Selective halodemethylation reactions of metallocene dimethyls with triphenylmethyl chloride and benzyl bromide

Article abstract of DOI:10.1021/om0305014

NMR-scale reactions of several group 4 metallocene dimethyls with either trityl chloride or benzyl bromide gave the corresponding L₂M(Me)X complexes selectively. Five reactions, including syntheses of Cp₂-Zr(Me)Cl and Ind₂Zr(Me)Cl, were conducted on a preparative scale to afford useful isolated yields of L₂M-(Me)Cl complexes.

Full text of DOI:10.1021/om0305014

Organometallics 2004, 23, 9-11
9
Communications
Selective Ha lod em eth yla tion Rea ction s of Meta llocen e
Dim eth yls w ith Tr ip h en ylm eth yl Ch lor id e a n d Ben zyl
Br om id e
Eric J . Hawrelak and Paul A. Deck*
Department of Chemistry, Virginia Tech, Blacksburg, Virginia 24061-0212
Received J une 30, 2003
Summary: NMR-scale reactions of several group 4 met-
allocene dimethyls with either trityl chloride or benzyl
bromide gave the corresponding L2M(Me)X complexes
selectively. Five reactions, including syntheses of Cp2-
Zr(Me)Cl and Ind2Zr(Me)Cl, were conducted on a pre-
parative scale to afford useful isolated yields of L2M-
(Me)Cl complexes.
(C6F5Cp)2Zr(Me)Cl, (C6F5Cp)CpZr(Me)Cl, and (C6F5-
Cp)2Hf(Me)Cl, first to assign these species conclusively
in spectra of complex reaction mixtures and second to
study their reactivity toward alkylaluminum species,
including MAO. Established routes1,2,20-24 to Cp2Zr(Me)-
Cl did not afford the target compounds selectively.
Because alkyl halides are known to react with transi-
tion-metal alkyls or hydrides to form the corresponding
metal halides, we surmised that the wide range of
reactivity among readily available alkyl halides might
allow us to find generally selective reagents and reaction
conditions for the monohalodemethylation of metal-
locene dimethyls.
Activation of group 4 metallocene dichlorides (L2MCl2)
toward olefin polymerization using methylalumoxane
(MAO) is believed to proceed by initial halodemethyla-
tion to form L2M(Me)Cl, followed by chloride abstraction
to afford ionic-like species formulated as [L2MMe]+-
[ClMAO]- and [L2MMe]+[MeMAO]-.3 However, the fate
of the halide upon activation and its subsequent role
during olefin polymerization catalysis are not yet fully
known.4-18 As part of our ongoing study of metallocene
activation using C6F5-substituted Cp ligands as 19F
NMR spectroscopic probes,19 we needed to prepare
Metallocene dimethyls19,25-30 were treated with 1
equiv of triphenylmethyl chloride (Ph3CCl) in either
benzene-d6 or toluene-d8 according to eq 1, where L )
Cp (or congener) and M is a group 4 metal. We chose
L2MMe2 + Ph3CCl )
L2M(Me)Cl (+L2MCl2) + Ph3CCH3 (1)
(1) Surtees, J . R. Chem. Commun. 1965, 567.
(2) Lisowsky, R. U.S. Patent No. 5523 435, J une 4, 1996, to Witco
GmbH.
Ph3CCl on the basis of its reactivity and ready avail-
ability. NMR-scale reactions enabled us to follow the
course of each reaction and thereby optimize both the
reaction time and the selectivity for L2M(Me)Cl. The
results are presented in Table 1.
An NMR-scale reaction was considered a promising
candidate for scale-up if no more than 5% of L2ZrMe2
remained and no more than 5% of L2MCl2 had formed
(3) Chen, E. Y. X.; Marks, T. J . Chem. Rev. 2000, 100, 1391-1434.
(4) Coevoet, D.; Cramail, H.; Deffieux, A. Macromol. Chem. Phys.
1998, 199, 1451-1457.
(5) Pe´deutour, J . N.; Cramail, H.; Deffieux, A. J . Mol. Catal. A:
Chem. 2001, 174, 81-87.
(6) Pe´deutour, J . N.; Cramail, H.; Deffieux, A. J . Mol. Catal. A:
Chem. 2001, 176, 87-94.
(7) Pe´deutour, J . N.; Radhakrishnan, K.; Cramail, H.; Deffieux, A.
Polym. Int. 2002, 51, 973-977.
(8) Babushkin, D. E.; Semikolenova, N. V.; Zakharov, V. A.; Talsi,
E. P. Macromol. Chem. Phys. 2000, 201, 558-567.
(9) Wieser, U.; Babushkin, D.; Brintzinger, H. H. Organometallics
2002, 21, 920-923.
