Chromatographic Techniques for
the Analysis of Complex Polyolefins
Prepared by Metallocene Catalysis
Sven Markus Graef
M.Sc. Polymer Science
Dissertation presented for the degree of
Doctor of Philosophy in Polymer Science
at the
University of Stellenbosch
Promoter: Co-promoter:
Prof. H. Pasch Prof. R. D. Sanderson
D eutsches K unststoff Institute Division o f Polymer Science
D arm stadt Department o f Chemistry
G erm any University o f Stellenbosch
Stellenbosch South Africa
D
e c l a r a t i o nI, the undersigned, hereby declare that the w ork contained in this thesis is m y own original w ork and has not previously in its entire or in part been subm itted at any university for a degree.
Innovative Spectroscopic and
C h r « ^ a t o g r a ^ ^ f #echniques for
tl$&4nalysis <iJ|£oinplex
Summary
The study focused on the analysis o f a variety o f synthesised tailored copolymers. D uring the investigation o f the samples new and innovative analytical techniques w ere developed to be able to identify the presence o f certain predicted or expected copolym erisation products.
Em phasis was placed on the versatility o f CRYSTAF as a m ethod for the analysis o f sem i-crystalline copolymers. Changes in the crystallisation tem perature w ere used as an indicator, while the type o f m ethod, solvent and sample w eight served as variables in the system. The percentage com onom er content distribution for an unknow n sam ple was determ ined from the standard curve plotted w ith the aid o f copolym ers w ith know n com onom er content.
E thylene/higher a-o lefin and propylene/higher a-o lefm copolym ers w ere synthesised by m eans o f a m etallocene precatalyst. In both cases, N M R spectroscopy, DSC, GPC, and C R Y STA F were used as analytical tools. In the ethylene series it was shown that the sam ple m ixture was hom ogenous in the m olar mass axis (G PC) but not in the chem ical com position axis (CRY STAF) regarding the com onom er content. For the propylene series, an increase in stereoerrors was observed by N M R and this was correlated w ith crystallisation on heating a DSC for the range o f copolym ers.
In the case w here ethylene/m ethyl m ethacrylate block copolym ers were synthesised using m etallocene precatalyst, novel detection and separation m ethods were used and developed. This included the use o f CRYSTAF fitted w ith a carbonyl filter, high tem perature gradient HPLC and high tem perature liquid chrom atography under critical conditions (LCCC). The last two techniques were the first w here separation could be achieved with samples only dissolving at high tem perature. All previous m entioned techniques, as well as the coupling o f FTIR to GPC and high tem perature gradient HPLC via LC-Transform revealed that the samples consisted o f varying ethylene and M M A block lengths.
Opsomming
Die doelstelling van die navorsing was die analise van spesiaalvervaardigde kopolimere. N uw e analitiese tegnieke is vir die bevestiging van sekere voorgestelde kopolimerisasie-produkte ontwikkel.
KJem is gele op die veelsydigheid van CRYSTAF as ’n m etode vir die analise van gedeeltelik-kristallyne kopolimere. V eranderinge in die kristallisasie-tem perature is as respons gebruik, terwyl die m etode van sintese, die oplosmiddel en die hoeveelheid m onster as veranderlikes in die sisteem beskou is. ‘n Standaardkurw e vir kom onom eerinhoud is opgestel met behulp van kopolimere met ‘n bekende kom onom eerinhoud. Hierdie kurw e is gebruik om die komonom eerinhouds- verspreiding van onbekende m onsters te bepaal.
Etileen/hoer a-olefien- en propileen/hoer a-olefien-kopolim ere is met behulp van ‘n metalloseen pre-katalis gesintetiseer. In beide gevalle is KM R spektroskopie, DSC, GPC en CRYSTAF gebruik om die analises uit te uitvoer. M et verwysing na kom onom eerinhoud is daar in die geval van die etileenreeks bevind dat die monstermengsel hom ogeen is met betrekking to t die molere massa, m aar nie met betrekking tot die chemiese samestelling nie. V ir die propileenreeks is ’n verhoging in die stereofoute met behulp van K M R waargeneem. Dit is gekorrelleer met kristallisasie weens verhitting tydens DSC-bepalings vir die reeks kopolimere.
In die geval van die sinteses van etileen/metielmetakrilaat-blokkopolim ere met metalloseen as pre-katalis, moes nuwe waam em ings- en skeidingstegnieke ontwikkel word. Dit het die gebruik van CRY STA F met ’n karbonielfilter, hoe- tem peratuurgradient-H PLC en hoe-tem peratuurvloeistofchrom atografie onder kritiese toestande ingesluit. Laasgenoemde tw ee tegnieke het vir die eerste keer skeiding van m onsters w at net by hoe tem perature oplos, moontlik gemaak. Bogenoem de tegnieke, sowel as die koppeling van FTIR met GPC en hoe-tem peratuur-gradient-H PLC via LC-transformasie het getoon dat die m onsters etileen- en M M A -blokke met verskillende lengtes be vat het.
List of contents
Sum m ary...i
O psom m ing... ii
List o f F igures... ix
List o f Schem es...xviii
List o f T ables... xx
List o f A bbreviations... xxiii
A cknow ledgem ents... xxv
Publications and posters that have resulted from this research... xxvii
Chapter 1
O verview 1.1 In tro d u ctio n ...11.2 O b jectiv es... 2
1.3 Layout o f this dissertatio n ...4
1.3.1 Polyolefin crystallisation in solution m easured by Crystallisation Analysis Fractionation (C R Y S T A F )... 4
1.3.2 C opolym erisation o f propylene with higher a-o lefin s in the presence o f the syndiospecific catalyst *'-Pr(Cp)(9-Flu)ZrCl2/ M A O ...4
1.3.3 Copolym erisation o f ethylene w ith higher a-o lefins in the presence o f the m etallocene catalyst Et(Ind)2ZrCl2/M A O ... 4
1.3.4 Ethylene and methyl methacrylate block copolym ers synthesised by m etallocene cataly sts...5
Chapter 2
H istorical and theoretical background
2.1 In tro d u c tio n ...7
2.2 T ransition m etal-based catalysts for polyethylene and polypropylene synthesis... 7
2.2.1 H isto ric a l... 7
2.2.2 Effective cocatalysts for hom ogeneous p olym erisatio n...10
A ctivation by m ethylalum inoxane... 10
A ctivation by borate species...11
2.2.3 Polym erisation m echanism ... 11
2.2.4 Significance o f catalyst s y s te m ... 13
2.3 T ransition m etal-based catalysts for the copolym erisation o f ethylene and methyl m e th a c ry la te ... 16
2.4 C hrom atographic techniques for polyolefin an aly sis... 20
2.4.1 Historical developm ent o f chrom atographic tech n iq u es...21
2.4.2 G radient High Performance Liquid Chrom atography (gradient-HPLC) and Liquid Chrom atography under Critical Conditions (L C C C )...24
2.4.3 GPC coupled to FTIR spectroscopy... 25
2.4.4 Crystallisation Analysis Fractionation (C R Y ST A F)...26
2.5 R efere n ces... 30
Chapter 3
Experim ental 3.1 In tro d u ctio n ... 36 3.2 M aterials...36 3.2.1 S o lv e n ts...36 3.2.2 M on om ers... 37 3.2.3 C a ta ly s ts ...37Synthesis o f isotactic polypropylene... 37
Synthesis o f syndiotactic polypropylene...37
Synthesis o f ethylene/M M A copolym er...38
3.2.4 C ocatalysts... 38
3.3 P oly m erisatio ns... 39
