1

Exercises From polymers to plastics

A.K. van der Vegt

DUP Blue Print

VSSD DUP Blue Print is an imprint of: Delft University Press P.O. Box 98, 2600 MG Delft, The Netherlands tel. +31 15 27 85678, telefax +31 15 27 85706, e-mail info@library.tudelft.nl internet: http://www.library.tudelft.nl/dup Published on behalf of: Vereniging voor Studie- en Studentenbelangen te Delft Poortlandplein 6, 2628 BM Delft, The Netherlands tel. +31 15 27 82124, telefax +31 15 27 87585, e-mail: hlf@vssd.nl internet: http://www.vssd.nl/hlf URL on this book: http://www.vssd.nl/hlf/m008.htm A collection of digital pictures, printed in the book, can be made available for lecturers who adopt this book. Please send an e-mail to hlf@vssd.nl. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photo-copying, recording, or otherwise, without the prior written permission of the publisher. NUGI 831 Key word: polymers

Contents
1. 2. 3. 4. 5. 6. 7. 8. 9. INTRODUCTION MOLECULAR COMPOSITION GLASSY STATE AND GLASS-RUBBER TRANSITION CRYSTALLINE POLYMERS RUBBERY AND LIQUID PHASES VISCO-ELASTICITY MECHANICAL PROPERTIES FURTHER PROPERTIES POLYMERIC COMPOUNDS AND COMPOSITES 5 7 13 15 19 23 26 30 32 36 41 41 44 52 56 63 67 73 78 81 86

11. PRINCIPLES OF PROCESSING ANSWERS 1. Introduction 2. Molecular composition 3. Glassy state and glass-rubber transition 4. Crystalline polymers 5. Rubbertoestand en vloeibare toestand 6. Visco-elasticity 7. Mechanical properties 8. Further properties 9. Polymeric compounds and composites 11. Processing techniques

1
Introduction
1.1. Give twenty high molecular materials from a natural (both from vegetable and animal) origin, that are used as technical materials 1.2. Give ten technically used materials based on half synthetic polymers and explain the addition half to it. 1.3. Why are hydrocarbons, obtained from a cracking process, very suitable for the production of synthetic polymers? 1.4. Why are polymers, built up from dienes, very suitable for application as technical rubbers? 1.5. Which kind of chainstructures can be formed in the polymerisation of butadiene (CH2 =CHCH=CH 2 )? 1.6. Why is it not possible to vulcanise ethylene-propylene copolymers and polyisobutylene in the usual way. In what way are these polymers modified in order to make vulcanisation possible anyway? 1.7. In what way can polymers be formed from saturated monomers? 1.8. In which two manners can a three dimensional network be formed? 1.9. What is the meaning of the name thermosets? Why can this name lead to misunderstanding? 1.10. What is the difference between thermoplastics on the one side and thermosets and rubbers on the other side? 1.11. What is the difference between thermosets and technical rubbers? 1.12. What is the difference between thermoplastic rubbers and ordinary rubbers? 1.13. What is the difference between thermoplastic rubbers and thermoplastics? 1.14. Give the most important thermoplastics, thermosets and synthetic rubbers. 1.15. Give the three maintypes of PE. What are the differences between these types. 1.16. Give a number of polymers (thermoplastics, thermosets and rubbers) in which

Problems Polymers

styrene plays a role. 1.17. Which polymer cannot be manufactured to an end product in a normal production process? In what way is this polymer being processed? Which similar polymers are more suitable for production processes? 1.18. Give a polymer with an extremely high temperature resistance and some members of the same family. 1.19. Give a number of composite plastics 1.20. When and for what purpose are the following materials added to a polymer? a. fine carbon black d. plasticizer g. lubricants j. wood flour m. UV-stabilisers b. chalk e. mica h. sulphur k. anti-oxidants n. antistatic agents c. glassfibres f. accelerators i. pentane l. silicium carbide o. rubber particles.

2
Molecular composition
2.1. In what aspects can the mainchains of polymers differ? 2.2. a. Give the name of four polymers with sidegroups only consisting of hydrogen. b. Give the name of some vinylpolymers of the type -CH 2 -CHR-. c. Give the name of some vinylidenepolymers, -CH2-CR2 -. 2.3. Of three polymers, PE, PS and PA-12 the degree of polymerisation is 500. Calculate their molar masses. Calculate for each polymer the contour length. For which types of PE and PS the contour length calculation is valid? 2.4. If we lay a chain with molar mass of 80,000 kg/kmol next to a chain with molar mass of 120,000 kg/kmol, why is the average molar mass not 100,000 kg/kmol? What is the real meaning of the value 100,000 kg/kmol and what is the real value of Mw? 2.5. Give a definition of n i , wi , z i , andMn , Mw en Mz. What is the value of ni , wi and zi ? 2.6. Show that Mn can be calculated from the weight fractions wi through 1/Mn = (wi/Mi). 2.7. Show that MwMn = niM2 i. 2.8. Show that MwMz = wiM2 i. 2.9. a. Give a description of a monodisperse polymer. b. Give the definition of the degree of polymerisation. 2.10. How will an originally very narrow molar mass distribution change if the polymer is being broken down gradually, for example under the influence of oxygen? 2.11. We mix two polydisperse polymers, A and B with average molar masses: A: (Mn )A, (Mw)A and (Mz)A; B: (Mn )B, (Mw)B and ( Mz)B. The mass fractions of A and B are wA and wB. Give a formula from which the Mw can be calculated. 2.12. We have three batches, A, B and C of a polymer, each with a Mw of of 60,000 kg/kmol, while the heterogenity indices (D = Mw/Mn ) are 6.3 and 2, respectively. 10

Problems Polymers

kg of A is mixed with 10 kg of B. How many kg of C must be added to this mixture to obtain a mixture with D= 4? 2.13. Equal weight quantities of two polydisperse polymers with Mn of 15,000 and 10,000 respectively, Mw of 50,000 and 60,000 respectively and Mz of 100,000 and 200,000 kg/kmol are mixed. Calculate the Mn , Mw and Mz of the mixture. 2.14. Three monodisperse polystyrene fractions A, B and C have molar masses of 76,000 , 250,000 and 850,000 kg/kmol respectively. The intrinsic viscosities [], as measured in toluene, are 0.382, 0.91 en 2.21 dl/g. Determine the values of the constants k and a in the Mark-Houwink equation from these data. 2.15. From the fractions in 2.14, 37 grammes of A, 18 grammes of B and 45 grammes of C are mixed. Calculate Mw and Mn of the mixture. 2.16. From the mixture of 2.15 solutions in toluene are being made with concentrations of 0, 0.1, 0.2, 0.3 and 0.4 g/dl. These fluids need 95.0, 107.7, 121.4, 136.3, and 152.8 seconds to run through an Ubbelohde viscosimeter. Determine the intrinsic viscosity of the mixture and from the intrinsic viscosity the the average molar mass. Compare the value with the values calculated in 2.15. 2.17. Calculate the Mv of the mixture, without the determination of the []. 2.18. Eight fractions of polystyrene with very narrow molar mass distributions are mixed fraction nr 1 2 3 4 5 6 7 8 mass (g) 10 18 25 17 12 8 6 4 M (kg/mol) 15 27 39 56 78 104 120 153 a. To which particular fraction can the mixture be compared best with respect to melt-viscosity? b. What is the difference between the injection moulding behaviour of this particular fraction in comparison with that of the mixture? c. To which particular fraction can the mixture be best compared with respect to impact strength? d. To which particular fraction can the mixture be compared best with respect to die-swell? e. To which particular fraction can the mixture be compared best with respect to modulus of elasticity in the solid state? 2.19. 50 kg of a monodisperse polymer with M = 60,000 kg/kmol is mixed with a certain quantity of a comparable, also monodisperse polymer with M = 600,000 kg/kmol

Molecular composition

a. How many kg of the latter polymer are needed to obtain a number average molar mass of 100,000? b. What will then be the values of weight average and z-average molar masses. 2.20. For the end-to-end distance in a coil of a polystyrene with M = 1000,000 2 1/2 kg/kmol it can be determined that ( r0 ) = 73,5 nm. Calculate the coil density and compare this value with the density of polystyrene 1050 kg/m3 . 2.21. Calculate the degree of dilution for a polystyrene with M= 10,000 kg/kmol. 2.22. What are the consequences from the results found in 2.20 and 2.21 for the socalled dilute solution (a solution in which the coils do not penetrate into each other)? 2.23. In what way will this influence the measurement of the intrinsic viscosity? 2.24. Two batches of PE, A and B, with Mw = 200,000 and 30,000 kg/kmol respectively are being mixed with the intention to obtain a mixture with Mw = 81,000. Both batches have the same form of molar mass distribution, with D = Mw /Mn = 5, and Mz/Mw = 3. Calculate how many kg of each component is needed to obtain 1000 kg of the desired mixture. Calculate the Mn and Mz of the mixture. To what extent will the properties of this mixture differ from a directly produced batch with Mw = 81,000 and the same distribution width as A and B? 2.25. The intrinsic viscosity is related to the molar mass by the Mark-Houwink equation: [] = kMa. a. In what range can the value of a usually be found? b. What is the relation of Mv of a polydisperse polymer, calculated from the Mark Houwink equation, to Mn and Mw? c. In which case can Mv be higher than Mw? 2.26. We have two white cubes. One cube is a polymer. The other one is a low molecular substance. Both cubes can be solved in the same solvent. Give some simple ways how to determine in the laboratory which of the two cubes is made of the polymer? 2.27. Explain why the heterogenity index D = Mw/Mn is approximately 2 for an ideal polycondensation process. 2.28. Equal molar quantities of 1,2 ethanediole and benzene-1,4-dicarbonic acid are mixed in order to react to a polymer. a. What is the mass ratio of the components? b. Give the equation of reaction. What is the name of the polymer? c. What is the meaning of the symbols in the formula P = 1/(1 p)?

10

Problems Polymers

d. Which conversion is needed to reach a Mn of 40,000 kg/kmol? What will then be the value of Mw? e. Which circumstances can prevent that the desired conversion will be reached? 2.29. Which type of polymerisation process will result in a polymer with a D-value close to 1? 2.30. Which methods can be use for the measurement of Mn , of Mw and Mz and give a description of the principles on which these methods are based. 2.31. For a polymer in solution the following data were found from a viscometer experiment: run time solvent run time solution 100 106.25 115 126.25 140 sec sec sec sec sec

0.05 g/dl 0.1 g/dl 0.15 g/dl 0.2 g/dl

Calculate the intrinsic viscosity from these data. 2.32. If the value of [ ] for the same type of polymer with M = 50,000 and M = 800,000 is 0.5 and 2.0 dl/g respectively, what is the value of M of the sample under investigation? 2.33. Calculate the values of the constants in the Mark-Houwink equation for the polymer solvent combination from 2.31. 2.34. a. A monodisperse polymer with M = 20,000 is contaminated with 2% (w/w) polymer of M = 200. Calculate the values of Mn , Mw and Mz. b. Now the monodisperse polymer contains 2% (w/w) chains with M = 2,000,000. Calculate also the values of Mn , Mw and Mz for this contaminated polymer. 2.35. Osmotic pressures are measured at 25 C from a dissolved polymer at various concentrations c and the following fluid heights were found: c (g/dl) 0.12 0.18 0.25 0.33 0.45

h (mm) 6.50 9.94 14.05 18.8 26.6

Determine Mn from these data. The solvent density is 903 kg/m3, R = 8.314 J/mol.K

Molecular composition

11

and g = 9.81 m/s2. 2.36 . The two figures give imaginary results of osmotic pressure measurements (/c versus c) and solution viscosity measurements ( red versus c), each with two different solvents A and B. Explain why in the case of the osmotic pressure measurements extrapolation leads to the same value and in the case of the viscosity measurements extrapolation does not lead to the same value.
c

red

c Figure 2.1. Figure for problem 2.36

2.37. Why does the sensitivity of the determination of the intrinsic viscosity increase with increasing molar mass but decrease with the determination of the osmotic pressure? 2.38. Nylon 6.6 is produced twice by an ideal polycondensation process from adipic acid, HOOC-(CH2 )4 -COOH, and hexamethylenediamine H2N-(CH2)6 -NH2. In the first case (A) the conversion p = 0.99, in the second case (B) the conversion is 0.995. Calculate Mn and Mw for both cases. 2.39. 30% of A and 70% of B are mixed. Calculate Mn and Mw for this mixture. 2.40. How can, without measuring the viscosities of solutions at various concentrations, a simple comparative measure be found for the average molar mass? 2.41. For PVC k-values are used, like k = 55, k = 70. What is the meaning of these values both molecular and technological? 2.42. Why is the melt-index, as used for e.g. PE, PP and ABS a useful and functional measure for the molar mass? 2.43. A light scattering experiment applied to two types, A and B, of the same polymer results in the same value for M for both types. The value of the intrinsic viscosity of type A is however higher than that of type B. What can be said about the ratio of the osmotic pressures of A and B?

12

Problems Polymers

2.44. Which principle is the basis of GPC (Gel Permeation Chromatography or SEC, Size Exclusion Chromatography) How can the various average molar masses like Mn , Mw and Mz be determined from the result of a GPC-measurement? 2.45. In what aspects can a polymer chain be built up irregularly? Give examples of each aspect. 2.46. Give schematically the difference in the structure of an isotactic, an atactic and a syndiotactic polymer formed from the monomer CH2 =CHX. 2.47. What are the consequences of the steric configuration of the chain for practical usability and for the properties if X is CH3 or C6 H5 ? (T g of PP and PS are 15 C and 95 C respectively) 2.48. Give two factors which cause a high chain stiffness. Give an example of each factor. 2.49. Give two strong and one weak interaction force between chains and give examples of polymers in which these forces occur. 2.50. What is being understood by the expression cross-link density? Give another measure to characterize the degree of crosslinking. How can these properties be determined? 2.51. What is the difference between chemical and physical cross-linking. Give examples of both types. In what way does the difference show up in the properties of the network?

13

3
Glassy state and glass-rubber transition
3.1. Give differences and similarities between ordinary anorganic glass and a polymeric glass like polystyrene. 3.2. Give three essential differences between a glass-rubber transition and a melting point, in particular concerning the temperature dependency of volume V, enthalpy H and entropy S around these transition points.. 3.3. Why can, just above the Tg , complete chains not yet move with respect to each other? When are they able to do so? 3.4. Give examples of the phenomenon that enlarging of the distance between the chains leads to lower Tg -values. 3.5. Why can the glass-rubber transition not be considered as thermodynamic second -order transition? 3.6. Polyhexene-1 and polymethylenepentene both have a C 4 H9 -sidegroup connected to the main chain. The Tg -values are 50 C and +30 C respectively. Explain the difference. 3.7. If we look at the following series: PE (-CH2 -CH2 -) PP (-CH2 -CHCH3-) PVC (-CH2 -CHCl-) Tg = 120 C = 15 C = 90 C

a strong rise of the glass-rubber transition temperatures can be seen. Explain the differences. Let us now consider the following series. BR (-CH2 -CH=CH-CH2 ) Tg = 90 C IR (-CH2 -CCH3 =CH-CH2 -) = 73 C CR (-CH2 -CCl=CH-CH2-) = 50 C Explain why the differences between the glass-rubber transition temperatures are now considerably smaller.

14

Problems polymers

3.8. We compare the Tg of PMMA (110 C) first with the Tg of poly methyl acrylate (5 C), then with the Tg of poly (n-butyl methacrylate)(25 C) and finally with the Tg of poly(n-octylmethacrylate)(-20 C). Explain the differences. 3.9. An amorphous plastic sometimes shows two transitions, within or from the glassy state. Give three possible causes for this phenomenum. 3.10. The Tg of an SBR is 65 C. Which factors determine this value? SBS shows two glass-rubber transitions, one at 90 C and one at +85 C. Explain these values if you know that the Tg of PS is 95 C. 3.11. What analogy is there, with respect to glass-rubber transitions, between copolymers and polymerblends? 3.12. If a single glass-rubber transition is present does this mean that the polymer is completely mixed on a molecular scale? 3.13. Why does crosslinking influence the value of Tg ? 3.14. A polybutadiene rubber (BR) with molar mass of 200,000 kg/kmol is being vulcanised with sulphur in such a way that the average number of connections that each chain has with another one is four. Explain why the Tg of this vulcanised rubber does not differ significantly from the Tg of the one that does not contain crosslinks. 3.15. Why is the Tg of a polyamide variable?

15

4
Crystalline polymers
4.1. Give a reason for the fact that some polymers with regular chains, like PC, PPE, PIB, cis-polyisoprene and cis-polybutadiene do not crystallize. 4.2. Why is not desirable that these polymer should crystallize? 4.3. In the production of IR and BR one aims at the highest possible cis-content or in other words, a maximum in chain regularity, although crystallization (see Question 4.2) is not desired! Explain. 4.4. Give the thermodynamic formula for the melting point and describe the symbols in this formula. Describe also the relation between these quantities and the chain properties. 4.5. Estimate the melting point of PVC (with a regular chain) when Tg = 85 C. Would a stereospecific PVC be a useful thermoplastic? 4.6. For the melting points of some members of the alkane-range (Cn H2n+2 ), unbranched, the following values were found : n = 8: 56.8 n = 10: 29.7 n = 15: 10.0 n = 20: 36.66 n = 32: 69.7

Estimate from these values the melting point of PE with the formula: b 1 =a+M Tm Discuss the question whether in practice the melting point of PE will be dependent on the chain length. 4.7. What is the reason for the fact that polymers cannot be entirely in a crystalline state? Give an example where, exceptionally, a level of almost entire crystallisation is achieved. How is this being realised? 4.8. Two samples of PP from the same batch have melting points of 161 C and 165 C respectively. What is the reason for this difference? 4.9. How would the width of the melting-range differ for the two samples from 4.8? 4.10. The melting points of PETB and PBTB are 260 C and 210 C respectively. What causes this difference? Why is PBTB preferred to PETB for a number of

16

Problems polymers

applications like injection moulding products? 4.11. How will it be possible to make semi-crystalline products anyway , in injection moulding products, from PETP? 4.12. In which processes can PETP be brought to rapid crystallisation in another way? 4.13. With respect to problem 4.12 which analogy can be found with rubbers? 4. 14. What is the critical nucleus size rc? Which, counteracting, phenomena determine r c? Which factor in the formula for rc is the most important? In what way is this property present in the formula? 4.15. The maximum crystallisation speed vcr for a polymer with M = 87,000 appeared to be 98 m/min. and with M = 143,000 it is 23 m/min. Explain the difference in a quantitave way. 4. 16. Polyamides (nylons) should crystallize. Why? As an exception, also amorphous polyamides exist. What are the requirements for the chain structure to achieve this? 4.17. Why do the values of Tg and Tm of polyamides (PA-s) decrease with increasing n?Why does the curve of Tm as a function of n show a zig-zag form? 4.18. A non-crystallizing polymer is being : A. Slowly cooled down from T1 = Tg + 50 K to T2 = Tg 50 K, B. Subsequently heated slowly to T1 . C. Subsequently cooled down rapidly to T2 . D. Held at T2 for a long time. - Show in a schematic diagram how the volumes changes as a function of T for A, B, C and D. - Do the same for a crystalllizable polymer, for which now T1 = Tm + 50 K and T2 = Tm 50 K. 4.19. Polyethylene-terephtalate is being treated from the melt in the following manner: A. Cooled down rapidly to room temperature: It appears to be amorphous now and to stay amorphous for an unlimited period of time. B. Raised to a higher temperature, e.g. 80 C and kept at this value for a longer period of time; it now crystallizes partly. C. Cooled down to20 C, D. Heated above the melting point.

