5 Solution, Suspension and Casting Processes

182 Principles of the Processing of Plastics

by interdiffusion of polymer molecules from separate particles to form a coherent

homogeneous mass. Success in the use of PVC pastes depends, amongst other

things, on the ability to keep plasticiser diffusion to a minimum at storage

temperatures but for it to be rapid at the ‘setting’ temperatures.

A third type of suspension is that in which polymer particles are suspended in

monomer which is then polymerised. This is, however, rather more a variant of

the casting process in which monomer or low molecular weight polymer is cast

into a mould and then polymerised in situ.

Polymerisation casting involves mixing monomer or low molecular weight

polymer with a polymerisation initiator, pouring the mix into the mould and

allowing polymerisation to occur in situ. A variation is to impregnate fibres with

initiated monomer or other low molecular weight material and polymerise to

produce composite structures. The main problem is due to the heat of

polymerisation. Unless heat transfer distances are kept short or unless the

reaction is carried out very slowly it can easily get out of hand.

An important development of polymerisation casting is that of reaction

injection moulding. Developed primarily for polyurethanes (and discussed

further in Chapter 27), the process has also found some use with polyamides and

with epoxide resins.



8.6 SUMMARY

A wide variety of methods is available for both shaping polymers and setting

the shapes once formed. In order to obtain the best products in the most

economic way processors need an understanding of the underlying polymer

science.

It is, however, much easier to write such statements than to achieve perfection

in practice. Some idea of the practical difficulties are vividly illustrated in Figure

8.18 due to Clegg13 which shows the complex interrelations of process variables

in an extrusion process.



ELASTICITY (DIE SWELLING)



VISCOSITY



DEGREES OF CONTROL



t



MAJOR



+ MINOR



Figure 8.18. Interrelationships of process variables in an extrusion process. (After Clegg”)



Bibliography



183



References

1. KENNAWAY, A., Brit. Plastics, 28, 18 (1955)

2. WHELAN, A. and COW, I. P., Paper presented to the PRI Mouldmaking '86 Conference at Solihull,

England, January 1986

3. WILLIAMS, M. L., LANDEL, R. F., and FERRY, J . D., J . Am. Chem. SOC.,77, 3 7 0 1 (1955)

4. COGSWELL, F. N., Plastics & Polymers, 41, 39 (1973)

5. DOWNS, G. Vinyl and Allied Polymers Vol. 1 (Ed. KITCHIE, P. D.), p. 128, Iliffe, London (1968)

6. BUECHE, F., J . Chem. Phys., 20, 1959 (1952); 25,599 (1956); see also a summary in BUECHE, E,

Physical Properties of Polymers, Interscience, New York (1962)

7. PECZIN, G., Pure Appl. Chem., 26 (2), 241 (1971)

8. BRYDSON, J . A,, Flow Properties ofpolymer Melts (2nd edn), Geo. Godwin, London (1981)

9. BEYNON, D. L. T., and CLYDE, B. s., Brit. Plastics Inst, 33, 414 (1960)

IO. HOWELLS, E. R., and BENBOW, J . I., Trans. Plastics Inst., 30, 242 (1962)

11. GROVES, W. R., Plastics Moulding Plant Vol. 1, IIiffe, London (1963)

12. ESTEVEZ, J . M. J . , and POWELL, D. c., Manipulation o Thermoplastic Sheet, Rod and Tube, Iliffe,

f

London (1969)

13. CLEGG, P. L., Chapter in Thermoplastics: Effects of Processing (Ed. OGOKKIEWICZ, K. M.), IIiffe,

London (1960)

14. BAKRIE, I. T ,Chapter in Polymer Rheology (Ed. LENK, K. s.), E l s e v i e r A p p l i e d Science Publishers,

.

