The fractionation and characterisation of propylene-ethylene random copolymers

Hele tekst

(1)The fractionation and characterisation of propylene-ethylene random copolymers by. Gareth Harding. Thesis presented in partial fulfilment of the requirements for the degree of Master of Science at the University of Stellenbosch. Study leader Dr. AJ van Reenen. Stellenbosch December 2005.

(2) I, the undersigned, hereby declare that the work contained in this thesis is my own original work and that I have not previously in its entirety or in part submitted it at any university for a degree.. Signature:…………………... Date:……………………….

(3) Abstract This study involves the fractionation and characterisation of three propyleneethylene random copolymers. The fractionation technique used in the study was temperature rising elution fractionation (TREF). The TREF fractions were subsequently analysed offline by crystallisation analysis fractionation (CRYSTAF), differential scanning calorimetry (DSC), 13C NMR, high-temperature gel-permeation chromatography (HT-GPC), and wide-angle x-ray diffraction (WAXD). The effect of the ethylene comonomer on the crystallisability of the propylene was investigated, along with the effect of the comonomer on the type of crystal phase formed during the crystallisation. The results show that the comonomer inhibits the crystallisation of the copolymer and that as the ethylene content increases, the crystallisation and melting points decrease. It was also shown that the higher the ethylene content, the more of the γ-phase crystal type is formed. The distribution of the comonomer throughout the copolymers was also investigated. The results show that there is an uneven distribution of the comonomer with most of the comonomer accumulating in the amorphous areas, and very little actually being incorporated in the crystalline regions. It was also observed that the fractions eluting at the highest temperatures had considerably higher polydispersities and lower molecular weights than the fractions eluting just before them. The highest temperature fractions also have lower melting and crystallisation temperatures than the preceding fractions. This has been attributed to a nucleation effect by the sand support used during the TREF fractionation..

(4) Opsomming Hierdie navorsing behels die fraksioneering en karakterisering van drie propileen-etileen statistiese kopolimere. Die fraksioneering tegniek wat gebruik is in die navorsing is temperatuurstyging elueering fraksioneering (TREF). Die TREF fraksies was toe geanaliseer deur kristallisasie analise fraksioneering (CRYSTAF), differensiële skandeer kalorimetrie (DSC), 13C kern magnetise resonans spektroskopie (NMR), hoë-temperatuur jel-permeasie kromatografie (HT-GPC), en wye-hoek xstraal diffraksie (WAXD). Die effek van die etileen ko-monomeer op die kristallisasie van die propileen word geondersoek, asook die effek van die ko-monomeer op die tipe kristal wat gevorm is gedurende die kristallisasie. Die resultate dui aan dat die komonomeer kristallisasie van die kopolimeer inhibeer en dat as die etileen inhoud verhoog word, dan daal die smelting and kristalisasie temperature. Dit is ook bewys dat hoe hoor die etileen inhoud, hoe meer van die γ-tipe kristal word gevorm. Die verspreiding van die ko-monomeer in die ko-polimere word ook ondersoek. Die resultate dui aan dat daar ‘n oneweredige verspreiding van die ko-monomeer is en dat die meeste van die ko-monomeer versamel in die amorfe gedeeltes van die kopolimere, met baie min wat eintlik in die kristallyn omgewing is. Dit was ook waargeneem dat die fraksies wat elueer by die hoogste temperature aansienlike hoë polidispersiteite en laer molekulêre massas het as die fraksies wat voor hulle geelueer is. Die fraksies van die hoogste temperature het ook lear smeltpunte en kristallisasie temperature as die vorige fraksies. Dit kan toegeskryf word aan ‘n kernvorming proses..

(5) Acknowledgements I would like to thank the following people for their help and support in getting me through this study.. Dr. AJ van Reenen - for his guidance throughout the study Valerie Grumel - for all the HT-GPC work Derick Mcauley - for all the CRYSTAF work Remy Bucher at Ithemba Labs for all the WAXD work Elsa Malherbe - for the NMR work The Olefins research group My parents - for their continued support throughout my studies.

(6) Abbreviations TREF. Temperature rising elution fractionation. P-TREF. Preparative temperature rising elution fractionation. A-TREF. Analytical temperature rising elution fractionation. CRYSTAF. Crystallization analysis fractionation. DSC. Differential scanning calorimetry. NMR. Nuclear magnetic resonance. HT-GPC. High-temperature gel permeation chromatography. WAXD. Wide-angle x-ray diffraction. LDPE. Low density polyethylene. LLDPE. Linear low density polyethylene. HDPE. High density polyethylene. PP. Polypropylene. MAO. Methylaluminoxane. IR. Infrared. RI. Refractive index. o-DCB. ortho-Dichlorobenzene. TCB. Trichlorobenzene. TMB. Trimethylbenzene. SEC. Size-exclusion chromatography. GFC. Gel filtration chromatography. DGMBE. Diethylene-glycol-monobutylether. DMP. Dimethyl phthalate. PD. Polydispersity.

(7) CONTENTS List of figures...............................................................................................................III List of tables............................................................................................................... VII Chapter 1. Introduction............................................................................................1 1.1 General introduction ......................................................................................1 1.2 Aims...............................................................................................................2 1.3 References......................................................................................................2 Chapter 2. Background ............................................................................................4 2.1 Polyolefins: A brief historical overview ........................................................4 2.2 Polymerisation chemistry: An overview........................................................5 2.2.1 General mechanism of transition metal catalysed polymerisation ........5 2.2.2 Polymerisation control mechanisms and stereochemistry .....................8 2.2.3 The evolution of the transition metal catalysts ....................................11 2.3 Commercial polypropylene..........................................................................13 2.3.1 Varieties of polypropylene manufactured............................................13 2.3.1.1 Polypropylene homopolymer...........................................................13 2.3.1.2 Impact copolymers...........................................................................14 2.3.1.3 Random copolymers ........................................................................15 2.3.2 Crystallinity types ................................................................................16 2.4 Fractionation techniques ..............................................................................18 2.4.1 Fractionation by crystallinity ...............................................................18 2.4.1.1 Fractionation mechanism and crystallisation theory........................18 2.4.1.2 TREF................................................................................................20 2.4.1.3 CRYSTAF........................................................................................26 2.4.2 Molecular weight fractionation............................................................27 2.4.2.1 Analytical techniques.......................................................................27 2.4.2.2 Preparative techniques .....................................................................28 2.4.3 Solvent extraction ................................................................................28 2.5 Concluding remarks and methodology ........................................................29 2.6 References....................................................................................................30 Chapter 3. Experimental ........................................................................................39 3.1 TREF............................................................................................................39 3.1.1 The crystallisation step ........................................................................39 3.1.2 The elution step....................................................................................40 3.2 High-temperature GPC ................................................................................42 3.3 CRYSTAF....................................................................................................42 3.4 DSC..............................................................................................................43 3.5 NMR ............................................................................................................43 3.6 WAXD .........................................................................................................43 3.7 References....................................................................................................44 Chapter 4. Results and Discussion ........................................................................45 4.1 The unfractionated samples .........................................................................45 4.1.1 Molecular structure analysis ................................................................45 4.1.2 Crystallisation and melting ..................................................................48 4.1.3 Crystal phase analysis ..........................................................................50 4.2 Optimising the TREF fractionation .............................................................53 4.3 The fractionated material .............................................................................58 4.3.1 TREF analysis......................................................................................58. I.

