Facing unexpected reactivity paths with ZrIV-pyridylamido polymerization catalysts

FULL PAPER

DOI: 10.1002/chem.201102194

Facing Unexpected Reactivity Paths with Zr^IV–Pyridylamido Polymerization

Catalysts

Lapo Luconi,^[a]Andrea Rossin,^[a]Giulia Tuci,^[a]Incoronata Tritto,^[b]Laura Boggioni,^[b]

Jerzy J. Klosin,^[c]Curt N. Theriault,^[c]and Giuliano Giambastiani*^[a]

Dedicated to Professor A. Brandi on the occasion of his 60th birthday

Abstract: This work provides original

insights to the better understanding of

the complex structure–activity relation-

ship of Zr^IV–pyridylamido-based olefin

polymerization catalysts and highlights

the importance of the metal-precursor

a five-coordinated tris-amido species

stabilized by bidentate monoanionic

{N,N^À} ligands to a six-coordinated bis-

amido–mono-amino complexes fea-

tured by tridentate dianionic {N^À,N,C^À}

ligands). Well-defined neutral Zr^IV–pyr-

idylamido complexes have been pre-

pared from Zr(Bn)₄as metal source.

Their cationic derivatives [Zr^IV-

{N^À,N,C^À}Bn]⁺[B(C₆F₅)₄]^Àhave been

ACHUTGTNRENNUG CAHTUNGTRENNUGN

successfully applied to the room-tem-

perature polymerization of 1-hexene

with productivities up to one order of

magnitude higher than those reported

for the related Hf^IVstate-of-the-art sys-

tems. Most importantly, a linear in-

choice (ZrACHTUNGTRENNUNG(NMe₂)₄vs. Zr(Bn)₄) to pre-

pare precatalysts of unambiguous iden-

tity. A temperature-controlled and re-

versible s-bond metathesis/protonolysis

reaction is found to take place on the

Zr^IV-amido complexes in the 298–383 K

temperature range, changing the metal

coordination sphere dramatically (from

crease of the polyACTHUGNTRENNU(G 1-hexene) M_nvalues

(30–250 kgmol^À1) has been observed

upon catalyst aging. According to that,

the major active species (responsible

for the increased M_npolymer values)

in the aged catalyst solution, has been

identified.

Keywords: homogeneous catalysis ·

metathesis · polymers · polymeri-

zation · pyridylamido · zirconium

Introduction

Dow Chemical Company followed by a fruitful collabora-

tion between industry and academia is based on the C₁-sym-

metric arylcyclometallated pyridylamido Hf^IVsystems,^[1]

whose optimization was made possible by the use of high-

throughput parallel screening techniques. This class of com-

pounds has been successfully applied to the production of

highly isotactic and high molecular weight polypropylenes in

high-temperature solution processes^[2]and olefin-based

block copolymers through chain-shuttling polymerization.^[3]

Due to the utility of these complexes in the production of

specialty polyolefins, many research efforts have been de-

voted to ascertain the identity of the true active species in-

volved in the polymerization process, as well as to provide

reliable mechanistic understandings of the polymerization

processes.^[2b,c]Overall, these studies have provided funda-

mental insights into the wealth of organometallic chemistry

of nitrogen-based early transition-metal complexes, high-

lighting at the same time their unique polymerization char-

acteristics.

In the past decades, “non-metallocene” or “post-metallo-

cene” complexes have attracted considerable attention in a-

olefin polymerization chemistry. Group 4 metal alkyls or

amides, in combination with a number of nitrogen-contain-

ing ligands, are the basis of non-conventional, highly effi-

cient, and selective catalyst precursors of great promise for

polyolefin production.

One of the most exciting classes of catalysts recently dis-

covered by researchers at Symyx Technologies and The

[a] Dr. L. Luconi, Dr. A. Rossin, Dr. G. Tuci, Dr. G. Giambastiani

Institute of Chemistry of OrganoMetallic Compounds

ICCOM-CNR, Via Madonna del Piano

10-50019 Sesto F.no, Florence (Italy)

E-mail: giuliano.giambastiani@iccom.cnr.it

[b] Dr. I. Tritto, Dr. L. Boggioni

Institute for Macromolecular Studies

ISMAC-CNR, Via E. Bassini

15, 20133 Milano (Italy)

In a study on related Zr^IV– and Hf^IV–pyridylamido com-

plexes, we have recently demonstrated the importance of

the metal precursor choice in the preparation of tailored

catalysts of unambiguous identity.^[4]An unquestionable

identification of the catalyst precursor along with the knowl-

edge of the true active species involved in the process is of

crucial importance for the design of improved catalytic

structures and the study of novel catalytic processes. Tetra-

[c] Dr. J. J. Klosin, Dr. C. N. Theriault

Corporate R&D

The Dow Chemical Company

1776 Building

Midland, MI 48674 (USA)

Supporting information for this article is available on the WWW

under http://dx.doi.org/10.1002/chem.201102194.

kis(dimethylamide) (MACHTNUTGRNEUNG

(NMe₂)₄, M=Zr^IV, Hf^IV) are com-

Chem. Eur. J. 2012, 18, 671 – 687

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

671

monly used metal precursors in the synthesis of organome-

tallic derivatives by transamination reactions. In addition,

Group 4 amido–metal complexes are known to undergo in-

tramolecular s-bond metathesis,^[1]providing unconventional

cyclometalated frameworks. Noteworthy, the combination of

these metal precursors with an aminopyridinate ligand bear-

ing a thienyl pendant group at the pyridine moiety has un-

veiled unexpected reactivity paths. In particular, we have

demonstrated how the dimethylamine generated upon the

cyclometalation reaction strongly coordinates to the metal

modifications of the originally activated species), a catalyst

modification occurring before the monomer addition adds

additional layers of complexity to the identification of the

active species operating in these polymerization processes.

Results and Discussion

IV

Temperature-controlled C^ArH/Zr NMe₂s-bond meta-

À

IV

Ar

À

C

thesis/Me₂N H/Zr

anyl-substituted aminopyridinate ligands: The transamina-

tion reactions between Zr(NMe₂)₄and the aminopyridine li-

protonolysis of phenyl- and 2-fur-

center as a non-innocent ligand, acting as a temperature-

Ar

À

controlled “proton shuttle” for the M C (Ar=aryl) bond

ACHTUNGTRENNUNG

protonolysis.^[4]This reactivity changes the metal coordina-

tion sphere significantly (from five-coordinated species sta-

bilized by bidentate monoanionic {N,N^À} ligands to six-coor-

dinated complexes with tridentate dianionic {N^À,N,C^À}

ones), making the precatalyst identity in solution rather elu-

sive (Scheme 1).

gands 1a and 1b proceeded smoothly already at room tem-

perature in toluene or benzene, with loss of only one dime-

thylamine molecule. In a typical procedure, the reaction

course was followed by sampling mixtures at different reac-

tion times until complete ligand consumption and analyzing

the crude product obtained after solvent removal by

¹H NMR spectroscopy. The reaction workup led to the tris-

amido complexes 2 and 3 as air- and moisture-sensitive,

dark-yellow, microcrystalline solids in more than 90% yields

(Scheme 2).

Scheme 1. Temperature-controlled intramolecular s-bond metathesis/pro-

tonolysis of Zr^IV-and/or Hf^IV–pyridylamido olefin polymerization catalyst

precursors.

Studies conducted on differently substituted aminopyri-

dine ligands in the presence of ZrACTHNUTRGNEUNG(NMe₂)₄led to systems be-

having similarly, revealing the presence of interconverting

pyridylamido tautomers. One distinguishing structural fea-

ture of these species is the formation of a temperature-in-

Scheme 2. Synthesis of the tautomeric 2/2’ and 3/3’ Zr^IVcomplexes.

Ar

duced and reversible Zr^IV

bond, which makes the com-

À

C

prehension of their activation and polymerization mecha-

nisms less evident. In the first part of this paper we report a

full account on the s-bond metathesis/protonolysis process

as observed by the combination of a series of 6-aryl-substi-

1

The H NMR spectra at room temperature (298 K) show

very similar patterns for both systems; only one broad reso-

À

nance for the Zr NMe₂methyl protons was observed (d=

tuted aminopyridine ligands with Zr^IV

N

2.85 (2), 2.86 ppm (3)), indicating a rapid equilibration of

the NMe₂ligands on the NMR time scale. The most relevant

spectral features are given by a pair of diastereotopic iso-

propyl–methyl resonances (two doublets at d=1.42 and

Aimed at highlighting the central role of the metal precur-

sor choice in the production of well-defined molecular cata-

lysts, a direct comparison between ZrACTHUNTGRNEG(UN NMe₂)₄and Zr(Bn)₄

À

has been systematically carried out. In particular, while

studying the 1-hexene polymerization upon activation of se-

lected dibenzyl catalyst precursors with strong Lewis acids,

we came across an unusual effect of the catalyst aging time

on both the polymer molecular mass (M_n) and the mass dis-

tribution (M_w/M_n) modality. An in-depth investigation of the

activated catalytic species present in solution at different

aging times (and in the absence of the monomer) has high-

lighted a progressive active species modification. If broad

molecular weight distributions are reasonably consistent

with polymers produced by multiple active sites (the latter

being typically generated by permanent monomer–ligand

1.58 ppm (2) and d=1.32 and 1.48 ppm (3)) with a single

CHMe₂septet (d=3.82 (2) and 3.73 ppm (3)) and a single

resonance for the bridging PyCH₂N moiety (d=4.92 (2) and

4.77 ppm (3)). Similarly to previously reported data for Zr^IV

and Hf^IVcomplexes stabilized with a 2-thienyl-substituted

pyridylamido ligand,^[4]the 1D variable-temperature (VT)

¹H NMR spectra ([D₈]toluene, 298–383 K) of 2 and 3 show

the coexistence of two distinct isomeric forms in solution at

high temperature, with the minor component being unam-

biguously attributed to the hexacoordinated, cyclometalated

[N₂^ArZr

(h¹-HNMe₂)] species 2’ and 3’ with a dime-

ACHUTNGTREN(NUG NMe₂)₂ACHTUTGNRENNUGN

thylamine ligand as part of the metal coordination sphere

672

www.chemeurj.org

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2012, 18, 671 – 687

Zr^IV–Pyridylamido Polymerization Catalysts

FULL PAPER

(Scheme 2). In contrast to the thienyl-containing system,^[4]

where the two equilibrating tautomers are already present

at room temperature in a molar ratio of 70:30, complexes 2

and 3 represent almost the only detectable species in solu-

tion at room temperature. The molar ratio of the cyclometa-

lated form at higher temperature depends on the different

À

ability of the aryl moiety to undergo intramolecular C H

bond activation. Accordingly, an interconversion of the iso-

meric mixtures from the initial 99:1 ratio to 58:42 (2/2’) and

38:62 (3/3’) takes place in the 298–383 K temperature range.

Notably, complete reversibility between the tautomeric

forms was maintained with no decomposition of the mix-

tures even after several thermal cycles (Figure 1). Contrary

to the behavior observed for the thienyl derivative,^[4]a pro-

longed heating (24–48 h) in toluene, which was heated to

reflux, did not eliminate the metal-coordinated NHMe₂but

a progressive catalyst decomposition took place instead.

Thermodynamic standard parameters for the tautomeric

interconversions of 2/2’ and 3/3’ were calculated from the re-

spective linear van ꢁt Hoff fits (Figure 2). The linearity of

the van’t Hoff plots indicates no appreciable heat capacity

changes for both processes over the 313–363 K temperature

range; their calculated thermodynamics at 298 K are report-

ed in Table 1.

^

Figure 2. The van ꢁt Hoff plots for the tautomeric equilibration of 2/2’ (

)

*

and 3/3’ ( ).

Table 1. Calculated thermodynamic parameters for the tautomeric equili-

bration of 2/2’ and 3/3’.

DH⁰[kcalmol^À1

(9.53Æ0.2)

(5.66Æ0.2)

]

DS⁰[calK^À1mol^À1

]

DG⁰[kcalmol^À1

]

2/2’

3/3’

N

(23.93Æ0.5)

(14.57Æ0.6)

(2.39Æ0.08)

(1.32Æ0.06)

Density functional theory calculations carried out on both

real systems at the M06/6-31G* level of theory (toluene, po-

larizable continuum model, PCM) have contributed to eluci-

date the mechanism of the observed tautomerism and to

confirm (whenever possible) the calculated thermodynamics.

Both processes present positive DE profiles associated with

the formation of the cyclometalated species (DE= +8.8 (2/

2’), +4.8 kcalmol^À1(3/3’); Figure 3), with higher energy bar-

riers (equal to 28.7 and 26.5 kcalmol^À1for TS_2/2’and TS_3/3’

,

Figure 3. Calculated internal energies (DE, toluene; top) and Gibbs free

energy (DG, gas phase) values (bottom) for the tautomeric equilibrations

2/2’, 3/3’ and the amine loss with generation of the diamido complexes 4

and 5. Please see the Supporting Information for the structures of the

corresponding optimized geometries.

1

Figure 1. Selected VT H NMR spectra (400 MHz, C₇D₈, 5ꢀd [ppm]ꢀ1)

of 2/2’ and 3/3’.

