Formation of trivalent zirconocene complexes from ansa-zirconocene-based olefin-polymerization precatalysts: An EPR- and NMR-spectroscopic study

Article abstract of DOI:10.1021/ja403170u

Reduction of Zr(IV) metallocenium cations with sodium amalgam (NaHg) produces EPR signals assignable to Zr(III) metallocene complexes. The chloro-bridged heterodinuclear ansa-zirconocenium cation [(SBI)Zr(μ-Cl) ₂AlMe₂]⁺ (S

Full text of DOI:10.1021/ja403170u

Subscriber access provided by George Mason University Libraries & VIVA (Virtual Library of Virginia)
Article
Formation of Trivalent Zirconocene Complexes from ansa-Zirconocene-Based
Olefin-Polymerization Precatalysts - an EPR- and NMR-Spectroscopic Study.
Taylor N Lenton, John Edward Bercaw, Valentina Nikolaevna Panchenko, Vladimir Alexandrovich
Zakharov, Dmitrii E. Babushkin, Igor E. Soshnikov, Evgenii P. Talsi, and Hans H. Brintzinger
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/ja403170u • Publication Date (Web): 10 Jun 2013
Downloaded from http://pubs.acs.org on June 11, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 32
Journal of the American Chemical Society
1
2
3
4
Formation of Trivalent Zirconocene Complexes from ansaꢀZirconoceneꢀBased
OlefinꢀPolymerization Precatalysts ꢀ an EPRꢀ and NMRꢀSpectroscopic Study.
5
6
7
8
9
Taylor N. Lenton,Η John E. Bercaw,*,Η Valentina N. Panchenko,Ι Vladimir A. Zakharov,Ι Dmitrii
E. Babushkin,*,Ι Igor E. Soshnikov,Ι Evgenii P. Talsi,Ι Hans H. Brintzinger*,§
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Η Arnold and Mabel Beckman Laboratories of Chemical Synthesis, California Institute of
Technology, Pasadena, California 91125, USA.
Ι Boreskov Institute of Catalysis, Russian Academy of Science, Siberian Branch, Pr. Lavrentieva
5, 630090 Novosibirsk, Russian Federation,
§ Fachbereich Chemie, Universität Konstanz, Dꢀ78457 Konstanz, Germany.
eꢀmail: Bercaw@caltech.edu; Dimi@catalysis.ru, hans.brintzinger@uniꢀkonstanz.de
Abstract. Reduction of Zr(IV) metallocenium cations with sodium amalgam (NaHg) produces
EPR signals assignable to Zr(III) metallocene complexes. The chloroꢀbridged heterodinuclear
ansaꢀzirconocenium cation [((SBI))Zr(ꢀꢀCl)2AlMe2]+ (SBI = racꢀdimethylsilylbis(1ꢀindenyl)),
−
present in toluene solution as its B(C6F5)4 salt, thus gives rise to an EPR signal assignable to the
complex (SBI)ZrIII(ꢀꢀCl)2AlMe2, while (SBI)ZrIIIꢀMe and (SBI)ZrIII(ꢀꢀH)2AliBu2 are formed by
−
−
reduction of [(SBI)Zr(ꢀꢀMe)2AlMe2]+ B(C6F5)4 and [(SBI)Zr(ꢀꢀH)3(AliBu2)2]+ B(C6F5)4 ,
respectively. These products can also be accessed, along with (SBI)ZrIIIꢀiBu and [(SBI)ZrIII]+
−
AlR4 , when (SBI)ZrMe2 is allowed to react with HAliBu2, eliminating isobutane en route to the
Zr(III) complex. Further studies concern interconversion reactions between these and other
(SBI)Zr(III) complexes and reaction mechanisms involved in their formation.
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 32
1
2
3
4
5
Introduction
6
7
8
9
Cationic, tetravalent group 4 metallocene complexes have been well established as active
α-olefin polymerization catalysts.1 Frequently, metallocene precatalysts are activated with
methylaluminoxane (MAO), producing complex reaction mixtures,2 in which metal species
reduced by one electron to trivalent compounds can sometimes also be observed. In situ
formation of Ti(III) species, which are apparently inactive, has been documented for titanoceneꢀ
based polymerization catalysts.3 While Ti(III) halfꢀsandwich complexes have been reported to
catalyze the polymerization of styrenes,4 the role of various Ti oxidation states in these systems
appears still controversial.5
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Much less attention has been directed at the formation of Zr(III) species in the course of
zirconoceneꢀcatalyzed olefinꢀpolymerizations. Despite some clear EPRꢀspectroscopic evidence
for the occurrence of Zr(III) species in zirconoceneꢀbased polymerization catalysts,6,7 and
substantial previous work on trivalent zirconocene complexes in general,8 information about the
nature and the reaction tendencies of Zr(III) complexes, which arise under polymerization
conditions, is still rather limited. Identifying these reduced species and determining the
mechanisms of their formation from Zr(IV) complexes present in active catalyst systems are first
steps towards elucidating whether they exist as temporary resting states or are permanently
removed from the catalytic cycle.
Additionally, we are interested in the role that aluminum hydrides play in polymerization
systems,9 as they are frequently introduced by loss of isobutene from triisobutylaluminum
(TIBA), which is added to increase the overall activity of the catalytic system.10 Previously it
was shown that [(SBI)Zr(ꢀꢀMe)2AlMe2]+ (SBI = racꢀdimethylsilylbis(1ꢀindenyl)), a highly
reactive intermediate in zirconoceneꢀbased polymerization catalysts,11 is converted to
2
ACS Paragon Plus Environment