(8) 4.3.2 Molecular structure analysis ................................................................61 4.3.3 Crystallisation and melting ..................................................................67 4.3.4 Crystal phase analysis ..........................................................................71 4.4 References....................................................................................................74 Chapter 5. Conclusions..........................................................................................78 5.1 Conclusions..................................................................................................78 5.2 Future work..................................................................................................79 Appendix A HT-GPC data .......................................................................................80 Appendix B 13C NMR data ......................................................................................82 Sample A..................................................................................................................82 Sample B..................................................................................................................83 Sample C..................................................................................................................84 Appendix C CRYSTAF data....................................................................................85 Sample A..................................................................................................................85 Sample B..................................................................................................................92 Sample C................................................................................................................100 Appendix D DSC data ............................................................................................108 Sample A................................................................................................................108 Sample B................................................................................................................115 Sample C................................................................................................................123 Appendix E WAXD data .......................................................................................131 Appendix F DSC data of the samples analysed by WAXD...................................133 Original samples ....................................................................................................133 Fractions of sample A ............................................................................................134 Fractions of sample B ............................................................................................136 Fractions of sample C ............................................................................................137. II.

(9) List of figures Figure 2.1 The Ziegler-Natta polymerisation mechanism. ............................................8 Figure 2.2 Polymerisation control mechanisms.............................................................9 Figure 2.3 Catalyst active sites on 1,0,0 and 1,1,0 cuts of the MgCl2 crystal..............10 Figure 2.4 Coordination of internal donors ensuring isospecific active sites. .............10 Figure 2.5 Types of polypropylene tacticity. ...............................................................14 Figure 3.3.1 Setup used for the crystallisation step of preparative TREF. ..................39 Figure 3.3.2 Temperature profile used for the slow cooling of the samples used for TREF............................................................................................................................40 Figure 3.3.3 An illustration of the elution column packing method............................41 Figure 3.3.4 The TREF elution setup...........................................................................41 Figure 4.1 13C NMR spectrum of original sample A in the region between 10 and 55 ppm. .............................................................................................................................46 Figure 4.2 13C NMR spectrum of original sample B in the region between 10 and 55 ppm. .............................................................................................................................46 Figure 4.3 13C NMR spectrum of original sample C in the region between 10 and 55 ppm, with peak assignments. .......................................................................................46 Figure 4.4 The structure of isotactic polypropylene with a single ethylene unit inserted between the regioregular propylene units. ...................................................................47 Figure 4.5 CRYSTAF curves of all three original samples A, B, and C. ....................49 Figure 4.6 DSC melting curves of the three original samples A, B, and C. ................49 Figure 4.7 WAXD analysis of the three original samples A, B, and C after melt pressing and slow cooling of the samples....................................................................51 Figure 4.8 TREF results of the first fractionation of sample A. The fractionation temperatures were 25, 50, 75, 95, and 120°C. .............................................................53 Figure 4.9 TREF results of the second fractionation of sample A. The fractionation temperatures were 25, 50, 75, 95, 120, and 140°C. .....................................................54 Figure 4.10 TREF results of the third fractionation of sample A. The fractionation temperatures were 25, 50, 75, 85, 95, 105, 120, and 140°C. .......................................54 Figure 4.11 TREF results of the fourth fractionation of sample A..............................56 Figure 4.12 The TREF results of the fifth, and final, fractionation of sample A. .......57 Figure 4.13 Comparison of eluting 200 mL v. 400 mL for sample B. ........................58 Figure 4.14 A comparison of the fractionation of all three samples, A, B, and C.......60 Figure 4.15 HT-GPC molecular weight results for the fractions of sample A illustrating the weight average molecular weight and polydispersity of the fractions.61 Figure 4.16 HT-GPC molecular weight results for the fractions of sample B illustrating the weight average molecular weight and polydispersity of the fractions.62 Figure 4.17 HT-GPC molecular weight results for the fractions of sample C illustrating the weight average molecular weight and polydispersity of the fractions.63 Figure 4.18 13C NMR spectrum of the 25°C (C1) fraction of sample in the region between 10 and 55 ppm. ..............................................................................................64 Figure 4.19 Suggested chain structures for the room temperature fraction of sample C. ......................................................................................................................................64 Figure 4.20 13C NMR spectra of fractions C7, C9, and C11 in the region between 10 and 55 ppm...................................................................................................................65. III.

(10) Figure 4.21 A waterfall plot of the DSC melting endotherms of the first 8 fractions of sample A. .....................................................................................................................67 Figure 4.22 A waterfall plot of the DSC melting endotherms of the last 7 fractions of sample A. .....................................................................................................................68 Figure 4.23 CRYSTAF curves of selected fractions of sample A...............................68 Figure 4.24 DSC melting endotherms of fraction A12 obtained at different heating rates. The rates were 5, 10, and 20°C/minute. .............................................................70 Figure 4.25 WAXD results for fractions A7, A9, and A11 of sample A.....................72 Figure 4.26 DSC melting endotherms of the first and second heating cycles of fraction A7 of the sample which was slow-cooled for WAXD analysis...................................73 Figure B.1 13C NMR spectrum of the 25°C fraction of sample A (A1). .....................82 Figure B.2 13C NMR spectra of the selected fractions of sample A (A7, A9, and A11). ......................................................................................................................................82 Figure B.3 13C NMR spectrum of the 25°C fraction of sample B (B1).......................83 Figure B.4 13C NMR spectra of the selected fractions of sample B (A6, A8, and A10). ......................................................................................................................................83 Figure B.5 13C NMR spectrum of the 25°C fraction of sample C (C1).......................84 Figure B.6 13C NMR spectra of the selected fractions of sample C (A7, A9, and A11). ......................................................................................................................................84 Figure C.1 CRYSTAF results for the 25°C fraction of sample A (A1).......................85 Figure C.2 CRYSTAF results for the 45°C fraction of sample A (A2).......................85 Figure C.3 CRYSTAF results for the 65°C fraction of sample A (A3).......................86 Figure C.4 CRYSTAF results for the 75°C fraction of sample A (A4).......................86 Figure C.5 CRYSTAF results for the 80°C fraction of sample A (A5).......................87 Figure C.6 CRYSTAF results for the 85°C fraction of sample A (A6).......................87 Figure C.7 CRYSTAF results for the 90°C fraction of sample A (A7).......................88 Figure C.8 CRYSTAF results for the 95°C fraction of sample A (A8).......................88 Figure C.9 CRYSTAF results for the 100°C fraction of sample A (A9).....................89 Figure C.10 CRYSTAF results for the 105°C fraction of sample A (A10).................89 Figure C.11 CRYSTAF results for the 110°C fraction of sample A (A11).................90 Figure C.12 CRYSTAF results for the 115°C fraction of sample A (A12).................90 Figure C.13 CRYSTAF results for the 120°C fraction of sample A (A13).................91 Figure C.14 CRYSTAF results for the 125°C fraction of sample A (A14).................91 Figure C.15 CRYSTAF results for the 140°C fraction of sample A (A15).................92 Figure C.16 CRYSTAF results for the 25°C fraction of sample B (B1). ....................92 Figure C.17 CRYSTAF results for the 45°C fraction of sample B (B2). ....................93 Figure C.18 CRYSTAF results for the 65°C fraction of sample B (B3). ....................93 Figure C.19 CRYSTAF results for the 75°C fraction of sample B (B4). ....................94 Figure C.20 CRYSTAF results for the 80°C fraction of sample B (B5). ....................94 Figure C.21 CRYSTAF results for the 85°C fraction of sample B (B6). ....................95 Figure C.22 CRYSTAF results for the 90°C fraction of sample B (B7). ....................95 Figure C.23 CRYSTAF results for the 95°C fraction of sample B (B8). ....................96 Figure C.24 CRYSTAF results for the 100°C fraction of sample B (B9). ..................96 Figure C.25 CRYSTAF results for the 105°C fraction of sample B (B10). ................97 Figure C.26 CRYSTAF results for the 110°C fraction of sample B (B11). ................97 Figure C.27 CRYSTAF results for the 115°C fraction of sample B (B12). ................98 Figure C.28 CRYSTAF results for the 120°C fraction of sample B (B13). ................98 Figure C.29 CRYSTAF results for the 125°C fraction of sample B (B14). ................99. IV.