Chem. Eur. J. 2012, 18, 671 – 687

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

673

G. Giambastiani et al.

respectively) compared with the one calculated for the thio-

phene analogue (19.2 kcalmol^À1).^[4]Similarly to what was

observed in the thienyl complex,^[4]in the optimized geome-

try of 3 the furyl substituent is not coordinated to the metal

À

the Zr^IV–pyridylamido precatalysts (from five-coordinated

species (2 and 3) stabilized by bidentate monoanionic

{N,N^À} ligands, to six-coordinated complexes (2’ and 3’) with

tridentate dianionic {N^À,N,C^À} ligands) limits the complete

understanding of the structure–activity relationship. Ethyl-

ene polymerization tests conducted on the thiophene-based

tautomeric mixture (Scheme 1) by using AlMe₃-free methyl-

alumoxane (PMAO)^[8]as activator have unambiguously

demonstrated that only the ortho-metalated species in the

mixture is responsible for the generation of catalytically

active forms.^[4,9]Accordingly, ethylene polymerization tests

conducted with the catalytic systems 2/2’/PMAO and 3/3’/

PMAO between 298 and 343 K (where the bidentate mono-

anionic {N,N^À} forms 2 and 3 largely prevail over the triden-

tate dianionic {N^À,N,C^À} ones (2’ and 3’)) resulted in only

moderately active to almost inactive catalytic systems.^[10]

center (final dACHTUNGTRENNUNG(Zr O)=3.21 ꢂ, Figure S37), the heterocyclic

ring being tilted away with respect to the pyridine central

core (q(N_Py-C-C-O_Fu)=268, Py=pyridine, Fu=furanyl; Fig-

ure S37). The computational results match reasonably well

with the experimental DH⁰values calculated from the

van ’t Hoff plots (9.53 and 5.66 kcalmol^À1for the phenyl and

furanyl system, respectively). At odds with the thiophene-

containing complex,^[4]the internal energy levels of the {4+

NHMe₂} or {5+NHMe₂} system obtained after dimethyl-

ACHTUNGTRENNUNGamine loss (and concomitant formation of the corresponding

diamido derivative) lie below the energy level of TS_2/2’or

TS_3/3’(see Figure 3, top). Such a result does not mirror the

experimental findings. Different entropic factors in the gas

phase and in solution are expected (S^gasvs. S^sol), mainly be-

cause of solvation phenomena that cannot be taken into ac-

count properly in the computation. According to this, the

Gibbs energy (G) rather than the internal energy (E) differ-

ences should be evaluated for this step. In fact, in the real

sample the solvent molecules surrounding the metal com-

plex severely limit the translational/rotational freedom of

molecular fragments. This ultimately translates into signifi-

cant DG^gas/DG^solvdifferences. The DG^gasvalues related to

the formation of the {diamido complex+NHMe₂} ensemble

reveal that the amine loss is only slightly endothermic

Synthesis of the pyridylamido Zr^IV(Bn)₂complexes (6–9)

from Zr(Bn)₄as alternative metal precursor: Aiming at the

development of tailored catalytic structures active as poly-

merization catalysts, we have investigated the reactivity of

the aminopyridine ligands 1a–1d with Zr(Bn)₄as an alterna-

tive Zr^IVmetal precursor (Scheme 3).

(DG^gas(2’/4+NHMe₂)= +1.5,

DG^gas(3’/5+NHMe₂)= +

4.0 kcalmol^À1; Figure 3, bottom). Smaller DG^gasthan DE^gas

values are in line with the positive entropy variation during

the amine loss. Unfortunately, the evaluation of DG^solvis not

trivial; no general procedure is given to assign a precise

value to DG^solvwithin the PCM approach.^[5]

Scheme 3. Synthesis of the Zr^IV(Bn)₂complexes 6–9 from dianionic tri-

dentate {N^À,N,C^À} pyridylamido ligands.

These results constitute a useful reference to gain addi-

À

tional insights into C H bond functionalization reactions of

early-transition-metal complexes, highlighting at the same

time unexpected reactivity paths in the field of polymeri-

zation catalysis. Indeed, the well-established efficiency of

Group IV pyridylamido catalysts (in high-temperature solu-

tion processes, in particular) for the production of specialty

polyolefins^[1,3,6,7]invariably relies on distinctive structural

features of these catalyst precursors. In recent papers, Mac-

chioni et al. have developed a detailed understanding of the

polymerization mechanisms with a series of arylcyclometal-

lated Hf^IV–pyridylamido complexes.^[2b,c]These studies pro-

The addition of Zr(Bn)₄to a solution of the ligand in tolu-

ene at room temperature did not generate any product,

even after several hours under stirring. This was confirmed

1

by the H NMR spectra of the reaction mixture recorded at

different reaction times, which showed distinctive signals of

the unreacted reagents. On the other hand, progressive

ligand consumption took place upon heating of the mixture

1

at 708C for 12–48 h. The H NMR spectroscopic analysis re-

vealed the direct formation of a Zr^IV-dibenzyl cyclometalat-

ed compound as an unique species with no evidence of puta-

tive tribenzyl intermediate generation. Such a result indi-

vide clear evidence of the fundamental role played by the

unusual Hf C bonds, emphasizing the importance of the

Ar

À

cates that at higher temperature N H deprotonation of the

true identification of the active species for the study of new

catalytic processes or the design of improved catalytic struc-

tures. On this ground, a reliable identification of the catalyst

precursor is also of crucial importance; in this context, the

reactivity outlined above (temperature-induced tautomeric

interconversions) is expected to add further layers of com-

plexity to the understanding of both the activation and the

polymerization mechanisms. Indeed, the temperature-in-

duced change occurring at the metal coordination sphere of

ligand and N,N chelation of the metal center facilitate the

immediate and irreversible ortho-metalation step, with con-

comitant elimination of a second toluene molecule. The abil-

À

ity of the ligand to undergo intramolecular C H bond acti-

vation provides clear evidence of the importance of the

metal precursor choice for the preparation of well-defined

catalysts.

Complexes 6–9, isolated as air- and moisture-sensitive,

yellow-brown solids and less light sensitive than the parent

674

www.chemeurj.org

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2012, 18, 671 – 687

Zr^IV–Pyridylamido Polymerization Catalysts

FULL PAPER

tetrabenzyl complexes, were completely characterized by

spectroscopic data and X-ray diffraction analyses. The most

1

relevant H NMR data for complexes containing isopropyl

moieties (6–8) are two sets of methyl groups, which appear

as two well-separated doublets^[11]centered at d=1.17/1.32

(6), 1.05/1.22 (7), and 1.15/1.32 ppm (8), whereas a singlet at

d=2.35 ppm is observed for the resonance of the methyl

groups of the m-xylyl fragment in 9. The methylene protons

of the residual benzyl groups bound to zirconium show an

AB pattern^[12]centered at d=1.91 (6, ²J

1.53 (7, ²J(H,H)=9.6 Hz), 1.74 (8, ²J

(H,H)=9.6 Hz), and

1.78 ppm (9, J

(H,H)=9.8 Hz). In the ¹³C{¹H} NMR spectra,

ACHTUNGTREN(NGNU H,H)=10.0 Hz),

A

ACHTUNGTRENNUNG

2

ACHTUNGTRENNUNG

only one carbon resonance around d=68 ppm (6–9) is pres-

ent for the two methylene groups directly bound to the Zr

atom, consistent with two isochronous benzyl groups in solu-

tion; ¹H and ¹³C{¹H} NMR patterns indicate that all com-

plexes possess C_ssymmetry in solution. Moreover, in the

¹³C NMR spectra, the methylene carbons attached to the

Figure 5. Crystal structure of N₂^FuZr

ACTHNUGTRNEUGN(CH₂Ph)₂(7). Thermal ellipsoids are

drawn at the 40% probability level. Hydrogen atoms are omitted for

clarity.

metal ion have a JACTHNUTRGNE(NUG C,H) value of 129–131 Hz (for 6–9),

which is typical for a weighted average value between an h¹-

(120 Hz) and h²-coordination mode (143 Hz).^[13,14]These

data confirm the existence of a rapid exchange in solution^[15]

between h¹- and h²-coordinated benzyl groups.^[16]The ¹H

and ¹³C resonances for the CH₂groups related to the

pyridylACHTUNGTRENNUNGamido framework appear at lower field when com-

pared to those of the corresponding free ligands, thus con-

firming ligand coordination to the zirconium center. Finally,

the Zr C^Arbonds generated upon intramolecular cyclometa-

À

lation are evidenced by a remarkable resonance shift to

lower fields of the aromatic carbon directly bound to the

metal center (189.8, 168.0, 195.4, and 189.7 ppm for 6–9).

All complexes are highly soluble in aromatic and aliphatic

hydrocarbons, whereas they are only sparingly soluble in di-

ethyl ether. Crystals suitable for X-ray analysis were isolated

from concentrated solutions of toluene (6 and 9) or diethyl

Figure 6. Crystal structure of N₂^ThZr

ACTHNUGTRENNUG(CH₂Ph)₂(8, Th=thienyl). Thermal

ellipsoids are drawn at the 40% probability level. Hydrogen atoms and

diethyl ether (crystallization solvent) are omitted for clarity.

ether (7 and 8). ORTEP diagrams of the crystal structures

of 6–9 are given in Figure 4–7. Table 2 lists all the main crys-

tal and structural refinement data, whereas selected bond

lengths and angles are summarized in Table S1 (see the Sup-

porting Information).

The coordination environment at the metal center consists

of two nitrogen atoms and one carbon atom from the di-

AHCTUNGTRENNUNG

anionic tridentate {N^À,N,C^À} pyridylamido ligand (k³-coordi-

nation) and two benzyl groups. All complexes show a (dis-

torted) trigonal bipyramidal coordination geometry, where

the three donor atoms from each ligand occupy the axial po-

sitions and the two residual benzyl fragments are placed at

Figure 4. Crystal structure of N₂^PhZr

ACTHNUGTRNEUGN(CH₂Ph)₂(6). Thermal ellipsoids are

drawn at the 40% probability level. Hydrogen atoms are omitted for

clarity.

Chem. Eur. J. 2012, 18, 671 – 687

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

675

G. Giambastiani et al.

observed for related structures.^[16,17]As clearly shown by the

X-ray study, in the solid state one benzyl group is h²-coordi-

À

nated (6: Zr···C(26) 2.634(4), Zr C(25) 2.271(3) ꢂ, Zr-

À

C(25)-C(26), 86.9(2)8; 7: Zr···C(24) 2.583(4), Zr C(23)

2.270(4) ꢂ, Zr-C(23)-C(24) 85.0(2)8; 8: Zr···C(24) 2.599(3),

À

Zr C(23) 2.268(3) ꢂ, Zr-C(23)-C(24) 85.70(18)8; 9:

À

Zr···C(22) 2.687(3), Zr C(21) 2.289(3) ꢂ, Zr-C(21)-C(22)

88.59(18)8), whereas the other approaches h¹-coordination

À

to the metal center (6: Zr C(32) 2.280(3), Zr C(33)

À

3.006(4) ꢂ, Zr-C(32)-C(33) 104.4(2)8; 7: Zr C(30) 2.279(4),

À

Zr C(31) 2.878(3) ꢂ, Zr-C(30)-C(31) 98.0(3)8; 8: Zr C(30)

2.284(3), Zr C(31) 2.977(2) ꢂ, Zr-C(30)-C(31) 102.40(19)8;

9: Zr C(28) 2.272(3), Zr C(29) 2.906(4) ꢂ, Zr-C(28)-C(29)

À

99.4(2)8).^[18]

Figure 7. Crystal structure of N_2(xylyl)^PhZr

ACTHNUGTRENNUG(CH₂Ph)₂(9). Thermal ellipsoids

are drawn at the 40% probability level. Hydrogen atoms and toluene

(crystallization solvent) are omitted for clarity.

In situ generation and characterization of cationic monoben-

zyl pyridylamido Zr^IV(Bn) complexes upon activation by

the equatorial positions. The k³-pyridylamido fragments are

almost planar, with an average deviation of the metal center

from the {N^À,N,C^À} plane of 0.289 ꢂ (6: 0.360(3), 7:

0.243(4), 8: 0.257(3), 9: 0.296(3) ꢂ). The final coordination

number at the metal is six, due to a close contact between

either B

complexes 6a–9a and 6b–9b were obtained by treatment of

the dibenzyl precursors 6–9 with B(C₆F₅)₃or [Ph₃C][B-

(C₆F₅)₄], respectively. The reaction of 6–9 with one equiva-

lent of B(C₆F₅)₃in [D₆]benzene at room temperature result-

ACHTUNGTRENU(NG C₆F₅)₃or [Ph₃C][BHCAUTNGTREN(NUGN C₆F₅)₄]: The cationic zirconium

AHCTUNGTRENNUNG

the zirconium atom and the ipso-carbon atom of one benzyl

ed in the clean abstraction of one benzyl group from each

complex,^[19]with the rapid formation of the red-orange zwit-

terionic species 6a–9a (Scheme 4).

All of the zwitterionic compounds were slightly soluble in

benzene or toluene and separated as dark-red semisolid

products upon standing within a few hours. The limited solu-

2

À

group (h -coordination). For all complexes, the Zr N(1)

À

(2.315(2)–2.329(3) ꢂ, 6–9) and Zr N(2) (2.090(2)–

2.103(2) ꢂ, 6–9) distances are comparable to those mea-

sured for similar Zr^IV–pyridylamido systems^[6]and the Zr

À

C^Arbonds (2.313(3)–2.325(4) ꢂ, 6–9) also fall in the range

Table 2. Crystal data and structure refinement for complexes 6–9.