Crystalline polymers

17

a. b. c. d.

What conclusion can be drawn from the observations about the value of Tg ? Estimate in which temperature range Tm will be. Explain the observations under B. Explain why at a certain temperature the crystallization process proceeds at a maximum rate. What is the postion of this temperature compared to the characteristic phase-transistion temperatures of the polymer? e. Draw in a volume-temperature diagram the volume change during A, B,C and D. f. Draw in a log E-temperature diagram the temperature dependency of E for the situations reached after A and C respectively. g. Give several technological consequences of the behaviour of PETP as mentioned above. 4.20. The dimensions of the orthorombic unit cell of crystalline PE are 0.740, 0.493 and 0.253 nm. Calculate the density of crystalline PE from these data. 4.21. Explain the formula for the crystalline volume fraction:

a =
c

What is the formula for the crystalline massfraction? 4.22. A quantity of 0.1 gramme of PP (melting point 165 C) is being melted in a dilatometer and subsequently cooled down rapidly to 150 C. At 150 C the following data were found for the volume as a function a time: time (sec) vol (mm3 ) 0 120.00 500 119.16 1000 114.97 1500 110.94 2000 110.04 2500 110.00

a. Investigate if this behaviour can be described with the Avrami-equation: V = (V) [1 exp(ct)], and determine the values of c and . b. Which conclusions can be drawn from the value of ? c. How would the values of c and have changed if the experiment would have been performed at 155 C? d. Determine the degree of crystallization in the end situation, assuming that crystalline PP has a density of 937 kg/m3 at 150 C. 4.23. Calculate the crystalline volume fractions for a LDPE with = 0.91 g/cm3 and for a HDPE with = 0.965 g/cm3. ( am = 0.855, cr = 1.00 g/cm3 ) 4.24. From the answer to problem 4.23 it appears that HDPE has a crystalline volume fraction twice as large as the crystalline volume fraction of LDPE. From this one could expect that the stiffness of HDPE would be twice as high as LDPE(the

18

Problems polymers

stiffness of the amorphous fraction, in the rubbery state is negligibly small). However the E-modulus of HDPE appears to be 10 times the value of the E-modulus of LDPE. What is the cause of this difference? 4.25. Why is a semi-crystalline polymer not transparant in most cases? Under which circumstances can it be transparant? 4.26. Why is a PE with a molar mass as high as possible used for the production of an extremely strong PE fibre? 4.27. When is the statement valid that semi-crystalline polymers show more creep than amorphous ones and when is it not valid? 4.28. What is the difference between lyotropic and thermotropic LCP-s (liquid crystalline polymers). In which cases is it essential to use either one type or the other? 4.29. To what extent can the three characteristic temperatures of a polymer Tg, Tm and T v be regarded as material constants? If this is not the case, what can cause changes to these temperatures? 4.30. Give a kind of polymers in which all three temperatures Tg, T m and Tv are important. 4.31. Give two sorts of polymers in which only two transitions occur. 4.32. Give two sorts of polymers in which only one transition occurs. 4.33. Why is a semi-crystalline polymer impact resistant between Tg and Tm?

19

5
Rubbery and liquid phases
5.1. The free energy of a stretched polymer bar is given by F = U TS. Explain the symbols used. 5.2. In what way will F change if the bar is being stretched? Describe the way dF/dl looks for a polymer in the glassy state and for a polymer in the rubbery phase. Explain the difference. 5.3. Why is dS/dl independent of the temperature? 5.4. A polyisoprene rubber IR is being vulcanised with 0.5 % sulphur (S = 32, C = 12, H = 1 g/mol). All sulphur is being used and the cross-links contain 8 S-atoms each. a. What is the average number of monomer units present between two cross-links. b. Show that the E -modulus at 25 C should have the value of 264 kPa if E is calculated from the formula E = 3RT/Mc ( = 0.91 g/cm3 , R = 8.31 J/molK, be aware of the units!) c. If the real E -value of the vulcanised rubber appears to be 1.5 MPa, how can the difference with the value calculated under b be explained? d. Give some general argumentation for the position of T and Mc in the formula. 5.5 . A nonvulcanised butadiene rubber shows at 20 C an E-value of 2.2 MPa in a fast stretching experiment. Its density = 0.91 g/cm3 . a. Calculate MC from the formula E = 3 RT/Mc ( R = 8.314 J/molK). What is the meaning of MC in this case? b. What would the value of E , according to the formula, be at 100 C? Why is the real E-value found significantly lower, even lower than the value at 20 C, than the calculated one? 5.6. Draw the log E vs T curve for : a. a nonvulcanised rubber, b. for the same rubber, but now lightly vulcanised. c. for the same rubber, but now strongly crosslinked. d. Discuss the differences in Tg in the three cases above. e. Discuss the differences in the E-level above Tg in the three cases above. f. What is the shape of the curve above Tg for case b and c?

20

Problems polymers

5.7. A small rubber balloon (ideal vulcanisate, ideal rubber behaviour, without leaks) is blown up and put in store for a long time at -20 C. The diameter is now 20 cm. Subsequently it is brought to a room where the temperature is 25 C. What will happen to the diameter? 5.8. Why is the flow temperature not a real, for a certain polymer characteristic, transition temperature from the rubbery state to the fluid state? 5.9. In figure 5.1 the apparent viscosity, , has been plotted against the shear stress , both on a logarithmic scale.
Pas 105 104 103 102 0,1 Figure 5.1. Figure for problem 5.9. 1 10 100 1000 (Pa)

a. Draw schematically the shape of the curve for a comparable polymer with a molar mass twice as large as the original polymer. Which average is the most important property? What is de range of shift of the curve? b. Do the same for a polymer type with the same molar mass as in the original curve, but with a broader molar mass distribution. c. The so-called Power Law is often used to describe Non-Newtonian behaviour: = . k()n . On which part of the original curve in Figure 5.1 can this approximation be applied and what will be the slope of that part of the curve for n = 0.5? d. Give some consequences of the width of the molar mass distribution for processing. Distinguish between various type of processes. . 5.10. Why is plotting of log versus log more useful than log versus log ? 5.11. a. What restriction has the value of the melt-index in predicting the processing behaviour of a polymer? b. In what way can the conditions in the determination of the melt-index be altered in order to obtain a better prediction? c. What does the ratio of the two melt-indices, measured as above, say about the chain structure of the polymer?

Rubbery and liquid phases

21

5.12. In Figure 5.2 published values of the impact strength are plotted against the melt-index for a number of PP-s (homopolymers). There is a clear trend of decrease of impact strength with increasing melt-index (not taking into account two types that have, as reported, a narrower molar mass distribution. Explain these data.
4 IZOD

1 1 Figure 5.2. Figure for problem 5.12. 10

m.i. 100

5.13. The same source as problem 5.12 gives the bending modulus as a function of the melt-index (see Figure 5.3). Here also a trend of increasing E with increasing melt-index can be observed, with the two types of narrower molar mass distribution showing a lower stiffnes. Explain these observations.
E MPa 2500

2000

1500

1000 1 Figure 5.3. Figure for problem 5.13 10

m.i. 100

5.14. Can these trends, that is to say a changing impact strenght and stiffness with changing molar mass, also be expected with PE? 5.15. Give four effects of the elastic behaviour of fluid polymers on the processing

22

Problems polymers

behaviour and the properties of the end product.

23

6
Visco-elasticity
6.1. To what extent can a Maxwell-element (spring and dashpot in series) be used as a model for the real behaviour of a polymer? 6.2. A fluid polymer behaves like a Maxwell-element: a spring with E = 105 Pa in series with a viscosity = 107 Pas. At t = 0 a constant stress of 104 Pa is applied to the material during a period of 500 s. From t = 500 s the deformation is kept constant, while the stress relaxation is recorded as a function of time. a. What is the value of the deformation immediately after applying the stress at t = 0? b. What is the value of the deformation at t = 500 s? c. What is the value of the stress at t = 800 s? d. Draw the curves of and as a function of time between t = 0 and t = 800 s. e. To what extent will a real polymer deviate from this behaviour? 6.3. To what extent can a Kelvin-Voigt element (spring and dashpot parallel) be used as a model for the real behaviour of a polymer? 6.4. A force of 600 N is applied in the length direction of a small plastic bar with a length of 10 cm, a width of 1 cm and a thickness of 1 mm. Under the influence of this force the bar shows an elongation as a function of time as given in figure 6.1 a. Which spring-dashpot model should be able to describe this behaviour? b Determine or estimate the values of parameters for this particular model (E moduli and viscosity). c. Draw the elongation as a function of time when the force is withdrawn after 250 seconds. d. Give a number of reasons why a polymer in reality will not behave like this pattern. 6.5. A polymer behaves like a Kelvin-Voigt element (spring E 1 = 3108 Pa parallel to viscosity = 31014 Pas) connected in serie with a spring E2= 3109 Pa. At t = 0 a stress = 6106 Pa is applied to the system; After 3.106 seconds the stress is removed. Draw the curve of the deformation as a function of time and calculate the dseformation (%)

24

Problems polymers

mm 10

0 0 100 200 300 400 500 600 sec Figure 6.1. Figure for problem 6.4.

a. Just after applying the stress. b. Just before the removal of the stress. c. Just after the removal of the stress. d. 3.106 seconds after the removal of the stress. 6.6. In problem 6.5 it was presumed that after the removal of the stress the strain as a function of time could be described with the formula = 0 exp( t/). Prove that this is true based on the superposition principle. 6.7. A force of 10,000 N is applied to a polymer bar with a length of 20 cm and a cross section of 1 cm2. After 1.5 hour the length has increased to 25 cm. From the moment the force has been removed the strain can be described by l = 4.exp(-2t) cm, with t expressed in hours from the moment of the withdrawal of the force. Develop a spring-dashpot model that will represent this behaviour. 6.8. The same problem as given in 6.7, but now the strain is described by l = 4.exp(-2t) + 0.5 cm. 6.9. A stretching experiment with constant rate of strain is applied to a Maxwellelement (spring E with dashpot in series). What will be the shape of the stress strain diagram? 6.10. The same question as posed in 6.9 , but now for a Kelvin-Voigt element (spring E with dashpot parallel).

Visco-elasticity

25

6.11. A visco-elastic system is subjected to an instantaneous deformation of 100 %. The stress, necessary for this deformation, is 2 MPa. After the deformation the stress decreases as a function of time as displayed in the table below. t (sec) 0.01 (MPa) 1.99 t (sec) 200 (MPa) 0.98 0.10 1.90 500 0.95 0.20 1.82 1,000 0.90 0.50 1.61 1 2 5 10 20 50 1.37 1.14 1.01 1.00 1.00 1.00 5,000 0.61 10,000 20,000 0.37 0.14 50,000 0.00 100 0.99

2,000 0.82

Give a model for this behaviour. 6.12. The modulus of elasticity of a polymer is often represented by a complex quantity: E* = E1 + iE2. a. What is the meaning of E1 and E2 in this formula? b. In what way can these quantities be measured, e.g. around 1 Hz? c. In what way does the dampingfactor, tan , depend on these quantities? d. Give a number of cases in which tan has a practical technological meaning. 6.13. In figure 6.2 the damping factor is plotted against the temperature for two situations (p and q). a. p and q are valid for which two types of polymers? b. Where is the first maximum for both of these two curves situated? c. Which value will the curve of p reach with a further increase of temperature?. b. Explain why the curve of situation q does not approach to zero, but to a finite value.
tan p

T Figure 6.2. Figure for problem 6.4.

26

7
Mechanical properties
7.1. A polymer bar, width 1 cm, thickness 1 mm and length 10 cm is being elongated by 2 mm. The necessary force is 600 N. What is the value of the modulus of elasticity (E)? 7.2. In general a value for the modulus of elasticity (E) of 3 GPa is found for hard , amorphous polymers in the glassy state. Why is the value for E for polycarbonate only 2.1 GPa? 7.3. Why do semi-crystalline polymers differ considerably in stiffness? Examples : PEEK; E = 4 GPa, PP: E = 1.3 GPa, LDPE even down to 0.15 Gpa. 7.4. In which two manners can the stiffness of a polymer be increased? 7.5. We have two equally formed bars; the first one is ordinary PP, the other one contains a small amount of copolymer (to increase the impact strength). What is the most simple way to determine the material which the bars are made of? How can the amount of dispersed rubber be estimated easily? 7.6. Why are fibres stiffer than the polymers that they are made of? What order of magnitude can the stiffness increase be? Give a number of extreme examples. 7.7. Why is it, practically spoken, not possible to say that a stiffer polymer chain leads to a higher stiffness of the polymer? Explain why extremely stiff chains , like aromatic polyamides, result in a high stiffness? 7.8. Why is it not possible to predict the creep of an amorphous glass-like polymer after a long time by shifting creep curves at higher temperatures along the log t - axis as suggested by the principle of time-temperature equivalenc? 7.9. Draw schematically a number of creep isochrones of : a. A polymer with a high short-term modulus of elasticity and much creep. b. A polymer with the same high short-term modulus of elasticity but little creep. c. A polymer with a short-term modulus of elasticity three times lower as in a and b and much creep. For which kinds of thermoplastic polymers can the cases of a, b and c be valid?

Mechanical properties

27

7.10. The creep formula of Kohlrausch: D(t) = D0 exp(t/t0)1/3 appears to describe the creep behaviour surprisingly well for a large number of materials, including polymers. What is the meaning of D0 in the formula? Why can this formula not be applied for the practical behaviour of polymers? 7.11. A glass-like polymer can show physical ageing. a. What is the real cause of this phenomenon? b. What is the influence on the creep behaviour? c. Draw bundles of isochrones with and without physical ageing. 7.12. Figure 7.1 shows published creep isochrones of Noryl (PPE/PS) and PBTP, both at 23 C. Explain the difference in shape of both isochrone bundles.
MPa 20 15 10 5 0 PPO/PS (Noryl) 23 C
1 10 102 103 104

MPa 20 15 10 5

10

102 103 104

hours

hours PBTP 23 C 0.5 1.0 % 1.5

0.5 1.0 % 1.5 0

Figure 7.1. Figure for problem 7.12.

7.13. In figure 7.2 creep isochrones are displayed at five different temperatures; Each temperature has its own vertical axis for the stress. a. What will be the deformation at 20 C under a stress of 8 MPa after 100 hours? b. What will be the deformation at 70 C under a stress of 5 MPa after 10 hours? c. What is the value of the short-term modulus of elasticity at 20 C? d. What is the value of the short-term modulus of elasticity at 80 C? e. What stress is allowed at 20 C, if the deformation after 1 year must not exceed 2%? f. What stress is allowed at 100 C, if the deformation after 1000 hours must not exceed 3%?

28

Problems polymers

18
5 6 10 7 8

MPa

1 102 hours 102

16 14

5 6 8

12 10

104

4 3

5 6 4

3 2 2 2 1 1 1 3

8 6 4 2 PP

4 3 2 1

20 C 1 2 3 4%

100

80

60

40

Figure 7.2. Figure for problem 7.13.

7.14. Breaking stress experiments on PB (poly-1-butene) pipes gave the following breaking times tb at various wall stresses and temperatures T:

(MPa) 189 183 T (C) 20 20 tb (hours) 0.02 1.1 T tb

180 175 163 157 151 135 128 123 119 107 105 20 20 40 40 40 60 60 60 60 80 80 12.9 850 0.15 40 3000 0.3 95 2000 8100 0.2 3

100 80 80 77 72 60 43 58 53 42 26.7 20 80 80 100 100 100 100 100 110 110 110 110 110 600 12000 0.01 20 1100 3800 9800 0.03 20 2000 5000 7500

Estimate from these data the safe wall stress for a period of use of 10 years at 80 C. 7.15. a . Why is the safe wall stress, as determined in 7.14, not safe enough for certain applications? b . Which information is needed in order to construct in such a way that never an irreversible material damage can occur? 7.16. Why is a standard impact test often performed with notched test bars?

Mechanical properties

29

7.17. a . How can a polymer with a low impact strenght as obtained from a test described in 7.16 be improved? b. Which mechanisms are involved in this improvement? c. To what property must be given in with this improvement? d. Give examples of this principle. 7.18. Why can the impact strenght of a polymer not be considered as a material constant? 7.19. a . Why is there a rather good correlation between the Shore-D hardness of polymers and their modulus of elasticity? b. Why can this correlation not be perfect? c. Why are polymers (like TPE-s) often characterised by their hardness (Shore-D) instead of their modulus of elasticity? 7.20. Why is the damping of of a vulcanised rubber lower with a lower Tg ? 7.21. By which factors is the tan of a rubber vulcanisate, far above Tg , controlled?

30

8
Further properties
8.1. Where would the brittleness temperature approximately lie for: - an amorphous thermoplastic? - a semi-crystalline thermoplastic? - a rubber-modified thermoplastic? 8.2. Why is the softening behaviour of an amorphous thermoplast different of that of a semi-crystalline thermoplast? 8. 3 . Look at the data for Vicat B and ISO/A (Chapter 10). Assume that the characteristic E-moduli for these determinations are 200 MPa and 1000 MPa respectively. Try to verify these values for PBTB, PTFE, PP, PMMA, PA 6.6, POM and HDPE from the E(T) curves (see end of chapter 10). What are the reasons for the differences found? 8.4. Two different softening temperatures of two different polymers may show an apparent contradiction, like (C): ISO/A 95 60 Vicat B 100 180

ABS PBTP

a. What causes this contradiction? b. Which softening temperature should be used in practice? 8.5. Manufacturers of semi-crystalline polymers sometimes claim that the addition of short glassfibres not only causes an increase in stiffness, but also cause a considerable rise in the softening temperature. Is this an extraordinary achievement or is it just in the nature of things? 8.6 . What are the disadvantages of the relatively high expansion coefficient of polymers? 8.7 . Give both advantages and disadvantages of the low thermal conduction of polymers. 8.8. The heat of conduction of a polymer foam is composed of four components. a. Which are these components? b. Point out the way changes with the cell size at constant density.

Further properties

31

c. Point out the way changes with the foam density. d. Point out the way changes with the type of gas. e. Which factor determines the maximum insulation value possible for a foam? 8.9. Why does a high softening temperature of a polymer not guarantee sufficiently that the polymer can be used in practice at high temperatures? 8.10. a. What is understood by stress corrosion? b. Why is this name in general not correct c. When is this name correct? d. Give a number of examples. 8.11. For what reason would one lower the excellent high electrical resistance? In what manners can this be done? 8.12. Which property influences the tendency to electrostatic charge? How can this tendency be surpressed? What is the importance of the time scale in this respect? 8.13 . What are the advantages and disadvantages of the extremely low dielectric losses of PE? 8.14. a. Why are polymers not or hardly suitable for fine optical applications? b . Polymers are frequently used in the manufacture of spectacles. What are the advantages and disadvantages in this application? 8.15 . Why would one choose a PC with a molar mass as low as possible for the manufacturing of compact discs? 8. 1 6 . The nitrogen-permeabilities of HDPE, PP and IIR are about equal. Nevertheless IIR is considered as very gastight, while PP and PE are considered as rather permeable. Explain this. 8.17. Why is the permeability of LDPE higher than of HDPE?

