High-yield ruthenium-catalyzed Friedel-Crafts-type allylation reactions using dicationic RuIV catalysts

Article abstract of DOI:10.1002/anie.200600447

(Chemical Equation Presented) Unexpected linkage: The use of dicationic Ru^IV complexes in the Friedel-Crafts-type allylation of aromatic alcohols leads unexpectedly to the formation of a new C-C bond and not to the phenolation product (see sche

Full text of DOI:10.1002/anie.200600447

Communications
Allylations
DOI: 10.1002/anie.200600447
High-Yield Ruthenium-Catalyzed Friedel–Crafts-
Type Allylation Reactions Using Dicationic
RuIV Catalysts**
Ignacio Fernµndez, RenØ Hermatschweiler,
Frank Breher, Paul S. Pregosin,* Luis F. Veiros, and
Maria JosØ Calhorda
The organometallic catalytic chemistry of ruthenium contin-
ues to expand with applications in organic synthesis involving
[*] Dr. I. Fernµndez, R. Hermatschweiler, Prof. Dr. P. S. Pregosin
Laboratory of Inorganic Chemistry
ETHZ
Hönggerberg, 8093 Zürich(Switzerland)
Fax: (+41)446-331-071
E-mail: pregosin@inorg.chem.ethz.ch
Prof. Dr. F. Breher
Institut für Anorganische Chemie
Universität Karlsruhe
Engesserstrasse 15, 76131 Karlsruhe (Germany)
Prof. Dr. L. F. Veiros
Centro de Química Estrutural
Complexo I
Instituto Superior TØcnico
Av. Rovisco Pais 1, 1049-001 Lisbon (Portugal)
Prof. Dr. M. J. Calhorda
Departamento de Química e Bioquímica
Faculdade de CiÞncias
Universidade de Lisboa, 1749-016 Lisbon (Portugal)
[**] P.S.P. thanks the Swiss National Science Foundation, the Bunde-
samt für Bildung und Wissenschaft, and the ETHZ for financial
support. I.F. thanks the Junta de Andalucía for a research contract.
We also thank Johnson Matthey for the loan of precious metals.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
6386
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6386 –6391

Angewandte
Chemie
hydroamination,[1] isomerization,[2–4] kinetic resolution,[5]
Table 1: Selected ruthenium-catalyzed carbon–carbon bond formations
hydrogenation,[6] metathesis,[7] a variety of multiple C C
in acetonitrile at 353 K using the catalyst precursor 4 with the branched
À
allyl carbonate substrate 6.[a]
bond-formation reactions,[8] and allylic alkylation.[9,10]
With respect to allylation chemistry, several Cp*Ru
precursors (Cp* = C5Me5) have been suggested as catalysts.
These include the tris(acetonitrile) complex [Ru(Cp*)-
(CH3CN)3]PF6 (1) from Trost et al.,[9] complexes containing
bidentate nitrogen ligands,[10] and the RuIV carbonate com-
plex 2 reported by us.[11] The latter complex[11b] was shown to
be an interesting alternative to 1 in the allylic alkylation of
À
Entry Arene
t [min] Product
% C C
(o/m/p)[b]
100
1
2
phenol
72
(10:6:84)
100
2,6-Me2-phenol 58
(0:0:100)
100
(82:18:0)
3
4
5
4-Me-phenol
2-naphthol
1
4
4
94
phenylallyl-tert-butyl carbonate with sodium dimethyl malo-
nate. Complex 2 can also be used successfully in allylic
amination[12a] and phenolation[12b] reactions.
6-Br-2-naphthol
(5 equiv)
100
If 2 were to decompose to afford CO2 and t-butoxide, the
RuIV solvate complexes, for example 3 and 4, would be
6-Br-2-naph-
thoxysilane
6
5
100
100
(5:2:93)
7
8
anisole
205
425
74
(16:0:84)
thioanisol
9
p-xylole[c]
o-xylole[c]
16 h
92
73
formed. Consequently, we prepared these dicationic com-
plexes and show herein that 3, in contrast to 2, represents a
novel catalyst for Friedel–Crafts-type allylation and catalyzes
10
81 (20:80)
[a] Conditions: 0.07 mmol of allyl substrate 6, 0.21 mmol of the
corresponding arene derivative, 0.002 mmol of catalyst (3 mol%), in
0.5 mL of acetonitrile. [b] The o/m/p ratios were determined by 1H NMR
spectroscopy. [c] 30 equivalents.