6

7

8·0.25Et₂O

9·0.5toluene

formula

molecular weight

T [K]

C₃₈H₄₀N₂Zr

614.22

100(2)

0.71069

monoclinic

P2₁/n

C₃₆H₃₈N₂OZr

605.90

120(2)

0.71069

triclinic

¯

P1

C₃₆H₃₈N₂SZr·¹=

640.49

120(2)

0.71069

tetragonal

P4/n

26.133(9)

9.573(5)

90

6538(7)

8

1.301

N

C₃₄H₃₂N₂Zr·¹=

605.90

120(2)

0.71069

Monoclinic

C2/c

38.208(3)

7.982(4)

19.880(1)

90

101.145(6)

90

5949(5)

8

1.353

0.398

₂ACHTUNGTRENNUNG(C₇H₈)

l [ꢂ]

crystal system

space group

a [ꢂ]

b [ꢂ]

c [ꢂ]

9.522(7)

9.842(3)

21.178(14)

16.076(17)

90

106.769(10)

90

3104(4)

4

1.318

0.383

12.628(6)

24.285(11)

89.928(4)

87.846(3)

83.873(3)

2999(2)

a [8]

b [8]

g [8]

V [ꢂ³]

Z

4

1_calcd[gm^À3

]

1.342

0.397

1264

m [mm^À1

(000)

]

0.428

2676

F

A

1288

2520

crystal size [mm]

V range [8]

limiting indices

0.01ꢃ0.05ꢃ0.05

3.85–26.46

À11ꢁhꢁ10

À22ꢁkꢁ26

À18ꢁlꢁ15

14873/4629

0.979

0.02ꢃ0.02ꢃ0.05

4.14–25.44

À11ꢁhꢁ11

À14ꢁkꢁ14

À29ꢁlꢁ27

20708/8633

0.836

0.05ꢃ0.10ꢃ0.20

4.26–26.47

0.05ꢃ0.05ꢃ0.1

4.16–27.68

À27ꢁhꢁ29

À30ꢁkꢁ30

À11ꢁlꢁ11

49810/6084

0.809

À48ꢁhꢁ49

À10ꢁkꢁ9

À24ꢁlꢁ25

31602/6173

0.926

reflns collected/unique

GOF on F²

data/restraints/parameters

final R indices [I>2s(I)]

4629/0/370

R1=0.0382

wR2=0.0822

R1=0.0626

wR2=0.0884

0.391/À0.392

5684/0/410

R1=0.0442

wR2=0.0766

R1=0.0932

wR2=0.0839

0.519/À0.409

6084/0/382

6173/0/361

R1=0.0404

wR2=0.0976

R1=0.0708

wR2=0.1032

1.189/À0.466

R1=0.0379

wR2=0.0670

R1=0.1060

wR2=0.0747

0.629/À0.460

R indices (all data)

largest diff. peak/hole [eꢂ^À3

]

676

www.chemeurj.org

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2012, 18, 671 – 687

Zr^IV–Pyridylamido Polymerization Catalysts

FULL PAPER

5.04 ppm,

assigned

to

ZrCH₂Ph and to PyCH₂N, re-

spectively. Although poorly

soluble in aromatic hydrocar-

bons such as benzene or tolu-

ene, the activated complex is

stable in these solvents for

days with no apparent decom-

position. On the other hand,

6b is highly soluble in polar ar-

omatics such as mono- or diha-

Scheme 4. Synthesis of the cationic Zr^IV(Bn) complexes 6a,b–9a,b from activation of the dibenzyl precursors

with either BACHTUNGTRENNUNG(C₆F₅)₃or [Ph₃C][BAHCTUNGTREN(NUGN C₆F₅)₄].

lobenzene

(halo=Cl,

Br)

where it forms a deep-orange

solution. Unfortunately, polar solvents cause the catalyst de-

composition within a few minutes resulting in the separation

of rubbery/untreatable dark-brown semisolid materials. The

low solubility of the activated complex 6b in [D₆]benzene or

[D₈]toluene, together with its rapid decomposition in polar

aromatics, did not allow the complete spectroscopic charac-

terization of this species. The cationic complexes7b–9b were

prepared similarly by treating compounds 7–9 in

bility of the ion pairs 6a–9a in aromatic solvents is sugges-

tive of weak ion pairing.^[20]The ¹H NMR spectrum of

[N₂^PhZr(CH₂Ph)]⁺[B (C₆F₅)₃]^À(6a) shows the pres-

ACHTUNGTRENNUNG ACHTUNRGTEG(NNUN CH₂Ph)ACHTGUNTRENNUGN

ence of four non-equivalent Me groups from the two iPr

fragments; such a reduced symmetry of the activated com-

plex is due to a h⁶-coordination of the benzyl-tris(penta-

fluorophenyl)borate anion to the metal center. Noteworthy,

two sets of diastereotopic methylene protons (AB systems)

from the residual ZrCH₂Ph and PyCH₂N fragments ap-

peared at d=2.55 and 4.44 ppm, respectively. Finally, the ex-

istence of non-equivalent proton signals of the methylene

group at the BCH₂Ph moiety (two broad signals at d=3.54

and 3.09 ppm) suggests the presence of a close contact be-

tween the borate anion and the metal center. As a matter of

fact, the ¹⁹F NMR spectrum of complex 6a shows a large

[D₆]benzene with one equivalent of [Ph₃C][BACHTUNGRTNEUNG(C₆F₅)₄]. The

¹H NMR spectra (in [D₆]benzene) of the supernatants show

signals almost exclusively attributed to the benzyltriphenyl-

methane derivative, whereas those recorded on the solid res-

1

idues (either in C₆D₅Br or in C₆D₅Cl) show similar H NMR

patterns to those recorded for 6b (dCAHTUNGTERN(NUNG ZrCH₂Ph)=2.73 (7b),

2.76 (8b), 2.72 ppm (9b); d(PyCH₂N)=4.90m (7b), 4.98

AHCTUNGTRENNUNG

Dd

G

E

(8b), 2.69 ppm (9b)).

a borate anions tightly bound to the metal center.^[21]In the

¹³C NMR spectrum of complex 6a the signal of the methyl-

ene carbon atom of ZrCH₂Ph appears at d=79.76 ppm with

1-Hexene polymerization by activated Zr^IV–pyridylamido

complexes: After the synthesis and characterization of the

activated forms 6a–9a and 6b–9b, we studied their polymer-

ization activity in the presence of 1-hexene as the monomer

(Table 3). Catalysts derived from BACHTNUTRGNE(UNG C₆F₅)₃activation (6a–

9a) showed very low catalytic activity, providing only traces

a JACHTUNGTRENNUNG

(C,H) value of 131 Hz, indicative of an h²-coordination of

the residual benzyl fragment in the cationic species. Finally,

the ¹³C{¹H} NMR signal for the methylene carbon atom of

BCH₂Ph appears, as expected, as a broad multiplet centered

at d=35.49 ppm. Similarly to the procedure reported above,

the zwitterionic complexes 7a–9a were prepared by treating

of the polymer products. This very low activity is likely due

to the strong h⁶-binding of [B (C₆F₅)₃]^Àto the cat-

ACHTUNGRTENN(UG CH₂C₆H₅)ACHUTNGTRENNGUN

complexes 7–9 with one equivalent of B

ionic Zr^IVcenter, which hampers monomer access to the

metal center.^[1b,22]

tion chemistry of 7–9 is similar to that described for 6, lead-

ing to the highly unsymmetrical species 7a–9a, which are

characterized by four non-equivalent Me groups (from the

two iPr fragments) and by well-defined AB systems for all

residual methylene moieties (d

(8a), 2.44 ppm (9a); d(PyCH₂N)=4.37 (7a), 4.39 (8a),

4.02 ppm (9a); d(BCH₂Ph)=3.51 (7a), 3.59 (8a), 3.35 ppm

(9a); see the Supporting Information).

The cationic monobenzyl complex 6b was prepared by

treating a solution of the dibenzyl precursor 6 in benzene

Table 3. 1-Hexene polymerization by activated pyridylamido complexes

6b–9b.^[a]

Catalyst

t

T

Yield

[g]

Conv

Activity

Activity^[e]

[d]

A

[8C]

[%]^[b,c]

[10^À3

ACHTUNGTERNNUNG ]

1

2

3

4

5

6

7

6b

7b

8b

9b

5

2

5

1

30

5

23

1.3

1.1

99.9

81.4

1.8

46.2

14.7

9.2

19.18

39.07

0.35

44.35

0.47

17.70

36.05

0.32

40.92

0.43

0^[f]

50^[f]

23

n.d.^[g]

0.6

with one equivalent of [Ph₃C][BACHTNUTRGNEUNG(C₆F₅)₄] (Scheme 4). The re-

0.2

0.1

1.2

23

0.29

17.34

0.27

16.00

action occurs rapidly and quantitatively at room tempera-

ture producing dark-green crystals, which precipitate within

a few minutes. Due to the weaker coordinating character of

90.3

[a] Reaction conditions: catalyst precursor (10 mmol), [Ph₃C][BACHTUNGTRENNUNG(C₆F₅)₄]

(10 mmol), 1-hexene (2 mL, 16 mmol), toluene (8 mL); [b] determined by

GC; [c] average values calculated over three independent runs; [d] mol

of converted 1-hexene (mol of Zr)^À1h^À1; [e] kg of 1-hexene converted (g

of Zr)^À1h^À1; [f] precatalyst activation was run at room temperature for

5 min prior to setting the final reaction temperature; [g] not determined.

BACHTUNGTRENNUNG(C₆F₅)₄, the H NMR spectrum of 6b (in C₆D₅Br) shows

the formation of a C_s-symmetric system, which is character-

ized by only two non-equivalent Me groups from the iPr

fragments and by two well-defined singlets at d=2.84 and

Chem. Eur. J. 2012, 18, 671 – 687

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

677

G. Giambastiani et al.

On the other hand, catalysts prepared by using [Ph₃C][B-

ACHTUNGTRENNUNG

ACHTUNGTRENNUNG(1-hexene) with complete monomer conversion in a few mi-

nutes and productivities that are over one order of magni-

tude higher than those previously reported in the literature

for related pyridylamido-Hf^IVsystems.^[1b]Although higher

polymerization temperatures (Table 3, entry 4) were not

found to affect the catalyst activity significantly, runs con-

ducted at 08C revealed almost complete deactivation of the

catalytic processes. All polyACTHNUGRTNEUNG(1-hexene)s produced by 6b

showed moderate isotacticities (m⁴=0.47) with approxi-

mately 5–6 mol% of regioerrors as evidenced by their

¹³C{¹H} NMR spectra (see the Supporting Information). Al-

though the observed isotacticity can be rationalized on the

basis of an in-situ ligand modification resulting from the mo-

nomer insertion into the Zr^IVAr bond (that translates into

À

a lowering of the symmetry of the active species),^[1b,2a–e]

there is no apparent reason to explain the dramatic activity

decrease observed in the case of the heteroaryl-containing

systems 7b and 8b.^[2a,c]Experimental evidence for the olefin

insertion into the Zr^IVAr bond was gathered by treating

À

the two representative activated species 6b and 7b with a

stoichiometric amount of 1-hexene at room temperature.

The GC-MS analysis of the hydrolyzed reaction mixtures re-

vealed the presence of two main peaks (m/z=344 and 428

for 6b/1-hexene; m/z=334 and 418 for 7b/1-hexene) in a

78:22 and 39:61 ratio, attributed to the unmodified ligands

and the hexyl-substituted ligands 10^[23]and 11, respectively

(Figure 8).

Figure 8. GC traces of the reaction solutions obtained from the reaction

of 6b and 7b with 1 equivalent of 1-hexene after quenching with MeOH

(GC program: 408C for 1 min, 158Cmin^À1, 2508C for 20 min).

To gain additional information on the olefin insertion

path, the GC-MS analysis was ultimately conducted on the

solutions obtained from the sub-stoichiometric runs (6 equiv

of 1-hexenes) with 6b and 7b. The presence of a monomer

excess did not change the ligand/modified-ligand ratio sig-

nificantly (71:29 for 6b and 28:72 for 7b); no detectable

amount of differently modified ligands (resulting from the

possible chain-growth continuation between the metal and

the aryl fragment) was found (see the Supporting Informa-

catalyst 6b, aiming at a better understanding of its polymeri-

zation ability. Despite moderate solubility in aromatic sol-

vents (benzene or toluene), compound 6b maintains its cata-

lytic activity virtually unchanged even after prolonged aging

times (Table 4, entries 1–4).^[1b]All poly

ACTHNUTRGNEUNG(1-hexene)s produced

by 6b were isotactically enriched, as evidenced by the

tion).^[2a]If the idea that the monomer insertion into the Zr

¹³C{¹H} NMR spectra (see the Supporting Information),

À

Ar bond of 6b generates either inactive or slow

polymerization species can be reasonably discarded

Table 4. Poly

(1-hexene)s from catalysts 6b and 9b at different catalyst aging times.^[a]

ACHTUNGTRENNUNG

on the basis of its high catalytic performance (6b

shows productivities up to 41 kg of poly(1-hexene)

Cata- t_age

t

A

Yield Conv

[%]^[c,d]

M_nvis^[e]h^[f]

[10^À3[dLg^À1

M_nrel^[g]M_w/M_n

[10^À3

AHCTUNGTRENNUNG

lyst

[h]

[min]^[b][g]

R

]

G

]

N

]

per g of Zr per h), a deactivation path associated

1

2

3

4

5

6

6b

9b

0

1

5

2

5

30

5

1.1

1.3

1.1

0.5

1.23

0.24

81.4

97.8

82.7

37.6

90.3

17.7

31

60

86

258

127

107

0.76

1.04

1.18

0.98

1.33

1.43

45

82

129

265

137

127

4.9

5.2

À

to the Zr Ar monomer insertion cannot be defini-

tively ruled out for 7b and 8b. On the contrary, it

is evident that minor differences in the ligand

frameworks (1a,d vs. 1b,c) would influence the

final catalysts performance dramatically.

4.3

16

0

2.5^[h]

3.4

5

30

3.6

[a] Reaction conditions: catalyst precursor (10 mmol), [Ph₃C][BACHTUNGRTNEUNG(C₆F₅)₄] (10 mmol), 1-

hexene (2 mL, 16 mmol), toluene (8 mL), 238C; [b] polymerization time; [c] deter-

mined by GC analysis; [d] average values calculated over three independent runs;

[e] determined from viscosimetric elaboration; [f] intrinsic viscosity; [g] relative to

polystyrene calibration; [h] calculated on the main gel-permeation chromatography

(GPC) curve corresponding to an M_nvalue of 265 kDa.