32

9
Polymeric compounds and composites
9.1. The Flory-Huggins formula is: 1 2 Gm V = kT [V1 ln 1 + V2 ln 2 + Vs 12] Show from this formula that the miscibility of two polymers decreases with increasing molar masses. 9.2. Which role does the interaction parameter play? 9.3. Simplify the Flory-Huggins equation for a 50/50 (volume) mixture of macromolecules of equal size and derive for this case a simple condition for the sign of the free enthalpy of mixing. 9.4. Show from the formula that, independent of the parameters, there will always be values for 1 and 2 for which the free energy of mixing is negative. 9.5. If after mixing in the melt a homogeneous mixture is obtained, it can segregate during cooling to the solid state. When will this be the case? 9.6. The reverse: One has mixed at a not too high temperature and obtains, after cooling, particles consisting of a homogeneous mixture. During processing in an injection moulding process segregation occurs resulting in an endproduct being a dispersion. Explain when this can happen. 9.7. Why does the postion of the G- curve not always guarantee, that there will indeed result a homogeneous mixture from a mixing process? 9.8 . How can it simply be explained that segregation in a block copolymer will become more perfect with increasing length of the blocks? 9.9. Which properties govern the mimimum droplet size in a mixing process? 9.10. Why do we find a much larger droplet size in real dispersions? 9.11 . Which two processes contribute to the reduction of the droplet size in a mixing process?

Polymeric compounds and composites

33

9.12. Why is it not possible to obtain the properties of a dispersion from a rule of mixing following from a serial or parallel arrangement? 9.13. Referring to problem 4.24 we look at the bending modulus E of a polyethene with density 0.935. The degree of crystallization, (volume fraction) can be calculated with : d dam =d d
cr

am

with dcr =1.00 and dam = 0.855 (see maintext 4.4.1) and so = 0.552. The Emodulus is 660 Mpa. Can we determine the morphology of PE from these data (either amorphous dispersed in crystalline,the reverse or co-continuous)? Use the formulae of Kerner and Nielsen ( 9.1.6 of the main text). 9.14. A three block copolymer SBS is built up a polybutadiene middle section of 2000 butadiene-units -(CH2-CH=CH-CH2 )- with two end sections, each consisting of 150 styrene units -(CH2-CHC6 H5 ))-. Complete segregation occurs; the PS domains are spherically shaped with a diameter of 25 nm. a. What is the mass fraction of polystyrene? b. How many PS chains are present in a domain ( = 1.05 g/cm3 and NA = 61023? c. Why does SBS lose its properties already around 10 C below the glasspoint of PS? d. Estimate with the Kerner-formula the modulus of elasticity of SBS, assuming that the E-moduli of polybutadiene and polystyrene are 3 MPa and 3 GPa respectively and the density of BR = 0.91 g/cm3. 9.15. Three foams of a glasslike thermoplastic polymer, each with a density of = 0.5 g/cm3 (density of the solid polymer = 1 g/cm3 , E modulus is 3200 Mpa) show E-moduli of 1100, 300 and 100 MPa respectively. Which conclusions can be drawn from these data about about the structure of these foams? 9.16. Show that the stiffness of a heavy foam (little foaming) is, roughly estimated, proportional to its density. 9.17. Why is the expression reinforcing fillers in general not really correct? How should these be called in a correct way? 9.18. Why is carbon black in particular a reinforcing filler and why is a technical SBR-vulcanisate not possible without a carbon black reinforcement and can a natural rubber vulcanisate do without? 9.19. Give a number of cases in which it is very desirable or even essential for the

34

Problems polymers

properties of a heterogenous mixture that the smallest phase is the continuous phase. 9.20. Two polymers A and B are mixed in two mixing ratios to M1 and M2. For the ISO heat deflection temperatures of A, M1, M2 and B values of 50, 80, 110 and 140 C are found respectively. The Vicat softening temperature however is more or less the same for all four materials, all values are around 145 C. Give a schematic drawing of log E versus T for all four materials. In what respect do A and B differ? 9.21. Now two polymers C and D are mixed in two mixing ratios to M3 and M4. For the ISO heat deflection temperatures of C, M3, M4 and D values 90, 130, 170 and 210 C are found. The Vicat softening temperatures are 95, 135, 180 and 220 C. Give also a schematic drawing of log E versus T for these four materials. What is the nature of the polymers C and D and what can be said about their miscibility? 9.22. The addition of glass fibres to a polymer increases the impact strenght of certain polymers (the brittle ones), but decreases the impact strenght of others (the tough ones) (see main text Figure 9.1.7). How can these contradicting observations be explained? 9.23. It is desired to enhance the impact strength of a glassy thermoplastic (E = 3 GPa, p = 1.4 g/cm3 ) by adding a dispersed rubbery phase (E = 2 MPa, r = 0.9 g/cm3 ). However the stiffness should not be reduced by more than 20% as a result of this addition. How many %(weight) rubber may be added? The Poisson number of the matrix is = 0.33. 9.24. The mixture made in problem 9.23 should be returned to its original stiffness with glassfibres (E = 75 GPa, g = 2.4 g/cm3 ). How many %(weight) of glassfibres must be added? 9.25. Why is the velocity of sound in a composite important? 9.26. In a certain composite glass fibres are dispersed in a parallel alignment in the polymer. The volume fraction of the fibres is 0.25. Imagine a number of cases in which the lenght of the fibres increases more and more. Calculate (without using the Formula of Kerner) to which limit the E- modulus of the composite will approach in the direction of the glassfibres. (the E -moduli of the polymer and the glass are 3 and 75 GPa respectively). 9.27. Estimate the value of the E -modulus in the direction perpendicular to the direction of the glass fibres with the aid of the Einstein formula. Why is it allowed to use this formula here? 9.28. Estimate, without too detailed calculations, the E -modulus of a similar composite in which the glass fibres are now dispersed at random.

Polymeric compounds and composites

35

9.29. To a semi-crystalline thermoplastic short glass fibres are added in one case and in another case rubber particles are added. In what way and why does each of the following properties alter for both cases? a. Transition temperature from glassy tot rubbery phase b. Vicat softening temperature. c. Melting point. d. ISO-HDT. e. Modulus of elasticity. f. Tensile strength. g. Conduction of heat. h. Impact strength.

36

11
Principles of processing
11.1. Mention four different principles to heat-up a polymer in a processing operation. Give an example of each of these. Which material properties are important for each of these principles? 11.2. Mention a few examples of thermosets and thermoplastics where casting is a possible shaping technique. 11.3. Which processing techniques are not- or less suited for semi-crystalline polymers? Why? 11.4. Why is compression moulding for rubbers and thermosets more attractive than for thermoplastics? 11.5. a. Give two different routes to manufacture PS (or TPS) cups from the raw material. b. How can these two cups be distinguished in a non-destructive way? c. How does their behaviour difference upon heating? d. Which factors govern the choice of the processing methods? 11.6. Figuur 11.1 gives, schematically, two "short-shot" lines for an injection moulding process.
T A B

p Figure 11.1. Figure to problem 11.6.

a. Explain the shape of the curves. b. If A and B hold for the same injection moulding machine and the same polymer, but for different moulds, in which respects do the moulds then differ?

Principles of processing

37

c. If A and B now hold for the same mould, but for different grades of the same polymer, which conclusion could, at first sight, be drawn about the difference between A and B? d. If the conclusion from the above question would appear to be wrong, what would then be the following conclusion about the difference between A and B? e. What can happen when, at a certain temperature, the pressure p is being set at a much higher value f. The same when, at a certain pressure, the temperature T is strongly increased? g. The same when T is strongly decreased? 11.7. In an injection moulding process an HDPE sheet of 10 cm 10 cm 3 mm is being produced. The injection moulding pressure is 800 bar, the melt temperature is 220 C. a. What is the minimum required clamping force on the mould? b. How many grammes of the polymer are present in the mould, immediately after filling? c. What would the dimensions of the product be if the filling gate would be closed at this moment? d. How can it be prevented that the end-product is too small? The compressibility of molten HDPE is 7.5 105 /bar. The volume-temperature relation at atmospheric pressure is given in figure 11.2.
1.4 specific volume

1.3

1.2

1.1 HDPE 1.0

0.9 0 50 100 150 200

T C 250

Figure 11.2. Figure to problem 11.7.

11.8. Why is the mould shrinkage for PE, PP and PA greater than for PS, PVC and

38

Problems polymers

PC? 11.9. Why is the density in the interior of a casted, compression-moulded, or injection-moulded block smaller than at the outside? 11.10. Describe the competition between flow and curing of powder resins, used for e.g. bicycle frames, 11.11. Why exists the risk in a calender process of uneven thickness over the width of the sheet? What can be done to prevent this? 11.12. Mention some processing techniques where the elastic behaviour of the molten polymer plays an essential part. Why is the word molten not appropriate? 11.13. In which two respects is a broad molar mass distribution, e.g. with PE, of importance when making shrink films? 11.14. In which two respects is a narrow MMD, e.g. with PE, important in the production of an oil tank in a rotational moulding process? 11.15. When a wire coating is extruded around a copper wire, a high throughput is required. This involves the risk of melt fracture, resulting in a rough surface. How can melt fracture be avoided? 11.16. Mention a few cases where the meeting of two streams of molten polymer may result in weak spots in the end-product. 11.17. What is meant by an "adiabatic" extruder? Why is this to be recommended? 11.18. Why can an extruder be used as a mixer? 11.19. The figure gives, for a certain combination of an extruder, a die and a polymer, two lines for Q (throughput) as a function of the pressure p, resp. Q = aN bp/ and Q = cp/. a. explain the symbols in these equations. b. What is, at the given combination of extruder screw and polymer, independent of the die, the maximum throughput and the maximum pressure? When are these maxima being reached? c. How do the throughput and the pressure change when the rotational speed of the screw is halved? d. Estimate the rate of throughput and the pressure when the die has a twice as large flow resistance. e. The temperature is lowered to such an extent that the viscosity becomes three times higher. How does the rate of throughput and the pressure change?

Principles of processing

39

Illustrate the points b. to e. with extra lines in the diagram. f. Why are the lines in the diagram not straight lines in reality? Sketch (only for the die) a more realistic curve.
Q (dm3/h) 80 70 60 50 40 30 20 10 0 0 100 200 300 p (bar)

Figure 11.3. Figure to roblem 11.18.

11.20. In an extruded pipe internal as well as frozen-in stresses may be present. a. Describe the difference and explain the origin of both. b. Give a simple detection method for each. c. What are the consequences of each of these kinds of stresses for the properties in practice? d. Why are internal stresses more pronounced with thick-walled pipes, and frozen-in stresses with thin-walled pipes? 11.21 . When making PET bottles, the first step is the injection moulding of an amorphous preform, which is later blown-up. a. Why is this pre-form amorphous? b. Why is this necessary for the further fabrication of the bottle? c. What happens more with the blowing-up? d. Why should the final bottle be (semi)-crystalline? e. Why does the material crystallise during the formation of the bottle? f. Why is the bottle beautifully transparant? g. What is the analogy with the manufacture of PETP (polyester) fibres? 11.22. How and why does, with extrusion, the die-swell change:

40

Problems polymers

a. b. c d. e.

with a polymer with a higher melt-flow-index m.f.i.? with a higher ratio of m.f.i.(10) / m.f.i.(2.16)? with a higher rate of throughput? at a higher melt temperature ? at a larger l/d ratio of the exit channel ?

11.23. Why does the extrusion of PVC require a longer compression zone than with PA? 11.24. What is a structural foam? Mention four advantages in its manufacturing and its use. 11.25. Which processing techniques could be applied to manufacture (from the raw material) of the following articles: a. 2-litre bottle of PE, b. ship propellor of PA, c. butter cup of PP, d. bottle crate of PP, e. floor tiles of PVC-compound, f. rubber ball, g. broad PVC film, h. tyre tread of SBR, i. waste bag of LDPE, j. packaging case of PS foam, k. rubber O-ring, l. rubber shoe-sole m. small rowing-boat of PVC, n. thick PMMA-sheets , o. wire-coating of PVC, p. sewer pipe of PVC, q. sewer pipe of epoxy/glass, r. 10 m3 vessel of PE, s. switch housing of UF, t. light armature of PMMA, u. toothrad of PA, v. thin-walled PP pipe of 3 m diameter, w. PU thermal insulation of a storage tank, x. rowing-boat of UP/glass, y. PET-bottle. z. thin flat sheets of PS foam?

41

Answers
1. Introduction
1.1. Vegetable: timber, cotton, cork, sisal, hemp, straw, reed, cane, capok, rattan, etc. Animal: wool, silk, fur, mohair, horse-hair, intestine, down, ivory, horn, sponge, beeswax etc. 1.2. With half-synthetic plastics the polymer chain has been formed in a living tissue, but it has been chemically modified afterwards. Cellophane is, after chemical modification, obtained from the cellulose in wood, just as paper (from cellulose and lignin), cellulose fibres (rayon), and cellulose plastics. Leather is made from animal hides in a tanning process. Natural rubber has to be vulcanised (with sulphur) to obtain a technically usable product. Ebonite is also made from natural rubber, but with a much higher amount of sulphur; it belongs, therefore, to the family of thermosets. Linen is obtained from flax after a chemical treatment (retting). The boundaries between half-synthetic and natural polymers are sometimes vague. 1.3. In a cracking process the long, saturated, hydrocarbon chains in the crude oil are transformed into shorter ones, which, for lack of hydrogen atoms, contain double bonds, (-C=C-), which provide polymerisability. 1.4. When a diene, C=C-C=C, is polymerised, only one of the two double bonds is used; an unsaturated chain results, -C-C=C-C-, which offers the possibility of vulcanisation with sulphur. 1.5.a. 1.4-polybutadiene, -CH2 -CH=CH-CH2 b. 1.2- of 3.4-polybutadiene, -CH-CH2-CH-CH2| | CH CH || || CH2 CH2 in which under a. two different structures, cis and trans can be formed. -C C\ / C=C C/ or C=C / -C

1.6. Ethylene-propylene copolymers, and also polyisobutylene, are rubbers with saturated chains, which cannot be vulcanised with sulphur. They are, therefore,

42

Problems Polymers

copolymerised with a small amount of diene: EP with a diene in a sidegroup (e.g. dicyclopentadiene), PIB with isoprene (IIR = isobutylene-isoprene rubber = butyl rubber). 1.7. a . Polycondensation is possible with saturated monomers, e.g. a diol with a bifunctional carbon acid, by ester formation on two sides. In this reaction a small molecule, e.g. water is also formed. b . After ring opening of saturated monomers these can also polymerise (e.g. caprolactam to PA-6). 1.8. a. By a reaction between tri-functional and bi-functional components, such as with the formation of phenol-formaldehyde or a polyurethane from glycerol and diisocyanate. b . By cross-linking polymer chains containing double bonds: with rubbers using small amounts of sulphur; with unsaturated polyesters using (poly)styrene. 1.9. Thermosets are cured into networks at elevated temperature from small molecules or from single chains. The name could suggest that thermosets become harder at temperature increase; on the contrary: they soften just as all polymers at their glass-rubber transition, though their stiffness remains much higher than that of a rubber. Moreover: curing is not always carried out at a high temperature; also cold-curing resins are used (e.g. in two-component adhesives or in large ship decks). 1.10. In thermoplasts the parts of different chains are bound to each other by weak secondary bonds, which are broken at higher temperatures. In thermosets and vulcanised rubbers the chains are permanently linked by primary chemical bonds, which, also at higher T, do not allow flow. 1.11. Thermosets have a tight network; with rubbers it is much looser; the crosslinks are just sufficient to connect all chains at a few points. 1.12. In thermoplastic rubbers the chains are not connected by primary bonds as is the case with sulphur-vulcanised rubbers, but by weaker secondary bonds (e.g. within a PS domain in SBS), which are loosened upon temperature increase). 1.13. Thermoplastic rubbers (TPEs) show rubbery behaviour at ambient temperature because the rubber phase plays the dominant part. In thermoplastics the behaviour of the glassy phase or of the crystalline part is dominant. 1.14. Thermoplasts: PE, PS, PVC, PP. Thermosets: PF, UP. Rubbers: SBR.

Answers

43

1.15. LDPE, HDPE and LLDPE. They differ in density: LDPE between 0.91 and 0.92, HDPE between 0.94 and 0.96, and LLDPE between 0.92 and 0.95 g/cm3. Since the density reflects the crystallinity, they also differ considerably in stiffness, between as well as within types. 1.16. thermoplasts: PS, TPS, SAN, ABS en ASA. Thermosets: UP (is cross-linked with styrene). Rubbers: SBR and SBS. 1.17. PTFE with its very high melting point (327 C) can only be transformed into an end product as a powder in a sintering process, since as a melt it would degrade at the high temperatures involved in other processing operations. Other fluor polymers such as FEP, PVDF and EFTE are suitable for the usual processing techniques. 1.18. PI, polyimide, can be used short-time up to 500 C. Related to PI are: polyetherimides, polyesterimides en polyamidimides. 1.19. a. a. Blends of polymers b. Particle- and short fibre- reinforced polymers c. Composites d. Foams 1.20. a. Carbon black is added to thermoplasts to decrease the electric resistance, in particular to prevent electrostatic charge, and also to shield the polymer from light, thus preventing ultraviolet degradation. Carbon black is added to rubbers to increase strength and abrasion resistance, b. Chalk is a reinforcing filler in thermoplasts and thermosets. c. Glass fibres are added to thermoplasts and thermosets to increase stiffness and strength, more than particles do. d. Plasticisers are added to a polymer to reduce the glass rubber transition temperature drastically (e.g. with PVC), so that the polymer behaves as a rubber at ambient temperature rather than as hard glassy thermoplast. e. Mica is a reinforcing filler which, due to its platelet structure, increases the stiffness more than spherical particles do. f. Accelerators are added to thermosets and rubbers to speed-up the curing or vulcanisation process to such an extent that an end product can be formed under optimal conditions. g. Lubricants are sometimes used to reduced the friction resistance in a shaping process. h. Sulphur is essential in the usual processing technology of rubbers; it brings about the formation of cross-links, which provide the necessary shape stability. i. Pentane is a blowing agent, applied in a.o. PS, to form foam.

44

Problems Polymers

j. Wood flour is a cheap filler, which brings about some stiffening (e.g. in PVC), and which reduces the price. k. Antioxidants can, in the first place, protect a polymer against degradation at the high temperatures during processing; in the second place they can (mostly different types) prevent chain rupture during long-term use at elevated temperatures. l. SiC is a very hard filler, mostly used to increase the abrasion resistance of a polymer. m. UV-stabilisers protect a polymer against chain break under the influence of the ultraviolet components of daylight, often in combination with pigments (e.g. with PP). n. Anti-static agents bring about a very thin layer of a conducting (hydrophilic) substance at the surface of the product, which allows electric charges, resulting from friction or contact, to be dissipated quickly. o. Rubber particles are blended into a brittle polymer to increase its impact strength, e.g. butadiene rubber in PS, etc.

