK,

1997.

2. Guo, Y.; Hossieny, N.; Chu, R. K. M.; Park, C. B.; Zhou, N. Critical processing parameters

for foamed bead manufacturing in a lab-scale autoclave system. Chem. Eng. J. 2013, 214,

180-188.

3. Zhai, W. T.; Wang, H. Y.; Yu, J.; Dong, J. Y.; He, J. S. Cell coalescence suppressed by

crosslinking structure in polypropylene microcellular foaming. Polym. Eng. Sci. 2008, 48,

1312.

4. Romania, F.; Corrierib, R.; Bragab, V.; Ciardelli, F. Monitoring the chemical crosslinking of

propylene polymers through rheology. Polymer 222, 43, 1115-1131.

5. Zhai, W. T.; Wang, H. Y.; Yu, J.; Dong, J. Y.; He, J. S. Foaming behaviour of isotacic

polypropylene in supercritical CO2 influenced by phase morphology via chain grafting.

Polymer 2008, 49, 3146.

6. Nam, G. J.; Yoo, J. H.; Lee, J. W. Effect of long-chain branches of polypropylene on

rheological properties and foam-extrusion performances. J. Appl. Polym. Sci. 2005, 96, 1793-

1800.

7. Gotsis, A. D.; Zeevenhoven, B. L. F.; Tsenoglou, C. Effect of long branches on the rheology

of polypropylene. J. Rheol. 2004, 48, 895-915.

8. Tian, J.; Yu, W.; Zhou, C. The preparation and rheology characterization of long chain

branching polypropylene. Polymer 2006, 23, 7962–7969.

9. Gruber, P. R.; Kolstad, J. J.; Witzke, D. R.; Hartmann, M. H.; Brosch, A. L. Viscosity-

modified lactide polymer composition and process for manufacture thereof. US 5594095 A,

July 27, 1994.

10

10. Okamoto, M.; Nam, P. H.; Maiti, P.; Kotaka, T.; Nakayama, T.; Takada, M.; Ohshima, M.;

Usuki, A.; Hasegawa, N.; Okamoto, H. Biaxial flow-induced alignment of silicate layers in

polypropylene/clay nanocomposite foam. Nano Lett. 2001, 1, 503-505.

11. Lee, L. J.; Zeng, C.; Cao, X.; Han, X.; Shen, J.; Xu, G. Polymer nanocomposite foams.

Comp. Sci. Tech. 2005, 65, 2344-2363.

12. Zheng, W.; Lee, Y. H.; Park, C. B. Use of nanoparticles for improving the foaming

behaviours of linear PP. J. App. Polym. Sci. 2010, 117, 2972-2979.

13. Leung, S. N.; Wong, A.; Park, C. B.; Zong, J. Ideal surface geometry of nucleating agents to

enhance cell nucleation in polymer foaming. J. App.Polym. Sci. 2008., 108, 3997-4003.

14. Leung, S. N.; Wong, A.; Guo, Q.; Park, C. B.; Zong, J. H. Change in the critical nucleation

radius and its impact on cell stability during polymeric foaming processes. Chem. Eng. Sci.

2009, 64, 4899-4907.

15. Zhai, W.; Kuboki, T.; Wang, L.; Park, C. B. Cell structure evolution and the crystallization

behaviour of polypropylene/clay nanocomposites foams blown in continuous extrusion. Ind.

Eng. Chem. Res. 2010, 49, 9834.

16. Giannelis, E. P. Polymer layered silicate nanocomposites. Adv. Mater. 1996, 8, 29-35.

17. Zeng, C.; Han, X.; Lee, L. J.; Koelling, K. W.; Tomasko, D. L. Polymer-clay nanocomposite

foams prepared using carbon dioxide. Adv. Mater. 2003, 15, 1743-2747.

18. Favis, B. D. Factors influencing the morphology of immiscible polymer blends, in melt

processing. In Polymer; Paul, D. R., Bucknall, C. B., Eds.; Wiley-Interscience: New York,

1999; Vol. 1.

19. Favis, B. D.; Chalifoux, J. P. Influence of composition on the morphology of

polypropylene/polycarbonate blends. Polymer 1988, 29, 1761-1767.

20. Scott, C. E.; Macosko, C. W. Morphology development during the initial stages of polymer-

11

polymer blending. Polymer 1995, 36, 461-470.

21. Sundararaj, U.; Dori, V.; Macosko, C. W. Sheet formation in immiscible polymer blends.

Polymer 1995, 36, 1957-1968.

22. Janssen, J. M. H.; Meijer, H. E. H. Dynamics of liquid-liquid mixing: A 2-zone model.

Polym. Eng. Sci. 1995, 35, 1766-1780.

23. Raghavan, S. R.; Douglas, J. F. The conundrum of gel formation by molecular nanofibers,

wormlike micelles, and filamentous proteins: gelation without cross-links? Soft Matter 2012,

8, 8539-8546.

24. Kharchenko, S. B.; Douglas, J. F.; Obrzut, J.; Grulke, E. A.; Migle, K. B. Flow-induced

properties of nanotube-filled polymer materials. Nature Materials 2004, 3, 564-568.

25. Bicerano, J.; Douglas, J. F.; Brune, D. A. Model for the viscosity of particle dispersions. J.

Macromol. Sci. Rev. Macromol. Chem. Phys. 1999, 561–642.

26. Ali, M. A. B. N.; Okamoto, K.; Yamaguchi, M.; Kasai, T.; Koshirai, A. Rheological

properties for polypropylene modified by polytetrafluoroethylene. Journal of Polymer

Science Part B: Polymer Physics 2009, 47, 2008-20014.

27. Fakirov, S.; Evstatiev, M. Microfibrillar reinforced composites—new materials from

polymer blends. Adv. Mater. 1994, 6, 395-398.

28. Rizvi, A.; Park, C. B.; Yamaguchi, M. Resin composition foam and method for producing

same. WO2013137301, March 13, 2013.

29. Rizvi, A.; Park, C. B.; Yamaguchi, M. Resin foams and production methods thereof.

JP2013514463, March 13, 2012.

30. Inkson, N. J.; McLeish, T. C. B.; Harlen, O. G.; Groves, D. J. Predicting low density

polyethylene melt rheology in elongational and shear flows with ―pom-pom‖ constitutive

equations. J. Rheol. 1999, 43, 873-896.

12

31. Tucker, C. L.; Moldenaers, P. Microstructural evolution in polymer blends. Annu. Rev. Fluid

Mech. 2002, 34, 177–210.

32. Stone, H. A.; Stroock, A. D.; Ajdari, A. Engineering flows in small devices: Microfluidics

toward a lab-on-a-chip. Annu. Rev. Fluid Mech. 2004, 36, 381-411.

33. Taylor, G. I. The formation of emulsions in definable fields of flow. Proc. R. Soc. London,

Ser. A 1934, 146, 501-523.

34. Grace, H. P. Dispersion phenomena in high viscosity immiscible fluid systems and

application of static mixers as dispersion devices in such systems. Chem. Eng. Comm. 1982,

14, 225-277.

35. Janssen, J. M. H.; Meijer, H. E. H. Droplet breakup mechanisms: Stepwise equilibrium

versus transient dispersion. J. Rheol. 1993, 37, 597-608.

36. Manas-Zloczower, I. In Mixing and Compounding of Polymers, 2nd ed.; Meijer, H. E. H.,

Janssen, J. M. H., Anderson, P. D., Eds.; Hanser: Munich, 2009.

13

Chapter 2

Dispersed polypropylene fibrils improve the foaming ability of a polyethylene matrix1

2.1 Abstract

We compare the foaming ability of metallocene catalyzed polyethylene (mPE) with that of mPE

containing fibrillated polypropylene (PP) domains (mPE/fibrillated-PP) and mPE containing

spherical PP domains (mPE/spherical-PP). We observe that mPE/fibrillated-PP shows the best

foam morphology with the highest number of bubbles per unit volume. We explain this

enhancement in foaming ability through rheological and crystallization studies. We identify that

the fibrils form a percolated network at a fibril content of 4.5 wt% using the Winter-Chambon

analysis. The network of entangled fibrils results in strain hardening in uniaxial extension. Shear

thickening responses are observed in shear flow. Such responses are not observed for mPE or

mPE/spherical-PP. Oscillatory shear flow investigation of the isothermal melt crystallization

reveals a two decade decrease in the time for the onset of crystallization in the mPE/fibrillated-

PP and a decade decrease in mPE/spherical-PP relative to neat mPE. We attribute the

enhancement in foaming ability of mPE/fibrillated-PP to the concurrent increase in strain-

induced hardening response and improved crystallization kinetics.

2.2 Introduction

Polymer blends represent a large and rapidly growing fraction of all plastics produced. Various

forms of polymer blends and filled polymers have been employed in electronic 1

and optical

devices, 2

and structural materials. 3

The components are chosen to tune or accentuate selected

properties which may be thermal, electro-optical, rheological and/or mechanical properties.

Fundamentally, the behaviour of a polymer matrix depends critically on the morphology of the

dispersed phase. 4

Divergence in performance from seemingly identical blend compositions is

often attributed to morphological differences. Therefore, morphological control is a critical

parameter in the optimization of the performance of a polymer matrix.

1 Reproduced From: Rizvi, A.; Park, C.B., ―Dispersed polypropylene fibrils improve the foaming ability of a

polyethylene matrix‖, Polymer 2014, 55, 4199-4205.

14

Of particular importance is the fibrillar morphology where the dispersed phase deforms into

extended submicron filamentous structures in a polymeric-host. Owing to their high aspect ratio,

the fibrils are able to bend substantially in response to interfibrillar interactions during flow. 5

At

a concentration greater than the random close packing of fibers, this bending causes the

formation of a disordered physical network characterized by superior mechanical properties. 6, 7, 8,

9, 10, 11 The presence of a fibrillar network defined by topological (entangled) interactions is

expected to create additional and large contributions to the viscoelasticity of the polymer matrix.

In this study, we characterize the effect of two solid-state dispersed phase morphologies, i.e., 1)

droplet/spherical domains, and 2) fibrillar structures, on the rheological and thermal properties of

polymer blends of metallocene polymerized semicrystalline ethylene-α-olefin copolymer (mPE)

and polypropylene (PP) where the mPE is the matrix and the PP is the dispersed phase. The

knowledge of the rheological and crystallization properties of polymers is of direct practical

importance for polymer processing operations. An example is polymer bead foaming which is

purposely performed at a temperature slightly below the melting temperature of a semicrystalline

polymer to facilitate physical gel formation through crystallization, 12

so that the polymer is stiff

enough to prevent foam bubbles from rupturing in the initial stages of the process but also soft

enough to deform under relatively small stresses during bubble growth. 13

Using polymer

foaming as a model for polymer processing, this study aims to compare how changes in

rheological response and crystallization behaviours induced by the PP dispersed phase

morphologies affects the foaming of the mPE matrix.

2.3 Experimental

2.3.1 Materials

The matrix polymer employed in this study is metallocene polymerized semicrystalline ethylene-

α-olefin copolymer commercially available as Dow Engage ® 8445 by Dow Chemicals with a

density of 0.902 g/cm3 at 23°C and a melt flow rate (MFR) of 3.5 g/10 min at 190°C/2.16 kg.

The fibrillated polymer in this study is isotactic polypropylene homopolymer commercially

available as Novatec ® PP FY4 by Japan Polypropylene with a MFR = 5 g/10 min at 230°C/2.16

kg. Carbon dioxide (CO2) is purchased from Linde Gas with purity in excess of 99%. Xylene is

purchased from Sigma-Aldrich Chemicals and used as received.

15

The melting temperatures ( ) of mPE and PP are determined using Differential Scanning

Calorimetry (DSC) under a nitrogen atmosphere. A heat-cool-heat experiment is performed from

25°C to 200°C. Heating and cooling rates of 10°C/min are used. The is measured at the

midpoint of the melting peak obtained during the second heating cycle. The of mPE and PP

are found to be 102°C and 165°C, respectively.

2.3.2 Sample Preparation

Fig. 2.1 Photograph of the co-rotating twin-screw extruder (DSM Xplore microcompounder)

with a built-in back-flow channel for material recirculation.

We blend mPE and PP using a co-rotating twin-screw extruder (DSM Xplore micro-

compounder) with a built-in back-flow channel which enables material recirculation. A

photograph of the twin-screw extruder is presented in Fig. 2.1. An initial blending temperature of

190°C is used and the polymers are blended for 8 min to get well dispersed spherical domains of

PP in the mPE matrix. To fibrillate the dispersed PP, the temperature is reduced to 155°C using

compressed-air active cooling and the polymers are mixed for another 8 min before collecting

the samples at a draw ratio of 1:5. The PP content is chosen to range from 0 to 10 wt%. In order

to prepare mPE blends with spherical PP domains, the obtained samples are compressed into a

flat sheet using a laboratory hydraulic press (Carver SC7620) at 180°C under a pressure of 10

MPa for 10 min followed by cooling at room temperature.

16

2.3.3 Measurements

The morphology of the blends is examined using a scanning electron microscope (SEM, JEOL

6060). Prior to the observation, the surface of cryogenically fractured samples is subjected to a

solvent vapour etching process using xylene. Samples are exposed to xylene vapours for 2 h at

65°C. This type of treatment preferentially etches the relatively less crystalline, low-melting,

mPE than the crystalline, high-melting PP, which leads to bringing up these domains and

increasing differentiation in observation. The etched samples are then coated with a thin layer of

platinum using a sputter coater.

Measurements of uniaxial extensional viscosity are made using an Extensional Viscosity Fixture

(EVF, TA Instruments) attached to an Advanced Rheometric Expansion System (ARES).

Samples are tested at strain rates of 0.01, 0.1 and 1 s-1

and at a sample environment temperature

of 130°C. We empirically find that this temperature provides us with the most reproducible and

consistent results. Strain rates higher than 1 s-1

are experimentally inaccessible because sample

breakage occurs in the initial stages and homogeneous extension cannot be accomplished.

To determine the linear viscoelastic properties, oscillatory shear experiments are performed

using the ARES rheometer with a controlled strain mode, and cone-and-plate geometry. The

diameter of the cone is 25 mm and the cone angle is 5°. The shear elastic modulus, , and

viscous modulus, , are evaluated at a temperature of 120°C under a nitrogen atmosphere to

avoid oxidative degradation. The temperature is chosen because it is above the melting

temperature of mPE but below the melting temperature of PP allowing preservation of the

dispersed phase morphology during experiments. Frequency sweeps between 0.1 and 100 rad/s

are carried out at strains within the linear viscoelastic range. Repeated sweeps with increasing

and decreasing frequencies show that the material is stable under the measurement conditions.

Prolonged heating at 120°C of the samples in the rheometer results in no change in the

viscoelastic data.

Small Amplitude Oscillatory Shear (SAOS) tests are conducted to follow the isothermal

crystallization behaviour of the samples. The elastic modulus is monitored as a function of time

using the cone-and-plate rheometer at a constant angular frequency of 10 rad/s. The amplitude of

the strain applied is 0.5%. Once the samples are melted at 140°C for 5 min under nitrogen the

gap is reduced to 700 µm, the samples are trimmed and allowed to relax for 5 min at 140°C.

17

Then the samples are cooled to the crystallization temperatures of 95°C by manually lowering

the temperature rapidly to 115°C followed by steps of -1°C to prevent undershoot below the

experimental temperature of 95°C. Otherwise, crystals would start forming earlier than at the

experimental temperature and shorten the crystallization time. The cooling process takes 2 min.

For the SAOS experiments, the sample temperature in the environmental chamber under active

flow of nitrogen is calibrated by placing a thermocouple between the upper and lower fixtures

and filling the gap with thermal paste to ensure thermal contact. Thermal expansion of the

rheometer tool during heating is accounted for by the rheometer software to ensure the gap size

remaines constant.

For the foaming experiments, samples are shaped into disks of 1 cm diameter and 400 μm

thickness. The shaped samples are placed in a custom-built, temperature-regulated, pressure

vessel connected to a syringe pump filled with CO2. The vessel is purged with CO2 prior to

pressurization. The samples are then saturated at a temperature of 95°C and at pressures 4.1

MPa, 5.5 MPa, 6.9 MPa, 10.3 MPa, 13.8 MPa and 17.3 MPa. After an exposure time of 1 h,

compressed CO2 is quickly released from the pressure vessel, the samples removed from the

vessel and immersed in a water bath to quench the foam structure. To characterize the obtained

foams, the foam volume expansion ratio, ⁄ , is calculated where and are the mass

densities of samples before and after foaming, respectively, determined using the water

displacement method according to ASTM-D792. Additionally, the number of bubbles per unit

volume, called the cell density, is also calculated using SEM images of the foams.

2.4 Results and Discussion

2.4.1 Morphology of fibrils

Polymer blending in a twin-screw extruder is a complex process. Accurate determination of the

nature of flow fields while bringing heat transfer considerations into account makes the

identification of parameters that are closely related to polymer deformation processes (e.g.

capillary number, viscosity ratio, applied stress, deformation rate and total strain), difficult.

Consequently, we will make use of the duration and temperature of mixing to discuss our

strategy of fibrillating the domains of a dispersed phase in a polymer matrix. While this is

unsatisfactory to a certain extent, these are the only accessible parameters that are accurately

controllable for all samples.

18

Fig. 2.2 SEM images of surface etched sample of mPE/PP: a) mPE/fibrillated-PP (95/5 wt%), b)

mPE/spherical-PP (95/5 wt%).

We employ the method of Fakirov and Evstatiev 9

for inducing in-situ fibrillation of a blend of

immiscible polymers in a two-step process: (1) Dispersion Stage: The two immiscible polymers

with different melting temperatures are blended in a twin-screw extruder at a temperature above

the melting temperatures of both components. This step aims to get reasonably well-dispersed

spherical domains of PP in the mPE matrix. (2) Fibrillation Stage: The temperature of the twin-

screw extruder is reduced to a temperature between the melting temperatures of the two

components of the blend while maintaining flow. The lower melting component remains as a

melt, whereas the dispersed phase with the higher melting temperature starts to increase in

viscosity with solidification. The shear and extensional flow fields applied during blending in the

twin-screw extruder cause the spherical domains of the dispersed phase to elongate. 14, 15

Subsequent cold drawing further extends the dispersed phase domains. Crystallization becomes

more active with elongation due to strain-induced crystallization. Eventually, the elongated

domains of the dispersed phase are too crystalline to undergo stretching and their fibrillar

structure is said to be frozen. Consequently, recoiling of the fibrils is suppressed and the matrix

reinforced with flexible fibrils of the higher melting component is obtained.

The morphology of mPE/fibrillated-PP is shown in Fig. 2.2a. After twin-screw extrusion at

155°C, string-like PP fibrils appear in the mPE matrix. The same treatment followed by

annealing at a temperature above the melting temperature of PP (180°C for 10 min) leads to the

19

formation of almost spherical domains of PP in the mPE matrix, shown in Fig 2.2b. These

micrographs are in agreement with our hypothesis that thermal treatment at temperatures

between the melting temperatures of the two components is a necessary step to accomplish

dispersed phase fibrillation during twin-screw extrusion. The diameters of the fibrils obtained by

this method range from 100 nm to 300 nm and the length of most fibrils exceed 100 μm.

Consequently, the aspect ratio is above 200. Subsequent heating of the fibrillar blend above the

melting temperature of PP leads to the recoiling of the fibrils into spherical domains distributed

randomly in the mPE matrix (Fig. 2.2b). The average diameter of the spherical domains is 3 μm.

The morphology that occurs in blends is decided by a complex interplay between the viscosity of

the individual components, interfacial properties, and composition. 14

The fundamental

understanding of the mechanism of morphology development during polymer blending comes

from the work of Taylor 16, 17

and others, 18, 19, 20

who showed that droplet deformation during

flow is controlled by two dimensionless parameters: 1) the viscosity ratio, , where

is the viscosity of the dispersed phase and is the viscosity of the continuous phase, and 2) the

capillary number ( ) which is the ratio between the applied hydrodynamic stress, , and the

interfacial stress, , where is the interfacial tension and is the unperturbed droplet

radius. Under certain conditions of and the dispersed phase can elongate into fibrils of high

aspect ratio. 21

It seems that the chosen blending method accomplishes such conditions causing

the PP domains to fibrillate in the mPE matrix.

2.4.2 Shear response in the linear viscoelastic regime

Fig. 2.3a shows the changes in the elastic modulus, , of mPE/fibrillated-PP at different

fibril contents. We observe that becomes increasingly independent of the frequency as the

fibril content is increased, indicating that the viscoelastic response is changing from ‗liquid-like‘

to ‗solid-like‘. 22, 23

This is one of the typical rheological features of physical gels and similar

observations are made in many studies. 6, 8, 11, 24, 25

The structural implication of the enhanced

elasticity observed with the increase in fibril content is that the solid-state fibrillar PP tends to

restrict the long range motion of the matrix polymer chains and prevents them from complete

relaxation when subjected to shear. 26

Frequency dependence of the loss tangent,

20

Fig. 2.3 Rheological behaviour of mPE/fibrillated-PP at 120°C; a) Frequency dependence of

elastic modulus for five samples with different fibril content; b) Frequency dependence

of loss tangent for the five samples. Samples with fibril contents of 0 and 3 wt% show a

negative slope hence show liquid-like response, and those with fibril contents of 5, 7 and 10 wt%

show a positive slope hence show solid-like response; 22

c) as a function of fibril content.

Different lines represent different frequencies ranging from 0.1 to 100 rad/s. The lines intersect

at the gel point content cg.

⁄ can be employed to accurately detect the gel point. The gel point is

reached when becomes independent of frequency. 27, 28, 29, 30

In the pre-gel regime,

decreases with an increase in , a typical behaviour of viscoelastic liquids, while in the post-gel

regime, increases with an increase in , a typical behaviour of pseudo-solids. 22

We show

the frequency dependence of in Fig. 2.3b and observe that for fibril contents of 3 wt% and

less, decreases with an increase in whereas for fibril contents of 5 wt% and more, a low

magnitude, positive gradient in occurs with an increase in . Thus, a fibril content

between 3 and 5 wt% should be independent of , i.e., the gel point must be between 3 and 5

wt% fibril content.

The Winter-Chambon criterion is the most widely used method to accurately determine the gel

point of physical and chemical gels from rheological data. 31, 32

We can identify the gel point

concentration by observing a frequency independent value of when is plotted as a

function of fibril content at several frequencies. This plot is shown in Fig. 2.3c. We observe that

10-1

100

101

102

100

101

102

b

tan

Frequency (rad/s)

0 wt%

3 wt%

5 wt%

7 wt%

10 wt%

10-1

100

101

102

10-1

100

101

102

103

104

105

106

10 wt%

7 wt%

5 wt%

3 wt%

0 wt%

G' (P

a)

Frequency (rad/s)

a

2 3 4 5 6 7 8 9 100.0

0.5

1.0

1.5

2.0

2.5

3.0

100 rad/s

tan

Fibril content (wt%)

gel point

0.1 rad/s

cg = 4.5 wt%

c

21

lines of obtained at different frequencies decay gradually with an increase in fibril content

and intersect at the gel point content, cg, of 4.5 wt% corresponding to a = 1.3. We further

notice that before reaching the cg the value of tan δ decreases with increasing but after cg the

trend reverses. Many systems involving physical and chemical gelation exhibit this behaviour

and is related to the magnitude of and before and after cg at different . Thus, the

Winter-Chambon theory is applicable to the mPE/fibrillated-PP system over a wide angular

frequency range. It seems that at PP fibril contents ≥ 4.5 wt%, the solid-like response originates

from the high aspect ratio of the PP fibrils which can form topological interactions

(entanglements). The entanglements between PP fibrils manifest into a network structure. The

network is stable for small deformations and is able to store the deformation energy over long

timescales transitioning the behaviour from liquid-like to gel-like. Previously, results bearing

resemblance to the present case have been reported for other filled polymer systems and their

elastic response to deformation under shear at low is also presumed to be due to the

development of a stable network structure. 24

In contrast to the mPE/fibrillated-PP, the

mPE/spherical-PP does not show any evidence of gelation and the rheological responses remain

largely unaltered when the spherical PP domains are present. Contrary to conventional wisdom

which states that solid-like rheological response requires a permanent three dimensional network

of polymer chains to be established via cross-linking/chemical bonds, we demonstrate that the in-

situ fibrillation of a dispersed polymeric phase in a polymer melt can induce such behaviours, in

the absence of cross-links.

2.4.3 Stress growth during elongation at a constant rate

The extensional viscosity measures the resistance against extensional flow. The transient

extensional viscosity is a function of both the strain rate and time and is expressed as the tensile

stress growth coefficient :

(2.1)

where is the tensile stress growth function and is the extensional strain rate.

According to the Trouton relation, 33

for nonlinear strain hardening response, the is equal

to 3 times the linear viscoelastic shear viscosity for a sufficiently small strain rate.

22

Fig. 2.4 Tensile stress growth curves of uniaxial extensional viscosity, , at strain rates of

0.01, 0.1 and 1 s-1

at a temperature of 130°C for a) mPE, b) mPE/spherical-PP (95/5 wt%), c)

mPE/fibrillated-PP (95/5 wt%). The solid lines represent the where is the growth

curve of shear viscosity in the linear region obtained from startup shear experiments for each

sample at a strain rate of 0.001 s-1

and temperature 130°C. The insert in c) shows the rheopectic

behaviour of mPE/fibrillated-PP where continues to increase with showing a time-

dependent thickening instead of reaching a steady state.

Deviation from the Trouton relation, given by , can be used to evaluate strain hardening

through,

(2.2)

Uniaxial elongation experiments are conducted to understand the strain hardening behaviour of

the network of entangled PP fibrils in the mPE matrix and the results are shown in Fig. 2.4. The

3 obtained from steady shear start-up experiments at a temperature of 130°C, and a strain

rate of 0.001 s-1

is included in the graphs to indicate the linear viscoelastic behaviour limit. The

accuracy of the measurements in extensional flow is confirmed in Fig. 2.4a-b by the fact that the

Trouton relation, 3 , is fulfilled at all studied strain rates for neat mPE and

mPE/spherical-PP (95/5 wt%). For the case of mPE/fibrillated-PP (95/5 wt%), shown in Fig.

2.4c, although the extensional viscosities do not follow the Trouton relation, the reproducibility

is good reflected by a low relative error of the measured extensional stress growth coefficient. At

0.01 0.1 1 10 10010

3

104

105

106

1 s-1

0.1 s-1

0.01 s-1

+

+ E(t

, P

a.s

]

Time (s)

a

0.01 0.1 1 10 10010

3

104

105

106

1 s-1

0.1 s-1

0.01 s-1

+

+ E(t

, P

a.s

]

Time (s)

b

0.01 0.1 1 10 10010

3

104

105

106

+

+ E(t

, P

a.s

]

Time (s)

1 s-1

0.1 s-1

0.01 s-1

slope~0.70

10-3

10-1

101

103

1.5x104

2.0x104

2.5x104

3t

) [P

a.s

]

Time (s)

c

23

strain rates of 0.01, 0.1 and 1 s-1

the relative errors are calculated to be 1.5, 1.1 and 1.1 %,

respectively.

The for neat mPE in Fig. 2.4a superposes for all applied extensional strain rates on the

Trouton prediction 3 , and no strain hardening is observed. A similar result is obtained for

mPE/spherical-PP (Fig. 2.4b). In contrast, the mPE/fibrillated-PP exhibits pronounced strain

hardening behaviour in the curves shown in Fig. 2.4c.

The divergence of the tensile stress growth behaviour of mPE/fibrillated-PP from mPE and

mPE/spherical-PP depicted in Fig. 2.4a-c is also observed in other particle-suspensions and is

thought to result from the formation of a network superstructure through physical entanglements

of flexible fibers. 24, 34

The origins of strain hardening in extensional flow for the fibrillar

network is analogous to that of long chain branched polymers, where strain hardening is

generally related to the inability of the macromolecules to disentangle quickly enough and follow

the deformation. 35, 36, 37

The degree of strain hardening is dependent upon the applied strain-rate. In general, an increase

in the strain-rate leads to a higher degree of strain hardening. However, a deviation to this is

evident in Fig. 2.4c where the degree of strain hardening appears to increase towards low strain-

rates. Specifically, the value of (calculated from Eq. 2.2) obtained at 0.01 s-1

, the lowest strain-

rate studied, is 21.7. This is larger than the value of 6.5 obtained at 1 s-1

, the highest strain-rate

studied. This behaviour is also reported for polyethylenes and polypropylenes with a small

amount of long-chain branching and has been attributed to topological factors. 38, 39, 40

Another unusual rheological response mPE/fibrillated-PP exhibits pertaining to the

experimentally-measured growth of the extensional viscosity is that it does not follow

the Trouton relation, 3 . Instead, it exceeds the Trouton prediction i.e.

3 . The unique behaviour indicates that flow induced changes in the network structure of

PP fibrils in the mPE melt take place in both shear and elongational flow, however, the structural

changes tend to be very different from each other as inferred from the difference in 3

prediction and the experimental in Fig. 2.4c. The insert in Fig. 2.4c shows that the linear

viscoelastic prediction as a function of time obtained from steady shear flow experiments

at a strain rate of 0.001 s-1

increases continually for mPE/fibrillated-PP and does not show a

24

tendency to plateau over the time examined (15 min). This response is also seen at other small

strain rates such as 0.002 and 0.005 s-1

. Neat mPE or mPE/spherical-PP does not show this

behaviour. Okamoto et al., have reported similar behaviour in polypropylene/clay

nanocomposites and call this time-dependent stiffening behaviour rheopexy. 41

The rheopectic

response of mPE/fibrillated-PP points to the fact that a long time-relaxation process is involved

in the structural changes of the entangled network that occur during shear flow.

2.4.4 Isothermal crystallization behaviour

100

101

102

103

104

105

106

95

mPE/fibrillated-PP

mPE/spherical-PP

mPE

G' (P

a)

Time (s)

10

140

25 Tem

pera

ture

(°C

)

Time (min)

Fig. 2.5 Build-up of elastic ( ) modulus as a function of time at a strain rate of 10 rad/s and an

isothermal crystallization temperature of 95°C, a temperature below the melting temperature of

mPE. Samples were heated to 140°C and left in the melt state for 10 min prior to the isothermal

treatment at 95°C. Crystallization behaviour of mPE is compared with mPE/spherical-PP (95/5

wt%); and mPE/fibrillated-PP (95/5 wt%). The insert shows the thermal treatment of the samples

prior to measurements.

Small amplitude oscillatory shear experiments are conducted to investigate the influence of well-

defined shearing in the melt state, on the subsequent crystallization of mPE/fibrillated-PP,

mPE/spherical-PP and neat mPE. A strong increase in the elastic modulus, , corresponds to the

onset of crystallization. 42

Fig. 2.5 shows that the onset of build-up of for mPE occurs around

25

104 s. The extremely slow crystallization is attributed to the branching units of mPE which are

excluded from the crystalline phase. 43

In the presence of spherical PP the onset of build-up of

occurs around 103 s, and in the presence of fibrillated PP, the onset of build- up of occurs

around 102 s. The earlier onset of crystallization in the samples containing the PP dispersed

phase suggests an easier nucleation and growth process due to the availability of heterogeneous

crystal nucleating sites. The submicron diameter of the fibrillated PP increases the surface-to-

volume ratio relative to the sample containing spherical PP domains and hence a larger area of

contact between the mPE melt is realized. Consequently, the rate of crystallization of

mPE/fibrillated-PP is rapid compared with mPE/spherical-PP.

2.4.5 Foam processing and cellular morphology

a

Fig. 2.6 Foam morphologies: a) SEM images of foamed samples of i) mPE, ii) mPE/spherical-PP

(95/5 wt%), iii) mPE/fibrillated-PP (95/5 wt%). The samples were subject to a temperature of

95°C and CO2 pressure of 5.5 MPa for 1 hour. b) Cell densities shown as a function of the

saturation pressure at a foaming temperature of 95°C. c) Foam volume expansion ratios shown as

a function of the saturation pressure at a foaming temperature of 95°C.

4 6 8 10 12 14 16 180

3

6

9

Exp

an

sio

n R

ati

o

Pressure (MPa)

mPE/fibrillated-PP

mPE/spherical-PP

mPEc

4 6 8 10 12 14 16 1810

3

104

105

106

107

108

109

1010 mPE/fibrillated PP

Cell d

en

sit

y (

cells/c

m3)

Pressure (MPa)

mPE

mPE/spherical PP

�b

26

The effect of morphology of the dispersed phase on the processability of mPE is studied using

polymer foaming. Neat mPE, mPE/spherical-PP, and mPE/fibrillated-PP are foamed using a

batch process. Foaming is conducted at a temperature of 95°C and CO2 pressures ranging from

4.1 MPa to 17.3 MPa. Fig. 2.6a shows typical foam morphologies of these samples obtained at a

CO2 pressure of 5.5 MPa. With the addition of the PP second phase, the cell size decreases and

the cell density increases. In the presence of spherical PP, the average cell size decreases from

700 μm to 90 μm, and the cell density increases from 5.9 × 104

cells/cm3 to 1.2 × 10

5 cells/cm

3.

The most significant improvements in foam morphology are seen in mPE/fibrillated-PP where

the average bubble size decreases to 27.6 μm and the bubble density increases to 9.7 × 108

cells/cm3. We attribute these improvements in foaming ability primarily to the improved tensile

stress growth in extensional flow observed for mPE/fibrillated-PP which is the primary mode of

deformation during foam processing. 25, 44

Fig. 2.6b plots the cell densities as a function of the saturation pressure at 95°C and clearly

shows that under all saturation pressures studied, cell densities follow the following trend

(highest to lowest): mPE/fibrillated-PP > mPE/spherical-PP > mPE. Recent studies have

demonstrated that the melt flow induced by the expansion of nucleated bubbles triggers the

formation of new cells through stress-induced cell nucleation. 45, 46

Expanding bubbles generate

flow which produces a complex and locally varying superposition of shear, compressive and

tensile stresses within the melt. In regions where a tensile stress is created, the energy barrier for

heterogeneous nucleation, , decreases. This, in turn, enhances the cell nucleation rate. 47

The

viscoelastic character of the polymer/gas mixture is a key parameter dictating the effectiveness

of stress-induced cell nucleation. 48, 49

The higher cell density observed for mPE/fibrillated-PP

relative to neat mPE and mPE/spherical-PP is attributed to its solid-like viscoelastic character

depicted in Fig. 2.3 and not observed for the other two samples. The enhanced solid-like

character increases the internal local pressure variations which make the bubble nucleation rate

faster. 50, 51

This effect is enhanced by the improved crystallization kinetics when fibrillated-PP is

present in the mPE matrix further increasing the solid-like character of mPE/fibrillated-PP.

Another reason for the highest cell-density observed for mPE/fibrillated-PP is that it has the most

rapid crystallization kinetics (see Fig. 2.5), i.e., mPE/fibrillated-PP is able to reach the highest

degree of crystallinity during the batch foaming experiments relative to neat mPE and

mPE/spherical-PP. The evolution of viscoelastic character for polymers during crystallization is

27

well known. 12, 52

The early stages of crystallization could be viewed as a physical gelation

process where crystallites act as the physical cross-links linking molecular chains together.

Consequently, improvements in crystallization are expected to increase the material stiffness and

suppress cell deterioration mechanisms such as cell wall rupture and cell coalescence. Fig. 2.6c

plots the foam volume expansion ratio as a function of the saturation pressure at 95°C. In

general, the expansion ratio of a foam is determined by the gas saturation pressure. At a given

temperature, an increase in the saturation pressure leads to a larger foam expansion ratio because

more gas is able to dissolve in the polymer melt for expansion. This can be realized in the case of

neat mPE in Fig. 2.6c where an increase in the CO2 saturation pressure from 4.1 MPa to 17.3

MPa leads to an increase in the expansion ratio. However, the expansion ratio of mPE/spherical-

PP and mPE/fibrillated-PP does not follow this generalization. We attribute this deviation from

expectation to the more rapid crystallization kinetics of mPE/fibrillated-PP and mPE/spherical-

PP relative to neat mPE (Fig. 2.5). It is well known that the crystallization rate is influenced by

the presence of dissolved gas. Often times, gas dissolution in a polymer matrix increases

crystallization rates by increasing free volume and promoting chain registry. 53

If the rate of

crystallization is increased, the matrix viscosity rises and the bubbles experience a larger degree

of resistance to expansion during the bubble growth stage of the foaming process. It seems that

the stiffness of the samples at the instant foaming is initiated vary in the following order (highest

to lowest): mPE/fibrillated-PP < mPE/spherical-PP < mPE due to increasingly active

crystallization kinetics which make foam expansion difficult.

