University of Groningen

Copolymerization of Ethene and Functionalized Comonomers with Cationic α-DiiminePalladium CatalystsLi, Weidong

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2008

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Li, W. (2008). Copolymerization of Ethene and Functionalized Comonomers with Cationic α-DiiminePalladium Catalysts s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 03-05-2018

Copolymerization of Ethene and Functionalized

Comonomers with Cationic α-Diimine Palladium Catalysts

The studies described in this thesis are part of the Research Programme of

the Dutch Polymer Institute (DPI), PO Box 902, 5600 AX, Eindhoven, The

Netherlands, projectnr. #110.

RIJKSUNIVERSITEIT GRONINGEN

Copolymerization of Ethene and Functionalized

Comonomers with Cationic α-Diimine Palladium Catalysts

Proefschrift

ter verkrijging van het doctoraat in de Wiskunde en Natuurwetenschappen aan de Rijksuniversiteit Groningen

op gezag van de Rector Magnificus, dr. F. Zwarts, in het openbaar te verdedigen op

vrijdag 28 maart 2008 om 16:15 uur

door

Weidong Li

geboren op 19 augustus 1968

te Shanxi, China

Promotor: Prof. dr. B. Hessen

Beoordelingscommissie: Prof. dr. C. J. Elsevier

Prof. dr. S. Mecking

Prof. dr. H. J. Heeres

ISBN: 978-90-367-3355-7

To my parents

Contents Chapter 1. Introduction 1 Chapter 2. Synthesis and stability of a cationic α-diimine 19

palladium methyl catalyst Chapter 3. Reactivity with oxygen-functionalized olefins 53 Chapter 4. Reactivity with nitrogen-functionalized olefins 93 Chapter 5. Reactivity with sulfur-functionalized olefins 117 Samenvatting 139 Acknowledgement 143

1

Chapter 1. Introduction

1.1. Polyolefins

Polyolefins form a highly important class of materials, with a wide range of applications, which are produced industrially on a huge scale (> 85 Mton per year). By far the largest volume of these polyolefins is linear polyethene (HDPE) and its copolymers with 1-alkenes (LLDPE), and isotactic polypropene (iPP). These polymers are all prepared by the catalytic (co-)polymerization of olefins using transition-metal catalysts. As polyethenes and polypropenes consist essentially of very long alkane molecules, these materials are very apolar and lipophilic. Although polyolefins serve a wide range of applications, their apolar nature limits their properties in certain areas, e.g. with respect to its barrier properties and permeability, dyeability, printability, and compatibility with more polar materials (such as glass fiber for reinforcement). It therefore would be desirable for certain applications to modify the polymer by introducing polar functional groups, with retention of the other favorable properties, like modulus, strength, solvent resistance, etc. 1,2 The conceptually most straightforward method to introduce polar functionalities into polyolefin materials is the incorporation into the polymer of comonomers with polar functional groups, such as alcohol, ester or amine functionalities. Once a copolymer bearing functionalities is obtained, further chemical derivatization using these reactive groups would be possible. Presently, the commercial copolymerization of ethene with polar monomers is only carried out in high pressure radical polymerization processes, used to generate highly branched low-density polyethene (LDPE).3,4,5

2

The metal-catalyzed copolymerization of simple alkenes, like ethene and propene, with functionalized olefin comonomers has to contend with certain inherent difficulties, which are related to the fundamentals of the catalyzed polymerization process. Transition-metal catalyzed olefin polymerization proceeds on metal alkyl species via a reaction sequence known as the Cossee-Arlman mechanism (Scheme 1.1). An electron-deficient metal alkyl species, with a vacant coordination site, can bind a molecule of olefin monomer, which then undergoes a migratory insertion reaction in which a new carbon-carbon bond is formed, and the vacant site is regenerated.

Scheme 1.1 When a comonomer contains a polar functionality, this usually is Lewis basic, and will compete for binding to the Lewis acidic catalyst metal center with the olefinic group. If this binding is reversible, it will in the best cases still lead to a certain catalyst inhibition, slowing down the catalysis. This inhibition can be significantly enhanced after insertion of the functionalized comonomer into the metal-carbon bond, when there is a possibility for intramolecular coordination of the functionality to the metal center. Chelate complexes thus formed can be very stable, and lead to a further decrease in catalyst performance. In addition, many polar functionalities react in an irreversible way with the polarized metal-alkyl bond. Protic functionalities like alcohols will protonate the alkyl group, destroying the metal-carbon bond that is essential for the polymerization catalysis. Carbonyl functionalities in aldehyde or ester groups are susceptible to nucleophilic attack by the alkyl group bound to the metal, especially when the metal is

MR R' Insertion

M

R'

RMR

R'

3

relatively electropositive. In each of these cases, irreversible, and often rapid, catalyst deactivation will result (Scheme 1.2). It is therefore clear that the catalytic copolymerization of ethene or propene with functional comonomers is by no means an easy undertaking.

Scheme 1.2 In recent years, considerable effort has been directed to finding strategies for incorporation functional comonomers into polyolefins by catalytic copolymerization. 6,7,8 In the next paragraphs, a brief survey is given of several approaches taken to this end. For further development in this area, an increased understanding of the strength and nature of the interactions between these functional comonomers and various catalyst types is desirable. The work described in this thesis seeks to contribute to this understanding.

f = functional group with O, N, S, etc.

M+ Pf

f M+

f

P

M+ P

M+ P

f

A D

CB

chain growthM+

f

P

M fother type of deactivation

monomer

P = polymer chain

4

1.2. Early transition metal catalysts

Since the discoveries by Ziegler and Natta that ethene and propene can be effectively polymerized by heterogeneous catalysts generated from combinations of titanium halides and alkyl aluminum compounds, this group 4 metal has been the mainstay of industrial polyolefin production. 9,10,11,12 The subsequent development of soluble, well-defined (“single site”) organometallic olefin polymerization catalysts also focused initially on the group 4 metals, especially titanium and zirconium. 13,14,15,16 As these metals are electropositive, hard Lewis acidic and “oxophilic” (i.e. with M-O bonds being considerably stronger than M-C bonds), they are not easily applied to copolymerizations involving heteroatom-containing comonomers. Many strategies were developed to avoid these problems. The most applied in copolymerizations of functionalized comonomers with early transition metal catalysts is the functional group protection strategy.17 The polar functionalities are protected by incorporating very high levels of Lewis acids, e.g. AlCl3, 18 tetramethylaluminium (TMA) 19 into the copolymerization system to weaken or prevent the strong interactions between the catalysts and the functionalities. For example, comonomers with -OH groups were reacted with an excess of Al-alkyl to generate protected functionalities -OAlR2. The -OH functionality was regenerated upon hydrolytic work-up after polymerization.20,21 A related strategy uses comonomers in which the functionalities are already protected by hydrolysable groups e.g. -OSiR3,22,23 and -N(SiMe3)2.24 Upon hydrolytic work-up after the polymerization, –OH or –NH2 functionalities are generated on the polymer. The introduction of sterically demanding groups on a heteroatom in the functionalized comonomer can be sufficient to disfavor metal-functionality interactions interfering with the polymerization process. An example is 5-(N,N-diisopropylamino)-1-pentene, 25 which is readily copolymerized

5

with olefins. Unfavorable interactions of a polar group with the catalyst metal center can also be decreased when the functional group on the monomer is located far away from the double bond (separated from it by a spacer group). This disfavors the formation of stable chelates by intramolecular interaction after insertion of the comonomer. This ‘spacer strategy’ is more effective when it is combined with other ‘protection strategies’ that diminish intermolecular metal-functionality interactions, e.g. in the copolymerization of 10-undecene derivatives.26,27 Alternative routes to functionalized polyolefin copolymers have been devised that avoid the use of insertion polymerization catalysis. Examples are the catalytic ring-opening metathesis polymerization (ROMP) of cyclic olefins and functionalization of the resulting unsaturated polymer, and catalytic ROMP of cycloolefins bearing polar functionalities (metathesis catalysts generally being more resistant to polar functionalities than catalysts for insertion polymerization), followed by hydrogenation to remove the remaining unsaturation in the polymer backbone.28, 29

1.3. Late transition metal catalysts

Compared with early transition metal catalysts, the less electropositive, softer Lewis acidic nature and lower oxophilicity of the later transition metal complexes generally lead to a greater functional-group tolerance.30,7 Nevertheless, late transition-metals have only relatively recently been successfully applied in the catalytic synthesis of high polymer polyolefins. The reason for this is that, for these metals, chain transfer processes are relatively facile, leading to the formation of low molecular-weight products (Scheme 1.3).31, 32

6

Scheme 1.3

Many efforts have been made on finding proper ligands for late transition metal complexes to prevent β-H elimination to achieve a polymerization. Sterically demanding α-diimine ligands are successful examples of ligands that can diminish the rate of chain-transfer relative to that of chain growth due to their steric hindrance. A number of well-defined catalysts based on complexes of iron and cobalt,32,33,34,35 rhodium and platinum,36 nickel and palladium,37,38 etc. were reported to be able to polymerize olefins to high polymers with reasonable activities. However, only few of them show an appreciable ability to incorporation of polar olefins.39,40,41 The various approaches taken with late transition-metal catalysts to copolymerize ethene with functionalized comonomers will be outlined in following sections. Many late transition-metal catalysts (but especially the ones based on palladium) display a peculiar behavior called “chain-walking”. Through sequential β-H elimination, olefin rotation and olefin re-insertion steps (Scheme 1.4), the metal center can “walk” along the carbon chain of the polymer.30 The result of this behavior is that these catalysts can homopolymerize ethene to branched polyethene, but also that the “spacer strategy”, mentioned in section 1.2, is ineffective: even when the olefinic moiety and the polar functionality in the comonomer are separated by a -(CH2)n- spacer, the metal center can “walk” along that chain to the functionality, to form the most stable (and therefore least reactive) chelate complex.

M+ R

HM+

H

RM+

HR

7

Scheme 1.4

A way to counter this problem is to place a quaternary center between the olefinic group and the functional group in the comonomer, which will prevent the catalyst from walking all the way up to the functionality. Copolymerizations of ethene with diethyl allylmalonate and methyl 2,2-dimethyl-4-pentenoate are examples of this blocked chain strategy.42

1.4. Cationic Pd and Ni α-diimine catalysts

Scheme 1.5

The cationic α-diimine Ni and Pd catalysts [(N^N)ML][BAF] (Scheme 1.5) were first reported by Brookhart and co-workers in 1995.37 By applying a bulky α-diimine ligand around the metal center to prevent chain-transfer, the catalysts can successfully catalyze homopolymerization of ethene, α-olefins and internal olefins to high molecular weight polymers. This type

BAF-NN

R R

M

L

+R = H , CH 3,

NN

M = N i, Pd

L = O (C 2H 5)2, N C M e, e tc .

B AF - = {B [3 ,5-(C F 3)2C 6H 3]4} -

M+

HP

PM+

HM+

H

PM+

H

P

8

of catalysts is ready to take chain-walking so the products are highly branched polymers, varying from semi crystalline to amorphous. 43,44 The high tolerance of these catalysts to functionalized comonomers generated a lot of research interest. Ethene polymerizations can be carried out in the presence of ethers, organic esters, and acids.45 ,46 And this functional-group tolerance extends to comonomers bearing polar functionality, e.g. acrylates. The copolymerizations of ethene and propene with acrylates in an unprotected form to give high molecular weight and random copolymers were reported by Brookhart and co-workers.39,47 This was really the first successful attempt to copolymerization of conventional polar comonomers, although the rate of incorporation of acrylates is still rather low. 48 , 49 , 50 Similar to corresponding ethene homopolymers synthesized with these catalysts, the copolymers are highly branched (~100 branches/1000 carbon atoms) and the ester groups are located predominantly at the ends of branches. 51 , 52 Details of this particular copolymerization will be illustrated in Chapter 3. The success of copolymerizations of ethene and propene with acrylates encourages more attempts to copolymerize other functionalized comonomers using α-diimine Ni or Pd catalysts. Combining the ‘spacer strategy’ and protection with trimethylaluminum, 5-hexen-1-ol, 10-undecen-1-oic acid etc. were copolymerized with ethene or propene by α-diimine nickel catalysts. 53 DuPont researchers claimed the copolymerization of a range of polar monomers, e.g. fluoro-functionalized comonomers (CH2=CH(CH2)3C(CF3)3, etc.), chlorosilanes (CH2=CH(CH2)6SiCl3, etc.) and siloxanes (CH2=CHSi(OR)3, etc.) with ethene by using these catalysts. The copolymers with silane functionality can be cross-linked by hydrolysis, e.g. when they are exposed to moist air. 54 , 55 , 56 , 57 However, some polar monomers were found to inhibit polymerization completely, e.g. vinyl acetate, acrylamide and acrylonitrile.7

9

1.5. Neutral Pd(II) and Ni(II)-based catalysts

Neutral nickel hydride complexes bearing monoanionic phosphine-alkoxide chelating ligands are well-known catalysts for SHOP (Shell Higher Olefin Process): the catalytic oligomerization of ethene to linear 1-alkenes. Apparently, the chain-transfer reaction competes with the chain-growth process to lead to oligomers. In order to achieve a polymerization, the steric demand strategy which was succeeded in the cationic α-diimine systems was extended to the neutral nickel catalysts. A series of salicylaldimine ligands with bulky aryl substituents lying adjacent to the phenoxide substituent were selected for the new nickel catalysts (Scheme 1.6) to polymerize ethene or propene to high molecular weights and highly linear polymers even in the presence of ethers, ketones, esters, alcohols, amines and water. 58 , 59 , 60 Furthermore, copolymerizations of ethene with functionalized norbornenes such as 5-norbornen-2-ol and 5-norbornene-2-yl acetate were reported by Grubbs and co-workers.61 However, attempts to copolymerize functionalized vinyl monomers and ethene were not succeeded. A deactivation mechanism involving hydrogen transfer from substrate to Ni complex was found in a reaction of the catalyst and deuterated methyl acrylate.62

Scheme 1.6

R

O NNi

L R'L = Ph3P, CH3CNR = H, Ph, 9-antheraceneR' = Ph, CH3, etc.

10

Neutral palladium complexes are the most recently studied catalyst family for the copolymerization of ethene with functional vinyl monomers. An early study by Drent and co-workers revealed the incorporation of methyl acrylate (MA) into linear polyethene using a neutral palladium catalyst with a chelating P-O (di(2-methoxyphenyl)phosphinobenzene-2-sulfonate) ligand. Random copolymers of ethene and acrylate (2-17 mol%) with modest molecular weight (Mn = 2,000 – 20,000) was obtained.63 Another neutral palladium catalyst, (PO-OMe)PdMe(pyridine) (PO-OMe: 2-[bis(2-methoxyphenyl)phosphino]-4-methyl-benzenesulfonate), was reported very recently to copolymerize ethene and alkyl vinyl ethers to linear polymers with in-chain and chain-end functional groups.64 At same time, this catalyst was also applied in copolymerization of ethene and acrylonitrile to yield similar polymer.65

1.6. Research objectives and survey of the thesis

As mentioned above, Pd α-diimine catalysts have shown considerable promise in the copolymerization of functionalized olefins with ethene. Nevertheless, most of the attention in this area has focused on monomers of direct commercial interest (methyl acrylate, acrylonitrile). These catalysts also provide a good platform for systematic studies that can give more fundamental information on the substrate-catalyst interactions of a wide range of substrates. This information can be useful for future endeavors in the search for new functional polyolefin materials. In this thesis the interactions between the cationic α-diimine palladium catalyst [(N^N)PdMe(OEt2)][BAF] (2, N^N: ArN=CMe-CMe=NAr with Ar: 2,6-diisopropylphenyl, BAF: B[3,5-(CF3)2C6H3]4) and heteroatom-containing momomers with different functional groups (ether, amine, thioether etc.) were studied. Copolymerizations of ethene with these functionalized comonomers were also explored.

11

In Chapter 2, the synthesis of the cationic α-diimine palladium catalyst 2 is revisited. A new, convenient synthesis of the catalyst precursor (N^N)PdMe2 (1) is presented. It turns out that the details of the reaction conditions are important in the generation of the ionic complex 2. A side product in the reaction was identified as the µ-methyl, µ-methylene complex [(N^N)Pd(µ-CH3)(µ-CH2)Pd(N^N)][BAF] (3). The compound is formed by reaction of 2 with the starting material 1. The molecular and electronic structure of 3 and some of its reactivity features were explored. Chapter 3 describes the reaction of the cationic Pd-catalyst 2 with olefins bearing oxygen-containing functionalities: acrolein dimethyl acetal (ADMA), allyl ethyl ether (AEE) and 2-vinyl-1,3-dioxolane (VDO). AEE and ADMA smoothly give 1,2-insertion into the Pd-Me bond of 2 to give the 5-membered chelate complexes [(N^N)Pd(CH2CHMeCH2OEt)][BAF] (9) and {(N^N)Pd[CH2CHMeCH(OMe)2]}[BAF] (11) respectively. Both are able to catalyze the homopolymerization of ethene. Attempted ethene/AEE copolymerization resulted in rapid catalyst deactivation, forming the allyl complex [(N^N)Pd(η3-1-CH2CHCH2)][BAF] (10) and ethanol. Nevertheless, the ethene/ADMA copolymerization successfully yielded a branched polyethene copolymer bearing acetal functionalities. Here too, gradual deactivation takes place through formation of an allylic complex, [(N^N)Pd(η3-1-CH2CHCHOMe)][BAF] (12), and methanol. This deactivation could be retarded by the addition of aliquots of methanol to the reaction mixture. In contrast, VDO very readily ring-opens in the presence of 2, and could not be copolymerized. Chapter 4 describes the reaction of the cationic Pd-catalyst 2 with olefins bearing nitrogen-containing functionalities: allyl dimethyl amine (ADA), N-allyl carbazole (NAC) and 5-pentenyl carbazole (NPC). ADA reacts with 2 via smooth 1,2-insertion to give the 5-membered chelate complex [(N^N)Pd(CH2CHMeCH2NMe2)][BAF] (14). In contrast to the ether chelates from chapter 3, this compound does not catalyze the homopolymerization of ethene, due to the stronger Lewis basicity of the

12

amine. NAC and NPC react with 2 to form unusual 3-membered chelate complexes after insertion into the Pd-Me bond and “chain-walking”. The product from NPC, {(N^N)Pd[Me2CHC2H4CHN(C6H4)2]}[BAF] (15), was structurally characterized. These 3-membered chelates readily react with ethene, and NPC was successfully copolymerized with ethene to give branched polyethene copolymers bearing carbazole functionalities. The fluorescence of the carbazole group in these copolymers is highly dependent on the comonomer content. In contrast to NPC, NAC is not incorporated upon attempted copolymerization, probably due to the steric hindrance imparted by the carbazole group close to the olefinic moiety. Chapter 5 describes the reaction of the cationic Pd-catalyst 2 with olefins bearing sulfur-containing functionalities: allyl methyl thioether (AMT), allyl tert-butyl thioether (ABT), 2-allyl-1,3-dithiane (ADT) and 2-pentenyl-2-methyl-1,3-dithiane (PMDT). AMT reacts with 2 to give the stable thioether adduct [(N^N)PdMe(κ1-MeSCH2CH=CH2)][BAF] (17): the soft Lewis basic thioether effectively competes with the olefinic moiety for coordination with the metal, blocking insertion into the Pd-Me bond. The substrates ABT and PMDT, where the S-atoms are more screened by steric hindrance, do insert into the Pd-Me bond to give 5-membered chelates, which were structurally characterized. Nevertheless, due to the strong intramolecular Pd-S interactions in these complexes, reactivity with ethene (and thus the possibility for copolymerization) is impeded.

1.7. References

1 Takemoto, K.; Ottenbrite, R. M.; Kamachi, M. Functional Monomers and polymers; Marcel Dekker, Inc.: New York; 1997.

2 Simonazzi, T,; Nicola, A. D.; Aglietto, M.; Ruggeri, G. Comprehensive Polymer Science, Pergamon Press, New York, 1992.

13

3 Padwa, A. R. Prog. Polym. Sci. 1989, 14, 811.

4 Hagman, J. F.; Crary, J. W. Encyclopedia of Polymer Science and Engineering, vol. 1, Wiley, New York, 1985, p325.

5 Doak, K. W. Encyclopedia of Polymer Science and Engineering, vol. 6, Wiley, New York, 1986, p386.

6 Boffa, L. S.; Novak, B. M. Chem. Rev. 2000, 100, 1479.

7 Ittel, S. D.; Johnson, L. K.; Brookhart, M. Chem. Rev. 2000, 100, 1169.

8 Sen, A.; Borkar, S. J. Organometallics Chem. 2007, 692, 3291.

9 Ziegler, K.; Holzkamp, E.; Breil, H.; Martin, H. Angew. Chem. 1955, 67, 541.

10 Ziegler, K. Angew. Chem. 1964, 76, 545.

11 Natta, G. Angew. Chem. 1956, 68, 393.

12 Natta, G. Angew. Chem. 1964, 76, 553.

13 Coates, G. W.; Waymouth, R. M. Science 1995, 267, 217.

14 Yang, X.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10015.

15 Coughlin, E. B.; Bercaw, J. E. J. Am. Chem. Soc. 1992, 114, 7606.

16 Yang, X.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10015.

17 Schulz, D. N.; Waymouth, R. M. Functional polymers: Modern Synthetic Methods and Novel Structures; Patil A. O., Eds.; ACS Symposium Series 704; American Chemical Society: Washington DC, 1998; pp 38-57.

18 Tanaka, M.; Machida, S. EP Patent 283972 (1988).

14

19 Marques, M. M.; Correia, S. G. J. Polym. Sci., Part A, 1999, 37, 2457.

20 Aaltonen, P.; Fink, G.; Lofgren, B.; Seppala, J. Macromol. 1996, 29, 5255.

21 Wilen, C.-E.; Nasman, J. H. Macromol. 1994, 27, 4051.

22 Waymouth, R. M.; Kesti, M. R.; Coates, G. W. WO Patent 9412547 (1994).

23 Chung, T. C. Macromol. 1988, 21, 865.

24 Schneider, M. J.; Schafer, R.; Mulhaupt R., Polymer 1997, 38, 2455.

25 Stehling, U. M.; Stein, K. M.; Fischer, D.; Waymouth, R. M. Macromol. 1999, 32, 14.

26 Cerrada, M. L.; Benavente, R.; Perez, E.; Moniz-Santos, J.; Campos, J. M.; Ribeiro, M. R. Macromol. Chem. Phys. 2007, 208, 841.

27 Zhang, X.; Chen, S.; Li, H.; Zhang, Z.; Lu, Y.; Wu, C.; Hu, Y. J. Polym. Sci., Part A, 2005, 43, 5944.

28 Mclain, S. J.; Brookhart, M. Polym. Mater. Sci. Eng. 1997, 76, 246.

29 Bansleben, D. A.; Huynh-Tran, T. T.; Blanski, R. L.; Hughes, P. A.; Roberts, W. P.; Grubbs, R. H.; Hatfield, G. R. WO Patent 9950331 (1998).

30 Mecking, S. Angew. Chem. Int. Ed. 2001, 40, 534.

31 Britovsek, G. J. P.; Gibson, V. C.; Wass, D. F. Angew. Chem. Int. Ed. Engl. 1999, 38, 428.

32 Small, B. L.; Brookhart, M.; Bennett, A. M. A. J. Am. Chem. Soc. 1998, 120, 4049.

33 Small, B. L.; Brookhart, M. Macromol. 1999, 32, 2120.

15

34 Brookhart, M.; DeSimone, J. M.; Grant, B. E.; Tanner, M. J. Macromol.

1995, 28, 5378.

35 Brookhart, M.; Volpe, A. F.; Lincoln, D. M.; Horvath, I. T.; Millar, J. M. J. Am. Chem. Soc. 1990, 112, 5634.

36 Timonen S., Pakkanen, T. T.; Pakkanen, T. A. J. Mol. Catal. 1996, 111, 267

37 Johnson, L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc. 1995, 117, 6414.

38 Mackenzie, P. B.; Moody, L. S.; Killian, C. M.; Lavoie, G. G. WO Patent 9962968 (1998).

39 Johnson, L. K.; Mecking, S.; Brookhart, M. J. Am. Chem. Soc. 1996, 118, 267.

40 Emert, J.; Rossi, A.; Gindelberger, D. E.; Stanat, J. E.; Stokes, J. P.; Sher, J. US Patent 6066603 (2000).

41 Heinemann, J.; Mulhaupt, R.; Brinkmann, P.; Luinstra, G. Macromol. Chem. Phys. 1999, 200, 384.

42 Marques, M. M.; Correia, S. G. ; Ascenso, J. ; Dias, A. R. ; Biais, M. ; Rausch, M. D. ; Chien, J. C. W. Metalorganic Catalysts for Synthesis and Polymerization ; Kaminsky, W., Ed,; Springer-Verlag: Heidelberg, 1999, p 576.

43 Derek, P. G.; Svejda, S. A.; Onate, E.; Killian, C. M.; Johnson, L. K.; White, P. S.; Brookhart, M. Macromol. 2000, 33, 2320.

44 Brookhart, M. S.; Johnson, L. K.; Arthur, S. D.; Feldman, J.; Kreutzer, K. A.; Bennett, A. M. A.; Coughlin, E. B.; Ittel, S. D.; Parthasarathy, A.; Tempel, D. J. US Patent 5886224 (1999).

16

45 Johnson, L. K.; Killian, C. M.; Arthur, S. D.; Feldman, J.; McCord, E. F.;

McLain, S. J.; Kreutzer, K. A.; Bennett, A. M. A.; Brookhart, M. S. WO Patent 9623010 (1995).

46 Brown, K. A. WO Patent 9717380 (1996).

47 Mecking, S.; Johnson, L. K.; Wang, L.; Brookhart, M. J. Am. Chem. Soc. 1998, 120, 888.

48 Arthur, S. D.; Brookhart, M. S. US Patent 5891963 (1999).

49 Geprags, M.; Mulhaupt R. WO Patent 9947569 (1998).

50 Suzuki, JP Patent 11292918 (1999).

51 Szabo, M. J.; Jordan, R. F.; Michalak, A.; Piers, W. E.; Weiss, T.; Yang, S.; Ziegler, T. Organometallics 2004, 23, 5565.

52 Szabo, M. J.; Galea, M. N.; Michalak, A.; Yang, S.; Groux, L. F.; Piers, W. E.; Ziegler, T. J. Am. Chem. Soc. 2005, 127, 14692.

53 Correia, S. G.; Marques, M. M.; Ascenzo, J. R.; Ribeiro, A. F. G.; Gomes, P. T.; Dias, A. R.; Blais, M.; Rausch, M. D.; Chien, J. C. W. J. Polym. Sci., Part A, 1999, 37, 2471.

54 Klabunde, U. US Patent 4698403 (1987).

55 Klabunde, U. US Patent 4716205 (1987).

56 Klabunde, U. US Patent 4906754 (1990).

57 Klabunde, U. US Patent 5175326 (1992).

58 Johnson, L. K.; Bennett, A. M. A.; Ittel, S. D.; Wang, L.; Parthasarathy, A.; Hauptman, E.; Simpson, R. D.; Feldman, J.; Coughlin, E. B. WO Patent 9830609 (1997).

59 Bansleben, D. A.; Grubbs, R. H. WO Patent 9842664 (1997).

17

60 Wang, C.; Friedrich, S.; Younkin, T. R.; Li, R. T.; Grubbs, R. H.;

Bansleben, D. A.; Day, M. W. Organometallics 1998, 17, 3194.