À
unexpected C C couplings. Specifically, different arene com-
pounds can be selectively allylated under relatively mild
conditions in very good yield without the use of Lewis acidic
main-group compounds or other additives.
Complexes 3 and 4 were prepared from our previously
reported RuIV chlorido complex [Ru(Cp*)Cl(CH3CN)(h3-
PhCHCHCH2)]PF6 [5, see Eq. (1)][11a] by treatment with
AgPF6 in the appropriate solvent.[13a] Table 1 shows the
Friedel–Crafts-type products from the reaction of the
branched phenylallyl carbonate PhCH(OCO2tBu)CHCH2
(6) with a variety of electron-rich aromatic substrates at
353 K with 4 as the catalyst (which in acetonitrile solution
reverts to 3).
In all the examples shown, the conversion to product is
high, and 1H NMR spectroscopic data show only linear and no
branched isomer. For a few of the arene compounds only one
isomer is observed in very high yield (Table 1, entries 2, 5, and
6); however, for several of the arene compounds, one does
find a mixture of ortho, meta, and para isomers. For the
phenol compounds,[13b] the reaction is quite rapid, and in some
cases complete conversion is observed in less than 10 minutes.
Angew. Chem. Int. Ed. 2006, 45, 6386 –6391ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6387

Communications
With chlorobenzene, however, there is no allylation after
about 24 hours. We were also able to show that the solvent can
have a significant influence on the reaction. With dimethyl-
formamide (DMF) as the solvent and precursor 4, the
À
reaction of a naphthol substrate affords both C C (36%)
À
and C O (64%) coupling products.
To further test the generality of the catalyst precursors for
Friedel–Crafts reactions, we used 3 to catalyze the reaction of
6-bromo-2-naphthol with several differently substituted car-
bonates [Eq. (2)]. For all of the reactions leading to 7, the
The molecular structures of the two dications of 3 and 4[13]
are shown in Figures 1 and 2. A selection of bond angles and
bond lengths are given in the captions of the figures. Both
Figure 1. Structure of the dication [Ru(Cp*)(h3-CH2CHCHPh)-
(MeCN)2]2+ in 3. Thermal ellipsoids are drawn at 30% probability; PF6
anions are omitted for clarity. Bond lengths [pm] and angles [8]: Ru1-
N1 207.8(3), Ru1-N2 208.1(3), Ru1-C1 218.0(3), Ru1-C2 218.9(3), Ru1-
C3 238.2(3), Ru1-Ct1 185.9(3), Ru1-C10 219.1(3), Ru1-C20 220.6(3),
Ru-C30 221.4(3), Ru1-C40 225.7(3), Ru1-C50 223.9(3), N1-C100
112.3(4), N2-C200 112.4(4); C1-C2-C3 119.4(3), N1-Ru1-N2 82.6(1);
f=70.38; Ct1=centroid of the Cp* ligand; f is the intersection of the
planes described by C1, C2, C3 and the five atoms of the Cp* ring.
conversion is 100%, and the yield of product is greater than
80%. The reactions leading to 7b–7d were complete within
less than 10 minutes, whereas 100% conversion to 7a
required about two hours.
The Friedel–Crafts reaction has
a
long history.[14]
Recently, it has been shown that a number of late-transi-
tion-metal complexes[15,16] are capable of condensing arene
compounds with a suitable electrophile (for example, hex-
anoic acid anhydride[15]), and this area has recently been
reviewed.[17] We are aware of several reports in which
Ru compounds have been employed.[15,18] Two of these
studies[18] use a dinuclear RuIII species at about 413 K (fairly
harsh conditions), and the yields are often only moderate.
Fürstner and co-workers have employed RuCl3 in the
acylation of hexanoic acid anhydride with anisole; however,
this catalyst seems to be relatively slow. Our catalyst
complexes show the Ru atom complexed by the Cp* ligand,
the phenylallyl ligand, and two solvent molecules. As
expected, the dmf ligands are complexed through the amide
oxygen atoms. The most interesting features in these dications
À
involve the various Ru C(allyl) bond lengths. Specifically, the
À
Ru C3 bonds (238.2(3) (3) and 232.3(4) pm (4)) are relatively
long with the former separation exceptionally large. We have
À
represents
a
significant improvement relative to these
observed this type of long Ru C(allyl) bond in both the
Ru sources in that we use less catalyst and have faster
reactions under milder conditions.