(11) Figure C.30 CRYSTAF results for the 140°C fraction of sample B (B15). ................99 Figure C.31 CRYSTAF results for the 25°C fraction of sample C (C1). ..................100 Figure C.32 CRYSTAF results for the 45°C fraction of sample C (C2). ..................100 Figure C.33 CRYSTAF results for the 65°C fraction of sample C (C3). ..................101 Figure C.34 CRYSTAF results for the 75°C fraction of sample C (C4). ..................101 Figure C.35 CRYSTAF results for the 80°C fraction of sample C (C5). ..................102 Figure C.36 CRYSTAF results for the 85°C fraction of sample C (C6). ..................102 Figure C.37 CRYSTAF results for the 90°C fraction of sample C (C7). ..................103 Figure C.38 CRYSTAF results for the 95°C fraction of sample C (C8). ..................103 Figure C.39 CRYSTAF results for the 100°C fraction of sample C (C9). ................104 Figure C.40 CRYSTAF results for the 105°C fraction of sample C (C10). ..............104 Figure C.41 CRYSTAF results for the 110°C fraction of sample C (C11). ..............105 Figure C.42 CRYSTAF results for the 115°C fraction of sample C (C12). ..............105 Figure C.43 CRYSTAF results for the 120°C fraction of sample C (C13). ..............106 Figure C.44 CRYSTAF results for the 125°C fraction of sample C (C14). ..............106 Figure C.45 CRYSTAF results for the 140°C fraction of sample C (C15). ..............107 Figure D.1 DSC data for the 25°C fraction of sample A (A1). .................................108 Figure D.2 DSC data for the 45°C fraction of sample A (A2). .................................108 Figure D.3 DSC data for the 65°C fraction of sample A (A3). .................................109 Figure D.4 DSC data for the 75°C fraction of sample A (A4). .................................109 Figure D.5 DSC data for the 80°C fraction of sample A (A5). .................................110 Figure D.6 DSC data for the 85°C fraction of sample A (A6). .................................110 Figure D.7 DSC data for the 90°C fraction of sample A (A7). .................................111 Figure D.8 DSC data for the 95°C fraction of sample A (A8). .................................111 Figure D.9 DSC data for the 100°C fraction of sample A (A9). ...............................112 Figure D.10 DSC data for the 105°C fraction of sample A (A10). ...........................112 Figure D.11 DSC data for the 110°C fraction of sample A (A11). ...........................113 Figure D.12 DSC data for the 115°C fraction of sample A (A12). ...........................113 Figure D.13 DSC data for the 120°C fraction of sample A (A13). ...........................114 Figure D.14 DSC data for the 125°C fraction of sample A (A14). ...........................114 Figure D.15 DSC data for the 140°C fraction of sample A (A15). ...........................115 Figure D.16 DSC data for the 25°C fraction of sample B (B1).................................115 Figure D.17 DSC data for the 45°C fraction of sample B (B2).................................116 Figure D.18 DSC data for the 65°C fraction of sample B (B3).................................116 Figure D.19 DSC data for the 75°C fraction of sample B (B4).................................117 Figure D.20 DSC data for the 80°C fraction of sample B (B5).................................117 Figure D.21 DSC data for the 85°C fraction of sample B (B6).................................118 Figure D.22 DSC data for the 90°C fraction of sample B (B7).................................118 Figure D.23 DSC data for the 95°C fraction of sample B (B8).................................119 Figure D.24 DSC data for the 100°C fraction of sample B (B9)...............................119 Figure D.25 DSC data for the 105°C fraction of sample B (B10).............................120 Figure D.26 DSC data for the 110°C fraction of sample B (B11).............................120 Figure D.27 DSC data for the 115°C fraction of sample B (B12).............................121 Figure D.28 DSC data for the 120°C fraction of sample B (B13).............................121 Figure D.29 DSC data for the 125°C fraction of sample B (B14).............................122 Figure D.30 DSC data for the 140°C fraction of sample B (B15).............................122 Figure D.31 DSC data for the 25°C fraction of sample C (C1).................................123 Figure D.32 DSC data for the 45°C fraction of sample C (C2).................................123 Figure D.33 DSC data for the 65°C fraction of sample C (C3).................................124. V.

(12) Figure D.34 DSC data for the 75°C fraction of sample C (C4).................................124 Figure D.35 DSC data for the 80°C fraction of sample C (C5).................................125 Figure D.36 DSC data for the 85°C fraction of sample C (C6).................................125 Figure D.37 DSC data for the 90°C fraction of sample C (C7).................................126 Figure D.38 DSC data for the 95°C fraction of sample C (C8).................................126 Figure D.39 DSC data for the 100°C fraction of sample C (C9)...............................127 Figure D.40 DSC data for the 105°C fraction of sample C (C10).............................127 Figure D.41 DSC data for the 110°C fraction of sample C (C11).............................128 Figure D.42 DSC data for the 115°C fraction of sample C (C12).............................128 Figure D.43 DSC data for the 120°C fraction of sample C (C13).............................129 Figure D.44 DSC data for the 125°C fraction of sample C (C14).............................129 Figure D.45 DSC data for the 140°C fraction of sample C (C15).............................130 Figure E.1 WAXD results for the selected fractions of sample A (A7, A9, and A11). ....................................................................................................................................131 Figure E.2 WAXD results for the selected fractions of sample B (B6, B8, and B10). ....................................................................................................................................131 Figure E.3 WAXD results for the selected fractions of sample C (C7, C9, and C11). ....................................................................................................................................132 Figure F.1 DSC melting endotherms of the first and second heating cycles of sample A of the samples which were slow cooled for WAXD analysis................................133 Figure F.2 DSC melting endotherms of the first and second heating cycles of sample B of the samples which were slow cooled for WAXD analysis. ...............................133 Figure F.3 DSC melting endotherms of the first and second heating cycles of sample C of the samples which were slow cooled for WAXD analysis. ...............................134 Figure F.4 DSC melting endotherms of the first and second heating cycles of fraction A7 of the sample which was slow cooled for WAXD analysis.................................134 Figure F.5 DSC melting endotherms of the first and second heating cycles of fraction A9 of the sample which was slow cooled for WAXD analysis.................................135 Figure F.6 DSC melting endotherms of the first and second heating cycles of fraction A11 of the sample which was slow cooled for WAXD analysis...............................135 Figure F.7 DSC melting endotherms of the first and second heating cycles of fraction B6 of the sample which was slow cooled for WAXD analysis. ................................136 Figure F.8 DSC melting endotherms of the first and second heating cycles of fraction B8 of the sample which was slow cooled for WAXD analysis. ................................136 Figure F.9 DSC melting endotherms of the first and second heating cycles of fraction B10 of the sample which was slow cooled for WAXD analysis. ..............................137 Figure F.10 DSC melting endotherms of the first and second heating cycles of fraction C7 of the sample which was slow cooled for WAXD analysis. ................................137 Figure F.11 DSC melting endotherms of the first and second heating cycles of fraction C9 of the sample which was slow cooled for WAXD analysis. ................................138 Figure F.12 DSC melting endotherms of the first and second heating cycles of fraction C11 of the sample which was slow cooled for WAXD analysis. ..............................138. VI.