61 Younkin, T.; Grubbs, R. H. Science 2000, 287, 460.

62 Waltman, A. W.; Younkin, T. R.; Grubbs, R. H. Organometallics 2004, 23, 5121.

63 Drent, E.; Dijk, R.; Ginkel, R.; Oort, B.; Pugh, R. I. Chem. Commun. 2002, 744.

64 Luo, S.; Vela, J.; Lief, G. R.; Jordan, R. F. J. Am. Chem. Soc. 2007, 129, 8946.

65 Kochi, T.; Noda, S.; Yoshimura, K.; Nozaki, K. J. Am. Chem. Soc. 2007, 129, 8948.

18

19

Chapter 2. Synthesis and stability of a cationic α-diimine

palladium methyl catalyst

2.1. Introduction

Since cationic α-diimine Ni and Pd alkyl complexes were reported by Brookhart and co-workers in 19951 to catalyze the polymerization of ethene (Scheme 1.5), the area of olefin polymerization catalysis by late transition metal complexes has developed strongly. In this catalyst family, sterically hindered ligand systems prevent fast chain transfer processes, which are responsible for the olefin dimerization or oligomerization catalysis found more usually for the late transition metals.2,3,4 The low reaction rates of these Pd species at low temperature and the lability of the coordinated ether molecule in the parent catalyst are very suitable for mechanistic studies by NMR spectroscopy. Olefin coordination, insertion and subsequent transformation (chain walking, chain transfer processes) for ethene, α-olefin and alkyl acrylates with these Pd species were studied via variable temperature NMR by Brookhart and co-workers.1,5,6,7,8 For this purpose, a convenient synthesis of the palladium catalyst [(N^N)PdMe(OEt2)][BAF] (2, N^N: ArN=CMe-CMe=NAr with Ar: 2,6-diisopropylphenyl, BAF: B[3,5-(CF3)2C6H3]4) is essential. A general synthesis procedure has been reported by Brookhart and co-workers1, involving protonation of one of the methyl groups of α-diimine palladium dimethyl complexes 1 by the strong acid H(Et2O)2BAF in ether (Scheme 2.1). The ionic methyl complex 2 was reportedly isolated as a relatively stable solid. Nevertheless, when this procedure was repeated by us, the formation of

20

side products was observed. As the subsequent purification of crude 2 proved difficult, the synthesis of the palladium catalyst 2 was studied in more detail.

1 2

BAF-

OEt2

PdMeN

NMePd

MeN

N

H+(OEt2)2BAF-

Et2OCH4

BAF- = {B[3,5-(CF3)2C6H3]4} -NN=N

N

Scheme 2.1

2.2. Syntheses of precursors: α-diimine palladium dimethyl complex and H(Et2O)2BAF

Various ways to prepare α-diimine palladium dimethyl complex 1, precursor to the ionic catalyst 2, have been reported. These are summarized in Scheme 2.2.8 All the methods begin from the same starting material (COD)PdCl2.9 First, (COD)PdCl2 was monomethylated by tetramethyl tin to give (COD)PdMeCl.10 After the COD ligand was displaced by N-aryl α-diimine (N^N), the second methyl group was introduced by reaction of (N^N)PdMeCl with MgMe2 to give the α-diimine dimethyl complex 1 in moderate yield, 24.6% based on (N^N)PdMeCl.1 A more effective published route is direct dimethylation of (COD)PdCl2 by 2 equiv of methyl cuprate, followed by workup with aqueous KCN, to give (COD)PdMe2. 11 (COD)PdMe2 is a thermally labile compound, so once prepared it should be

21

quickly reacted with N-aryl α-diimine to give the desired complex 1.12

(COD)PdCl2 (COD)PdMeClCl

PdMeN

N

(COD)PdMe2

MePd

MeN

N

1

MeLi 0.5MgMe2

1.methyl cuprate

2.KCN

SnMe4

- COD

N^N

- COD

N^N

Scheme 2.2 In our hands, it turned out that MeMgBr is a very efficient methylating agent for the synthesis of 1. Not only can it smoothly methylate intermediate (N^N)PdMeCl to give 1, it also allows a direct synthesis of 1 starting from (COD)PdCl2. From (COD)PdCl2, (COD)PdMe2 can be generated in situ through reaction of 2 equiv of MeMgBr with (COD)PdCl2 at –30 oC. Addition of the diimine, followed by extraction with toluene and crystallization from toluene/pentane affords pure 1 in an isolated yield of 46% based on (COD)PdCl2 (Scheme 2.3).

(COD)PdCl2 (COD)PdMe2

MePd

MeN

N

1

2 MeMgBr N^N

- 2 MgBrCl - COD

Scheme 2.3

22

The syntheses of NaBAF (BAF- = {B[3,5-(CF3)2C6H3]4}-) and the Brønsted acid H(Et2O)2BAF have been reported in literature. NaBAF was made by reaction of [3,5-bis(trifluoromethyl)phenyl]magnesium bromide with NaBF4.13 Brookhart reported that, after an ether solution of NaBAF was treated with HCl gas, H(Et2O)2BAF could be crystallized. 14 But subsequently it was found8 that the product is a mixture of H(Et2O)2BAF and H(H2O)2BAF, even after pre-drying NaBAF under vacuum and storing its solution in ether for 12 h over 4 Å molecular sieves. The H2O stems from incompletely dried NaBAF (which has been isolated from aqueous workup). Bahr et al. have reported that NaBAF could be dried at 110 oC for 6 h, although details of the process were not given.15 The room temperature 1H NMR spectrum of H(Et2O)2BAF in CD2Cl2 shows a very broad peak for [H(Et2O)2]+, which is difficult to observe and is unresolved from any [H(H2O)2]+ formed by remaining traces of water. We found that, when the 1H NMR spectrum of H(Et2O)2BAF in CD2Cl2 was measured at –60 oC, the proton in [H(Et2O)2]+ shows a sharp resonance at δ 16.63 ppm and that the [H(H2O)2]+ ion also shows a clear singlet at δ 13.05 ppm. The presence and quantity of the water adduct can thus be easily determined by the relative integral of these two peaks. After investigating various methods to dry NaBAF, we found that drying at 130 oC under vacuum for 6 h is sufficient and does not cause significant decomposition. The dried NaBAF was dissolved in ether and then treated with a 1M HCl solution in ether. The latter is easier to handle and to dose than HCl gas. The NaCl precipitate was filtered off, and H(Et2O)2BAF was crystallized from the solution by cooling. The H(Et2O)2BAF thus obtained contains less than 5 mol% of H(H2O)2BAF, based on low temperature 1H NMR spectroscopy.

23

2.3. Synthesis of the cationic α-diimine palladium catalyst 2

According to the procedure for the synthesis of the cationic α-diimine palladium complex 2 described by Brookhart et al.1, an equimolar mixture of (N^N)PdMe2 and H(Et2O)2BAF were cooled to –78 oC. Following the addition of ether, the solution was allowed to warm and stirred briefly for 15 min at room temperature. The solution was then filtered and the solvent was removed in vacuo to give a pale orange powder (94.5 % yield). When this published procedure was repeated, we observed the presence of a large amount (approximately 50 mol% Pd) of a side-product (compound 3, Scheme 2.4, the selective synthesis and characterization of which will be described in section 2.4) that could not readily be removed e.g. by recrystallization. We therefore explored this synthesis at different reaction conditions. The content of 3 in the crude product increased from about 10% to 50% when the reaction was performed in ether at room temperature instead of at –50 oC. Below –50 oC, no by-product 3 was found but the reaction of palladium dimethyl with H(Et2O)2BAF became very slow so the reaction time became prohibitively long. During reactions at this temperature, gradual decomposition of product 2 became significant and interfered with its final purity.

PdMeN

N CH2

PdN

N

BAF-

3

Scheme 2.4 When the reaction was carried out in dichloromethane, the C-Cl bond of CH2Cl2 was activated as well during the reaction. This side-reaction led to the formation of palladium chloride complexes 4 and 5 (Scheme 2.5). Complex 4 was previously reported as the product of the reaction of α-diimine palladium methyl chloride with NaBAF.8

24

5

4H+(OEt2)2BAF-

CH2Cl2

Me

Pd

ClN

N MePd

N

NBAF-

2

Cl

Pd

ClN

NPd

N

N2BAF-

MePd

MeN

N

Scheme 2.5 Complexes 4 and 5 were obtained from a reaction in which a CH2Cl2 solution of (N^N)PdMe2 was slowly added to a H(Et2O)2BAF solution in CH2Cl2 at –60 oC. After the reaction mixture was stored at –80 oC for around 20 h, some black precipitate (Pd) had formed. After filtering and concentrating the solution, and addition of some pentane, initially a small amount (3.5% yield) of 5 was obtained. Further concentrating the solution, addition of pentane and cooling to -30oC afforded orange crystals of 4 in 65% isolated yield. In addition it was observed that complex 5 can be produced by allowing 4 to stand in CH2Cl2 solution over prolonged times. Complex 5 was characterized by single-crystal X-ray diffraction (Figure 2.1). 16 Selected bond distances and angles are listed in Table 2.1. The dinuclear Pd2 molecule contains a center of symmetry. The two Pd centers are bridged by two chlorides to form a planar 4-membered ring that is essentially coplanar with the Pd-diimine plane (Pd1-N11-C113-C115-N12). The Pd-Cl bond distances (2.342 and 2.331 Å) are somewhat longer than typical Pd-Cl bonds in mononuclear complexes, e.g. in (N^N)PdMeCl (2.300 Å).8

25

Figure 2.1. ORTEP structure of complex 5 (cation)

Table 2.1. Selected bond distances (Å) and angles (o) of complex 5

_____________________________________________________________ Pd1-Cl1 2.3418(7) Pd1-N11 1.998(2) Pd1-Cl1a 2.3310(6) Pd1-N12 2.006(2) Pd1-Cl1- Pd1a 96.10(3) N11-Pd1-N12 79.17(9) Cl1-Pd1-Cl1a 83.90(2) Cl1a-Pd1-N12-C115 178.43(18) Cl1-Pd1-N11-C113 177.87(18) _____________________________________________________________

26

2.4. Synthesis and characterization of {[(N^N)Pd]2(µ-CH3)(µ-CH2)}[BAF] (3)

It turns out that the side-product {[(N^N)Pd]2(µ-CH2)(µ-CH3)}[BAF] (3), which is formed during the synthesis of the cationic methyl complex 2, derives from a reaction of 2 with the starting material, the Pd dimethyl complex 1 (Scheme 2.6). This was observed by making a CD2Cl2 solution at low temperature of equimolar amounts of 1 and 2 and monitoring this by NMR spectroscopy while gradually warming up. Above -50oC, slow formation of 3 was observed. Performing the reaction at ambient temperature resulted in immediate formation of 3 and methane (δ 0.16 ppm in 1H NMR spectra).

Scheme 2.6 In a preparative scale experiment, 1 equiv of H(Et2O)2BAF and 2 equiv of palladium dimethyl complex 1 were combined in CH2Cl2 at room temperature. Subsequent crystallization by concentrating the solution and layering with pentane afforded crystalline 3 in 69% isolated yield. The nature of 3 was established by NMR spectroscopy, elemental analysis and single

-Et2O H+(OEt2)2BAF-

MePd

MeN

N

-CH4

BAF-PdMeN

N OEt2

BAF-

CH2Pd

Me N

NH

PdMeN

NPd

MeN

N CH2

PdN

N

BAF-

1 2

3

-Et2OMe

PdMe N

N

27

crystal X-ray diffraction (Figure 2.2, Table 2.2). In NMR spectrum of 3, besides the normal assignments of the diimine ligand and the counteranion, there are two 1H resonances found at δ 5.71 and 0.04 ppm, which exhibit a 2:3 integral ratio. By HSQC (Heteronuclear Single Quantum Coherence), the correlated 13C resonances are found at δ 129.66 and -46.60 ppm respectively. Furthermore the 2-D spectrum HMBC (Heteronuclear Multiple Bond Coherence) indicates the 1H resonance at δ 5.71 ppm is from CH2 with a C-H coupling constant of 139 Hz, and the 1H resonance at δ 0.04 ppm is from CH3 with a C-H coupling constant of 120 Hz. Several methylene bridged dinuclear complexes were reviewed by Casey and Audett.17 The 1H and 13C chemical shifts (δ 5.71 and 129.66 ppm) of µ-CH2 in complex 3 are in the ranges (δ 5-11 and 100-210) observed for µ-CH2

in dinuclear complexes.18 Similar 1H and 13C chemical shifts of δ 5.19 and 158.4 ppm were also found in Cp2(Cl)Ti(µ-CH2)Pt(Me)(PMe2Ph)2 for the methylene bridge.21 The upfield 1H and 13C chemical shifts (δ 0.04 and -46.60 ppm) of µ-CH3 in complex 3 are also typical for the methyl bridge in dinuclear complexes, e.g. δ -5.65 (1H) and δ-27.1 ppm (13C) in Ru2(µ-CH3)(µ-CH2)2(PMe3)6.20

The X-ray structure of 3 shows a "butterfly" cation of two palladiums bridged by a methylene and a methyl group and paired with only one anion, which means the palladium dimer is a monocation. The hydrogen atoms of the bridging methyl and methylene groups were located in the refinement: 3 hydrogens on C57 and 2 hydrogens on C58. The Pd2C2 4-membered ring is folded, with a dihedral angle C57-Pd1-Pd2-C58 of 137.1(2)o. Such a folded ring geometry has been reported for some related methylene- and methyl-bridged metal dimers, e.g. (µ-C5H5)2Ti(µ-CH2)(µ-CH3)Rh(COD),19 Ru2(µ-CH3)(µ-CH2)2(PMe3)6

20 and Cp2(Cl)Ti(µ-CH2)Pt(Me)(PMe2Ph)2. 21 The Pd-N distances trans to the Pd-µ-CH2 bonds (2.149(2) and 2.192(3) Å) are longer than those trans to the Pd-µ-CH3 bonds (2.098(2) and 2.077(2) Å), corroborating the assignment of the two groups.

28

Figure 2.2. ORTEP structure of complex 3 (cation)

Table 2.2 Selected bond distances (Å) and angles (o) of complex 3.

_____________________________________________________________ Pd1-Pd2 2.7347(3) Pd1-N1 2.098(2) Pd1-N2 2.149(2) Pd1-C57 2.153(4) Pd1-C58 2.007(4) Pd2-N3 2.192(3) Pd2-N4 2.077(2) Pd2-C57 2.237(4) Pd2-C58 2.000(4) N1-Pd1-N2 76.03(9) N3-Pd2-N4 75.93(9) Pd1-C57-Pd2 77.04(13) Pd1-C58- Pd2 86.07(16)

29

Pd1-C57-H57 132(2) Pd2-C57-H57 74(2) Pd1-C57-H57’ 85(4) Pd2-C57-H57’ 76(4) Pd1-C57-H57’’ 78(2) Pd2-C57-H57’’ 135(3) Pd2-C58-H58 97(2) Pd2-C58-H58’ 121(2) C57- Pd1- Pd2-C58 137.1(2) C58-Pd2-C57-Pd1 29.92(15) N1- Pd1- Pd2-C58 44.3(2) N1- Pd1- Pd2-C57 178.62(18) _____________________________________________________________

2.5. Structure calculations on {[(N^N)Pd]2(µ-CH3)(µ-CH2)}[BAF] (3)

Dinuclear d8-d8 metal complexes (reviewed by Dedieu22) have, in some cases, been studied by quantumchemical calculations.23,24,25,26 In order to gain more understanding of the bonding in the binuclear core of 3, DFT calculations were performed by Dr. P. H. M. Budzelaar (Radboud University Nijmegen) on (µ-CH2)(µ-CH3)[Pd(ArNCRCRNAr)]2

+ for three different ligand substitution patterns: a "naked" system 3a (Ar = R = H), an intermediate system 3b (Ar = 2,6-Me2C6H3, R = Me) and the complete system 3c (Ar = 2,6-iPr2C6H3, R = Me). Pictures of the optimized geometries of the molecules are shown in Figure 2.3; relevant geometrical parameters for the core of the complexes are shown in Figure 2.4. The "butterfly" shape of the core is similar for the three complexes but becomes flatter with increasing steric demand of the ligand. This can be traced to increasing repulsion between the aryl substituents on different ligands, as is clear from Figures 2.3-B, C. In the series, the Pd-CH2 distances change very little, while the Pd-CH3 distances increase by nearly 0.04 Å and the Pd-Pd distance increases by over 0.1 Å. Clearly, the Pd-Pd interaction is very "soft" and, if bonding at all, represents much less than a single bond. Its softness also explains the rather large error in the length calculated for it.

30

A

B

C

Figure 2.3. Calculated structures of (A) 3a, (B) 3b and (C) 3c.

31

The most interesting aspect of the core concerns the location of the CH3 hydrogen atoms. Regardless of the nature of the substituents, we find a more or less symmetrical arrangement for the CH3 group, two of the hydrogens having weak agostic interactions with one palladium atom each (Pd H = 　

2.20-2.26 Å, C-H bond elongation ca 0.02 Å). Despite this symmetric arrangement, the highly substituted 3c shows a definite asymmetry in the Pd-CH3 distances (2.215 vs. 2.257 Å). This is not related to any agostic interaction but rather to repulsive interactions between the iPr groups on the Ar substituents. Loss of symmetry lowers these repulsions, allowing the iPr groups to interlock. However, asymmetry also results in bond length changes in the core, and the Pd-CH3 "half-bonds" respond more strongly to this (∆ = 0.042 Å) than the Pd-CH2 single bonds (∆ = 0.015 Å). The similarity of the X-ray structure of 3 in Figure 2.5 and the calculated structure in Figure 2.4 shows the agreement of experimental and calculation results.

1.9931.9891.982

1.9951.9921.997

2.6962.7542.818

2.1962.2192.215

2.2192.2412.2572.203

2.2382.235

2.2542.2592.228

1.1261.1221.120

1.1241.1211.122

1.1071.1071.106

∠PdCH Pd, PdCH Pd123.3128.0131.0

2 3

Figure 2.4. Key geometrical parameters (Å,°) calculated for core of 3a,3b, and 3c.

32

Figure 2.5. Core of X-ray structure of 3. (The other atoms are omitted for clarity)

All calculations were carried out with the Turbomole program27 coupled to the PQS Baker optimizer.28 Geometries were fully optimized as minima or transition states at the bp8629/RIDFT30 level using the Turbomole SV(P) basisset on all atoms (pseudopotential basis on the metal). Zero-point energy (ZPE) and thermal corrections were not included. For 3c, a large number of conformations of the interlocking iPr groups are possible. We have tested only a few of them, and the structure shown in Figure 2.5 probably does not represent the global minimum. However, for the purpose of the present analysis we think this is not a serious problem.

2.007(4) 2.000(4)

2.153(4) 2.237(4)

0.90(4) 0.98(3)

0.69(5)

2.7347(3

<Pd1C58Pd2, Pd1C57Pd2 = 137.1(2)o

33

2.6. Reaction of dipalladium complex 3 with 2-butyne.

Several µ-methylene complexes were reviewed by Puddephatt. 31 Most examples contain additional bridging ligands. Three types of organometallic reactions were mentioned: CO insertion in [Ru2Cp2(CO)4(µ-CH2)] to give a ketene derivative;32 reductive elimination in [Cp2Ti(µ-CH2)2TiCp2] to give ethene by coupling of the two methylene groups; 33 cleavage of [LAu(µ-CH2)PtMe2Cl(bipyridine)] with HCl to give [PtMe3Cl(bipyridine)].34 By analyzing the structure of dipalladium complex 3, two possible reaction pathways of this compound may be anticipated. One is the cleavage of the dinuclear species 3 by Lewis basic reagents to two mononuclear species: one fragment is a cationic Pd-Me species similar to 2, the other a neutral Pd alkylidene species (Scheme 2.7, left hand side). In this way, 3 could be a precursor of Pd=CH2 species, similar to Tebbe’s reagent Cp2Ti(µ-CH2)(µ-CH3)Al(CH3)2,35 which reacted with internal acetylenes to produce a titanacyclobutenes.36,37 At the same time, the cationic Pd-Me species can show its specific reactivity (3 is able to polymerize ethene, probably through this fragment45). On the other hand, 3 can retain its dinuclear character by the opening of just one side of the Pd2C2 ring by a Lewis basic reagent (Scheme 2.7, right hand side). The result is a species with three different Pd-C bonds that can each show reactivity. In order to gauge the character of 3, its reactivity with various reagents has briefly been explored.

Scheme 2.7

L2L

Cleaving Opening3

L = Lewis base

CH2

PdLN

N LPd

Me N

N CH2

PdMeN

NPd

N

N CH2

PdN

NPd

N

N

MeL

34

Dipalladium complex 3 is only moderately reactive. It does not react with benzophenone, THF, acetonitrile or even methanol at ambient temperature over 24 hours. Solid 3 can be heated at 50 oC for 12 h under N2 without notable decomposition. When the dipalladium complex 3 was reacted with 2-butyne in diethyl ether at ambient temperature, gradually a black precipitate of Pd(0) formed. After 12 h the brown solution was filtered and a yellow solid precipitated upon addition of pentane to the solution. Recrystallization from CH2Cl2 / pentane afforded yellow crystals of the cationic Pd-(η3-1,1,2-trimethylallyl) complex 6, (yield: 35 mol% on Pd). A suitable single-crystal was found for an X-ray structure determination (Figure 2.6, Table 2.3).

Figure 2.6. ORTEP structure of complex 6 (cation)

35

Table 2.3 Selected bond distances (Å) and angles (o) of complex 6

_____________________________________________________________ Pd11-C129 2.097(5) Pd11-C130 2.129(4) Pd11-C132 2.187(4) Pd11-N11 2.041(3) Pd11-N12 2.178(3) C129-C130 1.376(8) C130-C132 1.381(6) C129-Pd11-C130 38.0(2) C130-Pd11-C132 37.30(15) C129-C130-C132 124.3(5) C130-C132-C133 124.6(5) C130-C132-C134 117.5(5) C133-C132-C134 113.0(4) Pd11-C129-C130-C132 52.0(4) C129-C130-C132-C133 165.0(5) C129-C130-C132-C134 41.4(6) C131-C130-C132-C134 157.7(4) _____________________________________________________________ As with many Pd-complexes with this α-diimine ligand (see e.g. Chapter 3), the structure refinement of the X-ray diffraction data was complicated by a disorder problem: from the solution it was clear that the electron density of the Pd atom appeared to be spread out, indicating conformational disorder. A disorder model with two different positions for the Pd (Pd11: Pd12 = 50:50) was used in the refinement. This disorder must have obviously influenced the ligand positions, but this could not be satisfactorily modeled. Nevertheless, the non-hydrogen atom connectivity is unambiguous, indicating the presence of an η3-1,1,2-trimethylallyl ligand. The delocalization in the allyl fragment is seen from the C-C bond distances (C129-C130 = 1.376(8) Å and C130-C132 = 1.381(6) Å). The Pd-C bond distances ( Pd11-C129 = 2.097(5) Å, Pd11-C130 = 2.129(4) Å and Pd11-C132 = 2.187(4) Å, with the longest bond to the Me2-substituted carbon) are in the typical range of η3–allylpalladium complexes.38,39,40 The three methyl substituents (C131- C133-C134) are somewhat tilted out of the allyl plane as indicated by the torsion angle of 165.0(5)o for C129-C130-C132-C133 and 157.7(4)o for C131-C130-C132-C13. The 1H NMR chemical shifts of the allyl methylene protons (δ 3.06 ppm

36

H-syn and 3.01 ppm H-anti) and the corresponding 13C NMR resonance at δ 61.66 ppm (JCH = 143 Hz) in complex 6 are typical for η3-allyl palladium complexes, e.g. δ 3.49 ppm (H-syn) and 3.23 ppm (H-anti) and δ 63.8 ppm (13C) in an α-diimine η3-2-methylallyl palladium complex.41 Three different methyl substituents on the allyl group can be seen at δ(1H) 1.87, 1.10 and 0.51 ppm. These data are comparable with those of a related 1,1,2-trimethylallyl complex of the type [Pd(CH2-CMe-CMe2)(N-N)]BF4 with δ 4.18 and 3.54 ppm for syn and anti allylic protons and δ 2.03, 1.19 and 0.60 ppm for the methyl substituents.42

70%63

Pd(0)

PdMeN

N CH2

PdN

N

BAF-

ether

25oC 12 hrs

BAF-

NPd

N

insertion PdN

NPd

N

N

MePd

MeN

N CH2

PdN

N

Ligand

Scheme 2.8

The proposed reaction route is depicted in Scheme 2.8. The reaction can be initiated by coordination of 2-butyne to one of the Pd centers, involving the opening of one of the Pd(µ-Me)Pd sides of the Pd2C2 ring. The alkyne can

37

then insert into the adjacent Pd-CH2 bond. In the resulting species, a reductive elimination, connecting the vinyl carbon of the inserted alkyne with the Pd-bound methyl, can generate the 1,1,2-trimethylallyl group. The reduced Pd atom then loses its diimine ligand (free α-diimine is observed by NMR) and precipitates as Pd(0).

2.7. Reaction of dipalladium complex 3 with CO.

ether

CO

7

BAF-PdMeN

N CH2

PdN

N

BAF-

3

NPd

N CO

Me

Scheme 2.9 Dipalladium complex 3 was treated with CO in diethyl ether solution. After stirring at room temperature overnight, a black precipitate had formed. After filtration, the carbonyl methyl complex 7 was isolated by recrystallization from the filtrate in 40% yield based on Pd (Scheme 2.9). A suitable crystal of 7 was found for X-ray diffraction. The X-ray structure determination of complex 7 was complicated to some extent by positional disorder between the Me and CO groups, but does allow an identification of the compound (Figure 2.7 and Table 2.4). It is a regular square planar 4-coordinate Pd complex. The 13C NMR spectrum of complex 7 reveals a chemical shift of δ 176.1 for C≡O and its IR spectrum a C≡O stretching frequency at 2155 cm-1. A similar complex (phen)Pd(CH3)(CO)+ (phen = 1,10-phenanthroline) shows related features at δ 176.3 and 2130 cm-1, respectively.43

38

Figure 2.7. ORTEP structure of complex 7 (cation)

Table 2.4 Selected bond distances (Å) and angles (o) of complex 7

_____________________________________________________________ Pd11-C1291 1.94(3) Pd11-C1301 2.029(18) Pd11-N11 2.113(3) Pd11-N12 2.086(3) O112-C1301 0.90(2) C1291-Pd11-C1301 85.5(9) O112- C1301- Pd11 179.1(17) N11-Pd11-N12 77.11(11) C1291-Pd11-N11-C113 174.5(8) C1301-Pd11-N12-C115 175.5(6) _____________________________________________________________ In this reaction, the identified product is essentially the product expected to be formed in the reaction of the cationic Pd-Me complex 2 with CO. It thus seems likely that, in this case, the dipalladium complex 3 is cleaved in such a way as to liberate a “Pd=CH2” species as well. Nevertheless, the fate of this

39

part of the molecule could not be established, apart from the observation that it eventually leads to deposition of Pd(0).

2.8. Conclusions The synthesis of the cationic α-diimine palladium methyl complex [(N^N)PdMe(OEt2)][BAF] (2) has been facilitated by an improved preparation of the starting materials (N^N)PdMe2 and [H(OEt2)2][BAF]. It was established that the synthesis conditions have to be well-controlled, to prevent the formation of the dinuclear (µ-methyl)(µ-methylene) species 3 and attack of solvent C-Cl bonds. The dipalladium complex 3 is an unusual ionic (µ-methyl)(µ-methylene) species. Dinuclear palladium complexes have been much less investigated than mononuclear species.44 After completion of our work, Baird et al. reported some similar dipalladium complexes.45 These complexes have the same dipalladium cation as the 3 but with different anions. Nevertheless, the reactivity of the species was left unexplored. Our studies showed some interesting reactivity features of 3, in which the methyl and methylene groups either end up on the same metal (reaction with 2-butyne) or on different metals (reaction with CO). Especially the possibility to generate transient “(N^N)PdCH2” species is intriguing and would deserve further study.