RuIV chlorido complex
5
(Ru C3 235.1(2) pm)[11a] and
À
RuIV carbonate complex 2 (Ru C3 230.3(5) pm).
How-
[11b]
À
It is important to note that the related catalyst 2[11b,12b]
ever, for the acetonitrile complex 3, this bond is the longest of
the set.
À
À
forms only C O and no C C bonds with phenol and the same
branched allyl substrate [Eq. (3)], so that by a slight
modification of the catalyst, one can completely redirect the
nature of the organic product formed. Given this useful
flexibility, we considered it appropriate to study aspects of
these ruthenium catalysts.
We have previously suggested that this long bond arises
from relatively weak p back-bonding.[11a] The observed
13C NMR chemical shift of C3 in 3 (d = 103.3 ppm) appears
at higher frequency than that in the monocation 5 (d =
91.0 ppm) and is thus consistent with this suggestion.
6388
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6386 –6391

Angewandte
Chemie
Figure 3. a) Bonding Ru allyl p orbitals. Left: HOMOÀ4 of [Ru(Cp)-
(CH3CN)2(h3-PhCHCHCH2)]2+, withno contribution from the acetonitrile
orbitals; middle: HOMOÀ5 of [Ru(Cp)(dmf)2(h3-PhCHCHCH2)]2+; right:
HOMOÀ4 of [Ru(Cp)(CH3OCO2)(h3-PhCHCHCH2)]+. b) LUMO of
[Ru(Cp)(CH3CN)2(h3-PhCHCHCH2)]2+, witha large contribution from the
À
Figure 2. Structure of the dication [Ru(Cp*)(h3-CH2CHCHPh)(dmf)2]2+
in 4. Thermal ellipsoids are drawn at 30% probability; PF6 anions are
omitted for clarity. Bond lengths [pm] and angles [8]: Ru1-O1 211.0(3),
Ru1-O2 212.6(3), Ru1-C1 218.4(4), Ru1-C2 215.5(4), Ru1-C3 232.3(4),
Ru1-Ct1 186.0(4), Ru1-C10 220.1(4), Ru1-C20 216.3(4), Ru-C30
222.9(4), Ru1-C40 228.1(4), Ru1-C50 223.6(4), O1-C100 123.7(5), O2-
C200 125.8(5), N1-C100 133.2(6), N2-C200 129.6(6); C1-C2-C3
117.0(4), O1-Ru1-O2 79.7(1); f=68.28; Ct1=centroid of the Cp*
ligand; f is the intersection of the planes described by C1, C2, C3 and
the five atoms of the Cp* ring.
Ph C allyl carbon atom.
the acetonitrile ligand than either of the oxygen ligands.
Although the dicationic acetonitrile complex is substantially
different from the other two cations, it is not clear from the
À
X ray and DFT data why the linear allyl product (namely, Ph-
=
À
À
CH CH CH2 Ar) is formed preferentially. Indeed, relative
to palladium, ruthenium catalysts in allylation chemistry are
interesting because they can favor the formation of the
branched products.[9,10]
To understand the chemistry of these salts a bit further, we
performed DFT calculations[19] for the three model cations
[Ru(Cp)(CH3CN)2(h3-PhCHCHCH2)]2+, [Ru(Cp)(dmf)2(h3-
PhCHCHCH2)]2+, and [Ru(Cp)(CH3OCO2)(h3-PhCHCH-
A partial answer to this question is given by diffusion data
from pulsed-gradient spin echo (PGSE) NMR measurements
for complexes 3 and 5 (Table 2). This diffusion method-
CH2)]+ (Cp = C5H5). The results reveal that the Ru C bond
À
Table 2: D and rH values for the ruthenium complexes in acetonitrile at
lengths for the substituted, terminal allyl carbon atom
decrease as follows: 264 pm for the acetonitrile complex,
252 pm for the dmf complex, and 233 pm for the carbonate
299 K.[a]
À
species. The remaining Ru C(allyl) bond lengths are all in the
range 210–222 pm. Although the separations of 264 pm and
252 pm are longer than the X-ray values, they are qualita-
tively consistent with the solid-state structures. Furthermore,
natural population analysis (NPA) reveals the negative
charges on the three terminal phenylallyl carbon atoms to
be À0.05, À0.08, and À0.15, respectively, that is, the aceto-
nitrile dication has the least amount of negative charge on this
carbon atom. A possible reason can be deduced from the
three molecular orbitals in Figure 3a, which represent the
bonding metal–allyl p interactions. While the dmf and
carbonate donors contribute to the orbital of the p interac-
tion, this is not the case for the acetonitrile donors. This
observation suggests an electron flow from the oxygen donors
to the allyl ligand in the carbonate and dmf complexes, thus
rationalizing the charges on the allyl carbon atoms.