Page 3 of 32
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
[(SBI)Zr(ꢀꢀH)3(AliBu2)2]+ , itself a polymerization precatalyst,12 when even small amounts of
HAliBu2 are present in solution.13 Intriguingly, higher concentrations of paramagnetic
compounds are observed by EPRꢀspectroscopy upon activation with MAO/TIBA mixtures than
with MAO alone, which might result from Zr(III) species being formed by the introduction of
aluminum hydride compounds into these reaction systems.7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In light of the above information, we have sought to generate relevant zirconocene
complexes of oxidation state Zr(III) by alternative routes, in order to provide spectral and
reactivity data for comparison with zirconocene reaction systems during polymerization
catalysis. In this report, we present results of NMRꢀ and EPRꢀspectroscopic studies on reduction
reactions of several (SBI)Zr(IV) complexes with sodium amalgam and with diisobutylaluminum
hydride, which give access to a variety of trivalent zirconocene species, and on interconversion
reactions of some of these Zr(III) complexes. This should provide a more complete reference for
studying Zr(III) products, that might arise under catalytic conditions, and offer first insights into
possible mechanisms of their formation.
Results and Discussion
Due to the complex nature of these reaction systems and the limited structural information
that can be obtained by EPR spectroscopy alone, a combination of data is used here to identify
and characterize Zr(III) species present in these reaction media. First, we consider the products
that arise from the sodium amalgam reduction of various zirconocenium cations present under
catalytic reaction conditions. Further studies concern the formation of Zr(III) compounds that
might arise in catalytic systems upon interaction with dialkyl aluminum hydride compounds.
Finally, we investigate the reactivity patterns of these species upon addition of aluminumꢀ
chloride, aluminumꢀalkyl and aluminumꢀhydride compounds.
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 32
1
2
3
4
5
6
7
8
9
1. Reduction with Sodium Amalgam. While sodium amalgam (NaHg) has previously been a
useful reductant for various metallocene complexes,14ꢀ16 no change in the NMR spectra of C6D6
solutions of (SBI)ZrMe2 (1) was observed upon shaking with NaHg, containing 1% (wt./wt.) Na
for several hours. Prolonged stirring of a C6D6 solution of (SBI)ZrCl2 (2) with NaHg for several
days produced a color change to dark brown, but we were not able to isolate or identify any
products from this reaction system. Several cationic zirconocenium complexes, on the other
hand, proved to be more prone than these neutral species to give identifiable reduction products
upon treatment with sodium amalgam.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A particularly straightforward reduction is observed from the dichloroꢀbridged cation
ꢀ
[(SBI)Zr(ꢀꢀCl)2AlMe2]+ (3), present in solution as its B(C6F5)4 salt.17 Reaction of NaHg with a
toluene solution of 3 gives rise to an EPR spectrum with a signal at g = 1.958 (Figure 1). The
shape of this signal – a midꢀpoint inflection with reduced slope – indicates some unresolved
hyperfine splitting. As in a similar case reported by Lyakin et al.,7 a model spectrum calculated
for interaction of the unpaired electron with one 27Al and one 91Zr nucleus is in nearꢀperfect
agreement with the observed spectrum and consistent with an assigned structure of (SBI)ZrIII(ꢀꢀ
Cl)2AlMe2 (4). Crystallographic data which further confirms this assignment will be discussed in
a later section, as 4 was generated using alternate methods to avoid the presence of (rather
insoluble) [Na][B(C6F5)4] in the reaction system.
The doublyꢀintegrated intensity of this signal accounts, within accuracy limits, for the total Zr
content of the sample. In the absence of air, this signal is stable for days. Additionally, in 1H
NMR spectra of (SBI)ZrIII(ꢁꢀCl)2AlMe2, obtained by reduction of [(SBI)Zr(ꢁꢀCl)2AlMe2]+ with
the oneꢀelectron reductant cobaltocene,16 two paramagnetically broadened signals with very
4
ACS Paragon Plus Environment

Page 5 of 32
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
short relaxation times of ca. 4 ms appear, at 0.68 and –1.4 ppm, with intensities practically equal
to those of the (CH3)2Si and Al(CH3)2 signals seen for the initial cation [(SBI)Zr(ꢁꢀCl)2AlMe2]+
before reduction (Supporting Information). These observations indicate that reaction of NaHg
with the cation 3 furnishes the Zr(III) complex (SBI)ZrIII(ꢀꢀCl)2AlMe2 (4) according to eq. 1 in
practically quantitative yield.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15000
10000
x20
g = 1.958
5000
0
-5000
-10000
-15000
x20
3350
3400
3450
3500
B/G
Figure 1. EPR spectrum (Xꢀband, 25 °C) of 4 the ZrIII complex obtained by NaHgꢀinduced
reduction of a 3.5 mM solution of [(SBI)Zr(ꢀꢀCl)2AlMe2]+ B(C6F5)4 in C6D6 (solid line), and
−
EPR spectrum calculated for interaction of the unpaired electron with one 27Al and one 91Zr
nucleus (100% 27Al, I = 5/2, a(27Al) = 3.4 G; 11% 91Zr, I = 5/2, a(91Zr) = 18.6 G, line width 5 G,
dashed line).
-
B(C6F5)4
Cl
Cl
Cl
Cl
NaHg
[Na][B(C6F5)4]
AlMe2
+
(1)
Zr III
Si
Zr
AlMe2
Si
+
3
4
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 32
1
2
3
4
5
6
7
8
9
Quite a different course is taken by the AlMe3ꢀcomplexed methyl zirconocenium cation
[(SBI)Zr(ꢁꢀMe)2AlMe2]+ (5), a prominent component of (SBI)Zrꢀbased olefinꢀpolymerization
catalysts.9 When a C6D6 solution of this heterobinuclear cation, prepared by reaction of 1 equiv.
of trityl tetrakis(pentafluorophenyl)borate ([Ph3C][B(C6F5)4]) with 1 in the presence of AlMe3, is
shaken with sodium amalgam for ca. 1 h, it produces the EPR spectrum shown in Figure 2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10000
5000
g = 1.971
0
-5000
-10000
3350
3400
3450
B/G
Figure 2. EPR spectrum (Xꢀband, 25 °C) obtained after reduction of a 1 mM solution of
[(SBI)Zr(ꢁꢀMe)2AlMe2]+ B(C6F5)4 in C6D6 with 1%ꢀNaHg immediately (solid line) and ca. 3 h
ꢀ
after the end of the 1ꢀh reaction period (broken line).
The EPR spectrum contains as its main component a signal centered at g = 1.971. Based on
the absence of any 27Alꢀassociated hyperfine structure from the main signal in Figure 2, we can
tentatively assume that this signal is due to the simple methyl complex (SBI)ZrIIIꢀMe (6, eq. 2),19
rather than to the conceivable direct electronꢀuptake product (SBI)ZrIII(ꢁꢀMe)2AlMe2.
6
ACS Paragon Plus Environment