2.9. Experimental General considerations. All experiments were carried out under an atmosphere of purified nitrogen using standard Schlenk and glovebox techniques, unless mentioned otherwise. CH2Cl2 and CD2Cl2 (Aldrich) were distilled from CaH2 prior to use. Ether, toluene and pentane (Aldrich,

40

anhydrous, 99.8%) were dried by passing over columns of Al2O3, BASF R3-11 supported Cu oxygen scavenger, and molecular sieves (Aldrich, 4 Å) under a nitrogen atmosphere prior to use. NMR spectra were recorded on Varian Gemini 300/500 spectrometers in NMR tubes equipped with a Teflon (Young) valve. The 1H NMR spectra were referenced to resonances of residual protons in the deuterated solvents, δ 5.32 ppm for CD2Cl2. The 13C NMR spectra were referenced to the carbon resonances of the deuterated solvents, δ 53.8 ppm for CD2Cl2. Chemical shifts are given relative to tetramethylsilane (downfield shifts are positive); J values are given in Hertz. Assignments of the resonances of the organometallic products were made using information from COSY and HSQC (Heteronuclear Single Quantum Coherence) and/or HMBC (Heteronuclear Multiple Bond Coherence) spectra. NMR data for [BAF]- anion. The 1H and 13C NMR resonances of [BAF]- anion in CD2Cl2 are the same in the spectra for the different cationic palladium complexes at various temperatures. They are give here and will not be repeated in the listing of the other spectra: 1H NMR (CD2Cl2, 500 MHz, 25 oC): δ 7.73 (s, 8H, Ho), 7.57 (s, 4H, Hp); 13C NMR (CD2Cl2, 126 MHz, 25 oC): δ 162.13 (q, JCB = 50.6, Cipso), 135.20 (Co), 129.31 (qq, JCF = 31.5, JCB = 2.9, Cm), 124.99 (q, JCF = 273.1, CF3), 117.83 (septet, JCF = 3.9, Cp). Elemental analyses were performed by Kolbe Mikroanalytisches Laboratorium, Mülheim an der Ruhr, Germany. Starting materials. The compounds (N^N)PdMeCl,1 (COD)PdCl2 (COD = cyclooctadiene), 46 and NaBAF14 were prepared according to published procedures. NaBAF was dried at 130 oC under vacuum for 6 h. 2-Butyne was dried on molecular sieves (Aldrich, 4 Å) overnight. Other compounds were used as received. Modified synthesis of [H(Et2O)2][BAF]. NaBAF (3.13 g, 3.53 mmol) was

41

dissolved in 70 mL of Et2O. 7 mL of a 1M HCl solution in ether was added at 0 oC. A white precipitate was formed immediately. After stirring at 0 oC for 10 mins, the solution was filtered and cooled at –80 oC overnight. White crystals were isolated. Yield 3.14 g, 3.10 mmol, 88 %. By low-temperature 1H NMR, there is less than 5 % [H(H2O)2][BAF] presented in the final products. 1H NMR (CD2Cl2, 500MHz, -60 oC) δ 16.63 (s, [H(Et2O)2]) 13.05 (s, [H(H2O)2]), 7.73 (s, 8H, BAF: H0), 7.56 (s, 4H, BAF: Hp), 3.96 (q, 8H, J = 7.2, OCH2), 1.30 (t, 12H, J = 7.0, CH3). Modified synthesis (I) of (N^N)PdMe2 (1): From (N^N)PdMeCl and MeMgBr. To a suspension of (N^N)PdMeCl (3.37 g, 6.32 mmol) in 150 mL Et2O was added 2.1 mL of a 3 M solution of MeMgBr (6.32 mmol) in Et2O at 25 oC. The suspension was stirred for 2 h. The Et2O was removed by vacuum. The brown solid was extracted twice with 150 mL of toluene. The combined extracts were concentrated to 20 mL. After addition of 80 mL pentane, the mixture was cooled at –80 oC overnight. Brown crystals of 1 were isolated and washed by 10 mL pentane 3 times, then dried in vacuum. Yield 2.60 g, 4.80 mmol, 76 %. The 1H NMR of the product shows the product to be pure 1.1 Anal. Calcd for (C30H46N2Pd): C, 66.59; H, 8.57; N, 5.18. Found: C, 66.45; H, 8.71; N, 4.88. Modified synthesis (II) of (N^N)PdMe2 (1): From (COD)PdCl2 and MeMgBr. To a suspension of (COD)PdCl2 (610 mg, 2.1 mmol) in 20 mL Et2O, pre-cooled to –80 oC, was added 1.4 mL 3 M solution of MeMgBr (4.3 mmol) in Et2O. The suspension was warmed to –30 oC and stirred for 0.5 hr. Diimine ligand (N^N) (864 mg, 2.1 mmol) was then added. The brown mixture was stirred for 2 hr at –30 oC and for 1 h at room temperature. The solvent was removed by vacuum. The dark brown solid was extracted with 50 mL toluene. The extract was concentrated to 10 mL, followed by addition of 30 mL pentane. Cooling at –80 oC overnight afforded brown crystals of 1. Yield 537 mg, 0.99 mmol, 47%. The 1H NMR spectrum of the product shows it to be pure 1.1

42

Modified synthesis of [(N^N)PdMe(EtO)2][BAF] (2). (N^N)PdMe2 (1, 0.597 g, 1.10 mmol) and [H(Et2O)2][BAF] (1.117 g, 1.10 mmol) were weighed into a Schlenk flask. Ether (50 mL) was slowly condensed into the flask at –196 oC. The mixture was stirred at –80 oC for 1 h, at –60 oC for 0.5 h and at –50 oC for 1.5 h and then warmed up to room temperature. After the solution was filtered, the solvent was removed by vacuum. The solid was washed with 10 mL of pentane 3 times. Yield 2 (1.046 g, 0.71 mmol) 65%, containing about 10 mol% of 3 based on NMR data. Synthesis of [(N^N)2Pd2(µ-CH3)(µ-CH2)][BAF] (3). To (N^N)PdMe2 (1, 1.095 g, 2.02 mmol) and [H(Et2O)2][BAF] (1.045 g, 1.03 mmol) was added 25 mL of CH2Cl2 at room temperature. The mixture was stirred for 5 min. The solution was filtered and concentrated to 15 mL. Then 30 mL of pentane was added, and the mixture was cooled at –80 oC overnight. Red crystals of 3 were isolated (1.36 g, 0.71 mmol, 69%). 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 7.21 – 7.10 (m, 12H, Haryl), 5.71 (s, 2H, µ-CH2), 2.77 and 2.60 (septet, 4H each, J = 7.3, iPrH), 2.02 and 1.95 (s, 6H each, N=CMe), 1.12, 1.07 and 1.06 (d, 12H, 12H and 24H, J = 7.3, iPrMe), 0.04 (s, 3H, µ-CH3). 13C NMR (CD2Cl2, 126MHz, 25 oC) δ 173.41 and 170.93 (N=CMe), 144.82 and 140.86 (Ar Cipso), 137.65 and 137.09 (Ar Co), 129.66 (t, JCH = 139, µ-CH2), 127.44 and 127.18 (Ar Cp), 124.11 (Ar Cm), 28.77 and 28.50 (iPr CH), 23.78, 23.63, 23.43 and 20.48 (iPr Me), 20.05 and 17.62 (N=CMe), -46.60 (q, JCH = 120, µ-CH3). Anal. Calcd for (C90H97BF24N4Pd2): C, 56.47; H, 5.11; N, 2.93. Found: C, 56.12; H, 5.11; N, 2.85. [(N^N)2Pd2Me2(µ-Cl)][BAF] (4) and [(N^N)2Pd2(µ-Cl)2][BAF]2 (5). To a solution of [H(Et2O)2][BAF] (0.948 g, 0.94 mmol) in 40 mL CH2Cl2 pre-cooled to -60 oC was added a (N^N)PdMe2 (0.500 g, 0.92 mmol) solution in 10 mL CH2Cl2 dropwise over 0.5 h with stirring. After stirring at -50 oC for 1.5 h and at -30 oC for 1 h, the mixture was stored at -80 oC for 19 hours. Black precipitate was filtered off. The filtrate was concentrated to 15 mL by vacuum and 30 mL of pentane was added. The solution was cooled to -80 oC for 7h. Orange crystals were filtered off and washed with pentane to give

43

0.032 g (0.016 mmol) of isolated 5 in 3.5% yield. The filtrate was concentrated to 5 mL by vacuum and then combined with 30 mL of pentane. The brown solution was slowly cooled to -30 oC overnight. Crude product 4 was isolated by filtration and further recrystallized from CH2Cl2 / pentane to give pure (by NMR) complex 4 (0.583 g, 0.30 mmol, 65%). The 1H and 13C NMR are identical to the literature data.8 5 : 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 7.40 (t, 4H, J = 7.5, Ar: Hp), 7.16 (d, 8H, J = 7.5, Ar: Hm), 2.70 (septet, 8H, J = 7.0, iPrH), 2.20 (s, 12H, N=CMe), 1.29 and 1.16 (d, 24H, each, J = 7.0, iPr Me). 13C NMR (CD2Cl2, 126MHz, 25 oC) δ 184.34 (N=CMe), 139.18 (Ar: Co), 138.82 (Ar: Cipso), 132.08 (Ar: Cp), 125.32 (Ar: Cm), 30.32 (iPr CH, 23.71 and 23.58 (iPr Me), 21.48 (N=CMe). Reaction of [(N^N)2Pd2(µ-CH3)(µ-CH2)][BAF] (3) with 2-butyne: synthesis of 6. To a solution of 3 (210 mg, 0.11 mmol) in 5 mL Et2O was added 2-butyne (10 µL, 4.3 mmol). The solution was stirred overnight at room temperature. A black precipitate had formed. The brown solution was filtered. Yellow solid was precipitated by addition of 5 mL of pentane. The solid was recrystallized from CH2Cl2 / pentane to give complex 6 (56 mg, 0.038 mmol, 35% yield based on Pd). 1H NMR (CD2Cl2, 300MHz, 25 oC) δ 7.42 – 7.31 (m, 12H, Haryl), 3.06 and 3.01 (d, 1H each, J = 1.5, Hallyl), 2.90 and 2.70 (septet, 2H each, J = 6.9, iPr CH), 2.27 and 2.22 (s, 3H each, N=CMe), 1.87 (s, 3H, MeC=CMe2), 1.44, 1.36, 1.32, 1.31, 1.25, 1.19, 1.18 and 1.15 (d, 3H each, J = 6.9, iPr Me), 1.10 (s, 3H, MeC=CMe’Me), 0.51 (s, 3H, MeC=CMe’Me). 13C NMR (CD2Cl2, 126MHz, 25 oC) δ 175.71 and 175.09 (N=CMe), 144.03 and 142.09 (Ar Cipso), 137.89, 137.17, 136.97 and 136.79 (Ar Co), 128.83 and 128.76 (Ar Cp), 125.12, 125.08, 124.92 and 124.79 (Ar Cm), 92.04 (MeC=CMe2), 61.66 (t, JCH = 143, CH2(Me)C=CMe2), 29.65, 29.49, 29.44 and 29.23 (iPr CH), 24.55, 24.22, 23.69, 23.52, 23.43, 23.24, 23.20 and 22.92 (iPr Me), 23.69 (q, JCH = 126, MeC=CMe’Me), 21.70 and 20.97 (N=CMe), 20.64 (q, JCH = 124, MeC=CMe’Me), 20.11 (q, JCH = 130, MeC=CMe2). Anal. Calcd for (C66H63BF24N2Pd): C, 54.39; H, 4.36; N, 1.92. Found: C, 54.39; H, 4.39; N, 1.90.

44

Reaction of [(N^N)2Pd2(µ-CH3)(µ-CH2)][BAF] (3) with CO: synthesis of 7. On a vacuum line, about 1.4 mmol (10 eqv.) CO was condensed into a Schlenk flask pre-charged with a solution of 3 (274 mg, 0.14 mmol) in 10 mL CH2Cl2 at –196 oC. The solution was warmed to room temperature and stirred for 4 hours. Black precipitate was formed. After filtration, 20 mL pentane was added to the solution. A brown precipitate was filtered and recrystallized from CH2Cl2 / pentane to yield crystals of 7 (79 mg, 0.056 mmol, 40% based on Pd). 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 7.50 – 7.39 (m, 12H, Haryl), 2.84 and 2.68 (septet, 2H each, J = 6.6, iPr CH), 2.41 and 2.30 (s, 3H each, N=CMe), 1.40, 1.32, 1.29 and 1.24 (d, 6H each, J = 6.6, iPr Me), 0.86 (s, 3H, PdMe). 13C NMR (CD2Cl2, 126MHz, 25 oC) δ 176.10 (CO), 173.91 and 173.45 (N=CMe), 142.64 and 141.11 (Ar Cipso), 138.59 and 136.88 (Ar Co), 130.26 and 129.82 (Ar Cp), 125.41 (Ar Cm), 29.86 and 29.78 (iPr CH), 23.84, 23.78, 23.58 and 23.21 (iPr Me), 22.54 (PdMe), 20.40 (N=CMe). IR: 2155 cm-1, ν (CO). Anal. Calcd for (C62H55BOF24N2Pd): C, 52.54; H, 3.91; N, 1.98. Found: C, 52.38; H, 4.06; N, 1.92. X-ray Crystallographic analysis With inert-atmosphere handling techniques, suitable crystals were mounted on top of a glass fiber and aligned on a Bruker47 SMART APEX CCD diffractometer (Platform with full three-circle goniometer). Data integration and global cell refinement was performed with the program SAINT. The final unit cell was obtained from the xyz centroids of total reflections after integration. Intensity data were corrected for Lorentz and polarization effects, scale variation, for decay and absorption: a multi-scan absorption correction was applied, based on the intensities of symmetry-related reflections measured at different angular settings (SADABS)48, and reduced to Fo

2. The program suite SHELXTL was used for space group determination (XPREP)47. Reduced cell calculations did not indicate any higher metric lattice symmetry49 and examination of the final atomic coordinates of the structure did not yield extra crystallographic or metric symmetry elements50,51. The structures were solved by Patterson methods and extension of the model was

45

accomplished by direct methods applied to difference structure factors using the program DIRDIF 52 . The positional and anisotropic displacement parameters for the non-hydrogen atoms and isotropic displacement parameters for hydrogen atoms were refined on F2 with full-matrix least-squares procedures minimizing the function Q = ∑h[w(│(Fo

2) - k(Fc

2)│)2], where w = 1/[σ2(Fo2) + (aP)2 + bP], P = [max(Fo

2,0) + 2Fc2] / 3, F0

and Fc are the observed and calculated structure factor amplitudes. Crystallographic data can be found in Table 2.5 and Table 2.6. For complex 3, all hydrogen atoms were located from the difference Fourier map, and their coordinates and isotropic displacement parameters were refined; the hydrogen atoms connected to C57 and C58 were refined with one common isotropic displacement parameter. In complex 6, a disorder model with two slightly different positions for the Pd center (both with 0.5 occupancy factor, Pd-Pd’ = 0.25 Å) was used in the refinement. This disorder is likely to be correlated with a different positioning of the 1,1,2-trimethyl-allyl ligand, but this could not be satisfactorily modeled. It is reflected in the apparent large thermal ellipsoids of the carbon atoms in that fragment. For complex 7, refinement was complicated by a configurational-disorder problem: the -CH3 (C130) and the -C=O (C129-O110) ligand positions are alternatively occupied, and a 50:50 disorder model with bond restraints was used in the refinement. A residual peak of 3.2 e/Å3 between C113 and C115 was observed; this peak was refined as partly occupied by a Pd (refined s.o.f. = 0.064), implying a disorder of the cation by an approximately 180° rotation about the line N11-N12. The disorder phenomena may account for the observed unrealistic displacement parameters for some atoms when allowed to vary anisotropically, suggesting dynamic disorder as a consequence of the configurational- and rotational- disorder.

46

Table 2.5. Crystallographic data for complex 3 and 5.

____________________________________________________________ 3 5 Formula C90H97BF24N4Pd2 C120H104Cl2B2F48N4Pd2 FW 1914.40 2819.43 Crystal system triclinic triclinic Space group P-1 P-1 a (Å) 12.1587(6) 14.2360(5) b (Å) 16.9380(9) 15.8017(5) c (Å) 22.215(1) 28.040(1) α (deg) 96.781(1) 96.921(1) β (deg) 99.356(1) 93.878(1) γ (deg) 94.819(1) 97.930(1) V (Å3) 4457.7(4) 6179.0(4) θ range (deg) 2.44-28.53 2.32-29.60 Z 2 2 ρ calc (g.cm-3) 1.426 1.515 F(000) 1956 2840 µ(Mo Kα ), cm-1 5.00 4.56 Temp (K) 100(1) 100(1) Reflections 22466 31143 Parameters 1474 1623 wR(F2) 0.1272 0.1358 Weighting (a, b) 0.0556, 3.9093 0.0737, 7.4449 R(F) 0.0501 0.0481 GooF 1.033 1.037 _____________________________________________________________

47

Table 2.6. Crystallographic data for complex 6 and 7.

_____________________________________________________________ 6 7 Formula C66H63BF24N2Pd C62H55BOF24N2Pd FW 1457.43 1417.32 Crystal system monoclinic triclinic Space group P21/n P-1 a (Å) 19.257(1) 12.6166(6) b (Å) 16.8713(9) 15.3958(8) c (Å) 20.808(1) 16.6283(8) α (deg) 87.064(1) β (deg) 104.3700(10) 80.019(1) γ (deg) 81.913(1) V (Å3) 6548.8(6) 3148.2(3) θ range (deg) 2.35-28.94 2.23-25.78 Z 4 2 ρ calc (g.cm-3) 1.478 1.495 F(000) 2960 1432 µ(Mo Kα ), cm-1 3.93 4.08 Temp (K) 100(1) 100(1) Reflections 17412 12142 Parameters 869 864 wR(F2) 0.1587 0.1450 Weighting (a, b) 0.0786, 8.0395 0.0708, 4.3064 R(F) 0.0573 0.0542 GooF 1.026 1.022 _____________________________________________________________

48

2.10. References

1 Johnson, L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc. 1995, 117, 6414.

2 Wilke, G. Angew. Chem., Int. Ed. Engl. 1988, 27, 185.

3 Mohring, V. M.; Fink, G. Angew. Chem., Int. Ed. Engl. 1985, 24, 1001.

4 Peuckert, M.; Keim, W. Organometallics 1983, 2, 594.

5 Johnson, L. K.; Mecking, S.; Brookhart M. J. Am. Chem. Soc. 1996, 118, 267.

6 Mecking, S.; Johnson, L. K.; Wang, L.; Brookhart, M. J. Am. Chem. Soc. 1998, 120, 888.

7 Shultz, L. H.; Tempel, D. J.; Brookhart, M. J. Am. Chem. Soc. 2001, 123, 11539.

8 Tempel, D. J.; Johnson, L. K.; Huff, R. L.; White, P. S.; Brookhart, M. J. Am. Chem. Soc. 2000, 122, 6686.

9 J. Chatt, L. M. Vallarino, L. M. Venanzi, J. Chem. Soc., 1957, 3413.

10 Rulke, R. E.; Vrieze, K. Inorg. Chem. 1993, 32, 5769.

11 Rudler-Chauvin, M.; Rudler, H. J. Organomet. Chem. 1977, 134, 115.

12 Van Asselt, R.; Elsevier, C. J. Organometallics 1994, 13, 706.

13 Iwamoto, H.; Sonoda, T.; Kobayashi, H. Tetrahedron Lett. 1983, 24, 4703.

14 Brookhart, M.; Volpe, Jr. A. F. Organometallics 1992, 11, 3920.

15 Bahr, S. R.; Boudjouk, P. J. Org. Chem. 1992, 57, 5545.

49

16 A structure of the same cation but with the [AlCl4]- anion was reported

previously by Kang, M.; Sen, A. J. Am. Chem. Soc. 2002, 124, 12080.

17 Casey, C. P.; Audett, J. D. Chem. Rev. 1986, 86, 339.

18 Herrmann, W.A. Adv. Organomet. Chem. 1982, 20, 159.

19 Park, J. W.; Grubbs, R. H. J. Am. Chem. Soc. 1986, 108, 6402.

20 Hursthouse, M. B.; Wilkinson G. J. Am. Chem. Soc. 1979, 101, 4128.

21 Ozawa, F; Park, J; Grubbs, R. H. J. Am. Chem. Soc. 1989, 111, 1319.

22 Dedieu, A. Chem. Rev. 2000, 100, 543.

23 Krogmann, K. Angew. Chem., Int. Ed. Engl. 1969, 8, 35.

24 Aullon, G.; Alemany, P.; Alvarez, S. J. Organomet. Chem. 1994, 478, 75.

25 Capdevila, M.; Clegg, W.; Gonzales-Duarte, P.; Jarid, A.; Lledos, A. Inorg. Chem. 1996, 35, 490.

26 Aullon, G.; Ujaque, G.; Lledos, A.; Alvarez, S.; Alemany, P. Inorg. Chem. 1998, 37, 804.

27 (a) Ahlrichs, R.; Bär, M.; Baron, H.-P.; Bauernschmitt, R.; Böcker, S.; Ehrig, M.; Eichkorn, K.; Elliott, S.; Furche, F.; Haase, F.; Häser, M.; Hättig, C.; Horn, H.; Huber, C.; Huniar, U.; Kattannek, M.; Köhn, A.; Kölmel, C.; Kollwitz, M.; May, K.; Ochsenfeld, C.; Öhm, H.; Schäfer, A.; Schneider, U.; Treutler, O.; Tsereteli, K.; Unterreiner, B.; Von Arnim, M.; Weigend, F.; Weis, P.; Weiss, H. Turbomole Version 5, january 2002. Theoretical Chemistry Group, University of Karlsruhe; (b) Treutler, O.; Ahlrichs, R. J. Chem. Phys. 1995, 102, 346.

50

28 (a) PQS version 2.4, 2001, Parallel Quantum Solutions, Fayetteville,

Arkansas, USA (the Baker optimizer is available separately from PQS upon request); (b) Baker, J. J. Comput. Chem. 1986, 7, 385.

29 Becke, A.D. Phys. Rev. A 1988, 38, 3089; Perdew, J.P. Phys. Rev. B 1986, 33, 8822.

30 Eichkorn, K.; Weigend, F.; Treutler, O.; Ahlrichs, R. Theor. Chem. Acc. 1997, 97, 119.

31 Puddephatt, R. J. Polyhedron 1988, 7, 767.

32 Lin, Y. C.; Calabrese, J. C.; Wreford, S. S. J. Am. Chem. Soc. 1983, 105, 1680.

33 Ott, K. C.; Grubbs, R. H. J. Am. Chem. Soc. 1981, 103, 5922.

34 Arsenault, G. J.; Crespo, M.; Puddephatt, R. J. Organometallics 1987, 6, 2255.

35 Tebbe, F. N.; Parshall, G. W.; Reddy, G. S. J. Am. Chem. Soc. 1978, 100, 3611.

36 Tebbe, F. N.; Parshall, G. W.; Ovenall, D. W. J. Am. Chem. Soc. 1979, 101, 5074.

37 McKinney, R. J.; Tulip, T. H.; Thorn, D. L.; Coolbaugh, T. S.; Tebbe, F. N.; J. Am. Chem. Soc. 1981, 103, 5584.

38 Haaren, R. J.; Leeuwen, P. W. N. M. Inorg. Chem.2001, 40, 3363.

39 Burckhardt, U.; Togni, A. Organometallics 1996, 15, 3496.

40 Togni, A.; Salzmann, R. J. Am. Chem. Soc. 1996, 118, 1031.

41 Mechria, A.; Bouachir, F. Tetrahedron Lett. 2000, 41, 7199.

51

42 Felice, V.; Cucciolito, M. E.; Renzi, A. D.; Ruffo, F.; Tesauro, D. J.

Organomet. Chem. 1995, 493, 1.

43 Rix, F. C.; Brookhart, M.; White, P. S. J. Am. Chem. Soc. 1996, 118, 4746.

44 Murahashi, T.; Kurosawa, H. Coord. Chem. Rev. 2002, 231, 207.

45 Brownie, J. H.; Baird, M. C. Organometallics 2003, 22, 33.

46 Drew, D.; Doyle, J. R. Inorg. Syn. 1972, 13, 52.

47 Bruker (2000). SMART, SAINT, SADABS, XPREP and SHELXTL/NT. Area Detector Control and Integration Software. Smart Apex Software Reference Manuals. Bruker Analytical X-ray Instruments. Inc., Madison, Wisconsin, USA.

48 Sheldrick, G.M. (2001). SADABS. Version 2. Multi-Scan Absorption Correction Program. University of Göttingen, Germany.

49 Spek, A.L. J. Appl. Cryst. 1988, 21, 578-579.

50 Le Page, Y. J. Appl. Cryst. 1987, 20, 264-269.

51 Le Page, Y. J. Appl. Cryst. 1988, 21, 983-984.

52 Beurskens, P.T., Beurskens, G., Gelder, R. de, García-Granda, S., Gould, R.O., Israël, R. & Smits, J.M.M. (1999). The DIRDIF-99 program system, Crystallography Laboratory, University of Nijmegen, The Netherlands.

52

53

Chapter 3. Reactivity with oxygen-functionalized olefins

3.1. Introduction The incorporation of polar functionalities into polyolefin materials is a subject of considerable interest. It can provide a means to modify important properties such as the dyeability, compatibility, permeability and wettability of these intrinsically apolar materials.1,2 They can also be used as reactive groups for the cross-linking of polyolefin elastomers. The catalytic incorporation of functionalized comonomers would be the most versatile method for the synthesis of such materials. Most polar functionalities relevant to this purpose are Lewis bases (esters, ethers, amides etc.). As all olefin polymerization catalysts are Lewis acids (a property required to capture the olefin substrate), competition for binding to the catalyst center between the olefin and the polar functionality is a problem that needs to be addressed. The intramolecular coordination of the polar functionalities after comonomer insertion (giving stable chelates) exacerbates this problem. Add to this the possibility of irreversible reactions between the polar functionality and the polar metal-carbon bond of the catalyst (especially for Brønsted acid functionalities and polar unsaturated species such as carbonyls), and it is clear that the catalytic incorporation of polar comonomers into polyolefins is not an easy task to accomplish. One approach is to minimize the strength of the interaction between the Lewis basic functionality and the metal center by choosing the catalyst and substrate combination to obtain a hard-soft Lewis acid-base mismatch. As the functionalities of interest are mostly hard Lewis bases, catalysts based on the heavier late transition-metals are likely to be the most tolerant to these functionalities. Brookhart et al. have reported copolymerizations of ethene and acrylate comonomers with [(α-diimine)PdMe][BAF] ([BAF] = {B[3,5-(CF3)2C6H3]4})

54

catalysts to give branched copolymers. Nevertheless, the polymerization is rather slow and the functional groups are only found at the end of the branches, rather than incorporated in the polymer backbone.3 For this system, low-temperature NMR studies allowed the observation of all critical intermediates and reaction steps for this catalysis. As shown in Scheme 3.1, acrylate insertion occurs predominantly in a 2,1-fashion, yielding a strained four-membered chelate ring in which the carbonyl oxygen atom is coordinated to palladium. This insertion is followed by a series of β-H elimination steps and reinsertions, expanding the ring stepwise to a six-membered chelate complex. The coordination of the carbonyl oxygen to the palladium centre in this chelate resting state makes the subsequent monomer insertion more difficult. This was found to be responsible for the significantly lower polymerization rates of the ethene/acrylate copolymerization in comparison to ethene or propene homopolymerizations with the same catalyst system.4,5,6

Scheme 3.1 The ratio of incorporation of ethene and methyl acrylate into the copolymers is governed by both the equilibrium ratio of the alkyl ethene and alkyl methyl acrylate complexes and their relative rates of migratory insertion. Even though the rate of migratory insertion of methyl acrylate is faster than that of

NPd

OEt2

MeN

NPd

MeN

COOMe

2,1-insEt2OCOOMe

CD2Cl2

-60oCN

PdN O

OMe

Me

NPd

N O OMe

Me

-20oC

NPd

N OOMe

k=0.01

NPd

N O

OMe

55

ethene at low temperature, there is an overwhelming preference for ethene to bind into the electrophilic Pd(II) center relative to the electron-deficient olefin, methyl acrylate. In order to get significant incorporation of MA into the copolymer, a high concentration of MA has to be applied. But increasing MA concentrations decreases the overall rate of polymerization due to the stability of the chelate complex. Decreasing the steric demand of the substituents on the diimine ligand, or incorporating more electron-donating substituents on the diimine, increased the acrylate incorporation, probably through improved binding of MA to the catalyst center. But this also lowered the molecular weight of the copolymers produced.4

ORn

CopolymerOR

No polymerization

n

+

+

Pd-catalyst

Pd-catalyst

Scheme 3.2

A recent paper described the use of the cationic α-diimine Pd-catalysts for the copolymerization of ethene with α-olefinic ethers as comonomers.7 It was found that polymerization activity could only be observed when the olefin and the ether functionalities in the comonomers were separated by at least one quaternary carbon atom (Scheme 3.2). It was proposed that, after insertion of the comonomer, the chain-walking process translates the Pd-center to the carbon atom in the β-position relative to the ether group. Catalyst deactivation was then presumed to take place through β-alkoxy transfer to the metal center (Scheme 3.3). Although no clear evidence for this process was presented, there is some precedent for β-alkoxy transfer processes.8

56

Scheme 3.3 3.2. Approach taken in this chapter These observations prompted us to study the interaction of simple allylic ether substrates with cationic α-diimine Pd catalysts, and attempt their copolymerization with ethene. The substrates used in this study include AEE (Allyl Ethyl Ether), ADMA (Acrolein Dimethyl Acetal), and VDO (2-vinyl-1,3-dioxolane). As in these substrates the heteroatom is not adjacent to the double bond, 1,2-insertion into the Pd-alkyl bond of the catalyst is expected as the first reaction step. The reactions between (α-diimine)Pd methyl complexes and the allylic ether comonomers were investigated by two ways (Scheme 3.4). Method 1 involves the reaction of the comonomers with one equivalent of the pre-formed ionic species [(N^N)PdMe(Et2O)][BAF] (2). These reactions can be followed by variable temperature NMR in CD2Cl2 from –50 oC up to 25 oC, in which the substrate coordination and subsequent insertion can be observed step by step. For the isolation of the insertion products (usually chelate complexes) this method can also be used for reactions on a preparative scale. However, for this purpose this method is not very convenient, as the cationic Pd catalyst [(N^N)PdMe(Et2O)][BAF] is relatively cumbersome to prepare and isolate in pure form on a larger scale (see Chapter 2). Preparative scale reactions are more conveniently carried out using (N^N)PdMeCl as precursor, which is reacted with NaBAF in presence of excess of the functionalized monomers (Method 2). This method was previously used by Brookhart et al. to prepare chelates derived from methyl acrylate.3 In contrast with the

Insertion

Chain-walking NPd

N OMe

R

NPd

N OMe

R

57

reaction employing the preformed cationic methyl complex, these reactions need 1-2 days at ambient temperature to go to completion, so that the final products have to be thermally stable at ambient temperature. The compounds thus obtained were used for further study and for polymerization catalysis. A part of this study was published: W. Li, X. Zhang, A. Meetsma, B. Hessen, J. Am. Chem. Soc. 2004, 126, 12246.