(13) List of tables Table 2.1 Recent work carried out in the field of analytical TREF.............................22 Table 4.1 HT-GPC molecular weight data for the unfractionated samples A, B, and C ......................................................................................................................................45 Table 4.2 13C NMR Chemical shift data for sample C ................................................47 Table 4.3 The percentage of ethylene included in each of the original random copolymers...................................................................................................................48 Table 4.4 Crystallisation and melting data for the original samples as obtained by DSC and CRYSTAF analysis ......................................................................................50 Table 4.5 Crystallinity percentages calculated from DSC endotherms, and the amount of γ-phase crystals present in the original samples as determined from the WAXD spectra after slow cooling the samples.........................................................................52 Table 4.6 HT-GPC molecular weight and CRYSTAF data for the third fractionation of sample A ..................................................................................................................55 Table 4.7 HT-GPC molecular weight data for the fourth fractionation of sample A ..56 Table 4.8 TREF fractionation data for the fractions of samples A, B, and C..............59 Table 4.9 Ethylene content percentages for selected fractions of all three samples as determined by 13C NMR ..............................................................................................66 Table 4.10 Summary of all the DSC and CRYSTAF data for all fractions of sample A ......................................................................................................................................69 Table 4.11 Crystallinity percentages calculated from DSC endotherms and the amount of γ-phase crystals present in selected fractions of all samples, as determined from the WAXD spectra after slow cooling the samples ...........................................................72 Table A.1 HT-GPC data for the fractions of sample A. ..............................................80 Table A.2 HT-GPC data for the fractions of sample B................................................80 Table A.3 -GPC data for the fractions of sample C.....................................................81. VII.

(14) Chapter 1.. Introduction. 1.1 General introduction Propylene copolymers have received a great deal of attention in recent times due to the excellent properties that have been obtained by the introduction of a comonomer. These copolymers have become increasingly competitive in a variety of areas in which the polypropylene homopolymer was not, such as in flexible films [1]. The introduction of a comonomer has resulted in copolymers being developed with a lower degree of crystallinity than the propylene homopolymer, allowing the use of the copolymer in a broader spectrum of applications [2]. Polypropylene is also an extremely interesting polymer in that it can crystallise in a variety of crystal forms, each with its own properties [2]. This has meant that this material has become a viable option and a serious commercial commodity in certain areas of use traditionally dominated by other materials such as polyethylene. The effect of the introduction of a comonomer on the macroscopic properties of the material must be explained on a molecular level if the full benefits of this development are to be harnessed. This is vitally important and is the fundamental basis of material science. Without this knowledge, further development becomes far more difficult and much more of a lottery. This study looks at three commercial propylene-ethylene random copolymers. The materials are predominantly polypropylene with a small degree of ethylene included as comonomer. The material which comes out of the reactor during the polymers’ synthesis contains a variety of chains of varying lengths and with varying degrees of comonomer inclusion. Due to the complex nature of the manufactured copolymer it is necessary to fractionate the material before a full characterisation is possible. The technique employed during this study is temperature rising elution fractionation (TREF). This is an excellent technique for the fractionation, i.e. separation, of a semi-crystalline material into a number of fractions. The TREF technique is based on the separation of material according to its ability to crystallise [3]. The crystallisation temperature of a semi-crystalline polymer depends on a number of factors such as the molecular weight, molecular weight distribution, chemical composition, chemical composition distribution, tacticity, and the type of 1.

(15) internal ordering of the crystal unit cell. The internal ordering, or crystal phase, of polypropylene plays a large role in the properties of the polymer. Polypropylene can crystallise in different crystal forms, the formation of which is influenced by various internal factors such as chain defects, and external factors such as crystallisation temperature and pressure [4]. Comonomer inclusion in a random copolymer can also affect the type of crystal phase formed by acting as a chain defect.. 1.2 Aims The aims of this study are therefore as follows: •. The fractionation of three different propylene-ethylene random copolymers.. •. The full characterisation of the fractions as well as the original samples.. •. The determination of the effect of the inclusion of a comonomer on the ability of the chains to crystallise.. •. An investigation into the distribution of the ethylene comonomer in the copolymers, and the effect of the distribution on the properties of the copolymers.. •. The determination of the effect of the ethylene comonomer on the crystal phase formed in the original samples as well as the isolated fractions.. •. An examination of the effect of the crystal phase on the melting characteristics of the copolymer fractions.. 1.3 References 1.. Moore, E.P., Jr., & Larson, G.A., Introduction to PP in business, in Polypropylene handbook, E.P. Moore, Jr., Editor. 2002, Hanser: Munich. p. 257-285.. 2.. Phillips, R.A., & Wolkowicz, M.D., Structure and morphology, in Polypropylene Handbook, E.P. Moore, Jr., Editor. 2002, Hanser: Munich. p. 113-176.. 3.. Wild, L., Temperature rising elution fractionation. Advances in Polymer Science, 1990. 98(1): p. 1-47.. 2.

(16) 4.. Foresta, T., Piccarolo, S., & Goldbeck-Wood, G., Competition between alpha and gamma phases in isotactic polypropylene: effects of ethylene content and nucleating agents at different cooling rates. Polymer, 2001. 42: p. 1167-1176.. 3.

(17) Chapter 2.. Background. 2.1 Polyolefins: A brief historical overview The term olefin is a derivative of the word “olefiant”, meaning oil-forming gas, which was the term used by four Dutch chemists to describe the gas that produced an oil (ethylene dichloride) by the addition of chlorine [1]. It was as early as 1858 that Goryainov and Butlerov managed to polymerise pentene by the addition of boron trifluoride. This was followed soon after by Berthelot, in 1869, who managed to polymerise propylene by a reaction with concentrated sulphuric acid [2]. The product formed, a viscous oil, was of no industrial importance. In 1894 H. von Peckman produced a linear, low molecular weight, polyethylene by the decomposition of diazomethane, a technique also used for the production of polymethylene [1]. It was only much later, during the early part of the 1920’s, that the concept of a high molecular weight polymer emerged, meeting considerable resistance in scientific circles [3]. Taylor and Jones reported the polymerisation of ethylene in the presence of diethylmercury in 1930 [4]. The concept of stereoregular polymerisation was largely ignored and it was not until the stereoregular form of natural rubber was observed in the 1940’s that stereoregularity as a concept became more readily acceptable. It was in the 1950’s that real strides forward were taken in the development of polyolefins [5], when in 1953 high-density polyethylene was synthesised in the labs of Karl Ziegler. Early the following year Giulio Natta managed to synthesise polypropylene with Ziegler following suite only a few months later [3]. Fontana managed the cationic polymerisation of propylene in 1952 [1], producing an amorphous material which was useful as an additive for lubricating oil but lacked the strength necessary for structural applications. The first commercial production of a polyolefin was that of polyethylene by ICI in 1939 [3]. Licenses for the production of polyethylene were subsequently granted to Union Carbide and Du Pont [1]. Due to the fact that the branches in the low density polyethylene (LDPE) significantly lowered the density of the polymer, attempts were made to produce a polymer with a more linear structure, namely a linear low density polyethylene (LLDPE). Copolymers of ethylene and 1-butene were. 4.

