/smash/get/diva2:8665/FULLTEXT01.pdf

/smash/get/diva2:8665/FULLTEXT01.pdf
Recycling of Mixed Plastic Waste
– Is Separation Worthwhile?
Stefan Tall
Department of Polymer Technology
Royal Institute of Technology
Stockholm, Sweden
2000
Recycling of Mixed Plastic Waste
– Is Separation Worthwhile?
Stefan Tall
Department of Polymer Technology
Royal Institute of Technology
Stockholm, Sweden
2000
Akademisk avhandling
som med tillstånd av Kungliga Tekniska Högskolan framlägges till offentlig
granskning för avläggande av Teknologie Doktorsexamen måndagen den 6 mars
2000, kl. 10:00, i sal K1, Teknikringen 56, KTH, Stockholm.
Avhandlingen försvaras på engelska.
Recycling of Mixed Plastic Waste
- Is Separation Worthwhile?
Stefan Tall
Department of Polymer Technology, Royal Institute of Technology, SE-100 44 Stockholm, SWEDEN
ABSTRACT
The automated separation of plastic waste fractions intended for mechanical recycling is associated
with substantial investments. It is therefore essential to evaluate to what degree separation really brings
value to waste plastics as raw materials for new products. The possibility of reducing separation
requirements and broadening the range of possible applications for recycled materials through the
addition of elastomers, mineral fillers or other additives, should also taken into consideration.
Material from a Swedish collection system for rigid (non-film) plastic packaging waste was studied.
The non-film polyolefin fraction, which dominated the collected material, consisted of 55% polyethylene
(PE) and 45% polypropylene (PP). Mechanical tests for injection-moulded blends of varying
composition showed that complete separation of PE and PP is favourable for yield strength, impact
strength, tensile energy to break and tensile modulus. Yield strength exhibited a minimum at 80% PE
whereas fracture toughness was lowest for blends with 80% PP. The PE fraction, which was
dominated by blow-moulded high density polyethylene (HDPE) containers, could be made more
suitable for injection-moulding by commingling with the PP fraction. Nucleating agents present in the
recycled material were found to influence the microstructure by causing PP to crystallise at a higher
temperature than PE in PP-rich blends but not in PE-rich blends.
Studies of sheet-extruded multi-component polyolefin mixtures, containing some film plastics, showed
that fracture toughness was severely disfavoured if the PE-film component was dominated by low
density polyethylene (LDPE) rather than linear low density polyethylene (LLDPE). This trend was
reduced when the non-film component was dominated by bottle -grade HDPE. A modifier can be
added if it is desired to increase fracture toughness or if there are substantial variations in the
composition of the waste-stream. A very low density polyethylene (VLDPE) was found to be a more
effective modifier than poly(ethylene-co-vinyl acetate) and poly(1-butene). The addition of 20%
VLDPE to multi-component polyolefin mixtures increased the tensile strength and tear propagation
resistance by 30% on average, while standard deviations for mechanical properties were reduced by
50%, which would allow product quality to be kept more consistent.
ABS was found to be more sensitive to contamination by small amounts of talc-filled PP than viceversa. Contamination levels over 3% of talc -filled PP in ABS gave a very brittle material whereas talcfilled PP retained a ductile behaviour in blends with up to 9% ABS. Compatibility in blends of ABS,
high-impact polystyrene and talc -filled PP was poorer at high deformation rates, as opposed to blends
of PE and PP from rigid packaging waste where incompatibility was lower at fast deformation. This
difference was explained by a higher degree of interfacial interaction through chain entanglements in
PE/PP blends.
Keywords: Polyethylene, polypropylene, polyolefins, plastic packaging waste, recycling, compatibility,
crystallisation, morphology, modification, injection moulding, compounding, ABS, high-impact
polystyrene, polymer composites.
LIST OF ARTICLES
This thesis is a summary of the following papers:
I
“Recycling of Mixed Plastic Fractions: Mechanical Properties of Multicomponent Extruded
Polyolefin Blends Using Response Surface Methodology“, Stefan Tall, Ann-Christine
Albertsson and Sigbritt Karlsson, Journal of Applied Polymer Science, vol. 70(12),
p2381-2390 (1998).
II
“EPDM Elastomers as Impact Modifiers for Contaminated, Recycled HDPE“, Stefan Tall,
Ann-Christine Albertsson and Sigbritt Karlsson, Polymers & Polymer Composites, vol.
5(6), p417-422 (1997).
III
“Morphology and Compatibility of Blends of Recycled Polyethylene and Polypropylene
from Packaging Waste“, Stefan Tall, Ann-Christine Albertsson and Sigbritt Karlsson,
submitted to Polymer Engineering and Science.
IV
“Enhanced Rigidity of Recycled Polypropylene from Packaging Waste by Compounding
with Talc and High-Crystallinity Polypropylene“, Stefan Tall, Ann-Christine Albertsson and
Sigbritt Karlsson, submitted to Polymers for Advanced Technologies.
V
“Mechanical Properties, Morphology and Compatibility of Recycled Plastic Mixtures of
ABS, Talc-Filled Polypropylene and High-Impact Polystyrene”, Stefan Tall, Marcello
Colnaghi, Luigi Maffioli, Ann-Christine Albertsson and Sigbritt Karlsson, submitted to
Journal of Applied Polymer Science.
TABLE OF CONTENTS
LIST OF ABBREVIATIONS ................................................................................................................................................. 3
1
PURPOSE OF THE STUDY.......................................................................................................................................... 5
2
INTRODUCTION ........................................................................................................................................................... 6
2.1
BACKGROUND.......................................................................................................................................................6
2.2
WASTE MANAGEMENT METHODS................................................................................................................7
2.2.1
Landfilling .....................................................................................................................................................7
2.2.2
Primary recycling.........................................................................................................................................8
2.2.3
Secondary recycling ....................................................................................................................................8
2.2.4
Feedstock recycling .....................................................................................................................................8
2.2.5
Incineration with energy recovery ............................................................................................................8
2.2.6
Degradation and composting ....................................................................................................................9
2.3
RECYCLABILITY OF PLASTICS.........................................................................................................................9
2.4
MUTUAL COMPATIBILITY OF PLASTICS .....................................................................................................9
2.4.1
Polyethylene and polypropylene.............................................................................................................10
2.4.2
Other polyolefin blends.............................................................................................................................12
2.4.3
Polyolefins and other plastics..................................................................................................................12
2.4.4
Compatibilisation and modi fication ......................................................................................................13
2.5
SEPARATION TECHNIQUES FOR PLASTIC WASTE..................................................................................14
2.5.1
Manual sorting ...........................................................................................................................................14
2.5.2
Sorting by density.......................................................................................................................................14
2.5.3
Air classification.........................................................................................................................................15
2.5.4
Electrostatic separation............................................................................................................................15
2.5.5
Material identification techniques .........................................................................................................15
3
MATERIALS.................................................................................................................................................................16
3.1
VIRGIN PLASTICS................................................................................................................................................16
3.1.1
Polyolefins used in the multi-variable evaluation of sheet-extruded blends..................................16
3.1.2
PP reference and modifying grades ........................................................................................................16
3.1.3
Non-polyolefin plastics .............................................................................................................................17
3.2
MODIFIERS AND FILLERS................................................................................................................................17
3.3
POST-USE MATERIALS .....................................................................................................................................18
3.3.1
The Lunda plant..........................................................................................................................................18
3.3.2
Collection and sorting of waste samples ...............................................................................................19
4
EXPERIMENTAL METHODS.................................................................................................................................... 21
4.1
COMPUTER-AIDED EXPERIMENTAL DESIGN AND EVALUATION......................................................21
4.2
PREPARATION OF TEST SPECIMENS............................................................................................................21
4.2.1
Extrusion compounding............................................................................................................................21
4.2.2
Injection moulding .....................................................................................................................................21
4.2.3
Sheet extrusion............................................................................................................................................22
4.2.4
Simulated recycling of ABS and talc-filled PP:....................................................................................22
4.3
MECHANICAL TESTING....................................................................................................................................22
4.3.1
Tensile tests..................................................................................................................................................22
4.3.2
Impact tests ..................................................................................................................................................22
4.3.3
Dynamic mechanical tests.........................................................................................................................23
4.3.4
Heat deflection tests...................................................................................................................................23
4.3.5
Tear-propagation resistance ....................................................................................................................23
4.4
DIFFERENTIAL SCANNING CALORIMETRY (DSC) ....................................................................................23
4.5
SCANNING ELECTRON MICROSCOPY...........................................................................................................23
5
RESULTS AND DISCUS SION ..................................................................................................................................25
1
5.1
COMPATIBILITY OF SINGLE GRADE PE/PP BLENDS.................................................................................25
5.2
MECHANICAL PROPERTIES OF MULTI-COMPONENT POLYOLEFIN MIXTURES.............................26
5.3
MODIFICATION OF MULTI-COMPONENT POLYOLEFIN MIXTURES FOR IMPROVEMENT OF
FRACTURE TOUGHNESS ................................................................................................................................................28
5.3.1
Modification of multi-component polyolefin blends using EVA, VLDPE and poly(1 -butene) ....28
5.3.2
Modification of HDPE-rich blends using EPDM ..................................................................................29
5.4
CHARACTERISATION OF MATERIALS SAMPLED FROM THE SWEDISH COLLECTION SYSTEM
FOR RIGID PLASTIC PACKAGING WASTE................................................................................................................30
5.4.1
Non-film PE fraction ..................................................................................................................................31
5.4.2
Non-film, floating PP fraction ..................................................................................................................32
5.5
CRYSTALLISATION AND MICROPHASE STRUCTURE OF PE/PP FROM RIGID PLASTIC
PACKAGING WASTE.......................................................................................................................................................33
5.5.1
DSC-observations.......................................................................................................................................33
5.5.2
SEM-observations......................................................................................................................................36
5.6
MECHANICAL COMPATIBILITY OF PE/PP FROM RIGID PLASTIC PACKAGING WASTE..............40
5.6.1
Fracture toughness ....................................................................................................................................40
5.6.2
Other properties..........................................................................................................................................42
5.6.3
Should the PE and PP fractions be separated prior to mechanical recycling?.............................44
5.7
MODIFICATION OF PP FROM RIGID PACKAGING WASTE ....................................................................44
5.7.1
Modification with highly crystalline PP................................................................................................45
5.7.2
Modification with talc and highly crystalline PP................................................................................45
5.8
EFFECTS OF FOREIGN POLYMERS AS IMPURITIES IN RECYCLED PLASTICS ...................................47
5.8.1
EVOH from barrier layers in recycled PP ..............................................................................................47
5.8.2
Talc-filled PP and HIPS in ABS ...............................................................................................................48
5.8.3
ABS in talc-filled PP ..................................................................................................................................49
6
CONCLUSIONS........................................................................................................................................................... 51
7
SUGGESTIONS FOR FUTURE WORK................................................................................................................... 54
8
ACKNOWLEDGEMENTS.......................................................................................................................................... 55
9
REFERENCES............................................................................................................................................................... 56
APPENDIX I: Experimental design and raw data for sheet extruded polyolefin blends.
APPENDIX II: Classification of materials sampled at the Lunda plant.
2
LIST OF ABBREVIATIONS
ABS
Engineering plastic composed of the monomers Acrylo-nitrile, Butadiene and Styrene.
DSC
Differential Scanning Calorimetry.
EPDM
Rubber material produced by co-ordination polymerisation of Ethylene, Propylene and a
Diene Monomer.
EVA
Copolymer of Ethylene and Vinyl Acetate. Produced by free-radical high-pressure
polymerisation.
EVOH
poly(ethylene-co-vinyl alcohol). Produced by deacetylation of EVA.
HDPE
High-density polyethylene. Polyethylene resin with a density higher than 0.94 g/cm3.
HIPS
High-impact polystyrene. PS modified with an elastomer, normally polybutadiene.
LDPE
Low-density polyethylene. Produced by free-radical high-pressure polymerisation.
LLDPE Linear low-density polyethylene. Produced by co-ordination polymerisation of ethylene
and an α -olefin, normally 1-butene, 1-hexene or 1-octene.
MFR
Melt flow ratio, i.e. the amount of plastic resin that can flow through a standardised
capillary die at a given temperature and a given load.
MSW
Municipal Solid Waste
PB
Isotactic poly(1-butene) produced by co-ordination polymerisation.
PE
Polyethylene
PET
Poly(ethylene terephthalate). Thermoplastic polyester.
PMMA Poly(methyl methacrylate)
PP
Polypropylene
PS
Polystyrene
3
PVC
Poly(vinyl chloride)
RSM
Response Surface Methodology.
SBS
Thermoplastic elastomer composed of poly(styrene-block-butadiene-block-styrene).
SEBS
Thermoplastic elastomer with a similar structure to SBS but with ethylene incorporated
with the butadiene-block.
SEM
Scanning Electron Microscopy
TPR
Tear Propagation Resistance.
VLDPE Very low-density polyethylene. LLDPE-type polymer with higher content of α -olefin comonomer, giving it more elastic properties.
4
1 PURPOSE OF THE STUDY
The object of this work has been to evaluate the necessity of sorting waste plastics prior to
mechanical recycling. Modifications of recycled plastic fractions, aimed at reducing the purity
requirement or increasing the number of potential applications for the materials, are also assessed.
Materials studied include the most common types of thermoplastics used in packaging applications
and in durable products such as automotive components, office machines and domestic appliances.
Special emphasis is put on polyolefins; polyethylene (PE) and polypropylene (PP), because of the
vast abundance of these materials.
In the case of mixed polyolefin fractions containing both film plastics and rigid plastics, as well as
both PE and PP, the following issues are addressed:
• How do the mechanical properties of recycled materials alter with variations in the composition of
the waste-stream?
• How can fracture toughness be increased and how can variations in mechanical properties of a
recycled resin due to a varying composition of the waste stream from which the materials are
derived be reduced?
Studies of the polyolefin fraction of rigid (non-film) packaging waste, after separation by density, are
focused on the following questions:
• What is the microphase structure of a PE/PP blend and how is it formed?
• What is the mechanical compatibility between PE and PP and how does it relate to the question
of whether the polyolefins should be separated into pure PE and PP fractions or be recycled as a
mixed fraction?
• How can the range of possible applications for the recycled materials be broadened?
The following issues are addressed regarding the effect of foreign polymers as impurities in recycled
plastics:
• What are the possibilities of mechanical recycling of plastic containers which have a coextruded
barrier layer of a foreign polymer?
• What are the purity requirements in the mechanical recycling of mineral-filled PP and ABS
(acrylo -nitrile, butadiene, styrene plastic) used in durable products?
5
2 INTRODUCTION
2.1 BACKGROUND
The use of plastics in packaging applications is growing steadily. Most industrialised countries have
systems for the collection and recycling of plastic packaging waste, either implemented on a full scale
or on trial. An important aspect of plastic packaging recycling is that the types of plastics used for
most packaging applications are inexpensive commodity materials. The price of corresponding virgin
resins determines the ceiling for the price at which recycled materials can be sold for reprocessing.
Tab. 2-1 displays prices and consumption data for Western Europe (EU plus Norway and
Switzerland) of virgin plastics during 1998. Engineering plastics used in durable products are
generally more expensive than the most common packaging plastics. This relation promotes the
recovery of scrapped engineering plastics. For common packaging plastics such as polyethylene and
polypropylene, all steps in the recovery and recycling process need to be highly cost-effective, unless
sizeable subsidies are being paid [Bruder, 1997]. Techniques to facilitate the collection, sorting and
reprocessing of plastic packaging waste are therefore urgently needed, as well as methods that can
increase the value and the number of potential applications of the recovered materials. The price at
which secondary material can be sold is related to the price of the corresponding virgin material.
Price fluctuations therefore entail the need for an economic safety-margin (risk premium) in order to
make investments in recycling facilities viable [Brandrup, 1997].
Tab. 2-1: Market prices and consumption in Western Europe for virgin plastics. (Sources:
European Plastic News and Modern Plastics International, various issues)
Plastic type
Plastics common in packaging applications
Polyethylene, low density
Polyethylene, linear low density
Polyethylene, high density
Polypropylene
Polystyrene
Poly(vinyl chloride)
Poly(ethylene terephthalate)
Engineering plastics
ABS
Polycarbonate
Polyoxymethylene
Poly(methyl methacrylate)
Polyamide
6
Consumption, 1998
(tonnes)
Price, last quarter 1998
(DEM/kg)
4,650,000
1,856,000
4,162,000
6,152,000
2,802,000
5,618,000
1,118,000
1.25
1.20
1.30
1.05
1.30
0.95
1.55
640,000
351,000
145,000
261,000
551,000
2.50
5.90
3.80
3.70
4.90
An aspect that favours the recycling of packaging plastics is that their (primary) service life is
relatively short. The utilisation of plastics from scrapped durables is often constrained by the
depletion and migration of stabilisers and their long-term degradation [Gedde et al., 1994, Eriksson,
1997]. This is usually not a major problem associated with plastic packaging recycling.