2. Molecular composition
2.1. C-atoms only, or also other ones such as N, O, S or Si, saturated or unsaturated chains simple chains or chains containing rings, even composite rings (examples in MT 2.1.1) 2.2. a. PE -CH2 -CH2 BR -CH2 -CH=CH-CH2 polyacetylene -CH=CH-CH=CHPOM -CH2 -O-CH2-Ob. PP, PVC, PS, PB, PAN. c. PIB, PVDC, PVDF (see MT 2.1.2). 2.3. PE, -CH2 -CH2 -, monomer mass = 2C + 4H = 2 12 + 4 1 = 28 g/mole, PS, -CH2 -CH(C6 H5 )-, monomer mass = 8C + 8H = 8 12 +8 1 = 104 g/mole; PA-12, -(CH2 )11-CO-NH-, monomer mass = 12C + O + N + 23H = 12 12 + 1 16 + 1 14 + 23 1 = 197 g/mole. M(PE) = 500 28= 14,000 g/mole, M(PS) = 500 104 = 52,000 g/mole M(PA-12)= 500 197 = 98,500 g/mole. -C-C-C-C-C- chains lie zig-zag shaped, with C-C distances of 0.154 nm and with an angle between the bonds of 109.5. The contribution of a C-C link to the contour length is, therefore, 0.154 sin 54.2 = 0.126 nm. for PE the stretched length is thus 1000 0.126 nm = 126 nm (since the monomer

Answers

45

contains two C-atoms), for PS also 126 nm, for PA-12: 13 links (for simplicity taking the same length for CO-NH-), so 500 13 0.126 nm = 820 nm. The calculation is only valid for strictly linear PE, without branches. As to PS, it holds only for the syndio-tactic structure, since otherwise the neighbouring phenyl rings prevent a stretched conformation. (MT 2.5). 2.4. In this case equal numbers (both 1) are mixed. The number fractions are both 1 1 1/2 and the number average is 2 80,000 + 2 120,000 = 100,000. The weight fractions are, respectively, 80/200 and 120/200 or 0.4 and 0.6, so that the weight average is: 0.4 80,000 + 0.6 120,000 = 104,000. 2.5. ni = Ni /Ni = number of the species i divided by the total number = number fraction, wi = n i Mi /ni Mi = weight fraction, zi = w iMi /w i Mi = z-fractionMn = ni Mi , Mw = wiMi , Mz = zi Mi , ni = wi = zi = 1 (all normalised). 2.6. Mn = niMi ; ni = (wi /Mi)/(wi /Mi), Mn = ((wi/Mi))Mi )/(wi /Mi)) = wi /(wi /Mi)) = 1/(wi /Mi))

/ n ; Mw = (niM2 2.7. Mw = wiMi ; wi = (nMi )/(ni Mi ) = ni Mi M i )/Mn ; MwMn = niM2 i.

2 2.8. Mz = zi Mi ; zi = (wi Mi )/(wi Mi ) = wi MM i / w; Mz = (wiMii )/Mw; 2 MzMw = wiMi i .

2.9. a. In a mono-disperse polymer all chains are equally long (exception). Then Mn = Mw = Mz etc.. b. The degree of dispersion or heterogeneity index, D, is defined as Mw/Mn ,, and is, therefore, D = 1 for a mono-disperse polymer. 2.10. The chains break in a statistical way, by which shorter ones of various lengths are formed. The molar mass distribution is broadened in the direction of the lowmolecular side. 2.11. In the blend every (wi )A is replaced by: wA(wi )A and every (wi )B by wB(wi )B. So Mw of the blend is: Mw = (wA(wi )A(Mi )A) + (wB(wi )B(Mi )B) = wA(wi )A(Mi )A + wB(wi )B(Mi )B = wA(Mw)A + wB( Mw)B 2.12. The number of chains in one gramme of polymer is N /Mn (N = Avogadro number). One gramme of the mixture contains wA grammes of A, wB grammes of B and wC grammes of C, so that: N/Mn = wAN/(Mn )A + wBN/(Mn )B + wCN/(Mn )C

46

Problems Polymers

of 1/Mn = wA/(Mn )A + wB/(Mn )B + wC/(Mn )C. The Mn values of A, B and C are 10, 20 and 30 kg/mole respectively 10, 20 en 30 kg/mole; wA = wB = 10/(20 + x), wC = x/(20 + x), Mn = 15 kg/mole, so: 1/15 = 10/(20 + x )10 + 10/(20 + x)20 + x/(20 + x)30, which results in: x = 5.
2 2.13. Mz = wiM2 i /wi Mi of wi M i /Mw. Further evaluation is analogous to 2.11 and results in: MzMw = w1 (Mz)1 (Mw)1 + w2(Mz)2 (Mw)2 . So: Mz = 154.5 kg/mole.

2.14. 0.382 = k (76,000)a; 0.91 = k(250,000)a; 2.21 = k(850,000)a; we can plot log[ ] versus log M to determine the slope a ; we can also consider ratios of []: 0.91/0.382 = (250/76)a; a = 0.729; 2.21/0.91 = (850/250)a; a = 0.725; 2.21/0.382 = (850/76)a; a = 0.727; on average a = 0.727. With 0.91 = k250,0000,727 we find: k = 10.8105. 2.15. w A = 0.37; w B = 0.18; w C = 0.45. Mw (in kg/mole) = 0.3776 + 0.18250 + 0.45850 = 455.62 = 455,620 g/mole. 1/Mn = 0.37/76 + 0.18/250 + 0.45/850; Mn = 163.457 = 163,457 g/mole. 2.16. c 0 0.1 0.2 0.3 0.4 t 95.0 107.7 121.4 136.3 152.8 t to 0 12.7 26.4 41.3 57.8 (t to )/to 0.1337 0.2779 0.4347 0.6084 (t to )toc = red 1.337 1.390 1.449 1.521

When we plot red against c, it is clear that extrapolation of red to leads to 1.30 dl/g, so [ ] = 1.30 dl/g. From Mark-Houwink it follows (see 2.14): 1.30 = 10.8105M0.727, from which Mv = 410,000 g/mole, indeed slightly below Mw.
1.55 1.50 1.45 1.40 1.35 1.30 1.25 0 0.1 0.2 0.3 0.4 c (g/dl)

red (dl/g)

Figure A.1. Figure to the answer of problem 2.16.

Answers

47

1/a 2.17. With Mv = (wiMa i) : Mv = (0.3776,0000,727 + 0.18250,0000,727 + 0.45850,0000,727)1/0,727 = 409,340, close enough to the answer on 2.16 .

2.18. We first calculate the various averages: w1 = 0.10, w2 = 0.18, w3 = 0.25, w4 = 0.17, w5 = 0.12, w6 = 0.08, w7 = 0.06, w8 = 0.04. M1 = 15, M2 = 27, M3 = 39, M4 = 56, M5 = 78, M6 = 104, M7 = 120, M8 = 153. Mn = 1/(wi /Mi)) = 1/(0.10/15 + 0.18/27 + ) = 38.7 kg/mole. Mw = (wiMi ) = 0.1015 + 0.1827 + = 56.6 kg/mole. 2 2 Mz = (wiM2 i )/Mw = (0.1015 + 0.1827 + )/56.6 = 78.8 kg/mole. 3 3 Mz+1 = (wiMi )/MzMw = (0.1015 + 0.18273 + )/(78.856.6) = 99.2 kg/mole a. The melt viscosity (zero-shear) is governed by Mw, so as far as this is concerned the blend resembles fraction 4 . b. The blend has a broader molar mass distribution; it behaves, therefore, stronger non- Newtonian (see MT 5.3.2 ) and flows better at high rates of shear. c. The impact strength is mainly dependent on Mn , so the blend resembles fraction 3. d. The die-swell is a consequence of melt-elasticity which is governed by higher averages such as Mz and Mz+1 ; here the fractions 5 and 6 are of importance. e. The Youngs modulus in the glassy state is independent of M. 2.19. In kg/mole: M1 = 60, M2 = 600,Mn =1/(wi /M1) = 1/((1 w2 )/60 + w2 /600) = 100, so w2 = 4/9 and w1 = 5/9; so 40 kg are needed.Mw = (5/9)60 + (4/9)600 = 300 kg/mole = 300.000 g/mole. Mz = [(5/9)602 + (4/9)6002 )]/300 = 540 kg/mole = 540.000 g/mole.
1/2 = 73.5 nm then the 2.20. If we consider the coil as a sphere with a radius r = (r 2 o ) 3 21 3 volume of such a sphere is V = (4/3)r = 1.66310 m . The mass m of the chain is 1000/61023 = 1.6671021 kg. The density, , is then m/V = 1.00 kg/m3 . When we compare this with the density of PS, 1050 kg/m3, then the coil is 1050 times diluted 2 2.21. With M = 10,000 r 2 o = nb is 100 times smaller; so the radius of the spherical coil is 10 times smaller and the volume V is 1000 times smaller. The chain mass is 100 times smaller, so the density, r, is 10 times greater. The dilution is then 105 times.

2.22. In a tightest array of spheres the volume fraction is about 0.65; if we consider a maximum volume fraction of 0.3 to exclude all interaction effects, we find in the first case (M = 10 6 ) a maximum concentration of 0.3/1050 = 0.29 g/l = 0.028 g/dl (the densities of solvent and polymer considering as equal for convenience), and in the second case ten times 10 higher, so 0.28 g/dl.

48

Problems Polymers

2.23. The slope of sp versus c is greater as M is higher (at lower concentrations more effect of entanglements). 2.24. For convenience in kg/mole: wA200 + (1 wA)30 = 81 ; wA = 0.3; wB = 0.7, so 300 kg of A and 700 kg of B. (Mn )A = 40, (Mn )B = 6, (Mz)A = 600, (Mz)B = 90 kg/mole. 1/Mn = wA/Mn )A + wB/(Mn )B = 0.3/40 + 0.7/6; Mn = 8.054 = 8054 g/mole. Mz = [w A (Mz)A(Mw)A + w B ( Mz)B(Mw)B]/ Mw = (0.3600200 + 0.79030)/81 = 467.8 = 467,800 g/mole. A directly produced batch would show: Mn = 81/5 = 16.2 = 16,200 g/mole, and Mz = 381 = 243 = 243,000 g/mole. So the blend, though on-spec as regards its Mw, has a two times lower Mn and a two times higher Mz as a user would expect. A blown bottle is, therefore, less resistant against stress-corrosion and impact failure at low temperatures. As a result of the stronger pronounced melt-elasticity the die-swell is higher , so the bottle is too thick and its manufacture requires more material! 2.25. a. a is usually between 0.5 and 1.0. b. Mv is usually between Mn and Mw, but closer to Mw as a approaches to a = 1. c. When a = 1, Mv = Mw ; when a > 1, Mv > Mw (this can occur with very stiff chains). 2.26. When in solution, both substances will cause an increase in boiling point. At the same concentration this increase will, for a low-molecular substance, be considerably higher than for a polymer (readily a factor 100 of 1000 times greater), since the increase in boiling point is proportional to the number of molecules solved. An even simpler criterion is the viscosity of the solution: with a low-molecular substance hardly any change is noticed, where a polymer solution exhibits a significant increase in viscosity. 2.27. The number average degree of polymerisation P n = 1/(1 p), in which p is the degree of conversion. The weight average degree of polymerisation P w = (1 + p)/(1 p) (see MT 2.2.2). so the heterogeneity index Mw/Mn = P w/ P n = 1 + p, close to 2. 2.28. a. HO-C2 H4 -OH = 62; HOOC-C6 H4 -COOH = 166; the mass ratio is 62:166 = 1:2.68. b. HO-C2H4 -OH + HOOC-C 6 H4 -COOH = -O-CH 2 -CH2 -O-CO-C6H4 -CO- + 2H 2 O; the polymer formed is polyethylene terephtalate (PETP). c. P = number average degree of polymerisation, p = degree of conversion. d. The molar mass of one unit is 192 (= 166 + 62 218), so 40.000/192 = 1/(1 p) and p = 0.9952. Mw is 192(1 + p)/(1 p) = 79,800 g/mole, nearly twice Mn .

Answers

49

e. Growing chains can close to rings; small acid or basic contamination can terminate chain growth, etc. 2.29. With anionic polymerisation (living polymers) all chains start at the same time and they grow at the same rate until the monomer is exhausted; the chains are, therefore, equally long. 2.30. M n : ebulliometry (boiling point increase); measurement of end group concentration (both for relatively low molar masses); osmometry for higher, but not too high molar masses. Mw: light scattering. Mz: sedimentation in solution by ultracentrifugeing. Further: see MT 2.2.3. 2.31. c (g/dl) t (sec) (t to )to (t to )/to c (dl/g) 0 100 0.05 106.25 0.0625 1.25 0.1 115 0.15 1.5 0.15 126.25 0.2625 1.75 0.2 140 0.40 2.0 In this oversimplified case the limit for c = 0 can easily be determined without drawing a graph: [ ] = 1,0 dl/g. 0.5 = k(5104 )a 2 = k(8105 )a 1 = k(x)a (1) (2) (3)

2.32. [] = kMa, so:

From (1) and (2) it follows: 4 = 16a, so a = 0.5, and from (3) and (1): (x/5104)0.5 = 2 so x = 200,000. 2.33. a is already known (a = 0.5); k follows e.g. from: 0.5 = k(5104 )0.5, so k = 22.36 104. 2.34. a. Mn = 1/(0.98/20.000 + 0.02/200) = 6,711 Mw = 0.9820,000 + 0.02200 = 19,604 Mz = (0.9820,0002 + 0.022002 )/19,604 = 19.996 so: very strong effect on Mn , small effect on Mw, and hardly any effect on Mz.

b. Here it is easier in kg/mole: Mn = 1/(0.98/20 +0.02/2000) = 20,4 = 20,400 g/mol Mw = 0.9820 + 0.022000 =59,6 = 59,600 g/mol Mz = (0.98202 + 0.0220002 )/59,6 = 138,490 = 1.384,900 g/mol so: hardly any influence on Mn , considerable effect on Mw, very large effect on Mz.

50

Problems Polymers

2.35. We calculate /c, for convenience first in mm/(g/dl): c (g/dl) 0.12 0.18 0.25 0.33 0.45 (mm) 6.50 9.94 14.05 18.8 26.6 /c 54.2 55.2 56.2 57.0 59.1

Now we plot /c versus c. Extrapolation to c = 0 gives 52.8 mm/(g/dl). 1mm fluid pressure = gh = 903 kg/m3 9.81 m/sec2 10 3 m = 46.7 Pa and 1 g/dl = 10 kg/m3 . It follows that: /c = 46.77 m2 /sec2 . This equals RT/Mn = 8.314298/Mn , so Mn = 52.97 kg/mole or 52,970 g/mole. 2.36. Measurements of osmotic pressure provide an absolute determination (without calibration) of the number-average molar mass. This is independent of the type of solvent; for each solvent the extrapolation to zero concentration results in the same value: /c = RT/Mn . With viscosimetry the determination of M is not absolute; dependent on the solvent and the temperature one finds a value for the intrinsic viscosity, [], which is not unique but which has to be calibrated.
c

60 58 56 54 52 50 0

0.1

0.2

0.3

0.4

0.5

Figure A.2. Figure to the answer of problem 2.35.

2.37. The increase of viscosity from a certain concentration becomes more pronounced with increasing M (see Mark-Houwink), but the osmotic pressure is inversely proportional to the molar mass. 2.38. with p = 0.99, P n = 1/(1 p)= 100, with p = 0.995, P n = 200. The mass of the unit link is: (-CO-(CH2 )4 -CO-NH-(CH2 )6 -NH-) = 1212 + 214 + 216 + 221 = 226 g/mole. For A, therefore: Mn = 100226 = 22,600 g/mole, for B: Mn = 200226 =

Answers

51

45,200 mole. The weight-average degree of polymerisation is P w = (1 + p)/(1 p) = 199 for A and 399 for B. So the weight average molar masses are 199226 = 44,970 (for A) and 399226 = 90,170 (for B). (end groups disregarded for convenience). 2.39. wA = 0.3; wB = 0.7: (Mw)A = 44,970; ( Mw)B = 90,170; Mn = 1/(0.3/22.600 + 0.7/45,200) = 34,770 g/mole (see 2.12), Mw = 0.344,970 + 0.790,170 = 76,610 g/mole, 2.40. We can measure rel = /o or sp = ( )/ o at a single concentration c (from the run times to and t),and calculate red = ( o)/ o c. rel or sp are in itself already convenient data to rank polymers of the same kind after their molar mass. If a quantitative estimation of [] is desired, then these values can be corrected by the known general dependence of red on c. 2.41. The principle of Problem 2.40 is the basis of the k-value, used with PVC. A measurement of the running time through a viscometer at a fixed concentration, compared with that of the pure solvent, results, after a standard calculation, in the kvalue, which is, for that polymer, a simple index for the molecular mass. In technological terms one knows that a PVC with k = 70 (high M) does not flow easily, but it provides excellent end-use properties, e.g. in the long-term strength of pipes. In injection moulding a better flow is required; then a PVC with k = 55 is more appropriate. 2.42. The melt flow index is a useful indication of the molar mass, since it is a reciprocal measure of the melt viscosity . depends very strongly on: (:) Mw3,4, (doubling of Mw results in a 10.6 times higher !). This relation is valid for the zero-shear viscosity; the melt index is measured at a shear stress where the nonNewtonian behaviour, and thus the width of the molar mass distribution, is already playing a part (see MT 5.3.2). The melt index is a functional measure for the molar mass, because for a producer of end products the processability is often of primary importance. 2.43. From the results of the light scattering experiments it follows that A and B have the same Mw. The viscosity average, Mv , is for A higher than for B, and is thus closer to Mw. A has, therefore, a narrower molar mass distribution and a higher Mn . The osmotic pressure is therefore lower. 2.44. Large coils diffuse through the porous gel via the broadest channels, thus following the shortest way, and they arrive as the first at the end. Smaller coils are able to follow all side paths and have a longer travelling time. The result of a GPC-measurement is a curve of wi against Mi from which Mw can be calculated. Conversion to w i/Mi) against Mi and to wi Mi against Mi gives the possibility to calculate Mn and Mz as well.

52

Problems Polymers

2.45. head-head and head-tail polymerisation in irregular sequences (exception). atactic instead of iso- or syndiotactic structure (example: PP). irregular branching (example: PE). with unsaturated chains: cis-, trans- and 1.2 configurations distributed at random (examples: BR and IR). with copolymers: random distribution of building blocks (examples: SBR and E/P). 2.46. Isotactic: all R-groups are at the same side of the main chain; syndiotactic: in alternating positions; atactic: randomly situated. All of this has to be considered in 3D space due to the tetraedric structure round a C-atom (see MT 2.3). 2.47. An atactic structure is in both cases not crystallisable. Atactic PP is because of its glass-rubber transition temperature (Tg = 15 C) rubbery and technically of no use. Isotactic PP is able to crystallise and can, therefore, be used in practice. For PS atacticity is no objection; its properties as a glassy polymer are retained up to its Tg (95 C). 2.48. by a stiff main chain, e.g. with rings (example: aromatic polyamides). by big side groups (example: PS). 2.49. strong interactions: dipole forces, e.g. in PVC (-Cl) and PMMA (-COOCH3 ). also: hydrogen bridges, e.g. in PA and PU (NH....OC). weak interactions: dispersion forces, e.g. in PE and PP. 2.50. cross link density = number of connected units / total number of units. Another measure is the molar mass between cross-links, Mc, The degree of cross-linking can be determined experimentally from the swelling in a solvent or from the E-modulus. vernettingsdichtheid = aantal verbonden eenheden / totaal aantal eenheden. 2.51. Chemical cross linking is the formation of primary chemical bonds between chains, such as sulphur bridges (rubbers) or polystyrene bridges (UP), both between unsaturated chains. In saturated chains cross linking can be brought about after radical formation by irradiation or peroxides. Physical cross linking is based on secondary bonds (e.g. domains in thermoplastic elastomers) or entanglements.

