Methylaluminoxane-Catalyzed

Jul 10, 2015 - DFT modeling of the post-titanocene catalytic system LTiCl2-Bu2Mg-Et2AlCl for alkene polymerization: The role of alkyl bridge Mg–C–...
4 downloads 9 Views 6MB Size
Article pubs.acs.org/Organometallics

Toward Controlling the Metallocene/Methylaluminoxane-Catalyzed Olefin Polymerization Process by a Computational Approach Mikhail S. Kuklin,† Janne T. Hirvi,† Manfred Bochmann,‡ and Mikko Linnolahti*,† †

Department of Chemistry, University of Eastern Finland, Joensuu Campus, FI-80101 Joensuu, Finland Wolfson Materials and Catalysis Centre, School of Chemistry, University of East Anglia, Earlham Road, Norwich, NR4 7TJ, United Kingdom



S Supporting Information *

ABSTRACT: We describe and compare the proposed mechanisms of ethene polymerization by the metallocene/ methylaluminoxane (MAO) catalyst in terms of quantum chemical calculations. In combination with the Cp2ZrMe2 precatalyst, we employ two models for MAO, produced by hydrolysis of trimethylaluminum (TMA). Both MAOs contain associated TMA as a key ingredient for cocatalytic activity. The TMA association/dissociation equilibrium in the MAOs controls the mechanism of catalyst activation and suggests preference for catalyst activation via [AlMe2]+ abstraction from the MAO by the precatalyst rather than via Lewis-acidic abstraction of the leaving group from the precatalyst by the MAO. Solvent interactions increase the relative concentration of Lewis-acidic sites. Chlorination of MAO facilitates the catalytic processes. Studies as a function of precatalyst structure reproduce the general experimental observations of the easier catalyst activation by zirconocenes than by hafnocenes and the positive effects of adding a dimethylsilyl bridge and replacing the cyclopentadienyl with an indenyl ligand. This study provides a starting point for rational control of the behavior of the metallocene/MAO catalyst system.



INTRODUCTION Group 4 metallocenes are excellent catalysts for the preparation of polyolefins, methylaluminoxane (MAO) playing a key role in the process.1 As a cocatalyst, MAO activates the metallocene precatalyst in a series of reactions, which are not properly understood,2 eventually yielding [catalyst]+[MAO]− ion pairs as catalytically active species.3 The inability of precisely addressing the nature of the active species and the mechanism of its formation is a consequence of the complex structure of MAO. Moreover, the unknown composition of MAO, possibly a mixture, has led to its inefficient use: a large excess of MAO is typically needed for catalyst activation, and MAO is in fact the major contributor to catalyst costs.3−7 In the absence of structural characterization of the catalytically active species, two activation mechanisms have been proposed for the ion pair formation: (1) direct Lewis-acidic abstraction of a Me− or Cl− ligand from the precatalyst3,8,9 and (2) transfer of an [AlMe2]+ end group from MAO to the precatalyst.10,11 Various structural alternatives have been proposed for MAO, including chains, rings, cages, and extended structures, such as nanotubes.6,12−27 Size estimates vary, depending on the method of measurement and on assumptions on the packing density of MAO clusters.10,28−31 Spectroscopic studies,4,10,31 together with recent computational explorations on the preparation of MAO by hydrolysis of TMA,32−36 point to nonstoichiometric Al:O:Me ratios, which can be explained by the association of trimethylaluminum (TMA) into the MAO © XXXX American Chemical Society

structures. TMA association has been shown to lead to the formation of AlMe2 end groups,37 thereby presumably generating the sites in MAO that are required for catalyst activation.36 In principle, the single-site nature of the metallocene catalysts allows precise control over the polymerization process through systematic modification of the catalyst structure. In the control of the microstructure of the polyolefin product, the counterion typically has no significant role, and as a consequence, the stereo- and regiocontrol are well explained by consideration of the metallocene ligand structure.38 Generally, catalytic properties are difficult to control by consideration of the metallocene structure alone, as they are strongly dependent on many factors, including polymerization conditions and not least the structure of the cocatalyst.39,40 The latter is particularly difficult to address, because specific modification of MAO is not feasible, since its structure is unknown. Nonetheless, various modifications of MAO have been studied experimentally,41−44 and compositional modification is possible, e.g., using higher aluminoxanes, such as tetraisobutylaluminoxane,45,46 or modifying TMA content, which plays a significant role. Addition of TMA decreases the molecular weight of MAO due to Al−O bond cleavage.37 Catalyst activity is dependent on the TMA concentration. Increasing the TMA content has been reported Received: May 8, 2015

A

DOI: 10.1021/acs.organomet.5b00394 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