Scheme 3.4

3.3. Functional monomer - allyl ethyl ether (AEE)

3.3.1 Formation of the Pd-AEE chelate Following Method 1 as described above, the reaction of [(N^N)PdMe(Et2O)][BAF] (2) with AEE in CD2Cl2 was followed by variable temperature NMR experiments. (Scheme 3.5) The coordination of the C=C bond of AEE to the Pd(II) center is much stronger than that of diethyl ether. At –50 oC, the 1H NMR resonance at δ 3.20 ppm for the OCH2 protons of the

f

NPd

MeN

OEt2

NaBAFf

f = functional group

NPd

N

f1,2-insertion

Method 1

Method 2N

PdMeN

Cl

58

coordinated ether has already been replaced completely by the resonance at δ 3.38 ppm for non-coordinated ether. The three resonances for the protons of the olefinic group of AEE were shifted from δ 5.87 (m), 5.22 (d, J = 17.4 Hz) and 5.11 (d, J = 10.4 Hz) for the non-coordinated olefin to δ 4.91 (m), 4.39 (d, J = 15.7 Hz) and 4.31 (d, J = 8.4 Hz) for the coordinated olefin. At –30 oC, the product from AEE 1,2-insertion into the Pd-Me bond, chelate complex 9, was gradually formed, as seen e.g. from the doublet 1H NMR resonance of the former Pd-Me group at δ 0.79 (d, J = 7.0 Hz, PdCH2CHMeC). Upon increasing the temperature to 25 oC, full conversion to the chelate 9 was observed. The chelate 9 appears to be stable towards β-OEt elimination reaction: no evidence for β-OEt elimination was observed, even after prolonged reaction times.

1,2-insertion

9

ONPd

N+

O

NPd

N Me

+

O

+

OEt2NPd

N Me

Scheme 3.5

The same reaction performed on a preparative scale afforded [(N^N)Pd(CH2CHMeCH2OEt)][BAF] (9) in 61 % isolated yield. A single crystal X-ray structure determination (Figure 3.1 and Table 3.1) corroborated the proposed chelate structure of compound 9. It is the product of 1,2-insertion of AEE into the Pd-Me bond. The OEt group of the ether moiety is coordinated to the metal center, resulting in a 5-membered chelate ring. The crystal contains a substantial amount of disorder in which two orientations of the chelate are present. It is quite frequently observed in complexes of this highly substituted diimine ligand (see also in later chapters in this thesis), that it strongly dominates the crystal packing, and allows a certain amount of freedom to the other groups attached to the metal. Nevertheless, the non-hydrogen atom connectivity could be established unequivocally. The

59

methyl group of the original Pd-Me species occupies an equatorial position on the chelate ring.

Figure 3.1 ORTEP structure of complex 9 (cation), major conformation. Table 3.1 Selected bond distances (Å) and angles (o) of complex 9 _____________________________________________________________ Pd1-N11 2.071(5) Pd1-N12 2.113(5) Pd1-C129A 2.189(13) Pd1-O11A 2.008(9) O11A-C132A 1.45(2) O11A-C133A 1.39(2) C129A-C130A 1.529(13) C130A-C131A 1.55(2) C132A-C130A 1.53(2) C133A-C134A 1.42(3) N11-Pd1-N12 77.11(19) O11A-Pd1-C129A 79.8(4) C132A-O11A-C133A 114.3(11) Pd1-C129A-C130A 101.5(8) C129A-C130A-C131A 111.5(10) C129A-C130A-C132A 109.3(11) C132A-C130A-C131A 110.8(12) O11A-C132A-C130A 106.0(11)

60

Pd1-O11A-C132A 119.0(8) Pd1-O11A-C133A 125.8(9) O11A-C133A-C134A 116.6(16) C129A-Pd1-O11A-C132A 7.1(10) O11A-Pd1-C129A-C130A -31.3(8) C129A-C130A-C132A-O11A -48.3(15) C131A-C130A-C132A-O11A -171.5(11) C130A-C132A-O11A-C133A -170.4(12) C134A-C133A-O11A-C132A -58(2) _____________________________________________________________ The 1H and 13C NMR spectra of the chelate complex 9 were assigned by a combination of COSY, HSQC and HMBC techniques. The chelate Pd-CH2 group displays resonances at 1.67 ppm (1H, br. 2H) and 47.77 ppm (13C, t, JCH = 113 Hz), the chelate C-CH2 group at 3.53 and 3.14 ppm (2JHH = 7.6 Hz) and 83.69 (t, JCH = 143 Hz). 3.3.2 Polymerization attempts and catalyst deactivation To see whether the 5-membered chelate ring in 9 can be opened by ethene, the complex was tested for ethene homopolymerization. The compound was shown to be active, yielding 1.94 g of branched polyethene at a catalyst activity of 142.6 kg(PE)/mol.h under the applied conditions (15 mL CH2Cl2 solvent, 6.8 mol of 9, 5 bar ethene, room temperature, 2 h run time). Despite the fact that 9 is competent in the homopolymerization of ethene, the copolymerization of ethene (E) with AEE could not be achieved with this catalyst. Instead, rapid catalyst deactivation was observed. When applying a very high catalyst concentration (100 mg, 67.0 µmol of 9 in 5 mL of CH2Cl2 and 1 mL, 8.82 mmol of AEE), analysis of the reaction mixture from an attempted copolymerization with ethene (5 bar, ambient temperature) allowed identification of the organometallic product formed. Orange crystals of the ionic palladium allyl species [(N^N)Pd(η3-CH2CHCH2)][BAF] (10) were isolated in high yield (87 mg, 61.5 µmol, 92%). In addition, ethanol was

61

detected in the reaction mixture by GC/MS analysis. The nature of complex 10 was determined by NMR spectroscopy (Figure 3.2 and 3.3), elemental analysis and a single crystal structure determination (Figure 3.4 and Table 3.2).

ppm1234567843.12

22.1532.59

4.4274.82

9.5810.15

9.7510.01

30.6210.00

32.54101.30

Figure 3.2. 1H NMR (CD2Cl2, 500 MHz) spectrum of complex 10

ppm2040608010012014016018011.50

12.8911.40

4.3147.36

3.886.90

1.76

Figure 3.3. 13C NMR (CD2Cl2, 126 MHz) spectrum of complex 10

62

The Cs-symmetric nature of complex 10 was illustrated by its 1H and 13C spectra. Characteristic allylic 1H and 13C resonances are found at δ 5.64 ppm and 121.03 ppm (CH), 3.35 and 3.04 ppm, 65.78 ppm (CH2). The same allyl Pd cation was reported very recently as being formed as catalyst deactivation product in the copolymerization of silyl vinyl ethers with olefins.9

Figure 3.4 ORTEP structure of complex 10 (cation) Table 3.2 Selected bond distances (Å) and angles (o) of complex 10 _____________________________________________________________ Pd11-N11 2.093(2) Pd11-N12 2.097(2) Pd11-C129 2.105(4) Pd11-C1301 2.071(11) Pd11-C131 2.116(4) C129-C1301 1.271(18) C131-C1301 1.298(13) N11-Pd11-N12 76.92(9) C129-Pd11-C131 68.97(16) C129-Pd11-C1301 35.4(5) C131-Pd11-C1301 36.1(4) C129-C1301-C131 137.0(14) Pd11-C129-C1301-C131 -42.6(19) _____________________________________________________________

63

As in the chelate complex 9, disorder is also presented in the X-ray structure of 10. A fraction of the molecules is present in an orientation that corresponds to an approximate 180° rotation around the line N11-N12. Nevertheless, the η3-allylic nature of the compound was unambiguously established. It is unlikely that the fast deactivation and the formation palladium allyl species 10 and ethanol are resulting from β-OEt transfer to the metal center (although Jordan and coworkers have established that cationic Pd-OR complexes can slowly react with propene -which would be liberated by a β-OEt transfer- to generate ethanol and a Pd-allyl species10). An alternative route to the formation of this allylic species is proposed in Scheme 3.6. After initial insertion of ethene into the Pd-C bond of the chelate 9, the resulting Pd-alkyl species is in equilibrium with its derived hydride-olefin complex (implied in chain-transfer processes for these catalysts11,12,13). This olefin is exchanged for AEE comonomer which leads to a hydride-AEE complex. This intermediate can conduct either 1,2-insertion of the comonomer (reforming a chelate similar to 9) and then continue the polymerization with ethene, or could eliminate a molecule of ethanol, forming an η3-allyllic species 10. This ethanol elimination may proceed via intramolecular protonation of the ether by the cationic Pd-H species (which is known to have Brønsted acid properties14,15). This process was implied in the catalytic substitution of allyl alcohol with amines to give allylic amines, catalyzed by cationic palladium bis(phosphinidene)cyclobutene hydride complexes. 16 A similar allyl Pd complex was also formed from [(L2)PdCl][OTf] complexes and allylic alcohols. 17 The mechanism of this type of reactions is still being studied.18,19,20

64

R

ethene

OEt

10

OEtR

NPd

N R

H

NPd

N

H

+

+

+

+

NPd

N

9

+

ONPd

N

- EtOH

+

NPd

N

NPd

N

HOEtH

Scheme 3.6 Complex 10 could be recovered from the reaction mixtures of E/AEE copolymerization attempts in >90% yield, indicating that its formation is the main catalyst transformation process in operation. Attempts to effect ethene homopolymerization with 10 as catalyst showed that this species is not active under the present conditions used for catalytic ethene polymerization. It can thus be concluded that the deactivation of Pd catalyst in E/AEE copolymerization is associated with the formation of the palladium allyl species 10, possibly following the route in Scheme 3.6. The direct synthesis of the Pd-AEE 5-membered chelate 9 by reaction of (N^N)PdMeCl with NaBAF and excess AEE in CH2Cl2 (Method 2) was not

65

successful. Instead, this reaction already yielded predominantly the allyl complex 10. Apparently, at room temperature in the presence of excess AEE, and with the longer reaction times required for this reaction than for the reaction of 1 with AEE, decomposition to give the cationic palladium allyl complex can occur. Insufficient evidence is present to establish which mechanism is operative in this case. 3.4. Functional monomer - allyl dimethyl acetal (ADMA) 3.4.1 Formation of the Pd-ADMA chelate Having observed the catalyst deactivation process for allyl ethyl ether (AEE) substrate, the substrate allyl dimethyl acetal (ADMA) was investigated. This substrate is more sterically hindered than AAE, and has the advantage that, if successfully incorporated into polyolefins, under acid conditions the acetal functions can be converted into reactive aldehyde groups that could be used for cross-linking or further functionalization.21,22

1,2-insertion

NPd

N Me

+

11

O

O

O

O

+

NPd

N

O O

+

OEt2NPd

N Me

Scheme 3.7

Following Method 1 as described above, the reaction of [(N^N)PdMe(Et2O)][BAF] (2) with ADMA in CD2Cl2 was followed by variable temperature NMR experiments (Scheme 3.7). At –50 oC, the OCH2 resonance at δ 3.20 for coordinated ether has already completely been replaced by the resonance at δ 3.38 for non-coordinated ether. The proton

66

resonances of coordinated ADMA are found at δ 4.59, 4.38 (d, J = 15.4 Hz) and 4.32 (d, J = 8.5 Hz), indicating that the substrate is coordinated to the metal through the olefinic moiety. A small amount of Pd-ADMA chelate 11 was observed at -40 oC, showing the 1H NMR resonance of the former Pd-bound methyl group at δ 0.74 (d, J = 7.0 Hz, PdCH2CHMe). Upon warming, the concentration of 11 increased gradually, until at -20 oC chelate 11 is the only product visible. This product was stable in solution up to at least 50 oC. On a preparative scale, ADMA was reacted with (N^N)PdMeCl and NaBAF in CH2Cl2 solvent. The 5-membered chelate complex {(N^N)Pd[CH2CHMeCH(OMe)2]}[BAF] (11) could be obtained in 91% isolated yield from this reaction (Scheme 3.8). Apparently, under these conditions the possible decomposition to produce Pd-allylic species is much slower than for the analogous reaction with AEE (see previous section).

Cl

Me

NPd

N ADMA

NaBAF

CH2Cl2

11

NPd

N

O O BAF-

+

Scheme 3.8

The 1H and 13C NMR spectra of Pd-ADMA chelate 11 were interpreted by combination of COSY, HSQC and HMBC techniques. The characteristics of the chelate ring are similar to those in the Pd-AEE chelate 9. The methyne group CH(OMe)2 shows its resonances at δ 4.12 (1H, d, J = 5.5 Hz) and δ 119.48 (13C, d, JCH = 172 Hz). The coordinated and uncoordinated OMe groups show distinct resonances at δ 2.84 vs. 3.12 (1H) and δ 56.55 (q, JCH = 145) vs. 60.51 (JCH = 147) (13C). A single crystal X-ray structure of complex 11 was determined (Figure 3.5,

67

Table 3.3). Despite the presence of a certain amount of disorder in the chelate fragment, it can be clearly seen that a 5-membered chelate ring is formed through the intramolecular coordination of one of the OMe groups of the inserted ADMA molecule.

Figure 3.5 ORTEP structure of complex 11 (cation)

Table 3.3 Selected bond distances (Å) and angles (o) of complex 11 _____________________________________________________________ Pd11-O111 2.052(14) Pd11-C1321 2.01(3) O111-C1291 1.43(2) O111-C1301 1.41(2) C1311-C1321 1.56(3) C1301-C1311 1.52(3) C1311-C1341 1.55(3) O121-C1331 1.40(2) O121-C1301 1.381(16)

68

Pd11-O111-C1291 124.9(12) Pd11-O111-C1301 113.6(10) Pd11-C1321-C1311 102.2(15) O111-Pd11-C1321 83.3(6) O111-C1301-O121 108.5(12) O111-C1301-C1311 106.1(13) O121-C1301-C1311 122.5(16) C1301-C1311-C1321 107(2) C1301-O111-Pd11-C1321 -1.6(15) O111-C1301-C1311-C1321 54(2) C1291-O111-Pd11-C1301 151(2) C1331-O121-C1301-O111 78.5(18) C1341-C1311-C1321-Pd11 -168.6(16) _____________________________________________________________ 3.4.2 (Co)-polymerization of ethene and allyl dimethyl acetal Pd-ADMA chelate 11 was first tested as a catalyst for ethene homopolymerization (Table 3.4, top entry). Like the acrylate chelate complex described by Brookhart et al.,4 and the Pd-AEE chelate 9, the Pd-ADMA chelate 11 also readily polymerizes ethene to branched polyethene. For comparison with other polyethenes described later, a 1H NMR spectrum of this branched polyethene is shown in Figure 3.6. It should be noted that in the experiments described in Table 3.4, the catalyst activity as expressed in kg(PE) mol(Pd)-1 h-1 is low due to the high catalyst concentration and long reaction times employed, which aim at maximizing the absolute polymer yield. Table 3.4. Ethene/ADMA copolymerization catalyzed by 11.

Conditiona ADMA (mmol)

PolymerYield (g)

Activity (kg mol-1 h-1)

Mw (x10-3)

Mw/Mn ADMA Contentb (mol %)

A 0 3.50 13.0 186 1.5 0

A 4.2 1.21 4.5 105 1.7 0.29

69

Conditiona ADMA (mmol)

PolymerYield (g)

Activity (kg mol-1 h-1)

Mw (x10-3)

Mw/Mn ADMA Contentb (mol %)

A 8.4 0.64 2.4 52 2.0 0.32

A 17.4 0.44 1.6 31 2.3 0.66

B 4.2 1.60 1.2 50 1.7 0.72

B 8.4 1.50 1.1 32 1.7 1.30

B 19.1 1.20 0.9 16 1.7 1.67

B 26.1 0.52 0.4 13 1.6 2.02

a) A: 13.4 mol of 2, 15 mL CH2Cl2, 5 bar ethene, room temperature, 20 h run time; B: 67

mol of 　 2, 3 ml CH2Cl2, otherwise as A. b) determined by 1H NMR.

ppm12345678910

Figure 3.6 1H NMR (CDCl3, 500 MHz) spectrum of branched PE produced by catalyst 11.

70

Apparently the Pd-O dative interaction in the chelate is sufficiently labile to allow the coordination of ethene molecule that can subsequently insert into the Pd-C bond (Scheme 3.9).

Scheme 3.9

It is also possible to obtain highly branched PE/ADMA copolymers with the chelate catalyst 11, and representative data are collected in Table 3.4. The 1H and 13C NMR spectra of such a copolymer with 1.7 mol% ADMA comonomer incorporation are shown in Figures 3.7 and 3.8. In the spectra, there is evidence of two types of acetal group: one at the polymer chain- or branch-ends: -CH2-CH(OMe)2, and one inside the polymer main chain or branch: -CH-CH(OMe2)2. This can be seen from the multiplicity of the acetal methyne proton. This is a significant contrast with the branched ethene/methyl acrylate copolymers, where the comonomers are present only at the chain- or branch-ends.4 As expected due to the stability of the chelate, the copolymerization activity is significantly lower than that for ethene homopolymerization, and decreases with increasing comonomer concentration. An increasing comonomer concentration does result in an increased incorporation, but also in a drop of polymer molecular weight. In order to increase the polymer yield at higher ADMA concentrations, runs were performed at higher catalyst concentrations (conditions “B” in Table 3.4).

11

NPd

N O O

O

ON

PdN

Chain growing

71

Figure 3.7. 1H NMR (CDCl3, 500 MHz) of PE/ADMA (1.7 mol% comon.).

ppm102030405060708090100110 Figure 3.8. 13C NMR (CDCl3, 126 MHz) of PE/ADMA (1.7 mol% comon.)

ppm1.01.52.02.53.03.54.04.5

ppm3.903.954.004.054.104.154.204.254.304.354.404.45

b, b*

a, a*

CH2

CHOCH3

OCH3a

a*

b

b*CHOCH3

OCH3

CH2

CHOCH3

OCH3

c*CHOCH3

OCH3

c

c, c*

72

3.4.3 Deactivation of Pd catalysts in copolymerization of ethene and ADMA

Although ethene and ADMA could be successfully copolymerized, yielding a new type of polyethene copolymer, an irreversible catalyst deactivation process is also operative here. Upon work-up of the reaction mixtures, orange crystals of palladium species were isolated, which were characterized as a cationic palladium 1-methoxyallyl complex (12, see below). In addition, methanol was found in the reaction mixture by GC/MS analysis. NMR spectroscopic analysis and a single crystal structure determination (Figure 3.9, Table 3.5) confirmed the formation of the 1-methoxyallyl species [(N^N)Pd(η3-1-MeOCHCHCH2)][BAF] (12). In the structure solution, it was evident that again a certain amount of disorder is present in the crystal. Nevertheless, the atom connectivity was unequivocally established. The general features of the methoxyallyl group are similar to those in known methoxyallyl palladium complexes.23,24,25

Figure 3.9 ORTEP structure of complex 12 (cation)

73

Table 3.5 Selected bond distances (Å) and angles (o) of complex 12 _____________________________________________________________ C1172-C1182 1.54(3) C1182-C1192 1.27(2) Pd112-C1172 2.29(2) Pd112-C1182 2.119(14) Pd112-C1192 2.142(14) C1192-O112 1.20(2) O112-C1202 1.23(2) C1172-Pd112-C1182 40.6(8) C1182-Pd112-C1192 34.8(5) C1182-C1192-O112 141.0(17) C1192-O112-C1202 146.3(19) C1202-O112-C1192-Pd112 92(4) C1202-O112-C1192-C1182 -160(3) C1172-C1182-C1192-O112 -172(2) Pd112-C1172-C1182-C1192 45(2) _____________________________________________________________ Complex 12 could be recovered from the reaction mixtures in >80% yield after recrystallization (Scheme 3.10). Like allylic species 10, complex 12 showed no activity under the general conditions used for ethene catalytic polymerization. As was observed in the attempted E/AEE copolymerization, the formation of this allylic species is the main deactivation route of the Pd catalyst. Nevertheless, this process appears to be considerably slower for ADMA than for AEE, allowing the formation of appreciable amounts of copolymer.

Scheme 3.10

+ MeOHCH2Cl2

ADMAethene

11

NPd

N

O O

12

NPd

N

O

80%

74

3.4.4 Effect of added methanol on copolymerization of ethene and ADMA

Figure 3.10. Yields of ethene/ADMA copolymerization with and without addition of MeOH

Based on the proposed deactivation mechanism, the addition of methanol to the reaction mixture may either retard the formation of the inactive Pd methoxyallyl species 12 or regenerate an active species from it, so that catalyst lifetime is expected to be improved. A series of experiments using conditions A (defined in Table 3.4) with 4.2 mmol of ADMA and 37 equiv/Pd of methanol (0.5 mmol) showed that this does lead to improved overall productivity (Figure 3.10). Run times of 6, 12, 20 and 40 h gave polymer yields of 0.9, 1.21, 1.72 and 2.31 g respectively, corresponding to productivities of 11.2, 7.5, 6.4 and 4.3 kg mol-1 h-1. In the absence of methanol, a maximum polymer yield of 1.2 g was reached after 12 h, after which no additional polymer was formed. Although methanol addition improves overall productivity, it is clear that gradual catalyst deactivation still takes place. The addition of substantially larger amounts of methanol (500 equiv/Pd) again led to a decrease in catalyst productivity and formation of a

0

0,5

1

1,5

2

2,5

6 12 20 40

Run times (h)

Poly

mer

yie

ld (g

)

No MeOH37 eqv MeOH

75

black precipitate, probably Pd(0). Similarly, addition of methanol to solutions of methoxyallyl complex 12 resulted in Pd(0) formation, and in the presence of ethene no polymerization was observed. Thus it appears that the addition of methanol to the reaction mixture for ethene/ADMA copolymerization by catalyst 11 retards the catalyst deactivation reaction, and does not effect a reactivation of deactivated catalyst. 3.5. Further process of PE/ADMA copolymer The acetal groups in polymer chains allow chemical modification to change the properties of PE/ADMA copolymers. Acid-catalyzed acetal hydrolysis (Scheme 3.11) was performed as follows: PE/ADMA copolymer (0.3 g, 0.29 mol%) was mixed with p-toluene sulfonic acid monohydrate (0.5 g) in THF (40 mL) and stirred at 40 oC overnight. Subsequently the solvent was evaporated and the polymer was washed with methanol to remove the acid, then reprecipitated from petroleum ether/methanol. This yielded 0.26 g of material (87 % recovery).

Scheme 3.11

The 1H NMR spectrum of this material is shown in Figure 3.11. At least 50% of the acetal groups have disappeared and new resonances of two aldehyde species can be seen: at 9.75 ppm (s, 1H, -CH2C(O)H), and 2.40 ppm (t, 2H, J = 6.9, -CH2C(O)H) of an aldehyde group at the polymer chain- or branch-ends; and at 9.6 ppm (s, 1H, -(CH2)2CHC(O)H) and 2.30 ppm (quin, 1H, J = 7.0 , (-CH2)2CHC(O)H) of an aldehyde group inside the polymer main chain or branch. They correspond with the two types of acetals

OOH+

OH

76

described previously. Resulting from the strong acid treatment, a certain degree of degradation accompanied the hydrolysis of the acetal: 1H resonances of olefinic groups were observed in the 4.8-5.8 ppm spectral range. The polymer was partially crosslinked (as seen through the highly increased viscosity of the material), probably due to the reactivity of the aldehyde groups. Increasing the reaction time and temperature increases the degree of cross-linking: in another hydrolysis experiment at higher temperature (50 oC) overnight, the viscous oil of the PE/ADMA copolymer had crosslinked to give a hard solid.

Figure 3.11. 1H NMR (CDCl3, 500 MHz) spectrum of hydrolyzed

PE/ADMA copolymer by p-toluene sulfonic acid monohydrate in THF at 40 oC overnight.

CH2

CH

O

ppm12345678910

1.000.48

0.280.60

0.880.26

0.371.39

1571.03

77

3.6. Functional monomer - vinyl dioxolane (VDO) The successful copolymerization of ethene with ADMA prompted us to investigate a related monomer, vinyl dioxolane (VDO). When VDO is reacted with [(N^N)PdMe(EtO)2][BAF] (2) in CD2Cl2, after 5 min at -55 oC the coordinated ether has already been displaced completely by the olefinic moiety of VDO. Gradually increasing the temperature to -30oC leads to formation of the corresponding 1,2-insertion product, the reaction having gone to completion at around 0 oC (Scheme 3.12). Nevertheless, under these conditions the product appears to be unstable: upon warming to ambient temperature the NMR spectrum becomes increasingly complex and formation of Pd(0) was observed.

Scheme 3.12 A reaction of (N^N)PdMeCl with NaBAF and an excess VDO did not lead to isolation of the 1,2- insertion product. Instead, the cationic allylic species 13 was obtained from this reaction in 89% yield (Scheme 3.13). The product was characterized by NMR spectroscopy as well as an X-ray structure determination.

Scheme 3.13

CH2Cl2

NaBAFNPd

N Me

Cl BAF-

O

O

13

O

HO

PdN

N

O

O

-Et2OPd

N

N

Me

OEt2 O

OPdN

N

MePd

N

N OO

78

The 1H and 13C NMR spectra of 13 (Figure 3.12 and 3.13) show a great similarity with those of the methoxyallyl palladium complex 12. The difference lies in the hydroxoethyloxy-substituent on the allyl moiety. All the protons in this fragment are non-equivalent and could be assigned using 2D NMR techniques. The -OH proton resonance is found as a triplet (J = 5.3 Hz) at 1.47 ppm in dry CD2Cl2 solvent.

ppm123456782.08

0.961.51

0.213.52

0.460.99

0.480.49

1.531.03

4.84-0.13

Figure 3.12. 1H NMR (CD2Cl2, 500 MHz) spectrum of complex 13

ppm20406080100120140160 ppm20406080100120140160

Figure 3.13. 13C NMR (CD2Cl2, 126 MHz) spectrum of complex 13

79

The single crystal X-ray structure of complex 13 is shown in Figure 3.14 and the selected structure data are listed in Table 3.6. Again, disorder is present in the allylic fragment, and a refinement was made using two different conformations (only one of which is shown in the figure). Apparently, the acetal function in VDO is much more susceptible to C-O cleavage than the one in ADMA. It comes therefore as no surprise that in attempted copolymerizations of ethene with VDO (using chelate complex 11 as catalyst) rapid catalyst deactivation prevents the formation of copolymer.