ABS is likely to be contaminated with PP or impact-modified polystyrene (HIPS) when materials are
collected for recycling, because of the widespread use of such materials in durable products such as
automotive components, housings for office machines and domestic appliances. Unfilled PP-grades
can be separated from ABS by flotation, but mineral-filled PP-grades, common in durable products,
often have a density similar to that of ABS and this makes separation more difficult. Since the cost of
material separation tends to increase when a high purity of the output is desired, it is essential to
know how the presence of foreign polymers influences the performance of recycled materials.
2.2 WASTE MANAGEMENT METHODS
Human activities produce waste of many kinds. The following discussion will be focused on
“municipal solid waste“ (MSW). MSW is defined as non-hazardous waste generated in households,
commercial establishments and institutions; excluding industrial process waste, demolition waste,
agricultural waste, mining waste, abandoned automobiles, ashes and sewage sludge [Lund et al.,
1993]
Waste management can in a narrow sense be defined as “how to get rid of the trash“ but a broader
definition also includes issues such as:
• How to re-appraise materials previously considered as waste, by setting up in-house recycling
schemes.
• How to realise, and possibly also increase, the value of materials that is waste to us but may be
regarded as a resource by others.
• How to delay the point in time at which a product becomes waste by extending its lifetime or by
re-use.
• How to prevent materials from becoming waste by reducing material consumption through
modified designs and technology.
The plastic fraction of MSW can’t be discussed in isolation from MSW management in general
because it is only a minor fraction of the total MSW. This means that any special arrangements
regarding the plastic fraction will be associated with considerable costs, but also potential benefits.
An illustration of the interdependence of plastic waste management with waste management in
general is that, if incineration with energy recovery is the dominant method, it is likely to be attractive
not to sort out the plastic fraction for mechanical recycling. The reason is that plastics have a high fuel
value. If, on the other hand, landfilling is the prevalent method, separate recycling schemes for
plastics will be more attractive because plastic waste generally has a high specific volume (low
density).
2.2.1 Landfilling
The historically most common method for dealing with MSW, including plastic waste, is deposition in
landfills. In areas with a high population density such as Western Europe, North-Western USA and
7
Japan, landfilling is becoming more and more difficult and expensive because locations suitable for
such a purpose are scarce. An alternative approach to landfilling that has been proposed is the
establishment of strategic storage facilities [Pearson 1993]. The idea is to stockpile recyclable
materials until technological and economic circumstances have made their recovery viable some time
in the future.
2.2.2 Primary recycling
The processing of plastics often generates a considerable amount of production scrap. The
mechanical recycling of such material, i.e. material that has not been converted to a useful product, is
referred to as primary recycling. Examples of such plastic material that can be re-utilised are edgetrims, start-up and change-over scrap, finished products or parts that fails to meet required
standards, material solidified in mould runners, etc.
Primary recycling can be done in-house if the necessary equipment is available. Machines used for
primary recycling are shredders, grinders and extruders. The recycled material is often mixed with
virgin resin and fed back into the same process that generated it, but it is also possible to produce
other products. If a plastic processing industry does not recycle production scrap itself, the material
can be sold to other companies for primary recycling. This solution may be attractive to processors
who cannot afford to invest in the necessary recycling equipment. To sell production scrap on the
open market can however be risky, since it may make it possible for competitors to acquire cheap
raw-materials and thereby gain a competitive advantage.
The primary recycling of production scrap has been practised for a long time in order to save money.
It helped the plastic processors in Western Germany to reduce the amount of homogeneous plastics
that were lost as waste from 3.1% in 1974 to 0.8% in 1976 [Milgrom, 1982].
2.2.3 Secondary recycling
Secondary recycling is what we normally think of when the recycling of plastics is mentioned, i.e. the
reprocessing of material from used, discarded products into new products. It is also the major
concern of this thesis. The term “mechanical recycling“ is also used frequently. It refers to both
primary and secondary recycling
2.2.4 Feedstock recycling
The degradation of polymeric materials into low molecular weight compounds is referred to as
feedstock recycling or chemical recycling. The ideal form of feedstock recycling is to convert the
polymer back to monomers that can be purified using normal chemical methods and then repolymerised, yielding a material that is identical to virgin resin. This is technically possible for
condensation polymers such as PET and polyamides. For plastics such as polyethylene and
polypropylene there is no technique available that gives a particularly high yield of monomer, but it is
possible to produce fuel oils and synthesis gases by pyrolysis [Curran et al., 1996].
2.2.5 Incineration with energy recovery
Incineration is a waste management method that reduces the volume and recovers energy from
MSW. It has a prominent role in several countries including Switzerland Denmark, Sweden and
Japan [Huang, 1995]. A major environmental concern regarding the incineration of waste which
includes unsorted plastics is the formation of dioxins. This requires chorine and a major source of
8
chlorine is in PVC present in MSW. Modern incineration technology can reduce dioxin emissions to
practically zero [Scheirs, 1998] but the establishment of MSW incineration plants is still severely
hampered by the ”NIMBY-syndrome” (“not in my back-yard“) in many countries [Blom et al.
1998].
2.2.6 Degradation and composting
One way to reduce plastic waste and litter problems is to use materials that are designed to degrade
after their service-life by the action of micro-organisms, oxygen and sun-light [Albertsson, 1977,
Karlsson, 1988]. This strategy has been implemented in polyethylene film-products such as carrierbags and six-pack yokes. Such materials are incongruent with the prerequisite for mechanical
recycling that require the plastics to be resistant to degradation. Therefore it is generally an advantage
if sensitised materials are separated from waste streams considered for mechanical recycling.
2.3 RECYCLABILITY OF PLASTICS
Most plastics in use today are thermoplastics, which means that the material can be melted and reshaped. Some plastics are thermo-sets, which means that they cannot be melted without severe
chemical degradation. This makes the possibilities of recycling of thermo-sets very limited. This
special topic is not dealt with in this thesis.
Most thermoplastics are highly suitable for mechanical recycling, at least in theory. Numerous studies
have shown that the important properties of the most common plastics are fairly well preserved
throughout several cycles of processing and ageing. This means that any company that puts products
on the market that are made of thermoplastics can claim that its products are recyclable. For it to
become realistic that the material will be recovered and recycled, several other criteria must however
be fulfilled. There has to be an infra-structure available for collecting, sorting and reprocessing the
material and there have to be useful applications for the material in its second life.
Recycling infrastructures are characterised by the requirement of economy of scale. This means that
large amounts of material have to be recovered in order to sustain the system for recycling. A plastic
product is consequently not in practice recyclable unless there is enough discarded material of the
same kind to make recycling worthwhile. A possible strategy to overcome this dilemma is to recycle
waste plastics as “commingled plastics“, which means that the plastics are reprocessed without prior
sorting according to plastic type. Park-benches, poles and fences are examples of products that are
being made out of commingled plastics [Scheirs, 1998]. The value of materials made out of
commingled waste plastics is very low compared to that of virgin plastics and this type of recycling is
therefore sometimes referred to as “down-cycling“.
A critical factor is the depletion of antioxidants, but if the material is considered to have insufficient
protection against oxidative degradation, it is possible to add more stabilisers during reprocessing.
Antioxidant formulations are now commercially available that are designed specifically for the
purpose of re-stabilising recycled plastics.
2.4 MUTUAL COMPATIBILITY OF PLASTICS
Most polymer blends are phase-separated. There are only a few known examples of polymer pairs
that exhibit total miscibility. Phase-separation often makes polymer blends brittle due to poor
interfacial adhesion. This is not however always the case. Polymer engineers tend to desire material
9
properties that can be varied in a seemingly endless number of ways. To develop new polymers for
such purposes is very expensive, risky and time-consuming. A much more convenient way to
develop “new” materials is by blending those which are already commercially available. Sometimes
positive synergism occurs. An example of such polymer alloys, that have very attractive mechanical
properties even though the polymers are not completely miscible, is the polycarbonate/ABS-blends.
2.4.1 Polyethylene and polypropylene
The mechanical compatibility of polyethylene and polypropylene is a matter on which results and
conclusions presented in the scientific literature are contradictory [Teh et al. 1994a]. One reason for
the complexity of this matter is the very broad spectrum of properties exhibited by the materials that
we normally refer to as “polyethylene“ or “polypropylene“. There is a seemingly endless number of
possibilities for resin manufacturers to modify their properties and thereby extend their potential
range of applications. These techniques involve:
•
incorporation of co-monomers
•
branching
•
molecular weight distribution
•
degree of tacticity (for polypropylene)
•
molecular orientation through deformation [Lemstra & Kirschbaum, 1984]
•
blending
•
additives
The possibilities of producing specialised polyethylene and polypropylene grades are enhanced by
the rapid development of new polymerisation catalysts [Vogl, 1998].
The invention of co-ordination catalysis for the production of stereoregular α -olefin polymers [Natta
et al., 1955] led to a rapid commercialisation of isotactic polypropylene in the late 1950’s and early
1960’s. The use of the new material was however limited by its brittleness at low temperatures.
Polymer scientists suggested that this could be solved by blending with polyethylene. Between 1962
and 1969, 34 patents were issued on this subject, all assigned to resin manufacturers [Noel &
Carley, 1975]. The performance of such blends was however limited.
Today the most common method of improving the low-temperature toughness of PP is by the
addition of an elastomeric ethylene-propylene random copolymer that forms a separate rubbery
phase [Tselios et al., 1998]. The formation of the copolymer can be incorporated into the
polymerisation process so that no subsequent blending step is necessary. Such PP-grades are often
called “block-copolymers“ in the technical literature, although this is a somewhat misleading
designation. The term “hetero-phasic copolymers“ provides a better description. Rubber-modified
PP formulations can also include some polyethylene [Ha & Kim, 1989].
The basic characteristics of an immiscible blend such as separate melting peaks and unshifted glasstransitions were identified for polyethylene/polypropylene blends by Zakin et al. [1966]. Depending
on the mixing conditions, some degree of interpenetration does however occur [Kryszewski et al.,
1973]. Domain sizes for each phase in a 50/50 blend precipitated from a solution or melt-crystallised
were found to be in the range of 0.1 - 1 µm when the blend was studied with small-angle neutron
scattering [Wignall et al., 1982].
10
The degree of miscibility in the molten state has been studied theoretically by Rajasekaran et al.
[1995]. The PP/PE sys tem is described as being highly incompatible, exhibiting a UCST-behaviour
(Upper Critical Solution Temperature). It should be remembered however that such studies are
concerned with molten blends in thermodynamic equilibrium. When polymer blends are processed
using conventional techniques such as injection moulding, phase dissolution is enhanced by high
shear-forces [Hindawi et al., 1992] and non-equilibrium structures are frozen-in due to fast cooling
[Sano et al., 1998].
Experimental studies on unsheared melts have shown that structural dissimilarities between a
polyolefin pair do not have to be large for phase separation to be observable [Krishnamoorti et al.,
1994]. For blends of PP and ethylene/propylene random copolymers, an ethylene content of 8% in
the copolymer was sufficient for liquid -liquid phase-segregation to occur [Lohse, 1986]. Hill et al.
[1998] observed total solubility for up to 1% PP in HDPE or 0.5% HDPE in PP, even after 20
minutes of storage in the melt at 170 °C. Longer storage times and higher temperatures decreased
the level of solubility.
Similar behaviour was also observed for blends of PP and LLDPE [Hill et al., 1994]. Stachurski et
al. [1996] studied droplet growth of the dispersed phase in a melt blend of 80% PP and 20%
LLDPE. The growth was attributed to the diffusion of LLDPE species from the PP-rich matrix into
the droplets. The process was faster at higher temperatures due to a higher diffusivity. Diffusivity is
also dependent on molecular weight, which means that high molecular-weight species stay longer in
the PP-rich phase.
When about 5-20% HDPE is added to PP homopolymers, several researchers have found positive
synergistic effects regarding the tensile modulus [Noel & Carley, 1975, Deanin & Sansone, 1978,
Lovinger & Williams, 1980, Teh et al., 1985] and yield strength [Teh et al., 1994b]. Gupta et al.
[1982] also reported an increase in the tensile modulus when HDPE was added to glass-fibrereinforced PP. An increase in the rigidity when HDPE is added to PP can be attributed to the fact
that HDPE enhances the crystallisation of PP by acting as a nucleating agent, as was shown by
Lovinger & Williams [1980]. In contrast, Wenig & Meyer [1980] found that PP nucleation and
spherulite growth was virtually unaffected by the presence of HDPE. The difference can probably be
assigned to differences in experimental procedure regarding the degree of mixing. Bartczak et al.
[1986] studied the isothermal crystallisation of a blend of 80% PP and 20% HDPE in the presence
of two nucleating agents, sodium benzoate and magnesium sulphate. Synergistic effects between the
nucleating agents and HDPE on the ability to nucleate PP was observed. This was explained by the
ability of primary nuclei to diffuse between the phases.
The manner in which impact strength is affected by the addition of HDPE to PP is dependent on the
ratio between the impact strengths of the pure materials. This means that the impact strength can
increase slightly if the PP-grade used has poor impact strength in its pure form [Blom et al. 1995].
There is however a negative synergy, i.e. values for blends fall below the straight line connecting the
values for pure PP and pure HDPE [Galeski et al., 1984, Blom et al. 1995].
Assessments of the mechanical compatibility of HDPE and PP grades are also highly dependent on
how the test specimens have been prepared and which measurements are performed. In the study of
Blom et al. [1995], HDPE had only a small effect on stiffness, impact strength, yield strength, and
stress at break at levels up to 25% when tests were performed on standardised test specimens. In
contrast, gate puncture tests performed on thin-walled, injection-moulded containers made from the
same blends, revealed a 75% drop in the load at failure already at a HDPE-level of 10%.
11
The compatibility between the components in HDPE/PP-blends with HDPE as the main constituent
is an important issue related to packaging recycling, since HDPE bottles often have caps and
closures made of PP. Such blends are generally considered to have poor compatibility. Several
researchers have reported a reduction of 50-60% in the notched Charpy or Izod impact strength at
PP levels of about 20% [Bartlett et al., 1982, Hope et al., 1994, Blom et al., 1996b]. Orroth &
Malloy [1994], on the other hand, reported practically no changes at all in tensile or impact strength
up to a PP-addition level of 16% in bottle-grade HDPE. The designation “HDPE“ is not used
exclusively for linear homopolymers but for all polyethylene resins with a density above 0.94 g/cm3 .
Many HDPE grades have a considerable degree of short-chain branching, although less than in
LLDPE grades. This can partly explain why studies of HDPE/PP compatibility are contradictory.
The stiffness of the HDPE used in the study by Orroth & Malloy was rather low. It had a tensile
modulus of 0.7 GPa, which corresponds to polyethylene grades used for soft, easily squeezable
bottles. Such polyethylene grades are produced by incorporating α -olefin comonomers that give
short-chain branching, and this may enhance PE/PP-compatibility.
Another example of good compatibility with PP for a soft HDPE grade was presented by
Schürmann et al. [1998]. In this case, the density of the resin was 0.943 g/cm3 , i.e. very close to the
lower limit for the “HDPE“ designation (0.940 g/cm3). A rather substantial positive synergism
regarding the impact strength was observed, with a maximum at 60% HDPE. Positive synergism
regarding the tensile modulus was observed also in this case over the composition range of 5-20%
HDPE.
Crystallisation of a PP/LLDPE-blend was studied by Basset et al. [1998] by DSC and Transmission
electron microscopy. At PP-levels ≥20% there was a distinct crystallisation peak for PP at about
115 °C in the DSC thermogram. For compositions with ≤15% PP, PP crystallised at a temperature
of only 80 °C, which could be observed as a small peak after the peak for crystallisation of LLDPE
at 90 °C. The nature of the interface was found to be highly dependent on which polymer crystallises
first. When it was PP, crystal lamellae grew out of the PP-droplets into the LLDPE-rich phase,
reinforcing the interface. This behaviour was possible because PP was partially miscible in LLDPE.
At compositions where LLDPE crystallised first, no crystal interpenetration was observed.
2.4.2 Other polyolefin blends
Isotactic poly(1-butene), PB, is a polyolefin mainly used in pipes [Gedde et al., 1994]. It is also used
as a modifier in polyethylene film. The addition of PB to polyethylene film lowers the adhesive
strength of heat-seals which is desirable in easy-to-open packaging applications [De Clippeleir,
1997]. Blends of PB and PP crystallise as separate crystal phases but they have been said to be
completely miscible in the amorphous phase [Piloz et al., 1976, Siegmann, 1982]. This conclusion
has been based on observations of a single glass-transition, between the glass-transitions of pure PB
and pure PP. Other researchers have claimed that the amorphous phase miscibility is only partial
[Berticat et al., 1980, Gohil & Petermann, 1980, Hsu & Geil, 1987]. Cham et al. [1994] showed
that the degree of miscibility of PB and PP is dependent on the time available for the polymers to
segregate in an unsheared melt before solidification.