PolyACHTUNGTRENNUNG(1-hexene)s characterization and study on the

effect associated to catalysts aging: The effect of

the catalyst aging in solution before the monomer

introduction was investigated for the most efficient

678

www.chemeurj.org

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2012, 18, 671 – 687

Zr^IV–Pyridylamido Polymerization Catalysts

FULL PAPER

Table 5. Microstructures of the poly

ferent catalyst aging times.^[a]

ACHTUNGTRNE(NUNG 1-hexene)s prepared from 6b at dif-

with no appreciable influence of the catalyst aging time on

the isotacticity degree (m⁴). The observed isoselectivity from

the C_s-symmetric precursors can be explained by an in-situ

ligand modification brought about by monomer insertion

t_age

[h]

t

A

[m⁴]^[c][r⁴]+

ACHTUNGTRENNUNG

[min]^[b]

[mrrr]+

[rmrm]^[c]

A

ACHTUNGTRENNUNG

Ph

into the Zr^IV

bond.^[2a]Finally, relatively broad polydis-

À

C

1

2

3

4

0

1

5

8

2

5

30

0.47

0.48

0.27

0.24

0.26

0.08

0.09

0.06

0.017

0.007

n.d.

1

6

7

8

persity indices (PDIs) and the molecular weight distribution

profiles (from monomodal to polymodal) measured for all

polymer samples were consistent with a nonliving character

of the catalytic active species; living species were previously

reported by Coates et al. for related pyridylamido-Hf^IVsys-

tems.^[1b]

0.004

[a] Reaction conditions: catalyst 6b (10 mmol), [Ph₃C][BACHTUNGTRENNUNG(C₆F₅)₄]

(10 mmol), 1-hexene (2 mL, 16 mmol), toluene (8 mL), 238C; [b] polymer-

ization time; [c] determined by ¹H NMR spectroscopy; [d] terminal

groups determined by ¹³C{¹H} NMR spectroscopy.

Worthy of note, the prolonged catalyst aging resulted in

unexpected and almost linear increases of the M_nvalues

from 30 to 250 kgmol^À1(Table 4, entries 1–4 and Figure 9,

top). In addition, a slight decrease of both r⁴and mrrm pen-

tads, together with an increase of the regioerrors in the poly-

mer microstructure (up to 8 mol%) were also observed

(Table 5, entries 1–4 and Figure 9, bottom).

The relative areas of the methine pentads were analyzed

with theoretical statistical equations by using either the

chain-end-controlled mechanism or the enantiomorphic site

control, predicted by the site epimerization mechanism. The

best fit with the experimental areas was found by using the

site epimerization mechanism that confirmed the possibility

of the growing polymer chain to migrate from one site to

the other of the C₁-symmetric active species. All these data

taken together suggested the generation of new active forms

throughout the catalyst aging time. Accordingly, the GPC

profiles of the polymeric materials prepared from catalyst

6b at different catalyst aging times displayed either mono-

modal (for unaged catalyst) or bimodal (for any aged cata-

lyst species) molecular weight and composition distributions

(see the Supporting Information). In order to shed light on

this unusual polymerization trend, we were particularly in-

terested in characterizing the nature of the active specie(s)

responsible for the linear increase of the polymer M_nversus

the aging time. Noteworthy, stirring a suspension of 6b in

toluene prior to the addition of the monomer led to the for-

mation of the one major soluble species 12b (Scheme 5).

Scheme 5. Evolution of the activated species 6b into 12b upon simple

catalyst aging.

The key mechanism responsible for the generation of 12b

consists of the metalation of an isopropyl fragment with

concomitant elimination of one molecule toluene, a behav-

ior which is not without precedent in the literature.^[2b]The

¹H NMR spectrum of 12b exhibits two well-separated dou-

blets attributed to the unmodified isopropyl fragment at d=

1.18 and 1.36 ppm, whereas the metalated alkyl moiety gives

rise to two doublets (each integrating for one proton) cen-

tered at d=1.22 and 1.25 ppm, corresponding to the diaste-

reotopic methylene protons and the methyl group, which ap-

1

pears as a doublet at d=1.02 ppm. The two H NMR reso-

nances from the Zr-activated methyl group show a scalar

coupling with the methine residue centered at d=3.43 ppm

and correlate with the same carbon resonance at d=

84.8 ppm (Figure 10). The metalation of the isopropyl frag-

ment accounts for lower symmetry of the resulting species;

Figure 9. Top: M_nof poly

catalyst 6b. Bottom: Increased polymer microstructure regioerrors (2,1

regio insertion) in the polyACHTNUTRGNE(NUG 1-hexene)s obtained at different aging times

of catalyst 6b.

ACHTUNGTREN(NUNG 1-hexene)s prepared at different aging times of

Chem. Eur. J. 2012, 18, 671 – 687

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

679

G. Giambastiani et al.

in polar aromatic solvents (C₆D₅Br or C₆D₅Cl) did not allow

for an exact quantification of the 6b/12b ratio at different

catalyst aging times. However, it is evident, that the soluble

fraction of 6b is quantitatively converted into the more solu-

ble and stable cationic species 12b after 16 h.^[24]

The increasing M_nvalues measured for the polyACHTUNGTRENNUNG(1-hex-

ene)s produced at different 6b aging times, the reduction of

the degree of regioerrors in the polymer microstructure, and

the changes occurring on the polymer molecular weight dis-

tributions (from monomodal to bimodal) all together seem

to offer a reasonable explanation of the observed activated

species evolution (from 6b to 12b).

Indirect supporting evidence of the role played by 12b in

the catalytic performance of the aged catalyst 6b was given

by the analysis of the polyACTHNUTRGENU(GN 1-hexene)s produced by the xylyl

derivative 9b, where further intramolecular activation paths

cannot take place. As Table 4 shows, the catalytic activity of

9b rapidly decreases upon catalyst aging (Table 4, entries 5

vs. 6). Most importantly, the M_nvalues and molecular

weight distributions of the polyACTHNUGTRNEUNG(1-hexene)s produced by 9b

at different catalyst aging times did not show any relevant

variation.

High-temperature polymerization studies: To get further de-

tails about the catalytic performance of the newly prepared

Zr^IV(Bn)₂complexes, high-temperature polymerization and

co-polymerization tests have also been carried out. In addi-

tion to the polymerization study with selected Zr^IV(Bn)₂

complexes (6, 7, and 9), the highly optimized C₁-symmetric

Zr^IV–pyridylamido complex 13, recently proposed by some

of us,^[2g]has been included for comparison.

Overall, the Zr^IV(Bn)₂complexes

scrutinized under high-temperature

conditions (90–1208C), exhibited simi-

lar reactivity trends to those investigat-

ed in the 1-hexene polymerization re-

actions conducted at room tempera-

ture (productivities for catalyst 6 and 9

are higher than that of 7). The catalytic

activity measured for the 6/7/9 group,

under these harsher reaction conditions, was moderate

(15% at the highest) compared with that obtained with

compound 13. Finally, the highest polymer stereo regularity

(see polypropylene section) was obtained with the C₁-sym-

metric complex 13, as expected. This set of data is in line

with other previous studies, indicating how selected structur-

al parameters (i.e., the amido nitrogen and/or the bridging

methylene substitution and the nature of the 2-aryl substitu-

ent) and precatalyst symmetry, play a major role in control-

ling both the catalyst activity and the stereo regularity of

the resulting polymers.

Figure 10.

N

as a matter of fact, the bridging methylene group PyCH₂N

gives rise to a well-defined AB system centered at d=

4.98 ppm, which correlates with a single carbon resonance at

d=63.79 ppm (Figure 10). Unfortunately, the limited solu-

bility of 6b in polar aromatics and its rapid decomposition

Ethylene/1-octene copolymerization study: Ethylene/1-

octene copolymerizations were conducted at 1208C in a 2 L

batch reactor filled with ethylene (460 psi ) and 1-octene

(250 g ). All precatalysts were activated in situ with [HNMe-

ACHUTNERGNNU(G C₁₈H₃₇)₂][BACHTUGNTERN(NUNG C₆F₅)₄] (1.2 equiv, relative to precatalyst) as ac-

680

www.chemeurj.org

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2012, 18, 671 – 687

Zr^IV–Pyridylamido Polymerization Catalysts

FULL PAPER

tivator. Polymerization tests were carried out for 10 min and

stopped by venting the ethylene pressure. As shown in

Table 6, complexes 6, 7, and 9 were moderately active under

lowest stereoregularity, as expected.^[1e]Although isotactical-

ly enriched polypropylenes can be obtained from C_s-sym-

metric complexes as a result of an in-situ catalyst modifica-

À

tion (olefin insertion into the Zr aryl bond), the highest iso-

tacticity is obtained with the C₁-symmetric precursor 13

Table 6. Ethylene/1-octene copolymerization data.^[a]

(Table 7, entries 1–3 vs. 4).

Catalyst

[mmol]

Yield

[g]

Activity

M_w

[10^À3

M_w/M_n

Octene

content

[b]

G

]

N

]

AHCTUNGTRENGN[UN mol%]

Conclusion

1

2

3

4

6 (10)

7 (10)

9 (10)

16.7

3.3

11.5

12.3

10.02

1.94

6.90

542

268

377

2.4

21.8

2.5

4.2

6.6

6.4

8.4

Understanding the importance of the metal-precursor

choice as well as highlighting unexpected organometallic

paths for selected classes of polymerization catalysts is a

more important achievement than a mere chemical curiosity

to be addressed. In the last few years, the organometallic

chemistry and polymerization catalysis associated with

Group IV pyridylamido complexes have offered an excellent

ground for discussion to many research groups from indus-

try and academia around the world; under this perspective,

the present contribution is thought to shed further light on

this topical subject. In the first part of the manuscript, we

present a full account on the temperature-controlled s-bond

metathesis/protonolysis reaction occurring at Group 4 amido

complexes, an original outcome that highlights how the

chemistry of well-known metal precursors such as Zr^IV-

13 (0.7)

105.43

2010^[2g]

1.6

[a] Polymerization conditions: T=1208C, Isopar-E (533 mL), 1-octene

(250 g), ethylene (460 psi), precatalyst/activator/modified methylalumi-

noxane (MMAO)=1:1.2:10, hydrogen (10 mmol), polymerization time=

10 min; [b] g of polymer per mmol Zr per h.

these reaction conditions, leading to very high molecular

weight ethylene/1-octene copolymers ranging from 268 kDa

for 7 to 542 kDa for 6. 1-Octene incorporation, as measured

by IR spectroscopy, ranged from 4.2–6.6 mol% and was

somewhat lower than that found in copolymers produced by

complex 13 (8.4 mol%).^[25]

In all cases, complexes 6, 7, and 9 form copolymers with

higher polydispersities than those coming from the most

active catalyst 13; the least active precatalyst 7 produced co-

polymers with the highest polydispersity.

AHCTUNTGRENGUN(N NMe₂)₄may deserve unexpected reactivity paths. More im-

portantly, this issue provides a better understanding of the

complex structure–activity relationships of Group 4-amido-

based olefin polymerization catalysts. With the aim at em-

phasizing the role of the metal-precursor choice in the pro-

duction of tailored catalyst systems, a direct comparison be-

Propylene polymerization study: Propylene polymerization

reactions were conducted at 908C in a 1.8 L reactor filled

with propylene (180 g) in isoparaffins (Isopar-E) as the sol-

vent (700 mL). The data summarized in Table 7 indicate

tween Zr

spective of the unique nature of the newly synthesized Zr^IV-

{N^À,N,C^À}(Bn)₂precatalysts, and the unambiguous identity

of their activated forms (cationic [Zr^IV{N^À,N,C^À}]⁺[B-

(C₆F₅)₄]^Àand zwitterionic [Zr^IV{N^À,N,C^À}]⁺A (C₆F₅)₃]^À

[BnB

ACHTUGNRTNE(NUNG NMe₂)₄and Zr(Bn)₄has been carried out. Irre-

AHCTUNGTRENNUNG

Table 7. Propylene polymerization data.^[a]

A

T

G

ACHTUNGTRENNUNG

Catalyst

[mmol]

Yield

[g]

Activity

T_m

M_w

[10^À1

M_w/M_n

[m⁴]

species), the wealth of the organometallic chemistry associ-

ated to this class of compounds has unveiled unexpected

catalytic outcomes. Room-temperature polymerization ex-

periments of 1-hexene by using the cationic Zr complexes

[b]

G

N

]

[8C]^[c]

G

]

1

2

3

4

6 (10)

7 (10)

9 (10)

13 (1)

11.9

1.9

7.8

144.9

146.2

146.8

147.7

282

389

230

4.1

8.1

4.0

2.5

0.48

0.75

0.19

0.91

1.14

4.68

48.00

8.0

1052

have provided poly

ACHTUNGTRENNUNG

order of magnitude higher (as high as 44ꢃ10³kg of poly

ACHTUNGTRENNUNG

[a] Polymerization conditions: T=908C, Isopar-E (700 mL ), propylene

(180 g), precatalyst/activator/MMAO=1:1.2:10, reaction time=10 min.

[b] g of polymer per mmol Zr per h. [c] T_mis the melt transition of the

polypropylene samples.

hexene) per mmol Zr per h) than those previously reported

in the literature for related Hf^IV–pyridylamido systems.

Most importantly, catalyst aging tests (conducted prior to

the monomer addition) resulted into catalytic species capa-

ble of generating polymers featured by increased M_nvalues

(30–250 kgmol^À1) and increased regioerrors in the polymer

microstructure (up to 8 mol%). An in-depth investigation of

the catalyst aging in solution (in the absence of monomer)

has unambiguously demonstrated its progressive evolution

towards the formation of (at least) one major novel active

form with reduced symmetry; this finding adds additional

layers of complexity to the identification of the true active

species operating in the polymerization process.

that Zr^IVcomplexes 6 and 9 exhibit (among the C_s-symmet-

ric precatalysts) the highest catalytic activities, whereas com-

plex 7 shows moderate performance (Table 7, entries 1 and

3 vs. 2). Overall, the C_s-symmetric precursors show about

15% of the catalytic activity measured for complex 13, at

the highest. The polydispersity versus the catalytic activity

trend is almost identical to that described above for the eth-

ylene/1-octene copolymerization tests (see Table 6 vs.