(10) Kim, I.; J ordan, R. F. Macromolecules 1996, 29, 489-491.
(11) Ystenes, M.; Eilertsen, J . L.; Liu, J .; Ott, M.; Rytter, E.;
Stovneng, J . A. J . Polym. Sci., Polym. Chem. 2000, 38, 3106-3127.
(12) Pe´deutour, J . N.; Coevoet, D.; Cramail, H.; Deffieux, A. Mac-
romol. Chem. Phys. 1999, 200, 1215-1221.
(13) Wang, Q.; Song, L.; Zhao, Y.; Feng, L. Macromol. Rapid
Commun. 2001, 22, 1030-1034.
(14) Ma¨kela¨, N. I.; Knuuttila, H. R.; Linnolahti, M.; Pakkanen, T.
A.; Leskela¨, M. A. Macromolecules 2002, 35, 3395-3401.
(15) Bochmann, M.; Lancaster, S. J . Organometallics 1993, 12, 633-
640.
(16) Ferreira, M. L.; Belelli, P. G.; Damiani, D. E. Macromol. Chem.
Phys. 2001, 202, 495-511.
(20) Reid, A. F.; Shannon, J . S.; Swan, J . M.; Wailes, P. C. Aust. J .
Chem. 1965, 18, 173.
(21) Moore, E. J .; Strauss, D. A.; Armantrout, J .; Santarsiero, B.
D.; Grubbs, R. H.; Bercaw, J . E. J . Am. Chem. Soc. 1983, 105, 2068.
(22) J ordan, R. F. J . Organomet. Chem. 1985, 294, 321.
(23) Wailes, P. C.; Weigold, H.; Bell, A. P. J . Organomet. Chem.
1972, 34, 155.
(24) Wendt, O. F.; Bercaw, J . E. Organometallics 2001, 20, 3891.
(25) Diamond, G. M.; J ordan, R. F.; Petersen, J . L. J . Am. Chem.
Soc. 1996, 118, 8024-8033.
(26) Siedle, A. R.; Newmark, R. A.; Lamanna, W. M.; Schroepfer, J .
N. Polyhedron 1990, 9, 301-308.
(27) Horton, A. D.; Orpen, A. G. Organometallics 1991, 10, 3910-
3918.
(17) Makela, N. I.; Knuuttila, H. R.; Linnolahti, M.; Pakkanen, T.
A. J . Chem. Soc., Dalton Trans. 2001, 91-95.
(18) Smith, J . M.; Bott, S. G. J . Chem. Soc., Chem. Commun. 1996,
377-378.
(19) Hawrelak, E. J .; Deck, P. A. Organometallics 2003, 22, 3558-
3565.
(28) Samuel, E.; Rausch, M. D. J . Am. Chem. Soc. 1973, 95, 6263-
6267.
(29) Neithamer, D. R.; Stevens, J . C. U.S. Patent No. 5 399 635,
March 21, 1995, to Dow Chemical Co.
(30) Wolczanski, P. T.; Bercaw, J . E. Organometallics 1982, 1, 793-
799.
10.1021/om0305014 CCC: $27.50 © 2004 American Chemical Society
Publication on Web 12/06/2003

10 Organometallics, Vol. 23, No. 1, 2004
Ta ble 1. Ha lod em eth yla tion Rea ction s of Dim eth ylm eta llocen esa
Communications
reacn conditions
product, mol %b
d
entry
substrate
Cp2ZrMe2
Cp2ZrMe2
Cp2ZrMe2
Ind2ZrMe2
(C6F5Cp)2ZrMe2
(C6F5Cp)2ZrMe2
(C6F5Cp)2ZrMe2
(C6F5Cp)CpZrMe2
(C6F5Cp)2HfMe2
(Me3SiCp)2ZrMe2
(Me3SiCp)2ZrMe2
(Me5Cp)CpZrMe2
(Me5Cp)CpHfMe2
(C6F5Cp)(Me5Cp)ZrMe2
rac-C2H4(Ind)2ZrMe2
rac-C2H4(Ind)2ZrMe2
[(Me4C5)SiMe2NtBu]ZrMe2
[(Me4C5)SiMe2NtBu]ZrMe2
reagent
solvent
T, °C
time
18 h
15 h
2 days
18 h
18 h
23 h
26 h
18 h
MMe2
M(Me)Xc,d
MX2
1
2
3
4
5
6
7
8
Ph3CCl
PhCH2Br
0.5 PbCl2
Ph3CCl
Ph3CCl
Ph3SiCl
CD2Cl2
Ph3CCl
Ph3CCl
Ph3CCl
PhCH2Br
Ph3CCl
Ph3CCl
Ph3CCl
Ph3CCl
PhCH2Br
Ph3CCl
PhCH2Bre
C6D6
tol-d8
tol-d8
tol-d8
C6D6
C6D6
CD2Cl2
C6D6
C6D6
C6D6
tol-d8
tol-d8
tol-d8
C6D6
tol-d8
C6D6
C6D6
tol-d8
60
95
90
90
60
70
25
60
60
60
90
95
95
60
25
70
25
95
0
6
65
0
95 (80)
5
8
30
2
4
0
0
1
2
0
0
5