2.4.3 Polyolefins and other plastics
Polypropylene and polyethylene, the most common polyolefins, are non-polar hydrocarbons and
their compatibility with polar polymers such as PET, ABS and polyamides is therefore poor [Jabarin
12
et al., 1992, Boucher et al., 1996]. The mechanical strength of those poly mers is based on polarpolar interactions and their interfacial interactions with non-polar polymers is therefore very weak.
Blends of polyethylene and polystyrene also exhibit poor compatibility, due to inferior interfacial
adhesion [Fayt et al., 1981]. The degree of dispersion in incompatible polymer blends is dependent
on the shear forces to which the melt is subjected during processing. This was studied by Min et al.
[1984] for polyethylene/polystyrene blends. Higher shear rates gave a finer morphology.
2.4.4 Compatibilisation and modification
One way to increase the interfacial adhesion and achieve a finer dispersion of phases in an
incompatible polymer blend is to add a compound that has an affinity to both phases and therefore
aggregates at the interface. Such a compound is called a “compatibiliser“ [Paul et al., 1972] or a
“solid-phase dispersant“ [Scott et al., 1985]. Compatibilisers are used in alloys of incompatible
polymers and as adhesive layers in co-extruded films and sheets.
A large portion of the compatibilisers suggested for and used in recycling applications are targeted
at blends rich in polyethylene and/or polypropylene. These compounds are usually more or less
elastomeric polymers such as EPDM, SBS, SEBS, EVA or VLDPE [Bartlett et al, 1982, Hope et
al., 1994, Blom et al., 1995, Obieglo & Romer, 1996]. They generally act so that they shift the
properties of the mixture from stiff and brittle towards soft and tough by reducing the overall
crystallinity of the blend and/or forming a separate rubbery phase that can absorb deformations and
impact energy [Tall et al., 1998]. This is often considered to be beneficial since impact toughness is
the mechanical property that suffers most from the presence of incompatible polymeric contaminants.
The observation that an elastomeric additive improves the fracture toughness of an incompatible
blend does not alone prove that the additive is a true compatibiliser, i.e. a surface-active agent. The
same effect is often observed when an elastomeric additive is added to a single polymer, e.g. EPDM
to PP or SBS to PS.
The additives used can have some of the characteristics of surface-active agents, e.g.
styrene/butadiene block copolymers in polystyrene/polyolefin blends have been shown to have an
emulsifying effect [Fayt et al., 1982]. It is often the case however, that their action is targeted more
towards the bulk of one or more of the individual phases of the blend rather than to the interfaces.
The word “compatibiliser“ does not therefore really provide the best description in those cases. The
terms “modifier“ or “modifier agent“ are more correct for such additives [Tall et al., 1998, La
Mantia, 1993]. This does not however exclude a modifier from having a compatibilising effect as
well. The observed improvement in fracture toughness can very well be a combination of matrixmodification and interfacial adhesion improvement.
Modifications of 50/50-blends of LDPE and either PS, PP or PVC have been studied by Scott et al.
[1981], using EPDM, natural rubber, SBS, polybutadiene and chlorinated polyethylene. As
expected from the anticipated surface-active effect of the elastomeric additives, SBS was most
efficient in LDPE/PS while EPDM was most efficient in LDPE/PP. Following the same logic, it
would have been expected that chlorinated polyethylene would have been most efficient in
LDPE/PVC. This was not however the case. The study showed that EPDM improved the impact
resistance the most. Scott and his co-workers suggested that the degree of compatibility of the
LDPE/PVC-blend was governed by the formation of block or graft copolymers through radical
reactions during processing. The formation of such copolymers can be enhanced by the addition of a
free-radical initiator that decomposes during processing [Scott et al., 1984].
13
Most published studies of the effect of elastomeric modifiers are concerned with rather substantial
additions, normally between 10 and 20%, which cause a severe loss of stiffness. Blom et al. [1996a]
showed on the other hand that, for a blend of 90% PP homopolymer and 10% HDPE, an addition
of 1% EPDM was sufficient to yield a 45% increase in impact strength. The effect of the modifier
was somewhat reduced when the samples had been annealed at 75 °C for one week. An EVAgrade with 28 wt% vinyl acetate was also evaluated as an impact modifier. It was found to be
efficient when added to the PP homopolymer alone but not for the blend with 10% HDPE. In blends
of mixed post-consumer polyolefins with virgin PP or HDPE, EPDM was also considered to be
more effective than EVA [Blom et al., 1998].
A multi-variable analysis of the modification of complex blends of common packaging plastics using
elastomeric modifiers was conducted by Breant [1993], who observed that more additives were
needed to produce a fine morphology of a blend of HDPE, LDPE, PP, PS and PVC, than were
needed in binary blends. It was nevertheless possible to achieve substantial gains in toughness at a
moderate level of modifier addition (10%). For polyolefin-rich blends containing polar polymers such
as PET, the interfacial adhesion and fracture toughness can be increased drastically by adding a
maleic-anhydride-grafted polyolefin [Kalfoglou et al., 1995]. This compatibilisation technique is
employed for engineering alloys based on PP and polyamides [Gonzalezmontiel et al., 1995].
Yao and Beatty [1997] studied the compatibilisation of blends of ABS with 25, 50 or 75% PP using
a combination of poly(propylene-graft-maleic anhydride) and poly(styrene-co-acrylo nitrile-coglycidyl methacrylate). The blends were extremely brittle compared to pure ABS and pure PP and
large additions of the compatibilising system were needed to reduce this problem.
2.5 SEPARATION TECHNIQUES FOR PLASTIC WASTE
Mechanical recycling of plastics usually requires that the plastic material that is considered
worthwhile to recycle is separated from other materials. In many cases, there is also a need to
separate different plastic types (e.g. PVC, PET and polyethylene) from each other. In several cases,
plastics are also sorted by colour in order to improve the physical appearance of the products
derived from post-use material.
2.5.1 Manual sorting
To sort collected plastic waste manually is very labour-intensive. It can be facilitated by material
identification codes but the possibility of human error should not be neglected.
2.5.2 Sorting by density
Tab. 2-2 presents densities of common plastics in neat form. The overall density of a plastic material
can however be altered significantly by the incorporation of fillers or by foaming. Separation of
materials by density in float-sink tanks or hydrocyclones is commonly applied to ground waste
plastics. The polyolefins most commonly used in packaging applications, PP, LDPE and HDPE, are
notoriously difficult to separate efficiently because of the small difference between their densities
[Scheirs, 1998].
Tab. 2-2: Densities of common plastics
14
Plastic type
Density [g/cm3]
Plastic type
Density [g/cm 3]
Poly[1-butene]
0.90
0.90-0.91
0.91-0.94
0.94-0.97
1.03-1.07
1.04-1.07
1.05
1.07-1.08
1.08-1.20
Polyamide 6,6
Polyamide 6
Poly[vinyl acetate]
PMMA
PBT
PET
PVC
Poly[vinylidene chloride]
Polytetrafluoroethylene
1.14
1.14
1.19
1.22
1.30
1.38
1.39-1.43
1.65-1.72
1.70
PP
LDPE
HDPE
ABS
Polystyrene
PP, 20% talc-filled
SAN
Polycarbonate
2.5.3 Air classification
It is possible to sort materials by a combination of density shape using air streams. The technique is
called air classification or air sorting. It can be used to sort e.g. film plastics and paper residues from
ground plastic flakes [Fahrbach & Schnettler, 1996].
2.5.4 Electrostatic separation
Electrostatic charging of different plastics can be utilised to achieve separation. A wide variety of
equipment exists. The most common way of charging the materials is by triboelectric charging which
means that particles are tumbled against one another. This causes some materials to become
positively charged and others to become negatively charged. The materials can then be sorted by
letting them fall freely through an electric field [Stahl & Kleine-Kleffmann, 1997].
2.5.5 Material identification techniques
There are a lot of techniques available for making a fast and reliable identification of plastics, most
often through some kind of spectroscopic fingerprinting. The technology can be used in order to
assist manual sorting, or can be incorporated into automated sorting machines.
15
3 MATERIALS
This chapter defines the basic properties of all virgin plastics and modifiers used in the experimental
work. The source of waste samples is described as well as the procedures for the collection and
sorting of the waste plastics used.
3.1 VIRGIN PLASTICS
3.1.1 Polyolefins used in the multi-variable evaluation of sheet-extruded blends
Tab. 3-1 presents general information about the materials used and their abbreviations. The last three
materials were included in order to be evaluated as possible modifiers for polyolefin blends. Both
polypropylene grades are homopolymers but the poly[1-butene] is a random copolymer containing 2
wt% ethylene.
Tab. 3-1: General information about the materials used in the multi-variable evaluation of
sheet-extruded blends.
Material
Abbr. supplier
grade
HDPE (blow moulding)
LLDPE (1-butene based)
LDPE (high-pressure grade)
HDPE (injection moulding)
PP (extrusion)
PP (injection moulding)
Modifiers:
Poly[ethylene -co-vinyl acetate]a
Very low density polyethylene
(poly[etylene -co-1-octene])
Poly[1-butene]
HDb
LLD
LD
HDi
PPe
PPi
Borealis
Borealis
Borealis
Borealis
Neste
Borealis
EVA DuPont
VLD DuPont-Dow
Elastomers
PB
Shell
HE8331
LE6520
LE1804
HE7012
VB3247C
HF135M
density
2
[kg/m ]
955
919
922
962
908
908
MFR
[dg/min]
0.2b/24c
1.2b
2.1b
12b
3.2d
18d
Elvax 3165
Engage 8150
940
868
0.7b
0.5b
DP8220
897
2.0b
a) 18 wt% vinyl acetate
b) 190 °C, 2.16 kg
c) 190 °C, 21.6 kg
d) 230 °C, 2.16 kg
3.1.2 PP reference and modifying grades
Tab. 3-2 presents information about the virgin reference grades and modifying grades used. All
materials are commercial polypropylene grades supplied by Borealis. HC210P is a highly crystalline
grade intended for thin-walled thermo-formed packaging. In this case it was used as a modifying
grade for recycled PP in order to enhanc e the overall crystallinity, as a way of compensating for the
presence of less rigid PP copolymers in the collected packaging materials. ME210U is a talc -filled
16
compound intended for demanding engineering applications, such as automotive under-the-hood
components.
Tab. 3-2: Information about virgin PP-grades used, as supplied by the manufacturer.
Grade
HF135M
RE220P
BE160M
ME210U
HC210P
Type
homopolymer
908
18
1.55
random copolymer
905
12
1.1
heterophasic
co-polymer
901
13
1.15
20% talcfilled
1050
12
homopolymer
94
85
78
3
Density [kg/m ]
MFR at 230°C/2.16kg [g/10 min]
Tensile modulus [GPa]
Heat deflection temp. at 0.45MPa [°C]
1) Flexural modulus
1
2.65
125
4
2.1
3.1.3 Non-polyolefin plastics
For the evaluation of their compatibility with HDPE, a general purpose grade polystyrene (Neste PS128 with a MFR of 20 dg/min at 200 °C/5.0 kg), and a general purpose grade PET (ICI B95A
Laser+ ), were used.
For studies of polymer/polymer compatibility at low levels of addition, ABS and HIPS were used
together with the talc-filled PP, ME210U, presented in Tab. 3-2. Tab. 3-3 shows information about
the ABS and HIPS, as provided by the supplie rs.
Tab. 3-3: Information about ABS and HIPS grades used.
Material
Supplier
Trade name
Flexural modulus [GPa]
Notched Charpy impact strength [kJ/m2]
Melt flow rate [g/10 min]
Density [kg/m3]
ABS
HIPS
DSM
Polykemi AB
Ronfalin SRA36 POLYstrene 552
2.7
1.9
5
10
46 (220°C, 10kg) 7 (200°C, 5kg)
1040
1040
3.2 MODIFIERS AND FILLERS
Apart from the modifiers presented in Tab. 3-1 (EVA, VLD and PB), two EPDM-elastomers were
also evaluated as impact-modifiers for polyolefin-based blends. Tab. 3-4 presents the constitution of
the elastomers. In the following, they are referred to as EP72 and EP58 respectively to indicate their
different ethylene contents. The high ethylene content of EP72 makes it somewhat crystalline
whereas EP58 is amorphous.
17
Tab. 3-4: Constitution of EPDM elastomers used as modifiers.
EPDM grade
ethylene content [wt%]
propylene content [wt%]
diene content [wt%]
diene type
DSC melting peak [°C]
Mw [g/mol]
DuPont Nordel® 2722
(EP72)
72
22
6
1,4-hexadiene
51
180,000
DSM Keltan® 1446A
(EP58)
58
35
7
ENB (ethylidene norbornene)
none
125,000
In order to increase the stiffness and dimensional stability of recycled PP, talc was used as a mineral
filler in some formulations. Talc powder was supplied by Kebo AB, Sweden, as purum grade,
without any special surface treatment. Talc is a sheet-structure silicate with the chemical formula
Mg3 Si4 O10[OH]2 .
3.3 POST-USE MATERIALS
A Swedish collection system for hard packaging plastics was monitored during one year, from
August 1997 to July 1998. Packaging waste both from households and from industrial sectors was
included. At each collection spot, placed as close to the households as feasible, there are separate
bins for each type of recyclable material, one of which is for hard plastic packaging. Other bins, for
newspapers, glass, cardboard and metal packaging, are similar but marked with different labels. The
labels on the bins gives examples of the types of items for which they are intended. Collection from
the commercial sector and institutions follows the same basic principles regarding source separation
but the practical details can be tailored for individual companies.
Samples were taken at the site where collected plastics were gathered for further sorting by plastic
type. This site is hereafter referred to as “the Lunda Plant“. It is not considered worthwhile to
mechanically recycle film plastics from the household sector, so people are therefore instructed to
dispose of plastic film and bags with the residual household waste, which goes to an incineration
plant, and not in the recycling bins. Film plastics from industrial sectors, e.g. pallet wrapping, is
collected for mechanical recycling but is kept separate from the hard packaging plastics.
3.3.1 The Lunda plant
The plant is located in the industrial area of Lunda in the northern suburb of Stockholm. It is
operated as a joint-venture between two Swedish waste management companies, Ragn-Sells and
HA-industrier. Plastics collected within a 120 km radius around Stockholm, an area with about 2.5
million inhabitants, are treated. Hard plastics are sorted manually into four fractions for mechanical
recycling:
• Colourless HDPE
• PET
18
• PP
• HDPE, mixed colours, including a maximum of 15% PP
The sorted materials are compressed and baled for shipping to customers, i.e. recycling companies
who reprocess the material. The residue goes to a MSW incineration plant.
3.3.2 Collection and sorting of waste samples
Samples were taken at random from a large pile of collected material in a hall at the Lunda plant
where the collection trucks were unloading the collected hard plastic packaging. The materials were
identified on the basis of:
• recycling symbols and information on labels,
• general knowledge about the most common plastics e.g. appearance, mechanical properties,
density and applications,
• behaviour of different plastics when subjected to the flame of a cigarette lighter.
The amount of each type of material was established and weight-percentages were calculated. On
one occasion, in March 1998, the non-film polyethylene and polypropylene fractions of the sampled
materials were recovered for reprocessing. Results from experiments conducted on these materials
make up a significant part of this thesis. The object of the reprocessing was to acquire material
fractions similar to those which would have been the result if automated separation had been applied
on an industrial scale. Fig. 3-1 presents the material treatment scheme that was imitated. The reason
why film plastics were not included in this study was because:
• films are not supposed to be included in this waste stream,
• film plastics can, as well as other light plastics such as expanded polystyrene, be separated from
the rigid polyolefins by conventional sorting methods such as air classification.
In cases were the distinction between polyethylene and polypropylene was uncertain, an infra-red
absorption spectrum of the surface was recorded between 1300 and 1500 cm-1, where there are
two distinct absorption peaks for polypropylene and only one for polyethylene.
19
1: Material
collection
2: Brief manual
inspection
3: Grinding
Large metal
objects, etc.
6: Electrostatic
separation
PE+PP
PE
PP
5: Drying and air
classification
4: Washing and
density separation
Film plastics,
foam plastics,
paper, textiles.
Heavy materials
(PET, PVC, PS,
metals, etc.)
7: Extrusion compounding
with melt filtration
7: Extrusion compounding
with melt filtration
7: Extrusion compounding
with melt filtration
Granulated
materials for
conversion to
recycled
products
Fig. 3-1: Potential treatment scheme for material acquired through a collection system for
rigid plastic packaging waste
The non-film PE and PP fractions were washed with hot water, and paper labels were removed.