that spectroscopic studies on metallocene/MAO systems have shown the activation process to involve homobinuclear metallocene cations,53−55 which are ignored in this context due to computational complexity. Furthermore, there is no evidence that these species play a role once the polymerization process is under way. We start by showing how the two proposed activation mechanisms become linked through reactions involving TMA and then continue by investigation of ethene insertion and chain propagation, followed by evaluation of the effects of solvent, MAO chlorination, and catalyst structure. Catalyst Activation and Incorporation of the First Ethene Monomer. The structural connection between MAO1 and MAO2 is shown in Figure 1. The two MAOs are in equilibrium through an association/dissociation reaction with TMA. This is due to the presence of a potentially active three-coordinate Lewis-acidic Al center in MAO1, which binds TMA so that the Al becomes four-coordinate. TMA association leads to the formation of a terminal AlMe2 moiety, involving a bridging methyl group with a pentavalent carbon. MAO1, and its TMA association product, MAO2, exhibit two possible reaction mechanisms for catalyst activation: (1) direct Lewisacidic abstraction of [Ligand]− from the catalyst3,8,9 or (2) transfer of [AlMe2]+ from MAO to the precatalyst10,11 followed by displacement of TMA by the incoming olefin monomer.36 As a consequence, both activation mechanisms eventually lead to identical catalyst−MAO complexes, as will be demonstrated below, and are thus linked via a TMA association−dissociation equilibrium. The reaction energy profiles for Cp2ZrMe2 activation and initiation of the ethene polymerization process by both MAO1 and MAO2 are presented in Figure 2 relative to the neutral, noninteracting Cp2ZrMe2 and MAO2. The relative energies (ΔE) are tabulated in Table 1, and the optimized structures of all intermediate species are visualized in Figures 3 and 4. Figure 2a shows directly comparable reaction profiles for MAO1 and MAO2. The energy of MAO1-A, 122.8 kJ mol−1 higher than that of MAO2-A, reflects the cost of TMA dissociation from MAO2. While entropy effects lower the energy difference by ca. 40 kJ mol−1 at 25 °C, the equilibrium remains strongly on the side of the TMA-associated MAO2.36 For comparison and to illustrate the temperature effect, at 150 °C the equilibrium remains on the side of MAO2 by ca. 60 kJ mol−1. Catalyst activation by MAO1 takes place through direct Lewis-acidic abstraction of Me− from Cp2ZrMe2, with ΔE = −139.4 kJ mol−1 (MAO1-B), thus bringing the energy down to 16.6 kJ mol−1 below the reference level. The abstraction yields partial cation−anion separation, where the methyl group is clearly abstracted from Zr by the Lewis-acidic Al atom of MAO (Zr− Me = 2.51 Å; Al−Me = 2.08 Å). The incoming ethene monomer (MAO1-G* trans) then fully separates the catalyst− MAO complex, resulting in the formation of an ethene πcoordinated zirconocene cation and MAO anion (MAO1-H trans). Ethene insertion takes place with a barrier of only 19.1 kJ mol−1, the transition state being stabilized by an α-agostic interaction (MAO1-I* trans). Relaxation to the propyl product (MAO1-K trans) shows no agostic interaction due to stabilization of the metallocene cation by a stronger metallocene−MAO interaction via the abstracted methyl group. The product formation is accompanied by a significant decrease in energy. The route for catalyst activation significantly changes upon association of TMA into MAO1 to form MAO2. As we have

to improve ethene polymerization activity, possibly via stabilization of the active species and extending the catalyst lifetime. TMA may also play a positive role due to its function as impurity scavenger and for accelerating the alkylation step.5,47,48 On the other hand, increasing TMA concentrations have been reported to suppress catalytic activities, possibly if the precatalyst has already been alkylated.49,50 Moreover, the TMA content has been reported to affect the molecular weight of the polymer and the comonomer incorporation.5,48,51,52 The motivation of the present theoretical study lies in the need for deepening the understanding of the metallocene/ MAO catalyst system to enable control of the process as a whole. Utilizing previously reported models for MAO, which contain associated TMA and are thereby capable of generating the active species,36 we describe the mechanisms of the two proposed pathways for generation of the catalytically active species and for the subsequent growth of the polyethene chain. We then continue by evaluation of factors affecting the processes, taking into consideration the effects of solvent together with structural modifications taking place on either the cocatalyst or the metallocene precatalyst.



RESULTS AND DISCUSSION Previous computational studies on the preparation of MAO by hydrolysis of TMA have shown the formation of MAOs of general formula (MeAlO)n(AlMe3)m, where n is the degree of MAO oligomerization and m is the number of associated TMA molecules.32−36 At small values of n, the thermodynamically favored MAOs resemble chains, rings, or sheets, while the studies suggest that as the value of n increases, cage-like structures become dominant.35,36 The point of transition to cages is not known. In line with the experimentally observed Al:O:C ratios,4,10,31 all the thermodynamically feasible MAOs contain associated TMA, which has been suggested to be a key ingredient regarding cocatalytic activity.33,36 In this context, we employ two previously proposed TMAassociated MAO cages as model cocatalysts: (MeAlO)8(AlMe3) = MAO1 and (MeAlO)8(AlMe3)2 = MAO2 (Figure 1).36 While not the lowest energy compositions in terms of TMA content, they possess sites capable of catalyst activation either by direct methyl or chloride ligand abstraction from the metallocene precatalyst or via [AlMe2]+ transfer from the MAO to the precatalyst, and thus serve as useful model systems. Note

Figure 1. TMA association/dissociation equilibrium between MAO1 and MAO2. Active sites of MAO1 and MAO2 are colored yellow and purple, respectively. Schematic structures for the active sites are given at the top. B

DOI: 10.1021/acs.organomet.5b00394 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

Figure 2. Reaction energy profiles for Cp2ZrMe2 precatalyst activation and initiation of the ethene polymerization process by (a) MAO1 and MAO2 via trans routes and (b) MAO1 via cis outer and inner routes. The energies are given relative to the neutral, noninteracting Cp2ZrMe2 and MAO2. Optimized structures of the intermediates are visualized in Figures 3 and 4.

recently demonstrated,36 catalyst activation starts with the methyl ligand of Cp2ZrMe2 approaching the four-coordinate aluminum of the AlMe2 moiety in the MAO, and this results in

a Zr−Me−Al bridge structure (MAO2-B). Bending of the Zr− Me−Al moiety, which takes place via a shallow minimum (MAO2-D), enables cleavage of the [AlMe2]+ cation from the C