3. Glassy state and glass-rubber transition

3.1. Similarities: Both are not able to crystallise because of their irregular molecular structure. Both will, therefore, reach the glassy condition upon cooling down from the liquid phase.

Answers

53

Differences: When a low-molecular glass is heated it will, at its T g , immediately change into a liquid (though less abrupt as at a real melting point). A polymeric glass, however, first becomes rubbery as a result of chain entanglements. At a higher temperature (Tv) it passes very gradually into the fluid state. 3.2. All three quantities (V , H and S ) show an upward jump at Tm upon heating (apart from pathological liquids such as water as regards V ). At T g there is, however, no jump, so V = 0, H = 0 and S = 0, but a bend, or a jump in their first derivatives. For V, H and S this means, respectively, jumps in the coefficient of expansion , in the specific heat c and in the compressibility k.
V, H, S V, H, S

Tg

Tm

Figure A.3. Figure to the answer of problem 3.2.

3.3. Because of the many entanglements a long chain is still fixed at a number of points, so that only parts of the chain (of the order of 50 to 100 links) are free to move. At further temperature increase these entanglements are also loosened, so that the polymer is able to flow (Tv region). 3.4. Large side groups increase the distance between the chains and thus decrease the interaction forces and, consequently, the Tg ; example: polyhexene-1 (see also Question 3.6). Plasticisers, added in a sufficient quantity, also remove the chains from each other. Example: PVC with di-octylphtalate, where the polar attraction forces are strongly reduced and Tg may be lowered by more than 100 C. 3.5. A thermodynamically defined transition is a transition between two equilibrium states. The glassy state is, however, never in thermodynamic equilibrium; it is subjected to volume-retardation (rearrangement of chain segments) up to unimaginably long times. The glass-rubber transition should, therefore, be considered as a freezing-in phenomenon, governed by kinetics. 3.6. The side group in polyhexene-1 is a straight and flexible C4 chain; the one in PMP is, however, branched and rigid. Both side groups act as spacers between the chains, thus tending to a lower Tg , but with PMP this effect is overcompensated by the chain stiffening as a result of the rigid side group.

54

Problems Polymers

3.7. The difference between the Tg s of PP and PE is caused by the much greater chain stiffness of PP as a result of the hindered rotation round the C-C bonds in the main chain; the CH3 -groups are in the way. In PVC rotation is hindered in a similar way (the Cl atoms are of about the same size as the methyl groups). The Cl atoms, bound to the main chain, are strong electrical dipoles and they exert much larger attraction forces on each other than the methyl groups in PP, which are subjected to the much weaker dispersion forces only. The same reasoning applies to the series BR, IR, CR. Here again we see an increasing Tg . Now, however, less methyl- and Cl-groups are present, namely one on four C-atoms instead of one on two. There is no effect of the double bond: this is present in all three polymers in the same way. 3.8. PMMA: (110 C) CH3 | -CH2 -C- - - | C=O | O-CH3 H p.methylacrylate: | (5 C) -CH2 -C- - - | C=O | O-CH3

This difference in Tg is a result of the enormous effect of the methyl group at the main chain on the chain flexibility. poly(n-butylmethacrylate): (25 C) CH3 | -CH2 -C- - - | C=O | O-CH2-CH2 -CH2-CH3 CH3 | -CH2 -C- - - | C=O | O-CH2-CH2 -CH2-CH2 -CH2 -CH2 -CH2-CH3

poly(n-octylmethacrylate): (20 C)

Increase of the length of the (flexible) side group increases the distance between the chains and lowers Tg . Moreover, the polarity of the side group decreases, with the same effect on Tg . 3.9. The polymer may show a secondary transition in the glassy region as a result of small movements of parts of the chain or side groups (example: PC). A blend of two incompatible polymers also shows two transitions: the Tgs of both components (example: TPS, toughened PS, PS + BR). A block copolymer tends to segregation; then there are two phases present with each their own Tg (example: SBS).

Answers

55

3.10. SBR is a random copolymer of styrene and butadiene, and it has, therefore, a Tg between that of poly-butadiene (90 C) and polystyrene (+95 C). Its position depends on the S/B ratio, which is chosen in such a way that the polymer can function optimally as a technical rubber for tyres. SBS is a three-block copolymer of styrene and butadiene. Incompatibility of PS and poly-butadiene (BR) causes segregation of the chain parts: a two-phase system results in which PS and BR each have their own individuality and their own Tg . The fact that the highest Tg is lower than the Tg of normal PS, is a result of the relatively low molar mass of the PS chain ends. In general the effect of M on T g is 5 insignificant: Tg = T g - K/ M, but when K is, e.g. 10 and M = 10,000, T g can be 10 lower. 3.11. With copolymers it is of importance whether a one-phase or a two-phase structure is formed (random- or block-copolymers). In a similar way we see with polymer blends two (extreme) cases: real blends of wholly compatible components, and dispersions of incompatible ones. In the first case, which seldom occurs (e.g. PS with PPE) we find a single Tg ; in the second case two Tg s, namely those of the two components. 3.12. Since the glass-rubber transition is characterised by large chain parts becoming mobile (e.g. 50 monomer units), we can from a single Tg only conclude that the blend is homogeneous on that scale; at a smaller scale (of a few links) a two-phase system may still be present. 3.13. The glass transition is mainly governed by chain interactions and chain mobility. Cross-linking will hardly influence the interactions, but the mobility is hindered by the cross-links. Tg will, therefore, increase with increasing network density. 3.14. When a chain with M = 200,000 g/mole is linked to other chains at four points, the average molar mass between cross-links, Mc, amounts to 40,000. The mass of one unit is 4 12 + 61 = 54 g/mole; so the number of units between cross-links is about 740. At the glass-rubber transition no whole chains obtain free mobility, as a result of the entanglements, but chain parts of 30 to 100 monomer units. The chemical cross-links, therefore, hardly contribute to the restriction in chain mobility; the increase in Tg will, therefore, be negligible. 3.15. Polyamides (nylons) can absorb water up to a few percent, which acts as a plasticiser (the distance between the chain segments is increased), so that Tg is lowered. Tg is, therefore, dependent on the environment (dry air, moist air or water). The extremes may differ by several tens of degrees.

56

Problems Polymers

4. Crystalline polymers
4.1. A polymer which is in principle crystallisable because of its regular chains, may crystallise too slowly to develop, under normal circumstances, a detectable degree of crystallinity. In such a case the kinetics play the major role. 4.2. PC and PPE (Tg 150 and 215 C) would have melting points of, respectively: 3/2 (150 + 273) 273 = 361 C(asymmetric chain) and 2(215 + 273) 273 = 703 C (symmetric chain). In both cases processing above Tm would lead to strong degradation. Poly-isobutylene (PIB) is a very useful rubber because of its very low gas permeability. Co-polymerised with small amounts of isoprene (to enable vulcanisation with sulphur) to butyl rubber (IIR), it is the ideal rubber for inner tubes. If PIB would crystallise, it could not be used as a technical rubber! The same holds for the rubbers BR and IR. 4.3. Even with a high cis content the rate of crystallisation of BR and IR is so low, that generally speaking, no crystallisation occurs. However, upon stretching a few hundred percent, micro-crystallites are being formed spontaneously. These form extra cross-links between the chains, thus increasing the strength of the vulcanisate enormously. When the stress is released these crystallites disappear, so that the rubbery behaviour is completely restored. The same holds for butyl rubber, which has, apart from a few isoprene segments, a regular chain structure. 4.4. Tm = H/S, in which H is the increase of enthalpy upon melting, and S the entropy increase. H is related to the chain interactions, S to the chain flexibility. 4.5. Tm would be about 3/2 times Tg , so (3/2) (273+85) 273 = 264 C. Processing of semi-crystalline PVC would only be possible above Tm ; however, the polymer cannot withstand such high temperatures. 4.6. n 8 10 15 20 32 M 114 142 212 282 450 Tm (C) 56.8 29.7 10.0 36.8 69.7 Tm (K) 216.4 243.5 283.2 310.0 342.9 1000/M 8.77 7.04 4.72 3.55 2.22 1000/Tm 4.62 4.11 3.53 3.23 2.92

After plotting 1000/Tm against 1000/M, and extrapolating the slightly curved line to 1000/M = 0, it appears that the first part of the curve can be represented by: 1000/Tm = 2.44 + 220/M. So the melting point of PE with infinite chain length is 1000/2.44 = 410 K or 137 C. For Tm = 409 we find: M =37,000. With this low molar mass Tm appears to be only 1

Answers

57

C lower than the maximum; so for practical grades the differences are tenths of a degree.
5 1000/T m n=8 10 4 15 20 3 32

2 0 1 2 3 4 5 6 7 8 Figure A.4. Figure to the answer of problem 4.6

1000/M 9 10

4.7. The coiled structure with entanglements does not allow complete crystallisation to occur. From very dilute solutions perfect crystals can be formed, in particular when the chain molecules are lying fully stretched and oriented. A technical example is the high-strength PE fibre: Dyneema. 4.8. The difference in T m must be due to the previous history. The first sample has been quenched from the melt, and thus it is crystallised at a lower temperature, e.g. 130 C. The crystallites are, therefore, smaller (more nuclei at a lower temperature) and less perfect than in the second sample, which was cooled more slowly and which crystallised at, e.g. 145 C. 4.9. The lower the temperature of crystallisation, the lower the melting point of the best crystal will be, but also: the more imperfect crystallites are formed. The melting range of the sample with Tm = 161 C is, therefore, broader. 4.10. PBTP has about the same chain structure as PETP, but has in the main chain four CH2- groups instead of two with PETP. The chain is , therefore, more flexible, which results in a lower melting point. PBTP crystallises considerably more rapidly than PETP. In an injection moulding process, where short cycle times are required, and thus rapid crystallisation during the rapid cooling in the mould, PBTP may, therefore, be preferred. PETP would stay amorphous, and, therefore, not usable above its Tg (69 C). 4.11. Nucleating agents can be added to the polymer, which enhance the rate of

58

Problems Polymers

crystallisation considerably. It is also possible to take the amorphous product from the mould and heat it for some time in an oven at some temperature above its Tg , so that it crystallises to form an acceptable end product. 4.12. Chain orientation enhances the rate of crystallisation, because every volume element in which a few chain parts lie parallel forms a nucleus. This is the case in fibre spinning, film stretching and bottle blowing. 4.13. In a (stereo-specific) rubber spontaneous crystallisation occurs under strain; here also small regions in which chain parts lie parallel, act as nuclei.These nuclei, which improve the strength considerably, do, however, not grow out into a continuous phase; they disappear upon stress release. 4.14. Nuclei with a radius r smaller than rk are disrupted by the thermal motion; when r > rk the nuclei are stable and can grow. rk is determined by the competition between the formation of new interface (interfacial energy has to be supplied) and the production of crystallisation heat. In the formula for r k , (Tm T ) is in the denominator, which indicates that with stronger super-cooling a nucleus can grow more easily. 4.15. The ratio of the crystallisation rates is 98/23 = 4.26, the ratio of the molar masses is 143/87 = 1.64. The ratio of the melt viscosities (which govern the crystal growth) amounts to 1.643,4: 1 = 5.38. Apparently the rate of crystallisation is, next to the rate of nucleation, governed by the melt viscosity. 4.16. The Tg values of most of the polyamides are between 50 C (PA-6 en PA -6.6) up to 82 C (PA-4.6) (in ambient atmosphere a few tens of degrees lower due to moisture absorption). They, therefore, already lose their properties at some T increase. Crystallisation is, therefore, essential for technical applications Amorphous polyamides should have a higher Tg (e.g. Gelon A 100 of General Electric with Tg = 130 C) by the presence of aromatic rings in the main chain, and, moreover, an irregular chain structure (to prevent crystallisation), by the introduction of side groups. 4.17. With increasing n the -CO-NH- groups in the main chain are more and more diluted in the -CH2- sequences; their effect decreases and Tg and T m approach to the values of PE. The zig-zag effect is a matter of parity: the formation of hydrogen bridges is only possible if the groups are in the proper position. 4.18. - For a non-crystallisable polymer: a. and b. equal c. solidifies earlier d. volume-retardation

Answers

59

- For a crystallisable polymer: a. undercooling b. melting range c. more undercooling d. post-crystallisation.

V b c d b c a

T2

Tg

T1

Figure A.5. Figure to the first part of the answer to problem 4.18. V a b c a d T2 b Tm T1 T c

Figure A.6. Figure to the second part of the answer to problem 4.18.

4.19. a. Tg will certainly lie between 20 C and 80 C; probably between 30 en 70. b. PETP has an asymmetric chain, so Tm = (3/2) T g , i.e. between (3/2)(273+30)273 = 182 C en (3/2)(273+70)-273 = 241 C. c. Above Tg crystallisation can occur in the rising branch of the curve for crystallisation rate vs temperature. d . At T m the rate of nucleus formation is zero; at Tg (and at Tm) the rate of crystal growth is zero ; between these two a maximum occurs (see MT 4.3.2). e. See Figure A.7. f. See Figure A.8. g. In the rapidly cooled mould of an injection moulding machine crystallisation proceeds too slowly and needs help from nucleating agents or annealing above Tg after moulding. See also problems 4.10, 4.11 and 4.12 .

60

Problems Polymers

a d c 0 50

b b d 100 150 200 250 T C 300

Figure A.7. Figure to the answer of problem 4.19e. Log E

after C after A T C 0 50 100 150 200 250 300

Figure A.8. Figure to the answer of problem 4.19f.

4.20. The volume of the unit cell is 0.740 0.493 0.253 nm3 = 9.230 10-29 m3. When n CH2- groups are present in such a cell, then the mass within the cell is m = n 14 g/mole, so n 14/ 6 10 23 grammes = 14n/610 26 kg. The density is: = (14n1029)/(610269.23) = 253n kg/m3 . n must be an integer and (must be slightly higher than the density of a near-linear PE (960 kg/m3 ). Only n = 4 meets these conditions. 4.21. 1 cm3 of the polymer has a mass of grammes. It contains cm3 of crystalline polymer with a mass of c grammes and (1) cm3 amorphous with a mass of (1) a gr. The total mass is = ( a)/(c a).. 1 gramme of the polymer has a volume of 1/ cm3 and contains (1/ ) c grammes crystalline matter, so the crystalline mass fraction is = (c/) = (c/) ( a)/(c a). 4.22. The Avrami equation can be transformed into: log [ln(1 V/V)] = log c + log t.

Answers

61

We can now construct the following table: t 0 500 1000 1500 2000 2500 V 120.00 119.16 114.97 110.94 110.04 110.00 V 0 0.84 5.03 9.06 9.96 10.00 = V/V 0 0.084 0.503 0.906 0.996 1.000 1 1 0.916 0.497 0.094 0.004 0 ln(1 ) 0.0877 0.6992 2.364 5.521

Plotting ln(1 ) against t, both on a log scale, indeed results in a straight line with slope 3, so = 3. c can e.g. be determined at t = 1000: c = 71010 .
1.0 0.5 0 0.5 1.0 2.5 3 3.5 log t log(ln(1 ))

Figure A.9. Figure to the answer of problem 4.22.

b. =3 can mean: three-dimensional crystal growth with heterogeneous nucleation of two-dimensional growth with homogeneous nucleation. c. At 155 C the exponent will not change very much. c will, however, be considerably smaller as a result of much slower crystallisation. d. The volume of the amorphous polymer is 120 mm 3; so the amorphous density is a = 100/120 g/cm3. The density of the crystallised sample is = 100/110 g/cm3. With c = 0.937 g/cm3 the crystalline volume fraction can be calculated: c = ( a)/(c a) = 0,73. 4.23. From c= ( a)/(c a) it follows for LDPE: c = 0.38 and for HDPE: c= 0,76. 4.24. The cause can be found in the morphology of the polymer. The question is: is the amorphous phase continuous or the crystalline phase or both? This is further dealt with in Chapter 9. 4.25. A semi-crystalline polymer is a two-phase system; it consists of a crystalline

62

Problems Polymers

phase and an amorphous phase, which, in most cases, have a different refractive index. Light is, therefore, deflected at each interface, which results in strong light scattering. When, by exception, the refractive indices are equal (as in PMP), no light scattering occurs and the polymer is transparent. When the crystallites are small enough (small in comparison with the wave length of the light), only minor scattering occurs. This can be achieved with nucleating agents, but also with chain orientation. 4.26. In a super-strong PE fibre the chains lie nearly completely stretched, and are thus loaded under optimal conditions. They must, however, transmit the mechanical stress on each other, which can only happen via transverse bonds. These are relatively weak binding forces (London-Van der Waals bonds), so the number of these bonds should be made as large as possible, which means very long chains. 4.27. For a semi-crystalline polymer the E-modulus shows between Tg and Tm (in which region it is already lower than below Tg ), a rather strong decrease at increasing T, whereas with amorphous polymers, which are used below Tg , the stiffness is not much temperature dependent (apart from possible secondary transitions). The time dependency, or the creep, shows a similar behaviour. As a matter of fact, this does not hold for semi-crystalline polymers far below their Tg . (strong example: PEEK). The word far is here important, since creep proceeds up to very long times of loading, whereby Tg is lowered; the polymer can as it were creep towards its Tg ! 4.28. Lyotropic LCPs are processed from a solution, thermotropic ones from the melt. In both cases the flow pattern provides the necessary orientation of the stiff chains. Working with a solution is needed for polymers which above their melting point would degrade (example: aromatic polyamide fibres such as Kevlar and Twaron). For fibres the removal of the solvent is not too problematic. In e.g. injection moulding applications solvents cannot be used; here thermotropic LCPs have to be used. Since these would degrade during processing, they are diluted by copolymerisation (example: poly-hydroxy-benzoic acid - co - PETP) 4.29. The glass-rubber transition extends over a rather broad temperature region. Tg is, therefore, dependent on the chosen method of measurement and on the definition of the transition point. Moreover, Tg depends on the time scale of measurement. Though the melting point, Tm , is a well-defined quantity for an ideal crystal, melting also extends over a broad temperature region as a result of the differences in crystal perfection. The highest melting point found for a polymer depends on the quality of the best crystal, which depends on the regularity of the chain structure (compare

Answers

63

HDPE and LDPE) and on the rate at which the polymer has crystallised (so on the crystallisation temperature). 4.30. All three transitions are shown by a semi-crystalline thermoplastic with a chain length long enough to extend the rubbery region to above the melting point. 4.31. Only T g and T v are present in an amorphous thermoplastic and in a nonvulcanised rubber. A semi-crystalline thermoplast with short chains only shows Tg and Tm. In a (very exceptional) fully crystalline polymer Tg is not present 4.32. For vulcanised rubbers and for thermosets Tg is the only transition. 4.33. Between Tg and T m the polymer contains a crystalline and an amorphous, rubbery phase. The rubbery phase is responsible for the high impact strength.

