Novel Catalysts for Ethylene Polymerisation and Oligomerisation

Novel catalysts for ethylene polymerisation and oligomerisation

Citation for published version (APA): Vadake Kulangara, S. (2012). Novel catalysts for ethylene polymerisation and oligomerisation. Technische Universiteit Eindhoven. https://doi.org/10.6100/IR741099

DOI: 10.6100/IR741099

Document status and date: Published: 01/01/2012

Document Version: Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication

General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement: www.tue.nl/taverne

Take down policy If you believe that this document breaches copyright please contact us at: [email protected] providing details and we will investigate your claim.

Download date: 05. Oct. 2021 Novel catalysts for ethylene polymerisation and oligomerisation

PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Technische Universiteit Eindhoven, op gezag van de rector magnificus, prof.dr.ir. C.J. van Duijn, voor een commissie aangewezen door het College voor Promoties in het openbaar te verdedigen op woensdag 14 november 2012 om 16.00 uur

door

Shaneesh Vadake Kulangara

geboren te Panoor, Kerala, India

Dit proefschrift is goedgekeurd door de promotoren: prof.dr. C.E. Koning en prof.dr. S. Gambarotta

Copromotor: dr. R. Duchateau

Novel catalysts for ethylene polymerisation and oligomerisation by Shaneesh Vadake Kulangara

Technische Universiteit Eindhoven, 2012.

A catalogue record is available from Eindhoven University of Technology Library

ISBN: 978-90-889-1503-1

The research results described in this thesis form part of the research program of the Dutch Polymer institute (DPI, PO Box 902, 5600 AX Eindhoven, The Netherlands), DPI project #638.

Cover design: Rashija U.R & Proefschriftmaken.nl || printyourthesis.com

Printed by Proefschriftmaken.nl || printyourthesis.com

To my Mother, Santha

CONTENTS

Chapter 1. General introduction 1

Chapter 2. Synthesis, X-ray structural analysis and ethylene polymerisation studies of group (IV) metal heterobimetallic aluminium-pyrrolyl catalysts 39

Chapter 3. Synthesis and ethylene oligomerisation studies of pyrrole-based chromium catalysts 73

Chapter 4. Synthesis and ethylene oligomerisation capabilities of new chromium(II) and chromium(III) heteroscorpionate catalysts 105

Chapter 5. Synthesis and ethylene oligomerisation activity of modified bis(diphenylphosphino)methane chromium catalysts 133 131

Chapter 6. Technology assessment 169

Summary 179 Acknowledgement 183 List of publications 187 Curriculum Vitae 189

Chapter 1

General Introduction

Chapter 1

INTRODUCTION

Polyolefins, the generic name for the synthetic polymers based on ethylene, propylene, other non-functionalised α-olefins and reactive internal olefins are the most common commodity plastics. With a current annual capacity exceeding 130 × 106 metric tonnes per year, they constitute over 60 % of the entire production of plastics.1 The large volume production and broad utility of polyolefins, synthesised from a very simple set of inexpensive monomer units, are related to their wide range of properties allowing them to be used in a wide variety of applications, ranging from simple plastic bags to high strength fibres.2 The growth of the span of polymer properties has been fuelled by the continuous development of new catalyst systems as well as suitable process technologies, both from academic and industrial researchers.

1.1 α-Olefin polymerisation

With a total production of over 80 × 106 metric tonnes per year, polyethylene is the largest volume polymer produced worldwide. The three major classes of polyethylene are described by the acronyms HDPE, LDPE, and LLDPE.3 High- density polyethylene (HDPE) is a linear, semi-crystalline ethylene homopolymer

(Tm ≈ 135 °C) generally prepared by heterogeneous, multi-site Ziegler-Natta and chromium-based coordination polymerisation catalysts and is widely used in packaging, container and pipe applications. Linear low-density polyethylene (LLDPE) is a linear polymer with a significant number of randomly distributed short chain branches and is prepared by the copolymerisation of ethylene with short chain α-olefins (1-butene, 1-hexene or 1-octene). Due to their high toughness, flexibility and relative transparency, they are widely used in film applications. Low-density polyethylene (LDPE), prepared by a high pressure, high temperature

Chapter 1 free radical polymerisation process, usually contains a high degree of branches. The significant branching does not allow the chains to pack into the crystal structure very well, which results in low tensile strength and high ductility. These polymers are widely used in film wraps, plastic bags and for making rigid containers. The schematic diagram of three types of polyethylene is depicted in the Figure 1.

Figure 1. Schematic representation of different polyethylenes.3b

The first commercialisation of polyethylene production was established in 1939 by Imperial Chemical Industry (ICI), using a high pressure-high temperature free radical process for the production of LDPE.4 In 1951, Phillips Petroleum Company developed a low temperature, low pressure process for the production of polyethylene using a chromium trioxide-based catalyst supported on silica.5 This type of catalyst still constitutes around 30% of today’s production of HDPE and LLDPE. A significant breakthrough in polyolefin chemistry was the discovery of the Ziegler-Natta olefin polymerisation catalysts; a heterogeneous catalyst system originally based on titanium halides and organoaluminium reagents, which can

Chapter 1 produce HDPE, LLDPE and isotactic polypropylene under mild polymerisation conditions.6 Current generations of Ziegler-Natta catalysts consist of an inorganic support (typically MgCl2) impregnated with a titanium precursor, aluminium alkyls and internal and external donors (generally mono- or diesters).7

Next to polyethylene, polypropylene constitutes the second largest volume polyolefin with a global market of 45 million tonnes in 2008.7 The polymerisation of propylene is similar to that of ethylene, however, propylene being a prochiral monomer the resulting polymer chain may shows different stereochemical orientations (tacticity). There are three predominant microstructures known for polypropylene depending on the orientation of each methyl group relative to the neighbouring methyl group in the polymer chain, namely isotactic polypropylene, syndiotactic polypropylene and atactic polypropylene (Figure 2).

Figure 2. Schematic representation of polypropylenes with different tacticity.8

When the methyl group in the polymer chain has no preference towards any particular orientation with respect to the carbon-carbon backbone (polypropylene formed via random monomer insertions), polypropylene thus formed is called atactic polypropylene. Due to the lack of crystallinity, atactic polypropylene is an

Chapter 1 amorphous rubbery material with very little strength and is generally used as adhesives and roofing tars. The polypropylene in which all the side chain methyl groups have the same regular orientation along the polymer chain is referred as isotactic polypropylene. Due to its highly ordered structure isotactic polypropylene has got high crystallinity, resulting in good mechanical properties such as stiffness and tensile strength. In syndiotactic polypropylene, which can be prepared using

Cs-symmetric metallocenes (vide infra), the methyl groups alternate regularly from side to side in the polymer chain. Syndiotactic polypropylene chains are less stiff than isotactic polypropylene chains, but materials based thereon have better impact strength and clarity. Both the nature of the catalyst used and the growing polymer chain influence the stereochemistry of the resulting polymer. If the insertion of the incoming monomers is controlled by the stereochemistry of the last enchained monomer, it is called “chain end controlled polymerisation”, whereas if it is controlled by the chiral catalyst site, the mechanism is known as “enantiomorphic- site controlled polymerisation”. The correlation between catalyst structure and the polymer microstructure has been extensively reviewed.8 Apart from ethylene and propylene, higher α-olefins such as 1-hexene,9 norbornene,10 4-methyl-1-pentene11 etc. can also be (co-)polymerised to produce different polyolefins with enhanced material properties.

1.1.1 Mechanism of α-olefin polymerisation

The first well-accepted mechanism for olefin polymerisation was proposed by Cossee and Arlman in the mid 1960s.12 According to their proposed mechanism, the first step in the polymerisation is the formation of an olefin π-adduct, by the coordination of the monomer to the vacant site of the active catalyst (a highly electrophilic alkyl species). It is then followed by the migratory insertion of the olefin monomer into the metal-carbon bond, resulting in the chain growth and the

Chapter 1 regeneration of a new vacant site. In the case of propylene and higher α-olefins, the preferred insertion mode is 1,2-insertion (primary insertion). However, in some cases 2,1-insertion of the monomer (secondary insertion) can also occur, which generally results in the formation of a dormant site and reduces the catalytic activity.

Scheme 1. General mechanism for α-olefin polymerisation (P = polymer chain).

In the 1970s, based on an unexpected shift in the 1H NMR spectrum of an - CH group in their metal complex, Green and Rooney proposed a mechanism involving metal-alkylidine hydride intermediates, which react with ethylene to form a metallocyclobutane species.13 The metallocyclobutane undergoes a reductive elimination and regenerates the active species. However, due to the lack of d-electrons in the case of tetravalent group IV metal catalysts, the formal oxidative addition required by the Green-Rooney mechanism is not possible and an alternate mechanism involving α-agostic interactions has been proposed, the so- called modified Green-Rooney or modified Cossee-Arlman mechanism.14

Chapter 1

Scheme 2. Proposed mechanisms for ethylene polymerisation (P = polymer chain).14

1.1.2 Role of cocatalyst

As discussed in the previous section (Section 1.1.1), the presence of transition metal-carbon bond is an important prerequisite for initiating the polymerisation process (via the insertion of olefin into metal-carbon bond). However, in the case of precatalysts lacking transition metal-carbon bond (for instance, in the case of precatalyst containing dichloride), an external reagent known as activator or cocatalyst is usually added to generate the metal-carbon bond in situ and also to generate the electrophilic metal center by alkyl/halide abstraction. In the case of heterogeneous Ziegler-Natta catalyst system, triethylaluminium (TEA) is commonly used as activator. Methylaluminoxane is one of the most commonly used activators for single-site olefin polymerisation catalysts.15 The cocatalyst MAO, which is prepared by the partial hydrolysis of

AlMe3 (TMA), has multiple functions, being the alkylation of the metal centre,

Chapter 1 abstraction of a halide or alkyl ligand to produce a vacant site and also to act as scavenger for the impurities present in the system.15 Although very extensive research has been carried out to elucidate the structural features of MAO, its exact composition and structure is still not well understood. Nevertheless, in spite of its unresolved structural features MAO is represented as one dimensional linear chains (1) or cyclic rings (2), two dimensional structures (3) and three dimensional clusters (4), the latter most likely being the most realistic structure.16

Scheme 3. Proposed structures of MAO.15

Despite of its ability to increase the activity of the catalysts, MAO also has some serious disadvantages. Most of the catalysts require high aluminium to metal ratios to yield considerable polymerisation activities and stable kinetic profiles, 15 which lead to high cocatalyst costs and high ash contents (Al2O3) of the polymer. The unresolved structural features of the MAO also complicate the investigation of the actual active species. For several catalytic systems, the free TMA present in the MAO has a negative effect on the catalytic performance by forming bimetallic

Chapter 1

+ 17 inactive species such as [Cp2Zr(μ-Me)2AlMe2] . The free TMA can also act as a chain transfer agent decreasing the molecular weight of the resulting polymer.18 Fortunately, the deactivating effect of TMA can be eliminated by adding a bulky phenol, such as 2,6-di-tert-butyl-4-methylphenol (BHT), as a TMA scavenger.19 Apart from MAO, Lewis acidic perfluoroaryl boranes and trityl/ammonium perfluoroaryl borates can also be used to generate the active single-site catalytic species.15, 20 The synthesis and use of different boranes and borates as cocatalyst has been reviewed and will not be further discussed here.15, 20

1.1.3 Single-site olefin polymerisation catalysts

A drawback of both the heterogeneous Phillips and Ziegler-Natta catalysts is the presence of multiple catalytic sites, which result in the formation of polymers with broad molecular weight distributions and non-homogeneous comonomer incorporation.3 Albeit that these properties are quite beneficial with respect to processing, the ill-defined nature of these catalytic systems, along with their moderate and non-uniform comonomer incorporation ability, stimulated the discovery of new single-site olefin polymerisation catalysts.

Novel polyolefin materials with precise control over the polymer microstructure were prepared using these new generation single-site catalysts. The following section will shortly summarise different generations of single site olefin polymerisation catalysts, with special emphasis on ethylene polymerisation. Although, most of these catalysts are also capable of, or even especially developed for polymerising higher α-olefins such as propylene, we will limit ourselves in this thesis to ethylene polymerisation.

Chapter 1

1.1.3.1 Metallocenes

In general, a metallocene system consists of a group IV transition metal (Ti, Zr, Hf) sandwiched between 5-bonded carbocyclic ligands (substituted or non- substituted) such as cyclopentadienyl, fluorenyl or indenyl moieties.21 In the late 1950s, the earliest examples of these metallocenes showed poor catalytic performance towards olefin polymerisation once activated with different aluminium alkyls.22 Later, Kaminsky and coworkers accidently discovered that the addition of water to the Cp2ZrMe2/AlMe3 system significantly enhanced the catalytic performance. As already mentioned above (Section 1.1.2), it led to the discovery of a highly efficient activator or cocatalyst known as methylaluminoxane.23 Changing the nature of the substituents of the cyclopentadienyl-based ancillary ligands, the type of transition metal together with the application of different cocatalysts resulted in the discovery of a virtually unlimited number of metallocene catalysts with different steric and electronic characteristics, which in turn produced novel polyolefin materials with better control over the polymer microstructure and comonomer incorporation.22, 24

1.1.3.2 Ansa-metallocenes and constrained geometry catalysts

The well-defined nature of the metallocene systems have provided systematic opportunities for the better understanding of the steps involved in the coordination polymerisation mechanism and gave more insight into the nature of the active catalyst. The era of simple metallocene catalyst systems were followed up by ansa-metallocene catalysts.25 These catalysts were mainly designed to produce polypropylene and other poly--olefins with controlled tacticities. C2- symmetric catalysts typically produce isotactic polymers (Scheme 4).8 On the other hand, Cs-symmetric catalysts that have mirror planes reflecting two enantiotopic coordination sites produce syndiotactic polymers.

Chapter 1

Scheme 4. Typical examples of metallocene (A), ansa-metallocenes (B) and constrained geometry catalysts (C).8

The term constrained geometry catalysts (CGC) was forged by Stevens et al. for titanium complexes obtained by exchanging one cyclopentadienyl ring (Cp) of the ansa-metallocene by an amido moiety.1, 26 The CGC catalysts showed excellent comonomer incorporation ability and were capable of producing ethylene/α-olefin copolymers with a high polymerisation activity.27 The higher comonomer insertion is due to the fact that the reduced steric hindrance allows improved accessibility for bulky α-olefin comonomers.1 Another advantage is the higher thermal stability of CGC alkyl species compared to related metallocenes, which allow the use of high polymerisation temperatures for these catalysts.28 Detailed reviews about the synthesis, modifications and polymerisation behaviour

Chapter 1 of CGC catalysts have been published.1, 29 During the first half of 1990s, interest grew in developing a new generation of non-metallocene catalysts in order to circumvent the growing number of patents in group IV cyclopentadienyl systems and to explore the potentials of other ligands to polymerise ethylene and higher α- olefins. One of the successful strategies was to replace one of the cyclopentadienyl moieties by anionic ligands such as NR2, N=PR3, N=CR2 etc. and these new ‘half- sandwich’ catalysts have attracted considerable attention due to their ability to polymerise ethylene and to copolymerise ethylene and α-olefins with reasonable polymerisation activity.30 Apart from Cp-based post-metallocenes, non Cp-based catalysts such as chelating aryloxide catalysts,29 phenoxy-imine (FI) catalysts31 and tris(pyrazolyl)borates catalysts32 were also reported.

Chapter 1

Scheme 5. Typical examples of post-metallocene catalysts.29-32

1.1.3.3 Ethylene polymerisation by late transition metal catalysts

Even though the majority of olefin polymerisation catalysts are based on early transition metals, the single-site late transition metal catalysts also showed significant olefin polymerisation activity.33 Due to the lower oxophilicity and higher tolerance towards polar functional groups, they are potential candidates to be used in the polymerisation of polar monomers or in unconventional diluents 33 such as in aqueous emulsions and super critical CO2. Another significant characteristic of the late transition metal-catalysed ethylene polymerisation is their ability to in situ introduce short chain branches in the polyethylene chains via a chain walking mechanism, without the addition of any comonomer.34 Generally, α- diimine complexes of Pd(II) and Ni(II), reported by Brookhart et al. showed this

Chapter 1 interesting catalytic behaviour and the degree of branching can be controlled by catalyst structure, polymerisation temperature and ethylene pressure.34

Scheme 6. Schematic representation of chain walking mechanism (P = polymer chain, L = ligand, anions are omitted for clarity).33

Transition metal complexes of iron and cobalt, ligated with neutral bis(imino)pyridine ligands, are also known to polymerise ethylene to form HDPE.35 Unlike the nickel- or palladium-based catalysts, which are known to produce branched polyethylene via chain walking mechanisms, iron-based catalysts produced strictly linear polymers.35c Furthermore, the variation of steric hindrance of the substituents afforded products ranging from HDPE to a Schultz- Flory distribution of α-olefins. Synthesis, structural characterisation, ethylene polymerisation capabilities and mechanistic insights of different iron- and cobalt- based catalysts have been extensively reviewed.35a

Chapter 1

Scheme 7. Generic structures of late transition metal catalysts.

1.2 Ethylene oligomerisation

Oligomerisation of ethylene to higher linear α-olefins (LAOs) continues to be an area of high research interest due to the wide use of these α-olefins for example as comonomers (C4-C8), plasticizers (C6-C10), synthetic lubricants (C8-C18) 36, 42 and detergents (C10-C18). A typical transition metal-catalysed oligomerisation of ethylene usually results in the production of a statistical distribution of oligomers known as a Schulz-Flory or Poisson distribution, ranging from C4 to 37, 38 C26. However, as the separation of these mixtures into LAOs of specific chain length is tedious, and there is a considerable market demand for the comonomer grade LAOs, much effort from both industrial and academic communities has been devoted to search for more efficient catalysts that will produce -olefins of specific lengths.38 The selective oligomerisation of ethylene to 1-hexene and 1-octene is of special industrial significance, as they are widely used as comonomer in the preparation of LLDPE.38 Ethylene trimerisation has been reported for titanium,39 tantalum40 and chromium,41 the last one being the most selective and active one.

Chapter 1

Table 1. Applications of linear α-olefins (LAOs) and their market distribution in 2006.42

Chain length Application Market distribution (%)

C4-C8 Comonomer for LLDPE and HDPE 56

C6-C10 Plasticizer 8

C8-C18 Additives for synthetic lubricants 11

C10-C18 Intermediates for detergents 12 others Amines, mercaptens e.t.c 16

Industrially, LAOs are produced mainly via oligomerisation of ethylene and Fischer Tropsch synthesis followed by special purification techniques.43 Currently, there are five commercial processes for the non-selective ethylene oligomerisation namely, the Ethyl Corporation (Ineos) process, the Gulf process, the Shell Higher alpha olefin process (SHOP), the Idemitsu Petrochemical process and the Sabic- Linde (α-Sablin) process.43 The SHOP process, invented and exploited by Shell, utilises nickel-phosphine catalysts for ethylene oligomerisation whereas the Ineos (stoichiometric Ziegler process) and Gulf processes (catalytic Ziegler process) utilise triethyl aluminium as catalysts for oligomerisation. The α-Sablin and Idemitsu Petrochemical processes are based on zirconium-based catalysts activated with triethyl aluminium. The selective ethylene oligomerisation is so far commercialised by two companies, utilising the Chevron-Phillips (trimerisation of ethylene to 1-hexene) and IFP-Alphabutol processes (dimerisation of ethylene to 1- butene). A different route for the selective production of 1-octene was commercialised by Dow Chemicals in 2008. In their process, they telomerised 1,3- butadiene with methanol in presence of a palladium catalyst, and the resulting product on hydrogenation followed by cracking at high temperature resulted in the selective formation of 1-octene.43 This thesis mainly focuses on the synthesis of

Chapter 1 linear α-olefins via ethylene oligomerisation using homogeneous chromium-based catalysts. Various aspects of the chromium-catalysed ethylene oligomerisation reaction, such as the mechanism (selective and non-selective oligomerisation), the use of selective tri-/tetramerisation catalysts (chromium- and non-chromium-based) and the influence of the metal oxidation state on selectivity are discussed in the following sections.

1.2.1 Mechanism of ethylene oligomerisation

The non-selective ethylene oligomerisation process, which produces a statistical distribution of oligomers, is known to follow the Cossee-Arlman chain growth mechanism.12 The catalytically active species, containing a metal-carbon bond, is generated by the reaction of a precatalyst with suitable activators, usually organo-aluminium compounds such as AlMe3, AlEt3 or MAO. According to this mechanism, the chain growth occurs via ethylene coordination to the transition metal followed by its insertion into a metal alkyl bond. The chain termination occurs via β-hydrogen transfer to the metal to form a metal hydride species, thereby releasing the LAOs. It should be noted that, through all the steps of the Cossee-Arlman mechanism, the metal oxidation state remains unchanged. Therefore the cationisation of the metal centre and the polarity of the M-R bond are the key factors controlling the relative reaction rates and thereby the productivity and selectivity of the catalytic system.

Chapter 1

Scheme 8. Cossee-Arlman mechanism for non-selective ethylene oligomerisation.

The most widely accepted mechanism to explain the selective ethylene oligomerisation was postulated by Briggs et al (Scheme 9).44 According to this mechanism, ethylene oligomerisation proceeds through metallocycle intermediates. The catalytic cycle starts with the initial coordination of two equivalents of ethylene to a ligated Crn species, followed by oxidative coupling forming a metallocyclopentane of oxidation state Crn+2. The transition state for the β- hydrogen elimination from the metallocyclopentane is strained due to its stability. Hence, the ring expansion by ethylene insertion dominates, forming a metallocycloheptane. The seven-membered metallocycle can undergo β-H elimination due to the ring strain, forming a chromium–alkenyl-hydride species, which reductively eliminates 1-hexene and regenerates the Crn oxidation state. Hence, the selectivity of this process is controlled by the relative stability of differently sized metallocycles, especially their tendency to either decompose or grow via ethylene insertion. In agreement with this mechanism, Emrich et al. have reported well-characterised metallocyclopentane and metallocycloheptane

Chapter 1 complexes. The latter decomposes more readily and yields 1-hexene, supporting the viability of these intermediates in the catalytic cycle.45 The isotope labelling studies carried out by Bercaw46 and McGuiness47 also supported the metallocycle mechanism. An elegant deuterium labelling experiment carried out by Gibson and co-workers also demonstrated that the non-selective ethylene oligomerisation can also occur via the same metallocycle mechanism.48

Scheme 9. Metallocycle mechanism.44

1.2.2 Chromium-based selective trimerisation catalysts

The first selective trimerisation system producing 1-hexene with more than 90% selectivity was reported by the researchers of the Chevron-Phillips petroleum company.49 Reagan and coworkers discovered that the combination of chromium (III) alkanoate, such as chromium (III) 2-ethylhexanoate (Cr(III) 2-EH), with 2,5-

Chapter 1 dimethylpyrrole and triethylaluminium (TEA) in an aliphatic hydrocarbon solvent such as cyclohexane produced 1-hexene with high selectivity and activity. It was also found that the presence of halogen-containing compounds, such as diethyl aluminium chloride (DEAC), GeCl4 and SnCl4, during the catalyst preparation leads to marked improvements in both the catalyst activity and selectivity towards 1-hexene formation. Chevron-Phillips successfully commercialised this technology in Qatar and this is up to date the only example of a commercial scale ethylene trimerisation plant.

Scheme 10. Chevron-Phillips trimerisation system.49

Soon after the Phillips trimerisation system, Tosoh Corporation reported a trimerisation system based on maleimidyl ligands.50 By using Cr(III)2-EH, maleimide, TEA and DEAC in a molar ratio of 1:60:430:160 at 40 bar of ethylene pressure, catalyst activities of up to 278 kg/g Cr/h, but with a somewhat lower 1- hexene selectivity of 79.6 %, have been achieved.

Scheme 11. Tosoh trimerisation system.50

Chapter 1

Following this, several selective trimerisation systems have been developed, including the Sasol mixed heteroatomic systems,51 the BP diphosphine system52 and Tosoh’s tris(pyrazolyl) system.53 Selected trimerisation systems along with their 1- hexene selectivity and activity are depicted in Scheme 12.

Scheme 12. Selected examples of recently reported trimerisation catalysts.

Apart from chromium, a few other transition metal-based catalysts, capable of producing 1-hexene with good trimerisation activities, were also reported.39, 40 These are summarised in Scheme 13.

Chapter 1

Scheme 13. Selected examples of non-chromium based trimerisation catalysts.

1.2.3 Selective ethylene tetramerisation

Even though various selective trimerisation catalysts are known, selective tetramerisation catalysts remain rare. The first class of selective tetramerisation catalysts was reported by Sasol in 2004.54 The chromium (III) complexes supported by PNP ligands, [PPh2-NR-PPh2], upon activation with MAO produced up to 70 % 1-octene, with an activity of 272 kg/g Cr/h. Followed by this, SK Energy reported a selective ethylene tetramerisation catalyst system based on substituted

Chapter 1 bis(diphenylphosphino)ethane ligands producing 1-octene with 68 % selectivity.55 Wass and coworkers in 2011 reported that the chromium-tetracarbonyl complex of the ligand [PPh2N(i-Pr)P(Ph)N(i-Pr)PPh2] was also capable of producing 1-octene with reasonable selectivity (57 % C8 selectivity).56 In 2010, Gambarotta and coworkers reported a selective tetramerisation catalyst, which produces 1-octene with a selectivity greater than 99% as the sole oligomeric product along with a significant amount of low molecular weight PE.57 Recently, the same researchers have also found that the chromium (III) complexes of [PPh2NR(CH2)4NRPPh2] are capable of producing 1-octene with a maximum selectivity of 91 %.58 The authors reported that the oligomerisation solvent, cocatalyst and catalyst loading have profound influence on the selectivity and activity of these catalysts. Selected tetramerisation systems along with their 1-octene selectivity and activity are depicted in Scheme 14.

Chapter 1

Scheme 14. Selected examples of tetramerisation catalysts.

A detailed mechanistic investigation by Sasol researchers nicely explained that selective tetramerisation can also occur via a metallocycle mechanism, even though the structural proof is still missing.59 However, if a fourth molecule of ethylene can be inserted readily into a seven-membered ring, it is clearly very difficult to prevent the further expansion of the resulting nine-membered ring and thereby the formation of higher oligomers. Nevertheless, the amount of 1-decene or higher -olefins formed was very limited. Rosenthal and co-workers proposed an alternative concept-mechanism for the formation of 1-octene that might explain the absence of large amounts of higher -olefins.42 According to their hypothesis, a dimetallic system with two low-valent chromium centres may form two

Chapter 1 independent five-membered metallocycles. A cooperative dimetallic reductive elimination from this species can selectively produce 1-octene (see Scheme 15).

Scheme 15. Proposed mechanism for ethylene tetramerisation.42

Despite the development of several trimerisation and tetramerisation systems, there remains considerable debate on the mechanistic details, such as the dynamics of the chromium oxidation states at different stages of the catalytic cycle. Active trimerisation catalysts are typically generated in situ by the addition of a cocatalyst, and the individual activation steps and co-catalysts do vary from system to system. It has been proposed that the cocatalyst reduces the ligated chromium metal at the early stage of activation.60 The key issue here is to understand which oxidation state of the chromium is responsible for the higher selectivity. The direct characterisation of the catalytically active species nevertheless remains a challenge, mainly due to the paramagnetic nature of the chromium complexes as well as the faster catalyst decomposition.61 The possibility of different spin states for

Chapter 1 chromium also made the theoretical studies more complicated.62 Different redox couples such as Cr(I)/Cr(III),63 Cr(II)/Cr(IV)64 and Cr(III)/Cr(V)65 have been proposed to be responsible for selective ethylene oligomerisation and our recent work has demonstrated that the Cr(I)/Cr(III) redox couple is the most likely one to be responsible for the selectivity.66

1.3 Background, objective and outline of the thesis

The development of new single-site homogeneous catalysts for ethylene polymerisation is an area of ongoing research interest due to the versatility of the ligand fine-tuning, which allows the production of (co-)polymers with tailored microstructures and properties. Even though homogeneous single-site catalysts allow the production of polymer microstructures which cannot be obtained with heterogeneous Ziegler-Natta catalysts, they possess a serious disadvantage, viz. the lack of thermal stability, which in turn may result in reduced activity of the catalytic system. In order to tackle the issue of catalyst stability versus activity, our research group has developed completely new classes of thermally stable self- activating ethylene polymerisation/oligomerisation catalysts (vanadium- and chromium-based), supported by unprecedented aluminium-pyrrolyl ligand systems.67, 66a

The aim of this thesis is to develop novel thermally stable olefin polymerisation catalysts, which can undergo a reversible intramolecular alkyl shift from the main group metal (Al, Zn etc) to the transition metal centre, thereby creating a transition metal-carbon bond which is present while the catalyst is active and absent in the dormant state, thus creating a self activating catalyst. We also extended our studies on developing new selective ethylene oligomerisation

Chapter 1 catalysts, especially to get more insight into the issue of metal oxidation state versus selectivity.

Chapter 2 focuses on the synthesis and ethylene polymerisation capabilities of heterobimetallic aluminium-pyrrolyl complexes of group (IV) metals. Several novel heterobimetallic aluminium pyrrolyl complexes of titanium, zirconium and hafnium were synthesised and were fully characterised by means of single crystal X-ray diffraction, elemental analysis and NMR spectroscopy. Ethylene polymerisation studies of these catalysts showed that they were capable of producing ultra high molecular weight polyethylene (UHMWPE) with moderate activity. A possible mechanism for the activity of these catalysts is proposed based on DFT calculations.

In the second part of the thesis (Chapters 3, 4 and 5), we describe the synthesis and ethylene oligomerisation studies of novel chromium-based catalysts. The effects of experimental parameters such as type of solvent, nature of cocatalysts, polymerisation temperature, etc. on the activity and the selectivity were studied in detail and it was found that these parameters have a profound effect on the activity and selectivity of the catalyst system. We investigated three different classes of chromium-based homogeneous ethylene oligomerisation catalysts:

1) Pyrrole-based chromium catalysts (Chapter 3). 2) Chromium(II) and chromium(III) heteroscorpionate catalysts (Chapter 4). 3) Modified bis(diphenylphosphino)methane chromium catalysts (Chapter 5).

Chapter 6 is a technology assessment, highlighting the most important results along with perspectives for the future research in the area of polyolefin

Chapter 1 catalysis as well as presenting a view on the industrial feasibility of the findings reported in this thesis.

Chapter 1

1.4 REFERENCES

1. Braunschweig, H.; Breitling, F. M.. Coord. Chem. Rev. 2006, 250, 2691.

2. Kaminsky, W.; Arndt, M. Adv. Poly. Sci. 1997, 127, 143.

3. (a) McKnight, A. L.; Waymouth, R. M. Chem. Rev. 1998, 98, 2587. (b) Peacock, A. J.; Handbook of polyethylene: Structure, Properties and Applications, Marcel Deckcer: New York, 2000. (c) Ross, J. F.; MacAdams, J. L.; Polymeric Material Encyclopedia, Volume 8, Ed. Salamone, J. C.; CRC Presss, 1996, 5953.

4. Fawcett, E. W.; Gibson, R. O.; Perrin, M. W.; Patton, J. G.; Williams, E. G. (Imperial Chemical Industries Ltd.), GB 471590, 1937.

5. Hogan, J. P.; Banks, R. L. US 2825721 (Phillips Petroleum Company), 1958.

6. (a) Ziegler, K. Angew. Chem. 1952, 64, 323. (b) Ziegler, K.; Holzkamp, E.; Breil, H.; Martin, H. Angew. Chem. 1955, 67, 541. (c) Natta, G. Angew. Chem. 1956, 68, 393. (d) Ziegler, K. Angew. Chem. 1964, 76, 545. (e) Natta, G.; Pino, P.; Corradini, P.; Danusso, F.; Mantica, E.; Mazzanti, G.; Moraglio, G. J. Am. Chem. Soc. 1955, 77, 1708.

7. (a) Coates, G. W. Chem. Rev. 2000, 100, 1223. (b) Boor, J.; Ziegler Natta Catalysis and Polymerisation, academic Press, Inc. 1995, 47. (c) Mukhopadhyay, S.; Kulkarni, S. A.; Bhaduri, S. J. Organomet. Chem. 2005, 690, 1356. (d) Busico, V.; Cipullo, R.; Corradini, P.; Biasio, R. D. Macromol. Chem. Phys. 1995, 196, 491.

Chapter 1

8. Resconi, L.; Cavallo, L.; Fait, A.; Fabrizio, P. Chem. Rev. 2000, 100, 1253.

9. (a) Tshuva, E. Y.; Goldberg, I.; Kol, M.; Weitman, H.; Goldschmidt, Z. Chem. Commun. 2000, 5, 379. (b) Tshuva, E. Y.; Groysman, S.; Goldberg, I.; Kol, M.; Goldschmidt, Z. Organometallics 2002, 21, 662.

10. (a) Tritto, I.; Boggioni, L.; Ferro, D. R. Coord. Chem. Rev. 2006, 250, 212. (b) Gilliom, L.; Grubbs, R. H. J. Mol. Catal. 1988, 46 255.

11. Descour, C.; Duchateau, R.; Mosia, R. M.; Gruter, G. J. M.; Severn, J. R.; Rastogi, S. Polym. Chem. 2011, 2, 2261.

12. (a) P. Cossee. J. Catal., 1964, 3, 80. (b) Arlman, E. J.; Cossee, P. J. Catal., 1964, 3, 99.

13. Green, M. L. H. Pure Appl. Chem., 1978, 50, 27.

14. Grubbs, R. H.; Coates, G. W. Acc. Chem. Res., 1996, 29, 85.

15. Chen, E. Y. Chem. Rev. 2000, 100, 1391.

16. (a) H. Sinn, H.; Macromol. Symp. 1995, 97, 27. (b) Koide, Y.; Bott, S. G.; Barron, A. R. Organometallics 1996, 15, 2213. (c) Reddy, S. S.; Sivaram, S.; Prog. Polym. Sci. 1995, 20, 309.

17. Bochmann, M.; Lancaster, S. J. Angew. Chem. Int. Ed. Engl., 1994, 33, 1634.

18. Rieger, B.; Janiak, C. Angew. Makromol. Chem., 1994, 215, 35.

Chapter 1

19. Busico, V.; Cipullo, R.; Cutillo, F.; Friederichs, N.; Ronca, S.; Wang, B. J. Am. Chem. Soc., 2003, 125, 12402.

20. Piers, W. E.; Chivers, T.; Chem. Soc. Rev. 1997, 26, 345.

21. Wilkinson, G.; Birmingham, I. M. J. Am. Chem. Soc. 1954, 76, 4281.

22. Kaminsky, W.; Laban, A. Appl. Catal., A. 2001, 222, 47.

23. Sinn, H.; Kaminsky, W.; Vollmer, H. J.; Woldt, R. Angew. Chem. 1980, 92, 396.

24. Hlatky, G. G. Coord. Chem. Rev. 2000, 199, 235.

25. Wang, B.; Coord. Chem. Rev. 2006, 250, 242.

26. Stevens, P. N.; Timmers, F. J.; Wilson, D. R.; Schmidt, G. F.; Nickias, P. N.; Rosen, R. K.; Knight, G. W.; Lai, S. Eur. Pat. Appl. EP-416815-A2, 1991.

27. Canich, J. M.; Hlatky, G. G.; Turner, H. W.; PCT Appl. WO 92-00333, 1992.

28. Millward, D. B.; Cole, A. P.; Waymouth, R. M. Organometallics 2000, 19, 1870.

29. Nomura, K.; Liu, J.; Padmanabhan, S.; Kitiyanan. B. J. Mol. Catal. A: Chem. 2007, 267, 1.

30. Nomura, K.; Fujii, K. Macromolecules, 2003, 36, 2633. 31. Makio, H.; Terao, H.; Iwashita, A.; Fujita, T. Chem. Rev. 2011, 111, 2363.