Table 7). Finally, precatalyst 9, bearing the xylyl substituent

on the amido nitrogen atom, gave polypropylene with the

Finally, high-temperature polymerization studies have

shown that the cationic Zr complexes of the present study

Chem. Eur. J. 2012, 18, 671 – 687

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

681

G. Giambastiani et al.

are moderately active catalysts for propene polymerization

and ethylene/1-octene copolymerization.

Experimental Section

General considerations and materials characterization: All air- and/or

moisture-sensitive reactions were performed under inert atmosphere in

flame-dried flasks by using standard Schlenk-type techniques or in a

drybox filled with nitrogen. Benzene and toluene were purified by distil-

lation from sodium/triglyme benzophenone ketyl and stored over 4 ꢂ

molecular sieves or were obtained by means of a MBraun solvent purifi-

cation system. [D₆]Benzene and [D₈]toluene were dried over sodium/ben-

zophenone ketyl and condensed in vacuo over activated 4 ꢂ molecular

sieves prior to use. 1-Hexene was purified prior to use by distillation over

Na, stored at room temperature under a nitrogen atmosphere over acti-

vated 4 ꢂ molecular sieves, and used within 2 days. All the other reagents

and solvents were used (unless otherwise stated) as purchased from com-

mercial suppliers without further purification. The aminopyridinate li-

gands 1a–1d were obtained on a multigram scale according to standard

procedures reported in the literature^[26]or with slight modifications.^[27]1D

(¹H and ¹³C{¹H}) and 2D (COSY H–H, HETCOR H–C) NMR spectra of

all organometallic species were obtained on either a Bruker Avance

DRX-400 spectrometer (400.13 and 100.61 MHz for ¹H and ¹³C, respec-

tively) or a Bruker Avance 300 MHz instrument (300.13 and 75.47 MHz

Data for 2/2’ (ꢂ99:1): ¹H NMR (300 MHz, C₆D₆, 293 K, selected data):

d=1.33 (d, ³J (CH₃^ACH₃^B), H^22,24, 2), 1.49 (d, ³J-

ACTHNUTRGENN(GU H,H)=6.9 Hz, 6H; CHACHUTNGTRENNUGN

A

R

ACHTUNGTRENNUNG(CH₃)₂,

1

for H and ¹³C, respectively). Chemical shifts (d) are reported in ppm rel-

A

ACHTUNGTRENNUNG

ative to trimethylsilane (TMS), referenced to the chemical shifts of resid-

ual solvent resonances (¹H and ¹³C). The multiplicities of the

¹³C{¹H} NMR spectra were determined on the basis of the ¹³C{¹H} J-

modulated spin-echo sequence and quoted as CH₃, CH₂, CH, and C for

primary, secondary, tertiary, and quaternary carbon atoms, respectively.

The C, H, N, S elemental analyses were made at ICCOM-CNR by using

a Thermo FlashEA 1112 Series CHNS-O elemental analyzer with an ac-

cepted tolerance of Æ0.4 units on carbon (C), hydrogen (H), and nitro-

gen (N).

2), 6.89 (t, ³J

G

,

H^15,16,17, 2), 7.25–7.30 (3H; CH_Ar, H^8,9,10, 2), 7.44–7.47 ppm (2H; CH_Ar

H^7,11, 2); ¹³C{¹H} NMR (75 MHz, C₆D₆, 293 K, selected data): d=23.9

(CH

A

(C², 2), 123.2 (C^15,17, 2), 123.9 (C¹⁶, 2), 127.6 (C^8,10, 2), 128.4 (C⁹, 2), 128.5

(C^7,11, 2), 137.2 (C³, 2), 140.0 (C_,2), 144.6 (C, 2), 150.5 (C, 2), 160.9 (C,

2), 164.6 ppm (C, 2); elemental analysis calcd (%) for C₃₀H₄₅N₅Zr

(566.94): C 63.56, H 8.00, N 12.35; found: C 63.32, H 7.92, N 12.47.

X-ray data measurements: Single-crystal X-ray data were collected at

low temperature (100 or 120 K) on an Oxford Diffraction XCALIBUR 3

diffractometer equipped with a CCD area detector by using Mo_Karadia-

tion (l=0.7107 ꢂ). The program used for the data collection was Crys

Alis CCD 1.171.^[28]Data reduction was carried out with the program

Data for 3/3’ (ꢂ99:1): ¹H NMR (300 MHz, C₆D₆, 293 K, selected data):

d=1.32 (d, ³J (CH₃^ACH₃^B), H^20,22, 3), 1.48 (d, ³J-

ACTHNUTRGENN(GU H,H)=6.9 Hz, 6H; CHACHUTNGTRENNUGN

G

CrysACHTUNGTRENNUNG

lis RED 1.171^[29]and the absorption correction was applied with

the program ABSPACK 1.17.^[28]Direct methods implemented in Sir97^[30]

were used to solve the structures and the refinements were performed by

full-matrix least-squares against F²implemented in SHELX97.^[31]All the

non-hydrogen atoms were refined anisotropically (with the exception of

those of the disordered diethyl ether and toluene solvent molecules in 8

and 9, respectively), whereas the hydrogen atoms were fixed in calculated

positions and refined isotropically with the thermal factor depending on

the one of the atom to which they are bound (riding model). The pres-

ence of large accessible voids (107 ꢂ³) in the lattice of 8 is due to the

presence of disordered Et₂O molecules that cannot be located precisely

in the Fourier difference density maps. Molecular plots were produced by

the program ORTEP3.^[32]

AHCTUNGTRENNUNG

,

AHCTUNGTRENNUNG

U

ACHTUNGTRENNUNG

A

ACHTUNGTRENNUNG

112.5 (C⁸, 3), 119.2 (C², 3), 119.9 (C⁴, 3), 123.2 (C^13,15, 3), 124.1 (C¹⁴, 3),

137.0 (C³, 3), 143.2 (C⁷, 3), 144.8 (C, 3), 150.0 (C, 3), 150.1 (C, 3), 151.5

(C, 3), 164.2 ppm (C, 3); elemental analysis calcd (%) for C₂₈H₄₃N₅OZr

(556.90): C 60.39, H 7.78, N 12.58, O 2.87; found: C 60.20, H 7.52, N

12.43, O 2.95.

General procedure for the synthesis of Zr^IV–pyridylamido complexes 6–

9: A solution of tetrabenzylzirconium (95%, 1 equiv) in dry and degassed

C₆H₆(4 mL) was added dropwise to a solution of the proper aminopyri-

dinate ligand N₂H^R(1a–1d) (0.100 g) in dry and degassed C₆H₆(2 mL)

and heated at 708C, The reaction was maintained at the same tempera-

ture for several hours and the reaction course was periodiocally moni-

tored by sampling the mixture at different times and following the ligand

consumption by ¹H NMR spectroscopy (inside an NMR capillary filled

with C₆D₆) till completeness (6: 13, 7: 16, 8, 48, 9: 13 h). Afterwards, the

reaction mixture was concentrated in vacuo to afford red-brown solids.

The crude materials were re-crystallized by layering a toluene solution

with pentane and cooling down the mixtures overnight to À308C (6:

0.170 g, 95%; 7: 0.155 g, 89%; 8: 0.165 g, 91%; 9: 0.180 g, 95%). Crystals

suitable for X-ray diffraction analyses were obtained from concentrated

CCDC 829625 (6), 829624 (7), 829626 (8·0.25Et₂O), and 829623 (9·0.5tol-

uene) contain the supplementary crystallographic data for this paper.

These data can be obtained free of charge from The Cambridge Crystal-

lographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.

General procedure for the synthesis of the Zr^IV–pyridylamido tautomeric

mixtures 2/2’ and 3/3’: A solution of tetrakis(dimethylamido)zirconium

(99%, 1 equiv) in C₆H₆(3 mL) was added dropwise to a solution of the

proper aminopyridinate ligand N₂H^R(1a or 1b) (0.150 g) in C₆H₆(2 mL).

The reaction mixture was maintained at RT under stirring for 3 h and

then concentrated in vacuo to afford a yellowish solid residue. The crude

material was re-crystallized by layering a toluene solution with pentane

and cooling down the mixture overnight to À308C (2/2’: 0.230 g, 94%; 3/

3’: 0.232 g, 93%). Products were isolated as tautomeric mixtures (2/2’ and

3/3’) in a roughly 99:1 ratio at 298 K.

682

www.chemeurj.org

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2012, 18, 671 – 687

Zr^IV–Pyridylamido Polymerization Catalysts

FULL PAPER

140.5 (C), 146.3 (C), 146.8 (C), 149.4 (C), 157.9 (C), 164.2 (C), 195.4 ppm

(C⁷); elemental analysis calcd (%) for C₃₆H₃₈N₂SZr (621.99): C 69.52, H

6.16, N 4.50, S 5.16; found: C 69.25, H 6.09, N 4.43, S 5.64.

Data for 9: ¹H NMR (300 MHz, CD₂Cl₂, 293 K): d=1.52 (d, ABq, ²J-

ACHUTNGRENNUG CAHTUNGTRENNUNG

(H,H)=9.8 Hz, 2H; ZrCH₂Ph, H^21,28), 2.05 (d, ABq, ²J

A

ACHTUNGTRENNUNG

AHCTUNGTRENNUNG

A

,

¹³C{¹H} NMR (75 MHz, CD₂Cl₂, 293 K): d=19.0 (CH₃, C^19,20), 65.2 (C¹²),

67.1 (C^21,28), 115.2 (C⁴), 117.6 (C²), 122.3 (C¹⁰), 122.4 (C^25,32), 125.2 (C¹⁶),

127.6 (C⁸), 127.7 (C^23,27,30,34), 128.6 (C^15,17), 128.7 (C⁹), 129.1 (C^24,26,31,33),

134.2 (C⁷), 136.4 (C), 140.0 (C³), 141.0 (C), 145.8 (C), 148.6 (C), 163.6

(C), 164.9 (C), 189.7 ppm (C¹¹); elemental analysis calcd (%) for

C₃₄H₃₂N₂Zr (559.86): C 72.94, H 5.76, N 5.00; found: C 72.68, H 5.65, N

5.32.

General procedures for the synthesis of the zwitterionic pyridylamido

complexes 6a–9a upon B

ACHTNUGERTN(NGNU C₆F₅)₃activation of the dibenzyl precursors: A

solution of B(C₆F₅)₃(1 equiv) in C₆D₆(0.5 mL) was added in one portion

AHCTUNGTRENNUNG

to a dry and degassed solution of the proper dibenzyl complex 6–9

(0.025 g) in C₆D₆(0.5 mL). The addition of the Lewis acid causes the

rapid change of the color of the solutions from yellowish to dark-orange.

The final mixtures were stirred at room temperature for 5 min before

they were sealed in an NMR tube and spectroscopically characterized.

Activation yields were assumed to be quantitative by ¹H NMR spectros-

copy because of the absence of detectable amount of the signals attribut-

ed to the starting dibenzyl species.

solutions of each complex in toluene at room temperature (6), Et₂O

cooled to À308C (7 and 8), toluene cooled to À308C (9).

Data for 6: ¹H NMR (300 MHz, CD₂Cl₂, 293 K): d=1.17 (d, ³J

ACHTUNGTRENNUNG

6.8 Hz, 6H; CH

H^21,23), 1.59 (d, ABq, ²J

ABq, ²J

G

E

ACHTUNGTRENNUNG

E

ACHTUNGTRENNUNG

3

7.4 Hz, 4H; o-Ph, H^27,31,34,38), 6.71 (t, ³J

3

6.88 (t, J

(H,H)=7.6 Hz, 4H; m-Ph, H^28,30,35,37), 7.11 (d, J

N

1H; CH_Ar, H²), 7.22–7.27 (5H; CH_Ar, H^8,9,15,16,17), 7.53–7.59 (2H; CH_Ar

H^4,10), 7.73 (t, ³J(H,H)=7.7 Hz, 1H; CH_Ar, H³), 7.80 ppm (m, 1H; CH_Ar

H⁷); ¹³C{¹H} NMR (75 MHz, CD₂Cl₂, 293 K): d=22.6 (CH

27.4 (CH

(CH₃)₂, C^22,24), 27.8 (CH

,

ACHTUNGTRENNUNG

(CH₃)₂, C^21,23),

E

(CH₃)₂, C^19,20), 67.3 (C¹²), 68.2 (C^25,32),

115.2 (C⁴), 117.4 (C²), 122.3 (C¹⁰), 122.4 (C^29,36), 123.9 (C^15,17), 126.0 (C¹⁶),

127.6 (C^28,30,35,37), 127.7 (C⁸), 128.8 (C⁹), 129.0 (C^27,31,34,38), 134.5 (C⁷),

140.1 (C³), 140.8 (C), 145.6 (C), 146.7 (C), 146.9 (C), 163.6 (C), 164.6 (C),

189.8 ppm (C¹¹); elemental analysis calcd (%) for C₃₈H₄₀N₂Zr (615.96): C

74.10, H 6.55, N 4.55; found: C 74.02, H 6.49, N 4.61.