5. Rubbertoestand en vloeibare toestand

5.1. F is the free energy and is related to the Gibbs free enthalpy G, via G = U + pV TS = F + pV. When F is used instead of G the work exerted on the environment by volume change is thus neglected. U is the internal energy, which increases when work is done to overcome interaction forces. T is the temperature in Kelvin. S is the entropy of the system, and is a measure for the probability of the state, or for the number of possibilities (conformations) at which the state can be realised. 5.2. When the length l is increased the force needed is: dF dU dS K = dl = dl T dl In the glassy state interaction forces have to be overcome, so dU/dl is important. The entropy, however, hardly changes, since the chain conformations do not change at a small deformation. The force is, therefore: K = dU/dl. In the rubbery state, on the contrary, the chain interactions are not or hardly active: they have, from T g , been overcome by the thermal movement. The entropy, S , strongly depends on the deformation so the force is now given by: K = T dS/dl. 5.3. The entropy change upon elongation is a matter of conformations, so purely topological, and is not influenced by temperature. 5.4. a. The molar mass of the isoprene unit (-CH2 -C.CH3 =CH-CH2 -) is 68 g/mole, of a half cross-link: 4 32 =128 g/mole. The mass fraction S is 0.005. When n isoprene units are between crosslinks, then 128 = 0.005 n 68, so n = 376 and Mc = 376 68 = 25,600 g/mole. b. Substitute: = 910 kg/m3, R = 8.31 J/K.mole, T = 298 K, Mc = 25.6 kg/mole: E =

64

Problems Polymers

264.000 Pa. c. E appears to be more than 5 times higher. Strain-induced crystallisation cannot be an explanation, for E is measured at very small strains. Deviation from ideal rubber-elastic behaviour gives a very small effect only. The explanation is the existence of physical entanglements, which are much more numerous than the Sbridges. (without vulcanisation also a value of nearly 1.5 MPa! is found!) d. Entropy-elasticity dominates the behaviour: TdS/dl is the important parameter. S is a matter of order/disorder: the number of possible conformations, The difference between conformation possibilities in strained and unstrained condition is greater as the freely moving chain part is longer; therefore Mc is in the denominator. 5.5. a . Mc = 3 RT/E = 3 910 8.31 293 / 2.2 10 6 = 3.02 kg/mole = 3020 g/mole. This amounts to 3020/54 = 56 monomer units. The use of proper units is very important ! In this case, i.e. for non-vulcanised rubber: Mc = molar mass between physical, temporary entanglements. Not the whole chain! b. E(100) = E(20) 373/293 = 2.8 MPa, but: entanglements are loosened; there is a gradual transition to flow, so M c increases and E decreases. The three deviations from ideal rubberelastic behaviour, mentioned in MT 5.1, are here, for nonvulcanised rubber, of much less importance. 5.6. a., b., c. See Figure A.10. d. With light cross-linking Tg hardly increases, since the chemical cross-links are at much greater distances from each other than the physical entanglements. With tighter networks this is no longer the case, so that the chain mobility is hindered by the presence of the many cross-links and Tg is increased. e. The same reasoning as above: In the formula E = 3RT/Mc, the distance between cross-links, (expressed as molar mass M c) is, with light vulcanisation, mainly governed by the physical entanglements. With higher degrees of cross-linking the distance between chemical bridges plays an increasing role, so that E increases. f. For an ideal network T is in the numerator of the formula for E , so that E is proportional to the absolute temperature. The log E - T curve thus shows a positive slope (not a straight line because of the log-scale but slightly curved upward). In reality this simple picture is often disturbed by deviations from ideal rubber-elastic behaviour.

Answers

65

log E

c. b. a. Figure A.10. Figure to the answer of problem 5.6. T

5.7. When the rubber is heated from 253 K to 298 K, the E-modulus becomes 298/253 = 1.18 times higher. The wall stress in the balloon will increase by 18% and the diameter decreases to 20/1.18 = 17 cm. Gradually the air in the balloon will be heated up, and the pressure will increase proportionally to the absolute temperature. Eventually the old diameter of 20 cm will be reached. 5.8. First: Also in the rubbery condition flow will occur and in the fluid state the melt also shows elastic properties. Secondly: the transition from rubber to flow is strongly dependent on the molar mass: this transition is governed by the unravelling of chain entanglements, which are, with long chains, much more numerous than with short ones. Thirdly: the longer one takes the time to disentangle the coils, the easier this is performed. The time scale plays, therefore, also a large role. At short times a polymer may behave as a rubber, whereas it flows at longer times. (think e.g. of silly puttee). 5.9. a. en b.: see Figure A.11. A twice as high molar mass gives a 23.4 = 10.6 times higher zero-shear viscosity, in which the relevant average is Mw is ( (:) M 34 w ) The whole curve is thus shifted upward by a factor 10 (one unit on the log scale). Broadening of the molar mass distribution shifts the curve to the left; the zeroshear viscosity does not change. c. the power-law holds for the right-hand part of the curve, which often resembles . . a straight line. From = k()n it follows, together with = : (:)(11/n). When 1 n = 2 (:) 1, so on the log-log scale the slope is 1 (45 to the left). d. In injection moulding shear stresses and shear rates are high, so that a broad distribution is beneficial for easy mould-filling. However, the high Mz may be responsible for high melt-elasticity resulting in frozen-in orientations and anisotropy.

66

Problems Polymers Pas 105 a. 104 103 102 0,1 1 10 100 1000 Pa b.

Figure A.11. Figure to the answer of problem 5.9.

With rotation moulding is very small; a narrow distribution may then be preferred, also because a higher Mn favours the impact strength etc. With film extrusion a broad distribution is advantageous for making very thin films and for the elastic effects (Mz) required in shrink films; a narrow distribution results in better tear strength and impact resistance (Mn ). 5.10. In a plot of log vs. log the curves are shifted vertically upon a change in M or in T . (see MT Figure 5.12). In such a plot the melt-index can also be indicated easily, since the m.i. is measured at a fixed stress. 5.11. a. The melt flow index is measured at a shear stress which may strongly differ from the one present in a processing operation. In particular with injection moulding the latter is much higher than the former. b. By a higher value of the shear stress, so a greater force on the plunger, better agreement can be reached, e.g. instead of 2.16 kgf: 5 or 10 or even 21.6 kgf. c. At e.g.10 kgf the ratio of the melt flow indices is higher than 10 /2.16 ; this ratio increases with increasing width of the molar mass distribution (non-Newtonian behaviour or shear-thinning). 5.12. An increasing melt-flow index signifies decreasing molar mass; with polymers made in the same process both Mw and Mn are lower. Since the impact strength is governed by Mn , it decreases with increasing m.f.i. With a narrower MMD Mn is, however, higher than what would be expected on the basis of the m.f.i; therefore the impact strength is also higher. 5.13. A higher m.f.i. means a lower molar mass and a lower melt viscosity. The rate of crystallisation will then be higher, and a grade with a higher m.f.i. will crystallise better and reach a higher degree of crystallinity, thus a higher stiffness. For the grades with a narrower MMD one would expect, at the same m.f.i., a lower Mw, so a lower zero-shear viscosity (see MT, Figure 5.14), which would result in a higher

Answers

67

crystallinity and a higher stiffness. It is, however, the other way round. Possibly in the crystal growth process higher shear stresses and -strains are involved than under m.f.i. conditions; possibly other, unknown causes play a part. 5.14. Polyethylene crystallises much more rapidly than polypropylene; under usual cooling conditions PE will always reach its maximum possible crystallinity, irrespective of its chain length, contrary to PP. An exception is formed by the ultrahigh-molecular PE (UHMWPE), in which the chains are exceptionally long (M > 106 g/mole), the viscosity exceptionally high and the crystallinity (also the density) slightly lower. However: super- strong PE -fibres are made from the longest possible chains, but the crystallinity is extremely high as a result of the near-perfect alignment of the chains. 5.15. a. die-swell. i.e. recoil of the elastic deformation of the melt after it leaves the die of an extruder; b. melt fracture: the elastic strain may be so high that the material breaks; the product is distorted; c. crystallisation: with a crystallisable polymer (e.g. a stereospecific rubber) the elastic strain may reach such high values that spontaneous orientation-induced crystallisation occurs, causing considerable trouble in the process (see MT Figure 5.15). d. when the elastic deformations (chain orientations) are frozen in during cooling, an anisotropic end-product results, which shows different stiffness and strength in different directions. Also the refractive index may be dependent on the direction (bi-refrigence, which is of importance with e.g. compact discs).

6. Visco-elasticity
6.1. A Maxwell element shows an instantaneous elastic deformation and thereafter unlimited flow. For polymers in the solid condition the latter is not realistic. For a fluid polymer it is more relevant; moreover, the instantaneous elastic deformation is in accordance with the real behaviour: when the stress is released a polymer fluid shows an instantaneous recoil. 6.2. a. o = /E =104/105 = 0.1 = 10%. b. = o + t/ = 0.1 + 104500/107 = 0.6 = 60%. c. relaxes between t = 500 and t = 800 sec with = exp(t/), in which o = 104 , t = 300 sec and = /E = 100 sec, so = 104e3 = 500 Pa (e3 is close to 20). d. See Figure A.12. e. - Non-linearity. - spectrum instead of single relaxation time

68

Problems Polymers

kPa 10 8 6 4 2 0 0 0,6 0,4 0,2 0 0 200 400 600 800 t sec Figure A.12. Figure to the answer of problem 6.2 200 400 600 800 t sec

6.3. Evidently a fluid polymer cannot be considered; in the model the deformation approaches to a limit. For a solid polymer the model seems more appropriate, though is represents neither a spontaneous elastic deformation nor permanent flow. Therefore a combination of a Kelvin-Voigt element with a spring and with a dashpot in series is, in principle, more appropriate. 6.4. Cross-section of the bar = A = 102103 = 105 m2 ; the stress = K/A = 600 N/ 105 m2 = 60106 N/m2 = 60 MPa. The strain is l (mm)/l (mm) = l(mm)/100, so for 1 mm strain is = 0.01. a. The model shows instantaneous strain and retarded strain, and it can, therefore, in principle be represented by a spring E1 in series with a Kelvin-Voigt element with E2 and . b. From the instantaneous strain (2 mm) it follows: = 0.02. So E 1 is: E 1 = /o = 60/0.02 = 3000 MPa. The limit of the strain is 8 mm, so = 0.08, of which the Kelvin-Voigt element accounts for 0.08 - 0.02 = 0.06 , so in = /E1 + (/E2 )(1 exp(t/)) is /E2 = 0.06, so E2 = /0.06 = 60 MPa/0.06 = 1000 MPa. At last the retardation time : at t = 250 sec = 7,5 mm = 0.075. Substitution:

Answers

69

0.075 = 0.02 + 0.06(1 exp(250/) results in: 0.055 = 0.06 (1 exp(250/)); = 100 sec = /E2 , so = E2 = 100109 = 1011 Pas. c. When the stress is released after 250 sec the strain on E 1 disappears instantaneously; so the total strain decreases from 7.5 to 5.5 mm. Thereafter the strain decreases as: = 5.5 * exp(t/), where t starts at 250 sec. d. Non-linearity and spectrum rather than a single retardation time.
mm 10

0 0 100 200 300 400 500 600 sec Figure A.13. Figure to the answer of problem 6.4.

6.5. a. o = /E2 = (6106 )/(3109) = 0.002 = 0.2% (only the spring in series). b. (3106 ) = + 1 = 0.002 + ( /E1)(1 et/), ( = /E1 = 106 sec) = 0.002 + (6106 )/(3108)(1 e3) = 0.002 + 0.020.95 = 0.021 = 2.1% c. E2 recoils: = 2.1 0.2 = 1.9%. d. The Kelvin element recoils with: = 0.019exp(3106 /106) = 0.0190.05 = 0.095%.
2%

1%

t 0 0 1 2 3 4 5 6
10 6 sec

Figure A.14. Figure to the answer of problem 6.5.

70

Problems Polymers

6.6. The course of the strain can be described by superimposing the effects of the two stresses; the first one is , applied at t = 0, and continued after t1; the second one is , applied at t = t1 . We find then from t = t1: = (/E2 ) + (/E1 )(1 exp(t/)) (/E2 ) (/E1 )(1 exp((t t1)/ ) = = (/E1 ){exp((t t1 )/ ) exp(t/)} = = (/E1 ){1 exp(t1/ )}{exp((t t1)/ )} = = (t1) exp((t t1)/ ), in which (t1) is the strain at t1 immediately after the release of the stress. Without the superposition principle we find the same result : After taking away the stress the spring E2 is unloaded and we keep the deformed Kelvin-Voigt element with a strain: = (/E1 )(1 exp(t1 / )) = (t1). The spring E 1 now pulls back the Kelvin-Voigt element from t = t1 with a stress , proportional to the remaining deformation : = E 1 (t t1). The rate of deformation is given by the behaviour of the dashpot: d /d t = / (negative, since the deformation decreases). So = E 1(t t1) = d /dt, from which: = (t1)exp((t t1)/ ) with, again, = /E1 . 6.7. The strain after removal of the stress approaches tot 0, and there is also an elastic strain of 1 cm (retraction from 5 to 4 cm strain), A possible model is, therefore, a Kelvin-Voigt element (with E1 and parallel), in series with a spring E2. From the immediate jump in the strain after stress release ( = 104 N/10 4 m2 = 108 Pa), it follows: 2 = 1/20 = 0.05 = /E2 = 108 /E2 , so E2 = 20108 Pa = 2000 MPa. 1 Now the Kelvin-Voigt element: from l = 4exp(2t) it follows = 2 hour = 1800 sec = /E1 . The strain during the first 5400 sec can thus be represented by: = 2 + (/E1 )(1 exp(t/ )) = 0,05 + (108 /E1 )(1 exp(3)) = 0.5, from which (with e 3 = 20) follows: E1 = 475 MPa. With = 1800 sec this results in: = 1800 4.75108 = 855109 Pas. The model is shown in the figure. 6.8. Now the strain, after stress release, approaches to 0.5 cm, which is a permanent strain 3 = 0.5/20 = 0.025 as a result of the dashpot in series in the model. The viscosity, 2, of this dashpot is given by: 3 = t/2 or 0.025 = 108 5400/ 2 , so (2 = 21.61012 Pas. The elastic deformation is now 0.5 cm or 1 = 0.025,, so the spring in series has a stiffness / 1 = 108 /0.025 = 40108 Pa = 4000 MPa. For the Kelvin-Voigt element remains: 2(5400 sec) = (4 cm) = 0.2 = 108/E1 (1 - exp(5400/1800)) = 0.95108 /E1 , so E 1 = 475106 = 475 MPa. Since = 1800 sec, 1 = 1800475106 = 855109 Pas. The model is, therefore, as shown in Figure A.16.

Answers

71

E1 = 0.475 GPa

= 855 GPas

E1 = 0.475 GPa

1 = 855 GPas

E2 = 2 GPa

E2 = 4 GPa

2 = 21.600 GPas

Figure A.15. Figure to the answer of Figure A.16. Figure to the answer of problem 6.7. problem 6.8.

6.9. Let the deformation of the spring E be 1 and of the dashpot: 2. Their sum, = 1 + 2 is equal to vt, in which v is the constant rate of strain, d/dt. The stress is given by = E1 = (d2/dt). With (d 2 /dt) = v (d1/dt) we find: v = (E/)1 + d 1 /dt, from which, after solving, it follows that: 1 = v(1 exp(t/)) with = /E. So thhe - diagram (see Figure A.17) looks as follows: = Ev(1 exp(t/) = v (1 exp(t()).

E Figure A.17. Figure to the answer of problem 6.9.

6.10. Again: = vt. Now the stress on the spring is 1 = E = Evt, and on the dashpot: 2 = v . The total stress is = v (E t + ) = (/t)(E t + ) = (E + /t) = E + v.

72

Problems Polymers

Figuur A.18. Figure to the answer of problem 6.10.

6.11. Apparently the stress relaxation proceeds in two phases, each with a stress decrease of 1 MPa at a strain = 1. It seems logical to think of a parallel arrangement of two Maxwell elements, both with a spring constant E = 1 MPa, but with relaxation times which differ by a factor 10,000 . Inspection of the stress values (with = oexp(t/)) easily results in: 1 = 1 sec, 2 = 10,000 sec. The viscosities are then: ( = E): 106 and 1010 Pas. 6.12. a . E 1 is the storage modulus, and indicates how much reversible energy is stored in the material (purely elastic). E2 is the loss modulus; it is a measure for the energy lost during deformation i.e. dissipated into heat; it thus signifies the contribution of the fluid nature of the material. b. Vibration experiments, e.g. with a torsion pendulum, yield E1 from the vibration time and E2 from the damping of the vibration (see MT 6.2).
2 MPa

0 0,01 0,1 1 10 102 103 104 105 t sec Figure A.19. Figure to the answer of problem 6.11.

c. The ratio of these two quantities, E2/E1 , is tan , which is, therefore, a measure for the position of the material in the visco-elastic region ( = 0: purely elastic; = 2 (90): purely viscous) d. heating under repeated stress cycles (e.g. with fatigue). damping of shock waves, e.g. with rubber-modified polymers, use of a polymer for damping of vibrations, heating of an automotive tyres,

Answers

73

friction of a tyre on the road (see MT 7.3) 6.13. a. a. The curve p is typical for an amorphous, not cross-linked polymer; q for a cross-linked polymer, b. For both curves the maximum lies at the glass-rubber transition temperature Tg, c. For p the polymer gradually shifts into the liquid condition; increases up to 2 and tan to infinity, d. In a network (q) the loose chain ends (not held between two cross-links) do not contribute to the network elasticity, but they behave more or less as a fluid. Therefore, with increasing T, tan approaches to a finite value, dependent on the network structure.

7. Mechanical properties
7.1. The cross section of the bar is 102103 = 105 m2 . The stress is = 600/105 = 60106 Pa = 60 MPa. The elongation is = 2/100 = 0.02. The modulus of elasticity is E = / = 60/0.02 = 3000 MPa. = 3 GPa. 7.2. A value for E of about 3 GPa is normal for an amorphous polymer in the glassy state, unless below Tg a strong secondary transition occurs such as with PC, so that the E-modulus at ambient temperature is significantly lower. 7.3. With semi-crystalline polymers we should always carefully distinguish between the behaviour below Tg and above Tg . Below Tg (such as with PEEK) the crystalline fraction, which is somewhat stiffer than the amorphous glass, dominates, so that E is somewhat higher. Above Tg , such as with PE and PP, the amorphous fraction, which is in the rubbery condition, is responsible for a significantly lower E. 7.4. By chain orientation, such as in fibres, by reinforcing fillers, such as hard particles or short fibres. 7.5. Tighten both bars at equal length in a clamping-screw, bring them in a bending vibration and listen to the pitch. The highest tone is given by the stiffest bar, which is the bar made of pure PP. Musical people may be able to estimate the interval between the two tones; if this is, e.g. a whole-tone distance (12% in the frequency), we know that the ratio of their E-moduli is about (1/1.122) = 0.8. With the Kerner formula (Chapter 9) we can now estimate the rubber content. 7.6. Fibres are, as a result of the spinning process, molecularly oriented, and they have, therefore, a 2 to 3 times higher stiffness than the non-oriented polymer (e.g. polyamide and polyester textile fibres). With the highest attainable orientation, such as in aromatic polyamides (Twaron and Kevlar), and in the PE-fibre (Dyneema) the stiffness can be a hundred times higher than the one in the unoriented condition!