Page 7 of 32
Journal of the American Chemical Society
1
2
3
4
-
B(C6F5)4
Me
5
6
7
NaHg
III
(2)
AlMe2
Zr
+ AlMe3 + [Na][B(C6F5)4]
Me
Si
Zr
Si
+
Me
8
9
6
5
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The signal shown in Figure 2 accounts, however, only for a rather small fraction (ca. 10ꢀ
20%) of the total zirconocene concentration. In order to elucidate the course of the reduction
reaction in more detail, we followed the EPR and 1H NMR spectra of the reaction mixture in
parallel. The 1H NMR signals of the cation [(SBI)Zr(ꢀꢀMe)2AlMe2]+ (3), initially the only signals
discernible, completely disappear during the oneꢀhour reduction period. At the same time, the
appearance of the characteristic 1H NMR signals of the dimethyl complex (SBI)ZrMe2 (1) shows
that this complex is formed in substantial concentration – about one half of the initial
concentration of the cation 3 – during the reduction reaction (Supporting Information).20
Reactions which might lead to the rather unexpected formation of 1 – itself not a reduction
product – in the course of the reduction reaction will be discussed in a later section.
Finally, we have also studied the NaHgꢀinduced reduction of the hydrideꢀbridged cation
[(SBI)Zr(ꢀꢀH)3(AlR2)2]+ (7, R = Me or iBu), another possible component of (SBI)Zrꢀbased
catalyst systems.10 When a C6D6 solution of this cation, prepared by reaction of 1 with 1 equiv.
of [Ph3C][B(C6F5)4] in the presence of 4 equiv. of HAliBu2,13 is shaken for 1 hr with sodium
amalgam, it gives rise to the EPR spectrum shown in Figure 3. Apart from a minor, as yet not
assignable component in its central part, the spectrum shows a main signal at g = 1.970. The
shape of the main signal is indicative of a hyperfine splitting by a 27Al nucleus, which is about
twice as strong as that discussed above for the Clꢀbridged species 4 (Figure 1).21 Model spectra
calculated with a hyperfine interaction constant of a(27Al) = 7.5 G closely approach the main
signal shown in Figure 3, provided that the hyperfine splitting in its central part is not resolved,
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 32
1
2
3
4
5
6
7
8
9
either due to an unusually large line width of 10ꢀ12 G, or alternatively, by further unresolved
hyperfine interaction with two 1H nuclei, each with a(1H) ≈ 4 G and a more reasonable line width
of 7ꢀ8 G (Supporting Information). We can thus tentatively assume that the main signal shown in
Figure 3 is due to a dihydrideꢀbridged complex (SBI)ZrIII(ꢀꢀH)2AlR2 (8 eq. 3).22,23The intensity
of the signal indicates that the yield of complex 8 is ca. 30%.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. EPR spectrum (Xꢀband, 25 °C) obtained from a 1 mM C6D6 solution of [(SBI)Zr(ꢀꢀ
H)3(AlR2)2]+ after reduction with 1%ꢀNaHg (solid line) and EPR spectrum calculated for
interaction of the unpaired electron with one 27Al nucleus (100% 27Al, I = 5/2, a(27Al) = 7.6 G)
and two 1H nuclei (100% 1H, I = ½, a(1H) = 4 G, line width 7 G, broken line), assigned to
(SBI)ZrIII(ꢀꢀH)2AlR2 (8).
While the value of a(1H) ≈ 4 G in 8 might appear rather small for hydride ligands in direct
contact with a Zr(III) center,24ꢀ27 this assignment has precedence in the dihydrideꢀbridged
complex (C5H5)2ZrIII(ꢀꢀH)2Al(Me)(C6H2ꢀ2,4,6ꢀtBu), for which an EPR signal with a(27Al) = 8.7
8
ACS Paragon Plus Environment

Page 9 of 32
Journal of the American Chemical Society
1
2
G and a(1H) = 3.9 G has been reported by Wehmschulte and Power.28 It must be taken into
account that the ZrIII(ꢀꢀH)2Al bonding situation differs distinctly from that of a terminal ZrIIIꢀH
unit. Since Al(III) is likely to have a higher electronegativity than Zr(III), the complex
(SBI)ZrIII(ꢀꢀH)2AlR2 is likely to display some character of an innerꢀsphere ion pair
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
−
(SBI)ZrIII+…H2AlR2 . The diminished ZrꢀH covalency can thus be expected to result in reduced
values of a(1H);29 hence, a value of a(1H) ≈ 4 G would appear compatible with this assignment.
2. Reduction of (SBI)ZrMe2 by HAliBu2. While searching for causes for the incomplete
progress of the reduction to 8, described above, we noticed that EPR signals similar to that
shown in Figure 3 are observed in solutions containing the dimethyl complex (SBI)ZrMe2 (1)
and HAliBu2 even before the addition of [Ph3C][B(C6F5)4] and/or NaHg. More detailed studies
on this reaction are described below, along with subsequent reactivity studies aimed at further
characterization.
A 3.5 mM solution of 1 in C6D6, which contains a 10ꢀfold excess of HAliBu2, slowly
develops the EPR spectra shown in Figure 4. The spectrum obtained after about one day is
practically identical to that produced by reduction of the cation [(SBI)Zr(ꢀꢀH)3(AlR2)2]+ (7) with
NaHg (Figure 3) and is assigned as the same hydrideꢀbridged complex (SBI)ZrIII(ꢀꢀH)2AlR2 (8).
The intensity of this signal indicates that species 8 amounts to ca. 20 % of the total Zr
concentration at this time and does not grow any further at longer reaction times. Parallel 1H
NMR measurements reveal that the reaction is accompanied by formation of approximately one
equiv. of isobutane per unit of 1,30 together with traces of methane. Under these conditions, 8
thus appears to arise mainly by reductive elimination of isobutane (eq. 4).
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 32
1
2
3
4
5
6
7
8
9
20 h
93 h
10000
5000
0
g = 1.970
3.5 h
.25 h
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
-5000
-10000
3340
3360
3380
3400
B/G
3420
3440
3460
Figure 4. EPR spectrum (Xꢀband, 25 °C) of a 3.5 mM solution of (SBI)ZrMe2 in C6D6, which
contains a 10ꢀfold excess of HAliBu2 at various reaction times (solid lines) and, for comparison,
the spectrum assigned to 8, taken from Figure 3 (broken line).
In a related experiment, a C6D6 solution of 1 containing only 2 equiv. of HAliBu2, i.e. an
amount less than that required by the stoichiometry of eq. 4, developed the EPR spectra shown in
Figure 5. The spectra still show the signal of 8 but are now dominated by a central signal, also at
g = 1.970. This signal shows a partly resolved hyperfine splitting due to interaction of the
unpaired electron with two 1H nuclei, each with a(1H) = 5.7 G (Supporting Information). Based
on its g value, which is similar to that of the species (SBI)ZrIIIꢀMe discussed above, and on its
a(1H) value, which is within the range of values previously reported for alkyl zirconocene
10
ACS Paragon Plus Environment

Page 11 of 32
Journal of the American Chemical Society
1
2
complexes of oxidation stage Zr(III),7,31 we assign this signal to the isobutyl complex
(SBI)ZrIIIꢀiBu (9). After a reaction period of several hours, 8 and 9 account together for about
90% of the total zirconocene concentration in this reaction mixture.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7 h
20000
100 min
50 min
10000
30 min
15 min
0
g = 1.970
-10000
-20000
3340
3360
3380
3400
B/G
3420
3440
3460
Figure 5. EPR spectra (Xꢀband, 25 °C) of a 3.5 mM solution of (SBI)ZrMe2 in C6D6, which
contains 2 equiv. of HAliBu2, at various reaction times.
Parallel 1H NMR measurements of the reaction mixture reveal that the appearance of the
EPR signals shown in Figure 4 is accompanied again by substantial evolution of isobutane and
methane – in this case in comparable amounts. Formation of isobutyl complex 9 by reductive
alkane elimination can thus be described by eq. 5
Me
III
HAliBu2
Si
Zr
H R
iBu
+
AlR3
+
2
+
(5)
Si
Zr
2
Me
R = Me or iBu
9
1
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 32
1
2
Formation of (SBI)ZrIIIꢀiBu as the reduction product is further supported by the observation
that addition of dimethylphenylphosphine to the product mixture generates, at the expense of the
signals shown in Figure 5, a somewhat broadened EPR doublet with a hyperfine splitting of 10.6
G (Supporting Information), as would be expected for a phosphine complex of composition
(SBI)Zr(iBu)PMe2Ph, formed according to eq. 6.32
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
While we were not able to corroborate the structures of complexes 8 and 9 by crystal
structure determinations, our assignment of the “outerꢀwings” signal shown in Figure 4 to
(SBI)ZrIII(ꢀꢀH)2AliBu2 (8) and of the central signal to (SBI)ZrIIIꢀiBu (9) is supported by the
following observation: Addition of HAliBu2 to a mixture of the two causes the signal assigned to
8 to increase relative to that of 9, while the opposite is observed upon addition of AliBu3, i.e. an
equilibrium reaction as in eq. 7 appears to interconvert 8 and 9 (Supporting Information).
Formation of the dichloroꢀbridged complex 4 from 8 and 9 was studied by reacting a solution
containing both of these species with increasing increments of ClAlMe2. The EPR spectra
obtained in this manner (Figure 6) clearly reveal a twoꢀstep reaction: addition of only 1 equiv. of
12
ACS Paragon Plus Environment