Nucleus
D[b]
rH[c] [Š]
3
5
1H (cation)
12.07
23.50
40.35
13.72
24.76
40.34
5.4
2.8
1.6
4.7
2.6
1.6
19F (anion)
1H (free CH3CN)
1H (cation)
19F (anion)
1H (free CH3CN)
[a] All at 2 mm. [b] Æ2%, ”1010 m2 sÀ1. [c] h(CH3CN, 299 K)=0.3377”
10À3 kgsÀ1 mÀ1
.
The LUMO of the acetonitrile complex shows a greater
participation of the substituted carbon atom (Figure 3b). If
our reaction were orbital-controlled, one would expect attack
at the substituted, terminal carbon and not, as found, at the
allylic CH2 carbon atom. Interestingly, there is less positive
charge on the Ru atom in the acetonitrile complex (+ 0.34)
than for the dmf (+ 0.49) and carbonate complexes (+ 0.51).
This difference is consistent with the better donor ability of
ology[20] allows one to estimate molecular volumes and ion
pairing in solution. From the measured diffusion coefficient
values D for the cation of 5, we estimate the hydrodynamic
radius rH (using the Stokes–Einstein equation) to be about
4.7 Š, in good agreement with estimates of the size of this
cation (4.5 Š) from our crystallographic data.[21] For the
Angew. Chem. Int. Ed. 2006, 45, 6386 –6391ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6389

Communications
Synlett 2003, 408; c) M. D. Mbaye, J. L. Renaud, B. Demerse-
man, C. Bruneau, Chem. Commun. 2004, 1870; d) M. D. Mbaye,
B. Demerseman, J. L. Renaud, L. Toupet, C. Bruneau, Adv.
Synth. Catal. 2004, 346, 835; e) N. Gurbuz, I. Ozdemir, B.
Cetinkaya, J. L. Renaud, B. Demerseman, C. Bruneau, Tetrahe-
dron Lett. 2006, 47, 535.
cation of 3, we obtain a much larger rH value of about 5.4 Š,
which can be the result of ion pairing.[20] However, the
diffusion data from the PF6 anions of 3 suggest that this is not
the case. The rH values for the PF6 anions of both 3 and 5 (2.6–
2.8 Š) are typical of what one finds in a polar solvent such as
methanol. We suggest that the larger rH value of 5.4 Š for the
cation of 3 arises as a result of some charge-induced
aggregation.[20d,e] The presence of an aggregate may make it
more difficult to attackthe substituted allylic carbon, so that
we tentatively attribute the observed regioselectivity to steric
effects as a result of aggregation.[22]
[11] a) R. Hermatschweiler, I. Fernandez, P. S. Pregosin, E. J.
Watson, A. Albinati, S. Rizzato, L. F. Veiros, M. J. Calhorda,
Organometallics 2005, 24, 1809; b) R. Hermatschweiler, I.
Fernandez, F. Breher, P. S. Pregosin, L. F. Veiros, M. J. Calhorda,
Angew. Chem. 2005, 117, 4471; Angew. Chem. Int. Ed. 2005, 44,
4397.
[12] a) I. Fernandez, R. Hermatschweiler, P. S. Pregosin, A. Albinati,
S. Rizzato, Organometallics 2006, 25, 323; b) R. Hermatsch-
weiler, I. Fernandez, P. S. Pregosin, F. Breher, Organometallics
2006, 25, 1440.