Data for 7: ¹H NMR (300 MHz, CD₂Cl₂, 293 K): d=1.04–1.10 (m, 8H;

3

CH

H^19,21), 2.00 (d, ABq, ²J

³J

(H,H)=6.8 Hz, 2H; CH

6.34 (m, 5H; CH_Ar, o-Ph, H^{8,25,29,32,36}), 6.69 (t, ³J

G

N

E

7.7 Hz, 1H; CH_Ar, H⁴), 7.11–7.14 (m, 3H; CH_Ar, H^13,14,15), 7.28 (d, ³J-

3

G

N

,

H³); ¹³C{¹H} NMR (75 MHz, CD₂Cl₂, 293 K): d=22.5 (CH

ACHTUNGTRENNUNG

27.4 (CH

N

ACHTUNGTRENNUNG

112.4 (C⁴), 116.0 (C²), 118.9 (C⁸), 123.2 (C^27,34), 123.8 (C^13,15), 125.9 (C¹⁴),

127.5 (C^26,28,33,35), 129.5 (C^25,29,32,36), 140.4 (C³), 140.7 (C), 142.1 (C⁹), 146.1

(C), 146.8 (C), 152.1 (C), 162.1 (C), 164.6 (C), 168.0 ppm (C⁷); elemental

analysis calcd (%) for C₃₆H₃₈N₂OZr (605.92): C 71.36, H 6.32, N 4.62, O

2.64; found: C 71.22, H 6.28, N 4.45, O, 2.94.

Data for 6a: ¹H NMR (400 MHz, C₆D₆, 293 K, selected data): d=0.95

(d, ³J

CH

(H,H)=6.5 Hz, 3H; CH

ZrCHHPh, H^19,25b), 2.74 (d, ³J

3.09 (m, 1H; BCHHPh, H^32c), 3.54–3.61 (m, 2H; CH

H^20,32c’), 4.13 (d, ³J(H,H)=19.9 Hz, 1H; CHHN, H^12a), 4.76 ppm (d, ³J-

(H,H)=19.9 Hz, 1H; CHHN, H^12a’); ¹³C{¹H} NMR (100 MHz, C₇D₈,

293 K, selected data): d=21.5 (CH

(CH₃), C²³), 23.2 (CH(CH₃), C²⁴), 26.2

(CH (CH₃), CH

(CH₃), C²²), 27.3 (CH(CH₃), C²⁰), 27.4 (CH (CH₃), C^19,21),

69.3 (CH₂N, H¹²), 79.6 (ZrCH₂Ph, C²⁵), 185.9 ppm (C¹¹).

Data for 7a: ¹H NMR (400 MHz, C₇D₈, 293 K, selected data): d=0.78

(d, ³J (CH₃), H²⁰), 0.91 (d, ³J

(H,H)=6.8 Hz, 3H; CH (H,H)=6.8 Hz, 3H;

CH (H,H)=6.8 Hz, 3H; CH

(CH₃), H²²), 1.10 (d, ³J (CH₃), H¹⁹), 1.27 (d, J-

(H,H)=6.8 Hz, 3H; CH

(CH₃), H²¹), 2.14 (m, 1H; ZrCHHPh, H^23b), 2.25

(sept, ³J (CH₃)₂, H¹⁷), 2.41 (d, ³J

(H,H)=6.8 Hz, 1H; CH (H,H)=11.5 Hz,

A

(CH₃), H²²), 1.00 (d, ³J

ACHUTGTNRENNUG ACHUTNTGREN(NUGN H,H)=6.5 Hz, 3H;

3

E

A

ACHTUNGTRENNUNG

A

R

ACHTUNGTRENNUNG

Data for 8: ¹H NMR (300 MHz, CD₂Cl₂, 293 K): d=1.16 (d, ³J

6.8 Hz, 6H; CH (CH₃), ZrCH₂Ph,

(CH₃), H^20,22), 1.31–1.33 (m, 8H; CH

H^19,21,23,30), 2.16 (d, ABq, ²J(H,H)=9.6 Hz, 2H; ZrCH₂Ph, H^23,30), 3.57

(sept, ³J (CH₃)₂, H^17,18), 4.82 (s, 2H; CH₂N, H¹⁰),

(H,H)=6.8 Hz, 2H; CH

6.39 (d, ³J

AHCTUNGTRENGN(UN H,H)=

AHCTUNGTRENNUNG

A

ACHTUNGTRENNUNG

AHCTUNGTRENNUNG

A

ACHTUNGTRENNUNG

ACHTUNGTRENNUNG ACHTUGNTREN(NUGN H,H)=7.3 Hz,

(H,H)=7.5 Hz, 4H; o-Ph, H^25,29,32,36), 6.75 (t, ³J

A

ACHTUNGTRENNUNG

2H; p-Ph, H^27,34), 6.92–7.00 (m, 5H; m-Ph, CH_Ar, H^{2,26,28,33,35}), 7.08 (d, J-

3

A

E

N

ACHTUNGTRENNUNG

(H,H)=7.7 Hz, 1H; CH_Ar, H⁴), 7.13 (d, J

N

3

7.22–7.24 (m, 4H; CH_Ar, H^9,13,14,15), 7.62 ppm (t, ³J

ACHTUNGTRENNUNG

CH_Ar, H³); ¹³C{¹H} NMR (75 MHz, CD₂Cl₂, 293 K): d=22.5 (CH

AHCTUNGTRENNUNG

A

E

ACHTUNGTRENNUNG

C^19,21), 27.4 (CH(CH₃)₂, C^17,18), 27.8 (CH

ACHTUNGTRENNUNG ACHTUNGTRENNUNG

3

R

G

ACHTUNGTRENNUNG

(C^23,30), 114.8 (C⁴), 115.8 (C²), 123.0 (C^27,34), 123.9 (C^13,15), 125.9 (C¹⁴),

126.0 (C⁹), 127.5 (C^26,28,33,35), 129.4 (C^25,29,32,36), 135.5 (C⁸), 140.4 (C³),

A

ACHTUNGTRENNUNG

A

R

ACHTUNGTRENNUNG

Chem. Eur. J. 2012, 18, 671 – 687

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

683

G. Giambastiani et al.

1

1H; ZrCHHPh, H^23b’), 3.37–3.47 (m, 2H; CH

3.58 (m, 1H; BCHHPh, H^30c’), 3.98 (d, ³J

(H,H)=20.8 Hz, 1H; CHHN,

H^10a), 4.74 ppm (d, ³J(H,H)=20.8 Hz, 1H; CHHN, H^10a’); ¹³C{¹H} NMR

(100 MHz, C₇D₈, 293 K, selected data): d=21.9 (CH

(CH₃), C²¹), 23.0

(CH

(CH₃), C²²), 26.3 (CH(CH₃), C²⁰), 27.1 (CH(CH₃), C¹⁸), 27.2 (CH-

(CH₃), C¹⁹), 27.5 (CH(CH₃), C¹⁷), 68.6 (CH₂N, H¹⁰), 80.4 (ZrCH₂Ph, C²³),

168.9 ppm (C⁷).

Data for 8a: ¹H NMR (400 MHz, C₇D₈, 293 K, selected data): d=0.89

(d, ³J (CH₃), H²⁰), 0.93 (d, ³J

(H,H)=6.8 Hz, 3H; CH (H,H)=6.8 Hz, 3H;

CH (H,H)=6.8 Hz, 3H; CH

(CH₃), H²²), 1.16 (d, ³J (CH₃), H¹⁹), 1.40 (d, J-

(H,H)=6.8 Hz, 3H; CH (CH₃)₂,

(CH₃), H²¹), 2.22–2.30 (2H; CH

ZrCHHPh, H^17,23b), 2.67 (d, ³J(H,H)=11.2 Hz, 1H; ZrCHHPh, H^23b’),

3.34 (m, 1H; BCHHPh, H^30c), 3.60 (sept, ³J

(H,H)=6.8 Hz, 1H; CH-

(CH₃)₂, H¹⁸), 3.85 (m, 1H; BCHHPh, H^30c’), 4.06 (d, ³J

(H,H)=19.9 Hz,

1H; CHHN, H^10a), 4.87 ppm (d, ³J(H,H)=19.9 Hz, 1H; CHHN, H^10a’);

¹³C{¹H} NMR (100 MHz, C₇D₈, 293 K, selected data): d=21.6 (CH

(CH₃),

C²¹), 23.1 (CH(CH₃), C²²), 26.3 (CH(CH₃), C²⁰), 27.2 (CH(CH₃), C¹⁸),

(CH₃), C^17,19), 69.2 (CH₂N, H¹⁰), 81.6 (ZrCH₂Ph, C²³),

G

Data for 7b: H NMR (400 MHz, C₆D₅Br, 293 K, selected data): d=1.09

(d, ³J (CH₃), H^20,22), 1.22 (d, ³J

(H,H)=6.7 Hz, 6H; CH (H,H)=6.7 Hz,

6H; CH

(CH₃), H^19,21), 2.73 (s, 2H; ZrCH₂Ph, H²³), 3.22 (m, 2H; CH-

(CH₃), H^17,18), 4.90 ppm (s, 2H; CH₂N, H¹⁰).

E

R

ACHTUNGTRENNUNG

AHCTUNGTRENNUNG

ACHTUNGTRENNUNG

AHCTUNGTRENNUNG

1

N

G

Data for 8b: H NMR (400 MHz, C₆D₅Br, 293 K, selected data): d=1.10

(d, ³J (CH₃), H^20,22), 1.22 (d, ³J

(H,H)=6.7 Hz, 6H; CH (H,H)=6.7 Hz,

6H; CH

(CH₃), H^19,21), 2.76 (s, 2H; ZrCH₂Ph, H²³), 3.28 (m, 2H; CH-

(CH₃), H^17,18), 4.98 ppm (s, 2H; CH₂N, H¹⁰).

E

ACHTUNGTRENNUNG

A

R

ACHTUNGTRENNUNG

AHCTUNGTRENNUNG

1

N

Data for 9b: H NMR (400 MHz, C₆D₅Br, 293 K, selected data): d=2.15

(s, 6H; CH

(CH₃), H^19.20), 2.72 (s, 2H; ZrCH₂Ph, H²¹), 4.64 ppm (s, 2H;

CH₂N, H¹²).

N

AHCTUNGTRENNUNG

R

E

AHCTUNGTRENNUNG

Synthesis of 12b: In a dry-box filled with nitrogen, a solution of 6b

(0.025 g) in dry and degassed C₇D₈(1 mL) was maintained under vigo-

rous stirring in the dark at room temperature

for 16 h. The soluble fraction was isolated

from the residual 6b fraction and analyzed by

A

AHCTUNGTRENNUNG

N

G

1D and 2D NMR spectroscopy (see

Figure 10). ¹H NMR (400 MHz, C₇D₈, 293 K,

27.3 (CHACHTUNGTRENNUNG(CH₃), CHACHTUNGTRENNUNG

192.2 ppm (C⁷).

selected data): d=1.02 (d, ³J

ACHTUNGTRENNUNG

1

3H; CH

(CH₃), H²⁴), 1.18 (d, ³J

(CH₃), H²²), 1.22 (d, ³J

ACHTUNGTRENNUNG

Data for 9a: H NMR (400 MHz, C₇D₈, 293 K, selected data): d=1.55 (s,

3H; CH₃, H¹⁹), 2.23 (d, ³J

3H; CH₃, H²⁰), 2.66 (d, ³J

(m, 1H; BCHHPh, H^28c), 3.45 (m, 1H; BCHHPh, H^28c’), 3.56 (d, ³J-

(H,H)=11.0 Hz, 1H; ZrCHHPh, H^21b), 2.33 (s,

1H; ZrCHHCH, H^23b), 1.25 (d, ³J

ACHTUNGTRENNUNG

AHCTUNGTRENNUNG

ACHTUNGTRENNUNG ACHTUNTGREN(NUGN H,H)=20.7 Hz, 1H;

(H,H)=20.7 Hz, 1H; CHHN, H^12a), 4.49 ppm (d, ³J

A

E

ACHTUNGTRENNUNG

2

CHHN, H^12a’); ¹³C{¹H} NMR (100 MHz, C₇D₈, 293 K, selected data): d=

18.1 (CH₃, C²⁰), 21.0 (CH₃, C¹⁹), 37.8 (BCH₂Ph, C²⁸), 67.7 (CH₂N, C¹²),

79.8 (ZrCH₂Ph, C²¹), 185.2 ppm (C¹¹).

G

ACHTUNGTRENNUNG

AHCTUNGTRENNUNG

General procedures for the synthesis of cationic pyridylamido complexes

Polymerization procedures and polymer characterizations

ACHTUNGTRENNUNG

6b–9b upon Ph₃C⁺[B(C₆F₅)₄]^Àactivation of the dibenzyl precursors: All

Propylene polymerization: Propylene polymerizations were conducted in

a 1.8 L SS batch reactor (manufactured by Buchi AG and sold by Met-

tler). The reactor was heated/cooled through the vessel jacket and reac-

tor head. Syltherm 800 was used as the heat transfer fluid; it was con-

trolled by a separate heating/cooling skid. Both the reactor and the heat-

ing/cooling system were controlled and monitored by a Camile TG pro-

cess computer. The bottom of the reactor was fitted with a large orifice

bottom dump valve, which empties the reactor contents into a 6 L SS

dump pot. The dump pot was vented to a 30 gal blowndown tank, the pot

and the tank were purged with N₂. All chemicals used for polymerization

or catalyst makeup were run through purification columns, to remove

any impurities that may affect polymerization. Propylene, toluene, and

Isopar-E were passed through two columns, the first containing A2 alu-

mina, the second of which contained Q5 reactant. N₂and H₂were passed

through a single Q5 reactant column. The reactor was cooled to 508C for

chemical additions. The Camile controls the addition of Isopar-E by

using a micro-motion flowmeter to add the desired amount accurately.