3.3.1 Polym erisation o f ethylene and propylene w ith higher a-o lefin s... 39
3.3.2 Ethylene/M M A copolym erisation... 42
3.4 Com m ercial p o ly m e rs... 44
3.5 E x tractab les... 44
3.6 A nalytical equipm ent... 45
3.6.1 Therm al analysis... 45
D ifferential scanning calorim etry (D S C )...45
Dynamic mechanical analysis (D M A )... 46
3.6.2 M olar mass determ ination... 46
3.6.3 Crystallisation Analysis Fractionation (C R Y STA F)...46
General p ro c e d u re ...46
M M A d e te c tio n ... 47
3.6.4 l3C- and ‘H-NM R S pectroscopy ...48
3.6.5 High tem perature gradient H PL C ...48
3.6.6 Fourier Transform ed Infra-Red (FTIR) spectroscopy... 49
3.6.7 GPC coupled to FTIR spectroscopy (L C -T ransform )... 49
Instrum ental se t-u p ...49
Experim ental set-u p...52
3.6.8 Liquid Chrom atography under Critical Conditions (LC C C )...53
3.7 R efere n ces... 54
Chapter 4
Polyolefin crystallisation in solution m easured by Crystallisation A nalysis Fractionation (CRYSTAF)
4.1 A b stract...55
4.2 In tro d u c tio n ... 56
4.3 Theoretical back g rou nd ... 56
4.4 Experim ental and re s u lts ... 59
4.4.1 Basic CRYSTAF a n a ly sis... 59
4.4.2 T roubleshooting... 62 4.4.3 A verage Tc... 63 4.4.4 Sam ple w e ig h t...63 W eight in c re ase... 63 M odest amounts o f L D P E ...64 4.4.5 M ethod set-u p ...65
D issolution tem perature... 65
D issolution tim e... 65
Stabilisation tim e ... 66
C ooling rate during analy sis... 66
4.4.6 D ifference in so lven ts... 68
4.4.7 Extraction o f sam ple...70
4.4.8 C om onom er content distribution...71
4.5 S um m ary... 74
4.6 R e fe re n c e s...76
Chapter 5
C opolym erisation o f propylene with higher a-olefln s in the presence o f the syndiospecific catalyst /-Pr(Cp)(9-Flu)ZrCl2 / M AO 5.1 A b stract...775.2 In tro d u c tio n ...78
5.3 R esults and D iscu ssion ... 79
5.3.1 M icro stru ctu re... 81
5.3.2 Therm al A nalysis...91
5.5 R eferen ces...103
Chapter 6
C opolym erisation o f ethylene with higher a-olefins in the presence o f the m etallocene catalyst E t(Ind)2ZrCl2 / MAO 6.1 A bstract...1056.2 In tro d u ctio n ...106
6.3 Results and D iscussion... 107
6.3.1 S ynthesis... 107
6.3.2 M icro stru ctu re... 110
6.3.3 Therm al analysis...111
6.3.4 C R Y S T A F ... 117
Gradient copolym er fo rm atio n ... 117
Com parison o f CRYSTAF and DSC analysis... 120
6.3.5 FTIR spectroscopy...121
6.3.6 GPC coupled to FTIR spectroscopy via L C -T ransform ... 122
6.4 Sum m ary...124
6.5 R efere n ces... 125
Chapter 7
Ethylene and M ethyl M ethacrylate block copolym ers synthesised by the M e2C C pIndZrM e2/B (C6Fs)3 catalyst system
7.2 In tro d u c tio n ...129
7.3 T heory o f chrom atographic separation... 131
7.3.1 G radient H P L C ... 131
G e n e ra l...131
G radient steep ness... 133
G radient separation (Linear Solvent Strength (LSS) M odel)... 133
7.3.2 Liquid chrom atography under critical conditions (L C C C )...135
7.4 R esults and D iscussion...136
7.4.1 13C- and 'H -N M R spectroscopy...137
7.4.2 G P C ...141
7.4.3 Therm al analysis... 143
7.4.4 Infrared spectroscop y... 145
7.4.5 GPC coupled to FTIR spectroscopy (L C -T ransform )... 146
7.4.6 C R Y S T A F ... 149
7.4.7 H igh tem perature gradient HPLC (H T G -H P L C )...152
7.4.8 H igher Tem perature Liquid Chrom atography under Critical Conditions (HT- L C C C )...158 7.5 Sum m ary... 164 7.6 R e fe re n c e s...166
Chapter 8
C onclusions 8.1 C R Y S T A F ...1688.2 Copolym ers o f propylene/higher a-olefins...168
8.3 Copolym ers o f ethylene/higher a-o le fin s...169
8.4 B lock copolym ers o f ethylene and methyl m eth acry late... 170
List of Figures
Chapter 2
Figure 2.1. Steric control as a function o f metallocene symmetry (Ew en’s symmetry rules).
Figure 2.2. Control o f comonomer incorporation (ethylene copolymers) through the manipulation o f catalytic structure.
Figure 2.3. Control o f molar mass (ethylene copolymers) through the manipulation o f catalytic structure (high- and very high reactivity).
Figure 2.4. Catalysts used for activity studies by Hocker et al.
Figure 2.5. Scheme illustrating the build up o f TREF (A) and CRYSTAF (B) instruments.
Chapter 3
Figure 3.1. A) Photograph o f the autoclave reactor and B) Schematic illustration o f the different components o f the autoclave reactor.
Figure 3.2. Schematic illustration o f the reactor set-up used for the synthesis of ethylene/MMA copolymers, as employed at the University o f Aachen.
Figure 3.3. Solvent profile, as obtained for high temperature gradient HPLC separation.
Figure 3.4. Schematic illustration o f the optical sample press used for the IR analysis o f all powdery and solute samples.
Figure 3.5. Illustration o f the (A) collection module, after the separation by the HPLC system, and (B) the optical module in a FTIR spectrometer to allow the scanning o f the disk.
Figure 3.6. The heating chamber o f the LC-Transform (A), with an enlargement o f the ultrasonic transducer (B).
Chapter 4
Figure 4.1. Schematic representation o f the various steps in CRYSTAF analysis.
Figure 4.2. Photos o f the CRYSTAF instrum ent’s main oven unit with its five reactors (A) and a single reactor (B) with the sintered glass filter (1) and the magnetic stirrer (2).
Figure 4.3. Schematic layout o f the CRYSTAF instrument showing all the valves, the five reactors, the infrared cell and the procedure currently performed by the instrument.
Figure 4.4. Basic CRYSTAF diagram, with the infrared signal presented in percentage points and the first derivative forming the peak.
Figure 4.5. GPC chromatograms o f Lupolen 3020D (LDPE) and Lupolen 5261Z (HDPE) with a shoulder at 1.5xlO6 g/mol.
Figure 4.6. The same LDPE (Lupolen 3020 D) sample run at two different sample weights, showing the formation of a double peak.
Figure 4.7. Change in Tc, for 20 mg samples, with change in the cooling rate.
Figure 4.8. Comonomer content o f propylene/ 1-hexene copolymers, as a function o f temperature (at cooling rates o f 0.1 °C/minute and 0.4 °C/minute). Solvent: DCB.
Figure 4.9. Propylene/1-hexene copolymers crystallised from different solvents (TCB, DCB, CB and TClEt).
Figure 4.10. A) CRYSTAF plots and B) the first derivative curves o f an extracted (blue line) and a non-extracted (black line) ethylene/ 1-octadecene copolymer sample.
Figure 4.11. Size exclusion chromatogram o f the ethylene copolymer containing un-reacted 1-octadecene as comonomer (M n = 252 g/mol).
Figure 4.12. A) The linear calibration curve for a propylene/l-hexene copolymer series and a CRYSTAF chromatogram o f such copolymers on the same temperature scale, B) Comonomer content distribution curve calculated for the CRYSTAF curve using the calibration curve.
Chapter 5
Figure 5.1. 3-Dimensional illustration o f /-Pr(Cp)(9-Flu)ZrCl2 pre-catalyst (white = hydrogen atoms, grey = carbon atoms, cyan = zirconium and green = chlorine atoms).
F ig u re5.2. I3C-NMR spectrum o f the propylene/ 1-octadecene copolymer O il synthesised with 1. The indices PP and PC indicate propylene-propylene and propylene-comonomer dyads. Numbers refer to C-atoms o f the long chain branches as designated.