2.5 Conclusion

The morphology of a solid-state dispersed phase influences the viscoelastic as well as

crystallization behaviours of semicrystalline polymers. We demonstrate this by comparing

properties of mPE containing PP domains which maintain a spherical morphology and mPE

containing PP domains which have a fibrillated morphology with that of neat mPE. For the case

of mPE/fibrillated-PP, strain hardening is observed during uniaxial extensional flow.

Furthermore, a shear-induced thickening or rheopectic response is observed under shear flow.

Such viscoelastic behaviour is not observed for the case of mPE/spherical-PP or neat mPE. The

mPE/fibrillated-PP shows a two orders of magnitude earlier onset of crystallization and the

mPE/spherical-PP shows a one order of magnitude earlier onset of crystallization relative to neat

mPE. These enhancements in physical properties induced by the fibrillar solid-state dispersed

28

phase morphology are of both scientific and technological significance. Foam processing reveals

noticeable improvements in the foaming ability of mPE/fibrillated-PP relative to neat mPE and

mPE/spherical-PP. Specifically, microcellular foams are obtained at mild processing conditions

easily accessible in conventional foam manufacturing equipment i.e., 5.5 MPa CO2 pressure and

a temperature of 95°C. This means that in-situ fibrillation of a dispersed phase has wide

commercial appeal. Because the fibrillar blend is prepared from thermoplastic commodity

plastics, our processing strategy offers further advantages such as low raw material costs, easy

processability, and complete recyclability of individual components of the blend.

29

References

1. Coronado, E.; Galán-Mascarós, J. R.; Gómez-García, C. J.; Laukhin, V. Coexistence of

ferromagnetism and metallic conductivity in a molecule-based layered compound. Nature

2000, 408, 447-449.

2. Wang, Y.; Herron, N. X-Ray Photoconductive Nanocomposites. Science 1996, 273, 632-634.

3. Pinnavaia, T. J. Intercalated clay catalysts. Science 1983, 220, 365-371.

4. Hsu, W. Y.; Gierke, T. D.; Molnar, C. J. Morphological effects on the physical properties of

polymer composites. Macromolecules 1983, 16, 1945-1947.



8, 8539-8546.

6. Kharchenko, S. B.; Douglas, J. F.; Obrzut, J.; Grulke, E. A.; Migle, K. B. Flow-induced

properties of nanotube-filled polymer materials. Nature Mater. 2004, 3, 564-568.

7. Bicerano, J.; Douglas, J. F.; Brune, D. A. Model for the viscosity of particle dispersions. J.

Macromol. Sci. Polymer Rev. 1999, C39, 561–642.

8. Ali, M.; Okamoto, K.; Yamaguchi, M.; Kasai, T.; Koshirai, A. Rheological properties for

polypropylene modified by polytetrafluoroethylene. J. Polym. Sci. B 2009, 47, 2008-2014.




same. WO2013137301, March 13, 2013.


30

JP2013514463, March 13, 2012.

12. Schwittay, C.; Mours, M.; Winter, H. H. Rheological expression of physical gelation in

polymers. Farady Discuss. 1995, 101, 93-104.



180-188.

14. Scott, C. E.; Macosko, C. W. Morphology development during the initial stages of polymer-

polymer blending. Polymer 1995, 461-470.

15. Sundararaj, U.; Macosko, C. W.; Rolando, R. J.; Chan, H. T. Morphology development in

polymer blends. Polym. Eng. Sci. 1992, 32, 1814-1823.


Ser. A 1934, 146, 501-523.

17. Taylor, G. I. The viscosity of a fluid containing small drops. Proc. R. Soc. Lond. A 1932,

834, 41-48.



14, 225-277.

19. Wu, S. Formation of dispersed phase in incompatible polymer blends: Interfacial and

rheological effects. Polym. Eng. Sci. 1987, 27, 335-343.

20. Wu, S. Calculation of interfacial tension in polymer systems. J. Polym. Sci. Part C: Polym.

Symp. 1971, 34, 19-30.



31

22. Winter, H. H.; Mours, M. Rheology of polymers near their liquid–solid. Adv. Polym. Sci.

1997, 134, 165–234.

23. Larson, R. G. The structure and rheology of complex fluids. Oxford University Press: New

York, 1999; pp 14-15.

24. Yokohara, T.; Nobukawa, S.; Yamaguchi, M. Rheological properties of polymer composites

with flexible fine fibers. J. Rheol. 2011, 55, 1205-1218.

25. Rizvi, A.; Tabatabaei, A.; Barzegari, R.; Mahmood, H.; Park, C. B. In-situ fibrillation of

CO2-philc polymers: sustainable route to polymeric foams in a continuous process. Polymer

2013, 54, 4645–4652.

26. Ren, J.; Silva, A. S.; Krishnamoorti, R. Linear Viscoelasticity of Disordered

Polystyrene−Polyisoprene Block Copolymer Based Layered-Silicate Nanocomposites.

Macromolecules 2000, 33, 3739-3746.

27. Winter, H. H. Can the gel point of a cross-linking polymer be detected by the G′ – G″

crossover? Polym. Eng. Sci. 1987, 27, 1698–1702.

28. Hodgson, D. F.; Amis, E. J. Dynamic viscoelastic characterization of sol-gel reactions.


29. Muller, R.; Gerard, E.; Dugand, P.; Rempp, P.; Gnanou, Y. Rheological characterization of

the gel point: a new interpretation. Macromolecules 1991, 24, 1321–1326.

30. Holly, E. E.; Venkataraman, S. K.; Chambon, F.; Winter, H. H. Fourier transform mechanical

spectroscopy of viscoelastic materials with transient structure. J. Non-Newton Fluid Mech.

1988, 27, 17-26.

31. Chambon, F.; Winter, H. H. Linear viscoelasticity at the gel point of a crosslinking PDMS

with imbalanced stoichiometry. J. Rheol. 1987, 31, 683-697.

32. Winter, H. H.; Morganelli, P.; Chambon, F. Stoichiometry effects on rheology of model

32

polyurethanes at the gel point. Macromolecules 1988, 21, 532–535.

33. Trouton, F. T. On the coefficient of viscous traction and its relation to that of viscosity. Proc.

R. Soc. London Ser. A77 1906, 77, 426–440.

34. Yamaguchi, M.; Fukuda, K.; Yokohara, T.; Ali, M.; Nobukawa, S. Modification of

Rheological Properties Under Elongational Flow by Addition of Polymeric Fine Fibers.

Macromol. Mater. Eng. 2012, 297, 654-658.

35. Wagner, M. H.; Yamaguchi, M.; Takahashi, M. Quantitative assessment of strain hardening

of low-density polyethylene melts by the molecular stress function model. J. Rheol. 2003, 47,

779-793.

36. McLeish, T. C. B.; Allgaier, J.; Bick, D. K.; Bishko, G.; Biswas, P.; Blackwell, R.; Blottiere,

B.; Clarke, N.; Gibbs, B.; Groves, D. J.; Hakiki, A.; Heenan, R. K.; Johnson, J. M.; Kant, R.;

Read, D. J.; Young, R. N. Dynamics of entangled H-polymers: theory, rheology, and

neutron-scattering. Macromolecules 1999, 32, 6734-6758.

37. Masubuchi, Y.; Inaniruberto, G.; Greco, F.; Marrucci, G. Primitive chain network

simulations for branched polymers. Rheol. Acta. 2006, 46, 297-303.

38. Gabriel, C.; Munstedt, H. Strain hardening of various polyolefins in uniaxial elongational

flow. J. Rheol. 2003, 47, 619-630.

39. Malmberg, A.; Liimatta, J.; Lehtinen, A.; Löfgren, B. Characteristics of long chain branching

in ethene polymerization with single site catalysts. Macromolecules 1999, 32, 6687-6696.

40. Auhl, D.; Stange, J.; Münstedt, H. Long-Chain Branched Polypropylenes by Electron Beam

Irradiation and Their Rheological Properties. Macromolecules 2004, 37, 9465-0472.

41. Okamoto, M.; Nam, P. H.; Maiti, P.; Kotaka, T.; Hasegawa, N.; Usuki, A. A House of Cards

Structure in Polypropylene/Clay Nanocomposites under Elongational Flow. Nano Lett. 2001,

1, 295-298.

33

42. Wischnewski, A.; Monkenbuch, M.; Willner, L.; Richter, D.; Kali, G. Direct observation of

the transition from free to constrained single-segment motion in entangled polymer melts.

Phys. Rev. Lett. 2003, 90, 583021–583024.

43. Gelfer, M. Y.; Winter, H. H. Effect of branch distribution on rheology of LLDPE during

early stages of crystallization. Macromolecules 1999, 32, 8974-8941.

44. Xu, Z.; Zhang, Z.; Guan, Y.; Wei, D.; Zheng, A. Investigation of extensional rheological

behaviours of polypropylene for foaming. J. Cell. Plast. 2013, 0, 1-18.



2009, 64, 4899-4907.

46. Leung, S. N.; Wong, A.; Wang, C.; Park, C. B. Mechanism of extensional stress-induced cell

formation in polymeric foaming processes with the presence of nucleating agents‖. J.

Supercrit. Fluids 2012, 63, 187-198.

47. Wong, A.; Guo, Y.; Park, C. B. Fundamental Mechanisms of Cell Nucleation in

Polypropylene Foaming with Supercritical Carbon Dioxide - Effects of Extensional Stresses

and Crystals. J. Supercrit. Fluids 2013, 79, 142–151.

48. Leung, S. N.; Park, C. B.; Li, H. Effects of Nucleating Agents‘ Shapes and Interfacial

Properties on Cell Nucleation. J. Cell. Plast. 2010, 46, 441-460.

49. Wang, C.; Leung, S. N.; Bussmann, M.; Zhai, W. T.; Park, C. B. Numerical Investigation of

Nucleating Agent -Enhanced Heterogeneous Nucleation. Ind. Eng. Chem. Res. 2010, 49,

12783–12792.

50. Wong, A.; Wijnands, S. F. L.; Kuboki, T.; Park, C. B. Mechanisms of Nanoclay-Enhanced

Plastic Foaming Processes – Effects of Nanoclay Intercalation and Exfoliation. J. Nanopart.

Res. 2013, 15, 1815-1829.

34

51. Liu, W.; Wang, X.; Li, H.; Du, Z.; Zhang, C. Study on rheological and extrusion foaming

behaviours of chain-extended poly (lactic acid)/clay nanocomposites. J. Cell. Plast. 2013, 49,

535-554.

52. Garancher, J. P.; Fernyhough, A. Crystallinity effects in polylactic acid-based foams. J. Cell.

Plast. 2012, 48, 387-397.

53. Walker, T. A.; Frankowski, D. J.; Spontak, R. J. Thermodynamics and kinetic processes of

polymer blends and block copolymers in the presence of pressurized carbon dioxide. Adv.

Mater. 2008, 20, 879-898.

35

Chapter 3

Enhanced foaming ability of poly(propylene-co-ethylene) random copolymer with polyethylene terephthalate fibrils 2

3.1 Abstract

In this study, we show that in-situ generated fibrils of polyethylene terephthalate (PET) can

impart strain hardening behaviour to a poly(propylene-co-ethylene) random copolymer (PP) in

uniaxial extensional flow. The strain hardening response is attributed to the formation of a

disordered physical network of PET fibrils, confirmed by applying Winter-Chambon‘s analysis

on the linear viscoelastic oscillatory shear data. Differential Scanning Calorimetry (DSC) at

ambient pressures and at elevated CO2 pressures reveals that the PET fibrils enhance the kinetics

of isothermal crystallization of PP under both ambient and elevated CO2 pressures. Wide-angle

X-ray scattering (WAXS) study shows that isothermal crystallization of the PP/PET fibrillar

composite in the presence of pressurized CO2 facilitates the formation of γ-phase crystals in the

PP matrix. Foaming of the PP/PET fibrillar composite shows that the presence of fibrils lead to

marked enhancements in the foaming ability of PP. The improvements in foaming are explained

based on the rheological and crystallization results.

3.2 Introduction

Plastic resins employed for industrial applications are often blends of two or more polymers.

Polymer blending offers a remarkably rich range of new materials with enhanced physical and

chemical properties. During polymer blending, a morphology is reached in which one phase is

dispersed in another and the size and shape of the dispersed phase is governed by factors such as

the composition of the blend, the processing history, the rheological properties and the interfacial

properties. 1, 2, 3, 4, 5, 6

In particular, microfibrillar reinforced composites (MFCs), 7

where

immiscible polymeric dispersed phases deform into high-aspect-ratio fibrils inside a polymer

matrix, have attracted attention as versatile materials with wide commercial appeal. Fibrillar

composites are prepared through a two-stage process involving melt blending immiscible

2 Manuscript in preparation.

36

polymers followed by cold drawing of the blend to elongate the dispersed phase domains. 7, 8, 9

In

comparison with the neat polymer host, fibrillar composites show improvements in the solid-

state mechanical properties such as the tensile strength, 9

the flexural strength, 10

and the Young‘s

modulus. 11

However, most industrially viable processing of polymers is conducted in the melt

state or in the vicinity of the melt state, i.e., the processing temperature exceeds specific values

characteristic of the polymeric system such as the softening/heat-deflection temperature, or the

melting temperature. At such temperatures, the viscoelastic character dominates and is not

represented by the solid-state mechanical properties. When the concentration of the high-aspect-

ratio fibrils in a polymer host exceeds the percolation threshold, a three-dimensional fibrillar

network defined by topological (entangled) interactions is expected which creates large

contributions to the viscoelasticity of the matrix. 12, 13, 14

Specifically, the melt viscosity can be

tuned in fibrillar composites by simply controlling the fibril content in the matrix polymer. This

is useful from a technological standpoint because in many polymer processing applications, the

achievement of a suitable melt viscosity under the processing conditions is required to obtain a

high quality product. For example, in polymer foam processing, an appropriate melt viscosity is

required to obtain a highly expanded, stable foam structure with low variations in the foam

morphology. 15, 16, 17, 18

In this study, we evaluate the foaming behaviour of a fibrillar composites of poly(propylene-co-

ethylene) random copolymer (PP) with in-situ generated fibrils of polyethylene terephthalate

(PET). The most important mode of deformation during polymer foaming is extensional flow.

Therefore, the uniaxial extensional viscosity of PP/PET fibrillar composites is characterized and

the behaviour is explained through the formation of a percolated network of PET fibrils in the PP

matrix. The gel point 19

analysis of Winter-Chambon 20, 21

is used to identify the fibril content that

leads to the formation of a rheologically percolated network. The most important phase transition

that semicrystalline polymers undergo during polymer foaming is crystallization. Crystallization

can be thought of as a ‗physical crosslinking‘ process where crystals tend to act as the crosslink

junctions which increase the polymer melt strength. 22, 23

It has been shown that there is an

optimum degree of crystallinity for obtaining foams of PP: if the crystallinity is too low, gas loss

and bubble coalescence mechanisms will be active, and if the crystallinity is too high, the matrix

will be too stiff to undergo expansion. 18, 24, 25

Understanding the crystallization behaviour of

PP/PET fibrillar composites during the foaming process is paramount for controlling the quality

37

of the derived foams. Because CO2 is used as the blowing agent in our process, the

crystallization behaviour of the polymer system will be altered by the increased chain mobility

with the dissolution of CO2. Consequently, a further aim of this study is to examine the

isothermal crystallization kinetics of PP/PET fibrillar composites under ambient nitrogen

pressures and under pressurized CO2. Wide angle X-ray Scattering (WAXS) studies are also

conducted to examine the changes in the crystal structure when crystallization is carried out at

ambient pressure and at elevated CO2 pressures.

3.3 Experimental

3.3.1 Materials

The matrix polymer employed in this study is a poly(propylene-co-ethylene) random copolymer

(PP) commercially available as Sabic®

PP 670K with a melt flow index (MFI) of 10 g/10 min at

230°C/2.16 kg. The dispersed phase polymer employed in this study is a polyethylene

terephthalate (PET) copolymerized with 3.5 % 1,4-cyclohexanedimethanol (CHDM),

commercially available as Eastapak® PET 9921 (Mn =26 000 g mol

-1). CO2 is supplied by Linde

Gas with purity in excess of 99%. Xylene is purchased from Caledon Laboratories Ltd. and used

as received.

The melting transitions of PP and PET are determined using Differential Scanning Calorimetry

(DSC) under a nitrogen atmosphere. A heat-cool-heat experiment is performed from 25°C to

260°C. Heating and cooling rates of 10°C/min are used. The melting transition temperature is

derived from the melting peak maximum during the second heating cycle. The melting transition

peaks for PP and PET are seen at temperatures of 149°C and 240°C, respectively.

3.3.2 Sample preparation

PP and dried PET are blended using a co-rotating twin-screw extruder (DSM Xplore micro-

compounder) with a built-in back-flow channel which enables material recirculation. Various

PET contents are used ranging from 0 to 7 wt%. An initial blending temperature of 255°C is

used and the polymers are blended for 7 min to get well dispersed spherical domains of PET in

the PP matrix. Subsequently, the temperature of the extruder is reduced to 170°C using

compressed-air active cooling, and the polymers are blended for another 7 min. The samples are

then drawn at a drawing ratio of about 6.3:1.

38

3.3.3 Morphological observations

To observe the morphology of the PET domains in the PP/PET blends (Tprocessing = 255°C), a

solvent-vapor etching technique is used. The sample is exposed to xylene vapours for 45 min at

60°C. This type of treatment preferentially etches the PP matrix which is soluble in xylene,

leading to bringing up the PET domains and increasing differentiation in observation using a

Scanning Electron Microscope (SEM, JEOL 6060). To observe the morphology of the PET

fibrils in the PP/PET fibrillar composite (Tprocessing = 170°C), the extrudate from the twin-screw

microcompounder is immersed in hot xylene at 110°C to remove the PP matrix. The PET residue

is then examined using the SEM. To observe the morphology of the PP/PET fibrillar composite

foams, the foamed samples are cryo-fractured, and the exposed surface is examined using SEM.

3.3.4 Rheological characterization


(EVF, TA Instruments) attached to an Advanced Rheometric Expansion System (ARES).

Samples are tested at strain rates from 0.005 to 1 s-1

and at a sample environment temperature of

165°C. This temperature provides us with the most reproducible and consistent results. Strain

rates higher than 1 s-1

are experimentally inaccessible because sample breakage occurs in the

initial stages and homogeneous extension cannot be accomplished. Higher temperatures are also

inaccessible because of sagging of the polymer samples.


using the ARES rheometer in the controlled strain mode, and cone-and-plate geometry. The

diameter of the cone is 25 mm and the cone angle is 5°. The shear elastic modulus, G‘, and

viscous modulus, G‘‘, are evaluated at a temperature of 165°C under a nitrogen atmosphere to

avoid oxidative degradation. The temperature is chosen because it is above the melting

temperature of PP but below the melting temperature of PET, allowing preservation of the PET

phase morphology during the experiments. Frequency sweeps between 0.1 and 100 rad/s are

carried out at strains within the linear viscoelastic range. Repeated sweeps with increasing and

decreasing frequencies show that the material is stable under the measurement conditions.

Prolonged heating at 165°C of the samples in the rheometer results in no change in the

viscoelastic data.

39

3.3.5 Investigation of crystallization kinetics

The isothermal crystallization behaviour of the samples under ambient nitrogen pressures is

followed using a TA Instruments Q2000 DSC. The nitrogen atmosphere is used to prevent

oxidative degradation. The isothermal crystallization behaviour of the samples in the presence of

dissolved CO2 is measured using a Netzsch High Pressure DSC 204 HP. The specific DSC

procedure for the samples is kept the same for both the ambient pressure DSC experiments and

the high pressure DSC experiments. The procedure is defined as follows: 1) equilibrate at 25°C;

2) heat to 210°C at a heating rate of 10°C min−1

; 3) isothermal treatment at 210°C for 10 min; 4)

cool to the isothermal crystallization temperature of 135°C (under ambient pressure), or 126°C

(under pressurized CO2) at a cooling rate of -10°C min−1

; 5) isothermal treatment at the

crystallization temperature for 100 min. For the high pressure DSC experiments, samples are

exposed to a CO2 pressure of 3 MPa during the whole process.

3.3.6 Investigation of crystal structures of PP in PP/PET

The crystal structure of PP in the samples is studied using WAXS. The X-ray diffraction patterns

are collected on a Bruker AXS D2 Phaser diffractometer. High resolution CuKα source is used

operating at 30 KV and 10 mA. The system is equipped with Ni-filter for elimination of CuK

peaks and a solid state Lynxeye XE detector. WAXS traces are obtained for PP and PP/PET

isothermally crystallized at a temperature of 135°C (under ambient pressure) or at a temperature

of 126°C (under a CO2 pressure of 3 MPa).

3.3.7 Investigation of foaming behaviour

For the foaming experiments, samples are shaped into disks of 1 cm diameter and 400 μm

thickness. The shaped samples are placed in a custom-built, temperature-regulated, pressure

vessel connected to a syringe pump filled with CO2. The vessel is purged with CO2 prior to

pressurization. The samples are then saturated at a temperature of 137°C and at a pressure of 5.5

MPa. After an exposure time of 1 h, compressed CO2 is quickly released from the pressure

vessel, and the samples are removed from the vessel and immersed in a water bath to quench the

foam structure. To characterize the obtained foams, the foam volume expansion ratio, ⁄ , is

calculated, where and are the mass densities of samples before foaming and after foaming,

respectively, determined using the water displacement method according to ASTM-D792.

40

Additionally, the number of bubbles per unit volume, called the cell density, 26

is also calculated

using the SEM images of the foams.


3.4.1 Morphology of fibrils

Fig. 3.1 SEM images of PP/PET (95/5 wt%) blends: a) PET domains have spherical shapes after

blending at 255°C, a temperature above the Tm of PET. The image is obtained after solvent-

vapor etching using xylene; b) PET domains have a fibrillar structure obtained after blending at

170°C, a temperature between the Tm of the two blend components, and cold drawing at a draw

ratio of 6.3:1. The image is obtained after selective removal of PP using xylene.

The presence of high-aspect-ratio fibrils in a polymer matrix is expected to create additional and

large contributions to the viscoelasticity of the matrix. Recently, we have applied the

methodology proposed by Fikirov et al. 7

to fibrillate a dispersed phase in a polymer blend using

twin-screw extrusion. 15, 17

We showed that the strategy is effective in fibrillating a

polytetrafluoroethylene dispersed phase in isotactic polypropylene 15

and isotactic polypropylene

in metallocene-catalyzed polyethylene. 17

In general, the fibrillation of a dispersed phase involves

two stages: 1) Mixing stage: Obtaining well-dispersed spherical domains of the high melting

temperature polymer in the matrix by blending above the melting temperatures of both

components of the blend; 2) Fibrillation stage: Fibrillating the high melting temperature

component by blending at a temperature between the melting temperatures of the two

41

components of the immiscible blend. The hydrodynamic stress during blending facilitates

deformation of the solidifying dispersed phase. 27

The blend is subjected to cold drawing to

further extend the dispersed phase domains. 8

The fundamentals of the mechanism of

morphology development during polymer blending can be found elsewhere. 1, 2, 3, 4, 5

Fig. 3.1a shows the SEM micrographs of PP/PET (95/5 wt%) prior to the fibrillation of the PET

fraction. PET exists as spherical domains with average diameters of about 3 µm. Fig. 3.1b shows

the SEM micrographs after the fibrillation process. The SEM image is obtained after selective

removal of PP from the PP/PET fibrillar composite. The average fibril diameter is about 163 nm

and the average length is about 31 µm. Consequently, the aspect ratio of the fibrils is about 190.

In Table 3.1, the dimensions of the PET fibrils in PP/PET blends with various PET contents

determined from SEM images are presented. When the PP/PET samples are blended at a

temperature above the melting temperatures of both components (i.e., the mixing stage), we

observe that the average diameter of the spherical domains of PET gradually increases with an

increase in the PET content. Subsequent blending at a temperature between the melting

temperatures of both components (i.e., fibrillation stage) yields fibrils with an average diameter

that seems to increase with an increase in the PET content. The lengths of the fibrils do not seem

to show much change, with different PET contents.

Table 3.1 Dimensions of the PET domains in various PP/PET blends as revealed from SEM

images.

a The diameters of the PET spherical domains are measured after the mixing stage (Tprocessing =

255°C > Tm of both components); b the PET fibrils are measured after the fibrillation stage (Tm,

PP < Tprocessing =170°C < Tm, PET ); c Shown values are averages of > 200 measurements;

d Aspect

ratio is determined as the ratio of Length/Diameter.

Composition PET domain dimensions a PET fibril dimensions

b

PP/PET (wt%) Diameter (µm) c Diameter (nm)

c Length (µm)

c Aspect ratio

d

99/1 2.6 144 29 201

99/3 3.1 157 27 172

99/5 3.0 163 31 190

99/7 3.7 213 28 131

42

3.4.2 Uniaxial extensional flow response

Fig. 3.2 Uniaxial extensional viscosity of PP/spherical-PET (97/3 wt%) and PP/fibrillated-PET

composites (97/3 wt%), a) curves of uniaxial extensional viscosity, , measured at a

temperature of 165°C and at strain rates of 0.005, 0.01, 0.05, 0.1 and 1 s-1

. The solid lines

represent the where is the growth curve of shear viscosity in the linear region

obtained from startup shear experiments for each sample at a strain rate of 0.001 s-1

and

temperature 165°C. To prevent overlap, the data for PP/spherical-PET is offset by one order of

magnitude; b) Strain hardening factor, χ, as a function of time for the PP/fibrillated-PET

composite at different extensional strain rates determined using Eq. 3.1.

Fig. 3.2a-b illustrates the uniaxial extensional viscosity, , at various constant strain rates,

, for PP containing 3 wt% PET with spherical domains (PP/spherical-PET) and PP containing 3

wt% fibrillated PET (PP/fibrillated-PET). The solid line in the figures represents the linear

viscoelastic prediction of extensional viscosity, , where is the growth curve of

shear viscosity in the linear viscoelastic region obtained from startup shear experiments at a

strain rate of 0.001 s-1

. The uniaxial extensional flow behaviour of PP/spherical-PET (97/3 wt%)

is similar to that of linear PP. 15, 28

There is no upward deviation from the linear viscoelastic

prediction of the extensional viscosity, , within the strain rate scale studied, and the

extensional viscosity curves obtained at different strain rates superimpose. The experimentally

determined extensional viscosity, , for PP/spherical-PET (97/3 wt%) has a tendency to

follow the Trouton relationship: 3 ,

29 in the limit of steady state (t→∞). On the

10-1

100

101

102

102

103

104

105

+

+ 0.01 s

-1 0.1 s

-1

0.005 s-1 0.05 s

-1 1 s

-1

+ E(t

, P

a.s

]

Time (s)

+

-10x

PP/fibrillated-PET

PP/spherical-PET

a

10-1

100

101

102

0

1

2

3

4

5

3

0.005 s-1

0.01 s-1

0.05 s-1

0.1 s-1

1 s-1

Str

ain

-ha

rde

nin

g f

ac

tor,

Time (s)

b

43

other hand, for the PP/fibrillated-PET (97/3 wt%) composite exhibits upward deviation

in called ―strain hardening‖, and the magnitude of strain hardening depends on the constant

strain rate, . The magnitude of strain hardening can be defined by

(3.1)

where is the strain hardening factor of the extensional flow and the is the 3-fold linear

viscoelasticity. Fig. 3.2b shows a plot of versus time determined from the data presented in

Fig. 3.2a. From Fig. 3.2b, we observe that there is no strain hardening response for

PP/fibrillated-PET (97/3 wt%) at an extensional strain rate of 0.005 s-1

but at strain rates ≥ 0.01 s-

1, strain hardening is observed. The divergence in the extensional viscosity of the fibrillar

composite from the neat polymer matrix is also observed in other systems and is thought to result

from the formation of a network superstructure through physical entanglements of flexible fibers.

12, 30, 31 The origins of strain hardening in extensional flow for the fibrillar network is analogous

to that of long chain branched polymers, where strain hardening is generally related to the

inability of the macromolecules to disentangle quickly enough and follow the deformation. 13,

32,

33, 34 From the results shown in Fig. 3.2a-b, it seems that the fibrils are able to disentangle and

flow with the extensional deformation when a low strain rate such as 0.005 s-1

is applied

resulting in no marked strain hardening response. However, at strain rates ≥ 0.01 s-1

, the network

of fibrils is not able to disentangle readily in response to the strain and strain hardening responses

are observed. In the subsequent section, we study the linear viscoelastic behaviour of the

PP/fibrillated-PET to gain further understanding of the network structure that develops through

the topological interactions (entanglements) of the PET fibrils, in the PP/PET fibrillar composite.

3.4.3 Shear response in the linear viscoelastic regime

The melt state linear viscoelastic response of the fibrillar composites is characterized to

determine the rheological percolation behaviour of the PET fibrils. Interconnected

microstructures assumed by anisotropic fillers, such as high-aspect-ratio PET fibrils, affect the

frequency behaviour of the elastic, G‘, and viscous, G‘‘, moduli. 35, 36

When the fibril content is

high enough, the system may exhibit characteristics of a pseudo-solid or gel which is

44

Fig. 3.3 Rheological behaviour of PP/PET fibrillar composites at 165°C for five samples with

different PET fibril content; a) Frequency dependence of elastic modulus ; b) Frequency

dependence of viscous modulus ; c) Frequency dependence of loss tangent, .

Samples with fibril contents ≤ 1 wt% show a negative slope hence are liquids. Samples with

fibril contents ≥ 5 wt% show a positive slope hence are solids. The sample with fibril content of

3 wt% seems to have a zero slope (frequency independence) and therefore must be in the vicinity

of the critical gelation concentration; 19

d) as a function of fibril content. Different lines

represent different frequencies ranging from 0.1 to 100 rad/s. The lines intersect at the gel point

content cg. Direction of arrows indicates increasing frequency.

10-1

100

101

102

102

103

104

105

106

107

PET 7 wt%

PET 5 wt%

PET 3 wt%

PET 1 wt%

PET 0 wt%

G" (

Pa)

Angular frequency (rad/s)

b

-1 0 1 2 3 4 5 6 7

10-1

100

101

102

tan

PET content (wt%)

inc

rea

sin

g f

req

ue

nc

ygel point

Cg = 3 wt%

d

10-1

100

101

102

102

103

104

105

106

107

PET 7 wt%

PET 5 wt%

PET 3 wt%

PET 1 wt%

PET 0 wt%

G' (P

a)


a

10-1

100

101

102

10-1

100

101

102

tan


PET 0 wt%

PET 1 wt%

PET 3 wt%

PET 5 wt%

PET 7 wt%

c

45

characterized by a three dimensionally percolated network of fibrils spanning throughout the

sample volume. 37

Fig. 3.3a-b illustrates G‘ and G‘‘ as a function of frequency at different fibril

contents ranging from 0 to 7 wt%. Increasing the fibril content leads to a gradual increase in both

G‘ and G‘‘ but the increase in G‘ is more pronounced than that in G‘‘. Furthermore, the slope of

the double logarithmic plot of G‘ versus ω and G‘‘ versus ω progressively decreases with the

increase in the fibril content. That G‘ and G‘‘ become increasingly independent of ω as the fibril

content is increased, indicates that the viscoelastic response is changing from ‗liquid-like‘ to

‗gel-like‘. 19, 38

Fig. 3.3c plots the loss tangent, tan δ = G‘‘/G‘, as a function of frequency using

the data in Fig. 3.3a-b. The tan δ values decrease with an increase in the fibril content. This can

be explained by a lower viscous/elastic ratio when the fibril content is increased. At a fibril

content of 3 wt%, tan δ seems to exhibit frequency independence. The physical meaning of this

result is that the critical fibril concentration required to transition the behaviour of the PP matrix

from liquid-like to solid-like is estimated to be 3 wt% since frequency independence of tan δ

occurs in the vicinity of the gel point. 20, 21

A multi-frequency plot of tan δ versus the fibril

content reveals an intersection that precisely marks the gel point. The data in Fig. 3.3c can be re-

expressed to prepare such a plot, and is presented in Fig. 3.3d. The common point in Fig. 3.3d

occurs at a fibril content of 3 wt%. Thus, the gel point concentration is accurately identified to be

3 wt% for the PP/PET fibrillar composites.

3.4.4 Isothermal crystallization kinetics

In semi-crystalline polymer processing, e.g., extrusion, blow molding, fiber spinning, etc.,

crystallization is the most important phase transition. The duration of crystallization in these

processes influence the transient extensional viscosities, the heat transfer rates, and therefore, the

dynamics of the process. Enhancement of the kinetics of crystallization is common in most

polymers reinforced with high-aspect-ratio fillers such as polymer fibrils, carbon nanotubes,

glass fibers, etc. However, contradictory results are observed in some composites of

polypropylene/carbon nanotubes, 39

polyphenylene sulphide/glass fibers 40

and polyphenylene

sulphide/carbon fibers 40

where the kinetics of crystallization are enhanced initially but hindered

as the filler content is increased. The latter effect suggests a complex influence of high-aspect-

ratio fillers on the crystallization of polymers: at low contents, the fillers provide heterogeneous

nucleation sites for crystallization, but at high concentrations, the spatial confinement becomes

the overwhelming factor hindering kinetics of crystallization. 41

Consequently, the question arises

46

Fig. 3.4 Isothermal crystallization kinetics of PP/PET fibrillar composites with different PET

contents: a) Isothermal crystallization exotherms as a function of the crystallization time

calculated from DSC at 135°C; b) Relative crystallinity, X(t), as a function of the crystallization

time calculated from DSC at 135°C; c) Isothermal crystallization kinetics analyzed using the

Avrami equation; d) Half-times, t1/2, of crystallization determined from the Avrami analysis of

isothermal crystallization.

0 500 1000 1500 2000 2500 3000 3500

0.0

0.2

0.4

0.6

0.8

1.0 b

X(t

)

Time (s)

PET 7 wt%

PET 5 wt%

PET 3 wt%

PET 1 wt%

PET 0 wt%

0 500 1000 1500 2000 2500 3000 3500

PET 7 wt%

PET 5 wt%

PET 3 wt%

PET 1 wt%

PET 0 wt%

Exo

Time (s)

a

0 1 2 3 4 5 6 70

200

400

600

1800

2000

t 1/2 (

s)

PET fibril content (wt%)

d

2 3 4 5 6 7 8 9

-14

-12

-10

-8

-6

-4

-2

0

2

4

PET 7 wt%

PET 5 wt%

PET 3 wt%

PET 1 wt%

PET 0 wt%

ln ln

{1

/[1

-X(t

)]}

ln (Time)

c

47

Table 3.2 Summary of Avrami parameters for the isothermal crystallization kinetics of PP/PET

fibrillar composites with different PET contents at a crystallization temperature of 135°C.

PP/PET (wt%)

n a K (s

-n) b t1/2 (s)

c

100/0 2.30 2.88 × 10-8

1.94 × 103

99/1 2.58 8.05 × 10-8

4.90 × 102

97/3 2.89 5.57 × 10-8

2.86 × 102

95/5 2.97 9.23 × 10-8

2.06 × 102

93/7 2.28 4.97 × 10-8

1.79 × 102

a n is obtained from slopes of lines of best fit for the data plotted in Fig. 3.4c;

b K is obtained

from the vertical axis intercepts of lines of best fit for the data plotted in Fig. 3.4c; c

t1/2 is

determined using Eq. 3.5.

whether or not the PET fibrils accelerate the kinetics of crystallization of the PP matrix. The

isothermal DSC thermograms for samples containing varying PET fibril contents are shown in

Fig. 3.4a. All heat flow curves show a single peak as is typical for isothermal crystallization of

semicrystalline polymers. The relative crystallinity at different crystallization times, X(t), defined

as the ratio of crystallinity at time t to the crystallinity in the limit of steady state (t→∞), is given

by

∫ (

)

∫ (

)

(3.2)

taking t = 0 to be the instant of attainment of thermal equilibrium and t = to be the instant of

completion of crystallization. Then ∫ (

)

is the heat generated from the attainment of

thermal equilibrium to time t. Fig. 3.4b shows the plot of X(t) against the crystallization time. It

can be seen from Fig. 3.4b that all the curves have a similar sigmoidal shape. Additionally, the

increase in the PET fibril content shortens the duration for crystallization to complete. In order to

quantify the change in the isothermal crystallization kinetics, we make use of the Avrami

equation 42, 43

(3.3)

48

where n is the Avrami exponent and dictates the dimensionality of the crystal growth and K is

the overall crystallization rate constant. Eq. 3.3 can be rewritten as:

*

+ (3.4)

The crystallization rate constant, K, and the Avrami exponent, n, can be determined from the y-

intercept and the slope of the linear section in the plot of ln ln {1/[1-X(t)]} vs. ln (t), respectively.