DOI: 10.1021/acs.organomet.5b00394 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics Table 1. Relative Energies (in kJ mol−1) for Cp2ZrMe2 Precatalyst Activation and Initiation of the Ethene Polymerization Process by MAO1 and MAO2a MAO1

stabilizes the transition state for insertion with respect to the cis inner route. Nevertheless, the lowest energy transition state is obtained by the trans route. In the cis paths, β-agostic propyl products are formed first (MAO1-J cis, identical for inner and outer), but the agostic interaction ends up being abandoned due to the Zr−Me interaction (MAO1-K cis). Insertion of the Second Ethene Monomer. We studied both cis and trans routes for insertion of the second monomer for propyl products formed after insertion of the first monomer from either the cis or trans route. The energy profiles of the four combinations are given in Figure 5, optimized structures in Figure 6, and tabulated relative energies in Table 2. The propyl products formed after the first ethene insertion (MAO-K) are close in energy, the product formed from the trans route being favored by 4.8 kJ mol−1. Independent of the chain end, the πcomplex is more easily formed (MAO-K to MAO-M) via the cis route, because ethene approach from the trans direction pushes the propyl chain toward the MAO anion, which causes ion pair separation and thereby costs energy. However, the trans route becomes energetically favored upon ethene insertion (MAON*), where the cis route is now destabilized by the close contact of the MAO hindering the formation of the fourcentered transition state. In this case, metallocene−MAO separation, which would cost energy at the π-complex (MAOM), would hence assist the insertion step, bringing it closer to the energy level of the insertion transition state via the cis route. The net effect is that the two routes appear to have about equal probabilities, the rate-limiting step being ethene π-coordination for the trans route and ethene insertion for the cis route. The propyl products (MAO-O) are roughly equal in energy. They both feature close metallocene−MAO coordination and, hence, an absence of agostic interactions from the propyl chain to the metal. Note that the model does not account for chain migration, as it would require weak coordination of the counterion. However, longer alkyl chains may enforce reorientation of the relative ion position, particularly in combination with larger oligomers of MAO, which can better delocalize the negative charge. Effect of Solvent. The presence of solvent molecules weakens the Coulombic interactions in steps involving cation− anion separation, resulting in an activity increase as a function of solvent polarity.3,5,38,61−64 The solvent affects the energetics of both mechanisms described above. Focusing here on steps involving catalyst activation and incorporation of the first ethene monomer, we investigated the effects of two solvents typically used in the process, n-hexane and toluene, with dielectric constants of 1.88 and 2.37, respectively, used in the calculations. Figure 7 visualizes the energy profiles for Cp2ZrMe2 precatalyst activation and initiation of the ethene polymerization process by the trans route, by both MAOs with and without the influence of the solvents. The relative energies under vacuum (ΔE), in n-hexane (ΔEhex), and in toluene (ΔEtol) are given in Table 3. Regarding the equilibrium of MAO1 and MAO2 via TMA association/dissociation, the solvent interactions move the equilibrium toward MAO1, the difference in energy decreasing from 122.8 kJ mol−1 under vacuum to 117.4 kJ mol−1 in nhexane and 115.6 kJ mol−1 in toluene. Hence, solvents are generally expected to move MAO compositions toward structures with lower amounts of associated TMA, i.e., a higher concentration of Lewis-acidic sites. The effect increases with increasing dielectric constant of the solvent.

MAO2

intermediate

ΔEcis inner

ΔEcis outer

ΔEtrans

ΔEtrans

A B C* D E* F G* H I* J K

122.8 −16.6 N/A N/A N/A N/A −7.3 −7.6 70.1 −97.2 −114.1

122.8 −16.6 N/A N/A N/A N/A 7.0 6.7b/15.8c 61.3 −97.2 −114.1

122.8 −16.6 N/A N/A N/A N/A 45.3 36.6 55.7

0.0 −54.0 −23.3 −28.2 33.6 7.3 81.5 36.6 55.7

d

d

−118.9

−118.9

a

The energies are given relative to the neutral, noninteracting Cp2ZrMe2 and MAO2. The intermediates are visualized in Figures 3 and 4. bVertical (v) ethene π-complex. cTilted (t) ethene π-complex. d Not formed.

MAO and results in the formation of the heterobinuclear [Cp2Zr-(μ-Me)2-AlMe2]+ cation (MAO2-F), which is experimentally observed at high Al/Zr ratios that are typical for active catalyst systems.53−57 Cation−anion separation takes place at this point. Next, in the rate-limiting step, an incoming monomer displaces TMA (MAO2-G*) and yields an ethene πcomplex (MAO2-H). Note that MAO2-G* has a higher barrier than MAO1-G*, thus favoring the Lewis-acidic activation mechanism. However, the TMA association/dissociation equilibrium, as described above, shows that the TMAassociated MAO is a dominant species, which suggests preference for the [AlMe2]+ abstraction route. The π-complex is now precisely the same species as MAO1-H trans discussed above. The reaction path of MAO2 thus merges with the trans reaction path of MAO1 at the point of formation of an ethene π-complex. The subsequent ethene insertion yields the same propyl product without agostic interactions and proceeds with a lower barrier than the step of TMA displacement from the heterobinuclear complex. Olefin approach from the cis direction, i.e., from the side blocked by the MAO, is sterically blocked for MAO2. However, the situation is spatially less demanding for MAO1, giving rise to alternative routes for olefin insertion: cis inner and cis outer (Figure 2b). In comparison to the above-described trans route, the cis inner route leads to a 44.2 kJ mol−1 more stable ethene π-complex (MAO1-H cis inner), stabilized by the persistence of a direct metallocene−MAO interaction via the abstracted methyl group. Alternatively, the direct metallocene−MAO interaction is abandoned, leading to stronger metal−olefin bond. This isomer (MAO1-H cis outer vertical) lies 14.3 kJ mol−1 higher in energy. The subsequent insertion step proceeds with a higher barrier from the cis (MAO1-I* cis inner and MAO1-I* cis outer) than from the trans intermediate species (MAO1-I* trans), which is in agreement with previous reports and is independent of the counterion.9,58−60 The transition state of the cis inner route is destabilized by steric repulsion between the inserting ethene monomer and the abstracted methyl group. The cis outer route, on the other hand, requires reorientation of the coordinated ethene to maintain ion pair separation during insertion (MAO1-H cis outer tilted). Although the reorientation destabilizes the π-complex, it D

DOI: 10.1021/acs.organomet.5b00394 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

Figure 3. Intermediates for activation and initiation of the ethene polymerization process by MAO1.