The accurate addition of H₂was achieved by pressuring a 50 mL shot

tank to 240 psi, and adding H₂slowly until the desired decrease was re-

flected in the shot tank pressure. Propylene was then added through the

micro-motion flow meter. After the additions, the reactor was heated to

the polymerization temperature. The activator(s) and catalyst were han-

dled in an inert dry-box, mixed together in a vial, drawn into a syringe,

and pressure transferred into the catalyst shot tank. The polymerizations

were run for 10 min, then the agitator was stopped, the reactor pressured

to approximately 500 psi with N₂, and the bottom dump valve opened to

empty the reactor contents into the dump pot. The dump-pot contents

were poured into trays and placed in a lab hood where the solvent was

evaporated off overnight. The trays containing the polymeric residue

were then transferred to a vacuum oven, where they were heated to

1458C under vacuum to remove any residual solvent. After the trays

were cooled to ambient temperature, the polymers were weighed for

yield/efficiencies, and submitted for polymer testing.

samples were prepared at room temperature in a dry-box filled with ni-

trogen. A solution of Ph₃C⁺[B(C₆F₅)₄]^À(1 equiv) in C₆D₆(2 mL) was

AHCTUNGTRENNUNG

added in one portion to a dry and degassed solution of the proper diben-

zyl complex 6–9 (0.025 g) in C₆D₆(0.5 mL). Dark-green rubbery materi-

als precipitate out upon addition of the tritylborate and were decanted

from the supernatant. Afterwards, the solids were washed with C₆H₆(2ꢃ

2 mL) and dried under vacuum to constant weight before they were char-

acterized by ¹H NMR spectroscopy. Again, activation yields were as-

1

sumed to be quantitative from H NMR spectroscopy; the mother liquors

contained an equimolar amount of the tritylbenzyl [Ph₃CCH₂Ph] species.

Polypropylene (material characterization): The molecular weight distribu-

tion (M_w, M_n) information was determined by analysis on a custom Dow-

built robotic-assisted dilution high-temperature gel permeation chroma-

tographer (RAD-GPC). Polymer samples were dissolved by stirring them

for 90 min at 1608C at a concentration of 30 mgmL^À1in 1,2,4-trichloro-

benzene (TCB) stabilized by 300 ppm of butylated hydroxytoluene

1

Data for 6b: H NMR (400 MHz, C₆D₅Br, 293 K, selected data): d=1.11

(d, ³J

6H; CH

(CH₃), H^19,20), 5.05 ppm (s, 2H; CH₂N, H¹²).

A

(CH₃), H^22,24), 1.21 (d, ³J

ACHUTGTNRENNUG ACHUTTGNREN(NUGN H,H)=6.8 Hz,

ACHTUNGTRENNUNG

684

www.chemeurj.org

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2012, 18, 671 – 687

Zr^IV–Pyridylamido Polymerization Catalysts

FULL PAPER

(BHT). The samples were diluted to 1 mgmL^À1before an aliquot of the

sample (400 mL) was injected. The GPC utilized two Polymer Labs PLgel

10 mm MIXED-B columns (300ꢃ10 mm) at a flow rate of 2.0 mLmin^À1

at 1508C. Sample detection was performed by using a PolyChar IR4 de-

tector in the concentration mode. A conventional calibration of narrow

polystyrene (PS) standards was utilized, with apparent units adjusted to

homo-polyethylene (PE) by using the known Mark–Houwink coefficients

for PS and PE in TCB at this temperature. Absolute M_winformation was

calculated by using a PDI static low-angle light scatter detector. Melting

and crystallization temperatures of the polymers were measured by dif-

ferential scanning calorimetry (DSC 2910, TA Instruments, Inc.). The

samples were first heated from room temperature to 2108C at

108Cmin^À1. After being held at this temperature for 4 min, the samples

were cooled to À408C at 108Cmin^À1and were then heated to 2158C at

108Cmin^À1after being held at À408C for 4 min.

broadband decoupling was achieved with a 1D sequence by using bi_

waltz_16_32 power-gated decoupling. Chemical shifts for ¹H were re-

ferred to internal solvent resonances (5.86 ppm) and chemical shifts for

¹³C were referred to hexamethyldisiloxane.

Molecular weight measurement: GPC measurements were performed by

using 12 mg of the product in tetrahydrofuran at 358C by a GPCV2000

high-temperature size exclusion chromatography (SEC) system from

Waters (Millford, MA, USA) equipped with two online detectors: a visc-

ometer (DV) and a differential refractometer (DRI). The column set was

composed of three mixed TSK-Gel GMHXL-XT columns from Tosohaas.

Molar masses were based on polystyrene standards. The universal cali-

bration was constructed from 18 narrow MMD polystyrene standards,

with the molar mass ranging from 162–5.48ꢃ106 gmol^À1

.

Stoichiometric and sub-stoichiometric reactions of 6b and 7b with 1-

hexene: The catalyst precursor 6 or 7 (250 mmol) was dissolved in toluene

(8 mL) and treated with freshly distilled 1-hexene (either 35 mL or

¹³C NMR analysis of the polypropylene samples: The sample was pre-

pared by adding the stock solution (2.7 g) to a sample (0.21–1.2 g) in a

10 mm NMR tube, which was then purged it in an N₂box for 2 h. The

stock solution was made by dissolving para-dichlorobenzene (PDCB; 4 g)

in ortho-dichlorobenzene (39.2 g) with chromium acetylacetonate

(0.025m, relaxation agent). The sample was dissolved and homogenized

by heating the tube and its contents at 140–1508C. Samples were homo-

genized immediately prior to insertion into the heated (1258C) NMR

tube changer, and were allowed to thermally equilibrate in the probe for

7 min prior to data acquisition. The NMR polymer characterization was

carried out on a Bruker 400 MHz spectrometer equipped with a Bruker

Dual DUL high-temperature CryoProbe.^[33]The ¹³C NMR data were ac-

quired by using 320 transients per sample, a 7.3 s pulse repetition delay

(6 s delay+1.3 s acquisition time), 90 degree flip angles, and modified in-

verse gated decoupling^[34]with a sample temperature of 1208C. All meas-

urements were made on non-spinning samples in locked mode.

ACHTUNGTRENNUNG

210 mL, 1 equiv or 6 equiv). A solution of the activator Ph₃C⁺[B(C₆F₅)₄]^À

(250 mmol) in toluene (6 mL) was added in one portion to the resulting

mixture. The mixture was then maintained under vigorous stirring at

room temperature for 1 h. Afterwards, the reaction mixture was treated

with MeOH (3 mL) followed by a saturated aqueous solution of NaCl

(20 mL). The organic phase was separated and the mother liquor was ex-

tracted with AcOEt (Ac=acyl; 3ꢃ10 mL). The collected organic layers

were concentrated in vacuum and analyzed by GC-MS analysis (GC pro-

gram: 408Cmin^À, 158Cmin^À1, 2508C20 min^À).

Ethylene-1-octene copolymerization: A 2 L Parr reactor was used in the

copolymerization tests. All feeds were passed through columns of alumi-

na and Q-5 catalyst (available from Engelhard Chemicals Inc.) prior to

introduction into the reactor. The pre-catalyst and the co-catalyst (activa-

tor) solutions were handled in a nitrogen-filled dry-box. The reactor was

charged with mixed alkanes solvent (Isopar-E, ca. 533 g) and the co-mo-

nomer 1-octene (250 g). Hydrogen was added as a molecular weight con-

trol agent by differential pressure expansion from a 75 mL addition tank

at 300 psi (2070 kPa). The reactor contents were heated to the polymeri-

zation temperature of 1208C and saturated with ethylene (460 psi,

3.4 MPa). The catalysts and the co-catalysts, as dilute solutions in tolu-

ene, were mixed and transferred to a catalyst addition tank and injected

into the reactor. The polymerization conditions were maintained for

10 min with ethylene added on demand. Heat was continuously removed

from the reaction vessel through an internal cooling coil. The resulting

solution was removed from the reactor, quenched with isopropyl alcohol,

and stabilized by addition of a solution of a hindered phenol antioxidant

(Irganox 1010 from Ciba Geigy Corporation; ca. 67 mg) and a phospho-

rus stabilizer (Irgafos 168 from Ciba Geigy Corporation; 133 mg ) in tol-

uene (10 mL). Between the polymerization runs, a wash cycle was con-

ducted in which mixed alkanes (850 g) were added to the reactor and the

reactor was heated to 1508C. The reactor was then emptied of the heated

solvent immediately before beginning a new polymerization run. Poly-

mers were recovered by drying for about 12 h in a temperature-ramped

vacuum oven with a final set point of 1408C.

1-Hexene polymerization by the cationic pyridylamido complexes 6b–9b:

All polymerization tests were conducted under an inert atmosphere in a

nitrogen-filled dry-box. The catalyst precursor (10 mmol) was dissolved in

toluene (5 mL) and treated with freshly distilled 1-hexene (2 mL,

ACTHNUTRGNEUNG

0.016 mol). A solution of the activator Ph₃C⁺[B(C₆F₅)₄]^À(10 mmol) in tol-

uene (3 mL) was added in one portion to the resulting mixture. The mix-

ture was then maintained under vigorous stirring at room temperature

for the desired reaction time. Afterwards, polymerization was stopped by

quenching the mixture with MeOH (5 mL). The mixture was maintained

under vigorous stirring for three hours, while a sticky colorless solid sepa-

rated off. The solid material was decanted, washed with methanol (3ꢃ

5 mL) and dried at 508C under vacuum to constant weight. For the mo-

nomer conversion, n-heptane was used as the GC internal standard. Ac-

tivities were calculated on the basis of the isolated polyACTHNUTRGNE(NUG 1-hexene) mate-

rials and expressed as mol of 1-hexene converted per mol Zr per h. Cata-

lyst aging tests were carried out by adding the monomer (after the

proper aging time) to the pre-activated catalyst. Catalytic 1-hexene poly-

merization experiments were performed in a dry-box filled with nitrogen

by using a 50 mL Schlenk-type round flask equipped with a magnetic stir-

rer bar.

Ethylene-1-octene copolymers (material characterization): Melting and

crystallization temperatures of the ethylene–1-octene copolymers were

measured by differential scanning calorimetry (DSC 2910, TA Instru-

ments, Inc.). Samples were first heated from room temperature to 1808C

at 108Cmin^À1. After being held at this temperature for 2–4 min, the sam-

ples were cooled to À408C at 108Cmin^À1, held for 2–4 min at this tem-

perature, and were then heated to 1608C. Weight average molecular

weights (M_w) and polydispersity values (PDI) were determined by analy-

sis on a Viscotek HT-350 gel permeation chromatographer, equipped

with a low-angle/right-angle light scattering detector, a 4-capillary inline

viscometer, and a refractive index detector. The GPC utilized three Poly-

mer Labs PLgel 10 mm MIXED-B columns (300ꢃ7.5 mm) at a flow rate

of 1.0 mLmin^À1in 1,2,4-trichlorobenzene at either 145 or 1608C. To de-

termine the octene incorporation, a sample of each polymer solution

(140 mL) was deposited onto a silica wafer, heated at 1408C until the tri-

chlorobenzene had evaporated, and analyzed by using a Nicolet Nexus

670 FTIR with 7.1 version software equipped with an AutoPro auto sam-

pler.

Poly

the polyACHTUNGTRENNUNG

ACHTUNGTRENNUNG

calorimetry by using a Perkin–Elmer DSC-7 instrument equipped with a

CCA-7 cooling device and calibrated with the melting transition of

indium and n-heptane as references (156.1 and À90.618C, respectively).

The polyACHTUNGTRENNUNG(1-hexene) sample mass was 10 mg and aluminum pans were

used. Any thermal history in the polymers was eliminated by heating of

the specimen at a heating rate of 208Cmin^À1to 2008C, cooling with

208Cmin^À1to À1008C, and then recording the second scan from À100–

2008C.

NMR characterization: For ¹³C NMR spectroscopy about 100 mg of the

polymer was dissolved in C₂D₂Cl₄in a 10 mm tube. Hexamethyldisilox-

ane (HDMS) was used as internal chemical shift reference. The spectra

were recorded on a Bruker NMR AVANCE 400 spectrometer operating

at 100.58 MHz (¹³C) in the PFT mode at 1038C. The applied conditions

were the following: 10 mm probe, 908 pulse angle, 64 K data points, ac-

quisition time=5.56 s, relaxation delay=20 s, 3–4 K transient. Proton

Chem. Eur. J. 2012, 18, 671 – 687

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

685

G. Giambastiani et al.

Computational methodology: Density functional theory (DFT) calcula-

tions were performed by using the M06^[35]functional implemented within

the Gaussian 09 program.^[36]Structures were fully optimized by using this

functional. The theoretical study was carried out on the real system, with

the full pyridylamido ligand. The starting geometries were obtained from

the crystallographic atomic coordinates of 6 (L^Ph) and 7 (L^Fu) by replac-

ing the benzyl substituents with three NMe₂ligands and re-optimizing

the resulting structures. Core electrons of the Zr atom were described by

using the pseudo-potentials of Hay–Wadt,^[37]and their valence electrons

were expressed through a LANL2DZ basis.^[38]A 6-31G* basis set was

chosen for all the other atoms. Transition states were characterized by a

frequency calculation. Normal coordinate analysis on these stationary

points was also performed by intrinsic reaction coordinate (IRC) calcula-

tions^[39]in both directions to the correspondent minima. When the IRC

calculations failed to reach the minima, geometry optimizations from the

initial phase of the IRC path were performed. Finally, a continuum mod-

eling of the reaction medium was also included in the computational

treatment by using the same basis set. Bulk solvent effects (toluene, e=

2.38) were expressed through the polarizable continuum model (PCM-

UA0 solvation spheres).^[40]

[6] a) F. Guꢈrin, D. H. McConville, J. J. Vittal, G. A. P. Yap, Organome-

tallics 1998, 17, 5172; b) M. E. G. Skinner, Y. Li, P. Mountford,

Inorg. Chem. 2002 , 41 , 1110; c) Z. Ziniuk, I. Goldberg, M. Kol,

Inorg. Chem. Commun. 1999 , 2 , 549; d) M. E. G. Skinner, T. Tou-

pance, D. A. Cowhig, B. R. Tyrrell, P. Mountford, Organometallics

2005 , 24 , 5586; e) K. Nienkemper, G. Kehr, S. Kehr, R. Frçhlich, G.