Such a plot is shown in Fig. 3.4c. The obtained Avrami parameters of PP and PP/PET fibrillar

composites are summarized in Table 3.2. However, it is difficult to compare the overall

crystallization rate directly from the values of K because the unit of K is s-n

and n is not constant

for the samples. Thus, the half-time of crystallization, t1/2, the time required to achieve 50% of

the final crystallization of the samples is computed for the purpose of discussing the kinetics of

crystallization. The value of t1/2 is calculated using

(

)

(3.5)

Fig. 3.4d illustrates the variation of t1/2 as a function of the fibril content. It can be seen that t1/2

decreases with an increase in the fibril content. Such variation indicates that the overall

isothermal crystallization rate increases with an increase in fibril content due to the

heterogeneous crystal nucleation effect of the fibrils. For example, at 135°C, the t1/2 for

crystallization of PP/PET (99/1 wt%) is 490 s; however, the t1/2 for crystallization of PP/PET

(9/5 wt%) is reduced to 206 s.

3.4.5 Isothermal crystallization kinetics in presence of dissolved CO2

The dissolution of CO2 into the PP/PET fibrillar composite will influence the kinetics of

crystallization by enhancing chain mobility. 44, 45

We investigate the isothermal crystallization

kinetics of PP/PET fibrillar composites with dissolved CO2 using the Avrami analysis. The

isothermal crystallization thermograms for PP/PET fibrillar composites with different PET fibril

contents obtained at a temperature of 126°C and CO2 pressure of 3 MPa, are shown in Fig. 3.5a.

Neat PP does not show any signs of crystallization over the investigated isothermal treatment

period. However, when the PET fibrils are present, the time required for completion of

49

Fig. 3.5 Isothermal crystallization kinetics of PP/PET fibrillar composites in presence of

dissolved CO2: a) Isothermal crystallization exotherms as a function of crystallization time

calculated from DSC at a temperature of 126°C and CO2 pressure of 3 MPa; b) Relative

crystallinity, X(t), as a function of crystallization time calculated from DSC thermograms shown

in a); c) Isothermal crystallization kinetics analyzed using the Avrami equation; d) Half-times,

t1/2, of crystallization determined from the Avrami analysis of isothermal crystallization.

1 2 3 4 5 6 70

50

100

150

200

250

300

t 1/2

(s)

PET content (wt%)

d

1 2 3 4 5 6 7

-10

-8

-6

-4

-2

0

2

4

ln ln

{1/[

1-X

(t)]

}

ln (Time)

PET 7 wt%

PET 5 wt%

PET 3 wt%

PET 1 wt%

c

0 100 200 300 400 500 600

0.0

0.2

0.4

0.6

0.8

1.0

X(t

)

Time (s)

PET 7 wt%

PET 5 wt%

PET 3 wt%

PET 1 wt%

b

0 100 200 300 400 500 600

PET 7 wt%

PET 5 wt%

PET 3 wt%

PET 1 wt%

PET 0 wt%

Exo

Time (s)

a

50

Table 3.3 Summary of Avrami parameters for the isothermal crystallization kinetics of PP/PET

fibrillar composites with different PET contents at a crystallization temperature of 126°C and 3

MPa CO2 pressure.

PP/PET (wt%)

n a K (s

-n) b t1/2 (s)

c

99/1 2.20 2.98 × 10-6

2.76 × 102

97/3 2.53 1.12 × 10-6

1.93 × 102

95/5 1.98 7.83 × 10-5

97.7

93/7 2.60 7.31 × 10-5

34.0

a n is obtained from slopes of lines of best fit for the data plotted in Fig. 3.5c;

b K is obtained

from the vertical axis intercepts of lines of best fit for the data plotted in Fig. 3.5c; c

t1/2 is

determined using Eq. 3.5.

crystallization becomes increasingly shorter as the PET content is increased. This enhancement

in the crystallization rate can be seen clearly in Fig. 3.5b which shows the relative crystallinity,

X(t), as a function of crystallization time for PP/PET fibrillar composites with the PET fibril

content ranging from 1 to 7 wt%. Fig. 3.5c shows the corresponding Avrami plots to derive the

Avrami parameters, i.e., the Avrami exponent, n, and the rate constant, K, which are summarized

in Table 3.3 along with the crystallization half-times, t1/2, determined using Eq. 3.5. The

crystallization half-times, t1/2, are plotted in Fig. 3.5d as a function of the fibril content. It can be

seen that t1/2 decreases with an increase in the fibril content indicating that the kinetics of

crystallization increase with an increase in fibril content over the studied fibril content range.

DSC at a temperature of 126°C and CO2 pressure of 3 MPa; b) Relative crystallinity, X(t), as a

function of crystallization time calculated from DSC thermograms shown in a); c) Isothermal

crystallization kinetics analyzed using the Avrami equation; d) Half-times, t1/2, of crystallization

determined from the Avrami analysis of isothermal crystallization.

51

3.4.6 Effect of CO2 pressure on formation of γ-phase in PP/PET

Fig. 3.6 shows WAXS traces of PP and PP/PET fibrillar composites after isothermal

crystallization at a temperature of 135°C (under ambient pressure) or at a temperature of 126°C

(under 3 MPa CO2 pressure). Both neat PP and PP/PET fibrillar composites, isothermally

crystallized under ambient pressure, show α-polymorph Bragg peaks at 2θ = 14.3°, 17.1°, 18.7°,

21.4° and 21.9°. 46

High pressure isothermal crystallization of PP also shows peaks

corresponding to the α-polymorph, but for the case of high pressure isothermal crystallization of

the PP/PET fibrillar composite, two additional peaks specific to the γ-polymorph of PP are

observed at 2θ = 15° (corresponding to the 113 plane), and 20° (corresponding to the 117 plane)

and indicated with arrows in Fig. 3.6. 47

The γ-polymorph may occur when: 1) crystallization is

carried out at elevated pressures, 48

2) when the matrix polymer is a random copolymer of PP, 49

and 3) when high-aspect-ratio second phases are present in the matrix which can facilitate the

Fig. 3.6 Wide-angle X-ray scattering (WAXS) profiles of PP and PP/PET fibrillar composite

(95/5 wt%). For samples isothermally crystallized under ambient pressures, a temperature of

135°C is used, and for samples isothermally crystallized under CO2 pressure of 3 MPa, a

temperature of 126°C is used. The peaks corresponding to the γ-phase are marked with an arrow

(top trace).

5 10 15 20 25 30 35

PP

3 MPa

PP/PET ambient

pressure

PP/PET

3 MPa

PP ambiant

pressure

2(°)

52

growth of a transcrystalline morphology. 50

The combination of these three effects can explain

why the γ-polymorph develops in the PP/PET fibrillar composite when it is crystallized in the

presence of 3 MPa CO2 pressure and does not develop in the other cases, as presented in Fig. 3.6.

The occurrence of the γ-polymorph is representative of superior mechanical properties including

higher modulus, higher yield stress and higher flow stress. 51

3.4.7 Foaming behaviour of PP/PET fibrillar composites

Extensional flow is the primary mode of deformation during polymer foaming. When bubbles

are growing in a polymeric material, the bubble walls undergo a biaxial extension. 52

If the

matrix does not show a tendency to strain harden during extension, the bubbles are susceptible to

bursting during this growth. Polymeric materials that exhibit strain hardening show a better

capacity to withstand this stretching force due to the viscosity increment with strain. This

ultimately results in foams with a higher number of bubbles per unit volume, known as the cell

density, and also, larger volume expansion ratios since a larger amount of gas is retained within

the closed bubbles of the foam as opposed to escaping in the case of ruptured bubbles. 53, 54

From a technological standpoint, the enhanced strain hardening of PP/PET fibrillar composites is

of interest to the foam manufacturing industry. Consequently, we study the foaming behaviour of

PP/PET fibrillar composites with different PET fibril contents ranging from 0 wt% (neat PP) to 7

wt%. The samples are foamed at a CO2 pressure of 5.5 MPa and a temperature of 137°C for 1 h.

Fig. 3.7a shows representative morphologies of the obtained foams of PP/PET fibrillar

composites with four different fibril contents: i) 0 wt%, ii) 0.1 wt%, iii) 3 wt%, and iv) 7 wt%.

The cell densities of the foams of PP/PET fibrillar composites have been characterized in Fig.

3.7b. As can be seen in Fig. 3.7a-b, neat PP shows a low cell density and poor foam morphology

indicating that processes such as bubble coalescence and bubble wall rupture are highly active.

Increasing the PET content leads to an increase in the cell density and a concurrent decrease in

the average bubble size, indicating suppression in such bubble deterioration mechanisms. This

increase in the cell density with an increase in the PET content can be attributed to the strain

hardening behaviour of PP/PET fibrillar composites. 52

53

a

Fig. 3.7 Batch foaming of PP/PET fibrillar composites using CO2 as the foam blowing agent at a

pressure of 5.5 MPa, a temperature of 137°C and a saturation time of 1 h a) SEM images of

foamed samples of PP/PET fibrillar composites with different PET fibril contents: i) 0 wt% (neat

PP), ii) 0.1 wt%, iii) 3 wt%, iv) 7 wt%; b) Cell densities shown as a function of the PET fibril

content; c) Foam volume expansion ratios shown as a function of the PET fibril content.

0 1 2 3 4 5 6 710

0

102

104

106

108

1010

1012

Ce

ll d

en

sit

y (

cells/c

m3)


b

0 1 2 3 4 5 6 71

2

3

4

5

6

7

8

9

Exp

an

sio

n r

ati

o


c

54

Interestingly, however, improvements in the cell density are observed even when an extremely

low amount of PET fibrils, e.g., 0.1 wt%, are present in the PP matrix. At such low PET fibril

contents, strain hardening in extensional flow is not observed and the melt viscosity is

unaffected. We explain this observed enhancement in the cell density at such low PET contents

by the enhancement of crystallization kinetics of PP with dissolved CO2 when PET fibrils are

present because the samples are foamed at a temperature below the melting temperature of PP.

The presence of PET fibrils increases the kinetics of crystallization of PP. The nucleated crystals

can affect the foaming behaviour, in particular, the cell density through the following

mechanisms: i) the crystals can act as ‗physical crosslinks‘ that behave similar to chemical

crosslinks and improve the melt strength of the sample, thereby making bubble rupture more

difficult during the bubble growth stage of the foaming process, 55

ii) the crystals can act as

heterogeneous bubble nucleation site, by making available a surface where the activation energy

for bubble nucleation is substantially lower, 56

iii) the crystals can act as heterogeneities in the

polymeric material and can create large pressure variations in their vicinity which can lead to a

secondary bubble nucleation process. 57

In a study by Leung et al., 57

a melt flow induced by the

expansion of nucleated bubbles triggered the formation of new bubbles. They hypothesized that

the growing bubbles generate biaxial stress fields around heterogeneous sites, such as crystals or

PET fibrils, which results in microscopic pressure variations that decrease the free energy barrier

for heterogeneous bubble nucleation. 56, 58, 59

The expansion ratio of the foamed samples as a function of PET fibril content is shown in Fig. 3.

7c. Neat PP shows a poor volume expansion ratio at the foaming conditions used. The poor melt

elasticity and absence of strain hardening behaviour of the neat PP accelerates gas loss through

the ruptured bubble walls sacrificing the expansion ratio. Increasing the PET fibril content

initially leads to an increase in the foam volume expansion ratio until a maximum volume

expansion is reached. Subsequently, a further increase in the PET fibril content leads to a decline

in the foam volume expansion ratio. Thus, an optimum content of the PET fibril exists for the

foam expansion of PP/PET foams. At fibril contents below the optimum content, lower foam

expansion is observed most likely due to rapid gas loss, and at fibril contents above the optimum,

lower foam expansion is observed most likely due to stiffer walls which offer more resistance to

bubble growth.

55

3.5 Conclusion

The results of the present work indicate that the in-situ fibrillation of PET in a PP random

copolymer matrix can enhance the foaming ability of the matrix. At PET fibril contents ≥ 3 wt%,

PP/PET fibrillar composites exhibit pronounced strain hardening response which improves the

foaming behaviour because the viscosity increase with strain suppresses bubble wall rupture and

coalescence. At fibril contents < 3 wt%, no strain hardening is observed but still an improvement

in foaming behaviour is seen. This enhancement of foaming at PET fibril content < 3 wt% is

attributed to the enhanced kinetics of crystallization of PP/PET fibrillar composites, especially in

the presence of dissolved CO2. The presence of only 0.1 wt% PET fibrils in PP can yield foams

with a two orders of magnitude higher cell density compared to the neat PP matrix. An optimum

PET fibril content (approximately 4 wt%) exists for achieving PP foams with the highest volume

expansion ratio. Below this optimum content, low foam expansions are seen most likely due to

active gas loss, and above this optimum content, low foam expansions are seen most likely due

to the increased stiffness of the walls which offer resistance against bubble expansion. Our

results demonstrate that in-situ fibrillation of PET in a PP matrix using conventional

microfibrillar composite preparation methods can be effective in improving the foaming ability

of a PP matrix.

56

References



14, 225-277.

2. Wu, S. Formation of dispersed phase in incompatible polymer blends: Interfacial and

rheological effects. Polym. Eng. Sci. 1987, 27, 335-343.

3. Wu, S. Calculation of interfacial tension in polymer systems. J. Polym. Sci. Part C: Polym.

Symp. 1971, 34, 19-30.


Ser. A 1934, 146, 501-523.

5. Taylor, G. I. The viscosity of a fluid containing small drops. Proc. R. Soc. Lond. A 1932,

834, 41-48.





8. Fakirov, S.; Duhovic, M.; Maitrot, P.; Bhattacharyya, D. From PET nanofibrils to

nanofibrillar single-polymer composites. Macromol. Mater. Eng. 2010, 295, 515-518.

9. Fakirov, S.; Evstatiev, M.; Petrovich, S. Microfibrillar reinforced composites from binary

and ternary blends of polyesters and nylon 6. Macromolecules, 1993, 5219-5226.

10. Friedrich, K.; Evstatiev, M.; Fakirov, S.; Evstatiev, O.; Ishii, M.; Harrass, M. Microfibrillar

reinforced composites from PET/PP blends: processing, morphology and mechanical

properties. Comp. Sci. Tech. 2005, 65, 107-116.

57

11. Fakirov, S.; Bhattacharyya, D.; Shields, R. J. Nanofibril reinforced composites from polymer

blends. Colloids Surf., A., 2008, 313-314, 2-8.



13. Hong, J. S.; Ahn, K. H.; Lee, S. J. Strain hardening behaviour of polymer blends with fibril

morphology. Rheologica Acta 2005, 45, 202-208.

14. Jayanarayanana, K.; Joseb, T.; Thomasc, S.; Joseph, K. Effect of draw ratio on the

microstructure, thermal, tensile and dynamic rheological properties of insitu microfibrillar

composites. Eur. Polym. J. 2009, 45, 1738-1747.



2013, 54, 4645–4652.


same. WO2013137301, March 13, 2013.

17. Rizvi, A.; Park, C. B. Dispersed polypropylene fibrils improve the foaming ability of a

polyethylene matrix. Polymer 2014, 55, 4199-4205.

18. Naguib, H. E.; Park, C. B.; Reichelt, N. Fundamental foaming mechanisms governing the

volume expansion of extruded polypropylene foams. J. Appl. Polym. Sci. 2003, 91, 2661.


1997, 134, 165–234.

20. Chambon, F.; Winter, H. H. Linear viscoelasticity at the gel point of a crosslinking PDMS

with imbalanced stoichiometry. J. Rheol. 1987, 31, 683-697.

21. Winter, H. H.; Morganelli, P.; Chambon, F. Stoichiometry effects on rheology of model

polyurethanes at the gel point. Macromolecules 1988, 21, 532–535.

58

22. Nijenhuis, K. T.; Winter, H. H. Mechanical properties at the gel point of a crystallizing

poly(vinyl chloride) solution. Macromolecules 1989, 22 , 411–414.

23. Lin, Y. G.; Mallin, D. T.; Chien, J. C. W.; Winter, H. H. Dynamic mechanical measurement

of crystallization-induced gelation in thermoplastic elastomeric poly(propylene).

Macromolecules 1991, 24, 850– 854.

24. Tabatabaei, A.; Barzegari, M. R.; Nofar, M.; Park, C. B. In-situ visualization of

polypropylene crystallization during an extrusion process. Polym. Test. 2014, 33, 57-63.

25. Tabatabaei, A.; Barzegari, M. R.; Nofar, M. R.; Park, C. B. Effects of crystallinity on the

extrusion-foaming behaviours of polypropylene blown with CO2. Blowing Agents and

Foaming Processes, Mainz, 2013.

26. Xu, X.; Park, C. B.; Xu, D.; Pop-Iliev, R. Effect of die geometry on cell nucleation of PS

foams blown with CO2. Polym. Eng. Sci. 2003, 43 (7), 1378-1390.

27. Ali, M. A.; Nobukawa, S.; Yamaguchi, M. Morphology development of

polytetrafluoroethylene in a polypropylene melt (IUPAC Technical Report). Pure Appl.

Chem. 2011, 83, 1819-1830.




R. Soc. London Ser. A77 1906, 77, 426–440.






59



779-793.







35. Shenoy, A. V. Rheology of filled polymer systems; Kluwer Academic Publishers: Dordrecht,

1999.

36. Potschke, P.; Fornes, P.; Paul, D. R. Rheological behaviour of multiwalled carbon

nanotube/polycarbonate composites. Polymer 2002, 43, 3247–3255.



8, 8539-8546.

38. Larson, R. G. The structure and rheology of complex fluids. Oxford University Press: New

York, 1999; pp 14-15.

39. Li, L.; Li, C. Y.; Ni, C.; Rong, L.; Hsiao, B. Structure and crystallization behaviour of Nylon

66/multi-walled carbon nanotube nanocomposites at low carbon nanotube contents. Polymer

2007, 48, 3452-3460.

40. Auer, C.; Kalinka, G.; Krause, T.; Hinrichsen, G. Crystallization kinetics of pure and fiber-

reinforced poly(phenylene sulfide). J. Appl. Polym. Sci. 1994, 51, 407-413.

41. Piorkowska, E.; Galeskia, A.; Haudinb, J. Critical assessment of overall crystallization

60

kinetics theories and predictions. Prog. Polym. Sci. 2006, 31, 549-575.

42. Avrami, M. Kinetics of phase change-II Tranformation-time relations for random distribution

of nuclei. J. Chem. Phys. 1940, 8, 212–224.

43. Avrami, M. Kinetics of phase change-I. General theory. J. Chem. Phys. 1939, 7, 1103-1112.

44. Nofar, M.; Zhu, W.; Park, C. B. Effect of Dissolved CO2 on Crystallization Behaviour of

PLA. Polymer 2012, 53, 3341– 3353.

45. Nofar, M.; Tabatabaei, A.; Park, C. B. . Effects of Nano-/Micro-Sized Additives on the

Crystallization Behaviours of PLA and PLA/CO2 Mixtures. Polymer 2013, accepted.

46. Miller, R. L. On the existence of near-range order in isotactic polypropylenes. Polymer 1960,

1, 135-143.

47. Meille, S. V.; Brückner, S. Non-parallel chains in crystalline γ-isotactic polypropylene.

Nature 1989, 340, 455-457.

48. Hosier, I. L.; Alamo, R. G.; Exteso, P.; Isasi, J. R.; Mandelkern, L. Formation of the α and γ

polymorphs in random metallocene−propylene copolymers. Effect of concentration and type

of comonomer. Macromolecules 2003, 36, 5623-5636.

49. Pae, K. D.; Morrow, D. R.; Sauer, J. A. Interior morphology of bulk polypropylene. Nature

1966, 211, 514-515.

50. Dean, D. M.; Register, R. A. γ-Transcrystallization in isotactic polypropylene-based

composites promoted by aramid fibers. J. Polym. Sci. Part B: Polym. Phys. 1999, 37, 2534-

2538.

51. Lezak, E.; Bartczak, Z.; Galeski, A. Plastic deformation of the γ phase in isotactic

polypropylene in plane-strain compression. Macromolecules 2006, 39, 4811-4819.

52. Yamaguchi, M. Melt elasticity of polyolefins: impact of elastic properties of foam

61

processing. In Polymeric Foams: Mechanisms and Materials; Lee, S. T., Ramesh, N. S.,

Eds.; CRC Press: Boca Raton, 2004; pp 19-72.

53. Wang, J.; Zhu, W.; Zhang, H.; Park, C. B. Continuous processing of low-density,

microcellular poly(lactic acid) foams with controlled cell morphology and crystallinity.

Chem. Eng. Sci. 2012, 75, 390-399.

54. Zhai, W.; Wang, J.; Chen, N.; Naguib, H.; Park, C. B. The orientation of carbon nanotubes in

poly(ethylene-co-octene) microcellular foaming and its suppression effect on cell

coalescence. Polym. Eng. Sci. 2012, 52, 2078-2089.

55. Hingmann, R.; Rieger, J.; Kersting, M. Rheological Properties of a Partially Molten

Polypropylene Random Copolymer during Annealing. Macromolecules 1995, 28, 3801-3806.

56. Wong, A.; Guo, Y.; Park, C. B. Fundamental Mechanisms of Cell Nucleation in

Polypropylene Foaming with Supercritical Carbon Dioxide - Effects of Extensional Stresses

and Crystals. J. Supercrit. Fluids 2013, 79, 142–151.

57. Leung, S. N.; Wong, A.; Wang, C.; Park, C. B. Mechanism of extensional stress-induced cell

formation in polymeric foaming processes with the presence of nucleating agents‖. J.

Supercrit. Fluids 2012, 63, 187-198.

58. Wong, A.; Wijnands, S. F. L.; Kuboki, T.; Park, C. B. Mechanisms of Nanoclay-Enhanced

Plastic Foaming Processes – Effects of Nanoclay Intercalation and Exfoliation. J. Nanopart.

Res. 2013, 15, 1815-1829.

59. Wong, A.; Park, C. B. The effects of extensional stresses on the foamability of polystyrene-

talc composites blown with carbon dioxide. Chem. Eng. Sci. 2012, 75, 49-62.

62

Chapter 4

Preparing fibrillated polyethylene terephthalate in poly(propylene-co-ethylene) through fiber spinning for foam processing 3

4.1 Abstract

A fiber spinning process is used to fibrillate polyethylene terephthalate (PET) fibrils in

poly(propylene-co-ethylene) random copolymer (PP). Macroscopic extensional flows applied

during fiber spinning cause both the PP matrix and the PET domains to stretch and orient in the

flow direction. When these blend fibers are isothermally treated at a temperature between the

melting temperatures of the two components, PP relaxes while the PET fibrils maintain their

morphology. These fibrils influence the viscoelastic behaviour of the PP matrix. The uniaxial

extensional viscosity measurements show strain hardening behaviour for the PP/fibrillated-PET,

not observed for the neat PP matrix or for PP/PET blends when the PET domains are spherical.

This strain hardening behaviour is attributed to the formation of a disordered physical network of

flexible PET fibrils. The oscillatory shear behaviour in the linear viscoelastic region is studied to

understand the percolation properties of PET fibrils in the PP matrix. When the PET domains are

fibrillated, the storage (G‘) and loss (G‘‘) moduli show an increase and their slopes decrease at

low frequencies compared to neat PP or when the PET domains are spherical, indicating that the

fibril network responds elastically over long timescales. The wide-angle X-ray scattering

(WAXS) data shows the presence of γ-polymorph crystals of PP in both PP and PP/PET after the

fiber spinning process. Foam extrusion is used as a model polymer process to study the effect of

the PET fibrils on the processability of PP. Results reveal that the presence of PET fibrils in PP

yields foams that exhibit up to two orders of magnitude higher number of bubbles per unit

volume and up to a five-fold increase in the expansion ratio relative to the neat PP. Enhancing

the foaming ability of polymer blends by fibrillating the dispersed phase using fiber spinning is

technologically promising.


63

4.2 Introduction

Fiber-spinning has been established as an effective means of producing fibrillar morphologies of

immiscible polymer blends in a large-scale and low-cost process. 1 The following stages are

involved in preparing fibrillar blends through fiber-spinning: 1) An immiscible polymer blend

with components that exhibit distinctly different melting temperatures (Tm) is prepared by melt

blending in a twin-screw extruder to get well dispersed domains of the dispersed phase in the

continuous matrix; 2) The immiscible polymer blend is pumped through a spinneret of a fiber

spinning system. The applied macroscopic extensional flow thins down the blend, which

solidifies on cooling, resulting the formation of blend fibers. The microstructure of the blend

fibers is characterized by elongated domains of the dispersed phase, referred to as fibrils, 1, 2, 3

and oriented polymer chains of the continuous phase in the direction of the flow. 4

The extent of

molecular orientation of the matrix polymer and the degree of elongation of the fibrils is

governed by parameters such as speed of the draw rolls, dimensions of the spinneret, spinning

temperature, and rate of cooling of the spun material; 4 3) the blend fibers are isothermally

treated at a temperature between the Tm of the blend components to yield disordered polymer

fibrils in an isotropic continuous matrix.

The presence of high-aspect ratio fibrils in a polymeric-host can modify the properties and

processing characteristics of the matrix. 5, 6

The high aspect-ratio imparts flexibility to the fibrils

allowing them to bend substantially in response to the fibril-fibril interactions. Above a critical

concentration known as the rheological percolation threshold, this bending causes the formation

of a disordered physical network characterized by topological interactions (entanglements). 7

Such a network of fibrils is expected to create additional contributions to both the solid-state and

viscoelastic properties of the matrix. 6, 8

Thus, polymer blends with fibrillar morphologies

demonstrate fundamentally and commercially interesting properties.

In early papers, we examined the effects of the fibrillar dispersed phase morphologies of polymer

blends prepared on a small-scale microcompounder and found that they had a significant effect

on the rheological and crystallization properties, and processing characteristics. 9, 10

Strain

hardening in a uniaxial extensional flow was observed in the fibrillar blend, not seen in the host-

polymer. Analogous to the mechanism of strain hardening during the uniaxial extension of

entangled branched polymers, this effect was attributed to the formation of a network of

64

entangled fibrils which could not disentangle readily enough to follow the extension. The present

study extends this previous work by showing that large-scale production of fibrillar blends of

polyethylene terephthalate (PET) in a poly(propylene-co-ethylene) (PP) matrix through fiber

spinning can give rise to rheological properties such as long relaxation times, pronounced elastic

properties and strain hardening responses in extensional flows. Microstructure evolution of the

PP/PET fibrillar blend is characterized at different draw ratios of the spinning process. The

rheology of the PP/PET fibrillar blend is studied under a uniaxial extensional flow. A physical

explanation is proposed for the observed changes in the extensional viscosity based on the

underlying blend morphology and supported by measurements of the frequency-dependent linear

viscoelastic moduli. Wide angle X-ray scattering (WAXS) is employed to identify any changes

in the crystal structure of fiber-spun PP/PET. Understanding the rheological and crystallization

behaviours of a polymeric system is of practical relevance in polymer processing. An example is

polymer foam extrusion where strain hardening in an extensional flow can prevent foam bubbles

from rupturing and can stabilize the foam structure. 9, 11, 12

Using continuous foam extrusion as a

model for polymer processing, a further aim of this study is to show how changes in rheology

and crystallization induced by the morphology of PET domains, can alter the foaming ability of

the PP matrix that has inherently low melt strength. 13, 14

4.3 Experimental

4.3.1 Materials

The matrix polymer employed in this study is a poly(propylene-co-ethylene) random copolymer

(PP) commercially available as Sabic ® PP 670K with a melt flow index (MFI) of 10 g/10 min at

230°C/2.16 kg. The dispersed phase polymer employed in this study is an amorphous

polyethylene terephthalate (PET) copolymerized with 3.5 % 1,4-

cyclohexanedimethanol (CHDM), commercially available as Eastapak ® PET 9921 (Mn =26 000

g mol-1

). Carbon dioxide (CO2) is purchased from Linde Gas with purity in excess of 99%.

Xylene is purchased from Caledon Laboratories Ltd. and used as received.

The melting transition of PP and PET is determined using differential scanning calorimetry

(DSC) under a nitrogen atmosphere. A heat-cool-heat experiment is performed from 25°C to

260°C. Heating and cooling rates of 10°C/min are used. The melting transition temperature is

65

derived from the melting peak maxima during the second heating cycle. The melting transition

peaks for PP and PET are seen at temperatures of 149°C and 240°C, respectively.

4.3.2 Blend preparation

Blends of PP and dried PET are prepared using a Leistritz co-rotating twin-screw extruder with a

screw diameter of 27 mm and an aspect ratio of 40. The blend composition used is PP/PET (95/5

wt%). Neat PP is processed using the same method as the PP/PET blend to keep the

thermal/processing history consistent. The samples are fed in the main feed opening of the twin-

screw extruder using an automatic feeder. The temperatures of the feeding zone, the extruder

barrel, and the die are maintained at 140°C, 255°C, and 220°C, respectively. A screw speed of

100 rpm is used. The extrudate from the die is shaped into a cylindrical strand, led into a water

bath and pelletized using a pelletizer. The resultant blend shows well dispersed spherical

domains of PET in the PP matrix as shown in Fig. 4.2a.

4.3.3 Fiber spinning of blend

Fiber spinning of the PP/PET (95/5 wt%) blend prepared in the twin-screw extruder is conducted

on a custom-built fiber spinning line shown in Fig. 4.1a. Neat PP that has also been processed in

the twin-screw extruder is fiber-spun under the same conditions as the PP/PET blend to keep the

thermal and processing history consistent. The samples are fed into the hopper of the fiber

spinning system which comprises a 19 mm single screw extruder. The temperature of the

extruder barrel is maintained at a temperature of 237°C. At this temperature, the PET domains

are soft and deformable. Therefore, the PET domains can elongate when subjected to strong

extensional flows. The extruder screw speed is set to 15 rpm. An Omega FMX-84441-S six

element static mixer with a diameter of 6.8 mm and a Labcore H-04669-12 heat exchanger are

positioned downstream to the extruder. An Oerlikon Barmag ZP504-0-IZ gear pump is attached

after the heat exchanger to regulate the melt flow before it reaches the spinneret. The gear pump

speed is maintained at 15 rpm. The spinneret comprises of a capillary die with a diameter of 0.6

66

Fig. 4.1 Schematics of: a) the custom-built fiber spinning system used in the study. The draw

ratio of the fibers is controlled by controlling the rotation rate of the godet (draw roll); b) the

custom-built tandem foam extrusion system used to prepare foams of the samples in a continuous

process using CO2 as the foam blowing agent.

mm and a length of 3.6 mm. As the extrudate exits the spinneret, it passes through a cross-flow

ventilation system which cools the extrudate before it comes in contact with the draw rolls,

known as godets. The rotational motion of the godet draws the extrudate. By controlling the

rotational rate of the godet the extrudate draw ratio can be controlled. The draw ratios studied are

1:1 (corresponding to an undrawn sample), 10.2:1 and 20.4:1, which are easily achievable under

the specified conditions. The PET exhibits a well-dispersed, completely fibrillated morphology

a b

Sample

67

in the PP matrix when the PP/PET blend is subjected to fiber spinning at a draw ratio of 20.4:1,

as depicted in Fig. 4.2d.

4.3.4 Uniaxial extensional viscosity measurements


(EVF, TA Instruments) attached to an Advanced Rheometric Expansion System (ARES). The

samples are tested at strain rates from 0.01 to 1 s-1

and at a sample environment temperature of

170°C. This temperature provides us with the most reproducible and consistent results. The

strain rates higher than 1 s-1

are experimentally inaccessible because sample breakage occurs in

the initial stages and homogeneous extension cannot be accomplished. Higher temperatures are

also inaccessible because of sagging of the polymer matrix.

4.3.5 Linear viscoelastic shear response


using the strain-controlled ARES rotational rheometer (TA Instruments), and cone-and-plate

geometry. The diameter of the cone is 25 mm and the cone angle is 5°. The shear elastic

modulus, G‘ and viscous modulus, G‘‘, are evaluated at a temperature of 170°C under a nitrogen

atmosphere to avoid oxidative degradation. The temperature is chosen so that the PET dispersed

phase morphology can be preserved during experiments. Frequency sweeps between 0.1 and 100

rad/s are carried out at strains within the linear viscoelastic range. Repeated sweeps with

increasing and decreasing frequencies show that the material is stable under the measurement

conditions. Prolonged heating at 170°C of the samples in the rheometer results in no change in

the viscoelastic data.

4.3.6 Wide-angle X-ray scattering

The crystal structure of PP in the samples is studied using wide angle X-ray scattering (WAXS).

The X-ray diffraction patterns are collected on a Bruker AXS D2 Phaser diffractometer. High

resolution CuKα source is used operating at 30 KV and 10 mA. The system is equipped with Ni-

filter for elimination of CuK peaks and a solid state Lynxeye XE detector. WAXS traces are

obtained for samples of PP/PET before fiber spinning (i.e., after twin screw extrusion of

PP/PET) and for PP/PET after fiber spinning. WAXS traces for neat PP before fiber spinning

68

(i.e., after twin screw extrusion of neat PP) and for neat PP after fiber spinning are also obtained

for comparison.

4.3.7 Foam extrusion procedure

A tandem foam extrusion system similar to those employed by the foam manufacturing industry

is used to foam the samples. The tandem foam system comprises of two single-screw extruder

barrels. The first extruder is a 5 horsepower (hp) Brabender 05-25-00 consisting of a mixing

screw with a diameter of 19 mm and an aspect ratio of 30. The second extruder is a 15 hp Killion

KN-150 consisting of a mixing screw with a diameter of 38.1 mm and an aspect ratio of 30. Fig.

4.1b gives a schematic of the configuration of the extrusion system. A metered amount of CO2

gas is injected into the melt through an injection port positioned at the first extruder.

The lack of tumbling motion for the as-spun fibers prevents these materials from being pushed

forward by the screw of the tandem foam extrusion system. Therefore, the diameter of the fiber-

spun materials is increased to approximately 2 mm by making the samples pass through a

capillary die attached to a twin-screw extruder at a temperature of 170°C. It should be mentioned

that the PP is completely isotropic, whereas the PET domains continue to retain their fibrillar

morphology because they remain in solid-state throughout the process. The cylindrical strands of

2 mm diameter are led into a water bath and pelletized. These 2 mm pellets are able to tumble

and thereby flow in the feeding zone of the foam extrusion system. The first extruder barrel of

the foam extrusion system is maintained at 170°C. The matrix polymer in the sample melts

completely in the first extruder due to the temperature as well as the screw motion which causes

shear heating. 7 wt% CO2 is injected into the first extruder barrel at a constant flow rate using a

syringe pump. The high shear and high pressure caused by the rotating screw inside the first

extruder barrel dissolves the gas in the melt through convective diffusion. The temperature of the

second extruder is gradually reduced. Consequently, the viscosity of the sample increases in the

second extruder. Foaming of the polymer melt occurs at the die exit where the PP/PET + CO2 or

PP + CO2 system is subjected to rapid depressurization resulting in the gas to undergo phase

separation. We employ a brass capillary die comprising of a circular pinhole with a diameter of

1.2 mm and a channel length of 10 mm. The temperature of the second extruder barrel and the

die is brought down and the foamed samples collected at each set temperature only after the

system temperature has equilibrated. Care is taken to ensure the die pressure is above the

69

solubility pressure of CO2 in the PP matrix to prevent phase separation before foaming initiates

inside the die.

4.3.8 Morphology characterization

To observe the morphology development of PET in PP/PET (95/5 wt%) after blending or fiber

spinning at different draw ratios, the sample is compression molded into a 2 mm thick sheet at a

temperature of 170°C for 10 min. The compression molded sheet is cryogenically fractured and

the exposed surface is subjected to a solvent vapour etching process using xylene. The sample is

exposed to xylene vapors for 45 min at 60°C. This type of treatment preferentially etches the PP

matrix which is soluble in xylene, leading to bringing up the PET domains and increasing

differentiation in observation. A JEOL 6060 scanning electron microscope (SEM) is used to

examine the structure.

To observe the morphology of the foamed samples, the foamed filament is cryogenically

fractured to expose the foam structure. The fractured surface is sputter coated with platinum

before observation under SEM. Final images are recorded from randomly selected areas at the

magnifications indicated in the SEM micrographs. For each foamed sample, the foam density is

determined using the water displacement method according to ASTM-D792. Additionally, the

number of bubbles per unit volume of unfoamed polymer, 15

called the cell density, is also

estimated from the SEM images of the foams.