Perhaps there is a transition state in which the O (or
À
S) donor approaches the Ph C(allyl) position, but is hindered
À
so that the C C coupling is favored. To test this aggregation
assumption we have carried out a catalytic reaction at tenfold
dilution using 6-bromo-2-naphthol as the substrate. Indeed,
[13] a) Synthesis of 3: AgPF6 (48 mg, 0.174 mmol) was added to
a
solution of [Ru(Cp*)Cl(CH3CN)(h3-PhCHCHCH2)]PF6
(100 mg, 0.174 mmol) in a mixture of toluene and acetonitrile
(2 mL:2 mL). The reaction mixture was stirred for 16 h after
which time the solution was filtered and then slowly concen-
trated under vacuum. The resulting crude solid was washed with
diethyl ether to afford an oil, which was dissolved in dichloro-
methane, filtered, and dried under vacuum to afford a brown-
yellow solid. This solid was dissolved in dichloromethane,
filtered, and dried under vacuum. This sequence was repeated
one more time. Yield: 78% (98 mg). An acetone solution of this
solid was layered with n-pentane and stored at 58C to afford air-
sensitive crystals of 3, which were suitable for X-ray diffraction.
1H NMR ([D6]acetone, 298 K, 400.13 MHz): d = 1.96 (s, 15H),
2.37 (s, 3H), 2.68 (s, 3H), 3.41 (d, 1H, J = 10.4 Hz), 4.87 (d, 1H,
J = 6.4 Hz), 5.31 (d, 1H, J = 12.0 Hz), 6.59 (ddd, 1H, J = 12.0,
10.4, 6.4 Hz), 7.58 (m, 2H, J = 7.6, 7.2 Hz), 7.74 (m, 2H, J = 7.6,
1.4 Hz), 7.90 ppm (m, 2H, J = 7.2 Hz); 13C NMR: d = 3.7 (CH3),
4.0 (CH3), 9.2 (CH3), 66.3 (H2Callyl), 108.8 (C), 94.1 (HCallyl),
103.3 (HCallyl), 129.1 (Cnitrile), 129.2 (Cnitrile), 129.9 (HCAr), 131.1
(HCAr), 132.6 (HCAr), 133.6 ppm (Cipso). Elemental analysis
(%) calcd for C23H30F12N2P2Ru: C 38.08, H 4.17, N 3.86; found:
C 38.46, H 4.38, N 3.20; ESI-MS: m/z: 436.1 [M+], 354.1
[M+À2CH3CN], 237.0 [M+À2CH3CNÀPhCHCHCH2). Synthe-
sis of 4: AgPF6 (48 mg, 0.174 mmol) was added to a solution
À
whereas we had previously found no C O product, we found
À
13% C O bond formation in the diluted reaction mixture.
In conclusion, we have found that a new dicationic
RuIV salt is a very efficient and mild catalyst for the
Friedel–Crafts-type allylation of various electron-rich arene
substrates. This C C coupling reaction is in contrast to the
selective C O bond-forming reactions observed with related
À
À
catalysts.
Received: February 2, 2006
Revised: May 22, 2006
Published online: August 23, 2006
Keywords: allylations · density functional calculations ·
NMR spectroscopy · ruthenium · structure elucidation
.
[1] J. Takaya, J. F. Hartwig, J. Am. Chem. Soc. 2005, 127, 5756.
[2] B. M. Trost, M. T. Rudd, J. Am. Chem. Soc. 2005, 127, 4763;
B. M. Trost, M. U. Frederiksen, M. T. Rudd, Angew. Chem. 2005,
117, 6788; Angew. Chem. Int. Ed. 2005, 44, 6630.
[3] M. Ito, S. Itahara, T. Ikariya, J. Am. Chem. Soc. 2005, 127, 6172.
[4] R. C. van der Drift, M. Gagliardo, H. Kooijman, A. L. Spek, E.
Bouwman, E. Drent, J. Organomet. Chem. 2005, 690, 1044.
[5] S. Hashiguchi, A. Fujii, K. J. Haack, K. Matsumura, T. Ikariya,
R. Noyori, Angew. Chem. 1997, 109, 300; Angew. Chem. Int. Ed.
Engl. 1997, 36, 288.
[6] a) T. Ohkuma, H. Ooka, T. Ikariya, R. Noyori, J. Am. Chem. Soc.