Chapter 1

32. (a) Nakazawa, H.; Ikai, S.; Imaoka, K.; Kai, Y.; Yano, T. J. Mol. Catal. A: Chem. 1998, 132, 33. (b) Gil, M. P.; Santos. J. H. Z. D.; Jr, O. L. C. Macromol. Chem. Phys., 2001, 202, 319. (c) Furlan, L. G.; Gil, M. P.; Jr, O. L. C. Macromol. Rapid Commun., 2000, 21, 1054.

33. Ittel, D. S.; Johnson, L. K.; Chem. Rev. 2000, 100, 1169.

34. (a) Johnson, L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc. 1995, 117, 6414. (b). Small, B. L.; Brookhart, M.; Bennett, A. M. A. J. Am. Chem. Soc. 1998, 120, 4049. (c) Svejda, S. A.; Johnson, L. K.; Brookhart, M. J. Am. Chem. Soc. 1999, 121, 10634.

35. (a) Gibson, V. C.; Spitzmesser, S. K. Chem. Rev 2003, 103, 283. (b) Britovsek, G. J. P.; Bruce, M.; Gibson, V. C.; Kimberley, B. S.; Maddox, P. J.; Mastroianni, S.; McTavish, S. J.; Redshaw, C.; Solan, G. A.; Stromberg, S.; White, A. J. P.; Williams, D. J. J. Am. Chem. Soc. 1999, 121, 8728. (c) Britovsek, G. J. P.; Mastroianni, S.; Solan, G. A.; Baugh, S. P. D.; Redshaw, C.; Gibson, V. C.; White, A. J. P.; Williams, D. J.; Elsegood, M. R. J. Chem. Eur. J. 2000, 6, 2221. (d) Small, B. L.; Brookhart, M. J. Am. Chem. Soc. 1998, 120, 7143.

36. Vogt, D. Applied Homogeneous Catalysis with Organometallic Compounds; Cornils, B., Herrmann, W. A., Eds.; Wiley-VCH: Weinheim, Germany, 1996; Vol 1, P 245.

37. McGuinness, D. S. Chem. Rev. 2011, 111, 2321.

38. Skupinska, J. Chem. Rev. 1991, 91, 613.

Chapter 1

39. (a) W.K. Reagan, J.W. Freeman, B.K. Conroy, T.M. Pettijohn, E.A. Benham. US 5,451,645 (Phillips Petroleum Company), 1995. (b) Pellecchia, C.; Pappalardo, D.; Oliva, L.; Mazzeo, M.; Gruter, G. Macromolecules 2000, 33, 2807. (c) Deckers, P. J. W.; Hessen, B.; Teuben, J. H. Organometallics 2002, 21, 5122. (d) Deckers, P. J. W.; Hessen, B.; Teuben, J. H. Angew. Chem. Int. Ed. 2001, 40, 2516. (e) Deckers, P. J. W.; Hessen, B. WO 02/066404 (Stichting Dutch Polymer Institute), 2002. (f) Huang, J.; Wu, T.; Qian, Y. Chem. Commun. 2003, 2816. (g) Suzuki, Y.; Kinoshita, S.; Shibahara, A.; Ishii, S.; Kawamura, K.; Inoue Y.; Fujita, T. Organometallics 2010, 29, 2394.

40. (a) Andes, C.; Harkins, S. B.; Murtuza, S.; Oyler, K.; Sen, A. J. Am. Chem. Soc. 2001, 123, 7423. (b) Arteaga-Müller, R.; Tsurugi, H.; Saito, T.; Yanagawa, M.; Oda, S.; Mashima, K. J. Am. Chem. Soc. 2009, 131, 5370.

41. Agapie, T. Coord. Chem. Rev. 2010, 255, 861.

42. Peitz, S.; Aluri, B. R.; Peulecke, N.; Müller, B. H.; Wöhl, A.; Müller, W.; Al- Hazmi, M. H.; Mosa, F. M.; Rosenthal, U. Chem. Eur. J. 2010, 16, 7670.

43. (a) Forestiere, A.; Olivier-Bourbigou, H.; Saussine, L. Oil Gas Sci. Technol. Rev. IFP 2009, 64, 649. (b) Bohley, R. C.; Jacobsen, G. B.; Pelt, H. L.; Schaart, B. J.; Schenk, M.; Oeffelen, D. A. G. WO 92/10450 (DOW Benelux), 1992. (c) Leeuwen, P. W. N. M.; Clément, N. D.; Tschan, M. J. L. Coord. Chem. Rev. 2011, 255, 1499.

44. (a) Manyik, R. M.; Walker, W. E.; Wilson, T. P. J. Catal. 1977, 47, 197. (b) Briggs, J. R. Chem. Commun. 1989, 11, 674. (c) McDermott, J. X.; White, J. F.; Whitesides, G. M. J. Am. Chem. Soc. 1973, 95, 4451; (d) McDermott, J. X.; White, J. F.; Whitesides, G. M. J. Am. Chem. Soc. 1976, 98, 6521. (e)

Chapter 1

Rensburg, W. J.; Grove, C.; Steynberg, J. P.; Stark, K. B.; Huyser, J. J.; Steynberg, P. J. Organometallics 2004, 23, 1207. (f) Blom, B.; Klatt, G.; Fletcher, J. C. Q.; Moss, J. R. Inorg. Chim. Acta 2007, 360, 2890. (g) Bhaduri, S.; Mukhopadhyay, S.; Kulkarni, S. A. J. Organomet. Chem. 2009, 694, 1297.

45. (a) Emrich, R.; Heinemann, O.; Jolly, P. W.; Kruger, C.; Verhovnik, G. P. G. Organometallics 1997, 16, 1511. (b) Jolly, P. W. Acc. Chem. Res. 1996, 29, 544.

46. (a) Agapie, T.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2007, 129, 14281. (b) Agapie, T.; Schofer, S. J.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2004, 126, 1304.

47. McGuinness, D. S.; Suttil, J. A.; Gardiner, M. G.; Davies, N. W. Organometallics 2008, 27, 4238.

48. Tomov, A. K.; Chirinos, J. J.; Long, R. J.; Gibson, V. C.; Elsegood, M. R. J. J. Am. Chem. Soc. 2006, 128, 7704.

49. (a) Reagan, W. K. EP 0417477 (Phillips Petroleum Company), 1991. (b) Knudsen, R. D.; Freeman, J. W.; Lashier, M. E. US 5563312 (Phillips Petroleum Company), 1996. (c) Freeman, J. W.; Buster, J. L.; Knudsen, R. D. US 5856257 (Phillips Petroleum Company), 1999. (d) Reagan, W. K.; Freeman, J. W.; Conroy, B. K.; Pettijohn, T. M.; Benham, E. A. EP 0608447 (Phillips Petroleum Company), 1994.

50. (a) Aoyama, T.; Mimura, H.; Yamamoto, T.; Oguri, M.; Koie, Y. JP 09176299 (Tosoh Corporation), 1997. (b) Aoyama, T.; Mimura, H.; Yamamoto, T.; Oguri, M.; Koie, Y. JP 09268133 (Tosoh Corporation), 1997. (c) Aoyama, T.;

Chapter 1

Mimura, H.; Yamamoto, T.; Oguri, M.; Koie, Y. JP 09268134 (Tosoh Corporation), 1997.

51. (a) McGuinness, D. S.; Wasserscheid, P.; Keim, W.; Morgan, D. H.; Dixon, J. T.; Bollmann, A.; Maumela, H.; Hess, F. M.; Englert, U. J. Am. Chem. Soc. 2003, 125, 5272. (b) Dixon, J. T.; Wasserscheid, P.; McGuinness, D. S.; Hess, F. M.; Maumela, H.; Morgan, D. H.; Bollmann, A. (Sasol Technology (Pty) Ltd) WO 03053890, 2001.

52. (a) Wass, D. F. (BP Chemicals Ltd) WO 02/04119, 2002. (b) Carter, A.; Cohen, S. A.; Cooley, N. A.; Murphy, A.; Scutt, J.; Wass, D. F. Chem. Commun. 2002, 858.

53. Zhang, J.; Braunstein, P.; Hor, A. T. S. Organometallics 2008, 27, 4277.

54. Bollmann, A.; Blann, K.; Dixon, J. T.; Hess, F. M.; Killian, E.; Maumela, H.; McGuinness, D. S.; Morgan, D. H.; Neveling, A.; Otto, S.; Overett, M.; Slawin, A. M. Z.; Wasserscheid, P.; Kuhlmann S. J. Am. Chem. Soc. 2004, 126, 14712.

55. Han, T. K.; Ok, M. A.; Chae, S. S.; Kang, S. O. (SK Energy Corporation), WO 2008/ 088178, 2008.

56. Dulai, A.; McMullin, C. L.; Tenza, K.; Wass, D. F. Organometallics 2011, 30, 935.

57. Licciulli, S.; Thapa, I.; Albahily, K.; Korobkov, I.; Gambarotta, S.; Duchateau, R.; Chevalier. R.; Schuhen, K. Angew Chem. Int. Ed. Engl. 2010, 49, 9225.

Chapter 1

58. Shaikh, E.; Albahily, K.; Sutcliffe, M.; Fomitcheva. V.; Gambarotta, S.; Duchateau, R.; Korobkov, I. Angew. Chem., Int. Ed. 2012, 51, 1.

59. Overett, M. J.; Blann, K.; Bollmann, A.; Dixon, J. T.; Haasbroek, D.; Killian, E.; Maumela, H.; McGuinness, D. S., Morgan, D. H. J. Am. Chem. Soc, 2005, 127, 10723.

60. (a) Temple, C.; Jabri, A.; Crewdson, P.; Gambarotta, S.; Korobkov, I.; Duchateau, R. Angew. Chem. 2006, 118, 7208. (b) Jabri, A.; Temple, C.; Crewdson, P.; Gambarotta, S.; Korobkov, I.; Duchateau, R. J. Am. Chem. Soc. 2006, 128, 9238; (c) Jabri, A.; Crewdson, P.; Gambarotta, S.; Korobkov, I.; Duchateau, R. Organometallics 2006, 25, 715. (d) Zhang, J.; Li, A.; Hor, T. S. A. Organometallics 2009, 28, 2935.

61. Skobelev, I. Y.; Panchenko, N. P.; Lyakin, O. Y.; Bryliakov, K. P.; Zakharov, V. A.; Talsi, E. P. Organometallics 2010, 29, 2943.

62. Koohn, R. D. Angew. Chem., Int. Ed. 2008, 47, 245.

63. (a) Kohn. R. D.; Haufe, M.; Mihan, S.; Lilge, D. Chem. Commun. 2000 1927. (b) Fang, Y.; Liu, Y.; Ke, Y.; Guo, C.; Zhu, N.; Mi, X.; Ma, Z.; Hu, Y. Appl. Catal. A 2002, 235, 33. (c) Yang, Y.; Liu, Z.; Zhong, L.; Qiu, P.; Dong, Q.; Cheng, R.; Vanderbilt, J.; Liu, P. Organometallics 2011, 30, 5297.

64. (a) Morgan, D. H.; Schwikkard, S. L.; Dixon, J. T.; Nair, J. J.; Hunter, R. Adv. Synth. Catal. 2003, 345, 939. (b) Van Rensburg, W. J.; Grove, C.; Steynberg, J. P.; Stark, K. B.; Huyser, J. J.; Steynberg, P. J. Organometallics 2004, 23, 1207.

Chapter 1

65. Meijboom, N.; Schaverien, C. J.; Orpen, A. G. Organometallics 1990, 9, 774.

66. (a) Vidyaratna, I.; Nikiforov, G. B.; Gorelsky, S. I.; Gambarotta, S.; Duchateau, R.; Korobkov, I. Angew. Chem., Int. Ed. 2009, 48, 6552. (b) Albahily, K.; Fomitcheva. V.; Murugesu, M.; Gambarotta, S.; Korobkov, I.; Gorelsky, S. I. J. Am. Chem. Soc. 2011, 133, 6388.

67. Jabri, A.; Korobbkov, I.; Gambarotta, S.; Duchateau, R. Angew. Chem. Int. Ed. 2007, 46, 6119.

38 Chapter 2

Synthesis, X-ray structural analysis and ethylene polymerisation studies of group (IV) metal heterobimetallic aluminium-pyrrolyl catalysts

The content of this chapter has been published: Vadake Kulangara, S.; Jabri, A.; Yang, Y.; Korobkov, I.; Gambarotta, S.; Duchateau, R. Organometallics 2012, 31, 6085.

Chapter 2

Abstract The synthesis, structural characterisation and ethylene polymerisation performance of hetero-bimetallic aluminium-pyrrolyl complexes of group (IV) metals are described. The combination of MCl4 (M = Ti, Zr, Hf), aluminium alkyls and pyrroles leads, depending on stoichiometry, to mono- and bis(aluminium-pyrrolyl) complexes that are remiscent of the corresponding mono- and bis- cyclopentadienyl systems. The 5 5 bis(aluminium-pyrrolyl) complexes ( -2,5-Me2C4H2NAlClMe2)2TiMe2 (1), zirconium ( - 5 2,5-Me2C4H2NAlClMe2)2ZrClMe (2), ( -2,5-Me2C4H2NAlCl2Et)2ZrCl2 (3) and hafnium 5 ( -2,5-Me2C4H2NAlClMe2)2HfClMe (4) were found to be inactive towards ethylene polymerisation. By contrast, the electron deficient mono(aluminium-pyrrolyl) piano stool 5 5 complexes ( -2,5-Me2C4H2NAlCl2Me)TiCl2Me (5), ( -2,3-Me2C8H4NAlCl2Me)TiCl2Me 5 (6) and ( -3,4,5,6-C12H12NAlCl2Me)TiClMe2 (7), obtained by treatment of TiCl4 with equimolar amounts of trimethyl aluminium and the corresponding pyrrole ligands, were + - found to be moderately active for ethylene polymerisation with MAO or [Ph3C] [B(C6F5)4]

, in both cases producing UHMWPE. The NMR scale reaction of 5 with B(C6F5)3 showed the formation of a solvent-separated ion pair, formed by the abstraction by B(C6F5)3 of a 1 13 methyl group from Al-CH3 rather than from Ti-CH3. H and C NMR analysis of 6 and 7 revealed that several stable structural isomers exist in solution, with a slow inter- conversion on the NMR time scale.The dimeric zirconium complexes [(5,1-2,5-

5 1 Me2C4H2NAlClMe2)ZrMeCl(-Cl)]2 (8) and [( , -2,5-Me2C4H2NAlCl2Et)ZrCl2(-Cl)]2

(9), prepared by the reaction of 2 equivalents of ZrCl4, 1 equivalent of 2,5-dimethyl pyrrole and 1 equivalent of aluminium alkyl, showed structures containing a bridging chlorine between aluminium and zirconium, thereby giving structural evidence for the lack of catalytic behaviour of these complexes. A possible explanation for the moderate or absence of catalytic activity of the aluminium-pyrrolyl complexes was proposed based on DFT calculations.

Chapter 2

2.1 INTRODUCTION

Cyclopentadienyl and related ligands have played an important role in the development of modern organotransition metal chemistry and they have been widely used to stabilise homogeneous olefin polymerisation catalysts.1 The discovery of single-site group IV metallocene and half-metallocene catalysts is considered to be one of the most significant breakthroughs within the modern polyolefin science and chemical industry. Single-site catalysts show superior properties over conventional Ziegler-Natta catalysts in terms of homogeneous product distribution as well as good copolymerisation activity.1-2 Indeed, much effort is currently being exerted into developing new single-site catalysts that will provide even higher catalyst efficiency and greater control over polymer microstructures. As a result of this extensive research, a number of classes of highly active catalysts based on both early and late transition metals have been discovered and are described in various reviews.3-4

In comparison with the cyclopentadienyl ligand, the isoelectronic N- heterocyclic pyrrolyl ligand has received much less attention as an ancillary ligand for olefin polymerisation catalysts.5 The main reason is the preferential coordination of high valent early transition metals to hard -donors rather than to soft -donors. Consequently, pyrrolyl ligands are generally -bonded to the electrophilic transition metals.6 To avoid this, we reasoned that once another strong

Lewis acid, such as AlX3 (X = halogen, alkyl), would lock the nitrogen atom through -bonding, a -bonding mode with the transition metal could occur.

Based on this hypothesis, we recently reported a new class of olefin polymerisation and oligomerisation catalysts based on aluminium-substituted pyrrolyl ligands.7,8 For example mono- and bis(aluminium-pyrrolyl) chromium and vanadium species proved to be self-activating ethylene oligomerisation and

Chapter 2 polymerisation catalysts.8 Due to the paramagnetic nature of chromium and vanadium, little is known about their dynamic coordination chemistry, which limits the rational development of this class of ethylene oligomerisation and polymerisation catalysts. Herein we have extended this study to the synthesis of group IV metal-based aluminium-pyrrolyl catalysts to probe their ethylene polymerisation activity and mechanistic behaviour.

2.2 RESULTS AND DISCUSSION

Following the synthetic method furnishing aluminium-pyrrolyl vanadium and chromium species,[7,8] the bis(aluminium-pyrrolyl) complex of titanium, (5-

2,5-Me2C4H2NAlClMe2)2TiMe2 (1) was synthesised in good yield by the sequential addition of AlMe3 to a solution of 2,5-dimethylpyrrole (Ti : Al : Pyr = 1 : 4 : 2) in 1 toluene at room temperature, immediately followed by TiCl4 (Scheme 1). The H NMR of 1 shows one singlet for the two methyl groups bonded to titanium (0.61 ppm) and one singlet for the methyl groups bonded to aluminium (-0.61 ppm), suggesting that the structure of 1 is stable and does not undergo rearrangement in solution. Independent experiments in which 1 was exposed to heat in the dark and to light at room temperature resulted in the formation of a black product mixture, which strongly suggests that 1 is sensitive to light and heat, most probably as the result of homolytic Ti-CH3 bond breakage. The complex is stable for prolonged periods of time at low temperature in the dark. Yellow crystals of complex 1, suitable for an X-ray single crystal structure determination, were obtained by layering the toluene solution with an equal amount of petroleum-ether and storing it at -35 C.

The single crystal structure of 1 (Figure 1) revealed that the molecule has a

C2v symmetric, 16 electron sandwich structure analogous to the traditional 9 sandwich complex Cp2TiMe2. The coordination geometry around the transition

Chapter 2 metal is distorted tetrahedral, with the tetrahedron defined by two methyl groups and the centroid of two pyrrolyl ligands. The Cn-Ti(1)-Cn (Cn = centroid of pyrrolyl rings) bond angle [131.25] was found to be in between that of related

5 species Cp( -2,5-Me2C4H2N)TiCl2 [130.72(5)] and Cp2TiMe2 [Cp-Ti-Cp = 2n 134.54]. Also the Ti(1)-Cnpyrrole distance [2.136 Å] is significantly shorter than 5 2n the Ti-Cnpyrrole distance in Cp( -2,5-Me2C4H2N)TiCl2 [2.349(9) Å]. The C(l7)- Ti(1)-C(l8) bond angle [93.70(2)] is somewhat larger than the corresponding bond

9 angle in Cp2TiMe2 [C-Ti-C = 91.29(7)]. The titanium-carbon bond [Ti(1)-C(17) = 2.133(6) Å, Ti(1)-C(18) = 2.102(5) Å] is shorter than the Ti-C bond distance of 9 the dimethyl titanocene [Ti-C = 2.181(2) Å] and Cp*(-2,5-Me2C4H2N)TiMe2 [Ti- C = 2.184(2) Å].2n The geometry of the aluminium atom in the molecule is distorted tetrahedral and defined by two methyl groups and one chlorine [Cl(2)- Al(2)-C(16) = 110.40(2); C(l5)-Al(2)-C(16) = 121.00(3)].

5 Scheme 1. Synthesis of bispyrrolyl complex ( -2,5-Me2C4H2NAlClMe2)2TiMe2 (1).

Chapter 2

5 Figure 1. ORTEP drawings for ( -2,5-Me2C4H2NAlClMe2)2TiMe2 (1). Ellipsoids are drawn at 50% probability level and H atoms are omitted for clarity. Selected bond distances (Å) and angles (): Ti(1)-C(l7), 2.133(6); Ti(1)-C(l8), 2.102(5); Ti(1)-Cnt(1), 2.136; Al(1)-Cl(1), 2.172(2); Al(2)-Cl(2), 2.178(2); Al(1)-C(8), 1.967(7); Al(2)-C(16), 2.005(7); C(17)-Ti(1)-C(l8), 93.70(2); Cnt(1)-Ti(1)-Cnt(2), 131.25; C(l5)-Al(2)-C(16), 121.00(3); Cl(2)-Al(2)-C(16), 110.40(2).

Ethylene polymerisation experiments of 1 with either MAO or + - [CPh3] [B(C6F5)4] under various conditions (temperatures and ethylene pressures) invariably resulted in just traces of polymer. To understand the reason of this lack 1 of activity, the reaction of 1 with B(C6F5)3 was followed by H NMR. Interestingly, the borane abstracts an aluminium methyl rather than a titanium methyl group. Although rapid interchange of methyls and/or chlorine groups between the aluminium and titanium seems feasible, on the 1H NMR time scale this is not

Chapter 2 observed, indicating that a rather non-fluxional structure is formed. The inactivity of the catalyst is therefore assumed to be due to the formation of a stable chelate species after cationisation of 1. For example, a chlorine could well be bridging between the two aluminium or one aluminium and the titanium centers. Such a chelate is coordinatively saturated, leaving no vacant site for ethylene to coordinate (Scheme 2). This is in agreement with the fact that a solvent-separated ion pair has been formed, as can be concluded based on the chemical shifts in the 19F NMR.10 Unfortunately, attempts to crystallise the cationic species invariably failed.

Scheme 2. Proposed deactivation pathway of the cationic derivative of 1.

Structure optimisation by means of DFT calculations indeed supports the formation of a chlorine-bridged species (Figure 2). An input structure where one of the aluminate groups has been dealkylated, leaving a trigonal planar aluminium pyrrolyl-Al(Me)Cl group, is unstable and transforms into complex A where the titanium methyl is interacting with the aluminium center. The similar chlorine- bridged species B, obtained after methyl transfer from titanium to aluminium, is clearly more stable than the methyl-bridged A.

Chapter 2

Figure 2. Free energies and optimized structures of (5-2,5-

5 5 Me2C4H2NAlClMe2)2TiMe2 (1), [( -2,5-Me2C4H2NAlClMe2)( -2,5- + 5 + Me2C4H2NAlClMe)TiMe2] (A), [( -2,5-Me2C4H2NAlClMe2)2TiMe] (B) and 5 + [{( -2,5-Me2C4H2NAlMe2)2(-Cl)}TiMeCl] (C). Hydrogen atoms in the structures were omitted for clarity.

Structural proof of a bridging chlorine interaction between aluminium and the transition metal as in B was found for the zirconium complexes [(5, 1-2,5-

5 1 Me2C4H2NAlClMe2)ZrMeCl(-Cl)]2 (9) and [( ,  -2,5-

Me2C4H2NAlCl2Et)ZrCl2(-Cl)]2 (10). The structure (C) containing a Me2Al(-

Cl)AlMe2 bridged bis(pyrrolyl) ligand system seems the most stable structure. A related structure consisting of a thulium sandwhich complex containing a similar 11a Me2Al(-CH2)AlMe2 bridge between two pyrrolyl ligands has been reported. The chorine-bridge structure C also reminiscents the N→B bridged metallocenes reported by Starzewski and coworkers.11b

Chapter 2

In order to investigate whether the formation of a stable chelate upon cationisation is a general feature, the analogues bis(aluminium-pyrrolyl) complexes of zirconium (2, 3) and hafnium (4) were prepared in moderate yield, following the similar synthetic procedure as used for 1.

The X-ray structures of 2, 3 and 4 are reminiscent of their metallocene analogues, containing the transition metal in a distorted tetrahedral geometrical environment (Figure 3). The central metal atom is bonded to one methyl [Zr(1)- C(9) = 2.177(15) Å] and one chlorine [Zr(1)-Cl(2) = 2.425(4) Å] in the case of 2, two chlorine atoms [Zr(1)-Cl(1) = 2.3754(4) Å, Zr(1)-Cl(1a) = 2.3754(4) Å] in the case of 3 and one methyl [Hf(1)-C(7) = 2.226(15) Å] and one chlorine [Hf(1)-Cl(1) = 2.370(5) Å] in the case of hafnium complex 4. The nitrogen to aluminium bond distance in the case of complexes 2 and 4 is comparable [N(1)-Al(1) = 2.01 Å] and is slightly shorter than the corresponding bond distance in complex 3 [N(1)-Al(1) = 1.988(12) Å].

Chapter 2

2 3 4

5 5 Figure 3. ORTEP drawings of ( -2,5-Me2C4H2NAlClMe2)2ZrClMe (2) ( -2,5- 5 Me2C4H2NAlCl2Et)2ZrCl2 (3) and ( -2,5-Me2C4H2NAlClMe2)2HfClMe (4). Ellipsoids are drawn at 50% probability level and H atoms are omitted for clarity. Selected bond distances (Å) and angles (): 2 Zr(1)-Cl(2) , 2.425(4); Zr(1)-C(9), 2.177(15); Al(1)-N(1), 2.0190(13); Al(1)-Cl(1), 2.1797(6); Al(1)-C(7), 1.958(2); Cl(2)-Zr(1)-C(9),101.5(5); Cnt(1)-Zr(1)-Cnt(1a), 124.96. 3 Zr(1)-Cl(1), 2.3754(4); Al(1)-N(1), 1.9885(12); Al(1)-Cl(2), 2.1560(5); Al(1)-Cl(3), 2.1520(6); Al(1)- C(7), 1.9503(17); Cnt(1)-Zr(1)-Cnt(2), 124.05; Cnt(1)-Zr(1)-Cl(1), 106.19; Cl(1)- Zr(1)-Cl(1a), 103.24(2). 4: (): Hf(1)-Cl(1), 2.370(5); Hf(1)-C(7), 2.226(15); Al(1)-N(1), 2.0163(18); Al(1)-Cl(2), 2.1806(9); Al(1)-C(9), 1.961(3); Al(1)-C(8), 1.962(3); Cl(1)-Hf(1)-C(7), 99.6(4); Cnt(1)-Hf(1)-Cnt(1a), 125.59; N(1)-Al(1)- Cl(2), 103.93(6); N(1)-Al(1)-C(8), 106.78(11); N(1)-Al(1)-C(9), 111.22(11).

The catalytic performance of the complexes 2, 3 and 4 towards ethylene polymerisation were evenly disappointing as in the case of complex 1. This is most likely due to the formation of the same kind of chelate complexes as obtained upon cationisation of 1 (which has been supported by DFT calculations; see Supporting

Chapter 2

Information), confirming that the bis(aluminium-pyrrolyl) complexes of group IV metals are inactive towards ethylene polymerisation, irrespective of the nature of the transition metal.

In order to overcome the lack of catalytic activity of the 16 electron bispyrrolyl complexes (1 – 4), the synthesis of the more electron-deficient mono(aluminium-pyrrolyl) species was persued. The treatment of a toluene solution of 2,5-dimethylpyrrole with AlMe3, immediately followed by addition of

TiCl4 (Ti: Al: Pyr = 1:1:1) at room temperature, afforded a dark brown solution from which upon cooling to -30 C yellow crystals of (5-2,5- 1 Me2C4H2NAlCl2Me)TiCl2Me (5) were obtained. The H NMR spectrum of 5 showed the resonance of the methyl group attached to titanium at 1.97 ppm and the resonance of the methyl group attached to aluminium at -0.03 ppm.

5 Figure 4. ORTEP drawings for ( -(2,5-Me2C4H2NAlCl2Me)TiCl2Me (5). Ellipsoids are drawn at 50% probability level and H atoms are omitted for clarity. Selected bond distances (Å) and angles ():Ti(1)-Cl(1), 2.215(1); Ti(1)-Cl(2), 2.216(1); Ti(1)-C(7), 2.056(5); Ti(1)-N(1), 2.365(2); Ti(1)-C(2), 2.379(2); Ti(1)- C(3), 2.375(2); Al(1)-Cl(3), 2.146(1); Al(1)-Cl(4), 2.149(1); Cl(1)-Ti(1)-Cl(2), 107.98(5); Cl(1)-Ti(1)-C(7), 101.80(1); Cl(4)-Al(1)-C(8), 112.70(1); Cl(3)-Al(1)- Cl(4), 108.38(5).

Chapter 2

The X-ray structure (Figure 4) showed that the molecule has a distorted tetrahedral geometry around titanium, where the vertices of the tetrahedron are defined by a -bonded aluminium-pyrrolyl ligand, two chlorines and one methyl group around the titanium. The Ti(1)-Cl(1) bond distances [2.215(1) Å, 2.216(1) Å] in the case of complex 5 were found to be shorter than the Ti-Cl bond distances 13 of related compounds such as for example Cp(ArO)TiCl2 [2.259(8) Å] and 2n Cp*(2,4-Me2C4H2N)TiCl2 [2.245(2) Å]. The Ti(1)-Cntpyrrole distance [2.062 Å] in 5 is significantly shorter than in 1 [2.136 Å]. It is difficult to judge whether this is mainly due to the higher electrophilicity or to the lower steric hindrance of the system. The Ti(1)-N(1) bond length of 2.365(2) Å is only slightly shorter than the Ti(1)-C(2) [2.379(2) Å] and Ti(1)-C(3) [2.375(2) Å] bond distances, confirming the 5-coordination of the aluminium-pyrrolyl ligand to titanium. The aluminium atom in the molecule has a slightly distorted tetrahedral structure [Cl(3)-Al(1)- Cl(4) = 108.38(5); Cl(4)-Al(1)-C(8) = 112.70(1)].

5 The analogues 2,3-dimethylindolyl ( -2,3-Me2C8H4AlCl2Me)TiCl2Me (6) 5 and tetrahydrocarbazolyl ( -3,4,5,6-C12H12NAlCl2Me)TiClMe2 (7) complexes were synthesized by similar synthetic procedures in moderate yield. However, the 1H NMR and 13C NMR analysis of these complexes showed the existence of various isomers in solution as a result of Me/Cl exchange reactions. This contrasts with the 1H NMR spectrum of 5, which showed only one isomer even at elevated temperatures. The NMR spectra of complex 6 and 7 were difficult to unravel as the result of different isomers being present in the solution. This Me/Cl exchange is also confirmed with EXSY experiments, which revelead cross peaks for certain sets of isomers. However, the complete interpretations of the spectra were not successful due to the structural complexity of these fluxional molecules.

Chapter 2

5 Figure 5. ORTEP drawings of ( -2,3-Me2C8H4NAlCl2Me)TiCl2Me (6, left) and 5 ( -3,4,5,6-C12H12NAlCl2Me)TiClMe2] (7, right). Ellipsoids are drawn at 50% probability level and H atoms are omitted for clarity. Selected bond distances (Å) and angles () for 6: Ti(1)-C(12), 2.067(4); Ti(1)-Cl(3), 2.206(3); Ti(1)-Cl(4), 2.229(4); Ti(1)-N(1), 2.270(3); Ti(1)-C(6), 2.503(3); Ti(1)-C(7), 2.471(3); Al(1)- C(l1), 2.141(1); Al(1)-Cl(1), 2.149(1); Al(1)-Cl(2), 2.154(1); Al(1)-N(1), 1.982(4); Ti(1)-Cnt(1), 2.062; Cl(3)-Ti(1)-Cl(4), 105.88(5); Cl(3)-Ti(1)-C(12), 102.20(1); Cl(1)-Al(1)-Cl2, 109.78(6); Cl(1)-Al(1)-C(11), 115.00(1). For 7: Ti(1)-Cl(1), 2.210(2); Ti(1)-C(15), 2.138(4); Ti(1)-C(13), 2.139(4); Ti(1)-N(1), 2.256(3); Ti(1)- C(6), 2.514(4); Ti(1)-C(7), 2.499(5); Al(1)-Cl(2), 2.128(2); Al(1)-Cl(3), 2.093(2); Al(1)-C(14), 1.991(4); Al(1)-N(1), 1.992(4); Ti(1)-Cnt1, 2.088; C(15)-Ti(1)-C(13), 95.20(1); Cl(1)-Ti(1)-C(15), 101.70(1).

The structures of 6 and 7 were determined by X-ray analysis (Figure 5). Both molecules have a distorted tetrahedral piano stool configuration. In complex 6 the titanium is bonded to two chlorides and one methyl group, whereas in 7 the titanium is bonded to one chloride and two methyl groups. The Ti(1)-Cntpyrrole bond distance in the case of complex 6 [Ti(1)-Cnt = 2.062 Å] is identical to that of complex 5, but is slightly shorter than the Ti(1)-Cntpyrrole bond distance of complex 7 [2.088 Å]. This might be due to the slight tilt in the coplanarity of the more bulky

Chapter 2 tetrahydrocarbozyl ring compared to the more planar 2,5-dimethylpyrrolyl and 2,3- dimethylindolyl rings. The Ti(1)-Cl bond distances [2.229(4) Å; 2.206(3) Å for 6 and 7, respectively] and Cl-Ti(1)-Cl bond angle of 105.88(5)° for 6 were comparable to the corresponding bond distances and angle in the structurally 5 14 related ( -indolyl)TiCl2Me [2.229(2) Å, 2.259(2) Å, 104.80(6)°]. The comparison of the Ti(1)-N(1) bond distance [2.270(3) Å] with the Ti(1)-C(6) [2.503(3) Å] and Ti(1)-C(7) [2.471(3) Å] bonds in 6 suggests that titanium has a stronger interaction with the N atom resulting in an 4,-bonding mode of the aluminium-indolyl ligand. Accordingly, the C(6)-C(7) bond distance [1.433(4) Å] is slightly longer than the C(8)-C(9) bond length [1.398(1) Å], indicating that the delocalization of -electrons is less efficient in 6 compared to 5. The shorter Ti(1)- N(1) bond distance also reflects in the slightly longer N(1)-Al(1) bond distance [1.982(4) Å]. A similar trend was observed in the case of 7, where the Ti(1)-N(1) bond distance [2.256(3) Å] was slightly shorter than the Ti(1)-C(6) and Ti(1)-C(7) [(2.514(4) Å); (2.499(5) Å] bond distances.

The corresponding mono(aluminium-pyrrolyl) zirconium complexes 8 and 9 were synthesized in moderate yield by the sequential addition of ZrCl4 to a freshly prepared solution of 2,5-dimethylpyrrole and corresponding alkyl aluminium

(AlMe3 for 8 and AlEt2Cl for 9) in toluene at room temperature (Zr:Al:Pyr = 1:1:2). Colourless crystals of complex 8 and 9 were obtained by layering the toluene solution with an equal amount of petroleum-ether and storing it at room temperature for one week.

Chapter 2

8 9

5 1 Figure 6. ORTEP drawings of [( , -2,5-Me2C4H2NAlClMe2)ZrMeCl(-Cl)]2 (8, 5 1 top) and [( , -2,5-Me2C4H2NAlCl2Et)ZrCl2(-Cl)]2 (9, bottom). Ellipsoids are drawn at 50% probability level and H atoms are omitted for clarity. Selected bond distances (Å) and angles () for 8: Zr(1)-Zr(1a), 4.196; Zr(1)- Cl(1), 2.4060(11); Zr(1)-Cl(2), 2.6750(11); Zr(1)-Cl(3), 2.8547(12); Zr(1)-C(9), 2.254(18); N(1)- Al(1), 1.978(4); Cl(1)-Zr(1)-Cl(3), 84.33(4); Cl(1)-Zr(1)-Cl(2), 150.90(4); Cl(1)- Zr(1)-Cl(2a), 83.04(17); Cl(1)-Zr(1)-C(9), 98.9(4). For 9: Zr(1)-Zr(1a), 4.196; Zr(1)-Cl(1), 2.7336(9); Zr(1)-Cl(2), 2.5782(8); Zr(1)-Cl(3), 2.3832(8); Zr(1)-Cl(4), 2.3866(8); N(1)-Al(1), 1.950(3); Cl(3)-Zr(1)-Cl(4), 95.86(3); Cl(3)-Zr(1)-Cl(2), 88.15(3); Cl(3)-Zr(1)-Cl(2a), 84.03(3); Cl(3)-Zr(1)-Cl(1), 162.289(3); Cl(3)-Zr(1)- Cl(4), 95.86(3); Cl(2)-Zr(1)-Cl(2a), 73.39(2).