4.4.1 Effect of the draw ratio on the morphology of PP/PET

The morphology development in melt blended PP/PET and fiber spun PP/PET is examined using

SEM. Fig. 4.2a shows the morphology of twin-screw extruded PP/PET blends and Fig. 4.2b-d

shows the morphology of PP/PET blends after passing the blend through the fiber spinning

system. Fig. 4.2a shows well dispersed isolated domains of PET in the PP matrix with no

evidence of fibrillation. Fig. 4.2b shows the morphology of PP/PET after pumping the blend

through the spinneret of the fiber spinning line without drawing (i.e., the draw rate is 1:1). Some

70

Fig. 4.2 PP/PET (95/5 wt%) blend morphology after, a) melt-blending PP and PET in a twin-

screw extruder. The PET phase exists as isolated, spherical domains. b) passing the PP/PET

blend through the spinneret without drawing. Some degree of fibrillation is visible due to the

transverse contraction and longitudinal elongation the blend experiences upon passing through

the orifice of the spinneret; c) pumping the PP/PET blend through the spinneret and drawing at a

draw ratio of 10.2:1. An increase in the degree of fibrillation of the PET domains can be seen; d)

pumping the PP/PET blend through the spinneret and drawing using the godet (draw rolls) at a

higher draw ratio of 20.4:1. The PET domains seem to be completely fibrillated at this draw

ratio.

degree of fibrillation is noticed and captured in the electron micrograph. We attribute this

fibrillation of PET to the extensional flow that occurs in the capillary die of the spinneret where

the blend experiences a transverse contraction and longitudinal elongation. Fig. 4.2c shows the

morphology of PP/PET after pumping the blend through the spinneret of the fiber spinning line

a b

c d

71

and drawing at a draw ratio of 10.2:1 using the godets (draw rolls). A larger degree of fibrillation

is seen after this process and the fibrillation is attributed to two factors: 1) the longitudinal

elongation/transverse contraction in the spinneret die, and 2) the extension caused by the cold-

drawing performed using the godet. A distinctive morphology of the PET domains can be

observed in the electron micrograph shown in Fig. 4.2c comprising of a bead-in-a-thread type

structure. Such morphology is observed in other studies and may correspond to a transient

intermediate state during the morphological transition of the dispersed phase from a spherical

domain shape to a completely fibrillated one. 16, 17

When the draw ratio is increased to 20.4:1 by

increasing the rate of rotation of the godets, the resultant morphology, shown in Fig. 4.2d,

reveals that the PET domains are completely fibrillated in the PP matrix. The average fibril

diameter and length is derived from the electron micrograph images and are determined to be

210 nm, and 38 µm, respectively. Thus, the aspect ratio is about 181. High aspect ratio fibrils are

known to contribute substantially towards the viscoelasticity of the matrix polymer, particularly

in uniaxial extensional flow. 18

Consequently, in the subsequent section we evaluate the uniaxial

extensional flow behaviour of the blend where the PET domains are fibrillated and compare it

with the blend where the PET domains are spherical.

4.4.2 Uniaxial extensional flow behaviour of PP/PET

The extensional viscosity measures the resistance against extensional flow. The transient


stress growth coefficient ( ):

( )

( )

(4.1)

where ( ) is the tensile stress growth function and is the extensional strain rate.

According to the Trouton relation, 27

for nonlinear strain hardening response, the ( ) is

equal to 3 times the linear viscoelastic shear viscosity for a sufficiently small strain rate.

72

Fig. 4.3 Uniaxial extensional viscosity, , of melt blended PP/PET (95/5 wt%) with

spherical PET domains, and PP/PET fibrillar blends after fiber spinning process at a draw ratio

of 20.4:1. A temperature of 170°C and strain rates of 0.01, 0.1 and 1 s-1

are used. The solid lines

represent the where is the growth curve of shear viscosity in the linear region

obtained from startup shear experiments for each sample at a strain rate of 0.001 s-1

and a

temperature of 170°C. The curves for PP/spherical-PET are offset by a factor of -10 to prevent

overlap with the curves of the PP/fibrillated-PET.

Deviation from the Trouton relation can be used to evaluate the degree of strain hardening

through,

( )

(4.2)

Uniaxial extension experiments are conducted to understand the uniaxial extensional behaviour

of the PP/PET (95/5 wt%) fibrillar blends obtained after fiber spinning at a draw ratio of 20.4:1.

For comparison purposes the uniaxial extensional behaviour of PP/PET (95/5 wt%) after melt

blending in a twin-screw extruder is also included, where the PET domains maintain a spherical

10-1

100

101

102

103

104

105

106

PP/spherical-PET

+

+

1 s-1

0.5 s-1

0.1 s-1

0.01 s-1

+ E(t

, P

a.s

]

Time (s)

PP/fibrillated-PET

-10x

73

morphology. The uniaxial extensional viscosity as a function of time is plotted in Fig. 4.3. The

growth curves of uniaxial extension are obtained at 170°C, to ensure the PET domains remain in

solid-state and their morphology is retained during the experiments. 3 obtained from a

steady shear start-up experiment at a temperature of 170°C, and a strain rate of 0.001 s-1

is

included in the graphs to indicate the linear viscoelastic behaviour limit. When the PET domains

in PP are spherical, the uniaxial extensional viscosity for all applied extensional strain rates

superimpose on the Trouton prediction and no strain hardening is observed over the rate scale

studied (0.01 to 1 s-1

). In contrast, when PET domains in PP are fibrillated, pronounced strain

hardening behaviour in the time dependent uniaxial extensional viscosity measurements is

observed. From Fig. 4.3 it can be seen that at extensional strain rates ≥ 0.1 s-1

, the stress

component in the direction of extension grows above the linear viscoelastic prediction. This

behaviour is analogous to that of branched polymers, which undergo hardening during stretching.

19, 20, 21

The divergence of behaviour during uniaxial extensional flow in PP/spherical-PET and

PP/fibrillated-PET results from the formation of a network superstructure via physical

entanglements of high aspect ratio PET fibrils that occur in PP/fibrillated-PET. In response to an

extensional flow, the network of fibrils is not able to disentangle readily enough to follow the

deformation and strain hardening responses are observed. 5, 6, 22

Other proposed mechanisms for

the origins of strain hardening during uniaxial extension include the elastic response to bending

deformation and frictional force between fibrils. 6, 5

Therefore, the origin of the strain hardening

response in uniaxial extension is related to the percolation behaviour of the PET fibrils in the

PP/fibrillated-PET. The subsequent section uses linear viscoelastic shear data to study the

rheological percolation behaviour of the system.

4.4.3 Linear viscoelastic characterization of PP/PET

The presence of a percolated fibrillar network defined by topological interactions

(entanglements) can be detected by characterizing the storage (G‘) and loss (G‘‘) moduli as a

function of frequency ( . Fig. 4.4a illustrates G‘ and G‘‘ vs. for the following three samples:

neat PP, melt-blended PP/PET (95/5 wt%) before the application of the fiber spinning, and

74

Fig. 4.4 Linear viscoelastic behaviour of neat PP after fiber spinning, PP/spherical-PET (95/5

wt%) after melt blending and PP/fibrillated-PET (95/5 wt%) after fiber spinning at a draw ratio

of 20.4:1. a) Frequency dependence of the storage (G‘) and loss (G‘‘) moduli for the three

samples at 170°C. PP and PP/spherical-PET show low frequency slopes for G‘ and G‘‘ of

approximately 1 and 2, respectively, however, PP/fibrillated-PET, shows a lower slope; b)

Frequency dependence of loss tangent (tan δ = G‘‘/G‘) for the three samples. PP and

PP/spherical-PET show negative slopes over the entire frequency range, however, PP/fibrillated-

PET shows a positive slope at low frequencies but a negative slope at high frequencies; c) Han

Plots (G‘ vs. G‘‘) for the three samples. Such plots can be used to identify microstructural

differences in composites. 23

PP/PET (95/5 wt%) after fiber spinning at a drawing ratio of 20.4:1. Before fiber spinning, the

PET domains maintain isolated, spherical morphologies and after the fiber spinning process, the

PET domains are fibrillated. Neat PP exhibits the rheological behaviour of a homogeneous

isotropic polymeric melt with liquid-like properties, that is, the slope of the G‘‘ and G‘ curves is

about 1 and 2 respectively, and G‘‘ exceeds G‘ over the studied frequency range. PP/spherical-

PET shows a very similar G‘‘ curve to neat PP but the G‘ curve is higher, specifically at low ,

although the slope remains similar. We attribute this slight shift in G‘ at low frequencies to the

presence of solid spherical PET domains that behave as rigid heterogeneities and influence the

stress relaxation behaviour of the matrix. This effect is more pronounced in G‘ than in G‘‘. 24

PP/fibrillated-PET shows a different rheological behaviour and G‘ seems to reach a plateau value

at low frequencies, indicating a transition from liquid-like to gel-like viscoelastic behaviour. We

attribute this absence of relaxation behaviour at low to the formation of a network of PET

75

fibrils in the PP matrix. It seems that the solid-state, flexible PET fibrils tend to restrict the long

range motion of the matrix polymer chains and prevents them from complete relaxation when

subject to shear. 30

The frequency dependence of the loss tangent (tan δ = G‘‘/G‘) for the three samples is depicted

in Fig. 4.4b. According to the Winter-Chambon criteria, viscoelastic liquids exhibit a decrease in

tan δ with an increase in . Systems that exhibit an increase in tan δ with an increase possess a

pronounced solid-like character. 25

While neat PP and PP/spherical-PET show the frequency

dependence of tan δ typical of viscoelastic liquids with tan δ decreasing monotonically with an

increase in , PP/fibrillated-PET shows an increase in tan δ with an increase in at low

frequencies, which is a rheological consequence of a physical gel. At high frequencies however,

the PP/PET fibrillar blend shows a negative slope typical of a liquid-like system. We attribute

this viscoelastic liquid-like response at high frequencies to the breakdown of the fibrillar network

caused by the increase in shear stress at higher frequencies which transitions the solid-like

behaviour to that of a viscoelastic liquid.

Fig. 4.4c shows curves of G‘ vs. G‘‘ often referred to as Han plots for neat PP, PP/spherical-PET

and PP/fibrillated-PET. This plot is also used to study the microstructure differences in the

matrix polymer and the composite system at a given temperature. 23,

24

A slope of 2 at low

frequencies corresponds to an isotropic homogeneous system with liquid-like linear

viscoelasticity. 23

Such a behaviour can be seen for neat PP which has a slope of 1.9. The slope

decreases slightly for the case of the melt blended PP/PET with spherical PET domains to 1.7. A

substantial decrease in the slope is observable for the case of PP/PET fibrillar blends to 0.5.

Polymeric systems that exhibit a slope < 2 in the low frequency regime of the G‘ vs. G‘‘ plot

possess a higher degree of heterogeneous microphases. The significant decrease in the slope at

low frequencies for the case of PP/PET fibrillar blends after fiber spinning can be attributed to

the existence of a percolated structure originating from topological interactions between the PET

fibrils or from interactions between the PET fibrils and the polymer chains. At high frequencies,

the increase in shear stress causes the network structure in the PP/PET fibrillar blend to gradually

disintegrate and approach a homogeneous state. This explains why a higher degree of overlap in

the three curves is observed at high frequencies.

76

4.4.4 Effect of processing on PP crystal structure in PP/PET

10 15 20 25 30 35

PP before

fiber spinning

PP after

fiber spinning

PET

PP/spherical-PET

2(°)

PP/fibrillated-PET

Fig. 4.5 Wide-angle X-ray scattering (WAXS) traces of neat PP before fiber spinning, neat PP

after fiber spinning, PP/spherical-PET (95/5 wt%), and PP/fibrillated PET (95/5 wt%) after fiber

spinning. The draw ratio the fiber spinning process is 20.4:1. WAXS profile of neat PET is also

shown.

Fig. 4.5 shows Wide-angle X-ray scattering (WAXS) traces of PP/spherical-PET (95/5 wt%)

obtained after melt blending, and PP/fibrillated-PET (95/5 wt%) obtained after fiber spinning at a

draw ratio of 20.4:1. Neat PP that has undergone the exact same processing is included for

comparison purposes. The WAXS trace of neat PET is also included. The absence of peaks in

Fig. 4.5 for neat PET indicates it is an amorphous polymer. The PET used in this study contains

3.5% copolymerized 1,4-cyclohexanedimethanol (CHDM) which renders PET amorphous. 26

Melt blended PP before fiber spinning and melt blended PP/spherical-PET show α-polymorph

Bragg peaks at 2θ = 14.3°, 17.1°, 18.7°, 21.4° and 21.9°. 27

Two additional peaks for fiber-spun

PP and fiber-spun PP/fibrillated-PET specific to the γ-polymorph of PP are observed and marked

117 plane), with PP/fibrillated-PET showing a higher intensity of the γ-peaks than fiber-spun PP.

77

28 We attribute the appearance of the γ-peaks in fiber-spun PP to the extensional deformational

flow that occurs during fiber spinning, and in PP/fibrillated-PET to with arrows in Fig. 4.5 at 2θ

= 15° (corresponding to the 113 plane), and 20° (corresponding to the peaks for fiber-spun PP

and fiber-spun PP/fibrillated-PET specific to the γ-polymorph of PP are observed and marked

with

arrows in Fig. 4.5 at 2θ = 15° (corresponding to the 113 plane), and 20° (corresponding to the

117 plane), with PP/fibrillated-PET showing a higher intensity of the γ-peaks than fiber-spun PP.

28 We attribute the appearance of the γ-peaks in fiber-spun PP to the extensional deformational

flow that occurs during fiber spinning, and in PP/fibrillated-PET to two factors: 1) the strong

extensional deformation that occurs during fiber spinning, and 2) to the presence of high-aspect-

ratio fibrils which are known to facilitate γ-phase growth in PP. 29, 30, 31, 32

The occurrence of the

γ-polymorph is representative of superior mechanical properties. 33

In the subsequent section we

characterize the solid-state tensile properties of PP/fibrillated-PET, PP/spherical-PET and neat

PP.

4.4.5 Continuous foam extrusion of PP/PET

Extrusion foaming is conducted on the melt blended PP/spherical-PET (95/5 wt%) and

PP/fibrillated-PET (95/5 wt%) after fiber spinning at a draw ratio of 20.4:1. For comparison,

extrusion foaming is also conducted for neat PP that has underwent the exact same thermal and

processing history as the fiber-spun PP/PET. All processing is performed at temperatures where

the PET is in solid-state so that its morphology is maintained. Care is taken to ensure that the die

pressure is above the solubility pressure of CO2 in molten PP to prevent polymer-gas phase

separation prior to foaming at the die exit. Fig. 4.6a shows the bubble structure of these three

samples at three different die temperatures (140°C, 145°C and 150°C). A decrease in bubble size

and an improvement in the uniformity of the foam structure can be observed at the studied

temperatures in the following order: neat PP < PP/spherical-PET < PP/fibrillated-PET. To

quantify this improvement in the foam structure, the cell density, which is the number of bubbles

per unit volume, and the foam volume expansion ratio, which is the ratio between the density of

the polymer system before and after foaming, are calculated as a function of the extrusion die

temperature and are shown in Fig. 4.6b-c. Fig. 4.6b shows that over the studied temperature

78

a

Fig. 4.6 Characterization of the foam morphologies of PP/spherical-PET (95/5 wt%) and

PP/fibrillated-PET obtained using fiber spinning at a draw ratio of 20.4:1. For comparison

purposes the foam morphologies of PP that has undergone the same thermal and processing

history as the fiber-spun PP/PET is also included; a) SEM micrographs of foams obtained after

the continuous foam extrusion process of the three samples at three different temperatures:

140°C, 145°C, and 150°C; b) cell density as a function of the die temperature for the three

samples; c) Foam volume expansion ratio as a function of the die temperature for the three

samples.

130 135 140 145 150 155 160 1650

5

10

15

20

25

30

35

Exp

an

sio

n R

ati

o

Die Temperature (°C)

PP/fibrillar-PET (draw ratio 20.4:1)

PP/spherical-PET

PP

c

130 135 140 145 150 155 160 16510

3

104

105

106

107

108

109

1010

PP/fibrillar-PET (draw ratio 20.4:1)

PP/spherical-PET

PP

Bu

bb

le d

en

sit

y (

bu

bb

les/ cm

3)

Die Temperature (°C)

b

79

range, the cell density of the neat PP ranges from 105 to 10

7 cell/cm

3, PP/spherical-PET ranges

from 106 to 10

8 cells/cm

3, and PP/fibrillated-PET ranges from 10

8 to 10

9 cell/cm

3. PP/spherical-

PET shows higher cell densities than the neat PP due to the presence of the solid-state spherical

PET domains which act as heterogeneities and induce microscopic stress variations inside the

polymer matrix, thereby decreasing the energy barrier for bubble nucleation. 34, 35, 36

This effect is

enhanced for the case of PP/fibrillated-PET because the high aspect-ratio, submicron fibrils of

PET offer a larger surface-to-volume ratio compared to the spherical PET making available more

heterogeneous surfaces for inducing microscopic stress variations in the polymer and facilitating

bubble nucleation. 13, 37, 38

Additional surfaces for bubble nucleation are provided by the growth

of crystalline heterogeneities such as a transcrystalline layer around PET fibrils. 39

The

transcrystalline heterogeneities occur because strong flow fields are applied on the polymer

matrix at a temperature below the melting temperature of the matrix during the foam extrusion

process which facilitates crystal nucleation. 40, 41

The strain hardening in extensional flow

observed for PP/fibrillated-PET after fiber spinning at a draw ratio of 20.4:1 also influences the

cell density. 15

The primary mode of deformation during polymer foaming is extensional flow. A

biaxial extension is exerted on the polymer matrix when a nucleated bubble undergoes

expansion. Any occurrence of strain hardening during the extension from bubble growth can

decrease the susceptibility of the bubble walls to rupture. Consequently, bubble deterioration

mechanisms such as bubble coalescence and collapse are reduced in PP/fibrillated-PET. A two

orders of magnitude increase in bubble density for PP/fibrillated-PET relative to neat PP is

observed and is attributed to the concurrent effects of the presence of high surface-to-volume

ratio fibrillar heterogeneities, transcrystalline layer development around them and marked strain

hardening response in extensional flow.

Fig. 4.6c shows that the foam volume expansion ratio of neat PP ranges from 2 to 11 folds,

PP/spherical-PET ranges from 9 to 14 folds and PP/fibrillated-PETs after fiber spinning at a

draw ratio of 20.4:1 ranges from 7 to 16 folds. The expansion ratio of a foam is a measure of the

degree of gas retention within a foamed polymer. 14

A high expansion ratio indicates that a large

amount of gas is retained within the closed bubbles of the foam and gas escape from the polymer

is inhibited. The noticeable enhancement in expansion ratio for the PP/fibrillated-PET is

attributed to two factors: 1) the strain hardening response exhibited by PP/fibrillated-PET

prevents bubble-wall rupture during bubble growth due to the viscosity increase with strain, and

80

2) flow-induced crystallization of PP on PET fibrils in the form of a transcrystalline layer during

bubble growth. 10, 42, 43

44, 45

Through these two mechanisms, strain-induced by bubble growth

results in an increase in the bubble wall stiffness preventing them from rupturing. Thus, a larger

amount of CO2 is retained within PP/fibrillated-PET and up to a five-fold higher expansion ratio

is seen relative to neat PP. The observed enhancement in foaming ability is consistent with

previous reports on foam extrusion of polymer-fibrillar blends where the in-situ generation of of

polytetrafluoroethylene (PTFE) fibrils in a PP matrix led to similar enhancements in both

expansion ratio and cell densities. 11, 46, 47

4.5 Conclusion

The results demonstrate that enhancements in foaming ability of PP can be achieved from the in-

situ fibrillation of PET in a PP matrix using a fiber spinning process. High draw ratios can cause

the PET domains in a PP/PET blend to elongate into large aspect ratio fibrils. These fibrils form

a percolated fibrillar network in the polymer matrix, altering the viscoelastic, crystallization and

mechanical properties of the matrix. The process is easy-to-scale-up, since it utilizes low-cost

commodities materials and conventional equipment used by the textile and polymer foam

manufacturing industries. The strategy, which in principle is highly versatile and applicable to a

wide range of polymer blends, is technologically promising for improving the foaming ability of

difficult-to-foam-plastics.

81

References

1. Fakirov, S.; Bhattacharyya, D.; Shields, R. J. Nanofibril reinforced composites from polymer

blends. Colloids Surf., A., 2008, 2-8, 313-314.

2. Fakirov, S.; Duhovic, M.; Maitrot, P.; Bhattacharyya, D. From PET nanofibrils to

nanofibrillar single-polymer composites. Macromol. Mater. Eng. 2010, 295, 515-518.



4. Paradkar, R. P.; Sakhalkar, S. S.; He, X.; Ellison, M. S. On-line estimation of molecular

orientation in polypropylene fibers using polarized Raman spectroscopy. Appl. Spectrosc.

2001, 55, 534-539.





7. Rodney, D.; Fivel, M.; Dendievel, R. Discrete modeling of the mechanics of entangled

materials. Phys. Rev. Lett. 2005, 95, 108004.

8. Friedrich, K.; Evstatiev, M.; Fakirov, S.; Evstatiev, O.; Ishii, M.; Harrass, M. Microfibrillar

reinforced composites from PET/PP blends: processing, morphology and mechanical

properties. Comp. Sci. Tech. 2005, 65, 107-116.

9. Rizvi, A.; Park, C. B. Dispersed polypropylene fibrils improve the foaming ability of a

polyethylene matrix. Polymer 2014, 55, 4199-4205.

10. Rizvi, A.; Park, C. B. Preparation and characterization of microstructured blends from

polyethylene terephthalate and poly(propylene-co-ethylene) random copolymer. 2014,

Submitted.

82



2013, 54, 4645–4652.

12. Spitael, P.; Macosko, C. W. Strain hardening in polypropylenes and its role in extrusion

foaming. Polym. Eng. Sci. 2004, 44, 2090-2100.

13. Zhai, W.; Kuboki, T.; Wang, L.; Park, C. B. Cell structure evolution and the crystallization

behaviour of polypropylene/clay nanocomposites foams blown in continuous extrusion. Ind.

Eng. Chem. Res. 2010, 49, 9834.

14. Naguib, H. E.; Park, C. B.; Panzer, U.; Reichelt, N. Strategies for achieving ultra low-density

PP foams. Polymer Engineering and Science, 2002, 42, 1481- 1492.



16. Jayanarayanan, K.; Jose, T.; Thomas, S.; Joseph, K. Effect of draw ratio on the

microstructure, thermal, tensile and dynamic rheological properties of in-situ microfibrillar

composites. Eur. Polym. J. 2009, 45, 1738-1747.

17. Fakirov, S.; Bhattacharyya, D.; Lin, R. J. T.; Fuchs, C.; Friedrich, K. Contribution of

coalescence to microfibril formation in polymer blends during cold drawing. J. Macromol.

Sci. Phy. 2007, 46, 183-194.

18. Hong, J. S.; Ahn, K. H.; Lee, S. J. Strain hardening behaviour of polymer blends with fibril

morphology. Rheologica Acta 2005, 45, 202-208.



779-793.


83









23. Han, C. D.; Baek, D. M.; Kim, J. K.; Ogawa, T.; Sakamoto, N.; Hashimoto, T. Effect of

volume fraction on the order-disorder transition in low molecular weight polystyrene-block-

polyisoprene copolymers. 1. Order-disorder transition temperature determined by rheological

measurements. Macromolecules 1995, 28, 5043-5062.

24. Potschke, P.; Fornes, P.; Paul, D. R. Rheological behaviour of multiwalled carbon

nanotube/polycarbonate composites. Polymer 2002, 43, 3247–3255.


1997, 134, 165–234.

26. Kim, H. S.; Karger-Kocsis, J. Tearing resistance of some co-polyester sheets. Acta

Materialia 2004, 52, 3123-3233.

27. Miller, R. L. On the existence of near-range order in isotactic polypropylenes. Polymer 1960,

1, 135-143.

28. Meille, S. V.; Brückner, S. Non-parallel chains in crystalline γ-isotactic polypropylene.

Nature 1989, 340, 455-457.

29. Dean, D. M.; Register, R. A. Oriented γ-isotactic polypropylene crystallized at atmospheric

pressure. J. Polym. Sci. Part B Polym. Phys. 1998, 36, 2821–2827.

84

30. Dean, D. M.; Register, R. A. γ-Transcrystallization in isotactic polypropylene-based

composites promoted by aramid fibers. J. Polym. Sci. Part B: Polym. Phys. 1999, 37, 2534-

2538.

31. Wang, Y.; Pan, J. L.; Mao, Y.; Li, Z. M.; Li, L.; Hsiao, B. S. Spatial distribution of γ-crystals

in metallocene-made isotactic polypropylene crystallized under combined thermal and flow

fields. J. Phys. Chem. B. 2010, 114, 6806-6816.

32. Hong, S. M.; Seo, Y. Crystallization of a polypropylene terpolymer made by a Ziegler-Natta

catalyst: formation of gamma-phase. J Phys Chem B. 2007, 111, 3571-3575.

33. Lezak, E.; Bartczak, Z.; Galeski, A. Plastic deformation of the γ phase in isotactic

polypropylene in plane-strain compression. Macromolecules 2006, 39, 4811-4819.

34. Naguib, H. E.; Park, C. B.; Song, S.-W. Effect of Supercritical Gas on Crystallization of

Linear and Branched Propylene Resins with Foaming Additives. Ind. Eng. Chem. Res. 2005,

44, 6685-6691.

35. Leung, S. N.; Park, C. B.; Li, H. Effects of Nucleating Agents‘ Shapes and Interfacial

Properties on Cell Nucleation. J. Cell. Plast. 2010, 46, 441- 460.

36. Nofar, M.; Guo, Y.; Park, C. B. Double-Crystal Melting Peak Generation for EPP Bead

Foam Manufacturing. Ind. Eng. Chem. Res. 2013, 52, 2297-2303.

37. Park, C. B.; Cheung, L. K. A Study of Cell Nucleation in the Extrusion of Polypropylene

Foams. Polym. Eng. Sci. 1997, 37, 1-10.

38. Kuboki, T. Foaming behaviour of cellulose fiber-reinforced polypropylene composites in

extrusion. J. Cell. Plast. 2014, 50, 113-128.



40. Somani, R. H.; Yang, L.; Hsiao, B. S. Shear-induced molecular orientation and

85

crystallization in isotactic polypropylene: effects of the deformation rate and strain.


41. Wang, K.; Guo, M.; Zhao, D.; Zhang, Q.; Du, R.; Fu, Q.; Dong, X.; Han, C. C. Facilitating

transcrystallization of polypropylene/glass fiber composites by imposed shear during

injection molding. Polymer 2006, 47, 8374-8379.

42. Keshtkar, M.; Nofar, M.; Park, C. B.; Carreau, P. Extruded PLA/Clay Nanocomposite Foams

Blown with Supercritical CO2. Polymer 2014, 55, 4077-4090.



180-188.

44. Mihai, M.; Huneault, M. A.; Favis, B. D. Crystallinity development in cellular poly(lactic

acid) in the presence of supercritical carbon dioxide. J. Appl. Polym. Sci. 2009, 113, 2920-

2932.

45. Chen, J.; Sun, X.; Turng, L. S.; Zhao, L.; Liu, T.; Yuan, W. K. Investigation of crystallization

behaviour of solid and microcellular injection molded polypropylene/nano-calcium carbonate

composites using carbon dioxide as a blowing agent. J. Cell. Plast. 2013, 49, 459-475.


same. WO2013137301, March 13, 2013.


JP2013514463, March 13, 2012.

86

Chapter 5

In-situ fibrillation of CO2-philic polymers: sustainable route to polymer foams in a continuous process 4

5.1 Abstract

In this study, in-situ polymer-fibrillar blends of polypropylene (PP)/polytetrafluoroethylene

(PTFE) are prepared. Dynamic oscillatory shear experiments confirm that the PTFE elongates

into fibrils during blending and forms a physical network of entanglements in the melt which

results in gel-like properties. Uniaxial extensional flow experiments show strain-induced

hardening behaviour. CO2 solvency in the PP/PTFE fibrillar blend is enhanced due to the CO2-

philic character of PTFE. Remarkably, adding only 0.3 wt% of PTFE is sufficient to markedly

enhance the CO2 sorption capacity of the matrix. Continuous foam extrusion of the in-situ

fibrillar blend reveals a three orders of magnitude increase in bubble density, a ten-fold increase

in volume expansion ratio, and a marked broadening of the foaming window with respect to neat

PP. These improvements are attributed to the simultaneous enhancement in CO2 solvency and

strain hardening behaviour of the melt in the in-situ fibrillar blend.

5.2 Introduction

Commercially relevant polymeric materials, on their own, often do not have all the properties

prerequisite for a specific application. Secondary components are required for imparting

desirable or amending properties to the polymers. For instance, linear polypropylene (PP) is a

semi-crystalline polymer that has excellent physical and chemical properties as well as low

material cost in comparison with other thermoplastics, but lacks the melt strength and strain

hardening required for many industrial processes, such as blow molding and polymer foaming. 1,

2, 3

4 Reproduced from: Rizvi, A.; Tabatabaei, A.; Barzegari, R.; Mahmood, H.; Park, C.B., ―In situ fibrillation of CO2-

philic polymers: sustainable route to polymer foams in a continuous process‖, Polymer, 2013, 54, 4645–4652.

87

Although a high degree of chemical crosslinking of PP is an effective strategy for increasing its

melt strength and strain hardening, the approach is undesirable as it renders the polymer difficult

to process and recycle. 4, 5

Methods based on introducing long-chain branching have been

successful, however, the cost of commercially available resins of PP with long chain branching

are at least twice as expensive as linear PP. 6, 7, 8, 9

Exfoliation of organically modified layered

nanosilicates in polymers has been effective in improving the melt viscosity and elasticity, 10, 11,

12, 13, 14 however, exposure to such nanoparticles poses significant health hazard. Furthermore,

exfoliation of nanosilicates in polymers is challenging and requires fine tuning of the

nanoparticle‘s surface chemistry as well as its synthesis and processing conditions. 15, 16, 17

Despite the availability of a number of methods to compensate for the poor melt strength and

weak strain-induced hardening of linear PP, it still remains a challenge improve its rheological

properties for various polymer processing applications.

With respect to foam processing, a further hurdle to the broader use of PP is its poor solubility

for carbon dioxide (CO2). While CO2 has emerged as an environmentally-sustainable foam

blowing agent, it is a poor solvent for most high molecular weight polymers, although there are

some exceptions such as amorphous fluoropolymers and silicones. 18

Consequently, energy-

intensive processing conditions such as elevated pressures are required to enhance CO2 solvency

in polymer melts, as well as expensive equipment capable of withstanding such working

pressures are needed for preparing foams of PP in a continuous process. There is critical need of

to establish methods that enhance the solubility of CO2 in a polymer melt without raising

pressure significantly.

The aim of the present work is to simultaneously address the aforementioned challenges by

preparing in-situ blends of linear PP and polytetrafluoroethylene (PTFE). The tendency of PTFE

to deform into fibrillar structures of large aspect ratios during blending and develop topological

interactions (physical entanglements) is expected to reinforce the PP matrix and improve

rheological response. Linear dynamic frequency tests are conducted to characterize the

microstructure of PP/PTFE fibrillar blends. Non-linear uniaxial extension flow experiments are

also conducted to characterize its transient extensional viscosity, as well as its strain hardening

behaviour. The presence of fluoroalkyl functional groups in PTFE demonstrate strong

thermodynamic affinity for CO2 19

and are expected to promote CO2 dissolution in the matrix.

Thus, a magnetic suspension balance (MSB) is used in combination with a pressure-volume-

88

temperature (PVT) apparatus to characterize the CO2 solubility of the fibrillar blend. Foaming

ability of PP reinforced with the fibrillated PTFE is evaluated using a conventional tandem foam

extrusion system. For reasons of comparison, all aforementioned tests are also performed for

neat PP, without the presence of the fibrillated PTFE.

5.3 Experimental

5.3.1 Materials

A commercially available linear isotactic polypropylene (PP) homopolymer supplied by Japan

Polypropylene, Novatec-PP FY4, with melt flow rate (MFR) = 5 g/10 min (at 230°C/2.16 kg

load) was used as the matrix in this study. The melting temperature of PP is found using DSC to

be about 165°C. A commercially available polytetrafluoroethylene (PTFE) powder supplied by

Mitsubishi Rayon, MetablenTM

A-3000, was also used. All materials were used as received. CO2

which was used as the foam blowing agent was purchased from Linde Gas with purities in excess

of 99%.

5.3.2 Sample preparation

Dry blends of PP/PTFE were prepared in proportions of 100/0 (i.e. pure PP was processed using

exactly the same method as other samples for comparison), 99.7/0.3 and 97/3 in weight fraction.

The resulting mixtures were fed in the main feed opening of a co-rotational twin-screw extruder

manufactured by Toshiba Machine Co. Ltd. (Trade name: TEM-26SS). The screw diameter was

26 mm and its L/D ratio was 40. In the extruder, the fed mixture was melt-blended under a barrel

and die temperature of 200°C but the hopper zone was cooled with a cold-water sleeve. The

discharge rate was set at 20 kg/hr and the screw rotation speed was maintained at 200 rpm.

Extrudate from the die was shaped into a cylindrical strand, led into a water bath, and pelletized.

5.3.3 Measurements

To observe the morphology development of PTFE after blending with PP, the blend was

dissolved in xylene and the residual PTFE examined using a scanning electron microscope

(JEOL 6060). To characterize the dispersion of the PTFE fibrils in the PP matrix, the surface of

cryogenically fractured sample of the blend was subjected to a solvent vapour etching process

89

using xylene. Samples were exposed to xylene vapours for around 2 h at 65°C. This type of

treatment preferentially etches the low-melting-phase PP than the high-melting PTFE, which

leads to bringing up these domains and increasing differentiation in observation. The etched

samples were coated with a thin layer of platinum using a sputter coater prior to observation.

The frequency dependency of oscillatory shear storage modulus ) and loss modulus

were evaluated in the linear viscoelastic region at 190°C by a strain-controlled cone and plate

ARES rheometer from TA Instruments. The diameter of the cone is 25 mm and the cone angle is

5°.

Uniaxial extensional viscosity was measured using the ARES rheometer at a temperature of

170°C and at strain rates of 0.01, 0.1 and 1 s-1

. The Trouton prediction 28

of extensional viscosity

was obtained from steady shear start-up experiments at a temperature of 170˚C, and strain rate of

0.01 s-1

.

The mass of CO2 sorption by PP/PTFE fibrillar blend (99.7/0.3 wt%) and neat PP was

experimentally determined using an MSB and PVT apparatus. The complete method for the

determination of CO2 sorption is described elsewhere. 20

5.3.4 Continuous Foam Processing

A tandem continuous foam extrusion system commonly employed by the foam manufacturing

industry was used to foam the samples. The tandem foam system comprises of two single-screw

extruders. The first extruder consists of a 5-hp extruder drive; a 3/4‖ extruder (Brabender, 05-25-

000) that has a mixing screw (Brabender, 05-00-144) with an L/D ratio of 30/1. The second

extruder measures 3/2‖ (Killion, KN-150) with a built-in 15-hp variable speed unit with an L/D

ratio of 18/1. Fig. 5.1 gives a schematic of the configuration of the extrusion system. 10 wt% of

CO2 gas was injected into the melt through an injection point positioned at the first extruder.

Foaming of the polymer melt occurred at the die exit where the polymer/gas solution was

subjected to rapid depressurization resulting in the gas and the melt to undergo phase separation.

A brass capillary die was used comprising of a circular pinhole with a diameter of 1.2 mm and a

channel length of 10 mm. Temperature of the first extruder was kept constant but the temperature

of the second extruder was varied as shown in Table 5.1. A temperature gradient was applied

along the melt flow direction to ensure that the melt cools uniformly as it flows through the

90

second extruder. To characterize the volume expansion of the foams, the mass densities of

samples before ( ) and after ( ) foaming were determined using the water displacement method

according to ASTM-D792. It is assumed that the absorption of water by the sample is negligible

during the measurement due to a smooth unfoamed skin of all foamed samples. Subsequently,

the volume expansion , was determined using the following expression:

(5.1)

The bubble (-population) density was calculated as the number of bubbles per unit volume

with respect to the unfoamed sample. First the number of bubbles in a defined area was

determined and then the total number of bubbles per cubic centimetre was calculated using:

(

)

(5.2)

To calculate the bubble sizes, the diameter of the individual bubbles was measured

directly from SEM images of the cryo-fractured cross-section of the foamed samples using an

image processing software. For ellipsoidal bubbles, the length of the major axis of the bubble

was measured. The average length of two hundred bubbles was calculated along with the

standard deviation.