Page 13 of 32
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
ClAlMe2 causes the practically complete conversion of all (SBI)Zr(III) species to a new complex
with g = 1.968. Subsequent additions of ClAlMe2 gradually convert this new complex to 4.
“Isosbestic” points at fields of 3416 and 3436 G indicate that only two (SBI)Zr(III) species are
present under these conditions.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20000
[ClAlMe2]/[Zr]tot
1
2
15000
10000
5000
5
10
20
50
g = 1.958
0
-5000
-10000
-15000
-20000
g = 1.970
[ClAlMe2]/[Zr]tot
1
2
5
10
20
50
g = 1.968
3350
3400
3450
3500
B/G
Figure 6. EPR spectra (Xꢀband, 25 °C) of a 3.5 mM solution of (SBI)ZrMe2 (1) in C6D6, which
had reacted with 2 equiv. of HAliBu2 for 16 h (broken line), after stepwise addition of 1ꢀ50
equiv. of ClAliBu2 (solid lines, corrected for dilution).
Based on these observations and on the lack of obvious hyperfine interaction, the new signal
at g = 1.968 might prima facie be assigned to the monochloro complex (SBI)ZrIIIꢀCl. Since this
presumably Lewisꢀbasic species would arise, however, in the presence of highly Lewisꢀacidic
trialkyl aluminum compounds, it should most probably be designated as its AlR3 adduct
(SBI)ZrIIIꢀClAlR3 (10). The presumably increased Zr...Al bond distance in this type of ion pair
would diminish the resolution of any 27Al hyperfine interaction beyond detection. This complex
appears to be formed in practically quantitative yield under these conditions (eq. 8), and is then
converted to 4 by excess ClAliBu2 in an equilibrium reaction (eq. 9).
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 32
1
2
3
4
(SBI)ZrIII-iBu
(8)
ClAlMe2
(SBI)ZrIII-ClAlR3
+
9
10
5
6
7
8
(SBI)ZrIII-ClAlR3
(SBI)ZrIII(ꢀ-Cl)2AlR2
(9)
ClAlMe2
+
+ AlR3
9
4
10
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
R = Me, iBu
In accord with eq. 9, addition of excess AlMe3 to a solution of species 4 decreases its EPR
signal at g = 1.958 with concomitant appearance of the signal of Alꢀstabilized Cl species 10 at g
= 1.968 (Supporting Information). The observation that excess trialkyl aluminum causes
conversion of 4 to the monochloride complex supports our tentative proposal that the latter is
present in form of a trialkyl aluminum adduct.
Addition of increasing proportions of HAliBu2, on the other hand, to a solution containing 4
converts most of the latter to 8, but superimposed on the signal of this species at g = 1.970
remains the signal of 10 at g = 1.968 (Supporting Information). While the equilibrium
represented in eq. 10 thus appears to be movable to either side, the Cl ligand in (SBI)ZrIIIꢀClAlR3
obviously resists exchange with an Alꢀbound iBu group.
H
Cl
+ 2 HAliBu2
AlMe2
+ 2 ClAlR2
AlR2
(10)
III
Si
Zr
III
Si
Zr
Cl
H
8
4
3. Crystal and molecular structure of (SBI)ZrIII(ꢀ-Cl)2AlR2 (4).Taking advantage of these
equilibria (eqs. 9 and 10), 4 can be obtained as a solution in pentane by first allowing 1 to react
overnight with 2 equiv. of HAliBu2, and then treating the resulting mixture of 8 and 9 with excess
ClAlMe2. When such a solution was cooled slowly to –20°C, it produced crystals of Xꢀray
quality (Supporting Information). Structure determination of these crystals revealed a P2/c space
14
ACS Paragon Plus Environment

Page 15 of 32
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
group and a unit cell with ß = 92.126(2) and four molecules of 4 with both enantiomers in equal
proportions (Figure 7). The final anisotropic fullꢀmatrix leastꢀsquares refinement on F2 yielded a
molecular structure for 4 in agreement with the spectral data discussed above (Figure 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. Molecular structure of (SBI)ZrIII(ꢀꢀCl)2Al(CH3)2 (4) with thermal ellipsoids at the 50%
probability level. H atoms omitted for clarity. For selected bond lengths and angles see Table 1.
The coordination geometry of the Zr center of 4 can be compared to that of (SBI)ZrCl2 (2),33
as well as to those of several other zirconocene chloride complexes of oxidation state Zr(III) –
viz. [(Me3SiꢀC5H4)2ZrIII(ꢁꢀCl)]2,34 [(1,3ꢀ(Me3Si)2ꢀC5H3)2ZrIII(ꢁ ꢀCl)]2,35 and [(C5H5)ZrIII(ꢁ ꢀ
Cl)]2ꢀ ꢁ ꢀ(C5H4ꢀC5H4) 36 (Table 1). The distance of the C5ꢀring centroid from the Zr center is
found – somewhat unexpectedly – to be smaller by ca. 0.1 Å for the Zr(III) complex 4 than for
its Zr(IV) congener 2. This observation and the substantially longer ZrꢀCl distances in 4 than in 2
conform with the notion that the extra electron resides in a molecular orbital of 4 which is antiꢀ
bonding with regard to the Cl ligands but mainly nonꢀbonding with respect to the ring ligands.37
Additionally, ZrIIIꢀCl distances are found to be significantly longer for 4 than for the dimeric
Zr(III) chloride compounds listed in Table 1. The AlꢀCl distances, on the other hand are shorter
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 32
1
2
3
4
5
6
7
8
9
in 4 than those reported for a number of dimeric [R2Al(ꢀꢀCl)]2 compounds (2.32 ꢀ 2.35 Å) and
close to those in the electronꢀpoor perfluorinated compound [(C6F5)2AlCl]2 (Table 2).38ꢀ41 These
deviations of ZrꢀCl and AlꢀCl bond distances in the heterobinuclear species 4 from values
reported for related homobinuclear species – as well as its conspicuously diminished ClꢀZrꢀCl
and widened ClꢀAlꢀCl angles – can all be considered to suggest that (SBI)ZrIII(ꢀꢀCl)2Al(CH3)2 is
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
−
indeed approaching the bonding situation of a contact ion pair (SBI)ZrIII+…Cl2Al(CH3)2 .
ZrꢀCl
ClꢀZrꢀCl
77.07(3)
ZrꢀCT
CTꢀZrꢀCT ref.
(SBI)ZrIII(ꢁꢀCl)2AlMe2 (4)
2.650(1)
2.624(1)
2.431(<1)
2.1908
2.1836
2.293
130.70
this
work
33
(SBI)ZrCl2 (2)
98.76(1)
119.04
[(Me3SiꢀC5H4)2ZrIII(ꢀꢀCl)]2
[(1,3ꢀ(Me3Si)2ꢀC5H3)2ZrIII(ꢀꢀCl)]2
[(C5H5)ZrIII(ꢀꢀCl)]2ꢀ(ꢀꢀC5H4ꢀC5H4)
2.560±4a
2.602±6a
2.578
93.34±0.10b
82.77±1.10b
100
34
35
36
Table 1. Selected bond lengths (Å) and angles (deg) for a series of zirconocene compounds;
a) averages of multiple ZrꢀCl bond lengths; b) averages of multiple ClꢀZrꢀCl bond angles.
AlꢀCl
ClꢀAlꢀCl Al ꢀ R
Ref.
(SBI)Zr(ꢁꢀCl)2AlMe2 (4) 2.253(1)
93.37(5)
87.2(1)
85.2(1)
1.941(4) this
1.946(4) work
1.966(6) 38
1.982(9)
1.972(4) 39
1.966(5)
2.263(1)
[tBu2Al(ꢁꢀCl)]2
[Mes2Al(ꢁꢀCl)]2
2.316(3)
2.324(3)
2.346(2)
2.315(2)
[AlMe(CSiR3)(ꢁꢀCl)]2
[(C6F5)2AlCl]2
2.3216(12) 86.53(4)
1.938(4) 40
2.244(1)
2.273(1)
90.60
90.75
1.953(3) 41
1.944(3)
Table 2. Selected bond lengths (Å) and angles (deg) for a series of aluminum compounds.
Attempts to crystallize in a similar manner also other (SBI)Zr(III) complexes, in particular
complexes 8 and 9, were not successful, most likely due to the lower polarity of these compounds.
16
ACS Paragon Plus Environment