74

Problems Polymers

7.7. The chain stiffness influences the height of the glass-rubber transition temperature (and of the melting point), but not the stiffness of the polymer below Tg (in the glassy state). Extremely stiff chains show the effect of the formation of LCP's (liquid-crystalline polymers), by which very high stiffness is reached, but only in the direction of the orientation. 7.8. In principle a shifting rule could be applied, but the volume retardation (physical ageing) is responsible for a gradual increase of the resistance against creep, so that the rate of creep gradually decreases. Creep curves at e.g. 20 C thus proceed considerably less steep, and the real creep is therefore (fortunately!) much less than would follow from the T-log t shift (see MT 7.2.4). 7.9. See Figures A.20, A.21 and A.22. a. could possibly apply to a semi-crystalline polymer with a Tg not too far above ambient temperature, so that E (below Tg ) is still high; at longer times of loading the polymer A creeps gradually towards Tg" (example: PETP). b. could apply to an amorphous glassy polymer; E has its normal high value (about 3 GPa), while T g is high enough to be not approached at longer times of loading (example: PMMA).
MPa 102 1 20 102 104 hours MPa 20 102 1 102 104 hours MPa 20 102 1 102 10 10 10 104 hours 0 0 1 2% 0 0 1 0 0 1 2%

Figure A.20 22. Figures to the answer of problem 7.9.

c. The low E-modulus could indicate a semi-crystalline polymer above Tg such as PP; the strong creep of a crystalline - rubber two-phase system is also in accordance with this assumption. 7.10. The Kohlrausch formula is well suitable to represent the creep at small stresses and strains. D o is then the compliance at t = 0, and is a measure of the immediate elastic deformation. The formula, however, fails when the creep behaviour is nonlinear; this is, in general, the case with stresses occurring in practice. 7.11. a . After cooling to below Tg the volume of a glassy polymer continuously decreases gradually (see MT 3.3.): "volume-retardation". b. The resistance against creep increases strongly by volume retardation.

Answers

75

c. See Figure A.23.

without Figure A.23. Figure to the answer of 7.11 c. with physical ageing

7.12. Noryl (homogeneous blend of PPE and PS) is an amorphous polymer in the glassy state. It shows relatively little creep; the isochrones lie close to each other. With PBTP the distance between isochrones is much greater: there is more creep. This is related to the semi-crystalline structure of PBTP (contrary to PPE/PS). Though the glass-rubber transition temperature of PBTP, Tg, is about 65 C, this transition is already apparent at lower temperatures and longer loading times. The polymer belongs, therefore, to the category described in MT 4.5, Figure 4.20. 7.13. a. Can be read off directly from the 102 hour isochrone at 8 MPa: = 1.33 %.
18
5 6 10 7 8

MPa 102 hours

16 14

102 10 104

5 6 8

12 10

4 3

5 6 4

3 3 2 1 1 2

8 6 4 2 PP 20 C 1 2 3 4%

4 3

2 1 1

100

80

60

40

Figure A.24. Figure to the answer of problem 7.13a.

b. At 70 C we choose the axis half-way between 60 C en 80 C; for 10 hours an isochrone half-way between 1 and 100 hours, and we find at 5 MPa: = 1.94%.

76

Problems Polymers

c. For the short-time modulus we take the slope of the 102 hour isochrone, and we read off at 20, e.g. at 0.2% : 3.4 MPa. It follows then that E = 3.4 / 0.002 = 1700 MPa. d. With the same isochrone we find at 80 and 0.2%: 1.3 MPa, so E = 1.3/0.002 = 650 MPa. e. 1 year = 8766 hours, so close to the 104 hours isochrone. With = 2% we find at 20 C about 8 MPa as an allowable stress. f. At 100 C and 1000 hours (half-way between the 102 and 104 isochrones), we find for 3% about 3.7 MPa. 7.14. Plotting times to failure (log t, horizontal) versus applied stress (log , vertical, different scale), results in the following figure:
MPA 200 150

20 C 40 C 60 C 80 C 100 C

100 80 60 50 40 PB 30 110 C

time to failure 20 102 101 1 10 102 103 104 105 hours Figure A.25. Figure to the answer of problem 7.14.

From this figure, safe wall stresses at various temperatures and loading times can be determined by extrapolation. We extrapolate the curves found at higher T to lower Ts. Here a ten years life (87,660 hours) at 80 C is required. Extrapolation of the curve for 80 C results in an estimated stress to break of 28 MPa. To remain at the safe side, we stay somewhat below this value, so that 20 MPa looks a reasonably safe wall stress for this purpose.

Answers

77

7.15. a. First damage can occur by "crazes" (see MT 7.4.2) b. To avoid "crazes" completely, we have to stay below a "critical strain", which, for each polymer is about a material constant. To remain within this limit the use of creep isochrones is required.. 7.16. First: A well defined notch forces the sample to break at that place because of the stress concentration, and not at an accidental scratch or contamination. The results are, therefore, better reproducible. Secondly: At the location of the notch the rate of deformation is considerably higher than in the remainder of the sample. The higher rate of strain has a similar effect as a temperature decrease, so that one measures, as it were, at a lower T , with a much greater chance of brittle fracture (which is important for practical use) (see MT 7.4.3). 7.17. a . Improvement of impact strength can be reached by adding a dispersed rubbery phase. b. The rubbery phase introduces a glass-rubber transition at the Tg of the rubber, and, therefore, a damping peak which is able to absorb shock energy (analogous to some secondary transitions in the glassy region, such as with PC). Moreover we can think of the neutralisation of a just initiated crack in a dispersed rubber particle. c. At the Tg of the added rubber the E-modulus of the blend necessarily decreases; in the temperature region of use we, therefore, lose in stiffness. d. Examples: TPS (toughened polystyrene) or HIPS (high-impact PS) or SB (styrene-butadiene); also PP copolymer, a PP with "tails" of ethylene-propylene copolymer, which form a rubber phase, PVC with CPE (chlorinated PE), etc. etc. 7.18. The impact strength of a polymer is strongly dependent on the molar mass, in particular on Mn (number average, number of chain ends) 7.19. a . In a Shore-D hardness test the penetration of the pin is measured during loading; the load is thus prescribed as well as the resulting deformation (see MT 7.5.1). b. The time scales of both measurements may differ considerably; moreover, with a hardness test we may enter into the non-linear region. c. A hardness test is, with a simple apparatus, much easier than a measurement of the E-modulus, which requires the measurement of a stress-strain diagram and the determination of its initial slope. 7.20. The higher the temperature above Tg , the more we are removed from the damping peak, so the lower is tan . 7.21. With increasing T, tan approaches to a limiting value, which is governed by

78

Problems Polymers

the network perfection, in particular by the number and length of loose chain ends, which as it were represent a fluid component.

8. Further properties
8.1. With an amorphous thermoplastic the brittleness temperature is about Tg, unless a secondary transition temperature Tsec occurs in the glassy region; in that case the brittleness temperature may be in the neighbourhood of Tsec. - with a semi-crystalline polymer the brittleness temperature is about Tg. - with a rubber-modified thermoplast the brittleness temperature is about the Tg of the dispersed rubber. 8.2. With an amorphous thermoplast the polymer softens over a rather short temperature interval from the glassy to the rubbery state. With a semi-crystalline polymer a certain amount of softening takes place at Tg; with further T- increase the stiffness drops very gradually up to the melting point Tm. 8.3. ISO-A from table ISO-A estimated from E(T) Vicat B from table Vicat B estimated from E(T) PBTP 60 40 180 150 PTFE 55 25 110 130 PP 60 50 90 140 PMMA PA 6.6 95 90 80 90 100 90 230 240 POM HDPE 115 45 115 65 165 180 70 110

The agreement is reasonable, though not perfect. Differences can be accounted to: - small differences between the samples taken, - differences in the time scales of measurements, - non-linear behaviour (in particular with the Vicat test). 8.4. a . The contradiction is a result of the slope of the log E - T curves (see MT 8.1.2). b. The ISO-A test is carried out at a higher E-modulus level than the Vicat test, thus giving more information for applications where a higher E is required, i.e. in loaded condition. 8.5. The blending of short glass fibres results in an increase of the E-modulus to e.g. its threefold over the whole temperature region. If the slope of the log E - T curve is small. such as with semi-crystalline polymers between Tg and T m ,, a vertical shift causes a considerable horizontal shift (see MT 8.1.2.), so a strong increase of the softening points, which are of importance in applications at higher temperatures.

Answers

79

8.6. The high coefficient of expansion can cause problems with shape stability and dimension tolerances in precision products. Moreover the high expansion causes more shrinkage upon cooling after a shaping process (see e.g. MT 11.3.2). 8.7. Advantages: - various applications in which, e.g. a handle remains cool ; also in foams for thermal insulation it is important that the cell walls do not conduct too much heat. Disadvantages: - With repeated changes of load the dissipated heat cannot easily flow away, so that the temperature of the object increases and it loses its good mechanical properties. - In a processing operation the heating-up and, thereafter, the cooling-down, are troublesome operations because of the low thermal conduction. In heating-up use can be made of friction or dielectric effects etc. (see MT 11.1.1), but with cooling the low heat conduction mainly governs the cycle time. 8.8. a. (see MT 8.1.4): = p + g + r + k ( polymer conduction + gas conduction + radiation + gas convection). b. Larger cells : more convection; larger k , so increases. c. Lower density: lower p , but higher r (radiation passes through cell walls), therefore a minimum in . d. g depends on the type of gas (heavier gas: lower g). e. The lowest possible minimum is g , the coefficient of conduction of the gas in rest. 8.9. Firstly: Though a high softening temperature is attractive, we should realise that the polymer has to undergo processing operations, which are carried out, for an amorphous polymer, well above T g , and for a crystalline polymer, above Tm . At these temperatures the polymer should be sufficiently chemically stable (for i.a. that reason PPE is being blended with PS). Secondly: For long-term applications not only the softening point is of importance, but also the degradation behaviour of the polymer; it may quite well happen that the polymer suffers unallowable degradation considerably below its softening T (see MT 8.1.5) 8.10. a. The term stress corrosion signifies damage (crack formation) under simultaneous presence of a mechanical stress and a certain environment, which both in itself are not fatal. b. The word corrosion suggests that a chemical reaction is involved, but in several types of stress corrosion no chemical degradation occurs, but it is caused by a lowering of the interfacial energy (such as with PE with a surface active agent). Another example of physical stress corrosion is the effect of e.g. tetrachloromethane on PC.

80

Problems Polymers

c. In those cases when chains are actually broken we have to do with real stress corrosion, such as the effect of ozone on a rubber vulcanisate. d. see above. 8.11. To avoid risks of electrostatic charging (sparks, explosions) anti-static additives are sometimes added to a polymer; also carbon black may be used (see MT 8.2.1). Polymers are sometimes used to shield electromagnetic fields; for this purpose they can be filled with metal fibres. In extreme cases conjugated dienes with an adequate doting can reach an electric conductance similar to that of metals. 8.12. The electrostatic charging is mainly controlled by the specific resistance of the material (sometimes also by the surface resistance). Decrease of the tendency to build up electrostatic charges can, therefore, be reached by lowering the specific resistance or the surface resistance, e.g. by adding carbon black or short metal fibres, or by using anti-static agents. The time scale is important because each application presents its own requirements as regards the dissipation of the charge, e.g. the formation of sparks or the attraction of dust (see MT 8.2.1). 8.13. Advantages: - The low tan of PE at very high frequencies enabled the application of radar in the 1930s. Disadvantages: - PE cannot be heated in a high-frequency electric field for, e.g., welding films. Special tricks are then needed, such as coating the film with a thin layer of a polymer with a higher tan . 8.14. a. With precision optics dimensional tolerances are extremely narrow, so that even the smallest amounts of recovery are disastrous for the performance. b. Advantages for spectacle glasses are: reduced weight and higher impact resistance. Disadvantages are the lower resistance against scratching (which can be improved by a coating), and, on the long term, the dimensional stability. 8.15. One of the main problems with a compact disc is the birefringence; the laser beam is split-up into two components, thus mutilating the signal. It is, therefore, important to reduce the birefringence; this can be done by minimising the chain orientations by choosing a PC with a low M. This is, moreover, beneficial for the ease of processing (very tiny details!), but is at the cost of the impact strength, which is, fortunately, for this type of application already high enough. 8.16. IIR (butyl rubber) is compared with other rubbers (SBR, NR) and then has a

Answers

81

20 to 30 lower permeability (inner tubes!). PE and PP are much more permeable than most of the other plastics. These phenomena can be attributed to the fact that the rate of diffusion in the rubbery phase (large free volume) is considerably higher than in the glassy-, and, in particular, in the crystalline phase. PE en PP contain, next to a crystalline phase, a continuous amorphous rubbery phase. 8.17. In the first place the rate of diffusion of LDPE is higher, as a consequence of the higher amorphous fraction. Secondly, for the same reason, LDPE has a higher solubility (the solubility in the crystalline phase is very low).

9. Polymeric compounds and composites

9.1. The molar masses are reflected in V1 en V2 (the molecular volumes). Increase of these values causes the first two (negative) terms to approach more to zero, so that a negative value of Gm becomes less probable. 9.2. is related to the enthalpy of blending, and is positive when only dispersion forces are present between the chains, so that, in view of the small contribution of the entropy of blending (the first two terms), Gm will be, in most cases, positive. With strong interactions becomes negative, and, at the same time, Gm.. 9.3. For 1 = 2 = 0.5 and V2 = V1 the formula reads:: Gm/V = kT[(ln 2)/V1 + /4Vs]. This is negative when /4Vs < (ln 2)/V 1 of: < 2.77(Vs/V1 ). 9.4. At very small values of 1 , ln 1 is so deeply negative that Gm is negative, independent of the values of V 1 , V2, etc. The same holds for 2 . The curve for Gm as a function of f thus proceeds, at its extreme values, steeply downward, independent of how its further course looks like, e.g. as indicated in the figure.
Gm

Figure A.26. Figure to the answer of problem 9.4.

9.5. When there is a UCST (upper critical solution temperature), then above that temperature the components are fully mixable. When cooling down to below this UCST a region of segregation can be reached, in particular when the volume fractions do not differ too much (see MT Figure 9.3, lower part "upside down")

82

Problems Polymers

9.6. Now about an LCST (lower critical solution temperature, see MT Figure 9.3). If we blend below this temperature, we can obtain a homogeneous mixture. Subsequent heating to above the LCST can result in segregation; the rapid cooling in an injection moulding machine does not allow to undo this. 9.7. For thermodynamic reasons mixability may be all right, but the last step in a mixing process, when the droplets are sufficiently small, is a diffusion process; this needs a considerable time, which is not always available in a blending process (in this way a "multiphase" blend could be formed, see MT Figure 9.4.d) 9.8. With very short "blocks", such as in the random-copolymer SBR, there is only a very small difference in entropy between the segregated and the homogeneous condition: no or hardly any change of entropy upon segregation. As the block length increases, this difference, however, increases, so that (G decreases (see MT 9.1.4). Segregation is, therefore, less complete in multi-block copolymers (such as polyethers- polyesters). 9.9. On the one hand in a flow field shear stresses are exerted on a droplet which cause a deformation into an ellipsoid; on the other hand the surface area of the droplet is increased by this deformation, so that the interfacial energy /R increases. When the size of the droplet decreases the first effect becomes smaller in comparison to the second one, which results in an equilibrium at which no droplets are broken-up (expressed in the capillary number Ca = /(/R), (see MT 9.1.5). 9.10. The droplets may have been broken down to their minimum diameter, but in the flow field they meet, they collide, and they can coagulate into bigger ones. The blending process is, therefore, an ongoing competition between break-up of droplets and coagulation. 9.11. The first process is: the deformation of droplets into an ellipsoid, until the droplet breaks up. The second process is: A droplet is, at high speed, extended into a thread, which is subject to instabilities (according to Rayleigh and Tomatika), and which, therefore, breaks up into a row of droplets (see MT 9.1.5). 9.12. Series- and parallel arrangements provide a first approximation of the properties; the properties which follow from these arrangements, are widely different. Combinations of series- and parallel arrangements (Takayanagi) are able to provide good approximations, but they require a large number of parameters (see MT 9.1.6). 9.13. If we apply Kerners formula to this PE, with cr = 0.552, for a continuous crystalline phase, in which an amorphous phase is dispersed, we find = 0.448, A = 1.14 (with = 0.33), Eo = 6000 MPa, E 1 = 1 MPa (both estimated provisionally), so

Answers

83

= 1/6000.
E = 6000 1 + 0.4481.14(1/6000 1)/(1/6000 + 1.14) 1 0.448(1/6000 1)/(1/6000 + 1.14) = 2378 MPa

With the inversed morphology, so crystalline dispersed in amorphous, we have: = 0.552, Eo = 1 MPa, E1 = 6000 MPa, so = 6000, A = 1.5, and, according to Kerner: E = 1 1 + 0.5521.5(6000 1)/(6000 + 1.5) 1 0552(6000 1)/(6000 + 1.5) = 4.08 MPa

With Nielsens formula we find: E = [0.55260000,2 + 0.44810,2]5 = 598.5 MPa. Apparently the result of Nielsens formula is closest to the observed E-modulus. With a few trials the values of Ecr = 6400 MPa and E am = 1.5 MPa render the best agreement, viz. E = 667 MPa, compared to the observed value of 660 MPa. Conclusion: In PE the crystalline phase and the amorphous phase are both continuous and "interwoven". 9.14. a. 150S + 2000B + 150S = (with S = 104, B = 54) = 15,600 + 208,000 + 15,600 = 139,200 g/mole; the weight fraction styrene w(PS) = 31,200/139,200 = 22,4%. b. The mass of a PS end chain is 15,600/6.1023 g = 2.601020 g. The volume of a PS domain = ( /6)25 3 1021 cm 3 = 8.1810 18 cm 3 , its mass V 1.05 = 8.591018 g. Divide: the number of PS chains in a domain is 330 c. Main cause: T g is lower for PS domains because of its low M , compared to "normal" PS; the difference may be as high as 10 C. (MT 3.4.3). Therefore at Tg -10 C softening of the PS domains can already occur. d. First we calculate the volume fraction of PS: 1 gramme SBS contains 0.224 g PS, so 0.224/1.05 = 0.213 cm 3 PS, and 0.776 g BR or 0.776/0.91 = 0.853 cm3 BR. Total volume is 1.066 cm3 , of which 0.213 cm3 PS, so = 0.213/1.066 = 0.20. Substituting this, with = 1000 and A = 1.5 (rubbermatrix), into the Kerner equation, results in: E/Eo = 1.62 so E = 4.9 MPa. The PS domains not only act as cross-links, but they also increase the stiffness. 9.15. When a foam contains closed cells, we can apply Kerners formula with = 0, so: E/Eo = (1 )/( 1 + /A). With = 0.5, Eo = 3200 MPa and A = 1.14 (for = 0.33) we find: E = 1112 MPa. An open cell foam is a co-continuous structure, sot that Nielsens formula can be n applied: En = 1 E n 1 + 2 E 2 with n = 0.2, 1 = 0.5, E1 = 3200 and E 2 = 0 (gas!); this 1 5 results in: E = (2 ) 3200 = 100 MPa. The conclusion is that the first foam has closed cells, the third one is completely open-celled and the second one contains partly open, partly closed cells. 9.16. For a "heavy" foam the gas fraction is small. The Kerner formula for a foam

84

Problems Polymers

with closed cells: E/Eo = (1 )/(1 + /A) can then be approximated by: E/Eo = 1 /A = 1 (1 + 1/A). For rubbers with A = 1.5 this becomes: E/Eo = 1 1.7, for glassy polymers with A = 1.14: E/Eo = 1 1.9 . When we use the approximation to 1 2 or (1 )2 (always for small ), and we realise that 1 is proportional to the foam density, then from this approximation it indeed follows, in particular for hard thermoplastics, that: E (:) d2 The approximation seems somewhat rough, but it appears to fit reasonably. 9.17. With the introduction of particles in a polymer the stiffness increases, but the tensile strength and the impact strength are hardly increased. It is, therefore, better, to use the term: stiffening additives. 9.18. Carbon black increases the tensile strength of an SBR vulcanisate to its 10 to 20-fold (which would, otherwise, be very low), as well as its abrasion resistance. Natural rubber can, because of its stereospecific (cis) chain structure crystallise under strain, and, therefore, reach higher values of its tensile strength; for a good abrasion resistance carbon black is also of importance with NR. 9.19. A continuous phase forms a "skeleton" in the blend. This is important, e.g. in the following cases: a second phase with a considerably higher softening temperature; when the other phase softens, the skeleton retains its stiffness. an electrically conducting polymer as a second phase provides the wanted conduction. an impermeable polymer as the second phase renders the blend impermeable only if it forms a continuous network. 9.20. See Figure A.27.
log E ISO

M1

M2

VICAT T C 0 100 200

Figure A.27. Figure to the answer of problem 9.20.