Fig. 5.1 Schematic illustration of the continuous tandem foam extrusion system used in this

study.

91

Table 5.1 Temperature profiles of first and second extruder for collecting foam samples. The

temperatures in the first extruder were kept constant but the temperatures in the second extruder

were varied.

First Extruder Second Extruder

T1 (°C) T2 (°C) T3 (°C) T4 (°C) T5 (°C) T6 (°C) T7 (°C) T8 (°C) T9 (°C) Tdie (°C)

160 220 220 220 220 190 170 150 150 150

160 220 220 220 220 190 170 140 140 140

160 220 220 220 220 190 160 140 140 135

160 220 220 220 220 190 160 140 135 130

160 220 220 220 220 190 150 140 135 125

160 220 220 220 220 180 150 140 130 120

160 220 220 220 220 180 150 140 130 115


5.4.1 Morphology of PP/PTFE fibrillar blends

Fig. 5.2 SEM micrograph of the morphology of PTFE nanofibrils: a) obtained after solvent-

vapor etching; b) obtained after removal of PP using xylene. Scale bars correspond to 10 μm in

a) and 2 μm in b).

Fig. 5.2 shows the morphology of fibrillated PTFE in the PP matrix after blending. The

micrograph in Fig. 5.2a is obtained after solvent vapour etching of a fractured surface of the

PP/PTFE blend. The micrograph in Fig. 5.2b is obtained after dissolving PP in xylene and

92

0.01 0.1 1 10 10010

0

101

102

103

104

105

PP G'

PP G"

PP/PTFE G'

PP/PTFE G"

G', G

" (

Pa)

Frequency (rad/s)

G"

G'

Fig. 5.3 Frequency dependence of elastic modulus and viscous modulus obtained

at 190°C comparing PP (hollow symbols) with PP/PTFE fibrillar blends (solid symbols).

observing the PTFE residue. Fig. 5.2 confirms that PTFE undergoes fibrillation and deforms

into fibrillar structures of large aspect ratios (diameter of fibrils are less than 500 nm, and their

lengths seem exceed 100 μm). The fibrillar morphology adopted by PTFE in Fig. 5.2 is a

consequence of its low yield strength to undergo plastic deformation, particularly at elevated

temperatures. At about 19°C, PTFE crystals exhibit a first order transition from triclinic to

hexagonal form which is characterized by weak cohesive forces between neighbouring chains

and a small amount of shear can cause the PTFE molecules packed in the crystal to unwind. 21, 22

The mechanisms of fibrillation of solid PTFE have been investigated earlier, and are described

elsewhere. 23

As long as the blending temperature is not increased to above the melting

temperature of PTFE, the fibrillar morphology can be maintained in the matrix because the PTFE

molecules are prevented from undergoing molecular relaxation and recoiling back into spherical

domains.

93

In order to characterize the microstructure adopted by PTFE in the PP matrix, mechanical

response of the blend is studied in the frequency domain (oscillatory shear). The results are

presented in Fig. 5.3 which shows the storage and loss moduli ( , respectively) of a

PP/PTFE fibrillar blend and PP. Comparison of the linear viscoelastic response of the two

materials shows the significant effect of the PTFE fibrils, especially at low frequencies. The

apparent plateau in at low frequencies for the PP/PTFE fibrillar blend is consistent with the

response of a viscoelastic fluid with a large characteristic relaxation time that is about 102 s,

since the crossover of and in the flow transition regime is observable around ω = 0.01 rad

s-1

. The observed secondary plateau in the storage modulus in the low frequency regime could be

explained by the existence of a microstructure comprising of a percolated, physical network of

fibrils which stores the deformation energy over large timescales. 24

5.4.2 Strain-induced hardening in uniaxial extensional flow

0.1 1 10 10010

3

104

105

106

Strain rate 1 s-1

Strain rate 0.1 s-1

Strain rate 0.01 s-1

+

0(t)

+ E(t

, P

a.s

]

Time (s)

a)

Fig. 5.4 Tensile stress growth curves of uniaxial extensional viscosity at strain rates of

0.01, 0.1 and 1 s-1

at a temperature of 170°C for (a) PP, (b) PP/PTFE. The solid lines represent

the where is the Trouton prediction obtained from steady shear start-up

experiments for each sample at a strain rate of 0.01 s-1

and temperature 170°C.

0.1 1 10 10010

3

104

105

106

Strain rate 1 s-1

Strain rate 0.1 s-1

Strain rate 0.01 s-1

+

0(t)

+ E(t

, P

a.s

]

Time (s)

b)

94

Extensional viscosity measures the resistance against extensional flow. The transient


stress growth coefficient :

(5.3)

where is the tensile stress growth function and is the extensional strain rate. According to

the Trouton relation, for nonlinear strain hardening response, the is equal to 3 times the linear

viscoelastic shear viscosity for a sufficiently small strain rate. 25, 26

Deviation from the

Trouton relation can be used to evaluate the degree of strain hardening through,

(5.4)

When a polymer is foamed, growth of nucleated bubbles exerts a biaxial extension on the

polymer matrix. If the polymer matrix shows weak strain hardening, it is not able to withstand

this extensional force and the bubbles rupture leading to the diffusion of gas out of the polymer

and ultimately, the foam collapses. Consequently, strain hardening plays a critical role in

facilitating gas retention by preventing bubble opening and gas escape. While biaxial extension

is the primary mode of deformation during foam processing, uniaxial extensional measurements

are more readily available and are used here to measure the strain hardening behaviour of the

samples. The experimental tensile stress growth coefficient, is plotted along with the

Trouton prediction of extensional viscosity, , for PP/PTFE fibrillar blends and neat PP in

Fig. 5.4. Extensional viscosity measurements are conducted at 170°C so that PTFE nanofibrils

remain in solid-state and retain their morphology.

For PP, no strain-induced hardening can be observed over the strain rate scale studied (0.01 to 1

s-1

) and the stress growth coefficients superpose for all applied strain rates on the Trouton

prediction. In contrast, the presence of 3 wt% PTFE nanofibrils in the PP matrix results in

pronounced strain hardening. This enhancement in strain-induced hardening is attributed to the

formation of a physical network of the flexible PTFE nanofibrils. When a tensile stress is

applied, flexible fibrillar networks resist deformation due to topological constraints or physical

entanglements. 27

Consequently, the network generates excess stress which is the origin of the

strain hardening behaviour during uniaxial extension.

95

5.4.3 Enhancement of CO2 sorption in the PP/PTFE fibrillar blends

Fig. 5.5 compares the mass of CO2 localized in PP to that localized in PP/PTFE fibrillar blend

(99.7/0.3 wt%) at a CO2 pressure of 17.2 MPa and three different temperatures, 155°C, 180°C

and 210°C. The measurements are made using a Magnetic Suspension Balance (MSB) in

combination with a pressure-volume-temperature (PVT) apparatus. A complete description of

the method is described elsewhere. 28

It must be noted that the PTFE content in the fibrillar blend

used for solubility measurements is 0.3 wt% (as opposed to 3 wt% used for all other

measurements in this study). When the PTFE content in the fibrillar blend was 3 wt%, uneven

swelling upon exposure to elevated temperatures and CO2 pressures made PVT measurements

150 160 170 180 190 200 210

0.08

0.10

0.12

0.14

So

lub

ilit

y (

g o

f C

O2/

g o

f p

oly

me

r)

Temperature (°C)

PP/PTFE

PP

Fig. 5.5 Solubility of CO2 in PP and PP/PTFE fibrils (99.7/0.3 wt%) obtained at a pressure of

17.2 MPa and temperatures of 155°C, 180°C, and 210°C.

96

challenging. By using a more dilute PTFE concentration of 0.3 wt%, this problem of non-

uniform swelling was rectified. At a CO2 pressure of 17.2 MPa, CO2 sorption is consistently

higher for PP/PTFE fibrillated blends relative to neat PP over all studied temperatures. This

enhancement in CO2 solubility is consistent with previous reports that show polymers containing

fluorinated functional groups demonstrate significant CO2 solubility under mild conditions. 18, 19,

29, 30, 31, 32

Certain functional groups, have been found experimentally, to exhibit miscibility with CO2 at

relatively moderate pressures due to favorable specific solute-solvent interactions between the

polymer and CO2. Effective CO2-philes include fluroalkyl, fluoroether, and siloxane functional

materials. 33

It has been suggested that CO2 may either form a weak complex or that it

preferentially clusters near the fluorine of the C−F bonds that are more polar than C−H bonds. 33

Experimental determination of the sorption capacity of neat PTFE is challenging due to its

peculiar swelling behaviour upon heating under pressurized CO2 which makes PVT

measurements difficult. Therefore, the rule of mixtures which defines the sorption of CO2 in the

melt at equilibrium was employed to calculate the solubility of CO2 in neat PTFE (see Appendix

A). It was found that at a temperature of 155°C and a CO2 pressure of 17.2 MPa, the solubility of

CO2 in neat PTFE is 6.2 gCO2/g polymer. This is in clear contrast to the sorption capacity of PP

experimentally determined to be 0.1148 gCO2/g polymer under the same conditions of

temperature and CO2 pressure. From this result, it can be realized that the addition of PTFE to

polymeric matrices provides a convenient opportunity to systematically tune the amount of CO2

localized within a melt by simply increasing the PTFE content, without the need of increasing

the processing pressures. While the presence of the CO2-philic fluoroalkyl groups in PTFE is

expected to enhance CO2 sorption in the PP/PTFE fibrillar blend, the large matrix-fibrillar

interfacial area of contact due to the nano-scale dimensions of the PTFE fibrils may also have a

role in facilitating adsorption of CO2. Although the present calculation does not decouple the

adsorption of CO2 at the interface between the PTFE fibrils and the PP melt from the uptake of

CO2 by the PTFE fibrils, this consideration certainly merits future investigation.

97

5.4.4 Continuous foam processing of PP and PP/PTFE fibrillar blends

Fig. 5.6 Electron micrographs of cryo-fractured surfaces of extruded foams in a continuous

process. (a) PP; (b) PP/PTFE fibrillar blends. The temperatures shown in the micrographs

correspond to the die temperatures. Scale bars correspond to 100 μm.

98

Continuous foam processing of neat PP and PP/PTFE fibrillar blend is conducted to understand

the effect introducing PTFE fibrils has on the foaming behaviour of PP. It must be noted that the

PTFE fibrils remain in solid-state during the foam extrusion so that the PTFE retain their fibrillar

morphology during the process. If the PTFE undergoes melting, the morphology of the PTFE

will not be preserved and the fibrils will relax and recoil back into spherical domains, altering the

rheological properties of the blend. Fig. 5.6a shows foams of neat PP obtained at various

temperatures. Fig. 5.6b shows foams of PP/PTFE fibrillar blends obtained at various

temperatures. Substantial improvements in bubble morphology with the presence of the in-situ

fibrillated PTFE are realized. Fig. 5.7a shows the bubble densities as a function of temperature.

The bubble densities for neat PP foams measured on the order of 105 – 10

6 bubbles/cm

3, whereas

PP/PTFE foams measured on the order of 107 – 10

9 bubbles/cm

3 indicating up to a three orders

of magnitude increase in bubble densities. The classical nucleation theory 34

describes the

nucleation efficiency of bubbles, where gas bubbles larger than a critical radius continue to

grow spontaneously while those smaller tend to collapse. Assuming that the polymer/gas

solution that forms prior to bubble nucleation is a dilute polymer/gas solution, the expression for

the free energy needed for bubble nucleation and are given by: 13, 14, 34, 35

( ) (5.5)

(5.6)

where is the surface tension at the polymer/gas interface, is a geometric factor that

equals to the ratio of the volume of a heterogeneously nucleated bubble to that of a spherical

bubble having the same radius of curvature, is the Henry‘s Law constant, is the gas

concentration, is the system pressure and is the local pressure variations. It follows

from Eqs. 5.5 and 5.6, that an increase in the gas concentration, , is expected to reduce the

free-energy barrier for bubble nucleation and the critical radius. In such a case, a larger bubble

density will be obtained. Therefore, it is to no surprise that the CO2-philicity of PTFE fibrils

shown in Fig. 5.5 should increase the number of bubble nucleation sites in the PP matrix. In

contrast, the absence of the CO2-philic PTFE fibrils decreases the number of CO2 molecules

available for nucleation and fewer new stable nuclei develop. Bubble density improvements in

commodity plastics through increasing the gas concentration, , has been accomplished in

99

Fig. 5.7 Foam characterization: a) Representation of the bubble density dependence on the

foaming temperature of PP and PP/PTFE in a continuous foam extrusion process. The dotted

lines confine the foaming temperature window. b) Comparison of the foam expansion ratios of

PP and PP/PTFE fibrillar blends as a function foaming temperature c) Comparison of the bubble

diameters of PP and PP/PTFE fibrillar blends as a function of temperature.

previous studies by raising the CO2 pressure. At a given temperature and depressurization rate,

increasing the CO2 pressure localizes a larger amount of CO2 in the melt. Thus, a higher degree

of supersaturation is reached during the depressurization stage leading to the formation of more

nuclei. 36

To the best of our knowledge, this is the first study which demonstrates that adding

CO2-philic entities such as PTFE fibrils can also be effective in increasing gas concentration for

reaching a higher degree of supersaturation in a continuous foam process.

Bubble nucleation is influenced by microscopic pressure variations which affect the magnitude

of supersaturation. 37

The local entanglements of PTFE fibrils create topological constraints that

prevent fibril flow along with the melt. 27, 38

Consequently, the network of PTFE fibrils

undergoes a macroscopic strain which is the origin of a complex and locally varying

superposition of shear, compressive and tensile stresses. In regions where a tensile stress is

applied to the flowing matrix, a negative local pressure occurs, i.e. . This reduces the

activation energy and the critical radius for bubble nucleation (Eqs. 5.5 and 5.6). Consequently,

an enhancement in bubble nucleation efficiency for PP/PTFE fibrillar blends is realized.

Presence of a secondary phase, such as the solid-state PTFE fibrils, is able to reduce the

heterogeneous nucleation energy barrier ( ) by increasing the number of bubble nucleation

120 130 140 1500

50

100

150

200

250

300

Bu

bb

le d

iam

ete

r (µ

m)

Temperature (°C)

c)

PP

PP/PTFE

120 130 140 150

10

20

30

PP

Ex

pa

ns

ion

ra

tio

Temperature (°C)

PP/PTFE

b)

120 130 140 15010

4

105

106

107

108

109

PP

Bu

bb

le d

en

sit

y (

bu

bb

les

/cm

3)

Temperature (°C)

Foaming window

Foaming window

PP/PTFE

a)

100

sites. In PP/PTFE, the PTFE is present mostly as fibrils with diameters on the nano-scale

resulting in a large effective concentration which in turn increases the surface-area of contact

between the fibrils, the polymer melt and CO2 for bubble nucleation. This also helps in

increasing the foam bubble densities in PP/PTFE fibrillar blends relative to neat PP.

The expansion ratio is a measure of the degree of gas retention within a foamed polymer. A high

expansion ratio indicates that gas escape out of the polymer is substantially inhibited. Fig. 5.7b

shows the foam expansion ratios of neat PP and PP/PTFE fibrillar blends as a function of

foaming temperature. The expansion ratios for PP foams at various die temperatures range from

10 to 20 folds whereas PP/PTFE foams range from 10 to 30 folds. The marked increase in

expansion ratio of PP/PTFE fibrillar blends relative to PP shown in Fig. 5.7b is attributed to

mainly two factors. First, the PTFE fibrils have a high affinity for CO2 and act as reservoirs

making available a larger amount of the blowing agent in the melt for expansion. Secondly, the

PTFE fibrils increase strain-induced hardening and reinforce the walls of the bubbles suppressing

bubble-wall opening, thereby minimizing bubble wall rupture and escape of the CO2 molecules

out of the melt, although it must be noted that the degree of strain hardening in the presence of

pressurized CO2 may differ from that observed under atmospheric pressure and depicted in Fig.

5.4b. A classic mountain shaped expansion ratio vs. die temperature curve is observed for both

PP and PP/PTFE fibrillar blends indicating the competition between active gas loss at elevated

temperatures and excessively stiff matrix at low temperatures which reduces foam expansion. 39

Fig. 5.7b reveals that the optimum volume expansion shifts to a lower temperature in PP/PTFE

compared with neat PP. One plausible explanation for this shift stems from the higher bubble

density of PP/PTFE compared to neat PP (Fig. 5.7a). In the case of higher bubble densities, for a

given rate of diffusion, the individual growth rate of the bubbles would be the same, but the

overall foam expansion rate will increase. 40

Consequently, in PP/PTFE fibrillar blends, the

overall volume expansion occurs faster than neat PP. In order to preserve the foam structure

when optimum expansion occurs, polymer vitrification needs to take place at a faster rate. This

could be achieved by decreasing the melt temperature. By decreasing the melt temperature,

polymer vitrification can be initiated sooner allowing the maximally expanded foam to stabilize.

The average apparent diameters of bubbles estimated from SEM images are plotted in Fig. 5.7c.

Over the entire temperature range studied, the average bubble diameters of PP/PTFE fibrillar

101

blends are significantly lower than that of neat PP. The higher solubility of CO2 in PP/PTFE

fibrillar blends relative to neat PP influences the bubble sizes as it allows a larger number of

bubbles to nucleate. This effect, coupled with the suppression of cell coalescence mechanisms,

where large bubbles result from the rupturing of walls separating neighbouring bubbles, due to

strain-induced hardening from PTFE fibrils leads to smaller sized bubbles. In the absence of the

strain-induced hardening, as in the case of neat PP, such bubble deterioration mechanisms are

active as evident from the broad bubble size distribution observable in the foam morphology

depicted in Fig. 5.6 and the error bars in Fig. 5.7c. It must be noted however, that although the

average bubble size decreases with the addition of PTFE, the overall foam volume expansion

still increases due to the availability of a larger amount of CO2.

It is generally accepted that decreasing the bubble sizes improves the mechanical performance of

foams. In particular microcellular foams (with diameters less than 30 μm and bubble densities

exceeding 109 bubbles/cm

3) have shown to provide improved mechanical properties over

conventional foams. 41

On the other hand, increasing the foam expansion ratio decreases the

foam density and allows for low density foams to be obtained. A concurrent increase in bubble

density and expansion ratio upon incorporation of a filler, such as the PTFE fibrils, is rare. 17, 42

Often times, the inclusion of micro/nanoparticle fillers into polymers lead to foams of larger

bubble densities. However, the bubble growth and foam expansion are limited due to the

increased stiffness of the polymer composite. Consequently an increase in bubble density is often

accompanied with a decrease in volume expansion. Previously, ultrasonic irradiation has been

effective in accomplishing an increase in both bubble densities and expansion ratios in a batch

foaming process. 42, 43

However, extending this approach to a continuous foam extrusion process

requires intensive capital investment and major modification of conventional extrusion systems.

We demonstrate that the introduction of PTFE fibrils in the continuous foam processing of

polymers is a novel and scalable approach which not only increases the bubble density and

decrease the bubble size as most fillers, but also effectively increase the expansion ratio of the

foam by making a larger amount of the foam blowing agent available for foam expansion.

102

5.4.5 Minimum admissible temperature for PP and PP/PTFE fibrillar bends

0 5 10 15 20

110

115

120

125

130

PP

Min

imu

m a

dm

iss

ab

le t

em

pe

ratu

re (

°C)

Injected amount of CO2 (%)

PP/PTFE

Fig. 5.8 Comparison of the minimum temperature at which foams could be collected for PP and

PP/PTFE fibrillar blends as a function of CO2 content.

Fig. 5.8 shows the influence of the amount of injected CO2 on the minimum temperature at

which foams could be collected for PP and for PP/PTFE fibrillar blends. The ‗minimum

admissible temperature‘ is defined as the temperature below which the viscosity of the melt is

too high for flow to occur in the extruder. If a larger amount of CO2 dissolves in the melt, CO2-

induced plasticization causes the minimum admissible temperature to decrease. For the case of

neat PP, increasing the CO2 content did not affect the minimum admissible temperature and it

remained constant at 130°C over the injected CO2 range studied (3 to 15%). This result indicates

that while increasingly higher amounts of CO2 were successfully injected in the extrusion barrel,

the injected CO2 did not dissolve in the melt to form a homogeneous polymer/gas solution.

103

Instead, the CO2 continued to exist as a second phase. Conversely, incorporating PTFE fibrils

allows the minimum admissible temperature to be decreased as the injected amount of CO2 is

increased. For the case of 20% CO2 injection, foams could be collected at temperatures as low as

110°C. The marked decrease in the minimum admissible temperature for PP/PTFE fibrillar

blends with respect to PP is attributed to the lubrication effect of PTFE and the plasticization

effect of CO2 which suppresses the flow resistance. The affinity of the PTFE fibrils for CO2

localizes increasingly larger amounts of CO2 in the melt as the injected amount of CO2 is

increased, reducing the minimum admissible temperature and allowing foam collection at lower

temperatures without any sign of plateau. This offers the three-fold advantage of broadening the

foaming window of PP by allowing foam collection at lower temperatures; permitting processing

at lower temperatures which leads to a further increase in CO2 solubility; and reducing energy

consumption.

5.5 Conclusions

In conclusion, we demonstrate that in-situ fibrillation of CO2-philic polymers in a melt can be

effective in simultaneously addressing two of the most pressing challenges experienced during

foam processing of PP, namely, increasing CO2 dissolution without significant increases in

processing pressure, and improving the strain hardening behaviour of a melt. The CO2-philicity

of PTFE provides a convenient route to systematically increase the CO2 content in a melt simply

by increasing the PTFE content without the need of increasing processing pressures. In-situ

fibrillation of the PTFE allows the formation of an interdigitated network of fibrils that provide

strain-induced hardening, not observed in the neat PP. This strain induced hardening suppresses

bubble wall rupture during bubble growth and CO2 gas retention is improved within in the closed

bubbles of the foam. Foams manufactured from the in-situ PP/PTFE blend reveal up to a three

orders of magnitude increase in bubble density and a ten-fold increase in expansion ratio. The

lubricating action of PTFE itself and its affinity for CO2 broadens the foaming window by

plasticizing the melt and allowing foam collection at lower temperatures.

104

References

1. Vasile, C.; Seymour, R. B. Handbook of Polyolefins; Marcel Dekker: New York, 1993.

2. Naguib, H. E.; Park, C. B.; Panzer, U.; Reichelt, N. Strategies for achieving ultra low-density

polypropylene foams. Polym. Eng. Sci. 2002, 42, 1481-1492.

3. Park, J. J.; Katz, L.; Gaylord, N. G. Polypropylene foam sheet. US Patent 5,116,881, 1992.

4. Zhai, W. T.; Wang, H. Y.; Yu, J.; Dong, J. Y.; He, J. S. Cell coalescence suppressed by

crosslinking structure in polypropylene microcellular foaming. Polym. Eng. Sci. 2008, 48,

1312.

5. Romania, F.; Corrierib, R.; Bragab, V.; Ciardelli, F. Monitoring the chemical crosslinking of

propylene polymers through rheology. Polymer 222, 43, 1115-1131.

6. Zhai, W. T.; Wang, H. Y.; Yu, J.; Dong, J. Y.; He, J. S. Foaming behaviour of isotacic

polypropylene in supercritical CO2 influenced by phase morphology via chain grafting.

Polymer 2008, 49, 3146.

7. Nam, G. J.; Yoo, J. H.; Lee, J. W. Effect of long-chain branches of polypropylene on

rheological properties and foam-extrusion performances. J. Appl. Polym. Sci. 2005, 96, 1793-

1800.

8. Gotsis, A. D.; Zeevenhoven, B. L. F.; Tsenoglou, C. Effect of long branches on the rheology

of polypropylene. J. Rheol. 2004, 48, 895-915.

9. Tian, J.; Yu, W.; Zhou, C. The preparation and rheology characterization of long chain

branching polypropylene. Polymer 2006, 23, 7962–7969.

10. Okamoto, M.; Nam, P. H.; Maiti, P.; Kotaka, T.; Nakayama, T.; Takada, M.; Ohshima, M.;

Usuki, A.; Hasegawa, N.; Okamoto, H. Biaxial flow-induced alignment of silicate layers in

polypropylene/clay nanocomposite foam. Nano Lett. 2001, 1, 503-505.

105

11. Lee, L. J.; Zeng, C.; Cao, X.; Han, X.; Shen, J.; Xu, G. Polymer nanocomposite foams.

Comp. Sci. Tech. 2005, 65, 2344-2363.

12. Zheng, W.; Lee, Y. H.; Park, C. B. Use of nanoparticles for improving the foaming

behaviours of linear PP. J. App. Polym. Sci. 2010, 117, 2972-2979.

13. Leung, S. N.; Wong, A.; Park, C. B.; Zong, J. Ideal surface geometry of nucleating agents to

enhance cell nucleation in polymer foaming. J. App.Polym. Sci. 2008., 108, 3997-4003.



2009, 64, 4899-4907.

15. Zhai, W.; Kuboki, T.; Wang, L.; Park, C. B. Cell Structure Evolution and the crystallization

behaviour of polypropylene/clay nanocomposites Foams Blown in Continuous Extrusion.

Ind. Eng. Chem. Res. 2010, 49, 9834.

16. Giannelis, E. P. Polymer layered silicate nanocomposites. Adv. Mater. 1996, 8, 29-35.

17. Zeng, C.; Han, X.; Lee, L. J.; Koelling, K. W.; Tomasko, D. L. Polymer-clay nanocomposite

foams prepared using carbon dioxide. Adv. Mater. 2003, 15, 1743-2747.

18. Rindfleisch, F.; DiNoia, T. P.; McHugh, M. A. Solubility of polymers and copolymers in

supercritical CO2. J. Polym. Chem. 1996, 100, 15581.

19. Huang, Z.; Shi, C.; Enick, R.; Beckman, E. Generation of microcellular materials via self-

assembly in carbon dioxide. Chem. Mater. 2002, 14, 4273–4280.

20. Li, G.; Wang, J.; Park, C. B.; Simha, R. Measurement of gas solubility in linear/branched PP

melts. J. Polym. Sci. Part B- Polym. Phys. 2007, 45, 2497-2508.

21. Clark, E. S.; Muus, L. T. Partial disordering and crystal transitions in

polytetrafluoroethylene. Zeitschrift für Kristallographie - Crystalline Materials 1962, 117,

119- 127.

106

22. Browna, E. N.; Dattelbaumb, D. M. The role of crystalline phase on fracture and

microstructure evolution of polytetrafluoroethylene (PTFE). Polymer 2005, 46, 3056-3068.

23. Lee, P. C.; Wang, J.; Park, C. B. Extruded open-cell foams using two semicrystalline

polymers with different crystallization temperatures. Ind. Eng. Chem. Res. 2006, 45, 175-

181.



8, 8539-8546.


Roy. Soc. 1906, 77, 426-440.

26. Munstedt, H. Rheological properties and molecular structure of polymer melts. Soft Matter

2010, 7, 2273-2283.



28. Li, Y. G.; Park, C. B.; Li, H. B.; Wang, J. Measurement of the PVT property of PP/CO2

solution. Fluid Phase Equilib. 2008, 270, 15.

29. DeSimone, J. M.; Guan, Z.; Elsbernd, C. S. Synthesis of fluoropolymers in supercritical

carbon dioxide. Science 1992, 257, 945.

30. Tuminello, W. H.; Dee, G. T.; McHugh, M. A. Dissolving perfluoropolymers in supercritical

carbon dioxide. Macromolecules 1995, 28, 1506.

31. Wang, Q.; Cao, F.; Chen, Q. Formation of carbon micro-sphere chains by defluorination of

PTFE in a magnesium and supercritical carbon dioxide system. Green Chem. 2005, 7, 733-

736.

32. Raeissia, S.; Peters, C. J. A potential ionic liquid for CO2-separating gas membranes:

107

selection and gas solubility studies. Green Chem. 2008, 11, 185-192.

33. Kirby, C. F.; McHugh, M. A. Phase behaviour of polymers in supercritical fluid solvents.

Chem. Rev. 1999, 99, 565-602.

34. Gibbs, J. W. In The scientific papers of J. Willard Gibbs, Vol. I. Thermodynamics; Ox Bow

Press: Woodbridge, Conecticut, 1993; pp 252-258.

35. Ward, C. A.; Tucker, A. S. Thermodynamic theory of diffusion-controlled bubble growth or

dissolution and experimental examination of the predictions. J. Appl. Phys. 1975, 46, 233-

238.

36. Zhong, X.; Dehghani, F. Fabrication of biomimetic poly(propylene carbonate) scaffolds by

using carbon dioxide as a solvent, monomer and foaming agent. Green Chem. 2012, 14,

2523-2533.

37. Wong, A.; Park, C. B. The effects of extensional stresses on the foamability of polystyrene-

talc composites blown with carbon dioxide. Chem. Eng. Sci. 2012, 75, 49-62.

38. Laun, H. M. Orientation effects and rheology of short glass fiber-reinforced thermoplastics.

Colloid Polym. Sci. 1984, 262, 257-269.

39. Naguib, H. E.; Park, C. B.; Reichelt, N. Fundamental foaming mechanisms governing the

volume expansion of extruded polypropylene foams. J. Appl. Polym. Sci. 2003, 91, 2661.

40. Naguib, H. E.; Park, C. B.; Patrick, L. Effect of talc content on the volume expansion ratio of

extruded PP foams. J. Cell. Plast. 2003, 39, 499.

41. Barlow, C.; Kumar, V.; Flinn, B.; Bordia, R. K.; Weller, J. Impact strength of high density

solid-state microcellular polycarbonate foams. J. Eng. Mater. Technol. 2001, 123, 229-233.

42. Zhai, W.; Yu, J.; He, J. Ultrasonic irradiation enhanced cell nucleation: An effective

approach to microcellular foams of both high cell density and expansion ratio. Polymer 2008,

49, 2430-2434.

108

43. Wang, J.; Zhai, W.; Ling, J.; Shen, B.; Zheng, W.; Park, C. B. Ultrasonic irradiation

enhanced cell nucleation in microcellular poly(lactic acid): a novel approach to reduce cell

size distribution and increase foam expansion. Ind. Eng. Chem. Res. 2011, 50, 13840.

109

Chapter 6

Crystallization-induced structural heterogeneities for open-cell foam extrusion of polypropylene/polytetrafluoroethylene fibrillar

blend 5

6.1 Abstract

We report a novel approach to the continuous and scalable production of low-density open-cell

foams in a polymer extrusion device using supercritical carbon dioxide (CO2). Precise control of

processing temperature allows for the production of foams of isotactic polypropylene (PP)

containing polytetrafluoroethylene (PTFE) fibrils with open-cell content of 97.7% and mass

densities of 0.07 g/cm3. Crystallization of the PP matrix around the PTFE fibrils creates the

structural heterogeneity required for inducing cell opening. The thermodynamics and kinetics of

transcrystallization of PP on PTFE fibrils is investigated. The growth rate of the transcrystalline

layer on PTFE fibrils remains the same as the growth rate of spherulites in the bulk and the

application of Lauritzen-Hoffman‘s growth theory for polymer crystals yields similar results.

However, the induction time for transcrystallization is shorter than the induction time for bulk

crystallization. Ishida‘s approach is used to quantify the free energy difference for

transcrystallization and for spherulitic crystallization in the bulk and the magnitudes are

calculated to be 7.8×10-4

J∙m-2

and 1.3×10-3

J∙m-2

, respectively. The lower free energy difference

seen for transcrystallization suggests it is a more thermodynamically favorable crystallization

process than bulk crystallization. The uptake of CO2 by PP/PTFE is compared to neat PP using a

magnetic suspension balance (MSB) and a pressure-specific volume-temperature (PVT)

apparatus. It is found that the PP/PTFE (99.7/0.3 wt%) exhibits a higher uptake capacity for CO2

than PP, under all investigated temperatures. The higher uptake of CO2 by PP/PTFE is attributed

to the CO2-philic character of PTFE. The increased uptake of CO2 by PP/PTFE increases the

plasticization of the sample, and facilitates cell wall opening during the bubble growth stage of

the foaming process. While PP is used as the matrix and PTFE as the fibrils for the present study,


110

the method may be extended to other semicrystalline polymers for fabricating low-density open-

cell foams.

6.2 Introduction

A special class of low-density porous polymeric materials, called open-cell foams 1

has attracted

considerable attention. Open-cell foams comprise of a three-dimensional network of

interconnected solid struts that delineate cavities (known as cells) throughout the sample volume.

The unsequestered void architecture in open-cell foams gives rise to unique transport and

mechanical properties making them versatile materials suitable for applications such as

cushioning, 2

sound insulation, 3

separation media for filtration, 4

absorbents 5, 6, 7

and scaffolds

for tissue engineering. 8

A multitude of methodologies and synthetic routes have been directed at preparing open-porous

polymer materials with structures that resemble open-cell foams, 9

however, at the bulk densities

pertinent to open-cell foams, fabrication is both challenging and expensive. Cell-opening arising

from foam blowing of a structurally heterogeneous melt, 10, 11, 12

wherein well dispersed arrays

of hard and soft segments are present in the sample, can serve as a robust platform for the

fabrication of open-cell foams. This idea was originally proposed by Rossmy and coworkers 13

who attributed cell-opening in polyurethane to the urea precipitation which occurs a few seconds

prior to cell-opening. Solid urea particles are more rigid than the polymer matrix and may cause

rupture through the classical particle de-foaming mechanism 14

or may induce strong extensional

thinning. 15

While the precise route for cell opening caused by the rigid particles is still a matter

of debate, this idea of creating a structurally heterogeneous melt with well dispersed arrays of

hard and soft segments to obtain open-cell foams has been successfully extended to

thermoplastic materials. Park et al. 12

obtained open-cell foams with an open-cell content in

excess of 99% using partial crosslinking to create the structural heterogeneity in a polymer

matrix, where the crosslinked regions acted as the rigid segments in the soft matrix. Lee et al. 10

blended two semi-crystalline polymers with distinctly different melting temperatures and

processed the blend at a temperature between the melting temperatures of the two components to

create a heterogeneous structure where one component of the blend continued to remain as a

‗soft‘ melt while the other existed as a relatively rigid phase. When this system was foamed in a

continuous extrusion process, an open-cell content of up to 99% was achieved. Harikrishnan et

111

al. 16

dispersed rigid nanoclay particles in a polymer matrix to create the structural heterogeneity

required to facilitate cell-opening.

Our interest is to explore the feasibility of using flow-induced crystallization of semicrystalline

polymers on solid-state polymeric fibrils to generate structural heterogeneity required for

preparing open-cell foams. Recently, crystallites of semi-crystalline polymers have been

visualized in extrusion when the extrusion processing is conducted below the melting

temperature of the semi-crystalline polymer. 17

Additionally, presence of fibrils in a polymer host

has shown to exhibit favorable thermodynamics for inducing polymer crystallization. 18

Under

quiescent conditions, crystallization occurs preferentially around the fibrils resulting in a highly

oriented crystalline structure that envelops the fibril (termed transcrystalline layer). 19

Under

intense flow fields typically observed during polymer extrusion, crystallization in polymer melts

produces the shish-kebab structure, consisting of a central fibrillar core (termed shish) and a

series of folded-chain crystalline lamellae (termed kebabs) oriented normal to the shish. In

unfilled polymer systems, the shish generally constitutes a single assembly of oriented and/or

extended polymer chains, and in filled systems the shish can constitute a high-aspect-ratio

second phase such as carbon nanotubes 20

or polymeric fibrils. 21

These crystals are rigid relative

to the molten matrix and can serve as in-situ generated structural heterogeneities for preparing

foams with high open-cell contents.

In this study, a method to produce open-cell foams of a fibrillar blend of isotactic polypropylene

(PP) containing well-dispersed polytetrafluoroethylene (PTFE) fibrils in a continuous process is

described. The fibrillar blend of PP/PTFE is prepared using a previously reported method. 22, 23, 24

Cell opening is attributed to the formation of crystalline heterogeneities around the PTFE fibrils,

in the melt. Consequently, the thermodynamics of crystallization of PP/PTFE fibrillar blends is

investigated using a combination of Laurtizen-Hoffman‘s growth theory 25

of polymer

crystallization and Ishida‘s induction time approach. 18

Most previous investigation of the

thermodynamics of crystallization of PP on high-aspect ratio fibers have been limited to single

fiber studies where the diameter of the fibers is on the micrometer scale. 26, 27

The current

investigation determines the thermodynamics of crystallization of PP on in-situ generated fibrils

of PTFE during blending where the PTFE fibrils have diameters < 300 nm. The kinetics of

isothermal crystallization of the PP/PTFE with and without the presence of dissolved CO2 is

studied using the Avrami analysis. The CO2-philic character of fluorinated polymers can increase

112

the open-cell content by plasticizing the melt, therefore, the uptake of CO2 by the PP/PTFE is

also characterized.