(18) synthesised in order to combat this problem although there was little demand for them while the technology was still in its infancy [1]. The LLDPE which was eventually produced was a superior polymer to LDPE for many applications. It was mainly produced by Union Carbide’s Unipol process [6]. Crystalline polypropylene, on the other hand, first went into commercial production in 1957 in the plants of a number of companies, including Hercules, Montecatini, and Fabewerke-Hoechst [1]. The polyolefin industry took off from there, as new processes and products were constantly being developed. Growth in the field has been phenomenal, with production increasing at an excellent rate. Polypropylene production increased by 6.9% between 1993 and 2000, with the other thermoplastics also showing similar strong trends (LDPE: 3.5%, LLDPE: 9.7%, HDPE: 6.1%) [7]. The sheer number of applications and diverse fields of applicability of this class of materials is greatly due to the continued and substantial research that is undertaken each year. New markets for polyolefin materials are constantly being created through the designing of new materials such as copolymers and blends, obtaining improved and more interesting properties for so-called standard materials. Processing conditions have also changed greatly over the years and the development of new catalyst systems has lead to an exciting period in the lifetime of the industry, with new doors constantly being opened.. 2.2 Polymerisation chemistry: An overview A discussion of catalyst technology and polymerisation processes is necessary in order to understand why the polymers produced by heterogeneous catalysts have their unique characteristics. The very nature of the catalyst is the reason for the chemical composition distribution of the polymers produced. Consequently, the necessity for fractionating a polymer in order to fully characterise it is directly due to the polymerisation process itself.. 2.2.1 General mechanism of transition metal catalysed polymerisation A Ziegler-Natta catalyst can be defined as a transition metal compound incorporating a metal-carbon bond which is able to perform the repeated insertion of 5.

(19) olefin units [8]. The active centres of Ziegler-Natta catalysts are basically formed due to interaction between a transition metal compound and an organometallic cocatalyst [5, 9]. The exchange of a halogen atom from the transition metal compound and an alkyl group from the organometallic cocatalyst is a critical step in the formation of the active centre [5, 10], as illustrated in Equation (1) for a TiCl3/AlEt3 system: [TiCl3] + [AlEt3] → [Cl3TiEt] + AlEt2Cl. (1). The most important factor regarding the bond between the transition metal atom and the carbon atom is that it has the ability to react with the double bonds of αolefins [5]. The monomer first coordinates to the transition metal before the actual insertion occurs. This leads to the formation of a complex, four-member transition state from which the monomer unit is inserted into the growing chain. This mechanism has been proven by the presence of isobutyl chain-end groups formed in the first step of the polymerisation reaction using 13C-enriched Al(CH3)3 [11]: M-13CH3 + CH2=CH-CH3 → M-CH2-CH(CH3)-13CH3. (2). The insertion of the α-olefin into the metal-carbon bond can occur in two different ways [8]: M-P + CH2=Ch-CH3 → M-CH2-CH(CH3)-Polymer (1,2 primary insertion). (3). M-P + CH2=Ch-CH3 → M-CH(CH3)-CH2-Polymer (2,1 secondary insertion). (4). where P represents the polymer chain. This defines the regiochemistry of the polymer formed. Heterogeneous catalysts have extremely high regiospecificity, resulting in mainly 1,2 insertions [8]. The polymer chain is then grown through the repeated insertions of the monomer units. Secondary insertions can either be followed by a primary insertion, leading to vicinal methyl groups, or by isomerization of the secondary inserted unit, resulting in 1,3 insertion of the monomer [12]. The 1,3 insertions result in the following structure [2]:. 6.

(20) -CH2-CH(CH3)-CH2-CH2-CH2-CH2-CH(CH3)-CH2The isomerization is favoured by a higher polymerisation temperature [12]. Eventually each growing polymer chain is disengaged from the transition metal atom. There are a number of ways in which this chain termination occurs. The first method of chain termination is chain transfer to monomer [5]. This is the most important chain termination process for the polymerisation of propylene with heterogeneous catalysts (in the absence of hydrogen) [8, 9]. It involves the replacement of a long alkyl chain at the transition metal atom with a short alkyl group derived from the monomer as illustrated in Equation (6): M-CH2-CHR-Polymer + CH2=CH-R → M-CH2-CH2-R + CH2=CR-Polymer. (6). A second reaction which can occur is the alkyl group transfer between the active centre and the organometallic cocatalyst as illustrated in Equation (7): M-CH2-CHR-Polymer + AlEt3 → M-Et + Et2Al-CH2-CHR-Polymer. (7). The Al-C bond decomposes on exposure to air and moisture, leaving a polymer molecule [5]. The third way in which termination occurs is by means of a β-hydride elimination Equation (8), although this process is not considered important in propylene. polymerisation. with. heterogeneous. catalyst. systems. at. normal. polymerisation temperatures [8]: M-CH2-CHR-Polymer → M-H + CH2=CR-Polymer. (8). Equation (8) does however become a significant chain termination reaction in metallocene-based catalyst systems [8]. There is also the β-methyl elimination method of chain termination, although this process has never been observed during the polymerisation of propylene with heterogeneous catalyst systems [8]. It is however important during homogeneous polymerisations. The chain termination reactions occur very infrequently compared to the chain growth reactions [5]. In order to limit the molecular weight of the polymer formed, hydrogen is usually introduced. 7.

(21) to terminate a growing chain according to the so-called chain transfer to hydrogen reaction [13], as illustrated in Equation (9): M-CH2-CHR-Polymer + H2 → M-H + CH3-CHR-Polymer. (9). The chain transfer to hydrogen reaction is the most commercially important method of controlling the molecular weight [5].. 2.2.2 Polymerisation control mechanisms and stereochemistry When dealing with a prochiral monomer such as propylene, the question of stereospecificity as well as regiospecificity arises during the polymerisation with a given catalyst. Figure 2.1 shows the general mechanism of the polymerisation process. The manner of the coordination during the first step of the reaction determines the stereoand regiospecificity of the monomer unit in the chain.. Figure 2.1 The Ziegler-Natta polymerisation mechanism.. The regioselectivity of the Ziegler-Natta catalysts is generally better than that of the metallocenes catalysts [5, 8]. The majority of the monomer units will therefore be inserted in the 1,2 insertion mode during polymerisation with heterogeneous catalysts. This still leaves the question of preferential enantioface selectivity during the polymerisation wide open. Should alternating enantiofaces be inserted during 8.

(22) polymerisation, the polymer formed would have no configurational regularity, and would therefore be an atactic polymer. Multiple insertions of the same enantioface would therefore produce isotactic polymer. Two possibilities exist for governing the stereoselection of the enantioface [8]. The first involves control by the chiral induction of the last inserted unit and is referred to as chain-end control. The second possibility is the asymmetry of the initiating site and this is known as enantiomorphic site control. The differences in the polymerisations with the different control mechanisms are illustrated in Figure 2.2.. Figure 2.2 Polymerisation control mechanisms.. It is the nature of the active site itself on the catalyst which governs the type of polymeric chain that is produced. One can differentiate between aspecific sites producing atactic material, and isospecific sites producing highly isotactic material. Figure 2.3 illustrates this using a MgCl2 supported titanium based catalyst as an example. The crystalline MgCl2 has the same elementary structure as TiCl3 [5]. The 1,1,0 and 1,0,0 lateral cuts contain coordinatively unsaturated Mg2+ ions with coordination number 4 on the 1,1,0 cut and 5 on the 1,0,0 cut.. 9.

(23) Figure 2.3 Catalyst active sites on 1,0,0 and 1,1,0 cuts of the MgCl2 crystal.. During the development of the various catalyst generations (discussed in Section 2.2.3), it was discovered that the addition of a Lewis base to the heterogeneous catalysts resulted in an increase in catalytic activity and stereospecificity. These Lewis bases subsequently became known as ‘internal donors’, which were co-milled with the MgCl2 and TiCl4, and ‘external donors’ which were combined with the cocatalyst [8]. The job of an internal donor such as ethyl benzoate, is to prevent the formation of atactic material by adsorbing onto the surface and changing the aspecific site at the tetracoordinated Mg atom on the 1,1,0 plane to a more isospecific site. An external donor, such as ethyl benzoate which can act as both an internal and external electron donor, helps to prevent the extraction of the internal donor as well as converting aspecific sites on the 1,0,0 crystal plane, as can be seen in Figure 2.4.. Figure 2.4 Coordination of internal donors ensuring isospecific active sites.. 10.