The molecular structures of 8 and 9, as determined by single crystal X-ray diffraction, are very similar (Figure 6) as these complexes are dimeric with zirconium in a distorted octahedral geometrical environment. Both complexes contain two types of bridging chlorine atoms: between two zirconium centers and between aluminium and zirconium. The Zr-Cl bond distances in 8 [Zr(1)-Cl(2) = 2.6750(11) Å; Zr(1)-Cl(3) = 2.8547(12) Å] were both found to be slightly longer

Chapter 2 than the corresponding bond distances in 9 [Zr(1)-Cl(2) = 2.5782(8) Å; Zr(1)-Cl(1) = 2.7336(9) Å]. The coordination sphere of 8 is completed by a non-bridging chlorine [Zr(1)- Cl(1) = 2.4060(11)] and a methyl [Zr(1)-C(9) = 2.254(18) Å], whilst 9 binds two terminal chlorines, [Zr(1)-Cl(3) = 2.3832(8) Å, Zr(1)-Cl(4) = 2.3866(8) Å].

Chapter 2

Table 1. Ethylene polymerisation results for 5 – 9.

a Run Catalyst Cocatalyst Yield Activity Mw PDI Tm # (equiv) (g) (kg/mol) (°C)

1 5 MAO (700) 2.5 167 2300 2.5 135.5

2b 5 MAO (700) 6.1 407 260 13.1* 131.5

3 5 TBF20 3.1 207 1460 9.8* 137.1 (2)/TIBA (50)c

4 6 MAO (700) 4.4 293 6500 4.2 135.3

5 6 TBF20 5.3 353 2400 12.9* 137.4 (2)/TIBA (50)c

6 7 MAO (700) 6.5 433 5700 4.4 135.1

7 7 TBF20 2.1 140 490 5.1 136.7 (2)/TIBA (50)c

8 8 MAO (700) 0 - - - -

9 8 TBF20 0 - - - - (2)/TIBA (50)c

10 9 MAO (700) 0 - - - -

11 9 TBF20 0 - - - - (2)/TIBA (50)c

Polymerisation conditions: Solvent = 100 mL of toluene, ethylene pressure = 20 bar, cat = 30 µmol, reaction time = 30 min, reaction temperature = 30 °C. a) Activity = kg PEmol-1h- 1 b) ) c) + - . Temperature = 60 °C. * bimodal distribution. TBF20 = [Ph3C] [B(C6F5)4] , TIBA = Al(i-Bu)3.

Chapter 2

The potential of complexes 5 – 9 for catalytic ethylene polymerisation was investigated under different polymerisation conditions and the results are summarized in Table 1. Upon activation with MAO, 5 produces a single-site 6 - catalyst affording ultra high molecular weight polyethylene (Mw = 2.3  10 g·mol 1, PDI = 2.5). The catalyst is only moderately active and relatively high pressure is required to observe significant catalytic activity (vide infra). Interestingly, when 5 + - was activated with [Ph3C] [B(C6F5)4] and 50 equivalents of TIBA as scavenger at 30 C a clear bimodal distribution of two distinct narrow molecular weight polymer products was obtained, indicating that two distinct active species had been formed. A similar bimodal distribution was observed when the ethylene polymerisation of MAO-activated 5 was carried out at 60 °C instead of 30 °C. The polymer obtained at 60 C showed a dramatic decrease in molecular weight. It is assumed that the high molecular weight fraction is produced by the same species that is also active at low temperature and that the low molecular weight fraction is formed by a newly formed, thus far unidentified catalyst. The aluminium-indolyl and - tetrahydrocarbozolyl analogues 6 and 7 displayed, upon activation with MAO, a somewhat higher catalytic activity compared to 5 and also produced UHMWPE. The higher PDI obtained for 6 and 7 might be due to the existence of different isomers as a result of Me/Cl scrambling (vide supra), each of them giving a slightly different molecular weight. The melting temperatures of the polymers range from 135C to 138C, characteristic for linear polyethylene. Complex 6 shows the same + - bimodal behaviour when activated with [Ph3C] [B(C6F5)4] as found for 5. For 7 only a broadening of the molecular weight distribution was observed. However, + - catalysts 8 and 9 upon activating with either MAO or [CPh3] [B(C6F5)4] under various conditions (temperatures and ethylene pressures) invariably resulted in just traces of polymer. Possibly, the dimeric structure and the internal coordination of the chlorides is too strong to form an electron-deficient active species.

Chapter 2

In order to get mechanistic insight into the polymerisation mechanism of these heterobimetallic olefin polymerisation catalysts, the reaction of 5 with 15 B(C6F5)3 was investigated. Since the precatalysts 5 – 7 contain both titanium- and aluminium-bonded methyl groups, and since B(C6F5)3 is known to be able to abstract both titanium- and aluminium-bonded alkyls,15, 16 the experiment was primarily intended to determine which of the methyl groups would be abstracted.

An equimolar amount of 5 and B(C6F5)3 was dissolved in benzene-d6 at room temperature and the resulting yellow solution of the cationic species was characterized by 1H and 19F NMR spectroscopy. The 1H NMR spectrum clearly showed that the borane selectively abstracts an aluminium-methyl group rather than a titanium-bonded methyl group, forming the putative species [(2,5- + - - Me2C4H2NAlCl2)TiCl2Me] [[MeB(C6F5)3] . The formation of [MeB(C6F5)3] was confirmed by the presence of a characteristic broad B-Me peak at 1.16 ppm in the 1H NMR spectrum and three resonances, -131.8 ppm (o-F), -163.7 ppm (p-F) and - 166.5 ppm (m-F), in the 19F NMR spectrum. The difference in chemical shifts 19 between the meta and para fluorine resonances in the F NMR [(m-F - p-F)] was 2.8 ppm, indicative for a solvent-separated ion pair.10 From the NMR spectrum it is clear that the reaction is an equilibrium shifted towards the product by the addition 1 19 of a second equivalent of B(C6F5)3. The H and F NMR spectra also confirmed the formation of a side product in which a C6F5 group from the borane has been 1 transferred to the titanium center, resulting in the formation of MeB(C6F5)2 ( H NMR: 1.3 ppm; 19F NMR: -129.9 ppm (o-F), -147.1 ppm (p-F) and -161.4 ppm (m- F)).17 The 19F spectrum also showed a resonance at -111.2 ppm (o-F), which is 17 characteristic of a Ti-C6F5 group, possibly of [(2,5-Me2C4H2NAlCl2)TiCl2C6F5]. Similar exchange reactions have been reported by for monocyclopentadienyl titanium complexes.17a Since the ratio of the integrals of the major [(2,5- + - Me2C4H2NAlCl2)TiCl2Me] [[MeB(C6F5)3] and minor products remains the same over a period of 6 hours, it seems that once the cationic species is formed it does not undergo any C6F5 transfer from boron to titanium.

Chapter 2

Since the NMR experiments demonstrated that the Ti-CH3 was retained in the cationic species, the ability of the putative cationic species [(2,5- + - Me2C4H2NAlCl2)TiCl2Me] [[MeB(C6F5)3] to polymerise ethylene was tested.

Treating 5 with 10 equivalents of B(C6F5)3 under ethylene pressure (20 bar) at room temperature indeed resulted in rapid formation of polyethylene. Since it is not 5 + - likely that a [ ,-(2,5-Me2C4H2NAlCl2)TiCl2Me] [MeB(C6F5)3] with a triagonal planar pyrrolyl-aluminium dichloride and a coordinatively saturated titanium center would be an active ethylene polymerisation catalyst, we performed DFT calculations to optimise the putative structure of the cationic species and its ethylene adduct.

Chapter 2

Figure 7. Free energies and optimized structures of (5-2,5-

5 + Me2C4H2NAlCl2Me)TiCl2Me (5), [( -2,5-Me2C4H2NAlCl2)TiCl2Me] (A’) and 5 + [( -2,5-Me2C4H2NAlCl2)TiCl2Me(C2H4)] (B’). Hydrogen atoms in the structures were omitted for clarity.

The computational results proved that a constrained-type of geometry18 with chlorine bridging titanium and aluminium is the most stable configuration for the cationic species (Scheme 3). This should still not be an active ethylene polymerisation catalyst as it is coordinatively saturated. Hence, either release of the bridged chloride or ring slippage from 5- to -bonding between aluminium and titanium of the pyrrole moiety, as found for chromium aluminium- pyrrolylcomplexes,19 could create a system sufficiently coordinatively unsaturated to coordinate and polymerise ethylene (Scheme 3). Overcoming the intramolecular

Chapter 2 chlorine coordination requires 12.2 kcal·mol-1 (Figure 7), which explains the necessity of applying high ethylene pressure. In fact, it is likely that the competiton for coordination of ethylene with the intramolecular ligands is rather difficult. The DFT calculation for the ring slippage from 5- to -bonding of the aluminium- pyrrolyl ligand did not converge to a stable structure.

Scheme 3. Proposed activation of the cationic derivative of 5.

2.3 CONCLUDING REMARKS

The one-pot reactions of group IV metal halides with different alkyl aluminiums (AlMe3 and AlEt2Cl) and 2,5-dimethypyrrole resulted in various novel group IV metal aluminium-pyrrolyl species. Depending on the ratio of 2,5- dimethypyrrole, metal halides and alkyl aluminiums, several bis(alumino-pyrrolyl) or mono(aluminium-pyrrolyl) group IV complexes were isolated and characterised. The bis(aluminium-pyrrolyl) group IV metal species did not yield active ethylene + - polymerisation catalysts upon activation with MAO or [CPh3] [B(C6F5)4]

Chapter 2 irrespective of the nature of the group IV metal applied. On the other hand, the titanium mono(aluminium-pyrrolyl) complexes 5 – 7 did afford active ethylene polymerisation catalysts upon activation with either MAO or + - TIBA/[Ph3C] [B(C6F5)4] , albeit only at higher ethylene pressure. Mechanistic studies showed that the reaction of 1 and 5 with B(C6F5)3 results in abstraction of the aluminium-bonded methyl rather than the titanium-bonded one. DFT calculations confirmed the formation of coordinatively stable chloride-bridged cations of 1 and 5 corroborating the lack of, or only poor catalytic behaviour of this class of complexes. The deactivating effect of a bridging chlorine as proposed by DFT calculations was supported by the isolation of dimeric zirconium complexes 8 and 9, containing bridging chlorides between aluminium and zirconium. The bridging and the internal coordination of the chlorides in 8 and 9 is likely too strong to form an electron deficient active species.

2.4 EXPERIMENTAL SECTION

General Procedures. All air and/or water sensitive reactions were performed under a dry nitrogen atmosphere in oven-dried flasks using standard Schlenk techniques. Anhydrous solvents were obtained by means of a multiple column purification system. Reagent grade pyrrole, 2,5-dimethylpyrrole, 2,3-dimethylindole, 3,4,5,6- tetrahydrocarbazole, titanium tetrachloride and trimethylaluminium were purchased from Sigma-Aldrich and were used as received. All 1H and 13C NMR spectra were recorded on Varian Mercury 400 and 300 MHz spectrometers.

Benzene-d6 was distilled from Na/K alloy and stored under nitrogen atmosphere. Elemental analyses were carried out with a Perkin-Elmer 2400 CHN analyser. Data for X-ray structure determination were obtained with a Bruker diffractometer equipped with a 1K Smart CCD area detector. Differential scanning calorimetry (DSC) was carried out with a Q100 differential scanning calorimeter (TA Instruments). The samples (4-6 mg) were heated to 160 °C and subsequently

Chapter 2 cooled to 20 °C at a rate of 10 °C/min. A second heating cycle was used for the data analysis. Molecular weight and molecular weight-distributions of the polyethylenes were determined by means of gel permeation chromatography on a PL-GPC210 equipped with refractive index and viscosity detectors and a 3  PLgel 10 µm MIXED-B column set at 160 °C with 1,2,4-trichlorobenzene as solvent. Irganox 1010 has been used as stabiliser. The molecular weights were referenced to polyethylene standards.

General Polymerisation Procedure. All polymerisations were performed in a 250 mL steel Büchi autoclave. The reactor was dried in an oven at 120 °C for two hours prior to each run and then evacuated for half an hour and rinsed with argon three times. After that, the reactor was charged with toluene and the desired amount of cocatalyst. After stirring for 10 minutes, the solution was saturated with ethylene. The reactor was temporarily depressurised to allow injection of the catalyst solution into the reactor under argon flow, after which the reactor was immediately repressurised to the desired pressure. The temperature of the reactor was controlled by a thermostate bath. After 30 minutes, the reaction mixture was depressurised and a mixture of ethanol and diluted hydrochloric acid was injected to quench the reaction. The polymer was separated by filtration and dried at 60 °C for 18 hours under reduced pressure before the molecular weight was determined.

5 Synthesis of ( -2,5-Me2C4H2NAlClMe2)2TiMe2 (1). AlMe3 (288 mg, 4 mmol) and TiCl4 (1.0 mL, 1 M in toluene, 1 mmol) were subsequently added at room temperature to a freshly prepared solution of 2,5-dimethylpyrrole (190 mg, 2 mmol) in toluene (10 mL). The mixure was stirred for an additional five minutes resulting in a dark yellow solution, which was then layered with petroleum-ether (10 mL) and stored in the dark at -35 °C for one week to yield yellow crystals of 1, suitable for a single crystal X-ray diffraction analysis. The mother liquor was carefully decanted, and the crystals were dried in vacuo. Yield: (340 mg, 0.76

Chapter 2

1 mmol, 76.0 %). H NMR (benzene-d6, 400 MHz, 293 K): δ = 6.04 (s, 4H; Pyr CH), 13 1 2.53 (s, 12H; Pyr CH3), 0.61 (s, 6H; TiCH3), -0.61 (s, 12H; AlCH3). C{ H} NMR

(benzene-d6, 100 MHz, 293 K): δ = 148.8 (s, PyrC2,5) 116.6 (s, PyrC3,4), 70.8 (s,

TiCH3), 18.3 (s, PyrCH3), -4.0 (s, AlCH3). Anal. Calcd. (found) for

C18H34Al2Cl2N2Ti: C 47.89 (47.38), H 7.53 (7.31), N 6.20 (6.14).

5 Synthesis of ( -2,5-Me2C4H2NAlClMe2)2ZrClMe (2). A slurry of ZrCl4 (236 mg, 1.0 mmol) in toluene was added to a freshly prepared solution of 2,5- dimethylpyrrole (190 mg, 2.0 mmol) and AlMe3 (435 mg, 6.0 mmol) in toluene (5 mL). The resulting yellow solution was then stirred for 10 minutes. After centrifugation, the supernatant was separated, layered with hexane and stored at room temperature for two days to yield pale yellow crystals of 2, suitable for single crystal X-ray diffraction analysis. The mother liquor was carefully decanted and the crystals were dried in vacuo. Yield: (310 mg, 0.60 mmol, 60.0 %). 1H NMR

(CD2Cl2, 300 MHz, 293 K): δ = 6.42 (br, 2H; Pyr CH), 6.38( br, 2H; Pyr CH), 2.69 13 1 (s br, 12H; Pyr CH3), 0.85 (br, 3H; ZrCH3), -0.52 (br, 12H; AlCH3). C{ H}

NMR: 148.0 (s, PyrC2,5), 147.0 (s, PyrC2,5), 116.62 (s, PyrC3,4), 114.17(s, PyrC3,4),

76.52 (s, ZrCH3), 18.11(s, PyrCH3), 17.68(s, PyrCH3), -4.71 (br, AlCH3). Anal.

Calcd.(found) for C17H31Al2Cl3N2Zr: C 39.61 (39.52), H 6.01 (6.12), N 5.40 (5.30).

5 Synthesis of ( -2,5-Me2C4H2NAlCl2Et)2ZrCl2 (3). A slurry of ZrCl4 (236 mg, 1.0 mmol) in toluene was added to a freshly prepared solution of 2,5- dimethylpyrrole (190 mg, 2.0 mmol) and AlEt2Cl (6.0 mL, 1M solution in heptane, 6.0 mmol) in toluene (5 mL). The resulting pale yellow solution was stirred for 30 minute. After centrifugation, the supernatant liquid was separated, layered with hexane and stored at room temperature for one week to yield pale yellow crystals of 3, suitable for single crystal X-ray diffraction analysis. The mother liquor was carefully decanted and the crystals were dried in vacuo. Yield: (394 mg, 0.65 1 mmol, 65.2 %). H NMR (CD2Cl2, 300 MHz, 293 K): δ = 6.78 (br, 4H; Pyr CH),

Chapter 2

13 1 2.92 (s br, 12H; Pyr CH3), 1.07 (t, 6H; AlCH2CH3), 0.32 (q, 4H; AlCH2). C{ H}

NMR: 148.2 (s, PyrC2,5), 114.30 (s, PyrC3,4), 18.4 (s, PyrCH3), 3.71 (br, AlCH2),

2.1 (br, AlCH2CH3) Anal. Calcd.(found) for C16H26Al2Cl6N2Zr: C 31.70 (31.95), H 4.30 (6.41), N 4.63 (4.48).

5 Synthesis of ( -2,5-Me2C4H2NAlClMe2)2HfClMe (4). A slurry of HfCl4 (320 mg, 1.0 mmol) in toluene was added to a freshly prepared solution of 2,5- dimethylpyrrole (190 mg, 2.0 mmol) and AlMe3 (435 mg, 6.0 mmol) in toluene (4 mL). The resulting brown solution solution was stirred for one hour and centrifuged. The supernatant liquid was separated, layered with hexane and stored in room temperature for one week, to yield pale yellow crystals of 4, suitable for single crystal X-ray diffraction analysis. The mother liquor was carefully decanted and the crystals were dried in vacuo. Yield: (420 mg, 0.70 mmol, 70.0 %). 1H

NMR (CD2Cl2, 300 MHz, 293 K): δ = 6.44 (br, 2H; Pyr CH), 6.39 (br, 2H; Pyr

CH), 2.76 (s, 12H; Pyr CH3), 0.62 (br, 3H; HfCH3), -0.49 (br s, 12H; AlCH3). 13 1 C{ H} NMR: 148.3 (s, PyrC2,5), 147.76 (s, PyrC2,5), 115.02 (s, PyrC3,4), 113.41

(s, PyrC3,4), 49.92 (s, HfCH3), 18.06 (s, PyrCH3), 18.31 (s, PyrCH3), -4.64 (br s,

AlCH3). Anal. Calcd.(found) for C17H31Al2Cl3N2Hf: C 34.10 (34.08), H 5.18 (5.32), N 4.68 (4.62).

5 Synthesis of ( -2,5-Me2C4H2NAlCl2Me)TiCl2Me (5). AlMe3 (648 mg, 9.0 mmol) and TiCl4 (1.0 mL, 9.0 mmol) were subsequently added at room temperature to a freshly prepared solution of 2,5-dimethylpyrrole (1.0 mL, 9.8 mmol) in toluene (20 mL). The resulting yellow solution was stirred for five minutes and carefully layered with petroleum-ether and stored in the dark at -35 C for one week to yield 5 as yellow crystals, suitable for single crystal X-ray 1 diffraction analysis. Yield: (2.5 g, 7.3 mmol, 73.8 %) H NMR (benzene-d6, 400

MHz, 293 K): δ = -0.028 (s, 3H; AlCH3), 1.97 (s, 3H; TiCH3), 2.20 (s, 6H; 13 1 PyrCH3), 5.30 (s, 2H; PyrCH). C{ H} NMR (benzene-d6, 100 MHz, 293 K): δ = - 64

Chapter 2

0.367 (br, AlCH3), 18.58 (s, PyrCH3), 99.3 (s, TiCH3), 121.7 (PyrCH), 149.3

(PyrCCH3). Anal. Calcd. (found) C8H14AlCl4NTi: C 28.16 (28.01), H 4.10 (4.15), N 4.10 (4.13).

5 Synthesis of ( -2,3-Me2C8H4NAlCl2Me)TiCl2Me (6). To a solution of 2,3- dimethylindole (145 mg, 1 mmol) and AlMe3 (72 mg, 1 mmol) in toluene (10 mL)

TiCl4 (1 mL, 1 M in toluene, 1 mmol) was added at room temperature yielding a yellow solution, which after five minutes stirring was layered with petroleum-ether and stored overnight in the dark at -35 C to yield yellow crystals of 6. Yield: (374 1 mg, 0.79 mmol, 79.0 %). H NMR (benzene-d6, 400 MHz, 293 K): equilibrium of 5 ( -2,3-Me2C8H4NAlCl3-yMey)TiClxMe3-x, a (x, y = 1) : b (x, y = 2) : c (x, y = 3) = 64 : 24 : 12 via 1H integration  = 8.79 (d, 1H, IndCHa) , 8.68 (d, 1H, IndCHc),

8.50 ( d, 1H, IndCHb), 2.32 (s, IndCH3, a), 2.25 (s, IndCH3, b), 2.23 (s, IndCH3, c),

1.71 (s, TiCH3, a), 1.65 (s, TiCH3, b), 1.58 (s, TiCH3, c), 1.48 (s, IndCH3, a), 1.43

(s, IndCH3, b), 1.35 (s, IndCH3, c), 0.32 (s, AlCl2CH3, a), 0.26 (s, AlCl(CH3)2, b), 13 1 0.18 (s, Al(CH3)3, c). C{ H} NMR (benzene-d6, 100 MHz, 293 K): δ = 147.5 (s, Ind), 136.3 (s, Ind), 135.0 (s, Ind), 132.3 (s, Ind), 131.8 (s, Ind), 123.6 (s, Ind),

104.3 (s, TiCH3, a), 101.7 (s, TiCH3, b), 92.7 (s, TiCH3, c), 53.4 (s, IndCH3), 32.1

(s, IndCH3), 27.6 (s, IndCH3), 14.5 (s, IndCH3), -3.5 (br, AlCH3). Anal.

Calcd.(found) for C12H16AlCl4NTi: C 36.85 (36.38), H 4.13 (4.09), N 3.58 (3.2).

5 Synthesis of ( -3,4,5,6-C12H12NAlCl2Me)TiClMe2 (7). Neat TiCl4 (1 mL, 1 M in toluene, 1 mmol) was added at room temperature to a solution of 3,4,5,6- tetrahydrocarbazole (160 mg, 1 mmol) and AlMe3 (72 mg, 1 mmol) in toluene (10 mL). After stirring for five minutes, the resulting yellow solution was layered with petroleum-ether and stored in the dark at -35 °C to yield pale-yellow crystals of 7, suitable for X-ray diffraction. Yield: (337mg, 0.81 mmol, 81 %). 1H NMR

(benzene-d6, 400 MHz, 293 K): equilibrium of three isomers a : b : c = 60 : 25 : 15

Chapter 2

1 via H integration at: 8.91, 8.83, 8.71, 8.65 8.51 (d, 3JHH = 6 Hz, IndCH a, b, c, 3 1H), 7.07 (t, 3JHH = 11 Hz, IndCH, 1H), 6.88 (t, JHH = 6 Hz, IndCH, 1H), 6.80 (d, 3 JHH = 11 Hz, IndCH, 1H), 2.85 (m, CH2 c, 2H), 2.45 (m, CH2 b, 2H), 2.09 (m, CH2 a, 2H,), 1.65, 1.56, 1.49 (s, TiCH3, b, a, c, 3H), 1.31 (m, CH2, 2H), 1.18 (m, CH2, 13 1 2H), 0.324, 0.254, 0.230, 0.127 (s, AlCH3 a, b, b, c, 3H). C{ H} NMR (benzene- d6, 100 MHz, 293 K): δ = 148.7 (s, IndCCH2) 144.8 (s, IndCCH), 138.6 (s, IndCCH), 132.5 (s, IndCH), 131.3 (s, IndCH), 129.7 (s, IndCH), 129.6 (s, IndCH),

126.6 (s, IndC), 126.3 (s, IndCH), 124.2 (s, IndCH), 102.7 (s, TiCH3 a), 91.2 (s,

TiCH3 b), 87.9 (s, TiCH3 c), 51.6 (IndCH2), 32.1 (IndCH2), 18.4 (IndCH2CH2),

15.7 (IndCH2CH2), 13.7 (IndCH2CH2), -3.2 (br s, AlCH3). Anal. Calcd. (found) for

C14H21AlCl4NTi: C 40.02 (39.89), H 5.04 (5.13), N 3.34 (3.28).

5 1 Synthesis of [( , К -2,5-Me2C4H2NAlClMe2)ZrMeCl(-Cl)]2 (8). A slurry of

ZrCl4 (236.0 mg, 1.0 mmol) in toluene was added to a freshly prepared solution of

2,5-dimethylpyrrole (95 mg, 1.0 mmol) and AlMe3 (145 mg, 2.0 mmol) in toluene (5 mL). The resulting yellow solution was stirred for 10 minutes and centrifuged. The supernatant liquid was separated, layered with petroleum-ether and stored at room temperature for one week to yield colourless crystals of 8, suitable for single crystal X-ray diffraction analysis. The mother liquor was carefully decanted and the crystals were dried in vacuo. Yield: (400 mg, 0.55 mmol, 55.0 %). 1H NMR

(CD2Cl2, 300 MHz, 293 K): δ = 6.32 (br, 4H; Pyr CH), 2.49 (br, 12H; Pyr CH3), 13 1 0.65 (br, 6H; ZrCH3), -0.32 (br, 12H; AlCH3). C{ H} NMR: 149.0 (s, PyrC2,5),

118.6 (s, PyrC3,4), 76.32 (s, ZrCH3), 17.12 (s, PyrCH3), -2.71 (br s, AlCH3). Anal.

Calcd. (found) for C18H34Al2Cl6N2Zr2: C 29.78 (30.01), H 4.61 (4.52), N 3.85 (3.65).

5 1 Synthesis of [( , К -2,5-Me2C4H2NAlCl2Et)ZrCl2(-Cl)]2 (9). A slurry of ZrCl4 (236 mg, 1.0 mmol) in toluene was added to a freshly prepared solution of 2,5- dimethylpyrrole (95 mg, 1.0 mmol) in toluene (5 mL) containing AlEt2Cl (2.0 mL,

Chapter 2

1M solution in heptane, 2.0 mmol). The resulting yellow solution was stirred for 30 minutes. After centrifugation, the supernatant liquid was separated, layered with pet-ether and stored at room temperature for one week to yield pale yellow crystals of 9, suitable for single crystal X-ray diffraction analysis. The mother liquor was carefully decanted and the crystals were dried in vacuo. Yield: (392 mg, 0.47 1 mmol, 46.8 %). H NMR (CD2Cl2, 300 MHz, 293 K): δ = 6.30 (br, 4H; Pyr CH), 13 1 2.42 (br, 12H; Pyr CH3), 0.82 (t, 6H; AlCH2CH3), 0.32 (q, 4H; AlCH2). C{ H}

NMR: 148.4 (s, PyrC2,5), 114.60 (s, PyrC3,4), 18.1(s, PyrCH3), 2.71 (br s, AlCH2),

2.0 (br,s, AlCH2CH3). Anal. Calcd. (found) for C16H26Al2Cl10N2Zr2: C 22.93 (22.79), H 3.10 (3.18), N 3.14 (3.01).

Cationisation Experiments – Formation of [(µ5-(2,5- + - Me2C4H2N(AlCl2))TiCl2Me] [MeB(C6F5)3] . A benzene-d6 solution of 5 (20 mg,

58 µmol) was added to a benzene-d6 solution of B(C6F5)3 (30.8 mg, 58 µmol) to give a yellow solution consisting of a mixture of the starting materials and the cationic complex. The cationic complex was found to be reasonably stable and soluble in benzene-d6 at room temperature. Only after several hours at room temperature visible oiling out of the cationic product occurred. The reaction was monitored at regular time intervals by 1H NMR and 19F NMR. 1H NMR (benzene- d6, 100 MHz, 293 K): δ = 1.16 (br, 3H, BMe), 1.99 (s, 3H, TiCH3), 2.29 (s, 3H, 19 PyrCH3), 5.14 (s, 2H, PyrCH). F NMR (benzene-d6, 100 MHz, 293 K): δ = -131.8 (o-F), -163.8 (p-F), -166.5 (m-F).

Computational method. All the DFT calculations were performed using a Gaussian 09 package of programs with the unrestricted B3LYP hybrid functional, using the double-ζ basis set, LANL2DZ for Ti and 6-31G(d,p) for other elements. Geometry optimizations were carried out without symmetry constraints, and analytical frequency calculations were carried out on the resulting geometries to verify the nature of stationary points (no imaginary frequencies). All energies

Chapter 2 reported refer to Gibbs free energy corrections to the total electronic energies at 298.15 K.

Chapter 2

2.5 REFERENCES

1. (a) Andresen, A.; Cordes, H. G.; Herwig, H.; Kaminsky, W.; Merk, A.; Mottweiler, R.; Pein, J.; Sinn, H.; Vollmer, H. J. Angew. Chem. Int. Ed. Engl. 1976, 15, 630. (b) Sinn, H.; Kaminsky, W.; Vollmer, H. J.; Woldt, R. Angew. Chem. Int. Ed. 1980, 19, 390. (c) Brintzinger, H. H.; Fischer, D.; Mülhaupt, R.; Rieger, B.; Waymouth, R. M. Angew. Chem. Int. Ed. 1995, 34, 1143. (d) Kaminsky, W. Macromol. Chem. Phys. 1996, 197, 3907. (e) Schmid, M, A.; Alt, H, G.; Milius, W. J. Organomet. Chem. 1996, 45, 514. (f) Alt, H. G.; Zenk, R. J. Organomet. Chem. 1996, 51, 512.

2. (a) Jekel, C. T.; Teuben, J. H. J. Organomet. Chem. 1985, 286, 309. (b) Shah, S. A. A.; Dorn, H.; Voigt, A.; Roesky, H. W.; Parisini, E.; Schmidt, H. G.; Noltemeyer, M. Organometallics 1996, 15, 3176. (c) Doherty, S.; Errington, R. J.; Jarvis, A. P.; Collins, S.; Clegg, W.; Elsegood, M. R. J. Organometallics 1998, 17, 3408. (d) Chien, J. C. W.; Yu, Z. T.; Marques, M. M.; Flores, J. C.; Rausch, M. D. J. Polym. Sci. Polym. Chem. 1998, 36, 319. (e) D. W.; Stewart, J. C.; Guerin, F.; Spence, R. E; Xu, W.; Harrison, D. G. Organometallics 1999, 18, 1116. (f) Sinnema, P. J.; Spaniol, T. P.; Okuda, J. J. Organomet. Chem. 2000, 598, 179. (g) Deckers, P. J. W.; Hessen, B.; Teuben, J. H. Angew. Chem. Int. Ed. 2001, 40, 2516. (h) Alt, H. G.; Fottinger, K.; Milius, W. J. Organomet. Chem. 1999, 572, 21. (i) Kretschmer, W. P.; Dijkhuis, C.; Meetsma, A.; Hessen, B.; Teuben, J. H. Chem. Commun. 2002, 608. (j) Murtuza, S.; Casagrande, O. L.; Jordan, R. F. Organometallics 2002, 21, 1882. (k) Nomura, K.; Fujii, K. Organometallics 2002, 21, 3042. (l) Mahanthappa, M. K.; Cole, A. P.; Waymouth, R. M. Organometallics 2004, 23, 836. (m) Tamm, M.; Randoll, S.; Bannenberg, T.; Herdtweck, E. Chem. Commun. 2004, 876.

Chapter 2

3. (a) Spaleck, W.; Kueber, F.; Winter, A.; Rohrmann, J.; Bachmann, B.; Antberg, M.; Dolle, V.; Paulus, E. P. Organometallics 1994, 13, 954. (b) Alt, H. G.; Köppl, A. Chem. Rev. 2000, 100, 1205. (c) Groppo, E.; Lamberti, C.; Bordiga, S.; Spoto, G.; Zecchina, A. Chem. Rev. 2005, 105, 115. (d) Sakuma, A.; Weiser, M. S.; Fujita, T. Polym. J. 2007, 39, 193. (e) Matsugi, T.; Fujita, T. Chem. Rev. 2008, 37, 1264. (f) Kirillov, E.; Roisnel, R.; Razavi, A.; Carpentier, J.-F. Organometallics 2009, 28, 5036. (g) Tam, K. H.; Chan, M. C. W.; Kaneyoshi, H.; Makio, H.; Zhu, N. Organometallics 2009, 28, 5877. (h) Yuan, S.; Bai, S.; Liu, D.; Sun, W. Organometallics 2010, 29, 2132.

4. (a) Johnson, L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc. 1995, 117, 6414. (b) Gates, D. P.; Svejda, S. A.; Onate, E.; Killian, C. M.; Johnson, L. K.; White, P. S.; Brookhart, M. Macromolecules 2000, 33, 2320. (c) Ittel, S. D.; Johnson, L. K.; Brookhart, M. Chem. Rev. 2000, 100, 1169. (d) Boffa, L. S.; Novak, B. M. Chem. Rev. 2000, 100, 1479. (e) Mecking, S.; Held, A.; Bauers, F. M. Angew. Chem. Int. Ed. 2002, 41, 54. (f) Leatherman, M. D.; Svejda, S. A.; Johnson, L. K.; Brookhart, M. J. Am. Chem. Soc. 2003, 125, 3068. (g) Luo, S.; Jordan, R. F. J. Am. Chem. Soc. 2006, 128, 12072. (h) Kochi, T.; Noda, S.; Yoshimura, K.; Nozaki, K. J. Am. Chem. Soc. 2007, 129, 8948. (i) Skupov, K. M.; Marella, P. R.; Simard, M.; Yap, G. P. A.; Allen, N.; Conner, D.; Goodall, B. L.; Claverie, J. P. Macromol. Rapid Commun. 2007, 28, 2033. (j) Leung, D. H.; Guan, Z. B. J. Am. Chem. Soc. 2008, 130, 7538.

5. (a) Dias, A. R.; Galvao, A. M.; Galvao, A. C.; Salema, M. S. Dalton Trans. 1997, 6, 1055. (b) Seo, W. J. Organomet. Chem. 2001, 640, 79. (c) Huang, J, H.; Chi, L, S.; Yu, R. C.; Jiang, G. J., Yang, W. T.; Lee, G. H. Organometallics 2001, 20, 5788. (d) Novak, A.; Blake, A. J.; Wilson, C.; Love, J. B. Chem. Commun. 2002, 23, 2796. (e) Yasumoto, T.; Yamagata, T.; Mashima, K. Organometallics 2005, 24, 3375. (f) Barnea, E.; Odom, A. L.

Chapter 2

Dalton. Trans. 2008, 30, 4050. (g) Saeed, I.; Katao, S.; Nomura, K. Organometallics 2009, 28, 111.

6. (a) Yunlu, K.; Basolo, F.; Arnold, L. R. J. Organomet. Chem. 1987, 330, 221. (b) Dias, A. R.; Ferreira, A. P.; Veiros, L. F. Organometallics 2003, 22, 5114.

7. Jabri, A.; Korobkov, I.; Gambarotta, S.; Duchateau, R. Angew. Chem. Int. Ed. 2007, 46, 6119.

8. Vidyaratne, I.; Nikiforov, G. B.; Gorelsky, S. I.; Gambarotta, S.; Duchateau, R.; Korobkov, I. Angew. Chem. Int. Ed. 2009, 48, 6552.

9. Thewalt, U.; Wohrle, T. J. Organomet. Chem. 1994, 464, 17.

10. Horton, A. D.; With, J. L.; Vander, A. J.; Vande, W. H. Organometallics 1996, 15, 2672.

11. (a) Korobkov, I.; Gambarotta, S. Organometallics 2009, 28, 5560. (b) Ostoja Starzewski, K. A.; Xin, B. S.; Steinhauser, N.; Schweer, J.; Benet-Buchholtz, J. Angew. Chem. Int. Ed. 2006, 45, 1799. (b) Ostoja Starzewski, K. A.; Steinhauser, N.; Xin, B. S. Macromolecules 2008, 41, 4095.

12. You, Y.; Girolami, G. S. Organometallics 2008, 27, 3172.

13. Nielson, A. J.; Harrison, J.; Shen, C.; Waters, J. M. Polyhedron 2006, 25, 1729.

Chapter 2

14. Shaw, S. L.; Storhoff, J. J.; Cullison, S.; Davis, C. E.; Holloway, G.; Morris, R. J.; Huffman, J.C.; Bollinger, J. C. Inorg. Chim. Acta 1999, 292, 220.