6.3 Experimental

6.3.1 Materials

A commercially available linear isotactic polypropylene (PP) homopolymer supplied by Japan

Polypropylene, Novatec-PP FY4, with melt flow rate (MFR) = 5 g/10 min (at 503 K/2.16 kg

load) is used as the matrix in this study. The melting temperature of PP is found using a

differential scanning calorimeter (DSC) to be about 438 K. A commercially available

polytetrafluoroethylene (PTFE) powder which shows good dispersion in PP 22

and high affinity

for CO2 23

supplied by Mitsubishi Rayon, MetablenTM

A-3000, is also used. All materials are

used as received. Carbon dioxide and nitrogen are purchased from Linde Gas with purities in

excess of 99%.


Dry blends of PP/PTFE are prepared in proportions of 100/0 (i.e. pure PP is processed using

exactly the same method as other samples for comparison), 99.9/0.1, 99.7/0.3, 99/1 and 97/3 in

weight fraction (wt%). The resulting mixtures are fed in the hopper of a co-rotational twin-screw

extruder manufactured by Toshiba Machine Co. Ltd. (Trade name: TEM-26SS). The screw

diameter is 26 mm and its aspect ratio is 40. The mixture is fed into the extruder through the

hopper zone cooled with water, then melt blended inside the barrel and finally extruded through

a die held at a temperature of 473 K The discharge rate is set at 20 kg/hr. Extrudate from the die

is shaped into a cylindrical strand, led into a water bath at a temperature of 286 K, and pelletized.


To observe the morphology development of PTFE after blending with PP, the blend is dissolved

in xylene and the residual PTFE examined using a scanning electron microscope (JEOL 6060).

To characterize the dispersion of the PTFE fibrils in the PP matrix, the surface of cryogenically

fractured sample of the blend is subjected to a solvent vapour etching process using xylene.

Samples are exposed to xylene vapors for 1 h at 333 K. This type of treatment etches the PP

113

which is soluble in xylene, leading to bringing up the PTFE domains and increasing

differentiation in observation.

The growth of the transcrystalline layer or the growth of spherulites during isothermal

crystallization, is studied on a model sample with a low PTFE content, PP/PTFE (99.9/0.1 wt%).

PP/PTFE (99.9/0.1 wt%) is dilute enough for easy isolation of fibrils with similar dimensions for

visualization of transcrystalline growth on the PTFE fibrils or spherulitic growth in the bulk.

Thin films of PP/PTFE (99.9/0.1 wt%) are prepared by compression molding to a thickness of

100 µm. The film is placed on a glass slide. The slide is placed on a heat-controlled stage set at a

temperature of 483 K for a period of 10 min to erase the thermal history. Then, the stage is

cooled to a chosen crystallization temperature and treated isothermally for various time periods.

Subsequently, the slide is immersed in liquid nitrogen to quench the morphology and prevent any

changes in the crystal structure. The quenched sample is etched using a permanganate etching 28

technique before it is coated with a thin layer of platinum using a sputter coater for observation

using SEM. The thickness of the transcrystalline layer (defined as the width from one edge to the

other) or the diameter of the spherulites is measured from the SEM micrographs.

To observe the morphology of the foamed samples, the samples are dipped in liquid nitrogen and


before observation under SEM. Final images are recorded from randomly selected areas. For

each foamed sample, the foam volume expansion ratio, ⁄ , is calculated where and

are the mass densities of samples before and after foaming, respectively, determined using the

water displacement method according to ASTM-D792. Additionally, the number of bubbles per

unit volume, called the cell density, Nc, is also calculated using SEM images of the foams.

6.3.4 Differential Scanning Calorimetry (DSC) characterization

The isothermal crystallization behaviour of the samples at ambient nitrogen pressure is studied

on a DSC (TA Instruments Q2000) and the isothermal crystallization behaviour of the samples at

3 MPa CO2 pressure is followed using a high pressure DSC (NETZSCH DSC 204 HP). The

specific DSC procedure is as follows: 1) equilibrate at 298 K; 2) heat to 473 K at a heating rate

of 10 K min−1

; 3) isothermal treatment at 473 K for 10 min; 4) cool to the isothermal

crystallization temperature of 413 K (under ambient pressures) or 404 K (under 3 MPa CO2

114

pressures) at a cooling rate of -10 K∙min−1

; 5) isothermal treatment at the crystallization

temperature for 200 min. For the experiments with high pressure CO2, samples are exposed to a

CO2 pressure of 3 MPa during the entire process.

6.3.5 Mass uptake of CO2

The solubility of CO2 in the samples is determined using a Rubotherm GmbH magnetic

suspension balance (MSB). Details of the experimental apparatus and the measurement

procedure are described in previous publications. 29, 30

6.3.6 Open-cell content determination

Quantachrome Ultrapycnometer 1000 is used with nitrogen to determine the open-cell content of

the samples using the procedure explained in ASTM-D6226. Prior to placing the samples in the

gas pycnometer, the unfoamed skin layer of the foamed samples is removed using a steel blade

to make the porous structure accessible by nitrogen. The applied nitrogen pressure is set to a low

value of 0.07 MPa to minimize collapse of the open-celled structure and the measurements are

taken after 20 min to ensure the pressure has equilibrated.


We use a tandem continuous foam extrusion system similar to those employed by the foam

manufacturing industry to foam our samples. The tandem foam system comprises of two single-

screw extruder barrels. The first extruder consists of a 5 horse power (hp) drive, a 19 mm

extruder (Brabender, 05-25-000) that has a mixing screw (Brabender, 05-00-144) with an aspect

ratio of 30. The second extruder consists of a 15-hp variable speed drive, a 38 mm (Killion, KN-

150) extruder that has a mixing screw with an aspect ratio of 30. A metered amount of CO2 gas is

injected into the melt through an injection point positioned at the first extruder.

Pellets of PP/PTFE (100/0 wt% or 99.7/0.3 wt% or 97/3 wt%) are fed into the first extruder

barrel through the hopper. This barrel is maintained at 473 K, a temperature above the melting

temperature of PP but below that of PTFE. The PP in the samples melts completely in the first

extruder due to the temperature as well as the screw motion which causes shear heating. We also

inject 10 wt% CO2 into the first extruder barrel at a constant flow rate using a syringe pump. The

high shear and high pressure caused by the rotating screw inside the first extruder barrel

115

dissolves the gas in the polymer melt through convective diffusion. The second extrusion barrel

is maintained at temperatures below the melting temperature of PP. Consequently, it is in this

second extrusion barrel where crystallization is initiated in the samples. Foaming of the polymer

melt occurs at the die exit where the polymer/CO2 solution is subject to rapid depressurization

resulting in the gas to undergo phase separation. 31

We employ a brass capillary die comprising

of a circular pinhole with a diameter of 1.2 mm and a channel length of 10 mm. The temperature

of the second extruder barrel and the die is brought down and the foamed samples collected at

each set temperature only after the system temperature has equilibrated. The temperature range

we study is 438 to 393 K.


6.4.1 Preparation of open-cell foams of PP/PTFE fibrillar blends

The method to prepare open-cell foams of PP/PTFE fibrillar blends is illustrated schematically in

Fig. 6.1. First PP/PTFE blends are prepared in the ratios 100/0 wt%, 99.7/0.3 wt% and 97/3 wt%

in a twin-screw extruder (Fig. 6.1a). The hydrodynamic stresses during blending cause the PTFE

domains to extend into high-aspect-ratio fibrils (diameter < 500 nm and length > 100 µm, so the

aspect ratio is > 200) due to the high ultimate strain properties of PTFE. 23, 22

Characterizing the

morphology of PP/PTFE (97/3 wt%) is difficult because the PTFE domains tend to be too close

to each other, but the morphology of the PTFE can be seen in the dilute PP/PTFE (99.7/0.3 wt%)

sample. Fig. 6.1b is obtained by subjecting a sample of PP/PTFE (99.7/0.3 wt%) to solvent

vapour etching and Fig. 6.1c is obtained by dissolving away the PP in xylene and observing the

residue under SEM. Since all processing is conducted below the melting temperature of PTFE, it

can be assumed that the fibrillar morphology of PTFE is preserved throughout the foam

extrusion process. Fig. 6.1d shows a schematic of the tandem foam extrusion system used to

foam the PP/PTFE fibrillar blend. 31

In the first extruder, the blend is subjected to a temperature

of 473 K to melt the PP and 10 wt% CO2 is injected and dissolved in the melt. In the second

extruder, the PP/PTFE/CO2 system is subjected to temperatures below the melting temperature

of PP, so that the PP is able to develop crystalline heterogeneities in this extruder. In Fig. 6.1e

the SEM

116

Fig. 6.1 Schematic showing the preparation of open-cell foams from PP /PTFE fibrillar blends:

a) PP/PTFE blends are prepared in the ratios 100/0 wt%, 99.7/0.3 wt% and 97/3 wt% in a co-

rotating twin-screw extruder; b) SEM micrograph of the PP/PTFE (99.7/0/3 wt%) blend after

solvent-vapour etching using xylene. The PTFE exists as well-dispersed, high-aspect-ratio fibrils

in the blend; c) PTFE fibrils after selective removal of PP from the fibrillar blend using xylene.

The PTFE fibrils form an entangled mesh-like structure; d) The fibrillar blend is foamed in a

tandem single-screw foam extrusion system. 31

The temperature of the first extruder is

maintained at 473 K, a temperature above the Tm of PP but below that of PTFE. 10 wt% CO2 is

injected in the first extruder using a syringe pump. PP/PTFE/CO2 is subjected to various

temperatures (ranging from 438 to 393 K) below the melting temperature of PP in the second

extruder; e) SEM micrograph of a permanganate etched 28

sample of PP/PTFE (99.7/0.3 wt%)

after extrusion. A shish-kebab type structure of PP on PTFE fibrils can be seen. Such crystalline

heterogeneities create the heterogeneous melt structure 10, 11, 12

required for open-cell formation.

The temperature profile used from the start of the second extruder to the die exit is 463, 438,

433, 428 and 423 K, respectively. The crystalline morphology is quenched by immersing the

extrudate in liquid nitrogen immediately after it exits the die; f) SEM micrograph of the extruded

foam filament of PP/PTFE (97/3 wt%) when the temperature is 423 K. A highly porous, low-

density open-cell structure is obtained.

117

micrograph of a permanganate etched sample of PP/PTFE after extrusion without blowing agent

is shown. The crystalline morphology is quenched by immersing the extrudate in liquid nitrogen

immediately after it exits the die. The formation of shish-kebab 32

type structures on high-aspect-

ratio second phases such as on carbon nanotubes 20

or polymeric fibrils 21

has also been

previously reported. The crystallized regions create a large stiffness contrast with respect to the

molten matrix. When bubble growth occurs upon exit from the die, cell-opening is initiated.

There exists a processing temperature window within which the crystalline heterogeneities will

yield foams with the highest open-cell content. 33

The optimum temperature for preparing foams

of PP/PTFE with the highest open-cell content is identified to be 423 K. Fig. 6.1f shows the

SEM of the foam cross-section which exhibits a highly porous, interconnected, 3D framework

with struts that delineate cavities (or cells).

6.4.2 Free energy difference for PP transcrystallization on PTFE fibrils and spherulitic crystallization in the bulk

The free energy difference function, Δσ, has been used to describe the thermodynamics of

crystallization of a polymer. A lower value of Δσ indicates a thermodynamically favorable

crystallization process. 18

We study the thermodynamics of transcrystallization of PP on PTFE

fibrils and the thermodynamics of spherulitic crystallization of PP in the bulk under controlled,

quiescent conditions. A dilute PP/PTFE (99.9/0.1 wt%) sample is used for the study to easily

isolate fibrils of similar dimensions. Note that the crystalline morphology that occurs under

quiescent crystallization of polymers on a fiber is the transcrystalline morphology, instead of the

shish-kebab structure observed in the extrusion process.

Fig. 6.2a shows SEM micrographs of the transcrystalline layer of PP growing on PTFE fibrils

during isothermal treatment at a temperature of 410.5 K when the time for the isothermal

treatment, t, is i) 0 s, ii) 1550 s and iii) 2100 s. The thickness of the transcrystalline layer depends

on the isothermal crystallization temperature and time. Fig. 6.2b shows the average thickness

(defined as the width from one edge of the transcrystalline layer to the other edge of the

transcrystalline layer with the fibril in the centre) of the transcrystalline layer on a PTFE fibril

and Fig. 6.2c shows the average diameter of the spherulites in the bulk as a function of time

obtained at various temperatures ranging from 398 K to 413 K. The slopes of the lines presented

in Fig. 6.2b-c are used to determine the growth rate as a function of temperature for

118

Fig 6.2. Thermodynamics of crystallization of PP/PTFE fibrillar blends. a) SEM micrograph of

permanganate etched sample of PP/PTFE (99.9/0.1 wt%) isothermally crystallized for a period of i) 0 s ii)

1550 s, and iii) 2100 s at a temperature of 410.5 K; b) the average thickness (defined as the width from

one edge to the other with the fibril in the centre) of the transcrystalline layer of the PP/PTFE blend as a

function of time. The solid lines represent straight lines of best fit; c) The average diameter of the

spherulites in the PP bulk as a function of time. The solid lines represent straight lines of best fit; d)

growth rate expressed as a function of time for the growth of the transcrystalline layer of PP on PTFE

fibrils and the growth of PP spherulites in the bulk. The growth rates are obtained from the slopes of the

lines in a) and b); e) A plot of ln G + U*/[R(T-T∞)] vs. 1/(TΔTf) in accordance with Lauritzen-Hoffman

growth theory. The data points fit two straight lines corresponding to a Regime III Regime II transition

during crystal growth; 34

f) Induction time, ti, for the onset of crystallization for transcrystallization of PP

on PTFE fibrils and spherulitic crystallization of PP in the bulk. The data is obtained from the time-axis

intercept of the lines in a) and b); g) A plot of ln(1/ti) + U*/[R(T-T∞)] vs. 1/(TΔT²f²) in accordance with

Ishida‘s approach to calculate the free energy difference function, Δσ, for transcrystallization of PP on

PTFE fibrils and spherulitic crystallization of PP in the bulk.

i ii iii

a

𝐥𝐧 𝟏 𝐭 𝐢

𝐔∗

𝐑 𝐓

𝐓

∞

𝟏

𝐓 𝐓𝟐 𝐟 𝐊 𝟑

𝐥𝐧 𝐆

𝐔

∗

𝐑 𝐓

𝐓

∞

𝟏

𝐓 𝐓 𝐟 𝐊 𝟐

1.5x10-6

2.0x10-6

2.5x10-6

3.0x10-6

3.5x10-6

4.0x10-6

-3.0

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Spherulites

Transcrystalline layer

slope = -1.63 x 106

slope = -2.88 x 106

g

0 500 1000 1500 2000 2500 3000 3500

0

5

10

15

20

25

30

396 K 404.5 K

399.5 K 406.5 K

402 K 407 KDia

me

ter

of

sp

he

ruli

tes

(µ

m)

time (s)

c

6.0x10-5

7.0x10-5

8.0x10-5

9.0x10-5

1.0x10-4

-16

-15

-14

-13

-12

-11

slope = -1.85 x 105

slope = -9.17 x 104

e

396 398 400 402 404 406 408 410 412 414

0.0

2.0x10-8

4.0x10-8

6.0x10-8

8.0x10-8

1.0x10-7

1.2x10-7

1.4x10-7 Transcrystalline layer

Spherulites

Gro

wth

rate

(µ

m/s

)

Temperature (K)

d

0 500 1000 1500 2000 2500 3000 3500

0

5

10

15

20

25

30398 K 408 K

401 K 409 K

403.5 K 410.5 K

407 K 413 K

Th

ickn

ess o

f tr

an

scry

sta

llin

e l

ayer

(µm

)

time (s)

b

396 398 400 402 404 406 408 410 412 414

0

200

400

600

800

1000

Ind

uc

tio

n t

ime

(s

)

Temperature (K)

Transcrystalline layer

Spherulites

f

119

transcrystalline growth of PP on the PTFE fibrils or spherulitic growth of PP in the bulk and the

result is plotted in Fig. 6.2d. We observe that the growth rate of the transcrystalline layer and the

spherulites depend on the isothermal crystallization temperature and seem to be simillar for

transcrystalline growth of PP on PTFE fibrils and spherulitic growth of PP in the bulk. This

observation is consistant with other studies that characterize the growth rate of PP crystals on

carbon fibers, 35

Kevlar fibers, 36

and glass fibers 36

which suggest that the transcrystalline growth

rate on the fibers is the same as the spherulitic growth rate in the bulk.

The Lauritzen-Hoffman growth theory, which characterizes the way the chains are deposited on

a substrate during crystallization, is applied to the PP/PTFE system. According to this theory,

the general expression of crystal growth rate, G, is: 25, 34

∗

(6.1)

where G0 is the growth rate constant, U* is a universal constant which describes the activation

energy for polymer chain motion (reptation) in the melt, R is the gas constant, T is the isothermal

crystallization temperature (in K), T∞ = Tg - 30 (in K) where Tg is the matrix polymer glass

transition temperature, is the theoretical temperature at which chain motion ceases, ΔT = Tm0 - T

(in K) is the degree of undercooling, f = 2T/(T0

m + T) is a correcting factor and Kg is the

nucleation rate constant, given by:

(6.2)

where σ is the lateral surface free energy, σe is the fold surface free energy, k is the Boltzmann

constant, and is the thickness of a new layer and is a material constant. The parameter Δhf is

the enthalpy of fusion. The parameter j is determined by the operating growth rate regime

defined by the Lauritzen-Hoffman theory, which describes the occurrence of up to three growth

rate regimes during isothermal polymer crystallization, denoted I, II, and III, in order of

appearance with descending isothermal crystallization temperature. For regimes I and III, the

value of j is equal to 4, and for regime II, the value of j is equal to 2. Eq. 6.1 can be rewritten as:

∗

(6.3)

120

Table 6.1 Values of isotactic PP material constants.

Parameter Value used for calculation

(K) 458

a

∞ (K) 231.2 a

(MJ∙m-3

) 196 a

(Å) 6.26 a

∗ (J) 6280 b

(J∙K-1

) 1.38 x 10-23

a Reference

34;

b Reference

36

Thus, the value of the nucleation rate constant Kg can be determined from the slope of a plot of ln

G + U*/[R(T-T∞)] versus 1/(TΔTf). Such a graph is plotted in Fig. 6.2e using PP material

constants selected from data in the literature and listed in Table 6.1. The parameter σσe can then

be calculated with Eq. 6.2. The magnitude of σσe will be used with Ishida‘s approach, 18

which

utilizes information on the induction time of crystallization, ti, (expressed as the delay between t

= 0 when the temperature reaches the set isothermal crystallization temperature, T, and the time

of onset of nucleation), to estimate the values for free energy difference, Δσ, for the

crystallization of bulk spherulites of PP and transcrystallization of PP on PTFE fibrils. The

abrupt upward trend in growth rate as the crystallization temperature is lowered (Fig. 6.2d)

causes a departure from a straight line fit for the plot shown in Fig. 6.2e. 34

Instead, the data

seems to fit two straight lines with distinct slopes of-1.85 ×105 K

2 and -9.17 ×10

4 K

2, which

suggests the presence of a regime transition. The slope ratio is calculated to be about 2,

consistent with a regime II regime III transition, seen for the crystallization of PP. 37, 34

From

the slopes of the two straight line representing the two regimes, the values of σσe are calculated

as 4.54×10-4

J2∙m

-4 and 4.53×10

-4 J

2∙m

-4 corresponding to an average value of σσe = 4.535×10

-4

J2∙m

-4 which is on the same order of magnitude as is previously estimated for the

transcrystallization of PP on fibers and spherulitic crystallization of PP in the bulk, in an

independent study where the average value of 8.2 ×10-4

J2∙m

-4 is reported.

36 Thus, the value of

121

σσe from crystal growth rate data can be used safely for the determination of the free energy

difference function, Δσ.

The induction time, ti, is taken as the time-axis intercept from the growth versus time data in Fig.

6.2b-c. Fig. 6.2f shows the induction time, ti, as a function of isothermal crystallization

temperature for transcrystallization of PP on PTFE fibrils and spherulitic crystallization of PP in

the bulk. The crystal nucleation rate, I, can be described by the following equation: 36

∗

∗

(6.4)

where ΔG* is the critical free energy for crystal nucleation: 36

∗

(6.5)

and Δσ is the free energy difference at the interface of the matrix and the fibril. According to

Ishida, 18

the induction time and the crystal nucleation rate are linked by the relation

(6.6)

By substituting Eqs. 6.5 and 6.6 into Eq. 6.4, Eq. 6.4 can be rewritten in the form:

∗

(6.7)

where A is a constant. The magnitude of σσeΔσ can be deduced by plotting a graph of ln(1/ti) +

U*/[R(T-T∞)] versus 1/(TΔT²f²) shown in Fig. 6.2g. From Eq. 6.7 the slope of the straight line is

given by

. The slope for transcrystallization of PP on PTFE fibrils is -1.63 × 10

8

K3

which corresponds to a σσeΔσ = 5.3 × 10-7

J3∙m

-6 and the slope for spherulitic crystallization of

PP in the bulk is -2.88 × 108

K3

which corresponds to a σσeΔσ = 8.5 × 10-7

J3∙m

-6, respectively.

Hence, Δσ is calculated by dividing values of σσeΔσ with σσe determined earlier. The magnitude

of Δσ is found to be 7.8 × 10-4

J∙m-2

for the transcrystallization of PP on PTFE fibrils and 1.3 ×

10-3

J∙m-2

for the spherulitic crystallization of PP in the bulk. These values are in good agreement

with other independent estimates of the free energy difference where the Δσ for PP bulk

crystallization is estimated to be 2.1 × 10-3

J∙m-2

, 36

and PP transcrystallization on a carbon fiber

122

is estimated to be 1.5 × 10-3

J∙m-2

, 36

PP transcrystallization on a Kevlar fiber is 1.3 × 10-3

J∙m-2

,

36 and PP transcrystallization on a polytetrafluoroethylene fiber is 7.5 × 10

-4 J∙m

-2. 27

That PP

transcrystallization on PTFE exhibits lower Δσ than spherulitic crystallization of PP in the bulk,

indicates that transcrystalline growth is favored from a thermodynamic standpoint.

6.4.3 Isothermal crystallization kinetics of PP/PTFE fibrillar blends

The isothermal DSC thermograms for samples containing varying PTFE fibril contents are

shown in Fig. 6.3a. The heat flow traces exhibit a single crystallization peak typically observed

for isothermal crystallization of semicrystalline polymers. Due to the slow kinetics of

crystallization of neat PP at the isothermal crystallization temperature of 413 K, an insert is

included in Fig. 6.3a which shows that PP completes crystallization over a longer time than

when PTFE fibrils are present in the PP. The relative crystallinity at different crystallization

times, X(t), defined as the ratio of crystallinity at time t to the crystallinity in the limit of steady

state (t→∞), is given as

∫ (

)

∫ (

)

(6.8)

taking t = 0 to be the instant of attainment of thermal equilibrium and t = to be the instant of

completion of crystallization. Fig. 6.3b shows the plot of X(t) versus t. It can be seen from Fig.

6.3b that all the curves have a similar sigmoidal shape. Additionally, the increase in the PTFE

fibril content shortens the duration for crystallization to complete. In order to quantify the change

in the isothermal crystallization kinetics, the Avrami equation 38, 39

(6.9)

is used, where n is the Avrami exponent and dictates the dimensionality of the crystal growth and

K is the overall crystallization rate constant. Eq. 6.3 can be rewritten as:

*

+ (6.10)

123

Fig. 6.3 Isothermal crystallization kinetics of PP/PTFE fibrillar blends containing different PTFE

contents: a) Isothermal crystallization vs. crystallization time calculated from DSC at 413 K; b)

Relative crystallinity, X(t), as a function of crystallization time calculated from the DSC data in

a); c) Avrami plots for the isothermal melt crystallization kinetics derived from the data in b); d)

Half-time, t1/2,of crystallization determined for different PTFE fibril contents using Eq. 6.11.

Increasing the PTFE content over the studied fibril content range leads to decrease in the

isothermal crystallization t1/2.

0 2000 4000 6000 8000 10000 12000

0.0

0.2

0.4

0.6

0.8

1.0

X(t

)

time (s)

PTFE 0 wt%

PTFE 0.3 wt%

PTFE 1 wt%

PTFE 3 wt%

b

2 3 4 5 6 7 8 9 10

-8

-6

-4

-2

0

2

4

PTFE 3 wt%

PTFE 1 wt%

PTFE 0.3 wt%

PTFE 0 wt%

ln ln

{1/[

1-X

(t)]

}

ln(time)

c0.0 0.5 1.0 1.5 2.0 2.5 3.0

0

1000

2000

5000

6000

t 1/2 (

s)

PTFE fibril content (wt%)

d

124

The crystallization rate constant K and the Avrami exponent n can be determined from the y-

intercept and the slope of the linear section in the plot of ln ln {1/[1-X(t)]} vs. ln (t), respectively.

Such a plot is shown in Fig. 6.3c. The obtained Avrami parameters of PP and PP/PTFE fibrillar

blends are summarized in Table 6.2. However, it is difficult to compare the overall

crystallization rate directly from the values of K because the unit of K is s-n

and n is not constant

for the samples. Thus, the half-time of crystallization, t1/2, the time required to achieve 50% of

the final crystallization of the samples is computed for the purpose of discussing the kinetics of

crystallization. The value of t1/2 is calculated using

(

)

(6.11)

Fig. 6.3d illustrates the variation of t1/2 as a function of the fibril content. It can be seen that t1/2

decreases with an increase in the fibril content. Such variation indicates that the overall

isothermal crystallization rate increases with an increase in fibril content due to the

heterogeneous crystal nucleation effect of the fibrils. For example, at 413 K, the t1/2 for

crystallization of PP/PTFE (99/0.3 wt%) is 1516 s; however, the t1/2 for crystallization of

PP/PTFE (9/3 wt%) is reduced to 754 s.

Table 6.2 Summary of Avrami parameters for the isothermal crystallization kinetics of PP/PTFE

fibrillar blends with different PTFE contents at a crystallization temperature of 413 K.

Sample

PP/PTFE (wt%) n

a K

(s

-n) b

t1/2 (s) c

100/0 3.46 6.74 × 10-14

5.73 × 103

99.7/0.3 3.47 6.35 × 10-12

1.51 × 103

99/1 3.48 2.53 × 10-11

9.90 × 102

97/3 3.67 1.92 × 10-11

7.53 × 102

a obtained from slopes of lines of best fit for the Avrami plots in Fig. 6.3c;

b obtained from the

time axis intercept of lines of best fit for the Avrami plots in Fig. 6.3c; c

determined using Eq.

6.11.

125

6.4.4 Isothermal crystallization kinetics of PP/PTFE fibrillar blends with dissolved CO2

The dissolution of CO2 in a melt influences polymer chain mobility, and consequently, the

crystallization behaviour. We study the isothermal crystallization behaviour of PP/PTFE with

different PTFE fibril contents at 3 MPa CO2 pressure. Note that the experiment is conducted at a

different isothermal crystallization temperature (i.e. 404 K) than the ambient pressure isothermal

crystallization (i.e. 413 K), because the crystallization process for PP/PTFE in the presence of

CO2 cannot be accurately captured at the same temperature as PP/PTFE in the absence of CO2.

While CO2 pressures in the range of 10 to 20 MPa are more representative of the pressures in our

foam extrusion process, we are experimentally limited by the maximum pressure reachable in the

high pressure DSC instrument. Fig. 6.4a shows the isothermal crystallization thermograms as a

function of time for PP/PTFE fibrillar blends with different PTFE fibril contents. Neat PP does

not show any signs of crystallization over the investigated isothermal treatment period. The time

for crystallization to complete is shortened as the PTFE fibril content is increased. The

isothermal crystallization thermograms in Fig. 6.4a are used to compute the relative crystallinity,

X(t), as a function of time, t, and is shown in Fig. 6.4b. Avrami plots corresponding to the

sigmoidal curves in Fig. 6.4b are plotted in Fig. 6.4c to derive the Avrami parameters, i.e., the

Avrami exponent, n, and the rate constant, K, which are summarized in Table 6.3 along with the

crystallization half-times, t1/2, determined using Eq. 6.11. The crystallization half-times, t1/2, are

plotted in Fig. 6.4d as a function of the fibril content. It can be seen that t1/2 decreases with an

increase in the fibril content indicating that the PTFE fibrils continue to behave as crystal

nucleating agents in the presence of compressed CO2.

126

Fig. 6.4 Isothermal crystallization kinetics of PP/PTFE fibrillar blends containing different PTFE

contents under 3 MPa CO2 pressure: a) Isothermal crystallization vs. crystallization time

calculated from DSC at 404 K; b) Relative crystallinity, X(t), as a function of crystallization time

calculated from the DSC data in a); c) Avrami plots for the isothermal melt crystallization

kinetics derived from the data in b); d) Half-time, t1/2,of crystallization determined for different

PTFE fibril contents using Eq. 6.11. Increasing the PTFE content over the studied fibril content

range leads to decrease in the isothermal crystallization t1/2.

0 200 400 600 800 1000

0.0

0.2

0.4

0.6

0.8

1.0

X(t

)

time (s)

PTFE 0.3 wt%

PTFE 1 wt%

PTFE 3 wt%

b

2 3 4 5 6 7-12

-10

-8

-6

-4

-2

0

2

4

PTFE 3 wt%

PTFE 1 wt%

PTFE 0.3 wt%

ln ln

{1/[

1-X

(t)]

}

ln(time)

c

0.0 0.5 1.0 1.5 2.0 2.5 3.0200

250

300

350

400

450

500

t 1/2 (

s)

PTFE fibril content (wt%)

d

127

Table 6.3 Summary of Avrami parameters for the isothermal crystallization kinetics of PP/PTFE

fibrillar blends with different PTFE contents at a crystallization temperature of 404 K and a CO2

pressure of 3 MPa.

Sample PP/PTFE (wt%)

n

a K

(s

-n) b

t1/2 (s) c

100/0 - - -

99.7/0.3 3.08 3.50 × 10-9

4.97 × 102

99/1 2.27 1.26 × 10-6

3.34 × 102

97/3 1.80 3.73 × 10-5

2.35 × 102

a obtained from slopes of lines of best fit for the Avrami plots in Fig. 6.4c;

b obtained from the

time axis intercept of lines of best fit for the Avrami plots in Fig. 6.4c; c

determined using Eq.

6.11.

128

6.4.5 CO2-philicity of PTFE in PP/PTFE fibrillar blends

0 20 40 60 80 100 120 1400.06

0.07

0.08

0.09

0.10

0.11

0.12

0.13

0.14

PP/PTFE 428 K

PP/PTFE 453 K

PP/PTFE 483 K

Lorentzian Fit

PP 428 K

PP 453 K

PP 483 K

Ma

ss

up

tak

e o

f C

O2(g

/g o

f p

oly

me

r)

time (min)

Fig. 6.5 Mass uptake of CO2 in PP/PTFE (99.7/0.3 wt%) as a function of time measured at three

temperatures: 428 K, 453 K, and 483 K. Presence of PTFE results in a higher CO2 uptake by the

samples at all studied temperatures. The solid lines correspond to the Lorentzian fit to the data.

Fluoropolymers exhibit strong thermodynamic affinity for CO2. 40, 41, 42, 43

Fig. 6.5 compares the

mass uptake of CO2 in neat PP and PP/PTFE (99.7/0.3 wt%) as a function of time measured at

three temperatures: 428 K, 453 K, and 483 K. The mass uptake of CO2 is reported as mass in

grams of the dissolved CO2 per gram of the polymeric sample (g g-1

). The CO2 uptake kinetics in

the studied samples can be described by the Lorentzian function:

129

Table 6.4 Fitting parameters for Eq. 6.12 describing the uptake kinetics in neat PP and PP/PTFE

(99.7/0.3 wt%) at a three different temperatures of 428 K, 453 K, and 483 K, and a CO2 pressure

of 17.2 MPa.

(6.12)

Where S is the mass uptake, t is the time, S0, t0, A, and w, are fitting parameters. The curves of

best-fit are included in Fig. 6.5 and the corresponding fitting values of S0, t0, A and w are

presented in Table 6.4.

When 0.3 wt% PTFE is added to the PP matrix, the amount of CO2 localized within the sample

increases at all studied temperatures. For example, at a temperature of 428 K, the solubility of

CO2 in neat PP is about 0.115 g g-1

but when 0.3 wt% PTFE is present in the PP the CO2 uptake

increases to about 0.135 g g-1

.

The increased uptake of CO2 by PP when PTFE is present can influence the final morphology of

the extruded foam. A higher uptake of CO2 will result in a greater degree of CO2-induced

plasticization which will, in turn, reduce the viscosity of the sample. Consequently, bubble wall

opening will occur more readily in the sample yielding foams with a higher open-cell content.

The increase in open-cell content through gas-induced plasticization has previously been

demonstrated. 44

Butane in the gas-phase shows a high solubility in most polyolefins and has

been effectively applied as a secondary foam blowing agent to increase the open-cell content in

partially cross-linked polyethylene foams. 45

Sample

PP/PTFE (wt%)

Temperature

(K) S0 t0 A w

100/0

428 0.112 1.89 -0.99 5.90

453 0.101 4.34 -0.66 3.82

483 0.090 -8.35 -69.37 0.49

99.7/0.3

428 0.133 3.35 -1.36 8.69

453 0.117 0.54 -1.43 5.95

483 0.089 0.48 -1.89 1.70

130

6.4.6 Characterization of open-cell foams

Fig. 6.6 Characterization of foams: a) Cell density as a function of temperature; b) Foam volume

expansion ratio as a function of temperature; c) Cell wall thickness expressed as a function of

cell density and volume expansion ratio using Eqs. 6.13 and 6.14. If the cell density and/or

volume expansion ratio can be increased, the cell wall thickness can be effectively decreased; d)

Cell wall thickness calculated using data from a) and b) and Eqs. 6.13 and 6.14; e) Open-cell

content determined using nitrogen pycnometry; f) SEM micrograph showing the open-cell foam

morphology for extruded PP/PTFE (97/3 wt%) obtained at 423 K.

c

f

390 400 410 420 430 440

100

101

102

Ce

ll W

all T

hic

kn

ess (

µm

)

Temperature (K)

PTFE 0 wt%

PTFE 0.3 wt%

PTFE 3 wt%

d

390 400 410 420 430 4400

5

10

15

20

25

30

35

Exp

an

sio

n R

ati

o

Temperature (°C)

PTFE 3 wt%

PTFE 0.3 wt%

PTFE 0 wt%

b390 400 410 420 430 440

103

104

105

106

107

108

109

1010

Ce

ll D

en

sit

y (

cells/c

m3)

Temperature (K)

PTFE 3 wt%

PTFE 0.3 wt%

PTFE 0 wt%

a

390 400 410 420 430 44010

20

30

40

50

60

70

80

90

100

110

Op

en

-Ce

ll C

on

ten

t (%

)

Temperature (K)

PTFE 3 wt%

PTFE 0.3 wt%

PTFE 0 wt%

e

131

Extrusion foaming is conducted on the neat PP and PP/PTFE fibrillar blends with PTFE fibril

contents of 0.3 wt% and 3 wt%. Since the foaming temperature is below the melting temperature

of PP (Tm = 438 K), the development of crystalline heterogeneities under the processing

conditions are expected (see Fig. 6.1e). 17

All processing is performed below the melting

temperature of the PTFE to ensure that the PTFE fibrils maintain their fibrillar structure. Care is

taken to ensure that the die pressure is above the solubility pressure of CO2 in molten PP to

prevent phase separation of CO2 from the polymeric system prior to foaming at the die exit.

Fig. 6.6a shows the cell density as a function of temperature for the three samples. Compared to

neat PP, PP/PTFE (99.7/0.3 wt%) exhibits up to a one order of magnitude increase in cell density

(between 107 to 10

8 cells cm

-3), and PP/PTFE (97/3 wt%) exhibits up to a two orders of

magnitude increase in cell density (between 108 to 10

9 cells cm

-3)over the studied temperature

range. Fig. 6.6b shows the foam volume expansion ratio as a function of temperature for the

three samples. While PP/PTFE (99.7/0.3 wt%) shows very similar volume expansion ratios to

neat PP (maximum expansion ratio is around 20-fold), PP/PTFE (97/3 wt%) shows up to a ten-

fold increase in the foam volume expansion ratio (maximum expansion ratio of 30-fold at a

temperature of 403 K) compared to neat PP. Cell densities and expansion ratios play a key role in

cell-opening during extrusion foaming by affecting the cell wall thickness. The mean cell wall

thickness (δ) can be described in terms of cell density ( ) and expansion ratio ( by: 46

(

√

) (6.13)

and Nc is given by: 46

(6.14)

where d is the mean cell size. Fig. 6.6c shows a three-dimensional graph of cell wall thickness

determined as a function of volume expansion ratio and the cell density. It can be seen that an

increase in cell density and/or expansion ratio is expected to yield foams with thinner cell walls.