Page 17 of 32
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
4. Formation of a Cationic Zirconium(III) complex. The Zr(III) compounds discussed above
seem to be relatively stable, but isobutyl compound 9 undergoes a slow further transformation to
give a signal centered at g = 1.983. When the reaction between 1 and 2 equiv. of HAliBu2 (Figure
5), is observed for longer reaction times, it yields the spectral changes shown in Figure 8. Over
the course of several days, the signals of 8 and 9 both diminish to ca. 10 percent of their
previous size, while a new signal grows in which is centered at g = 1.984. After about one week,
little if any further changes occur. Some loss of overall signal intensity appears to occur over this
period of time.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
g = 1.983
20000
10000
0
7 h
2 d
6 d
9 d
2 d
-10000
-20000
6 d
9 d
g = 1.970
3340
3360
3380
3400
B/G
3420
3440
3460
Figure 8. EPR spectra (Xꢀband, 25 °C) of a 3.5 mM solution of (SBI)ZrMe2 in C6D6, which
contains 2 equiv. of HAliBu2 (cf. Figure 5), measured after increasing reaction times.
Spectral changes of the same kind occur to a more complete extent, if a 3.5 mM solution of 1
is allowed to react with only 1 equiv. of HAliBu2. In this case, we observe, after ca. 6 days, only
the signal centered at g = 1.983 (Figure 9).42 This rather narrow signal shows a hyperfine
splitting due to the 91Zr nucleus with a(91Zr) = 17 G, but no signs of other hyperfine interactions.
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 32
1
2
3
4
5
6
7
8
9
g = 1.983
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 9. EPR spectrum (Xꢀband, 25 °C) of 11, generated from a 3.5 mM solution of 1 in C6D6,
which contains 1 equiv. of HAliBu2, after a reaction time of ca. 6 days (solid line) and spectrum
calculated for a hyperfine splitting due to a 91Zr nucleus (11% 91Zr, I = 5/2, a(91Zr) = 17 G).
Signals with similar gꢀvalues and hyperfine splitting, previously observed after reduction of
(C5H5)2ZrCl2 with BuLi and in MAOꢀactivated toluene solutions of (2ꢀPhꢀind)2ZrCl2 and
(SBI)ZrCl2, were assigned to the cations [Cp2ZrIII]+, [(2ꢀPhꢀind)2ZrIII]+ and [(SBI)ZrIII]+,
respectively, based on their lack of ligand hyperfine structure and their lowꢀfield position.7,43
We follow this proposal in tentatively assigning the signal at g = 1.983, shown in Figure 7, to the
“free” cation [(SBI)ZrIII]+ (11, eq. 11) .
18
ACS Paragon Plus Environment

Page 19 of 32
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
This assignment raises the question of how such a cationic species could arise in a reaction
system free of any of the typical cationization reagents, such as MAO or trityl cation, and which
kind of anion would be associated with it. Inspection of eq. 11 suggests that the “spontaneous
ionization” represented there is likely to be exergonic, since an aluminum trialkyl is most likely a
stronger Lewis acid than the [(SBI)ZrIII]+ cation and thus prone to abstract an iBu− ligand from
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
−
the latter. The resulting ion pair [(SBI)ZrIII]+ AlR4 would then differ from innerꢀsphere ion pairs
−
−
such as (SBI)ZrIII+…Cl2AlR2 and (SBI)ZrIII+…H2AlR2 , discussed above as representations of the
chlorideꢀ and hydrideꢀbridged complexes 4 and 8. Apparently, the alkyl periphery of an anion
−
AlR4 has a much lower tendency to coordinate to a Zr(III) center than the chloride or hydride
units of a Cl2AlR2 or an H2AlR2 anion, such that the cation [(SBI)ZrIII]+ associates with the
−
−
−
anion AlR4 predominantly in an outerꢀsphere ion pair.
Consistent with this proposal is the observation that addition of HAliBu2 (50 equiv.) to a
solution containing the ion pair 11 causes the signal at g = 1.983 to disappear, while regenerating
the signal pattern of the dihydrideꢀbridged complex 8 (Supporting Information). Apparently,
scrambling of alkyl and hydride units between neutral and anionic aluminum alkyls leads to
−
−
formation of H2AliBu2 , which upon association with [(SBI)ZrIII]+ gives (SBI)ZrIII+…H2AlR2
≡
(SBI)ZrIII(ꢀꢀH)2AlR2.
This notion would also provide an explanation for the rather unexpected formation of
(SBI)ZrMe2 (1) – itself not a reduction product – in the course of the NaHgꢀinduced reduction of
the dimethylꢀbridged cation 5, noted in section 1. Formation of 1 is not explicable by a
disproportionation of (SBI)ZrIIIꢀMe (6), one of the reduction products, to 1 and some (SBI)Zr
complex of oxidation stage Zr(II), since 6 is found to be quite stable when prepared, in analogy
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 32
1
2
to the reaction in eq. 7, by reaction of 1 with 5 equiv. of HAlMe2 (instead of HAliBu2) followed
by excess AlMe3 (Supporting Information).
3
4
5
6
7
8
9
If, however, the cation [(SBI)ZrIII(ꢀꢀMe)2AlMe2]+ (5) would be initially reduced directly to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(SBI)ZrIII(ꢀꢀMe)2AlMe2, this species with its unfavorable alkyl bridges might arise only as a
−
shortꢀlived intermediate and rapidly dissociate to [(SBI)ZrIII]+ and AlMe4 (eq. 12). The rather
−
basic anion AlMe4 would then most likely react with the highly Lewisꢀacidic Zr(IV) cation 5 so
as to release 1 (eq. 13). This reaction sequence would generate, in accord with our experimental
results (Supporting Information, Figure S1) 0.5 equiv. of 1 for each cation [(SBI)Zr(ꢀꢀ
Me)2AlMe2]+ reduced, while the coordinatively unsaturated cation 11 could react further with
NaHg to yield some yet unidentified zirconocene species of oxidation state Zr(II).
−
[(SBI)Zr(ꢀ-Me)2AlMe2]+ + NaHg
[(SBI)ZrIII]+ + [AlMe4] + Na+
(12)
11
5
−
[(SBI)Zr(ꢀ-Me)2AlMe2]+ + [AlMe4]
(SBI)ZrMe2 + 2 AlMe3
(13)
1
5
5. Mechanisms of Zr(III) Formation The dihydrideꢀbridged complex (SBI)ZrIII(ꢁꢀH)2AlR2 (8)
and the isobutyl complex (SBI)ZrIIIꢀiBu (9) are formed, according to equations 4 and 5, when
(SBI)ZrMe2 (1) is treated with varying amounts of HAliBu2. Under these conditions, the Zr(III)
products thus appear to arise mainly by reductive elimination of the alkanes isobutane and
methane. Conceivable mechanisms for a reduction of Zr(IV) to Zr(III) centers by alkane
elimination might be gleaned from the following observations: Immediately upon addition of
HAliBu2 to solutions of 1, the HAl and ZrMe 1HꢀNMR signals of the reagents disappear and are
replaced by broad signals, the intensities and positions of which depend on the
[HAliBu2]init/[Zr]tot ratios used (Supporting Information). These changes indicate the formation of
20
ACS Paragon Plus Environment