These two fractions were then ground using a Moretto ML 18/10 granulator equipped with a 6 mm
screen. If automated separation methods are to be applied to this type of waste stream, it will most
certainly include some kind of density-based sorting technique. Therefore all non-floating PP was
removed by putting the ground flakes into a water-bath. No non-floating PE was detected.
20
4 EXPERIMENTAL METHODS
4.1 COMPUTER-AIDED EXPERIMENTAL DESIGN AND EVALUATION
The experimental planning and evaluation was done using the software “Modde 3.0“, supplied by
Umetri AB. This si a Windows-based software for response surface methodology, i.e. statistical
experimental design and multivariate analysis. It is an aid to the investigation and optimisation of
complex products and processes where many factors affect the results. The factors that were varied
were the percentage of each of the nine polymers in the blends, presented in APPENDIX I. An
experimental set-up was designed with the following constraints: 1) The sum of polypropylene grades
was not allowed to exceed 30%. 2) The sum of injection-moulding grades was not allowed to
exceed 30%. 3) The sum of the modifiers, EVA, PB and VLD, was not allowed to exceed 20%.
The composition of the blends included in the studies of sheet-extruded polyolefin blends is
presented in APPENDIX I. Based on the experimental results, models for each response (tensile
strength, modulus, etc.) were generated by “Modde“. The models are polynomial functions of the
factors (HDb, LD, LLD, etc.) with a constant, first degree coefficients, second degree coefficients
and coefficients for interaction parameters (e.g. HDb x LLD).
4.2 PREPARATION OF TEST SPECIMENS
4.2.1 Extrusion compounding
Recycled PE and PP flakes and their blends were compounded and repelletised using a counterrotating, intermeshing twin-screw extruder; Brabender DSK 35/9D. The barrel temperature was 200
°C and the screw speed was 75 rpm (screw diameter = 35 mm, L/D=9). The extruder was
equipped with a die having four circular holes with a diameter of 4 mm each. After exiting the die, the
extrudate was cooled in a water bath and cut into pellets. A compound composed of 50 wt%
recycled PP flakes and 50 wt% talc was also prepared using the same parameters by dry-blending
the components in the extruder hopper. The talc-compound was then oven-dried at 125 °C for 30
min. Blends of virgin polyolefins for subsequent sheet extrusion were compounded at a screw speed
of 60 rpm and a barrel temperature of 200-210 °C.
4.2.2 Injection moulding
Virgin reference materials were moulded into tensile test bars (ASTM standard D 638M type M-I,
3.2 mm thick) using a Battenfeld PLUS 250/50 reciprocating screw type injection-moulder (screw
diameter = 22 mm, L/D=16). The barrel temperature was 200-220 °C and the mould temperature
was 30 °C. Total cycle time was 29 s. The reprocessed pellets were converted into test bars by
injection-moulding using the same procedures as for the virgin reference materials. In cases where
the PP-modifying grade (HC210P) was used, this was added by dry-blending in the injection-
21
moulder hopper, without prior compounding. The mould also contained a cavity for impact test
specimens (4x6x50 mm).
In the case of blends based on the HDPE-grade HE7012 (Paper II), compounding and injectionmoulding was done in one step in the Battenfeld machine. The barrel temperature settings were 200
°C (first heating section) and 250 °C (second heating section). This higher temperature was used
because some blends included up to 20% PET, the crystalline melting-point of which is in that region.
Studies of the moulded specimens using scanning electron microscopy showed that this was sufficient
to cause the PET to disperse. The mould temperature was 20 °C and the total cycle time was 40 s.
4.2.3 Sheet extrusion
The compounded materials where then extruded into sheets of approximately 0.5 mm thickness using
an Axon single -screw extruder (screw diameter =18 mm, L/D=20, barrel temp. profile: 170-190200-200 °C, screw-speed: 120 rpm) equipped with a slot die (width: 52 mm, gap: 0.55 mm,
temperature: 180 °C). The haul-off speed was 2.3 m/min and the throughput was approximately 2.7
kg/h.
4.2.4 Simulated recycling of ABS and talc-filled PP:
Tensile test bars of ABS and talc -filled PP were injection moulded at a barrel temperature of 220 230°C. Injection moulded ABS was subjected to thermo-oxidative ageing in air at 80°C for two
months and then ground into flakes for reprocessing. PP and HIPS were ground and reprocessed
without thermo-oxidative ageing. Ground ABS and PP were commingled with defined proportions of
the foreign materials and then compounded and repelletised using a the twin-screw extruder (same as
in section 4.2.1). The barrel temperature was 210°C and the screw speed was 45 rpm. ABS-based
blends with 3, 6, and 9% PP and with 4, 8, and 12% HIPS were prepared as well as PP-based
blends with 3, 6, and 9% ABS. The reprocessed pellets were converted into test bars using the
same parameters as for the first injection-moulding step.
4.3 MECHANICAL TESTING
4.3.1 Tensile tests
Tensile testing of sheet-extruded materials was performed at 23 °C according to ISO 1184. Dumbbell shaped specimens, with 33 mm long and 6 mm wide narrow sections, were punched out of the
sheets with the long axis coinciding with the machine direction. The rate of grip separation was 200
mm/min. The yield strength was calculated at an offset of 5% whereas yield properties reported for
injection-moulded specimens were calculated at the zero-slope point of the stress-strain curve.
4.3.2 Impact tests
Notched Charpy impact tests were performed according to ISO 179/2B. A pendulum impact tester
with impact energy of 3.8 J was employed.
Tensile impact tests were also performed on the same type of injection-moulded test specimens as
those used for the normal tensile tests. The energy to break was measured with a pendulum impact
testing instrument, Amsler RKP 300. The impact velocity was 1.8 m/s and the impact energy was 19
J for ABS, talc-filled PP and blends based on these materials. For blends based on HDPE, the
22
impact velocity was 2.7 m/s and the impact energy was 40 J. Ten specimens of each material
composition were tested.
4.3.3 Dynamic mechanical tests
Dynamic mechanical properties in the tensile mode were measured by scanning from -50 to 100 °C
using a Polymer Laboratories Mk II Dynamic Mechanical Thermal Analyser. The frequency was 1
Hz and the heating rate was 3 °C/min. Test strips with a rectangular cross-section were cut in the
machine direction of sheet-extruded materials.
4.3.4 Heat deflection tests
The dimensional stability of the materials at elevated temperatures was assessed using a Polymer
Laboratories Mk II dynamic mechanical thermal analyser. The samples, 40x3.2x2.6mm were cut
from the central part of tensile test bars. No cyclic load was applied, only a static load corresponding
to a tensile stress of 0.45 MPa. The initial distance between the grips was 25 mm. The displacement
was recorded during heating from room temperature at a rate of 2 °C/min.
4.3.5 Tear-propagation resistance
Tear propagation resistance (TPR) was measured in accordance with ASTM D1938 on sheetextruded polyethylenes and blends. The longitudinal slits in the test specimens from which the tears
propagate were cut parallel to the machine direction. Five specimens of each material composition
were tested. An Instron 5566 tensile tester was employed.
4.4 DIFFERENTIAL SCANNING CALORIMETRY (DSC)
The crystallisation of PP, PE and their blends was studied using a Mettler Toledo DSC820. 40 µl
aluminium pans with 4 mg material cut from the test-bars were tested in an inert atmosphere (N2).
The samples were quickly heated to 180 °C, held at that temperature for 1 min and then cooled at
either − 2, −10, − 20, − 40 or −60 °C/min. The onset temperature of crystallisation was calculated
using the “STARe“-software system supplied by Mettler Toledo.
The melting of virgin and recycled PP was also studied. The samples were quickly heated to 180 °C,
held at that temperature for 1 min and then cooled at −10°C/min. After this treatment, the samples
were stored at room temperature for 7 days and then put back into the DSC apparatus for another
heating at a rate of 10°C/min. The melt enthalpy ( ∆ Hm) and melting peak temperature for this second
heating were calculated using the “STARe“-software.
The morphology of the HDPE-EPDM system was studied using a Perkin-Elmer DSC 7. In this case,
10 mg of material from tensile test bars was placed in 50 µl pans with holes. The samples were
quickly heated to 180 °C, held at that temperature for 1 min and then cooled to 20 °C at a rate of
either 10 °C/min or 50 °C/min. The pans were then taken out of the apparatus and stored at room
temperature for 3 days before the DSC heating scan was recorded at a rate of 10 °C/min.
4.5 SCANNING ELECTRON MICROSCOPY
Fracture surfaces of blends of recycled PE and PP were observed with a Jeol JSM-5400 scanning
electron microscope, after sputtering with a gold/palladium alloy. The applied accelerator-voltage
23
was 10 or 15 kV. Notched tensile test-bars were fractured for this purpose using a Charpy-type
impact tester. Some of the specimens were fractured at room-temperature and some after cooling in
liquid nitrogen.
The collected PP packaging material included ketchup bottles with a co-extruded EVOH-layer
(poly[ethylene-co-vinyl alcohol]) as oxygen barrier. In order to study the distribution of EVOH in the
recycled PP, thin slices of the test-bars were cut with a microtome. The slices were immersed in a
well stirred 50/50 mixture of acetic anhydride and acetic acid at 110 °C for 30 min. The purpose of
this treatment was to acetylate the hydroxyl groups of EVOH according to the reaction shown in Fig.
4-1. This causes EVOH domains to swell permanently because acetyl groups are more bulky than
hydroxyl groups. After sputtering with a gold/palladium alloy, the slices were examined in the
scanning electron microscope at an accelerator voltage of 15 kV.
O
O o
O
||
||
||
OH + CH3-C-O-C- CH3 ⇒
O
o
||
O-C-CH3 + CH3 -C-OH
Fig. 4-1: Reaction scheme for acetylation of hydroxyl groups of EVOH using acetic
anhydride.
24
5 RESULTS AND DISCUSSION
5.1 COMPATIBILITY OF SINGLE GRADE PE/PP BLENDS
Tab. 5-1 shows mechanical properties of sheet-extruded HDPE and blends with two different PP
grades. Both PP grades increased the stiffness and yield strength but there were large differences in
their effects on ultimate properties, elongation at break (εb) and tensile strength at break (σb). The
injection-moulding grade caused very severe embrittlement whereas the extrusion grade only
hampered the tear propagation resistance (TPR) to a limited extent. The results illustrate a very
significant difference between HDPE/PP blends and blends of HDPE and non-polyolefins. In Paper
II it was observed that polystyrene and PET reduced the fracture toughness of HDPE to about the
same extent, despite the fact that pure polystyrene is much more brittle than pure PET. This suggests
that the minor component acts only as a passive filler in such blends and that the mechanical
properties are governed by the matrix and the level of interfacial adhesion. The results presented in
Tab. 5-1 show that this is not the case for HDPE/PP blends.
Tab. 5-1: Mechanical properties of bottle grade HDPE, and its blends with PP grades.
Material
Pure HDPE
30% PP, extrusion grade
30% PP, injection moulding
E-modulus
[MPa]
1239±168
1411±76
1330±94
σy
[MPa]
24.69±0.42
26.69±0.24
26.92±0.41
σb
[MPa]
18.08±1.53
19.92±1.57
10.21±1.49
εb
[%]
634±306
706±172
76±25
TPR
[N/mm]
111.1±15.4
48.3±2.8
10.1±5.6
Tab. 5-2 presents mechanical properties of sheet-extruded LDPE and blends with PP. The two PP
grades affected the material in essentially the same way. The TPR was very low for both blends
despite the fact that elongation at break was fairly high. This may be a result of anisotropy effects,
since the tensile test was performed in the machine direction while the tear test entailed a local
deformation in the transversal direction at the front of the propagating tear. The results emphasise
that conclusions regarding polymer compatibility should not be drawn on the basis of a single type of
experiment.
Tab. 5-2: Mechanical properties of film blowing grade LDPE, and its blends with PP
grades.
Material
Pure LDPE
30% PP, extrusion grade
30% PP, injection moulding
E-modulus
[MPa]
238±19
489±32
531±45
σy
[MPa]
8.79±0.14
14.33±0.24
15.19±0.20
25
σb
[MPa]
13.54±0.52
13.12±0.42
12.78±0.27
εb
[%]
769±29
496±40
442±99
TPR
[N/mm]
54.6±4.0
7.4±1.7
8.2±1.6
Tab. 5-3 presents mechanical properties of sheet-extruded LLDPE (poly[ethylene-co-1-butene])
and blends with PP. This type of polyethylene is much tougher than high-pressure LDPE in its pure
form. The compatibility with PP is very good, especially for the injection-moulding grade. This may
seem rather surprising in view of the results for the blends with HDPE, who showed the best
compatibility with the extrusion grade. A possible explanation could be that the blend of LLDPE and
the injection moulding grade represented a better melt-viscosity match, and this provide better
mixing.
Tab. 5-3: Mechanical properties of film blowing grade (butene-based) LLDPE, and its
blends with PP grades.
Material
Pure LLDPE
30% PP, extrusion grade
30% PP, injection moulding
E-modulus
[MPa]
301±34
540±36
528±22
σy
[MPa]
9.71±0.13
14.31±0.38
14.13±0.34
σb
[MPa]
26.05±0.94
28.56±1.00
24.14±0.94
εb
[%]
1290±42
1160±51
1300±55
TPR
[N/mm]
108.0±3.2
81.2±4.3
131.9±10.4
5.2 MECHANICAL PROPERTIES OF MULTI-COMPONENT POLYOLEFIN
MIXTURES
In section 5.1, it was concluded that LLDPE was not only tougher than LDPE in the pure form but
also much more compatible than LDPE with PP in binary mixtures. This suggests that the mechanical
properties of multi-component polyolefin mixtures containing both PE and PP grades as well as both
film plastics and rigid plastics, would be favoured if the PE-film component were dominated by
LLDPE. In order to investigate this issue further, computer simulations were made for a set of
hypothetical polyolefin blends consisting of 50% mixed HDPE and PP grades and 50% LD-grades
(LDPE, LLDPE or blends thereof). The simulations were made in “Modde 3.0“ (see section 4.1),
on the basis of experimental results presented in APPENDIX I.
Fig. 5-1 shows how tensile strength (σ b) was affected by the relation between LDPE and LLDPE in
a complex polyolefin mixture. Blends with high levels of PP showed a strong dependence on the
LDPE/LLDPE ratio which is in agreement with observations for binary mixtures. When the
HDPE/PP component was dominated by HDPE, especially the blow-moulding grade, the reduction
in the tensile strength was less substantial when LLDPE was substituted by LDPE.
26
Tensile strength [MPa]
28
26
24
22
20
35% HDb, 5% HDi, 5% PPe, 5% PPi
15% HDb, 25% HDi, 5% PPe, 5% PPi
10% HDb, 5% HDi, 25% PPe, 10% PPi
15% HDb, 5% HDi, 5% PPe, 25% PPi
18
50%0 LLDPE
0% LDPE
20
40
60
80
0% LLDPE
100
50% LDPE
Fig. 5-1: Computer-predicted values for the tensile strength of po lyolefin mixtures
containing 50% LDPE/LLDPE as a function of the portions of LDPE and LLDPE.
Fig. 5-2 shows how the TPR was affected by the LDPE/LLDPE ratio for the same mixtures as in
Fig. 5-1. In this case also, the reduction with increasing LDPE content was less prominent when the
HDPE/PP component was dominated by the HDPE blow-moulding grade. Substitution of LLDPE
by LDPE in blends with a large amount of PP injection-moulding grade gave a reduction by over
50% in TPR.
120
35% HDb, 5% HDi, 5% PPe, 5% PPi
15% HDb, 25% HDi, 5% PPe, 5% PPi
10% HDb, 5% HDi, 25% PPe, 10% PPi
15% HDb, 5% HDi, 5% PPe, 25% PPi
TPR [N/mm]
100
80
60
40
50%
0 LLDPE
0% LDPE
20
40
60
80
0% LLDPE
100
50% LDPE
Fig. 5-2: Computer-predicted values for the tear propagation resistance (TPR) of polyolefin
mixtures containing 50% LDPE/LLDPE, as a function of the portions of LDPE and
LLDPE.
27
5.3 MODIFICATION OF MULTI-COMPONENT POLYOLEFIN MIXTURES FOR
IMPROVEMENT OF FRACTURE TOUGHNESS
This section describes the effect of various additives that can be added to recycled polyolefin blends
in order to counteract embrittlement caused by poor compatibility.
5.3.1
Modification of multi-component polyolefin blends using EVA, VLDPE and poly(1butene)
Tab. 5-4 and Tab. 5-5 presents mechanical properties for 8 examples of mixed polyolefin fractions
that may be encountered in recycling operations, before and after the addition of 20% modifier. The
data are computer-generated predictions based on the experimental results presented in
APPENDIX I. The general effect of the modifiers is that tensile modulus is reduced while the ultimate
tensile strength and TPR are increased. This effect is most pronounced for VLDPE.