Solvent has a major influence at the transition states from where the cation−anion separation begins (MAO1-G* and MAO2-E*). The weakened Coulombic interactions make cation−anion separation more feasible, thus facilitating the entire process. For MAO1, the cation−anion separation starts from the transition state that leads to the formation of an ethene π-complex. For MAO2, the cation−anion separation

starts from the transition state that leads to the formation of a heterobinuclear complex. Overall, both solvents have a systematic effect in lowering the barriers compared to the gas phase, in accordance with previous findings,9 not affecting the relative barrier heights. The rate-limiting transition states are stabilized by ca. 20 and 30 kJ mol−1 for n-hexane and toluene, respectively. E

DOI: 10.1021/acs.organomet.5b00394 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

Cp2ZrCl(Me) + MAO−Cl alters the electronic environment of the active sites, consequently affecting the catalyst behavior, as the chlorinated MAO can participate in the subsequent catalyst activation reaction. 65 To explore the effect of MAO chlorination on the activation mechanisms, single methyl groups of MAO1 and MAO2 were substituted by chlorides in sites labeled in Figure 8. Due to symmetry, there are only four substitution sites in MAO1. Association of TMA breaks the symmetry in MAO2, resulting in 12 overall sites. The effect of chlorination was screened by calculation of relative methyl affinities of the chloride-substituted MAO1 structures and relative anionization potentials of the chloridesubstituted MAO2 structures. The methyl affinities represent a simplification of the Lewis-acidic catalyst activation and are relative energies for the reaction MAO + [CH3]− → [MAO− CH3]−.35 In a similar way, relative anionization potentials represent a simplification of the activation via [AlMe2]+ cleavage and are relative energies for the reaction Me2Al− MAO → [AlMe2]+ + [MAO]−.10 Note that due to TMA association/dissociation equilibrium, both reactions result in the formation of an equivalent MAO anion. The calculated methyl affinities of the chloride-substituted MAO1 structures and the anionization potentials of the chloride-substituted MAO2 structures are given in Table 4. Overall, chlorination stabilizes the MAO anion. The only exceptions are sites P2A-f, P2B-f, and P3A-f in MAO2, where the substituent interacts either directly or indirectly through the lone electron pair with the Al of the leaving AlMe2+ moiety. The strength of stabilization is the strongest in site P1 of MAO1, which is directly bound to the active three-coordinate Al. We further examined the effect of chlorination by repeating the study of Cp2ZrMe2 precatalyst activation and initiation of the ethene polymerization process by the trans route for the MAOs. Methyl to chloride substitutions were considered taking place at site P3 of MAO1 and P3B-b of MAO2, which do not directly interact with the active sites. The energy profiles are given in Figure 9 with the unmodified MAO1 and MAO2 included for comparison. The relative total energies (ΔECl) along the reactions pathways are tabulated in Table 3. The main message of the above-described study of methyl affinities and anionization potentials is directly transferred into the reaction energy profiles: MAO chlorination leads to overall facilitation of both catalyst activation processes. Besides signaling an influential factor for the overall performance of the real MAO, the result suggests a potential pathway for improving the catalyst performance through chemical modification of the cocatalyst structure. Effect of Catalyst Structure. As a step toward understanding the role of the metallocene precatalyst itself, we studied the effect of (1) ligand bridge, (2) central metal, and (3) ancillary ligand. Structural variations typical in real polymerization catalysts were employed: A dimethylsilyl bridge was added first, followed by changing Zr to Hf or by changing Cp2 to Ind2 (Figure 10). Energy profiles for precatalyst activation and initiation of the ethene polymerization process by the trans route were calculated for both MAOs and for each precatalyst (Figure 11 and Table 5). Effect of the Bridge. Regarding the Lewis-acidic activation route by MAO1, the effect of the Me2Si bridge is mainly seen at transition states MAO1-G* and I*, which both are lowered by ca. 5 kJ mol−1. Such an energy difference would correspond to a reaction rate constant that is ca. 8 times larger. The bridge

Figure 4. Intermediates for activation and initiation of the ethene polymerization process by MAO2. After MAO2-G*, the route merges with the trans route of MAO1 (Figure 3).

Figure 5. Reaction energy profiles for insertion of the second ethene monomer into propyl products formed after cis and trans insertions of the first ethene monomer. The energies are given relative to the neutral, noninteracting Cp2ZrMe2 and MAO2. Optimized structures of the intermediates are visualized in Figure 6.

Effect of MAO Chlorination. Alkylation of the zirconocene dichloride precursor by the reaction Cp2ZrCl2 + MAO → F

DOI: 10.1021/acs.organomet.5b00394 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

Figure 6. Intermediates for insertion of the second ethene monomer into propyl products formed after cis and trans insertions of the first ethene monomer.

Table 2. Relative Energies (in kJ mol−1) for Insertion of the Second Ethene Monomer into Propyl Products Formed after Cis and Trans Insertions of the First Ethene Monomera intermediate

ΔEcis/cis

ΔEcis/trans

ΔEtrans/trans

ΔEtrans/cis

K L* M N* O

−114.1 −109.0 −109.1 2.3 −227.1

−114.1 −51.1 −69.9 −64.7 −232.7

−118.9 −53.6 −72.2 −59.9 −232.7

−118.9 −97.1 −98.7 −30.4 −227.1

roughly equal, as they were for the corresponding methyl form (MAO1-B). This further supports the effect of the bridge being more of steric than electronic origin, and as such, its absence would more strongly complicate incorporation of larger monomers than ethene. The overall beneficial effect of the bridge for catalyst activity is also seen in experiments.28,66 Effect of the Metal. Changing the central metal from Zr to Hf complicates the Lewis-acidic abstraction of the methyl group from the precatalyst (MAO1-B), due to the stronger metal− ligand σ-bond.67 This directly translates into the general observation of zirconocenes being more active than their hafnocene counterparts.28,57,68−70 The relative destabilization by Hf is the strongest at the point of monomer insertion (I*), as it involves breaking of yet another metal−carbon bond. Similar destabilizations of the catalyst intermediates take place through the [AlMe2+] transfer mechanism. In addition to monomer insertion the relative destabilization by Hf is particularly strong at the rate-limiting step involving TMA dissociation from the heterobinuclear complex (MAO2-G*). Along with reduced catalytic activities obtained for hafnocenes, the high energy required for TMA dissociation possibly contributes to the experimental identification of the preceding heterobinuclear intermediates for several hafnocenes.57 Effect of the Ancillary Ligand. Regarding Lewis-acidic methyl abstraction by MAO1, the effect of changing the ancillary Cp2 ligand to Ind2 has the contrary effect to that of

a

The energies are given relative to the neutral, noninteracting Cp2ZrMe2 and MAO2. The intermediates are visualized in Figure 6.