Figure 3.14. ORTEP structure of complex 13 (cation)

80

Table 3.6 Selected bond distances (Å) and angles (o) of complex 13 _____________________________________________________________ Pd1-C129 2.093(7) Pd1-C1301 2.157(11) Pd1-C1311 2.011(13) C129-C1301 1.483(15) C1311-C1301 1.249(18) O111-C1321 1.445(13) O121-C1331 1.431(18) O111-C1311 1.396(15) C1301-Pd1-C1311 34.6(5) C1301-Pd1-C129 40.8(4) C1311-Pd1-C129 66.2(4) C1311-O111-C1321 120.2(13) C1311-C1301-C129 110.1(14) O111-C1311-C1301 144.4(16) O111-C1321-C1331 111.1(10) O121-C1331-C1321 114.0(12) C1311-C1301-C129-Pd1 -51.8(10) C1321-O111-C1311-C1301 -63(3) C1311-O111-C1321-C1331 -178.6(10) O111-C1321-C1331-O121 -57.2(16) _____________________________________________________________ 3.7. Conclusions A new branched polyethene copolymer, PE/ADMA, was obtained by palladium-catalyzed copolymerization of ethene with acrolein dimethyl acetal. Despite this success, it became clear that the copolymerization of ethene with allylic ethers has to contend with a catalyst deactivation process that proceeds via alcohol elimination and the formation of inactive η3-allyl palladium species. For allyl ethyl ether and vinyl dioxolane comonomers, this process is so fast that it prevented copolymer formation. For ADMA the process could be retarded by the addition of small quantities of methanol, allowing the new PE/ADMA copolymer to be produced readily in multigram quantities.

81

3.8. Experimental General considerations. Same as outlined for Chapter 2 except for the items specified below. GPC analysis of the polymers was performed by J. Vorenkamp at ambient temperature on a Spectra Physics AS 1000 LC-system using a Viscotek H-502 viscometer and a Shodex RI-71 refractive index detector, using THF as eluent and universal calibration with polystyrene standards. Starting materials. Allyl dimethyl acetal (ADMA, Aldrich 98%) and allyl ethyl ether (AEE, Acros 95%) were purified by distillation from CaH2. Others (Aldrich) were used as received. [(N^N)Pd(CH2CHMeCH2OEt)][BAF] (9). To a solution of [(N^N)PdMe(Et2O)][BAF] (2, 105 mg, 72 µmol) in 5 mL CH2Cl2 was added allyl ethyl ether (8.2 µL, 72 µmol). The red solution was stirred for 2 h at room temperature. After addition of hexane (10 mL), the solution was left at –30 oC overnight. Orange crystals were isolated. Yield 65 mg of 9 (0.044 mmol, 61%). 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 7.42–7.30 (m, 6H, Haryl), 3.53 (dd, J = 7.6, J’ = 6, CHH’OEt), 3.14 (t, J = 7.6, CHH’OEt), 2.97 (m, 4H, iPr CH), 2.85 (m, 2H, OCH2Me), 2.20 and 2.15 (s, 3H each, N=CMe), 1.67 (m, 2H, PdCH2C), 1.29 (m, PdCH2CHMeC), 1.42, 1.41, 1.40, 1.39, 1.25, 1.24, 1.20 and 1.19 (d, 3H each, J = 7.0, iPr Me), 0.82 (t, J = 7.0, OCH2Me), 0.79 (d, J = 7.0, PdCH2CHMeC). 13C NMR (CD2Cl2, 126MHz, 25 oC) δ 177.92 and 171.97 (N=CMe), 142.17 and 141.55 (Ar Cipso), 138.27, 138.24, 137.74 and 137.55 (Ar Co), 129.30 and 128.90 (Ar Cp), 125.20, 125.10, 125.03 and 124.94 (Ar Cm), 83.69 (t, JCH = 143, CH2OEt), 72.84 (t, JCH = 136, OCH2Me), 47.77 (t, JCH = 113, PdCH2C), 37.45 (d, JCH = 137, PdCH2CHMeC), 29.56, 29.45, 29.13 and 29.13 (iPr CH), 24.02, 23.88, 23.63, 23.59, 23.44, 23.43, 23.30 and 23.15 (iPr Me), 21.59 and 20.35 (q, JCH = 129, N=CMe), 14.62 (q, JCH = 127, PdCH2CHMeC), 13.69 (q, JCH = 127, OCH2Me). Anal. Calcd for (C66H65ON2BF24Pd): C, 53.73; H, 4.44; N, 1.90. Found: C, 53.59; H, 4.41; N,

82

1.88. [(N^N)Pd(η3-CH2CHCH2)][BAF] (10). Synthesized from (N^N)PdMeCl. To a suspension of (N^N)PdMeCl (204 mg, 0.36 mmol) and NaBAF (341 mg, 0.39 mmol) in 20 mL Et2O was added allyl ethyl ether (0.2 mL, 1.8 mmol). After the mixture was stirred for 2 days at room temperature, ether was removed by vacuum. The brown solid was extracted by 10 mL of toluene. Pentane (30 mL) was added into the solution, then the mixture was cooled to –80 oC overnight. Orange crystals of 10 were isolated (330 mg, 0.233 mmol, 59%). 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 7.42–7.30 (m, 6H, Haryl), 5.64 (m, allyl CH), 3.35 (d, 2H, J = 7.0, allyl CHH’), 3.04 (d, 2H, J = 12.8, allyl CHH’), 2.96 and 2.70 (septet, 2H each, J = 6.8, iPr CH), 2.25 (s, 6H, N=CMe), 1.35, 1.26, 1.23 and 1.22 (d, 6H each, J = 6.8, iPr Me). 13C NMR (CD2Cl2, 126MHz, 25 oC) δ 176.69 (N=CMe), 162.10 (Ar Cipso), 136.95 and 136.91 (Ar Co), 129.15 (Ar Cp), 125.03 and 124.98 (Ar Cm), 121.03 (d, JCH = 170, allyl CH), 65.78 (t, JCH = 185, allyl CH2), 29.82 and 29.44 (iPr CH), 23.65, 23.42, 23.33 and 23.30 (iPr Me), 20.19 (N=CMe). Anal. Calcd for (C66H65ON2BF24Pd): C, 53.46; H, 4.06; N, 1.98. Found: C, 52.9; H, 4.1; N, 1.8. Recovered from attempted copolymerization of ethene and AEE. From an attempted ethene/AEE copolymerization experiment (20 h run time) as described below. 100 mg of Pd catalyst 9 (67.0 µmol) with 1.0 ml of AEE (0.76 g, 8.82 mmol) in 5 ml of dichloromethane was stirred under 5 bar ethene. After 20 hours of reaction, the solvent was evaporated under vacuum at 40 °C to give an orange solid which was purified further by recrystallization from dichloromethane and petroleum ether. The recovered compound was identified as pure 10 (87 mg, 61.5 µmol, yield: 91.8%) by NMR spectroscopy. Synthesis of {(N^N)Pd[CH2CHMeCH(OMe)2]}[BAF] (11). Allyl dimethyl acetal (0.5 mL, 4.2 mmol) was added to a solution of (N^N)PdMeCl (1.130 g,

83

2.01 mmol) and NaBAF (1.752 g, 2.00 mmol) in 40 mL of Et2O. The mixture was stirred at ambient temperature for 24 h. Filtration followed by evaporation of the solvent in vacuo yielded an orange solid. After rinsing with pentane (3 x 30 mL) and drying in vacuo, 2.72 g (3.8 mmol, 91%) of the pure title compound was obtained. 1H NMR (CD2Cl2, 500MHz, 25 oC): δ 7.42–7.30 (m, 6H, Ar), 4.12 (d, J = 5.5, CH(OMe)2), 3.12 (s, 3H, OMe), 3.02 – 2.90 (m, 4H, iPr CH), 2.87 (m, 2H, PdCH2), 2.84 (s, 3H, OMe), 2.20 and 2.16 (s, 3H each, N=CMe), 1.44, 1.42, 1.39, 1.39, 1.27, 1.25, 1.22 and 1.18 (d, 3H each, J = 6.8, iPr Me), 1.3 (m, PdCH2CHMe), 0.74 (d, 3H, J = 7.0, PdCH2CHMe). 13C NMR (CD2Cl2, 126MHz, 25 oC): δ 178.03 and 172.18 (N=CMe), 141.74 and 141.71 (Ar Cipso), 138.39, 138.30, 137.76 and 137.38 (Ar Co), 129.37 and 129.01 (Ar Cp), 125.24, 125.17, 125.02 and 124.93 (Ar Cm), 119.48 (d, JCH = 172, CH(OMe)2), 60.51 (q, JCH = 147, OMe), 56.55 (q, JCH = 145, OMe), 38.80 (t, JCH = 140, PdCH2), 38.75 (d, JCH = 112, PdCH2CHMe), 29.61, 29.43, 29.18 and 29.13 (iPr CH), 24.05, 23.88, 23.79, 23.70, 23.40, 23.19, 23.12 and 23.03 (iPr Me), 21.60 and 20.13 (N=CMe), 14.65 (q, JCH = 131, PdCH2CHMe). Anal. Calcd for (C66H65O2N2BF24Pd): C, 53.15; H, 4.39; N, 1.88. Found: C, 52.9; H, 4.2; N, 1.8. Isolation of [(N^N)Pd(η3-1-CH2CHCHOMe)][BAF] (12). From an ethene/ADMA copolymerization experiment (20 h run time) as described below, with 100 mg (67 µmol) of 11, 1.0 mL (0.86 g, 8.44 mmol) of ADMA and 3 mL of dichloromethane, the crude complex was filtered off from the petroleum ether extract (that contains the copolymer), and was rinsed three times with petroleum ether to remove traces of remaining polymer. The orange powder was then recrystallized from a dichloromethane solution by the addition of petroleum ether. This yielded 80 mg (55 µmol, 82%) of 3 as deep red-orange crystals. 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 7.42 – 7.30 (m, 6H, Ar), 5.66 (d, J = 9.6, CHOMe), 5.27 (m, CH2CHCHOMe), 2.95 (m, CHH’CHCHOMe), 2.91 and 2.76 (septet, 2H each, iPr CH), 2.62 (s, 3H, OMe), 2.44 (dd, J = 12.8 and 2.2, CHH’CH=CHOMe), 2.24 and 2.23 (s, 3H each, N=CMe), 1.43, 1.31, 1.31, 1.30, 1.27, 1.24, 1.22 and 1.11 (d, 3H each, J = 6.9, iPr Me). 13C NMR (CD2Cl2, 126MHz, 25 oC): δ 175.06 and 174.93

84

(N=CMe), 144.14 and 143.07 (Ar Cipso), 138.44 and 136.86 (Ar Co), 128.88 and 128.69 (Ar Cp), 125.01 and 124.84 (Ar Cm), 120.33 (d, JCH = 168, CH2CHCHOMe), 98.67 (d, JCH = 149, CH2CHCHOMe), 60.18 (JCH = 141, OMe), 54.50 (t, JCH = 156, CH2CHCHOMe), 29.85, 29.73, 29.45 and 29.35 (iPr CH), 23.84, 23.81, 23.59, 23.37, 23.33, 23.30, 22.79 and 22.72 (iPr Me), 20.33 and 20.02 (N=CMe). Anal. Calcd for (C64H59ON2BF24Pd): C, 53.18; H, 4.11; N, 1.94. Found: C, 53.28; H, 4.05; N, 1.88. [(N^N)Pd(η3-CH2CHCHOC2H4OH)][BAF] (13). Vinyldioxolane (0.2 mL, 2 mmol, 50 eqv.) was added to a solution of (N^N)PdMeCl (219 mg, 0.39 mmol) and NaBAF (342 mg, 0.39 mmol) in 20 mL of Et2O. The mixture was stirred at 0 oC for 8 h and then at 25 oC for 40 h. Filtration was followed by evaporation of the solvent in vacuum until the solution was reduced to 2 mL. After addition of pentane (10 mL) the solution was kept at –30 oC overnight. 517 mg (0.35 mmol, 89%) of the title compound was isolated. 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 7.42-7.30 (m, 6H, Haryl), 5.64 (d, J = 9.55, CH2CHCHO-), 5.33 (m, CH2CHCHO-), 3.40 and 3.33 (m, 1H each, -OCH2CH2OH), 2.94 (m, CHH’CHCHO-), 2.93 (m, -OCHH’CH2OH), 2.93 and 2.76 (septet, 2H each, J = 6.6, iPr CH), 2.49 (dd, J = 12.6 and 2.1, CHH’CHCHO-), 2.42 (m, -OCHH’CH2OH), 2.25 and 2.23 (s, 3H each, N=CMe), 1.47 (t, J = 5.3, -OCHH’CH2OH), 1.43, 1.34-1.20 and 1.13 (d, 24H, iPr Me). 13C NMR (CD2Cl2, 126MHz, 25 oC) δ 175.1 (overlapping, N=CMe), 144.1 and 143.1 (Ar Cipso), 138.4 and 136.9 (Ar Co), 128.7 (Ar Cp), 125.0 and 124.8 (Ar Cm), 119.7 (d, JCH = 171, CH2CHCHO-), 99.4 (d, JCH = 174, CH2CHCHO-), 74.7 (t, JCH = 142, -OCHH’CH2OH), 61.6 (t, JCH = 142, -OCHH’CH2OH), 54.9 (t, JCH = 154, CH2CHCHO-), 29.8, 29.5 and 29.4 (iPr CH), 23.9, 23.8, 23.4, 23.3 and 22.8 (iPr Me), 20.3 and 20.0 (N=CMe). IR: 3485 cm-1, ν (OH). Anal. Calcd for (C64H61O2N2BF24Pd): C, 52.53; H, 4.20; N, 1.91. Found: C, 52.60; H, 4.28; N, 1.88. General procedure for VT NMR experiments The α-diimine palladium complex [(N^N)PdMe(EtO)2][BAF] (2, about 0.01

85

mmol) was weighed into an NMR tube in a drybox under N2 atmosphere. The tube was then capped with a latex septum, brought out from the drybox and cooled to –78 oC. The desired olefin (1 eqv.) and 0.7 mL of CD2Cl2 were injected into the NMR tube via a syringe. The tube was shaken briefly and then transferred to the pre-cooled probe of the NMR spectrometer. Spectra were acquired at regular temperature intervals. Complex 2 and allyl ethyl ether (AEE). At –50 oC, the coordinated ether was replaced completely by the olefinic part of AEE. Selected data of the AEE complex observed: 1H NMR (CD2Cl2, 500MHz, -50 oC) δ 4.91 (m, CHH’=CH), 4.39 (d, J = 15.7, CHH’=CH), 4.31 (d, J = 8.4, CHH’=CH), 0.30 (s, 3H, PdMe). At –30 oC, small amount of Pd-AEE chelate 9 was observed. This amount increased with increasing temperature. At 25 oC, all of 2 was converted to chelate 9. Complex 2 and allyl dimethyl acetal (ADMA). At –50 oC, the coordinated ether was replaced completely by the olefinic part of ADMA. Selected data of the ADMA complex: 1H NMR (CD2Cl2, 500MHz, -50 oC) δ 4.59 (m, CHH’=CH), 4.38 (d, J=15.4, CHH’=CH), 4.32 (d, J = 8.5, CHH’=CH), 4.11 (d, J = 7.0, CH(OCH3)2), 0.29 (s, 3H, PdMe). At –40 oC, small amount of Pd-ADMA chelate 11 was observed. This amount increased with increasing temperature. At -20 oC, all of 2 was converted to chelate 11. Complex 2 and vinyldioxolane (VDO). At –55 oC, the coordinated ether was replaced completely by the olefinic part of VDO. Selected data of the VDO complex: 1H NMR (CD2Cl2, 500MHz, -55 oC) δ 5.75 (m, CHH’=CH), 5.45 (d, J = 17.3, CHH’=CH), 5.34 (d, J = 14.4, CHH’=CH), 5.10 (d, J = 6.9, CH(OCH2C’H2O)), 3.95 and 3.85 (t, 2H each, J = 6.8, -OCH2C’H2O-), 2.35 and 2.23 (s, 3H each, N=CMe), 0.35 (s, 3H, PdMe). At –30 oC, formation of the 1,2-insertion product was observed. Selected data: 1H NMR (CD2Cl2, 500MHz, -55 oC) δ 3.95 and 3.85 (t, 2H each, -OCH2C’H2O-), 2.21 and 2.16 (s, 3H each, N=CMe), 0.78 (d, 3H, J = 6.4,

86

PdCH2CHMe). At 0 oC the insertion was completed. Decomposition of the insertion product occurred at 25 oC accompanied by the formation of Pd(0). At this stage the 1H NMR spectrum was complicated and could not be satisfactorily interpreted. General procedure for polymerizations The homo- and co-polymerizations were performed in a 50-mL glass mini-clave (Büchi AG, Switzerland) with a Teflon coated magnetic stirrer. Before use, the reactor was dried at 80 °C in a vacuum oven for 2 hours. A typical reaction procedure was as follows: In a nitrogen-filled glove-box, the mini-clave was charged sequentially with 1) Pd catalyst, 2) the desired amount of comonomers (for copolymerizations), and 3) dichloromethane. The reactor was taken out of the glove-box, put on a magnetic stirrer at room temperature and pressurized with ethene. The ethene pressure was kept constant during reaction by replenishing flow. After specified reaction time, the reactor was vented, and the (co)polymerization was terminated by addition of excess methanol. Further work-up was performed under aerobic conditions. After the reaction mixture was stirred at room temperature for 1 hour, the volatiles were evaporated in vacuum at 40 °C. This residue was extracted with petroleum ether (leaving solid air-stable allyl palladium species to be recovered, see above) and then the polymer was precipitated from the extract with methanol. This sequence may be repeated to remove any remaining impurities. Finally the viscous branched polymer product was dried under vacuum at 80 °C overnight. The ethene/ADMA copolymerizations can be run under aerobic conditions without any ill effect on productivity and polymer properties. To facilitate the experiments, the ethene/ADMA copolymerizations in the presence of added MeOH were performed under aerobic conditions. In these experiments, aliquots of MeOH were added last to the reaction mixture before pressurizing with ethene. Subsequent work-up was as described above.

87

. The PE/ADMA copolymers have the general NMR spectroscopic characteristics of branched polyethene materials produced by cationic palladium diimine catalysts,26 but also show the signatures of two distinct acetal functions: 1H NMR (CDCl3, 500MHz, 25 oC): δ 5.7 (t, J = 4.33 -CH2CH(OMe)2), 4.0 (d, J = 6.7, (-CH2)2CHCH(OMe)2), 3.33 (s, OCH3), 3.29 (s, OCH3). 13C NMR: δ 107.4 (-CH2CH(OMe)2) and 103.0 ((-CH2)2CHCH(OMe)2), 50.9 (OCH3). Acid treatment of PE/ADMA copolymer A three-necked flask with magnetic stirrer was charged with 0.3 g of PE/ADMA and 40 ml of THF. When the polymer had dissolved completely, 5 ml of acetone and 0.5 g of p-toluene sulfonic acid monohydrate were added while stirring. The mixture was then warmed to 40 °C and stirred overnight. The solvent was evaporated and the recovered polymer was washed with methanol to remove the acid. Then the polymer was re-dissolved in petroleum ether and precipitated with methanol. Finally, the product was dried under vacuum (yield: 0.26 g, 87 %). In addition to the normal features of branched polyethene, the polymer shows in its 1H NMR spectrum features of aldehyde groups: (CDCl3, 500MHz, 25 oC): δ 9.75 (s, 1H, -CH2C(O)H), 9.60 (s, 1H, (-CH2)2CHC(O)H), 2.40 (t, 2H, J = 6.9, -CH2C(O)H), 2.30 (quin, 1H, J = 7.0 , (-CH2)2CHC(O)H). X-ray Crystallographic analysis For the general features of the structure determinations, see the experimental section of Chapter 2. Crystallographic data can be found in Tables 3.7 and 3.8. The structure of 9 revealed a disorder in the chelate moiety, which was modeled by two orientations related by a twofold rotation about the vector Pd1-C132. The coordinates of the alternative setting were introduced in the

88

refinement and the site occupancy factors refined, using restrain instructions (DFIX, DELU, SIMU) for the disordered part (the disorder is compensated by the large displacement parameters). The s.o.f. of the major fraction of this component of the disorder model refined to a value of 0.60(1). In complex 10, disorder in the position of C130 (corresponding to the presence of two orientations of the allyl ligand) was modeled with two alternative positions that refined to an occupancy of 0.51(2) for the major orientation. A pentane solvent molecule in the cell was highly disordered over an inversion center. No satisfactory discrete model could be fitted in this density. The BYPASS procedure27 was used to take into account the electron density in the potential solvent area, which resulted in an electron count of 18 within a volume of 100.3 Å3 in the unit cell (probably the cavities are partly occupied). The presence of a residual of 1.77( 10) e/Å3 between C113 and C115 suggests a slight contribution of a disorder over a 180° rotation about the line N11-N12, similar (but less severe) as observed in the structure of 7 (Chapter 2). In complex 11, the chelate moiety (O11-O12-C129-C134 positions) is highly disordered. A disorder model with two alternative conformations with bond restraints was used in the final refinement (in which the disorder is compensated by the large thermal displacement parameters). Although the disorder model was not satisfactory in all respects, it was the best of several models tested. The cation in complex 12 is located over an inversion axis, implying disorder for a part of the cation (i.e. the part of the cation which has no inversion symmetry is alternatively occupied). Additionally, the CF3-groups in the anion were highly disordered. Restraints (DFIX) were used were applied in the refinement (in which the disorder is compensated by the large thermal displacement parameters). In complex 13 the ligand C5H9O2 (C129-C130-C131-O11-C132-C133-O12)

89

is highly disordered. The best model corresponded to a rotation of nearly 180° along the length axis of this ligand. A disorder model (50:50) with two alternative conformations and with bond restraints (DFIX, SIMU) was used in the refinement. Table 3.7. Crystallographic data for complex 9, 10 and 11. _____________________________________________________________ 9 10 11 Formula C66H65ON2BF24Pd C63H57N2BF24Pd C66H65N2O2BF24Pd FW 1475.45 1451.43 1491.45 Crystal system triclinic triclinic monoclinic Space group P-1 P-1 P21/n a (Å) 12.5293(8) 12.6191(5) 12.8098(7) b (Å) 15.592(1) 14.6417(6) 28.259(2) c (Å) 18.468(1) 18.3813(8) 19.154(1) α (deg) 77.686(1) 108.590(1) β (deg) 70.452(1) 95.619(1) 103.450(1) γ (deg) 82.915(1) 94.228(1) V (Å3) 3316.4(4) 3183.7(2) 6743.4(7) θ range (deg) 2.38-21.67 2.25-27.30 2.30-24.35 Z 2 2 4 ρ calc (g.cm-3) 1.478 1.514 1.469 F(000) 1500 1474 3032 µ(Mo Kα ), cm-1 3.9 4.04 3.86 Temp (K) 100(1) 100(1) 100(1) Reflections 11523 15198 11901 Parameters 962 851 950 wR(F2) 0.2318 0.1609 0.2107 Weighting (a, b) 0.1180, 0 0.1015, 0.0 0.1057, 15.8251 R(F) 0.0786 0.0554 0.0726 GooF 1.041 1.101 1.024 _____________________________________________________________

90

Table 3.8. Crystallographic data for complex 12 and 13. _____________________________________________________________ 12 13 Formula C64H59N2OBF24Pd C65H61N2O2BF24Pd FW 1445.37 1475.39 Crystal system monoclinic monoclinic Space group C2/c Cc a (Å) 21.9494(9) 20.9487(9) b (Å) 12.8691(5) 12.8804(5) c (Å) 25.207(1) 25.656(1) β (deg) 115.522(1) 109.770(1) V (Å3) 6425.4(4) 6514.7(5) θ range (deg) 2.27-28.02 2.21-25.56 Z 4 4 ρ calc (g.cm-3) 1.494 1.504 F(000) 2928 2992 µ(Mo Kα ), cm-1 4.01 3.99 Temp (K) 100(1) 100(1) Reflections 7900 13492 Parameters 597 920 wR(F2) 0.1874 0.1367 Weighting (a, b) 0.0838, 17.3084 0.081, 0.0 R(F) 0.0675 0.0538 GooF 1.036 0.997 _____________________________________________________________ 3.9. References

1 Ittel, S. D.; Johnson, L. K.; Brookhart, M. Chem. Rev. 2000, 100, 1169.

2 Boffa, L. S.; Novak, B. M. Chem. Rev. 2000, 100, 1479.

91

3 Johnson, L. K.; Mecking, S.; Brookhart, M. J. Am. Chem. Soc. 1996, 118,

267.

4 Mecking, S.; Johnson, L. K.; Wang, L.; Brookhart, M. J. Am. Chem. Soc. 1998, 120, 888.

5 Szabo, M. J.; Jordan, R. F.; Michalak, A.; Piers, W. E.; Weiss, T.; Yang, S.; Ziegler, T. Organometallics 2004, 23, 5565.

6 Szabo, M. J.; Galea, M. N.; Michalak, A.; Yang, S.; Groux, L. F.; Piers, W. E.; Ziegler, T. J. Am. Chem. Soc. 2005, 127, 14692.

7 Chen, G.; Ma, X. S.; Guan, Z. J. Am. Chem. Soc. 2003, 125, 6697.

8 Strazisar, S. A.; Wolczanski, P. T. J. Am. Chem. Soc. 2001, 123, 4278.

9 Luo, S.; Jordan, R. F. J. Am. Chem. Soc. 2006, 128, 12072.

10 Luo, S.; Jordan, R. F. 232nd ACS National Meeting, San Francisco, CA, USA, September 10-14, 2006, contribution INOR 549.

11 Wilke, G. Angew. Chem., Int. Ed. Engl. 1988, 27, 185.

12 Svedja, S. A.; Brookhart, M. Organometallics 1999, 18, 65.

13 Shultz, L. H.; Tempel, D. J.; Brookhart, M. J. Am. Chem. Soc. 2001, 123, 11539.

14 Curtis, C. J.; Miedaner, A.; Raebiger, J. W.; DuBois; D. L. Organometallics 2004, 23, 511.

15 Qi, X.; Liu, L.; Fu, Y.; Guo, Q. Organometallics 2006, 25, 5879.

16 Ozawa, F.; Ishiyama, T.; Yamamoto, S.; Kawagishi, S.; Murakami, H.; Yoshifuji, M. Organometallics 2004, 23, 1698.

17 Hosokawa, T.; Tsuji, T.; Mizumoto, Y.; Murahashi, S. I. J. Organomet. Chem. 1999, 574, 99.

92

18 Piechaczyk, O.; Thoumazet, C.; Jean, Y.; Floch, P. J. Am. Chem. Soc.

2006, 128, 14306.