Figure 5.3. Abundance o f the rrrr pentad as function o f the comonomer content for copolymers o f propylene with higher a-olefins synthesized with 1.
Figure 5.4A. Expansion o f the methyl region ( 13C NMR) o f the propylene building block in the propylene/1-hexene copolymer H5 with 1.88 mol-% comonomer content, synthesised with 1.
Figure 5.4B. Expansion o f the methyl region ( 13C NMR) o f the reference polypropylene sample (s-PP) synthesised with 1. The pentad assignment as carried out according to Busico et al [19].
Figure 5.5. Abundance o f the error pentads rm m r and rrmr (%) as function o f the comonomer content.
Figure 5.6A-D. Monomer insertion at alternative sites o f the active catalyst center 1.
Figure 5.7A-D. Chain migration to the unoccupied site without monom er insertion, leading to skipped insertion error.
Figure 5.8A-D. Monomer units, with wrong orientation towards the ring structure, are sterically hindered and will lead to reversed enantioface insertion.
Figure 5.9. 13C-NMR spectrum o f a copolymer o f propylene and 1-hexene, with a 1- hexene content o f 11.67 mol-%.
Figure 5.10A. DSC heating curves (2nd heating cycle, heating rate 10 °C/min) o f syndiotactic propylene/ 1-hexene copolymers with varying 1-hexene content.
Figure 5.10B. DSC cooling curves (cooling rate 10 °C/min) o f syndiotactic propylene/1- hexene copolymers with varying 1-hexene content.
Figure 5.11. Melting temperature, Tm (dsc), determined by DSC and crystallisation temperature, Tc(c r y s t a f), determined by CRYSTAF o f syndiotactic propylene/higher
a-olefm copolymers synthesised with 1 as function o f the comonomer content.
Figure 5.12. Heat o f fusion for propylene/higher a-olefm copolymers synthesised with 1 as function o f the comonomer content.
Figure 5.13. DSC heating curves (2nd heating cycle) o f the propylene/1-hexene copolymer H8, the propylene/1-dodecene copolymer (D8) and the propylene/1- octadecene (0 5 ) copolymer at heating rates o f 10 °C/min and 5 °C/min.
Figure 5.14. First derivative o f the concentration profile for propylene/1-hexene copolymers, determined in the CRYSTAF experiment as a function o f temperature.
Figure 5.15A. Melting (Tm) and crystallisation (Tc) temperatures determined by DSC and CRYSTAF, respectively, overlaid in such a way that the temperature difference for the s-PP reference sample is cancelled out for propylene/1-hexene copolymers synthesised with 1.
Figure 5.15B. Melting (Tm) and crystallisation (Tc) temperatures determined by DSC and CRYSTAF, respectively, overlaid in such a w ay that the temperature difference for the s-PP reference sample is cancelled out for proplyene/l-octadecene copolymers synthesised with 1.
Chapter 6
Figure 6.1. 3-Dimensional illustration o f rac-Et(Ind)2ZrCl2 pre-catalyst (white = hydrogen atoms, yellow = carbon atoms, cyan = zirconium atom and green = chlorine atoms).
Figure 6.2. The amount o f comonomer added under feed conditions as a function o f the comonomer incorporation in the copolymer, as determined by l3C NMR spectroscopy.
Figure 6.3. 13C NM R spectrum o f the ethylene/1-octadecene copolymer (E 0 6 ) synthesised with 1.
Figure 6.4. DSC heating curves (2nd heating cycle, heating rate 10 °C/min) o f ethylene/1-tetradecene copolymer samples with varying 1-tetradecene content.
Figure 6.5. DSC cooling curves (cooling rate 10 °C/min) o f ethylene/1-tetradecene copolymer samples with varying 1-tetradecene content (increasing from top to bottom).
Figure 6.6. M elting temperature, Tm (Dsc), and crystallisation temperature, Tc (Dsc)> determined by DSC, as well as the crystallisation temperature, Tc (c r y s t a f),
determined by CRYSTAF (o f ethylene/higher a-olefin copolymers synthesised with 1 as function o f the comonomer content.
Figure 6.7. Melting and crystallisation temperatures determined by DSC and CRYSTAF, respectively, overlaid in such a way that the temperature difference for the s-PP or PE reference samples are cancelled out.
Figure 6.8. Hypothetical illustration o f a broad and narrow chemical composition distribution, as observed from CRYSTAF analysis.
Figure 6.9. CRYSTAF curves o f ethylene/l-tetradecene (A) and ethylene/1-octadecene (B) copolymers synthesised with 1, containing different amounts o f comonomer (see Table 1).
Figure 6.10. Comparison o f the curves o f A = E 0 2 (1.50 mol-%) and B = E 0 6 (3.55 mol-%), as obtained by CRYSTAF and DSC analysis.
Figure 6.11. The ATR-FTIR spectrum o f ET4, with peaks o f interest allocated numbers (1) 2915 cm '1, (2) 2847 cm '1, (3) 1471 cm '1, (4) 1462 cm '1, (5) 1377 cm '1, (6) 731 cm '1 and (7) 721 cm '1.
Figure 6.12. Chromatograms produced by the coupling o f FTIR to GPC via LC- Transform where peaks at 1462-1471 cm '1 and 1377 cm '1 were used to plot the area chemigrams and this data was used to determine the percentage methyl groups in the samples.
Chapter 7
Figure 7.1. The active catalysts species for the synthesis o f isotactic PMMA, discussed in Chapter 2, as reported by Hocker et al.
Figure 7.2. An illustration o f A) the three dimensional projection and B) Newman- projection o f a MMA unit in an isotactic PM MA homopolymer. (The methylene protons (He and H,) are not in the same chemical environment.)
Figure 7.3. The 300-MHz proton NMR spectrum o f isotactic PMMA, where the peak shift can be attributed to the tacticity o f the polymer.
Figure 7.4. 300 MHz proton spectra o f A) polyethylene, B) an ethylene and MMA copolymers (T8) and C) PMMA.
Figure 7.5. Representation o f an ethylene and methyl methacrylate block copolymer.
Figure 7.6. 13C NMR spectrum o f sample T8 where both the poly(methyl methacrylate) and the poly(ethylene) can be seen.
Figure 7.7. Molar mass distribution o f TO synthesised via 2.
Figure 7.8. Mono-, di- and trimodal molar mass distribution GPC curves for samples T 1,T 8 , T9 and T i l .
Figure 7.9. DSC heating curves o f the copolymer samples TO (homopolymer), T l, T8, T9 and T i l .
Figure 7.10. Peak identification o f IR spectra for the poly(ethylene/M M A) sample T8.
Figure 7.11. LC-Transform chromatograms for (A) T l, (B) T8, (C) T9 and (D) T i l , with data given in the form o f Gram-Schmidt chromatograms, wavenumber height chemigrams at 1730 cm '1 and 720 cm’1, as well as the relative amount o f ethylene in the samples.
Figure 7.12. An illustration o f the proposed monomer shielding through the formation o f hydrophilic ethylene chains around the active catalyst site. A) depicts the effect o f long ethylene sequences, while B) represents the effect o f shorter ethylene sequences.
Figure 7.13. CRYSTAF curves for ethylene/MM A copolymers, containing increased amounts o f ethylene from T l 1, T9, T8 to T l.
Figure 7.14. Carbonyl and total concentration detection for T1 (A), T8 (B) and T9 (C), with a 5.82 micron filter (in perchloroethylene as solvent).
Figure 7.15. A) Chromatogram for PE and PMMA standards with MM as outlined in the legend, as well as B) Chromatograms o f T l, T8 and T9 with different elution regions marked 7-10. (Method: Gradient: 100% DMF to 100% TCB, column: Nucleosil Cig, flow speed 0.5ml/min, ELSD-PL director.)