Page 21 of 32
Journal of the American Chemical Society
1
2
H/Me exchange products and/or adducts between HAliBu2 and 1. While our data do not allow an
unequivocal identification of these intermediary reaction products, we assume that they are
analogs, with Me in place of Cl, of those formed from (SBI)ZrCl2 and HAliBu2,9c i.e.
heterobinuclear dihydrideꢀbridged Zr(IV) species of the type shown in eq. 14.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
R
H
H
R
AlR2
2 HAliBu2
- AlR3
Me
Me
Zr
Si
Si
(14)
Zr
H
Si
Zr
H
AlR2
R = Me, iBu
1
While these species would fulfill the formal requirement for reductive alkane elimination, viz.
cisꢀpositioned hydride and alkyl ligands, alternative and possibly more facile reaction paths
might start from homodinuclear complexes which should be accessible from these
heterobinuclear species via equilibria of the type represented in eq. 15.
Rather than leading to Zr(II) products, which would arise by alkane elimination from
heterobinuclear Zr(IV) species of the type shown in eq. 14, alkane elimination from
homobinuclear species, as in eq. 15, would lead directly to Zr(III) reduction products (eq. 16),
which could then yield the final products, (SBI)ZrIII(ꢀꢀH)2AlR2 and (SBI)ZrIIIꢀiBu, by subsequent
association or ligand exchange with HAliBu2.
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 32
1
2
3
4
5
6
R
H
R
Zr
Si
III
III
(16)
Si
+ R H + R
H
Zr
Zr
Si
7
8
9
H
Zr
Si
R = Me, iBu
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Our proposal, that homobinuclear complexes might be involved in the reduction process, is
supported by the following observation: At a ratio of [HAliBu2]init/[Zr]tot = 2, formation of Zr(III)
products from (SBI)ZrMe2 and HAliBu2 occurs with a halfꢀlife of about 1.5 h. The halfꢀlife of
this reaction increases unexpectedly to ca. 5.5 h when – under otherwise identical conditions –
HAliBu2 is used at a higher ratio of [HAliBu2]init/[Zr]tot = 10. This counterꢀintuitive observation
would be explained by the involvement of HAliBu2 in the equilibrium represented in eq. 9: An
excess of HAliBu2 would favor formation of heterobinuclear species at the expense of
homobinuclear complexes and would thus be expected to decrease the rate of Zr(III) formation,
if homobinuclear complexes are indeed required for efficient reductive alkane elimination.
Conclusions
Based on the evidence described above, six (SBI)Zr(III) species appear to be accessible by
reduction processes starting from (SBI)ZrMe2 (1) in the presence of Lewisꢀacidic
organoaluminum compounds. These are two complexes with
η
2ꢀbound aluminate ligands,
1ꢀbound
(SBI)ZrIII(ꢁꢀCl)2AlR2 (4) and (SBI)ZrIII(ꢁꢀH)2AlR2 (8), three complexes with simple
η
ligands, (SBI)ZrIIIꢀMe (6), (SBI)ZrIIIꢀiBu (9), and (SBI)ZrIIIꢀClAlR3 (10), and the cation
[(SBI)ZrIII]+ (11). Our EPR results indicate that these species are connected to one another and/or
to (SBI)ZrMe2 (1, by the reactions summarily represented in Scheme 1.
22
ACS Paragon Plus Environment

Page 23 of 32
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Most of these complexes appear to be coordinatively unsaturated, with valenceꢀelectron
counts of 13 for [(SBI)ZrIII]+, and 15 for SBZrIIIꢀX (with X = ClAlR3, Me or iBu). Even for
(SBI)ZrIII(ꢁꢀX)2AlR2, (with X = H or Cl), the effective electron density at the metal center is
probably lower than suggested by their formal 17ꢀelectron count, due to the polar character of
ꢀ
their Zrꢀligand bonding, which might be written as (SBI)ZrIII+…X2AlR2 (with X = H or Cl). The
trivalent entities encountered in this study appear to have metal centers of lower electronegativity
than the Zr(IV) centers of their tetravalent precursors, and hence an increased tendency to
transfer their ligands to Lewisꢀacidic aluminum compounds present in the reaction medium.
In the course of these studies, we have not been able to obtain any clear indication of stable
homobinuclear dimers of the type (SBI)ZrIII(ꢁꢀX)2ZrIII(SBI), with X = Cl, H or alkyl.
Monomeric and dimeric forms of trivalent zirconocene chlorides have both been reported with
complexes [(Me3SiꢀC5H4)2ZrIII(ꢀꢀCl)]2,34 and [(1,3ꢀ(Me3Si)2ꢀC5H3)2ZrIII(ꢀꢀCl)]2, 35 existing as
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 32
1
2
dimers in the solid state, whereas (1,3ꢀ(Me3C)2ꢀC5H3)2ZrIIICl, with doubly Me3Cꢀsubstituted
rings, is monomeric.44 The propensity of a trivalent zirconocene chloride to associate to a Clꢀ
bridged dimer thus appears to be determined by rather subtle changes in the steric load of its ring
ligands. While we cannot assess how a “free” (SBI)ZrIIICl species would range in this regard, its
monomeric form is apparently stabilized by formation of the AlR3 adduct (SBI)ZrIIIClAlR3 (10).
Similar considerations are likely to apply also to (SBI)ZrIIIꢀhydride species, where adduct
formation with a dialkylaluminum hydride is evident from the EPR spectra of 7. For (SBI)ZrIIIꢀ
alkyl species, on the other hand, no tendency to form any stable alkylꢀbridged heterodinuclear
adducts with alkylaluminum compounds is apparent from our data. Here, we can assume that a
similarly low propensity of alkyl ligands to occupy a bridging position would also thwart any
tendency of (SBI)ZrIIIꢀalkyl species to form alkylꢀbridged dimers.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
At any rate, the EPR signal of (SBI)ZrIII(ꢀꢀCl)AlR3 (10) at g = 1.968 accounts for practically
the entire zircononocene concentration, as does that of (SBI)ZrIII(ꢁꢀCl)2AlMe2 (4), with which it
is in equilibrium according to eq. 9 (cf. Figure 8), and the Zr(III) complexes 8 and 9, produced
by reaction of 1 with 2 equiv. of HAliBu2. Any EPRꢀsilent dimers of the type [(SBI)ZrIII(ꢁꢀX)]2,
with X = Cl, H or alkyl, could thus represent only insignificant fractions of the zirconocene
concentration in these reaction systems.
For some reaction systems, however, such as those where 1 reacts with 10 equiv. of HAliBu2,
the resulting EPR signals account only for rather small fractions (20ꢀ30%) of the total
zirconocene content. In these as in some of the other reaction systems, the appearance of new 1H
NMR signals indicates the formation of some diamagnetic products, which we have not been
able to identify so far. It must thus remain open, to which degree e.g. products of the (SBI)Zr(II)
oxidation stage accumulate in the course of these reactions.
24
ACS Paragon Plus Environment