19 Amatore, C.; Jutand, A.; Mensah, L.; Meyer, G.; Fiaud, J. C.; Legros, J. Y. Eur. J. Org. Chem. 2006, 1185.

20 Tang, D. Y.; Luo, X. L.; Shen, W.; Li, M. J. Mol. Struct. (THEOCHEM) 2005, 716, 79.

21 Themistou, E.; Patrickios, C. S. PMSE Preprints, 2006, 94, 160.

22 Login, R. B.; Shih, J. S.; Chuang, J. C. US Patent 5219950 (1993).

23 Morita, M.; Inoue, K.; Ogoshi, S.; Kurosawa, H. Organometallics 2003, 22, 5468.

24 Milani, B.; Paronetto, F.; Zangrando, E. J. Chem. Soc., Dalton Trans. 2000, 3055.

25 Vicente, J.; Abad, J. A.; Gilrubio, J.; Jones, P. G. Inorg. Chim. Acta 1994, 222, 1

26 Johnson, L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc. 1995, 117, 6414.

27 Sluis, P. van der; Spek, A.L. Acta Cryst. 1990, A46, 194-201.

93

Chapter 4. Reactivity with nitrogen-functionalized olefins

4.1. Introduction

Like the oxygen-based ether groups described in the previous chapter, nitrogen-based functionalities generally are also hard Lewis bases (especially the amines). They thus should also benefit from the hard-soft Lewis acid-base mismatch when combined with palladium catalysts in olefin copolymerization. Nevertheless, they are usually stronger Lewis bases than the related ethers, which should lead to a stronger metal-donor interaction. The most important nitrogen-containing olefin monomer is acrylonitrile (ACN). Polyacrylonitrile and related copolymers are commercially prepared by radical or anionic polymerization and extensively used in acrylic fibers, nitrile rubbers, etc.1,2,3,4 The interaction of ACN with cationic Pd-based catalysts was studied by Jordan and co-workers5,6 and was the subject of theoretical studies by Ziegler and co-workers.7,8 A strong interaction of the nitrile nitrogen donor atom with cationic Pd center is evident, but relatively electron-rich systems with strong donor ligands do show insertion of ACN into the Pd-C bond. Nevertheless, subsequent insertion of ethene or other alkenes after ACN insertion was not observed. Polyolefin materials bearing nitrogen-based functionalities can be of interest for various reasons. Pendant amine groups can be protonated or quaternized to give ionic groups on apolar backbones, as was shown e.g. by zirconocene/borate catalyzed copolymerization of 5-N,N-diisopropylaminopentene with 1-hexene or 4-methyl-1-pentene.9 In addition, polymers bearing N-based heterocycles can show redox behavior and/or display interesting electro-optical properties.10,11

94

In this chapter, the interactions of the cationic palladium methyl complex [(N^N)PdMe(OEt2)]+ (2) with two types of olefin bearing nitrogen-based functionalized are studied: allyl dimethyl amine (ADA) and N-alkenyl carbazoles (alkenyl = 1-pentenyl, allyl), together with attempts to copolymerize these monomers with ethene.

4.2. Reaction with Allyl Dimethyl Amide (ADA)

In a similar fashion as described in Chapter 3 for the ADMA monomer, a 5-membered chelate complex, [(N^N)Pd(CH2CHMeCH2NMe2][BAF] (14), was obtained in 52% isolated yield from the reaction of (N^N)PdMeCl with NaBAF and allyl dimethyl amine (ADA) in Et2O solvent (Scheme 4.1).

Scheme 4.1

The molecular structure of the Pd-ADA chelate 14 was established by single-crystal X-ray diffraction (Figure 4.1, Table 4.1). It shows a 5-membered chelate ring with intramolecular coordination of the amine nitrogen atom to the Pd-center. Although the structure contains less severe disorder than those in the related chelate complexes derived from AEE or ADMA insertion (Chapter 3), some conformational disorder in the chelate fragment is evident. The Pd11-N13 distance of 2.091(5) Å is slightly longer than the Pd-O distances in the AEE chelate 9 (2.008(9) Å) and the ADMA chelate 11 (2.052(14) Å).

14

NaBAF

ADANo activity

BAF-

NPd

N

N

Me

MeMe

ethene

NPd

N Me

Cl

95

Figure 4.1 ORTEP structure of complex 14 (cation)

Table 4.1 Selected bond distances (Å) and angles (o) of complex 14

_____________________________________________________________ Pd11-N11 2.078(4) Pd11-N12 2.152(4) Pd11-N13 2.091(5) Pd11-C134 2.023(6) N13-C129 1.459(9) N13-C130 1.449(11) N13-C131 1.505(10) C132-C133 1.540(15) C132-C134 1.406(13) C131-C132 1.364(15) N11-Pd11-N12 76.67(15) N11-Pd11-N13 174.94(17) N12-Pd11-N13 104.66(17) N12-Pd11-C134 169.74(19) N13-Pd11-C134 82.2(2) C129-N13-C130 110.1(6) C129-N13-C131 106.2(5) C130-N13-C131 106.3(6) N13-Pd11-N12-C113 -174.9(3) N12-Pd11-N13-C130 85.8(5)

96

C134-Pd11-N13-C131 11.9(4) N13-Pd11-C134-C132 1.3(6) C130-N13-C131-C132 90.1(9) N13-C131-C132-C133 166.4(8) _____________________________________________________________ The NMR spectra of the Pd-ADA chelate 14 are in agreement with the structure as determined by X-ray diffraction. The methyl substituents of the coordinated dimethylamino group show two resonances of 3H each (2.41 and 1.88 ppm). Despite the structural similarities to the ADMA- and AEE-derived chelates, there is a significant difference. Whereas it was seen that the chelates with a coordinated ether moiety readily initiated the catalytic polymerization of ethene, the ADA chelate does not react with ethene under the same conditions (20 oC, 5 bar ethene in CH2Cl2). This indicates that the palladium-amine dative bond is noticeably stronger than the palladium-ether dative bond, consistent with the notion that amines are the stronger σ-donors.

4.3. Reaction with N-pentenylcarbazole (NPC)

In the previous section it was seen that the tertiary amine function in allyl dimethyl amine (ADA) allows the smooth insertion of the alkene into the Pd-Me bond of the cationic catalyst species, but that subsequent olefin insertion is prevented by the strength of the intramolecular coordination of the amine group in the resulting chelate. In this section, the carbazole function is introduced as nitrogen-based functional group. Not only is the nitrogen atom this group a weaker Lewis base than the tertiary amine due to the unsaturation of the heterocyclic ring (making pyramidalization of the nitrogen atom unfavourable), the benzannelation also introduces substantial steric hindrance. Both of these features should reduce the strength of the Pd-N interaction. In addition, the carbazole group is well-known for its fluorescent and electron donor properties.12 , 13 The application in xerography of the

97

semiconducting polymer of N-vinylcarbazole prepared by radical polymerization14 was investigated extensively.15,16,17 A titanium complex has been found to initiate the polymerization of N-vinylcarbazole via a carbocationic (rather than a metal catalyzed) mechanism.18,19

Scheme 4.2

The reaction of [(N^N)PdMe(Et2O)][BAF] (2) and N-pentenyl carbazole (NPC) in CD2Cl2 was followed by VT NMR from -50 oC to 25 oC. After 5 min at –50 oC, all the ether that was coordinated to the cationic Pd center had been liberated. No intermediates (e.g. an η2 alkene adduct) could be observed, only the end-product, which was identified as the unusual three-membered chelate 15 (vide infra, Scheme 4.2), was visible. Performing the same reaction on a 60 µmol scale allowed isolation of 15 in 56% yield after crystallization. The molecular structure of complex 15 was determined by single-crystal

NPd

OEt2

MeNN

- Et2ON

NPd

MeN

NPd

N

N

chain walking

1,2-insertion

15

N

NPd

NMe

98

X-ray diffraction (Figure 4.2, Table 4.2). In contrast to most of the other chelate complexes of the Pd-diimine moiety, it is free of conformational disorder. Pd1, N13 and C141 form a 3-membered ring that is almost coplanar with the diimine ring (C141-Pd1-N11-C113 = 178.4(4)o). Despite the strain in the ring compared to the 5-membered chelates, the Pd1-N13 (2.104(5) Å) and Pd1-C141 (2.015(7) Å) distances are not significantly different to those in the Pd-ADA chelate (Pd11-N13 = 2.091(5) Å and Pd11-C134 = 2.023(6) Å). The Pd-N(diimine) distance trans to the chelate Pd-C bond is elongated by 0.06 Å relative to the other one. The carbazole ring is perpendicular to the 3-membered chelate ring (C140-C129-C141-Pd1 = 89.8(3)o). The carbon atom of the chelate ring bears a 3-methyl-but-1-yl substituent.

Figure 4.2 ORTEP structure of complex 15 (cation)

99

Table 4.2 Selected bond distances (Å) and angles (o) of complex 15

_____________________________________________________________ Pd1-N11 2.085(3) Pd1-N12 2.143(4) Pd1-N13 2.104(5) Pd1-C141 2.015(7) N13-C129 1.453(7) N13-C140 1.450(8) N13-C141 1.449(8) N11-Pd1-N12 75.32(14) N11-Pd1-N13 159.09(17) N12-Pd1-C141 165.2(2) N13-Pd1-C141 41.1(2) Pd1-N13-C141 66.1(3) C129-N13-C140 106.6(5) C129-N13-C141 127.0(5) C140-N13-C141 118.3(5) N13-C141-C142 122.4(6) N13-Pd1-N11-C113 -174.2(5) N12-Pd1-N13-C141 171.1(3) C141-Pd1-N11-C113 178.4(4) N13-Pd1-N12-C115 176.3(3) Pd1-C141-C142-C143 -86.1(8) Pd1-N12-C115-C113 1.5(5) Pd1-N13-C129-C134 132.2(4) C140-N13-C129-C134 -1.2(6) N13-C141-C142-C143 179.1(6) C141-C142-C143-C144 -164.2(7) C140-C129-C141-Pd1 89.8(3) C142-C141-Pd1-N13 -118.2(8) _____________________________________________________________

ppm1234567833.09

6.387.81

0.960.71

0.542.14

4.513.50

7.7310.82

21.7813.07

Figure 4.3. 1H NMR (CD2Cl2, 500 MHz) spectrum of complex 15

100

With the aid of COSY, HSQC and HMBC (Heteronuclear Multiple Bond Coherence) methodologies, the NMR spectra of 15 could be fully assigned (Figure 4.3 and Figure 4.4). This includes the connectivity within the 3-methyl-but-1-yl substituent. The CH group in the 3-membered chelate ring was observed at 4.29 ppm (1H, dd, J = 5.7 and 7.8 Hz) and 63.80 ppm (13C, d, JCH = 178 Hz).

ppm20406080100120140160

Figure 4.4. 13C NMR (CD2Cl2, 126 MHz) spectrum of complex 15 Complex 15 is likely to be formed via chain walking steps after initial 1,2-insertion of the alkene (which is expected to be the electronically preferred insertion for this alkene). In most cases, the chain-walking process after insertion of a functionalized monomer leads to 4-, 5- or 6-membered chelates. Brookhart observed the formation of all three types of chelate in the reaction of Pd complex (2) with methyl acrylate (MA) by variable temperature NMR, providing at room a mixture composed of 87% of the 6-membered chelate and 13% of the 5-membered chelate. When t-butyl acrylate is used as substrate, the 5-membered chelate becomes the main product (74%).20 In our previous studies (with AEE and ADMA in Chapter 3 and ADA in the previous section) only 5-membered chelates were found.

101

For the carbazole function, however, formation of a 5-membered chelate would require substantial pyramidalization of the carbazole nitrogen atom, and involve substantial steric interference between the carbazole group and the ligand 2,6-diisopropylphenyl groups. This causes the 3-membered chelate to be the most stable geometry for this particular substrate.

4.4. Reactions of diimine Pd complex and N-allylcarbazole (NAC)

Scheme 4.3

As described earlier for the reaction with NPC, the ionic Pd-Me complex [(N^N)Pd(Et2O)Me][BAF] (2) was reacted with N-allylcarbazole (NAC) in CD2Cl2 at -50oC and monitored by VT NMR spectroscopy. It appears that

NPd

OEt2

MeN

NPd

N

N

chain walking1,2-insertion

16b

NN

PdMeN

N

- Et2O16a

chain walking2,1-insertion

16c

NPd

N

N

102

NAC has a higher barrier to insertion into the Pd-Me bond than NPC. After 5 min at -50 oC, the ether adduct had been converted completely to the η2 olefin complex 16a, and free diethyl ether. Characteristically upfield shifted 1H NMR resonances for the coordinated olefinic moiety were found at δ 4.95 ppm (m, CHH’=CHCH2), 4.58 ppm (d, J = 14.6 Hz, CHH’=CHCH2), 4.44 ppm (m, CHH’=CHCH2), and 4.27 ppm (d, 2H, J = 8.4 Hz, CH2N(C6H4)2), comparing with the free monomer at δ 6.07 ppm (m, CHH’=CHCH2), 5.22 ppm (d, J = 10.3 Hz, CHH’=CHCH2), 5.08 ppm (d, J = 17.1 Hz, CHH’=CHCH2), and 4.95 ppm (d, 2H, J = 4.8 Hz, CH2N(C6H4)2). The Pd-Me 1H NMR resonance in the olefin adduct lies at δ 0.47 ppm. Subsequent insertion of the coordinated olefin is very slow at this temperature, but proceeds essentially to completion at -30oC to give a mixture of two products in approximately a 1:1 ratio. By comparison with the data for 15 and 1H,1H COSY NMR, the two products were characterized as the three-membered chelate complexes 16b and 16c (Scheme 4.3). They differ by having an iPr and an nPr substituent respectively on the carbon atom of the chelate ring. In 16b, the CH proton of the chelate ring is a doublet (4.18 ppm, J = 10.9 Hz) due to coupling with the iPr CH proton, whereas in 16c this is a (pseudo) triplet (4.32 ppm, J = 6.4 Hz). Due to the formation of the mixture of 16b and 16c, and their reasonable characterization based on NMR spectroscopy, it was not attempted to prepare and crystallize these species on a preparative scale. The difference in behavior between NAC and NPC is notable. Not only is for NAC an intermediate olefin adduct observable (indicating a higher free energy barrier for insertion into the Pd-C bond), it also displays both 1,2- and 2,1-insertion modes. The branch in 16b originates from initial 1,2-insertion (followed by chain-walking), whereas the absence of a branch in 16c indicates initial 2,1-insertion. It is likely that the observed difference in behaviour between NAC and NPC is predominantly due to steric factors.

103

4.5. Copolymerization of ethene and functionalized comonomers containing carbazole groups

The 3-membered chelate complex 15 was tested for ethene homopolymerization. Using 2.9 mg (1.8 µmol) of 15 in 10 mL of CH2Cl2, 5.0 bar ethene atmosphere and a 17.5 h reaction time at room temperature, the polymerization yielded 1.42 g of polyethene (45.5 kg/ h · mol Pd). For copolymerization experiments with ethene and the carbazole-containing comonomers NAC and NPC, the Pd-ADMA chelate 11 was used as the catalyst, as it is available in quantity, and has been shown to be effective in other copolymerizations (see Chapter 3). Attempted copolymerization of ethene and N-allyl-carbazole (NAC) using the Pd-ADMA chelate 11 as introduced catalyst precursor under standard conditions (4.5 h at room temperature, 5 bar ethene, 15 mL of CH2Cl2 solvent and 0.2 M of NAC) and more forcing conditions (1.5 h at 50 oC, 5 bar ethene, 5 mL of CH2Cl2 and 1.1 M of NAC), only produced amorphous branched homopolyethene, with the usual branch density (132 branches/1000 C, as determined by NMR, Figure 4.5) observed for Pd-catalysts with this α-diimine ligand.21 The presence of NAC comonomer (0.2 M and 1.1 M) has no significant effect on the activity of the catalyst (53.0 kg/mol.h and 59.6 kg/mol.h, compared to 51.1 kg/mol.h for ethene homopolymerization with the same catalyst in 3.5 h at room temperature, 5 bar ethene, 15 mL of CH2Cl2 solvent). Apparently, the carbazole substituent on the allylic position in NAC effectively prevents it from being incorporated. As seen from the NMR tube scale experiments, the barrier to insertion of NAC into the Pd-alkyl bond is relatively high, and in the presence of excess ethene NAC apparently cannot compete with ethene for productive interaction with the Pd center.

104

ppm12345678910

Figure 4.5 1H NMR (CDCl3, 500 MHz) spectrum of branched PE homopolymer from an E/NAC copolymerization run.

In contrast to the results presented above for NAC, catalytic copolymerization of ethene with N-pentenyl-carbazole (NPC) was successfully catalysed by the Pd(α-diimine) catalyst system. In Table 4.3, a series of experiments is listed using the ADMA chelate 11 as catalyst at varying concentrations of NPC comonomer. The catalyst activity is relatively little affected by the increase in comonomer content, and is close to the activity for ethene homopolymerization under these conditions. The longer alkyl chain between the double bonds and the sterically demanding carbazole group in NPC probably reduces the steric hindrance in the vicinity of the olefinic moiety so that it allows the comonomer to compete with ethene for coordination with Pd, and effect subsequent insertion into the Pd-alkyl bond. When NPC insertion takes place, reactivity with ethene can still proceed, even in the case that the chain-walking processes bring the carbazole close to the metal center to form a three-membered chelate. NMR spectroscopy on the polymers reveals that copolymers with up to 18 wt % of

105

NPC comonomer can be obtained (see last entry in Table 4.3, from an experiment with a lowered ethene pressure). The 1H and 13C NMR spectra of this copolymer are shown in the Figures 4.6 and 4.7 respectively. The 1H NMR spectrum indicates that essentially all carbazole groups in the copolymer have at least two methylene groups between the carbazole nitrogen and the rest of the polymer chain: the resonances for the α-CH2 group (4.25 ppm) and the β-CH2 group (1.85 ppm) have intensities that are equal to that of the carbazole 4,5-protons (8.10 ppm). This implies that the carbazole functions are not incorporated into the polymer via direct ethene insertion into a 3-membered chelate species like 15, as this would create a secondary carbon adjacent to the carbazole nitrogen atom. Thus, either the chain-walking process after NPC insertion is slower than the capture of a new ethene molecule, or chain walking is fast, and can occur to, as well as from, the 3-membered chelate (in the latter case enabling ethene capture and insertion). In view of available data on (co-)polymerization with Pd α-diimine catalysts, the latter option is deemed more likely.22,23,24 Table 4.3. Ethene/NPC copolymerization using chelate 11 as introduced

catalyst precursor.

a) A: 13.4 µmol 11, 15 mL CH2Cl2, 5 bar ethene, room temperature, 20 h; B: 13.4 µmol 11, 15 mL

CH2Cl2, 2 bar ethene, room temperature, 4 h. b) Due to the high catalyst concentration and long reaction

time employed to maximize the amount of produced polymer, the catalyst activities expressed in kg(PE)

mol(Pd)-1h-1 appear relatively low. c) Determined by 1H NMR.

Conditionsa NPC (mmol)

PolymerYield (g)

Activityb

(kg mol-1 h-1)Mw

(x10-3)Mw/Mn

NPC Contentc

(wt %)

A 0 3.50 13.0 186 1.5 0

A 2.1 3.50 13.0 350 1.7 1.6

A 3.8 4.03 15.0 259 1.6 5.0

B 6.0 0.63 11.8 279 3.4 18.8

106

Figure 4.6 1H NMR (CDCl3, 300 MHz) spectrum of PE/NPC (18.8 wt% comonomer)

Figure 4.7. 13C NMR (CDCl3, MHz) spectrum of PE/NPC (18.8 wt% comonomer)

12345678

0.491.19

1.030.07

0.520.05

0.7195.55

0

ppm20406080100120140

107

The carbazole group is known to exhibit aggregation-dependent fluorescence behavior. Isolated N-alkylated carbazole groups show a fluorescence around 365 nm, but carbazole groups that are closely associated show fluorescence at longer wavelengths (“monomer” versus “excimer” fluorescence).25,26 The amorphous ethene/NPC copolymers show visible fluorescence under UV irradiation. Two of these copolymers, with 5.0 wt% and 18.8 wt% NPC contents, were studied by emission fluorescence spectrometry (Figure 4.8 and 4.9). For the copolymer with low NPC content (5.0 wt%), mainly the locally excited state of the carbazole chromophore at 365 nm was observed (Figure 4.8). For the copolymer with higher NPC content (18.8 wt%), a significantly increased emission intensity is seen around 410 nm (Figure 4.9). As the copolymers produced with this catalyst are amorphous, the carbazole groups are essentially distributed evenly over the volume of the polymer. The ratio of monomer vs. excimer fluorescence decreases with increasing NPC content of the copolymer.

Figure 4.8. Emission fluorescence spectrum of amorphous PE/NPC (NPC: 5.0 wt %), Excitation wavelength = 295 nm

330 380 430 480

108

330 380 430 480 Figure 4.9. Emission fluorescence spectrum of amorphous PE/NPC (NPC:

18.8 wt %), Excitation wavelength = 295 nm

4.6. Conclusions

The trialkyl amine group is a stronger σ-donor group than the ether functionalities described in Chapter 3, and as a consequence the 5-membered chelate palladium complex derived from allyl dimethyl amine (ADA) insertion is unable to give subsequent ethene insertion. Thus, ethene/ADA copolymerization is impossible with this catalyst. In contrast, the carbazole function has steric and electronic properties that allow it to be incorporated in polyethene copolymers, provided the carbazole in the comonomer is sufficiently far removed from the olefinic moiety to allow ready coordination and insertion of the olefin. Due to the properties of the carbazole group, a 3-membered (rather than 5-membered) chelate is formed after insertion of the comonomer, allowing relatively facile reactivity with ethene. Thus, new amorphous branched polyethene/N-pentenylcarbazole copolymers can be

109

produced with the α-diimine palladium catalyst. These copolymers show fluorescent properties that vary with the concentration of the carbazole groups in the copolymer.

4.7. Experimental

General considerations. Same as outlined for Chapter 2 and 3 except for the items specified below. The emission fluorescence spectrum of E-NPC copolymer was recorded by Dr. V. Krasnikov on a Perkin-Elmer LS 50B instrument from a pressed solid sample of 13 mm diameter and 2 mm thickness at ambient temperature. The excitation wavelength was 300 nm, with excitation and emission widths of 10 nm and a scan rate of 100 nm/min. Starting materials. Allyl dimethyl amide (ADA, Acros 98%), 5-bromopentene-1 (Aldrich 95%), allyl bromide (Aldrich 97%), carbazole (Acros 95%) and NaH (Aldrich 60% in mineral oil) were used as received. N-pentenylcarbazole (NPC) and N-allylcarbazole (NAC) were synthesized from carbazole, NaH and 5-bromopentene-1 or allyl bromide following the procedure described by Kim et al.27 Synthesis of [(N^N)Pd(CH2CHMeCH2NMe2)][BAF] (14). Allyl dimethyl amine (0.1 mL, 1.6 mmol) was added to a solution of (N^N)PdMeCl (157 mg, 0.29 mmol) and NaBAF (210 mg, 0.24 mmol) in 20 mL of Et2O. The mixture was stirred at 20 oC for 3 days. After filtration, the solvent was evaporated in vacuum. The orange solid was washed with pentane (10 mL) 3 times, affording compound 1 (225 mg, 0.15 mmol, 52%). Crystals suitable for X-ray diffraction were obtained by recrystallization from CH2Cl2/pentane. 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 7.44–7.30 (m, 6H, Haryl), 3.00 and 2.88 (m, 4H, iPr CH), 2.41 (s, 3H, NMeMe’), 2.28 (m, 2H, CH2N), 2.21 and 2.13 (s,

110

3H each, N=CMe), 1.91 (m, CH), 1.88 (s, 3H, NMeMe’), 1.75 (dd, J = 8.8, J’ = 11.4, PdCHH’), 1.41 and 1.28-1.10 (m, 24H, iPr Me), 1.24 (PdCHH’), 0.67 (d, 3H, J = 6.8, CHMe). 13C NMR (CD2Cl2, 126MHz, 25 oC) δ 178.9 and 173.41 (N=CMe), 142.53 and 141.87 (Ar Cipso), 138.27, 137.99, 137.87 and 137.61 (Ar Co), 129.01 and 128.89 (Ar Cp), 125.53, 125.34, 124.91 and 124.84 (Ar Cm), 76.03 (t, JCH = 123, CH2N), 52.51 (q, JCH = 138, NMe), 49.65 (q, JCH = 138, NMe), 43.66 (t, JCH = 129, PdCH2), 35.61 (d, JCH = 122, CH), 29.41, 29.22, 29.12 and 29.08 (iPr CH), 23.0 and 21.70 (N=CMe), 15.50 (q, JCH = 129, MeCH2). Anal. Calcd for (C70H76N3BF24Pd): C, 53.76; H, 4.51; N, 2.85. Found: C, 53.4; H, 4.3; N, 2.9. Synthesis of {(N^N)Pd[Me2CHC2H4CHN(C6H4)2]}[BAF] (15). N-pentenylcarbazole (13.6 mg, 0.06 mmol) was added to a solution of [(N^N)PdMe(EtO)2][BAF] (2, 84.6 mg, 0.06 mmol) in CH2Cl2 (10 mL). The solution was stirred at 20 oC overnight. After addition of pentane (50 mL) the solution was kept at 20 oC for 2 days. Dark red crystals (53 mg, 0.033 mmol, 55%) were isolated. 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 7.77, 7.64, 6.99,6.89 and 6.86 (t, d, t, d, d, 1H each, J = 7.8, carb), 7.5 – 7.3 (m, 9H, Haryl and carb), 4.29 (dd, J = 5.7, J = 7.8, PdCH), 3.59, 2.94, 2.69 and 2.56 (septet, 1H each, J = 6.8, iPr CH), 2.24 and 2.04 (s, 3H each, N=CMe), 1.65, 1.64, 1.33, 1.30, 1.02, 0.99, 0.88 and 0.48 (d, 3H each, iPr Me), 1.5 (overlapping, CHHCHPd and Me2CH), 1.1 (CHHCHMe2), 1.0 (CHHCHPd), 0.53 and 0.52 (d, 3H each, J = 8.3, Me2CH), 0.48 (CHHCHMe2). 13C NMR (CD2Cl2, 126MHz, 25 oC) δ 175.55 and 172.04 (N=CMe), 146.63, 144.47, 135.18, 132.22, 129.07, 127.66, 125.44, 124.02, 121.35, 117.85, 113.68 and 99.89 (Ar : carbazole), 143.18 and 142.72 (Ar Cipso), 139.34, 138.05, 137.14 and 136.96 (Ar Co), 129.07 and 128.39 (Ar Cp), 128.07, 127.51, 125.23 and 124.9 (Ar Cm), 63.81 (d, JCH = 160, PdCH), 35.98 (JCH = 124, CH2CHMe2), 30.29, 29.80, 29.37 and 29.29 (iPr CH), 28.55 (t, JCH = 146.6, CH2CHPd), 23.16 and 21.83 (q, JCH = 123, Me2CH), 24.59–21.83 (8C, iPr Me), 20.69 and 20.04 (q, JCH = 127, N=CMe). Anal. Calcd for (C78H72N3BF24Pd): C, 57.67; H, 4.47; N, 2.59. Found: C, 57.62; H, 4.42; N, 2.71.

111

General VT NMR experiments of [(N^N)PdMe][BAF] (2) and functionalized monomers. The α-diimine palladium complex [(N^N)PdMe(EtO)2][ BAF] (2, about 0.02 mmol) was weighed into an NMR tube in a drybox under N2 atmosphere. The tube was then capped with a latex septum and brought out from the drybox. Solutions of monomers (1 eqv.) in CD2Cl2 (0.7 mL) were injected by a syringe into the NMR tube that was cooled at -196 oC by liquid N2 and then the septum was wrapped with Parafilm. Upon thawing out the tube was shaken briefly to dissolve the palladium complex and transferred to the pre-cooled probe of the NMR spectrometer. Spectra were acquired at regular temperature intervals. Complex 2 and N-pentenylcarbazole. After 5 min at -50 oC, the coordinated ether had been released and the NPC 1,2-insertion product (3-membered chelate 15) was observed. Complex 2 and N-allylcarbazole. After 5 min at -50 oC, the coordinated ether had been displaced by the olefinic moiety of the substrate, yielding the olefin complex 16a. Incipient insertion of NAC into the Pd-Me bond (producing the 3-membered chelate complexes 16b and 16c) was observed at this temperature. Upon increasing the temperature the insertion reaction progressed, until at 0 oC all of 16a had disappeared. The ratio of the 3-membered chelate complexes 16b and 16c at all stages of the reaction was about 1:1. The assignments of the NMR spectra of 16b and 16c are supported by 2D NMR (COSY). {(N^N)Pd(Me)[η2-CH2=CHCH2N(C6H4)2]}[BAF] (16a): 1H NMR (CD2Cl2, 500MHz, -55 oC) δ 8.06 (d, J = 7.8, N(C6H3H)2), 4.95 (m, CHH’=CH), 4.58 (d, J = 14.6, CHH’=CH), 4.44 (m, CHH’=CH), 4.27 (d, 2H, J = 8.4, CH2N(C6H4)2), 0.47 (s, 3H, PdMe). {(N^N)Pd[Me2CHCHN(C6H4)2]}[BAF] (16b): 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 4.18 (d, J = 10.9, PdCH), 2.33 (m, Me2CH), 0.25 and 0.11 (d, 3H each, J = 6.4, Me2CH).