Apparently all four curves intersect at the same point, namely at the temperature at which A and B have the same E-modulus. Each blend has the same modulus at this temperature. This situation may occur when B is a glassy amorphous polymer and A

Answers

85

is semi-crystalline. The behaviour of Xenoy (PC + PBTP) resembles this picture. 9.21. See Figure A.28. The four curves have about the same shape; they are shifted along the T-axis. C and D are typically amorphous polymers. They are wholly miscible. The pattern resembles the one of Noryl (PPE + PS).
log E ISO C M3 M4 D VICAT

100

200

T C

Figure A.28. Figure to the answer of problem 9.21.

9.22. Apparently two opposing effects play a part: On the one hand the glass fibres take over a part of the stress, thus partially releasing the stress on the polymer. On the other hand, fibres cause, in particular at their ends, stress concentrations, which can initiate crack formation. 9.23. E o = 3000 MPa, E1 = 2 MPa, so = 1/1500, so can safely be assumed as zero in Kerners formula. From = 0.33 it follows: A = 1,14. The value aimed at is E = 2400 MPa. From Kerner it follows: 0.8 = (1 )/(1 + /1.14), so = 0.1175. easy rule of thumb: the percentage of modulus reduction is about twice the volume percentage of rubber, or also gas bubbles). The mass fraction is: = r/pr in which pr is the density of the polymer - rubber blend: = r/[(1 ) p + r] = 0.11750.9/(0.88251.4 + 0.11750.9) = 0.106/1.34 = 0.079 = 7.9 % w. 9.24. In the blend of the previous question we blend glass fibres. Now Eo = 2400 MPa and E1= 75,000 MPa, so = 31.25. E should be increased up to 3000 MPa so E/Eo = 1.25. When we substitute this into the Kerner equation, again with A = 1.14 (not wholly correct since n has become somewhat higher because of the blending with rubber), Kerners formula then becomes: 1.25 = [1 + (1.1430.25/32.29)]/[1 (30.25/32.39)], from which it follows:: = 0.112 of 11.2 %. The mass fraction is now, calculated on the polymer-rubber blend ( pr = 1.34): = 0.1122.4/(0.8881.34 + 0.1122.4) = 0.184 = 18.4 %w. 9.25. In a composite one aims at the highest possible stiffness E at the lowest possible , so at a high as possible value of E/, and this is the square of the sound

86

Problems Polymers

velocity in the material. 9.26. The limit is reached when the glass fibres are so long that they can be considered as a continuous phase. We have then a parallel arrangement of polymer and glass, with 1 = 0.75, 2 = 0.25, E1 = 3 and E 2 = 75 GPa. The E-modulus of the composite then follows from: E = 0.75 3 + 0.25 75 = 21 GPa. 9.27. The Einstein formula E = E o (1 + 2.5 ) is valid if is rather high and is rather small, and it gives, for the transverse direction: E = 3 (1 + 2.5 0.25) = 4.9 GPa. (The calculation with Kerner is better, but we now aim at a simple first approximation). 9.28. In an isotropic composite (chaotic distribution of fibre direction) E is given by: E = 0.2 Epar + 0.8 Etransv. The first term follows from the answer to Question 19, the second one from 20, and the result is: E = 0.2 21 + 0.8 4.9 = 8.1 GPa. This value, therefore, approximates the highest attainable value with 25 vol% of glass fibres. 9.29. a. Tg glass fibres no change rubber particles no change (there is an extra transition, namely of the rubber phase) lower (shift effect E(T) curves, see MT 8.1.2) no change lower lower, due to soft inclusions lower about the same considerably higher (see MT 9.3)

b. Vicat c. T m d. ISO-HDT e. E-modulus

higher

no change higher (see b.) higher due to hard inclusions f. tensile strength slightly higher g. heat conduction higher due to glass fibres h. impact strength sometimes better, sometimes worse

11. Processing techniques

11.1. a. Heating from the outside (metal or infrared); controlled by heat conduction ; b. Friction (between granules, controlled by the friction coefficient ), or in the liquid, (controlled by the viscosity ); c. High-frequency dielectric heating, controlled by the dielectric loss factor tan ; d. Ultrasonic sound, controlled by the mechanical loss factor tan .

Answers

87

11.2. Casting is, of course, possible when the material is fluid enough. This is the case with some thermosets (casting resins), which, together with a second component, can be cast into a mould and then cured at a higher temperature. With thermoplastics casting is, in general, not possible because of the high melt viscosity, except when polymerisation has still to take place; the monomer or prepolymer is polymerised in the mould. This is, e.g. possible with PMMA and PA-6 (from methyl-methacrylate and caprolactam, respectively). 11.3. With semi-crystalline polymers the rubbery region is partly or wholly masked by the crystalline phase. Only very high-molecular grades types have part of the rubbery region left above their Tm. Processing techniques in which the rubber-elastic behaviour is essential, such as calendering and vacuum forming, are therefore less suitable for crystalline polymers, unless they have a very high molar mass. 11.4. When moulding thermoplastics the mould has to be heated, and, after shaping. cooled down to enable the taking-out of the product. This requires repeated changes in temperature of large blocks of metal. With thermosets and rubbers the temperature can remain high after shaping to enable curing or vulcanisation. 11.5. a. Two methods may be applied: injection moulding, forming from an extruded sheet. Both can lead to economic production. With the popular brown or white coffeecup the second method (two steps) has, apparently, won the competition. Transparent PS cups are mostly injection-moulded. b. The bottom of the injection-moulded cup shows the point of injection. Moreover, after sheet forming the wall thickness is less uniform than after injectionmoulding. c. An injection-moulded cup will gradually soften upon heating, whereas a vacuumformed one first contracts into a flat sheet. d. The choice between the possible processing methods is i.a. governed by the economy of production; this is strongly related to the production size (see MT 11.1.2). 11.6. a . At a certain injection pressure p , a minimum temperature T is needed to reduce the viscosity sufficiently to enable complete mould filling. A lower T requires a higher pressure. b. The mould B requires a higher pressure to be filled; so it has a higher flow resistance (narrower) or a longer flow path (l/d). c. In B the melt flows less easily than in A (higher viscosity); at first sight we could suppose that B has a lower melt flow index (m.f.i.) d. If, however, the m.f.i. of B appears to be higher than that of A, then A has a much

88

Problems Polymers

broader molar mass distribution; it is stronger non-Newtonian, so that it flows better at the high shear rates in the machine, despite of its lower m.f.i. e. At further increase of the pressure the mould halves may be pushed apart, because the clamping force is insufficient ("flash"). f. A too high temperature may lead to degradation of the polymer. g. A too low T involves the risk of incomplete or non-homogeneous melting of the material. 11.7. a. When the pressure in the mould is 800 bar (1 bar = 105 N/m2 ), the material exerts on the area of 100 cm2 a force of 800 105 10-2 N = 800 kN ("80 tonnes"). This is thus the minimum required clamping force. b. In the mould a volume is present of 10100.3 = 30 cm3. At atmospheric pressure and 220 C this would mean 30/1.34 = 22.4 grammes. At 800 bar the specific volume is, however, a fraction 800 7.5 10-5 = 0.06 or 6% smaller, the mass is 6% greater, or 22.4 1.06 = 23.7 g. For the end product, at 1 bar and 20 C this would mean a volume of 23.7 1.05 = 24.9 cm3 or 24.9 / 30 = 0.83 times the volume of the mould. This means a volume shrinkage of 17% or a linear shrinkage of about 5%. The dimensions would be about 9.5 cm 9.5 cm 2.85 mm. d. After filling of the mould extra space becomes available by cooling and solidification along the walls; maintaining the pressure until the polymer at the injection gate is also solidified will then enable several percents extra supply (after-pressure). 11.8. PE, PP and PA are semi-crystalline polymers; melting and solidification go accompanied by a (though gradual) volume jump. PS, PVC and PC are amorphous thermoplastics; upon solidification they show no volume jump, but only a bend in the V-T relation. 11.9. The outer side cools and solidifies first; the inner part remains fluid the longest; its shrinkage in the final stage of solidification is constrained. Its density will, therefore, be lower than at the outer side and tensile stresses are being built up, even with the risk of void formation. 11.10. Powder (blend of resin and curing agent) is deposited on the metal via a fluid bed or electrostatic. Upon heating it flows and, at the same time, it is cured. The course of the viscosity in an analogous situation is given in MT Figure 11.8. Now, however, too slow curing results in dripping of the fluid, while a too rapid cure stagnates the flow so that the surface is not smooth. 11.11. Between the rolls of a calendar enormously high pressures are built up, needed to transport and shape the material at high speed. These pressures cause

Answers

89

bending of the rolls, even though they are very heavily constructed. The result is that the sheet is too thick in the middle. By crossing the axles of the rolls this can be largely compensated (see MT 11.4.1). 11.12. With calendering and also with vacuum forming the polymer must be in the rubbery condition; in the first case the sheet must be taken off from the last roll under a certain stress, which it can withstand in the rubbery phase only. With vacuum forming the heated sheet is placed on the mould; this is only possible when it possesses enough coherence and handability. The word molten is therefore not relevant in these cases. In other processes such as blow-moulding and film extrusion some rubber-elastic response also plays a part, though considerably less than in the examples mentioned before. 11.13. First: a broad molar mass distribution helps to lower the apparent viscosity at high shear stresses and shear rates, so that thinner films can be produced at the same pressure. Secondly: a broader MMD leads to a higher value of Mz, the average which is mainly responsible for the elastic behaviour of the melt, and thus for the recoil after deformation and solidification. These frozen-in rubber-elastic deformations cause the behaviour of shrink-films. 11.14. In rotation moulding the shear stresses and shear rates are extremely low; the polymer particles flow together into a homogeneous wall under the influence of gravity and surface stresses only. The relevant viscosity is, therefore, the "Newtonian" viscosity, without the complication of shear-thinning. So the width of the MMD is not of primary importance: only the weight average, Mw counts. A consequence is, that for a given m.f.i. the distribution should be narrow (see MT 5.4). At the same time, with a narrow distribution Mn is higher since it closer to Mw, This is important for the resistance against stress corrosion, since this is related to Mn (good for oil tanks). 11.15. Melt fracture results from a too large rubber-elastic deformation. This is governed by the polymer behaviour and by the rate of deformation. The strain rate must be lowered, which may be accomplished by a lower, and therefore unattractive, rate of production. The time scale of the elastic deformation can also be increased in another way, namely by making the conical channel to the die more pointed; the same strain is then reached after a longer time. This goes at the cost of extra pressure and power, since the resistance of the channel is increased. 11.16. When a pipe is extruded, the melt has to flow round the spider at which the

90

Problems Polymers

torpedo is suspended, and, thereafter, flow together perfectly. The risk is an imperfect welding, causing weaker spots at the weld lines or spider lines. Also with injection moulding weld lines occur frequently. The best known example is the eye for the handle of a pail, which is the spot at which the stresses are highest during use! 11.17. Adiabatic: no exchange of heat with the environment. For an extruder this means that no heat is supplied from the outside, but the energy for melting is wholly supplied by the internal friction. This is ideal for homogeneous heating of the material, without the trouble of low heat conduction. This can be realised with larger diameters. In most cases the heating-up is a combination of wall heating and internal friction. 11.18. Along the screw of an extruder both transport flow and transverse flow occur. The latter provides a circulation flow, which promotes mixing and homogeneity. 11.19. a. Q is the throughput in volume per unit time. p is the pressure in front of the die, N is the rotational speed of the screw. is the viscosity of the melt, a is a geometrical constant, related to the screw geometry. b ditto, c is a geometrical constant related to the die dimensions. b. The maximum throughput is 80 dm3/hour; this is reached when there is no die at the end of the extruder (drag flow only). The maximum pressure (for a given viscosity of the polymer) is 400 bar (extrapolated); this is reached when the die is closed. c. When the rotation rate is halved, the screw characteristic shifts down in a parallel way, and it then intersects the vertical axis at 40 dm3 /hour. Both the throughput and the pressure are then halved. d. When the die resistance is doubled, the slope of the line through the origin is halved. From the graph it can then be deduced: the rate of throughput, Q is lowered from 80 to 48 dm3 /hour, the pressure p increases from 100 to 160 bar. e. When the viscosity becomes three times higher, both slopes are three times smaller; the rate of throughput, Q, remains 60 dm3 /hour (this also follows from Q = N a c/(b + c), in which and p do not appear), but the pressure p becomes 3 times higher, so 300 bar. f. As a result of the non-Newtonian behaviour both expressions for the pressure flows (bp/ and cp/) are no longer valid. The curve for the die is now curled upward since the apparent viscosity decreases with increasing shear stress. Also the shape of the screw characteristic changes.

Answers

91

Q (dm3/h) 80 70 60 50 40 30 20 (c) 10 0 0 100 200 300 400 Figure A.29. Figure to the answer of problem 11.19 f. (b) (f) (d) (e) (e)

p (bar)

11.20. a . Internal stresses are physically present stresses; frozen-in stresses are frozen-in chain orientations. The former result from uneven cooling from the melt: solidification first takes place at the cooled outer wall, while the inner parts solidify much more slowly, thus creating internal tensile stresses. The stress is, averaged over the cross-section, zero, but there is a stress distribution, composed of compression- and tensile stresses. Frozen-in stresses originate from the rubber-elastic behaviour of the melt: the rubber- elastic deformations (chain orientations) are frozen-in upon cooling and remain present as latent stresses. b. Internal stresses (cooling- or shrinkage stresses) become apparent when from a ring, cut from a pipe, a small segment is removed; the ring will tend to close because the tensile stresses are mainly at the inside. Frozen-in stresses can be detected by heating a piece of a pipe to above Tg or Tm; the rubber-elastic stresses then get their chance to deform the sample to shorter and thicker dimensions. c. Shrinkage stresses contribute to the total stress on a volume element in a pressure pipe. In the pipe a circumferential stress is present, on which, at the inner side, the shrinkage stress is superimposed. This may limit the life-time of the pipe. Frozen-in orientations are present in the length direction of the pipe and may cause anisotropy: in the transverse- (circumferential-) direction the material is weaker, and this direction is just important for a pressure pipe! Moreover, at higher temperatures, the pipe may deform. d. With thick-walled pipes cooling proceeds more slowly so that the shrinkage

92

Problems Polymers

differences and the internal stresses become greater. Thin-walled pipes are extruded at a higher rate; the deformation of the melt takes place at a shorter timescale, resulting in a higher rubber-elastic deformation. Moreover, the chain orientation is more readily frozen-in because of the quicker cooling. 11.21. a . PETP crystallises only slowly, and it stays amorphous in a normal injection-moulding process, after rapid cooling of the melt. b. PETP has a Tg of about 70 C and a Tm of 260 C. If the pre-form would be semicrystalline, then it could only be blown-up at a temperature above Tm. At such a T the viscosity is so low (condensation polymer, so not a very high M!), that blowing is not possible. Amorphous PETP can, at some distance above its Tg , easily be deformed in the rubbery condition. c. During the blowing process the bottle is simultaneously stretched in the length direction so that a bi-axial orientation is reached, which is useful for the properties. d. An amorphous bottle would soften completely already above 70 C, so it would be hardly usable. e. Chain orientation, also bi-axial, promotes rapid formation of nuclei, which is necessary to allow the polymer to crystallise sufficiently at a rather short time scale. f. Because of the very large number of nuclei formed upon chain orientation, the crystallites remain small enough to avoid appreciable light scattering. g. During spinning of polyester fibres such a high uni-axial orientation is brought away that rapid crystallisation occurs. This is, of course, essential for a good fibre. 11.22. a. Higher m.f.i.: lower M, so, in general, less elastic behaviour of the melt and a lower die-swell. b. Larger ratio of m.f.i. at 10 and at 2.16 kg indicates a more strongly pronounced non-Newtonian behaviour, so a broader MMD with relatively more long chains (higher Mz), resulting in stronger elastic response of the melt and a higher dieswell. c. At a higher rate of throughput the rate of deformation of the melt is higher, resulting in more elastic response and a higher die-swell. d. At a higher temperature the relaxation times are shorter, so that there is less chance of chain orientation when the melt leaves the die. e. Most of the rubber-elastic strain is brought about in the conical channel leading to the die (elongational flow).The longer the melt stays in the (straight) end channel, the more the orientations are able to relax, and the lower the die-swell will be. 11.23. PVC is an amorphous thermoplastic; during the heating-up along the screw of an extruder the viscosity is, after passing Tg , still very high and it decreases only slowly. Compression of the mass must, therefore, take place over a considerable part

Answers

93

of the screw length. PA, on the contrary, is slightly above its Tm already a relatively low-viscous fluid (also since its molar mass is not very high). The compression zone may, therefore, be short. 11.24. Structural foam (or integral foam) is a foam where the walls are solid. Advantages: Though its production, e.g. in an injection moulding machine, requires special tricks, the injection pressures and the mould clamping forces required are much lower than with a solid product. The surface is, compared with a normal foam, smooth. The product is considerably lighter than a solid product (e.g. half). The solid skin provides extra bending- and buckling resistance. 11.25. a. blow extrusion, b. casting of monomer (caprolactam) and in-situ polymerisation, c. sheet extrusion, followed by vacuum forming, or rather SPPF slightly below Tm. d. injection moulding, e. calendering and cutting, f. centrifugal casting from a paste, followed by vulcanisation, g. calendering, h. compounding, extrusion, moulding, vulcanisation, i. film extrusion (balloon), cutting, welding, j. pre-foaming of pentane containing beads, final foaming in the mould, k. moulding, vulcanisation, l. moulding and vulcanisation, or: injection moulding and vulcanisation, m. sheet extrusion or calendering, thereafter sheet forming (vacuum or with extra help), n. casting from monomer, in-situ polymerisation, o. wire-coating extrusion round a copper wire, p. extrusion, q. impregnating of glass wires with a blend of resin and curing agent, winding round a mandrel, curing, r. rotation moulding from powder, s. moulding of a powdery compound, curing, t. sheet extrusion, vacuum forming, u. machining of a moulded block or moulding or injection moulding (i.a dependent on production size), v. sheet extrusion, bending, welding, w. spraying a powdery mixture of the two components and a blowing agent onto the wall; it then foams and hardens, x. resin/curing agent blend apply to glass mat (hand-lay-up), curing,

94

Problems Polymers

y. injection moulding of pre-form; then blowing-up with bi-axial stretching, z. sheet extrusion with pentane-injection, vacuum forming.