Tab. 5-4: Computer-generated predictions of ultimate tensile strength for eight
hypothetical polyolefin blends, before and after the addition of 20% modifier.
1
2
3
4
5
6
7
8
HDb
LLD
30
40
40
20
40
60
30
30
10
10
0
10
5
5
10
25
Polyolefin composition (%)
LD
HDi
PPe
20
15
15
40
10
15
10
15
10
5
10
0
10
10
30
0
Ppi
Unmodified
10
10
15
10
0
0
10
30
20
20
20
20
35
10
10
0
Mean value
Standard deviation
22.9
21.7
22.0
21.8
17.5
20.3
23.2
24.4
21.7
2.1
σb (MPa)
20%
20%
EVA
VLD
24.8
28.2
25.8
28.9
25.7
28.8
23.2
27.3
25.3
29.7
26.7
29.3
22.5
27.4
23.3
29.7
24.7
28.6
1.5
0.9
20%
PB
26.3
27.2
28.3
23.3
26.8
26.9
26.8
28.6
26.8
1.6
Tab. 5-5: Predictions of tensile modulus and tear propagation resistance for eight
hypothetical polyolefin blends, before and after the addition of 20% modifier (blend
compositions given in Tab. 5-4).
1
2
3
4
5
6
7
8
Mean
Std. dev.
Unmodified
656
725
736
614
934
810
767
825
758
101
Tensile Modulus (MPa)
20% EVA 20% VLD
504
539
546
467
684
577
575
560
556
64
399
430
427
380
554
457
458
482
449
54
20% PB
497
539
545
466
658
572
563
606
556
60
28
Unmodified
68
67
69
58
36
82
90
54
65
17
TPR (N/mm)
20% EVA 20% VLD
67
66
65
59
42
86
88
61
67
15
82
83
87
77
70
93
94
79
83
8
20% PB
64
63
66
62
55
50
64
75
62
7
Another important finding of Tab. 5-4 and Tab. 5-5 is that the standard deviations decrease when
modifiers are added. This implies that modifiers also have the ability to smooth out the effects of
variations in the composition of the primary polyolefin mixture. VLDPE is also the most effective
modifier in this aspect.
5.3.2 Modification of HDPE-rich blends using EPDM
Paper II deals with the modification of HDPE-rich blends using two different EPDM elastomers.
One elastomer, called EP58, had a low ethylene content and was therefore fully amorphous. The
other elastomer, called EP72, had a higher ethylene content which made it more similar to
polyethylene and it could therefore either crystallise by itself or co-crystallise with HDPE.
The effect of EPDM elastomers on HDPE crystallinity was studied by DSC. Fig. 5-3 shows the melt
enthalpy, expressed as J/g HDPE, for two different thermal histories. Theoretically, EP58 should not
affect this value at all since its irregular structure prevents it from crystallising, and this was instead the
case for samples cooled at the faster rate.
235
230
225
220
215
EP72 10 °C/min
EP58 10 °C/min
EP72 50 °C/min
EP58 50 °C/min
210
205
0
1
2
3
EPDM-level [%]
4
5
Fig. 5-3: Melt enthalpy, adjusted relative to the HDPE-content of the samples, as a
function of EPDM-content of binary blends.
For slowly cooled samples, the general level of crystallinity was higher because of the longer time
available for chain-segments to organise. A clearly distinguishable inhibitory effect of EP58 on the
ability of HDPE to crystallise was observed. This observation suggests that some otherwise
crystallisable HDPE was “trapped“ within the amorphous elastomer. For EP72, that contains
crystallisable segments due to its higher ethylene-content, the effect of different thermal histories was
29
more pronounced. No melting peak was seen in the region of 50 ° C, where pure EP72 melts, for the
blends of EP72 and HDPE, and this indicates that co-crystallisation had occurred.
The blends that where modified with EPDM were based on a HDPE injection moulding grade,
HE7012 (see Tab. 3-1), and included up to 20% of either PP, PS or PET. The effect of EPDM
addition was about the same in all cases, an increase in the impact strength at the expense of a
reduction in rigidity (reduction of tensile modulus and yield strength). Fig. 5-4 shows a summary of
data from mechanical tests in Paper II. Data for the addition of 2% EP58 fall in approximately the
same region as data for addition of 5% EP72. This means that the desired toughening effect can be
reached at a lower level of addition using EP58. As in the results presented in section 5.3.1, the
modifier that reduced rigidity the most also increased fracture toughness the most.
60
5% EP72
2% EP58
50
5% EP58
40
30
y = -2.91x - 0.539
20
10
0
-20
-15
-10
Change of yield strength [%]
-5
-10
0
Change of impact strength [%]
70
2% EP72
-20
Fig. 5-4: Effect of additions of EPDM to HDPE and HDPE-rich blends on yield strength
and tensile impact strength (data from Paper II).
5.4 CHARACTERISATION OF MATERIALS SAMPLED FROM THE SWEDISH
COLLECTION SYSTEM FOR RIGID PLASTIC PACKAGING WASTE
Tab. 5-6 presents results from all the samplings during 1997 and 1998. Tables for individual
samplings are presented in APPENDIX II. PET was not so frequent as in similar collection schemes
in other countries because there is a separate bring-back system, based on deposit fees, for PETsoda bottles in Sweden. PET that ended up in this waste stream was mostly bottles for other than
carbonated soft-drinks, e.g. vegetable oils, concentrated lemonade and household chemicals. The
amount of correctly sorted materials (rigid plastic packaging) varied between 75.5% and 87.7% in
the individual samplings.
30
Tab. 5-6: Composition of samples from the collection system for rigid plastic packaging in
the Stockholm region. August 1997 - July 1998. Total amount analysed: 197 kg.
MATERIAL:
PET; bottles
PET; other packaging(trays, blisters)
HDPE; bottles, jugs
HDPE; film
HDPE; other packaging (trays, caps, buckets, cups)
HDPE; non-packaging
PVC; flexible packaging
PVC; rigid packaging
PVC; non-packaging
LDPE/LLDPE; film
LDPE/LLDPE; lids
PP; bottles
PP; buckets, trays, cups
PP; lids, caps
PP; film
PP; non-packaging
PS; cups, lids, etc.
PS; expanded
Other plastics (ABS, PMMA, PU, etc.)
Total plastics
Total rigid plastic packaging
Paper
Other organic materials
Glass
Metal
Ceramics
Fraction of total (wt%)
6.0
1.0
7.0
32.9
0.2
1.6
0.6
35.3
0.1
1.4
1.8
3.3
6.4
1.4
7.8
3.2
18.7
7.2
0.2
1.9
31.2
6.5
1.8
8.3
1.3
94.2
81.7
3.7
0.6
0.2
1.2
0.1
5.4.1 Non-film PE fraction
The non-film polyethylene fraction was dominated by extrusion blow-moulded HDPE containers
such as bottles and jugs. Not all non-film polyethylene was high-density grade material. There were
also some flexible lids for jars, ice-cream boxes, etc., made out of LDPE or LLDPE. These made up
about 4% of the non-film polyethylene fraction. Tab. 5-7 presents some properties of the non-film
PE fraction and virgin reference materials, normally used for extrusion blow-moulded HDPE
containers. Because the recycled PE contains some flexible grades, mainly LDPE lids, it was less stiff
than the references but the yield strength was actually higher. For these materials, the yield strength is
equivalent to the maximum load during the tensile test.
31
Tab. 5-7: Properties of the non-film PE fraction in packaging waste, and virgin HDPE
reference grades. Tensile properties tested at 500 mm/min.
HE8361
3
HE8331
3
PE
Density=963 kg/m
density=955 kg/m
recycled
1.67
31.9
7.8
47.3
65.4
1.55
32.7
9.2
39.0
59.4
1.46
33.2
11.2
38.3
57.6
Tensile modulus [GPa]
Yield strength [MPa]
Elongation at yield [%]
Elongation at break [%]
Crystallinity, DSC [%]
5.4.2 Non-film, floating PP fraction
The polypropylene fraction was dominated by injection-moulded items such as lids, caps and jars.
The non-floating PP-fraction (PP including mineral fillers) made up 3% of all non-film PP. Fig. 5-5
shows stress-strain curves from tensile tests on recycled PP and virgin reference grades. The figure
illustrates the broad range of properties that common PP-grades display. The recycled PP, which is
a mixture of many different grades, has mechanical properties intermediate between those of
homopolymers, random copolymers and heterophasic copolymers.
40
35
Stress [MPa]
30
25
20
15
Homopolymer
Random copolymer
Heterophasic copolymer
Recycled
10
5
0
0
10
20
30
40
50
Strain [%]
60
70
80
Fig. 5-5: Stress-strain curves at a cross-head speed of 500 mm/min for recycled PP and for
virgin reference grades.
The relation between the non-film PE and PP fractions, after removal of heavy materials by flotation,
was 56/44 for the sampling conducted in March 1998. No exact value for this ratio can be
calculated for the entire series of samplings because no distinction between floating and non-floating
polyolefins was made on other sampling occasions. If it is assumed that the non-floating PP fraction
was 3% of all non-film PP for all samplings, the ratio can be calculated to 55/45.
32
5.5 CRYSTALLISATION AND MICROPHASE STRUCTURE OF PE/PP FROM
RIGID PLASTIC PACKAGING WASTE
The morphology of semi-crystalline plastics is largely dependent on the conditions under which the
material has crystallised. For blends of PP and HDPE, this is highly relevant since the two polymers
affect the crystallisation of each other [Olley et al., 1979]. It is therefore appropriate to study the
crystallisation of such blends in order to be able to interpret morphological observations.
5.5.1 DSC-observations
Fig. 5-6 shows the onset of crystallisation, measured by DSC, as a function of composition and
cooling rate for PE/PP derived from packaging waste. Pure PP normally crystallises at about the
same temperature as HDPE when cooled at the same rate [Paper II]. In this case, the onset
temperature for PP is higher due to the presence of nucleating agents in many PP-grades used in
packaging applications. The role of nucle ating agents is to enhance clarity, modify mechanical
properties and/or reduce cycle times during processing. Common nucleating agents for PP are talc
[Menczel & Varga, 1983], sodium benzoate [Bartczak et al., 1985] and various sorbitol derivatives
[Kim & Kim, 1991, Shepard et al., 1997]. Pigments can also have a nucleating effect on PP
[Silberman et al., 1995].
132
-2°C/min
-20°C/min
-60°C/min
Onset temperature [°C]
130
-10°C/min
-40°C/min
128
126
124
122
120
118
116
0
20
40
60
PP-content [%]
80
100
Fig. 5-6: Crystallisation onset temperature for the first peak (either the PP peak or
combined PE and PP peaks) as a function of cooling rate and blend composition.
An interesting feature of Fig. 5-6 is that the effect of nucleating agents on the onset temperature
vanished when the PP-content was 40% or lower (at − 10°C/min). This can be observed as a step in
the curve of onset temperature versus composition. At faster cooling rates, this step decreased in
magnitude and was shifted towards a higher PP-content. Tab. 5-8 presents three theories for why
the nucleating effect on the onset temperature for crystallisation of PP vanished in PE-rich blends.
33
Tab. 5-8: Possible theories for the observed depression of PP nucleation in PE-rich blends.
“Dilution theory“
When the PE-content increases, the overall concentration of nucleating agents
decreases and eventually becomes insufficient for nucleation of the PP.
“Entropic theory“
When PP crystallises, the entropy decreases with the formation of an ordered
structure and enthalpy decreases because crystallisation is an exothermic
reaction. Both these factors, which counteract each other in the expression for
the free energy change (∆G=∆H-T∆S), are proportional to the PP-content of
the blend. When PP crystallises out of a mixed melt containing PE, there is yet
another factor that causes the entropy of the system to decrease; namely that
crystallisation implies phase separation since the polymers cannot cocrystallise. When the PP content decreases below a certain point, the ∆ H
becomes to small to make up for this extra entropy effect.
“Kinetic theory“
Due to entanglements between PE and PP chains, there is a retardation of the
speed with which the PP-chains can organise. This effect is more pronounced
the larger the PE-content of the blend.
In order to test the dilution theory, a 70/30 recycled PE/PP blend was prepared, to which 0.5% talc
was added in the compounding step. Talc is well known to be a powerful nucleating agent for
polypropylene [Menczel & Varga, 1983]. Fig. 5-7 shows DSC-curves illustrating the crystallisation
of this blend and the corresponding blend without the further addition of nucleating agent.
160
140
120
Heat flow
100
1 W/g
80
60
70% PE, 30% PP
40
70% PE, 30% PP + talc
20
70% PE, 30% PP, hypothetical
0
105
110
115
120
Temperature [°C]
34
125
130
Fig. 5-7: DSC cooling scans for 70/30 PE/PP blends, with and without further addition of
nucleating agent (0.5% talc), compared with a hypothetical curve based on the
corresponding curves for pure recycled PE and PP.
In Fig. 5-7 there is also a theoretical curve, calculated from measurements of pure recycled PE and
pure recycled PP, that shows how the curve for the blend would have looked if the two polymers
had crystallised independently. Fig. 5-7 illustrates clearly that further addition of nucleating agents did
not affect the onset temperature but that the rate of crystallisation was enhanced once the process
was initiated. This can be concluded from the higher relative magnitude of the crystallisation peak in
the presence of additional nucleating agent The observed effect can possibly be explained by an
enhanced thermal conductivity due to the addition of talc. From the experiment, it can be concluded
that the dilution theory has virtually no relevance.
The kinetic theory seems to be highly relevant when studying the curve of onset temperature versus
composition at a cooling rate of − 2 °C/min (also shown in Fig. 5-6). At this very slow cooling rate,
there was enough time for the PP to crystallise ahead of the PE. These experimental findings which
support the kinetic theory do not however exclude the entropic effect from having a significant
influence on the suppression of PP-nucleation.
A reason why a slow cooling rate allows PP to crystallise ahead of PE at a lower PP-content may be
because this provides time for the melt to separate and form PE-rich and PP-rich phases. This
behaviour has been studied by Cham et al. [1994] for blends of polypropylene and poly(1-butene).
In order to test this hypothesis, the DSC-scan at − 2 °C/min for the blend with 30% recycled PP was
repeated, but with a 20 minutes halt at 140 °C. Fig. 5-8 shows a comparison of the results of these
experiments.
160
140
166
Heat flow
120
164
100
162
1 W/g
160
80
158
60
70% PE, -2°C/min
40
70% PE, -2°C/min,
20 min [email protected]°C
156
154
152
150
20
0112
123
114
116
118
120
122
124
125
126
127
128
129
130
131
132
133
134
136
Temperature [°C]
Fig. 5-8: DSC cooling scans for 70/30 PE/PP blends at − 2 °C/min, with and without a 20 min
halt at 140 °C. The inserted box features an enlargement of the PP crystallisation peaks
(out of scale).
35
It is evident that the halt at 140 °C made the PP-fraction more prone to crystallisation. The onset
temperature for PP increased from 129.5 to 130.3 °C and there was a larger separation of the PP
and PE crystallisation peaks. The onset temperature of the large peak, the one for crystallisation of
PE, was not shifted but it is nevertheless apparent that the crystallisation of PE was also facilitated by
the extra time available for segregation of the polymers. This conclusion can be drawn from the fact
that the PE-peak of Fig. 5-8 reaches its maximum value at a higher temperature (122.31 vs. 122.18
°C) and a greater magnitude (1.176 vs. 1.132 W/g) and levels off faster for the experiment with the
20 minutes halt.
5.5.2 SEM-observations
Fig. 5-9 shows a fracture surface of a test-bar with 50% recycled PP and 50% recycled PE,
fractured at room temperature. Droplets of the dispersed phase, as well as holes left from torn out
droplets, are clearly visible. Further SEM-studies confirmed that a droplet structure was prevalent
for all blends with a PE-content equal to or less than 50%. It was further observed that the droplet
structure was not clearly revealed if specimens were fractured at cryogenic temperatures, after
cooling in liquid nitrogen [Paper III]. This indicates that there must have been a considerable amount
of interaction between PP and PE, so that when the mobility of chain-segments was frozen in, the
crack could propagate without any particular preference for droplet/matrix interfaces.
Fig. 5-9: SEM-image of a 50/50 PE/PP blend derived from packaging waste and fractured
at room temperature.
Fig. 5-10 presents a SEM -image of the blend with 60% recycled PE, fractured in the same way as
the specimen portrayed in Fig. 5-9. Note that the magnification is greater in Fig. 5-9. The SEMimages show that there was a major change in microphase structure, which coincided with the
36
composition range at which the nucleating effect for PP ceased to cause PP to crystallise ahead of
PE at a cooling rate of − 40 °C/min (see Fig. 5-6). The conditions under which the material was
solidified during injection moulding cannot however be exactly imitated in a DSC experiment.