112

{(N^N)Pd[MeCH2CH2CHN(C6H4)2]}[BAF] (16c): 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 4.32 (t, J = 6.4, PdCH), 1.48 (m, 2H, MeCH2CH2), 1.08 (m, 2H, MeCH2), 0.46 (t, 3H, J = 6.8, MeCH2). General Procedure for polymerizations (E, E/NPC and E/NAC) The homo- and co-polymerizations were performed in a 50-mL glass mini-clave (Büchi AG, Switzerland) with a Teflon coated magnetic stirrer. Before use, the reactor was dried at 80 °C in a vacuum oven for 2 hours. A typical reaction procedure was as follows: In a nitrogen-filled glove-box, the mini-clave was charged sequentially with 1) Pd catalyst, 2) the desired amount of comonomers (for copolymerizations), and 3) dichloromethane. The reactor was taken out of the glove-box, put on a magnetic stirrer at room temperature and pressurized with ethene. The ethene pressure was kept constant during the reaction by replenishing flow. After the specified reaction time, the reactor was vented, and the (co)polymerization was terminated by addition of excess methanol. Further work-up was performed under aerobic conditions. After the reaction mixture was stirred at room temperature for 1 hour, the volatiles were evaporated in vacuum at 40 °C and the residue was extracted with petroleum ether. Subsequently the polymer was precipitated with methanol. Finally the viscous product was dried in vacuum at 80 °C overnight. Polymerization of ethene by {(N^N)Pd[Me2CHC2H4CHN(C6H4)2]}[BAF] (15). The above polymerization procedure was followed using 2.9 mg (1.8 µmol) of 15 in 10 mL of CH2Cl2, 5.0 bar ethene atmosphere and a 17.5 h reaction time at room temperature. Polymer yield: 1.42 g of polyethene (45.5 kg/ h · mol Pd). X-ray Crystallographic analysis For the general features of the structure determinations, see the experimental

113

section of Chapter 2. Crystallographic data can be found in Table 4.4. In complex 14, the CF3-groups of the anion were heavily disordered. In some cases this was modeled by two alternative positions with partial occupancy. The thermal parameters of the chelate moiety also suggest some degree of disorder, but this could not be modeled. In complex 15, one of the CF3-groups in the anion was rotationally disordered, which was modeled by two orientations with 0.5 occupancy factors.

Table 4.4. Crystallographic data for complex 14 and 15.

_____________________________________________________________ 14 15 Formula C66H66N3BF24Pd C78H71N3BF24Pd FW 1474.46 1623.61 Crystal system monoclinic monoclinic Space group C2/c P21/n a (Å) 41.414(2) 12.8846(6) b (Å) 12.7965(5) 15.7270(7) c (Å) 26.010(1) 36.571(2) β (deg) 103.000(1) 91.951(1) V (Å3) 13430.8(10) 7406.3(6) θ range (deg) 2.20-20.53 2.26-24.27 Z 8 4 ρ calc (g.cm-3) 1.458 1.456 F(000) 6000 3308 µ(Mo Kα ), cm-1 3.85 3.57 Temp (K) 100(1) 100(1) Reflections 11866 15141 Parameters 907 1003 wR(F2) 0.1664 0.2061 Weighting (a, b) 0.0644, 65.10 0.0926, 20.0630 R(F) 0.0618 0.0762 GooF 1.028 1.057 _____________________________________________________________

114

4.8. References

1 Wade, B.; Knorr, R.; (Eds. Masson, J. C.) Acrylic Fiber Technology and Applications; Dekker: New York, 1995; p 37.

2 Peng, F. M. Encycl. Polym. Sci. Eng. 1985, 1, 426.

3 Alger, M. Polymer Science Dictionary, 2nd ed.; Chapmen Hall: New York, 1997; p 6.

4 Acrylonitrile and Acrylonitrile Polymers. In Kirk-Othmer Encyclopedia of Chemical Technology, 3rd ed.; Wiley & Sons: New York, 1983; Vol.1, p 414.

5 Wu, F.; Foley, S. R.; Burns, C. T.; Jordan, R. F. J. Am. Chem. Soc. 2005, 127, 1841.

6 Schaper, F.; Foley, S. R.; Jordan, R. F. J. Am. Chem. Soc. 2004, 126, 2114.

7 Szabo, M. J.; Galea, N. M.; Michalak A.; Yang, S.; Groux, L. F.; Piers, W. E.; Ziegler, T. J. Am. Chem. Soc. 2005, 127, 14692.

8 Yang, S.; Szabo, M. J.; Michalak A.; Weiss, T.; Piers, W. E.; Jordan, R. F.; Ziegler, T. Organometallics 2005, 24, 1242.

9 Stehling, U. M.; Stein, K. M.; Fischer, D.; Waymouth, R. M. Macromolecules 1999, 32, 14.

10 Gill, W. D.; (Eds. Mort, J.; Pai, D. M.) Photo conductivity and Related Phenomena; Elsevier: Amsterdam; 1976. p 303.

11 Takemoto, K.; Ottenbrite, R. M.; Kamachi, M.; Functional Monomers and polymers; Marcel Dekker, Inc.: New York; 1997. p 133.

115

12 Wise, D. L.; Wnek, G. E.; Tromtolo, D. J.; Cooper, T. M.; Gresser, J. D.

Photonic polymer systems; Marcel Dekker, Inc.: New York; 1998. p 235.

13 Kanega, H.; Shirota, Y.; Mikawa, H. J. Chem. Soc. Chem. Commun. 1984, 158..

14 Partridge, R. H. Chem. Abstr. 1974, 86, 11261.

15 Partridge, R. H. Polymer, 1983, 24, 748.

16 Seanor, D. A. Electrical Properties of Polymers; Academic Press: New York; 1982. p 116.

17 Mort, J.; Pfister, G.; Electronic Properties of Polymers; Wiley: New York; 1982. p 187.

18 Wang, Q.; Baird, M. C. Macromolecules 1995, 28, 8021.

19 Baird, M. C.; Wang, Q. US 5439996, 1995.

20 Johnson, L. K.; Mecking, S.; Brookhart M. J. Am. Chem. Soc. 1996, 118, 267.

21 Johnson, L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc. 1995, 117, 6414.

22 Shultz, L. H.; Tempel, D. J.; Brookhart, M. J. Am. Chem. Soc. 2001, 123, 11539.

23 Tempel, D. J.; Johnson, L. K.; Huff, R. L.; White, P. S.; Brookhart, M. J. Am. Chem. Soc. 2000, 122, 6686.

24 Mecking, S.; Johnson, L. K.; Wang, L.; Brookhart, M. J. Am. Chem. Soc. 1998, 120, 888.

25 Solaro, R.; Galli, G.; Ledwith, A.; Chiellini, E. In Polymer Photophysics; (Ed. Phillips, D.) Chapman and Hall: New York, 1985; p 377.

116

26 Miyashita, T.; Matsuda, M.; Auweraer, M.; Schryver, F. C.

Macromolecules 1994, 27, 513.

27 Kim, J. K.; Hong, S. I.; Cho, H. N. Polym. Bull. 1997, 38, 169.

117

Chapter 5. Reactivity with sulfur-functionalized olefins

5.1. Introduction

In the previous chapters we have described the reactivity of cationic palladium alkyl species with various heteroatom-functionalized olefins. The common feature of the substrates discussed so far is that these heteroatom functionalities were essentially all hard Lewis bases: ethers (Chapter 3) and amines (Chapter 4). It is likely that, due to the mismatch between the soft Lewis acidic nature of the palladium metal center with these hard Lewis basic moieties, the soft Lewis basic olefinic moiety of the substrate could effectively compete with the heteroatom functionalities for coordination to the metal center. As a result, insertion of the olefinic moiety of the substrate into the palladium-methyl bond of the [(N^N)PdMe]+ species was readily observed. The ability of the resulting chelate complexes to effect the catalytic polymerization of ethene was then only dependent on the donor strength of the heteroatom functionality in the chelate. Thus the strongly σ-donating trialkylamine group prevented subsequent insertion of ethene, whereas the more weakly donating ether and carbazole moieties readily allowed this, allowing the catalytic synthesis of new heteroatom-containing polyethene copolymers. In this chapter the interaction of the [{N^N}PdMe]+ species with olefins bearing a soft Lewis basic functional group is described. As suitable heteroatom functionalities we have chosen the thioether and 1,3-dithiane moieties. This will allow a direct comparison with the ether and acetal functions described in Chapter 3, and should give us information on the competition of the olefin with other soft Lewis basic groups and how this can be modified.

118

5.2. Reaction of [(N^N)PdMe(Et2O)][BAF] (2) with allyl methyl thioether (AMT)

The reaction of the cationic Pd-methyl species [(N^N)PdMe(Et2O)][BAF] (2) and allyl methyl thioether (AMT) in CD2Cl2 solvent was followed by VT NMR. After 5 min at -60 oC, the ether originally coordinated to the Pd-center in 2 is displaced by the substrate, as can be seen from the shift in the 1H resonance of the ether OCH2-protons from δ 3.20 ppm to δ 3.38 ppm for free ether. The Pd-Me 1H NMR resonance is still a singlet, shifted slightly upfield from δ 0.33 ppm in 2 to δ 0.28 ppm. In contrast to the intermediate adducts observed e.g. for the ADMA (Section 3.4.1) or NAC (Section 4.4) substrates, the resonances of the olefinic part of the substrate are not significantly shifted compared to the free AMT monomer: δ 5.49 (CHH’=CH), 5.27 (CHH’=CH) and 5.16 (CHH’=CH) ppm in the adduct compared to δ 5.69, 5.32 and 5.20 ppm respectively in free AMT. This suggests that the C=C bond of the substrate is not coordinated to the metal, but that AMT is coordinated to Pd via the thioether sulfur atom instead. The product can thus be formulated as the thioether adduct [(N^N)Pd(Me)(κ1-MeSCH2CH=CH2)][BAF] (17, Scheme 5.1). It is related to simple thioether complexes reported by Brookhart et al. from trapping experiments of cationic Pd-alkyl species, e.g. [(α-diimine)Pd(n-Pr)(SMe2)][BAF] and [(phen)Pd(Me)(MeSC6H5)][BAF] (phen = 1,10-phenanthroline).1,2 Upon warming the solution to room temperature, no further reaction was observed. Compound 17 was synthesized on a preparative scale by the reaction of an excess of AMT with [(N^N)PdMe(EtO)2][BAF] (2) in CH2Cl2 at ambient temperature. Addition of pentane to the solution and cooling to -30 oC afforded 17 as yellow crystals in 87% isolated yield. A suitable single crystal of 17 was found for X-ray diffraction. Its structure is shown in Figure 5.1, selected geometrical data are listed in Table 5.1. It establishes unequivocally that only the thioether sulfur atom of the substrate is coordinated to the metal, and that no insertion of the olefin into the

119

Pd-methyl bond has taken place. In contrast to many other Pd-diimine compounds in this thesis, the crystal structure of 17 shows no conformational disorder. The substrate C=C bond (C131-C132 = 1.310(6) Å) is retained, and not interacting with the metal. The Pd-methyl bond distance Pd1-C133 of 2.054(3) Å is slightly longer than the Pd-Me distance in the neutral (but less sterically hindered) α-diimine palladium dimethyl complex 1 (2.033(5) Å).3 Again, the diimine nitrogen trans to the methyl group shows the largest Pd-N bond distance (2.143(4) Å, versus 2.078(2) Å for the nitrogen trans to the thioether).The Pd-S dative bond distance is 2.2893(9) Å, which is similar as for example in reported square planar [PdCl2(MeSC6H4-CH2PPh2)] (Pd-S: 2.295(1) Å) 4 and [Pd(L)][PF6]2 (L = 2,5,8-trithia[9](2,9)-1,10- phenanthrolinophane) (Pd-S: 2.293(1) and 2.291(1) Å).5 Thus, the thioether adduct 17 is formed exclusively in the reaction of 2 with AMT, and even at or above room temperature no olefin insertion into the Pd-Me bond is observed. This shows that the soft-soft Lewis acid-base interaction between the Pd center and the thioether is too strong to allow competition with the olefinic part of the same substrate. One possibility to weaken this Pd-S dative interaction, and give the olefin more opportunity to react, is by increasing the steric demand of the substituent on sulfur. The next section describes the reaction of 2 with allyl tert-butyl thioether (ABT).

Scheme 5.1

SPd

MeN

N-Et2O

S

17

Pd

MeN

N OEt2

2

120

Figure 5.1 ORTEP structure of complex 17 (cation)

Table 5.1 Selected bond distances (Å) and angles (o) of complex 17

_____________________________________________________________ Pd1-N11 2.146(2) Pd1-N12 2.078(2) Pd1-S1 2.2893(9) Pd1-C133 2.054(3) S1-C129 1.806(4) S1-C130 1.824(3) C130-C131 1.503(5) C131-C132 1.310(6) N11-Pd1-N12 77.20(9) S1-Pd1-C133 89.75(8) Pd1-S1-C129 111.01(14) Pd1-S1-C130 107.93(11) C129-S1-C130 99.87(18) S1-C130-C131 110.8(2) N11-Pd1-S1-C130 132.79(14) C133-Pd1-S1-C130 -46.89(15) Pd1-S1-C130-C131 -170.9(2) C133-Pd1-S1-C129 61.61(16)

121

S1-Pd1-N11-C113 179.68(19) C129-S1-C130-C131 73.1(3) _____________________________________________________________

5.3. Reaction of [(N^N)PdMe(Et2O)][BAF] (2) with allyl tert-butyl thioether (ABT)

18a

Pd

MeN

N OEt2

2

-Et2O

S

18b

2,1-insertion1,2-insertion

NPd

N

S

Me

chain-walking

S Pd

MeN

N

NPd

N

S

Me

Scheme 5.2 The reaction of [(N^N)PdMe(Et2O)][BAF] (2) and allyl tert-butyl thioether (ABT) in CD2Cl2 solvent was followed by VT NMR. After 5 min at -60 oC, the diethyl ether originally coordinated to 2 was displaced completely from the metal. Concomitantly, the singlet resonance for the Pd-Me group at δ 0.35 ppm disappeared, and two doublets at δ 0.35 ppm (J = 7.0 Hz) and δ 0.84 ppm

122

(J = 6.7 Hz) appeared in a ratio of 2 : 3. This suggests insertion of the olefin substrate into the Pd-alkyl bond. Further NMR data, and comparison with isolated products described later in this chapter, indicated the formation of one product that derives from 2,1-insertion, [(N^N)PdCHMeCH2CH2StBu)][BAF] (18a), and one that derives from 1,2-insertion, [(N^N)PdCH2CHMeCH2StBu)][BAF] (18b) (Scheme 5.2). During the reaction sequence, neither a thioether adduct nor an intermediate alkene adduct could be observed. The product thus formed remained unchanged upon warming the reaction mixture to room temperature. Addition of excess hexane to this solution resulted in gradual formation of orange crystals over a period of two days, allowing the isolation of products in 68% yield. NMR analysis (combination of 1D and 2D NMR techniques) of the products showed that also in this material two different compounds are present in a 2:3 ratio. The minor product shows the presence of two adjacent CH2 groups, at δ 2.74 and 2.49 (CHHS) and 1.84 and 0.66 (PdCHCHH), and a doublet resonance for the Pd-bound alkyl carbon (δ 55.57 ppm, d, JCH = 132 Hz), and can be identified as 18a. The major product shows a triplet resonance for the Pd-bound alkyl carbon (δ 50.15 ppm, t, JCH = 135 Hz), consistent with its identification as 18a. A suitable crystal was found for X-ray diffraction. Its resolved structure (Figure 5.2, Table 5.2) revealed that it is the 1,2-insertion product 18b. The methyl group is attached on the β-carbon of Pd in the 5-membered chelate ring and appears to occupy the equatorial position (although conformational disorder is present in the chelate moiety). The dative Pd-S bond of 2.3209(16) Å in the chelate is elongated relative to the Pd-S bond in the AMT adduct 17 (2.2893(9) Å), probably due to the steric interference of the tBu substituent with the ligand aryl group. A suitable crystal of the product 18a could not be found, although the compounds clearly do not co-crystallize.

123

Figure 5.2 ORTEP structure of complex 18 (cation)

Table 5.2 Selected bond distances (Å) and angles (o) of complex 18

_____________________________________________________________ Pd1-N11 2.087(4) Pd1-N12 2.182(4) Pd1-S1 2.3209(16) Pd1-C129 2.041(7) S1-C131 1.828(8) S1-C132 1.872(6) N11-Pd1-N12 75.90(14) S1-Pd1-C129 82.2(2) Pd1-S1-C131 101.3(3) Pd1-S1-C132 118.9(2) Pd1-C129-C130 110.6(6) S1-C131-C130 107.6(5) N12-Pd1-S1-C131 -169.8(4) C129-Pd1-S1-C131 5.5(5) C129-Pd1-S1-C132 -106.7(4) S1-Pd1-N12-C115 163.2(3) N11-Pd1-N12-C115 -6.6(3) Pd1-S1-C131-C130 25.8(9) S1-Pd1-C129-C130 -35.3(5) N11-Pd1-C129-C130 135.0(6) C132-S1-C131-C130 149.4(8) C129-C130-C131-S1 -54.6(10) Pd1-C129-C130-C136 178.(1) _____________________________________________________ ________

124

From the contrast in product formation in the reactions of the Pd-Me cation 2 with ABT and with AMT, it appears that in the former the coordination of the thioether sulfur atom cannot compete with the coordination of the olefinic moiety and subsequent insertion into the Pd-Me bond. As the thioether in ABT electronically is certainly not a weaker Lewis base than that in AMT, it is likely that this difference is caused by steric factors. When ABT would bind to the Pd center through its sulfur atom, the sterically demanding tert-butyl group will interfere with the 2,6-diisopropylphenyl substituents on the diimine ligand. This steric hindrance will weaken the palladium-thioether interaction, allowing the olefinic moiety of the substrate to compete effectively for coordination to the metal. This then will lead to insertion of the olefin into the Pd-methyl bond. Apparently, the tert-butyl group can sterically be sufficiently accommodated when coordinating in an intra-molecular fashion, thus forming a 5-membered chelate structure. The fact that, with ABT, products derived both from 1,2- and 2,1-insertion are formed, may be related with the similar observation made in section 4.4 for N-allylcarbazole (NAC) insertion. In both substrates, a sterically demanding substituent is present not far removed from the olefin.

5.4. Reaction of [(N^N)PdMe(Et2O)][BAF] (2) with 2-allyl-1,3-dithiane (ADT)

1,3-Dithiane derivatives are often used in organic synthesis as a type of sulfur-stabilized carbanions, which can be hydrolyzed to the corresponding carbonyl compound. 6 , 7 , 8 In previous work in this research group, the reactivity of several 2-(alkenyl)-1,3-dithianes, including 2-allyl-1,3-dithiane (ADT) and 2-pentenyl-2-methyl-1,3-dithiane (PMDT), with cationic zirconocene catalysts was studied.9 Although evidence for alkene insertion was found, polymerization of these monomers was not achieved. Here we describe the interactions of these monomers with the cationic Pd species 2.

125

Scheme 5.3 The reaction of [(N^N)PdMe(Et2O)][BAF] (2) with 2-allyl-1,3-dithiane (ADT) in CD2Cl2 solvent was followed by VT NMR. After 5 min at -60 oC, the ether is again completely displaced from the Pd-center. In the product the resonances of the olefinic part of the substrate are only slightly shifted relative to those of free ADT (δ 5.63 (CHH’=CH), 5.14 (CHH’=CH) and 5.11 ppm (CHH’=CH) in the adduct versus δ 5.85 (CHH’=CH), 5.15 (CHH’=CH) and 5.11 ppm (CHH’=CH) in free ADT). This, together with the presence of a singlet for Pd-Me group (indicating that no insertion has taken place), shows that ADT coordinates via one of the dithiane sulfur atoms to the Pd-center. Upon warming the solution to room temperature, no further reaction occurred. Thus a stable 2-allyl-1,3-dithiane adduct (19, Scheme 5.3) is formed, which could be isolated as a brownish solid in 74% yield from the reaction mixture. The behavior of ADT is essentially the same as that of AMT described in section 5.2: it indicates that a secondary center adjacent to the coordinating sulfur atom is insufficient to disfavor its coordination to palladium relative to that of the olefinic moiety.

5.5. Reaction of [(N^N)PdMe(Et2O)][BAF] (2) with 2-pentenyl-2-methyl-1,3-dithiane (PMDT)

Following the same strategies of increasing the steric hindrance around the sulfur atom (as with AMT versus ABT), and increasing the spacer length

Pd

MeN

N OEt2 -Et2O

SS

192

SS

PdMeN

N

126

between the sterically demanding group and the olefinic moiety (as with NPC versus NAC, shown in Chapter 4) we studied the substrate 2-pentenyl-2-methyl-1,3-dithiane (PMDT). This provides a tertiary center near the sulfur atoms, and a C3-spacer between the tertiary center and the olefinic group. Thus PMDT is expected both to coordinate to the Pd-center via the double bond instead of sulfur, and to be able to effect insertion into the Pd-Me bond. It should provide the best possibility to aim at polyethene copolymers bearing 1,3-dithiane functionalities. The reaction of [(N^N)PdMe(Et2O)][BAF] (2) with 2-pentenyl-2-methyl-1,3-dithiane (PMDT) in CD2Cl2 was followed by VT NMR. After 5 min at -60 oC, the ether originally coordinated to Pd in 2 was liberated, and the olefinic substrate resonances had disappeared, together with the singlet of the Pd-Me group. This indicates that an insertion of the olefin into the Pd-Me bond has taken place. 1D and 2D 1H NMR spectroscopy showed that the product contains an iPr-group that is not part of the diimine ligand: δ 0.04 and 0.58 ppm (d, J = 6.8 Hz, 3H each) for the Me-groups and the methyne proton around δ 1.5 ppm (overlapped with other resonances). This is similar to the isopropyl group in the Pd-NPC three-membered chelate 15, which is derived from a 1,2-insertion followed by chain walking (section 4.3). Upon warming to room temperature, no further reaction was observed. Addition of hexane to the reaction mixture led to gradual crystallization of the product, which was isolated as red crystals in 85% yield. A single crystal structure was determined (Figure 5.3 and Table 5.3) and revealed that the compound can be formulated as {(N^N)Pd[CH(CH2iPr)CH2CMe(S2C3H6)]}[BAF] (20). It contains a 5-membered chelate ring, containing palladium, one of the dithiane sulfur atoms and the α, β and γ-carbon atoms of the substrate. The Pd-bound carbon atom bears a 2-methylpropyl substituent. The Pd-S distance in 20 of 2.2555(9) Å is noticeably shorter than the one in the other sulfur-containing 5-membered chelate, the ABT insertion product 18 (2.3209(16) Å). As in 20

127

the quaternary center is part of the chelate ring (instead of being a substituent on one of the ring atoms), it imparts much less steric hindrance around the Pd-S interaction, and the intramolecular Pd-S coordination is expected to be stronger in 20 than in 18.

Figure 5.3 ORTEP structure of complex 20 (cation)

Table 5.3 Selected bond distances (Å) and angles (o) of complex 20

_____________________________________________________________ Pd1-N11 2.146(2) Pd1-N12 2.122(2) Pd1-S11 2.2555(9) Pd1-C135 2.063(4) S11-C129 1.812(4) S11-C132 1.823(3) S12A-C131 1.844(4) S12A-C132 1.776(4) C135-C136 1.493(6) C137-C136 1.526(6)

128

N11-Pd1-N12 76.04(9) S11-Pd1-C135 86.67(11) Pd1-S11-C129 107.87(14) Pd1-S11-C132 99.85(12) C129-S11-C130 103.54(18) C131-S12A-C132 100.51(17) S11-C129-C130 110.4(3) C129-C130-C131 112.1(4) S12A-C131-C130 116.1(3) C132-C134-C135 113.6(3) S11-C132-S12A 111.6(2) S11-C132-C134 105.6(2) Pd1-C135-C134 113.4(2) Pd1-C135-C136 110.0(3) C135-Pd1-S11-C129 81.77(18) C135-Pd1-S11-C132 -26.02(16) S11-Pd1-C135-C134 4.3(3) Pd1-S11-C129-C130 -166.1(3) C132-S11-C129-C130 -60.9(3) Pd1-S11-C132-S12A 171.10(17) Pd1-S11-C132-C134 44.6(2) C129-S11-C132-S12A 59.9(2) C132-S12A-C131-C130 58.5(3) S11-C129-C130-C131 66.1(4) S11-C132-C134-C135 -47.1(4) C132-C134-C135-Pd1 25.2(4) Pd1-C135-C136-C137 -173.8(3) C135-C136-C137-C138 179.1(5) _____________________________________________________________

Scheme 5.4

-Et2O

SS

PdMeN

N OEt2

20

Insertion

PdMeN

N

SS

Chain walking

S

S

PdN

NSN

NPd

S

2

129

The formation of the 5-membered chelate complex 20 can be explained by the following sequence (Scheme 5.4). First, the olefinic moiety of the substrate displaces the ether molecule in 2 and coordinates to the Pd-center. Then, 1,2-insertion into the Pd-Me bond generates a Pd-alkyl species with a β-methyl substituent. Subsequent chain-walking via β-H elimination/rotation/insertion sequences allows the complex to reach a 5-membered chelate structure with a 2-methylpropyl -substituent on the α-carbon of the chelate ring. Again, apparently the steric hindrance imparted by the tertiary center next to the sulfur disfavors intermolecular coordination of the thioketal sulfur atoms relative to the olefinic group. The steric demand of the tertiary center is more readily accomodated in the intramolecular coordination in the 5-membered chelate.

5.6. Reactions with ethene

The reactivity of the complexes 17-20 with ethene was studied as a first evaluation to see whether any of the sulfur-containing substrates offer possibilities for copolymerization with ethene. For this it is necessary that the alkene moiety of the substrate can insert into the Pd-alkyl bond, and that the resulting insertion product (chelate complex) can react with incoming ethene substrate. The results from ethene homopolymerization attempts with 17-20 (CH2Cl2 solvent, room temperature, 5 bar ethene) are listed in Table 5.4. Although the coordination of the double bond in AMT to Pd (and subsequent insertion) was blocked by preferential thioether coordination, the S-Pd coordination in complex 17 is not so strong as to prevent the coordination of ethene to Pd and subsequent insertion into the Pd-C bond under the applied conditions. Although the productivity of 17 in ethene homopolymerization is modest (certainly considering the relatively low catalyst concentration), it was noted that, unlike in other ethene homopolymerization with (α-diimine)Pd-catalysts, with 17 there was no visible gradual precipitation of

130

black Pd(0), which is associated with catalyst deactivation. It may be that the (intermolecular) AMT thioether coordination stabilizes the catalyst to some extent, possibly depressing the polymerization rate, but also compensating that by suppressing catalyst deactivation. The molecular weight of the branched polyethene produced by 17 is higher than that of other catalysts (Mw = 564 x 103 compared to 186 x 103 of PE produced by 11 under similar conditions). The related (but more sterically encumbered) 1,3-dithiane adduct 19, formed by the coordination of ADT to the cationic Pd-Me species, is also able to give ethene homopolymerization, but considerably less efficiently than 17, also producing polyethene with a substantially lower molecular weight. The intramolecular thioether coordination in the 5-membered chelate formed by ABT insertion (18) is, despite the steric hindrance imparted by the tert-butyl substituent, apparently too strong to allow the coordination and insertion of ethene. This is also the case with the chelate 20, formed by PMDT insertion. Consequently, ethene/PMDT copolymerization attempts did not yield any polymer.

Table 5.4 Ethene homopolymerization catalyzed by Pd-S complexes

Catalyst Code

Catalyst µmol

PolymerYield (g)

Activity (kg mol-1 h-1)

Mw (x10-3)

Mw/Mn

17 2.0 1.2 25.0 564 1.98 18 2.0 0 0 19 1.9 0.23 5.0 83 2.17 20 2.0 0 0 Conditions: 15 mL CH2Cl2, 5 bar ethene, room temperature, 24 h.