opens up the coordination site,40 which results in slight facilitation of the cation−anion separation of the metallocene− MAO complex by the incoming monomer as well as the subsequent monomer insertion. In the route of activation via the [AlMe2+] transfer from the MAO to the precatalyst by MAO2, the initial metallocene−MAO coordination (MAO2-B) is likewise assisted by the bridge. This initial gain in energy is carried out through the entire process, and the energy difference is the largest at the rate-limiting step MAO2-G*, with additional spatial requirements due to involvement of the monomer, which displaces TMA from the complex. Upon reaching the propyl product (K), the steric effects no longer play a role and the relative stabilities of the products are G

DOI: 10.1021/acs.organomet.5b00394 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

Figure 7. Total energy profiles of Cp2ZrMe2 precatalyst activation and initiation of the ethene polymerization process by the trans route for (a) MAO1 and (b) MAO2 in a vacuum, n-hexane, and toluene. The energies are given relative to the neutral, noninteracting Cp2ZrMe2 and MAO2. The catalyst intermediates are illustrated in Figures 3 and 4.

Table 3. Relative Energies (in kJ mol−1) for Cp2ZrMe2 Precatalyst Activation and Initiation of the Ethene Polymerization Process by the Trans Route for MAO1 and MAO2 under Vacuum (ΔE), in n-Hexane Solvent (ΔEhex), in Toluene Solvent (ΔEtol), and under Vacuum upon MAO Chlorination (ΔECl)a MAO1

a b

MAO2

intermediate

ΔE

ΔEhex

ΔEtol

ΔECl

A B C* D E* F G* H I* K

122.8 −16.6 N/A N/A N/A N/A 45.3 36.6 55.7 −118.9

117.4 −25.4 N/A N/A N/A N/A 26.4 9.2 33.6 −126.5

115.6 −26.6 N/A N/A N/A N/A 20.0 −0.2 26.0 −130.0

119.0 −25.5 N/A N/A N/A N/A 30.6 19.0 41.5 −127.9

b

ΔE

ΔEhex

ΔEtol

ΔEClc

0.0 −54.0 −23.3 −28.2 33.6 7.3 81.5 36.6 55.7 −118.9

0.0 −63.2 −25.3 −34.2 10.3 −11.8 57.9 9.2 33.6 −126.5

0.0 −67.7 −27.0 −38.5 2.6 −18.8 49.6 −0.2 26.0 −130.0

0.0 −58.7 −28.7 −33.9 25.9 1.2 74.7 19.0 41.5 −127.9

The energies are given relative to the neutral, noninteracting Cp2ZrMe2 and MAO2. The catalyst intermediates are visualized in Figures 3 and 4. Substitution of methyl by chloride in site P3 of MAO1 in Figure 8. cSubstitution of methyl by chloride in site P3B-b of MAO2 in Figure 8.

Table 4. Methyl Affinities of the Chloride-Substituted MAO1 Structures and Anionization Potentials of the ChlorideSubstituted MAO2 Structuresa

Figure 8. Labeling of sites employed in methyl group substitutions. For MAO2, sublabels f and b refer to substitution sites pointing forward and backward. For MAO1, the f- and b-sites are identical due to symmetry.

position

methyl affinity

position

anionization potential

P1

−67.8

P2

−14.5

P3

−8.4

P4

−12.5

P1A P1B P2A-f P2A-b P2B-f P2B-b P3A-f P3A-b P3B-f P3B-b P4-f P4-b

−12.8 −18.8 13.0 −12.5 13.7 −12.5 7.8 −17.2 −2.2 −12.2 −13.5 −12.4

a The energies are given in kJ mol−1 and relative to the unmodified MAOs. The positions are illustrated in Figure 8.

changing Zr to Hf, as the more electron-rich indenyl ligand can better stabilize the metal cation and thereby facilitates the breaking of the metal−carbon bond and assists in the ion pair separation.40 In an apparent relation, many of the highperformance zirconocene catalysts possess large aromatic

ancillary ligands.28,71 On a contrary note, the larger ligand is sterically more demanding, as is best seen at the point of the incoming monomer fully separating the cation and the anion H

DOI: 10.1021/acs.organomet.5b00394 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

Figure 9. Total energy profiles of Cp2ZrMe2 precatalyst activation and initiation of the ethene polymerization process by the trans route for chlorinated MAOs: (a) Cl at site P3 of MAO1; (b) Cl at site P3B-b of MAO2. The energies are given relative to the neutral, noninteracting Cp2ZrMe2 and MAO2. Schematic pictures of the intermediates are given in Figure 2. The substitution sites are illustrated in Figure 8.

(MAO1-G*). The extra steric repulsion arises from the close contact between the six-membered ring of the indenyl ligand and the methyl group abstracted from the metal by the MAO. The repulsion becomes relieved upon subsequent insertion of ethene (I*), which pushes the metallocene and MAO farther away from each other, and ethene insertion thus proceeds with a lower overall barrier due to the general electronic stabilization by indenyl. In the route of catalyst activation via [AlMe2+] transfer from MAO2, the stabilizing effect of Ind2 is particularly strong at the transition state involving the full ion pair separation (MAO2-E*), which precedes the formation of the heterobinuclear complex. However, the subsequent TMA dissociation (MAO2-G*) is sterically complicated by the ligand and, as such, comparable to the route of Lewis-acidic activation (MAO1-G*), resulting in a slight overall increase in the barrier for the rate-limiting step.