London (1978)



Bibliography

c. (Ed.), Processing of Thermoplastic Materials, Reinhold, New York (1959)

c., High Polymer Lntices, Maclaren, London (1966)

BRYDSON, J . A,, Flow Properties o Polymer Melts, (2nd Edn), Geo. G o d w i n , London (1981)

f

BRYDSON, J . A,, Handbook for Plastics Processors, Butterworth-Heinemann, Oxford (1990)

CRAWFORD, R. J., Plastics Engineering, 2nd Edn, Pergamon, Oxford (1990)

GOGSWELL, F. N., Polymer Melt Rheology, Geo. Godwin, London (1981)

FENNER, K. T., Principles o Polymer Processing, Macmillan, London (1979)

f

MCKELVEY, J . M., Polymer Processing, W i l e y , New York (1962)

OGORKIEWICZ, R. M. (Ed.), Thermoplastics: Effects of Processing, Iliffe, London (1969)

SARVETNICK, H. A. (Ed.), Plastics and Organosols, Van Nostrand-Reinhold, New York ( 1 9 7 2 )

TADMOK, z., and GOGOS, c. G., Principles o Polymer Processing, Wiley, New York (1979)

f

BEKNHARDT, E.



BLACKLEY, D.



9

Principles of Product Design



9.1



INTRODUCTION



Successful product design requires knowledge, intelligence and flair. The

knowledge requirement may in turn be subdivided into:

(1) A knowledge of the requirements of the product.



(2) A knowledge of the behaviour of plastics materials.

(3) A knowledge of plastics processes.

(4) A knowledge of all relevant economic and psychological factors.

Intelligence is required to relate this knowledge, and flair to bring the design

to a successful reality. It is not a function of this book to consider points 1, 3 and

4 above but it is a function to consider those material properties that are relevant.

Probably enough has been said already about chemical, electrical and optical

properties but some additional thoughts on mechanical and thermal properties

relevant to design are worth considering.

For successful use a polymer must have appropriate:



(1)

(2)

(3)

(4)

(5)



Rigidity.

Toughness.

Resistance to long-term deformation (creep).

Recovery from deformation on release of stress.

Resistance to thermal degradation adversely affecting properties.



over the range of operating conditions likely to be encountered. This chapter will

consider how the diverse properties of plastics in respect of the above properties

can be explained by reasonably elementary yet fundamental considerations.

9.2 RIGIDITY OF PLASTICS MATERIALS



The rigidity of a polymer is determined by the ease with which polymer

molecules are deformed under load. In a polymer at -273°C all load is taken by

184



Rigidity of Plastics Materials



185



bond bending and stretching and for a polymer with no secondary transitions this

state of affairs more or less exists up to the region of the glass transition

temperature Tg. Several polymers do, however, show additional transitions below

the Tg at which point movement of small moieties in the polymer become

possible. This allows more response of the polymer to stress and there is a

decrease in modulus. Such decreases are usually small but the change at the Tg

in an amorphous polymer is considerable as the modulus drops from values of the

order of 500 000 lbf/in2 (3500MPa) to values of about 100 lbf/in2 (0.7 MPa).

Further heating of a polymer such as a commercial polystyrene would rapidly

cause a drop of the modulus towards zero but in a high molecular weight polymer

such as a cast poly(methy1 methacrylate) the entanglements would enable the

material to maintain a significant rubbery modulus up to its decomposition

temperature. Similar effects are achieved when the polymer is cross-linked and,

as might be expected, the more the cross-linking, the higher the modulus.

Molecular movement above the Tg is restricted by crystallinity and, as with

chemical cross-linking, the more the crystallinity, the more rigid the polymer.

Some polymers tend to melt over a wide temperature range, in which case the

modulus may fall over a range of temperatures leading up to the melting point

T,. The above effects are summarised in Figure 9.1.

Plastics materials, in general are blends of polymers with additives and the

latter may well affect the modulus. One simple law of use here relates the

modulus of the blend or composite E, to the modulus of the polymer Ep and of

the additive E , by the equation



E, = VFEp+ (1 - VF)E,



(9.1)



where V, is the volume fraction of polymer in the blend. Such a relationship only

holds when there is no real interaction between polymer and additive such as



. n.

Y



\



\



E+



\ \

T

O



Tm



TEMPERATURE



_c_



Figure 9.1. Schematic illustration of dependence of the modulus of a polymer on a variety of factors.