2
5
9
5
7
0
86
5
98 (78)
96 (85)
0
0
100
100
3
0
2
3
0
3
10
2
0
96 (88)
98 (64)
98
9
18 h
15 h
10
11
12
13
14
15
16
17
18
60 days
18 h
2 days
2 days
6 h
5 days
5 min
2 days
97
95
95
85
89
95
91
98
0
2
2
a
Thermolyses were carried out in NMR tubes in the dark without agitation of the solutions. Representative experiments are described
in ref 35. See the Supporting Information for complete experimental details. Relative amounts were determined by either 1H NMR or
b
19F NMR integration. c Quantities in parentheses are isolated yields from analogous preparative-scale reactions. Where reagent ) Ph3CCl,
d
PbCl2, X ) Cl; where reagent ) PhCH2Br, X ) Br. e An excess (10 equiv) of PhCH2Br was used.
(eq 1). Accordingly, entries 1, 4, 5, 8, and 9 also provide
the isolated yields for corresponding preparative-scale
reactions. For entries 10, 12, and 13, the L2M(Me)Cl
products were also formed selectively but were too
soluble in hexane to be isolated efficiently from the
byproducts (see below) by crystallization. Ph3SiCl did
not react under similar conditions (entry 6). Our failure
using PbCl2 was surprising, considering the success
reported previously by Wailes et al.23 However, the
reaction is heterogeneous and Wailes’s procedure indi-
cated “vigorous stirring”; thus, perhaps we did not
agitate our reactions enough.
an SH2 radical substitution (eqs 4 and 5).31,32
Cp2ZrRX + 2tBu2NO• )
Cp2Zr(X)(ONtBu2) + tBu2NOR (3)
tBu2NO• + Cp2ZrRX ) Cp2Zr(X)(ONtBu2) + R• (4)
R• + tBu2NO• ) tBu2NOR
(5)
Interestingly, Cp2ZrMe2 reacted much more rapidly
than Cp2Zr(Me)Cl, foreshadowing the selectivity re-
ported here. The analogous process using Ph3CCl would
require thermal initiation (not photochemicalsour reac-
tions were run in the dark) followed by substitution (eqs
6-8).
We then tried using benzyl bromide for a few reac-
tions (eq 2, results in Table 1). We selected two
L2M(Me)2 + PhCH2Br )
L2M(Me)Br (+L2MBr2) + PhCH2Me (2)
Ph3CCl ) Ph3C• + Cl• (initiation)
Cp2ZrMe2 + Cl• ) Cp2Zr(Me)Cl + Me•
Me• + Ph3C• ) Ph3CMe
(6)
(7)
(8)
substrates for which reactions with Ph3CCl were rapid
but unselective (Table 1, entries 15 and 17). Generally
the reactions with benzyl bromide were more selective
but also rather slow (especially entries 11 and 16).
The observation of long induction periods (up to 48
h) suggests a free-radical process. Although 1,1,1-
triphenylethane is the major organic byproduct (δMe 2.03
ppm), significant quantities of Ph3CH (δCH 5.50 ppm)
and Ph2(4-Me-C6H4)CCH3 (δMe 2.08 and 2.15 ppm) were
also observed (product assignments were confirmed by
GC-MS analysis), suggesting methyl- and hydrogen-
radical-transfer events. In contrast, the reactions with
benzyl bromide showed ethylbenzene as the only sig-
nificant organic byproduct. Ethylbenzene should be
more easily separated, e.g., by evaporation or crystal-
lization than 1,1,1-triphenylethanesa potential practi-
cal advantage.