(24) When dealing with copolymers, such as the propylene-ethylene random copolymers used in this study, there is the added factor of the comonomer to be considered when examining the polymerisation. The two different monomers have completely different reactivity ratios [14]. The propylene reactivity ratio, r1, multiplied by the ethylene reactivity ratio, r2, should be close to 1 for a random copolymer (r1r2 ≈ 1). A blocky structure is present if r1r2 > 1 and an alternating structure is present if r1r2 < 1 [8]. Ethylene monomer is far more reactive than propylene monomer [14], although it is only present in very small amounts in the random copolymers used in this study. Cheng and Kakugo [15] investigated ethylenepropylene random copolymers and applied Bernoullian and first-order Markovian models to the data they obtained, as well as the MIXCO.TRIAD and MIXCO.TRIADX programs for analysis of the triad data. They found that due to the heterogeneous catalyst, with three or four active catalytic sites, the polymer formed was an in situ blend of three or four random copolymer components. It was also observed that active sites with similarly high stereospecificities possess a broad spectrum of reactivities towards the comonomer during a copolymerisation [8, 15]. It is therefore clear that there are a number of factors that influence the nature of a propylene-ethylene random copolymer during its polymerisation and that a fractionation method is required for a full characterisation of the polymer.. 2.2.3 The evolution of the transition metal catalysts The development of the so-called Ziegler-Natta catalysts began around 1950 with Karl Ziegler’s work on the “Aufbau” reaction which involved the insertion of ethylene into the Al-C bond of trialkyl aluminium and the subsequent growth of linear alkyl chains [1, 8]. It was in 1953 when the major breakthrough occurred with the production of high-density polyethylene (HDPE) [8] in Ziegler’s laboratories, in which Giulio Natta had placed three of his assistants. In the years following this breakthrough, Ziegler and Natta, at the forefront of the polyolefin industry, were able to make polypropylene and even define the stereo conformations of the polypropylene. These original polypropylenes only contained up to approximately 40% isotactic material [8]. The development of the catalyst technology is best. 11.

(25) described by referring to the so-called generations of catalysts as they were conceived. The early work involving Ziegler-Natta catalysts involved a combination of TiCl3 as the catalyst and AlEt2Cl as the cocatalyst. The productivity was relatively low as was the isotacticity (around 90%). Removal of the atactic material as well as the catalyst residues (a process known as de-ashing) was necessary [8]. It was eventually realised that prolonged ball milling of TiCl3 and AlCl3 produced a more active catalyst than pure TiCl3. This catalyst became known as AA-TiCl3 (Al-reduced and activated) and is regarded as being the first generation of Ziegler-Natta catalysts. One of the main problems associated with the first generation of catalyst was the limited use of the titanium atoms as only those on the surface could take part in the polymerisation. This led to the development of a second generation TiCl3 catalyst with a much larger surface area [8], thus increasing the productivity and isotacticity. De-ashing to remove catalyst residues and atactic material removal was still necessary however. Supported catalysts became known as the third generation and involved the use of supports with functional groups onto which the TiCl4 could be attached. MgCl2 emerged as the main support used [5] and, with the aid of a Lewis base (benzoic acid esters) acting as an internal and external electron donor, a highly active [9] and stereospecific catalyst was born. The purpose of the electron donors is to aid in the formation of highly isospecific active sites as well as to selectively poison the nonstereospecific sites and convert them into isospecific sites [16]. The removal of atactic material was still required though, and this led to further developments. The fourth generation of catalysts (also known as super-active third generation catalysts) was brought about by the use of alkylphthalates and alkoxysilanes as internal and external electron donors respectively. There was thus a further improvement in the catalyst performance, in terms of increased isotacticity and productivity. The latest development in the chain was the discovery that 1,3-diethers could be used as internal electron donors, giving highly active sites and isotacticities without an external Lewis base being required [16]. Homogeneous stereospecific catalysts gained importance when it was discovered that metallocenes of Zirconium and Hafnium with methylaluminoxane. 12.

(26) (MAO) could be used to synthesise highly isotactic or syndiotactic polymers in very high yields [8].. 2.3 Commercial polypropylene 2.3.1 Varieties of polypropylene manufactured Despite the fact that polypropylene already has such a huge number and variety of applications, there are constantly more being developed. Due to the competitiveness of the commercial polymer industry, companies are constantly searching for new areas in which they can apply their products. This has led to extensive research in the field of copolymerisation, including both new copolymers and polymerisation conditions. One can differentiate between a statistical or random copolymerisation and a sequential copolymerisation [8]. The versatility of polypropylene is demonstrated when one looks at the different commercial types that are manufactured, namely the homopolymer, random copolymers, and the so-called impact copolymers.. 2.3.1.1 Polypropylene homopolymer The main structural factors that influence the properties of the polypropylene homopolymer are tacticity, molecular weight and molecular weight distribution [17]. The different types of tacticity are illustrated in Figure 2.5.. 13.

(27) Figure 2.5 Types of polypropylene tacticity.. The main influence of the tacticity is on the crystallinity of the polymer. Isotactic polypropylene homopolymer is highly crystalline. It has a correspondingly high melting point of approximately 186°C, although this value has been a matter of controversy for a number of years [18], and a Tg of approximately 0°C, which results in a brittle polymer below this temperature [8]. The product is however extremely versatile, which, coupled with the low monomer cost and efficient polymerisation technology, makes it one of the most important commercial thermoplastics [19]. The homopolymer can be too rigid for certain applications. A lower melting point would improve weldability and improved impact resistance at low temperature is also necessary for certain applications. Requirements such as this have led to the development of the copolymers of polypropylene, tailor-made for specific applications.. 2.3.1.2 Impact copolymers The fact that the polypropylene homopolymer has such a poor impact resistance, especially at low temperature, has led to the development of the so-called impact copolymers produced by means of a sequential polymerisation reaction. A sequential polymerisation involves a two-stage process, whereby propylene is first polymerised on its own, followed by a second stage where both propylene and ethylene are polymerised in the presence of the originally polymerised material from the first stage [8]. Two reactors are required, connected in series, for the production of these heterophasic copolymers [20]. The rubber phase is usually an ethylene-. 14.

(28) propylene rubber although an ethylene-propylene-diene monomer elastomer is also often used [21]. The result is an elastomeric poly(propylene-co-ethylene) copolymer dispersed. in. a. matrix. of. polypropylene. homopolymer.. These. sequential. polymerisation reactions yield polymers with greatly improved impact strength [22]; hence they are often referred to as impact copolymers. Various factors influence the performance of these copolymers, including the amount of elastomer included in the polymer, the size of the rubber particles, the chemical affinity of the elastomer for the polypropylene matrix, as well as the distribution of the rubber particles [21, 23]. A homogeneous distribution of the rubber particles provides the best dispersion of energy, giving the best stiffness-to-impact balance [20]. An homogeneous distribution of the rubber particles is also necessary in order to avoid reactor fouling [24]. An optimum rubber particle size often exists for a given matrix/rubber system. The optimum size for the PP/EPR system is approximately 0.4 μm [21]. The composition of the elastomer is also important [25]; for example, varying the ethylene/propylene ratio in an ethylene-propylene rubber can have a large effect on the copolymers properties. A high propylene content would result in poorer impact resistance, better interfacial adhesion and less shrinkage stresses, due to the polypropylene crystallinity, than a rubber with a lower propylene content [21]. Increasing the ethylene content would reduce the polypropylene crystallinity while improving the polyethylene crystallinity, thereby improving impact resistance up to a maximum value, after which the interfacial adhesion would decrease too much, and reduce the impact strength [21]. The optimum ethylene concentration for the best impact resistance is approximately 50 to 60 mol% [26].. 2.3.1.3 Random copolymers In order to harness the strength of polypropylene and to improve the properties of the material for certain applications, it is necessary to reduce the crystallinity slightly so as to improve properties such as flexibility and optical clarity [8]. This is done by introducing a comonomer, such as ethylene or 1-butene, into the polymerisation medium so as to create a discontinuity in the polymer chain, disrupting the crystal structure of the polypropylene slightly, thereby altering the morphology and structure in order to improve these properties [17]. The properties of the copolymer are largely dependent on the amount of comonomer included as well as 15.