A is an amorphous polymer of moderate molecular weight whereas B is of such a high molecular

weight that entanglements inhibit flow. Similar effects are shown in C and D, where the polymer is

respectively lightly and highly cross-linked. In E and F the polymer is capable of crystallisation, F

being more highly crystalline than E



186 Principles o Product Design

f

occurs between diene rubbers and carbon black where a form of cross-linking

may be considered to occur.

In practice one is basically concerned with the rigidity of the product and this

involves not only the modulus of the material but also the shape and size of the

product. From the points of view of weight saving, economics in material and

ease of processing, it is an important aim to keep section thicknesses down in

size. Since flat or singly curved surfaces have a minimum rigidity the designer

may wish to incorporate domed or other doubly curved surfaces or ribbing into

the product in order to increase stiffness. Corrugation can also enhance stiffness

but in this case the enhancement varies with position, being greatest when

measured at right angles to the corrugation.



9.2.1 The Assessment of Maximum Service Temperature

The design engineer often requires to know the maximum temperature for which

a polymer can be used in a given application. This depends largely on two

independent factors:

(1) The thermal stability of the polymer, particularly in air.

(2) The softening behaviour of the polymer.



Let us consider two polymers A and B. Let A ‘soften’ at 120°C but have longterm thermal stability to 200°C. On the other hand polymer B softens at 200°C

but degrades ‘at a measurable rate’ above 90°C. Consideration of these figures,

even allowing for the loose terminology, indicates that material A could not be

used much above 90°C for either long or short periods. In the case of polymer B

short-term service might be possible up to about 160-170°C but it could not be

used for prolonged periods much above 70-80°C.



9.2.1.1



Assessment of thermal stubility



Over the years many attempts have been made to provide some measure of the

maximum service temperature which a material will be able to withstand without

thermal degradation rendering it unfit for service. Quite clearly any figure will

depend on the time the material is likely to be exposed to elevated temperatures.

One assessment that is being increasingly quoted is the UL 746B Relative

Temperature Index Test of the Underwriters Laboratories (previously known as

the Continuous Use Temperature Rating or Index).

In order to obtain a temperature index rating a large number of samples are

subjected to oven aging at a variety of temperatures for periods up to a year.

During the course of this time samples are periodically withdrawn and tested. A

plot is then made of the percentage retention in the value of the property

measured (compared to its original control value) against time. A note is then

made of the time, at each temperature tested, which gives a 50% reduction in

value of the property. Somewhat arbitrarily this is taken as the failure time at that

temperature. Using the data from experiments carried out after aging at various

temperatures, the logarithm of the failure time is plotted against 1/K (where K is

the temperature in kelvin). The resultant linear Arrhenius plot is then extrapolated

to the arbitrary time of 10000 hours. The temperature at which the failure time

(as defined above) is 10000 hours is known as the relative temperature index

(RTI).



Rigidity of Plastics Materials



187



This long-term thermal performance of a material is tested alongside a second,

control, material which already has an established RTI and which exhibits a good

performance. Such a control is necessary because thermal degradation characteristics are sensitive to variables in the testing programme. Since the control

material will also be affected by the same unique combination of these factors

during the tests, there is a valid basis for comparison of test and control

materials.

It is to be expected that the RTI obtained would depend on the property

assessed and in UL 746B three properties are assessed:



(1) ‘Mechanical with impact’-by measuring tensile impact strength.

(2) ‘Mechanical without impact’-by measuring tensile strength.

( 3 ) ‘Electrical’-by measuring dielectric strength.



A value for the RTI is provided for each of these tests although in common

experience it is found that similar numerical values are obtained.

In addition, the RTI may be affected by the thickness of the sample, so this

should be given in any RTI specification.

Such a value for relative temperature index will be specific to a particular

grade of a polymer, sometimes even to a specific colour. The difference between

grades of a particular species of polymer can be substantial, depending both on

the variation in the inherent stability of a material between differing

manufacturing methods and also on the type and amount of additives used. It is

possible to obtain from the Laboratories a Generic Temperature Index to cover a

species of material but this will usually be considerably lower than for many of

the individual grades within that species.