However, in contrast to eq 5, where the nitroxyl
radical is present in stoichiometric quantities, eq 6
requires thermal dissociation of Ph3CCl first. Assuming
Ea(6) ≈ DC-Cl ≈ 70 kcal mol-1 and a reasonable preex-
ponential factor (A ≈ 1011 s-1), the rate of dissociation
at 100 °C would be too slow (k6 ≈ 10-30 s-1) to observe
product formation during the intervals reported in Table
1. Furthermore, eq 8 requires efficient recombination
of two dilute radicals. In the absence of kinetic evidence
Although we have not elucidated the mechanism of
these halodemethylation reactions, a related process
involving nitroxyl radicals (eq 3, where R ) CH2Ph, Me
and X ) CH2Ph, Me, Cl, Br) was found to proceed by
(31) Brindley, P. B.; Scotton, M. J . J . Organomet. Chem. 1981, 222,
89-96.
(32) Brindley, P. B.; Scotton, M. J . J . Chem. Soc., Perkin Trans. 2
1981, 419-423.

Communications
Organometallics, Vol. 23, No. 1, 2004 11
or meaningful substituent effect data, a chain process
(eqs 9 and 10) seems more plausible to us.33,34
Ack n ow led gm en t. This work was funded by the
NSF (Grant No. CHE-9875446). We thank W. F. J ack-
son for the synthesis of (C6F5Cp)(Me5Cp)ZrCl2, Prof. E.
Y.-X. Chen for a sample of rac-C2H4(Ind)2ZrMe2, and
Prof. J . M. Tanko for helpful discussions.
Ph3C• + Cp2ZrMe2 ) Ph3CMe + [Cp2ZrMe]• (9)
[Cp2ZrMe]• + Ph3CCl ) Cp2Zr(Me)Cl + Ph3C• (10)
Note Ad d ed in P r oof. Resconi and coworkers re-
cently reported clean conversions of Ind2TiMe2 to
Ind2Ti(Me)Cl using SiCl4 (4 equiv, toluene solvent, 50
°C, 6 h) and of Ind2ZrMe2 to Ind2Zr(Me)Cl using FeCl3
(1 equiv, CH2Cl2 solvent, 40 °C, 6 h).36
In conclusion, we find that reactions of group 4
metallocene dimethyls with either trityl chloride or
benzyl bromide are generally selective for the formation
of L2M(Me)Cl. NMR-scale experiments facilitate selec-
tion of the reagent and optimization of the reaction
conditions. As long as the ancillary ligands do not
increase the solubility of the metallocene too much, the
product may be isolated from the organic byproducts in
high yield and purity by crystallization. Long induction
times and minor organic byproducts suggest a free
radical mechanism for halodemethylation.
Su p p or tin g In for m a tion Ava ila ble: Complete experi-
mental details. This material is available free of charge via
the Internet at http://pubs.acs.org.
OM0305014
(35) (a) In a typical NMR experiment, a J . Young NMR tube was
charged in
a nitrogen glovebox with 0.05 mmol (nominal) of the
metallocene dimethyl, 1.0-1.1 equiv of either Ph3CCl or PhCH2Br, and
about 1 mL of solvent (see Table 1). The tube was placed in an oil
bath preheated to the prescribed temperature (see Table 1). At several
intervals, the sample was analyzed by 1H NMR (400 MHz) and, where
appropriate, 19F NMR (376 MHz). (b) In a representative preparative-
scale experiment, a mixture of Cp2ZrMe2 (188 mg, 0.749 mmol), Ph3-
CCl (211 mg, 0.756 mmol), and benzene (15 mL) was stirred under
argon at 60 °C for 18 h. The solvent was evaporated, and the residue
was recrystallized from hexane under argon to afford 138 mg (0.597
mmol, 80%) of Cp2Zr(Me)Cl as pale yellow crystals. 1H NMR (C6D6) δ
5.73 (s, 10 H, Cp), 0.439 (s, 3 H, ZrMe).
(33) The mixed alkyl complex Cp2Zr(CHdCHPh)(CH2Ph) reacts with
PhCH2Cl under photolysis conditions to give Cp2Zr(CHdCHPh)(Cl) and
PhCH2CH2Ph. Photolysis of Cp2ZrMe2 affords Cp2ZrMe, which may
be trapped as the P(OMe)3 adduct and characterized by EPR, sup-
porting the proposition of a Cp2ZrMe radical intermediate. See: Czisch,
P.; Erker, G.; Korth, H.-G.; Sustmann, R. Organometallics 1984, 3,
945-947.
(34) Edelbach, B. L.; Kraft, B. M.; J ones, W. D. J . Am. Chem. Soc.
1999, 121, 10327-10331.
(36) Balboni, D.; Camurati, I.; Ingurgio, A. C.; Guidotti, S.; Focante,
F.; Resconi, L. J . Organomet. Chem. 2003, 683, 2-10.