(29) the distribution of the comonomer throughout the polymer [20]. Generally the inclusion of the comonomer results in a reduction in the crystallisation rate, a lower degree of crystallinity, and a lower melting point [17]. The lower melting point of the copolymer is often required for the heat-sealable layer on a film, and it is the comonomer content that has the greatest influence on the melting point. The crystals produced are not as perfect as those of the homopolymer, which means that the difference in the refractive index between the crystalline and amorphous areas is less. Light is therefore not refracted as easily, resulting in lower haze and higher clarity for the copolymer [17]. These propylene-ethylene random copolymers are synthesised using the socalled statistical copolymerisation method, whereby the propylene is polymerised in the presence of small quantities of ethylene [8] in a single reactor. The degree of randomness of the polymerisation often varies due to factors such as the polymerisation conditions, the catalyst system, and the reactivity ratio of the comonomer relative to propylene [17]. The type of chains produced, or more specifically the amount of extractables produced by a polymerisation system, is a critical factor for food contact applications. For example, ethylene lowers the melting point to a greater degree than a comonomer such as 1-butene but also produces a higher level of extractables [17]. If the random copolymer is subjected to slow cooling from the melt then the actual form of the crystals produced is altered. There are substantial amounts of the γform crystals formed as well as the α-form. The specific crystal phase formed by these random copolymers is discussed in more detail in the next section, as the crystalline morphology (type of crystal) is of some importance in the application of these materials.. 2.3.2 Crystallinity types With the crystallinity of a polymer being the key factor which influences the physical properties of a polymer, information regarding the crystallinity becomes of paramount importance when assessing a polymer’s applicability. The crystallinity of polypropylene homopolymer is governed mainly by the tacticity of the chains [17].. 16.

(30) When investigating a propylene-ethylene random copolymer there is the added factor of the comonomer to consider, as this also influences the degree of crystallinity. The ethylene comonomer serves to reduce the crystallinity of the copolymer, thus improving properties such as the flexibility and optical clarity [17]. It is possible to look at the crystallinity of such a copolymer on various levels, from the skin-core morphology on a visual scale to the spherulitic scale, lamellar scale, and finally the crystallographic scale where the actual unit cell is examined [8, 27]. Polypropylene has four crystal forms, namely the α-form (monoclinic), β-form (trigonal), γ-form (orthorhombic) and a metastable mesomorphic form, often referred to as the smectic form [27, 28]. The smectic form is formed by fast cooling of the polymer melt at low temperatures [19] and represents a state of order intermediate between the amorphous and crystalline states [29]. All forms of the crystal contain chains in the characteristic 31 helix conformation of polypropylene [28]. There are in fact two types of monoclinic unit cells: the α1-form originally indexed by Natta and Corradini in the C2/c space group, and the α2-form in the P21/c space group [29]. It is well known that an increase in the comonomer content increases the number of ‘defects’ in the chains, thereby reducing the length of the isotactic sequences [30, 31]. The amount of the γ-phase is proportional to the number of short isotactic segments, caused by the interruption of the isotactic sequences by the comonomer [20]. An increase in the comonomer (such as ethylene) content therefore causes an increase in the growth of the γ-phase crystals [32]. A random terpolymer with ethylene and 1-butene yields an even higher percentage of the γ-phase at the same molar comonomer content as a similar copolymer [20]. The γ-phase is also known to be enhanced by crystallisation at high pressures, low molecular weight, and the presence of chain defects or chemical heterogeneity caused by atacticity [32, 33]. The α-phase is however the most stable and heating of the γ-phase results in conversion to the α-phase [29]. The γ-phase has an epitaxial relationship with the αphase and either phase can grow onto the lamellae of the other phase [28, 32]. The γphase crystals consist of bilayers in which the adjacent layers are at an angle of 80° to each other as opposed to being parallel [29, 34, 35]. The presence of these nonparallel chains in the crystal structure of γ-phase polypropylene is unique in crystallisable synthetic polymers [29, 36]. The γ-phase also displays screw dislocations and nucleates on the α-form crystal on the (010) contact plane [37]. The 17.

(31) γ-phase crystals are elongated in the b-axis direction and their chains are inclined at an angle of 40° to the lamellar surface [38]. The importance of the crystal type comes into play when one considers the applications of the polymer. The γ-phase crystal has a lower melting point than the αphase, and also produces polymer with improved optical properties [17, 34, 39, 40].. 2.4 Fractionation techniques The necessity of a suitable fractionation technique has evolved from the need to fully understand how the polymer chain architecture influences the physical properties of the material. The use of fractionation enables one to obtain many fractions of a much more narrow distribution than the unfractionated material, be it a chemical composition or a molecular weight distribution. The three main techniques used for fractionating a polymer are fractionation according to crystallisability, molecular weight, and solubility. These shall be discussed separately in the following sections.. 2.4.1 Fractionation by crystallinity Crystallinity is one of the most important characteristics of a polymer, greatly influencing the physical properties, and is therefore a key basis for fractionating a polymer. The fractionation reveals exactly how much material can crystallise and to what extent. This information is vital for the development of new materials and catalyst systems for subsequent better product performance.. 2.4.1.1 Fractionation mechanism and crystallisation theory Fractionation of the propylene-ethylene random copolymer by temperature rising elution fractionation (TREF) is based on separation according to crystallisability [41-45]. In other words the actual molecular structure and composition directly affects the ability of the chains to crystallise [43]. The longest crystallisable isotactic sequence in the propylene-ethylene random copolymer will therefore determine at what temperature the particular chain will crystallise. The 18.

(32) effect of the comonomer upon the crystallisation and melting point of the copolymer is complicated when examined at a molecular level [46]. The crystallisation and melting point will be affected according to the degree to which the comonomer disrupts the crystal lattice. The melting point of the copolymer will definitely be lower than that of the homopolymer [46]. The chain ends and the diluent also contribute to the lowering of the melting point [46]. An approximation for the depression of the melting point has been given by Flory and is shown here as Equation (1).. ⎛ R 1 1 − =⎜ Tm Tm 0 ⎜⎝ ΔH f. ⎞⎛ V ⎞ ⎟⎜ ' ⎟ v1 − χv12 ⎟⎝ V ⎠ ⎠. (. ). (1). In this equation Tm0 represents the melting point of a perfect crystal, Tm is the melting point of the polymer-diluent mixture, V and V’ are the molar volumes of the polymer repeat unit and diluent respectively, R is the gas constant, χ is a polymer-solvent interaction parameter, and ν is the volume fraction of diluent [41, 46]. According to Flory if the non-crystallisable comonomer causes the depression of the melting point then for a comonomer unit randomly distributed along a polymer chain the melting point becomes: ⎛ R 1 1 − = −⎜ ⎜ ΔH Tm Tm 0 f ⎝. ⎞ ⎟ ln N A ⎟ ⎠. (2). where NA is the mole fraction of comonomer units in the random copolymer [41, 46]. It has been found that the degree of melting point depression is actually greater than that predicted by the theory [41]. Shirayama, Kita, and Watabe [47] discovered an almost linear relationship between the melting point and the percentage of comonomer. Zhang, Wu, and Zu [48] assumed that the melting temperature of a copolymer, Tm, is close to that of the homopolymer, Tm0, such that Tm x Tm0 ≈ (Tm0)2, and that ΔH is constant in that temperature range. They thus reduced Equation (1) to Equation (3) and obtained a relationship between melting temperature and comonomer content.. ( ). R Tm0 Tm ≅ T − ΔH 0 m. 2. XE. (3). 19.