Some collected values for RTI taken from the literature are given in Table 9.1.

(These are given for guidance only and should not be taken to imply official UL

ratings.)

Table 9.1 Some collected values for Relative Temperature Index (RTI) (Unless otherwise stated,

data are for ‘mechanical without impact’ and for unreinforced grades)



RTI (“C)



Polymer

ABS

Nylons

Polyacetal

Styrenic PPO

Polycarbonate

Polyarylate

Poly(buty1ene terephthalate)

Poly(ethy1ene terephthalate)

Polysulphone

Pol yetherimide

Polyphthalamide

Polyethersulphone

Poly(pheny1ene sulphide)

Aromatic polyester

Liquid crystal polyester

Polyether ether ketone



(homopolymer and copolymer)

(Noryl731)

(Lexan 101)

(Lexan 3414R)

(Ardel D100)

(Pocan B1305)

(Pocan B3235)

(Petlon 4630)

(Ultern 1000)

(Amodel A1133HS)

(Victrex 200P)

(Supec G401)

(Ekkcel 1-2000)

(Victrex PEEK)



60-80

75

90

105

125

130 (40% glass filled)

130

140

140 (30% g/f)

150 (30% g/f)

160

170

180

180

200 (40% g/D

220

220

240



188 Principles of Product Design

9.2.1.2 Assessment of softening point

As will be seen from curves A, B and C of Figure 9.1, the 'softening point' of

an amorphous polymer, i.e. the temperature at which the modulus drops

catastrophically, is closely associated with the T g . (Such softening does not of

course occur in highly cross-linked polymers, as in type D, unless degradation

also takes place.)

In the case of crystalline polymers such as types E and F the situation is

somewhat more complicated. There is some change in modulus around the Tg

which decreases with increasing crystallinity and a catastrophic change around

the T,,, . Furthermore there are many polymers that soften progressively between

the Tg and the T,,, due to the wide melting range of the crystalline structures, and

the value determined for the softening point can depend very considerably on the

test method used.

Two particular test methods have become very widely used. They are the Vicat

softening point test (VSP test) and the heat deflection temperature under load test

(HDT test) (which is also widely known by the earlier name of heat distortion

temperature test). In the Vicat test a sample of the plastics material is heated at

a specified rate of temperature increase and the temperature is noted at which a

needle of specified dimensions indents into the material a specified distance

under a specified load. In the most common method (method A) a load of 10 N

is used, the needle indentor has a cross-sectional area of 1 mm2, the specified

penetration distance is 1 mm and the rate of temperature rise is 50°C per hour.

For details see the relevant standards (IS0 306; BS 2782 method 120; ASTM

D1525 and DIN 53460). (IS0 306 describes two methods, method A with a load

of 10N and method B with a load of 50N, each with two possible rates of

temperature rise, 50"C/h and 120"C/h. This results in I S 0 values quoted as A50,

A120, B50 or B120. Many of the results quoted in this book predate the IS0

standard and unless otherwise stated may be assumed to correspond to A50.)

In the deflection temperature under load test (heat distortion temperature test)

the temperature is noted at which a bar of material subjected to a three-point

bending stress is deformed a specified amount. The load ( F ) applied to the

sample will vary with the thickness ( t ) and width (w) the samples and is

of

determined by the maximum stress specified at the mid-point of the beam ( P )

which may be either 0.45 MPa (66 lbf/in2) or 1.82 MPa (264 Ibf/in2).

The formula used for the calculation is:

F = 2Pwt2/3L



Where L is the distance between the outer supports (loading points). For details

see the relevant standards (IS0 75; BS 2782 method 121; ASTM D648; DIN

5346 1).

Whilst the Vicat test usually gives the higher values the differences are quite

modest with many polymers (e.g. those of types A, B and C). For example, in the

case of the polycarbonate of bis-phenol A (Chapter 20) the heat distortion

temperatures are 135-140°C and 140-146°C for the high and low stress levels

respectively and the Vicat softening point is about 165°C. In the case of an acetal

homopolymer the temperatures are 100, 170 and 185°C respectively. With nylon

66 the two ASTM heat distortion tests give values as different as 75 and 200°C.