15. (a) Gillis, D. J.; Tudoret, M. J.; Baird, M. C. J. Am. Chem. Soc. 1993, 115, 2543. (b) Quyoum, R.; Wang, Q.; Tudoret, M. J.; Baird, M. C.; Gillis, D. J. Am. Chem. Soc. 1994, 116, 6435. (c) Bochmann, M.; Lancaster, S. J.; Hursthouse, M. B.; Malik, K. M. A. Organometallics 1994, 3, 235. (d) Bochmann, M.; Sarsfield, M. J. Organometallics 1998, 17, 5908. (e) Chen, E. Y.; Marks, T. J. Chem. Rev. 2000, 100, 1391. (f) Beringhelli, T.; Donghi, D.; Maggioni, D.; Dalfonso, G. Coord. Chem. Rev. 2008, 252, 2292.

16. (a) Bochman, M.; Dawson, D. M. Angew. Chem. Int. Ed. Engl. 1996, 35, 2226. (b) Coles, M. P.; Jordan, R. F. J. Am. Chem. Soc. 1997, 119, 8125. (c) Ihara, E.; Young, V. G.; Jordan, R. F. J. Am. Chem. Soc. 1998, 120, 8277.

17. (a) Sarsfield, M. J.; Ewarr, S. W.; Tremblay, L. T.; Baird, M. C. Dalton Trans. 1997, 3097. (b) Thorn, M. G.; Etheridge, Z. C.; Fanwick, P. E.; Rothwell, I. P. Organometallics 1998, 17, 3636. (c) Qian, B.; Ward, D. L.; Smith, M. R Organometallics 1998, 17, 3070. (d) Pindado, G. J.; Thoronton-Pett, M.; Hursthouse, M. B.; Coles, S. J.; Bochmann, M. J. Chem. Soc. Dalton Trans. 1999, 1663. (e) Yang, X.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10015.

18. Braunschweig, H.; Breitling, F. M. Coord. Chem. Rev. 2006, 250, 2691.

19. Jabri, A.; Mason, C. B.; Sim, Y.; Gambarotta, S.; Burchell, T. J.; Duchateau, R. Angew. Chem. Int. Ed. 2008, 47, 9717.

Chapter 3

Synthesis and ethylene oligomerisation studies of pyrrole-based chromium catalysts

Vadake Kulangara, S.; Haveman, D.; Yang, Y.; Vidjayacoumar, B.; Gambarotta, S.; Duchateau, R. “Cocatalyst effect on selective ethylene oligomerisation”. To be submitted.

Chapter 3

Abstract Ethylene oligomerisation capabilities of chromium catalysts stabilised by different dipyrrole-based ancillary ligands, [(Ph2C(C4H4N)2)] (2), [Ph2C(C4H4N)(C5H6N)]

(3) [(Et)2C(C4H4N)2] (4) and [(C6H5)(C5H4N)C(C4H4N)(C5H6N)] (5), were investigated using different activation methods and the results were compared with the commercial Phillips trimerisation system. Upon activation with triethylaluminium (TEA), chromium catalysts stabilised by dipyrrole-based ligands 2-5 showed a lower activity and selectivity compared to the Phillips trimerisation system based on 2,5-dimethylpyrrole (1) as the ancillary ligand. Unprecedented increases in the activity and selectivity were observed by carrying out the oligomerisation in methylcyclohexane using depleted-methylaluminoxane (DMAO) along with triisobutylaluminium (TIBA) (1:2 ratio) as cocatalyst. Apart from selectively producing 1-hexene, the catalyst system 3/Cr(acac)3/DMAO was also capable of incorporating the in situ produced α-olefins into the polymer chain to form different short chain branches (methyl, ethyl and butyl). Furthermore, the effect of ethylene pressure, chromium precursor and solvent on oligomerisation activity and selectivity were investigated. Well-defined chromium complexes, [(Ph2C(C4H3N)2)Cr(Cl)(THF)3] (6) and

{[Ph2C(C4H3N)(C5H6N]Cr(THF)(-Cl)}2 (7) were synthesised and fully characterised. Upon activating with MAO, catalyst 7 produced a Schultz-Flory distribution of oligomers, whereas under the identical oligomerisation conditions catalyst 6/MAO was found to be inactive. The use of MeAlCl2 as cocatalyst to activate 7 resulted in the switching of the catalyst behaviour from producing a Schultz-Flory distribution to a selective trimerisation catalyst and the addition of dialkylzinc along with MAO resulted in an unprecedented increase in the activity of the catalyst. Finally, the influence of varying Al/Cr ratio and temperature on activity and selectivity of 7/MAO was investigated and the highest activity was observed at 90 °C and Al/Cr = 700.

Chapter 3

3.1 INTRODUCTION

Selective trimerisation of ethylene to 1-hexene is an area of high research interest as 1-hexene is widely used as comonomer for the production of LLDPE.1 The first selective trimerisation system producing 1-hexene with more than 90% selectivity was reported by researchers of Chevron-Phillips petroleum company.2 Reagan and coworkers discovered that the combination of chromium (III) alkanoanate, such as chromium (III) 2-ethylhexanoate (Cr(III) 2-EH), with 2,5- dimethylpyrrole and triethylaluminium (TEA) in an aliphatic hydrocarbon solvent such as cyclohexane produced 1-hexene with high selectivity and activity. It was also found that the presence of halogen-containing compounds such as diethyl aluminium chloride (DEAC), GeCl4 and SnCl4 during the catalyst preparation leads to marked improvements in both the catalyst activity and selectivity towards 1- hexene formation.3

Scheme 1. Chevron-Phillips trimerisation system.

Following this discovery, researchers of Mitsubishi Chemical Corporation found that the catalysts prepared with Lewis acids, such as B(C6F5)3, in addition to Cr(III) 2-EH, 2,5-dimethylpyrrole and TEA exhibited significantly higher activities than the catalysts prepared without Lewis acids.4 Even though the pyrrolyl-based catalyst systems were known for decades, questions such as how the ancillary ligand system determines the remarkably high selectivity and activity, as well as which oxidation state of the chromium is

Chapter 3 responsible for the selectivity, remained unanswered for a long time. Recently, our research group has started exploring the versatility of new types of pyrrole-based ligands for olefin polymerisation and/or oligomerisation.5 The key of success is to use a strong Lewis acid, which binds to the N-atom of the pyrrole through σ- bonding leaving the transition metal with no other choice than to bind to the ligand via π-bonding (Figure 1).5 Whereas complex 1 is a self-activating polymerisation 6 catalyst producing UHMWPE (Mw = 2.6 × 10 g/mol, PDI = 1.8), complex 2 is a self-activating trimerisation catalyst. The difference in catalytic behaviour of these two Cr(II) precursors allowed us to correlate the metal oxidation state and the selectivity: chromium(I) is responsible for selective trimerisation, whilst divalent chromium produces Schultz-Flory distributions of oligomers. Polyethylene can either be produced by di- or trivalent chromium species. The hypothesis is that dimeric chromium complex 2 disproportionates in the presence of ethylene to form a transient monovalent species that is responsible for selective trimerisation. This argument was also supported by DFT calculations.

Figure 1. Self-activating polymerisation (right) and trimerisation catalysts (left).

So far, only mono-pyrrolyl systems have been studied as ligands. In this chapter we evaluated the ethylene oligomerisation capabilities of various dipyrrole- based ancillary ligands and the results are compared with the Chevron-Phillips

Chapter 3 trimerisation system. The effect of experimental parameters such as the nature of the cocatalysts, type of solvent, reaction temperature and ethylene pressure on catalyst activity and selectivity were studied in detail.

3.2 RESULTS AND DISCUSSION

Different types of dipyrrole-based ancillary ligands (Scheme 2) were synthesised according to known literature procedures.6 The corresponding chromium catalysts based on these ligands were tested towards their ethylene oligomerisation capabilities under different oligomerisation conditions and the results are described in the following sections.

Scheme 2. Different types of dipyrrole-based ancillary ligands.

Chapter 3

3.2.1 Triethylaluminium (TEA) as cocatalyst

In order to compare the effect of ligand modification on the activity and the selectivity, Cr(acac)3 was activated in the presence of 2,5-dimethylpyrrole (1) and the dipyrrole-based ligands (2-5) using TEA as cocatalyst and methylcyclohexane as solvent at 120 C. During the activation process a chlorine-containing reagent (hexachloroethane) was also added as a promoter, which is known to increase the activity of the catalyst.3 It was reported by Yang et al. that chlorine-containing compounds can act as a Lewis base to break down the dimeric TEA to monomeric TEA.7 It is believed that the monomeric TEA is advantageous to the formation of the active chromium species due to its stronger reducing power. Ethylene oligomerisation results for ligands 1-5 upon activating with TEA are given in the Table 1.

Chapter 3

Table 1. Ethylene oligomerisation results for 1-5/Cr(acac)3 + TEA. Ligand Yield Oligomers PE Activity (g) (mol %) (g) (g/(mmol Cr.h))

C6 C8 ≥C10 1 14.5 90.4 3.8 5.8 0.19 979 2 4.32 74.5 9.2 16.3 - 288 3 4.2 75.9 14.6 9.5 - 280 4 4.2 50.3 10.5 39.2 - 280 5 1.5 88.5 3.1 8.4 0.64 140 a1 10.8 94.9 3.8 1.3 - 645 a2 9.6 46.6 13.4 40.0 0.10 647 a3 7.1 49.5 35.8 10.7 - 511 a4 8.3 60.7 14.4 24.9 0.15 563 a5 2.6 75.4 10.2 14.4 0.12 181 Oligomerisation conditions: Solvent = 100 mL of methylcyclohexane, ethylene pressure = 30 bar, reaction time = 30 min, reaction temperature = 120 °C. Catalyst loading = 30 µmol a) Cr(acac)3, Cr/lig/HCE/TEA = 1/3/5/100. Cr/lig/HCE/TEA = 1/1.5/5/100.

As can be seen from Table 1 and Figure 2, the ethylene oligomerisation activities of the catalyst system containing the ligands 2-5 were considerably lower than the original system containing the ligand 1, however, the selectivity towards 1-hexene was comparable in all cases (Cr:ligand = 1:3) except for the system stabilised by ligand 4. This system did not show selective trimerisation behaviour at Cr:ligand ratios of 1:1.5 or 1:3, suggesting that the presence of a phenyl group is crucial to obtain selectivity. Varying the metal to ligand ratio from 1:3 to 1:1.5 in the case of ligands 2-5 enhances the oligomerisation activity; however, the selectivity drops, indicating that there is a delicate balance between selectivity and activity depending on the metal to ligand ratio. The chromium system stabilised by ligand 5, containing one pyridyl and one phenyl group on the carbon atom = connecting the two pyrrole moieties, showed higher 1-C6 selectivity both at

Chapter 3

Cr/ligand ratios of 1:1.5 or 1:3. Carrying out the oligomerisation at 60 C or in toluene resulted in either no oligomers or trace amount of polymers in all cases. Apart from the formation of even-numbered linear α-olefins, also small amounts of even-numbered alkanes were obtained. The formation of even-numbered alkanes can be explained by the chain transfer reaction to TEA, followed by hydrolysis. A similar observation on the formation of odd-numbered alkanes was reported by

Bazan and coworkers in the case of a Cp*CrMe2(PMe3) catalyst upon activation with MAO (In this case chain transfer reaction occurs to the Al-Me bond of the 8 free AlMe3 present in the MAO). The polyethylene obtained with ligand 1-

5/Cr(acac)3 in all cases showed narrow molecular weight distributions with an Mw <5000, indicating that low molecular weight polyethylene waxes were produced.

100 Cr/lig = 1/3 1200 Cr/lig = 1/3 90 Cr/lig = 1/1.5 80 1000 Cr/lig = 1/1.5 70 800 60 50 600 40 30 400

20 200

Activity (g/(mmol Cr.h)) (g/(mmol Activity hexene selectivity selectivity %) (mol hexene

- 10 1 0 0 1 2 3 4 5 0 1 2 3 4 5 Ligand Ligand

Figure 2. Comparison of 1-hexene selectivity (left) and activity (right) at different Cr/ligand ratios.

3.2.2 Aluminoxanes as cocatalyst

As we found that TEA is not a promising cocatalyst for activating the catalysts systems consisting of Cr(acac)3 and the ligands 2-5, different types of aluminoxane-based cocatalysts were used to activate 3/Cr(acac)3. When activated with MAO and MMAO in toluene, the catalyst produced a Schultz-Flory

Chapter 3 distribution of oligomers (α ≈ 0.6) along with a considerable amount of polyethylene. The DSC analysis of the polymer showed that it has a melting temperature of 122.3 °C and 123.4 °C for MAO- and MMAO-activated systems, respectively. HT-SEC analysis was not possible due to the poor solubility of the polymer in trichlorobenzene at 150 °C. Either the polymer has an extremely high molecular weight or it is cross-linked. In order to clarify the effect of free trimethylaluminium (TMA) present in the MAO on the catalytic behaviour, oligomerisation experiments were also carried out using DMAO in toluene and it was found that removal of free TMA has a detrimental effect on the catalytic activity. In order to verify this, an additional amount of 100 equivalents of TMA was added along with MAO. Indeed, an increase in the activity was observed without considerably affecting the selectivity. Interestingly, when oligomerisation experiments were carried out in methylcyclohexane using DMAO as cocatalyst, a switch in the catalytic behaviour was observed. The catalyst produced a large 5 amount of polyethylene (Mw = 1.2 × 10 g/mol, PDI = 2.3, Tm = 122.2 °C) along with small amounts of oligomers. However, the oligomers showed excellent selectivity towards 1-hexene (95.4 %). The switching of the catalyst behaviour upon changing the polymerisation solvent was previously reported by our group in the case of chromium-based oligomerisation catalysts, supported by aluminium- pyrrolyl and amidophoshine ligands.9, 5 A possible explanation for the switch in catalytic behaviour is ascribed to the poisoning of the catalytically active Cr(I) species by its coordination with toluene to form the corresponding µ6-arene complex.10, 5 Use of MAO or MMAO as cocatalysts in methylcyclohexane resulted in an inactive catalyst. Also when CrCl3(THF)3 was used as metal precursor instead of Cr(acac)3 along with DMAO in methylcyclohexane, the catalyst did not show any oligomerisation activity.

Chapter 3

Table 2. Effect of solvent on ethylene oligomerisation, 3/Cr(acac)3.

Solvent Cocatalyst Yield Oligomers PE Activity (equiv) (g) (mol %) (g) (g/(mmol Cr.h))

C6 C8 ≥C10

Toluene MAO (250) 9.1 41.7 27.5 30.8 3.3 2480

Toluene MMAO (250) 13.7 49.9 28.0 23.1 1.2 2980

Toluene DMAO (250) 2.1 50.3 25.2 24.5 1.2 660 MAO (250)/ Toluene 14.6 43.2 27.5 31.1 4.2 3760 TMA (100)

MeCy MAO (250) ------

MeCy MMAO (250) ------

MeCy DMAO (250) 0.80 95.2 2.7 2.1 10.5 2260 Solvent = 100 mL of toluene, ethylene pressure = 30 bar, reaction time = 30 min, reaction temperature = 60 °C, Cr(acac)3/ligand = 1/1. Cat loading = 10 µmol. MeCy = methylcyclohexane.

13 The high temperature C NMR (C2D2Cl4) analysis results showed that, apart from the methyl end groups (6.8 methyl end group/1000 carbon), the polymer chains also contain a significant number of vinyl end groups (2.7 vinyl end group/1000 carbon), which is typical for chromium-catalysed polymerisation products. Also, the NMR analysis revealed that the polymer is significantly branched and it contains methyl (9.9 methyl branch/1000 carbon), ethyl (1 ethyl branch/1000 carbon) and butyl branches (2.2 butyl branch/1000 carbon), indicating that part of the in situ produced α-olefins were incorporated into the polymer chain (see the appendix for the 13C NMR spectrum). A similar observation was also reported by Manyik et al. in the case of TEA-activated Phillips trimerisation catalysts, where in situ produced 1-hexene was incorporated into the polymer chain to form butyl branches.11 Indeed, this branching is also reflected in a rather low melting temperature of the polymer (Tm = 122.2 °C).

Chapter 3

The formation of ethyl and butyl branches is due to the incorporation of in situ generated 1-butene and 1-hexene, respectively, into the polymer chain.12 However, the origin of a significant number of methyl branching in the polymer chain is rather surprising. A possible explanation for this might be a chain walking mechanism via a β-hydride elimination/reinsertion/migration process. This type of chain walking mechanism, producing a significant number of methyl branches in polyethylene, was also observed in the case of a nickel-catalysed ethylene polymerisation system.13 However, to the best of our knowledge a chromium-based catalyst system producing methyl branches is not known yet.

Scheme 3. Proposed mechanism for the formation of methyl branches (L = ligand, P = polymer chain. Counter anions are omitted for clarity).

Ethylene oligomerisation of 2/Cr(acac)3 in methylcyclohexane showed an interesting catalytic behaviour. Upon activating with MAO and MMAO in methylcyclohexane, 2/Cr(acac)3 only produced UHMWPE with narrow polydispersity indexes, implying the formation of well-defined active species (Mw 6 6 = 4.3×10 g/mol, PDI = 3.4 for an MAO-activated system; Mw = 6.1×10 g/mol, PDI = 2.3 for an MMAO activated system). By using DMAO as cocatalyst in methylcyclohexane, the same catalyst selectively produced 1-hexene with comparable amounts of polyethylene. Again the HT-SEC analysis of the polymer

Chapter 3

6 showed a very high molecular weight (Mw = 1.1×10 g/mol) and a narrow polydispersity index (PDI = 2.6). Polyethylene obtained with MAO and MMAO

(where no oligomers were produced) showed higher melting temperatures (Tm = 134.2 °C, 135.1 °C, respectively) compared to the polymer produced with DMAO

(Tm = 125.8 °C). The lower melting temperature observed in the latter case is likely to be due to the incorporation of the in situ generated 1-hexene into the polymer chain, producing branches on the polymer chain. The 13C NMR analysis of the polymer was not successful due its poor solubility in typical NMR solvents.

Table 3. Oligomerisation of 2/Cr(acac)3 in methylcyclohexane

Entry Cocatalyst Oligomers C6 PE Activity Tm (equiv) (g) (mol%) (g) (g/(mmol Cr.h)) (°C)

1 MAO (250) - - 1.0 53 134.2

2 MMAO (250) - - 2.4 160 135.1

3 DMAO (250) 1.6 95.4 1.1 175 125.8

Solvent = 100 mL of methylcyclohexane, ethylene pressure = 30 bar, reaction time = 30 min, reaction temperature = 60 °C, Cr(acac)3/ligand = 1/1. Cat loading = 30 µmol.

The high temperature NMR analysis of the polymer obtained with MAO (Table 3, entry 1) showed that it contains both methyl (3.7 methyl end group/1000 carbon) and vinyl end groups (2.3 vinyl end group/1000 carbon). The methyl end groups are probably formed via the chain transfer to aluminium followed by the hydrolysis forming saturated chain ends, whereas the vinyl groups are probably formed via -hydride transfer reactions. However, no short chain branches were present along the chain, implying that the catalyst is purely a polymerisation catalyst.

Chapter 3

Even though the activation of 2/Cr(acac)3 and 3/Cr(acac)3 with DMAO mainly produces PE (this implies that probably the reducing power of DMAO is not good enough to produce monovalent chromium species), the small amount of 1-hexene formed along with the PE prompted us to add external reducing agents such as triisobutylaluminium (TIBA) and triethylaluminium (TEA) together with DMAO. The influence of the addition of external reducing agents on activity and selectivity is given in Table 4.

Chapter 3

Table 4. Effect of alkylating agents on activity and selectivity of 1-3/Cr(acac)3

Ligand Cocatalyst Oligomers C6 PE Activity Tm (equiv) (g) (mol%) (g) (g/(mmol Cr.h)) (°C)

2 DMAO (250) 1.6 95.4 1.1 175 125.8

2 DMAO (250)/ 7.1 98.6 2.3 625 123.1

TIBA (500)

2 DMAO (250)/ 3.2 95.1 0.6 253 127.6

TEA (500)

3 DMAO (250) 0.80 95.2 10.5 2260 122.2

3 DMAO (250)/ 8.7 99.2 2.4 2226 124.2

TIBA (500)

3 DMAO (250)/ 3.7 92.1 2.2 1182 124.5

TEA (500)

1 DMAO (250) 8.4 93.1 3.8 2446 125.2

1 DMAO (250)/ 9.68 96.1 - 19360 -

TIBA (500)

Solvent = 100 mL of methylcyclohexane, ethylene pressure = 30 bar, reaction time = 30 min, reaction temperature = 60 °C, Cr(acac)3/ligand = 1/1. Cat loading adjusted according to the exotherm.

As it can be seen from the Table 4, the amount of oligomers produced significantly increases upon the addition of TIBA and TEA and the 1-hexene selectivity is maintained nearly the same. The amount of polymer produced with

3/Cr(acac)3 + DMAO (Table 2) was considerably reduced upon the addition of TEA or TIBA, implying that more monovalent chromium is formed. However, in the case of 2/Cr(acac)3, the yield of polymer increased upon adding TIBA, whereas TEA significantly reduces the amount of polymer.

Chapter 3

This novel and highly promising activation method was also used to activate the Phillips trimerisation system (1/Cr(acac)3). Upon activation with DMAO the system showed moderate trimerisation activity with a significant formation of polymer; however the addition of TIBA along with DMAO resulted in a highly active polymer-free selective trimerisation system. The activity obtained with this new activation method (with ligand 1) was nearly two times higher than the values reported in the Chevron-Phillips patent with a slight improvement in the 1-hexene selectivity.2 Moreover, the higher activity and selectivity was obtained at relatively lower ethylene pressure and temperature (30 bar, 60 °C), whereas in the case of the commercial Phillips trimerisation system high pressure is required (50-100 bar) to observe higher activity.2 A chromium catalyst containing ligand 1 activated with

B(C6F5)3 was also patented by Mitsubishi. Activation of this catalyst with B(C6F5)3 under ethylene pressure of 35 bar, doubled the activity.4 Furthermore, Mitsubishi has significantly improved the performance of the same catalyst via precise control of the process and by slightly modifying the catalyst composition.14 A schematic comparison of the activity and selectivity of the new systems along with some existing trimerisation systems are given in Scheme 4.

Chapter 3

Scheme 4. Comparison of activity and selectivity of different pyrrole-based catalysts.

Keeping in mind that a chlorine-containing reagent can enhance the activity and selectivity of the Phillips trimerisation catalyst, we also added diethylaluminiumchloride (DEAC) as external reducing agent along with DMAO to activate ligands 1-3/Cr(acac)3 in methylcyclohexane. In all the cases no catalytic activity was observed, probably due to the formation of some stable chlorine-

Chapter 3 bridged chelating complex. Addition of other chlorine-containing reagents, such as

MeAlCl2 or EtAlCl2 or using chromium halides CrCl3(THF)3 or CrCl2(THF)2 as metal precursors with DMAO or DMAO/TIBA, all resulted in inactive species. This clearly indicates that all the chlorine-containing reagents somehow poison the catalytic system. When Cr(III) 2-EH was used as metal precursor in combination with ligands 1, 2 and 3 this afforded moderately active trimerisation catalysts. Using Cr(III) 2-EH, however, produced higher amounts of polymer in all the cases compared to Cr(acac)3.

Table 5. Cr(III) 2-EH as metal precursor with ligands 1-3 Ligand Oligomers C6 PE Activity (g) (mol%) (g) ((g/(mmol Cr.h))

1 3.8 97.7 0.12 2080

2 1.3 86.5 0.52 121

3 2.5 90.9 6.2 1744

Solvent = 100 mL of methylcyclohexane, ethylene pressure = 30 bar, reaction time = 30 min, reaction temperature = 60 °C, Cr(III) 2-EH/ligand = 1/1. Cat loading adjusted according to the exotherm.

3.2.3 Effect of ethylene pressure on activity and selectivity

In order to investigate the effect of ethylene pressure on activity and selectivity of ligand 3, catalytic runs were performed at different ethylene pressure using two different metal precursors, namely Cr(acac)3 and Cr(III)2-EH, and the results are summarised in Figure 3. We found that the catalytic activity (amount of oligomers as well as polymer) increases with increase in ethylene pressure in the case of both 3/Cr(acac)3 and 3/Cr(III)2-EH. At lower ethylene pressure (5 bar), both 3/Cr(acac)3 and 3/Cr(III)2-EH do not show any catalytic activity. The observed increase in activity upon increasing the ethylene pressure can be

Chapter 3 explained on the basis of the high solubility of ethylene in methylcyclohexane at higher ethylene pressures. This observation of increased activity at elevated pressure is in accordance with previously reported results for some homogeneous chromium-based catalysts.15 However, the selectivity of the catalysts towards 1- hexene is independent of the ethylene pressure.

2500 120 3/Cr(accac)3 2000 3/Cr(III) 2-EH 100 80 1500 60 3/Cr(accac)3 1000 3/Cr(III) 2-EH 40

500 20

Activity (g/(mmol Cr.h)) (g/(mmol Activity

hexene selectivity selectivity %) (mol hexene - 0 1 0 0 10 20 30 40 0 10 20 30 40 Ethylene pressure (bar) Ethylene pressure (bar)

Figure 3. Effect of ethylene pressure on activity (left) and selectivity (right) of 3/Cr(acac)3 and 3/Cr(III) 2-EH. Oligomerisation conditions: Solvent = 100 mL of methylcyclohexane, reaction time = 30 min, reaction temperature = 60 °C, cocatalyst = DMAO (250)/TIBA (500), Cat loading = 10 µmol.

3.2.4 Synthesis, structural characterisation and ethylene oligomerisation studies of Cr(III) and Cr(II) complexes of dipyrrole ligands

Reaction of in situ prepared potassium salt of ligand 2 in THF with one equivalent of CrCl3(THF)3 resulted in the formation of brown crystals of the octahedral chromium complex [(Ph2C(C4H3N)2)Cr(Cl)(THF)3] (6), whereas the same reaction with CrCl2(THF)2 afforded only microcrystalline powder, not suitable for a single crystal X-ray analysis. The single crystal X-ray structure of 6 showed that the chromium atom is octahedrally surrounded by one chlorine [Cr(1)- Cl(1) = 2.3569(7) Å], three THF molecules [Cr(1)-O(1) = 2.0272(14) Å, Cr(1)- O(2) = 2.1154(15) Å, Cr(1)-O(3) = 2.0696(13) Å] and two nitrogen atoms of the

Chapter 3 dipyrrole ligand [Cr(1)-N(1) = 1.9881(16) Å, Cr(1)-N(2) = 1.9793(17) Å]. The molecular structure of 6 along with selected bond angles and distances is depicted in Figure 4.

Figure 4. ORTEP drawing of [(Ph2C(C4H3N)2)Cr(Cl)(THF)3] (6). Ellipsoids are drawn at 50% probability level and H atoms are omitted for clarity. Selected bond distances (Å) and angles (°) for complex 6: Cr(1)-Cl(1) = 2.3569(7), Cr(1)-N(1) = 1.9881(16), Cr(1)-N(2) = 1.9793(17), Cr(1)-O(1) = 2.0272(14), Cr(1)-O(2) = 2.1154(15), Cr(1)-O(3) = 2.0696(13); N(2)-Cr(1)-Cl(1) = 95.41(5), N(2)-Cr(1)- N(1) = 90.52(6), N(2)-Cr(1)-O(1) = 92.53(6), N(2)-Cr(1)-O(3) = 90.37(6), N(2)- Cr(1)-O(2) = 178.34(6).

The reaction of in situ prepared potassium salt of ligand 3 with CrCl2(THF)2 resulted in the formation of a square planar dimeric chromium complex

{[Ph2C(C4H3N)(C5H6N]Cr(THF)(-Cl)}2 (7), while its reaction with CrCl3(THF)3 resulted only in the formation of oily products. The X-ray structure of the complex (Figure 5) revealed that it is a dimeric chromium complex exhibiting a relatively long chromium-chromium bond distance [Cr(1)-Cr(1a) = 3.490 Å].16 The two chromium atoms are connected via two bridging chlorine atoms of nearly equal bond distances [Cr(1)-Cl(1) = 2.3876(14) Å; Cr(1)-Cl(1a) = 2.4057(14) Å]. The remaining two coordination sites of each chromium atom are occupied by one THF

Chapter 3 molecule [Cr(1)-O(1) = 2.066(3) Å] and the nitrogen atom [Cr(1)-N(1) = 2.025(4) Å] of the ligand moiety. Also the chromium atom has some -interaction with the pyrrolyl ring of the N-methylpyrrolyl moiety with a Cr-Centroid bond distance of 2.95 Å.

Figure 5. ORTEP drawing of {[Ph2C(C4H3N)(C5H6N]Cr(THF)(-Cl)}2 (7). Ellipsoids are drawn at 50% probability level and H atoms are omitted for clarity. Selected bond distances (Å) and angles (°) for complex 7: Cr(1)-N(1), 2.025(4); Cr(1)-N(2), 2.950; Cr(1)-O(1), 2.066(3); Cr(1)-Cl(1), 2.3876(14); Cr(1)-Cl(1a), 2.4057(14); N(1)-Cr(1)-O(1), 91.07(14) ; N(1)-Cr(1)-Cl(1), 92.23(11); Cl(1)-Cr(1)- Cl(1a), 86.53(5); O(1)-Cr(1)-Cl(1a), 88.15(10).

The precatalysts 6 and 7 were assessed under different ethylene oligomerisation conditions and the results are described in the following sessions.

3.2.4.1 Effect of cocatalyst

The nature of the cocatalyst is known to have a profound effect on oligomerisation activity and selectivity.17 To explore the effect of different cocatalysts on the catalytic activity and selectivity of 6 and 7, different activation

Chapter 3 methods were adopted to activate both precatalysts 6 and 7 and the results are summarised in Table 5. Table 5. Effect of cocatalyst on activity and selectivity of catalyst 7

Entry Cocatalyst Yield Oligomers PE Activity (equiv) (g) (mol%) (g) (g/(mmol Cr.h))

C6 C8 ≥C10 1 MAO (700) 15.5 47.6 24.6 27.8 2.0 1750

2 MMAO (700) 11.7 44.2 23.2 32.5 1.6 1330

3 DMAO (700) 14.2 46.1 21.1 32.8 0.2 1440

4 TMA (200) - - - - 1.2 120

5 MAO (700)/TMA 14.5 43.1 28.1 28.8 3.3 1780 (200)

a 6 MAO (700)/ZnEt2 13.5 53.9 18.3 27.8 0.3 9200 (100)

a7 MAO (700)/ 14.8 48.2 22.2 29.6 0.6 10273 ZnMe2 (100)

8 MeAlCl2 (100) 3.5 100 - - 0.2 370

b 9 MeAlCl2 (100) 4.8 100 - - 0.3 510

10 MeAlCl2 ------(100)/ZnEt2 (100)

Solvent = 100 mL of toluene, ethylene pressure = 30 bar, reaction time = 30 min, catalyst loading = 20 µmol, temperature = 60 °C, Al/Cr = 700. a) cat loading = 3 µmol. b) 100 mL of methylcyclohexane as solvent.

Upon activating with MAO, 7 produced a Schultz-Flory distribution of oligomers, whereas under the identical oligomerisation conditions catalyst 6/MAO was found to be inactive. Attempts to activate the precatalyst 6 with MMAO or DMAO only resulted in trace amount of oligomers. The use of different types of aluminoxanes to activate 7 resulted in the formation of a comparable Schultz-Flory distribution of oligomers. However, the use of DMAO as cocatalyst resulted in less

Chapter 3 polymer formation whilst the oligomerisation activity remained unaffected, probably implying that the free TMA present in MAO might be responsible for the formation of polymer. In order to verify this, 7 was activated with pure TMA and the thus obtained catalysts exclusively produced polyethylene. The HT-SEC analysis of the polymer obtained with TMA showed that it is rather a low molecular weight polyethylene with a narrow molecular weight distribution (PDI =

1.9, Mw = 2490 g/mol), whereas the polymer obtained with MAO and DMAO showed very broad molecular weight distributions. Also the addition of an extra amount of free TMA along with MAO enhances the performance of the catalyst system.

Activation of 7 with a combination of MAO and diethyl zinc or dimethyl zinc resulted in a considerable enhancement of the activity of the catalyst. Also the addition of dialkyl zinc resulted in a significant decrease in the amount of polymer along with a slight increase in the C6 selectivity. Carrying out the oligomerisation using 20 µmol catalyst loading resulted in an extremely high exotherm (>30 °C) indicating that catalyst is highly active, whereas the oligomerisation with MAO alone showed a controllable exotherm (3 °C). In order to control the temperature, oligomerisations were carried out using 3 µmol catalyst loading. Even though the effect of dialkyl zinc is well-documented and studied in the case of olefin polymerisation, its effect on selective/non-selective ethylene oligomerisation has only recently been investigated. Waymouth and coworkers studied a selective ethylene oligomerisation process in the presence of ZnR2 as a strategy to make a new class of functionalised ethylene oligomers and also to understand the reactivity of the metallocycle with the dialkylzinc.18 However, the addition of dialkyl zinc to an MAO-catalysed polymerisation/oligomerisation system makes the chemistry of the system complicated, because of the known alkyl exchange reactions between the dialkyl zinc and TMA present in MAO.19 So the exact reason for the increase in activity was difficult to explain.

Chapter 3

Interestingly, a complete switch in the selectivity of 7 from a Schultz-Flory distribution to 1-hexene formation was observed upon changing the cocatalyst to

MeAlCl2. Activation of 7 with MeAlCl2, resulted in the formation of exclusively 1- hexene with the formation of small amount of polymer. Carrying out the oligomerisation in methylcyclohexane using MeAlCl2 resulted in a slight increase in the oligomerisation activity of the catalyst. It was well documented that chlorine-containing reagents can enhance the performance of pyrrole-based trimerisation catalysts through the “chlorine effect”, even though the exact reason for the promoting effect is still not fully understood.3, 7 However, this observation is not in line with the findings that the use of chlorine-containing reagents poison the

2-3/Cr(acac)3/DMAO catalyst system (see the section 3.2.2). Varying the amount of MeAlCl2 or carrying out the oligomerisation at higher temperature also did not considerably enhance the performance of the catalyst. The combination of dialkyl zinc or DMAO with MeAlCl2 or the use of AlEtCl2 as cocatalyst invariably resulted in inactive catalyst systems.

3.2.4.2 Effect of Al/Cr molar ratio and temperature

The influence of the Al/Cr ratio and polymerisation temperature on the activity and selectivity of 6/MAO and 7/MAO was investigated in detail and the results are summarised in Table 6. Varying the Al/Cr ratio from 100 to 1200 in the case of 7 led to a significant enhancement in the activity of the catalyst, whereas the selectivity towards oligomers remains nearly the same (see Figure 6). A further increase of the amount of MAO (Al/Cr = 1500) resulted in a decreased amount of oligomers, whereas it promoted the formation of polyethylene. The HT-SEC analysis of the polymers showed broad PDIs in all cases, implying the formation of multiple catalytically active species.

Chapter 3

2500 60

2000 50 Al/Cr = 100 Al/Cr = 700 40 1500 Al/Cr = 1500 30 1000 Mol (%) Mol 20

500 10 Activity (g/(mmol Cr.h)) (g/(mmol Activity 0 0 0 500 1000 1500 2000 C6 C8 C10 C12 C14 C16 C18 Al/Cr ratio Carbon number

Figure 6. Effect of Al/Cr ratio on activity (left) and distribution of oligomers at different Al/Cr ratios for 7.