Figure 7.16. High temperature gradient HPLC, with LC-Transform as detector, for samples T l (A), T8 (B) and T9 (C). Peak-height chemigrams were chosen for wavenumbers 1730 cm '1 (carbonyl group) and 720 cm '1 (ethylene group). The factors 8x (A), 5x (B) and 3x (C) are the ratios by which the cheimgrams at 720 cm '1 were enlarged.
Figure 7.17. Illustration o f the separation o f different MM samples by changing the solvent composition to show size exclusion-, critical- and adsorption modes respectively.
Figure 7.18. Critical diagram o f MM vs retention time for PM M A and PE standards under different solvent/non-solvent conditions. Stationary phase: Nucleosil 300 silica gel; mobile phase TCB and CH; separation temperature 140 °C.
Figure 7.19. Critical chromatograms for a range o f PM M A standards at the CP o f 34.5/65.5 (V/V%) TCB/CH.
Figure 7.20. Chromatogram under critical conditions o f PE standards (M p = 2030 g/mol and 22 000 g/mol) (A), T l (B), T8 (C) and T9 as well as T l 1 (D). Stationary phase: Nucleosil 300
A
silica gel; mobile phase 34.5/65.5 (V/V%) TCB/CH; analysis temperature 140 °C.List of Schemes
Chapter 2
Schem e 2.1. Activation o f precatalyst with (i) methylalumoxane and (ii) a borate species.
Schem e 2.2. Schematic representation o f coordination and insertion during olefin polymerisation with transition metals.
Schem e 2.3. Chain-end and enantiomorphic site mechanisms o f stereocontrol [18].
Schem e 2.4. Reaction diagram o f the ring formation mechanism resulting from two M M A monomers during polymerisation with 1.
Schem e 2.5. Reaction mechanism for a two-component catalyst system for the polymerisation o f MMA.
Schem e 2.6. Reaction scheme as proposed by H ocker et al. [29], for the copolymerisation o f ethylene and MMA with 12.
Chapter 3
Schem e 3.1. Pre-catalysts used in the polymerisations o f monomers considered in this study: Et(Ind)2ZrCl2 [1], (CH3)2Si(2-methylbenz[e]indenyl)2ZrCl2 [2], /-Pr(Cp)(9- Flu)ZrCl2 [3] and Me2 CCpIndZrM e2 [4],
Chapter 5
Schem e 5.1. Synthesis o f copolymers o f propylene with 1-hexene, 1-dodecene or 1-octadecene, respectively, using the catalyst system 1/MAO.
Schem e 5.2. Pentad distribution o f stereoerrors in syndiotactic polypropylene characteristic for catalytic-site control and for chain-end stereocontrol.
Chapter 6
Schem e 6.1. Reaction diagram for the synthesis o f ethylene/1-decene, /1-tetradecene or /I-octadecene copolymers via the 1/MAO catalyst system.
Chapter 7
Schem e 7.1. Reaction scheme as proposed by Hocker et al. [3], for the copolymerisation o f ethylene and MMA with 2.
List of Tables
Chapter 2
Table 2.1. Development o f metal based catalysts for the polymerisation o f polyolefins in the past 50 years [1].
Chapter 3
Table 3.1. Comonomer contents as calculated from 13C N M R spectroscopy. Quantities of propylene and comonomer added under feed conditions.
Table 3.2. Comonomer contents as calculated from l3C N M R spectroscopy. Quantities of propylene and comonomer added under feed conditions.
Table 3.3. Comonomer content as calculated from 13C N M R spectroscopy for the isotcatic propylene/1-hexene copolymers.
Table 3.4. Reaction conditions for the synthesis o f poly(ethylene-co-M M A), with reference to the monomers, B(C6F5)3 as cocatalyst and the pre-catalyst 4 added, as well as the polymerisation times for each reaction.
Table 3.5. PMMA and PE standards used for the calibration o f the HTG-HPLC and the LCCC systems.
Table 3.6. Parameter changes for the TCB and DCB temperature programs in CRYSTAF analysis.
Table 3.7. Instrumental settings as used for sample collection, with different samples and separation methods.
Chapter 4
Table 4.1. Tc measurements for different samples weights, where the method o f the runs was changed for every second row.
Table 4.2. Tc (c r y s t a f) for propylene/1-hexene copolymers as obtained in
trichlorobenzene (TCB), dichlorobenzene (DCB), chlorobenzene (CD) and tetrachlorobenzene (TClEt) as solvents. Mw and polydispersity o f these samples are also shown.
Chapter 5
Table 5.1. Comonomer content in the feed and in the copolymer for propylene/higher a- olefin copolymers synthesised with 1. The abundance o f the [rrrr] pentad and the main [rmmr] and [rrmr] error pentads are listed in percent. Mn is the num ber average m olar mass in g/mol and M J M n the polydispersity.
Table 5.2. Assignment o f signals ( 13C NMR) to the respective C-atoms o f propylene/higher a-olefin copolymers synthesised with 1. The indices PP, PC and CC indicate propylene-propylene, propylene-comonomer and comonomer- com onom er dyads. Numbers refer to C-atoms o f the long chain branches, as designated in Figure 5.2.
Table 5.4. Melting temperatures, Tm (d s c), and crystallisation temperatures, Tc (c r y s t a f),
o f propylene/higher a-olefin copolymers synthesised with 1, determined by DSC and CRYSTAF, respectively. Tc (dsc, heat) and Tc (Dsc) are the exothermic peaks during the second heating cycle and the cooling cycle, respectively, AH the heat o f fusion (J/g) and Tg the glass transition temperature. All temperatures are stated in °C.
Chapter 6
Table 6.1. Reaction conditions for the preparation o f ethylene copolymers, with reference to the quantities o f (co)monomers added during the polymerisation process, as well as data regarding the subsequent analysis o f the copolymers. Values regarding comonomer content ( 13C-NMR spectroscopy), crystallization temperature Tc (c r y s t a f), m elting temperature Tm (Ds q, crystallisation temperatureTc (d s c), AH the heat o f fusion (DSC), Tg (DMA), M n and M w/M n are given.
Table 6.2. Chemical shifts (ppm) for ethylene/higher 1-olefin (1-decene, 1-tetadecene and 1-octadecene) copolymers synthesized with 1 and the shift predictions, as obtained by Paul Grant calculations. Code numbering is the same as in Figure 6.3.
Chapter 7
Table 7.1. Tetrad and pentad assignments o f the synthesised PM M A compared to Ferguson’s predictions.
Table 7.2. The MMA content o f the ethylene/MMA copolymers synthesised via 2, as well as their peak maximum molar mass values (Mp) and further results obtained from thermal analysis.
13c
C arbon thirteen‘H Proton
A T R A ttenuated total reflection
B r Backbone carbon from w here side branches originate
CB C hlorobenzene
CC
Com onom er-com onom er interaction C C D Chem ical com position distributionCF Cross fractionation
CH Cyclohexanone
CP Critical point
C R Y STA F Crystallisation analysis fractionation
D CB 1,2-Dichlorobenzene
D M A Dynamic m echanical analysis D M F N ,N -Dim ethylform am ide
D SC Differential scanning calorim etry FTD Functionality type distribution F TIR Fourier transform infrared
G C Gas chrom atography
G PC Gel perm eation chrom atography H D PE High density polyethylene
H PLC High perform ance liquid chrom atography
H TG -H LPC H igh tem perature gradient high perform ance liquid chrom atography
ID Internal diam eter
i-PP Isotactic polypropylene
IR Infrared
LC C C Liquid chrom atography under critical conditions LD PE Low density polyethylene
LES Longest ethylene sequence
L L D PE Linear low density polyethylene m M eso (tacticity in N M R)
M AO M ethylalum oxane M M A M ethyl m ethacrylate M M D M olar mass distribution
M n Num ber-average molar num ber (g/mol) M p M olar mass peak m axim um (g/mol)
MP M elting point
M w W eight-average m olar mass (g/mol) N M R N uclear m agnetic resonance
PC Propylene-com onom er interaction
PE Polyethylene
PM M A Poly(methyl methacrylate)
PP Polypropylene
r Racem ic (tacticity in N M R) SCB Short chain branching
SCBD Short chain branch distribution SEC Size exclusion chrom atography s-PP Syndiotactic polypropylene
Tc Crystallisation point tem perature (°C)
Tc (c r y s t a f) Crystallisation point tem perature in solution m easured by CRYSTAF
Tc (d s c) Crystallisation point tem perature m easured by DSC (°C)
Tc (d s c, heat) Crystallisation tem perature on heating in D SC (cold crystallisation) (°C)
TCB 1,2,4-Trichlorobenzene TC lEt 1,1,2,2-Tetrachloroethylene Tg Glass transition tem perature (°C) TLC Thin layer chrom atography T m M elting point tem perature (°C)
T m (d s c) M elting point tem perature m easured by D SC (°C)
TREF Tem perature rising elution fractionation AH H eat o f fusion (J/g)
Acknowledgements
I would like to thank the following people and institutions that made this study possible:
1) Prof. H Pasch, for his leadership during the course o f my study, showing me how best to approach analytical problems and the methods available for solving them. His unique insight into a problem and the critical manner in which he taught me to look at my data has proved to be very valuable.