Fig. 6.6d shows the cell wall thickness for PP, PP/PTFE (99.7/0.3 wt%) and PP/PTFE (97/3

wt%) calculated from the cell density and expansion ratio data presented in Figs. 6.6a-b and the

Eqs. 6.13 and 6.14. The figure reveals that foams with thinner cell walls are obtained as the

132

temperature is decreased from 438 to 393 K. This occurs because of an increase in the cell

density as well as an increase in the expansion ratio with a decrease in temperature as shown in

Figs. 6.6a-b.

Fig. 6.6e shows the open-cell content of the foams obtained at various temperatures within the

investigated temperature range. An optimum temperature window exists when foams with

maximum open-cell contents can be obtained and corresponds to the temperature range 416 to

430 K. Below the optimum temperature, the ‗stiffness‘ of the cell wall increases and becomes the

governing parameter preventing cell opening. Above the optimum temperature, the cell wall

thickness increases (see Fig. 6.6d) due to the high rate of blowing agent loss (high gas diffusivity

at high temperatures) through the foam and becomes the dominant factor preventing cell-

opening. Fig. 6.6f shows the morphology of the open-cell foam obtained for PP/PTFE (97/3

wt%) at a temperature of 423 K. Reticulated foam morphology can be seen in Fig. 6.6f which

exhibits an open-cell content of 97.7%, a cell density of 107 cells cm

-3 and a foam volume

expansion ratio of 13.

6.5 Conclusion

We report an easy-to-scale extrusion process to prepare low-density open-cell foams of PP

containing in-situ generated fibrils of PTFE. Fundamental investigation of the thermodynamics

of crystallization of neat PP and of PP/PTFE, and the kinetics of crystallization of neat PP and of

PP/PTFE suggest that by controlling the fibril content and the processing conditions, crystalline

heterogeneities can be created around the PTFE fibrils during the foaming process. These

crystalline heterogeneities create the heterogeneous melt structure required for fabricating open-

cell foams. At a processing temperature of 423 K, foams of PP/PTFE (97/3 wt%) with open-cell

contents of 97.7% and bulk densities of 0.07 g cm-3

are produced. The method can be

implemented to prepare open-cell foams well-suited in applications such as sound insulation,

filtration membranes, absorbents, and scaffolds for tissue engineering.

133

References

1. Lee, S. T.; Ramesh, N. S. Foam stability in flexible polyurethane foam systems. In Polymeric

Foams, mechanisms and materials; Lee, S. T., Ramesh, N. S., Eds.; CRC Press: Boca Raton,

2004; pp 173-251.

2. Gong, L.; Kyriakides, S.; Jang, W. Y. Compressive response of open-cell foams. Part I:

Morphology and elastic properties. Int. J. Solids. Struct. 2005, 42, 1355-1379.

3. Park, C. P. Acoustical open-cell polylefins and process for making. US6541105 B1, April 1,

2003.

4. Beard, K. W. Production of high porosity open-cell membranes. US20070100012 A1,

December 6, 2006.

5. Krueger, J. J. Low density flexible resilient absorbent open-cell thermoplastic foam.

US10950541, November 6, 2007.

6. Browers, S. D.; Wiegand, D. E. Oil sorbent material made by opening cells of a closed cell

foam. US4183984, January 15, 1980.

7. Rizvi, A.; Chu, R. K. M.; Lee, J. H.; Park, C. B. Superhydrophobic and oleophilic open-cell

foams from polymer-fibrillar blends of polypropylene and polytetrafluoroethylene. ACS

Appl. Mater. Interfaces 2014, 6, 21131-21140.

8. Vyakamam, M. N.; Zimmerman, M. C.; Gosiewiska, A. Porous tissue scaffolds for the repair

and regeneration of dermal tissue. US20040146544 A1, July 24, 2004.

9. Wu, D.; Xu, F.; Sun, B.; Fu, R.; He, H.; Matyjaszewski, K. Design and preparation of porous

polymers. Chem. Rev. 2012, 112, 3959-4015.

10. Lee, P. C.; Wang, J.; Park, C. B. Extruded open-cell foams using two semicrystalline

polymers with different crystallization temperatures. Ind. Eng. Chem. Res. 2006, 45, 175-

134

181.

11. Lee, P. C.; Wang, J.; Park, C. B. Extrusion of microcellular open-cell LDPE-based sheet

foams. J. Appl. Polym. Sci 2006, 102, 3376-3384.

12. Park, C. B.; Padareva, V.; Lee, P. C.; Naguib, H. E. Extruded open-celled LDPE-Based

foams using non-homogeneous melt structure. J. Polym. Eng. 2005, 25, 239.

13. Rossmy, G. R.; Kollmeier, H. J.; Lidy, W.; Schator, H.; Wiemann, M. Cell-opening in one-

shot flexible polyether based polyurethane foams. The role of silicone surfactant and its

foundation in the chemistry of foam formation. J. Cell. Plast. 1977, 13, 26-35.



2013, 54, 4645–4652.

15. Zhang, X. D.; Davis, H. T.; Macosko, C. W. A new cell opening mechanism in flexible

polyurethane foams. J. Cell. Plast. 1998, 10, 2037-2040.

16. Harikrishnan, G.; Patro, T. U.; Khakhar, D. V. Polyurethane foam−clay nanocomposites:

nanoclays as cell openers. Ind. Eng. Chem. Res. 2006, 45, 7126-7134.



18. Ishida, H.; Bussi, P. Surface induced crystallization in ultrahigh-modulus polyethylene fiber-

reinforced polyethylene composites. Macromolecules 1991, 24, 3569-3577.

19. Keller, A. Unusual orientation phenomena in polyethylene interpreted in terms of the

morphology. J Polym Chem 1955, 15, 31-49.

20. Li, L.; Li, C. Y.; Ni, C. Polymer crystallization-driven, periodic patterning on carbon

nanotubes. J. Am. Chem. Soc. 2006, 128, 1692-1699.

135

21. Shen, K. Z.; Yang, W.; Huang, R.; Yang, M. B. Transcrystalline morphology of an in-situ

microfibrillar poly(ethylene terephthalate)/poly(propylene) blend fabricated through a slit

extrusion hot stretching-quenching process. Macromol. Rapid Commun. 2004, 25, 553-558.

22. Ali, M. A.; Nobukawa, S.; Yamaguchi, M. Morphology development of

polytetrafluoroethylene in a polypropylene melt (IUPAC Technical Report). Pure Appl.

Chem. 2011, 83, 1819-1830.



2013, 54, 4645–4652.



25. Hoffman, J. D.; Davis, G. T.; Lauritzen, J. I. Treatise on solid state chemistry. Hannay, N. B.,

Ed.; Plenum Press: New York, 1976.

26. Wang, C.; Liu, C.-R. Transcrystallization of polypropylene composites: nucleating ability of

fibres. Polymer 1999, 40, 289.

27. Wang, C.; Hwang, L. M. Transcrystallization of PTFE fiber/PP composites (I) crystallization

kinetics and morphology. J. Polym. Sci. Part B: Polym. Phys. 1996, 34, 47-56.

28. Olley, R. H.; Bassett, D. C. An improved permanganic etchant for polyolefines. Polymer

1982, 23, 1707-1710.

29. Li, G.; Wang, J.; Park, C. B.; Simha, R. Measurement of gas solubility in linear/branched PP

melts. J. Polym. Sci. B 2007, 45, 2497-2508.

30. Li, Y. G.; Park, C. B.; Li, H. B.; Wang, J. Measurement of the PVT property of PP/CO2

solution. Fluid Phase Equilib. 2008, 270, 15-22.

31. Park, C. B.; Suh, N. P. Filamentary extrusion of microcellular polymers. Polym. Eng. Sci.

136

1996, 36, 34.

32. Binsbergen, F. L. Orientation-induced nucleation in polymer crystallization. Nature 1966,

211, 516 - 517.

33. Park, C. B.; Padareva, V.; Lee, P. C.; Naguib, H. E. Extruded open-celled LDPE-based

foams using non-homogeneous melt structure. J. Polym. Eng. 2005, 25, 239-260.

34. Clark, E. J.; Hoffman, J. D. Regime III crystallization in polypropylene. Macromolecules

1984, 17, 878-885.

35. Thomason, J. L.; Van Rooyen, A. A. Transcrystallized interphase in thermoplastic

composites. J. Mater. Sci. 1992, 27, 889-896.

36. Assouline, E.; Pohl, S.; Fulchiron, R.; Gerard, J. F.; Lustiger, A.; Wagner, H. D.; Marom, G.

The kinetics of α and β transcrystallization in fibre-reinforced polypropylene. Polymer 2000,

41, 7843-7854.

37. Hoffman, J. D. Regime III crystallization in melt-crystallized polymers: The variable cluster

model of chain folding. Polymer 1983, 24, 3-26.

38. Avrami, M. Kinetics of phase change-II Tranformation-time relations for random distribution

of nuclei. J. Chem. Phys. 1940, 8, 212–224.

39. Avrami, M. Kinetics of phase change-I. General theory. J. Chem. Phys. 1939, 7, 1103-1112.

40. Tuminello, W. H.; Dee, G. T.; McHugh, M. A. Dissolving perfluoropolymers in supercritical

carbon dioxide. Macromolecules 1995, 28, 1506.

41. DeSimone, J. M.; Guan, Z.; Elsbernd, C. S. Synthesis of fluoropolymers in supercritical

carbon dioxide. Science 1992, 257, 945.

42. Shieh, Y. T.; Hsiao, T. T.; Chang, S. K. CO2 pressure effects on melting, crystallization, and

morphology of poly(vinylidene fluoride). Polymer 2006, 47, 5929-5937.

137

43. Raveendran, P.; Wallen, S. L. Exploring CO2-philicity: effects of stepwise fluorination. J.

Phys. Chem. B 2003, 107, 1473-1477.

44. Lee, P. C.; Naguib, H. E.; Park, C. B.; Wang, J. Increase of open-cell content by plasticizing

soft regions with secondary blowing agent. Polym. Eng. Sci. 2005, 45, 1445-1451.

45. Lee, P. C.; Naguib, H. E.; Park, C. B.; Wang, J. Increase of Open-Cell Content by

Plasticizing Soft Regions with Secondary Blowing Agent. Polym. Eng. Sci. 2005, 45, 1445-

1451.

46. Klempner, D.; Frisch, K. C. Handbook of Polymeric Foams and Foam Technology; Hanser

Publishers: Munich, Vienna, and New York, 1991.

138

Chapter 7

Superhydrophobic and oleophilic open-cell foams from fibrillar blends of polypropylene and polytetrafluoroethylene 6

7.1 Abstract

Effective removal of oils from water is of global significance for environmental protection. In

this study, we study the hydrophobicity and oleophilicity of open-cell polymer foams prepared in

a continuous and scalable extrusion process. The material used to prepare the open-cell foams is a

fibrillar blend of polypropylene (PP) and polytetrafluoroethylene (PTFE). Scanning electron

microscopy (SEM) images of the morphology of the PP/PTFE fibrillar blend reveal that the

PTFE has a fibrillar morphology in the PP matrix. SEM micrograph of the extruded foam shows

the formation of an interconnected open-cell structure. Using nitrogen pycnometry, the open-cell

content is estimated to be 97.7%. A typical bulk density of the open-cell foam is measured to be

about 0.07 g cm-3

corresponding to a void fraction of 92%. Thus, a large three-dimensional space

is made available for oil storage. A drop of water on the cross-section of the extruded open-cell

foam forms a contact angle of 160° suggesting that the open-cell foam exhibits

superhydrophobicity. The open-cell foam can selectively absorb various petroleum products such

as octane, gasoline, diesel, kerosene, light crude oil and heavy crude oil from water and the

uptake capacities range from about 5 to 24 g g-1

. The uptake kinetics can be enhanced by

exposing the open-cell foam to high intensity ultrasound which increases the surface porosity of

the thin, impervious, foam ‗skin‘ layer. The reusability of the foam can be improved by using a

matrix polymer which demonstrates superior elastic properties and prevents the foams from

undergoing a large permanent deformation upon compression to ‗squeeze out‘ the oil. For

example, when the PP homopolymer matrix is replaced with a PP random copolymer, the

permanent deformation for 10 compressive cycles is reduced from about 30% to 10%. To the

best of our knowledge, these PP-based open-cell foams outperform PP-based absorbents

6 Reproduced from: Rizvi, A.; Chu, R.K.M.; Lee, J.H.; Park, C.B., "Superhydrophobic and oleophilic open-cell

foams from fibrillar blends of polypropylene and polytetrafluoroethylene", ACS Appl. Mater. Interfaces 2014, 6,

21131-21140.

139

conventionally used for oil-spill cleanup applications such as nonwoven PP fibers or melt-blown

PP pads.

7.2 Introduction

Crude oil fulfills a substantial portion of the world‘s energy needs. Oil exploration, extraction,

transportation, storage and usage create a risk of spillage which can be damaging to the

environment. The world spends $10 billion dollars annually on oil spill remediation. 1

There is an

urgent need to develop new methods for oil spill cleanup to mitigate the adverse environmental

implications particularly because the effectiveness of the treatment depends on several factors

such as oil type and weather conditions. 2

One of the most efficient and economical methods for

oil spill cleanup is oil sequestering by sorption. The properties of an ideal absorbent for oil spill

remediation include high oleophilicity and hydrophobicity to maximize the oil/water selectivity,

high oil uptake capacity, high buoyancy, and low cost. Natural absorbents for oil spill cleanup

include cotton, 3

milkweed, 3 wool,

4 and vegetable fibers.

5 Unfortunately, these absorbents

exhibit poor oil/water selectivity due to their affinity for water limiting their effectiveness in the

cleanup of oil spills. Polyurethane foams are synthetic absorbents which have characteristics like

high uptake capacities, easily scalable fabrication, and low density but also show an affinity for

water. 6, 7, 8

Furthermore, they pose a waste disposal challenge as polyurethane foams are non-

recyclable and thermoset materials. Recently, porous hydrophobic and oleophilic materials

(PHOMs) such as carbon-based sponges, 9,

10

porous boron nitride, 11

metal-organic frameworks

(MOFs), 12

and polymethylsilsesquioxane aerogels 13

have emerged as attractive alternatives in

treating releases because of their excellent oil uptake capacity and selectivity. However, present

studies ignore the high material cost and complex synthesis procedures to fabricate these

PHOMs. The deficiencies of these existing technologies call for an urgent need to develop

methods to fabricate absorbents with inexpensive, recyclable materials in a high throughput

process.

Herein, we study the hydrophobic and oleophilic properties of open-cell foams prepared from

fibrillar blends of polypropylene (PP) and polytetrafluoroethylene (PTFE) on a large-scale in a

continuous extrusion process 14

capable of producing the open-cell foam at a rate of 2 kg/h (See

Appendix B, V1: Extrusion of open-cell PP/PTFE foams). The resulting open-cell foams

comprise of a three dimensional network of interconnected solid struts that delineate cavities

140

(known as cells), throughout the sample volume. A facile method to fabricate such foams is

described and their hydrophobic properties are characterized. The uptake capacities of these

foams are studied for several petroleum products such as octane, gasoline, diesel, kerosene, light

crude oil and heavy crude oil. Ultrasound irradiation 15, 16, 17

treatment is used to increase the

porosity of the thin, impervious ‗skin‘ layer of the extruded open-cell foams, and the change in

the kinetics of oil uptake by the foams before and after treatment is characterized. The reusability

of the open-cell foams after oil absorption is investigated by subjecting the foams to a cyclic-

compression stress-strain test as a controlled analogue for mechanical ―squeezing‖ required for

extracting the oil from the open-cell foam.

141

7.3 Experimental

7.3.1 Materials

A commercially available linear isotactic polypropylene homopolymer (PP-homopolymer)

supplied by Japan Polypropylene, Novatec-PP FY4, with melt flow rate (MFR) = 5 g/10 min (at

230°C/2.16 kg load) and poly(propylene-co-ethylene) random copolymer (PP-copolymer)

commercially available as Sabic ® PP 670K with a melt flow index (MFI) of 10 g/10 min (at

230°C/2.16 kg load) are used as the matrix polymers in this study. The melting temperature of

the PP-homopolymer is found using DSC to be about 165°C and the PP-copolymer is found to

be around 149°C. A commercially available polytetrafluoroethylene (PTFE) powder which

shows good dispersion in PP 18

and high CO2 19

solubility supplied by Mitsubishi Rayon

Company, MetablenTM

A-3000, is also used. Carbon dioxide and nitrogen are purchased from

Linde Gas with purities in excess of 99%. Diesel (Shell Ultra Low Sulphur Diesel) and gasoline

(Shell Bronze Octane Number 87) are purchased from a Shell Canada Ltd. gas station. Light

Crude Oil (Paraffinic light) and Heavy Crude Oil (Aromatic-Naphthenic heavy) are purchased

from ONTA Inc. Kerosene Fuel is purchased from Canadian Tire Corporation Ltd. Octane is

purchased from Caledon Laboratories Ltd. Sudan III red dye is purchased from EMD Millipore

Corp.


Dry blends of the matrix polymer (either PP homopolymer or PP-copolymer) and PTFE are

prepared in proportion of 97/3 in weight fraction (wt%). The resulting mixture is fed in the

hopper of a co-rotational twin-screw extruder manufactured by Toshiba Machine Co. Ltd. (Trade

name: TEM-26SS). The screw diameter is 26 mm and its aspect ratio (Length/diameter) is 40. In

the extruder, the fed mixture is melt-blended under a barrel and a die temperature of 200°C but

the hopper zone is cooled with a cold-water sleeve. The discharge rate is set at 20 kg/hr.

Extrudate from the die is shaped into a cylindrical filament, led into a cool water bath, and

pelletized.

142


A tandem foam extrusion system similar to those employed by the foam manufacturing industry

is used to foam the samples. The tandem foam system comprises of two single-screw extruder

barrels. The first extruder is a 5 horsepower (hp) Brabender 05-25-00 consisting of a mixing

screw with a diameter of 19 mm and an aspect ratio of 30. The second extruder is a 15 hp Killion

KN-150 consisting of a mixing screw with a diameter of 38.1 mm and an aspect ratio of 30. Fig.

7.1 gives a schematic of the configuration of the extrusion system. A metered amount of CO2 gas

is injected into the melt through an injection port positioned at the first extruder. 20

First, we feed pellets of the fibrillar blend (either PP-homopolymer/PTFE or PP-

copolymer/PTFE) into the first extruder barrel through the hopper. This barrel is maintained at

200°C, a temperature above the melting temperature of the matrix polymer but below that of

PTFE fibrils. The matrix polymer in the samples melts completely in the first extruder due to the

temperature as well as the screw motion which causes shear heating. Then, we inject 10 wt%

CO2 into the first extruder barrel at a constant flow rate using a syringe pump. The high shear

and high pressure caused by the rotating screw inside the first extruder barrel facilitates the

dissolution of CO2 in the melt through convective diffusion. This homogeneous solution enters

the second extrusion barrel which is maintained at temperatures below the melting temperature

of the matrix polymer. Consequently, it is in this second extrusion barrel where crystallization is

initiated in the samples. Foaming of the polymer melt occurs at the die exit where the

polymer/CO2 solution is subject to rapid depressurization resulting in the gas to undergo phase

separation. 14

We employ a brass capillary die comprising of a circular pinhole with a diameter of

1.2 mm and a channel length of 10 mm. The temperature of the second extruder barrel and the

die is brought down and the foamed samples collected at each set temperature only after the

system temperature has equilibrated. The temperature range we study is 165 to 120°C.


To observe the morphology development of PTFE after blending with the matrix polymer, the

blend is dissolved in xylene and the residual PTFE examined using a scanning electron

microscope (SEM, JEOL 6060). To characterize the dispersion of the PTFE fibrils in the matrix,

the surface of cryogenically fractured sample of the blend is subjected to a solvent vapour

etching process using xylene. Samples are exposed to xylene vapors for 2 h at 65°C. This type of

143

treatment preferentially etches the matrix polymer which is soluble in xylene, leading to bringing

up the PTFE domains and increasing differentiation in observation.

To observe the morphology of the foamed samples, the samples are dipped in liquid nitrogen and


before observation under SEM. Final images are recorded from randomly selected areas at the

magnifications indicated in the SEM micrographs. For each foamed sample, the foam density is

determined using the water displacement method according to ASTM-D792. Additionally, the

number of bubbles per unit volume, called the cell density, is also estimated from the SEM

images of the foams using the method described elsewhere. 21

7.3.5 Oil absorption test

The oil uptake capacity, defined as the mass in grams of the oil absorbed per gram of the foam at

equilibrium, is measured for various petroleum products including octane, gasoline, diesel, light

crude oil, heavy crude oil and kerosene. In a typical test, a dry open-cell foam sample of known

mass is immersed completely into the respective petroleum product for 24 h using an external

force. Subsequently, the foam is taken out and rapidly weighed to minimize the effect of

evaporation of the absorbate. The absorbate uptake capacity of the foam is calculated as (mass

after saturation in oil - initial mass of foam) / (initial mass of foam). The petroleum product is

then extracted from the foam with the aid of a vacuum pump. This process is repeated ten times

and the average value along with the standard deviation is calculated.

To measure the kinetics of oil absorption, dry foam samples of known masses are immersed

completely using an external force into the petroleum product for different time periods after

which the foam samples are removed, weighed, and their uptake capacities at their respective

times calculated.

7.3.6 Mechanical properties of the foam

To test the reusability of the foams, the foams are subjected to a cyclic compression test as a

controlled analogue for mechanical squeezing for oil recovery. Stress-strain measurements are

performed for 10 cycles on an Instron 5848 microtester equipped with a 50 N load cell. The

maximum strain is set to 60% and the compression rate is set at 0.25 mm/min.

144

7.3.7 Contact angle determination

The contact angles of droplets (10-30 µL) of deionized water or gasoline on the cross-section of

the cylindrical open-cell foams are obtained using a custom-built contact angle measurement

system under room temperature and pressure. The details of the system used can be found

elsewhere. 22

After capturing the boundary profile of the droplet, ImageJ software is used to

determine the contact angles.

7.3.8 Ultrasound treatment

The extruded foam filament is cut into cylinders of about 3 mm length and subjected to high

intensity ultrasound irradiation using a commercial ultrasonic source (Sonics VCX750) with a

probe diameter of 13 mm, operating at a frequency of 20 kHz and a maximum power of 750 W.

The ultrasonic probe is positioned about 1 mm above the foam held in place using a rubber

fixture in a bath filled with distilled water. Ultrasound pulses with an on/off ratio of 3:3 are

applied to the foam with an ultrasound power intensity of 750 W for 10 min.

145


7.4.1 Preparation of open-cell foams of PP/PTFE fibrillar blends

Fig. 7.1 Schematic showing the preparation of open-cell foams from PP-homopolymer/PTFE

fibrillar blends: a) PP-homopolymer is blended with PTFE in the ratio 97 wt% and 3 wt%,

respectively in a twin-screw extruder; b) SEM micrograph of the twin-screw extruded blend after

solvent-vapour etching using xylene. The PTFE exists as well-dispersed, high-aspect-ratio fibrils

in the blend; c) PTFE fibrils after selective removal of PP from the fibrillar blend using xylene.

The PTFE fibrils form an entangled mesh-like structure; d) The fibrillar blend is foamed in a

tandem single-screw foam extrusion system. 14

The temperature of the first extruder is

maintained above the Tm of PP but below that of PTFE. 10 wt% CO2 is injected in the first

extruder using a syringe pump; e) Photograph of a continuous open-cell foam filament produced

by the foam extrusion system. The open-cell foam filament is produced with a diameter of about

4 mm and can be produced at a rate of 2 kg/h. The continuous nature of the process allows the

foam filament to be of a very large length exceeding hundreds of meters. The SEM micrograph

(dashed outline) shows the morphology of the extruded foam filament. The foamed filament

‗skin‘ layer does not develop a porous structure and is 2 ± 0.6 µm thick.

146

We prepare open-cell foams of PP and PTFE fibrillar blends. The procedure for fabricating the

open-cell foams of PP-homopolymer/PTFE is illustrated schematically in Fig. 7.1. First, PP-

homopolymer/PTFE (97/3 wt%) is blended in a twin-screw extruder at 200°C (Fig. 7.1a). The

SEM image in Fig. 7.1b reveals that the PTFE deforms into high aspect ratio fibrils during

blending, as reported previously. 18, 19, 23, 24

The image is obtained after solvent-vapour etching

using xylene. Fig. 7.1c shows the SEM micrograph of PTFE fibrils after the selective removal of

PP from the PP/PTFE blend with xylene. The PTFE fibrils form an entangled mesh with fibrils

of a high aspect-ratio (diameter < 500 nm and length > 100 µm, so the aspect ratio is > 200). Fig.

7.1d shows a schematic of the tandem foam extrusion system used to foam the PP/PTFE fibrillar

blend. A photograph of the extrusion system can be found in Appendix B, Fig. S1. The system

is able to extrude a foam filament of 4 mm diameter at a rate of 2 kg/h. In the first extruder, the

blend is subjected to a temperature of 200°C to melt the PP and 10 wt% CO2 is injected and

dissolved in the melt. In the second extruder, the PP/PTFE/CO2 system is subjected to

temperatures below the melting temperature of PP, so that the PP is able to develop crystalline

heterogeneities in this extruder. Specifically, the occurrence of flow-induced shish-kebab 25

type

structures during extrusion of semi-crystalline polymers is reported. We also observe such

crystalline morphologies in the extruded PP/PTFE (see Appendix B, Fig. S2). Upon foaming,

cell opening is initiated when a heterogeneous-melt structure 26, 27, 28

characterized by well

dispersed arrays of rigid and soft segments exists in the sample; the rigid segments preserve the

strut while the soft segment rupture upon foam expansion. 29,

30

The crystalline heterogeneities of

PP that develop in the second extruder are rigid relative to the molten matrix and form the

structural heterogeneities needed to achieve high open-cell contents. Furthermore, the dissolution

of 10 wt% CO2 in PP/PTFE sample plasticizes the molten sections, thereby increasing the

chances of opening the soft segments of the bubble walls. 31, 27

There exists a processing

temperature window within which the crystalline heterogeneities will yield foams with the

highest open-cell content. 32

The optimum temperature for preparing foams of PP-

homopolymer/PTFE with the highest open-cell content (i.e. open cell content of about 97.7%) is

identified to be 150°C. While the bulk density of the unfoamed fibrillar blend (ρunfoamed) is

measured to be about 0.92 g cm-3

, the bulk density after foam extrusion (ρfoam) at 150°C drops to

about 0.07 g cm-3

. Fig. 7.1e shows a photograph of the extruded open-cell foam filament and the

blow-up image shows the SEM of the foam cross-section which exhibits a highly porous,

interconnected, 3D framework with struts that delineate cavities (or cells). The number of cells

147

per unit volume, or cell density, is estimated from the SEM images to be between 107 to 10

8 cells

cm-3

. The cell sizes are in the range of 50 to 500 µm. The void fraction, (Vf = 1 – ρfoam / ρunfoamed),

is calculated to be about 0.92. It is expected that the large void fraction of the highly porous,

largely expanded, interconnected structure makes available a large 3D space for oil storage.

7.4.2 Superhydrophobicity of the open-cell foams

Superhydrophobic materials are difficult to wet with water and are arbitrarily defined to exhibit a

contact angle, θ, >150°. 33

In order to study the hydrophobicity of the extruded open-cell foam,

the foam filament is cryogenically fractured into cylinders for the contact angle measurements,

as shown in Fig. 7.2a. The apparent contact angle of the cross-sectional surface of the foam with

water is shown in Fig. 7.2b and is determined to be 160° ± 0.9°. In order to compare the

oleophilicity (i.e., the affinity for oil) and superhydrophobicity of the open-cell foam, a

photograph is shown in Fig. 7.2c where a water droplet forms a large contact angle on the foam,

and a drop of gasoline (~ 12 µL) labelled with red dye is absorbed rapidly by the same foam

sample. The time for absorbing a drop (~ 12 µL) of gasoline is only 0.73 ± 0.08 s (Fig. 7.2d).

Such rapid absorption kinetics of the open-cell foam is attributed to its highly porous structure,

oleophilicity, capillary action and the low viscosity of the gasoline. After absorbing the oil, the

open-cell foam continues to float on the water surface due to its low density and

superhydrophobicity, thereby making it easy to collect the foams and extract the oil. One striking

observation is the change in appearance of the open-cell foam upon immersion in water using an

external force. Fig. 7.2e shows that when the white open-cell foam is immersed in water and

viewed at a glancing angle, it appears to have a silver mirror-like surface. This occurs because of

the Cassie-Baxter 34

non-wetting behaviour which creates a continuous air gap at the interface

between the hydrophobic open-cell foam surface and water. The effect is strongest in the cross-

sectional area, possibly because of a stronger hydrophobic surface than the foam skin, since a

higher degree of roughness exhibits more hydrophobicity. 35

It is evident that the open-cell foam

structure naturally generates sufficient roughness to impart superhydrophobicity.

The thermal stability of the open-cell foam is studied and presented in Fig. 7.2f-g. The foam is

subjected to extreme temperature environments including freezing in liquid nitrogen at -196°C

for 1 h and heating to 120°C for 1 h. No change in the foam morphology is observed after these

treatments and the open-cell foam continues to retain its superhydrophobic and oleophilic nature.

148

Fig. 7.2 Superhydrophobic properties of the PP-homopolymer/PTFE open-cell foams: a) Open-

cell foam filament is cut into cylindrical shapes for testing physical properties of the open-cell

foam. The diameter of the cylindrical foam sample is about 4 mm and lengths about 2 to 5 mm;

b) Measurement of contact angle of a water droplet placed on the cross-section of the foam. The

water droplet forms a contact angle of 160° ± 0.9° with foam; c) Photograph of a water drop on

the cross-sectional surface of the superhydrophobic open-cell foam. A drop of gasoline labelled

with a red dye is readily absorbed into the open-cell foam; d) Drop of gasoline taken up by the

oil in 0.73 ± 0.08 s as observed from a video recording; e) Photograph of the open-cell foam

149

immersed in water creates silver mirror-like surface when looked at from a glancing angle;

Photograph of water droplet and corresponding water contact angle after subjecting the sample

to: f) -196°C and g) 120°C for 1 h. Superhydrophobicity of the open-cell foam is preserved after

these treatments.

When the foam is subjected to a temperature of 125°C for 1 h, the foam morphology is not

retained as the material begins to soften and distort, sacrificing the superhydrophobic

characteristic of the foam. Nonetheless, the foam has a broad service temperature range of -196

to 120°C, a feature which may make the foam attractive for industrial applications beyond oil-

spill cleanup and recovery.

7.4.3 Oil uptake study

PP-based oil absorbents such as nonwoven PP fibers or melt-blown PP pads are widely used for

oil spill cleaning. However, these absorbents exhibit low oil uptake capacities. 36, 37, 38, 39, 40, 41

Our PP-based open-cell foams are an ideal substitute for the sorption of oils and other liquid-

phase organic pollutants. When a cylindrical open-cell foam sample, such as the one shown in

Fig. 7.2a, is placed on the surface of a gasoline-water mixture, the gasoline layer is readily and

selectively taken up by the foam. Fig. 7.3b shows the time lapse images of the video of oil

uptake (See Appendix B, V2: Oil uptake by extruded open-cell foam). To differentiate the

gasoline from water, the gasoline is labelled with a red dye.

The uptake capacity of the open-cell foam is expressed in terms of the mass in grams of oil

absorbed / mass in gram of foam (g g-1

). We study the uptake capacity of the open-cell foams for

various petroleum products including octane, gasoline, diesel, light crude oil, heavy crude oil and

kerosene. First, the open-cell foam is saturated with one of these petroleum products and the gain

in mass is determined to calculate the uptake capacity. Subsequently, the petroleum product is

extracted with the aid of a vacuum pump. The petroleum product is recovered, and the open-cell

foam regenerated for a subsequent cycle. This process is repeated ten times for all studied

petroleum products. The results are shown in Fig. 7.3b and the corresponding data is presented

in Table 7.1. The data presented in Table 7.1 is also compared with literature values of uptake

capacities obtained for two PP-based materials commonly used for oil spill cleanup, namely,

150

a b

Fig. 7.3 Uptake of various petroleum products by the open-cell polymer foams: a) Few drops of

gasoline labelled with a red dye for clear differentiation with water. The Open-cell foam readily,

and selectively absorbs the gasoline; b) Uptake capacity of open-cell polymer foam for various

petroleum products in terms of mass in grams of oil per mass in grams of the open-cell foam (g

g-1

). The oil is removed from the open-cell foam using vacuum pump and reused over ten cycles

to study the reusability of the foams.

nonwoven PP fibers and melt-blown PP pads. The physical characteristics and specifications of

these PP-based materials can be found elsewhere. 39, 40, 42

With the exception of heavy crude oil,

and kerosene, the open-cell foam outperforms these PP-based absorbents for the uptake of all

other studied petroleum products. In general, the uptake capacity of the open-cell foam varies

from oil to oil and ranges from around 5 to 24 g g-1

of foam. Furthermore, over the ten

absorption/vacuum extraction cycles, the open-cell foam demonstrates consistent performance.

Within experimental error, the differences in the measured uptake capacity does not change

substantially as reflected from a maximum standard deviation of 1.35 g g-1

of foam. This result

indicates that a simple vacuum extraction after saturation of the foam with the petroleum product

is sufficient to recover active open-cell foam for a subsequent absorption cycle.

1 2 3 4 5 6 7 8 9 100

5

10

15

20

25

30

35

Up

take C

ap

acit

y (

g g

-1)

Cycle

Octane Diesel

Gasoline Heavy Crude Oil

Light Crude Oil Kerosene

Gasoline labelled with dye Open-cell foam

151

Table 7.1 Measured uptake capacities of the open-cell foam for various petroleum products and

organic liquids a, b

a The absorption time was kept constant at 24 h;

b the uptake capacity of the foam is reported as g

g-1

of foam; c Average of uptake capacities are calculated for the ten cycles;

d σ = standard

deviation; e Uptake capacities of various commercially available PP-based products from

literature for comparison purposes; f Uptake capacity of commercial melt-blown PP pad;

g Uptake

capacity of commercial nonwoven PP fibers.

Cycle Octane Gasoline Light Crude Oil Diesel Heavy Crude Oil Kerosene

1 26.1 24.2 14.7 12.6 8.4 6.6

2 25.1 25.4 14.4 11.2 8.2 5.9

3 24.8 22.11 13.7 11.4 8.9 5.5

4 23.5 21.91 14.2 12.2 9.1 6.1

5 21.8 21.4 14.8 11.8 8.2 5.1

6 24.1 20.9 15.0 11.4 8.6 4.9

7 23.8 22.5 14.9 11.0 8.1 5.8

8 22.9 21.8 14.3 10.5 8.6 5.1

9 24.1 23.3 15.8 10.1 8.2 3.9

10 23.2 23.0 14.1 10.9 8.3 4.8

Average c 23.9 22.6 14.6 11.3 8.5 5.4

σ d 1.22 1.35 0.59 0.75 0.33 0.76

Literature e - 9.2

40, f 8.3 39, g 9.1

40, f 9.1 39, g 10

42, g

152

7.4.4 Effect of mechanical squeezing for oil recovery

Fig. 7.4 Mechanical behaviour of PP-homopolymer/PTFE and PP-copolymer/PTFE open-cell

foam: a) Stress-strain curves of the open-cell foams made from PP-homopolymer/PTFE in the

process of ten cyclic compressions. The insert shows the morphology of the PP-

homopolymer/PTFE open-cell foam; b) Stress-strain curves of the open-cell foams made from

poly(propylene-co-ethylene)/PTFE in the process of ten cyclic compressions. The insert shows

the morphology of the PP-copolymer/PTFE open-cell foam; c) Permanent strain undergone by

the PP-homopolymer/PTFE open-cell foam and PP-copolymer/PTFE open-cell foam as a

function of cycle number. PP-homopolymer/PTFE open-cell foams undergo a permanent

deformation of 30% while PP-copolymer/PTFE open-cell foams undergo a permanent

deformation of 10% over the ten compressive cycles.