Erker, J. Organomet. Chem. 2008 , 693 , 1572.

[7] G. Giambastiani, L. Luconi, R. L. Kuhlman, P. D. Hustad in Olefin

Upgrading Catalysis by Nitrogen-based Metal Complexes I (Eds.: G.

Giambastiani, J. Campora), Springer, London, 2011, pp. 197–282.

[8] K. P. Bryliakov, E. P. Talsi, N. V. Semikolenova, V. A. Zakharov, Or-

ganometallics 2009 , 28 , 322 5 –3 232.

[9] These data are in line with the scarce catalytic activity of related

{N,N^À} pyridylammido-Zr^IV

ACTHUNTGRENNG(U triamido)complexes containing bulky

À

alkyl substituents (not suitable for intramolecular C H bond activa-

tion) on the 6-position of the pyridine ring, see Ref. [6e].

[10] Polymerization conditions used: catalyst precursor 2/2’ or 3/3’

(20 mmol), PMAO (1000 equiv), C₂H₄pressure (15 bar), toluene

(50 mL), reaction time=30 min, 1700 rpm, productivity=7.2 kg

(high-density polyethylene (HDPE)) mol^À1h^À1at the highest.

[11] All ligands showed four equivalent methyl groups for the isopropyl

moiety: a doublet centered at d=1.27 (1a), 1.19 (1b), and 1.20 ppm

(1c).

Acknowledgements

[12] H. Tsurugi, Y. Matsuo, T. Yamagata, K. Mashima, Organometallics

2004 , 23 , 2797.

Thanks are due to the GDRI project (Homogeneous Catalysis for the

Sustainable Development) for support. A.R. and G.G. would also like to

thank Prof. Agustꢄ Lledꢅs (Universitat Autꢆnoma de Barcelona, UAB)

for fruitful discussions. The Centre de Serveis Cientꢄfics i Acadꢇmics de

Catalunya (CESCA, Barcelona) is also acknowledged for providing the

computational resources. Dr. Zhe Zhou of The Dow Chemical Company

is acknowledged for performing ¹³C polymer analysis.

[13] a) M. Bochmann, S. J. Lancaster, Organometallics 1993 , 12 , 633;

b) S. L. Latesky, A. K. McMullen, G. P. Niccolai, I. P. Rothwell, Or-

ganometallics 1985 , 4 , 902; c) R. F. Jordan, R. E. LaPointe, N. Baen-

ziger, G. D. Hinch, Organometallics 1990 , 9 , 1539; d) R. F. Jordan,

R. E. LaPointe, S. B. Chandrasekhar, S. F. Echols, R. Willett, J. Am.

Chem. Soc. 1987 , 109 , 4111.

À

[14] All dibenzyl complexes (6–9) show different Zr C_a/Zr C_ipsodistan-

ces at the “h²-coordinated” benzyl groups. The definition of “h²-co-

ordination” has been conventionally adopted (according to the liter-

ature, see Ref. [12],[13],[22]) to allow one to distinguish among the

[1] a) R. T. Boussie, G. M. Diamond, C. Goh, K. A. Hall, A. M. La-

Pointe, M. K. Leclerc, V. Murphy, J. A. W. Shoemaker, H. Turner,

R. K. Rosen, J. C. Stevens, F. Alfano, V. Busico, R. Cipullo, G. Talar-

ico, Angew. Chem. 2006 , 118 , 335 6 –3 361; Angew. Chem. Int. Ed.

2006 , 45 , 327 8 –3 283; b) G. J. Domski, E. B. Lobkovsky, G. W.

Coates, Macromolecules 2007 , 40 , 351 0 –3 513; c) V. Busico, R. Cipul-

lo, R. Pellecchia, L. Rongo, G. Talarico, A. Macchioni, C. Zuccaccia,

R. D. J. Froese, P. D. Hustad, Macromolecules 2009 , 42 , 436 9 –4 373;

d) T. R. Boussie, G. M. Diamond, C. Goh, K. A. Hall, A. M. La-

Pointe, M. K. Leclerc, C. Lund, V. Murphy, (Symyx Technologies,

Inc.) U.S. Pat. Appl. Publ. No. US 2006/0135722A1, 2006; e) T. R.

Boussie, G. M. Diamond, C. Goh, A. M. LaPointe, M. K. Leclerc, C.

Lund, V. Murphy, (Symyx Technologies, Inc.) U.S. Pat. 6750345,

2004.

[2] a) R. D. J. Froese, P. D. Hustad, R. L. Kuhlman, T. T. Wenzel, J. Am.

Chem. Soc. 2007 , 129 , 783 1 –7 840; b) C. Zuccaccia, A. Macchioni, V.

Busico, R. Cipullo, G. Talarico, F. Alfano, H. W. Boone, K. A. Frazi-

er, P. D. Hustad, J. C. Stevens, P. C. Vosejpka, K. A. Abboud, J. Am.

Chem. Soc. 2008 , 130 , 1035 4 –1 0368; c) C. Zuccaccia, V. Busico, R.

Cipullo, G. Talarico, R. D. J. Froese, P. C. Vosejpka, P. D. Hustad, A.

Macchioni, Organometallics 2009 , 28 , 544 5 –5 458; d) K. A. Frazier,

H. W. Boone, P. C. Vosejpka, J. C. Stevens, (DOW Global Techolo-

gies Inc.), U.S. Pat. Appl. Publ. No. US 2004/0220050A1, 2004;

e) G. J. Domski, J. B. Edson, I. Keresztes, E. B. Lobkovsky, G. W.

Coates, Chem. Commun. 2008 , 613 7 –6 139; f) J. R. Hagadorn, (EX-

XONMOBIL Chemical Patents Inc.), WO 2009/114209A1, 2009;

g) K. A. Frazier, R. D. Froese, Y. He, J. Klosin, C. N. Theriault, P. C.

Vosejpka, Z. Zhou, K. A. Abboud, Organometallics 2011 , 30 , 331 8 –

3329.

1

2

À

differently coordinated CH₂Ph groups in the solid state (h - vs. h -

À

coordination), although a “C C a-agostic interaction” should be

most properly invoked for the latter,

[15] Attempts to “freeze” this rapid interconversion by ¹H NMR spec-

troscopy at 193 K did not allow the authors to differentiate the two

bonding modes.

[16] B. Qian, W. J. Scanlon IV, M. R. Smith III, D. H. Motry, Organome-

tallics 1999 , 18 , 1693.

[17] a) M. Bouwkamp, D. van Leusen, A. Meetsma, B. Hessen, Organo-

metallics 1998 , 17 , 3645; b) P. Shao, R. A. L. Gendron, D. J. Berg,

G. W. Bushnell, Organometallics 2000 , 19 , 509; c) M. C. W. Chan,

S. C. F. Kui, J. M. Cole, G. J. McIntyre, S. Matsui, N. Zhu, K.-H.

Tam, Chem. Eur. J. 2006 , 12 , 2607; d) S. C. F. Kui, N. Zhu, M. C. W.

Chan, Angew. Chem. 2003 , 115 , 1666; Angew. Chem. Int. Ed. 2003 ,

42 , 1628.

[18] For similar distances [ꢂ] and angles [8] see also reference [12] and

[22b].

À

[19] Conversion of the neutral BCATHUNGNERTN(NUG C₆F₅)₃to the anionic BAHCUTNTRGE(NGNUN CH₂Ph)ACHTUNGTRENNUNG(C₆F₅)₃

is confirmed by the upfield shift of the ¹¹B resonance from >d=

59.8 to À12.9 (6a), À12.7 (7a), À12.4 (8a), and À13.0 ppm (9a).

[20] a) S. Beck, M. H. Prosenc, H. H. Brintzinger, J. Mol. Catal. A 1998,

128, 41–52; b) R. A. Lee, R. J. Lachicotte, G. C. Bazan, J. Am.

Chem. Soc. 1998 , 120 , 603 7 –6 046; c) S. Y. Desjardins, A. A. Way,

M. C. Murray, D. Adirim, M. C. Baird, Organometallics 1998 , 17 ,

238 2 –2 384; d) C. Pellecchia, A. Immirzi, A. Grassi, A. Zambelli,

Organometallics 1993 , 12 , 447 3 –4 478; e) L. Jia, X. Yang, C. L.

Stern, T. J. Marks, Organometallics 1997 , 16 , 84 2 –8 57; f) C. L. Bes-

wick, T. J. Marks, Organometallics 1999 , 18 , 2410.

[3] D. J. Arriola, E. M. Carnahan, P. D. Hustad, R. L. Kuhlman, T. T.

Wenzel, Science 2006 , 312 , 71 4 –7 19.

[4] L. Luconi, G. Giambastiani, A. Rossin, C. Bianchini, A. Lledꢅs,

Inorg. Chem. 2010 , 49 , 681 1 –6 813.

[21] a) A. D. Horton, J. de With, Chem. Commun. 1996 , 1375; b) A. D.

Horton, J. de With, A. J. van der Linden, H. van de Weg, Organome-

tallics 1996 , 15 , 2672.

[22] a) J. Saßmannshausen, A. K. Powell, C. E. Anson, S. Wocadlo, M.

Bochmann, J. Organomet. Chem. 1999 , 592 , 8 4 –9 4; b) A. Noor,

[5] A. A. C. Braga, G. Ujaque, F. Maseras, Organometallics 2006 , 25 ,

364 7 –3 658.

686

www.chemeurj.org

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2012, 18, 671 – 687

Zr^IV–Pyridylamido Polymerization Catalysts

FULL PAPER

W. P. Kretschmer, G. Glatz, A. Meetsma, R. Kempe, Eur. J. Inorg.

Chem. 2008 , 508 8 –5 098.

[35] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008 , 120 , 21 5 –2 41.

[36] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schle-

gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V.

Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X.

Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-

nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,

M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,

J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J.

Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J.

Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,

J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,

J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.

Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.

Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A.

Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, ꢊ.

Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian,

Inc., Wallingford CT, 2009.

[23] The ¹H NMR spectrum of the isolated modified ligand 10 is provid-

ed in the Supporting Information (Figure S30).

[24] The complete ¹³C{¹H} NMR assignment for complex 12b was ham-

pered by the partial signals overlap caused by the intense residual

solvent resonances.

[25] During this study the precatalysts 13 was employed under identical

experimental conditions to those previously described in referen-

ce [2g] except for the use of hydrogen to control the polymer molec-

ular weights.

[26] L. Luconi, D. M. Lyubov, C. Bianchini, A. Rossin, C. Faggi, G. K.

Fukin, A. V. Cherkasov, A. S. Shavyrin, A. A. Trifonov, G. Giambas-

tiani, Eur. J. Inorg. Chem. 2010 , 60 8 –6 20, and references therein.

[27] Ligand 1d containing the N-xylyl fragment was prepared according

to the procedure described for the analogue 1a (see Ref. [26], and

references therein), by using 2,6-dimethyl-aniline instead of 2,6-iPr-

aniline as reactant.

[28] CrysAlisCCD 1.171.31.2, CrysAlis171.NET, Oxford Diffraction Ltd,

2006.

[29] CrysAlis RED 1.171.31.2, CrysAlis171.NET, Oxford Diffraction Ltd,

2006.

[30] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giaco-

vazzo, A. Guagliardi, A. G. Moliterni, G. Polidori, R. Spagna, J.

Appl. Crystallogr. 1999 , 32 , 11 5 –1 19.

[37] a) P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985 , 82 , 27 0 –2 83; b) W. R.

Wadt, P. J. Hay, J. Chem. Phys. 1985 , 82 , 28 4 –2 98.

[38] a) A. Hçllwarth, M. Bçhme, S. Dapprich, A. W. Ehlers, A. Gobbi,

V. Jonas, K. F. Kçhler, R. Stegmann, A. Veldkamp, G. Frenking,

Chem. Phys. Lett. 1993 , 208 , 23 7 –2 40; b) A. W. Ehlers, M. Bçhme,

S. Dapprich, A. Gobbi, A. Hçllwarth, V. Jonas, K. F. Kçhler, R.

Stegmann, A. Veldkamp, G. Frenking, Chem. Phys. Lett. 1993 , 208 ,

11 1 –1 14.

[31] SHELXL97, G. M. Sheldrick, University of Gçttingen, Gçttingen

(Germany), 1997.

[32] L. J. Farrugia, J. Appl. Crystallogr. 1997 , 30 , 565.

[33] Z. Zhou, R. Kꢉemmerle, J. C. Stevens, D. Redwine, Y. He, X. Qiu,

R. Cong, J. Klosin, N. Montanez, G. Roof, J. Magn. Reson. 2009 ,

200 , 32 8 –3 31.

[39] K. Fukui, Acc. Chem. Res. 1981 , 14 , 36 3 –3 68.

[40] a) S. Miertus, E. Scrocco, J. Tomasi, J. Chem. Phys. 1981, 55, 117–

129; b) V. Barone, M. Cossi, J. Tomasi, J. Chem. Phys. 1997 , 107 ,

321 0 –3 221.

[34] Z. Zhou, R. Kꢉemmerle, X. Qiu, D. Redwine, R. Cong, A. Taha, D.

Baugh, B. Winniford, J. Magn. Reson. 2007 , 187 , 22 5 –2 33.

Received: July 18, 2011

Published online: December 6, 2011

Chem. Eur. J. 2012, 18, 671 – 687

ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

687

Article Doi

Facing unexpected reactivity paths with ZrIV-pyridylamido polymerization catalysts

DOI: 10.1002/chem.201102194

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.201102194

Full text of DOI:10.1002/chem.201102194

Products guided by the article

R&D Labs maybe for 1362250-82-5

Relevant to this article

Hot Product