(33) The effect of molecular weight on the fractionation was also considered by Wild [49]. The data obtained by Wild indicated that if the polymer chain ends are considered to be the equivalent of a branch point then the molecular weight dependence on the fractionation mostly disappears. They also showed that the molecular weight dependence falls away as soon as the molecular weight reaches approximately 104 g/mol. Zhang, Wu, and Zu [48] also noted that there were two types of chains with a low melting point, those with a low molecular weight and those with a high comonomer content (ethylene in their case). They also noted that for a chain to have a high melting point it must have a high molecular weight as well as a low ethylene content. It is therefore clear that although molecular weight effects cannot be ignored, the fractionation of a copolymer such as the propylene-ethylene random copolymers is dependant on the ability of the chains to crystallise. The two main techniques that are used to fractionate semi-crystalline polymers according to crystallisability are temperature rising elution fractionation (TREF) and crystallisability analysis fractionation (CRYSTAF).. 2.4.1.2 TREF The ability to fully characterise a polymer material in order to fully understand where it gets its macroscopic properties from has been the goal of many researchers over the past fifty years. Much of the early work in this field was focused on ways to establish molecular weight distributions. Desreux and Spiegels [50] were the first to realise that a semi-crystalline polymer could be fractionated according to solubility at a given temperature, and that this fractionation was based on the ability of the polymer to crystallise and not simply on its molecular weight. Their pioneering work involved the elution of fractions of polyethylene at successively higher temperatures. Further development and refinement occurred in the field, but it was not until Shirayama et al. [51] described the method of fractionating low density polyethylene according to the degree of short chain branching that the term “temperature rising elution fractionation” was born. At this time See and Smith [52] were investigating the effect of different solvent/non-solvent mixtures of varying compositions on the elution of linear polyethylene and isotactic polypropylene. Their experimental setup was essentially the same as that used for TREF, with the exception that they 20.

(34) maintained a constant elution temperature and varied the strength of the eluting solvent. This was similar to the work of Guillet et al. [53] on polyethylene. With the development of size exclusion chromatography as an excellent method for determining molecular weight distributions, fractionation according to crystallisability became the new area of interest. The development of the TREF experimental setup occurred for very practical reasons. It is far easier to dissolve polymer off a support than it is to collect fractions which crystallise at successively lower temperatures. The TREF technique separates material on the basis of molecular structure or composition [41]. Changes at a molecular level influence the crystallisability of the chains and therefore the solubility at a given temperature. The general TREF technique can be divided into two main steps, namely a crystallisation step and an elution step. During the crystallisation step, the semi-crystalline polymer that is being analysed is first dissolved at high temperature, and then allowed to cool slowly under the control of a programmed temperature profile. According to Wild [41] the maximum cooling rate that should be used for achieving a good separation is 2°C/hour. Various media have been utilised for the crystallisation step of TREF with the most common being a temperature controlled oil bath [49, 54]. Alternatives do exist such as the oven from a GPC setup [55], although in this case heat transfer is not as good. One advantage of an oven however is the decreased cycle rotation time due to the fact that the oven can be cooled far quicker than an oil bath, in preparation for the next fractionation [41]. Problems associated with temperature gradients in the column as well as poor heat transfer have been noted by Wild [41]. A single medium can be used for both the crystallisation and elution steps as in the setup of Bergstrom and Avela [56] and Nakano and Goto [57]. It is often the case that two separate media are used [49, 54], enabling the simultaneous crystallisation of a number of samples, seeing as this is the time-limiting step of TREF [41]. The operations utilising a separate step usually use an oil bath for the crystallisation step followed by either another oil bath or an oven for the elution step. The importance of the crystallisation step was not fully recognised at first, although it gradually gained importance as it was eventually recognised as the critical step necessary to obtain good reproducible separations [45]. The cooling step can either be done in the presence of a support [48, 49, 55, 58], or simply in solution [54, 59, 60], which is then later slurried with a. 21.

(35) support before the elution step. The addition of 0.1% of an antioxidant is advised in order to prevent polymer degradation [61]. During the elution step the polymer is dissolved off the support at successively higher temperatures. Columns thus became an integral part of the experimental setup as they provided a simple medium in which to perform the fractionation. Initially constructed from glass [62, 63] and later from stainless steel [64, 65], the columns developed were of many different sizes. There are a few good reviews in the literature that cover all aspects of TREF [41, 42, 44, 45, 61, 66]. As the experimental techniques were improved and refined a distinction could be drawn between the technique involving an on-line detector for continuous signal detection (analytical TREF), and the technique involving the collection of much larger fractions for subsequent offline analysis (preparative TREF). These techniques are now discussed separately.. Analytical TREF (A-TREF) Analytical TREF is a relatively recent development in the experimental setup of TREF, with workers such as Usami, Gotoh, and Takayama [55] being among the first to describe their systems in detail. Analytical TREF involves the same slow, controlled crystallisation step as in the preparative version of the fractionation, but instead of collecting the fractions for offline analysis the eluent is sent to an RI/IR detector which constantly monitors the polymer being eluted. Recently the trend has been to use an IR detector set at 3.41 μm (C-H stretch), as this presents less of a problem when compared to an RI detector, with respect to with baseline noise [43, 45], due to the relative insensitivity of IR to temperature fluctuations [41, 42]. Table 2.1 contains a detailed list of various analytical TREF systems and their corresponding variables that have been utilised recently.. Table 2.1 Recent work carried out in the field of analytical TREF. Polymer type. Sample size (mg). Support material. Cooling Solvent. rate (°C/h). Heating rate. Flow rate. Reference. (ml/min). 22.

(36) Polymer type. Sample size (mg). Support material. Cooling Solvent. rate (°C/h). Heating rate. Flow rate. Reference. (ml/min). LLDPE. -. -. o-DCB. 5. 4°C/min. -. [60]. HDPE/EVA. -. -. -. 5. 4°C/min. -. [60]. PP-co-PE. 2-5. Xylene. 5. 4°C/min. 3. [54]. LLDPE. 2. -. o-DCB. 1. [55]. LLDPE. -. -. TCB. 0.1 - 1. 0.2. [67]. LLDPE. -. HDPE. -. LLDPE. 2. PE. -. C104H210. Diatomaceous earth. Diatomaceous. 0.05 0.5. Xylene. 5. 4°C/min. 3. [68]. Xylene. 5. 4°C/min. 3. [68]. TCB. 5.6. 4°C/min. 3. [65]. Glass beads. TCB. 1. 20°C/h. 1. [69]. -. Glass beads. TCB. 1. 20°C/h. 1. [69]. LLDPE. 100. Chromosorb-P. TCB. 1.5. 20. 4. [49]. LLDPE. -. Chromosorb-P. TCB. 1.5. 20. 2. [58]. LDPE. -. Chromosorb-P. TCB. 1.5. 20. 2. [58]. HDPE. -. Chromosorb-P. TCB. 1.5. 20. 2. [58]. LLDPE. -. Glass beads. o-DCB. 1.5. 1. 1. [70]. LDPE. -. Chromosorb-P. o-DCB. 1.5. 20. 0.3. [71]. earth Diatomaceous earth Diatomaceous earth. Analytical TREF development has advanced quite prodigiously in recent years due to the possibility for automation, reducing the manpower required to obtain results.. Preparative TREF (P-TREF) Similar to analytical TREF in many ways, the preparative variation of the technique is a means to obtain the greatest amount of information regarding the composition of a semi-crystalline polymer. There is less possibility for automation in. 23.

No results found