A low-density polyethylene may have a Vicat temperature of 90°C but a heat

distortion temperature below normal ambient temperatures.



Rigidity of Plastics Materials



189



The differences in the assesement of softening point between the tests is

clearly largely a matter that the ‘end point’ of the test measures a different

modulus. Reference to Figure 9.1 shows that with some materials (e.g. of type A)

this will not be of great importance but with other types (e.g. types E or F) the

difference could be very large.

At the risk of oversimplification it might be said that the Vicat test gives a

measure of the temperature at which a material loses its ‘form stability’ whilst the

higher stress level heat distortion temperature (1.82 MPa) test provides a measure

of the temperature at which a material loses its load-bearing capacity. The lower

stress (0.45 MPa) heat distortion temperature test gives some rather intermediate

figures and it is perhaps not surprising that it is today less often quoted than the

other two tests.

Some interesting differences are noted between amorphous and crystalline

polymers when glass fibre reinforcement is incorporated into the polymer. In

Figure 9.2 (ref. 10) it will be seen that incorporation of glass fibre has a minimal

effect on the heat deflection temperature of amorphous polymers (polystyrene,



‘5-1

110



g

120



n

.

0



Figure 9.2. Heat deflection temperatures under a load of 1.82MPa for selected polymers. Note that

incorporation of glass fibre has a much greater effect with crystalline polymers than with amorphous

ones (after Whelan and Craft courtesy of British PIustics and Rubber)



ABS, polycarbonate and polysulphone) but large effects on crystalline polymers.

It is particularly interesting, as well as being technically important, that for many

crystalline polymers the unfilled polymer has a heat deflection temperature (at

1.82 MPa stress) similar to the Tg, whereas the filled polymers have values close

to the T , (Table 9.2).

Other tests occasionally quoted are the BS softening point test and the Martens

test. These involve the bending under load of samples held at one end as they are

subjected to a rise in temperature.



190 Principles of Product Design

Table 9.2 Comparison of T x . T , and heat deflection temperatures of polymers with and without

glass fibre reinforcement (All values in "C)

~~



Polymer



Hear deflection

temperature

(unfilled)



Polyether ether ketone

Polyether ketone

Polyphenylene sulphide

Polyethylene terephthalate

Polybutylene terephthalate



143

165

85

70

22-43



334

365

285

255

225



(filled)



150

165

135

85

54



3 15

340

260

210



210



9.3 TOUGHNESS

For many applications the resistance to impact is the most important property of

a plastics material. It is also notoriously one of the most difficult to assess.

If a polymer with no secondary transitions is struck a blow at some

temperature well below its glass transitions temperature, deformation will be

very limited before fracture occurs. Nevertheless because of the high modulus

quite high tensile strengths will be recorded, of the order of 80001bf/in2

(55 MPa). The energy to break will be given by the area under the stress-strain

curve and will not be very large (see Figure 9.3).



(b)



(a)



Figure 9.3. Stress-strain curves for (a) rigid amorphous plastics material showing brittle fracture and

(b) rubbery polymer. The area under the curve gives a measure of the energy required to break the

test piece.



On the other hand, if an amorphous polymer is struck above the T g ,i.e. in the

rubbery state, large extensions are possible before fracture occurs and, although

the tensile strength will be much lower, the energy to break (viz. the area under

the curve) will be much more, so that for many purposes the material will be

regarded as tough.

A common requirement is to produce a rigid plastics material with the

toughness of a rubber. This can be achieved in a number of ways:



(1) By the use of a moderately crystalline polymer with a Tg well below the

expected service temperature (e.g. polyethylene).



Toughness 191

(2) By block copolymerisation so that one component of the block copolymer

has a Tg well below the expected service temperature range (e.g

polypropylene with small blocks of polyethylene or preferably polypropylene with small amorphous blocks of ethylene-propylene copolymer).

(3) By blending with semi-compatible materials which have a Tg well below the

expected service temperature range (e.g. high-impact polystyrene-as

described in Chapter 3).