2) Prof. R D Sanderson, for allowing me the opportunity to do part o f my studies overseas to acquire the necessary skills and knowledge, not only to finish m y Ph D but also enabling me to assist other students in their research.
3) SASOL Polymers, the National Research Foundation o f South Africa, the Jiilich fund in Germany, as well as the BMBF in Germany, for their financial support during my studies both in South Africa and Germany.
4) S Balk and H Keul at the Institut fur Technische Chemie und Makromolekulare Chemie at RWTH Aachen, Germany. The ethylene/methyl methacrylate copolymer samples where obtained through collaboration with this group. Valuable information was exchanged in meetings regarding the reaction mechanism for the synthesis o f ethylene/M M A with metallocene catalysts and the possible products formed during the reactions.
5) B Monrabal, from Polymer Char, Valencia Spain, who assisted me in method development on the CRYSTAF. Thanks to him I also had the opportunity to visit Spain and see the manufacturing process o f the CRYSTAF. Further assistance was rendered in the analysis o f the ethylene/methyl methacrylate copolymer samples by means o f a special carbonyl detector on the CRYSTAF.
6) C Brinkmann, for assistance with the gel permeation chromatography, R Briill, with the LC-Transform data analysis, and T Macko, for analyses o f the ethylene/methyl methatcrylate copolymers by liquid chromatography under critical conditions at the Deutsches Kunststoff Institut in Darmstadt Germany. I also acknowledge all the other staff and students at the Institute who assisted me during my visits.
7) S de Goede at SASOL Polymers, my mentor for the duration o f my studies, on whom I could always rely for advice.
8) I would like to thank my fellow students and friends at the Institute for Polymer Science that were there in the good as well as the bad times. In this regard, I would especially like to mention A van Zyl and J M cLeary for their stimulating discussions regarding the obstacles I faced during the course o f my study.
9) M y friends, that made these years as memorable as they were.
10) A special thanks also goes to M Smit for her support during my studies and her assistance with correcting the manuscript.
11) Dr. A van Reneen, for advising me on changes that could be made in the method o f the reaction setup, as well as the interpretation o f the N M R data.
12) A Fourie and E Cooper for assistance with official matters.
13) Dr M J Humdall, who assisted me with the grammatical corrections o f both my articles and the final manuscript.
14) Last, but by no means least, I would like to thank m y parents who always had confidence in my abilities and believed that I could achieve anything I set my mind to, despite my dyslexia. I would like to thank them for teaching me the moral o f life, what is right and what is wrong, how to behave and to respect other people.
Publications and posters that have resulted from this
research
1. Copolymerization o f Propylene with Higher a-Olefins in the Presence o f the Syndiospecific Catalyst i-Pr(Cp)(9-Flu)ZrCl2/MAO\ SM Graef, U W ahner, R Brull, AJ van Reenen, RD Sanderson and H Pasch; Journal o f Polym er Science Part A; Polymer Chem istry; Volum e 40, N um ber 1, issue dated January 1, 2002.
2. Synthesis and Characterisation o f Syndiotactic Polypropylene Copolymers with Higher a-Olefins\ SM Graef, H Pasch, U W ahner, R Brull, J van Reenen, RD Sanderson; 4th annual U N ESCO school and IUPAC conference; 2001, Stellenbosch, South Africa.
3. Synthesis and Characterisation o f Syndiotactic Polypropylene Copolymers with Higher a-Olefins; SM Graef, H Pasch, U W ahner, R Brull, J van Reenen, RD Sanderson; Europolym er Congress, 15-20 July 2001; Eindhoven U niversity o f Technology; Eindhoven; Netherlands.
4. Synthesis and Characterisation o f Syndiotactic Polypropylene Copolymers with Higher a-Olefms\ SM Graef, H Pasch, U W ahner, R Brull, J van Reenen, RD Sanderson; 222nd ACS N ational M eeting; 26-30 A ugust 2001; Chicago; Illinois; U nited States o f America.
5. Polyolefinanalytik mit der Kristallisationsfraktionierung\ R Brull, SM G raef, H Pasch, K Rode, U W ahner; GDCh Fachgruppen Tagung; 14 M arch; 2002; D arm stadt; Germany.
6. Copolymers o f Methyl Methacrylate and Ethylene Utilizing Methylated Metallocene Catalysis-, SM Graef, S Balk, H Keul, RD Sanderson, H Pasch; 5nd annual UNESCO school and IUPAC conference; 25-28 M arch; 2002; Stellenbosch; South Africa.
CHAPTER 1
Overview
1.1 Introduction
D evelopm ents in the polyolefin field have seen explosive growth over the last two decades for m etallocenes catalysts, advancing from academ ic interest to industrial applications. Statistics recently published show that low density (LD PE)-, linear low density (LLDPE)-, high density polyethylene (HDPE) and polypropylene (PP) make up the bulk mass o f the 120 m illion tonnes o f polym ers that are produced annually [1].
T he exponential growth in the num ber and diversity o f polyolefin products has not been m et w ith the same technological advances in the field o f analysis. O nly in recent years have scientists developed advanced chrom atographic m ethods for the tailor- m ade analysis o f these sem i-crystalline polymers i.e. high tem perature gel perm eation chrom atography (H T-GPC), G PC -FTIR (Fourier transform infrared spectroscopy) coupling, tem perature rising elution fractionation (TREF) and crystallisation analysis fractionation (CRYSTAF). Further advances in polyolefin research, w ith special reference to the once thought im possible copolym erisation o f a-olefins with polar m onom ers [2], has lead to further challenges for the analyst in this field.
The potential o f various chrom atographic methods, w ith reference to the analysis o f polyolefins and ethylene/ methyl methacrylate copolym ers prepared by using m etallocene catalysts, as well as their limitations are the focus o f this study. C R Y STA F is a new technique that was developed for the determ ination o f the crystallisation tem perature o f sem i-crystalline polym ers in solution. The effect o f
under discussion was studied. B y studying the crystallisation curves obtained by CRYSTAF and the spectra obtained by N M R spectroscopy, deductions regarding the reaction m echanisms could be made. We report here, for the first tim e in literature, the separation o f ethylene/m ethyl methacrylate copolym ers b y gradient HPLC (high perform ance liquid chrom atography) at elevated tem peratures. High tem perature gradient HPLC was used in the separation o f ethylene/m ethyl methacrylate copolymers to determ ine the presence o f block structures. U se o f LC-Transform enabled for the unique coupling o f the GPC or high tem perature gradient HPLC to FTIR spectroscopy. This coupling allowed for the sim ultaneous detection o f the different m onom er units (ethylene/a-olefins and ethylene/m ethyl m ethacrylate) in a chromatogram.
1.2 Objectives
1. CRYSTAF is a new analytical technique with room for m ethod improvement. This was to be addressed in terms o f the following.
• Instrum ental param eter settings for analyses w ill be investigated to show w hich param eters should be im proved for better and faster analysis.
• The influence o f the tared am ount o f samples used for analysis and the effect o f solvent on the crystallisation tem perature in CRY STA F will be investigated.
• By using copolym ers o f known com onom er com position and draw ing up a standard calibration curve, the com onom er content distribution o f an unknow n copolym er will be determined.
2. The product o f the copolym erisation o f propylene/ higher a-olefins w ith the /-Pr(Cp)(9-Flu)ZrCl2/ M AO catalyst will be analysed.
• N M R spectroscopy will be used to the study the reaction m echanism o f the catalyst system.
• DSC curves will be analysed to ascertain w hether there is any relationship between the m elting tem peratures and the com onom er incorporation.
• Investigation will also focus on the different m elting and crystallisation peaks that occur on heating and cooling in DSC, with regarding to the phenom enon o f cold crystallisation.
• A relationship between the crystallisation in CRY STA F and DSC results will be checked to see if the data show any correlations.
3. The product o f the copolym erisation o f ethylene/ higher a-olefins with Et(Ind)2ZrC l2 / M AO catalyst will be analysed.
• The relationship between the crystallisation in CRY STA F and DSC results will be followed for copolym ers w ith different com onom er concentrations to see if the data show any correlation.
• CRYSTAF diagram s will be studied to determ ine the heterogeneity in different copolym er samples.
• By coupling FITR spectroscopy to GPC, this investigation will focus on the com onom er distribution according to m olar mass.
4. D ifferent analytical techniques will be em ployed to prove the presence o f block copolym er form ation during the synthesis o f ethylene and M M A w ith the M e2C(Cp)(Ind)ZrM e2 / B(C6F5)3 catalyst. These include:
• GPC • DSC
• GPC coupled to FITR spectroscopy via LC-Transfrom • High tem perature gradient HPLC
• High tem perature LCCC
1.3 Layout of this dissertation
The range and limitations o f the above-m entioned chrom atographic m ethods will be discussed under the following headings while, for ease o f reference, the experimental details are all given in Chapter 3.
1.3.1
Polyolefin
crystallisation
in
solution
measured
by
Crystallisation Analysis Fractionation (CRYSTAF)
The separation o f polym er fractions, as a function o f a tem perature profile, was introduced by B. M onrabal in 1991 [3]. This method has been successfully used for the analysis o f a-o lefin s, providing inform ation regarding a particular polym er’s chem ical com position distribution. The purpose o f this investigation is to assess the effect o f instrum ental param eters on experim ental results obtained by this m ethod, processing the obtained data and looking at further developing this m ethod for the analysis o f specific polym ers (Chapter 4).
1.3.2 Copolymerisation of propylene with higher a-olefins in the
presence of the syndiospecific catalyst /-Pr(Cp)(9-Flu)ZrCl2/ MAO
D etailed analysis o f propylene and higher a-o lefm copolym ers (1-hexene, 1-dodecene and 1-octadecene) by nuclear m agnetic resonance (NM R) spectroscopy, differential scanning calorim etry (DSC) and C R Y STA F analysis is described. Special attention was given to the catalyst m echanism, resolved by N M R spectroscopy (C hapter 5).1.3.3 Copolymerisation of ethylene with higher a-olefins in the
presence of the metallocene catalyst Et(Ind)2ZrCl2/ MAO
This section can be regarded as supplem entary to the LLDPE sam ples analysed by CRY STA F, as seen in C hapter 4, as well as results o f existing data as published in literature on ethylene copolym ers [4]. Similarities betw een the results obtained in this section and that o f the polypropylene analysis (in C hapter 5) w ill be highlighted. The
com onom er distribution heterogeneity o f the samples will be investigated in term s o f both the m olar mass distribution (GPC-FTIR) and the chem ical com position distribution (CRYSTAF) (Chapter 6).
1.3.4
Ethylene
and
methyl
methacrylate
block
copolymers
synthesised by metallocene catalysts
The unique copolym erisation o f a polar monom er w ith an a-olefm by means o f a m etallocene precatalyst and a borate cocatalyst was first reported by H ocker et al. [2]. Polym er sam ples (ethylene/M M A copolym ers) supplied by this research group were analysed via spectroscopic and chrom atographic m ethods to obtain inform ation regarding their chem ical com position distribution and microstructure. The subsequent analysis o f these polym ers did not allow for the use o f existing standard and high tem perature chrom atographic methods. For this reason new and innovative analytical methods had to be developed, nam ely high tem perature gradient HPLC and high tem perature liquid chrom atography under critical conditions. These m ethods and results o f analyses are described in C hapter 7.
1.4 References
1. DPI, Annual report, 2001.
2. Frauenrath H, Balk S, Keul H, H ocker H, M acromol. Rapid. Commun., 2001, 22, 1147.
3. M onrabal B, US Patent 5 222 390, 1991.
CHAPTER 2
Historical and theoretical background
2.1 Introduction
The aim o f this chapter is to introduce the field o f transition m etal catalysis and the theory behind these transition metal catalysed polyolefin system s, as well as to discuss the significance o f analytical m ethods for polyolefm analysis.
2.2 Transition metal-based catalysts for polyethylene and
polypropylene synthesis
2.2.1 Historical
A rguably, one o f the most significant discoveries in the field o f polym er science in the last 40 years was the polym erisation o f olefins using transition m etal-based catalysts. The details regarding the chronological developm ent o f these system s is sum m arised in T able 2.1 [1].
Karl Z iegler was the first person to discover that transition m etal halides (T1CI4, TiCl3 and Z rC l4), in com bination w ith alkylalum inum com pounds as cocatalysts, could be successfully used to synthesise linear high m olar m ass polyethylene and isotactic polypropylene [2, 3]. These catalysts were known as heterogeneous, multi active site catalysts.
Table 2.1. Development of metal based catalysts for the polymerisation o f polyolefins in the past 50 years [1].
Y ear Catalyst system Polym er A ctivity
(kg-polym er / g-metal)
Tacticity
r% i
1953-1955 TiC l4-Et3Al Polyethylene 10-15
-Polypropylene 5-10 50-60
T iC l3-Et3Al Polyethylene 5-10
-Polypropylene 1-3 60-80
1955-1960 TiC l3 electron donors-Et3Al Polyethylene -
-Polypropylene 1-10 90-95
1970-Present Supported M gCl2 Polyethylene 500-1000
-TiCl4-Et3Al Polypropylene 500-1000 560-70
Supported M gC l2 electron Polyethylene 500-1000
-D onor TiC l4-Et3Al Polypropylene 300-500 90-99
1980-Present H om ogeneous m etallocene-m ethylalum inoxane Polyethylene 400-500
-1985-Present Stereorigid m etallocene-m ethylalum inoxane Polypropylene 150-170 96-98
H om ogeneous alum inum free Polypropylene 0-1 97-98
M etallocene catalysts Polypropylene
Supported m etallocene Polypropylene 10-20 80-90
Catalysts/A1R3 or M AO Polypropylene
1986-Present Ti(O R )4-M AO Syndiospecific polystyrene 0.4-0.5 80-96
C pTi(O R )3-M AO Syndiospecific polystyrene 3300 82
M g(O H )2/Ti(O bu)4- M AO Syndiospecific polystyrene 0.3 100
1988-Present Cp2ZrC l2-M AO Ethylene-propylene copolym er 50-100 A specific
Et[IndH 2]ZrC l2-M AO Ethylene-propylene copolym er 5-15 Isospecific
/-Pr(C p)(Flu)ZrC l2-M AO Ethylene-propylene copolym er Syndiospecific
C p2ZrC l2-M AO Ethylene-hexene copolym er - Aspecific
Et[IndH 2]ZrC l2-M AO Ethylene-hexene copolym er Isospecific
/-Pr(C p)(Flu)ZrC l2-M AO Ethylene-hexene copolym er Syndiospecific
-Shortly thereafter, N atta independently discovered the stereoregular nature o f polym ers o f a-olefins such as propylene, 1-butene and styrene [4-7]. In 1963, the N oble Prize for chem istry was awarded to Ziegler and N atta for their contribution to the field o f polym er science.
The transition metal catalyst range was broadened by the introduction o f more transition metal com pounds from groups IV-VII [8], Im provem ent w ith regard to the efficiency o f the current systems, as well as the econom ics o f polyolefm processing through elim ination o f catalyst removal and solvent purification/rem oval steps, was later found with the introduction o f M gCl2 supported catalyst systems [9, 10].
The next revolution in polyolefin synthesis via transition metal catalysts was led by Breslow and N ew burg [11] with the identification o f the homogeneous (Cp)2TiC l2/alkylalum inum catalyst species. The significance o f the homogenous character o f a catalyst is its single site activity that produces polyethylenes o f high m olar m asses and narrow m olar mass distributions. H owever, the discovery o f hom ogeneous catalyst systems did not attract w idespread interest due to their poor catalytic activity, their short lifetime and lack o f product stereospecificity.
The next m ilestone in the synthesis o f polyolefm s was the discovery o f highly active and stereospecific catalyst species with m ethylalum oxane (M AO ) as cocatalyst. Reichert and M eyer observed an increase in catalytic activity w hen w ater was added to the C p2TiEtC l/A lEtC l2 catalytic system [12, 13]. This was follow ed by the discovery o f a significantly higher polym erisation activity o f 5 000 000 g PE/g Ti after the addition o f C p2TiM e2 to two equivalents o f trim ethylalum inium , previously treated w ith one equivalent o f w ater [14]. Further studies b y Sinn and Kaminsky showed that the increase in catalyst activity was a result o f the presence o f an oligom eric com pound, know n as m ethylalum inoxane (M AO).
The use o f w eakly bonded anion species to serve as cocatalysts in the activation o f transition metal catalyst species renew ed both com m ercial and scientific interest in hom ogeneous catalyst systems [15]. A nother type o f com pound also used lately with great success to obtain cationic active metal species for olefin polym erisation is the borate species.
A ctivation o f the homogeneous precatalyst can be obtained w ith one o f two com pounds, as illustrated in Scheme 2.1.
Scheme 2.1. Activation of precatalyst with (i) methylalumoxane and (ii) a borate species.
A ctivation by m ethylalum inoxane
A lthough the properties o f M AO have been studied extensively since its discovery, quite a few questions still remain. It is know n that M AO has a m olar m ass o f 1500 g/mol. The exact structure is, however, not known. A fter various suggestions, the proposal by Sinn et al. [16] was accepted as the w idely used model. The hypothesis is that M AO consists o f a cage-like structure, w ith TM A dispersed w ithin the molecule.
MAO is an ideal cocatalyst, as it serves as activator for transition metal catalyst species, as seen in Scheme 2.1. It scavenges the im m ediate surroundings o f the active catalyst species for m oisture and polar com pounds, to prevent the deactivation o f the catalyst. Its last function is the activation o f inactive species during the polym erization process. The disadvantage o f this system is that M AO is used in
m olar ratios o f at least 1000:1 with regard to the catalyst species. This makes it a very expensive cocatalyst to use in industry and resulted in a further search for other effective cocatalysts. The result was the introduction o f the borate species as possible cocatalysts for transition metal based catalysts.
A ctivation by borate species
These species have the advantage that they are used in a 1:1 or 2:1 m olar ratio with the desired precatalyst. (A precatalyst is the transition metal species before activation by a cocatalyst). The most simple borate species used is B(C6F5)3. The draw back o f borate activated systems is that they do not function as scavengers and, in m ost cases, an extra activator is required.
2.2.3 Polymerisation mechanism
Polyolefins are produced by m ultiple insertions o f olefins into a m etal-carbon bond. Olefin insertion occurs by the m -o p e n in g o f the double bond (both new bonds are on the same side o f the inserting olefin) and w ith chain m igratory insertion (it is the alkyl group on the m etal that migrates to the olefin with a net exchange o f two available coordination positions on the metal centre). See Scheme 2.2.
W hen we have prim ary insertion (1,2), then the 1-alkene enantioface that is inserted preferentially is the one which, in the transition state, places its substituent anti to the first C-C bond o f the grow ing polym er chain (m inim ises non-bonded interactions).
The active m etal centre bearing the grow ing alkyl chain m ust have an available co-ordination site for the incom ing m onom er. Insertion occurs via chain m igration to the closest carbon o f the olefin double bond, w hich undergoes cw-opening, w ith the formation o f the new m etal-carbon and carbon-carbon bonds. The new C-C bond is then on the site previously occupied by the co-ordinated m onom er molecule.
■ i
Mt---\ %
Polymer
4-center transition state with cis opening o f the double bond
Polymer
Mt
Chain migration
Scheme 2.2. Schematic representation of coordination and insertion during olefin polymerisation with transition metals. (Mt = metal atom).
Points o f im portance in transition metal catalysed polym erisation o f olefins: • The metal atom must have a site for coordination
• Insertion occurs via chain m igration to the closest carbon o f the olefin double bond.
The four proposed m echanism s for polym erisation are sum m arised in the w ork o f Resconi et al. [17],
In short, the acceptable m echanism s all agree that:
• M onom er insertion is a tw o-step process, nam ely coordination followed by insertion;
• The active m etal m ust have an available coordination site;
• O lefin insertion occurs by c/s-opening o f the double bond, follow ed by chain m igratory insertion;
• The olefin has to coordinate face-on to the metal, w ith its double bond parallel to the m etal-carbon bond.
\
Mt---Polymer
1
Primary, anti coordination
The ligands and their substituents play an im portant role in the insertion o f the m onom er with reference to steric hindrance. The bite angle that the ligands form with each other is thus a vital variable and brings us to the concept o f fl«sa-metallocenes. Control o f the tacticity o f the formed polym er is possible.
A distinction between the active centre and the active site o f the catalyst precursor should be made: a m etallocene type active centre has a m inim um o f tw o sites on w hich chain grow th can take place. The nature o f the active site is determ ined by the metal, the Cp ligands, the geom etry and structure o f the m etal-bonded chain-end. D ifferent types o f last inserted m onom er will thus increase the num ber o f possible active sites. A difference in reactivity, regioselectivity and enantioface selectivity is therefore possible. The result is that the active centre itself changes during a single chain growth but, statistically, behaves the same from one chain to another. Such a species can therefore be described as a single-site catalyst [17].
2.2.4 Significance of catalyst system
The versatility o f m etallocene catalysts is seen in the various m odifications one can m ake to the catalyst system in order to create polym ers w ith specific properties that meet the needs o f the scientist and industry. Tw o possible sources o f enantioface selectivity in olefin insertion are possible (Scheme 2.3):
a) Stereoregularity o f the metal active site is based on an enantiom orphic site control m echanism for stereoselection. The relationship o f the chirality o f the two co-ordination sites o f the catalytic com plex determ ines the stereochem istry o f the polymer.
b) The last inserted m onom er unit is based on a chain-end control m echanism for stereoselection. Every m onom er insertion generates a new stereo centre.
m onom er unit.
A predictable relationship between com plex sym m etry and polym er tacticity exists. This can be seen from E w en’s sym m etry rules (Figure 2.1), com piled by Ewen et al.
[18-23] and K am insky et al. [24],
chiral coordination site
Pm and Pr refer to the probability of m e so and ra cem ic placements, respectively M = transition metal
P = polymer L = ligand
a = 1 or 0, then polymer is isotactic, for a = 0,5, the polymer is atactic
Scheme 2.3. Chain-end and enantiomorphic site mechanisms o f stereocontrol [17].
Single-site polym erisation catalysts can be divided into five m ajor sym m etry categories. It is assum ed that the polym er rapidly equilibrates w ith the available co-ordination site for the purposes o f assigning sym m etry. C2V symmetric m etallocene catalysts produce atactic polym ers or m oderately stereoregular polym ers by chain-end control mechanisms. Catalysts exhibiting Cs sym m etry, consisting o f m irror planes containing two diastereotopic co-ordination sites, behave sim ilarly and produce atactic polym ers. Cs sym m etric catalysts that have a m irror plane reflecting tw o enantiotopic co-ordination sites frequently produce syndiotactic polym ers. C2 sym m etric com plexes, both racemic and enantiom erically pure ones, typically
produce isotactic polymers via a site-control m echanism. The fifth sym m etry category, not mentioned in Figure 2.1, is the “so-called” oscillating m etallocenes.
The structure o f the catalyst precursor also has a significant influence on the com onom er incorporation and m olar mass o f the obtained copolym er, as can be seen in Figures 2.2 and 2.3.
Symmetry Example Polymer (ie Poiy(propyiene))
^ S 3
C2v O / *
Achiral *Zr""//, Atactic —I — i— i— i—
2 ( ^ \ Si i
C hiral S r H ^ """/a Isotactic _ | __I__ I__ |___I___ 1 racem icform
s C~S) S i ^ * Z r . I
A chiral V Z \ _ / ""a A tactic _ J ___________
m esom eric form
^ 3 ^ *CI
r H % / , i i
° s f ) z'-'ma
P rochiral S yn d io ta ctic _I_
<om
i i
__-C H3
Hfe T \ /
^ 1 'Z r -.IIIIQI
C hiral H em i-iso ta ctic
tBu 1 / ’
H3C i / A tactic
Z f N i i i a h3c
Figure 2.1. Steric control as a function of metallocene symmetry (Ewen’s symmetry rules) [17,25].
rCl
fair good excellent
Figure 2.2. Control of comonomer incorporation in ethylene copolymers by the manipulation of catalytic structure [26] (fair-, good- and excellent activity).
Figure 2.3. Control o f molar mass (ethylene copolymers) by the manipulation of catalytic structure [26] (high- and very high reactivity).
2.3
Transition
metal-based
catalysts
for
the
copolymerisation of ethylene and methyl methacrylate
In general, Ziegler-N atta catalysts such as T iC V A lR j and the chlorinated K am insky catalysts such as Cp2ZrCl2/(M A O) do not polym erise polar m onom ers. However, Collins et al. [27] show ed that som e m ethylated K am insky species like C p2ZrM e2 2 and [Cp2ZrM e(TH F)]+[BPh4]' can be used to polym erise M M A. These polym ers w ere syndiotactic and had m olar m asses betw een 62 900 g/m ol and 158 300 g/mol w ith polydispersities between 1.19 and 1.40. Soga et al. [28] reported on the synthesis o f m ostly syndiotactic poly(m ethyl m ethacrylate) (PM M A ) with Cp2Z r(C H3)2/P h3C B(C6F5)4/B (C6F5)3 as catalyst system, as w ell as isotactic PM M A
with rac-E t[Ind2Zr(CH3)2] as catalyst precursor. It was determ ined that the syndiospecific polym erisation proceeds via a chain-end controlled m echanism , while the isospecific polym erisation proceeds via an enantiom orphic site controlled mechanism.
The lanthanide elements can be used as metal ions in a m etallocene catalyst for the synthesis o f both polar and nonpolar monomers (acrylate and olefin based polym ers). Y asuda et al. [29, 30] exam ined the existence o f an eight-m em bered structure, in the M M A polym erisation with Sm(CsM e5)2(M M A)2H 1 as catalyst precursor, via single crystal X -ray analysis. They found that one M M A unit links to the metal in an enolate form, w hile at the other end the penultim ate M M A unit is attached to the metal through its C = 0 group. Two M M A units are thus used in this reaction m echanism. The assum ption for the initiation step was that the hydride attacks the C H2 group o f the M M A and a transient Sm O C(O C H3)=C(CH3)2 is formed, as illustrated in Scheme 2.4. The next M M A unit to be inserted at the reactive catalyst site attaches in a 1,4-addition, to produce an eight-m em bered cyclic interm ediate species. M arks et al. [31] used a chiral a«sa-lanthanide m etallocene to obtain isotactic PM M A.
Scheme 2.4. Reaction diagram of the ring formation mechanism resulting from two MMA monomers during polymerisation with 1.
m echanism s o f the two-com ponent catalyst systems 2 and 3. It was found that the m echanism involves two types o f zirconocene com plexes: a cationic methyl com plex that activates a monom er as an acceptor and, secondly, the grow ing chain methyl ester enolate com plex as a donor. The formation o f the carbon bond proceeds via a M ichael-type addition between the activated m onom er and activated grow ing chain, far aw ay from the zirconium centre (Scheme 2.5). The m arked methyl (* in Schem e 2.5) is the one end o f the propagating PM M A chain, w hile 4 and 5 are the enolate com plexes which are form ed in the interm ediate stages o f polym erisation. The reactions were carried out at 0 °C, where this specific catalyst lends itself for the production o f syndiotactic PMMA.
Initiation --- * * Me Me
2
@ Cp2ZrMe2 ZrCp23
~ m m a4
_ © ' 5 + Cp2ZrMe Propagation N © — Zr (Me) Cp23
~ m m a Zr(Me)Cp2Scheme 2.5. Reaction mechanism for a two-component catalyst system for the polymerisation of MMA.
H ocker et al. [34] investigated different types o f catalyst system s w ith respect to their reactivity w ith M M A. Polym erisation with 6 yields highly isotactic PM M A, whereas
7 yields syndiotactic PM M A at low reaction tem perature (Figure 2.4). These results are in agreem ent with the findings o f Cam eron et al. [35], who used in situ generated catalysts. Cam eron et al [35] were, however, also able to generate active catalysts sim ilar to 8 {in situ), w hile H ocker et al. [34] reported cations 8-10 to be inactive for polym erisation. H ocker et al. [34] proposed that the inactivity could be related to the nature o f the angle o f the ligands around the zirconium centre. This change in angle can entail a change in the steric dem and o f the com plexes or their electronic properties. The activity o f the site is also influenced by the sterical dim ensions o f the acrylate m onom er (the bulkier the side groups the lower the reactivity).
Sustm ann et al. [36] considered the reaction m echanism proposed by Collins et al.
[27, 32, 33] and Soga et al. [28, 37] in the polym erisation o f acrylic acid, instead o f M M A. The energies for the different reaction coordinates were also m easured to see w hich m echanism has the best energy configuration. It was found that all m echanism s have the ability to polym erise, although some o f the interm ediate species show ed higher activation energies than others.
© ©
B P h ,
9
'
10
V 'O /-
11
Scheme 2.6. Reaction scheme as proposed by Hocker et al. [34], for the copolymerisation of ethylene and MMA with 12.
The model for the formation o f isotactic PMMA by means o f the metallocene catalyst 6 [34] can be related to that o f the metallocene catalyst Me2C(Cp)(Ind)ZrMe2 12. The two catalysts 6 and Me2C(Cp)(Ind)ZrMe2 12 have the same ligands, the only difference being that one methyl group in Me2C(Cp)(Ind)ZrMe2 is replaced by a tetrahydrofurane (THF) in 6.
The polymerisation mechanism o f Me2C(Cp)(Ind)ZrMe2, as proposed by Hocker et
al. [34], will be discussed in further detail in Chapter 7 where the analysis o f the
poly(ethylene-co-MMA) samples will be reported on. Scheme 2.6 gives only a brief outline o f this reaction.
2.4 Chromatographic techniques for polyolefin analysis
The main analytical tools used in this study were based on chromatographic methods. It is therefore vital to understand, firstly, the development and the theory behind these