0 10 20 30 40 50 60 70-5

0

5

10

15

20

25

30

35

40

20 22 24

4.8

5.4

6.0

Co

mp

ressiv

e S

tress (

KP

a)

Compressive Strain (%)

109

87

6

5

4

3

2

10th

cycle

Co

mp

ress

ive

Str

ess

(K

Pa)


cycle sequence

1st cycle

b

0 10 20 30 40 50 60 70-10

0

10

20

30

40

50

60

70

28 32 36

2

3

4

1098

76

5

4

3C

om

pre

ssiv

e S

tress (

KP

a)


2

10th

cycle

Co

mp

ressiv

e S

tress (

KP

a)


cycle sequence

1st

cycle

a

1 2 3 4 5 6 7 8 9 100

5

10

15

20

25

30

Perm

an

en

t S

train

(%

)

Cycle

Homopolymer

Random copolymer

c

153

The previous section demonstrates that vacuum extraction is effective in removing the absorbed

petroleum product from the open-cell foam, to regenerate the open-cell foam for a subsequent

absorption cycle. However, vacuum extraction is not ideal for oil recovery, in particular, for low

viscosity petroleum products which are volatile organic liquids. For them, a sorption-heating-

condensation 43

or a mechanical ―squeezing‖ process needs to be employed to maximize

petroleum product recovery. From a practical standpoint, the mechanical squeezing of the foam

for oil extraction is a more cost-effective and scalable methodology for oil recovery. Thus, we

investigate the mechanical properties of the PP open-cell foams by cyclic compression

measurements as a controlled analogue of mechanical ―squeezing‖ needed for oil recovery. Fig.

7.4a shows cyclic compression stress-strain curves of the PP-homopolymer/PTFE open-cell

foam for ten compressive cycles. The inset in Fig. 7.4a shows the morphology of the open-cell

foam. A typical stress curve increases, reaches a plateau and then increases with a gradually

increasing slope until a maximum strain of 60%. The foam is able to be compressed to the

maximum strain of 60% at a relatively low stress 70 kPa, owing to the low density of the open-

cell foam. Upon removal of the compressive force, the foam shows poor recovery to its original

shape. After the first compressive cycle, the foam undergoes a permanent strain of about 20%.

This poor compressive fatigue behaviour is attributed to the brittle nature of the PP-

homopolymer matrix used in the PP- homopolymer/PTFE fibrillar blend. The brittle

characteristics are a consequence of the high degree of crystallinity in linear isotactic PP-

homopolymers which occurs due to the stereoregularity of the polymer chains. 44

The impact

properties of PP-homopolymers can be improved by copolymerization with ethylene. The

random insertion of ethylene molecules in the PP backbone disrupts the regular, repeating

structure and reduces the crystallinity compered to isotactic PP-homopolymers. 45

This reduction

in copolymer crystallinity is responsible for the modified physical properties, such as reduced

stiffness, higher impact resistance, and better elastic properties than PP-homopolymers.

The inadequate compressive fatigue of the PP-homopolymer/PTFE open-cell foam limits its

reusability for oil uptake applications because mechanical squeezing for oil recovery will

damage the morphology of the foam. When the matrix polymer used to prepare the open-cell

foams is replaced from the brittle PP-homopolymer to a poly(propylene-co-ethylene) random

copolymer (PP-copolymer) known to exhibit better elastic properties, a noticeable improvement

in the compressive fatigue behaviour of the open-cell foam is seen. Fig. 7.4b shows the stress-

154

strain behaviour of open-cell foams prepared from PP-copolymer/PTFE fibrillar blend and the

insert shows the morphology of the obtained open-cell foams. Upon 60% compression, PP-

copolymer/PTFE open-cell foam reaches a maximum stress of 45 kPa, which is lower than the

maximum stress of the PP-homopolymer/PTFE open-cell foam (70 kPa), indicating that the PP-

copolymer/PTFE open-cell foam is softer compared to the PP-homopolymer/PTFE open-cell

foam. Fig. 7.4c compares the permanent strain undergone by the PP-homopolymer/PTFE open-

cell foam sample and the permanent strain undergone by the PP-copolymer/PTFE open-cell foam

sample, as a function of the compressive cycle number. The PP-copolymer/PTFE foam

undergoes a smaller percentage permanent strain over the ten compressive cycles (about 10%)

than the PP-homopolymer/PTFE foam (about 30%). This indicates that the PP-copolymer/PTFE

foam demonstrates better recovery to its original shape after a compressive cycle than the PP-

homopolymer/PTFE open-cell foam and has a superior reusability after mechanical ‗squeezing‘

required for extracting oil from the open-cell foam. The PP-copolymer/PTFE foam demonstrates

similar uptake capacities, morphological characteristics and can be prepared under similar

processing conditions as the PP-homopolymer/PTFE open-cell foams (See Appendix B: Fig. S3,

Table S1, and Table S2).

155

7.4.5 Kinetics of oil uptake

Fig. 7.5 Enhancing kinetics of oil absorption of PP-homopolymer/PTFE open-cell foams: a)

SEM micrographs of the ‗skin‘ layer of the extruded open-cell foams, i) before high intensity

ultrasound treatment, ii) after high intensity ultrasound treatment; b) Gasoline uptake as a

function of time for the open-cell foam before and after ultrasound treatment; c) Diesel uptake as

a function of time for the open-cell foam before and after ultrasound treatment.

0 500 1000 1500 2000 25000

3

6

9

12U

pta

ke C

ap

acit

y (

g g

-1)

Time (s)

After ultrasound treatment

Before ultrasound treatment

Second order fit

c

0 20 40 600

5

10

15

20

25

Up

take C

ap

acit

y (

g g

-1)

Time (s)

After ultrasound treatment

Before ultrasound treatment

Second order fit

b

156

The extruded open-cell foams have a thin, impervious ‗skin‘ layer which offers low oil

permeability. The uptake of oil occurs predominantly from the open-cell core which becomes

exposed when the filament is cut in cross-sections. In order to increase the porosity of the

unfoamed skin layer and enhance the rate of oil uptake, high intensity ultrasound has been

successfully employed previously. 15, 16, 17,

46, 47, 48

Such morphological changes occur due to a

phenomenon known as ultrasonic cavitation 47

near solid surfaces in a liquid. Ultrasonic

cavitation is the formation and violent collapse of small bubbles in the liquid as a result of

pressure changes in the medium. The collapse of bubbles near solid surfaces can generate fast-

moving liquid microjets directed towards the surface. The impact of the jets on the surface can

create localized damage. We employ high intensity ultrasound to increase the porosity of the

impervious ‗skin‘ layer of the extruded foam. Fig. 7.5a shows the SEM micrographs of the PP-

homopolymer/PTFE open-cell foam skin layer before ultrasound irradiation treatment and after

the ultrasound irradiation treatment. Prior to treatment, the surface shows a relatively non-porous

structure. However, post-treated samples exhibit pronounced surface porosities. While ultrasonic

irradiation treatment has been reported to cause site-specific chain scission in polymers by

cleaving weak peroxide, 49

azo 50

and coordination bonds, 51

the chain scission rate coefficients

for PP are low. 52

Measurements of the complex viscosity of the open-cell PP/PTFE foam before

and after the ultrasonic irradiation treatment reveal that ultrasonic irradiation does not affect the

complex viscosity of PP/PTFE and chemical changes such as chain scission and degradation can

therefore be ignored.

Fig. 7.5b-c show the uptake rate of gasoline and the uptake rate of diesel, respectively, for the

open-cell foams before ultrasound treatment and after ultrasound treatment. The open-cell foam

takes a shorter time to saturate with gasoline than to saturate with diesel, both before and after

ultrasound irradiation treatment. The more rapid uptake of gasoline compared to diesel is due to

its lower viscosity than diesel and is able to infiltrate into the open-cell foam structure more

easily. After ultrasound irradiation treatment, the open-cell foams demonstrate more rapid oil

uptake rates for both gasoline and diesel compared to before the ultrasound irradiation treatment

due to the increased permeability of oil through the opened ‗skin‘ layer. The uptake kinetics can

be described by the second order equation:

(7.1)

157

where is the equilibrium uptake capacity corresponding to when the foam is completely

saturated, is the uptake capacity at time , and is the uptake capacity rate constant. The

curves of best-fit are included in Fig. 7.5b-c and the corresponding fitting values of and for

the two petroleum products are presented in Table 7.2. The coefficients of determination R2

, for

all curves are > 0.99 indicating excellent agreement between the second order uptake kinetics

model and the experimental data.

Table 7.2 Fitting parameters for uptake kinetics of gasoline and diesel.

7.5 Conclusion

Superior superhydrophobic and oleophilic open-cell foams of PP/PTFE (97/3 wt%) fibrillar

blends, are prepared. The open-cell foams display significant improvements in oil uptake

capacities compared to commercially available PP-based materials currently used for oil-spill

cleanup applications, such as nonwoven PP fibers and melt-blown PP pads. These open-cell

foams exhibit superhydrophobicity and oleophilicity which results in a high oil/water selectivity.

The open-cell foams have a low density of about 0.07 g cm-3

which makes the foams highly

buoyant in water. The open-cell foams also have a high void fraction of 0.92 which makes

available a large 3D space for oil storage. The open-cell foams offer robust stability at

temperatures ranging from -196 to 120°C. Subjecting the open-cell foams to these temperature

extremes for prolonged periods does not influence the superhydrophobic and oleophilic

properties of the foams. The kinetics of oil uptake can be enhanced by increasing the ‗skin‘ layer

Petroleum

product

Foam

treatment Qe (g g

-1) k × 10

3 (g g

-1s

-1)

Gasoline Unsonicated 23.8 6

Gasoline Sonicated 23.2 19

Diesel Unsonicated 11.6 0.9

Diesel Sonicated 11.9 2.7

158

porosity of the extruded foams using high intensity ultrasound irradiation treatment. If the PP

matrix in the PP/PTFE fibrillar blend is a PP-homopolymer, the reusability of the foams is poor

because PP-homopolymer/PTFE undergoes a large permanent deformation after mechanical

―squeezing‖ to extract the oil (30% permanent deformation over ten compression cycles). This

damages the open-cell morphology and consequently, the superhydrophobic and oleophilic

properties. If the PP matrix in the PP/PTFE fibrillar blend is a PP random-copolymer, the

reusability of the foams is drastically improved because the resultant PP-copolymer/PTFE open-

cell foams undergo a small permanent deformation after mechanical squeezing to extract the oil

(10% permanent deformation over ten compressive cycles). The open-cell foams are prepared

using conventional foam extrusion equipment and low-cost, commercially-available raw

materials. Consequently, the fabrication process is easy to scale-up. The open-cell foams are

technologically promising for applications such as oil-spill cleanup, organic pollutant removal

and field water remediation.

159

References

1. Kontovas, C. A.; Psaraftis, H. N.; Ventikos, N. P. An Empirical Analysis of IOPCF Oil Spill

Cost Data. Mar. Pollut. Bull. 2010, 60, 1455-1466.

2. Fingas, M. The Basics of Oil Spill Cleanup, 3rd ed.; CRC Press, 2012.

3. Choi, H. M.; Cloud, R. M. Natural Sorbents in Oil Spill Cleanup. Environ. Sci. Technol.

1992, 26, 772-776.

4. Radetić, M. M.; Jocić, D. M.; Jovančić, P. M.; Petrović, Z. L.; Thomas, H. F. Recycled

Wool-Based Nonwoven Material as an Oil Sorbent. Environ. Sci. Technol. 2003, 37, 1008-

1012.

5. Annunciado, T. R.; Sydenstricker, T. H. D.; Amico, S. C. Experimental Investigation of

Various Vegetable Fibers as Sorbent Materials for Oil Spills. Mar. Pollut. Bull. 2005, 50,

1940-1346.

6. Duong, H. T. T.; Burford, R. P. Effect of Foam Density, Oil Viscosity, and Temperature on

Oil Sorption Behaviour of Polyurethane. K. Appl. Polym. Sci. 2006, 99, 360-367.

7. Zhu, Q.; Pan, Q.; Liu, F. Facile Removal and Collection of Oils from Water Surfaces through

Superhydrophobic and Superoleophilic Sponges. J. Phys. Chem. C 2011, 115, 17464-17470.

8. Zhu, Q.; Chu, Y.; Wang, Z.; Chen, N.; Lin, L.; Liu, F.; Pan, Q. Robust Superhydrophobic

Polyurethane Sponge as a Highly Reusable Oil-Absorption Material. J. Mater. Chem. A

2013, 1, 5386-5393.

9. Camilli; L.; Pisani, C.; Gautron, E.; Scarselli, M.; Castrucci, P.; D'Orazio, F.; Passacantando,

M.; Moscone, D.; Crescenzi, M. D. A Three-Dimensional Carbon Nanotube Network for

Water Treatment. Nanotechnology 2014, 25, 1-7.

10. Bi, H.; Xie, X.; Yin, K.; Zhou, Y.; Wan, S.; He, L.; Xu, F.; Banhart, F.; Sun, L.; Ruoff, R. S.

160

Spongy Graphene as a Highly Efficient and Recyclable Sorbent for Oils and Organic

Solvents. Adv. Funct. Mater. 2012, 22, 4421-4425.

11. Lei, W.; Portehault, D.; Liu, D.; Qin, S.; Chen, Y. Porous Boron Nitride Nanosheets for

Effective Water Cleaning. Nat. Commun. 2013, 4, 1-7.

12. Yang, C.; Kaipa, U.; Mather, Q. Z.; Wang, X.; Nesterov, V.; Venero, A. F.; Omary, M. A.

Fluorous Metal–Organic Frameworks with Superior Adsorption and Hydrophobic Properties

toward Oil Spill Cleanup and Hydrocarbon Storage. J. Am. Chem. Soc. 2011, 133, 18094-

18097.

13. Hayase, G.; Kanamori, K.; Fukuchi, M.; Kaji, H.; Nakanishi, K. Facile Synthesis of

Marshmallow-Like Macroporous Gels Usable Under Harsh Conditions for the Separation of

Oil and Water. Angew. Chem. 2013, 52, 1986-1989.

14. Park, C. B.; Suh, N. P. Filamentary Extrusion of Microcellular Polymers. Polym. Eng. Sci.

1996, 36 (1), 34.

15. Wang, X.; Li, W.; Kumar, V. A Method for Solvent-Free Fabrication of Porous Polymer

using Solid-State Foaming and Ultrasound for Tissue Engineering Applications. Biomaterials

2006, 27, 1924-1929.

16. Guo, G.; Ma, Q.; Zhao, B.; Zhang, D. Ultrasound-Assisted Permeability Improvement and

Acoustic Characterization for Solid-State Fabricated PLA Foams. Ultrason. Sonochem. 2013,

20, 137-143.

17. Watson, N. J.; Johal, R. K.; Glover, Z.; Reinwald, Y.; White, L. J.; Ghaemmaghami, A. M.;

Morgan, S. P.; Rose, F. R. A. J.; Povey, M. J. W.; Parker, N. G. Post-Processing of Polymer

Foam Tissue Scaffolds with High Power Ultrasound: A Route to Increased Pore

Interconnectivity, Pore Size and Fluid Transport. Mater. Sci.Eng. C 2013, 33, 4825-4832.

18. Ali, M. A.; Nobukawa, S.; Yamaguchi, M. Morphology Development of

Polytetrafluoroethylene in a Polypropylene Melt (IUPAC Technical Report). Pure Appl.

Chem. 2011, 83, 1819-1830.

161

19. Rizvi, A.; Tabatabaei, A.; Barzegari, R.; Mahmood, H.; Park, C. B. In-situ Fibrillation of

CO2-Philc Polymers: Sustainable Route to Polymeric Foams in a Continuous Process.

Polymer 2013, 54, 4645–4652.

20. Park, C. B.; Baldwin, D. F.; Suh, N. P. Effect of the Pressure Drop Rate on Cell Nucleation

in Continuous Processing of Microcellular Polymers. Polym. Eng. Sci. 1995, 35, 432-440.

21. Xu, X.; Park, C. B.; Xu, D.; Pop-Iliev, R. Effect of die geometry on cell nucleation of PS

foams blown with CO2. Polym. Eng. Sci. 2003, 43 (7), 1378-1390.

22. Li, Y. G.; Park, C. B.; Li, H. B.; Wang, J. Measurement of the PVT Property of PP/CO2

Solution. Fluid Phase Equilibr. 2008, 270, 15-22.

23. Ali, M.; Okamoto, K.; Yamaguchi, M.; Kasai, T.; Koshirai, A. Rheological Properties for

Polypropylene Modified by Polytetrafluoroethylene. J. Polym. Sci. Part B: Polym. Phys.

2009, 47, 2008-2014.

24. Rizvi, A.; Park, C. B.; Yamaguchi, M. Resin Composition Foam and Method for Producing

Same. WO2013137301, March 13, 2013.

25. Binsbergen, F. L. Orientation-Induced Nucleation in Polymer Crystallization. Nature 1966,

211, 516 - 517.

26. Lee, P. C.; Wang, J.; Park, C. B. Extruded Open-cell Foams Using two Semicrystalline

Polymers with different Crystallization Temperatures. Ind. Eng. Chem. Res. 2006, 45, 175-

181.

27. Lee, P. C.; Wang, J.; Park, C. B. Extrusion of Microcellular Open-Cell LDPE-Based Sheet

Foams. J. Appl. Polym. Sci. 2006, 102, 3376-3384.

28. Harikrishnan, G.; Patro, T. U.; Khakhar, D. V. Polyurethane Foam−Clay

Nanocomposites: Nanoclays as Cell Openers. Ind. Eng. Chem. Res. 2006, 45, 7126-7134.

29. Lee, S. T.; Ramesh, N. S. Flexible Polyurethane Foams. In Polymeric Foams, mechanisms

162

and materials; Lee, S. T., Ramesh, N. S., Eds.; CRC Press: Boca Raton, 2004; pp 173-251.

30. Keller, A. Unusual Orientation Phenomena in Polyethylene Interpreted in terms of the

Morphology. J. Polym. Chem. 1955, 15, 31-49.

31. Lee, P. C.; Naguib, H. E.; Park, C. B.; Wang, J. Increase of Open-Cell Content by

Plasticizing Soft Regions with Secondary Blowing Agent. Polym. Eng. Sci. 2005, 45, 1445-

1451.

32. Park, C. B.; Padareva, V.; Lee, P. C.; Naguib, H. E. Extruded Open-Celled LDPE-Based

Foams using Non-Homogeneous Melt Structure. J. Polym. Eng. 2005, 25, 239-260.

33. Wang, S.; Jiang, L. Definition of Superhydrophobic States. Adv. Mater. 2007, 19, 3423-3424.

34. Cassie, A. B. D.; Baxter, S. Wettability of Porous Surfaces. Trans. Faraday Soc. 1944, 40,

546-551.

35. Li, X. M.; Reinhoudt, D.; Crego-Calama, M. What Do We Need for a Superhydrophobic

Surface? A Review on the Recent Progress in the Preparation of Superhydrophobic Surfaces.

Chem. Soc. Rev. 2007, 36, 1350-1368.

36. Adebajo, M. O.; Frost, R. L.; Kloprogge, J. T.; Carmody, O.; Kokot, S. Porous Materials for

Oil Spill Cleanup: A Review of Synthesis and Absorbing Properties. J. Porous Mater. 2003,

10, 159– 170.

37. Wei, Q. F.; Mather, R. R.; Fotheringham, A. F.; Yang, R. D. Evaluation of Nonwoven

Polypropylene Oil Sorbents in Marine Oil-Spill Recovery. Mar. Pollut. Bull. 2003, 46, 780–

783.

38. Teas, C.; Kalligeros, S.; Zanikos, F.; Stournas, S.; Lois, E.; Anastopoulos, G. Investigation of

the Effectiveness of Absorbent Materials in Oil Spills Clean Up. Desalination 2001, 140,

259– 264.

39. Bayat, A.; Aghamiri, S. F.; Moheb, A.; Vakili-Nezhaad, G. R. Oil Spill Cleanup from Sea

163

Water by Sorbent Materials. Chem. Eng. Technol. 2005, 28, 1525-1528.

40. Chung, T. C.; Yuan, X. P. Methods and Materials for Hydrocarbon Recovery. WO

2012024614 A2, February 23, 2012.

41. Zhao, J.; Xiao; C.; Xu, N. Evaluation of Polypropylene and Poly(butylmethacrylate-co-

hydroxyethylmethacrylate) Nonwoven Material as Oil Absorbent. Environ. Sci. Pollut. Res.

2013, 20, 4137-4145.

42. Pan, Y.; Shi, K.; Peng, C.; Wang, W.; Liu, Z. J. X. Evaluation of Hydrophobic Polyvinyl-

Alcohol Formaldehyde Sponges as Absorbents for Oil Spill. ACS Appl. Mater. Interfaces

2014, 6, 8651–8659.

43. Wu, Z. Y.; Li, C.; Liang, H. W.; Zhang, Y. N.; Wang, X.; Chen, J. F.; Yu, S. H. Carbon

Nanofiber Aerogels for Emergent Cleanup of Oil Spillage and Chemical Leakage Under

Harsh Conditions. Sci. Rep. 2014, 4, 1-6.

44. Dasari, A.; Rohrmann, J.; Misra, R. D. K. Microstructural Aspects of Surface Deformation

Processes and Fracture of Tensile Strained High Isotactic Polypropylene. Mater. Sci. Eng. A-

Struct. 2003, 358, 372-383.

45. Maier, C.; Calafut, T. Polypropylene - The Definitive User's Guide and Databook, 1st ed.;

William Andrew Publishing/Plastics Design Library, 1998.

46. Gandhi, A.; Asija, N.; Gaur, K. K.; Rizvi, S. J. A.; Tiwari, V.; Bhatnagar, N. Ultrasound

Assisted Cyclic Solid-State Foaming for Fabricating Ultra-low Density Porous Acrylonitrile–

Butadiene–Styrene Foams. Mater. Lett. 2013, 94, 76-78.

47. Benjamin, T. B.; Ellis, A. T. The Collapse of Cavitation Bubbles and the Pressure Thereby

Produced Against Solid Boundaries. Phil. Trans. R. Soc. Lond. A 1966, 260, 221-245.

48. Wang, X.; Li, W.; Kumar, V. Creating Open-Celled Solid-State Foams using Ultrasound. J.

Cell. Plast. 2009, 45, 353-369.

164

49. Encina, M. V.; Lissi, E.; Sarasua, M.; Gargallo, L.; Radic, D. Ultrasonic Degradation of

Polyvinylpyrrolidone—Effect of Peroxide Linkages. J. Polym. Sci., Polym. Lett. Ed. 1980,

18, 757–760.

50. Berkowski, K. L.; Potisek, S. L.; Hickenboth, C. R.; Moore, J. S. Ultrasound-Induced Site-

Specific Cleavage of Azo-Functionalized Poly(ethylene glycol). Macromolecules 2005, 38,

8975–8978.

51. Paulusse, J. M. J.; Sijbesma, R. P. Reversible Mechanochemistry of a Pd-II Coordination

Polymer. Angew. Chem. Int. Ed. 2004, 49, 4460–4462.

52. Oh, J. S.; Isayev, A. I.; Rogunova, M. A. Continuous Ultrasonic Process for In-situ

Compatibilization of Polypropylene/Natural Rubber Blends. Polymer 2003, 44, 2337-2349.

165

Chapter 8

Conclusions and Recommendations

8.1 Conclusions

Viscoelastic properties and crystallization kinetics of semicrystalline polymers are critical in

polymer foam processing. Improvements in viscoelastic properties such as melt strength, and

strain hardening in extensional flow and rapid crystallization rates can stabilize the cell

morphology and suppress cell rupture, effectively broadening the foam processing window.

Using fibrillar blends of semicrystalline polymers we demonstrate that the presence of fibrillated

dispersed phases has the following influence on the polymer-host:

Strain hardening in uniaxial extensional flow: It is demonstrated that when the fibril in the

semicrystalline polymer host exceeds the rheological percolation threshold, the high aspect ratio

fibrils develop topological interactions. The formation of a network of entangled fibrils markedly

enhances the strain hardening response in uniaxial extensional flow of the matrix. The strain

hardening response is attributed to the inability of the fibrils to disentangle readily enough and

follow the deformation upon experiencing an extensional flow.

Enhanced kinetics of crystallization: The fibrillar dispersed phases increase the kinetics of

crystallization of the polymer-host. Not only do the high aspect ratio fibrils provide a large

surface to volume ratio to facilitate crystal nucleation, the thermodynamics of crystallization on

the fiber in the form of a transcrystalline layer are favorable compared to crystallization in the

bulk.

Foam processing of fibrillar blends reveals noticeable improvements in the foaming ability

compared to the neat polymer-host or when the dispersed phase domains are spherical.

Specifically, microcellular foams are obtained at mild processing conditions easily accessible in

conventional foam manufacturing equipment making the strategy commercially appealing in

foam processing.

A fiber-spinning line can be effectively used to produce fibrillar blends on a large-scale. The

applied macroscopic extensional flow at high draw ratios can cause the dispersed phase domains

in a polymer blend to elongate into large aspect ratio fibrils. This was successfully demonstrated

166

using a blend of PP containing 5 wt% PET. After fiber spinning the spherical PET domains in

the blend elongated into high aspect ratio fibrils.

PTFE has a strong tendency to fibrillate during blending with another polymer. The chains

packed in PTFE crystals are characterized by weak cohesive forces. Upon application of a

hydrodynamic stress, the chains can unpack from the crystals. Consequently, PTFE has a low

yield strength to undergo plastic deformation. This behaviour allows PTFE to fibrillate during

blending in a twin-screw extruder without the need of the fiber-spinning process, typically

required for fibrillating dispersed phases in polymer blends on a large scale.

It is also observed that PTFE exhibits a CO2-philic character. The presence of PTFE fibrils in a

polymer-host can increase the CO2 solubility. For example, at a temperature of 155°C and a CO2

pressure of 17.2 MPa, the solubility of CO2 in neat PP is about 0.115 g g-1

but when 0.3 wt%

PTFE is present in the PP, the CO2 uptake increases to about 0.135 g g-1

. The increased solubility

of CO2 with the addition of PTFE makes available a larger amount of the blowing agent for

increasing the foam volume expansion ratio. Furthermore, the presence of PTFE in a polymer-

host can induce a larger degree of CO2-induced plasticization permitting the fibrillar blend to be

effectively processed at lower temperatures.

Open-cell foams of PP/PTFE can be prepared within a narrow processing temperature window.

At the optimum conditions, open cell contents as high as 97.7% and bulk densities as low as 0.07

g cm-3

are produced in an extrusion process. The open-cell formation is attributed to the growth

of crystalline heterogeneities which create the heterogeneous melt structure required for

fabricating open-cell foams.

These open-cell foams of PP/PTFE demonstrate superhydrophobic and oleophilic properties

which results in a high oil/water selectivity. The open-cell foams display significant

improvements in oil uptake capacities compared to commercially available PP-based materials

currently used for oil-spill cleanup applications, such as nonwoven PP fibers and melt-blown PP

pads. The higher uptake capacities are attributed to a high void fraction of 92% which makes

available a large 3D space for oil storage. The open-cell foams offer robust stability at

temperatures ranging from -196 to 120°C. Subjecting the open-cell foams to these temperature

extremes for prolonged periods does not influence the superhydrophobic and oleophilic

properties of the foams. The kinetics of oil uptake can be enhanced by increasing the ‗skin‘ layer

167

porosity of the extruded foams using high intensity ultrasound irradiation treatment. If the PP

matrix in the PP/PTFE fibrillar blend is a PP-homopolymer, the reusability of the foams is poor

because PP-homopolymer/PTFE undergoes a large permanent deformation after mechanical

―squeezing‖ to extract the oil (30% permanent deformation over ten compression cycles). This

damages the open-cell morphology and consequently, the superhydrophobic and oleophilic

properties. If the PP matrix in the PP/PTFE fibrillar blend is a PP random-copolymer, the

reusability of the foams is drastically improved because the resultant PP-copolymer/PTFE open-

cell foams undergo a small permanent deformation after mechanical squeezing to extract the oil

(10% permanent deformation over ten compressive cycles). The open-cell foams are

technologically promising for applications such as oil-spill cleanup, organic pollutant removal

and field water remediation.

8.2 Recommendations

1. Viscoelastic properties of polymeric systems are greatly affected by the dissolution of a

blowing agent. Measurement of rheological properties of polymer fibrillar blends with a

blowing agent under high-pressure and intense flow fields typically observed during

foam extrusion should be conducted to derive a more complete comprehension of the

effect of the fibrils on the rheological characteristics during foam extrusion.

2. The main mode of deformation during polymer foaming is biaxial extension.

Consequently, biaxial extensional properties of the fibrillar blends should be

characterized to provide a more accurate relation between the degree of strain hardening

and foam processing.

3. The current investigation uses laboratory-scale foam extrusion as an effective screening

tool for research and development. However, laboratory-scale foam extruders do not

scale-up to larger-volume foam extruders because equipment design attributes and the

processing parameters are fundamentally different. Factors that may differ include screw

size and configuration, material residence time, flow rate, heat transfer rate, die design,

etc. Consequently, a separate study is needed to evaluate the feasibility of producing

168

foams of fibrillar blends on a pilot-scale foam extrusion system which can then be scaled-

up to the industrial-scale with relatively less process modification.

169

Appendix A

Calculation of solubility of neat PTFE using the rule of mixtures from experimental data

presented in Fig. 5.5

The rule of mixtures which defines the sorption of CO2 in the melt at equilibrium was employed

to calculate the solubility of CO2 in neat PTFE from the experimental data obtained for the

PP/PTFE fibrillar blend (99.7/0.3 wt%). Based on the rule of mixtures, sorption capacity of CO2

in the melt ( ) at a given condition of temperature and CO2 pressure can be described as

(1)

where is the CO2 sorption in PP, is the CO2 sorption in PTFE and is the weight

fraction of PTFE. Crystalline segments are present in both PP and PTFE. Thus, Eq. 1 can be

rewritten as:

[ ] [ ] (2)

where , , , is the sorption capacity of CO2 in the amorphous and

crystalline PP and PTFE respectively; and and are the crystalline fractions of PP and

PTFE. Crystalline segments are assumed to have negligible CO2 sorption capacity. Therefore

Eq. 2 can be rewritten as:

(3)

At 17.2 MPa17.2 MPa and 155°C, from the experimental data shown in Fig. 5.5, the sorption

capacity of PP, given by is 0.1148 gCO2/g polymer, the total sorption

capacity of PP/PTFE (99.7/0.3 wt%) is given by which is 0.1331 gCO2/g polymer and is 0.3

wt%. Thus:

(4)

170

Based on this, the expected sorption capacity of CO2 for PTFE at a temperature of 155°C and a

CO2 pressure of 17.2 MPa, , is calculated to be 6.2 gCO2/g polymer. This is

in clear contrast to the sorption capacity of PP experimentally determined to be 0.1148 gCO2/g

polymer under the same conditions of temperature and pressure. Thus, the CO2 sorption capacity

of PTFE fibers within the PP matrix seems to be fifty four-folds higher than that of PP at a

temperature of 155°C and a CO2 pressure of 17.2 MPa.

171

Appendix B

Photograph of the tandem foam extrusion system used to prepare open-cell foam filaments

Fig. S1 A photograph of the foam extrusion system used to produce the PP/PTFE open-cell

foams with carbon dioxide (CO2) as the foam blowing agent. The white arrows indicate the

direction of material flow. Gas is injected in the first extruder. Foaming occurs when the material

exits through the die. The system is capable of producing a continuous open-cell foam filament

at a rate of 2 kg/h.

172

Creation of a heterogeneous melt from crystalline heterogeneities during extrusion

Fig. S2 An SEM micrograph of permanganate etched 1

sample of PP-homopolymer/PTFE (97/3

wt%) after extrusion showing a shish-kebab type structure of PP on PTFE fibrils for creation of a

heterogeneous melt. The temperature profile used from the start of the second extruder to the die

exit is 190, 165, 160, 155 and 150°C, respectively. The crystalline morphology is quenched by

immersing the extrudate in liquid nitrogen immediately after it exits the die. Previously,

crystallization of PP has been observed in extrusion when the extrusion processing is conducted

below the Tm of PP 2 and the thermodynamics of crystallization of PP on PTFE fibers is more

favorable than bulk crystallization of PP. 3

The formation of shish-kebab type structures on high

aspect ratio second phases has also been previously reported such as on carbon nanotubes 4

or

polymeric fibrils. 5

Thus, the growth of such crystalline morphology during our extrusion process

is expected, since Tprocessing < Tm of the PP matrix.

173

Comparison of PP-homopolymer/PTFE and PP-copolymer/PTFE open-cell foams:

Fig. S3 SEM micrographs comparing the open-cell foam morphologies of a) PP-

homopolymer/PTFE (97/3 wt%), and b) PP-copolymer/PTFE (97/3 wt%).

The method described in the Experimental section to prepare open-cell foams of PP-

homopolymer/PTFE is employed for preparing open-cell foams of PP-random-copolymer/PTFE

(97/3 wt%) fibrillar blend. Open-cell foams are successfully obtained at a temperature of 146°C

with a bulk density of 0.055 g cm-3

, a cell density on the order of 107 cells cm

-3, a volume

expansion ratio of 16-fold, and a void fraction of 0.94. The open-cell content is determined using

gas pycnometry and found to be 98.1%. The resultant PP random-copolymer open-cell foam

forms a reticulated structure.

174

Table S1 Comparison of morphological characteristics of the open-cell foams of PP/PTFE (97/3

wt%) when PP is either a homopolymer, or a random copolymer.

Characteristic PP-homopolymer/PTFE

open-cell foam

PP-copolymer/PTFE

open-cell foam

Open-cell extrusion temperature (°C) 150 146

Cell density (cells/cm3) 10

7 to 10

8 10

7 to 10

8

Volume expansion ratio 13.1 16.0

Void fraction (%) 92 94

Open-cell content (%) 97.7 98.1

Foam density (g/cm3) 0.07 0.055

Table S2 Comparison of oil uptake capacities of open-cell foams of PP/PTFE (97/3 wt%) when

PP is either a homopolymer, or a random copolymer. The values shown are the average uptake

capacities of 10 measurements. The oil uptake performance remains largely unaltered when the

polymer matrix in the PP/PTFE is changed from a PP homopolymer to a PP-random copolymer

indicating absorption is mainly a function of morphology.

Material Octane Gasoline Light Crude Oil Diesel Heavy Crude Oil Kerosene

Homopolymer-average 23.9 22.6 14.6 11.3 8.5 5.4

Copolymer-average 21.7 22.7 12.6 12.1 9.8 6.1

Literature - 9.2 6 8.3

7 9.1

6 9.1

7 10.0

8

Video:

V1: Extrusion of open-cell PP/PTFE foams

V2: Oil uptake by extruded open-cell foam

175

References

1. Olley, R. H.; Bassett, D. C. An improved permanganic etchant for polyolefines. Polymer

1982, 23, 1707-1710.

2. Tabatabaei, A.; Barzegari, M. R.; Nofar, M.; Park, C. B. In-situ visualization of polypropylene

crystallization during an extrusion process. Polym. Test. 2014, 33, 57-63.

3. Wang, C.; Hwang, L. M. Transcrystallization of PTFE fiber/PP composites (I) crystallization

kinetics and morphology. J. Polym. Sci. Part B: Polym. Phys. 1996, 34 (1), 47-56.

4. Li, L.; Li, C. Y.; Ni, C. Polymer crystallization-driven, periodic patterning on carbon

nanotubes. J. Am. Chem. Soc. 2006, 128, 1692-1699.

5. Shen, K. Z.; Yang, W.; Huang, R.; Yang, M. B. Transcrystalline morphology of an in-situ

microfibrillar poly(ethylene terephthalate)/poly(propylene) blend fabricated through a slit

extrusion hot stretching-quenching process. Macromol. Rapid Commun. 2004, 25, 553-558.

6. Chung, T. C.; Yuan, X. P. Methods and materials for hydrocarbon recovery. WO 2012024614

A2, February 23, 2012.

7. Bayat, A.; Aghamiri, S. F.; Moheb, A.; Vakili-Nezhaad, G. R. Oil spill cleanup from sea water

by sorbent materials. Chem. Eng. Technol. 2005, 28, 1525-1528.

8. Pan, Y.; Shi, K.; Peng, C.; Wang, W.; Liu, Z. J. X. Evaluation of hydrophobic polyvinyl-

alcohol formaldehyde sponges as absorbents for oil spill. ACS Appl. Mater. Interfaces 2014, 6,

8651–8659.

functional polymer foams from in-situ fibrillated …...ali rizvi doctor of philosophy department of...

Documents