131

5.7. Conclusions

Although the studies in this chapter did not result in new sulfur-containing polyethene copolymers, they do provide an interesting insight into the factors that influence metal-substrate interactions and olefin insertion of functionalized substrates. When a sterically unencumbered thioether and an olefin are present in the same substrate, the thiolate binds preferentially to the palladium center. This effectively prevents insertion of the olefinic moiety into the Pd-Me bond (although it does not prevent coordination and insertion of ethene when the latter is present in excess). This binding preference can switch to the olefin when the thioether is bearing a sterically demanding substituent. The coordination of the olefin to Pd then leads to 1,2-insertion into the Pd-Me bond. A 5-membered chelate appears to be thermodynamically the most stable product: in the case of the PMDT substrate, chain-walking leads to the 5-membered chelate exclusively. The combination of a soft thioether donor and a 5-membered chelate ring then is too stable to allow interaction of the metal center with other substrates (ethene or functionalized olefin).

5.8. Experimental

General considerations. Same as outlined for Chapters 2-4 except for the items specified below. Materials. Allyl methyl thioether (AMT, Acros 99%), was used as received. Allyl tert-butyl thioether (ABT), 10 2-allyl-1,3-dithiane (ADT) 11 and 2-pentenyl-2-methyl-1,3-dithiane (PMDT)11,12 were synthesized according to literature procedures. General VT NMR experiments of [(N^N)PdMe(Et2O)][BAF] (2) and functionalized monomers. The α-diimine palladium complex

132

[(N^N)PdMe(EtO)2][BAF] (2, about 0.02 mmol) was weighed into an NMR tube in a drybox under N2 atmosphere. The tube was then capped with a latex septum and brought out from the drybox. Solutions of monomers (1 eqv.) in CD2Cl2 (0.7 mL) were injected by a syringe into the NMR tube that was cooled at -196 oC by liquid N2 and then the septum was wrapped with Parafilm. Upon thawing out the tube was shaken briefly to dissolve the palladium complex and transferred to the pre-cooled probe of the NMR spectrometer. Spectra were acquired at regular temperature intervals. Complex 2 and allyl methyl thioether. The reaction of [(N^N)PdMe(Et2O)][BAF] (2, 14 mg, 0.01 mmol) and allyl methyl thioether (AMT, 0.1 mL, 1.4 mmol) in CD2Cl2 was followed by VT NMR. At -60 oC, the coordinated ether was fully displaced by AMT. The spectral characteristics of the product are those of 17 (described below). Upon warming to room temperature no further reaction ensued. Synthesis of [(N^N)PdMe(κ1-MeSCH2CH=CH2)][BAF] (17). Allyl methyl thioether (AMT, 0.01 mL, 0.14 mmol) was added to a solution of [(N^N)PdMe(Et2O)][BAF] (2, 66.1 mg, 0.045 mmol) in CH2Cl2 (10 mL). The solution was stirred at 20 oC overnight. After addition of pentane (20 mL) the solution was kept at -30 oC for 2 days. Yellow crystals of 17 (58 mg, 0.039 mmol, 87%) were isolated. 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 7.44 – 7.32 (m, 6H, Haryl), 5.49 (m, 1H, CHH’=CH), 5.27 (d, J = 10.1, 1H, CHH’=CH), 5.16 (d, J = 17.1, 1H, CHH’=CH), 2.95 (d, 2H, J = 7.7, CH2=CHCH2), 2.83 (septet, 4H, J = 6.8, iPr CH), 2.27 and 2.23 (s, 3H each, N=CMe), 1.88 (s, 3H, SMe), 1.35, 1.31, 1.22 and 1.19 (d, 6H each, J = 6.8, iPr Me), 0.42 (s, 3H, PdMe). 13C NMR (CD2Cl2, 126MHz, 25 oC) δ 179.08 and 174.11 (N=CMe), 141.19 and 140.44 (Ar Cipso), 138.54 and 137.71 (Ar Co), 129.29 (d, JCH =162, CH2=CH), 129.03 and 128.92 (Ar Cp), 124.97 and 124.91 (Ar Cm), 123.00 (t, JCH =159, CH2=CH), 40.96 (t, JCH =144, CH2=CHCH2), 29.62 and 29.39 (iPr CH), 23.76, 23.67 and 23.11 (iPr Me), 21.85 and 21.12 (N=CMe), 17.89 (q, JCH =144, SMe), 12.74 (q, JCH =137, PdMe). Anal. Calcd for (C65H63N2BF24PdS): C, 52.84; H, 4.30; N, 1.90. Found: C, 52.75; H, 4.25; N,

133

1.82. [(N^N)Pd(CHMeCH2CH2StBu)][BAF] (18a) and [(N^N)Pd(CH2CHMeCH2StBu)][BAF] (18b). The reaction of [(N^N)PdMe(Et2O)][BAF] (2, 38 mg, 0.026 mmol) and allyl tert-butyl thioether (ABT, 0.059 mL, 1.5 eqv., 0.039 mmol) in CD2Cl2 was followed by VT NMR. After 5 min at -60 oC, the coordinated ether had been liberated and the palladium species had been converted completely to a 2 : 3 mixture of two products 18a and 18b derived from ABT 1,2-insertion and 2,1-insertion. Subsequent warming to room temperature did not induce further reaction. 2 mL of hexane was layered on top of the mixture and the tube was left to stand at room temperature for 2 days. Orange crystals (27 mg, 0.018 mmol, 68%) were isolated. The crystalline products also consisted of 18a and 18b in a 2 : 3 ratio. 1H NMR (CD2Cl2, 500MHz, 25 oC) 18a: δ 7.45 – 7.23 (m, 6H, Haryl), 3.84, 3.51, and 3.13 (1H, 1H and 2H, iPr CH), 2.74 (dd, J = 5.5, J = 10.6, CHHS), 2.49 (overlapping, 2H, CHHS and PdCH), 2.27 and 2.18 (s, 3H each, N=CMe), 1.84 (PdCHCHH), 1.49 – 1.08 (m, iPr Me), 0.95 (s, 9H, tBu), 0.66 (m, PdCHCHH), 0.35 (d, 3H, J = 7.0, Me); 18b: δ 7.45 – 7.23 (m, 6H, Haryl), 3.22, 2.95 and 2.6 (1H, 2H and 1H, iPr CH), 2.6 and 2.16 (CHHS), 2.24 and 2.22 (s, 3H each, N=CMe), 1.80 (t, J = 8.8, PdCHH), 1.67 (m, PdCH2CH), 1.57 (m, PdCHH), 1.49 – 1.08 (m, iPr Me), 0.97 (s, 9H, tBu), 0.84 (d, 3H, J = 6.7, Me). 13C NMR (CD2Cl2, 126MHz, 25 oC) 18a: δ 177.55 and 173.76 (N=CMe), 55.57 (d, JCH =132, PdCH), 38.94 (t, JCH =117, PdCHCH2), 30.45 (t, JCH =142, CH2S), 29.23 (q, JCH =120, tBu Me), 21.36 and 20.98 (N=CMe), 19.36 (q, JCH =124, Me); 18b: δ 177.96 and 173.44 (N=CMe), 50.15 (t, JCH =135, PdCH2), 40.51 (t, JCH =129, CH2S), 39.75 (d, JCH =125, PdCH2CH), 29.43 (q, JCH =124, tBu Me), 22.11 and 21.47 (N=CMe), 19.62 (q, JCH =124, Me). Anal. Calcd for (C68H69N2BF24PdS): C, 53.75; H, 4.58; N, 1.84. Found: C, 53.88; H, 4.65; N, 1.79. {(N^N)PdMe[κ1-(S2C4H6)CH2CH=CH2]}[BAF] (19). The reaction of [(N^N)PdMe(Et2O)][BAF] (2, 48 mg, 0.033 mmol) and 2-allyl-1,3-dithiane (ADT, 5.4 mg, 1.0 eqv., 0.033 mmol) was followed by VT NMR. After 5 min

134

at -60 oC, all of the coordinated ether had been displaced by ADT to form adduct 19. Subsequent warming to room temperature did not induce further reaction. After a small amount of black precipitate was filtered off, the solution was layered with 2 mL of hexane and the tube was left to stand at room temperature for a week. Brownish solid 19 was isolated (38 mg, 0.024 mmol, 74%). 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 7.44 – 7.32 (m, 6H, Haryl), 5.63 (m, CHH=CH), 5.14 (d, J = 9.4, CHH=CH), 5.11 (d, J = 16.6, CHH=CH), 3.93 (dd, J = 9.9, J = 3.6, CHS2), 3.0 – 2.6 (m, 4H, iPr CH),2.9 and 2.6 (SCHH), 2.7 (S2CHCHH), 2.4 (PdSCHH), 2.3 (S2CHCHH), 2.27 and 2.24 (s, 3H each, N=CMe), 2.07 and 1.65 (m, SCH2CHH), 1.3 (PdSCHH), 1.4 ~ 1.1 (24H, iPr Me), 0.43 (s, 3H, PdMe). 13C NMR (CD2Cl2, 126MHz, 25 oC) δ 179.23 and 174.49 (N=CMe), 143.19 and 140.83 (Ar Cipso), 140.13, 139.27 and 138.62 (Ar Co), 131.59 (d, JCH =160, CH2=CH), 129.37 and 129.13 (Ar Cp), 125.15, 125.10, 125.03 and 124.99 (Ar Cm), 119.72 (t, JCH =155, CH2=CH), 51.25 (d, JCH = 137, CHS2), 39.68 (t, JCH = 123, CH2=CHCH2), 37.79 (t, JCH = 137, SCH2), 34.88 (t, JCH = 141, PdSCH2), 29.93, 29.43 and 29.11 (iPr CH), 26.56 (t, JCH = 124, SCH2CH2), 24.20, 23.86, 23.72, 23.62, 23.53, 23.35, 23.24 and 23.11 (iPr Me), 21.88 and 21.56 (N=CMe), 10.58 (q, JCH =137, PdMe). Anal. Calcd for (C68H67N2BF24PdS2): C, 52.71; H, 4.36; N, 1.81. Found: C, 52.84; H, 4.31; N, 1.74. {(N^N)Pd[Me2CH2CHCH2Me(S2C4H6)]}[BAF] (20). The reaction of [(N^N)PdMe(Et2O)][BAF] (2, 17 mg, 0.012 mmol) and 2-pentenyl-2-methyl-1,3-dithiane (PMDT, 80 mg, 0.40 mmol) was followed by VT NMR. After 5 min at -60 oC, all of the coordinated ether had been displaced and the Pd species was converted completely to the chelate complex 20. Subsequent warming to room temperature did not induce further reaction. After a small amount of black precipitate was filtered off, the solution was layered with 2 mL of hexane and the tube was left to stand at room temperature for a week. Red crystals of 20 (16 mg, 0.01 mmol, 85%) were isolated. 1H NMR (CD2Cl2, 500MHz, 25 oC) δ 7.5 – 7.3 (m, 6H, Haryl), 3.45, 2.88, 2.68 and 2.50 (1H each, iPr CH), 3.0 (PdSCHH), 2.9 and 2.5 (SCHH), 2.6 (S2CCHH), 2.29 and 2.21 (s, 3H each, N=CMe), 2.13 (m,

135

PdCH), 2.00 (m, SCH2CHH), 1.85 (d, J = 13.5, PdSCHH), 1.72 (s, 3H, MeCS2), 1.5 (CHMe2), 1.5 - 1.1 (24H, iPr Me), 1.4 (SCH2CHH), 1.2 and 0.5 (CHHCHMe2), 1.1 (S2CCHH), 0.58 and 0.04 (d, 3H each, J = 6.8, CHMe2). 13C NMR (CD2Cl2, 126MHz, 25 oC) δ 177.51 and 173.30 (N=CMe), 143.92 and 142.71 (Ar and Ar’: Cipso), 139.11, 138.76, 138.24 and 137.62 (Ar and Ar’: Co), 129.38 and 128.80 (Ar and Ar’: Cp), 125.55, 125.36, 125.17 and 124.63 (Ar and Ar’: Cm), 101.79 (CS2), 47.88 (t, JCH = 127, S2CCH2), 45.25 (t, JCH = 121, CH2CHMe2), 33.11 (t, JCH = 131, PdSCH2), 42.05 (d, JCH = 106, PdCH), 29.93, 29.36, 28.91 and 27.89 (iPr CH), 27.72 (q, JCH = 122, MeCS2), 25.40, 25.28, 24.27, 24.20, 24.04 and 22.99 (iPr Me), 25.13 (t, JCH = 125, SCH2), 24.89 (t, JCH = 121, SCH2CH2), 23.97 and 19.23 (q, JCH = 120, CHMe2), 23.21 and 21.04 (N=CMe). Anal. Calcd for (C71H73N2BF24PdS2): C, 53.58; H, 4.62; N, 1.76. Found: C, 53.52; H, 4.57; N, 1.76. General procedure for polymerizations Following the same procedure as described in the experimental section of chapter 4, compounds 17, 18, 19 and 20 (2.0 µmol) were tested as catalysts for ethene polymerizations in 15 mL of CH2Cl2 under 5 bar ethene at room temperature over 24 h. Only in the case of 17 and 19 was polymer product formed, in both cases branched ethene homopolymers, as seen by NMR spectroscopy. X-ray Crystallographic analysis For the general features of the structure determinations, see the experimental section of Chapter 2. Crystallographic data can be found in Table 5.5. For complex 17, the structure was solved by direct methods with SIR-9713. The structure of 18 was determined at a relatively high temperature (200 K) to avoid a low-temperature phase transition. Some atoms showed unrealistic displacement parameters when allowed to vary anisotropically, suggesting

136

dynamic disorder (dynamic means that the smeared electron density is due to fluctuations of the atomic positions within each unit cell, in which the disorder is compensated by the large displacement parameters). Disorder is especially apparent in the anion CF3-groups, but also in the chelate moiety. This could not be satisfactorily modeled, however. In the structure of complex 20, the 1,3-dithiane ring displays conformational disorder around S2. This was modeled in two fractions, the s.o.f. of the major fraction refined to a value of 0.797(8).

Table 5.5. Crystallographic data for complex 17, 18 and 20.

_____________________________________________________________ 17 18 20 Formula C65H63N2BSF24Pd C68H69N2BSF24Pd C71H73N2BS2F24Pd FW 1477.49 1519.57 1591.70 Crystal system triclinic monoclinic monoclinic Space group P-1 P21/n P21/c a (Å) 12.6582(7) 18.8508(8) 13.6269(8) b (Å) 14.1099(8) 18.1612(8) 12.8646(7) c (Å) 19.248(1) 21.3696(9) 41.322(2) α (deg) 101.151(1) β (deg) 100.654(1) 103.372(1) 98.936(1) γ (deg) 92.389(1) V (Å3) 3304.1(3) 7117.6(5) 7156.0(7) θ range (deg) 2.29-24.82 2.45-23.49 2.18-27.54 Z 2 4 4 ρ calc (g.cm-3) 1.485 1.418 1.477 F(000) 1500 3096 3248 µ(Mo Kα ), cm-1 4.21 3.93 4.23 Temp (K) 100(1) 200(1) 100(1) Reflections 15757 16105 19101 Parameters 859 888 933 wR(F2) 0.1363 0.2648 0.1428 Weighting (a, b) 0.0744, 0.0 0.1478, 10.8586 0.0545, 13.4719 R(F) 0.0524 0.0839 0.0570

137

GooF 1.037 1.049 1.058 _____________________________________________________________

5.9. References

1 Tempel, D. J.; Johnson, L. K.; Huff, R. L.; White, P. S.; Brookhart, M. J. Am. Chem. Soc. 2000, 122, 6686.

2 Rix, F. C.; Brookhart, M.; White, P. S. J. Am. Chem. Soc. 1996, 118, 4746.

3 Tempel, D. J.; Brookhart, M. J. Am. Chem. Soc. 2000, 122, 6686.

4 Dilworth, J. R.; von Beckh W., C. A. M.; Pascu, S. I. Dalton Trans. 2005, 2151.

5 Contu, F.; Demartin, F.; Devillanova, F. A.; Garau, A.; Isaia, F.; Lippolis, V.; Salis, A.; Verani, G. J. Chem. Soc. Dalton Trans. 1997, 4401.

6 Seebach, D. Synthesis 1969, 17.

7 Seebach, D. Angew. Chem., Int. Ed. Engl. 1979, 18, 239.

8 Bulman Page, P. C.; Van Niel, M. B.; Prodger, J. C. Tetrahedron 1989, 45, 7643.

9 Bijpost, E. Ph. D. Thesis, University of Groningen, 1996.

10 Kim, J. K.; Caserio, M. C. J. Org. Chem. 1979, 44, 1897.

11 Seebach, D.; Corey, E. J. J. Org. Chem. 1975, 40, 231.

12 Turpin, J. A.; Weigel, L. O. Tetrahedron Lett. 1992, 33, 6563.

138

13 Altomare, A.; Burla, M.C.; Camalli, M.; Cascarano, G. L.; Giacovazzo,

C.; Guagliardi, A.; Moliterni, A.G.G.; Polidori, G. & Spagna, R. J. Appl. Cryst. 1999, 32, 115-119. SIR-97. A Package for crystal structure solution by direct methods and refinement. Univ. of Bari, Univ. of Perugia and Univ. of Roma, Italy.

139

Samenvatting Polyolefines vormen een zeer belangrijke klasse van polymere materialen, met een breed scala aan toepassingen. Ze worden industrieël geproduceerd op een enorme schaal (meer dan 85 miljoen ton per jaar). Het grootste deel van deze productie betreft lineair polyetheen (HDPE), copolymeren van etheen met 1-alkenen (LLDPE) en isotactisch polypropeen (iPP). Deze (co-)polymeren worden alle vervaardigd door katalytische (co-)polymerisatie van olefines met behulp van overgangsmetaal-katalysatoren. Daar deze polyolefines eigenlijk uit hele lange alkaanmoleculen bestaan, zijn deze materialen alle zeer apolair. Het kan gunstig zijn om, met behoud van de andere (met name mechanische) eigenschappen, deze materialen minder apolair te maken. Dit kan onder meer het verven en kleuren van het materiaal vergemakkelijken en ook de hechting aan meer polaire materialen (b.v. glasvezel of andere polymeren zoals polyacrylaat of polycarbonaat) bevorderen. De meest voor de hand liggende en flexibele methode om dit te bereiken is in principe het inbouwen van olefine-monomeren met polaire groepen (b.v. esters, ethers, hydroxylgroepen) in de polyolefines via katalytische co-polymerisatie. Dit blijkt in de praktijk echter zeer moeilijk te zijn, doordat de katalysatoren die etheen en propeen kunnen polymeriseren, bijzonder reactief zijn: zij reagereren meestal op ongewenste wijze met de polaire functionaliteiten, wat leidt tot een zeer sterke vermindering van de productiviteit van de katalysatoren, tot complete deactivering aan toe. De laatste jaren is een aanzienlijke onderzoeksinspanning besteed aan het zoeken van geschikte strategieën voor de inbouw van gefunctionaliseerde comonomeren in polyolefines door middel van katalytische copolymerisatie. Voor een verdere ontwikkeling van dit gebied is het van belang, meer te weten te komen over de aard en de sterkte van de interacties van de diverse polaire functionaliteiten met polymerisatie-katalysatoren. Het werk, dat in dit proefschrift beschreven staat, probeert hieraan een bijdrage te leveren.

140

Een groep van katalysatoren voor olefinepolymerisatie, die een aanzienlijke tolerantie voor polaire functionaliteiten heeft, bestaat uit kationische alkyl complexen van de groep 10 metalen nikkel en palladium, met bidentaat Lewis base liganden. In dit proefschrift wordt een katalysatorsysteem behorende tot deze familie als test-katalysator gebruikt: het palladium α-diimine complex [(N^N)PdMe(OEt2)][BAF] (2, N^N = ArN=CMe-CMe=NAr with Ar = 2,6-diisopropylphenyl; BAF = B[3,5-(CF3)2C6H3]4). De reacties van deze katalysator met een reeks van heteroatoom-gefunctionaliseerde olefines zijn onderzocht, alsmede het gedrag in de katalytische co-polymerisatie van etheen met deze olefines. Hoofdstuk 1 geeft een overzicht van de verschillende problemen die spelen bij het inbouwen van polaire functionaliteiten in polyolefine-materialen, en de diverse strategieën die hiervoor in het verleden ontwikkeld zijn. In Hoofdstuk 2 wordt een nieuwe blik geworpen op de synthese van de kationische α-diimine palladium katalysator 2. Een nieuwe, vereenvoudigde synthese van de uitgangsstof (N^N)PdMe2 (1) wordt beschreven. Het blijkt dat de reactieomstandigheden zeer sterk de vorming van complex 2 beïnvloeden. Als bijproduct in deze reactie wordt het µ-methyl,µ-methyleen complex [(N^N)Pd(µ-CH3)(µ-CH2)Pd(N^N)][BAF] (3) beschreven. Het ontstaat door de reactie van 2 met de uitgangsstof 1. De moleculaire and electronische structuur van 3 wordt beschreven, alsmede enige aspecten van de reactiviteit. Hoofdstuk 3 beschrijft de reactiviteit van de kationische Pd-katalysator 2 met olefines die zuurstof-bevattende functionaliteiten hebben: acroleïne dimethyl acetaal (ADMA), allyl ethyl ether (AEE) and 2-vinyl-1,3-dioxolaan (VDO). AEE en ADMA geven 1,2-insertie in de Pd-methyl binding van 2 onder vorming van de 5-ring chelaatcomplexen [(N^N)Pd(CH2CHMeCH2OEt)][BAF] (9) en {(N^N)Pd[CH2CHMeCH(OMe)2]}[BAF] (11). Beide chelaatcomplexen

141

kunnen de homopolymerisatie van etheen katalyseren. Pogingen tot etheen/AEE copolymerisatie leidden echter tot snelle katalysator-deactivering onder vorming van het allylcomplex [(N^N)Pd(η3-1-CH2CHCH2)][BAF] (10) en ethanol. Desondanks konden etheen en ADMA succesvol worden gecopolymeriseerd tot een vertakt polyetheen copolymeer met acetaal-functionaliteiten. Ook hier treedt (langzame) katalysatordeactivering op onder vorming van het allylcomplex [(N^N)Pd(η3-1-CH2CHCHOMe)][BAF] (12) en methanol. Deze deactivering kon worden vertraagd door het toevoegen van kleine hoeveelheden methanol aan het reactiemengsel. Daarentegen werd het monomeer VDO snel door 2 aangetast via opening van de dioxolaanring; het kon daardoor niet worden gecopolymeriseerd. Hoofdstuk 4 beschrijft de reactiviteit van de kationische Pd-katalysator 2 met olefines die stikstof-bevattende functionaliteiten hebben: allyl dimethyl amine (ADA), N-allyl carbazool (NAC) and 5-pentenyl carbazool (NPC). ADA reageert met 2 via 1,2-insertie tot het 5-ring chelaatcomplex [(N^N)Pd(CH2CHMeCH2NMe2)][BAF] (14). In tegenstelling tot de chelaatcomplexen beschreven in hoofdstuk 2, katalyseert dit complex niet de homopolymerisatie van etheen. Dit hangt samen met de sterke Lewis base donoreigenschappen van de tertiaire amine groep. NAC en NPC reageren met 2 tot ongebruikelijke 3-ring chelaatcomplexen, via insertie van het olefine in de Pd-Me binding gevolgd door “ketenwandelen” van het metaalcentrum. De moleculaire structuur van het product van 2 met NPC, {(N^N)Pd[Me2CHC2H4CHN(C6H4)2]}[BAF] (15), werd bepaald. Het 3-ring chelaatcomplex reageert snel met etheen, en etheen en NPC konden succesvol worden gecopolymeriseerd tot vertakte polyetheen copolymeren met carbazool-functionaliteiten. De fluorescentie van de carbazoolgroepen in deze copolymeren is sterk afhankelijk van de hoeveelheid ingebouwd comonomeer. NAC wordt, in tegenstelling tot NPC, niet makkelijk met etheen gecopolymeriseerd. Dit ligt waarschijnlijk aan de sterische hinder uitgeoefend door de carbazoolgroep, die zich in NAC in de nabijheid van de olefinegroep bevindt.

142

Hoofdstuk 5 beschrijft de reactiviteit van de kationische Pd-katalysator 2 met olefines die zwavel bevattende functionaliteiten hebben: allyl methyl thioether (AMT), allyl tert-butyl thioether (ABT), 2-allyl-1,3-dithiaan (ADT) en 2-pentenyl-2-methyl-1,3-dithiaan (PMDT). AMT reageert met 2 tot het stabiele thioether adduct [(N^N)PdMe(κ1-MeSCH2CH=CH2)][BAF] (17): de zachte Lewis basische thioether groep bindt sterk aan het zachte metaalcentrum en blokkeert zo de insertie van het olefine in de Pd-methyl binding. In de substraten ABT en PMDT is het zwavelatoom beter afgeschermd door sterische hinder en deze monomeren inserteren in de Pd-Me binding tot 5-ring chelaatcomplexen, waarvan de moleculaire structuur kon worden bepaald. In deze chelaatcomplexen is de intramoleculaire Pd-S interactie echter zo sterk, dat dit verdere reactie met etheen (en dus ook de mogelijkheid tot copolymerisatie) verhindert.

143

Acknowledgment When I started to recall all the people whom I appreciate for help me make this thesis realized during the 5 years, the first one came up to my mind is Prof. Jan Teuben. Without his strong support, I would even not be able to stay in the Netherlands so long. From the moment he selected me as a Ph.D candidate in his group, there were so many things inscribed in my memory, research guidance, teaching me how to scientifically analysis and resolve problem, improving my English, helping me on my orientation in the Netherlands. His kind help made me adapt for Dutch cultural much easier and feel warm as home. What I learned from him is far more than chemistry. I was deep influenced by his kindness, intelligence, well organizing and humors. I feel regret I could not finish my thesis before his retirement. However, I will remember him forever as my supervisor, a wise elder and a good friend who I can always rely on. Prof. Bart Hessen is the second person I would like to acknowledge. This thesis would never been finished without his help. I am grateful to him for the insightful discussion, the strict requirement on the research and the multi-modification of my manuscript. His painstaking was marked on every page of this book. I was so benefited from his broad knowledge and precise attitude on study. I am proud of being guided by him. I also would like to thank the people in Teuben/Hessen group and Hummelen group, who provided a friendly and enjoyable working environment during my Ph.D study. Within the limitation of the page, I try to name some of them who gave me so much help on my work, Xiaochun Zhang, Marco Bouwkamp, Dirk Beetstra, Sergio Bambirra, Winfried Kretschmer, Patrick Deckers, Timo Sciarone, Shaozhong Ge, Elena Novarino, Victor Quiroga, Cindy Visser, Edwin Otten, Steven Boot, Andries Jekel, Daan van Leusen, Guohua Liu, Aurora Batinas, Thomas Koch, Erica Jellema, Menno Brandsma, Niels Tazelaar, Peter Dijkstra, Bodo Richter, Piet-Jan Sinnema, Coen van de Brom,

144

Pavel Shutov, Christian Nijhuis, Jan Helmantel, Wolter Beukema, Oetze Staal, Alfred Boonstra, Marten Hettinga, and many persons who have worked in Teuben/Hessen group for a long or short term. I am also indebted to Dr. Peter Budzelaar (Radboud University Nijmegen, presently at University of Winnipeg) for the DFT calculations, Auke Meetsma for the X-ray structures, Klaas Dijkstra for help on NMR and Joop Vorenkamp for GPC measurement. Besides chemistry, I got numerous aids from my Chinese friends in Groningen as well. Zhongxiang Zhang, Guoxin Li, Yi Liao, family Li (Miao) and many kind Chinese have lent their hands on me and made my life more pleasurable. I could not mention all their names here but they will be remembered by my heart forever. In the early period when I just arrived in the Netherlands, I had been helped by many Dutch people in airport, train stations, streets, campus, supermarket, hospital and many places where I have been. Especially my neighbor, Mr. W. Weemhoff (Bill), helped me a lot on the settlement of my family. I would like to thank all of them although I do not know all their names. Last but not least, I heartily thank my son, Lemeng, and my wife, Jichi, who shared all my happiness and sadness in the years and constantly supported me as much as they could. I could not get through the hard time during my Ph.D study without their love. They are my motivity for all the time.