Page 25 of 32
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
In our present study we have obtained first evidence concerning some relevant structures and
reaction patterns of trivalent ansaꢀzirconocene complexes occurring in catalysisꢀrelated reaction
media. Further studies will be aimed at elucidating the role which these Zr(III) species play in
zirconoceneꢀbased olefinꢀpolymerization catalysis and if, once formed under polymerization
conditions, they exist permanently in this oxidation state or can react further to regenerate Zr(IV)
precatalysts.45
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental Section
All operations were carried out under a protective dinitrogen atmosphere, either in a glovebox or
on a vacuum manifold. Solvents (benzene, toluene, benzeneꢀd6, tolueneꢀd8) were degassed and
dried over 3ꢀÅ molecular sieves for 12 hours or filtered through a plug of activated alumina.
(SBI)ZrMe2 was obtained from Dr. M. Ringwald, MCAT, Konstanz. Trimethylaluminum,
diisobutylaluminum hydride, aluminum chloride, triisobutylaluminum (1 M in toluene), and
dimethylphenylphosphine were used as obtained from Aldrich Chemical Co., Milwaukee.
Dimethylaluminum hydride was synthesized as previously reported.11 Stock solutions of
diisobutylaluminum hydride, dimethylphenylphosphine, and dimethylaluminum chloride were
made at two concentrations (280 mM and 28 mM) in toluene and benzeneꢀd6.
Solutions of [(SBI)Zr(ꢁꢀCl)2AlMe2][B(C6F5)4],13, 17 [(SBI)Zr(ꢁꢀH)3(AliBu2)2][B(C6F5)4] 13 and
[(SBI)Zr(ꢁꢀMe)2AlMe2][B(C6F5)4] 11 were prepared in analogy to previous reports.
Samples for NMR and EPR spectra were prepared in the glove box by transferring 0.8 mL of
prepared sample solutions into 5ꢀmm OD glass NMR tubes. These were closed with rubber
stoppers which had been washed with pentane and dried extensively in vacuo. For each
25
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 32
1
2
3
4
5
6
measurement, these sample tubes were taken from the glovebox to the respective spectrometer
and immediately transferred back to the glovebox for storage.
7
8
9
EPR spectra were measured using a Magnettech Miniscope MS400 spectrometer at an Xꢀ
band frequency of 9.41 GHz. Hyperfine interaction parameters and g values were analyzed by
model calculations using the software EASYSPIN.46 Concentration of Zr(III) species were
obtained by double integration of the EPR signal, compared to reference samples of TEMPO.
NMR spectra were measured using Varian Inova 500, 600, or Mercury 300 spectrometers.
Chemical shifts are referenced to residual solvents peaks, 7.16 ppm for benzene and 2.08 ppm
for CD2H proton resonances of toluene. Further experimental details are described in the
supporting information.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CAUTION: alkylaluminum compounds are pyrophoric and must be handled with special
precautions (see, e.g., Shriver, D.F. The Manipulation of Airꢀsensitive Compounds; Robert E.
Krieger Publishing Company; Malabar, Florida, 1982).
26
ACS Paragon Plus Environment