Table 6. Effect of Al/Cr ratio and temperature for 7

Entry Al/Cr Temperature Oligomers PE  Activity (°C) (g) (g) (g/(mmol Cr.h))

1 100 60 3.1 0.80 0.58 318

2 400 60 9.2 1.2 0.60 920

3 700 60 15.5 2.0 0.62 1750

4 1200 60 19.2 2.8 0.58 2200

5 1500 60 10.6 6.2 0.58 1680

6 700 30 1.5 0.2 0.61 170

7 700 90 19.2 3.3 0.63 2250

8 700 120 8.9 3.6 0.60 1250

Solvent = 100 mL of toluene, ethylene pressure = 30 bar, reaction time = 30 min, catalyst loading = 20 µmol.

Changing the polymerisation temperature from 30 °C to 120 °C showed a significant influence on the catalytic activity. Both the amount of oligomers and polymer increases with raising the temperature from 30 °C to 90 °C (Figure 7). At a

Chapter 3 temperature of 120 °C the oligomerisation activity of the catalyst drops significantly, however, the amount of polymer produced remains nearly the same at high temperature, indicating that the catalytically active species which produced polymer at 60 °C retains its activity even at higher temperatures. The pronounced decrease in the catalytic activity at still higher temperatures might be due to the faster thermal deactivation of the catalyst at such high temperatures.20, 15b

60 2500 30 °C 50 60 °C 2000 90 °C 40 120 °C 1500 30

1000 (%) Mol 20 500 Activity (g/(mmol (g/(mmol Cr.h)) Activity 10

0 0 50 100 150 0 C6 C8 C10 C12 C14 C16 C18 Temperature (°C) Carbon number Figure 7. Variation of activity with temperature (left) and distribution of oligomers at different temperatures (right) for 7.

3.3 CONCLUDING REMARKS

In conclusion, we evaluated the ethylene oligomerisation capabilities of different dipyrrole-based ancillary ligands using TEA as cocatalyst and found that they are less active and selective compared to the commercial Phillips trimerisation system. However, by using an alternative activation method (via combination of DMAO and TIBA), an efficient way to increase both the activity and selectivity was discovered. Moreover, we found that the catalyst 3/Cr(acac)3/DMAO is capable of incorporating in situ produced α-olefins into the polymer chain to form different short chain branches (ethyl and butyl). Interestingly, a significant amount of methyl branches were also produced probably via chain walking mechanism.

Chapter 3

Furthermore, well-defined chromium complexes (6, 7) of dipyrrole ligands were synthesised (ligands 2 and 3), fully characterised and assessed for their ethylene oligomerisation activities. The use of MeAlCl2 as cocatalyst to activate precatalyst 7 resulted in a switch of the catalyst behaviour from non-selective to selective trimerisation and the addition of dialkyl zinc along with MAO resulted in an unprecedented increase of the activity of the catalyst. Increasing the polymerisation temperature increases the activity of the catalyst and optimum was obtained at 90 C. Varying the Al/Cr ratio from 100 to 1200 in the case of 7 led to a significant enhancement in the activity of the catalyst (both the amount of oligomers and polymers), whereas further increase (Al/Cr = 1500) only enhances the formation of the polymer.

3.4 EXPERIMENTAL SECTION

General Procedures. All air and/or water sensitive reactions were performed under a nitrogen atmosphere, in oven dried flasks using standard Schlenk type techniques. Anhydrous reaction solvents were obtained by means of a multiple column purification system. CrCl2(THF)2 and CrCl3(THF)3 were prepared according to a published procedure.21 Elemental analyses were carried out with a Perkin-Elmer 2400 CHN analyser. Data for X-ray structure determination were obtained with a Bruker diffractometer equipped with a 1K Smart CCD area detector. Molecular weights and molecular weight distributions of the polyethylenes were determined by means of high temperature SEC on a PL- GPC210, equipped with refractive index and viscosity detectors and a 3  PLgel 10 µm MIXED-B column set, at 160 °C with 1,2,4-trichlorobenzene as solvent. BHT and Irganox have been used as antioxidants. The molecular weights of the polyethylenes produced were referenced to linear polyethylene standards. Results of the oligomerisation reactions were assessed by 1H NMR spectroscopy for

Chapter 3 activity and by GC-MS for reaction mixture composition. Gas chromatography of oligomerisation products was conducted on a Varian 450-GC equipped with an auto sampler.

General Oligomerisation Procedure. All oligomerisations were performed in a 250 mL Büchi reactor. The reactor was dried in an oven at 120 °C for two hours prior to each run and then evacuated for half an hour and rinsed with argon three times. After that, the reactor was charged with toluene and the desired amount of cocatalyst. After the solution was stirred for 10 minutes it was saturated with ethylene. The reactor was temporarily depressurised to allow injection of the catalyst solution into the reactor under argon flow, after which the reactor was immediately repressurised to the desired set point. The temperature of the reactor was kept as constant as possible by a thermostat bath. After 30 minutes reaction time and cooling to 0 °C, the reaction mixture was depressurised and a mixture of ethanol and diluted hydrochloric acid was subsequently injected to quench the reaction. The polymer was separated by filtration and dried at 60 °C for 18 hours under reduced pressure before the molecular weight was determined.

Synthesis of [(Ph2C(C4H3N)2)Cr(Cl)(THF)3] (6). A solution of ligand 3,

[Ph2C(C4H4N)2] (0.298 g, 1.0 mmol) in THF (7 mL) was treated with KH (0.088 g, 2.2 mmol) and stirred overnight to give a colourless solution. Addition of

CrCl3(THF)3 (0.375 g, 1.0 mmol) to the above solution changed the colour to brown and the stirring was continued for an additional 6 hours. The solvent was completely removed in vacuum, and the residue was redissolved in THF (5 mL). The mixture was then centrifuged to discard the insoluble colourless solid, and the centrifugate was reduced to half of its original volume and kept in the freezer at -30 °C for two days to yield X-ray quality brown crystals of the product. Yield (0.320 g, 0.53 mmol, 53%). Anal. Calcd (found) for C33H40N2O3ClCr: C, 66.05 (66.29); H,

6.68 (6.74); N, 4.65 (4.80). [μeff = 3.82 μB].

Chapter 3

Synthesis of {[Ph2C(C4H3N)(C5H6N]Cr(THF)(-Cl)}2 (7). A solution of ligand 2,

[Ph2C(C4H4N)(C5H6N] (0.312 g, 1.0 mmol) in THF (10 mL) was treated with KH (0.044 g, 1.1 mmol) and stirred overnight to give a light yellow solution. Addition of CrCl2(THF)2 (0.267 g, 1.0 mmol) to the above solution changed the colour to blue and the stirring was continued for an additional 12 h. The solvent was completely removed in vacuum, and the residue was redissolved in THF (5 mL). The mixture was then centrifuged to discard the insoluble colourless white solid, and the centrifugate was reduced to half of its original volume and layered with n- hexane to yield X-ray quality pale blue crystals of the product. Yield (0.263 g, 0.28 mmol, 56%). Anal. Calcd (found) for C52H54N4O2Cl2Cr2: C, 66.31 (66.59); H, 5.78

(5.72); N, 5.95 (5.98). [μeff = 4.96 μB].

100

Chapter 3

3.5 REFERENCES

1. Vogt, D. Applied Homogeneous Catalysis with Organometallic Compounds; Cornils, B., Herrmann, W. A., Eds.; Wiley-VCH: Weinheim, Germany, 1996; Vol 1, P 245.

3. Reagan, W. K.; Freeman, J. W.; Conroy, B. K.; Pettijohn, T. M.; Benham, E. A. EP 0608447 (Phillips Petroleum Company), 1994.

4. Tanaka, E.; Urata, H.; Oshiki, T.; Aoshima, T.; Kawashima, R.; Iwade, S.; Nakamura, H.; Katsuki, S.; Okanu, T. EP 0 611 743 (Mitsubishi Chemical Corporation), 1994.

5. (a) Jabri, A.; Korobbkov, I.; Gambarotta, S.; Duchateau, R. Angew. Chem. Int. Ed. 2007, 46, 6119. (b) Vidyaratna, I.; Nikiforov, G. B.; Gorelsky, S. I.; Gambarotta, S.; Duchateau, R.; Korobkov, I. Angew. Chem., Int. Ed. 2009, 48, 6552.

6. (a) Berube, C. D.; Yazdanbakhsh, M.; Gambarotta, S.; Yap, G. P. A. Organometallics 2003, 22, 3742. (b) Freckmann, D. M. M.; Dube, T.; Berube, C. D.; Gambarotta, S.; Yap, G. P. A. Organometallics 2002, 21, 1240. (c) Athimoolam, A.; Korobkov, I.; Gambarotta, S. Can. J. Chem. 2005, 83, 832. (d) Athimoolam, A.; Crewdson, P.; Korobkov, I.; Gambarotta, S. Organometallics 2006, 25, 3856.

101

Chapter 3

7. Yang, Y.; Kim, H.; Lee, J.; Paik, H.; Jang, H. G. Appl. Catal. A, 2000, 193, 29.

8. Bazan, G. C.; Rogers, J. S.; Fang, C. C. Organometallics 2001, 20, 2059.

9. Thapa, I.; Gambarotta, S.; Duchateau, R.; Kulangara, S. V.; Chevalier, R. Organometallics 2010, 29, 4080.

10. (a) Macchia, G. L.; Gagliardi, L.; Power, P. P.; Brynda, M. J. Am. Chem. Soc. 2008, 130, 5104. (b) Kreisel, K. A.; Yap, G. P. A.; Dmitrenko, O.; Landis, C. R.; Theopold, K. H. J. Am. Chem. Soc. 2007, 129, 14162.

11. Manyik, R. M.; Walker, W. E.; Wilson, T. P. J. Catal. 1977, 47, 197.

12. (a) Zhu, F.; Huang, Y.; Yang, Y.; Lin, S. J. Polym. Sci., Part A: Polym. Chem .2000, 38, 4258.

13. (a) Ittel, D. S.; Johnson, L. K.; Chem. Rev. 2000, 100, 1169. (b) Svejda, S. A.; Johnson, L. K.; Brookhart, M. J. Am. Chem. Soc. 1999, 121, 10634. (c) Okada, M.; Nakayama, Y.; Ikeda, T.; Shiono, T. Macromol. Rapid Commun. 2006, 27, 1418. (d) Jurkiewicz, A.; Eilerts, W. N.; Hsieh, E. T. Macromolecules 1999, 32, 5471.

14. Araki, Y.; Nakamura, H.; Nanba, Y.; Okanu, T. US 5856612 (Mitsubishi Chemical Corporation), 1999.

15. (a) Bluhm, M. E.; Walter, O.; Doring, M. J. Organomet. Chem. 2005, 690, 713. (b) Zhang, W.; Sun, W. H.; Zhang, S.; Hou, J.; Wedeking, K.; Schultz, S.; Frohlich, R.; Song, H. Organometallics 2006, 25, 1961.

102

Chapter 3

16. (a) Cotton, F. A.; Walton, R. A. Metal-metal Multiple bonds in dinuclear clusters. Struct. Bonding (Berlin) 1985, 62, 1. (b) Horvath, S.; Gorelsky, S. I.; Gambarotta, S.; Korobkov, I. Angew. Chem., Int. Ed. 2008, 47, 9937.

17. McGuinness, D. S.; Rucklidge, A. J.; Tooze, R. P.; Slawin, A. M. Z. Organometallics 2007, 26, 2561.

18. Son, K.; Waymouth, R. Organometallics 2010, 29, 3515.

19. Kempe, R. Chem. Eur. J. 2007, 13, 2764.

20. Svejda, S. A.; Brookhart, M. Organometallics 1999, 18, 65.

21. Kern, R. J. J. Inorg. Nucl. Chem. 1962, 24, 1105.

103

104 Chapter 4

Synthesis and ethylene trimerisation capability of chromium(II) and chromium(III) heteroscorpionate catalysts

The content of this chapter has been published: Kilpatrick, A.; Vadake Kulangara, S.; Cushion, M.; Duchateau, R.; Mountford, P. Dalton Trans, 2010, 39, 3653.

105

Chapter 4

Abstract

i Reaction of (Me2pz)2CHSiMe2N(H)R (R = Pr or Ph) or (Me2pz)2CHSiMe2NMe2 1 2 1 2 i with CrCl3(THF)3 or CrCl2(THF)2 gave Cr{(Me2pz)2CHSiMe2NR R }Cl3 (R = H, R = Pr 1 2 1 2 1 2 (2) or Ph (3); R = R = Me (7)) or Cr{(Me2pz)2CHSiMe2NR R }Cl2(THF) (R = H, R = iPr (4) or Ph (5); R1 = R2 = Me (8)), respectively. Compounds 2 and 3 were crystallographically characterised and the magnetic behaviour of all the new compounds was evaluated using SQUID magnetometry. Reaction of CrCl3(THF)3 with

Li{C(Me2pz)3}(THF) gave the zwitterionic complex Cr{C(Me2pz)3}Cl2(THF) (9) containing an apical carbanion. Reaction of the analogous phenol-based ligand (Me2pz)2CHArOH t (ArO = 2-O-3,5-C6H2 Bu2) with CrCl3 gave Cr{(Me2pz)2CHArOH}Cl3 (11) whereas the corresponding reaction with CrCl2(THF)2 unexpectedly gave the Cr(III) phenolate derivative Cr{(Me2pz)2CHArO}Cl2(THF) (12) which could also be prepared from

CrCl3(THF)3 and the sodiated ligand [Na{(Me2pz)2CHArO}(THF)]2. Reaction of the corresponding ether (Me2pz)2CHArOMe with CrCl3(THF)3 or CrCl2(THF)2 gave

Cr{(Me2pz)2CHArOMe}Cl3 (15) and Cr{(Me2pz)2CHArOMe}Cl2(THF) (16), respectively. The catalytic performance in ethylene oligomerisation/polymerisation of all of the new Cr(II) and Cr(III) complexes was evaluated. Most of the complexes showed high activity, but produced a Schultz-Flory distribution of -olefins. Compound 15 had an exceptionally low -value of 0.37 and showed a preference for 1-hexene and 1-octene formation. While replacing a secondary amine (2-5) for a tertiary amine (7-8) resulted in loss of catalytic activity, replacing a phenol (11) for an anisole (15) group afforded a more selective and more active catalyst. Changing from MAO to DIBAL-O as cocatalyst induced a switch in selectivity to ethylene polymerisation.

All catalysts mentioned in this chapter have been synthesised in the group of Prof. Philip Mountford, Department of Chemistry, University of Oxford, UK.

106

Chapter 4

4.1 INTRODUCTION

Scorpionate ligands are poly(pyrazolyl)-based species generally containing two or three N-deprotonated pyrazolyl ring bound to a main group element through one of the ring nitrogens.1 The main group element is predominantly B or C, but examples with other bridging atoms such as Si, P, N, Ge, Sn etc. are also known.2 Homoscorpionate ligands have the same three moieties appended to the apical group (e.g., BH, CH); while in the case of heteroscorpionate ligands one of the pyrazolyl groups has been replaced by a different C-, O-, S- or N-donor moiety. While much of the early work with heteroscorpionate complexes was concerned with synthetic and structural studies, several reports have focussed more on their applications in the areas of ring-opening polymerisation of cyclic esters3 and Ziegler-Natta polymerisation catalysis.4, 7

In 2002, Tosoh Corporation patented a homoscorpionate tris(dimethylpyrazolyl)methane chromium catalyst, Cr{CH(Me2pz)3}Cl3, which produces 1-hexene with very high selectivity (99 %) and activity (40.1 kg1- -1 -1 5 hexene/(gCr h )) upon activation with MAO in toluene. Later, Zhang et al. reported a series of chromium-based trimerisation catalysts stabilised by fac-coordinated bis(pyrazolyl)methane based ligands.6 Preliminary ethylene trimerisation results for these Cr(III) precatalysts were competitive with that of the “parent” system 1, but the N2S donor derivatives were significantly less active than their N2O and N2N´ analogues.

107

Chapter 4

Scheme 1. Tosoh (left) and Hor (right) trimerisation catalysts.

In this chapter, we investigated the synthesis and ethylene oligomerisation capabilities of several different types of heteroscorpionate Cr(III) and Cr(II) complexes containing secondary or tertiary amine side arms and phenyl ether, phenol or phenolate side arms.

4.2 RESULTS AND DISCUSSIONS

4.2.1 Synthesis of N2N´ and N3 donor scorpionate Cr(II) and Cr(III) complexes

i 7 Reaction of CrCl3(THF)3 with (Me2pz)2CHSiMe2N(H)R (R = Pr or Ph) in THF at room temperature gave a colour change from purple to pale green or pink. After work up and crystallisation from acetonitrile the Cr(III) complexes i Cr{(Me2pz)2CHSiMe2N(H)R}Cl3 (R = Pr (2) or Ph (3)) were obtained in ca 85% yield (Scheme 2). Cr(II) analogues of 2 and 3 were obtained from the reaction of in o situ prepared CrCl2(THF)2 with the same ligands at 70 C in THF, forming the sky i blue THF adducts Cr{(Me2pz)2CHSiMe2N(H)R}Cl2(THF) (R = Pr (4) or Ph (5)) in ca. 60% yield.

108

Chapter 4

Scheme 2. Synthesis of new Cr(II) and Cr(III) N2N΄-donor heteroscorpionate complexes.

The solid state structures of 2 and 3 confirm the presence of neutral ligands bound in a fac-coordinated manner to CrCl3 moieties. The IR spectra show (N-H) absorbances (ca. 3220 cm-1) red-shifted from those of the free ligands, and the elemental analyses are consistent with the expected composition. The solid state structures of 2 and 3 are shown in Figure 1, and selected bond distances and angles are compared in Table 1. In general the distances and angles are within the

3 expected ranges. Each complex contains a  N–bound (Me2pz)2CHSiMe2N(H)R ligand and an approximately octahedral chromium. The Cr-N(1,3) distances are slightly shorter than their Cr-N(5) counterparts, possibly reflecting the difference in

109

Chapter 4 hybridisation of the two types of nitrogen donor (sp2 vs sp3). Somewhat surprisingly, the Cr-Cl(3) (trans to N(H)R) distance is slightly longer than the average Cr(1)-Cl(1,2) distances (trans to the shorter Cr-Npz bonds).

Figure 1. Displacement ellipsoid plots (20% probability) of i Cr{(Me2pz)2CHSiMe2N(H) Pr}Cl3 (2, left) and Cr{(Me2pz)2CHSiMe2N(H)Ph}Cl3 (3, right). C-bound H atoms, dichloromethane of crystallization and minor disorder (for 2) omitted for clarity.

110

Chapter 4

Table 1. Selected distances (Å) and angles (°) for i Cr{(Me2pz)2CHSiMe2N(H) Pr}Cl3 (2) and Cr{(Me2pz)2CHSiMe2N(H)Ph}Cl3 (3). The values in brackets for 2 are for the other orientation of N(5) which is positionally disordered over two sites).

Parameter 2 3 ______

Cr(1)-N(1) 2.129(2) 2.129(3) Cr(1)-N(3) 2.125(3) 2.141(2) Cr(1)-N(5) 2.163(10) [2.182(11)] 2.190(3) Cr(1)-Cl(1) 2.3023(9) 2.2747(9) Cr(1)-Cl(2) 2.3005(8) 2.3291(9) Cr(1)-Cl(3) 2.3292(9) 2.3280(8)

N(1)-Cr(1)-N(3) 86.72(9) 86.54(9) N(1)-Cr(1)-Cl(2) 174.95(7) 169.78(7) N(3)-Cr(1)-Cl(1) 175.88(8) 178.49(8) N(5)-Cr(1)-Cl(1) 82.44(18) [93.6(2)] 90.83(8) N(5)-Cr(1)-Cl(2) 91.6(2) [83.10(19)] 83.33(9) N(5)-Cr(1)-Cl(3) 173.6(2) [172.47(19)] 178.00(8) Cl(1)-Cr(1)-Cl(2) 92.97(3) 91.06(3)

In order to avoid the possible deprotonation of the N-bound H atom during the activation process8 (implying reaction with different alkyl aluminiums) in the case of complexes 2-5, a new ligand (Me2pz)2CHSiMe2NMe2 (6), containing a pendant NMe2 donor in place of the N(H)R was synthesised. Compound 6 was obtained in 48% yield by reaction of the previously reported7 chlorosilane

(Me2pz)2CHSiMe2Cl with LiNMe2 in THF. Examination of the crude reaction

111

Chapter 4 mixture showed evidence for the formation of (Me2pz)2CHLi during this process,

Thus (Me2pz)2CHLi elimination appears to compete with LiCl elimination in this reaction, contributing to the relatively modest yield of 6. Reaction of 6 with

CrCl3(THF)3 and in situ generated CrCl2(THF)2 required slightly more forcing conditions than for the secondary amine analogues (higher temperatures or longer reaction times). The target complexes (Scheme 2) Cr{(Me2pz)2CHSiMe2NMe2}Cl3

(7, 87% yield) and Cr{(Me2pz)2CHSiMe2NMe2}Cl2(THF) (8, 67% yield) were nonetheless satisfactorily obtained in analytically pure form. The solid state magnetic moments of 3.79 μB and 5.03 μB were close to those expected according 3 to the spin-only formula for S = /2 and S = 2 ground states, respectively.

Under the trimerisation reaction conditions reported for 1, o Cr{HC(Me2pz)3}Cl3 (MAO, 80 C), there is a possibility that deprotonation of the apical C–H bond occurs. Therefore, in addition to the heteroscorpionate complexes, we also prepared the zwitterionic, homoscorpionate complex 8a, 8b Cr{C(Me2pz)3}Cl2(THF) (9) from Li{C(Me2pz)3}(THF) and CrCl3(THF)3 (Scheme 3). Compound 9 was obtained as a green microcrystalline solid in 40% crystallised yield.

Scheme 3. Synthesis of zwitterionic, homoscorpionate complex

Cr{C(Me2pz)3}Cl2(THF) (9).

112

Chapter 4

4.2.2 Synthesis of N2O donor heteroscorpionate Cr(II) and Cr(III) complexes

Encouraged by Braunstein and Hor’s preliminary reports6 concerning the

N2O-donor heteroscorpionate Cr(III) complexes, we also extended our collection of Cr(III) and Cr(II) complexes to include derivatives of the bulky 9,3d bis(pyrazolyl)methane-phenol (Me2pz)2CHArOH (10). Reaction of o (Me2pz)2CHArOH (10) with CrCl3(THF)3 in THF for 12 h at 60 C gave the phenol complex Cr{(Me2pz)2CHArOH}Cl3 (11) as a dark green solid in 73% yield (Scheme 2). The IR spectrum of 11 showed a (O-H) band and the solid state magnetic moment was consistent with Cr(III). Reaction of 10 with CrCl2(THF)2 also gave a dark green product. However, although the elemental analysis was consistent with that anticipated for the targeted phenol complex

Cr{(Me2pz)2CHArOH}Cl2(THF), the expected (O-H) band was absent in the IR spectrum. Furthermore, both the solid state (SQUID) and solution (Evans method10) room temperature magnetic moments (3.72 μB and 3.66 μB, respectively) were considerably closer to the magnetic moment expected for S = 3/2 (Cr(III), 3.87 μB) than to the value expected for S = 2 (high spin Cr(II), 4.89

μB). Thus it appears that reaction of 10 with CrCl2(THF)2 spontaneously forms the

Cr(III) phenolate complex Cr{(Me2pz)2CHArO}Cl2(THF) (12, Scheme 4), perhaps by H atom loss from a first-formed phenol intermediate

Cr{(Me2pz)2CHArOH}Cl2(THF). By way of verification, we also prepared 12 3d from CrCl3(THF)3 and the sodiated ligand [Na{(Me2pz)2CHArO}(THF)]2 (13). The samples of 12 prepared by the two routes were indistinguishable.

113

Chapter 4

Scheme 4. Synthesis of new Cr(II) and Cr(III) N2O-donor heteroscorpionate complexes.

114

Chapter 4

In order to block the spontaneous oxidation process leading to 12, and to evaluate the influence of the ligand-bound O-H group in 11 on oligomerisation catalysis, we prepared the new ligand (Me2pz)2CHArOMe (14) from 13 using MeI in THF (Scheme 2). Reaction of 14 with CrCl3(THF)3 or CrCl2(THF)2 in THF at o 70 C for 12 hours gave dark red Cr{(Me2pz)2CHArOMe}Cl3 (15) or dark blue

Cr{(Me2pz)2CHArOMe}Cl2(THF) (16) in very good yields.

4.2.3 Ethylene oligomerisation studies

All of the new N2N΄ and N2O donor heteroscorpionate complexes, together with the zwitterionic homoscorpionate Cr{C(Me2pz)3}Cl2(THF) (9), were evaluated for their ethylene oligomerisation capabilities. We also tested the 5 previously reported complex Cr{HC(Me2pz)3}Cl3 (1) under our experimental conditions in order to make valid comparisons. Catalytic runs were carried out in toluene solvent with MAO activation (Al:Cr = 500:1) under 35 bar ethylene pressure at 60 oC. The most relevant oligomerisation results are listed in Table 2.

115

Chapter 4

Table 2. Ethylene oligomerisation results for 1-5, 11, 12, 15 and 16.

Pre- Yield Oligomers (mol %) α PE Activity -1 catalyst (mL) (g) (golig mmol C6 C8 C10 ≥C12 h-1) 2 18.2 30.2 27.2 19.6 22.9 0.61 0.70 1320 3 6.3 37.1 26.8 16.9 20.2 0.59 0.62 770 4 60.7 36.6 25.5 16.8 21.2 0.67 0.20 4400 5 48.5 36.8 26.4 17.8 19.0 0.64 0.28 3492 11 17.4 39.2 25.4 15.5 19.8 0.61 0.35 1280 12 17.0 42.2 25.7 15.1 17.1 0.59 0.10 1300 15 37.0 52.4 34.7 11.8 1.1 0.37 1.80 2310 15a 15.7 41.8 25.6 13.7 18.8 0.64 1.6 4520 15b ------16 23.9 37.6 25.9 16.2 20.3 0.61 0.38 1580 1 8.9 91.5 0.1 2.4 6.0 0.79 0.16 690

Conditions: 100 mL of toluene, 35 bar C2H4, Al(MAO):Cr = 500; 20 µmol pre-catalyst loading, 30 min, 60 ºC. a) 5 µmol precatalyst loading and DMAO as cocatalyst. b) DMAO as cocatalyst and methylcyclohexane as solvent.

The first observation that can be made is that all new complexes are more active than 1 but invariably produce a Schultz-Flory distribution of oligomers. Whereas most complexes display a typical α-value between 0.59 and 0.67, compound 15 with an α-value of 0.37 shows a significantly higher selectivity for 1- hexene (52.4%) and 1-octene (34.7%). In general, an initial temperature rise of 4-7 °C occurred upon injection of pre-catalyst solution, which remained constant until the catalyst deactivated. Reaction profile assessment of selected active catalysts (1– 3) showed that catalytic activity had ceased after 30 minutes, and that either increasing the reactor temperature to 80 oC or reducing it to 30 oC decreased the

116

Chapter 4 yield of oligomers while leaving the mass of undesirable polyethylene co-product approximately constant. The polyethylene obtained in all cases shows an extremely broad multimodal product distribution, typical for multiple site chromium catalysts. When comparing the catalytic behaviour of the various catalysts in more detail, several observations can be made. The divalent complexes 4 and 5 are clearly more active than their trivalent equivalents 2 and 3, but the selectivity of all four is rather similar. This suggests that both Cr(II) and Cr(III) complexes (2/4 and 3/5, respectively) are precursors to the same active species. Based on the lack of selectivity the active species most likely contains chromium in the divalent state.11

The effect of changing from a pendant secondary amine to pendant tertiary amine has also been evaluated. Surprisingly, both the trivalent 7 and divalent 8 lack any form of activity. There has been some debate about the role and importance of the NH-functionality in mer-SNS and fac-NNX ligand systems.12 Although it was initially assumed that deprotonation of the NH-functionality of the SNS ligand occurs upon activation with MAO, this was found not to be the case. The deprotonated version of the ligand actually afforded an inactive complex and pyridine-based SNS ligands lacking an NH functionality afford active trimerisation catalysts.11g-i Hor and coworkers demonstrated that treating a bis(pyrazolyl)(methyleneamine)methane chromium complex containing a secondary amine with AlMe3 resulted in deprotonation of the amine, which did not affect the catalytic behaviour.6b

The virtually identical yield and product distribution for 11 and 12 suggests that they are both precursors of the same active species, in which the phenol is obviously deprotonated. Furthermore, the THF present in 12 does not inhibit activity, presumably due to the scavenging effects of MAO. The GC analysis of the product mixtures formed from 11 and 12 revealed the unexpected formation of a statistical distribution of odd-numbered α-olefins and saturated even-numbered

117

Chapter 4 oligomers as by-products of the expected distribution of even-numbered α- olefins.13 The amount of these odd-numbered oligomers (approx. 1.3 mL, 6 mmol) is of the order of the amount of MAO in the reaction mixture, which might suggest the mechanism by which they are formed is non-catalytic. Wass et al. indeed proved that free AlMe3 present in the MAO is responsible for the formation of odd- numbered olefins.14

Replacing the phenolic hydrogen in 11 for a methyl substituent (15) has an unexpected positive effect on the catalytic activity and selectivity. Although somewhat more polyethylene is formed, the α-value of 0.37 for complex 15 shows an increased selectivity for 1-hexene and 1-octene. In order to understand whether the free AlMe3 present in the MAO is poisoning the activity and selectivity of catalyst 15, oligomerisation was carried out using DMAO as cocatalyst. Interestingly, the activity of the catalyst increased significantly, however, the selectivity drops to a normal Schulz-Flory distribution with an α-value of 0.64. Carrying out the polymerisation in methylcyclohexane using DMAO as cocatalyst resulted only in the formation of a trace amount of polyethylene. Interestingly, the divalent 16 displays a somewhat lower activity than 15 but moreover, the α-value is significantly higher yielding a broader product distribution compared to 15. This strongly suggests that 15 and 16 do not lead to the same active species.

As mentioned above, free tris(dimethylpyrazolyl)methane is readily deprotonated at the apical C-H bond by strong Brønsted bases. It might therefore be possible that deprotonation of the coordinated ligand in Cr{HC(Me2pz)3}Cl3 (1) occurs under trimerisation reaction conditions. The catalytic behaviour of the apically-deprotonated Cr{C(Me2pz)3}Cl2(THF) (9) was therefore compared with that of 1 under identical conditions. Surprisingly, 9 shows no catalytic activity which strongly suggests that deprotonation of the apical C-H bond of 1 under

118

Chapter 4 trimerisation conditions is very unlikely (assuming that the coordinated THF in 9 would be effectively scavenged by the MAO under polymerisation).

As earlier reports showed that the type of cocatalyst can have a profound effect on the catalytic selectivity,15 we tested several pre-catalysts (2, 5, 7, 11, 15) with AlMe3 (TMA), AliBu3 (TIBA), DIBAL-O and MMAO-7. TMA and TIBA did not afford any active catalyst. In agreement with earlier reports,15 when activated with DIBAL-O and MMAO-7 all complexes tested produced exclusively polyethylene (Table 3).

Table 3. Polymerisation results for selected precatalysts using DIBAL-O as cocatalyst.

Pre-catalyst Polyethylene Activity (g) (kgmol-1h-1) 2 0.43 86 5 16.22 3240 7 0.55 110 11 3.40 680 15 5.31 1060

Conditions: 100 mL toluene, 35 bar C2H4, Al:Cr = 500 with 10 μmol pre-catalyst loading, 30 min, 50ºC.

4.3 CONCLUDING REMARKS

N2N΄ and N2O donor heteroscorpionate ligands with secondary and tertiary amine donors, as well as phenol and phenyl ether donors, were successfully complexed to Cr(II) and Cr(III). Their solid state structures and magnetic properties have been determined. The tris(pyrazolyl)methanide analogue (9) was

119

Chapter 4 also prepared. The ethylene oligomerisation capability of the new complexes was compared with that of the “parent” system Cr{HC(Me2pz)3}Cl3 (1) under identical reaction conditions. The results show that very small changes to the nature of the pendant donor atom (N, O) and/or the presence of ionisable hydrogen (N-H, O-H) can have a large effect on activity and/or selectivity. Although in almost all cases the new complexes gave an increase in productivity compared to 1, disappointingly only one of them (15) had any particular selectivity for 1-hexene and 1-octene. Removal of the apical C-H proton from 1 (forming 9) gave a switch from a highly selective and productive catalyst to an inactive one.

4.4 EXPERIMENTAL SECTION

General methods and instrumentation. All manipulations were carried out using standard Schlenk line or dry-box techniques under an atmosphere of argon or nitrogen. Solvents were degassed by sparging with nitrogen and dried by passing through a column of the appropriate drying agent. Deuterated solvents were refluxed over the appropriate drying agent, distilled and stored under nitrogen in Teflon valve ampoules. NMR samples were prepared under nitrogen in 5 mm Wilmad 507-PP tubes fitted with J. Young Teflon valves. 1H and 13C-{1H} NMR spectra were recorded on Varian Mercury-VX 300 and Varian Unity Plus 500 spectrometers and referenced internally to residual protio-solvent (1H) or solvent (13C) resonances, and are reported relative to tetramethylsilane ( = 0 ppm). Assignments were confirmed as necessary with the use of DEPT-135, DEPT-90, and two dimensional 1H-1H and 13C-1H NMR correlation experiments. Chemical shifts are quoted in  (ppm) and coupling constants in Hz. IR spectra were recorded on a Nicolet Magna 560 E.S.P. FTIR spectrometer. Samples were prepared in a dry-box as Nujol mulls between NaCl plates, and the data are quoted in wavenumbers (cm-1). Mass spectra were recorded by the mass spectrometry service

120

Chapter 4 of Oxford University's Department of Chemistry. Elemental analyses were carried out by the Elemental Analysis Service at the London Metropolitan University. SQUID Magnetometry measurements were carried out on a Quantum Design MPMSXL at 1000 Oe. Gas chromatography of oligomerisation products was conducted on a Varian 450-GC equipped with an autosampler and a factor four capillary column VF-5ms 25M×0.25MM. A gradient oven temperature program, starting from 50 °C (for 2 min) to 280 °C at a rate of 10 °C min-1 and holding at the final temperature for 3 min was employed. The chromatograms were obtained via flame ionization detector (FID). As a transport gas ethanol was used. The following compounds were prepared according to (or by analogy with) 16 t 7 literature procedures: CrCl3(THF)3, (R2pz)2CHSiMe2Cl (R = Me or Bu), i i 7 3d, 9b (Me2pz)2CHSiMe2N(H)R (R = Pr, Ph or 2,6-C6H3 Pr2), (Me2pz)2CHArOH, 3d 17 5 [Na{(Me2pz)2CHArO}(THF)]2, HC(Me2pz)3, Cr{HC(Me2pz)3}Cl3, and 8a,b n Li{C(Me2pz)3}(THF). LiNMe2 was prepared from HNMe2 and BuLi in hexanes.18 Other reagents were obtained commercially and used as received.

General procedure for ethylene oligomerisation. Catalytic runs were carried out in a 200 mL high-pressure Büchi reactor containing a heating/cooling jacket. A preweighed amount of catalyst was dissolved in 6 mL of toluene in a glove box and pre-activated with 1 mL of cocatalyst solution to increase solubility. The activated catalyst solution was sonicated for 1 min and immediately injected into the preheated reactor already charged with co-catalyst and toluene (total volume 100 mL) and saturated with ethylene. Solutions were heated using a thermostatic bath and charged with 35 bar of ethylene, the stirring rate was increased to 1000 rpm and the ethylene pressure was maintained throughout the run. The oligo- /polymerisations were quenched by venting the reactor (after cooling to below 0 °C for oligomerisation runs) and addition of EtOH and HCl. The resulting polymer was isolated by ﬁltration, sonicated with an acidiﬁed ethanol solution, rinsed, and

121

Chapter 4 thoroughly dried prior to mass determination and GPC analysis. Oligomers were analysed using GC and 1H NMR spectroscopy.

i Synthesis of Cr{(Me2pz)2CHSiMe2N(H) Pr}Cl3 (2). To a purple solution of

CrCl3(THF)3 (1.02 g, 2.71 mmol) in THF (30 mL) was added a solution of i (Me2pz)2CHSiMe2N(H) Pr (0.91 g, 2.86 mmol) in THF (30 mL), giving a green solution with a small amount of solid. The resulting mixture was stirred for 4 h at RT. The volume was concentrated by half and hexanes (10 mL) were added to complete crystallisation. The green solid was collected by filtration, washed with hexanes (3  5 mL) and dried in vacuo. Recrystallisation from a saturated acetonitrile (50 mL) solution at -30 ºC afforded 2 as a pale green, microcrystalline -1 solid. Yield: 1.11 g (86 %). IR (NaCl plates, Nujol mull, cm ): 3222 (m, N-H), 3134 (w), 1558(m), 1417(s), 1304 (w), 1236(m), 1199(m), 1168 (s), 1125 (w), 1048 (w), 967 (w), 918 (m), 889 (m), 860 (s), 845 (s), 829 (m), 791(s), 765 (w), + + 687 (w). EI-MS: m/z = 440 (5 %), [M - HCl] ; 405 (20 %), [M – HCl - Cl] . Anal. found (calcd. for C16H29Cl3CrN5Si): C, 40.26 (40.21); H, 6.05 (6.11); N, 14.57

(14.66) %. μeff = 3.90 μB.