Fig. 5-10: SEM -image of a 60/40 PE/PP blend derived from packaging waste and fractured
at room temperature.
Further SEM studies showed that a layered structure very similar to that in Fig. 5-10 was also
present at PE-levels of 70% and 80%. Fig. 5-11 shows a SEM image of the blend with 80% PE.
For blends with 90% or 95% PE, it was difficult to produce distinct fracture surfaces at room
temperature because of the great ductility of these materials. No comparable SEM study could
therefore be conducted. As was the case for the droplet structure of PP-rich blends, the layered
structure of PE-rich blends was not observed if specimens were fractured at cryogenic temperatures.
37
Fig. 5-11: SEM -image of an 80/20 PE/PP blend derived from packaging waste and fractured
at room temperature.
Fig. 5-10 indic ates that, at a PE-content of 60%, the morphology consisted of layers, about 0.2 µm
thick and 0.6µm apart, who apparently had been pulled out slightly during the propagation of the
fracture. It can therefore be concluded that these layers must have been more ductile than the
intermediate material. When the specimen was cooled in liquid nitrogen, the ductility must have been
lost so that the “pull-out effect“ observed in Fig. 5-10 was absent. This explains why the layered
structure could not be detected on cryogenically fractured specimens. The layers should not be
confused with crystal lamellae, which are normally about 10 nm thick for both polyethylene [Gedde,
1995] and polypropylene [Dolgopolsky et al., 1995] as well as for PP random copolymers with
minor amounts of ethylene [Laihonen et al., 1997]. Blending of polyethylene and polypropylene has
been observed to decrease the lamellar thickness [Flaris et al., 1993]. It is more likely that the layers
represent stacks of crystal lamellae.
The formation of the layered structure seen in Fig. 5-10 can possibly be explained as follows: The
crystallisation starts with PE while the concentration of PP in the melt is too low for the formation of
stable PP crystallites. This can be either because of lack of time (kinetic theory) or because of
unfavourable thermodynamics (entropic theory) or a because of combination thereof (see section
5.5.1). PE forms stacks of crystal lamellae. Crystallisation extracts PE from the melt and
consequently the local concentration of PP increases near the surface of the stack of PE lamellae.
When the local PP concentration has increased sufficiently, PP starts to crystallise. This does not
terminate the crystallisation of PE so the result is a mixture of PP and PE crystallites and amorphous
material at the surface of the original stack of lamellae. This process lowers the local PP-content of
the melt back towards the original value, making the crystallisation of PP halt.
38
If this theory holds, the more ductile layers would be those composed of stacks of PE-lamellae. The
reason why a layered structure was not produced during solidification of PP-rich blends is apparent
in view of the DSC-measurements presented in Fig. 5-6. At compositions where PP crystallised
ahead of PE, the temperature at the start of the crystallisation process was considerably higher than
the onset temperature for the crystallisation of PE, even in its pure form. Since crystallisation of PP is
an exothermic process, heat was evolved that counteracted the temperature reduction caused by
heat transfer to the mould surface. This provided time for crystallisation of PP before the temperature
could be reduced to the level at which stable PE-crystallites could be formed. That the PE-content of
the melt increased, as crystallisable PP-molecules were extracted from the melt, did not change the
situation. This means that phase segregation took place while PE still was completely molten and
therefore formable. The PE-phase (or rather the PE-rich phase) consequently assumed the shape at
which the surface free energy was minimised, i.e. spherical droplets.
Indications of a layered structure were also reported by Gupta et al. [1982] for injection-moulded
blends of PP with 5-25% HDPE, observed by SEM after slow deformation in tensile tests.
It should be pointed out that the layered structure, unlike the morphology of the blends with PEcontent ≤ 50%, was anisotropic. The fracture surfaces of Fig. 5-10 and Fig. 5-11 were
perpendicular to the long axis of the tensile test bars. Fig. 5-12 shows a fracture surface of the blend
with 70% PE, parallel to the long axis of the tensile test bar. This picture further confirms the layered
structure. In the middle of Fig. 5-12 it can be seen that the top layer had been disrupted, but in
general the crack propagated between layers. It is therefore most likely that the mechanical strength
of the material was less in the transverse direction than in the parallel direction. It is not however easy
to measure mechanical properties in the transverse direction of a tensile test bar so this was not
further investigated.
39
Fig. 5-12: SEM -image of a 70/30 PE/PP blend fractured at room temperature. The fracture
surface is parallel to the long axis of the injection-moulded tensile test-bar.
5.6 MECHANICAL COMPATIBILITY OF PE/PP FROM RIGID PLASTIC
PACKAGING WASTE
5.6.1 Fracture toughness
Fig. 5-13 and Fig. 5-14 show the energy to break during tensile testing for the recycled PE/PP
blends, at cross-head speeds of 50 and 500 mm/min respectively. The graphs confirm that the
mechanical compatibility of the blends was poorest in the composition range where the droplet
morphology was prevailing.
160
Energy to break [J]
140
120
100
80
60
40
20
0
0
20
40
60
PP-content [%]
80
100
Fig. 5-13: Tensile energy to break, as a function of blend composition for recycled PE/PP, at
a cross-head speed of 50 mm/min.
40
24
22
20
Energy to break [J]
18
16
14
12
10
8
6
4
2
0
0
20
40
60
PP-content [%]
80
100
Fig. 5-14: Tensile energy to break, as a function of blend composition for recycled PE/PP, at
a cross-head speed of 500 mm/min.
40
Impact energy [J/m ]
35
30
25
20
15
10
5
0
0
20
40
60
PP-content [%]
80
100
Fig. 5-15: Notched Charpy impact energy as a function of blend composition for recycled
PE/PP.
Fig. 5-15 shows the impact resistance of the blends as a function of composition. Pure recycled PE
specimens did not break. When Fig. 5-13, Fig. 5-14 and Fig. 5-15 are compared, it becomes clear
that the deformation rate had a very strong influence on the compatibility between PE and PP. The
reduction in fracture toughness caused by the presence of 10-20% PE in recycled PP was smaller
the faster the deformation rate. In the impact test it was not even statistically significant. This
41
observation, together with the SEM observations described above gives the impression that there
must have been a great deal of interaction between droplet and matrix in the droplet morphology.
Neither polyethylene nor polypropylene has any functional groups capable of forming strong
intermolecular bonds other than bonding through crystalline domains. Since it is well known that
polyethylene and polypropylene are unable to co-crystallise1, the most likely type of interfacial
interaction is chain entanglement. If the interaction had been Van der Waal-forces alone, there would
definitely have been a greater reduction of impact energy caused by minor amounts of PE in PP,
because of the general weakness of such forces. Chain entanglements require some time to
disentangle. Therefore it is logical that the incompatibility was less at a higher deformation rate.
34
12
33
11
32
10
31
9
30
8
29
7
28
0
20
40
60
PP-content (%)
80
Elongation at yield (%)
Yield strength (MPa)
5.6.2 Other properties
Fig. 5-16 presents yield strength and yield elongation for the blends. In the region where the layered
morphology was observed, 20-40% PP, the yield strength exhibited a minimum which coincided
with the local maximum for the impact strength of Fig. 5-15. The change in yield elongation was
almost linear.
6
100
Fig. 5-16: Tensile yield properties, measured at a cross-head speed of 500 mm/min, as a
function of blend composition for recycled PE/PP.
Fig. 5-17 shows the dependence on blend composition of the tensile modulus and heat deflection
temperature, i.e. the point where the heat deflection curve reaches 2% elongation. For the tensile
modulus there was a somewhat negative synergism throughout the whole composition range. The
positive synergism reported in the literature for blends with 80-95% PP was not observed. This is
hardly surprising because that was an effect of polyethylene as a nucleating agent for polypropylene
(see section 2.4.1). When the PP fraction already contains nucleating agents, as in this case, the
1
This statement refers to the fact that PE/PP blends do not form any regularly repeating crystal structure that has
both PE and PP segments in the unit cell. In some articles, the frequently observed phenomenon of mutual
nucleation has been called “co-crystallisation“.
42
stiffness of unblended PP is enhanced and any nucleating effect of PE will be of less significance. The
change in heat deflection temperature was almost linear.
110
105
Tensile modulus (GPa)
1,8
100
95
1,7
90
1,6
85
80
1,5
75
70
1,4
65
1,3
0
20
40
60
PP-content (%)
80
Heat deflection temperature (°C)
1,9
60
100
Fig. 5-17: Tensile modulus, at a cross-head speed of 500 mm/min, and heat deflection
temperature as a function of blend composition for recycled PE/PP.
Since the PE-fraction was dominated by blow-moulded HDPE containers, it was not very suitable
for injection-moulding. Resins used for such applications have a high melt viscosity since extrusionblow-moulding requires high melt strength. Using blow-moulding grades for injection-moulding
results in large internal stresses, causing distortion of the moulded item. In the case of a simple, ”one
dimensional” object such as a tensile test-bar, this phenomenon can be observed as a contraction of
the bar after moulding, but for more complex shaped objects, warping may be the result [Lavieri,
1994]. Fig. 5-18 shows the mould shrinkage, i.e. the percentage contraction relative to the length of
the mould cavity, for recycled PE/PP. Mould shrinkage decreases with increasing PP-content, which
is hardly surprising since the PP-fraction is dominated by injection-moulding grades. The blend
containing 50% PP had a shrinkage factor similar to that of a HDPE injection-moulding grade with
MFR=12.
43
Mould shrinkage [%]
3
2
1
recycled PE/PP
HDPE, injection moulding grade
0
0
20
40
60
PP-content [%]
80
100
Fig. 5-18: Mould shrinkage factor for injection-moulded test-bars as a function of blend
composition for recycled PE/PP, compared with a virgin HDPE injection-moulding grade
(HE 7012, described in Tab. 3-1)
5.6.3 Should the PE and PP fractions be separated prior to mechanical recycling?
The mechanical tests showed that there are many advantages in separating the PE and PP fractions.
A significant negative synergy between PE and PP was found regarding yield strength, impact
strength, tensile energy to break and tensile modulus. It is not however clear that the benefits of
separating PE and PP will outweigh the costs involved. Waste management companies wishing to sell
recycled polyolefins to plastics processing industries should carefully consider what degree of
separation actually brings value to the customers. This is dependent on the intended application. Both
present customers and potential customers should be considered. A factor that severely narrows the
range of potential applications of recycled HDPE bottles is that the material is not suitable for
injection moulding [Lavieri, 1994]. This means that an investment in separation equipment in order to
satisfy the purity requirements of a bottle producer may not be the best solution, since recycling PE
together with PP makes the material more suitable for injection moulding [section 5.6.2]. A middle
way could be to make a smaller investment in separation equipment with a smaller capacity in order
to a produce both mixed and segregated grades. This strategy, called “product differentiation”,
makes the supplier of recycled material less dependent on a single customer or on a small number of
customers [Porter, 1980].
5.7 MODIFICATION OF PP FROM RIGID PACKAGING WASTE
This section deals with modifications of PP derived from packaging waste, aimed at enhancing
stiffness and dimensional stability. The intent has been to produce material with properties
comparable to those of PP-based engineering compounds. A more appealing designation of such
modifications is “upgrading“.
44
5.7.1 Modification with highly crystalline PP
Tab. 5-9 displays results from mechanical tests and DSC measurements on the recycled PP and on
blends with the highly crystalline PP grade, HC210P. Modification with this grade increases the
tensile modulus and yield strength significantly even though the increase in crystallinity, as expressed
by the ∆Hm-value, is moderate. The melting-peak temperature also increased which indicates that
more stable crystals where formed. Even though HC210P has a much lower impact strength than the
recycled PP, due to the lack of impact modifier, the addition of this grade to the recycled material
did not lead to a lower impact strength. The impact strength did in fact increase slightly for all the
modified blends, even though these changes are not statistically significant.
Tab. 5-9: Properties of recycled PP, virgin highly crystalline PP (HC210P) and blends of the
two materials. Tensile properties measured at 500 mm/min.
Pure recycled
Tensile modulus [GPa]
Yield strength [Mpa]
Yield elongation [%]
Elongation at break [%]
2
Impact strength [J/m ]
DSC, ∆Hm [J/g]
DSC, peak temp. [°C]
1.84
33.83±0.34
7.05±0.19
50.6±5.0
7.61±0.55
92.1
163.4
20%
HC210P
1.94
36.36±0.15
7.03±0.11
44.7±8.1
8.38±0.60
94.0
164.5
30%
HC210P
1.96
37.18±0.19
7.05±0.05
46.8±9.8
8.17±0.64
93.7
164.1
40%
HC210P
2.02
38.29±0.13
7.02±0.15
41.7±8.1
8.17±0.44
94.6
165.1
100%
HC210P
2.38
43.79±0.04
6.64±0.13
30.3±2.3
3.71±0.24
101.6
166.4
5.7.2 Modification with talc and highly crystalline PP
Tab. 5-10 presents results of mechanical tests and DSC measurements on talc -filled materials based
on recycled PP, and on the talc -filled reference grade ME210U. Also in this case, modification with
HC210P increased the crystallinity, melting-peak temperature, stiffness and yield strength. In
comparison with the reference grade, all the composites based on recycled PP had a lower tensile
modulus but a higher elongation at break and a higher impact strength. Modification with 20%
HC210P almost raised the yield strength to the level of the reference grade, although 60% of the
material was recycled material derived from mixed plastic packaging waste.
Tab. 5-10: Properties of talc -filled recycled PP, talc-filled recycled PP modified with virgin
highly crystalline PP (HC210P) and a commercial talc-filled PP-composite (ME210U). All
materials in the table had a talc-content of 20% by weight. Tensile properties measured at
50 mm/min.
Tensile modulus [GPa]
Yield strength [MPa]
Yield elongation [%]
Elongation at break [%]
2
Impact strength [J/m ]
DSC, ∆Hm [J/g]
DSC, peak temp. [°C]
talc-filled recycled
2.55
32.61±0.30
4.79±0.30
28.7±5.8
5.29±0.29
10% HC210P
2.66
34.21±0.42
5.19±0.27
30.9±3.6
5.01±0.65
20% HC210P
2.81
35.16±0.29
4.98±0.19
29.9±3.4
5.26±0.26
ME210U
3.18
35.62±0.24
5.28±0.14
21.9±2.5
4.64±0.22
89.6
94.3
96.9
96.2
163.6
164.9
165.2
163.9
45
The flake-shape of talc causes it to orient in the flow-direction during injection moulding. McGenity
et al. [1992] argued that this lowers the impact strength measured in the Charpy-type test because of
the nucleating effect of talc, causing PP crystallites to grow perpendicular to the preferred direction
of the flakes. When talc is added to PP derived from packaging waste, such effects are probably of
minor importance because the material is already nucleated and consequently the importance of the
filler as provider of nucleating sites is reduced.
Fig. 5-19 shows heat-deflection curves for some of the materials based on PP packaging waste,
together with those for two of the reference grades. It is striking that the unmodified recycled PP,
even though it contains copolymers which reduce the rigidity, had a much higher heat deflection
temperature than the virgin homopolymer, HF135M. The copolymer reference grades, RE220P and
BE160M, have even lower heat deflection temperatures (see Tab. 3-2). Heat deflection
temperatures presented in Tab. 3-2 are equivalent to the temperature at which the heat deflection
curve reaches 2% elongation. A part of the reason why the dimensional stability was greater in the
case of the recycled PP is that HF135M did not contain any nucleating agents. Modification with
HC210P increased the dimensional stability further. So did compounding with talc, but it was only
through the combined addition of HC210P and talc that the very good dimensional stability of the
talc-filled reference grade was reached.
Fig. 5-19: Heat deflection curves for materials based on recycled PP and for reference
grades.
It was also observed empirically that the level of odour decreased when talc was added to recycled
PP. No qua ntitative data was produced however. This might have been a result of adsorption of
odorous compounds onto the filler particles.
46
5.8 EFFECTS OF FOREIGN POLYMERS AS IMPURITIES IN RECYCLED
PLASTICS
5.8.1 EVOH from barrier layers in recycled PP
The reprocessed PP from the March 1998 sampling contained 11.4% material originating from PP
bottles. Most of this was ketchup bottles. Since ketchup is oxygen sensitive, such bottles contain a
barrier layer of EVOH (poly[ethylene-co-vinyl alcohol]). This is an important factor that has to be
taken into account when automated separation methods are considered. When separation is done
manually, the staff can be instructed not to include ketchup bottles in the PP fraction in order to avoid
the EVOH, if this is the wish of the customer. If sorting is done by some automated method, it is not
easy to remove material from ketchup bottles since both the inner and outer surfaces are pure PP.