Page 27 of 32
Journal of the American Chemical Society
1
2
3
4
Acknowlegement.
5
6
7
8
9
This research was supported by Deutsche Forschhungsgemeinschaft (Grant Me1388/9ꢀ1), by
Russian Foundation of Basic Research (grant 10ꢀ03ꢀ91330) and by USDOE Office of Basic
Energy Sciences (grant DEꢀFG03ꢀ85ER13431). We thank Prof. Jay A. Labinger (California
Institute of Technology), Dr. Konstantin P. Bryliakov (Boreskov Institute of Catalysis) and Prof.
Heiko Möller (University of Konstanz) for helpful discussions, Anke Friemel, Evelyn Wuttke,
Martin Spitzmesser (University of Konstanz), David Van der Velde and Angelo Di Bilio
(California Institute of Technology) for help with the acquisition and analysis of NMR and EPR
spectra, and Lawrence M. Henling and Thomas S. Teets for assistance with mounting crystals
and solving cyrstal structures. The Bruker KAPPA APEXII Xꢀray diffractometer was purchased
via an NSF CRIF:MU award to the California Institute of Technology (CHEꢀ0639094).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Supporting Information Available Further 1H NMR and EPR data, simulations of EPR spectra
and experimental details. This material is available free of charge via the Internet at
http://pubs.acs.org. Crystallographic data for crystals containing the compound (SBI)Zr(ꢂꢀ
Cl)2AlMe2 (TNL17) have been deposited with the CCDC, 12 Union Road, Cambridge CB2 1EZ,
UK; copies can be obtained on request, free of charge, by quoting the publication reference and
deposition number 941568.
27
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 32
1
2
3
4
References and Footnotes.
5
6
7
8
9
1. Reviews: a) Resconi, L.; Cavallo, L.; Fait; A:, Piemontesi, F. Chem. Rev. 2000, 100, 1253ꢀ
1345. b) Resconi, L.; Chadwick, J. C.; Cavallo, L., In Comprehensive Organometallic Chemistry
III, Crabtree, R. H.; Mingos, D. M. P., Eds. Elsevier: Oxford, 2007; Vol. 4, Bochmann, M., Ed.,
Chapter 4.09, p. 1005ꢀ1166.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2. Reviews: a) Chen, E. Y. X.; Marks, T. J. Chem. Rev. 2000, 100, 1391ꢀ1434. b) Bochmann, M.
J. Organomet. Chem. 2004, 689, 3982ꢀ3998. c) Bochmann, M. Organometallics 2010, 29, 4711ꢀ
4740.
3. a) Huang, Y. H.; Yu, Q.; Zhu, S.; Rempel, G. L.; Li, L. J. Polym. Sci. A, Polym. Chem. 1999,
37, 1465ꢀ1472. b) Bryliakov, K. P.; Babushkin, D. E.; Talsi, E. P.; Voskoboynikov, A. Z.;
Gritzo, H.; Schröder, L.; Damrau,H.ꢀR. H.; Wieser, U.; Schaper, F.; Brintzinger, H. H.
Organometallics 2005, 24, 894ꢀ904.
4. a) Grassi, A.; Saccheo, S.; Zambelli, A.; Laschi, F. Macromolecules 1998, 31, 5588ꢀ5591. b)
Chien, J. W. C.; Salajka, Z.; Dong, S. Macromolecules, 1992, 25, 3199. c) Mahanthappa, M. K.;
Waymouth, R. M. J. Am. Chem. Soc. 2001, 123, 12093ꢀ12094.
5. a) Williams, E. F.; Murray, M. C.; Baird, M. C. Macromolecules 2000, 33, 261ꢀ268. b)
Bryliakov, K.P.; Semikolenova, N. V.; Zakharov, V. A.; Talsi, E. P. J. Organomet. Chem. 2003,
683, 23ꢀ28. c) Napoli,M.; Grisi, F.; Longo, P. Macromolecules, 2009, 42, 2516ꢀ2522.
6. Cram, D.; Sartori, F.; Maldotti, A. Macromol. Chem.Phys. 1994, 195, 2817ꢀ2826.
7. Lyakin, O. Y.; Bryliakov, K. P.; Panchenko, V. N.; Semikolenova, N. V.; Zakharov, V. A.;
Talsi, E. P. Macromol. Chem.Phys. 2007, 208, 1168ꢀ1175.
8. Review: Lancaster, S. J.; In Comprehensive Organometallic Chemistry III, Crabtree, R. H.;
Mingos, D. M. P. , Eds.; Elsevier, Oxford, 2007, Vol. 4, Bochmann, M., Ed., Chapter 4.07, p.
741ꢀ757.
9. a) Babushkin, D. E.; Brintzinger, H. H. Chem. Eur. J. 2007, 13, 5294ꢀ5299. b) Babushkin, D.
E.; Panchenko, V. N.; Timofeeva, M. N.; Zakharov, V. A.; Brintzinger, H. H. Macromol. Chem.
Phys. 2008, 209, 1210ꢀ1219. c) Baldwin, S. M.; Bercaw, J. E.; Brintzinger, H. H. J. Am. Chem.
Soc. 2008, 130, 17423ꢀ17433.
10. Kleinschmidt, R.; van der Leek, Y.; Reffke, M.; Fink, G. J. Mol. Catal. A-Chem. 1999, 148,
29ꢀ41.
11. Bochmann, M.; Lancaster, S. J. Angew. Chem. Int. Ed. Engl. 1994, 33, 1634ꢀ1637.
12. Baldwin, S. M.; Bercaw, J. E.; Brintzinger, H. H. J. Am. Chem. Soc. 2010, 132, 13969ꢀ
13971.
28
ACS Paragon Plus Environment

Page 29 of 32
Journal of the American Chemical Society
1
2
3
4
5
13. Baldwin, S. M.; Bercaw, J. E.; Henling, L. M.; Day, M. W.; Brintzinger, H. H. J. Am. Chem.
Soc. 2011, 133, 1805ꢀ1813.
6
7
8
9
14. Wailes, P. C.; Weigold, H. J. Organomet. Chem. 1971, 28, 91ꢀ95.
15. Cuenca, T., Royo, P. J. Organomet. Chem. 1985, 293, 61ꢀ67.
16. Review: Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877ꢀ910.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
17. Formation of the dichloroꢀbridged cation [(SBI)Zr(ꢀꢀCl)2AliBu2]+ by reaction of ClAliBu2
with the cation [(SBI)Zr(ꢀꢀH)3(AliBu2)2]+ or with the dimeric diꢀcation [((SBI)Zr)2(ꢀꢀCl)2]2+,
−
both of which have been crystallographically characterized in form of their B(C6F5)4 salts (ref.
13 and ref. 18, respectively), has been described in ref. 13. For the present study, we have
prepared solutions of [(SBI)Zr(ꢀꢀCl)2AlMe2]+ by reaction of ClAlMe2 with the cation
[(SBI)Zr(ꢀꢀMe)2AlMe2]+, in order to avoid the presence of potentially hydrideꢀgenerating
aluminum isoꢀbutyl species (Supporting Information).
18. Bryliakov, K. P.; Talsi, E. P.; Semikolenova, N. V.; Zakharov, V. A.; Brand, J.; Alonsoꢀ
Moreno, C.; Bochmann, M. J. Organomet. Chem. 2007, 692, 859ꢀ868.
19. The expected hyperfine splitting by the ZrꢀCH3 group (ca. 5 G) remains unresolved under the
line width (ca. 8 G) of the main signal in Figure 2.
20. Attempts to reduce the cation [(SBI)Zr(ꢁꢀMe)2AlMe2]+ with cobaltocene likewise yielded
substantial amounts of (SBI)ZrMe2.
21. Hyperfine splitting due to the minor isotope 91Zr (I = 5/2, 11%) is not observed in this case,
probably due to overlap with multiple splitting by 27Al and 1H nuclei.
22. The EPR signal shown in Figure 3 is not compatible with a hyperfine splitting due to two
27Al nuclei, e.g. in a complex (SBI)Zr(ꢀꢀH)3(AlR2)2.
23. A similar EPR spectrum with hyperfine splitting by one Al nucleus, with a(27Al) = 5.9 G, and
two hydrogen nuclei, with a(1H) = 3.1 G, in reaction media containing (2ꢀPhꢀind)2ZrCl2 and
AliBu3, had been assigned to a species (2ꢀPhꢀind)2ZrIII(ꢁꢀCl)(ꢁꢀiBu)Al iBu2 by Lyakin et al. (ref.
7). Such an assignment can be excluded in our case, due to the absence of Cl− ions.
24. For zirconocene hydride complexes with terminal ZrIIIꢀH units, a(1H) values in the range of
6ꢀ7 G have been reported in ref. 23aꢀd. Higher values of 7ꢀ9 G have been found for anionic
hydride complexes (ref. 24), while a(1H) values as low as 3ꢀ4 G have been observed for
zirconocene phosphine hydride complexes (ref. 25).
25. a) Samuel, E.; Maillard, P.; Gianotti, C. J. Organomet. Chem. 1977, 142, 289ꢀ298. b)
Hudson, A.; Lappert, M. F.; Pichon, R. J. Chem. Soc., Chem. Comm. 1983, 374. c) Bajgur, C. S.;
Jones, S. B., Petersen, J. L. Organometallics 1985, 4, 1929ꢀ1936. d) Dioumaev, V.K.; Harrod, J.
F. Organometallics 1997, 16, 1452ꢀ1464.
29
ACS Paragon Plus Environment