Synthesis of Cr{(Me2pz)2CHSiMe2N(H)Ph}Cl3 (3). To a purple solution of

CrCl3(THF)3 (0.30 g, 0.81 mmol) in THF (20 mL) a solution of

(Me2pz)2CHSiMe2N(H)Ph (0.30 g, 0.85 mmol) in THF (20 mL) was added, giving a light pink solution with a small amount of solid. The resulting mixture was stirred for 4 hours at RT. The solution was concentrated to half its volume and hexanes (10 mL) were added to complete crystallisation. The pale pink solid was collected by filtration, washed with hexanes (3  5 mL) and dried in vacuo. Recrystallisation from a saturated acetonitrile (30 mL) solution at -30 ºC afforded 3 as a microcrystalline pink solid. Yield: 0.33 g (84 %). IR (NaCl plates, Nujol mull, cm- 1 ): 3221 (m, N-H), 3036 (w), 1557 (m), 1493 (w), 1415 (w), 1394 (w), 1045 (m),

122

Chapter 4

851 (s), 797 (m), 698 (m). EI-MS: m/z = 439 (50 %), [M – HCl - Cl]+. Anal. found

(calcd. for C19H27Cl3CrN5Si): C, 44.49 (44.58); H, 5.37 (5.32); N, 13.68 (13.67) %.

μeff = 3.99 μB. i Synthesis of Cr{(Me2pz)2CHSiMe2N(H) Pr}Cl2(THF) (4). Anhydrous CrCl2 (0.43 g, 3.49 mmol) in THF (40 mL) was heated at 70 ºC for 12 h, allowed to cool i to RT and a solution of (Me2pz)2CHSiMe2N(H) Pr (1.09 g, 3.67 mmol) in THF (30 mL) was added. A colour change from pale green to sky blue occurred and a small amount of solid was formed. The resulting mixture was stirred for 4 h at 70 ºC. The solution was concentrated to one third of its volume and 10 mL of hexanes were added to complete crystallisation. The solid was collected by filtration, washed with cold hexanes (3  5 mL) and dried in vacuo, yielding 4 as a sky blue solid.

-1 Yield: 1.96 g (62 %). IR (NaCl plates, Nujol mull, cm ): 3225 (m, N-H), 3084 (w), 3122 (w), 1556 (m), 1416 (m), 1316 (m), 1285 (w), 1239 (w), 953 (w), 876 (m),

855 (m), 788 (m), 752 (w), 724 (w). Anal. found (calcd. for C20H37Cl2CrN5OSi): C,

46.69 (46.62); H, 7.25 (7.18); N, 13.61 (13.61) %. μeff = 4.92 μB.

Synthesis of Cr{(Me2pz)2CHSiMe2N(H)Ph}Cl2(THF) (5). Anhydrous CrCl2 (0.28 g, 2.28 mmol) in THF (40 mL) was heated at 70 ºC for 12 h, allowed to cool to RT and a solution of HC(Me2pz)2CHSiMe2N(H)Ph (0.85 g, 2.39 mmol) in THF (30 mL) was added. A colour change from pale green to sky blue occurred and a small amount of solid formed. The resulting mixture was stirred for 4 h at 70 ºC. The solution was concentrated to one third of its volume and 10 mL of hexanes were added to complete crystallisation. The solid was collected by filtration, washed with cold hexanes (3  5 mL) and dried in vacuo, yielding 5 as a sky blue

-1 solid. Yield: 1.35 g (61%). IR (NaCl plates, Nujol mull, cm ): 3202 (m, N-H), 3085 (w), 1600 (w), 1591.4 (w), 1554 (m), 1418 (m), 1316 (m), 1234 (m), 1203 (m), 1067 (w), 903 (w), 855 (m), 865 (s), 837 (m), 796 (s), 763 (m), 697 (m), 666

123

Chapter 4

(m). Anal. found (calcd. for C23H35Cl2CrN5OSi): C, 50.29 (50.36); H, 6.56 (6.43);

N, 12.86 (12.77) %. μeff = 5.08 μB.

Synthesis of (Me2pz)2CHSiMe2NMe2 (6). To a solution of (Me2pz)2CHSiMe2Cl (2.66 g, 8.56 mmol) in THF (40 mL) dropwise a solution of lithium dimethylamide (0.46 g, 8.56 mmol) in THF (20 mL) was added. The resulting yellow solution was stirred for 1 h. The volatiles were removed under reduced pressure to give an off white solid which was washed with Et2O (3  50 mL) and dried in vacuo. Extraction into hot benzene (3  50 mL, 70 ºC) followed by removal of the volatiles under reduced pressure gave a yellow oil as the crude product. Recrystallisation from pentane (20 mL) at -30 ºC afforded 6 as an analytically pure 1 white solid. Yield: 1.26 g (48 %). H NMR (C6D6, 299.9MHz, 293K): 5.91 (1 H, s,

(Me2pz)2CHSiMe2NMe2), 5.67 (2 H, s, N2C3HMe2), 2.46 (6 H, s, CHSiMe2NMe2),

2.21 (6 H, s, N2C3HMe2), 1.84 (6 H, s, N2C3HMe2), 0.48 (6 H, s, 13 1 (Me2pz)2CHSiMe2NMe2) ppm. C-{ H} NMR (C6D6, 75.4 MHz, 293K): 146.8 (5-

N2C3HMe2), 140.1 (3-N2C3HMe2), 106.6 (4-N2C3HMe2), 69.0

((Me2pz)2CHSiMe2NMe2), 38.7 (CHSiMe2NMe2), 14.1 (N2C3HMe2), 11.0 - (N2C3HMe2), -1.5 ((Me2pz)2CHSiMe2NMe2) ppm. IR (NaCl plates, Nujol mull, cm 1): 1553 (m), 1419 (m), 1357 (s), 1319 (w), 1290 (m), 1276 (m), 1175 (m), 1027 + (m), 1002 (w), 785 (m), 667 (w), 661 (m). EI-MS: m/z = 261 (15 %), [M – NMe2] ; + + 210 (30%), [M – Me2pz] ; 102 (60%), [M – HC(Me2pz)2] . Anal. found (calcd. for

C15H27N5Si): C, 58.97 (58.88); H, 8.91 (8.86); N, 22.92 (22.96) %.

Synthesis of Cr{(Me2pz)2CHSiMe2NMe2}Cl3 (7). To a purple solution of

CrCl3(THF)3 (1.17 g, 3.14 mmol) in THF (30 mL) a solution of

(Me2pz)2CHSiMe2NMe2 (6) (1.01 g, 3.31 mmol) in THF (20 mL) was added, giving a pink solution with a small amount of solid. The resulting mixture was stirred for 4 h at 70 ºC, concentrated to half its volume and hexanes (10 mL) were added to complete crystallisation. The solid was collected by filtration, washed

124

Chapter 4 with hexanes (3  5 mL) and dried in vacuo, affording 7 as a pale pink solid. Yield: 1.27g (87%). IR (NaCl plates, Nujol mull, cm-1): 3087 (w), 1560 (m), 1415 (w),

1304 (w), 1045 (m), 895 (m), 847 (m). Anal. found (calcd. for C15H27Cl3CrN5Si):

C, 38.90 (38.84); H, 5.86 (5.87); N, 15.01 (15.10) %. μeff = 3.79 μB.

Synthesis of Cr{(Me2pz)2CHSiMe2NMe2}Cl2(THF) (8). Anhydrous CrCl2 (0.22 g, 1.78 mmol) in THF (40 mL) was heated at 70 ºC for 12 h, allowed to cool to RT and a solution of (Me2pz)2CHSiMe2NMe2 (6) (0.57 g, 1.87 mmol) in THF (15 mL) was added. A colour change from pale green to sky blue occurred and a small amount of solid formed. The resulting mixture was stirred for 4 h at 70 ºC, concentrated to one third of its volume and 10 mL of hexanes were added to complete crystallisation. The solid was collected by filtration, washed with cold hexanes (3  5 mL) and dried in vacuo, affording 8 as a turquoise solid. Yield: 0.60 g (67 %). IR (NaCl plates, Nujol mull, cm-1): 3085 (w), 1551 (w), 1317 (w), 1238 (w), 901 (m), 848 (m), 760 (w), 700 (w). Anal. found (calcd. for

C19H35Cl2CrN5OSi): C, 45.49 (45.59); H, 6.93 (7.05); N, 13.86 (13.99) %. μeff =

5.03 μB.

Synthesis of Cr{C(Me2pz)3}Cl2(THF) (9). To a purple solution of CrCl3(THF)3

(0.59 g, 1.57 mmol) in THF (20 mL) a solution of [Li{C(Me2pz)3}(THF)] (0.62 g, 1.65 mmol) in THF (30 mL) was added. The mixture was stirred for 12 h at 60 ºC, giving an orange solution and a small amount of solid. The volatiles were removed under reduced pressure and the residues were extracted with hot benzene (4  20 mL, 60 ºC). Volatiles were again removed under reduced pressure giving a green solid. Crystallisation from a minimum volume of toluene at -30 ºC yielded 9 as a dark green microcrystalline solid. Yield: 0.27 g (40%). IR (NaCl plates, Nujol mull, cm-1): 3127(w), 1556 (m), 1210 (w), 871 (m), 667 (m). EI-MS: m/z = 418 (20 + %), [M - THF] . Anal. found (calcd. for C20H28Cl2CrN6O): C, 48.85 (48.89); H,

5.74 (5.96); N, 16.99 (17.10) %. μeff = 3.76 μB.

125

Chapter 4

Synthesis of Cr{(Me2pz)2CHArOH}Cl3 (11). To a purple solution of CrCl3(THF)3

(0.82 g, 2.20 mmol) in THF (30 mL) a solution of (Me2pz)2CHArOH (10) (0.95 g, 2.32 mmol) in THF (20 mL) was added. The mixture was stirred for 12 h at 60 ºC, giving a dark green solution. The volatiles were removed under reduced pressure giving a dark green solid that was washed with hexanes (3  10 mL) and dried in vacuo. Extraction with toluene (3  10 mL) and recrystallisation at -30 ºC afforded 11 as a dark green microcrystalline solid. Yield: 0.91 g (73 %). IR (NaCl plates, Nujol mull, cm-1): 3367 (br, w), 3191 (w), 3120 (w), 3093 (w), 1596 (w), 1563 (m), 1415 (m), 1343 (s), 1297 (m), 1200 (m), 1161 (m), 1046 (m), 1016 (m), 918 (w), 854 (m), 683 (w). EI-MS: m/z = 530 (20 %), [M - HCl]+. Anal. found (calcd. for

C25H36Cl3CrN4O): C, 52.96 (53.03); H, 6.42 (6.40); N, 9.88 (9.89) %. μeff = 3.39

μB.

Synthesis of Cr{(Me2pz)2CHArO}Cl2(THF) (12). Method 1 – from CrCl2.

Anhydrous CrCl2 (0.075 g, 0.62 mmol) in THF (30 mL) was heated at 70 ºC for 12 h, allowed to cool to RT and a solution of (Me2pz)2CHArOH (0.36 g, 0.63 mmol) in THF (10 mL) was added. A colour change from pale green to dark green occurred and the mixture was stirred for 12 h at 70 ºC. The volatiles were removed under reduced pressure to give a green solid that was washed with hexanes (3  10 mL) and dried in vacuo. Extraction with THF (3  10 mL), concentration, and cooling to -30 ºC afforded 12 as a dark green microcrystalline solid. Yield: 0.30 g

(81%). Method 2 – from CrCl3(THF)3. To a purple solution of CrCl3(THF)3 (0.62 g, 1.7 mmol) in THF (30 mL) a solution of [Na{(Me2pz)2CHArO}(THF)]2 (13) (0.88 g, 1.8 mmol) in THF (20 mL) was added. The mixture was stirred for 12 h at 60 ºC, giving a dark green solution. The volatiles were removed under reduced pressure giving a dark green solid that was washed with hexanes (3  10 mL) and dried in vacuo. Extraction with toluene (3  10 mL) and cooling to -30 ºC afforded 12 as a dark green microcrystalline solid. Yield; 0.47 g (47 %). IR (NaCl plates,

126

Chapter 4

Nujol mull, cm-1): 3091 (w), 1600 (w), 1562 (m), 1342 (s), 1299 (m), 1201 (w), 1159 (w), 1120 (w), 972 (w), 918 (w), 855 (w), 842 (m), 776 (s), 753 (w), 682 (s). EI-MS: m/z = 566 (20%), [M – Cl]+; 494 (100%), [M – Cl - THF]+. Anal. found

(calcd. for C29H44Cl2CrN4O2): C, 57.79 (57.71); H, 7.30 (7.35); N, 9.33 (9.28) %. 27, 28 μeff = 3.72 μB (solid state); 3.66 μB (Evans method, toluene-d8).

Synthesis of (Me2pz)2CHArOMe (14). To a suspension of

[Na{(Me2pz)2CHArO}(THF)]2 (1.68 g, 3.34 mmol) in THF (40 mL) at -20 ºC MeI (0.23 mL, 3.67 mmol) was added dropwise. The mixture was allowed to warm to RT and was stirred for a further 6 h, giving a clear yellow solution. The volatiles were removed under reduced pressure and the resultant solid was extracted with pentane (3  15 mL). Removal of the volatiles under reduced pressure yielded 14 as a yellow, glassy solid which was recrystallised from pentane at -30 ºC to afford 1 14 as a yellow microcrystalline solid. Yield: 0.99 g (70%). H NMR (C6D6, 299.9 3 MHz, 293 K): 7.97 (1 H, s, (Me2pz)2CHArOMe), 7.57 (1 H, d, J = 3 Hz, 4- t 3 t C6H2 Bu2), 7.51 (1 H, d, J = 3 Hz, 6-C6H2 Bu2), 5.70 (2 H, s, N2C3HMe2), 3.42 (3

H, s, (Me2pz)2CHArOMe), 2.16 (6 H, s, 3 or 5 N2C3HMe2), 2.01 (6 H, s, 3 or 5

N2C3HMe2), 1.44 (9 H, s, 3-C6H2(CMe3)2), 1.25 (9 H, s, 5-C6H2(CMe3)2) ppm. 13 1 t C-{ H} NMR (C6D6, 75.4 MHz, 293 K): 153.3 (2-C6H2 Bu2), 148.2 (3 or 5- t t N2C3HMe2), 142.2 (5-C6H2 Bu2), 141.2 (3 or 5-N2C3HMe2), 140.0 (3-C6H2 Bu2), t t t 125.3 (4-C6H2 Bu2), 125.1 (6-C6H2 Bu2), 124.7 (1-C6H2( Bu)2), 106.9 (4-

N2C3HMe2), 74.7 ((Me2pz)2CHArOH), 35.6 (3-C6H2(CMe3)2), 34.4 (5-

C6H2(CMe3)2), 31.7 (5-C6H2(CMe3)2), 30.1 (3-C6H2(CMe3)2), 13.6 (N2C3HMe2), -1 11.2 (N2C3HMe2) ppm. IR (NaCl plates, Nujol mull, cm ): 3118 (w), 3086 (w), 1557 (m), 1334 (s), 1309 (m), 1293 (m), 1229 (m), 1116 (m), 1158 (m), 1028 (m), 869 (m), 834 (m), 781 (m), 631 (w), 607 (w), 682 (s). EI-MS: m/z = 391 (80 %), + + [M - OMe] ; 327 (100 %), [M – Me2pz] . Anal. found (calcd. for C26H38N4O): C, 73.79 (73.89); H, 9.01 (9.06); N, 13.30 (13.26) %.

127

Chapter 4

Synthesis of Cr{(Me2pz)2CHArOMe}Cl3 (15). To a purple solution of

CrCl3(THF)3 (0.65 g, 1.70 mmol) in THF (10 mL) a solution of

(Me2pz)2CHArOMe (0.77 g, 1.80 mmol) in THF (10 mL) was added. The mixture was stirred for 12 h at 70 ºC, to form a dark red solution. The volatiles were removed under reduced pressure to give a solid that was washed with hexanes (310 mL) and dried in vacuo. Extraction with toluene (20 mL), concentration and recrystallisation at -30 ºC afforded 15 as a red microcrystalline solid. Yield: 0.87 g (86%). IR (NaCl plates, Nujol mull, cm-1): 3300 (w), 1571 (w), 1230 (w), 1202

(w), 869 (w), 667 (m). Anal. found (calcd. for C26H38Cl3CrN4O): C, 53.73 (53.75);

H, 6.65 (6.59); N, 9.61 (9.64) %. μeff = 3.56 μB.

Synthesis of Cr{(Me2pz)2CHArOMe}Cl2(THF) (16). Anhydrous CrCl2 (0.14 g, 1.14 mmol) in THF (15 mL) was heated at 70 ºC for 12 h, allowed to cool to RT and a solution of (Me2pz)2CHArOMe (0.51 g, 1.21 mmol) in THF (10 mL) was added. A colour change from pale blue to dark blue occurred and the mixture was stirred for a further 12 h at 70 ºC. The volatiles were removed under reduced pressure to give a blue solid that was washed with hexanes (3  10 mL) and dried in vacuo. Extraction with THF (3  10 mL), concentration and cooling to -30 ºC afforded 16 as a blue, microcrystalline solid. Yield: 0.51 g (72%). IR (NaCl plates, Nujol mull, cm-1): 1700 (w), 1653 (m), 1497 (s), 1458 (s), 1378 (w), 1249 (s), 1120 (w), 964 (w), 922 (w), 842 (m), 737 (w), 664 (s). Anal. found (calcd. for

C30H47Cl2Cr1N4O2): C, 57.96 (58.32); H, 7.38 (7.67); N, 9.31 (9.07) %. μeff = 4.83

μB (solid state).

128

Chapter 4

4.5 REFERENCES

1. (a) Trofimenko, S. J. Am. Chem. Soc. 1966, 88, 1842. (b) Trofimenko, S. J. Am. Chem. Soc. 1967, 89, 3170. (c) Trofimenko, S. Acc. Chem. Res. 1971, 4, 17. (d) Trofimenko, S. Chem. Rev. 1993, 93, 943.

2. (a) Pullen, E. E.; Rheingold, A. L.; Rabinovich, D. Inorg. Chem. Commun. 1999, 2, 194. (b) Steiner, A.; Stalke, D. Inorg. Chem. 1995, 34, 4846. (c) Sorrell, T. N.; Allen, W. E.; White, P. S. Inorg. Chem. 1995, 34, 952. (d) Szczepura, L. F.; Witham, L. M.; Takeuchi, K. J. Coord. Chem. Rev. 1998, 174, 5.

3. (a) Milione, S.; Grisi, F.; Centore, R.; Tuzi, A. Organometallics 2006, 25, 266. (b) Sánchez-Barba, L. F.; Garcés, A.; Fajardo, M.; Alonso-Moreno, C.; Fernández-Baeza, J.; Otero, A.; Antinolo, A.; Tejeda, J.; Lara-Sánchez, A.; López-Solera, M. I. Organometallics 2007, 26, 6403. (c) Otero, A.; Fernández-Baeza, J.; Antinolo, A.; Lara-Sánchez, A.; Martinez-Caballero, E.; Tejeda, J.; Sánchez-Barba, L. F.; Alonso-Moreno. C.; López-Solera, M. I. Organometallics 2008, 27, 976. (d) Schofield, A. D.; Luna Barros, M.; Cushion, M. G.; Schwarz, A. D.; Mountford, P. Dalton Trans., 2009, 85.

4. (a) Otero, A.; Fernández-Baeza, J.; Antinolo, A.; Carrillo-Hermosilla, F.; Tejeda, J.; Díez-Barra, E.; Lara-Sánchez, A.; Sanchez-Barba, L.; López- Solera, I. Organometallics 2001, 20, 2428. (b) Milione, S.; Montefusco, S.; Cuenca, T.; Grassi, A. Chem. Commun., 2003, 1176. (c) Milione, S.; Bertolasi, V.; Cuenca, T.; Grassi, A.; Organometallics 2005, 24, 4915.

129

Chapter 4

5. Yoshida, T.; Yamamoto, H.; Okada, H.; Murakita, H. US 0035029 (Tosoh Corporation), 2002.

6. (a) Zhang, J.; Braunstein, P. J.; Hor, T. S. A.; Organometallics 2008, 27, 4277. (b) Zhang, J.; Li, A.; Hor, T. S. A. Organometallics 2009, 28, 2935. (c) Zhang, J.; Li, A.; Hor, T. S. A. Dalton Trans., 2009, 9327.

7. Howe, R. G.; Tredget, C. S.; Lawrence, S. C.; Subongkoj, S.; Cowley, A. R.; Mountford, P. Chem. Commun., 2005, 223.

8. (a) Breher, F.; Grunenberg, J.; Lawrence, S. C.; Mountford, P.; Rüegger, H. Angew. Chem. Int. Ed., 2004, 43, 2521. (b) Lawrence, S. C.; Skinner, M. E. G.; Green, J. C.; Mountford, P. Chem. Commun., 2001, 705. (c) Bigmore, H. R.; Meyer, J.; Krummenacher, I.; Rüegger, H.; Clot, E.; Mountford, P.; Breher, F. Chem. Eur. J., 2008, 14, 5918. (d) Bigmore, H. R.; Dubberley, S. R.; Kranenburg, M.; Lawrence, S. C.; Sealey, A. J.; Selby, J. D.; Zuideveld, M.; Cowley, A. R.; Mountford, P. Chem.Commun., 2006, 436.

9. Milione, S.; Bertolasi, V.; Cuenca, T.; Grassi, A.; Organometallics 2005, 24, 4915.

10. (a) Evans, D. J. J. Chem. Soc., 1959, 2003. (b) Schubert, E. M. J. Chem. Ed., 1992, 69, 62.

11. (a) Agapie, T.; Schofer, S. J.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc., 2004, 126, 1304. (b) Elowe, P. R.; McCann, C.; Pringle, P. G.; Spitzmesser, S. K.; Bercaw, J. E. Organometallics 2006, 25, 5255. (c) Schofer, S. J.; Day, M. W.; Henling, L. M.; Labinger, J. A.; Bercaw, J. E.; Organometallics 2006, 25, 2743. (d) Agapie, T.; Labinger, J. A.; Bercaw, J. E.

130

Chapter 4

J. Am. Chem. Soc., 2007, 129, 14281. (e) Agapie, T.; Day, M. W.; Henling, L. M.; Labinger, J. A.; Bercaw, J. E. Organometallics 2006, 25, 2733. (f) McGuinnes, D. S.; Suttil, J. A.; Gardiner, M. G.; Davies, N. W. Organometallics 2008, 27, 4238. (g) Temple, C.; Jabri, A.; Crewdson, P.; Gambarotta, S.; Korobkov, I.; Duchateau, R. Angew. Chem., Int. Ed., 2006, 45, 7050. (h) Jabri, A.; Temple, C.; Crewdson, P.; Gambarotta, S.; Korobkov, I.; Duchateau, R. J. Am. Chem. Soc., 2006, 128, 9238. (i) Temple, C.; Gambarotta, S.; Korobkov, I.; Duchateau, R. Organometallics, 2007, 26, 4598. (j) Jabri, A.; Mason, C. B.; Sim, Y.; Gambarotta, S.; Burchell, T. J.; Duchateau, R. Angew. Chem. Int. Ed., 2008, 47, 9717. (k) Vidyaratna, I.; Nikiforov, G. B.; Gorelski, S. I.; Gambarotta, S.; Duchateau, R.; Korobkov, I. Angew. Chem. Int. Ed., 2009, 48, 6552.

12. (a) Dixon, J. T.; Wasserscheid, P.; McGuinness, D. S.; Hess, F. M.; Maumela, H.; Morgan, D. H.; Bollmann, A. (Sasol Technology (Pty) Ltd) WO 03053890, 2001. (b) McGuinness, D. S.; Wasserscheid, P.; Keim, W.; Hu, C.;Englert, U.; Dixon, J. T.; Grove, C. Chem. Commun. 2003, 334. (c) McGuinness, D. S.; Wasserscheid, P.; Keim, W.; Morgan, D. H.; Dixon, J. T.; Bollmann, A.; Maumela, H.; Hess, F.; Englert, U. J. Am. Chem. Soc. 2003, 125, 5272. (d) McGuinness, D. S.; Wasserscheid, P.; Morgan, D. H.; Dixon, J. T. Organometallics 2005, 24, 552.

13. Wöhl, A.; Müller, W.; Peulecke, N.; Müller, B. S.; Peitz, S.; Heller, D.; Rosenthal, U. J. Mol. Catal. A Chem., 2009, 297, 1.

14. Thomas, W. H.; Wass, D. F. Organometallics 2010, 29, 3676.

15. (a) Crewdson, P.; Gambarotta, S.; Djoman, M. C.; Korobkov, I.; Duchateau, R. Organometallics 2005, 24, 5214. (b) Albahily, K.; Koç, E.; Al-Baldawi, D.;

131

Chapter 4

Savard, D.; Gambarotta, S.; Burchell, T. J.; Duchateau, R. Angew. Chem. Int. Ed. 2008, 47, 5816. (c) McGuinness, D. S.; Rucklidge, A. J.; Tooze, R. P.; Slawin, A. M. Z. Organometallics 2007, 26, 2561.

16. Boujounk, P.; So, J. H.; Inorg. Synth., 1992, 29, 108.

17. Reger, D. L.; Grattan, T. C.; Brown, K. J.; Little, C. A.; Lamba, J. J. S.; Rheingold, A. L.; Sommer, R. D. J. Organomet. Chem., 2000, 607, 120.

18. Chisholm, M. H.; Cotton, F. A.; Frenz, B. A.; Reichert, W. W.; Shive, L. W.; Stults, B. R. J. Am. Chem. Soc., 1976, 98, 4469.

132

Chapter 5

Synthesis and ethylene oligomerisation activity of modified bis(diphenylphosphino)methane chromium catalysts

The content of this chapter has been published. Vadake Kulangara, S.; Mason, C.; Juba, M.; Yang, Y.; Thapa, I.; Korobkov, I.; Gambarotta, S.; Duchateau, R. Organometallics 2012, 31, 6438.

133

Chapter 5

Abstract

Nucleophilic attack of in situ generated bis(diphenylphoshino)methane, (DPPM-) anion to CO2, CyNCO, t-BuNCO and 2,6-(i-Pr)2PhNCO resulted in the formation of novel anionic ligands {[(Ph2P)2CHCO2]Li(THF)2}2 (1), {[(Ph2P)2C=CNH(R)O] Li(OEt)2}2 (R =

Cy (2), R = t-Bu (3)) and [Ph2PCH=P(Ph2)C=N(2,6-i-Pr2C6H3)O]Li(OEt2)2 (4), respectively. Ligand 4, however, showed a connectivity resulting from a “non-classical” type of attack where the P atom acted as a nucleophilc center, thus affording a mixed valent P(III)/P(V) species. The reaction of the in situ generated DPPM- anion with 1 and 0.5 equivalents of CrCl3(THF)3 gave the chelated chromium complexes [HC(PPh2)2]Cr[(-

Cl)2Li(THF)2]2 (5) and [HC(PPh2)2]2Cr(-Cl)2Li(THF)2]·(THF)1.5 (6), respectively. The reaction of ligand 1 with CrCl2(THF)2 afforded the dimeric

[{[(Ph2P)2C(H)CO2]2}Cr(THF)]2 (7), whereas the reaction of 3 with CrCl3(THF)3 resulted in the octahedral complex [(Ph2P)2C(H)C=N(t-Bu)O]CrCl2(THF)2·(THF)0.5·(toluene)0.5 (8).

The complexation of ligand 4 with CrCl3(THF)3 switched the connectivity to “classical” form and afforded the octahedral chromium complex [(Ph2P)C(H)C=N(2,6-i-

Pr2C6H3)O]CrCl2(THF)2·(THF)1.5 (9). The catalytic behaviour of all of these complexes has been assessed under different oligomerisation conditions and it was found that the modification of DPPM framework with cumulenes considerably enhances their catalytic performance compared to catalysts 5 and 6. In any event, a Schultz-Flory distribution of oligomers was obtained. However, the in situ catalytic testing of ligands 2-4 using

Cr(acac)3 as metal precursor and DMAO as cocatalyst, in methylcyclohexane switched the catalytic behaviour to selective formation of 1-hexene and 1-octene (no higher liquid oligomers) along with significant amount of narrowly-dispersed, low molecular weight polyethylene wax.

134

Chapter 5

5.1 INTRODUCTION

Catalytic ethylene oligomerisation is an industrially relevant process.1 The fact that the fractionation of the oligomerisation mixture to extract the most requested α-olefins (1-hexene and 1-octene) is an energy intensive process, motivates the search for selective catalytic systems. As a result of intense research activity in this field, several selective ethylene trimerisation catalysts,2-6 as well as a few ethylene tetramerisation systems have been discovered.7

Among the elements displaying catalytic behaviour for ethylene oligomerisation/polymerisation, chromium occupies a unique position since it provides commercially viable poly- tri- and tetramerisation catalytic systems.6 The commonly accepted mechanism for selective ethylene oligomerisation is a redox metallacycle mechanism,8 where a two-electron oxidative addition of ethylene to chromium forms an expandable chroma-metallacycle. The active oxidation state of chromium responsible for the selectivity has been the center of a debate due to the ambiguities generated by the redox dynamism of this metal.8f,9-13 Different redox couples10-12 have been proposed but recent work has clearly pointed out that Cr(I) is most likely the species responsible for catalyst selective behaviour.13 Therefore, if a selective tri- or tetramerisation system is being sought, the main challenge is to stabilise the highly reactive mono-valent oxidation state. In turn, this relates to a judicious choice of the ancillary ligand framework and donor atom combination. However, designing catalysts which may distinguish between selective tri- and tetramerisation remains a great challenge.

The BP Chemicals’ and Sasol’s PNP chromium complexes, stabilised by neutral RN(PAr2)2 ligands, have marked a milestone in this field. These catalysts oligomerise ethylene with high selectivity towards either 1-hexene or 1-octene, 4,7a depending on the ligand substituents (Ar = 2-OMe-C6H4 and C6H5 respectively).

135

Chapter 5

Further replacements of the heteroatom combinations or modifications of the ligand frameworks14 also produced highly selective ethylene trimerisation catalysts. We also have reported a series of Cr(III) and Cr(II) complexes of NPN15 and NP16 ligands, which showed switchable catalytic behaviour. Last but not least, SK- Energy reported a selective ethylene tetramerisation system based on substituted bis(diphenylphosphino)ethane ligands producing 1-octene with 77% selectivity.7b

Recently, we found that the chromium (III) complexes of [PPh2NR(CH2)3NRPPh2] ligand are also capable of producing 1-octene with record selectivity.7c

From this collection of results, it is clear that the presence of phosphorus donor atoms in the ligand scaffold is an important prerequisite. However, the simpler bis(diphenylphos-phino)methane (DPPM) ligand, while in combination with CrCl3(THF)3 and activated with MMAO only produces a Schultz-Flory distribution of oligomers with very low activity.17 This in spite of its similarity to the PNP system in terms of bite angle, steric encumbrance and donor atoms and also to the relationship with the SK-Energy diphosphine system. Conversely, Wass and coworkers18 reported that activation of bis(diphenylphosphino)methane- stabilized tetracarbonylchromium via one-electron oxidation provide a selective ethylene oligomerisation catalyst producing mainly 1-hexene and 1-octene with a higher selectivity towards 1-octene compared to 1-hexene.

We are currently engaged in a systematic screening of families of diversified ligand systems containing various combinations of donor functions and ligand scaffolds containing the ability for retaining alkyl aluminate residues. This last point is of interest in the ultimate view of obtaining self-activating catalytic systems.9,13,15 For this work, we have examined deprotonated bis(diphenylphosphino)methane as starting point and reacted with CO2 and isocyanate to probe geometrical arrangements not previously examined. It was argued that the phosphorus donor atoms might be central to the stabilisation of the

136

Chapter 5 highly desirable lower oxidation states of chromium. At the same time, hard donors such as oxygen and nitrogen could aid in the retention of aluminium residues.13,14c,15,16 The anionic character of these ligands, in combination with the ability of the system to form zwitterionic species, was also regarded as beneficial for imparting robustness to the complexes and providing sufficient electrophilicity to the metal center. Herein we describe our observations.

5.2 RESULTS AND DISCUSSION

The bis(diphenylphoshino)methane anion [DPPM-] was prepared in situ by treating commercially available DPPM with one equivalent of n-BuLi in THF or

Et2O. Reacting the anion with CO2, CyNCO and t-BuNCO afforded the corresponding lithium salts {[(Ph2P)2CHCO2]Li(THF)2}2 (1),

{[(Ph2P)2C=CNH(Cy)O]Li(OEt)2}2 (2) and {[(Ph2P)2C=CNH(t-Bu)O]Li(OEt)2}2 (3), respectively (Scheme 1). The 31P{1H} NMR spectra of the ligands 1, 2 and 3 showed a single resonance at -16.13 ppm, -12.04 ppm and -13.9 ppm, respectively, in agreement with the presence of two identical P atoms. In these reactions, the carbon atom located between two phosphorus atoms of the DPPM- anion acted as a nucleophile attacking the electrophilic carbon atom of the cumulenes, in the process forming a carbon-carbon bond. Interestingly, this species may exist in a tautomeric form where the hydrogen atom may have been transferred from the C to the N atom (Scheme 1) and the double bond from the C=N to the C=C position. In the cases of 1 and 2 it was possible to obtain single crystals of suitable quality for X-ray diffraction.

137

Chapter 5

Scheme 1. Schematic presentation of the synthesis of ligands 1-3.

The X-ray structure of 1 consists of a dimer (Figure 1) with two ligand moieties bridged by two tetrahedrally coordinated lithium atoms. Two of the coordination sites around each lithium are occupied by the oxygen atoms of two THF molecules [Li(1)-O(3) = 1.947(4) Å, Li(1)-O(4) = 1.992(5) Å] while the other two coordination sites are filled by two oxygen atoms, each from one carboxylate moiety [Li(1)-O(1) = 1.891(4) Å, Li(1)-O(2) = 1.991(4) Å].

138

Chapter 5

Figure 1. ORTEP drawings of {[(Ph2P)2CHCO2]Li(THF)2}2 (1). Selected bond distances (Å) and angles (°) for 1: P(1)-C(1), 1.866(2); P(2)-C(1), 1.865(2); C(1)- C(26), 1.542(3); O(1)-C(26), 1.245(2); O(2)-C(26), 1.239(3); Li(1)-O(3), 1.947(4); Li(1)-O(4), 1.992(5); O(3)-Li(1)-O(4), 100.6(2); O(1)-Li(1)-O(2), 134.6(2); O(1)- Li(1)-O(3), 103.30(19); O(2)-Li(1)-O(3), 102.7(2).