Multilayer bottles also contain a tie-layer between PP and EVOH in order to prevent de-lamination.
The tie -layer material is designed to have strong adhesion to both the unpolar PP and the highly polar
EVOH and it therefore also has the potential to act as a compatibiliser in the recycled material. A
fine dispersion of EVOH was not however achieved when the mixed PP-flakes from the March
1998 sampling were extruded. Instead EVOH aggregated into large droplets that distorted the
extruded strands, making their thickness uneven. EVOH domains could easily be observed when a
cross-section of a strand was cut, since they were colourless while the surrounding PP was purple,
as a result of the mixture of pigments present in the recycled material. Sizes of EVOH droplets were
up to about 3 mm.
When the extruded material was repelletised and injection moulded, the high shear forces facilitated
dispersion of the EVOH. Fig. 5-20 displays a cross-section of an injection-moulded test-bar after
treatment with acetic anhydride. The figure shows that the EVOH had been dispersed finely when
the repelletised PP was injection-moulded. The bumps visible in Fig. 5-20, 0.5 - 4 µm in size, were
EVOH-domains which had been swollen by the treatment with acetic anhydride.
47
Fig. 5-20: SEM -image of recycled PP treated with acetic anhydride.
5.8.2 Talc-filled PP and HIPS in ABS
Tab. 5-11 presents results from mechanical tests on recycled ABS and blends with talc-filled ABS
or HIPS. ABS was significantly embrittled by the incorporation of PP, already at the 3% level. At
higher levels of PP, the material failed immediately after the yield point in the tensile test. The results
show that a contamination level of 3% can be acceptable for many applications but not more. In the
case of blends with 6% and 9% PP it was not even possible to perform the tensile impact test
because the material was so brittle that it broke when being mounted onto the test apparatus (see
Paper V)
Tab. 5-11: Mechanical properties of recycled ABS, HIPS and blends of ABS with minor
amounts of talc-filled PP or HIPS. The crosshead speed for tensile tests was 50 mm/min.
Yield strength [MPa]
Elongation at yield [%]
Stress at break [MPa]
Elongation at break [%]
Tensile impact strength [kJ/m2]
Yield strength [MPa]
Elongation at yield [%]
Stress at break [MPa]
Elongation at break [%]
Tensile impact strength [kJ/m2]
recycled ABS
46.6±0.3
2.22±0.05
38.5±0.6
10.2±1.1
286±26
4% HIPS
41.9±0.1
2.07±0.05
36.0±0.2
18.4±4.3
260±26
3% PP
45.2±0.1
2.22±0.04
37.5±0.7
9.6±2.3
216±27
8% HIPS
40.5±0.4
2.20±0.08
35.9±0.3
14.5±4.0
255±12
48
6% PP
42.1±0.2
2.20±0.03
40.8±0.5
2.6±0.1
not measured
12% HIPS
38.5±0.5
2.23±0.11
35.0±0.3
9.9±1.8
199±11
9% PP
39.3±0.2
2.19±0.04
39.2±0.4
2.3±0.1
not measured
100% HIPS
19.5±0.3
1.23±0.02
18.4±0.5
45.5±4.2
not measured
Observations by SEM showed that when talc-filled PP was incorporated as an impurity in ABS, talc
particles where distributed in the ABS matrix. In order to establish the role of talc particles as an
impurity in ABS, a blend containing 9% PP without filler was also prepared. Fig. 5-21 presents
results of tensile tests on blends of ABS with 9% filled or unfilled PP. Since unfilled PP caused even
more severe embrittlement to ABS it is concluded that the embrittlement observed when talc -filled
PP was incorporated in ABS was primarily due to the inherent incompatibility between the polymers,
not to the presence of filler particles.
40
35
20
With filler
25
Unfilled
Stress [MPa]
30
15
10
5
0
0
0,5
1
1,5
2
2,5
Strain [%]
Fig. 5-21: Tensile test curves for recycled ABS contaminated with 9% filled or 9% unfilled
PP. The crosshead speed was 50 mm/min.
The addition of HIPS to recycled ABS presented a more comple x case. The tensile impact strength
was decreased by the presence of HIPS but in the tensile test, which represents the same mode of
deformation but at a much slower rate, the elongation at break was increased from 10.2% to 18.4%
by the presence of 4% HIPS. This was explained as being due to an increased number of craze
initiation sites in the matrix of ABS [Paper V], which also caused the yield strength to drop by 10%.
Larger additions of HIPS led to a further deterioration of the yield strength accompanie d by a
reduced ductility, regardless of deformation rate.
5.8.3 ABS in talc -filled PP
Tab. 5-12 displays results from mechanical tests on recycled talc-filled PP and blends containing
minor amounts of ABS. No significant changes were detected in the elongation at break. The
incompatibility was more pronounced at a fast deformation rate (tensile impact test), similar to the
case of ABS blends containing minor amounts of HIPS or talc -filled PP. It is clear that talc-filled PP
was less sensitive to contamination by ABS than vice-versa. The data do not indicate any certain
limit for an acceptable contamination level since the changes are gradual. Purity requirements should
therefore be set in relation to design criteria for the intended application of the recycled material.
49
Tab. 5-12: Mechanical properties of recycled talc-filled PP and blends with minor amounts
of ABS. The crosshead speed for tensile tests was 50 mm/min.
Yield strength [MPa]
Elongation at yield [%]
Ultimate tensile strength [MPa]
Elongation at break [%]
Tensile impact strength [kJ/m2]
recycled PP
34.1±0.3
5.63±0.15
29.6±1.2
19.0±3.9
318±18
3% ABS
32.6±0.2
5.21±0.12
27.9±0.8
19.2±4.0
280±20
50
6% ABS
31.4±0.2
4.86±0.06
26.3±0.8
18.7±3.3
245±8
9% ABS
30.2±0.4
4.59±0.10
25.0±0.7
19.2±4.9
220±14
6 CONCLUSIONS
• Complex polyolefin mixtures containing large amounts of bottle -grade HDPE are tough
and ductile and suitable for recycling through extrusion.
Tensile tests and the measurement of tear propagation resistance showed that LLDPE had very
much better mechanical compatibility with PP than LDPE in binary mixtures of film-grade PE and
PP. In the case of more complex polyolefin blends, containing also considerable amounts of
bottle-grade HDPE, the relation between fracture toughness and composition of the PE film
component (LDPE/LLDPE) was less significant. This implies that such mixed polyolefin fractions
can be recycled without further separation regardless of whether the film-component is dominated
by LDPE or LLDPE, an that they still produce a composite material of high ductility and
toughness.
• VLDPE is effective for enhancing toughness and reducing variations in the mechanical
properties of complex polyolefin mixtures.
If it is desired to increase the fracture toughness of a mixed polyolefin fraction or if there are
substantial variations in the composition of the waste stream, it is recommended to add a modifier.
A very low density polyethylene (VLDPE) was found to be a more effective modifier than EVA
or poly(1-butene). The addition of 20% VLDPE to a set of hypothetical polyolefin fractions of
different compositions increased tensile strength and tear propagation resistance by on average
30%. At the same time, the standard deviations for the same properties were reduced by over
50%, which is valuable for commercial applications of recycled materials because it allows
product quality to be kept more consistent. In general, modifiers that reduce stiffness the most
also increases fracture toughness the most. In the case of EPDM elastomers added to HDPE-rich
blends, this means that a fully amorphous EPDM is preferred over a semi-crystalline EPDM
capable of co-crystallising with HDPE. The reason is that the desired toughening effect can be
achieved using a smaller amount of additive.
• Microphase structure is shifted from droplets to layers in recycled rigid PE/PP blends at
a PP-content ≤40%.
Material from a collection system for rigid (non-film) plastic packaging waste was studied. The
non-film polyolefin fraction, which dominated the collected material, consisted of approximately
45% PP and 55% PE, after removal of non-floating materials. Whether or not PP crystallised
ahead of PE was found to be important for the microphase structure and consequently also for
the mechanical compatibility of PE/PP blends, after reprocessing by injection moulding. DSC
studies showed that nucleating agents present in the recycled material caused PP to crystallise
ahead of PE, instead of the simultaneous crystallisation which was found for PE/PP blends
without nucleating agents. In blends with low PP-content, the nucleating effect disappeared so
that crystallisation was simultaneous, despite the presence of nucleating agents. The nucleating
effect disappeared at a PP-content of about 40% for fast cooling rates. This coincided with a shift
in microphase structure observed by SEM, from a droplet structure for PP-rich blends to a
layered structure for PE-rich blends.
51
• For rigid plastic packing waste separation of PE and PP from each other is desirable but
absolutely not necessary for successful recycling.
Mechanical tests showed that the compatibility between PE and PP was poorest in the
composition range with 60-90% PP which confirmed that mechanical compatibility was
disfavoured by the droplet morphology. It was, however, not a sudden change coinciding with the
change in microphase structure, but rather a gradual deterioration in fracture toughness with
increasing PP-content, down to a minimum at 80% PP. A significant negative synergy between
PE and PP was found regarding yield strength, impact strength, tensile energy to break and tensile
modulus. These factors speak in favour of full separation of PE and PP. For yield elongation and
heat deflection temperature there were no significant synergistic effects, i.e. values for blends fall
approximately on a straight line joining the corresponding values for 100% PE and 100% PP.
Studies of non-film PE and PP packaging waste and their blends showed that it is not necessary
to invest in separation equipment a with capacity to separate the entire polyolefin fraction into PE
and PP of very high purity. The range of possible applications for the recycled material is
broadened if mixed fractions are also supplied. Injection moulding of recycled PE, dominated by
HDPE blow moulding grades, gives rise to large internal stresses that distort the physical shape of
the moulded article. Measurements of mould shrinkage showed that commingling the PE-fraction
with PP could reduce such problems.
• Rigidity can be enhanced without loosing impact resistance by adding a highly
crystalline virgin PP to mixed-grade recycled PP.
A highly crystalline virgin PP was found to be suitable for modification of the PP fraction from the
collection system for rigid plastic packaging waste, in order to increase rigidity through cocrystallisation. This gave rise to a positive synergism in the sense that the tensile modulus, yield
strength and heat deflection temperature were increased substantially without loss of impact
strength, despite the fact that the impact strength of the virgin PP was only about half the
corresponding value for unmodified recycled PP. A composite material consisting of 20% highly
crystalline PP, 20% talc and 60% recycled PP gave mechanical properties similar to those of a
commercial talc-filled PP-compound used for advanced engineering applications.
• EVOH from barrier layers can aggregate when recycled PP is extruded, but it is finely
dispersed during injection moulding.
Bottles accounted for 11% of the PP fraction of rigid plastic packaging waste. Most of this was
ketchup bottles with an EVOH barrier layer. When the PP fraction was extruded, EVOH
aggregated and formed macroscopic droplets that distorted the surface structure of the extrudate.
When the recycled PP was injection moulded, EVOH was finely dispersed into domains with a
size of 0.5 - 4 µm. This makes injection moulding a preferred conversion process for PP waste
including bottles with an EVOH barrier layer. The distribution of EVOH was studied by treating
the material with acetic anhydride, causing EVOH to swell permanently due to acetylation of
hydroxyl groups. EVOH domains were thereafter perceived as elevations above the surface of
the matrix material when it was studied by SEM.
• PP as a contaminant in recycled ABS causes severe embrittlement.
The compatibility of recycled ABS, HIPS and talc-filled PP used in durable products was also
studied. This issue resembles the case of PE/PP from packaging waste because small density
differences complicate automated separation. A maximum contamination level of 3% talc -filled
52
PP in ABS was found to be acceptable. The reason for embrittlement by talc -filled PP in ABS
was the incompatibility of the polymers, not the presence of filler particles. Talc-filled PP was less
sensitive to contamination by ABS than vice-versa. In blends of ABS, HIPS and talc -filled PP,
the degree of incompatibility increased at high deformation rates, in contrast to the case of blends
of PE and PP from packaging waste, where the degree of incompatibility was reduced at high
deformation rates. This difference was explainedas being due to a higher degree of interfacial
interaction through chain entanglements in blends of PE and PP.
53
7 SUGGESTIONS FOR FUTURE WORK
Studies of single grade mixtures, i.e. blends containing one grade of each polymer type, make up the
bulk of scientific studies on polymer/polymer compatibility. Such studies are highly relevant for the
understanding of material fractions encountered in waste-streams considered for mechanical
recycling, but they do not give the complete picture since they represent simplified models of
mixtures encountered in common waste-streams. Studies of single grade mixtures must therefore be
complemented by further studies of more complex plastic mixtures in order better to reach an
understanding of the problems and possibilities of plastics recycling. Research aimed at gaining a
deeper understanding of the relation between melt-processing, microphase structure, mechanical
properties and dimensional stability of polymer blends would be valuable for engineers engaged in
the design and development of products derived from mixed plastic waste.
54
8 ACKNOWLEDGEMENTS
I gratefully acknowledge my supervisors, Prof. Ann-Christine Albertsson and Prof. Sigbritt Karlsson
for accepting me as their student as well as for excellent guidance of the Department of Polymer
Technology in general and of my work in particular.
The work was supported financially by the European Commission, through the EU-project BRPRCT96-0247, “Electrostatic recovery of paper and plastic packaging waste“ (ELREC). I thank
everybody involved in the ELREC project, especially Claudio Fernandez of KTH, Chris Robertson
and Jeremy Smallwood of ERA Technology, Staffan Ågren and Susanne Eriksson of Ragn-Sells and
Rafael Miguel of Gaiker.
I also would like to thank my family for all their support they have shown me.
All present and previous scientists, administrative staff and students of the Department of Polymer
Technology have contributed to this work in one way or another and they are therefore also thanked.
55
9 REFERENCES
Albertsson, 1977
Albertsson A-C, “Studies on the Mineralization of 14C -labelled
Polyethylenes in Aerobic Biodegradation and Aqueous Ageing“, PhDthesis, Royal Institute of Technology, Stockholm, Sweden.
Bartczak et al., 1985
Bartczak Z, Galeski A, Martuscelli E, Janik H, Polymer, 26, p1843
Bartczak et al., 1986
Bartczak Z, Galeski A, Pracella M, Polymer, 27, p537
Bartlett et al., 1982
Bartlett D W, Barlow J W, Paul D R, Journal of Applied Polymer
Science, 27, p2351
Basset et al., 1998
Basset D C, Dong L, Olley R H, Journal of Macromolecular Science Physics, B37 (4), 527
Berticat et al., 1980
Berticat P, Boiteux G, Dallox J, Douillard A, Guillet J, Seytre G,
European Polymer Journal, 16, p479
Blom et al. 1998
Blom H P, Teh J W, Rudin A, Journal of Applied Polymer Science,
70, p2081
Blom et al., 1995
Blom H P, Teh J W, Rudin A, Journal of Applied Polymer Science,
58, p995
Blom et al., 1996a
Blom H P, Teh J W, Rudin A, Journal of Applied Polymer Science,
60, p1405
Blom et al., 1996b
Blom H P, Teh J W, Rudin A, Journal of Applied Polymer Science,
61, p959
Boucher et al., 1996
Boucher E, Folkers J P, Hervet H, Leger L, Creton C, Macromolecules,
29, p774
Brandrup, 1996
Brandrup J, in “Recycling and Recovery of Plastics“ ed. by Brandrup
J, Carl Hanser Verlag, p31-45
Breant, 1993
Breant P, Recycle’93, Davos, Switzerland, p 8/4
Bruder, 1996
Bruder J, in “Recycling and Recovery of Plastics“ ed. by Brandrup J,
Carl Hanser Verlag, p46-61
Cham et al., 1994
Cham P M, Lee T H, Marand H, Macromolecules 27(15), p4263
56
Curran et al., 1996
Curran P F, van Tubergen J, McGuire S O, Proceedings Globec’96,
Davos, Switzerland,p6/4
De Clippeleir, 1997
De Clippeleir J, paper presented at Polypropylene’97, Zürich,
Zwitzerland.
Deanin & Sansone, 1978
Deanin R D, Sansone M F, ACS Polymer Preprints, 19(1), p211
Dolgopolsky et al., 1995
Dolgopolsky I, Silberman A, Kenig S, Polymers for Advanced
Technologies, 6(10), p653
Eriksson, 1997
Eriksson P -A, “Mechanical Recycling of Glass Fibre Reinforced
Polyamide 66“, PhD-thesis, Royal Institute of Technology, Stockholm,
Sweden.