(4) By the use of a polymer which has effective transitions at or below the

expected service temperature range and which is able to respond to stress by

extensive deformation (e.g. polycarbonates).

(5) By plasticisation. This in effect reduces the Tg and in the case of nylon which

has absorbed small quantities of water the toughening effect can be quite

substantial. It should, however, be noted that in the case of PVC small

amounts of plasticiser actually reduce the impact strength.

The above procedures will vary in their efficiency with the extent to which

they are employed. For example, too little of a stereoblock in case (2) or too little

rubber in case (3) may not be enough to make the material tough and there is thus

a somewhat critical tough-brittle transition range. In terms of a stress-strain

curve a brittle material may be considered to be one that breaks without a yield

whilst a tough material yields to give a substantial energy to break. (It is perhaps

worth noting that if a material has not broken after being struck simply because

it yielded irrecoverably the product may still be useless.)

What is important to realise is that a polymer may be tough when exposed to

tensile load but brittle when assessed by an Izod-type test where a notched

sample is subjected to a bending load. Table 9.3 attempts to summarise the

behaviour of typical polymers to different stresses.



Table 9.3

Type of stress



Polymers ducrile a1 25°C and at

I min-' strain rate



1. Around Izod notch



Low-density polyethylene

Ethylene-propylene block copolymers

Cellulose nitrate and propionate

ABS and high-impact polystyrene

Bis-phenol A polycarbonate



2. Tension



Above materials plus:

High-density polyethylene

Polypropylene

Acetal polymers

Aliphatic polyamides (nylons)

PPO

Poly(ethy1ene terephthalate)

Polysulphones



3. Simple shear



Above materials plus:

Poly(methy1 methacrylate)



4. Compression



Above materials plus:

Polystyrene



I92 Principles o Product Design

f

Toughness is not simply a function of polymer structure or the mode of stressing.

It clearly will also depend on the temperature and the rate of striking but more

important still it will depend on the product design and method of manufacture.

Consider two products A and B (Figure 9.4) where the molecules are randomly

disposed as in A and aligned as in B. In A there will be no planes of weakness

whereas if B is struck a blow at point P failure is much more likely to occur along

XY than at any angle to this line (or plane if we consider a three-dimensional

system). For optimum toughness the aim of the processor must therefore be to

minimise such orientation. (It should be noted that in Chapter 3 it was found that

the impact strength increased to a limit with increasing biaxial orientation. This

would only apply when the impact blow was made at approximately right angles

to the plane of orientation.)



B



A



Figure 9.4



The presence of notches or sharp angles or of a few holes, voids, particle

inclusions or small inserts tends to concentrate the stress. Different polymers

vary in their 'notch sensitivity' and this is presumably a reflection of how close

they are to their tough-brittle transitions. The aim of the designer and processor

must be to reduce such stress concentration to a minimum.



9.3.1 The Assessment of Impact Strength



'



It is probably most useful to consider toughness as a property of a plastics part

under some specified conditions of service. Whilst it is possible to devise impact

tests and to rank a series of plastics materials according to the results obtained in

such tests it remains almost impossible to use such tests to try to predict whether

or not an article made from a specific material will or will not be satisfactory in

service.

Amongst the factors that will influence service performance are the effect of

additives and impurities, temperature, detailed geometric size and shape,

orientation and morphology, surface condition, energy and speed of any

impacting blow, the shape of the impacting instrument, the environment, and

strains in the article due to external loads. For this reason it is desirable, but not

always feasible, to test prototype articles under conditions as close to service

conditions as possible.

Impact tests are, however, used to try to compare the impact strength of

different materials. Of these tests four require specific mention. These are the

Izod test, the Charpy test, the falling weight tests and the tensile impact test.

Of these the most well known is the Zzod test. This consists of a bar, one end

of which is held in a vice, the sample being held vertically. The bar is then struck

by a striking device under controlled conditions at a specified point above the

vice. The energy required to break the sample is noted. It is common to have a

notch in the bar which is located during the test at the top of the vice and on the