The structure of 2 (Figure 2) is also dimeric with two monomeric units linked together by two tetrahedrally coordinated lithium atoms [O(1)-Li(1)-O(2) = 115.4(2)°, O(2)-Li(1)-O(1a) = 113.1(2)°, O(1)-Li(1)-O(1a) = 93.62(18)°, O(2)- Li(1)-P(1a) = 112.80(18)°]. Each lithium atom is bonded to two bridging oxygen atom of the isocyanate moieties [Li(1)-O(1) = 1.885(4) Å, Li(1)-O(1a) = 1.904(4) Å], a phosphorus atom [Li(1)-P(1a) = 2.569(4) Å] and one oxygen atom of diethyl ether [Li(1)-O(2) = 1.917(5) Å]. The trigonal planar geometry of the carbon bridging the two phosphorus atoms [P(1)-C(1)-P(2) = 119.14(12)°, C(2)-C(1)-P(1) = 115.11(15)°, C(2)-C(1)-P(2) = 125.48(15)°] and the rather short C-C distance [C(1)-C(2) = 1.415(3) Å] also implies some multiple bond character with the isocyanate residue carbon atom.

139

Chapter 5

Figure 2. ORTEP drawings of {[(Ph2P)2CCNH(Cy)O]Li(OEt)2}2 (2). Selected bond distances (Å) and angles (°) for 2: P(1)-C(1), 1.784(2); P(2)-C(1), 1.789(2); C(1)-C(2), 1.415(3); Li(1)-O(1), 1.885(4); Li(1)-O(2), 1.917(5); N(1)-C(2), 1.359(2); Li(1)-P(1), 2.569(4); O(2)-Li(1)-O(1a), 113.1(2); O(1)-Li(1)-O(2), 115.4(2); O(1)-Li(1)-O(1a), 93.62(18); O(2)-Li(1)-P(1a), 112.80(18).

The crystallographic parameters clearly indicate that the hydrogen atom, originally present at the methyne carbon atom of the DPPM- anion, has been transferred to the nitrogen atom of the isocyanate, forming an N-H and a C=C bond. The presence of this as a predominant tautomeric form is further confirmed by the absence of methyne correlation peaks in the 1H{13C}-HMQC experiments (See supporting info) and the presence in the 1H-NMR spectrum of a broad peak at 4.8 ppm without carbon correlation characteristic of the N-H function.

When t-butyl isocyanate was used, the {[(Ph2P)2C=CNH(t-Bu)O]Li(OEt)2}2 (3) lithiated ligand, possessing a similar structure as 2, was isolated and fully characterised by NMR. Instead, when 2,6-di-isopropylphenyl isocyanate was employed, the reaction took a different pathway. The structure of the lithium salt of

[Ph2PCH=P(Ph2)C=N(2,6-i-Pr2C6H3)O]Li(OEt2)2 (4), showed that a “non-classical”

140

Chapter 5 type of nucleophilic attack of the DPPM- phosphorus instead of the carbon atom had occurred (Scheme 2). This resulted in the formation of a phosphorus-carbon bond instead of the expected carbon-carbon bond. In addition, the formal oxidation state of one of the two phosphorus atoms has changed from +3 to +5 as a result of the consequent formation of the C=P double bond within the DPPM- anion residue. Accordingly, the 31P-NMR spectra of the ligand display two equally intense peaks at -8.0 ppm and -24.0 ppm as expected for the presence of two non-equivalent P atoms in different oxidation states.

Scheme 2. Schematic presentation of the synthesis of the non-classical ligand 4.

The crystal structure of 4 (Figure 3) shows the distorted tetrahedral lithium atom surrounded [O(2)-Li(1)-O(3) = 106.75(19)°, O(1)-Li(1)-O(3) = 109.20(2)°, O(1)-Li(1)-O(2) = 113.20(2)°] by one of the two phosphorus [Li(1)-P(1) = 2.703(4) Å] atoms, the oxygen atom [Li(1)-O(1) = 1.813(4) Å] of the isocyanate moiety and two coordinated diethylether molecules [Li(1)-O(2) = 1.948(4) Å, Li(1)-O(3) =1.977(4) Å]. The two fairly short20 carbon-phosphorus distances are in agreement with the presence of significant extent of double bond character and substantial electronic delocalisation within the P-C-P frame [C(1)-P(1) = 1.751(18) Å, C(1)- P(2) = 1.6932(18) Å].

141

Chapter 5

Figure 3. ORTEP drawing of [Ph2PCH=P(Ph2)C=N(2,6-i-Pr2C6H3)O]Li(OEt2)2 (4). Ellipsoids are drawn at 50% probability level and H atoms are omitted for clarity. Selected bond distances (Å) and angles (): P(1)-C(1), 1.7510(18); P(2)-C(1), 1.6932(18); P(2)-C(26), 1.8557(17); C(26)-O(1), 1.261(2); N(1)-C(26), 1.285(2); O(2)-Li(1)-O(3), 106.75(19); O(1)-Li(1)-O(3), 109.20(2); O(1)-Li(1)-O(2), 113.20(2); O(1)-Li(1)-P(1), 91.03(14).

To rationalise the discrepancy of behaviour between the aliphatic and aromatic isocyanates, DFT calculations were performed to understand the origin of the non-classical bonding mode. It was found that the two different bonding modes may be attributed to the different electronic configurations of the two isocyanates (Figure 4) and steric interactions within the final products. The straight O=C=N- functionality of the aromatic isocyanate tends to reside in plane with the rigid aromatic ring forming a conjugated system. On the other hand, in the case of aliphatic isocyanate, the alkyl group is bended away allowing the carbon atom to approach the central carbon of the DPPM- anion. Instead the carbon atom of the aromatic isocyanate can only reach the DPPM- anion through the phosphorus atom to minimise, the steric repulsions.

142

Chapter 5

Figure 4. Different ways of approaching of aliphatic and aromatic isocyanates to the DPPM anion (B3LYP-optimised structures).

To understand the behaviour of this family of ligands, we have preliminary - explored the behaviour of the basic DPPM anion with CrCl3(THF)3 with variable stoichiometric ratios. The reaction with a 1:1 molar ratio resulted in the formation of the chromium complex [HC(PPh2)2]Cr[(-Cl)2Li(THF)2]2 (5), in which the octahedral coordination sphere of chromium contains one bidentate monoanionic chelating DPPM- frame and four chlorine atoms bridging THF solvated lithium atoms (Figure 5). Instead, the reaction of the DPPM- anion with 0.5 equivalents of

CrCl3(THF)3 afforded an octahedral chromium complex [HC(PPh2)2]2Cr(-

Cl)2Li(THF)2]·(THF)1.5 (6), whose six coordination sites are occupied by two bidentate-monoanionic chelating DPPM frames and two chlorine atoms bridging THF solvated lithium atoms.

143

Chapter 5

Figure 5. ORTEP drawing of [HC(PPh2)2]Cr[(-Cl)2Li(THF)2]2 (5, left) and

[HC(PPh2)2]2Cr(-Cl)2Li(THF)2]·(THF)1.5 (6, right). Ellipsoids are drawn at 50% probability level and H atoms are omitted for clarity. Selected bond distances (Å) and angles (°) for 5: Cr(1)-Cl(1), 2.3500(6); Cr(1)-Cl(2), 2.3959(6); Cr(1)-P(1), 2.4560(6), Cr(1)-P(1a), 2.4560(6); P(1)-C(1), 1.7236, Cl(1)-Cr(1)-Cl(1a), 177.97(4); Cl(1)-Cr(1)-Cl(2a), 90.76(2); Cl(1)-Cr(1)-Cl(2), 90.58(2); Cl(1)-Cr(1)- P(1), 89.14(2); Cl(1)-Cr(1)-P(1a), 89.14(2); P(1a)-Cr(1)-P(1), 67.13(3). For 6: Cr(1)-Cl(1), 2.3634(12); Cr(1)-Cl(2), 2.3741(12); Cr(1)-P(1), 2.4626(12); Cr(1)- P(2), 2.4978(12); Cr(1)-P(3), 2.4700(12); Cr(1)-P(4), 2.5149(12); Cl(1)-Cr(1)- Cl(2), 90.31(4); Cl(1)-Cr(1)-P(1), 95.64(4); Cl(1)-Cr(1)-P(3), 93.10(4); Cl(1)- Cr(1)-P(4), 92.52(4).

The X-ray structures of the two complexes showed similarly distorted octahedral geometries around chromium, with four chlorine atoms bridging between two solvated lithium centers in the case of complex 5 [Cr(1)-Cl(1) = 2.3500(6) Å, Cr(1)-Cl(2) = 2.3959(6) Å]. In the case of 6, only two chlorine atoms bridge between a THF solvated lithium atom [Cr(1)-Cl(1) = 2.3634(12) Å, Cr(1)- Cl(2) = 2.3741(12) Å]. The Cr(1)-Li(1) distance in 5 [Cr(1)-Li(1) = 3.206(4) Å] was found to be slightly shorter than the corresponding bond distance in 6 [Cr(1)- Li(1) = 3.2989(1) Å]. Also the chromium to phosphorus distances in the case of 5

144

Chapter 5

[Cr(1)-P(1) = 2.4506(6) Å, Cr(1)- P(1a) = 2.4506(6) Å], were shorter than the corresponding bond distance in 6 [Cr(1)-P(1) = 2.4626(12) Å, Cr(1)-P(3) = 2.4700(12) Å, Cr(1)-P(2) = 2.4978(12) Å, Cr(1)-P(4) = 2.5149(12) Å].

The complexation of 1 with CrCl2(THF)2 resulted in the formation of the dimeric chromium complex {[((Ph2P)2CHCO2)2]Cr(THF)}2 (7).

Figure 6. ORTEP drawing of {[((Ph2P)2CHCO2)2]Cr(THF)}2 (7). Ellipsoids are drawn at 50% probability level and H atoms are omitted for clarity. Selected bond distances (Å) and angles (°) for complex 7: Cr(1)-Cr(1a), 2.3337(8); Cr(1)-O(3), 1.9985(18); Cr(1)-O(1), 2.016(2); Cr(1)-O(5), 2.247(2); O(3)-Cr(1)-O(5), 89.40(8); O(1)-Cr(1)-O(5), 91.78(8); O(3)-Cr(1)-O(1), 93.59(8); O(3)-Cr(1)-O(2), 86.99(8).

The molecular structure of 7 (Figure 6) showed a dinuclear chromium complex with a rather short Cr-Cr bond distance [Cr-Cr = 2.337(8) Å] in the characteristic arrangement of the axially-coordinated, paddlewheel complexes of divalent chromium. Four of the coordination sites around each chromium atom are

145

Chapter 5 occupied by four oxygen atoms of the carboxylate moiety [Cr(1)-O(3) = 1.9985(18) Å, Cr(1)-O(1) = 2.016(2) Å] while the fifth coordination site is occupied by oxygen atoms of the THF molecule [Cr(1)-O(5) = 2.247(2) Å]. Complex 8 shows the reduced paramagnetism typical of the paddlewheel divalent chromium dimers [µeff = 3.91 µB per dimer].

Attempts to crystallise complexes of CrCl3(THF)3 with 2, invariably led to powdery products. Conversely, reaction of 3 with CrCl3(THF)3 afforded

[(Ph2P)2C(H)C=N(t-Bu)O]CrCl2(THF)2·(THF)0.5·(toluene)0.5 (8) in crystalline form. The molecular structure of 8 (Figure 7) showed the octahedral coordination geometry of the chromium atom as defined by one phosphorus [Cr(1)-P(1) = 2.4280(12) Å], one oxygen [Cr(1)-O(1) = 1.937(3) Å], two chlorine atoms [(Cr(1)-Cl(1) = 2.2991(12) Å, Cr(1)-Cl(2) = 2.2998(12) Å] and two molecules of THF [Cr(1)-O(2) = 2.069(3) Å, Cr(1)-O(3) = 2.079(3) Å]. The long bond distance of the C-C bond connecting the two DPPM- and isocyanate residues as well as the pyramidality of the PCP’s C atom indicate that the ligand is present in the tautomeric form containing the C=N double bond and the hydrogen atom on the central carbon of the DPPM- residue. Unfortunately, the isolation of single crystals of the corresponding Cr(II) complex was not successful due to the poor solubility of this blue microcrystalline material including in highly polar solvent such as dichloromethane and acetonitrile.

146

Chapter 5

Figure 7. ORTEP drawing of [(Ph2P)2C(H)C=N(t-

Bu)O]CrCl2(THF)2·(THF)0.5·(toluene)0.5 (8). Ellipsoids are drawn at 50% probability level and H atoms are omitted for clarity. Selected bond distances (Å) and angles (°) for complex 8: Cr(1)-Cl(1), 2.2991(12); Cr(1)-Cl(2), 2.2998(12); Cr(1)-P(1), 2.4280(12); Cr(1)-O(1), 1.937(3); Cr(1)-O(2), 2.069(3); Cr(1)-O(3), 2.079(3); C(1)-C(26), 1.542(5); O(1)-Cr(1)-Cl(2), 91.47(8); O(3)-Cr(1)-Cl(2), 89.71(9); O(1)-Cr(1)-Cl(2), 91.47(8); Cl(2)-Cr(1)-P(1), 92.87(4); Cl(2)-Cr(1)- Cl(1), 91.67(5); O(2)-Cr(1)-Cl(2), 174.17(9).

Complexation of ligand 4, which resulted from a “non-classical” isocyanate/diphosphine anion reaction, with CrCl3(THF)3 resulted in the formation of complex [(Ph2P)C(H)C=N(2,6-i-Pr2C6H3)O]CrCl2(THF)2·(THF)1.5 (9), where the ligand has apparently switched the connectivity towards the classical type of isocyanate-diphosphino anion aggregation (Scheme 3). The resulting octahedral chromium complex is very similar to 8. Accordingly, the phosphorus oxidation state has been reduced from +5 to +3 through an internal redox transformation.

147

Chapter 5

Scheme 3. Synthesis of [(Ph2P)C(H)C=N(2,6-i-Pr2C6H3)O]CrCl2(THF)2·(THF)1.5 (9). The crystal structure of 9 (Figure 8) shows an octahedrally coordinated chromium atom bonded to one oxygen atom [Cr(1)-O(1) = 1.939(2) Å, O(1)-Cr(1)- Cl(1) = 90.41(7)°] and one phosphorus atom [Cr(1)-P(1) = 2.4334(9) Å, Cl(1)- Cr(1)-P(1) = 91.44(3)°] of the ligand, two chlorine [Cr(1)-Cl(2) = 2.2980(10) Å, Cr(1)-Cl(1) = 2.3016(10) Å, Cl(2)-Cr(1)-Cl(1) = 92.09(4)°] and two THF molecules [O(2)-Cr(1)-Cl(1) = 174.60(8)°]. The change of the phosphorus oxidation state from +5 to +3 also reflects in the lengthening of P-C bond distances [P(1)-C(1) = 1.844(3) Å, P(2)-C(1) = 1.870(3) Å] compared to the corresponding bond distances in the ligand 4 [P(1)-C(1) = 1.751(18) Å, P(2)-C(1) = 1.693(2) Å]. Also the C-C bond distance of the link between the isocyanate and diphosphine residues of complex 9 [C(1)-C(26) = 1.539(4) Å] compares well to that of 9 [C(1)- C(26) = 1.542(5) Å]. The pyramidality of the PCP’s central C atom also indicates the presence of the H atom in that position and consequent C=N double bond character.

148

Chapter 5

Figure 8. ORTEP drawing of [(Ph2P)C(H)C=N(2,6-i-

Pr2C6H3)O]CrCl2(THF)2·(THF)1.5 (9). Ellipsoids are drawn at 50% probability level and H atoms are omitted for clarity. Selected bond distances (Å) and angles (degree) for complex 9: P(1)-C(1), 1.844(3); P(2)-C(1), 1.870(3); C(1)-C(26), 1.539(4); Cr(1)-Cl(2), 2.2980(10); Cr(1)-Cl(1), 2.3016(10); Cr(1)-O(1), 1.939(2); Cr(1)-P(1), 2.4334(9); O(1)-Cr(1)-Cl(1), 90.41(7); Cl(1)-Cr(1)-P(1), 91.44(3); O(3)-Cr(1)-Cl(1), 90.52(9); Cl(2)-Cr(1)-Cl(1), 92.09(4); O(2)-Cr(1)-Cl(1), 174.60(8).

All the chromium complexes were tested for ethylene oligomerisation activity under different reaction conditions. The DPPM-chromium complexes 5 and 6, upon activation with MAO in toluene, showed moderate catalytic activity, producing a S-F distribution of oligomers (Table 1). Complex 6, which contains two monoanionic DPPM ligands, displays a lower catalytic activity [663 g/(mmolCrh)] compared to 5 [1180 g/(mmolCrh)] ligating one DPPM fragment. However, whereas 5 produced a considerable amount of waxy low molecular weight PE, 6 was found to be a polymer-free oligomerisation catalyst under identical reaction condition. The activity of catalysts 5 and 6 was found to be higher than the activity of the in situ generated DPPM–Cr(III) complex reported by

149

Chapter 5

Overett et al17, the poor catalytic performance of which was attributed to the deprotonation of highly acidic bridging methylene proton by metal alkyl species.21 Both catalysts 5 and 6 were found to be inactive upon activating with different activators such as TEA, DMAO or MMAO either in toluene or in methylcylcohexane.

The modification of the anionic DPPM framework with isocyanates or CO2 considerably enhanced the catalytic performance of the complexes (Table 1-3). Activation of 8 with MAO in toluene resulted in a highly active ethylene oligomerisation catalyst, again producing a Schultz-Flory distribution of oligomers with a maximum activity of 29,820 g/(mmolCrh). Interestingly, not even traces of polymer were detected. The activity and the selectivity of this system was 2 comparable to the previously reported NPN system, [(t-Bu)NP(Ph)2N(t-Bu)]Cr(μ - 22 Cl)2Li(THF)2, which also showed polymer-free oligomerisation behaviour. Varying the Al:Cr ratio affected the catalytic activity with a maximum activity at a low Al:Cr ratio. Increasing the Al:Cr ratio had a detrimental effect on the catalytic performance and resulted in the formation of a considerable amount of low molecular weight PE. The lower oligomerisation activity at higher Al:Cr ratio might be due to the poisoning effect of free TMA present in MAO. Indeed, the addition of 250 equivalents of TMA along with MAO switches the catalytic behaviour to polymerisation, producing polyethylene with a broad molecular weight distribution. Use of DMAO as cocatalyst in methylcyclohexane or in toluene resulted in the formation of different types of PE. In methylcyclohexane, 6 UHMWPE with narrow molecular weight distribution (PDI = 3.2, Mw = 4.1 × 10 g/mol) was obtained while the polymer obtained in toluene showed the characteristic broad molecular weight waxy PE. However, the addition of 250 equivalents of TEA along with DMAO in methylcyclohexane resulted in the formation of a S-F distribution of oligomers along with waxy low molecular weight PE, implying that yet another different active species is generated upon the

150

Chapter 5 addition of TEA. Upon activation with MAO (Al:Cr = 1000), 9 also showed high oligomerisation activity (21,480 g/(mmolCrh)). The catalyst was inactive upon activation with MMAO in toluene. However, with DMAO as cocatalyst in methylcyclohexane, 9 showed moderate oligomerisation activity with a slight excess of 1-hexene. This could be due to the formation of monovalent chromium in parallel to the divalent species, as a result of the redox dynamism occurring in presence of DMAO.

151

Chapter 5

Table 1. Ethylene oligomerisation results for complexes 5-9.

Cat ID Cocatalyst Oligomers PE Activity C6 C8 ≥C10  (loading in µmol) (equiv) (mL) (g) (g/(mmolCrh)) (mol%) (mol%) (mol%) 5 (30) MAO (500) 21.7 3.0 1180 48.2 23.6 28.2 0.58 6 (30) MAO (500) 14.3 0 663 49.8 26.5 23.7 0.55 8 (1) MAO (300) 21.4 Trace 29820 47.4 28.8 23.8 0.60 8 (1) MAO (500) 17 Trace 24000 47.6 29.1 23.1 0.63 8 (1) MAO (1000) 7.2 4.2 18340 46.8 28.7 24.5 0.69 MAO (1000) 8 (1) - 6 12000 - - - - + TMA (250) 8 (1) DMAO (500) - 1.2 2400 - - - - a8 (10) DMAO (500) - 2.2 440 - - - - DMAO (500) a8 (10) 6.2 2.8 1410 50.7 29.4 19.8 0.65 + TEAL (250) 8 (10) MMAO (500) - 0.7 140 - - - - 9 (1) MAO (500) 6.4 0.5 9520 42.9 31.4 25.2 0.68 9 (1) MAO (1000) 14.6 0.8 21480 43.7 30.8 25.3 0.65 9 (10) MMAO(500) - - 0 a9 (10) DMAO 18.6 0.2 2596 53.5 30 15.63 0.63 Oligomerisation conditions: Solvent = 100 mL of toluene, ethylene pressure = 30 bar, reaction time = 30 min, reaction temperature = 60 °C. a) Solvent is methylcyclohexane.

152

Chapter 5

Since the isolations of well-defined Cr(III) complexes of ligand 1 and 2 were not successful, these ligands were tested in situ by mixing with CrCl3(THF)3 and the results are summarised in Table 2.

Table 2. Ethylene oligomerisation results of in-situ catalyst testing, 1-2/CrCl3(THF)3.

Cat ID Cocatalyst Oligomers PE Activity C6 C8 ≥C10 

(loading in µmol) (equiv) (mL) (g) (g/(mmolCrh)) (mol%) (mol%) (mol%) 1 (5) MAO(500) 12 11 7,808 47.6 29.1 23.2 0.59 1 (5) MAO (1000) 18 9 8,680 48.2 28.7 23.1 0.61 1 (5) MMAO 7 23 11,160 44.5 24.9 30.2 0.59 (500) a1 (10) MMAO ------(500) 2 (5) MAO(500) 12 0.4 3,560 49.1 31.7 18.9 0.62 2 (5) MAO (1000) 30 0.7 8,800 53.8 30.2 15.2 0.55 2 (5) MMAO 6 - 1,704 51.9 31.9 16.1 0.53 (500) a2 (10) MMAO ------

Reaction conditions: CrCl3(THF)3 and 1 or 2 were mixed in toluene and the in situ prepared complexes were injected into the reactor. Solvent = 100 mL of toluene, ethylene pressure = 30 bar, reaction time = 30 min, reaction temperature = 60 °C. a) Solvent is methylcyclohexane.

153

Chapter 5

The in situ generated chromium complex of ligand 1 showed, upon activation with MAO (Al/Cr = 1000) in toluene, an activity of 8680 g/(mmolCrh), producing a S-F distribution of oligomers along with considerable amount of PE. The SEC analysis of the PE showed that it is low molecular weight wax with a narrow molecular weight distribution (PDI =

1.7, Mw = 2050 g/mol). When the CrCl3(THF)3/1 system was tested using MMAO as a cocatalyst in toluene, the amount of wax considerably increased

(PDI = 1.7, Mw = 2430 g/mol) along with small amounts of oligomers. However, the catalyst was completely inactive when the reaction was carried out in methylcyclohexane using MMAO as cocatalyst. The in situ testing of ligand 2 with CrCl3(THF)3 also produced Schulz-Flory distribution of oligomers at high Al:Cr ratio, with an activity of 8800 g/(mmolCrh), but with less PE formation compared to CrCl3(THF)3/1.

The in situ catalytic testing of ligand 1-4 using Cr(acac)3 as metal precursor and DMAO as cocatalyst in methylcyclohexane showed some interesting differences (Table 3). The ligand 1 at an Al/Cr ratio of 500, produced only low molecular weight wax with slightly broad polydispersity index (PDI = 5.1, Mw = 12000 g/mol). The IR analysis of the resulting PE showed that it is vinyl terminated. The addition of 100 equivalent of TEA along with DMAO only increased the amount of PE. Interestingly, whreas 8 and 9 produced a S-F distribution of oligomers, when activated with DMAO

(+ TEAL) the combination of Cr(acac)3 and ligand 2-4 produced 1-hexene/1- octene selective catalysts (no higher liquid oligomers), albeit with considerable amount of low molecular weight waxes. The formation of higher amount of wax compared to hexene and octene might be due to the predominant stabilisation of divalent chromium compared to the monovalent oxidation state. Ligand 3 showed slighly higher oliogomerisation activity.

154

Chapter 5

The addition of TEA as an external alkylating agent/reducing agent, did not result in any considerable improvement in terms of oligomerisation activity and selectivity, however it only helped to increasse the amount of wax in all the cases.

155

Chapter 5

Table 3. Effect of catalyst precursor and cocatalyst on selectivity.

Ligand Cocatalyst Oligomers PE Activity C6 C8 ≥C10

(loading in µmol) (equiv) (mL) (g) (g/mmolCrh)) (mol%) (mol%) (mol%) 1 (1) DMAO(500) - 7.8 15,600 0 0 0 1 (1) DMAO (500) - 9.4 18,800 0 0 0 100 TEA 2 (5) DMAO (500) 1.6 6.2 3,120 52.8 47.2 0 2 (5) DMAO 1.6 9.4 4,400 51.9 48.1 - (500)/100 TEA 3 (5) DMAO(500) 3.2 1.2 1,760 61.8 38.9 0 3 (5) DMAO 3.2 2.8 2,400 61.7 38.3 0 (500)/100 TEA 4 (5) DMAO (500) 1.6 1.9 1,400 50.5 49.5 0 4 (10) DMAO 1.6 2.4 1,600 48.8 51.2 0 (500)/100 TEA

Oligomerisation conditions: Solvent = 100 mL of Methylcyclohexane, ethylene pressure = 30 bar, reaction time = 30 min, reaction temperature = 60 °C. Cr(acac)3/ligand = 1/1.

156

Chapter 5

5.3 CONCLUDING REMARKS

Novel DPPM-based mono anionic ligands were prepared by the reaction of an in situ-generated DPPM- anion with different cumulenes and the ligands were characterised by single crystal XRD analysis and NMR techniques. The use of aliphatic isocyanates as cumulenes resulted in an unprecedented proton transfer reaction, whereas an aromatic isocyanate resulted in the formation of mixed valent P(III)/P(V) ligands. However, this arrangement disappears upon complexation with the chromium atom. Well-defined chromium complexes of these novel ligands were synthesised and characterised by single crystal XRD analysis and ethylene oligomerisation studies showed interesting catalytic behaviour. In particular, catalyst [[(Ph2P)2C(H)C=N(t-Bu)O]CrCl2(THF)2·(THF)0.5·(toluene)0.5 (8) exhibited polymer-free ethylene oligomerisation behaviour along with very high activity, which is highly important in an industrial environment. Moreover, we demonstrated that the modification of the DPPM frame work with different cumulenes (additional donor atoms) significantly enhances the catalytic performance, without considerably changing the selectivity of the oligomerisation.

The use of Cr(acac)3 as metal precursor and DMAO as cocatalyst in methylcylohexane resulted in the formation of hexene and octene (no higher liquid oligomers) along with low molecular weight polyethylene wax.

5.4 EXPERIMENTAL SECTION

CyNCO, t-BuNCO and 2,6-(i-Pr)2PhNCO, lithium and aluminium reagents were

157

Chapter 5 purchased from Sigma Aldrich and used as such. Unless otherwise stated, the 1H, 13C{1H}, and 31P{1H} NMR spectra were recorded on a Bruker 300 spectrometer at

300.13, 75.47, and 121.49 MHz, respectively at 298 K. DMSO-d6, dried with 4 Å molecular sieves and stored under nitrogen was used for all the NMR measurements. All chemical shifts are given in ppm and referenced to SiMe4. Elemental analyses were carried out with a Perkin-Elmer 2400 CHN analyser. Data for X-ray structure determination were obtained with a Bruker diffractometer equipped with a 1K Smart CCD area detector. Molecular weights and molecular weight distributions of the polyethylenes were determined by means of high temperature SEC on a PL-GPC210, equipped with refractive index and viscosity detectors and a 3  PLgel 10 µm MIXED-B column set, at 160 °C with 1,2,4- trichlorobenzene as solvent. BHT and Irganox have been used as antioxidants. The molecular weights of the polyethylenes produced were referenced to linear polyethylene standards. Oligomerisation results were analysed by 1HNMR spectroscopy for activity and GC-MS for reaction mixture composition. Gas chromatography of oligomerisation products was conducted on a Varian 450-GC equipped with an auto sampler.

General Oligomerisation Procedure. All oligomerisation were performed in a 250 mL Büchi reactor. The reactor was dried in an oven at 120 °C for two hours prior to each run and then evacuated for half an hour and rinsed with argon three times. After that, the reactor was loaded with toluene and the desired amount of cocatalyst. After the solution was stirred for 10 minutes, it was saturated with ethylene. The reactor was temporarily depressurised to allow injection of the catalyst solution into the reactor under argon flow, after which the reactor was immediately repressurised to the desired set point. The temperature of the reactor was kept as constant as possible by a thermostat bath. After 30 minutes reaction time and cooling to 0 °C, the reaction mixture was depressurised and a mixture of ethanol and diluted hydrochloric acid were subsequently injected to quench the

158

Chapter 5 reaction. The polymer was separated by filtration and dried at 60 °C for 18 hours under reduced pressure prior to molecular weight determination.

Computational Method. Geometry optimisations were performed using the Gaussian 09 program package without symmetry constraints, using the unrestricted B3LYP hybrid functional and double-ζ basis set 6-31G (d, p). Frequency analyses were carried out on the resulting geometries to verify the nature of stationary points (no imaginary frequencies).

Synthesis of {[(Ph2P)2CHCO2]Li(THF)2}2 (1). A solution of bis(diphenylphosphine)methane (13.4 mmol, 5.1 g) in THF (130 mL) was treated with BuLi (13.4 mmol, 2.5 M, 5.3 mL). The resulting yellow solution was stirred for three hours. The solution was exposed to CO2 (bubbling through) for a period of three hours and then kept at -35 C for two days, thus affording colourless crystals of 1. Yield (3.2 g, 5.5 mmol, 41.1 %). 1H NMR: δ = 7.08-7.55 13 (m, 20H; ArH), 4.00 (br, 1H; P2CH). C{H} NMR: δ = 48.05 (PCP), 128.3, 133.8, 31 139.7 (ArC, Ph), 170.2 ( t-C, CO2). P{H} NMR: δ = -16.13.

Synthesis of {[(Ph2P)2CCNH(Cy)O]Li(OEt)2}2 (2). A solution of bis(diphenylphosphine)-methane (10 mmol, 3.8 g) in diethylether (100 mL) was treated with BuLi (10.1 mmol, 2.5 M, 4.1 mL) at room temperature. After the addition was completed, the solution was allowed to stir for three hours. Cyclohexyl isocyanate (10 mmol, 1.3 mL) was added to the yellow solution and the mixture was stirred for three hours, and subsequently kept at -35 C for two days, resulting in the formation of colourless crystals of 2. Yield (5.4 g, 4.6 mmol, 46.0 %). 1H NMR: δ = 0.5-1.3 (m, 10H; CyH ), 3.3 (m, 1H; ipsoH Cy), 4.8( br, 13 1H; NH), 7.1-7.3 (m, 20H; ArH). C{H} NMR: δ = 25.9 (Cy CH2) 47.5 (ipsoC Cy), 132, 126, 125 (ArC), 144.8(PCP), 172.8 (t-C-NCO). 31P{H} NMR: δ = - 12.04.

159

Chapter 5

Synthesis of {[(Ph2P)2CCNH(t-Bu)O]Li(OEt)2}2 (3). A solution of bis(diphenylphosphine)methane (12.8 mmol, 4.9 g) in THF (100 mL) was treated with BuLi (13.5 mmol, 2.5 M, 5.6 mL). The resulting yellow solution was allowed to stir for three hours. Tertiary butyl isocyanate (12.8 mmol, 1.3 g) was then added, resulting in the immediate precipitation of a white powder. The precipitate was washed three times with cold hexanes and analysed after drying in vacuum for 12 hours. Yield (4.8 g, 9.8 mmol, 76.5 %). 1H NMR: δ = 0.796 (s, 9H; 13 (CH3)3C), 3.79 (br, 1H; NH), 7.1-7.6 (m, 20H; ArH). C{H} NMR: δ = 29.16

(CH3), 49.4 (CCH3), 126.85, 127.34,132.58 (ArC, Ph), 145.23 (PCP), 174.0 (NCO). 31P{1H} NMR: δ = -13.9.

Synthesis of [Ph2PCH=P(Ph2)C=N(2,6-i-Pr2C6H3)O]Li(OEt2)2 (4). A solution of bis(diphenylphosphine)methane (10 mmol, 3.8 g) in diethylether (100 mL) was treated with BuLi (10.1 mmol, 2.5 M, 4.1 mL). After the addition, the suspension was stirred for one hour resulting in a clear yellow solution and 2,6-di- isopropylphenyl isocyanate (10 mmol, 2.1 g) was added after which the mixture was stirred for another five hours. The resulting yellow solution was then kept at - 35 C for two days, affording colourless crystals of 4. Yield: (3.4 g, 4.6 mmol, 46.0 1 %). H NMR: δ = 0.61 (d, 12H; CH3), 2.63 (m, 2H; ipso-CH), 4.26 (t, 1H; PCHP), 13 6.5-7.7 (m, 23H; ArH). C{H} NMR: δ = 24.2 (i-pr CH3), 26.89 (i-pr CHCH3), 46.01(PCP), 164.9 (t-C, ArNCO), 151.2, 140.6, 134.4,127.8 (ArCNCO), 127.8, 127.4, 133.8 (PPh). 31P{H} NMR: δ = -8, -24.

Synthesis of [HC(PPh2)2]Cr[(-Cl)2Li(THF)2]2 (5). A solution of bis(diphenylphosphine)methane (1.0 mmol, 382 mg) in THF (6 mL) was treated with BuLi (1.1 mmol, 2.5 M, 0.4 mL) and the resulting yellow solution was stirred for 3 hours at room temperature. A suspension of CrCl3(THF)3 (1 mmol, 375 mg) in THF was then added, resulting in a green mixture which was stirred for 2 hours. After centrifugation, the resulting green solution was then kept

160

Chapter 5 at room temperature for 4 days, forming pale green crystals of 5. Yield (580 mg,

0.66 mmol, 66.0 %). Anal. Calcd. (found) for C41H53Cl4CrLi2O4P2: C 55.9 (56.01),

H 6.02 (6.10). [µeffec = 3.92 µB].

Synthesis of [HC(PPh2)2]2Cr(-Cl)2Li(THF)2]·(THF)1.5 (6). A solution of bis(diphenylphosphine)methane (1.0 mmol, 382 mg) in THF (6 mL) was treated with n-BuLi (1.1 mmol, 2.5 M, 0.4 mL) and the resulting yellow solution was stirred for 3 hours at room temperature. A THF suspension of

CrCl3(THF)3 (0.5 mmol, 190 mg) was then added to this solution, affording a brown mixture which was stirred for 8 hours. The solvent was removed in vacuum and the residue was redissolved in toluene. After centrifugation, the supernatant liquid was evaporated. The dark brown residue was redissolved in THF and kept at -35 C for seven days, affording brown crystals of 7. Yield (280 mg, 0.24 mmol,

24%). Anal. Calcd. (found) for C64H70Cl2CrLiO3.50P4: C 66.82 (66.90), H 6.09

(6.10).[µeff = 3.82 µB].

Synthesis of {[((Ph2P)2CHCO2)2]Cr(THF)}2 (7). A colourless solution of the ligand 1 (1157 mg, 1.0 mmol) in THF was mixed with a suspension of CrCl2(THF)2 (266 mg, 1 mmol) in THF and the resulting pale blue solution was stirred for 4 hours. The solvent was then removed in vacuum, the residue redissolved in fresh THF (4 mL) and the resulting mixture centrifuged. After centrifugation, the supernatant liquid was separated and concentrated, and kept at -35 C thus affording pale blue crystals of 7. Yield (837 mg, 0.35 mmol,

35.0 %). Anal. Calcd. (found) for C136H148Cr2O16P8: C 68.28 (68.21), 6.19 (6.21).

[µeffec= 3.91 µB].

161

Chapter 5

Synthesis of [(Ph2P)2C(H)C=N(t-Bu)O]CrCl2(THF)2·(THF)0.5·(toluene)0.5 (8). A white suspension of 3 (489 mg, 1.0 mmol) in THF was mixed with a suspension of CrCl3(THF)3 (375 mg, 1 mmol) in THF and the resulting pale green solution was stirred for 1 hour. The solvent was then removed in vacuum and the residue was suspended in toluene. After centrifugation, the supernatant liquid was separated, concentrated and kept at room temperature for seven days, thus affording of pale green crystals of 9 (400 mg, 0.48 mmol, 48%). Anal. Calcd.