CONCLUSIONS We have reported a quantum chemical study comparing the two proposed mechanisms of metallocene/MAO catalyst

Figure 10. Schematic structures of metallocene precatalysts included in the study of the effect of catalyst structure.

Figure 11. Effect of catalyst structure on activation and initiation of the ethene polymerization process by the trans route for (a) MAO1 and (b) MAO2. The energies are given relative to the neutral, noninteracting metallocenes and MAO2. I

DOI: 10.1021/acs.organomet.5b00394 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

Table 5. Relative Energies (in kJ mol−1) for Cp2ZrMe2, Me2SiCp2ZrMe2, Me2SiCp2HfMe2, and Me2SiInd2ZrMe2 Precatalyst Activation and Initiation of the Ethene Polymerization Process by the Trans Route for MAO1 and MAO2a MAO1

a

MAO2

intermediate

Cp

Me2SiCp

Me2SiCpHf

Me2SiInd

Cp

Me2SiCp

Me2SiCpHf

Me2SiInd

A B C* D E* F G* H I* K

122.8 −16.6 N/A N/A N/A N/A 45.3 36.6 55.7 −118.9

122.8 −17.0 N/A N/A N/A N/A 40.8 35.3 49.9 −116.1

122.8 −9.9 N/A N/A N/A N/A 53.2 48.5 70.3 −108.1

122.8 −29.1 N/A N/A N/A N/A 65.4 22.6 41.5 −131.3

0.0 −54.0 −23.3 −28.2 33.6 7.3 81.5 36.6 55.7 −118.9

0.0 −64.3 −27.7 −37.8 27.5 0.4 66.7 35.3 49.9 −116.1

0.0 −58.4 −26.3 −30.7 35.9 11.5 82.7 48.5 70.3 −108.1

0.0 −71.0 N/A −40.3 5.0 −4.1 68.7 22.6 41.5 −131.3

The energies are given relative to the neutral, noninteracting metallocenes and MAO2.

gaining new insight into rational control of this complex, but industrially important catalyst.

activation and continued the investigations for the subsequent ethene polymerization reactions. Cp2ZrMe2 precatalyst and two models for the MAO cocatalyst were employed: (AlOMe)8(AlMe3) = MAO1 and (AlOMe)8(AlMe3)2 = MAO2. An essential structural ingredient of the model MAOs is associated TMA, which is responsible for the cocatalytic activity of MAO. The two model cocatalysts are in equilibrium through TMA association/dissociation. MAO1, with a three-coordinate Al, activates the precatalyst by direct Me− abstraction, whereas MAO2, with a four-coordinate Al, activates via [AlMe2]+ abstraction from the MAO by the precatalyst to yield a heterobinuclear [Cp2ZrMe2-μ-AlMe2]+ complex and subsequent dissociation of TMA by the incoming monomer. Followed by a detailed description of the two mechanisms and their energetics, we report the first efforts toward analyzing the behavior of the catalytic system as functions of solvent and structural properties of both catalyst components. TMA reacts with three-coordinate Lewis-acidic sites of MAOs, generating sites active in the [AlMe2]+ abstraction mechanism. For the employed model cocatalysts the equilibrium is strongly on the side of MAO2, suggesting the [AlMe2]+ abstraction mechanism to dominate over the mechanism of Lewis-acidic Me− abstraction by MAO1. Solvent expectedly facilitates the reactions by weakening the Coulombic interactions in steps involving cation−anion separation, as shown by using toluene and n-hexane as a solvent. Moreover, solvent increases the relative concentration of Lewis-acidic sites, and the effect is more pronounced for toluene, which has a higher dielectric constant. Regarding structural modification of the catalyst components, chlorination of MAO during alkylation of the metallocene dichloride catalyst precursor leads to overall facilitation of both activation routes. The structural effects of the metallocene precatalyst were investigated by adding a dimethylsilyl bridge, together with changing the central metal from Zr to Hf and the ancillary ligand from Cp2 to Ind2. In line with experiments, the results suggest that Zr is superior to Hf in terms of catalytic activity, while both ligand modifications generally facilitate the activation processes. The effect of the bridge is mostly of steric origin, and the effect of the ancillary ligand is mostly of electronic origin. Overall, the reported study provides the means for a systematic evaluation of factors affecting the behavior of metallocene/MAO catalysts. The approach is readily extendable to a wider range of catalyst structures. Correlated with parallel experiments, this could provide a starting point for



EXPERIMENTAL SECTION



ASSOCIATED CONTENT

Computational Details. Theoretical treatment of MAOs is complicated by dispersive interactions due to pentavalent carbon atoms, introduced by association of TMA into the MAO core.33,72 The M06 series of functionals73 has been shown to provide a cost-effective alternative for correlated ab initio methods, providing results comparable to the MP2 level of theory.26,74 All intermediates for metallocene activation by the MAOs were fully optimized at the M062X73/def-TZVP75,76 level of theory using the Gaussian09 program package.77 Relativistic effective core potentials of 28 and 60 electrons were used to describe the core electrons of Zr and Hf, respectively.78 Vibrational frequencies were calculated to confirm the nature of local minima and transition states. Solvent interactions were taken into account by the polarizable continuum model approach.79

* Supporting Information S

The supplemental file contains the computed Cartesian coordinates of all of the molecules reported in this study and their electronic energies and Gibbs energies. The file may be opened as a text file to read the coordinates or opened directly by a molecular modeling program such as Mercury (version 3.3 or later, http://www.ccdc.cam.ac.uk/pages/Home.aspx) for visualization and analysis. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.organomet.5b00394.