Fahrbach & Schnettler,
1996
Fahrbach G, Schnettler H R, Proceedings Globec’96, Davos
Switzerland,p17
Fayt et al., 1981
Fayt R, Jérôme R, Teyssié Ph, Journal of Polymer Science, Polymer
Physics Edition, 19, p1269
Fayt et al., 1982
Fayt R, Jérôme R, Teyssié Ph, Journal of Polymer Science, Polymer
Physics Edition, 20, p2209
Flaris et al., 1993
Flaris V, Wasiak A, Wenig W, Journal of Materials Science, 28(6),
p1685
Galeski et al., 1984
Galeski A, Pracella M, Martuscelli E, Journal of Polymer Science,
Polymer Physics Edition, 22, 739
Gedde et al., 1994
Gedde U W, Viebke J, Leijström H, Ifwarson M, Polymer Engineering
and Science, 34(24), p1773
Gedde, 1995
Gedde U W, “Polymer Physics“, Chapman & Hall
Gohil & Petermann, 1980
Gohil R M, Petermann J, Journal of Polymer Science, Physics, B18,
p217
Gonzalezmontiel et al., 1995 Gonzalezmontiel A, Keskkula H, Paul D R, Polymer, 36(24), p4587
Gupta et al., 1982
Gupta A K, Gupta V B, Peters R H, Harland W G, Berry J P, Journal
of Applied Polymer Science, 27, p4669
Ha & Kim, 1989
Ha C S, Kim S C, Journal of Applied Polymer Science, 37, p317
Hill et al., 1994
Hill M J, Oiarabal L, Higgins J S, Polymer, 35, p3332
Hill et al., 1998
Hill M J, Montes P, Rafiq Y A, Polymer, 39(5), p6669
Hindawi et al., 1992
Hindawi I A, Higgins J S, Weiss R A, Polymer, 33, p2522
57
Hope et al., 1994
Hope P S, Bonner J G, Miles A F, Plastics, Rubbers & Composites,
Processing & Applications, 22(3), p147
Hsu & Geil, 1987
Hsu C C, Geil P H, Polymer Engineering and Science, 27, p1542
Huang, 1995
Huang S J, Journal of Macromolecular Science - Pure Applied
Chemistry, A32(4), p593
Jabarin et al., 1992
Jabarin S A, Lofgren E A, Shah S B, “ACS Symposium Series 513:
Emerging technologies in plastics recycling “, p215
Kalfoglou et al., 1995
Kalfoglou N K, Skafidas D S, Kallitsis J K, Lambert J C, Vandersta ppen
L, Polymer, 36(23), p4453
Karlsson, 1988
Karlsson S, “Some Techniques for studying the Degradation of Native
and Synthetic Polymers in Abiotic abd Biotic Environments“, PhD -thesis,
Royal Institute of Technology, Stockholm, Sweden.
Kim & Kim, 1991
Kim Y S, Kim C Y, Polymer Engineering and Science, 31(14), p1009
Krishnamoorti et al., 1994
Krishnamoorti R, Graessley W W, Balsara N P, Lohse D J,
Macromolecules, 27, p3073
Kryszewski et al., 1973
Kryszewski M, Galeski A, Pakula T, Grebowicz J, Journal of Colloid
Interface Science, 44, p85
La Mantia, 1993
La Mantia F P, Polymer Degradation and Stability, 42(2), p213
Laihonen et al., 1997
Laihonen S, Gedde U W, Werner P-E, Westdahl M, Jääskeläinen P,
Martinez-Salazar J, Polymer, 38(2), p371
Lemstra & Kirschbaum,
1984
Lemstra P J, Kirschbaum R, Polymer, 26(9),p1372
Lavieri, 1994
Lavieri N, Annual Technology Conference-Society of Plastics
Engineers, 52, p2856
Lohse, 1986
Lohse D J, Polymer Engineering & Science , 26, p1500
Lovinger & Williams, 1980
Lovinger A J, Williams M L, Journal of Applied Polymer Science, 25,
p1703
Lund et al., 1993
Lund H F, Lund B, in “The McGraw-Hill Recycling Handbook“ ed.
by Lund H F, McGraw-Hill, pB21
McGenity et al., 1992
McGenity P M, Hooper J J, Paynter C D, Riley A M, Nutbeem C, Elton
N J, Adams J M, Polymer, 33(24), p5215
Menczel & Varga, 1983
Menczel J, Varga J, Journal of Thermal Analysis, 28, p161
58
Milgrom, 1982
Milgrom J, Polymer and Plastics Technology and Engineering, 18(2),
p167
Min et al., 1984
Min K, White J L, Fellers J F, Polymer Engineering & Science ,
24(17), p1327
Natta et al., 1955
Natta G, Pino P, Corradini P, Danusso F, Mantica E, Mazzantl G,
Moranglio G, Journal of the Americal Chemical Society, 77, p1708
Noel & Carley, 1975
Noel O F, Carley J F, Polymer Engineering & Science , 15(2), p117
Obieglo & Romer, 1996
Obieglo C, Romer K, Kunststoffe, 86(4), p546
Olley et al., 1979
Olley R H, Hodge A M, Bassett J J, Journal of Polymer Science,
Polymer Physics Edition, 17, p627
Orroth & Malloy, 1994
Orroth M J, Malloy R A, Annual Technology Conference-Society of
Plastics Engineers, 52, p3059
Paul et al., 1972
Paul D R, Locke C E, Vinson C E, Polymer Engineering and Science,
12, p157
Pearson, 1993
Pearson W, in “The McGraw-Hill Recycling Handbook“ ed. by Lund
H F, McGraw-Hill, chapter 14
Piloz et al., 1976
Piloz A, Decroix J-Y, May J-F, Angew. Makromol. Chemie, 54, p77
Porter, 1980
Porter M E, “Competitive Strategy: Techniques for Analyzing
Industries and Competitors”, New York Free Press
Rajasekaran et al.,1995
Rajasekaran J J, Curro J G, Honeycutt J D, Macromolecules, 28(20),
p6843
Sano et al., 1998
Sano H, Yui H, Li H, Inoue T, Polymer, 39(21), p5265
Scheirs, 1998
Scheirs J, “Polymer Recycling“, John Wiley & Sons.
Schürmann et al., 1998
Schürmann B L, Niebergall U, Severin N, Burger C, Stocker W, Rabe J
P, Polymer, 39(22), p5283
Scott et al., 1981
Scott G, Ghaffar A, Sadrmohaghegh C, European Polymer Journal,
17, 941
Scott et al., 1984
Scott G, Hajian M, Sadrmohaghegh C, European Polymer Journal,
20(2), p135
Scott et al., 1985
Scott G, Sadrmohaghegh C, Setudeh E, Polymers and Plastics
Technology and Engineering, 24(2&3), p149
59
Shepard et al., 1997
Shepard T A, Delsorbo C R, Louth R M, Walborn J L, Norman D A,
Harvey N G, Spontak R J, Journal of Polymer Science, Physics,
B35(16), p2617
Siegmann, 1982
Siegmann A, Journal of Applied Polymer Science, 27, p1053
Silberman et al., 1995
Silberman A, Raninson E, Dolgopolsky I, Kenig S, Polymers for
Advanced Technologies, 6(10), p643
Stachurski et al., 1996
Stachurski Z H, Edward G H, Yin M, Long Y, Macromolecules, 29(6),
p2131
Stahl & Kleine-Kleffmann,
1997
Stahl I, Kleine-Kleffmann U, in “Recycling and Recovery of Plastics“
ed. by Brandrup J, Carl Hanser Verlag, p265-273
Tall et al., 1998
Tall S, Karlsson S, Albertsson A-C, Polymers & Polymer Composites,
6(5), p261
Teh et al., 1985
The J W, Noordin R, Low A K Y, Kok C M, Proceedings,
International Plastics Conference, Kuala Lumpur, 22-24 October
Teh J W, Rudin A, Keung J C, Advances in Polymer Technology,
13(1), p1
Teh et al., 1994a
Teh et al., 1994b
Teh J W, Blom H P, Rudin A, Polymer, 35(8), p1680
Tselios et al., 1998
Tselios C, Bikiaris D, Maslis V, Panayiotou C, Polymer, 39(26), p6807
Wenig & Meyer, 1980
Wenig W, Meyer K, Colloid & Polymer Science, 258, p1009
Wignall et al., 1982
Wignall G D, Child H R, Samuels R J, Polymer, 23, p957
Vogl, 1998
Vogl O, Journal of Macromolecular Science - Pure Applied
Chemistry, A35(7&8), p1017
Yao & Beatty, 1997
Yao L., Beatty C., Annual Technology Conference-Society of
Plastics Engineers, 55(2), p2582
Zakin J L, Simha R, Herskey H, Journal of Applied Polymer Science,
10, p1455
Zakin et al., 1966
60
APPENDIX I: Experimental design and raw data for sheet extruded polyolefin blends. The portions of each polymer type in each blend are
shown to the left and results from mechanical tests to the right.
Blend
#01
#02
#03
#04
#05
#06
#07
#08
#09
#10
#11
#12
#13
#14
#15
#16
#17
#18
#19
#20
#21
#22
#23
#24
HDb LLD
[%] [%]
100
100
LD
[%]
HDi
[%]
PPe
[%]
PPi
[%]
EVA VLD
[%] [%]
PB
[%]
100
70
30
30
70
70
30
30
30
70
70
70
70
70
40
30
80
30
30
30
30
20
20
20
80
80
80
20
20
20
80
80
80
20
20
20
80
80
50
50
50
30
30
30
20
20
20
E-modulus
[MPa]
1239±168
301±34
238±19
1211±107
528±22
1330±94
635±55
531±45
1411±76
540±36
489±32
825±89
805±57
204±11
182±18
683±36
166±22
139±17
852±47
223±36
224±47
920±65
389±29
336±16
σy
[MPa]
24.69±0.42
9.71±0.13
8.79±0.14
22.76±0.59
14.13±0.34
26.92±0.41
16.23±0.18
15.19±0.20
26.69±0.24
14.31±0.38
14.33±0.24
19.61±0.31
18.28±0.08
7.94±0.17
7.64±0.17
16.89±0.15
6.75±0.10
6.28±0.08
18.61±0.26
8.22±0.18
7.81±0.11
19.76±0.41
11.80±0.14
10.45±0.08
σb
[MPa]
18.08±1.53
26.05±0.94
13.54±0.52
17.80±3.18
24.14±0.94
10.21±1.49
28.99±0.42
12.78±0.27
19.92±1.57
28.56±1.00
13.12±0.42
18.34±0.88
26.12±2.99
20.18±1.98
14.53±0.11
32.37±0.76
20.67±3.75
18.73±1.86
27.78±1.87
20.88±1.24
13.06±0.56
23.02±3.96
11.76±1.30
26.06±0.75
eb
[%]
634±306
1290±42
769±29
679±481
1300±55
76±25
1189±27
442±99
706±172
1160±51
496±40
26±6
1006±127
1035±84
815±12
1112±39
1084±128
1056±101
883±68
1059±82
715±66
1196±164
699±115
1256±17
TPR
[N/mm]
111.1±15.4
108.0±3.2
54.6±4.0
116.0±19.3
131.9±10.4
10.1±5.6
106.5±4.4
8.2±1.6
48.3±2.8
81.2±4.3
7.4±1.7
4.7±1.4
82.9±2.6
98.4±4.9
69.3±8.2
108.1±2.2
83.8±3.3
70.4±2.9
20.7±4.1
94.2±5.7
42.7±5.7
119.1±4.1
70.5±5.3
101.2±6.2
#25
#26
#27
#28
#29
#30
#31
#32
#33
#34
#35
#36
#37
#38
#39
#40
#41
#42
#43
#44
#45
#46
#47
#48
#49
#50
#51
#52
#53
50
50
30
30
20
20
50
30
30
30
30
30
50
50
50
50
50
50
50
20
30
30
30
20
20
33.3
35
27.5
35
28.3
35
33.3
30
20
28.3
50
50
20
16.7
16.7
23.3
35
33.3
30
20
28.3
33.3
35
27.5
35
28.3
30
30
30
30
30
30
10
15
15
15
10
12
15
20
20
20
20
10
15
15
10
6
15
28.3
20
16.7
16.7
23.3
15
15
15
12
15
15
15
12
20
20
30
30
20
16.7
16.7
23.3
20
20
20
15
15
12
15
20
20
6.7
5
5
5
6.7
5
6.7
5
5
5
5
5
5
5
5
10
5
5
20
10
10
10
10
20
10
6
6.7
5
5
5
6.7
5
6.7
395±15
401±28
538±49
452±10
795±51
364±17
438±31
442±35
365±25
445±9
839±133
837±60
1099±96
349±32
338±48
422±15
535±25
651±29
568±25
570±27
769±42
852±22
359±27
951±54
1018±61
630±38
486±26
601±44
815±63
12.14±0.16
12.59±0.22
14.25±0.19
13.42±0.22
19.16±0.27
10.97±0.18
13.78±0.25
13.65±0.14
11.36±0.15
13.72±0.31
18.28±0.11
19.04±0.22
23.94±0.27
10.85±0.18
9.41±0.15
12.13±0.15
15.33±0.13
16.49±0.19
15.46±0.20
15.32±0.30
19.21±0.32
20.16±0.42
10.91±0.10
21.47±0.26
21.42±0.31
16.51±0.16
13.92±0.12
16.10±0.22
18.41±0.38
20.10±0.63
10.01±0.26
23.90±0.69
14.72±0.31
31.72±3.18
23.83±0.71
14.35±0.23
15.39±0.71
33.96±0.93
19.45±3.79
13.30±0.11
28.57±2.11
37.71±2.05
23.57±2.02
18.74±1.36
21.94±0.53
23.01±1.29
22.32±1.92
28.79±0.85
27.90±0.91
31.51±1.45
29.90±1.07
22.54±0.56
30.00±0.71
27.91±2.01
25.81±0.65
25.49±0.93
23.80±0.88
25.00±0.59
1113±59
102±16
982±18
606±11
998±90
1001±22
472±28
590±21
1229±39
707±200
158±44
1061±58
1231±67
1014±88
1046±86
1041±23
981±45
947±84
1129±28
1127±38
1148±38
1167±22
1162±25
1009±33
965±44
1174±23
1063±33
1022±51
1100±42
98.5±9.2
15.5±2.6
52.5±4.2
5.3±1.2
67.4±5.3
69.5±2.6
15.9±7.7
25.7±8.4
90.6±2.7
54.0±10.9
54.4±7.3
94.7±3.4
70.3±3.5
77.1±4.3
100.8±5.8
71.7±2.8
55.0±1.7
76.9±3.8
93.9±2.9
104.0±2.2
96.6±16.0
63.8±2.6
97.1±5.0
52.7±5.9
40.1±6.2
60.4±3.1
80.0±2.8
58.0±2.1
49.9±2.7
#54
18.3
18.3
18.3
12
12
6
5
5
5
622±38
15.05±0.16
24.49±1.10
1094±50
74.5±1.1
APPENDIX II: Classification of materials sampled at the Lunda Plant. August 1997 - July 1998
Sampling time:
Total amount sampled:
PET; bottles
PET; other packaging(trays, blisters)
HDPE; bottles, jugs
HDPE; film
HDPE; other packaging (trays, caps, buckets, cups)
HDPE; non-packaging
PVC; flexible packaging
PVC; rigid packaging
PVC; non-packaging
LDPE/LLDPE; film
LDPE/LLDPE; lids
PP; bottles
PP; buckets, trays, cups
PP; lids, caps
PP; film
PP; non-packaging
PS; cups, lids, etc.
PS; expanded
Other plastics (ABS, PMMA, PU, etc.)
Total plastics
Total rigid plastic packaging
Paper
Other organics
Glass
Metal
Autumn-97
42.5 kg
6.1
1.1
26.9
0.3
2.5
0.8
0.2
0.6
2.1
7.5
1.2
3.2
22.1
6.1
0.5
3.4
5.8
0.9
2.0
93.2
76.6
3.4
1.4
0.0
1.9
Jan-98
36.5 kg
6.0
1.4
31.6
0.7
0.5
1.4
0.1
1.9
0.8
11.3
1.6
2.4
13.1
7.5
0.4
1.9
7.6
2.0
2.0
94.1
75.5
5.2
0.2
0.0
0.5
Mar-98
46.3 kg
6.0
1.1
32.8
0.3
2.3
0.3
0.1
2.0
1.0
6.8
1.5
3.3
17.4
7.1
0.2
2.0
6.7
1.6
1.5
93.9
81.8
4.3
1.0
0.0
0.8
May-98
34.4 kg
5.8
0.6
37.3
0.1
1.3
0.3
0.1
1.2
2.6
3.3
1.2
3.7
20.6
7.7
0.1
0.5
6.0
2.4
0.3
95.1
87.7
2.1
0.1
1.1
1.6
Jul-98
36.7 kg
6.0
0.8
36.9
0.1
1.4
0.2
0.1
1.4
2.3
3.0
1.3
3.2
19.9
7.8
0.2
1.2
6.3
2.2
0.4
94.7
87.2
3.0
0.2
0.1
1.9
Ceramics
0.1
0.0
0.0
0.1
0.1
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement