Crompton - Physical Testing of Plastics

1

1.1 Introduction

In the past, results of standard tests such as tensile strength, Izod impact strength and

softening point have been given major emphasis in the technical literature on plastics.

More recently, however, with the increasing use of plastics in more critical applications,

there has been a growing awareness of the need to supplement such information with

data obtained from tests more closely simulating operational conditions.

For many applications, the choice of material used depends upon a balance of stiffness,

toughness, processability and price. For a particular application, a compromise

between these features will usually be necessary. For example, it is generally true

that, within a given family of grades of a particular polymer, the rigidity increases as

impact strength decreases. Again, processing requirements may place an upper or a

lower limit on the molecular weight (MW) of the polymer that can be used, and this

will frequently influence the mechanical properties quite markedly. Single test values

of a given property are a less reliable guide to operational behaviour with plastics

than with metals. This is because for plastics the great mass of empirical experience

and tradition of effective design which has been built up over centuries in the field of

metals is not yet available. Furthermore, no single value can be placed on the stiffness

or the toughness of a plastic because:

the mould, processing conditions and the temperature of use.

and oxidative ageing, ultraviolet (UV) ageing and chemical attack (including the

special case of environmental stress corrosion).

In addition, a change in a specific polymer parameter may affect processability and

basic physical properties. Both of these factors can interact in governing the behaviour

of a fabricated article. Comprehensive experimental data are therefore necessary to

understand effectively the behaviour of plastic materials, and to give a realistic and

reliable guide to the selection of material and grade.

1 Mechanical Properties of Polymers

2

Physical Testing of Plastics

In many applications, plastics are replacing traditional materials. Hence, there is

often a natural tendency to apply to plastics tests similar to those that have been

found suitable for gauging the performance of the traditional material. Dangers

can obviously arise if plastics are selected on the basis of these tests without clearly

recognising that the correlation between values of laboratory performance and field

performance may be quite different for the two classes of materials.

It is therefore important to realise that standard tests are not devised to give direct

prediction of end-use performance. In general, the reverse is the case. That is, if a

particular grade of a particular polymer is found to perform satisfactorily in a given

end use, it can then be characterised with reasonable accuracy by standard tests, and

the latter tests then used to ensure the maintenance of the required end-use quality.

Availability of instrumentation for mechanical testing is discussed in Appendix 1.

Typical mechanical properties of a range of polymers are listed in Table 1.1.

Table 1.1 Mechanical properties of polymers

Polymer

Tensile

strength

(MPa)

Flexural

modulus/

(modulus of

elasticity)

(GPa)

Elongation

at break

(%)

Strain

at yield

(%)

Notched

Izod impact

strength

(kJ/m)

Surface

hardness

Carbon/hydrogen-containing polymers

Low-density

polyethylene (LDPE)10 0.25 400 19 1.064 SD 48

High-density

polyethylene (HDPE)32 1.25 150 15 0.15 SD 68

Crosslinked

polyethylene (PE)18 0.5 350 N/Y 1.064 SD 58

Polypropylene (PP) 26 2 80 N/Y 0.05 RR 85

Ethylene-propylene 26 0.6 500 N/Y 0.15 RR75

Polymethyl pentene 28 1.5 15 6 0.04 RR 70

Styrene-butadiene 28 1.6 50 N/Y 0.08 SD 75

Styrene-ethylene-

butylene-styrene6 0.02 800 N/Y 1.064 SA 45

High-impact

polystyrene (PS)42 2.1 2.5 1.8 0.1 RM 30

PS, general purpose 34 3 1.6 RM 80

3

Mechanical Properties of Polymers

Oxygen-containing polymers

Epoxies, general

purpose600 80 1.3 N/A 0.5 RM 113

Acetal

(polyoxymethylene)50 27 20 8 0.10 RM 109

Polyesters (bisphenol),

polyester laminate

(glass filled)

280 16 1,5 N/A 1.064 RM 125

Polyester (electrical

grade)40 9 2 N/A 0.4 RM 125

Polybutylene phthalate 52 2.1 250 4 0.06 RM 70

Polyethylene

terephthalate (PET)55 2.3 300 3.5 0,02 RM 30

Polyether ether ketone

(PEEK)92 3.7 50 4.3 0,083 RM 99

Diallyliosphthalate 82 11.3 0.9 N/A 0.37 RM 112

Diallyl phthalate 70 10.6 0.9 N/A 0.41 RM 112

Alkyd resin glass fibre,

reinforced72 8.6 0.8 N/A 0.24 RM 125

Polyarylates 68 2.2 50 8.8 0.29 RR 125

Polycarbonate (PC) 50 2.1 200 3.5 0.05 RM 70

Polyphenylene oxide 65 2.5 60 4.5 0.16 RR 119

Phenol-formaldehyde 45 6.5 1.2 N/A 0.024 RM 114

Styrene-maleic

anhydride52 3 1.8 2 0.03 RL 105

Cellulose acetate 30 1.7 60 4 0.26 RR 71

Cellulose propionate 35 1.76 60 4 0.13 RR 94

Cellulose acetate

butyrate acrylics70 2.9 2.5 N/A 0.02 RM 92

Ethylene vinyl acetate 17 0.02 750 N/A 1.064 SA 85

Nitrogen-containing polymers

Polyamide (PA) 6 40 1 60 4.5 0.25 SD 75

PA 4,6 100 1 30 11 0.1 SD 85

PA 11 52 0.9 320 20 0.05 RR 105

PA 6,9 50 1.4 15 10 0.06 SD 78

PA 12 50 1.4 200 6 0.06 RR 105

PA 6,6 59 1.2 60 4.5 0.11 RR 90

PA 6, 12 51 1.4 300 7 0.04 RR 105

Nylon/acrylonitrile-

butadiene-styrene (ABS)

alloy

47 2.14 270 6 0.85 RR 99

PA-imide 185 4.58 12 8 0.13 RM 109

Polyimide 72 2.45 8 4 0.08 RM 100

Polyetherimide 105 3.3 60 8 0.1 RM 109

4


Polyurethane (PU)

thermoplastic elastomer24 0.003 700 N/Y 1.064 SA 70

Ether ester amide

elastomer

Urea formaldehyde

57 10 0.6 N/A 0.02 RM 115

Styrene acrylonitrile 72 3.6 2.4 3.5 0.02 RM 80

ABS 34 2.1 6 2 0.18 RR 96

Acrylate-styrene-

acrylonitrile35 2,.5 10 3.3 0.1 RR 106

Fluorine-containing polymers

Polytetrafluroethylene 25 0.70 400 70 0.16 RM 69

Poyvinylfluoride 40 1.4 150 30 0.18 SD 80

Polyvinylidene fluoride 100 5.5 6 N/A 0.12 SD 90

Perfluoroalkoxyethylene 29 0.7 300 85 1.064 SD 60

Ethylene tetrafluoro

ethylene28 1.4 150 15 1.064 RR 50

Ethylene chlorotrifluoro

ethylene30 1.7 200 5 1.064 RR 93

Fluorinated ethylene

propylene14 0.6 150 6 1.064 RR 45

Chlorine-containing polymers

Chlorinated poly(vinyl

chloride) PVC58 3.1 30 5 0.06 SA 70

Unplasticised PVC

(UPVC)51 3 60 3.5 0.08 RR 110

Plasticised PVC 14-20 0.007-0.03 280-95 N/Y 1.05+ SA 85

Sulfur-containing polymers

Polyphenylene sulfide 91 13.8 0.6 N/A 0.6 RR121

Polysulfone 70 2.65 80 5.5 0.07 RM69

Polyethersulfone 84 2.6 60 6.6 0.084 RM 85

Silicon-containing polymers

Silicones 28 3.5 2 N/A 0.02 RM 80

*These are typical room-temperature values of Notched Izod impact strength. A material that

does not break in the Izod test is given a value of 1.06 + kJ/m; the + indicates that it has a

higher impact energy than the test can generate.

N/A = if material is brittle and does not exhibit yield point

N/Y = if material is ductile and does not exhibit yield point

RM = Rockwell M 123 hardness (hard)

RR = Rockwell R 112 hardness

SA 65 = Shore A 65 hardness (soft)

SD 75 = Shore D75 hardness

Source: Author’s own files

5


Each of those measurements will now be discussed in further detail.

1.2 Tensile Strength

This is the room temperature tensile strength at yield for ductile materials and at

break for brittle materials. An excellent rating indicates high tensile strength, and

a very poor rating indicates low tensile strength. The values given in Table 1.1 are

typical room-temperature values taken at break for brittle materials and at yield for

ductile materials.

1.2.1 Electronic Dynamometer Testing of Tensile Properties

ATS FAAR supplies the Series TC200 computer-governed dynamometer, which can

carry out tensile, compression, and flexural tests on a variety of material. All details

of the tests are managed, including computing the final results and presenting them in

alphanumerical or graphical form. Moreover, the use of a personal computer (PC) for

the management of dynamometers gives the extra advantage of storing (at will) the

results of the tests in an electronic data file which can be utilised to obtain organised

print-outs, historical series, and go-no-go checks. Full automation of all operations

(including computations) increases the precision, reproducibility, and speed of analysis

due to the elimination of otherwise unavoidable human errors.

Tests that can be carried out by this dynamometer include those listed below and

those shown in Table 1.2:

specifications of ASTM D638-03 [1], DIN EN ISO, 527-1 [2] and DIN EN ISO

527-2 [3].

D695-02a [4] and DIN EN ISO 179-1 [5].

[6], ASTM D732 [7] and DIN EN ISO 178 [8].

6


Table 1.2 Mechanical properties of polymers using the Dyanometer test

Method ATS FAAR

Apparatus

code number

Measurement

units

Test suitable for meeting the

following standards

Tensile impact

strength

16.10050 kJ/m2 DIN EN ISO 8256 [9]

Tensile strength

(tensile modulus)

16.00121

16.00122

16.00126

16.00127

N/mm2

Psi

kg/cm2

ASTM D638-03 [1]

DIN EN ISO 527-1 [2]


Compression

strength and

compressive stress

16.00121

16.00122

16.00126

16.00127

N/mm2

Psi

N/mm2

ASTM D695-02a [4]

DIN EN ISO 604 [23]

Flexural strength

(Dyanometer)

16.00121

16.00122

16.00126

16.00127

N/mm2

Psi

kg/cm2

ASTM D790-03 [6]

ASTM D732 [7]

DIN EN ISO 178 [8]

ASTM = American Society for Testing and Materials

DIN EN ISO = Deutsches Institut für Normung Europa Norm International

Standards Organization


The TC 100 instrument can also undertake cycles between preset limits, creep and

relaxation studies, peeling experiments, and measurement of friction co-efficient and

puncture resistance of films.

Typical tensile strength data are quoted in Table 1.1.

Polymers of excellent tensile strength (i.e., >100 MPa) include: epoxies, polyester

laminates, polyamide (PA) 4/6, PA-imide, polyetherimide, polyvinylidene fluoride.

Polymers with very good impact strength (i.e., 60-100 MPa) include: acetals,

polybutylene terephthalete, polyethylene terephthalate (PET), polyallylphthalate,

polyetherketone, polycarbonate (PC), polyphenylene, styrene-maleic acid, copolymer,

acrylics, PA 11, PA 6/9, PA 12, polyimide 6/6 PA 6/12, polyimide: urea-formaldehyde,

styrene-acrylonitrile copolymer, chlorinated polyvinyl chloride (PVC), polyphenylene

sulfide, polysulfone, silicones, and alkyd resins.

7

Mechanical P

roperties of Polym

ers

Table 1.3 Identification of polymers with outstanding tensile strength and flexural modulus

Polymer Tensile strength (MPa)

Flexuralmodulus

(GPa)

Elongation at break

(%)

Strainat yield

(%)

NotchedIzod

(kJ/m)

Excellent or good performance in the

following

Poor performance in the following

Epoxies, general purpose

600 80 13 Notapplicable

(N/A)

0.5 Tensile strength, flexural modulus, detergent resistance, resistance to gamma radiation, heat distortion temperature (HDT), wear properties, low resin shrinkage during cure

Poor surface finish, high cost, elongation at break and volume resistivity

Polyesters(bisphenol polyester laminate) glass filled

280 16 1.5 N/A 1.06 + Impact strength and cheap when compared with epoxies

Heat resistance, solvent resistance, elongational at break

Polyester, sheet mouldingcompound

70 11 2.5 N/A 0.8 High flexural modulus, heat distortion temperature (HDT), resistance to UV and gamma irradiation, toughness

More expensive than polyesters. Dielectric constant, dielectric strength, flame spread, hydrolytic stability, elongation stability, HDT

PA 4,6 100 1 30 11 0.1 Reasonable HDT, good chemical resistance

Moisture absorption

Polyvinylidenefluoride (20% carbon fibre-reinforced)

100 5.5 6 N/A 0.12 Tensile strength, flexural modulus, HDT, detergent resistance, hydrolytic stability

Elongation at break, expensive, gamma irradiation resistance, dielectric properties, surface finish, toughness

8 Physical T

esting of Plastics

Polyetherimide 105 3.3 60 8 0.10 Tensile strength, cost, oxygen index, detergent resistance, gamma irradiation and UV resistance

Stress cracking with chlorinated solvents, notched impact strength, high cost, tracking, resistance, flexural modulus, toughness

Polyphenylenesulfide (glass fibre-reinforced)

91 13.8 0.6 N/A 0.031 Flexural strength, flexural modulus

Electrical properties, notched Izod

Diallyl iso-phthalate (long glass fibre-reinforced)

82 11.3 0.9 N/A 0.37 Tensile strength, flexural modulus, notched Izod, detergent resistance, HDT, tracking resistance, UV radiation

Volume resistance, dielectric strength, dissipation factor, flammability, hydrolytic stability, high cost, elongation at break

Diallyl phthalate 70 10.6 0.9 0.41 Tensile strength, flexural modulus, notched Izod, detergent resistance, HDT, tracking resistance, UV radiation

Volume resistance, dielectric strength, dissipation factor, flammability, hydrolytic stability, high cost, elongation at break

Polysulfone 70 2.65 80 5.5 0.07 High temperature performance, electrical properties, hydrolytic resistance, tensile strength, dimensional stability

Surface finish

Polyacrylates 68 2.2 50 15 1.06 + Impact strength, heat resistance and UV resistance

Stress cracking with hydrocarbons, needs 350 °C processing temperature.


9


In Table 1.3 are listed those polymers of high impact tensile strength and also those

which have a high flexural modulus. This shows that a high tensile strength is no

guarantee of a high flexural modulus.

In selecting a polymer for a particular end-use application, it is frequently necessary

to compromise on polymer properties. Thus, while epoxy resins have excellent tensile

strength and flexible modulus as well as detergent resistance, they have low resistance

to gamma radiation, poor heat distortion performance, and a poor wear properties.

They are also expensive and have a poor volume sensitivity and surface finish.

Son and co-workers [11] showed that the addition of a thermothropic liquid crystalline

aromatic polyester produces an improvement in the tensile strength and modulus of

blends with polyether ketone while simultaneously producing a significant decrease

in elongation at break.

Tasdemir and Yildirim [12] showed that rigid plastics such as polystyrene (PS), PVC,

polymethyl methacrylate, polypropylene (PP), Nylons and PC, epoxies, unsaturated

polyester resins, and PA can be toughened and their impact properties improved by

incorporation in the formulation of 5-20% of a rubbery elasomer such as acrylonitrile-

butadiene-styrene (ABS) terpolymer.

1.3 Flexural Modulus (Modulus of Elasticity)

This is the short-term modulus of materials at specified temperatures. Ratings for

flexural modulus have been assigned at 20 ºC and are usually determined at ~1%

strain. An ‘excellent’ rating indicates a high flexural modulus; a ‘very poor’ rating

indicates a low flexural modulus. A ‘not applicable’ status indicate that the material

has a modulus of limited practical use.

1.3.1 Torsion Test

AST FARR supplies a torsion test instrument for the measurement of the apparent

modulus of elasticity versus temperature of plastics and elastomers (Table 1.4 and

Appendix) following specifications ASTM D1043 [14], DIN 53477 [15], ISO 498

[22] and BS 2782 [13].

10


Table 1.4 Modulus of elasticity of polymers

Method ATS FAAR

Apparatus

code

number

Measurement

units


following standards

Modulus

elasticity

versustemperature

(torsion test)

10.22000

10.22001

GPa ASTM D1043 [14]

DIN 53477 [15]

ISO 458 [16]

BS 2782 [13]

Modulus

of elasticity

or flexural

modulus

(bend test)

10.29000 kN/mm2

psi

kg/cm2

ASTM D1896 [17]

ASTM D3419 [18]

ASTM D3641 [19]

ASTM D4703 [20]

ASTM D5227 [21]

DIN EN ISO 178 [8]



DIN EN ISO 604 [26]


This apparatus consists of an assembly of interdependent units. It is compact and

very easy to operate. It features an accurate torque applications system with a broad

angular measurement range.

The temperature is electronically controlled and the fully automatic system of pre-

setting heating periods and torque is electronic. The apparatus is equipped with a

set of weights allowing a wide range of torques to be applied to specimens having

difference dimensions. The distance between the clamps can be varied in the range

20-100 mm.

It is thus possible to measure accurately the apparent modulus of elasticity of specimens

obtained from a wide range of materials. The modulus is, in this case, defined as

‘apparent’ because it is obtained by measuring the angular torsion of the specimens

under test.

In its regular version, the apparatus can conduct tests in the temperature range -100

ºC to +100 ºC. A special version, however, allows one to run tests between 100 ºC

and +300 ºC. The torque application system is assembled on ball bearings of very

11


reduced dimensions and utilises the same ‘pivot’ contrivance used in watch making.

As a consequence, all radial frictions are reduced to zero and the main pulley can be

rotated even if applying torques <0.0005 N/m. The initial zero point can be adjusted

so as to allow tests to be undertaken even on specimens having planar deformation

between +5º and -5º. Cooling is obtained by solid carbon dioxide or by liquid nitrogen

so as to achieve maximum operational flexibility.

1.3.2 Hand Test

ATS FAAR also supply hand test apparatus for the measurement of the modulus of

elasticity (flexural modulus) according to specifications ASTM D1869 [17], ASTM

D3419 [18], ASTM D3641 [19], ASTM D4703 [20], D ASTM 5227 [21], DIN EN

ISO 178, [8] DIN EN ISO 527-1 [2], DIN EN ISO 527-2 [3] and DIN ISO 604 [23]

(see Table 1.4).

Table 1.4 Modulus of elasticity of polymers

Method ATS FAAR

Apparatus code

number

Measurement

units

Test suitable for meeting

the following standards

Modulus

elasticity versustemperature

(torsion test)

10.22000

10.22001

GPa ASTM D1043 [14]

DIN 53477 [15]

ISO 458 [16]

BS 2782 [13]

Modulus

of elasticity

or flexural

modulus (bend

test)

10.29000 kN/mm2

psi

kg/cm2

ASTM D1896 [17]

ASTM D3419 [18]

ASTM D3641 [19]

ASTM D4703 [20]

ASTM D5227 [21]

DIN EN ISO 178 [8]



DIN EN ISO 604 [26]


Janick and Krolikowski [24] investigated the effect of charpy notched impact strength

on the flexural modulus of PE and PET. Polymers with good flexible modulus (between

12


6 GPa and 80 GPa) include polydiallyephthalate (11.3 GPa), phenol-formaldehyde

(6.5 GPa), alkyd resins (8.6 GPa) and polyphenylene sulfide (13.8 GPa) as well as

glass-filled polyesterlaminate (16 GPa) epoxy resins (80 GPa), silica-filled epoxies

(15 GPa) and acetals containing 30% carbon fibre (17.2 GPa).

1.4 Elongation at Break

Elongation at break is the strain at which a polymer breaks when tested in tension at

a controlled temperature (i.e., the tensile elongation at specimen break).

An increasing number of applications are being developed for thermoplastics in which

a fabricated article is subjected to a prolonged continuous stress. Typical examples

are pipes, crates, cold water tanks, and engine cooling fans. Under such conditions

of constant stress, materials exhibit (to varying extents) continuous deformation with

increasing time. This phenomenon is termed ‘creep’.

A wide variety of materials will, under appropriate conditions of stress and

temperature, exhibit a characteristic type of creep behaviour (Figure 1.1).

Stra

in

A

B

C

D

0Time

Figure 1.1 Creep behaviour of PE. Source: Author’s own files

13


The general form of this creep curve can be described as follows. Upon application

of the load, an instantaneous elastic deformation occurs (O-A). This is followed

by an increase in deformation with time as represented by the portion of the curve

A-D. This is the generally accepted classic creep curve, and is usually considered to

be divided into three parts:

(i.e., a constant creep rate).

With thermoplastics, the secondary stage is often only a point of inflection, and the

final tertiary stage is usually accompanied by crazing or cracking of the specimen

or, at higher stresses, by the onset of necking (i.e., marked local reduction in cross-

sectional area). In assessing the practical suitability of plastics, we are interested in

the earlier portions of the curve, prior to the onset of the tertiary stage, as well as in

the rupture behaviour.

As the values of the stress, temperature, and time of loading vary with differing

applications, the type of information extracted from these basic creep data will also

vary with the application. There are several alternative methods (some of which are

discussed later in this book) of expressing this derived information, but whichever

form is used (whether tabular of graphical), it is unlikely to cover all eventualities.

It is thus considered of most value to give the actual creep curves and to discuss by

means of examples the methods by which particular data can be extracted from these

basic creep curves.

Typical creep curves for PE and PP at 23 ºC are shown in Figures 1.2 and 1.3. Lin

and co-workers [25] discussed the accuracy of creep phenomena in reinforced PA

and PC composites. The phenomenon of increasing dynamic creep and temperature

under tension-tension fatigue loading are compared between semi-crystalline and

amorphous composites.

14


6

5

4

3

2

1

00.1 1 10

62 MPa

48 MPa

34 MPa

21 MPa

1000

Total st

rain

(%)

Time (hours)

Figure 1.2 Creep of high-density polyethylene (HDPE) at 23 ºC. Source: Author’s

own files

6

5

4

3

2

1

00.1 1 10

97 MPa

117 MPa

138 MPa

159 MPa

55 MPa

34 MPa

76 MPa

1000100

Total st

rain

(%)

Time (hours)

Figure 1.3 Creep of PP homopolymer at 23 ºC. Source: Author’s own files

15


To minimise as far as possible the influence of processing variables, studies have

been carried out using tensile creep tests on carefully prepared compression moulded

specimens. With injection moulded articles, creep properties will also be subject to

variation with the amount and direction of residual flow orientation. However, with

crystalline polymers such as the polyolefins, creep effects will also be influenced

by variations in density caused by a combination of flow orientation, compressive

packaging and cooling effects. Stresses will generally be complex and often involve

compressive and flexural components. However, articles should be designed to limit the

strains occurring to quite low levels, where a reasonable correlation can be expected

between tensile, compressive and flexural creep data.

1.4.1 Basic Creep Data

In general, the first step in building up a comprehensive picture of creep behaviour

is to obtain creep curves (elongation versus time) at a series of stress levels, each at a

series of test temperatures. It is common practice to plot strain on a linear scale against

time on a logarithmic scale. Each curve should preferably cover several decades on the

logarithmic time scale so that subsequent extrapolations will have a firm foundation.

Wong and co workers [26] combined a commercially available research-grade

dynamic mechanical analyser with a Fourier transform infrared (FTIR) spectrometer

for the simultaneous mechanical analysis and dynamic IR spectral measurement of

films of a polyester urethane elastomer under large amplitude creep deformation.

Differential orientation of various segments of the molecule was observed during

the creep and recovery process. Permanent damage to the elastomer observed after

a large displacement was attributed to the irreversible destruction of its microscopic

network structure.

Table 1.5 lists polymers with outstanding percentage elongation at break. High

elongation at break is not necessarily accompanied by high tensile strength. Also

listed in Table 1.5 are the particularly good and particularly poor performance

characteristics which accompany high elongation at break.

Janick and Krolikowski [24] studied the effect of properties such as elongation at

break, tensile strength, flexural strength, elastic modulus and fleural modulus on

the elongation at break performance of PET–low-density polyethylene (LDPE) and

PET-PP blends.

16

Physical T

esting of Plastics

Table 1.5 Identification of polymers with outstanding elongation at break

Polymer Tensile strength(MPa)

Flexuralmodulus

(GPa)

Elongation at break

(%)

Strainat yield

(%)

Notched Izod

(kJ/m)

Excellent or very good performance

Poor performance

Low-densitypolyethylene

10 0.25 400 19 1.06 + Elongation at break, chemical resistance, electrical properties, cost, toughness

Tensile strength, stiffness, flammable, UV resistance, environmental stress cracking

Styrene-ethylene-butylene-styrene

6 0.02 800 Noyield(NY)

1.06 + Elongation at break, low-temperatureproperties, electrical properties, toughness

Abrasion resistance, tensile strength, flame spread, flexural modulus

Polyethyleneterephthalate

55 2.3 300 3.5 0.02 Elongation at break, stiffness, detergent resistance, electrical properties, elongation at break

Limited hydrolysis resistance, high mould shrinkage

Ethylene-vinyl acetate (25% vinyla acetate)

17 0.02 750 NY 1.06 + Elongation at break, cost, strain at yield, toughness

Tensile strength, chemical resistance, dissipation factor, flammability, heat distortion temperature, flexural modulus

Polyurethane (thermo-plasticelastomer)

24 0.003 700 NY 1.06 + Elongation at break, low temperature characteristics, detergent resistance, fatigue index, strain at yield, toughness

Tensile strength, flexural modulus

17

Mechanical P

roperties of Polym

ers

Polyamide12

50 1.4 200 6 0.06 Elongation at break, low water absorption, toughness, dielectric strength

High cost, strength and heat resistance

Polyamide6,12

51 1.4 300 7 0.04 Low water absorption, detergent resistance, elongation at break,

Low heat distortion temperature, dissipation factor, flammability, flexural modulus

Polyamide11

52 0.9 320 20 0.05 Elongation at break, impact strength, heat resistance, fatigue index, toughness

Cost, impact strength

Polyamide6,9

50 1.4 - - - - -


18


Lyons [27] obtained creep rupture data and tensile behaviour for glass-filled PA and

polyphthalamide at 23-150 ºC.

A polyphthamide with 33% glass reinforcement exhibited a good combination of

creep resistance, strength and ductility.

1.5 Strain at Yield

This is an indication of the degree of strain that a material can accept without

yielding. The values given are typical room-temperature values. If a material is brittle

and does not exhibit a yield point, then the value given in Table 1.1 is termed ‘NA’.

If the material is ductile and does not exhibit a yield point, then the value is given

as ‘NY’ in Table 1.1. An excellent rating indicates a ductile material with no yield

and high elongation to break. ‘Not applicable’ indicates a brittle material with low

elongation to break.

Materials rated ‘very good’ to ‘very poor’ include all those exhibiting yield, ranked

according to their strain at yield, as well as ductile materials with lower elongation

to break than those rated as ‘excellent’, and brittle materials with higher elongation

to break than those rated as ‘not applicable’. In general, ductile materials range from

‘excellent’ to ‘very poor’ by descending elongation to break, and brittle materials

from ‘poor’ to ‘not applicable’ by descending elongation to break.

1.5.1 Isochronous Stress-strain Curves

Where it is necessary to compare several different materials, basic creep curves alone

are not completely satisfactory. This is particularly so if the stress levels used are

not the same for each material. If the stress endurance time relevant to a particular

application can be agreed, a much simpler comparison of materials for a specific

application can be made by means of isochronous stress-strain curves.

As an example, let us consider a milk crate in which we can assume that the moulding

will not be continuously loaded for times >100 hours. From the basic creep curves, the

strain values at the 100 hours point for various stresses can be readily determined. For

each material, a stress-strain curve can now be drawn corresponding to the selected

loading time of 100 hours. Now let us suppose that for this application it is further

stipulated that after 100 hours of continuous loading the strain shall be ≤1%. The

stress that can be sustained without violating the strain-endurance stipulation can then

be conveniently read off for each material from the isochronous stress-strain curves,

thereby indicating with reasonable confidence the stress than can be used for design

19


purposes for each of the materials considered. Examples of these interpolations are

demonstrated in Figure 1.4.

4

3

2

1

0 0

250

500

750

1000

0.1 1 0 1 2 3 410

5.3 MPa

7 MPa

3.5 MPa

1.8 MPa

102 103

Tota

l str

ain

(%)

Stre

ss (

psi)

Total strain (%)Time (hours)

Figure 1.4 Isochronous stress-strain curve of PE. Source: Author’s own files

Hay and co-workers [28] developed an approximate method for the theoretical

treatment of pressure and viscous heating effects on the flow of a power-law fluid

through a slit die. The flow was assumed to remain one-dimensional and the

accuracy of this approximation was checked via finite element simulations of the

complete momentum and energy equations. For pressures typically achieved in the

laboratory, it was seen that the one-dimensional approximation compared well with

the simulations. The model therefore offered a method of including pressure and

viscous heating effect in the analysis of experiments, and was used to rationalise

experimentally pressure profiles obtained for the flow of polymer melts through a slit

die. Data for the flow of a LDPE and a PS melt slit die showed that these two effects

were significant under normal laboratory conditions. The shear stress-strain rate

curves would thus be affected to the point of being inaccurate at high shear rates. In

addition, it was found that the typical technique to correct for a pressure-dependent

viscosity was also inaccurate, being affected by the viscous heating and heat transfer

from the melt to the die.

Thermomechanical analysis is an ideal technique for analysing fibres because the

measured parameters - dimension change, temperature and stress - are major variables

that affect fibre processing. Figure 1.5 shows the thermal stress analysis curves for

20


a polyolefin fibre as received and after cold drawing. In this experiment, fibres are

subjected to initial strain (1% of initial length), and the force required to maintain

that fibre length is monitored. As the fibre tries to shrink, more force must be exerted

to maintain a constant length. The result is direct measurement of the shrink force of

the fibre. Shrink force reflects the orientation frozen into the fibre during processing,

which is primarily related to the amorphous portions of the fibre. Techniques that track

fibre crystallinity, therefore, are not as sensitive a measure of processing conditions

as thermomechanical analysis. In this case, the onset of the peak of the shrink force

indicates the draw temperature, whereas the magnitude of the peak is related to the

draw ratio of the fibre. It has been shown that the area under the shrink force curve

(from the onset to maxima) can be correlated to properties such as elongation at

break and knot strength. Other portions of the thermomechanical analysis thermal

stress plot can yield additional information. For example, the initial decreasing slope

is related to the expansion properties of the fibre, and the appearance of secondary

force peaks can be used to determine values such as heat set temperature in Nylon.

0.0

0.2

0.4

0.6

Force

(N)

Temperature (°C)20 40 60 80 100 120 140 160 180 200

As received

Cold drawn

Figure 1.5 Application of thermomechanical analysis to thermal stress analysis of

polyolefin films (as received and cold drawn). Source: Author’s own files

21


1.5.2 Stress-time Curves

With thermoplastic materials, above a certain stress, the stress-strain curve shows a

marked departure from linearity (i.e., above such values further increases in stress lead

to disproportionately greater increases in elongation). By study of the isochronous

stress-strain curves of a material it is possible to decide upon a certain strain that

should not be exceeded in a given application. The stress required to produce this

critical strain will vary with the time of application of the load, so that the longer

the time of application, the lower will be the permissible stress.

From isochronous stress-strain curves relating to endurance times of, for example,

1, 10, 100, and 1,000 hours, the magnitude of stress to give the critical strain at

each duration of loading can be readily deduced. This procedure can be repeated for

different selected levels of strain. In general, the more critical the application and the

longer the time factor involved, the lower will be the maximum permissible strain.

For the polyolefin family of materials, the upper limit of critical strain will always be

governed by the onset of brittle-type rupture (e.g., hair cracking at elongations that

are low compared with those expected from the short-term ductility of the material).

With thermoplastics of intermediate modules (e.g., HDPE, PP) the stress-strain curves

depart slightly from linearity at quite low strains. Thus, if accurate results are to be

obtained from standard formulae, it is sometimes necessary to limit the critical strain

and the corresponding design stress to low values. One commonly selected criterion

with high-modulus thermoplastics (e.g., polyacetal) is to base the design stress and the

design modulus on that point on the stress-strain curve at which the secant modulus

falls to 85% of the initial tangent modulus. This procedure is illustrated in Figure 1.6

It would be wrong, however, in many applications to limit the critical strain to a low

value if the ductility of the material is to be fully utilised. Extreme examples are those

such as pipes where rupture behaviour is the controlling factor.

The above type of data can be conveniently summarised by the presentation of stress-

time curves corresponding to different levels of permissible strain. Such curves enable

the time-dependency of different materials to be conveniently compared. If possible,

rupture data should also be included on such plots so that at longer times an adequate

safety margin over the rupture is always maintained.

As with normal creep curves, the time scale will normally be logarithmic. For

extrapolation purposes, however, a linear relationship can be obtained by also using

a logarithmic scale for the stress axis. In fact, if rupture data are to be included, a

double logarithmic plot is preferred.

22


C

B

Stre

ss

StrainO A

ABAC

85100

=

Figures 1.6 Stress-time curve. Source: Author’s own files

1.5.3 Stress-temperature Curves

The methods discussed so far have been concerned with the presentation of combined

creep data obtained at a single temperature. Thus, a further method is required to

indicate the influence of temperature.

One convenient method is to combine the information available from isochronous

stress-strain curves or stress-time curves obtained on the same materials at different

temperatures. For example, suppose the performance criterion for a particular

application is that the total strain should be ≤2% in 1,000 hours. Using the 1,000

hours isochronous stress-strain curve for each temperature and erecting an ordinate at

the 2% point on the strain axis, the individual working stresses for each temperature

can be obtained. Alternatively, by erecting an ordinate at the 1,000 hours point on

the stress-time curve for 2% strain for each temperature investigated, the individual

working stress can be similarly obtained. From these interpolated results, the stress-

temperature curve can be drawn.

By similar procedures stress-temperature plots for other specific ‘time-permissible

strain’ combinations can be obtained.

23


1.5.4 Extrapolation Techniques

Normal creep curves in which percentage total strain (or total strain) on a linear scale

is plotted against time on a logarithmic scale show increasing curvature with time,

particularly at higher stress levels. Due to this curvature it is not possible to extrapolate

curves to longer times with certainty. If, however, a double-logarithmic plot is used

for a series of stresses at a given temperature, a family of parallel straight lines is

obtained. These can then be extrapolated quite easily. With PE and PP, the slope of

such curves remains constant or tends to decrease at very long times. Furthermore,

an increase in temperature also leads to a slight reduction in the slope of the family of

curves for a given material. A straightforward linear extrapolation of the log (strain)/

log (time) creep curves may thus involve an additional safety factor.

As mentioned above, stress-curves on a double-logarithmic scale are also linear and

can be easily extrapolated. In both cases, however, over-extrapolation should be

avoided (i.e., one decade and preferably not more than two).

1.5.5 Basic Parameters

Within a given family of materials of the same basic polymer, further condensation

of data can be achieved if the principal polymer parameters influencing creep can be

separated. For example, with PE, density and the melt index (MI) would be obvious

possibilities.

If such factors can be found then all the results can be combined in suitable trend

curves and the results for any intermediate grade interpolated. If good correlations

are obtained, the reliability in the individual results is automatically increased and

one can feel confident of predicting the creep behaviour of any material within a

particular family.

1.5.6 Recovery in Stress Phenomena

In some applications, the load, instead of being constantly applied, will be applied

only intermittently for limited periods. In the period between consecutive loadings

the part will thus have an opportunity to recover. The extent to which this occurs will

depend upon the ratio of the loaded time to the unloaded time. The recovery will be

approximately exponential. Thus, to achieve virtually complete recovery, the time

without load must be substantially longer than the time under load. The residual strain

for a fixed ratio of loaded to unloaded time will also depend upon the magnitude of

the strain experienced at the time of removal of the load.

24


With intermittent loading for relatively long time periods, higher permissible stresses

will be possible than with continuous loading. With repeated loading over short time

periods, however, fatigue effects could considerably reduce the value of the permissible

stress for a long total endurance.

1.5.7 Stress Relaxation

In many applications such as those involving an interference fit (e.g., pipe couplings,

closures), we are concerned more with the decrease in stress with time at constant

strain (stress relaxation) rather than with the increase in strain with time under

constant stress (creep).

To a first approximation, such data can be obtained from the basic family of creep

curves by sectioning through them at the relevant strain value parallel to the time

axis. That is, similar procedures are used as for obtaining stress-time curves.

1.5.8 Rupture Data

In applications in which strains up to the order of a few percent do not interfere with

the serviceability of the component, the controlling factor as far as permissible stress

is concerned will probably be the rupture behaviour of the material. The rupture

curve forms the limiting envelope of any family of creep curves. Care must therefore

be exercised in extrapolating creep data to ensure that for long-term applications a

suitable safety factor is always allowed over the corresponding extrapolated rupture

stress. In general, with thermoplastics, an increase in MW leads to improved rupture

resistance, which is not necessarily reflected by an improvement in creep resistance

and in fact may be associated with a decrease in creep resistance. An increase in

MW will also, in general, be associated with a decrease in processability, which will

tend to give higher residual stresses in fabricated articles which can in turn influence

rupture properties. In designing injection moulded articles, particular care should

be taken to ensure that the principal stresses experienced by the article in service

are (wherever possible) parallel to the direction of flow of material in the mould.

Avoidance of stress concentration effects, by appropriate design, will tend to minimise

the interference of rupture phenomena and allow greater freedom to design on the

basis of the creep modulus.

Certain environments reduce rupture performance with particular materials (e.g., PE

in contact with detergents) and allowance must be made for this, where necessary,

by the incorporation of an additional safety factor to the rupture stress. In general,

25


the resistance to rupture will be greater if the stress is applied in compression than

in tension.

Differing stress levels have been used for HDPE and LDPE for the reasons previously

discussed, so an easier comparison between grades can be made by the use of

isochronous stress-strain curves. These are shown in Figures 1.7-1.10 for times of

1, 10, 100, 1,000 hours, respectively.

Stre

ss (

MPa

)

Total strain at 1 hour (%)

00

0.7

1.4

2.1

2.8

3.4

4.1

4.8

65-045

58-04550-075

30-005

25-040

18-070

Actual Densities - g/ml

65-045 0.963058-045 0.956550-075 0.9505

30-005 0.927025-040 0.922018-070 0.9160

5.5

6.2

1 2 3 4 5 6 7 8

}High Density

}Low Density

Figure 1.7 Isochronous stress-strain curves of various grades of LDPE and HDPE

at 23 ºC (1 hour data). Source: Author’s own files

26


The data have been further combined in Figure 1.7, which demonstrates the marked

(but systematic) influence of density on the creep behaviour of PE. The curves in

Figure 1.11 are relevant to a total strain of 1%, but similar plots for other permissible

strains can be readily derived from the isochronous stress-strain curves. The linear

relationship between creep and density for PE at room temperature, irrespective of

the melt index over the range investigated (i.e., 0.2 to 5.5), has enabled the stress-time

curve of Figure 1.12 to be interpolated for the complete range of PE and for a range

of PP (Figure 1.13). In this case, the data have been based on a permissible strain of

2% but, as previously explained, data for other permissible strains can be similarly

interpolated from the creep curves. The effect of temperature on the creep properties

of PE is summarised in Figure 1.14.

Stre

ss (

MPa

)


00

0.7

1.4

2.1

2.8

3.4

4.1

4.8

65-045

58-04550-075

30-005

25-040

18-070


65-045 0.963058-045 0.956550-075 0.9505

30-005 0.927025-040 0.922018-070 0.9160

5.5

6.2

1 2 3 4 5 6 7 8

}High Density

}Low Density


at 23 ºC (10 hours data). Source: Author’s own files

27

Mechanical Properties of PolymersSt

ress

(M

Pa)


00

0.7

1.4

2.1

2.8

3.4

4.1

4.8

65-045

58-04550-075

30-005

25-040

18-070


65-045 0.963058-045 0.956550-075 0.9505

30-005 0.927025-040 0.922018-070 0.9160

5.5

6.2

1 2 3 4 5 6 7 8

}High Density

}Low Density


at 23 ºC (100 hours data). Source: Author’s own files

28

Physical Testing of PlasticsSt

ress

(M

Pa)


00

0.7

1.4

2.1

2.8

3.4

4.1

4.8

65-045

58-04550-075

30-005

25-040

18-070


65-045 0.963058-045 0.956550-075 0.9505

30-005 0.927025-040 0.922018-070 0.9160

5.5

6.2

1 2 3 4 5 6 7 8

}High Density

}Low Density


at 23 ºC (1,000 hours data). Source: Author’s own files

29

Mechanical Properties of PolymersSt

ress

to

give

1 %

tot

al s

trai

n (M

Pa)

Actual density of polyethylenes (g/ml)

0.7

1.4

2.1

2.8

3.4

4.1

4.8

1000 hours

100 hours

10 hours

1 hours

0.920 0.930 0.940 0.950 0.960 0.970

Figure 1.11 Effect of density on creep of PE at 23 ºC. Source: Author’s own files

30


Stre

ss fo

r 2%

tot

al s

trai

n (M

Pa)

Time (hours)

1.4

0

2.8

4.1

5.5

6.9

8.3

100 100

0.9650.960

0.955

0.950

0.940

0.930

0.925

0.920

1,000 10,000

Figure 1.12 Interpolated stress-time curves of PE of various densities (2% total

strain). Source: Author’s own files

Stre

ss fo

r 2%

tot

al s

trai

n (M

Pa)

Time (hours)

00

0

00

00

00

00

00

100 100

KM 61

KMT 61

DE 61

GM 61

GMT 61

1,000 10,000

Figure 1.13 Interpolated stress-time curves of various grades of PP at 23 ºC (2%

total strain). KMT 61 and GMT 61= PP copolymers. KM 61, GM 61 and DE 61

= PP homopolymer. Source: Author’s own files

31


Stre

ss t

o gi

ve 2%

tot

al s

trai

n in

100

0 ho

urs

(MPa

)

Temperature (ºC)

200

0.7

1.4

2.1

2.8

3.4

4.1

4.8

5.5

30 40 50 60

0.960

0.950

0.940

0.930

0.920

70 80

Figure 1.14 Interpolated stress-temperature curves of PE of different densities (2%

total strain at 1,000 hours). Source: Author’s own files

32


For PE, the principal factor controlling creep is density but the principal factor

influencing rupture behaviour is the melt index. Thus, for a given density, for long-term

applications, the higher the melt index the lower will be the maximum permissible

strain. This is particularly important at elevated temperatures.

The effect of temperature on the creep of PP is also conveniently summarised in

Figures 1.15 and 1.16. Although these curves refer specifically to 0.5% and 2% total

strain in 1,000 hours, similar interpolations have shown that to a first approximation

the fall off in stress with temperature is similar for other strain-time combinations.

In contrast to PE, an excellent correlation has been found between the creep and

melt index of PP. An increase in the melt index is associated with an increase in creep

resistance. This has been found for homopolymers and copolymers, and is illustrated

in Figure 1.17. For similar melt indices the inferior creep resistance of the copolymers

is also indicated, this difference being more marked for grades of lower melt index.

1.5.9 Long-term Strain-time Data

To illustrate the linearity of double-logarithmic plots of percentage total strain against

time, Figure 1.18 shows a typical family of curves at a range of stress levels for a

HDPE and Figure 1.19 shows that for a PP copolymer. From the uniform pattern

of behaviour, the linear extrapolation of the curves at lower stresses to longer times

appears well justified. As emphasised above, with the higher-melt-index materials at

higher temperatures, over-extrapolation is dangerous because of the possible onset

of rupture at comparatively low strains.

Figures 1.20 and 1.25 shows creep curves, isochronous stress-strain curves and

stress-time curves for a typical ABS terpolymer and a typical unplasticised polyvinyl

chloride (UPVC).

33


Stre

ss t

o gi

ve 0

.5%

tot

al s

trai

n in

100

0 ho

urs

(MPa

)

Temperature (ºC)

200

0.7

1.4

2.1

2.8

3.4

GM 61

KM 61

KMT 61

GMT 61

DE 61

30 40 50 60 70 80

Figure 1.15 Interpolated stress-temperature curves for various grades of PP (0.5%

total strain at 1,000 hours). DE 61 = PP copolymers. Source: Author’s own files

34

Physical Testing of PlasticsSt

ress

to

give

0.5

% t

otal

str

ain

in 1

000

hour

s (M

Pa)

Temperature (ºC)

200

1.4

2.8

4.1

5.5

6.9

8.3

9.7

GM 61

KM 61

KMT 61

GMT 61

DE 61

30 40 50 60 70 80

Figure 1.16 Interpolated stress-temperature curves for various grades of PP (2%

total strain at 1,000 hours). DE 61 = PP homopolymers. Source: Author’s own files

35


Stre

ss fo

r 2%

tot

al s

trai

n (M

Pa)

Melt index (g/10 min)

2.8

0

5.5

8.3

11.0

13.8

18.5

100 100

1 hour

1 hour

10 hours

10 hours

100 hours

100 hours

1000 hours

1000 hours

Homopolymers

Copolymers

1,000

x

x

x

x

x

Figure 1.17 Effect of melt index on creep of PP at 23 ºC. Source: Author’s own

files

Tota

l str

ain

(%)

Time (hours)

0.10.1

1

10

1 10 100 1,000

6.2 MPa4.8 MPa3.4 MPa

2.1 MPa

1 yea

r

10 ye

ars

10,000 100,000

Density 0.96 gm/mlMelt index 0.3 gm/10 min (190 °C 2.16 kg)

Figure 1.18 Long-term creep of HDPE at 23 ºC. Source: Author’s own files

36

Physical T

esting of Plastics

10

1

0.1

0.1 1 10 100 1000

Time (hours)

Tota

l str

ain

(%)

1 yea

r

3.4 MPa

5.5 MPa

7.6 MPa

9.7 MPa

11.7 MPa

13.8 MPa

19.5 MPa

10,000

Figure 1.19 Long-term creep of high-density PP copolymer at 23 ºC. Source: Author’s own files

37


0.1 1 10 100 1000

Time (hours)

6.9 MPa

13.8 MPa

20.7 MPa

27.6 MPa

Tota

l str

ain

(%)

1

2

3

4

5

6

Figure 1.20 Creep of typical ABS terpolymer (normal impact grade) at 23 ºC.


10

0

0

0

0

0

0

10 100 1000

3% strain

2% strain

1% strain

0.5% strain

10,000

Time (hours)

Stre

ss t

o pr

oduce

a gi

ven

stra

in (

MPa

)

Figure 1.21 Interpolated stress-time curves for a typical ABS terpolymer (normal

impact grade) at 23 ºC. Source: Author’s own files

38


10.1 10 100 1000

34.5 MPa

25.9 MPa

17.2 MPa

8.6 MPa

Time (hours)

Tota

l str

ain

(%)

0

1

2

3

4

5

6

Figure 1.22 Creep of typical UPVC pipe formulation at 23 ºC. Source: Author’s

own files

Time (hours)

0.5% strain

1% strain

2% strain

3% strain

Stre

ss t

o pr

oduce

a gi

ven

stra

in (

MPa

)

0

6.9

13.8

20.7

27.6

34.5

41.4

0 10 100 1000 10,000

Figure 1.23 Interpolated stress-time curves for a typical UPVC pipe formulation at

23 ºC. Source: Author’s own files

39


Figure 1.24 Injection moulded dish used in measurement of falling weight strength.


Figures 1.25 Injection moulded ruler used in the study of weld effects (length, 37.5

cm). Source: Author’s own files

Stress-strain data have been reported for several polymers, including ultra-high MW

PE [29-30] syndiotactic PP [31] PA 6 [32], PP [33], PA 6,6 [34] and LDPE aluminium

foil laminates [35].

40


1.6 Impact Strength Characteristics of Polymers

Of the many properties of a plastic that influence its choice for a particular article or

application, the ability to resist the inevitable sharp blows and drops met in day-to-

day use is one of the most important. The prime object of impact testing should be

to give a reliable guide to practical impact performance. However, the performance

requirements and the design and size of articles can be vary considerably. The method

of fabrication can also vary and, because all these factors can influence impact

performance, a reasonably wide range of data are required if most of these eventualities

are to be covered and materials and grades compared sensibly.

Available instrumentation is listed in Table 1.6.

Table 1.6 Impact resistance (Izod Charpy) and notched impact resistance

ATS FAAR

Number

Units Suitable test

Izod Charpy impact test

Pendulum impact strength

(Izod and Charpy) and notched

impact strength

16.10000

16.10500

16.10600

ft.lb/in

kJ/m2

Nm/m

ASTM D26-56 [36]

DIN EN ISO 179 [37]

ISO 179-1 [5]

ISO 180 [38]

BS 2782 [13]

UNI EN ISO 180 [39]

Falling dart impact test (Fall-o-Scope ATS FAAR)

Impact resistance (free falling

dart)

ASTM D5420 [40]

ASTM D5628 [41]

DIN EN ISO 6603 [42]

Guided universal impact test (Fall-o-Scope ATS FAAR)

Impact resistance at low

temperature (cable)


1.6.1 Notched Izod Impact Strength

Of all the types of mechanical test data normally presented for grade characterisation

and quality-control purposes, the standard notched Izod test is probably the most

41


vigorously criticised on the grounds that it may give quite misleading indications

regarding impact behaviour in service. In most cases, however, the criticism should

be levelled at too wide an interpretation of the results, particularly the tendency to

overlook the highly important influence of the differences between the standard and

operational conditions in terms of dimensions, notch effects, and processing strains.

For example, in the Izod test, the specimens are of comparatively thick section (e.g.,

6.25 × 1.25 × 1.25 cm, or 6.25 × 1.25 × 0.625 cm) and are consequently substantially

unoriented. Even with injection moulded specimens (where some orientation does

exist), specimens are usually tested in their strongest direction (with the crack

propagating at a right angle to the orientation direction), which generally leads to high

values. Furthermore, specimens are notched with the specific intention of locating the

point of fracture and ensuring that as wide a range of materials as possible fracture

in a brittle manner in the test. It is thus reasonably safe to infer that, if a material

breaks in a ductile manner at the high impact energies involved in the Izod test it

will, in nearly all circumstances involving thick sections, behave in a ductile manner

in practice. Because of the severe nature of the test, it would be wrong, however, to

assume that because a material gives a brittle-type fracture in this test it will necessarily

fail in a brittle manner at low impact energies in service.

In cases in which high residual orientation is present, additional weakening can result

from the easier propagation of cracks in the direction parallel to the orientation.

A low Izod value does give a warning that care should be exercised in design to avoid

sharp corners or similar points of stress concentration. It does not by itself, however,

necessarily indicate high notch sensitivity.

1.6.2 Falling Weight Impact Test

The normal day-to-day abuse experienced by an article is more closely simulated by

the falling weight impact type test (see BS 2782) [13]. It is also much easier to vary the

type and thickness of specimens in this test. Furthermore, any directional weakness

existing in the plane of the specimens is easily detected.

ATS FAAR supplies the Fall-o-Scope Universal apparatus for free-falling dart impact

resistance tests (Table 1.6). The apparatus can be set to obtain the rate of energy

absorption during the impact. The apparatus can be used to conduct tests to different

specifications with computerised operation in the range 70-200 ºC.

One would normally expect the impact energy required for failure to increase with

specimen thickness, but the rate of increase will not be the same for all materials. In

addition, the effect of the fabricating process cannot be over-emphasised. I mentioned

42


above that residual orientation can lead to marked weakening and susceptibility to

cracking parallel to the orientation direction. Such residual orientation is extremely

likely to be brought about in the injection moulding process, where a hot plastic melt

is forced at a high shear rate into a narrow, relatively cold cavity. Naturally, the more

viscous the melt and the longer and narrower the flow path, the greater the degree

of residual orientation expected. Within a given family of grades the melt viscosity,

besides being influenced by temperature and shear rates, will also be affected by

polymer parameters such as MW and MW distribution. Thus, it cannot be assumed

that the magnitude of this drop in impact strength will be the same for differing

materials or for differing grades of the same material. For a satisfactory comparison

it is, therefore, essential to know that variation in impact strength with thickness of

unoriented compression moulded samples and injection moulded samples moulded

over a range of conditions.

In some investigations, a flat, circular, centre-grated, 15 cm diameter moulded dish

as (Figure 1.24) has been used. The dish is mounted in the falling weight impact

tester in its inverted position and located so that the striker hits the base 1.875 cm

from the centre because the area near the sprue is one of the recognised weak spots

of injection moulding. By means of inserts in the mould, the thickness of the base

of the moulding can be varied. For each material a family of curves is obtained for

variation of impact strength with thickness of compression moulded sheets moulded

under standard conditions, and of injection moulded dishes produced at a series of

temperatures. An estimate of the weakening effects of residual orientation and of

the tolerance that can be allowed on injection moulding conditions for satisfactory

impact performance is thus obtained.

A further factor that can adversely affect impact performance is often encountered

in injection moulding. Whether a single gate or multiple gates are used, it is usually

inevitable due to a non-uniform flow pattern during the mould-filling operation that

separate flow fronts meet to form a weld line. The strength at the weld line will vary

with the material, the injection moulding conditions, and the length and thickness of

the flow path, and can be considerably weaker that adjacent parts of the moulding.

Weld effects have been studied by means of a flat 37.5 × 3.125 cm ruler mould as

shown in Figure 1.25 which can be gated at one or both ends. The thickness of the

specimens can be varied by means of inserts. In this way, a series of comparisons

can be made between the falling weight impact strength of welded and non-welded

specimens at the same distance from the gate for varying path length to cavity thickness

ratios and differing moulding conditions.

Besides the average level at which failures occur in the falling weight impact test, the

type of failure can also be of importance in judging the relative impact performance

of a material in practice. There are generally three types of failure, as listed below:

43


in which the material yields and flows at the point of

impact, producing a hemispherical depression, which at sufficiently high impact

energies eventually tears through the complete thickness.

in which the specimen shatters or cracks through its complete

thickness with no visible signs of any yielding having taken place prior to the

initiation of the fracture. (With specimens possessing a high degree of residual

orientation a single crack parallel to the orientation direction is obtained.)

in which some yielding or cold flow of the

specimen occurs at the point of impact prior to the initiation and propagation of

a brittle-type crack (or cracks) through the complete thickness of the specimen.

These types of failure are shown in Figure 1.26.

An idea of the impact energy associated with these types of failure can be ascertained

by consideration of a typical load-deformation curve. This in general will be of the

form shown in Figure 1.27. From the definitions given above, a brittle-type failure

will occur on the initial, essentially linear part of the curve, prior to the yield point.

Figure 1.26 Examples of types of failure occurring in the falling weight impact

strength test. Source: Author’s own files

44


Deformation

Brittle Bructile

Ductile

Yield Point

Loa

d

Figure 1.27 Typical load-deformation curve. Source: Author’s own files

A bructile failure will occur on that part of the curve near to but following the yield

point. A ductile failure will be associated with high elongations on the final part of

the curve which follows the yield point. With the last type of failure, the deformation

at failure will be reasonably constant for a strike of given diameter. The area beneath

the curve up to failure gives a measure of the impact energy to cause failure. Hence, in

general, for a given thickness, tough-type failure is associated with high and reasonably

constant impact energy. Bructile failure is also associated with relatively high but

more variable energy. Brittle failures are generally associated with a low impact level.

If, in a given test, failures are all of the tough type, a reliable and reproducible measure

of the impact strength of the sample can be obtained. If, however, differing types of

failures are observed, a much wider variation in impact values can be obtained from

repeat tests.

45


However, the difference in energy level required to produce tough and brittle-type

failure varies for differing materials. In practice, the lowest level at which a failure

is likely to occur is as important as the estimated level at which 50% of the samples

are likely to fail, so it is necessary to have a measure of this variability. The 50%

failure level (F50

) is the value usually reported in a falling weight impact test. This

can be obtained by repeating the falling weight test using the ‘probit’ rather than the

‘staircase’ procedure. In the former, sets of samples are tested at each of a series of

increasing energy levels between the limits at which, none or all the specimens in a

given set fail. If percentage failure is then plotted against impact energy on probability

paper, the slope of the curve gives an estimate of the variable and the likelihood of

occasional samples failing at low impact levels.

A final important factor that requires consideration for many applications is the

influence of temperature because some thermoplastics undergo a tough-brittle

transition, with a correspondingly marked drop in impact performance, just below

ambient temperature. Changes in notch sensitivity with temperature have been

ascertained by the use of notched bar tests (as described above) and change in

impact strength with temperature by the falling weight test and standard Izod test.

Comparatively thick compression- and/or injection moulded specimens have been

used in these investigations to ensure that the starting value of the impact strength at

room temperature is high and that change in this value is easily detected.

Lavach [45] discussed the factors that affect results obtained by the Gardner impact

test. This test is used by plastics producers to approximate the mean failure energy

for many plastics. The test is inexpensive and easy to operate. The test equipment can

be placed close to the manufacturing equipment, permitting fast and nearly online

determination of the impact resistance of an article. The test is useful for finding the

mean failure energy for brittle thermoplastics such as acrylic and high-impact PS,

with standard deviations between 8% and 10%.

1.6.3 Notch Sensitivity

The sharpness of a notch is well known to have a strong influence on the impact

performance of notched specimens. The extent of this effect is found to vary

considerably from polymer to polymer. To investigate this point, impact values

obtained with the BS notch of 0.10 cm tip radius have been compared with those

obtained with other tip radii, namely 0.025 cm (the ASTM standard), 0.050 cm and

0.20 cm. The results for notch sensitivity are conveniently expressed by plotting the

impact value as a ratio of the BS value against the notch tip radius or alternatively

46


plotting the actual Izod value on a logarithmic scale against notch tip radius on a

linear scale. The steeper the slope of the curve, the greater the notch sensitivity of

the material and the more essential it is to remove sharp radii and points of stress

concentration in design if optimum performance is to be obtained.

If the limitations described above are remembered, the results of notched bar tests

give useful information. However, for realistic comparisons of materials, additional

impact data are required.

Reference was made above as to how a low Izod value gives a warning regarding

the care that should be exercised in design to remove points of stress concentration

if optimum impact behaviour is to be obtained with a component in practice. It was

also mentioned that a more reliable measure of notch sensitivity can be obtained by

carrying out tests using a series of notch radii rather than a single standard notch.

Figure 1.28 shows the relative notch sensitivities of some polyolefins and a few other

well-known thermoplastics. Polyolefins are of intermediate notch sensitivity, this being

more marked with the higher melt-index grades. Among the polyolefins, the low melt-

index PP copolymer designated GMT 61 has the lowest notch sensitivity, whereas the

higher-melt-index PE and PP homopolymers (e.g., 65-045, KM 61) have the highest.

Figure 1.29 shows the variation in standard Izod impact strength (0.10 cm notch tip

radius) with temperature for the same range of materials. By far the most temperature-

sensitive is the low melt-index PP copolymer. It is interesting to compare the sharp

fall-off in notched impact strength with this material between 23 ºC and 0 ºC with

the steady values obtained for falling weight impact strength for the same temperature

range (Figure 1.30). The removal of points of stress concentration is thus the most

vital factor governing the impact performance of this grade at temperature of 0 ºC

and below. Figure 1.31a shows in more detail the notched impact behaviour of PP

at room temperature. A semi-logarithmic plot has been used so that the overall level

and notch sensitivity can be determined. The marked influence of the melt index and

the superiority of the copolymers over the homopolymers are clearly demonstrated.

Figure 1.31b and c show similar data for 0 ºC and -20 ºC, respectively. With the

exception of the low melt index PP copolymer, notch sensitivities are similar to those

at room temperature, but the marked drop in overall level brought about by a decrease

in temperature is again clearly demonstrated.

47


0.25 0.50 1.0

20

40

60

80

120

140

160

Not

ch s

ensi

tivi

ty

180

200

Notch tip radius (mm)

2.0

Figure 1.28 Typical notch sensitivities of some thermoplastics at 23 ºC. Source:

Author’s own files

48


30 –20 –10 00

267

534

BSI Izod

impa

ct s

tren

gth (

1 m

m n

otch

) (J/m

not

ch)

801

1068

Temperature (°C)

10 20

Toughened polystyrene

PP GM 61

PP (MT 61)

ABS (high impact)

E (MI = 1.0)

DE (low MI) 60004

PP KMT 61

HDPE65045

BS (normal impact)

30

Figure 1.29 Influence of temperature on Izod impact strength of various polymers.

HDPE low melt index and high melt index. Source: Author’s own files

49


–30 –20 –10 00

107

214

320

427

534

641

GMT 61

BS (high impact)

ABS (normal impact)


KM 61

65-045

MT 61

748

Falling

weight

impa

ct s

tren

gth (

1/m F

50)

Temperature (°C)

10 20 30

Figure 1.30 Effect of temperature on falling weight impact strength of 0.060 inch

compression moulded specimens of various polymers. Source: Author’s own files

50


50.3 0.5 1.0 2.0

11

27

53

107

267

534

1068

MT 61

GM 61

BS notchTM notch

(a)

MT 61

KM 61

PM 61

Izod im

pact

str

ength

(ft

lb/inch

not

ch)


50.3 0.5 1.0 2.0

11

27

53

107

267

534

MT 61

GM 61

BS notchTM notch

(b)

MT 61

Izod im

pact

str

ength

(ft

lb/inch

not

ch)


51


50.3 0.5 1.0 2.0

11

27

53

107

267

534

MT 61

GM 61

BS notchATM notch

(c)

MT 61

Izod im

pact

str

ength

(J/m

not

ch)


Figure 1.31 Variation of Izod impact strengths of various grades of PP with notch

tip radius at (a) 23 ºC; (b) 0 ºC; and (c) -20 ºC. PM 61 = PP homopolymer. Source:


Figures 1.32 and 1.33 show the combined data for room temperature and -20 ºC for

HDPE and for ABS terpolymer and toughened PS, respectively. Of particular note

is the very small effect of temperature on the impact level and notch sensitivity of

HDPE, although the effect of MW (or melt index) is again clearly demonstrated. This

indicates the suitability of HDPE for low-temperature applications.

One important practical example of the controlling influence of notch sensitivity in

governing impact behaviour is that of embossed surfaces. Table 1.7 gives the results

of falling weight impact tests in which a series of different textured finishes have been

investigated for a range of PP. The injection moulded disc specimens possessed one

texture and one plain, smooth surface. If the embossed surfaces were struck, high

impact values were obtained. However, if the smooth surface was struck (so that the

textured surface was placed in tension), very low impact values were obtained, with

the magnitude of the drop depending upon the depth and sharpness of the embossing.

52


The sharper the pattern, the lower the value obtained. In practice, it will generally be

the textured surface that is struck but, if the reverse applies, considerable care should

be exercised in the choice of the finish.

50.3 0.5 1.0 2.0

11

27

53

23 ºC

23 ºC

23 ºC

23 ºC

–20 ºC

–20 ºC

–20 ºC

–20 ºC

MI = 0.4

MI = 1.0

MI = 2.0

MI = 4.5

107

267

534

1068

2670

BS notchASTM notch

Izod im

pact

str

ength

(J/m

not

ch)


Figure 1.32 Variation of Izod impact strength of 0.96 HDPE with notch tip radius

at 23 ºC and -20 ºC. Source: Author’s own files

53

Mechanical P

roperties of Polym

ers

Table 1.7 Effect of embossing on falling weight impact strength at 23 oC

Polypropylene

grade

Surface

towards

striker

Falling weight impact strength at 23 oC – kJ/m2

Surface 1 Surface 2 Surface 3 Surface 4 Surface 5 Specimen

thickness

(mm)Smooth

grain (mould

decoration

incorporated)

Rough

finish

(Martin’s)

Rough

finish

(Dornbush)

Smooth

grain (mould

decoration

incorporated)

Medium

finish

(Dornbush)

PM 61 Smooth 0.4 0.4 0.4 0.5 0.4 2.7

Embossed 2.2 4.5 1.9 3.8 2.0 2.7

GM 61 Smooth 0.5 0.4 0.4 0.5 0.4 2.8

Embossed 11.7 12.5 1.9 21.7 10.7 2.8

KMT 61 Smooth 1.5 0.8 0.9 4.6 1.9 2.7

Embossed 24.4 20.3 13.6 >24.4 >24.4 2.7


54


50.3 0.5 1.0 2.0

11

27

5323 ºC

23 ºC

23 ºC

–20 ºC

–20 ºC

–20 ºC

ABS (high impact)

ABS (normal impact)


107

267

534

1068

BS notchASTM notch

Izod

impa

ct s

tren

gth

(J/m

not

ch)


Figure 1.33 Variation of Izod impact strength of ABS terpolymer and toughened PS

with notch tip radius and temperature. Source: Author’s own files

1.6.4 Falling Weight Impact Tests: Further Discussion

Figure 1.34 gives the variation in falling weight impact strength with thickness at 23 ºC

of a range of materials in substantially stress-free compression moulded samples.

55


However, in practice, fabricated articles are seldom stress-free (particularly injection

moulded articles). Residual orientation can result in a marked reduction in impact

strength. This is illustrated in Figures 1.35-1.41 covering high- and low melt-index

PP (Figures 1.35 and 1.38), PP copolymer (Figure 1.37) HDPE (Figure 1.38), normal-

impact and high-impact ABS terpolymers (Figures 1.39 and 1.40) and toughened PS

(i.e., styrene-butadiene copolymer, Figure 1.41).

2.0

Polypropylene copolymer (low MI)Acry onitrile butadiene styrene

terpolymer (high impact)High density polyethylene

Polypropylene copolymer (high MI)

Unplasticised PVC sheet

Acrylonitrile butadiene styrene(normal impact)Polypropylene homopolymer (low MI)


Polypropylene hornopolymer (high MI)

2.5 3.0 3.51.51.00.50

41

27

14

0

Falli

ng w

eigh

t im

pact

str

engt

h (J

(F 50

))

Thickness (mm)

Figure 1.34 Effect of thickness on falling weight impact strength of some typical

thermoplastics at 23 ºC (compression moulded samples).


56


2.0 2.5 3.0 3.51.51.0

IM = Injection mouldedCM = Compression moulded

IM 280 ºC

CM

IM 200 ºC

0.50

41

27

14

0

Falli

ng w

eigh

t im

pact

str

engt

h (J

(F 50

))

Thickness (mm)

Figure 1.35 Falling weight impact strength of high melt index PP copolymer at 23

ºC. Source: Author’s own files

57


2.0 2.5 3.0 3.51.51.0

IM = Injection mouldedCM = Compression moulded

IM 280 ºC

IM 240 ºCCM

IM 200 ºC

0.50

41

27

14

0

Falli

ng w

eigh

t im

pact

str

engt

h (J

(F 50

))

Thickness (mm)

Figure 1.36 Falling weight impact strength of low melt index PP copolymer at

23 ºC. Source: Author’s own files

58


2.0 2.5 3.0

MI = 3.0 (nucleated)

All samples wheninjection moulded

MI = 0.6

CM = Compression Moulded

MI = 3.0

CM

CMCM

3.51.51.00.50

14

0

3

5

8

11

Falli

ng w

eigh

t im

pact

str

engt

h (J

(F 50

))

Thickness (mm)

Figure 1.37 Falling weight impact strength of PP copolymers (comparison of

moulded and injection moulded specimens at 23 ºC. Source: Author’s own files

2.0 2.5 3.0


IM = Injection moulded CM

IM 280 ºC

IM 200 ºC

IM 240 ºC

3.51.51.00.50

27

0

14

Falli

ng w

eigh

t im

pact

str

engt

h (J

(F 50

))

Thickness (mm)

Figure 1.38 Falling weight impact strength of HDPE at 23 ºC.


59


2.0 2.5 3.0


IM = Injection moulded

CM

IM 260 ºC

IM 220 ºC

IM 240 ºC

3.51.51.00.50

27

41

0

14

Falli

ng w

eigh

t im

pact

str

engt

h (J

(F 50

))

Thickness (mm)

Figure 1.39 Falling weight impact strength of ABS terpolymer (normal impact

grade) at 23 ºC. Source: Author’s own files

2.0 2.5 3.0



CM

IM 260 ºC

IM 220 ºC

IM 240 ºC

3.51.51.00.50

27

41

0

14

Falli

ng w

eigh

t im

pact

str

engt

h (J

(F 50

))

Thickness (mm)

Figure 1.40 Falling weight impact strength of ABS terpolymer (high impact grade)

at 23 ºC. Source: Author’s own files

60


2.0 2.5 3.0



CM

IM 260 ºC

IM 220 ºCIM 240 ºC

3.51.51.00.50

3

5

1

2

Falli

ng w

eigh

t im

pact

str

engt

h (J

(F 50

))

Thickness (thousandth of inch)

4

Figure 1.41 Falling weight impact strength of typical toughened PS at 23 ºC.


The change in impact strength with injection moulding temperature gives an indication

of the process adaptability of each material. With high-impact ABS terpolymers,

there is little difference in the level of falling weight impact strength over the range

of injection moulding temperatures. With all materials except PP homopolymers and

toughened PS, the impact strength of injection moulded samples produced at higher

temperatures is similar to that of the virtually non-oriented compression moulded

samples. With toughened PS, the impact strength of injection moulded samples is

markedly lower than that of compression moulded samples, although a slight effect

of injection moulding temperature has been observed.

The marked drop in impact strength that can result from weld lines is demonstrated

in Figure 1.42 for a PP copolymer. At lower thickness, where the flow path to cavity

61


thickness ratio is correspondingly higher, the impact strength measured at the weld

is very much lower than that measured with a similar non-welded specimen. The

positioning of gates to minimise weld effects is thus of vital importance in obtaining

optimum behaviour in an injection moulded article of a given design.

Figures 1.30 and 1.43 show the effect of temperature on the falling weight impact

strength for compression moulded and injection moulded specimens, respectively, of

a range of materials. The reduction in impact strength with decrease in temperature is

particularly marked with the injection moulded samples of the two otherwise higher

impact-resistant types: PP copolymers and ABS terpolymers. The superior impact

strength at low temperatures of injection moulded HDPE is clearly illustrated.

2.0 2.5 3.0 3.51.51.00.50

27Non-weld specimens

Compression moulded

Weld specimens

41

0

14

Falli

ng w

eigh

t im

pact

str

engt

h (J

(F 50

))

Thickness (mm)

Melt temperature: ∆ 200 °C240 °C280 °C

Figure 1.42 Effect of weld lines on the falling weight impact strength of PP

copolymer at 23 ºC. Source: Author’s own files

62


10 20


GM 61 KM 61

KMT 61

GMT 61

65-045

Typical ABS (high impact)

Typical ABS (normal impact)

300–10–20–30

5

19

0

3

Falli

ng w

eigh

t im

pact

str

engt

h (J

(F 50

))

Temperature (°C)

8

11

14

16

Figure 1.43 Effect of temperature on falling weight impact strength of 2.0 mm

thick injection moulded specimens of various polymers moulded at 240 ºC. 65-

045-HDPE. Source: Author’s own files

As mentioned above with regard to end-use performance, the minimum level at which

failures are likely to occur is of equal importance to the usually quoted F50

level. This

is particularly true if different types of failure are obtained in the falling weight impact

63


test. Such variations in the type of failure frequently occur as temperature is reduced

below ambient temperature. The probit method is then the preferred procedure of

evaluation of impact strength behaviour.

The additional information that can be obtained by this method is demonstrated in

Figures 1.44-1.46. For a low melt index HPDE (Figure 1.44) the probit is almost

horizontal at ambient temperature and at -20 ºC. For a typical high-impact ABS

terpolymer (Figure 1.45), the slopes of the curves are relatively steep compared with

HDPE unless the impact strength is very low. The probability of failure occurring at

low temperatures and at a low impact levels is thus higher with PP copolymers and

ABS than with HDPE. In fact, the results indicate that the impact strength of the

latter material at -20 ºC is slightly superior to that at 23 ºC.

22

19

16

14

11

8

5

3

00.01

0.050.1

0.20.5 2 10 30 50 70 90 98 9.8 99.99

99.99995

Polyethylene0.947 density0.3 melt ndex

–20 °C

–20 °C tough 23 °C

23 °C tough

806040

Failiure (%)

Imp

act

ener

gy (J

)

2051

Figure 1.44 Probit curves for a low melt-index HDPE (nominal thickness: 2.0 mm,

injection moulded at 270 ºC). Source: Author’s own files

64


22

19

16

14

11

8

5

3

00.01

0.050.1

0.20.5 2 10 30 50 70 90 98 9.8 99.99

99.99995

ToughBructile + 10 °C

ToughBructileBrittle + 10 °C

–20 °C brittle

–10 °C brittle0 °C

10 °C

23 °C23 °C tough

806040

Failiure (%)

Imp

act

ener

gy (J

)

2051

Figure 1.45 Probit curves for a typical high-impact ABS terpolymer (nominal

thickness: 2.00 mm, injection moulded at 240 ºC). Source: Author’s own files

22

19

16

14

11

8

5

3

00.01

0.050.1

0.20.5 2 10 30 50 70 90 98 9.8 99.99

99.99995

ToughBructile 23 °C

Bructile + 10 °C

Bructile 0 °C

Bructile – 10 °C

Bructile – 20 °C

806040

Failiure (%)

Imp

act

ener

gy (J

)

2051

Figure 1.46 Probit curves for a PP copolymer (nominal thickness: 2.0 mm,

injection moulded at 240 °C). Source: Author’s own files

65


1.6.5 Effect of Molecular Parameters

The major influence of MW on the impact behaviour of polyolefins with regard to

notched Izod and falling weight impact tests is emphasised. With HDPE, MW is the

prime factor controlling impact behaviour, and density is a secondary factor. This is

clearly demonstrated in Figures 1.47 and 1.48, which show the results for commercial

HDPE. However, it is generally thought that for any given density there is a critical

melt index which increases as the density decreases. Above this value, a marked fall-off

in impact strength occurs. Furthermore, within a given family of grades, a decrease

in the melt index is usually accompanied by a slight reduction in density.

A comparison of the impact data with the creep data also confirms, particularly with

PP polymers, that an increase in impact strength is accompanied by a decrease in

creep resistance.

14

12

11

9

8

70.1 1

0.961

0.9510.958

0.960

0.9610.958

0.951

0.960

0.964

0.957

0.955

10 100

Melt index (g/10 min at 190 °C/2.16 kg)

Fal

lin

g w

eigh

t im

pac

t st

ren

gth

(J(

F50))

Figure 1.47 Effect of the melt index on falling weight impact strength of various

HDPE at 23 ºC (1.5 mm compression moulded samples).


66


14

12

11

9

8

70.950 0.955

0.28

0.28

2.33.2

1.1

4.4

5.5

5.9

Nominal 5.0 melt index

Nominal 0.3 melt index

4.4

14.0

0.960 0.965

Density (g/ml)

Fal

ling

wei

ght

impac

t st

rengt

h (

J(F

50))

Figure 1.48 Effect of density on falling weight impact strength of HDPE in various

melt indices at 23 ºC (1.5 mm compression moulded samples). Source: Author’s

own files

Janik and Krolikowski [24] investigated the effect of Charpy notched impact strength

and other mechanical properties (tensile strength, flexural strength, elastic modulus,

flexural modulus, melt viscosity, fracture) on the mechanical and rheological properties

of PE-PET blends.

Pick and Harkin-Jones [44, 45] and others investigated the relationship between

the impact performance of rotationally moulded PE products and their dynamic

mechanical properties.

Comparisons with regard to impact performance were made between metallocene-

catalysed linear low-density polyethylene (LLDPE) and Ziegler-Natta LLDPE. The

transitions evident in dynamic mechanical thermal analysis (DMTA) results are related

67


to changes in impact performance with temperature. The beta transition is found to

fall in the transition region between high impact performance at low temperatures

and lower impact performance at ambient temperatures.

1.7 Shear Strength

Apparatus for the measurement of this property according to ASTM D732 [71] is

available from ATS FAAR (Table 1.8). Briatico-Vangosa and co-workers [33] discussed

the relationship between surface roughness and shear strength in ultra high MW PE.

Table 1.8 Instrumentation, compressive set and mould shrinkage

Method ATS FAAR

Apparatus code

number

Units Test suitable for

meeting the following

standards

Compressive set to measure

permanent deformation

10.65000 (stress)

10.07000 (strain)

-

-

ASTM D395 [46]

ISO 815 [48]

UNI 6121 [49]

Mould shrinkage

injection

compression and post-

shrinkage

- % DIN 53464 [50]


1.8 Elongation in Tension

Apparatus for the measurement of this property according to ASTM D32 [73] is

available from ATS FAAR (Table 1.8).

1.9 Deformation Under Load

The use of light-scattering and X-ray beams to measure polymer deformation has

been reported by Lee and co-workers [51] and Riekel and co-workers [52].

68


1.10 Compressive Set (Permanent Deformation)

The equipment described in the following section is supplied by ATS FAAR (Table 1.8).

Tests can be carried out under constant load Method A according to ASTM D395-03

[53]. The test is used to determine the capability of rubber compounds and elastomers

in general to maintain their elastic properties after prolonged action of compression

stresses. According to Method A of ASTM D395 [53], this stress is applied by a

constant load.

The tester incorporates a spring with a calibration curve and is accurately calibrated

to have a slope of 70 ± 3.5 kN/m at 1.8 kN force. Spring deflection is indicated by

a vernier system with an accuracy of 0.05 mm applied directly to the moving plate.

The tester may be placed in an oven or in a cryogenic chamber to conduct tests in

the range of +70 ºC to 30 ºC as indicated by ASTM specifications.

Tests can be carried out under constant deflection Method B according to ASTM D395

[53], ISO 815 [48] and UNI 6121 [49]. This method applies a constant compression

stress to specimens having a diameter of 29 ± 0.5 mm or 13 ± 0.2 mm with a thickness

of 12.5 ± 0.5 mm or 6 ± 0.2 mm. After introduction of the specimens, the apparatus

is placed in a conditioning chamber for a preset time at a preset temperature. At the

end of the heating period, the final thickness of the specimen is measured.

1.11 Mould Shrinkage

Apparatus for the measurement of this property according to DIN 53464 [50] is

available from ATS FAAR (Table 1.8). Bertacchi and co-workers [54] reported on

computer-simulated mould shrinkage studies on talc-filled PP, glass-reinforced PA

and a PC/ABS blend.

1.12 Coefficient of Friction

This is an assessment of the coefficient of friction of materials in terms of dynamic

(sliding) friction against steel. Friction is influenced by temperature, surface

contamination and, most importantly, by the two material surfaces:

An excellent rating indicates a low coefficient of friction.

A very poor rating indicates a high coefficient of friction.

69


Miyata and Yamodka [55] used scanning probe microscopy (SPM) to determine the

micro-scale friction force of silicone-treated polymer film surfaces. Polyurethane (PU)

acrylates cured by an electron beam were used as polymer films. The micro-scale

friction force obtained by SPM was compared with macro-scale data, such as surface

free energy as determined by the Owens-Wendt method and the macro-scale friction

coefficient determined by the ASTM method. These comparisons showed a good linear

relationship between the surface free energy and friction force, which was insensitive

to the nature of polymer specimens or to silicone treatment methods. Good linearity

was also observed between the macro-scale and the micro-scale friction force. It was

concluded that SPM could be a powerful tool in this field of polymer science. Everar

and co-workers [56] reported coefficient of friction measurements for nitrile rubber.

Frictional properties of PA, polyacetals, PET/poly tetrafluoroethylene (PTFE) [33] as

well as polyesters and acrylics [353] have also been studied.

1.13 Fatigue Index

This is an assessment of the ability of materials to resist oscillating (or dynamic), load

or deflection-controlled deformation:

An excellent rating indicates excellent resistance to fatigue loading.

A very poor rating indicates poor resistance to fatigue loading.

ATS FAAR supply flexing machines (De Mattia) for the measurement of resistance

to dynamic fatigue to ASTM D430-95 [57] and ASTM D813-95 [58] (Table 1.9).

These test methods are utilised to test the resistance to dynamic fatigue of rubber-like

materials if subjected to repeated bending. The tests simulate the stresses in tension or

in compression of inflexion or in a combination of the three modes of load application

to which the materials will be subjected when in use.

The failure of tested specimens is indicated by cracking of the surface or, as prescribed

by ASTM D813 [58] specifications, by the dimensional increase of a nick made on

the specimen before starting the test. In the case of composite materials, the failure

can show up as separation of the different layers.

Lin and co-workers [59] also investigated the static tensile strength and fatigue

behaviour of long glass-fibre-reinforced semi-crystalline PA (Nylon 6,6) and

amorphous PC composites. The static tensile measurement at various temperatures and

tension-tension fatigue loading tests at various levels of stress amplitudes were studied.

The two-parameter Weibull distribution function and the pooled Weibull distribution

function was applied to obtain the statistical probability distribution of experimental

data of static tensile strength and fatigue life under different stress amplitude tests.

70


The increasing dynamic creep property and temperature under tension-tension

fatigue loading are compared between semi-crystalline and amorphous composites.

Results show that the static tensile strength of PA composites is higher than that of

PC composites, with lower fatigue life and more sensitivity to temperature. The slope

of S-N0 curves of long glass-fibre-reinforced semi-crystalline PA and amorphous PC

composites are almost identical.

Table 1.9 Instrumentation for measurement of dynamic fatigue

Method ATS FAAR Apparatus

code number


following standards

Resistance to dynamic

fatigue on bending

(De Mattia apparatus)

10.23000

10.23001

10.23003

ASTM D430 [57]

ASTM D813 [55]

Cold bend test (cables) 10.65053 ASTM D430 [57]

ASTM D813 [58]


1.14 Toughness

Toughness tests are traditionally carried out at 40 ºC (-40 ºF) or at 20 ºC (68 ºF).

This quantity is related to the notched Izod impact strength. It includes an assessment

of the overall fracture toughness of the polymer.

1.15 Abrasion Resistance or Wear

This is an assessment of the rate at which material is lost from the surface of the

specimen when abraded against a steel face. Wear rates are dependent upon numerous

factors, including contact pressure, relative velocity, temperature and initial surface

roughness. This rating considers the inherent ability of a material to resist wear:

An excellent rating indicates excellent resistance to wear.

A very poor rating indicates poor resistance to wear.

71


Table 1.10 Comparison of fatigue index, abrasion resistance and coefficient

of friction

Polymer Type Fatigue

index

Wear Coefficient of

friction

LDPE General

purpose (GP)

Very good Poor Very poor

Crosslinked polyethylene GP Good Good Very poor

HDPE GP Very good Good Good

Polypropylene GP Excellent Good Good

Polybutylene GP Very good Very good Very poor

Polymethyl pentene GP Very good Good Poor

Ethylene-propylene

copolymer

GP Excellent Good Very poor

Styrene-ethylene-butylene

copolymer

GP Good Good Very poor

High-impact polystyrene GP Poor Very poor Very poor

Polystyrene GP Good Very poor Very poor

Epoxy resins GP Poor Good Very poor

Acetal copolymer GP Very good - -

Polyesters GP Good Poor Very poor

Polybutylene terephthalate GP Good Good Poor

Polyethylene terephthalate GP Poor Good Very poor

Polyether ether ketone GP Very good Good Good

Ethylene tetrafluoro ethylene GP Very good Poor Very poor

Polycarbonate GP Poor Poor Poor

Polyphenylene oxide GP Very poor Very poor Good

Acrylonitrile-butadiene-

styrene terpolymer

30% glass

fibre-reinforced

Poor Poor Very poor

Phenol-formaldehyde GP Poor Good Very poor

Perfluoroalkoxy ethylene GP Very good Very poor Poor

Styrene-maleic anhydride

copolymer

GP Poor Very poor Very poor

Polymethylmethacrylate GP Poor Good Very poor

Ethylene vinylacetate

copolymer

25% vinyl

acetate

Very good Poor Very poor

Polyamide 11 GP Very good Good Good


Polyamide 6,6 GP Very good Good Good




Polyamide 6, 12 GP Very good Good Good

Polyamide-imide GP Poor Very good Poor

72


Polyimide GP Poor Very good Good

Polyetherimide GP Very poor Poor Poor

Polyurethane GP Excellent Very good Very poor

Polyetherester amide GP Very good Very good Very poor

Urea formaldehyde Foam Very poor Very poor Very poor

Styrene-acrylonitrile

copolymer

High impact Very poor Very poor Very poor

Acrylate-styrene-acrylonitrile

terpolymer

GP Very poor Very poor Very poor

Polytetrafluoroethylene GP Very good Very good Excellent

Polyvinyl fluoride GP Very good Poor Good

Polyvinylidene fluoride GP Excellent Good Very good

Perfluoroalkoxy ethylene 20% glass

fibre-reinforced

Good Poor Good

Ethylene chloro-

trifluoroethylene

Glass fibre-

reinforced

Very good Poor Very good

Fluorinated ethylene

propylene copolymer

GP Very good Very poor Poor

Chlorinated PVC GP Very poor Poor Very poor

Unplasticised PVC GP Poor Poor Very poor

Plasticised PVC GP Good Poor Very poor

Polyphenylene sulfide Glass fibre-

reinforced

Poor Good Poor

Polysulfone 10% glass

fibre-reinforced

Poor Good Good

Polyether sulfone GP Very poor Good Good

Silicones GP Poor Very good Very good


Apparatus for the measurement of this property according to the DIN specification

is available from ATS FAAR (ATS FAAR apparatus code number 10.55000).

In Table 1.10 is recorded the fatigue index, abrasion resistance, and coefficient of

friction for a range of polymers. This information is useful if combined with the major

mechanical requirement of the polymer in reaching a decision on compromises that

need to be made when choosing a polymer that is suitable for a particular application.

Polymers which, for example, have a combination of a very good category for all

three characteristics include HDPE, various PA, PE, fluoroethylene and polyvinylidine

fluoride. Recent work on the measurement of mechanical properties of these properties

is reviewed in Table 1.11.

73


Table 1.11 Review of work on the determination of mechanical and thermal

properties of polymers

Polymer Property discussed Reference

Polylactic acid Mechanical [65]

Polyethylene Mechanical [66]

Polypropylene Mechanical [67, 68]

Ethylene-1-octene Mechanical [69]

Polypropylene Mechanical and thermal [70, 72, 77, 79]

Ethylene-vinyl acetate Mechanical and thermal [71]

High-density polyethylene Mechanical and thermal [72]

Polyimide-modified epoxy Mechanical and thermal [73]

High-impact polystyrene Mechanical and thermal [74]

Polypropylene wood composite Mechanical and thermal [75]

Polycarbonate Mechanical and thermal [76-79]

Polycarbonate diols Mechanical and thermal [80]

Poly(3-hydroxybutyrate-co-

hydroxyvalerate)

Mechanical and thermal [81]

PVC-polymethylmethacrylate

blends


PVC-wood composites Mechanical and thermal [83]

Polyurethane Mechanical and thermal [84]

Epoxy resin Mechanical and thermal [95]

Polycarbonate-PVC blend Mechanical and thermal [86]

Polyethylene-ethylene vinyl

acetate mixtures


Polyethylene-lignin polyblends Mechanical and thermal [88]

Polystyrene Mechanical and thermal [89]

Polypropylene-low-density

polyethylene blends


Polyethylene Mechanical and thermal [91]

Bisphenol epoxy vinyl ester Mechanical and thermal [92]

Fibre-reinforced polypropylene Mechanical and thermal [93]

Fluorocarbon elastomer Mechanical and thermal [95]

Acrylic elastomer Mechanical and thermal [94]

Polymethylmethacrylate Mechanical and thermal [96]

Polyethylene-terephthalate-

polybutylene terephthalate

blends



74


1.16 Effect of Reinforcing Agents and Fillers on Mechanical Properties

Several typical reinforcing agents and filler have been used to improve or alter the

mechanical properties of polymers. These include glass fibre, glass beads, calcium

carbonate, minerals, mica, talc, clay, carbon fibres, carbon nanotubes, aluminium or

other metal powders, silica and silicones [80-94, 96].

1.16.1 Glass Fibres

1.16.1.1 Poly Tetrafluoroethylene

The incorporation of 25% of glass fibre increases the tensile strength from 25 MPa

to 180 MPa and the flexural modulus of PTFE from 0.07 GPa to 1.03 GPa while

reducing the elongation at break from 400% to 240%.

1.16.2 Polyethylene Terephthalate

Incorporation of 30% glass fibres increases the tensile strength of PET from 55 MPa

to 100 MPa and the flexural modulus from 2.3 GPa to 9.5 GPa while reducing the

elongation at break from 300% to 2.2%. Glass-filled PP has been the subject of several

studies involving tensile testing and the measurement of notched Izod impact strength.

Gupta and co-workers [97] observed a maximum improvement of the mechanical and

thermal properties in PP at 1% of a chemical coupling agent. However, deterioration

in impact strength and elongation at break occurred with increasing contents of glass

fibre in the polymer Type B in Table 1.12.

Other publications cover the effect of glass fibres on the mechanical and electronical

properties of epoxies [109, 110], the crystallinity of PA 6 [111] and the thermal

properties of PET [112].

75


Table 1.12 Effect of fillers on the mechanical properties of polymers

Polymer Filler or no-

filler (general

purpose)

Glass fibre

reinforcement

CaCO3

Mineral Carbon

fibre

Aluminium Silica Silicone

lubricant

Type A improvement in tensile strength

PEEK T = 92

F = 3.7

E = 50

I = 0.08

T = 151

F = 10.3

E = 2.2

I = 0.09

Polyamide-

imide

T = 18.5

F = 0.6

E = 12

I = 0.13

T = 195

F = 11.1

E = 5

I = 0.10

PET T = 55

F = 2.3

E = 300

I = 0.02

T = 100

F = 9.5

E = 2.2

I = 0.06

(30% glass)

PTFE T = 25

F = 0.07

E = 400

I = 0.11

T = 180

F = 1.03

E = 240

(25% glass)

Nylon 6 T = 40

F = 1

E = 4

I = 0.25

T = 145

F = 1.6

E = 3.15

I = 0.6

PC T = 65

F = 2.8

E = 110

T =

165

F = 13

E = 2.7

(30%

carbon

fibre)

Type B deterioration and impact strength

PI T = 72

F = 2.46

E = 8

I = 0.08

T = 79

F = 132

E = 1.2

I = 0.24

(40% glass)

PP T = 33

F = 1.5

E = 150

I = 0.07

T = 2.6

F = 2.0

E = 80

I = 0.05

(20%

CaCO3)

76


Polyalkyl-

phthalate

T = 70

F = 10.6

E = 0.9

I = 0.40

T = 57

F = 9.5

E = 0.9

I = 0.35

(30%

mineral)

Epoxy

resins

T = 600

F = 80

E = 1.3

T = 68

F = 1.1

T = 58

F = 1.1

E = 0.8

T = 48

F = 9.5

E = 1.0

T =

72

F =

15

E =

1

Acetal T = 73

F = 2.6

E = 65

I = 0.06

T = 85

F =

17.2

E = 1

I =

0.04

T = 55

F = 2.1

E = 50

I =

0.085

E: Elongation (%)

F: Flexural modulus (GPa),

I: Izod impact strength (kJ/m)

PC: Polycarbonate

PEEK: Polyether ether ketone

PET: Polyethylene terephthalate

PI: Polyimide

PP: Polypropylene

PTFE: Polytetrafluoroethylene

T: Tensile strength (MPa)


Some of these fillers or reinforcing agents (e.g., filler glass) improve the tensile strength

of the blend over that of the general glass-free polymer (Type A, Table 1.12). Glass-

fibre additions in general lead to an increase in the flexural modulus (or modulus of

elasticity) and a decrease in the elongation at break. In the case of other polymers

(Type B, Table 1.12), addition of fillers such as glass fibre and calcium carbonate

causes a distinct decrease in tensile strength (usually with an accompanying decrease

in flexural modulus, elongation and impact strength). Some particular examples of

these effects are discussed in more detail below.

The effects of fillers on mechanical properties are reviewed at Table 1.12.

1.16.2.1 Polyether Ether Ketone

As seen in Table 1.12, the incorporation of 50% glass fibre into polyether ether

ketone (PEEK) increases the tensile strength from 92 MPa to 151 MPa and increases

77


the flexural modulus from 3.7 to GPa 10.3 GPa while decreasing the elongation at

break from 50% to 2.2%.

1.16.2.2 Polyimide

In the case of polyimide, the incorporation of 40% glass fibre increases the flexural

modulus from 2.46 GPa to 132 GPa with hardly any effect on tensile strength. The

elongation at break falls from 8% to 1.2%.

1.16.2.3 Polyamide Imide

The incorporation of glass fibre increases the tensile strength of PA-imide from 18.5

MPa to 195 MPa (Table 1.2) and the flexural modulus from 0.6 GPa to 11.1 GPa

while the elongation at break falls from 12% to 5%.

The effect of glass beads on the mechanical properties of PP has been studied [114].

Various workers have studied the effect of glass fibres on the influence of various

factors such as melt temperature [351], weld line strength [352] and on annealing

[353] on glass-reinforced polymers.

1.16.3 Calcium Carbonate

Calcium carbonate has been used as a reinforcing agent for polyetherether ketone

[104], ABS terpolymer [85], PP [106], and PP-vinyl acetate (PPVA) copolymer [107].

Tang and co-workers [99] showed that the tensile modulus of the PPVA copolymer

increased with increase in filler weight fraction. The impact strength decreased rapidly

when the weight fraction of calcium carbonate fell below 10% and then decreased

gradually with increasing weight fraction of calcium carbonate.

1.16.4 Modified Clays

Various workers have investigated the mechanical and thermal properties of polymers

in which had been incorporated organically modified clays and Montmorillonites

(Table 1.13). In general, it was found that increasing the clay contents of a polymer

increased the storage and loss modulus as well as Young’s modulus and reduced

crystallinity. The glass transition temperature (Tg) increased and thermal stability

tended to improve.

78


Table 1.13 Influence of incorporation of organically modified clays and

montmorillonite nanocomposites in polymers

Polymer Property investigated Reference

A Montmorillonite

Polyurethane Glass transition (Tg), particle

size, electrical

[334]

PVDF Thermal [335]

Polytrimethylene triphthalate Heat stability, mechanical [336]

Polyimide-amide Rheological [337]

Acrylonitrile-butadiene-styrene Thermal [338]

PE Thermal [339]

Polypropylene-ethylene-propylene diene Mechanical and viscoelastic [340]

Hydroxy benzoic acid-

hydroxynaphthoic acid copolymers

Thermal [341]

LDPE Rheological [342]

Polyamide Mechanical and rheological [343]

Polystyrene nanocomposites Thermal [344]

B Organically modified clays

PLA Rheological [345]

PS Rheological [346, 347]

PE Thermal [278]

Ethylene-propylene diene elastomers Mechanical and rheological [279]

PP Mechanical and thermal [280]

PP Viscoelastic

Ethylene-vinyl acetate Mechanical and thermal [281]

Maleic anhydride-based PP composites Rheological [282]

Epoxies Thermomechanical [283]

Nylon 12 Thermal and rheological [284]

High-impact PS-PET nanocomposite Mechanical and rheological [285]

Fluroroelastomers Thermal [286]

LDPE = Low-density polyethylene

PE = Polyethylene

PLA = Polylactic acid

PP = Polypropylene

PS = Polystyrene

PVDF = Polyvinylidene fluoride


79


Clay-polyvinylidene fluoride composites have a significantly improved storage modulus

to the base polymer over the temperature range -100 ºC to 150 ºC [102]. Gupta and

co-workers [103] and Yan and co-workers [104] reported on the mechanical and

thermal behaviour of glass-fibre filled or glass bead-filled polyvinylidene-bentonite-

clay nanocomposites, which were shown to have a significantly improved storage

modulus to the base polymers [95].

1.16.5 Polymer-silicon Nanocompsites

It has been found that the Tg of ABS, silica [145] and polymerthyl methacrylates silica

nanocomposites [146] increased with silica content. Also, the thermal properties

were enhanced. Thus, the degradation temperature at 10% weight loss was ~30 ºC

higher than that of pristine polymethyl methacrylate. Other properties that have been

studied include mechanical, viscoelastic, thermal and optical properties [136-144]

(Table 1.14).

Table 1.14 Properties of polymer-silica nanocomposites

Polymer Property studied Reference

Ethylene diamine and maleic

anhydride-grafted PP

Mechanical [147]

Epoxies Mechanical and thermal [148]

PS Viscoelastic [149]

Polyacrylonitrile Thermal [150]

Poly(3,4 ethylene-dioxythiophene) Optical [151]

Phenylene-vinylene oligomers Optical [152]

PP Crystallinity [153]

PET Crystallinity [154]


1.16.6 Carbon Fibres

Various mechanical properties have been determined on carbon fibres epoxy

composites [155], as well as LDPE [156] and polyvinylidene fluoride [158].

80


Carbon fibre has been used in PC blends. For instance, 30% carbon fibre increased the

tensile strength of PC from 68 MPa to 165 MPa and increased the flexural modulus

from 2.8 GPa to 13 GPa while decreasing the elongation at break from 110% to

2.7%. Carbon fibre has also been used as a reinforcing agent in epoxy resins [106].

1.16.7 Carbon Nanotubes

As will be discussed in Chapter 3, the incorporation of carbon nanotubes into

polymers can have beneficial effects on electrical properties such as conductivity. The

measurement of mechanical properties of these polymers is reviewed in Table 1.15.

Table 1.15 Effect of carbon nanotubes on mechanical properties

Polymer Property measured Reference

Epoxy composites Tensile and thermal [159]

Epoxy matrices Mechanical and thermal [160]

Polyurethanes Mechanical and rheological [161]

Polyethylene composites Mechanical and rheological [162]

Polyethylene glycol Morphological and rheological [163]

Polycaprolactone Rheological [164]

Nylon 10,10 Morphological and rheological [165]

PET Crystallinity [166]

PHB Morphological and thermal [167]

PC Rheological [168]

Polyetheramide and

epoxy resins

Mechanical and electrical properties [169]

Epoxy resins Mechanical and electrical [170]

PEEK Rheological and electrical [171]

PI Crystallinity [172]

Ethylene-vinyl acetate Morphological and rheological [173]

Polysilsesquioxane Morphological and thermal [174]


81


1.16.8 Miscellaneous Fillers/Reinforcing Agents

Natural materials that have been used to modify polymer properties include wood

sawdust [126, 127], coconut fibre [128] carbon fibres, [129], oat husks, cocao, shells

[130], sugarcane fibres, [131] and banana fibre [132]. Mohanty and co-workers [101]

observed a 70% increase in the flexural strength of polyorpylene to which had been

added 30% jute with a maleic anhydride-grafted PP coupling agent.

Other materials that have been incorporated into polymers to modify mechanical and

other properties include calcium sulfate in styrene-butadiene rubber [115]; barium

sulfate in PE [115] and PP; [116] talc in PP [117]; aluminium in epoxies [118];

kaolinite-muscovite in PVC-polybutyl acrylate [119]; aluminium in PE [120]; nickel

in epoxies [121]; copper in PE [122]; gold in PS [123]; caesium bromide in polyvinyl

alcohol [124]; nickel-cobalt-zinc ferrite in natural rubber [125]; and carbon black in

various polymers [133-134].

1.16.9 Test Methods for Fibre Reinforced Plastics

Sims [175] reviewed some UK contributions to the development of fibre-reinforced

plastics such as those used for glass reinforced plastic pressure vessels and pipes. He

also discussed the need for harmonising and validating the many variants of property

testing methods currently in use.

Sims [175] has pointed out that a redraft of ISO 527 tensile testing for plastics has

resulted in two complementary parts for composites: Part 4 based on ISO 3268 [176]

covers isotropic and orthotropic materials and Part 5 covers unidirectional materials

[175]. There is a need to harmonise these two methods into a new part 5. Inputs

were included from ISO, ASTM, CRAG and EN aerospace methods. The remaining

three parts are: Part 1 - general principles; Part 2 - plastics and moulding materials

(including short fibre-reinforced materials); and Part 3 - films. ISO DIS 527-5 has in

parallel been voted as a European Committee for Standardization (CEN) standard

(EN 527-5) and should, with part 4, replace EN61 [185] and possibly two Aerospace

EN standards [149-180].

Obtaining agreement between ISO, CEN General series, EN aerospace-carbon only,

ASTM and JIS in specimen sizes for unidirectional reinforced materials was a notable

first step towards harmonisation of the test methods, particularly because the 1.25

cm wide ASTM specimen [182] has been in extensive use for many years. The agreed

dimensions are 15 × 1 mm for 0º and 25 × 2 mm for 90º fibre direction specimens

[179]. After agreement on the specimen sizes (which is the same in all standards series

82


except EN Aerospace for glass-fibre systems [180]), several other aspects of the test

method were decided upon during drafting of the new standard.

A complete list of recommended mechanical test methods for fibre-reinforced plastics

has been reported [183-213].

1.17 Application of Dynamic Mechanical Analysis

1.17.1 Theory

Dynamic mechanical analysis (DMA) provides putative information on the viscoelastic

and rheological properties (modulus and damping) of materials. Viscoelasticity is the

characteristic behaviour of most materials in which a combination of elastic properties

(stress proportional to strain rate) are observed. DMA simultaneously measures the

elastic properties (modulus) and viscous properties (damping) of materials.

DMA measures changes in mechanical behaviour such as modulus and damping as a

function of temperature, time, frequency, stress, or combinations of these parameters.

The technique also measures the modulus (stiffness) and damping (energy dissipation)

properties of materials as they are deformed under periodic stress. Such measurements

provide quantitative and qualitative information about the performance of materials.

The technique can be used to evaluate elastomers, viscous thermoset liquids, composite

coatings and adhesives, and materials that exhibit time, frequency and temperature

effects or mechanical properties because of their viscoelastic behaviour.

Some of the viscoelastic and rheological properties of polymers that can be measured

by DMA are [223, 224, 354]:

Modulus and strength (elastic properties)

Viscosity (stress-strain rate)

Damping characteristics

Low- and high-temperature behaviour (stress-strain)

Viscoelastic behaviour

Compliance

Stress relaxation and stress relaxation modulus

83


Creep

Gelation

Projection of material behaviour

Polymer lifetime prediction

Basically, this technique involves measurement of the mechanical response of a polymer

as it is deformed under periodic stress. It is used to characterise the viscoelastic and

rheological properties of polymers.

DMA is measurement of the mechanical response of a material as it is deformed

under periodic stress. Material properties of primary interest include modulus (E´),

loss modulus (E´´), tan (E´´/E´), compliance, viscosity, stress relaxation and creep.

These properties characterise the viscoelastic performance of a material.

DMA provides material scientists and engineers with the information necessary to

predict the performance of a material over a wide range of conditions. Test variables

include temperature, time, stress, strain and deformation frequency. Because of

the rapid growth in the use of engineering plastics and the need to monitor their

performance and consistency, DMA has become the fastest-growing thermal analysis

technique.

Recent advances in materials sciences have been the cause and effect of advances in

the technology of materials characterisation.

The theory of DMA has been understood for many years. However, because of the

complexity of measurement mechanics and the mathematics required to translate

theory into application, DMA did not become a practical tool until the late 1970s

when DuPont developed a device for reproducibly subjecting a sample to appropriate

mechanical and environmental conditions. The addition of computer hardware and

software capabilities several years later made DMA a viable tool for the industrial

scientist because it greatly reduced analysis times and labour-intensity of the technique.

DMA provides information on the viscoelastic properties of materials. Viscoelasticity

is the characteristic behaviour of most materials in which a combination of elastic

properties (stress proportional to strain) and viscous properties (stress proportional to

strain rate) are observed. DMA simultaneously measures elastic properties (modulus)

and viscous properties (damping) of a material. Such data are particularly useful

because of the growing trend towards the use of polymeric materials as replacements

for metal and structural applications.

84


The DuPont 9900 system controls the test temperature and collects, stores and analyses

the resulting data. Data analysis software for the DuPont 983/9900 DMA system

provides hardcopy reports of all measured and calculated viscoelastic properties,

including time/temperature superpositioning of the data for the creation of master

curves. A separate calibration programme automates the instrument calibration

routine and ensures highly accurate, reproducible results. In addition to measuring

the specific viscoelastic properties of interest, the four modes provide a more complete

characterisation of materials (including their structural and end-use performance

properties). The system can also be used to simulate and optimise processing

conditions, such as those used with thermoset resins.

The DuPont 983 DMA system can evaluate a wide variety of materials, ranging

from the very soft (e.g., elastomers, supported vicious liquids) to the very hard (e.g.,

reinforced composites, ceramics, metals). The clamping system can accommodate a

range of sample geometries, including rectangles, rods, films and supported liquids.

The DuPont 983 instrument produces quantitative information on the viscoelastic

and rheological properties of a material by measuring the mechanical response of a

sample as it is deformed under periodic stress.

1.17.2 Instrumentation (Appendix 1)

Seiko Instruments supplies the Exstar 6000 Series, DMM 6100 [225]. The instrument

can determine characteristics such as Tgdamping intensity, heat resistance and creep

and stress relaxation of various materials. This allows the user to obtain complete

characterisation of a processed material. The instrument can also be used to evaluate

the compatibility, anisotropy, vibration absorbency, MW, degree of crystallinity and

degree of orientation of polymeric and elastomeric materials.

Recently introduced instruments from TA Instruments include the DMA Q800 analyser

[226, 227]. The instrument can be used for the testing of mechanical properties of a

broad range of viscoelastic materials at temperatures ranging from -150 ºC to 600 ºC.

The DMA Q800 system is claimed to provide unmatched performance in stress-strain

control and measurement. It uses a proprietary non-contact linear motor to provide

precise stress control and optical encoder technology for unmatched sensitivity and

resolution in strain deflection.

The Mettler-Toledo Incorporated SDTA 861 DMA analyser [235] can provide

measurements over wide frequency ranges (1 mHz to 1 kHz) and large dynamic

stiffness ranges. A schematic of the instrument is illustrated and examples represented

of test data for various rubbers obtained using this equipment. Other suppliers include

Rheometric Scientific and Thermocahn [229].

85


Arm-locking pins

LVDT adjustment screwLVDT

Flexure pivot

Sample arm

Sample

Clamp

Control and samplethermocouples

Mechanical slide

Vernier adjustment knob

Slide lock

Electromagneticdriver

Figure 1.49 DMA Electromechanical System, linear variable differential

transformer. Source: Author’s own files

Figure 1.49 shows the mechanical components of the DuPont 983 DMA system. The

clamping mechanism for holding samples in a vertical configuration consists of two

parallel arms, each with its own flexure point, an electromagnetic driver to apply

stress to the sample, a linear variable differential transformer for measuring sample

strain, and a thermocouple for monitoring sample temperature. A sample is clamped

between the arms and the system is enclosed in a radiant heater and Dewar flask to

provide precise temperature control.

Easily interchanged clamp faces are designed to accommodate various geometries:

smooth and serrated faces for rectangular shapes and notched faces to hold cylinders

or tubing. A horizontal clamping system with smooth clam faces is available to hold

soft samples and supported liquids.

The sample is clamped between the ends of two parallel arms, which are mounted

on low-force flexure pivots, allowing motion in only the horizontal plane. The

distance between the arms is adjustable by means of a precision mechanical slide to

accommodate a wide range of sample lengths. An electromagnetic motor attached

to one arm drives the arm/sample system to a strain selected by the operator. As the

86


original sample system is displaced, the sample undergoes flexural deformation. A

linear variable differential transformer mounted on the driven arm measures the

response of the sample (strain and frequency) to the applied stress, and provides

feedback control to the motor.

The sample is positioned in a temperature-controlled chamber which contains a

radiant heater and a coolant distribution system. The radiant heater provides precise

and accurate control of sample temperature. The coolant distribution system uses

cold nitrogen gas for smooth, controlled sub-ambient operation and for quench

cooling at the start or end of a run. The radiant heater and the coolant accessory

are controlled automatically by the DuPont 983 system to ensure reproducible

temperature programming. An adjustable thermocouple, mounted close to the sample,

provides precise feedback information to the temperature controllers, as well as the

readout of sample temperature.

1.17.3 Fixed Frequency Mode

This mode is used for accurate determination of the frequency-dependence of materials

and prediction of end-use product performance. In the fixed frequency mode, applied

stress (i.e., force per unit area that tends to deform the body, usually expressed in

Pa (N/m)) forces the sample to undergo sinusoidal oscillation at a frequency and

amplitude (strain), i.e., deformation from a specified reference state, measured as the

ratio of the deformation to the total value of the dimension in which the strain occurs.

Strain is non-dimensional, but is frequently expressed in reference values (such as

percentage strain) selected by the operator. Energy dissipation in the sample causes

the sample strain to be out of phase with the applied stress (Figure 1.50a). That is,

the sample is viscoelastic, so the maximum strain does not occur at the same instant

as maximum stress. This phase shift or phase lag, defined as the phase angle (δ), is

measured and used with known sample geometry and driver energy to calculate the

viscoelastic properties of the sample.

The DuPont 983 analyser can be programmed to measure the viscoelastic

characteristics of a sample at up to 57 frequencies during a single test. In such

multiplexing experiments, an isothermal step method is used to hold the sample

temperature constant while the frequencies are scanned. The sample is allowed to

reach mechanical and thermal equilibrium at each frequency before data are collected.

After all the frequencies have been scanned, the sample is automatically stepped to

the next temperature and the frequency scan repeated. Multiplexing provides a more

complete rheological assessment than is possible with a single frequency.

87


1.17.3.1 Resonant Frequency Mode

This mode is used for detection of subtle transitions, which are essential for

understanding the molecular behaviour of materials and structure-property

relationships. Allowing a sample to oscillate at its natural resonance provides higher

damping sensitivity than at a fixed frequency, making possible detection of subtle

transitions in the material. Because of its high sensitivity, the resonance mode is

particularly useful in analysing polymer blends and filled polymers, such as reinforced

plastics and composites.

In the resonance mode, the DuPont 983 system operates on the mechanical principle of

forced resonant vibratory motion at fixed amplitude (strain), which is selected by the

operator. The arms and sample are displaced by the electromagnetic driver, subjecting

the sample to a fixed deformation and setting the system into resonant oscillation.

In free vibration mode, a sample will oscillate at its resonant frequency with decreasing

amplitude of oscillation (dashed line in Figure 1.50b). The resonance mode of the

DuPont 983 analyser differs from the free vibration mode in that the electromagnetic

driver puts energy into the system to maintain fixed amplitude (Figure 1.50b).

Time

Phase angle (δ) Fixed frequency(a)

Stress (σ)

Strain (ε)

Stre

ss

Stra

in

88


Time

Resonant frequency

Free vibration983 DMA resonant mode

(b)Osc

illa

tion a

mplitu

de

Time

t0 t1 t2 t3 t4 t5

(c)

Stra

in

Strain

Strainrecovery

Stre

ss

Stress

Stress relaxationT

emper

ature

(°C)

Temperature

89


Time

t0 t1t2 t3 t4

t5

(d)

Stra

in

Strain

Strainrecovery

Stre

ssStress Creep

Tem

per

atu

re (

°C)Temperature

Figure 1.50 Modes of operation in the DuPont 983 dynamic mechanical analyser.


The make-up energy, oscillation frequency and sample geometry are used by the

DMA software to calculate the desired viscoelastic properties.

1.17.3.2 Stress Relaxation Mode

Stress relaxation is defined as a long-term property measured by deforming a sample

at a constant displacement (strain) and monitoring decay over a period of time. Stress

is a very significant variable that can dramatically influence the properties of plastics

and composites.

Induced stress is always a factor in structural applications, and can result from

processing conditions, thermal history, phase transitions, surface degradation and

variations in the expansion coefficient of components in a composite. The modulus

of a material is not only temperature-dependent but also time-dependent, so the

stress relaxation behaviour of polymers and composites is of great importance to

the structural engineer. Stress relaxation is also important to the polymer chemist

developing new engineering plastics because relaxation times and moduli are affected

by polymer structures and transition temperatures. Therefore, it is essential that

90


polymeric materials that will be subjected to loading stress be characterised for stress

relaxation and creep behaviour.

Figure 1.51 shows the stress relaxation curves obtained for PC. The large drop in the

modulus of the material >130 ºC is caused by the increased mobility of the polymer

molecules as the Tg is approached.

The stress relaxation mode is used for measurement of ultra low-frequency relaxation in

polymers and composite structures (a primary indicator of the long-term performance

of materials). This mode of operation provides definitive information for prediction

of the long-term performance of materials by measuring the stress decay of a sample

as a function of time and temperature at an operator-selected displacement (strain).

Figure 1.50c is a schematic representation of the process of measuring stress relaxation

in a viscoelastic material. Using an isothermal step method, the sample is allowed to

equilibrate at each temperature in an unstressed state. The sample is then stressed

to the selected strain. The amount of driver energy (stress) required to maintain that

displacement is recorded as a function of time for a period selected by the operator.

After the measurements are recorded, the driver stress is removed and the sample is

allowed to recover in an unstressed state. This sample recovery (strain) can be recorded

as a function of time for a period selected by the operator. If the measurements at one

temperature are complete, the temperature is increased and the experiment repeated.

1.17.3.3 Creep Mode

Creep is defined as a long-term property measured by deforming a sample at a constant

stress and monitoring the flow (strain) over a period of time. Viscoelastic materials

flow or deform if subjected to loading (stress). In a creep experiment, a constant stress

is applied and the resulting deformation measured as a function of temperature and

time. Just as stress relaxation is an important property to structural engineers and

polymer scientists, so is creep behaviour.

The elastic modulus is a quantitative measure of the stiffness or rigidity of a material.

For example, for homogeneous isotropic substances in tension, the strain (ı0) is related

to the applied stress (o′) by the equation E = o′/ı0, where E is defined as the elastic

modulus. A similar definition of shear modulus (g) applies if the strain is shear.

For a PEEK laminate below the Tg, the laminate exhibits <2% creep, whereas above

the Tg creep increases to >10%.

91


0

0 200

2 90110

130

150

170 190 210 230250

270 290

310 ºCMode: CreepSize: 35.90 × 12.43

× 1.00 mm

400 600 800

Time (min)

(–)

Per

cent

cree

p

1000 1200 1400

4

6

8

10

12

14

16

Figure 1.51 Stress relaxation modulus of PC. Source: Author’s own files

The thermal curve in Figure 1.52 shows the flexural storage modulus-and-loss

properties of a rigid pultruded oriented-fibre-reinforced vinyl ester composite. The

flexural modulus and Tg increase dramatically with post-curing, so the test can be used

to evaluate the degree of cure, as well as to identify the high-temperature mechanical

integrity of the composite.

The creep mode is used for the measurement of flow at constant stress to determine

the load-bearing stability of materials, which is a key to prediction of product

performance. The creep mode is used to measure sample creep (strain) as a function

of time and temperature at a selected stress. Using the isothermal step method,

the sample is allowed to equilibrate at each temperature in a relaxed state. After

equilibration, the sample is subjected to a constant stress (Figure 1.52). The resulting

sample deformation (strain) is recorded as a function of time for a period selected by

the operator. After the first set of measurements is made, the driver stress is removed

and the sample is allowed to recover in an unstressed state. Sample recovery (strain)

can be recorded as a function of time for any desired period. If the measurement at one

92


temperature is complete, the temperature is changed and the measurement repeated.

Creep studies are particularly valuable for determining the long-term flow properties

of a material and its ability to withstand loading and deformation influences.

Temperature (ºC)

1st temperature scan(undercured as received)

2nd temperature scan(post cured on 983 DMAduring 1st scan)

Amplitude (p–p) = 0.20 mm

(– –

–)

E flexural

storage

modulus (

GPa)

(– –

–)

E flexural

storage

modulus (

GPa)

50

40

30

20

10

0

50

40

30

20

10

050 90 110 130 150 170 190 210 230 25070

Figure 1.52 Flexural storage modulus-and-loss properties of rigid extruded

oriented fibre reinforce vinyl ester composites. Source: Author’s own files

Whereas elastic modulus at a single temperature (as measured by a static-mode

instrument) is indicative of the properties and quality of metals, this is not necessarily

true of polymeric materials in which molecular relaxations dramatically influence

the temperature and time (reciprocal frequency) dependences of material properties.

DMA is the preferred tool for evaluation of the effects of the following variables on

the mechanical properties of materials:

remain within the specifications at end-use temperatures.

93


frequency dependence of material properties.

mechanical properties of materials and their ability to withstand loading and

deformation influences.

The DuPont 983 DMA is a versatile laboratory instrument for characterising the

viscoelastic and rheological behaviour of materials. Various applications of DMA are

reviewed in Sections 1.17.2.5-1.17.2.17. Various applications of DMA are reviewed

below.

1.17.3.4 Projection of Material Behaviour using Superpositioning

The effects of time and temperature on polymers can be predicted using time-

temperature superpositioning principles. The latter are based upon the premise that

the process involved in molecular relaxation or rearrangements occur at greater rates

at higher temperatures. The time over which these processes occur can be reduced by

conducting the measurement at elevated temperatures and transposing the data to

lower temperatures. Thus, viscoelastic changes that occur relatively quickly at higher

temperatures can be made to appear as if they occurred at longer times simply by

shifting the data with respect to time.

Viscoelastic data can be collected by condcuting static measurements under isothermal

conditions (e.g., creep or stress relaxation) or by undertaking frequency multiplexing

experiments in which a material is analysed at a series of frequencies. By selecting

a reference curve and then shifting the other data with respect to time, a ‘master

curve’ can be generated. A master curve is of great value because it covers times or

frequencies outside the range readily accessible by experiment.

Figure 1.53a shows the DMA frequency multiplexing results for an epoxy/fibreglass

laminate. The plot shows the flexural storage modulus (E´) and the loss modulus

(E´´) as a function of temperature at various analysis frequencies. The loss modulus

peak temperatures show that the Tg moves to higher temperatures as the analysis

frequency increases. Through suitable calculations, the storage modulus data can

be used to generate a master curve (Figure 1.53b). The reference temperature of

this is 125 ºC and shows the effects of frequency on the modulus of the laminate

at this temperature. At very low frequencies (or long times), the material exhibits a

low modulus and would behave similarly to a rubber. At high frequencies (or short

times), the laminate behaves like an elastic solid and has a high modulus. This master

curve demonstrates that data collected over only two decades of frequency can be

transformed to cover more than nine decades.

94


(–––

–) F

lex

ura

l st

ora

ge m

od

ulu

s (G

Pa)

(– –

–)

Fle

xu

ral

sto

rage

mo

du

lus

(GP

a)

10

9

8

7

Mode: Fixed frequency(0.1, 0.32, 1.0, 3.2, 8.0 Hz)

Temperature (°C)

6

5

4

3

2

180 90 100 110 120 130 140 150

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

(a)

Log

[E (Pa)

]

10

9

8

7

6

5

4

3

2

1–6 –4 –2 0 2 4 6

(b)

Figure 1.53 Dynamic mechanical analysis (a) frequency multiplexing results for an

epoxy/fibreglass laminate; and (b) the master curve generated from fixed frequency

multiplexing data for epoxy fibreglass laminate. Source: Author’s own files

95


Example of the practical uses for superpositioning are:

integrity over time.

joint failure.

beam deflection under load over time.

by the fastener.

of the hose.

in specific frequency ranges.

vibration damping with engines, missiles and other heavy equipment.

of molecular transitions to high temperatures and can result in impact failure or

microcracking.

1.17.3.5 Prediction of Polymer Impact Resistance

Impact resistance is critical in the end uses of many commercial plastic products.

Measurement of impact properties, however, requires lengthy sample preparation

and is often irreproducible. For example, the ASTM drop weight (falling dart)

impact (DWI) method D5420 [40] and D5628 [41] requires an impact measurement

on as many as 30 samples. Each sample must be prepared by high-quality injection

moulding, followed by temperature conditioning at -29 ºC for 24 hours.

Low-temperature loss peak measured by DMA correlates with the ASTM DWI

values. DMA values are more precise than DWI measurements, so as few as four

determinations can be used to rank impact resistance. Sample preparation for DMA

is not lengthy and does not require sophisticated processing equipment because

surface effects are not critical and smaller samples are used. Figure 1.54 shows the

comparative DMA profiles for a series of impact-modified PP. The intensity of the

damping peak at -110 ºC correlates well with the DWI values at -29 ºC. Using these

96


types of data, a suitable calibration curve can be developed so that future formulations

can be rapidly screened by DMA.

1.17.3.6 Effect of Processing on Loss Modulus

PEEK is an important matrix material for thermoplastic composite applications. The

properties of PEEK laminate depend primarily on this morphology developed during

processing. Cooling rate, time in the melt, and sub-melt annealing are all critical

processing variables that can be simulated and the effects of changes immediately

evaluated by DMA. Figure 1.55 illustrates, for example, the effects of cooling rate

on loss modulus results. The peak maximum temperature, reflective of the Tg of the

matrix, increases significantly as the cooling rates is decreased. The magnitude of the

loss peak decreases and the peak width increases significantly as the cooling rate is

decreased. These changes in the loss modulus Tg peak can be explained in terms of

increased free volume as well as decreased level of crystallinity as the cooling rates

increase. A very fast cooling rate results in greater free volume in the amorphous

phase and imparts greater mobility to the polymeric backbone. A decrease in the

matrix crystalline content also results in greater mobility when passing through the Tg.

0.02

0.03

6

6

5

5

4

4

3

3

2

2

11

Impact resistanceDWI at –29 ºCCurve J

>61602717.27.22.4

0.04

–130 –90

Temperature (°C)

Tan

δ

–50 –10

Figure 1.54 Prediction of polymer impact resistance. Comparative dynamic

mechanical analysis profiles for a series of impact-modified PP.


97


0.5

1.0

1.5

2.0

2.5

3.0

A – “Quick cooled”B – 20 ºC/minC – 5 ºC/minD – 2 ºC/minE – Standard processing

120 130 140 150 160 170 180 190 200 210

Temperature (°C)

Fle

xura

l lo

ss m

odulu

s (G

Pa)

Mode: Flxed frequency (1.0 Hz)

A

B

C

D

E

Figure 1.55 Effect of polymer processing on loss modulus by dynamic mechanical

analysis. Effect of cooling rate of PEEK composite loss modulus at the Tg.


1.17.3.7 Material Selection for Elevated-temperature Applications

The task of evaluating new materials and projecting their performance for specific

applications is a challenging one for engineers and designers. Often, materials are

supplied with short-term test information such as deflection temperature under load

(DTUL) (ASTM D648) [356], which is used to project long-term, high-temperature

performance. However, because of factors such as polymer structure, filler loading and

type, oxidative stability, part geometry and moulded-in stresses, the actual maximum

long-term use temperatures may be as much as 150 ºC below or above the DTUL.

DMA continuously monitors material modulus with temperature and hence provides

a better indication of long-term elevated-temperature performance.

Figure 1.56 illustrates the DMA modulus curves for three resins with nearly identical

DTUL: a PET, a polyethersulfone and an epoxy. The PET begins to lose modulus

98


rapidly at 60 ºC as the material enters the glass transition. The amorphous component

of the polymer achieves an increased degree of freedom and, at the end of the Tg,

the modulus of the material has declined by ~50% from room-temperature values.

Because of its crystalline component, the material then exhibits a region of relative

stability. The modulus again drops rapidly as the crystalline structure approaches the

melting point. Because the Tg in a semi-crystalline thermoplastic is 150 ºC below its

melting point, the actual modulus of a resin of this type at the DTUL is only 10-30%

of the room-temperature value. The DTUL of highly filled systems based on these

resins is more closely related to the melting point than to the significant structural

changes associated with the Tg.

18

16

14

12

10

8

6

4

2

00 50

Epoxy

PESDTUL218 °C

DTUL224 °C

DTUL218 °C

PET

100 150 200 250 300

Temperature (°C)

[–––

] E

′ (G

Pa)

Figure 1.56 Comparison of DTUL and DMA results. Source: Author’s own files

1.17.3.8 Storage Modulus

One of the values obtained in a DMA is the storage modulus which approximately

quantifies the flexural or tensile strength of a material. Figure 1.57 shows the

99


modulus of some common materials analysed over three decades of frequency using

the frequency scan mode. In this mode, the temperature is held constant and the

frequency at which the sample is oscillated is scanned from low to high, or from

high to low, frequencies.

In polymer materials exhibiting viscoelastic behaviour, the modulus and viscosity are

dependent upon the frequency of the DMA measurement. This frequency dependence

is quite different for materials with different degrees of molecular branching,

crosslinking, or MW distribution. The use of DMA in the frequency scan mode

of operation is particularly important in material analysis because these types of

molecular differences are very difficult to distinguish using any other thermal analysis

techniques. For example, a frequency scan can show clear differences between PE

samples having a difference in average MW of only a few percent.

The effect of frequency on materials and the corresponding rate dependence of

materials on it can supply additional information about the viscoelastic characteristics

of a polymer.

When frequency is being scanned, small differences in MW (and the distribution of

MW) can be detected by shifts in the viscosity and modulus curves.

Storage modulus data have been reported for PP-wood composites [234], basalt

fibre-reinforced PP [237], ethylene-propylene-diene terpolymer [238], polyvinyl

fluoride-clay nanocomposites [239] thermosetting resins [240] and water-based

adhesives [235, 236].

1.17.3.9 Frequency Dependence of Modulation and Elasticity

As shown in Figure 1.58, the frequency dependence of modulus and viscosity can be

measured at different temperatures through the Tg to predict long-term behaviour of

polymers. This information can be used to calculate how the polymer will perform

over long durations at various temperatures and mechanical stress.

100


1011

Steel alloy

Uncured rubber

1010

109

108

107

106

105

104

103

102 101

Frequency (Hz)

Modulu

s (P

a)

101

1212

PTFE (25°C)

Chocolate (–5°C)

PDMS (20°C)

Figure 1.57 Examples of modulus range. Source: Author’s own files

Frequency (Hz)

Visco

sity

(Pas

)

Mo

du

lus

(Pa)

10110010–110–2

107

108

109

1010

1011

109

1010

1011

PC Board 130°C

125°C130°C

135°C

145°C

Figure 1.58 Viscoelastic behaviour through the Tg of epoxy PC board.


101


1.17.3.10 Elastomer Low Temperature Properties

Flexibility is an important end-use property for elastomers. The Tg (which is the

point at which an elastomer (on cooling) goes from a flexible more rubber-like form

to a more rigid inflexible form) is a critical parameter in determining the suitability

of the elastomer for specific applications. Plots of flexural storage modulus (in GPa)

versus specimen temperature by DMA are very useful in evaluating the stiffness and

flexibility of polymeric materials.

1.17.3.11 Tensile Modulus

The tensile modulus of poly-p-phenylene [230], relaxation modulus in LDPE

[231], diglycidyl ether bisphenol A epoxy resins [232] and styrene-butadiene block

copolymers with doped polyaniline [233] has been determined by DMA.

1.17.3.12 Stress-strain Relationships

Choosing the best conditions of dynamic stress and strain is often difficult if creating

mechanical analysis methods for a series of samples. Sample behaviour will often

change if the stress or strain imposed on a sample is increased or decreased, so a

curve of stress versus strain is very valuable. The Newtonian or linear region for a

material can be measured with the stress scan mode of the DMA. This is the region

in which quantitative data can be obtained.

The stress scan (dynamic stress (10,000 Pa) versus strain (%)) of an unvulcanised

rubber material depicted the linear region of this material to be within 0.8-0.27%

strain and 2,000-4,200 Pa stress. Without this capability, this linear region of stress

and strain can only be approximated.

Stress-strain curves have been reported for PE [241], PC [241], PA [241], polybutylene

phthalate [250] and styrene-butadiene block copolymers [242].

A stress scan (i.e., dynamic stress (Pa) versus strain (5%) plots) will show the effect

of increasing stress on a polymer. There is usually an initial region in which the strain

is proportional to stress. Then, with increasing strain there can be deviations from

linearity due to various molecular effects. Calculations can determine proportional

limits, yield modulus, draw strength and ultimate modulus.

Kim and White [243] used DMA to study the stress relaxation behaviour of 3501-6

epoxy resin during cure.

102


Specimens were tested at several cure states to develop master curves of stress

relaxation behaviour during cure. Using the experimental data, the relaxation modulus

was then modelled in a thermorheologically complex manner. A exponential series

was used to describe the relaxation times. Shift functions used to obtain reduced

times were empirically derived based on curve fits to the data. The data showed

that the cure state had a significant effect on the stress relaxation of epoxy resin.

Furthermore, the relaxation behaviour above gelation was shown to be quite sensitive

to the degree of cure.

Venneman and co-workers [244] reported the results of a study of the elastomeric

properties of thermoplastic elastomers determined using temperature scanning

stress relaxation analysis and recoverable strain after hysteresis. The thermoplastic

elastomers tested were styrene-ethylene/butylene-styrene-based thermoplastic

elastomers blended with PP, polyphenylene ether (PPE) and PPE/PA 12.

1.17.3.13 Viscosity

Sepe [240] showed that DMA allows the modulus and viscosity of a polymeric

thermoset resin to be calculated, making it possible to follow crosslinking process

by continuously monitoring these properties as the sample is heated.

Various workers [245-248] have discussed the application of DMA to the study

of aspects of viscosity, elasticity and morphology. Bouton and Rheo [249] studied

viscosity effects in blends of PC with styrene-acrylonitrile and ABS terpolymer.

1.17.3.14 Miscellaneous Applications of Dynamic Mechanical Analysis

DMA has also been used to determine the mechanical and thermal properties of

LDPE and ethylene-propylene diene terpolymer containing jute filler, which had

improved flexural and impact properties to those of the base polymer [280]. Jeong

and co-workers [251] investigated the dynamic mechanical properties of a series of

polyhexamethylene terephthalate, poly (1,4-cyclohexylenedimethylene terephthalate)

and random copolymers thereof in the amorphous state as a function of temperature

and frequency. The effect of copolymer composition on dynamic mechanical properties

was examined and the dynamic mechanical properties interpreted in terms of the

cooperativity of segmental motions.

Costa and co-workers [252] characterised polimides using gel permeation

chromatography, differential scanning calorimetry (DSC), DMA optical microscopy

and small- and wide-angle X-ray scattering. Stability of liquid crystalline phases

103


was discussed with respect to the number of ethylene oxide units and enthalpies of

transition between crystalline and liquid crystalline phases as well as between liquid

crystalline and isotropic liquid phases. The technique has also been applied to the

study of thermal properties of epoxy-bis maleimide composites [253], the dynamic

mechanical and thermal properties of maleic anhydride-treated jute HDPE composites

[254], the curing behaviour and viscoelastic properties of epoxy-anhydride networks

[255] and the mechanical alpha- and gamma-relaxation process of short-chain PE

[256].

1.18 Rheology and Viscoelasticity

There are several rheological studies specific to particular polymers. These include

dynamic rheological measurements and capillary theometry of rubbers [257],

capillary rheometry of PP [258], degradation of PP [259], torsion rheometry of PE

[260], viscosity effects in blends of PC with styrene-acrylonitrile and ABS [249], peel

adhesion of rubber-based adhesives [261] and the effect of composition of melamine-

formaldehyde resins on rheological properties [262].

As discussed in Section 1.17, DMA offers an enhanced means of evaluating the

performance of polymeric systems at elevated temperatures. It provides a complete

profile of modulus versus temperature as well as measurement of mechanical damping.

Operating in the creep mode and coupled with the careful use of time-temperature

superpositioning, projections can be made regarding the long-term time-dependent

behaviour under constant load. This provides a much more realistic evaluation of the

short- and long-term capabilities of a resin system than the values of DTUL.

Rheology is concerned with the flow and deformation of matter. Viscoelastic properties

are more concerned with the flow and elasticity of matter.

Numerous references to rheology and viscoelasticity occur. Thus, 250 references have

been identified in the five-year period between 2004 and 2008.

In addition to various moduli (storage, loss, loss shear, plateau) the following

rheological properties of polymers can be determined by a range of techniques:

104


g

Particular applications of the techniques in studies of the viscoelastic and rheological

properties of polymers are reviewed in Table 1.16.

Table 1.16 Measurement of rheological and viscoelastic properties of

polymers

Polymer Property measured Technique Reference

LLDPE Storage modulus (alpha

relaxation)

Effects of βirradiation

[263]

PS Storage shear modulus DMTA [264]

PBT and polyethylene

naphthalate

Stress relaxation

behaviour

Instron tester [274]

Polyvinyl alcohol

hydrogels and ferrogels

Storage modulus [265]

PEEK (fibre-reinforced) Specimens loaded in

flexure

DMA [266]

Poly(styrene-butadiene-

isoprene)

Order transition Dynamic

mechanical

spectroscopy

[267]

PMMA and LDPE Shear modulus and

clamping

Broadband

viscoelastic

spectroscopy

[268]

Polybutadienes Zero shear rate

dynamic viscosity and

Tg

GPC, FTIR,

rheometry

[269]

105


Poly(phenylene

sulfide)-Fe3O

4 magnetic

composite

Tg, steady state shear

deformation

DSC, DMA,

TGA

[270]

Ethylene-propylene

random copolymer

Plateau modulus and

entanglement

GPC, DSC,

oscillatory

rheometry

[271]

PP nanopolymer Structure and

viscoelastic properties

X-ray

diffraction,

DMA, TGA

[272]

LLDPE Creep strain stress

relaxation

DSC [273]

Polyethylene oxide-

polypropylene oxide

Steady-state shear and

oscillatory shear

- [274]

Branched polystyrenes Entanglements Oscillatory

shear

measurement

[275, 276]

Tetrafluoroethylene-

hexafluoropropylene

copolymer

Zero shear rate

viscosity

Linear

rheological

properties

[277]

Poly(N-

vinylpyrrolidone-co-

dimethethylamino-

propylmethacrylamide)

Radii of gyration GPC, multi-

angle light

scattering

[278]

Polyvinyl acetate Apparent viscosity,

viscous flow activation

energy

Viscometry [279]

Lightly crosslinked

acrylic networks

Bulk rheological

properties, elastic

modulus, resistance

to interfacial crack

propagation

Dynamic

mechanical

spectroscopy

[280]

Polyethylacrylates Zero rate shear,

viscosity dependence

Rheometric

measurements

[281]

Polyalkylether ketone

and polyaryl ether

ketone

Dynamic rheological

behaviour in oscillatory

shear mode

Viscosity

composition

studies

[282]

106


PVDF Shear rate Apparent

viscosity,

activation

energy

[283]

PP modified by

ultrasonic waves

Structure of modified

PP

High-intensity

ultrasonic wave

studies

[284]

Chlorinated

polyethylene/natural

rubber blends

Viscoelastic changes

and thermal

degradation

- [284]

DMA = Dynamic mechanical analysis

DMTA = Dynamic mechanical thermal analysis

DSC = Dynamic scanning calorimetry

FTIR = Fourier-transform infrared spectroscopy

GPC = Gel permeation chromatography

LLDPE = Linear low-density polyethylene

PMMA = Polymethylmethacrylate

TGA = Thermogravimetric analysis


1.19 Physical Testing of Rubbers and Elastomers

A wide variety of physical tests are available for rubbers and elastomers (Table 1.17).

1.19.1 Measurement of Rheological Properties

Negretti Automation offers a range of rheological testing instruments for rubber

elastomers. Probably the top-of-the-market instrument is the THS rheometer. This

is a unique and versatile instrument offering a complete characterisation of flow

behaviour of a polymer in the unvulcanised state. Variable shear rates between 0.1/s

and 100/s are available using a die and rotor principle with a preheated chamber. A

typical curve obtained with this instrument is illustrated in Figure 1.59, which shows

the curve obtained in the measurement of rheological behaviour.

107


Table 1.17 Physical testing of rubbers and elastomers

Properties measured Instrument Testing

to the

following

standards

Supplier Notes

Rheology

Measurement of

processability and

rheological behaviour

of raw (unvulcanised)

elastomers at variable

shear rates and

temperatures (viscosity and

elasticity)

THS Rheometer BS EN ISO

9001 [287]

Negretti

Automation

Oscillating disc

rheometry. Full analysis of

vulcanisation curing

Rheo-Check 100C

Barco

Rheomicrocomputer

package

BS

ASTM

ISO

Negretti

Automation

Oscillating die rheometer Oscillating die

rheometer

BS

ASTM

ISO

Gibitre, Italy,

available from

Negretti

Automation

Viscometry

Viscosity measurement of

rubbers and elastomers

Mark III Mooney

viscometer

BS

ASTM

ISO

NFT

DIN

JIN

Negretti

Automation

Mark 8 Mooney

viscometer

BS

ASTM

ISO

NFT

DIN

JIN

Negretti

Automation

Weight-density-volume

variation

Balance check

and Rapid Direct

Reading density

check

Gibitre, Italy,

available from

Negretti

Automation

Hardness IRHD Micro

durometer and

durometer support

Gibitre, Italy,

available from

Negretti

Automation

108


Brittleness point Low temperature

brittleness check

ASTM

D1329 [288]

Gibitre, Italy,

available from

Negretti

Automation

and ATS

FAAR

Code

Nos.

10.2900

10.29005

10.12010

Ross flexing test Measurement of

cut growth when

subject to repeat

bend flexing

ASTM

D1052 [289]

ATS FAAR Code no.

10.64002

De Mattia type flexing test De Mattia type

dynamic testing

machine

ASTM

DIN

Gibitre, Italy,

available from

Negretti

Automation

Mechanical properties

Resilience

Yersley mechanical

oscillograph

ASTM D945

[290]

ATS FAAR Code no.

16.65100

Universal tensile testing

Mechanical stability of

natural and synthetic

lattices

Abrasion test

Peel adhesion test

Ozone degradation test Ozone check Gibitre, Italy,

available from

Negretti

Automation

and ATS

FAAR

IRHD = International rubber hardness degrees


109


Time

C

B

A

Shea

r st

ress

(k

Pa)

Figure 1.59 Characterisation of polymer flow behaviour with a TMS rheometer

showing: (A) viscosity at very low shear rates, the stress measured is a function of

viscosity; (B) transient flow-step changes from low the high shear rates generate

a peak stress value which is a product of the thixotropic and structural features

of the sample; and (C) elasticity - high shear rates produce results that represent

elastic behaviour. Source: Author’s own files

The Rheocheck 100C is a classic oscillating disc rheometer offering a full analysis of

the vulcanisation curve with pre-set limits for quality-control purposes. It is available

from the standard cost-effective computer/printer/plotter package. The instrument

can be extended to include a Mooney viscometer.

The Barco Rheomicro computer package enables existing oscillating disc rheometers

to be converted to a powerful dynamic testing system.

The Gibitre oscillating die Rheocheck system available from Negretti Automation

is a moving die rheometer that enables the vulcanisation curve for any compound

to be produced rapidly at closely controlled temperatures. It is designed for rapid

computer-controlled quality checks in a production environment.

110


1.19.2 Viscosity and Elasticity

Depending upon temperature, shear rate and other flexural factors, the elements

of viscosity and elasticity impose a varying influence on processing behaviour.

A rheometer measures these in isolation in a single test, allowing prediction of

processing behaviour at an early stage. It also allows close control of raw polymers

of nominally identical specification from various sources, which often behave very

differently when processed.

Negretti Automation supplies the Negretti Mooney viscometer which meets the

requirements of all international standards. It is available with a digital control unit

and recorder to provide semi-automatic operation.

In this instrument, the viscosity of a material is derived by measuring the torque

required to rotate a disc squeezed between two samples of material at a defined

pressure. The diagram of the system (Figure 1.60) shows how the presence of the

sample around the shearing rotor causes a breaking reaction which can be measured

as a thrust against the calibrated U-beam. The net displacement of the U-beam (which

is proportional to the torque required to keep the rotor turning) is measured directly

on a gauge and can also be plotted on a chart recorder. The results are read or plotted

directly in Mooney units, which are universally used for viscosity measurement in

the rubber industry.

In accordance with the standards, the sample is preheated for a set time to a

temperature suitable for the test. For example, viscosity may be measured at 100 ºC,

whereas scorch tests normally use a higher temperature, typically 125 ºC.

In the mechanical layout of the viscometer (Figure 1.60), the rotor is driven by the

motor through the gear train and worm wheel. The axle carrying the ‘worm’ can

move laterally in the direction of the arrow. When the motor is started with the rotor

surrounded by a sample, rotation is resisted by the viscosity of the material. As a

result, the worm rotates against the worm wheel and tries to move in the direction

of the arrow. This movement is resisted by the U-beam and eventually the rotor is

forced to turn.

The deflection of the U-beam is related to the torque required to turn the rotor and

thus to the viscosity of the sample. The deflection can be read directly on the gauge,

which is calibrated in Mooney units. It can also control the linear deflection transducer

whose output drives a chart recorder.

Negretti Automation also supplies the Negretti Mark 8 Mooney viscomer which is a

computer-controlled instrument offering improved temperature control, repeatability,

hard copy of results, statistical memory and peripheral computer interface.

111


Shearingrotor

Rotation

Worm andworm wheel

Drivemotor

Transducer

CalibratedU-beam

Resultantthrust

Figure 1.60 Working details of a Negretti Mark 3 Mooney viscometer. Source:


1.19.3 Brittleness Point (Low-temperature Crystallisation)

Negretti Automation supplies the computer-controlled Gibitre low-temperature check

apparatus designed to test the low-temperature properties (-73 ºC, liquid carbon

dioxide), i.e., crystallisation effects and elastic recovery, of rubbers and elastomer

according to ASTM D1329 [288] by means of the temperature tests for rubber (TR)

test and brittleness point. ATS FAAR supplies an instrument for carrying out the

same measurements.

TR Test: After TR test positioning the test pieces in the relative mechanical system

using the special test piece extensions, they are then placed in the test chamber. The

chamber contains liquid cooled with carbon dioxide (-73 ºC) which is then gradually

112


heated to 20 ºC (2 ºC/min). The test assesses the temperature difference between the

two return values as a percentage.

After positioning the test pieces in a relative mechanical system, they are placed in the

test chamber. This contains liquid cooled with carbon dioxide (-73 ºC) to the required

temperature. A ‘ram’ comes into impact with the test pieces at a speed established

by the standards and this appears on the display. The test is repeated several times

using new test pieces to establish the brittleness temperature.

1.19.4 Flexing Test

The Ross flexing apparatus supplied by ATS FAAR measures, according to ASTM

D1052 [289], the cut growth of rubber or elastomer specimens when subjected to

repeated bend flexing. The machine allows the pierced flexed area of the specimen

to bend freely over a 10 mm-diameter rod at 90º with a frequency of 100 ± 5 or 50

± 5 cycles/min. The test is considered to have ended if the length of the cut (made

with a special nicking tool) has increased by 500%.

The test is widely used by all industries manufacturing rubber or elastomer goods

subjected (when in use) to repeated bending, such as shoes, tyres and conveyor belts.

Negretti Automation supplies the Gibitre De Mattia type dynamo tester which meets

the requirements of ASTM, BS and ISO standards. The equipment has been specially

designed by Gibitre to meet different requirements and can be used for other tests

besides the De Mattia type. Repeated flexing tests can be carried out on De Mattia-

type pieces or other types.

1.19.5 Deformation

The Yerseley mechanical Oscillograph supplied by ATS FAAR measures, according

to ASTM D945 [290], the mechanical properties of rubber vulcanisations in the

small range of deformation that characterises many technical applications. These

properties include resilience, dynamic modulus, static modulus, kinetic energy, creep

and set under a given force.

1.19.6 Tensile Properties

Negretti Automation supplies universal testing machines with capacities up to 5 kN

(500 kg) and 10 kN (1,000 kg) complete with computer systems and printers.

113


1.19.7 Mechanical Stability of Natural and Synthetic Lattices

ATS FAAR supplies equipment for carrying out this test to ASTM specifications.

1.19.8 Abrasion Test

Negretti Automation supplies equipment to carry out abrasion tests on rubbers to

the DIN specification.

1.19.9 Peel Adhesion Test

Negretti Automation supplies equipment for carrying out 180º peel tests for adhesive

tapes and similar products to BS specifications.

1.19.10 Ozone Resistance Test

The Ozone Check supplied by Negretti Automation determines the resistance to

ozone of elsatomers used in the tyre industry.

1.20 Physical Testing of Polymer Powders

Methods for the testing of polymer powders have been reviewed by Chambers [291]

and Klaren [292], and are listed in Table 1.18. With the exception of particle-size

distribution and thermal analysis measurements (by DSC and differential thermal

analysis), none of these tests requires commercially produced equipment (i.e., they

are simple laboratory tests). Full ASTM test specifications for polymer powders are

described in ASTM D3451 [293]. Brief comments on the various tests on powders

included in Table 1.18 are given below.

114


Table 1.18 Test methods for polymer powders

Measurement Standard test method(s)

Polymer powders

Particle-size distribution ASTM D1921 [295]

BS ISO 13319 [296]

Storage stability ASTM D609 [297]

DIN ISO 8130 [298]

Ultraviolet and outdoor stability

Reactivity

ASTM D822-01 [299]

ASTM G155 [300]

ASTM D3451 [293]

DIN ISO 8130 [298]

Melt viscosity ASTM D3451 [293]

Volatiles (stoving) ASTM D3451 [293]

DIN ISO 8130 [298]

True density DIN ISO 8130 [298]

ASTM D792 [301]

Bulk density BS 2782 [13]

BS 2701 [302]

ASTM D1895 [303]

Powder flow ASTM D1895 [303]

Test for cure

Electrical properties

Thermal analysis

By DSC and DTA

Tests on fused powder coatings

Defects

Glass

Whiteness

Film thickness

Hardness

ASTM D523-89 [304]

IN 53157 [305]

Adhesion

Flow deformation ASTM D2794-93 [306]

Bend test ASTM D522-93 [307]

Chemical resistance

Corrosion resistance ASTM B117-64 [308]

Glass transition

Melting

Cure


115


1.20.1 Ultraviolet and Outdoor Resistance

Although the physical properties of a powder coating may be studied to predict

its probable behaviour in practice, there are occasions when UV and/or outdoor

resistance are required. There are two main procedures: artificial weathering and

natural weathering (Appendix 5).

1.20.2 Artificial Weathering

The Atlas Xenon Arc Weather-O-Meter is used to calculate artificial weathering

according to ASTM G155-04 a [300] and ASTM D822-01 [299]. The coated panel

is exposed to a cycle of 102 minutes of light, followed by 18 minutes of light together

with a water spray.

1.20.3 Natural Weathering

Various sites have been used for the evaluation of coatings e.g., Miami (FL, USA)

or Phoenix (AZ, USA). The site at Phoenix greatly differs from that at Miami with

respect to atmospheric conditions. Because of the elevation (2000 feet above sea

level) and the clear dry atmosphere (38% relative humidity as the annual average),

the percentage of UV in the solar radiation is higher in Phoenix than in the humid

climate at sea level in Miami. (Arizona has 4,000 sun hours per year compared with

the UK, which has with 1,000 sun hours.) The yearly average climate data of Arizona

and Florida are as follows:

temperature: 21 ºC.

temperature: 24.5 ºC.

1.20.4 Reactivity

Thermosetting powders vary in their degree of reactivity, and some means of

quantifying this property is required. The basic method described in DIN ISO 8130

[298] and ASTM D3451 [293] involves melting the powder on a heating block at

160-200 ºC and stirring the molten material until gelation occurs.

There is undoubtedly some objectivity in this test, and slight differences in test method

could have a significant effect on the final result. Hence, elastomer repeatability is

116


reasonable, but there is considerable doubt as to how reproducible the method is

from one laboratory to another. Thus, the gelation test should be treated only as a

useful laboratory tool until sufficient evidence is available to justify inclusion in a

national standard.

1.20.5 Melt Viscosity

The fluidity of a material during the stoving process determines, to a large degree, the

smoothness of the final coating and the degree of edge coverage achieved. It may be

argued that the final appearance is the sole criterion for excellence, but some means

of quantitative measurement of melt viscosity is required for research work, and

is useful for quality control. A method is described in ASTM D3451 [294] which

employs a cone and plate viscometer. A simpler method is to use an inclined plane

whereby a pellet of the compressed powder is allowed to flow down on heated glass

plate set at a suitable angle. The total length of flow after a fixed time (usually 10

minutes) is related to the melt viscosity.

1.20.6 Loss on Stoving

Although stoving losses of coating powders are small when compared with

conventional finishes, some emission occurs, particularly with PU powders. Basically,

there are two methods of determination:

oven.

The first approach is favoured by the French (NFT 30-502) [310], whereas DIN ISO

8130-10 [298] and ASTM D3451-01 [293], describe dish methods. The dish method

has the advantage that it follows normal laboratory practice. The temperature of the

test and its duration can conveniently be the recommended cure schedule.

1.20.7 True Density

For powders of specific gravity above unity a pyknometer with water plus a wetting

agent may be used. DIN ISO 8130-7 [298], describes an air pyknometer which caters

for any powder (Appendix 5).

117


1.20.8 Bulk Density

Bulk density refers to the mass per unit volume of the powder (including air trapped

between the particles). Thus, this measurement enables the calculation to be made

of appropriate capacities for containers and hoppers. It is essential to ensure that the

sample is not compacted during the test, and this is normally achieved by pouring

the powder through a funnel into a measuring container. Methods are described in

BS 2782 [13] and ASTM D1895 [303]. A Rees-Hugill powder density flask is also

described in BS 2701 [302], (Appendix 5).

1.20.9 Powder Flow

The flow characteristics of a powder bear a complex relationship with the size,

distribution, shape and structure of particles. It is of particular significance in

connection with the ability of a powder to pass easily along feed lines to give the

necessary consistency of applications when using electrostatic spraying equipment.

The flow may be measured by determining the time for a prescribed volume of powder

to flow through a standard funnel, as described in ASTM D1895 [303] and DIN EN

ISO 813-10 [309].

This method has poor reproducibility in the case of thermosetting powders. This has

led to the development of an empirical technique by the Netherlands Organization

for Applied Scientific Research (TNO), which is described in the French standard

NFT 30-500 [311].

The method consists of placing the powder in a standard fluidised bed with a plugged

orifice in the side of the chamber. The height of the powder during and after fluidisation

is measured. The powder is then refluidised and the plug removed for 30 seconds,

allowing powder to flow into a tared container. The powder is weighed and the flow

factor calculated as the product of this mass and the height difference. This is then

compared with flow factors previously obtained on a wide range of powders, the

flow characteristics of which are known. This enables a prediction of flow behaviour

to be obtained.

1.20.10 Test for Cure

Unlike coatings made from thermoplastic powders, which require only visual

inspection to see if the coating is satisfactory, thermosetting powder coatings require

additional examination to determine if the coating has achieved its full physical

properties. Simple mechanical tests may be used (as is the practice with conventional

118


paint finishes) but, particularly for epoxy powder coatings, it is not normal to use

chemical resistance as a criterion for suitability. A common method is to swab the

coating continuously for 30 seconds with a cotton wool pad soaked in methyl isobutyl

ketone. Any deterioration of the film (with the exception of a minor loss) may be

taken as in indication that the coating is insufficiently cured.

1.20.11 Electrical Properties

Application of the electrostatic technique to powders has drawn attention to the

need for powders to possess appropriate electrical characteristics if they are to hold

an electrical charge and be successfully processed. This has led to development of

special apparatus which in addition to measuring resistivity and powder charge also

acts as a diagnostic kit for the electrostatic spraying equipment. This apparatus was

developed by the Wolfson Unit at Southampton University (Southampton, UK) and

is manufactured by Industrial Development (Bangor) at University College, North

Wales (Bangor, Wales).

1.20.12 Thermal Analysis

The performance of powder coatings during fusion and subsequent cure may be

followed in detail by thermal analytical procedures such as differential thermal analysis

and DSC. These techniques offer an opportunity for the study of phase transitions

and cure behaviour under closely controlled conditions. DSC is a requirement in the

British Gas standard for epoxy powder-coated materials for the external coatings of

steel pipes (where it is used as a guide to ascertain optimum cure conditions).

1.20.13 Particle-size Distribution

Procedures based on several principles are used for the measurement of the particle-

size distribution of polymer powders [312].

1.20.13.1 Methods Based on Electrical Sensing Zone (Coulter Principle)

Figure 1.61 shows, schematically, a simple form of apparatus. Particles, suspended

homogeneously at a low concentration in an electrolyte solution, are made to flow

through a small aperture (or orifice) in the wall of an electrical insulator, which is

commonly called an aperture (or orifice) tube – the aperture creates the sensing zone.

In addition, a current path is established between two immersed electrodes across

119


this aperture, setting a certain base impedance to the electrical detection circuitry. A

direct current is generally used. As each particle enters the aperture, it has effectively

displaced a volume of electrolyte solution equal to its own immersed volume. The

base impedance is therefore modulated by an amount proportional to the displaced

volume of the particle. This results in an electrical pulse of short duration being

created by each particle, the height of the pulse being essentially proportional to the

volume of the particle. The pulse may be measured, for example, as the change in

resistance, current, or voltage across the electrodes.

To vacuum

Stirred suspensionof particles inelectrolyte

Manometercontacts

Mainamplifier

ThresholdCircuit

Pulseamplifier

Counterdriver

Digitalregister

Oscilloscope

Horizontalsweep

Aperture

Mercurymanometer

– +

Counter stop/start

Electrodes

Tap

Figure 1.61 Simple form of particle size analyser (schematic). Source: Author’s

own files

The passage of several particles produces a ‘train’ of pulses that can be observed on

an oscilloscope and analysed by counter and pulse height analyser circuits to produce

a number against particle volume (or equivalent spherical diameter) distribution. A

volume or mass (‘weight’) against size distribution can also be measured, calculated,

120


or computed. The ‘weight’ percentages are possible if all of the particles have uniform

density or have a known density distribution across their size range.

Simple models have only one counter and size level circuit (and so are called ‘single-

channel models’); more complex instruments can obtain number and/or mass (weight)

distributions automatically in up to 256 size channels within a few seconds. Counting

and sizing rates of up to some 10,000 particles per second are possible, with each

pulse height being measured to within 1% or 2%.

Most particulate materials are irregularly shaped, so the volumetric response is

invaluable because volume is the only single measurement that can be made of an

irregular particle to characterise its size. In biological applications, the size response is

usually left calibrated in volume units (femtolitres, or ‘cubic microns’), but industrially

it is conventional to report the equivalent spherical diameter calculated from it. This

volumetric method makes no assumptions about particle shape and indeed is not

greatly affected by particle shape except in extreme cases such as flaky materials such

as certain clays. To count the number of particles in a known volume of suspension,

such as for particular contamination studies or for a blood cell count, the sample

volume is accurately metered by means of a calibrated ‘monometer’. Figure 1.61

illustrates the original simple mercury siphon and metering system.

The shape of the aperture tube can vary according to application. For example, a very

narrow design, which can be inserted into glass ampoules as small as 1 ml capacity,

allows the particle contamination of injectable solutions to be measured.

Instrument designs have a range of extra features, including embedded microprocessors

and various data reduction handling and presentation methods, and manufacturers

should be contacted for details. The volumetric sizing resolution, speed of data

collection, statistical accuracy of counting, freedom from any optical response effects

and the reliability and simplicity of calibration make these devised unsurpassed for

providing particle counting and size-distribution analysis.

Coulter-type instrumentation available from Coulter Electronics Limited is listed in

Table 1.19 (multisiser and Model 2M) and Appendix 1.

1.20.13.2 Laser Particle Size Analysers

Light diffraction is one of the most widely used techniques for measuring the size of

particles in the range 0.1-1,000 µm. This popularity is partly due to the way precise

measurements can be made quickly and easily. It also stems from the flexibility of

121


the technique, particularly the way it can be adapted to measure samples presented

in various forms.

The method depends on analysing the diffraction patterns produced when particles of

different sizes are exposed to a collimated beam of light. The patterns are characteristic

of the particle size, so mathematical analysis can produce an accurate, reproducible

picture of size distribution.

Coulter Electronics Limited supplies a range of laser particle size distribution

apparatus (LS100 and LS130) (Table 1.19).

Christianson Scientific Equipment supplies the Fritsch Analysette 22 Laser Particle

Sizer, which operates over the particle size range 0.16–1250 µm. This is a universally

applicable instrument for determining particle-size distributions of all kinds of solids

which can be analysed in suspension in a measuring cell or dry feeding through a

solid particle feeder. In the Fritsch Analysette 22 system, laser diffraction apparatus

measuring the particle-size distribution is displayed on the monitor in various forms,

either as a frequency distribution or as a summary curve in tabular form. These can

be subsequently recorded on a plotter, stored on hard disc, or transferred to a central

computer via an interface. The time required for one measurement is ~2 minutes.

Table 1.20 lists some suppliers and working ranges of various particle-size distribution

methods.

Terray [313] examined laser diffraction as a method for the inline measurement of

particle size. Examples of the application of Insitec laser granulometers developed

by Malvern Instruments are described. These include their use for particle-size

measurement in the cryogenic grinding of plastics.

1.20.13.3 Photon Correlation Spectroscopy (Autocorrelation Spectroscopy)

The Coulter N4 and its predecessor, the Nano-Sizer, use autocorrelation spectroscopy

of scattered light to determine the time-dependent fluctuations of the scattered light

which result from the Brownian motion of particles in suspension. The technique is

also known as photon correlation spectroscopy.

122


Table 1.19 Particle-size measurement instrumentation available from Coulter

Electronics Limited

Instrument Type Particle-size range Analysis

time (s)

Measurement

Coulter Multisiser Coulter 0.4-1200 µm

(overall)

0.4-336 µm for

distributions to be

expressed as volume

(weight) against size

30-90 Counting and

sizing

Model ZM (more basic

than the Multisiser)

Coulter 0.4-1200 µm - Counting and

sizing

Coulter LS series particle

size distribution analyser

Laser light

diffraction

0.1-810 µm 60 Particle sizing

LS130

LS100

-

Coulter series model N4 Photon

correlation

spectroscopy

(auto

correlation

spectroscopy)

0.003-3 µm -

Model N45 measures

coefficients of particles by

laser light scattering and

converts them to size or

molecular weight

- Sub-micro

particle

distribution

and molecular

weight

Model N4MD automated

laser based submicron

particle size analyser

0.003-3 µm -

Model N45D sub-micron

particle size analyser with

size distribution analysis

capability

-


1.20.13.4 Sedimentation

The use of sedimentation, whereby the particle size is determined by the application

of Stokes’ law, is well established. The time taken for fine particles to complete the

sedimentation process is far too long for practical utility. However, the development

123


of the wide-scanning sedimentometer (whereby a suspension of particles that are

separated under gravity is scanned by a light beam which is attenuated accordingly)

has largely overcome this difficulty. A result may be obtained in 30 minutes by a

competent operator using hand calculation methods.

Table 1.20 Suppliers and working ranges of particle-size distribution

methods

Method Particle size

range

Equipment supplier Model

Dry sieving 6-63 µm Fritsch Analysette 3 (20-25

mm)

Wet sieving 20-200 µm Fritsch Analysette 18

Microsieving 5-100 µm Fritsch

Sedimentation in

gravitational field

0.5-500 µm Fritsch Analysette 20

Laser diffraction 0.1-1100 µm Fritsch Analysette 22

Electrical zone

sensing

0.4-1200 µm Coulter Model ZM Coulter

Multisizer

Electron microscopy 0.5-100 µm - -

Photocorrelation

spectroscopy

0.5-5 µm - -

Sedimentation in

centrifugal field

0.5-10 µm Fritsch Analysette 21

(Anderson Pipette

centrifuge)

Diffraction

spectroscopy

1 µm-1 mm - -

Optical microscopy 0.5 µm-1

mm

- -

Projection

microscopy

0.05 µm-1

mm

- -

Image analysis

systems

0.8-150 µm

down to 0.5

µm

Joyce Leobl

Instruments,

Leitz, Karl Zeiss,

Cambridge

Instruments

Magiscan and

Magiscan P,

Autoscope P,

Videoplan II,

Quantimet 520


124


1.20.13.5 Acoustic Spectroscopy

Particle-size measurement results obtained from the deconvolution of acoustic

attenuation over a broad frequency range is showing increasing potential in the

characterisation of emulsions and slurries at process concentrations over particle

sizes ranging from 0.01 µm to 1,000 µm [314]. The results of this study reveal

that the particle-size distribution in high and low-density contrast materials can be

accurately measured by acoustic spectroscopy. The technique is distinct from other

size-characterisation technologies in three ways: there are no optics involved because

the instrument can measure through opaque high-concentration suspensions; it

has an extremely wide dynamic range; and it is well suited for online applications

because it is a relatively robust instrument and dilution is not required. Trottier and

co-workers [317] discussed the applications of acoustic spectroscopy to particle-size

measurement for LDPE and HDPE.

1.20.13.6 Capillary Hydrodynamic Fractionation

Venkatesan and Silebi [315] used capillary hydrodynamic fractionation to monitor

an emulsion polymerisation of styrene monomer as a model system. A sample taken

from the reactor at different time intervals is injected into the capillary hydrodynamic

fractionation system to follow the evolution of the particle-size distribution of the

polymer particles formed in the emulsion polymerisation. After the colloidal particles

have been fractionated by capillary hydrodynamic fractionation, they pass through

a photo-diode array detector which measures the turbidity at several wavelengths

instantaneously, thereby enabling utilisation of turbidimetric methods to determine

particle-size distribution. The particle size instrument is not hindered by the presence

of monomer-swollen particles. The shrinkage effect due to monomer swelling is found

to be accurately reflected in particle-size measurements.

1.20.13.7 Small-angle Light Scattering

Boerschig and co-workers [316] mentioned that light scattering should prove useful

for the direct measurement of particle growth during flow-driven coalescence of PS/

polymethyl methacrylate blends.

1.21 Plastic Pipe Materials

Testing of polymer pipe materials is done to international standards. Impact resistance

(falling dart) is done to ASTM D2944 [367] and UNI EN ISO 748 [337]. Hydrostatic

125


pressure tests have been conducted on polyolefin pipe exposed to chlorinated water

[318-319] and failure mechanisms discussed.

The effect of welding [320] and long-term exposure to soil [321] of PE pipes has

been investigated.

Physical performance testing has also been carried out on rigid PVC [334] and isotatic

polybutene [335] as well as polyacrylate [337] pipes.

Sims [326] reviewed the international standardisation of test methods according to

BS4994 and BS 6464 for general-purpose pipes and pressure vessels.

1.22 Plastic Film

Additional tests for polymer films that have been mentioned in the literature are the

tear strength at ethylene-vinylacelate plasticiser [344], the coefficient of friction of

polyolefin plasticiser blown film [342] and oxygen absorption of food packaging

film [343].

Michaelie and co-workers [332] studied the gas barrier properties of PET films

coated with microwave-enhanced plasma, polymerised barrier layer polyacetylene

or polyhdifluoroethylene.

References

1. ASTM D638-03,

Plastics, 2003.

2. DIN EN ISO 527-1, Plastics-Determination of Tensile Properties, Part 1: , 1996.

3. DIN ISO 527-2, Plastics-Determination of Tensile Properties, Part 2: Test , 1996.

4. ASTM D695-02a,

Plastics, 2002.

5. DIN EN ISO 179-1,

, 2001.

126


6. ASTM D790-03,

,

2003.

7. ASTM D732,

Tool, 2002.

8. DIN EN ISO 178, , 2003.

9. DIN EN ISO 8256, , 2005

10. DIN EN ISO 604, Plastics-Determination of Compressive Properties, 2003

11. Y. Son, Y.S. Chun and R.A. Weiss, , 2004,

44, 541.

12. M. Tasdemir and H. Yildirim, , 2003, 83,

2967.

13. BS 2782-0, , 1995.

14. ASTM D1043,

, 2002.

15. DIN 53477,

, 1992.

16. ISO 458,

, 1985.

17. ASTM D1869-99,

, 2004.

18. ASTM D3419,

, 2000.

19. ASTM D3641,

, 2002.

20. ASTM D4703,

, 2003.

21. ASTM D5227-01,

, 2001.

127


22. ISO 498, ,

1992.

23. DIN ISO 604, Plastics Determination of Compressive Properties, 2003.

24. J. Janick and W. Krolikowski, Polymery, 2002, 47, 250.

25. S. H. Lin, C.C.M. Ma, N.H. Tai and I.H. Perng,

, Anaheim, CA, USA,

1995, 40, 2, 1046.

26. H. Wong, P.G. Thompson, J.R. Schoonover, S.R. Ambuchon and R.A.

Palmer, Polymer Macromolecules, 2001, 34, 7084.

27. J.J. Lyons, Polymer Testing, 1998, 17, 237.

28. G. Hay, P.E. Mackey, P.N. Awati and Y. Park, , 1999, 43,

1099.

29. C. Nakafuku and K. Nishimura, , 2003,

87, 1962.

30. H-C. Wong, C.M. Sung and J. Hamilton, ,

1998, 7, 41.

31. T.C. Barany, E. Foldes, T. Czigany and J. Korgar-Kosis,

, 2004, 91, 3462.

32. J.F. Mana and J.C.A. Viana, Polymer Testing, 2001, 20, 937.

33. F. Briatico-Vangosa, M.D. Rink, F. D’Oria and A. Verzelli, Polymer , 2000, 40, 1553.

34. A. Marcellon, A.R. Bunsell, R. Piques and P. Colomban, Journal of Materials , 2003, 38, 2117.

35. S. Kao-Walter, J. Pahlstrom, T. Karlsson and A. Magnuson, Mechanics of Composite Material, 2004, 40, 29.

36. ASTM D256-56,

, 2005.

37. ISO 179-1 Plastics,

, 2001.

128


38. ISO 180, Plastics: , 2000.

39. UNI EN ISO 180, Plastics: , 2001.

40. ASTM D5420,

, 2004.

41. ASTM D5628, 96 e-I,

, 2001.

42. DIN EN ISO 6603, Plastics:, 2000.

43. M. Lavach, , 1999, 55, 41.

44. L.T. Pick and E. Harkin-Jones, , 2003, 43,

905.

45. Y. Kayano, H. Keskkula and D.R. Paul, Polymer, 1997, 38, 1885.

46. ASTM D395-03,

, 2003.

47. ISO 815,

Compression , 1993.

48. UNI ISO 815,

, 2001.

49. UNI 6121,

, 1967.

50. DIN 53464,

, 1962.

51. E.C. Lee, M.J. Solomon and S.J. Muller, Macromolecules, 1997, 30, 7313.

52. C. Riekel, M. Burghammer, M.C. Garcia, A. Gourier and S. Roth, PolymerPreprints, 2002, 43, 215.

53. ASTM D395-03,

, 2003.

54. G. Beracchi, A. Pipino and G. Boero, Macplas, 2001, 26, 78.

129


55. T. Miyata and T. Yamaoka, , 2002, 59, 415.

56. G. Evrard, A. Belgrine, F. Carpier, E. Valot and P. Dang,

, 1999, 777, 95.

57. ASTM D430-95,

, 2000.

58. ASTM D813-95,

, 2000.

59. S.H. Lin, C.C.M. Ma, N.H. Tai and L.H. Pering in the Proceedings of the

SAMPE Conference -

Anaheim, CA, USA, 1995, Volume 40, Book 2, p.1046.

60. A.M. Harris and F.C. Lee, , 2008, 107,

2246.

61. A. Menzur, , 2008, 108, 1574.

62. J. Jia, D. Raabe and W-M. Miao, , 2006,

24, 403.

63. T. Barany, E. Folden, T. Czigany and J. Karger-Kocsis,

, 2004, 91, 3462.

64. A. Flores, V.B.F. Mathot, G.H. Michler, R. Adhikari and F.J. Balta Calleja,

Polymer, 2006, 47, 5602.

65. S. Houshyai and R.A. Shanks, , 2007,

105, 390.

66. T. Kuila, H. Archarya, S.K. Shrivastava and A.K. Bhowmick, Journal of , 2008, 108, 1329.

67. B.G. Girija and R.R.N. Sailaja, , 2006,

101, 1109.

68. F-L. Jin and S-J. Park,

, 2006, 44, 3348.

69. F. Parres, R. Balart, J. Lopez and D. Garcia, ,

2008, 43, 3203.

130


70. R. Bouza, A. Lasagabasta, M.J. Abad and L. Bareal,

, 2008, 109, 1197.

71. F. Rouabah, M. Fuis, A. Bourdenne, C. Picard, D. Dadach and H. Hodduoni,

, 2008, 109, 1505.

72. S. Houshyar and R.A. Shonks, ,

2006, 291, 59.

73. F. Rouabah, M. Fois, L. Ibos, A. Bourdenne, D. Dadache, N. Haddaounis and

P Ausset, , 2007, 106, 2710.

74. M.G. Ahangani and A. Fereidoon, Polymers, 2008, 153, 1.

75. A Eceiza, M. Martin, K. De la Caba, G. Kortaberria, N. Gabilondo, M.

Curcnera and I. Mondragon, , 2008, 48,

297.

76. K.L. Pagnon, H.H. Chen, M.H. Innocentini-Mei and N.A. D’Souza, th , Milwauki WI,

USA, SPE, 2008, p.610.

77. R. Chakrabarti M. Das and D. Chakrabarti,

, 2004, 93, 272.

78. S. Pop and K. Chaochanchaikul, , 2004, 53, 1210.

79. C.V. Mythili, A.M. Retna and S. Gapalakrishanan,

, 2005, 98, 284.

80. M. Funabashi, S. Hirose and H. Hatcheyame,

, Sydney

Australia, Cambridge, Woodhead Publishing, Cambridge, UK, 2004, p.57.

81. A. Li, J. Chang, K. Wang, L. Lu, X. Yang and Y. Wang, Polymer, 2006, 55, 565.

82. A.J. Zattari, O. Bionchi, R.V.B. Oliveiri, L.B. Canto, C.A. Ferreiro and M.

Zeni, , 2006, 22, 69.

83. R.R.N. Sailaja, , 2005, 54, 1589.

84. F.R. Kogler, T. Kich, H. Peterlick, S. Seidler and U. Schubert, Journal of , 2007, 45, 2215.

131


85. A.P. Gupta, U.K. Saroop and M. Verma,

, 2004, 43, 937.

86. D. Sheivastava, , 2005, 96, 1691.

87. Ming-Zhang, C-Z. Wang, D-Z. Wu, Y-H. Yu and X-P. Yang, Polymer, 2005, 21, 196.

88. S. Houshyar, R.A. Shanks and A. Hodzik,

, 2005, 96, 2260.

89. E. Shrivakumar, K.N. Pandeu and S. Alam and C.K. Das, Polymer, 2005, 54, 1458.

90. Nohalen, L. Solar, I. Porcar, C.I. Vollo and G.M. Gomez,

Journal, 2006, 42, 3093.

91. F. Rouhbah, K. Ayadi and N. Hoddaoui,

Materials, 2006, 55, 975.

92. M. Szostah, , 2004, 416, 209.

93. I. Nigam, D. Nigam and G.N. Mathur,

, 2005, 44, 815.

94. Y.H. Nien and J. Chen, , 2006, 100, 3727.

95. L. Priya and J.P. Jog,

, 2003, 41, 31.

96. Q. Yuan, W. Jiang, L. An and L.K.Y. Li,

Technologies, 2004, 15, 409.

97. A.P. Gupta, U.K. Saroop, G.S. Jha and M. Verma, Polymer Plastics, , 2003, 42, 297.

98. B. Zhou, X. Ji, Y. Sheng, L. Wang and Z. Jiang, ,

2004, 40, 2357.

99. C.Y. Tang, L.C. Chan, J.Z. Liang, K.W.E. Cheng and T. Wong, Journal of , 2002, 21, 1337.

100. P.J.M. Von den Heuvrel, S. Goutianos, R.J. Young and T. Peijs, Composite, 2004, 64, 645.

132


101. S. Mohonty, S.K. Nayak, S.K. Verma and S.S. Tripathyl, Journal of , 2004, 23, 625.

102. L. Priya and J.P. Jog,

, 2003, 41, 31.

103. A.P. Gupta, U.K. Saroop, G.S. Tha and M. Verma, Polymer Plastics, , 2003, 42, 297.

104. Q. Yan, W. Jiang, L. An and R.K.Y. Li, ,

2004, 15, 409.

105. H. Xiaoming Xu, Y. Song, Q. Zheng and G. Ha,

, 2007, 103, 2027.

106. W. Leong, Z.A.M. Ishak and A. Ariffin, , 2004,

915, 3327.

107. O. Okzuk and H. Yildirim, , 2005, 96,

1126.

108. J. Zhaoliang, , 2007, 104, 1692.

109. F.H. Gojny, M.M.G. Wichman, B. Fiedler, W. Bauhofer and K. Schulte,

, 2005, 11, 1525.

110. Y. Nischitani, I. Sekiguchi, B. Hausnerova, N. Zolpazilova and T. Kitono,

, 2007, 15, 111.

111. E. Klata, S. Barysiak, K. Van De Velde, J. Garbarczyk and I. Krucinska,

, 2004, 12, 64.

112. A. Pegoretti and A. Pencalti, , 2004, 86,

233.

113. L. Huang, Q. Yuan, W. Jiang, L. An and S.Y. Jiang,

, 2004, 94, 1885.

114. S. Mishra and N.G. Shimpi, , 2007, 104, 2018.

115. F. Bianchi, A. Lazzeri, M. Pracella, A. Aguinu and G. Ligeri, Macromolecular, 2004, 218, 191.

116. K. Wang, J. Wu and H. Zeng, , 2006, 14,

473.

133


117. D. Yun-Sheng, G. Hong-Wei and W. Seng-Shan,

, 2005, 21, 90.

118. E.S.A. Rachid, K. Ariffin, H. Akit and C.C. Kooi

, 2008, 27, 1573.

119. Z-H. Chen and K-C. Gung, , 2007,

23, 203.

120. B. Mary, C. Dubois, P.J. Garreau and D. Brousgeay, , 2006,

45, 561.

121. J. Burghardt, N. Hansen, L. Hansen and G. Hansen in

Long Beach, CA, USA, 2006, Paper No.131.

122. A.S. Luyt, S.A. Molefi and H. Krump, ,

2006, 91, 1629.

123. J. Liu, B. Cankurtaran, L. Weizorek, M.J. Ford and M. Cortie,

, 2006, 16, 1457.

124. N. Abdelaziz, , 2004, 94, 2178.

125. D. Puryanti, S.H. Amed, H.M. Abdullah and A.N.H. Yushaft,

Journal of Polymeric Materials, 2007, 56, 327.

126. N. Sombatsompop, A. Kositchaiyong and W. Wimolala,

, 2006, 102, 1896.

127. B.L. Shah, S.E. Selke, M.B. Walters and P.A. Heiden, Polymer Composites,2008, 29, 655.

128. C.Y. Lai, S.M. Sapuan, M. Ahmid, N. Yahya and K.Z.H.M. Dahlan, Polymer, 2005, 44, 619.

129. E.M. Wouterson, F.C.Y. Boey, X. Hu and S-C. Wang, Polymer, 2007, 48,

3183.

130. E. Lezak, Z. Kulinski, R. Masirek, E. Piorkowgka and M. Pracella and K.

Gadzinowika, , 2008, 8, 1190.

131. S.M. Luz, A.R. Goncalves, A.P. Del-Arco and D.M.C. Ferrao, Composite, 2008, 15, 841.

134


132. S. Joseph and S. J. Lumas, , 2008, 109,

256.

133. H. Hu, J. Lin, Q. Zheng and X. Xu, ,

2006, 99, 3477.

134. S. Radhakrichnan, B.T.S. Ramanujam, A. Adhikara and S. Sivaram, Journalof , 2007, 163, 702.

135. K.I. Jung, S.W. Yoon, S.J. Sung and J.K. Park,

, 2004, 94, 678.

136. H. Mirzazadeh, A.A. Katbab and S. Bazgir in the 8th

,

Budapest, Hungary, 2005, Paper No.41.

137. O. Kusmon, Z.A. Mohd Ishek, W.S. Chow and T. Tekeichi,

Polymer Journal, 2008, 44, 1023.

138. E. Nazockdast, H. Nazockdost and F. Goharpey,

, 2007, 26, 789.

139. H. Ardhyananta, H. Ismail and T. Takeichi,

, 2002, 21, 822.

140. S. Hwan-Lee and J. Ryoun Youn, Polymer, 2007, 34, 1.

141. F. Kaya, M. Tanoglu and S. Okur, , 2008,

109, 834.

142. C.Y. Lew, W.R. Murphy and G.M. McNally in the nd , Chicago, IL, SPE, USA, 2004, p.304.

143. G. Sanchez-Olivares, A. Sanchez-Solis and O. Manerv,

of Polymeric Materials, 2008, 57, 417.

144. M.A. Kadei, M.Y. Lyn and C. Nah, , 2006,

66, 1431.

145. K. Zheng, L. Chen, Y. Li and P. Cui, , 2004,

44, 1077.

146. Y-H. Hu, C-Y. Chen and C-C. Wong, Macromolecules, 2004, 37, 2411.

135


147. S.U. Ryadiansyah, H. Ismail and B. Azhari, Polymer Composites, 2008, 29,

1169.

148. V. Paseual, V. Sanchez and J.M. Martin-Martinez, ,

2006, 233, 137.

149. C. Bartholom, E. Beyou, E. Bourgeat-Hamy, P. Cassagnau, P. Chanmonr, L.

David and N. Zydowicz, Polymer, 2005, 46, 9965.

150. O. Ruzimuradov, G. Rajan and J. Mark, , 2006,

245-256, 322.

151. J.C.S. Norton, M.G. Han, P. Jiang, G.H. Shim, Y. Ying, S. Greager and S.H.

Foulger, Chemistry of Materials, 2006, 18, 4570.

152. T.P. Ngnyen, S.H. Yang, J. Gomes and M.S. Wong, , 2005,

15, 269.

153. D.S. Chaudhary, M.C. Jalland and F. Cser,

Composites, 2004, 12, 383.

154. D.W. Chae and B.C. Kim, , 2007, 42, 1238.

155. J.L. Abot, A. Yasmin, A.J. Jacobson and I.M. Daniel,

Technology, 2004, 64, 263.

156. B. Hausner Nova, N. Zdrajilova, T. Kitano and P. Saha, Polymeri, 2006, 5,

33.

157. C-H. Chen and K-C. Lein, , 2006, 14,

155.

158. S. Vidhate, E. Ogunsona, J. Chung and N.A. D’Souza in theth Milwaukee, WI, USA, 2008, p.74.

159. M.R. Loos, S.H. Pezzin, S.C. Amico, C.P. Bergmann and L.A.F. Coelho,

, 2008, 18, 6064.

160. M.R. Loos, L.A.F. Coelho, S.H. Pezzin and S.C. Amico, Journal of Materials 2008, 11, 347.

161. H-C. Kuan, C-C.M. Ma, W-P. Chong, S-M. Yuen, H-H. Win and T-M. Lee,

, 2005, 65, 1703.

136


162. K.Q. Xiao, L.C. Zhang and I. Zarudi, ,

2007, 67, 177.

163. Y.S. Song, , 2006, 46, 1350.

164. D. Wu, L. Wu, Y. Sun and M. Zhang,

, 2007, 45, 3137.

165. B. Wang, G. Sun and J. Liu, , 2007, 47,

1610.

166. S. Tzavalas, D.E. Mouzakis, V. Drakonakis and V.E. Gregoriou, Journal of , 2008, 46, 668.

167. M. Lai, J. Li, J. Yang, J. Liu, X. Tong and H. Cheng, ,

2006, 53, 1479.

168. Y.T. Sung, M.S. Han, K.H. Song, J.W. Jung, H.S. Lee, C.K. Kum, J. Loo and

W.N. Kim, Polymer, 2006, 47, 4434.

169. T. Aoki, Y. Ono and T. Ogasawara in the st Technical Conference, Dearborn, MI, USA, 2006,

Paper No.31.

170. S-M. Yuen, C-C. M Ma, C-Y. Chuang, Y-H. Hsiao, C.L. Chiang and A-D.

Yu, , 2008, 39, 119.

171. D.S. Bangarusam Path, V. Alstradt, H. Ruckdeseschel, J.K.W. Sandler and

M.S.P. Shaffer, Journal of Plastics Technology, 2008, 6, 19.

172. V.E. Yudin, V.M. Svetlidnya, A.N. Shumkaov, D.G. Letenko, A.Y. Feldman

and G. Marom, , 2005, 26, 885.

173. S. Peeterbroveck, M. Alexandre, J.B. Nagy, N. Moreau, A. Destrec, F.

Monteverde, A. Pulmont, R.J. Jerome and P. Dubois, Macromolecular, 2005, 22, 115.

174. S-M. Ynen, C-C. M Ma, C-C. Teng, H-H. Wu, H-C. Kuan and C-L. Chiang,

, 2008, 46, 472.

175. G.D. Sims,

, Hamburg,

Germany, 1994.

137


176. ISO 3268, , 1978.

177. ISO/DIS 527 – Part 5, Plastics – Determination of Tensile Properties – Test , 1994.

178. EN 61:1977,

Properties, 1977.

179. prEN 2561, Tensile , 1989.

180. prEN 2747,

Direction, 1990.

181. M. Hojo, Y Sawada and H. Miyairi,

,

Amsterdam, The Netherlands, 1992.

182. ASTM D3039, Composites,1993.

183. G.D. Sims,

, Amsterdam, The

Netherlands, 1992.

184. G D Sims, Composites, 1991, 22, 4, 267.

185. BS 4994:1987,

Plastics, 1987.

186. BS 6464:1984,

for Process Plant, 1984.

187. ISO 10119:1992, , 1992.

188. ISO 10120:1991, , 1991.

189. ISO/DIS 10548,

, 2002.

190. ISO/DIS 10617, , 2010.

191. ISO 10352, Prepregs,, 1991.

138


192. prEN 2557,

, 1998.

193. EN 2339,

, 1993.

194. P.T. Curtis, Properties, 3rd Edition, RAE TR 88012, 1988.

195. G.D. Sims,

, DMM(A)8, 1990.

196. G.D. Sims,

DMM(M)6, 1990.

197. G.D. Sims,

DMM(M)7, 1990.

198. ISO 178, Plastics,1993.

199. ISO 8515, of Comparison , 1991.

200. ISO 4585,

, 1989.

201. ASTM D3518,

, 1991.

202. S. Kellas, J. Morton and K.E. Jackson, Mechanisms in ASTM STP 1156, 257-280, 1993.

203. prEN 6031, + TensileTest, 1995.

204. ISO 5725, (BSI 5497, 1987), for

, 1986.

205. ISO CD 13586,

, 2000.

139


206. R. Khoshravan and C. Bathias,

VAMAS

WG5, 1990.

207. W.R. Broughton and A. Vamas,

DMM(A)47, 1992.

208. G.D. Sims and A. Vamas

Material, DMM(A)180, 1989.

209. G.D. Sims,

.

210. ASTM D3044, 1976.

211. G.D. Sims, W. Nimmo, F.A. Johnson and D.H. Ferriss,

NPL Report

DMM(A)54, 1992.

212. S.W. Tsai and H.T. Hahn, Technomic

Publishing Company, Lancaster, PA, USA, 1980.

213. G.G. Sims,

ECCM-Composites: Testing and Standardisation 2, Hamburg,

Germany, 1994.

214. ASTM D5379/D3579M – 1993, Properties, 1993.

215. ISO 6603,

Plastics, ,

1989.

216. W.R. Broughton and G.D. Sims,

, NPL Report DMM(A)148, 1994.

217. G.D Sims and M.J. Lodeiro,

, NPL Report, 1994.

218. G.D. Sims, D.H. Ferriss and D. Payne,

, NPL Report, 1994.

140


219. G.D. Sims and W. Nimmo,

, NPL Report, 1994.

220. ISO/CD3 1172,

1994.

221. ISO/CD2 11667,

Dissolution, 1994.

222. BS 18, Materials,1987. [Withdrawn, replaced by BS EN 10002:1990]

223. C. Treny and B. Dupenray in the th

Conference, Toronto, Canada, 1997, p.2805.

224. P.C. Haschke, Machine Design, 2001, 73, 61.

225. Exstae 6000,

226. H.T. Jones, , 2002, 38, 91.

227. H. Wong, D.G. Thompson, J.R. Schoonover, S.R. Ambushon and R.A.

Palmer, Macromolecules, 2001, 34, 7084.

228. J. Foreman, J.E.K. Schawe and M.W. Wagner in nd

, Pittsburgh, PA, USA, 2002, Paper No.13.

229. F. Gouin, , 2001, 53, 17.

230. D. Dean, M. Husband and M. Trimmer,

, 1998, 36, 2971.

231. E. Kontou and G. Spathis, , 2003, 88,

1942.

232. L. Heux, J.J. Halary, F. Laupretre and I. Monnerie, Polymer, 1997, 38, 767.

233. M.E. Leyva, B.G. Soares and D. Khastgir, Polymer, 2002, 43, 7505.

234. J. Son, D.J. Gardner, S. O’Neill and C. Metaxas,

, 2003, 89, 1638.

141


235. D.W. Bamborough in the

Milan, Italy, 2001, Paper No.10.

236. D.W. Bamborough in the

, Basel, Switzerland, 1993, Section 111, p.2.

237. M. Botov, H. Betchev, D. Bikiaris and D. Panayiotou,

, 1999, 74, 523.

238. D. Singh, V.P. Malhotra and J.L. Vats, ,

1999, 71, 1959.

239. L. Priya and J.P. Jog, ,

2003, 41, 31.

240. M. Sepe, 1999, 7, 52.

241. O. Schroder and E. Schmachtenberg in the

Conference, Dallas, TX, USA, 2001, Paper No.405.

242. R.A. Palmer, P. Chen and M.H. Gilson in the

, Orlando, FL, USA,

Fall 1996, 75, p.43.

243. Y.K. Kim and S.R. White, , 1996, 36, 2852.

244. W. Venneman, S. Barbe, K. Kummerlow, K.G. Cai, A.C.A. Reid, S. Srinivan

and S. de Vogel in the

, Brussels, Belgium, 2004, p.141.

245. P. Maltese, 1995, 4, 225.

246. , 2002, 10-11, 68.

247. M. Exrin in the th

San Francisco, CA, USA, 2002, Session M40, Paper No.202.

248. J.L. Leblanc, , 1995, 745, 41.

249. J. Bouton and S.A. Rheo, , Magazine, 2000, 52, 34

250. G. Sarrkel and A. Choudhury, , 2008,

108, 3442.

251. Y.G. Jeong, S.C. Lee and W.H. Jo, , 2006, 14, 416.

142


252. G. Costa, G.C. Eastwood, J.P.A. Fairclough, J. Paprotny, A.J. Ryan and P.

Stagnaro, Macromolecules, 2008, 41, 1034.

253. J. Seo, W. Jang and H. Han, , 2007, 15, 10.

254. S. Mohanty, S.K. Verma and S.K. Nayak,

Technology, 2006, 66, 538.

255. P. Musto, M. Abbate, G. Ragosta and G. Scarinzi, Polymer, 2007, 48, 3703.

256. I.S. Kolesov, R. Androsch and H.J. Radusch, Macromolecules, 2005, 38, 445.

257. F. Declerq in the

, Puchov, Slovakia, 1996, p.173.

258. M. Fujiyama, Y. Kitajima and H. Inata, ,

2002, 84, 2128.

259. B. Vergnes and F. Berzin, , 2000, 158, 77.

260. H. Kwag, D. Rana, K. Cho, J. Rhee, T. Woo, B.H. Lee and S. Choe, Polymer, 2000, 40, 1672.

261. A.B. Kummer in the rd

New Orleans, LA, USA, 2000, p.25.

262. M. Doyle and J. A.E. Manson in the ANTEC 2000

Conference, Orlando, FL, USA, 2000, Paper No.169.

263. A.L.A Bobovich, Y, Unigovski, E.M. Gutman and E. Kolmakov in the rd , SPE, Boston, MA, USA,

2005.

264. M. Patel, Polymer Testing, 2004, 231, 107.

265. P.D.B. Jung, D. Bhattacharyya and A.J. Easteel, ,

2005, 40, 4775.

266. R. Hernandez, A. Sarafian, D. Dopez and C. Myangos, Polymer, 2004, 45,

5543.

267. J.D.O. Mela and R.W. Radford, Journal of Composite Materials, 2004, 38,

1815.

143


268. W. Lifeng, E.W. Cochran, T.P. Lodge and F.S. Bates, Macromolecules, 2004,

37, 3360.

269. J. Capodagli and R. Lakes, 2008, 47, 777.

270. O. Robles-Vadquez, A. Gonzales-Alvarez, J.E. Puig and O. Manero,

2006, 79, 859.

271. D. Wu, L. Wu, F. Goa, M. Zhong and C. Yan,

, 2008, 48, 966.

272. J. Ding, X. Ding, X. Riwei and Y. Ding Sheng, Journal of Macromolecular , 2005, 1344, 308.

273. K.K. Kar, P. Paik and J.U. Otaigbe, in the nd SPE

Conference - ANTEC 2004, Chicago, IL, USA, 2004, p.2495.

274. A.L. Bobovitch, Y. Linigovski, E.M. Gutman, E. Kolmakov and S.

Yyazovkin, , 2007, 103, 3218.

275. J.P. Habas, E. Pavie, A. Happ and J. Perelasse, , 2008, 47,

765.

276. J. Hepperle, H. Muenstedt, P.K. Haug and C.D. Eisenbach, ,

2005, 45, 151.

277. J. Hepperle and H. Muenstedt, , 2006, 45, 717.

278. J. Strange, S. Waechtex, H. Muenstedt and H. Kaspar, Macromolecules, 2007,

40, 2409.

279. J. Jachowicz, R. McMuller, C. Wu, L. Senak and D. Koelmel, Journal of , 2007, 105, 190.

280. F. Yang, X-H. Xiong, Q. Wang and L. Li,

, 2007, 23, 190.

281. A. Lindner, B. Lestriez, S. Mariot, C. Creton, T. Maevis, B. Luehmann and R.

Brummer, , 2006, 82, 267.

282. L. Andreozzi, V. Castelvetro, M. Faetti, M. Giordano and F. Zulli,

Macromolecules, 2006, 39, 1880.

283. J. Chan, X. Liu and D. Yang, , 2006,

102, 4040.

144


284. J-L. Zhang, L. Shen, W. Chun Li and Q. Zheng,

, 2006, 22, 123.

285. K.Y. Kim, G.J. Nam, S.M. Lee and J.W. Lee,

, 2006, 99, 2132.

286. P. Prasopnatre, P. Saroui and C. Sirisinha,

, 2009, 111, 1051.

287. ISO 9001, , 2001.

288. ASTM D1329,

, 2002.

289. ASTM D1052,

, 2005.

290. ASTM D945 – 92el,

, 2001.

291. R. H. Chambers, , 1979, 169, 1109.

292. K.H.J. Klaren, , 1977,

60, 205.

293. ASTM D3451-01,

Coatings, 2001.

294. ASTM D3451-0,

Materials, 2001.

295. ASTM D1921, , 2001.

296. BS ISO 13319,

, 2000.

297. ASTM D609,

, 2000.

298. DIN ISO 8130-10, , 2005.

299. ASTM D822-01,

, 2001.

145


300. ASTM G155-04a,

, 2004.

301. ASTM E308,

, 2008.

302. BS 2701, , 1965.

303. ASTM D1895-96,

, 2003.

304. ASTM D523, , 2008.

305. DIN 53157, , 2000.

306. ASTM D2794-93,

, 2004.

307. ASTM D522-93a,

, 2001.

308. ASTM B117-64, ,

2003.

309. DIN EN ISO 8130-10,

, 2011.

310. NFT 30 – 5-2,

, 1977.

311. NFT 30 – 500,

, 1976.

312. ASTM D1921,

Plastic Materials, 2001.

313. M. Terray, , 1998, 398, 121.

314. R. Trottier, J. Szalanski, C. Dobbs and A. Felix in the

Orlando, FL, USA,

Fall 1996, 75, 58.

315. J. Venkatesan and C.A. Silebi in the ACS Polymeric Orlando, FL, USA, 1996,

Volume 75, p.99.

146


316. C. Boerschig, B. Fries, W. Gronski, C. Weis and C. Friedrich, Polymer, 2000,

41, 3029.

317. R. Trottier, J. Szalanski, C. Dobbs and A. Felix in the

Orlando, FL, 1996, Volume 75,

p.58.

318. M. Ifwarson and K. Aoyama in

, Goteborg, Sweden, 1982, p.731.

319. R. Vegas, , 2002, 83, 589.

320. C. Bonter and E. Schmechtenberg, , 2000, 90, 38.

321. D.V. Reddy and S. Ataoglu, 2004, 36, 43.

322. S. Ghoch and D.K. Tully, , 1982, 9, 12.

323. M.A. Gomez, G. Ellis and C. Marco, , 2002,

83, 582.

324. Y. Germain, , 1998, 38, 657.

325. UNI 7448, Tubi di PVC Rigido (Non Plastificato), , 1975.

326. G.D. Sims in the

Composites, London, UK, 1994, Publication No. 293/9, p.101.

327. ASTM D1709,

, 2004.

328. UNI EN ISO 7765-1,

,

2005.

329. R.M. Patel, P. Saavedra, J. de Groot, C. Hinton and R. Guerra in the

TAPPI:

Toronto, Ontario, Canada, 1997, Book 1, p.239.

330. S.S. Wood and A.V. Pocius in the

Orlando, FL, USA, 2000, Paper No.33.

331. M.G. Cernak and W.L. Chiang in the Polymers,, Toronto, Ontario, Canada, 1997,

Book 2, p.479.

147


332. W. Michaeli, S. Goehel and R. Dahlmann, ,

2004, 24, 107.

333. R.M. Patel, P Saavedra, J. de Grat, C. Hinton and R. Guerra in the

Conference,

Toronto, Ontario, Canada, 1997, Book 1, p.239.

334. H-T. Lee and L-H. Lin, Macromolecules, 2006, 39, 6133.

335. Y.P. Wang, X. Goa, R.M. Wong, H.G. Liu, C. Yang and Y.B. Xiong,

, 2008, 687, 1170.

336. Z. Liu, K. Chen and D. Yan, Polymer Testing, 2004, 23, 323.

337. T. Mikolajczyk and M. Olejnik, , 2007,

15, 26.

338. B. Pourabbas and H. Azimi, Journal of Composite Materials, 2008, 42, 2499.

339. K. Stoeffler, P.G. Lafleur and J. Denault, ,

2008, 48, 1449.

340. H. Mirzazadeh and A.A. Katbaly, ,

2006, 17, 975.

341. J. Bandyopadhyaly, S.S. Ray and M. Bousmina, Macromolecular Chemistry , 2007, 208, 1979.

342. Y.C. Kim, S.J. Lee, J.C. Kim and H. Cho, , 2005, 37,

206.

343. W.S. Chaw and Z.A. Mohd Ishak, , 2007, 1, 77.

344. D-R. Yei, S-W. Kuo, H-K. Fe and F-C. Chang, Polymer, 2005, 46, 741.

345. D.Y. Wei, S. Innace, E. Di Maio and L. Nicolais,

, 2005, 43, 689.

346. S. Tanoue, L.A. Utracki, A. Garcia Rejon, P. Sammut, I. Pesneau, M.R. Kamal

and I. Lyngaae-Jorgensen, , 2004, 44, 1061.

347. K. Stoeffler, P.G. Lafleur and J. Renault in the th SPE

2006, Charlotte, NC, USA, 2006, p.263.

148


348. ASTM D693-03a,

.

349. L. Zsedai, G. Kalacska, K. Vercommen and K. Van Acber,

, 2002, 29, T17.

350. P.M. Taylor and D.M. Pollet,

, 2000, 91, 1.

351. E.A. Poppe, K. Leidig, K. Schirmer and L. Jayle,

, 1997, 760, 41.

352. R. Selden, , 1997, 37, 205.

353. M.Y. Keating, L.B. Malone and W.D. Saunders,

, 2002, 69, 37.

354. S.C. Shik, , 2003, 48, 55.

355. ASTM D2944-71,

Materials, 1998.

356. ASTM D648,

, 2007.

Documents

Crompton - Physical Testing of Plastics