(found) for C43.50H53Cl2Cr NO3.50P2: C 62.83 (62.86), H 6.38 (6.29), N 1.68 (1.70).

[µeffec = 3.89 µB].

Synthesis of [(Ph2P)C(H)C=N(2,6-i-Pr2C6H3)O]CrCl2(THF)2·(THF)1.5 (9).

A colourless solution of 4 (741 mg, 1 mmol) in THF was treated with CrCl3(THF)3

(375 mg, 1 mmol) and the resulting green mixture was stirred for 4 hours. The solvent was removed in vacuum and the residue was suspended in toluene. After centrifugation, the supernatant liquid was evaporated and the green solid redissolved in THF. Green crystals of 10 were obtained by keeping the solution at - 30 C for a week (560 mg, 0.58 mmol, 58 %). Anal. Calcd. (found) for

C52H66Cl2CrNO4.50P2: C 64.89 (64.72), H 6.90 (6.88), N 1.45 (1.49). [µeff = 3.82

µB].

162

Chapter 5

5.5 REFERENCES

1. Vogt, D. Applied Homogeneous Catalysis with Organometallic Compounds; Cornils, B., Herrmann, W. A., Eds.; Wiley-VCH: Weinheim, Germany, 1996; Vol 1, P 245.

2. (a) Reagan, W. K. EP 0417477 (Phillips Petroleum Company), 1991. (b) Knudsen, R. D.; Freeman, J. W.; Lashier, M. E. US 5563312 (Phillips Petroleum Company), 1996. (c) Freeman, J. W.; Buster, J. L.; Knudsen, R. D. US 5856257 (Phillips Petroleum Company), 1999. (d) Reagan, W. K.; Freeman, J. W.; Conroy, B. K.; Pettijohn, T. M.; Benham, E. A. EP 0608447 (Phillips Petroleum Company), 1994.

3. (a) McGuinness, D. S.; Wasserscheid, P.; Keim, W.; Morgan, D. H.; Dixon, J. T.; Bollmann, A.; Maumela, H.; Hess, F. M.; Englert, U. J. Am. Chem. Soc. 2003, 125, 5272. (b) Dixon, J. T.; Wasserscheid, P.; McGuinness, D. S.; Hess, F. M.; Maumela, H.; Morgan, D. H.; Bollmann, A. (Sasol Technology (Pty) Ltd) WO 03053890, 2001. (c) McGuinness, D. S.; Wasserscheid, P.; Keim, W.; Hu, C.; Englert, U.; Dixon, J. T.; Grove, C. Chem. Commun. 2003, 334.

4. (a) Wass, D. F. (BP Chemicals Ltd) WO 02/04119, 2002; (b) Carter, A.; Cohen, S. A.; Cooley, N. A.; Murphy, A.; Scutt, J.; Wass, D. F. Chem. Commun. 2002, 858.

5. Zhang, J.; Braunstein, P.; Hor, A. T. S. Organometallics 2008, 27, 4277.

6. (a) Agapie, T. Coord. Chem. Rev. 2010, 255, 861. (b) McGuinness, D. S. Chem. Rev. 2011, 111, 2321.

163

Chapter 5

8. (a) Manyik, R. M.; Walker, W. E.; Wilson, T. P. J. Catal. 1977, 47, 197. (b) Briggs, J. R. Chem. Commun. 1989, 11, 674. (c) McDermott, J. X.; White, J. F.; Whitesides, G. M. J. Am. Chem. Soc. 1973, 95, 4451; (d) McDermott, J. X.; White, J. F.; Whitesides, G. M. J. Am. Chem. Soc. 1976, 98, 6521. (e) Agapie, T.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2007, 129, 14281; (f) Agapie, T.; Schofer, S. J.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2004, 126, 1304. (g) Tomov, A. K.; Chirinos, J. J.; Long, R. J.; Gibson, V. C.; Elsegood, M. R. J. J. Am. Chem. Soc. 2006, 128, 7704. (h) McGuinness, D. S.; Suttil, J. A.; Gardiner, M. G.; Davies, N. W. Organometallics 2008, 27, 4238. (i) Rensburg, W. J.; Grove, C.; Steynberg, J. P.; Stark, K. B.; Huyser, J. J.; Steynberg, P. J. Organometallics 2004, 23, 1207. (j) Blom, B.; Klatt, G.; Fletcher, J. C. Q.; Moss, J. R. Inorg. Chim. Acta 2007, 360, 2890. (k) Bhaduri, S.; Mukhopadhyay, S.; Kulkarni, S. A. J. Organomet. Chem. 2009, 694, 1297.

9. (a) Temple, C.; Jabri, A.; Crewdson, P.; Gambarotta, S.; Korobkov, I.; Duchateau, R. Angew. Chem. 2006, 118, 7208. (b) Jabri, A.; Temple, C.; Crewdson, P.; Gambarotta, S.; Korobkov, I.; Duchateau, R. J. Am. Chem. Soc. 2006, 128, 9238; (c) Jabri, A.; Crewdson, P.; Gambarotta, S.; Korobkov, I.;

164

Chapter 5

Duchateau, R. Organometallics 2006, 25, 715. (d) Zhang, J.; Li, A.; Hor, T. S. A. Organometallics 2009, 28, 2935.

10. (a) Kohn. R. D.; Haufe, M.; Mihan, S.; Lilge, D. Chem. Commun. 2000 1927. (b) Fang, Y.; Liu, Y.; Ke, Y.; Guo, C.; Zhu, N.; Mi, X.; Ma, Z.; Hu, Y. Appl. Catal. A 2002, 235, 33. (c) Yang, Y.; Liu, Z.; Zhong, L.; Qiu, P.; Dong, Q.; Cheng, R.; Vanderbilt, J.; Liu, P. Organometallics 2011, 30, 5297. (d) Rucklidge, A. J.; McGuinness, D. S.; Tooze, R. P.; Slawin, A. M.Z.; Pelletier, J. D. A.; Hanton,M.J.; Webb, P. B. Organometallics 2007, 26, 2782.

11. (a) Morgan, D. H.; Schwikkard, S. L.; Dixon, J. T.; Nair, J. J.; Hunter, R. Adv. Synth. Catal. 2003, 345, 939. (b) Van Rensburg, W. J.; Grove, C.; Steynberg, J. P.; Stark, K. B.; Huyser, J. J.; Steynberg, P. J. Organometallics 2004, 23, 1207.

12. Meijboom, N.; Schaverien, C. J.; Orpen, A. G. Organometallics 1990, 9, 774.

13. (a) Vidyaratna, I.; Nikiforov, G. B.; Gorelsky, S. I.; Gambarotta, S.; Duchateau, R.; Korobkov, I. Angew. Chem., Int. Ed. 2009, 48, 6552. (b) Albahily, K.; Fomitcheva. V.; Murugesu, M.; Gambarotta, S.; Korobkov, I.; Gorelsky, S. I. J. Am. Chem. Soc. 2011, 133, 6388.

14. (a) Peitz, S.; Peulecke, N.; Aluri, B. P.; Hansen, S.; Muller, B. H.; Spannenberg, A.; Rosenthal, U.; Al-Hazmi, M. H.; Mosa, F. M.; Wohl, A.; Muller, W. Eur. J. Inorg. Chem. 2010, 1167. (b) Wohl, A.; Muller,W. Peitz, S.; Peulecke, N.; Aluri, B. P.; Muller, B. H.; Heller, D.; Rosenthal, U.; Al- Hazmi, M. H.; Mosa, F. M. Chem. Eur. J. 2010, 16, 7837. (c) Peitz, S.; Peulecke, N.; Muller, B. H.; Spannenberg, A.; Drexler, H. J.; Rosenthal, U.; Al-Hazmi, M. H.; Al-Eidan, K. E.; Wohl, A.; Muller, W. Organometallics

165

Chapter 5

2011, 30, 2364. (d) Aluri, B. R.; Peulecke, N.; Peitz, S.; Spannenberg, A.; Muller, B. H.; Schulz, S.; Drexler, H.-J.; Heller, D.; Al-Hazmi, M. H.; Mosa, F. M.; Wohl, A.; Muller, W. Dalton Trans. 2010, 39, 7911. (e) Peitz, S.; Peulecke, N.; Aluri, B. R.;Muller, B. H.; Spannenberg, A.; Rosenthal, U.; Al- Hazmi, M. H.; Mosa, F. M.; Wohl, A.; Muller, W. Organometallics 2010, 29, 5263.

15. (a) Albahily, K.; Koc, E.; Al-Baldawi, D.; Savard, D.; Gambarotta, S.; Burchell, T. J.; Duchateau, R. Angew. Chem., Int. Ed. 2008, 47, 5816. (b) Albahily, K.; Koc, E.; Al-Baldawi, D.; Savard, D.; Gambarotta,S.; Burchell, T. J.; Duchateau, R. Angew. Chem., Int. Ed. 2008, 47, 5816. (c) Albahily, K.; Al-Baldawi, D.; Gambarotta, S.; Koc, E.; Duchateau, R. Organometallics 2008, 27, 5708. (d) Albahily, K.; Al-Baldawi, D.; Gambarotta, S.; Koc-, E.; Duchateau, R. Organometallics 2008, 27, 5943.

16. Thapa, I.; Gambarotta, S.; Duchateau, R.; Kulangara, S. V.; Chevalier, R. Organometallics 2010, 29, 4080.

17. Overett, M. J.; Blann, K.; Bollmann, A.; Villiers, R.; Dixon, J. T.; Killian E.; Maumela, M. C.; Maumela, H.; McGuinness, D. S.; Morgan, D. H.; Rucklidge, A.; Slawin. M. Z. A. J. Mol. Catal. A: Chem. 2008, 283, 114.

18. Dulai, A.; de Bod, H.; Hanton, M. J.; Smith, D. M.; Downing, S.; Mansell, S. M.; Wass, D. F. Organometallics 2009, 28, 4613.

19. Kern, R. J.; J. Inorg. Nucl. Chem. 1962, 24, 1105. 20. (a) Zhang, S.; Pattacini, R.; Braunstein, P. Inorg. Chem. 2011, 50, 3511. (b) Block, M.; Kluge, T.; Bette, M.; Schmidt, J.; Steinborn, D. Organometallics 2010, 29, 6749.

166

Chapter 5

21. Al-Jibori, S.; Shaw, B. L.; J. Chem. Soc., Chem. Commun. 1982, 286.

22. Albahily, K.; Licciulli, S.; Gambarotta, S.; Korobkov, I.; Chevalier, R.; Schuhen, K.; Duchateau, R. Organometallics 2011, 30, 3346.

167

168

Chapter 6

Technology Assessment

169

Chapter 6

The development of new single-site catalysts for ethylene polymerisation is an area of ongoing research interest due to the versatility of the ligand fine-tuning, which allows the production of polymers with tailored microstructures and properties.1 Even though homogeneous single-site catalysts allow the production of polymers with microstructures that cannot be obtained with heterogeneous Ziegler- Natta catalysts, they possess a serious disadvantage, viz. the lack of thermal stability, which in turn may result in reduced activity of the catalytic system. In order to tackle the issue of catalyst stability versus activity, our research group has developed completely new classes of thermally stable self-activating ethylene polymerisation/oligomerisation catalysts (vanadium- and chromium-based), ligated by unprecedented aluminium-pyrrolyl ligand systems.2 Due to the paramagnetic nature of chromium and vanadium-based aluminium-pyrrolyl catalysts, little is known about their dynamic coordination chemistry, which limits the rational development of this class of promising ethylene oligomerisation/polymerisation catalysts.

In Chapter 2, we focussed on extending the aluminium-pyrrolyl chemistry to group (IV) metals (Ti, Zr and Hf), mainly to get more mechanistic insight into the nature of the active species generated during the catalysis and to understand the dynamic coordination chemistry of this class of complexes. Several novel heterobimetallic aluminium-pyrrolyl complexes of titanium, zirconium and hafnium were synthesised and were fully characterised by means of single crystal X-ray diffraction, elemental analysis and NMR spectroscopy. However, in comparison to vanadium- and chromium-based aluminium-pyrrolyl catalysts, the catalytic performance of the group (IV)-based aluminium-pyrrolyl catalysts were not promising because none of these catalysts showed any self-activating catalytic behaviour or high thermal stability. The mono(aluminium-pyrrolyl) complexes afforded moderately active ethylene polymerisation catalysts (producing + - UHMWPE) upon activation with either MAO or TIBA/[Ph3C] [B(C6F5)4] , albeit

170

Chapter 6 only at higher ethylene pressure. On the other hand, the bis(aluminium-pyrrolyl) group IV metal species did not yield active ethylene polymerisation catalysts, irrespective of the nature of the group IV metal applied. As expected, these catalysts did not show any catalytic activity towards the polymerisation of higher α-olefins such as propylene or 1-hexene either. The mechanistic investigation using NMR spectroscopy provided us a better understanding of the reason for the poor or lack of catalytic behaviour. A possible mechanism for lack of, or only poor catalytic behaviour of these catalysts was proposed and supported by DFT calculations.

Even though, the results obtained in Chapter 2 provided us mechanistic insight into the underlying chemistry of aluminium-pyrrolyl complexes, the catalytic performance of these catalysts was rather disappointing. Therefore, in Chapter 3 we changed our focus on pyrrole-based chromium catalysts. The chemistry of pyrrole-based chromium catalysts is more relevant from an industrial perspective, because the only existing commercial selective trimerisation system today (Chevron-Phillips trimerisation system, Qatar) consists of a pyrrole-based chromium catalyst system. Moreover, our past research works on chromium-based aluminium-pyrrolyl catalysts allowed us to correlate the metal oxidation state with the selectivity of ethylene oligomerisation.2

In order to investigate how the ancillary ligand system determines the selectivity and activity, different types of dipyrrole-based ligands were synthesised. Upon activation with triethylaluminium (TEA), dipyrrole-based chromium catalysts showed less activity and selectivity compared to the Chevron-Phillips ethylene trimerisation system based on 2,5-dimethylpyrrole-stabilised chromium catalysts.3 However, by using an alternate activation method (depleted- methylaluminoxane (DMAO) in combination with triisobutylaluminium (TIBA) (1:2 ratio) as cocatalyst), we succeeded to improve the selectivity of the dipyyrrole-

171

Chapter 6 based chromium catalysts significantly. Applying this novel activation method to the 2,5-dimethypyrrole-based system resulted in a nearly two fold increase in the activity [372.3 kg/(g Cr.h)] compared to the activity reported in the Chevron- Phillips patent [156.7 kg/(g Cr.h)], with a slight improvement in 1-hexene selectivity.3c The fact that the new activation method resulted in a polymer free- trimerisation system also gives significant added value to this finding. A selective ethylene trimerisation system without any polymer is of great industrial relevance, because it completely solves the issue of reactor fouling. Moreover, the higher activity and selectivity was obtained at relatively lower ethylene pressure and temperature (30 bar, 60 °C), whereas in the case of the commercial Phillips ethylene trimerisation system high pressure is required (50-100 bar) to get higher activity. The same ligand system (2,5-dimethylpyrrole) was activated with

B(C6F5)3 by Mitsubishi and also patented and the reported activity was comparable to the activity mentioned in the Chevron-Phillips patent [140.7 kg/(g Cr.h)].4 Further optimisation (increasing the ethylene pressure, oligomerisation temperature etc.) of the catalytic system and scaling up the oligomerisation process needs to be carried out. Also, this novel activation method will be extended to activate the existing trimerisation and tetramerisation catalysts, in order to understand its effect on activity and selectivity. This was unfortunately not possible during the course of this doctoral study due to the limited time available. A schematic comparison of the activity, selectivity and oligomerisation conditions of the newly developed system along with patented trimerisation systems is depicted below.

172

Chapter 6

Figure 1. Comparison of activity and selectivity of different pyrrole-based catalysts.

In Chapter 4, the synthesis and the ethylene trimerisation capability of chromium(II) and chromium(III) heteroscorpionate complexes has been discussed. In 2002, the Tosoh Corporation patented a homoscorpionate tris(dimethylpyrazolyl)-methane chromium catalyst, Cr{CH(Me2pz)3}Cl3, which produces 1-hexene with very high selectivity (99 %) and a moderate activity (40.1 5 kg/(g Cr. h)) upon activation with MAO in toluene. This was followed by a series of chromium-based trimerisation catalysts stabilised by fac-coordinated bis(pyrazolyl)methane based ligands.6 The objective of our investigation was to understand the effect of ligand modification on activity and selectivity. N2N΄ and

N2O donor heteroscorpionate ligands with secondary and tertiary amine donors, as well as phenol and phenyl ether donors, were successfully complexed to Cr(II) and Cr(III). The ethylene oligomerisation results show that very small changes in the nature of the pendant donor atom (N, O) and/or the presence of ionisable hydrogen (N-H, O-H) can have a large effect on activity and/or selectivity. In almost all cases, the new complexes gave an increase in productivity compared to

Cr{CH(Me2pz)3}Cl3, but disappointingly the selectivity towards 1-hexene was significantly lower than the selectivity obtained with Cr{CH(Me2pz)3}Cl3.

The current literature scenario of new catalyst search for selective trimerisation and tetramerisation is still dominated by serendipity and a clear

173

Chapter 6 relationship between catalytic behaviour (selective versus non-selective) and ligand features is yet to be found. However, based on the current state of the art, it is obvious that the heteroatom-containing ligands (especially ligands containing nitrogen and phosphorous donor atoms) seem to be most successful so far.7 It was argued that the phosphorus donor atoms might be crucial to the stabilisation of the highly desirable lower oxidation states of chromium. At the same time, hard donors such as oxygen and nitrogen could aid in the retention of aluminium residues.8

It is with this aim that we were engaged in a systematic screening of families of diversified ligand systems containing various combinations of donor functions, ligand scaffolds and with ability of retaining aluminates residues. In chapter 5, novel bis(diphenylphoshino)methane (DPPM)-based mono anionic ligands were prepared by the reaction of the in situ-generated DPPM- anion with different - cumulenes. The reaction of DPPM with CO2 and aliphatic isocyanates (CyNCO, t- BuNCO) occurred via the attack of the nucleophilic DPPM- anion to the electrophilic carbon atom of the cumulenes, in the process forming a carbon-carbon bond (classical addition). However, the use of aromatic isocyanate resulted in a “non-classical” type addition, forming a mixed valent P(III)/P(V)-containing ligand. We observed that this arrangement disappears upon complexation with chromium through an internal redox transformation. The connectivity may switch back and forward between “classical” and “non-classical” forms, depending on the steric requirement of the final complexes. While we found no evidence that this may be a fluxional behaviour (at least in the case of the diamagnetic lithium salt), this strange dynamic behaviour for a ligand system in principle holds some promises for catalysis. To this end, the catalytic testing of all the species presented in this work showed mainly non-selective behaviour with high catalytic activity. However, fine tuning the oligomerisation condition (DMAO as cocatalyst,

Cr(acac)3 as metal precursor, methyl cyclohexane as solvent) resulted in the selective formation of 1-hexene and 1-octene as the only liquid product.

174

Chapter 6

Nevertheless, the considerable amount of waxy polyethylene produced along with the liquid products makes the system less attractive from an industrial point of view.

175

Chapter 6

REFERENCES

1. (a) McKnight, A. L.; Waymouth, R. M. Chem. Rev. 1998, 98, 2587. (b) Peacock, A. J.; Handbook of polyethylene: Structure, Properties and Applications, Marcel Deckcer: New York, 2000. (c) Ross, J. F.; MacAdams, J. L.; Polymeric Material Encyclopedia, Volume 8, Ed. Salamone, J. C.; CRC Presss, 1996, 5953.

2. (a) Jabri, A.; Korobbkov, I.; Gambarotta, S.; Duchateau, R. Angew. Chem. Int. Ed. 2007, 46, 6119. (b) Vidyaratna, I.; Nikiforov, G. B.; Gorelsky, S. I.; Gambarotta, S.; Duchateau, R.; Korobkov, I. Angew. Chem., Int. Ed. 2009, 48, 6552.

3. (a) Reagan, W. K. EP 0417477 (Phillips Petroleum Company), 1991. (b) Knudsen, R. D.; Freeman, J. W.; Lashier, M. E. US 5563312 (Phillips Petroleum Company), 1996. (c) Freeman, J. W.; Buster, J. L.; Knudsen, R. D. US 5856257 (Phillips Petroleum Company), 1999. (d) Reagan, W. K.; Freeman, J. W.; Conroy, B. K.; Pettijohn, T. M.; Benham, E. A. EP 0608447 (Phillips Petroleum Company), 1994.

4. Tanaka, E.; Urata, H.; Oshiki, T.; Aoshima, T.; Kawashima, R.; Iwade, S.; Nakamura, H.; Katsuki, S.; Okanu, T. EP 0 611 743 (Mitsubishi Chemical Corporation), 1994.

5. Yoshida, T.; Yamamoto, H.; Okada, H.; Murakita, H. US 0035029 (Tosoh Corporation), 2002.

176

Chapter 6

7. (a) Bollmann, A.; Blann, K.; Dixon, J. T.; Hess, F. M.; Killian, E.; Maumela, H.; McGuinness, D. S.; Morgan, D. H.; Neveling, A.; Otto, S.; Overett, M.; Slawin, A. M. Z.; Wasserscheid, P.; Kuhlmann, S. J. Am. Chem. Soc. 2004, 126, 14712 (b) Han, T. K.; Ok, M. A.; Chae, S. S.; Kang, S. O. (SK Energy Corporation), WO 2008/ 088178, 2008. (c) Shaikh, E.; Albahily, K.; Sutcliffe, M.; Fomitcheva. V.; Gambarotta, S.; Duchateau, R.; Korobkov, I. Angew. Chem., Int. Ed. 2012, 51, 1. (d) Wass, D. F. (BP Chemicals Ltd) WO 02/04119, 2002; (e) Carter, A.; Cohen, S. A.; Cooley, N. A.; Murphy, A.; Scutt, J.; Wass, D. F. Chem. Commun. 2002, 858. (f) Peitz, S.; Peulecke, N.; Aluri, B. P.; Hansen, S.; Muller, B. H.; Spannenberg, A.; Rosenthal, U.; Al- Hazmi, M. H.; Mosa, F. M.; Wohl, A.; Muller, W. Eur. J. Inorg. Chem. 2010, 1167. (g) Wohl, A.; Muller,W. Peitz, S.; Peulecke, N.; Aluri, B. P.; Muller, B. H.; Heller, D.; Rosenthal, U.; Al-Hazmi, M. H.; Mosa, F. M. Chem. Eur. J. 2010, 16, 7837.

8. (a) Albahily, K.; Koc, E.; Al-Baldawi, D.; Savard, D.; Gambarotta, S.; Burchell, T. J.; Duchateau, R. Angew. Chem., Int. Ed. 2008, 47, 5816. (b) Albahily, K.; Koc, E.; Al-Baldawi, D.; Savard, D.; Gambarotta,S.; Burchell, T. J.; Duchateau, R. Angew. Chem., Int. Ed. 2008, 47, 5816. (c) Albahily, K.; Al-Baldawi, D.; Gambarotta, S.; Koc, E.; Duchateau, R. Organometallics 2008, 27, 5708. (d) Albahily, K.; Al-Baldawi, D.; Gambarotta, S.; Koc-, E.; Duchateau, R. Organometallics 2008, 27, 5943. (e) Thapa, I.; Gambarotta, S.; Duchateau, R.; Kulangara, S. V.; Chevalier, R. Organometallics 2010, 29, 4080.

177

178

Summary

Novel catalysts for ethylene polymerisation and oligomerisation

Polyolefins, the generic name for the synthetic polymers based on ethylene, propylene, other non-functionalised α-olefins and reactive internal olefins, are the most common commodity plastics. The development of new single-site catalysts for olefin polymerisation is an area of ongoing research interest due to the versatility of the ligand fine-tuning, which allows the production of polymers with tailored microstructures and properties. Likewise, selective ethylene oligomerisation to 1-hexene or 1-octene can only be achieved with homogeneous single-site oligomerisation catalysts. The design of the ancillary ligand, the metal and the cocatalyst are key factors determining the selectivity of these catalysts.

In the first part of the thesis (Chapter 2), we have studied the chemistry of group IV metal-based aluminium-pyrrolyl catalysts to probe their ethylene polymerisation activity and mechanistic behaviour. Several novel heterobimetallic aluminium-pyrrolyl complexes of titanium, zirconium and hafnium were synthesised and were fully characterised by means of single crystal X-ray diffraction, elemental analysis and NMR spectroscopy. However, in comparison to vanadium- and chromium-based aluminium-pyrrolyl catalysts, the catalytic performances of the group (IV)-based aluminium-pyrrolyl catalysts were not promising because none of these catalysts showed any self-activating catalytic behaviour or high thermal stability. The mono(aluminium-pyrrolyl) complexes afforded moderately active ethylene polymerisation catalysts (producing UHMWPE) upon activation with either methylaluminoxane (MAO) or + - triisobutylaluminium (TIBA)/[Ph3C] [B(C6F5)4] , albeit only at higher ethylene pressure. On the other hand, the bis(aluminium-pyrrolyl) group IV metal species

179

did not yield active ethylene polymerisation catalysts, irrespective of the nature of the group IV metal applied. Intramolecular coordination of a chlorine to the active site as a possible reason for lack of, or only poor, catalytic behaviour of these catalysts was supported by density functional theory (DFT) calculations.

The second objective of the research described in this thesis was to develop novel catalysts for ethylene oligomerisation. The fact that the fractionation of the oligomerisation mixture to extract the most requested α-olefins (1-hexene and 1- octene) is an energy-intensive process motivates the search for selective catalytic systems. However, designing catalysts which can produce 1-hexene and 1-octene selectively remains a great challenge. We investigated the ethylene oligomerisation capabilities of three different classes of chromium-based catalysts, namely: pyrrole-based chromium catalysts, pyrazole-based chromium catalysts and modified bis(diphenylphosphino)methane chromium catalysts.

In chapter 3, we evaluated the ethylene oligomerisation capabilities of various dipyrrole-based ancillary ligands and the results were compared with the Chevron-Phillips trimerisation system. Activation of dipyrrole-chromium catalysts using triethylaluminium (TEA) as cocatalyst showed that these systems are less active and selective compared to the commercial Phillips trimerisation system. However, by using an alternative activation method (via combination of depleted- MAO (DMAO) and TIBA), an efficient way to increase both the activity and selectivity was discovered. Applying this novel activation method to the 2,5- dimethypyrrole-based system resulted in a nearly two-fold increase in the activity compared to the activity of the original Chevron-Phillips system. The fact that the new activation method resulted in a polymer free-trimerisation system also gives significant added value to this finding. Moreover, the higher activity and selectivity was obtained at relatively lower ethylene pressure and temperature (30 bar, 60 °C), whereas in the case of the commercial Phillips ethylene trimerisation system high

180

pressure is required (50-100 bar) to get high activity. Furthermore, well-defined chromium complexes of dipyrrole ligands were synthesised, fully characterised and assessed for their ethylene oligomerisation activities.

In the course of Chapter 4, the synthesis and the ethylene trimerisation capability of chromium(II) and chromium(III) heteroscorpionate complexes has been discussed. N2N΄ and N2O donor heteroscorpionate ligands with secondary and tertiary amine donors, as well as phenol and phenyl ether donors, were successfully coordinated to Cr(II) and Cr(III). The ethylene oligomerisation results show that very small change in the nature of the pendant donor atom (N, O) and/or the presence of ionisable hydrogen (N-H, O-H) can have a large effect on activity and/or selectivity. In almost all cases, the new complexes gave an increase in productivity compared to Cr{CH(Me2pz)3}Cl3, but disappointingly the selectivity towards 1-hexene was significantly lower than the selectivity obtained with

Cr{CH(Me2pz)3}Cl3.

In chapter 5, novel bis(diphenylphoshino)methane (DPPM)-based mono anionic ligands were prepared by the reaction of the in situ-generated DPPM- anion - with different cumulenes. The reaction of DPPM with CO2 and aliphatic isocyanates (CyNCO, t-BuNCO) occurred via the attack of the nucleophilic DPPM- anion to the electrophilic carbon atom of the cumulenes, in the process forming a carbon-carbon bond (classical addition). However, the use of aromatic isocyanate resulted in a “non-classical” type addition, forming a mixed valent P(III)/P(V)- containing ligand. We observed that this arrangement disappears upon complexation with chromium through an internal redox transformation. The connectivity may switch back and forward between “classical” and “non-classical” forms, depending on the steric requirement of the final complexes. While we found no evidence that this may be a fluxional behaviour (at least in the case of the diamagnetic lithium salt), this strange dynamic behaviour for a ligand system in

181

principle holds some promises for catalysis. To this end, the catalytic testing of all the species presented in this chapter showed mainly non-selective behaviour with high catalytic activity. However, fine tuning the oligomerisation condition (DMAO as cocatalyst, Cr(acac)3 as metal precursor, methylcyclohexane as solvent) resulted in the selective formation of 1-hexene and 1-octene as the only liquid product. Nevertheless, the considerable amount of waxy polyethylene produced along with the liquid products makes the system less attractive from an industrial point of view.

The research described in this thesis has given more insights into the structure-property relationship of several new ethylene oligo/polymerisation catalysts.

182

Acknowledgements

Finally, it is time to acknowledge those people who supported me for last four years to complete this mission. It was not an easy task, but with all your support, care and advice, I managed to finish it in an excellent way.

Firstly, I would like to thank Dr. Rob Duchateau, my supervisor and co- promoter without whom this thesis would not be possible. Dear Rob, we started our research in SPC with a “nice fire” announcing our research journey. Thanks a lot for introducing me to the world of organometallic chemistry, teaching me how to work with pyrophoric chemicals and Schlenk techniques, guiding me patiently when things were not going smoothly and especially for sharing your enormous knowledge in the area of polyolefin catalysis. Indeed, we had a very pleasant time at TU/e and I am sure it will continue in our new work place-SABIC.

I would like to thank Prof. Cor Koning, my first promoter for giving me an opportunity to carry out my PhD in his excellent research group. Dear Cor, it was really a pleasure to work in your research group and many thanks for your constant motivation and care throughout the last four years.

I am grateful to Prof. Sandro Gambarotta, my second promoter for his support, advice and encouragements during my PhD life. Sandro, thanks for giving me an opportunity to work in your research lab (or should I say “crystal making factory”?) in Canada. Those six months in Canada were absolutely fruitful (especially, I learned how to make single crystals in a high-throughput way) and I was very fortunate to discuss my chemistry with you. Your contribution to the thesis is invaluable, even though most of our communications were mainly via e- mails and Skype. I am also very grateful to Dutch Polymer Institute (DPI) for their excellent support.

183

I would like to express my sincere gratitude to the reading committee: Prof. Philip Mountford (University of Oxford), Prof. Dieter Vogt (University of Eindhoven) and Prof. Gerrit Luinstra (University of Hamburg) for reading the thesis by spending their precious time and approving the manuscript in such a short period of time. Special thanks to Prof. Duncan Wass (University of Bristol) for kindly accepting our invitation to be a part of the defence committee.

Many thanks to Ilia Korobkov (University of Ottawa), Martin Fijten, Rinske Knoop for their technical support and Pleunie Smits and Caroline van Os- Rovers for helping me with all administrative issues. My sincere gratitude to Wouter Gerristen for his kind help during my first year in Netherlands. The thesis would not be completed without the contribution from Daniel Haveman, Indira Thapa, Amir Jabri, Alexander Kilpatrick (Sandy), and Balamurugan Vidjayacoumar. Guys, your help and contribution to my thesis are remarkable and highly appreciated.

Working in the Cactus lab was really a great pleasure and I am really proud to be a member of the “Cactus Family”. Timo, Dirk, Raf and Miloud, thanks for sharing your experience and knowledge with us during our cactus meetings. Yun, “Our DFT girl”, many thanks for your significant contributions with DFT calculations and good luck with making some nice crystals. Dear Camille, don’t work too hard in the lab, please keep some energy for the future. Your friendly character and enthusiasm helped me a lot to keep our cactus family pleasant and I wish you all the best to finish your thesis. Elham, many thanks for being a good friend and I will never forget our chats during NIOK course. Gemma “the organiser”, your friendly character and jokes in the lab always helped to keep the lab active and I wish you all the success with your new job. Dear Fabian, I will never forget our good old time with “Breathing exercise” and best wishes with finishing your project. Mark, best wishes with your project and thank you for your

184 help with SEC analysis. I also would like to thank former cactus members Ece, Simona, Saskia, Saied and Maurice for all their help.

I am very thankful to all my office mates Inge, Gijs, Hector, Evgeniy, Erik and Julien for making our office a pleasant place to work and also for all your support and help during the last four years. Many thanks to Jey for his care and valuable advice. Special thanks to Bahar, Lidia, Jing, Donglin, Stefan and Monique for their smiling faces. I also would like to thank all my colleagues Hemanth, Hans, Bert, Alex, Hanneke, Gozde, Martin, O, Yingyuan, Mark, B, Seda, Pooja, Judith, Dogan, Jerome, Tom, Ingeborg, Carin, Ali and Mohammed for their support.

I am very fortunate to have bunch of good friends in my life, who really keeps the life going smoothly. Dear Ajin and Deepak, no formal words of thanks to you as the word itself is not enough to express the feeling. Many thanks to Eldhose, Sumesh, Smitha, Vinod, Babu and Indu for making my life in Netherlands enjoyable. Dear all, the good time we spent together will stay in my mind for the rest of my life. Sincere thanks to Meera and Shafeer kalathil for their support during the past years and I wish you all the best for completing their theses.

This thesis would not be possible without the support and prayers from my family members. My deep gratitude to my father who is no longer with me to witness my achievement. My sincere thanks to my mother, Santha whom I owe everything for what I am today. Deep appreciations to my brother Suneesh and sister Sunisha for their prayers, support and care throughout my life. I would like to express my gratitude to my brother-in-law Santhosh, sister-in-law Vijina and my little brother Ashin for keeping me happy with their care and smiles.

Finally, my heartfelt thanks to my beloved wife Rashija, without her support, encouragement, patience and love I would not have finished this thesis.

Shaneesh

185

186

List of publications

1. Vadake Kulangara, S.; Jabri, A.; Yang, Y.; Korobkov, I.; Gambarotta, S.; Duchateau, R. Organometallics 2012, 31, 6085.

2. Vadake Kulangara, S.; Mason, C.; Juba, M.; Yang, Y.; Thapa, I.; Korobkov, I.; Gambarotta, S.; Duchateau, R. Organometallics 2012, 31, 6438.

3. Vadake Kulangara, S.; Haveman, D.; Yang, Y.; Vidjayacoumar, B.; Gambarotta, S.; Duchateau, R. “Cocatalyst effect on selective ethylene oligomerisation”. To be submitted.

4. Kilpatrick, A.; Vadake Kulangara, S.; Cushion, M.; Duchateau, R.; Mountford, P. Dalton Trans. 2010, 39, 3653.

5. Thapa, I.; Gambarotta, S.; Duchateau, R.; Vadake Kulangara, S.; Chevalier, R. Organometallics 2010, 29, 4080.

187

188

Curriculum Vitae

Shaneesh Vadake Kulangara was born on 19th January 1986 in Panoor, Kerala, India. After completing his high school studies in 2003, he started his undergraduate study in Polymer Chemistry at Kannur University, Kerala, India. In 2006, he completed his Bachelor of Science degree in Polymer Chemistry with first rank from Kannur University. In the following two years, he completed his Master of Science in Applied Chemistry from Cochin University of Science and Technology (CUSAT), Kerala, India.

In August 2008, Shaneesh Vadake Kulangara commenced his PhD in the group of prof.dr. Cor Koning at the Eindhoven University of Technology. He worked in a research project funded by Dutch Polymer Institute (DPI) under the supervision of dr. Rob Duchateau (Eindhoven University of Technology) and Prof. Sandro Gambarotta (University of Ottawa, Canada). The results of his PhD research are presented in this thesis. Since May 2012, Shaneesh is employed as a Research Scientist at Technology and Innovation Center, SABIC, Netherlands.

189