AUTHOR INFORMATION

Corresponding Author

*E-mail: mikko.linnolahti@uef.fi. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This research was supported by the European Commission (Grant NMP4-SL-2010-246274) and by the Academy of Finland (Project 251448). The computations were made possible by use of the Finnish Grid Infrastructure resources.



REFERENCES

(1) Busico, V. Dalton Trans. 2009, 8794−8802. (2) Bochmann, M. Organometallics 2010, 29, 4711−4740. (3) Chen, E. Y.; Marks, T. J. Chem. Rev. 2000, 100, 1391−1434.

J

DOI: 10.1021/acs.organomet.5b00394 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

(40) Linnolahti, M.; Pakkanen, T. A. Macromolecules 2000, 33, 9205−9214. (41) Daugulis, O.; Brookhart, M.; White, P. S. Organometallics 2003, 22, 4699−4704. (42) Pan, L.; Liu, Y.; Zhang, K.; Bo, S.; Li, Y. Polymer 2006, 47, 1465−1472. (43) Yoon, K.; Choi, Y. S.; Noh, S. K.; Lee, D. Macromol. Symp. 2007, 260, 27−33. (44) Hofman, M.; Nomura, K. J. Mol. Catal. A: Chem. 2010, 319, 85− 91. (45) Charpentier, P. A.; Zhu, S.; Hamielec, A. E.; Brook, M. A. Polymer 1998, 39, 6501−6511. (46) Galimberti, M.; Destro, M.; Fusco, O.; Piemontesi, F.; Camurati, I. Macromolecules 1999, 32, 258−263. (47) Reddy, S. S.; Shashidhar, G.; Sivaram, S. Macromolecules 1993, 26, 1180−1182. (48) Ghiotto, F.; Pateraki, C.; Severn, J. R.; Friedrichs, N.; Bochmann, M. Dalton Trans. 2013, 42, 9040−9048. (49) Bochmann, M.; Lancaster, S. Angew. Chem., Int. Ed. Engl. 1994, 33, 1634−1637. (50) Bochmann, M.; Lancaster, S. J. J. Organomet. Chem. 1995, 497, 55−59. (51) Chu, K.-J.; Soares, J. B. P.; Penlidis, A. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 1803−1810. (52) Xu, G.; Lin, S. Macromolecules 1997, 30, 685−693. (53) Tritto, I.; Donetti, R.; Sacchi, M. C.; Locatelli, P.; Zannoni, G. Macromolecules 1997, 30, 1247−1252. (54) Bryliakov, K. P.; Talsi, E. P.; Bochmann, M. Organometallics 2004, 23, 149−152. (55) Talsi, E. P.; Bryliakov, K. P.; Semikolenova, N. V.; Zakharov, V. A.; Bochmann, M. Kinet. Catal. 2007, 48, 490−504. (56) Babushkin, D. E.; Semikolenova, N. V.; Zakharov, V. A.; Talsi, E. P. Macromol. Chem. Phys. 2000, 201, 558−567. (57) Bryliakov, K. P.; Talsi, E. P.; Voskoboynikov, A. Z.; Lancaster, S. J.; Bochmann, M. Organometallics 2008, 27, 6333−6342. (58) Lohrenz, J. C. W.; Woo, T. K.; Fan, L.; Ziegler, T. J. Organomet. Chem. 1995, 497, 91−104. (59) Margl, P.; Lohrenz, J. C. W.; Ziegler, T.; Blochl, P. E. J. Am. Chem. Soc. 1996, 118, 4434−4441. (60) Laine, A.; Coussens, B. B.; Hirvi, J. T.; Berthoud, A.; Friederichs, N.; Severn, J. R.; Linnolahti, M. Organometallics 2015, 34, 2415−2421. (61) Longo, P.; Oliva, L.; Grassi, A.; Pellecchia, C. Makromol. Chem. 1989, 190, 2357−2361. (62) Herfert, N.; Fink, G. Makromol. Chem. 1992, 193, 773−778. (63) Eisch, J. J.; Pombrik, S. I.; Zheng, G. X. Organometallics 1993, 12, 3856−3863. (64) Forlini, F.; Fan, Z. Q.; Tritto, I.; Locatelli, P.; Sacchi, M. C. Macromol. Chem. Phys. 1997, 198, 2397−2408. (65) Trefz, T. K.; Henderson, M. A.; Linnolahti, M.; Collins, S.; McIndoe, J. S. Chem. - Eur. J. 2015, 21, 2980−2991. (66) Wang, B. Coord. Chem. Rev. 2006, 250, 242−258. (67) Martinho Simoes, J. A.; Beauchamp, J. L. Chem. Rev. 1990, 90, 629−688. (68) Mallin, D. T.; Rausch, M. D.; Chien, J. C. W. Polym. Bull. 1988, 20, 421−425. (69) D’Agnillo, L.; Soares, J. B. P.; Penlidis, A. J. Polym. Sci., Part A: Polym. Chem. 1998, 36, 831−840. (70) Silveira, F.; Simplício, L. M. T.; da Rocha, Z. N.; dos Santos, J. H. Z. Appl. Catal., A 2008, 344, 98−106. (71) Spaleck, W.; Küber, F.; Winter, A.; Rohrmann, J.; Bachmann, B.; Antberg, M.; Dolle, V.; Paulus, E. F. Organometallics 1994, 13, 954− 963. (72) Willis, B. G.; Jensen, K. F. J. Phys. Chem. A 1998, 102, 2613− 2623. (73) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215−241. (74) Ehm, C.; Antinucci, G.; Budzelaar, P. H. M.; Busico, V. J. Organomet. Chem. 2014, 772−773, 161−171. (75) Schaefer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100, 5829−5835.

(4) Imhoff, D. W.; Simeral, L. S.; Sangokoya, S. A.; Peel, J. H. Organometallics 1998, 17, 1941−1945. (5) Pédeutour, J.-N.; Radhakrisknan, K.; Cramail, H.; Deffieux, A. Macromol. Rapid Commun. 2001, 22, 1095−1123. (6) Zurek, E.; Ziegler, T. Prog. Polym. Sci. 2004, 29, 107−148. (7) Malpass, D. B. In Handbook of Transition Metal Polymerization Catalysts; Hoff, R.; Mathers, R. T., Eds.; John Wiley & Sons, Inc.: NJ, 2010; pp 1−28. (8) Harlan, C. J.; Bott, S. G.; Barron, A. R. J. Am. Chem. Soc. 1995, 117, 6465−6474. (9) Zurek, E.; Ziegler, T. Faraday Discuss. 2003, 124, 93−109. (10) Ghiotto, F.; Pateraki, C.; Tanskanen, J.; Severn, J. R.; Lühmann, N.; Kusmin, A.; Stellbrink, J.; Linnolahti, M.; Bochmann, M. Organometallics 2013, 32, 3354−3362. (11) Luo, L.; Sangokoya, S. A.; Wu, X.; Diefenbach, S. P.; Kneale, B. U.S. Pat. Appl. 2009/0062492 A1. (12) Pasynkiewicz, S. Polyhedron 1990, 9, 429−453. (13) Sinn, H. Macromol. Symp. 1995, 97, 27−52. (14) Babushkin, D. E.; Semikolenova, N. V.; Panchenko, V. N.; Sobolev, A. P.; Zakharov, V. A.; Talsi, E. P. Macromol. Chem. Phys. 1997, 198, 3845−3854. (15) Zakharov, I. I.; Zakharov, V. A.; Potapov, A. G.; Zhidomirov, G. M. Macromol. Theory Simul. 1999, 8, 272−278. (16) Ystenes, M.; Eilertsen, J. L.; Liu, J.; Ott, M.; Rytter, E.; Støvneng, J. A. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 3106− 3127. (17) Bryant, P. L.; Harwell, C. R.; Mrse, A. A.; Emery, E. F.; Gan, Z.; Caldwell, T.; Reyes, A. P.; Kuhns, P.; Hoyt, D. W.; Simeral, L. S.; Hall, R. W.; Butler, L. G. J. Am. Chem. Soc. 2001, 123, 12009−12017. (18) Zakharov, I. I.; Zakharov, V. A. Macromol. Theory Simul. 2001, 10, 108−116. (19) Rytter, E.; Støvneng, J. A.; Eilertsen, J. L.; Ystenes, M. Organometallics 2001, 20, 4466−4468. (20) Zurek, E.; Woo, T. K.; Firman, T. K.; Ziegler, T. Inorg. Chem. 2001, 40, 361−370. (21) Zurek, E.; Ziegler, T. Inorg. Chem. 2001, 40, 3279−3292. (22) Luhtanen, T. N. P.; Linnolahti, M.; Pakkanen, T. A. J. Organomet. Chem. 2002, 648, 49−54. (23) Linnolahti, M.; Luhtanen, T. N. P.; Pakkanen, T. A. Chem. - Eur. J. 2004, 10, 5977−5987. (24) Linnolahti, M.; Severn, J. R.; Pakkanen, T. A. Angew. Chem., Int. Ed. 2006, 45, 3331−3334. (25) Earley, C. W. J. Mol. Struct.: THEOCHEM 2007, 805, 101−109. (26) Boudene, Z.; De Bruin, T.; Toulhoat, H.; Raybaud, P. Organometallics 2012, 31, 8312−8322. (27) Falls, Z.; Tyminska, N.; Zurek, E. Macromolecules 2014, 47, 8556−8569. (28) Kaminsky, W. Macromolecules 2012, 45, 3289−3297. (29) Babushkin, D. E.; Brintzinger, H.- H. J. Am. Chem. Soc. 2002, 124, 12869−12873. (30) Stellbrink, J.; Niu, A.; Allgaier, J.; Richter, D.; Koenig, B. W.; Hartmann, R.; Coates, G. W.; Fetters, L. J. Macromolecules 2007, 40, 4972−4981. (31) Trefz, T. K.; Henderson, M. A.; Wang, M. Y.; Collins, S.; McIndoe, J. S. Organometallics 2013, 32, 3149−3152. (32) Negureanu, L.; Hall, R. W.; Butler, L. G.; Simeral, L. A. J. Am. Chem. Soc. 2006, 128, 16816−16826. (33) Linnolahti, M.; Severn, J.; Pakkanen, T. Angew. Chem., Int. Ed. 2008, 47, 9279−9283. (34) Glaser, R.; Sun, X. J. Am. Chem. Soc. 2011, 133, 13323−13336. (35) Linnolahti, M.; Laine, A.; Pakkanen, T. A. Chem. - Eur. J. 2013, 19, 7133−7142. (36) Hirvi, J. T.; Bochmann, M.; Severn, J. R.; Linnolahti, M. ChemPhysChem 2014, 15, 2732−2742. (37) Tritto, I.; Sacchi, M. C.; Locatelli, P.; Li, S. X. Macromol. Chem. Phys. 1996, 197, 1537−1544. (38) Resconi, L.; Cavallo, L.; Fait, A.; Piemontesi, F. Chem. Rev. 2000, 100, 1253−1345. (39) Alt, H. G.; Köppl, A. Chem. Rev. 2000, 100, 1205−1221. K

DOI: 10.1021/acs.organomet.5b00394 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics (76) Eichkorn, K.; Weigend, F.; Treutler, O.; Ahlrichs, R. Theor. Chem. Acc. 1997, 97, 119−124. (77) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Veven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision A.02; Gaussian, Inc.: Wallingford, CT, 2009. Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H. Theor. Chim. Acta 1990, 77, 123−141. (78) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H. Theor. Chim. Acta 1990, 77, 123−141. (79) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105, 2999−3093.

L

DOI: 10.1021/acs.organomet.5b00394 Organometallics XXXX, XXX, XXX−XXX