1995, 117, 10417; b) T. Ohkuma, H. Ooka, M. Yamakawa, T.
Ikariya, R. Noyori, J. Org. Chem. 1996, 61, 4872; c) K. J. Haack,
S. Hashiguchi, A. Fujii, T. Ikariya, R. Noyori, Angew. Chem.
1997, 109, 297; Angew. Chem. Int. Ed. Engl. 1997, 36, 285.
[7] R. H. Grubbs, Tetrahedron 2004, 60, 7117; A. Hejl, O. A.
Scherman, R. H. Grubbs, Macromolecules 2005, 38, 7214.
[8] a) P. H. Dixneuf, C. Bruneau in Organic Synthesis via Organo-
metallics OSM 5 (Eds.: D. Helmchen, J. Dibo, J. Flubacher, B.
Wiese), Friedr. Vieweg, Braunschwewig, 1997, p. 1; b) S. Derien,
P. H. Dixneuf, J. Organomet. Chem. 2004, 689, 1382; c) E.
Bustelo, P. H. Dixneuf, Adv. Synth. Catal. 2005, 347, 393; T.
Murahashi, Ruthenium in Organic Synthesis, Wiley-VCH, Wein-
heim, 2004.
of
[Ru(Cp*)Cl(CH3CN)(h3-PhCHCHCH2)]PF6
(100 mg,
0.174 mmol) in DMF (2 mL). The reaction mixture was stirred
for 16 h after which time the solution was filtered and then
slowly concentrated under vacuum. The resulting crude mixture
was washed with diethyl ether. The resulting red oil was
dissolved in dichloromethane, filtered, and dried under
vacuum to afford a red-purple solid. This solid was dissolved in
dichloromethane, filtered, and dried under vacuum. This
sequence was repeated one more time. Yield: 84% (115 mg).
A dichloromethane solution of this solid was then layered with
n-pentane and stored at À308C to afford red air-sensitive
crystals of 4, which were suitable for X-ray diffraction. 1H NMR
([D6]acetone, 298 K, 400.13 MHz): d = 1.76 (s, 15H), 2.60 (d, 3H,
J = 1.0 Hz), 3.05 (s, 3H), 3.13 (d, 3H, J = 1.0 Hz), 3.28 (s, 3H),
3.64 (dd, 1H, J = 10.0, 0.9 Hz), 4.70 (dd, 1H, J = 6.5, 0.9 Hz), 5.49
(d, 1H, J = 11.0 Hz), 6.48 (ddd, 1H, J = 11.0, 10.0, 6.5 Hz), 7.12
(s, 1H), 7.50 (m, 2H, J = 7.8, 7.5 Hz), 7.72 (m, 1H, J = 7.8,
1.5 Hz), 7.75 (m, 2H, J = 7.5 Hz), 7.95 ppm (s, 1H); 13C NMR:
d = 8.9 (CH3), 33.2 (CH3), 33.8 (CH3), 39.2 (CH3), 39.4 (CH3),
66.0 (H2Callyl), 107.9 (C), 96.9 (HCallyl), 98.4 (HCallyl), 129.9
(HCAr), 131.5 (HCAr), 131.6 (HCAr), 134.0 (Cipso), 166.5 (Cdmf),
[9] a) B. M. Trost, P. L. Fraisse, Z. T. Ball, Angew. Chem. 2002, 114,
1101; Angew. Chem. Int. Ed. 2002, 41, 1059; b) B. M. Trost, M. L.
Crawley, Chem. Rev. 2003, 103, 2921.
[10] a) M. D. Mbaye, B. Demerseman, J. L. Renaud, L. Toupet, C.
Bruneau, Angew. Chem. 2003, 115, 5220; Angew. Chem. Int. Ed.
2003, 42, 5066; b) J. L. Renaud, C. Bruneau, B. Demerseman,
167.8 ppm
(Cdmf);
ESI-MS:
427.1
m/z:
459.1
[M+
354.1
ÀMe2NCHO+MeOH),
[M+ÀMe2NCHO],
[M+À2Me2NCHO]. Catalysis: In a typical experiment, the
Ru catalyst precursor 3 or 4 (0.002 mmol, 3 mol%) was added
to a mixture consisting of acetonitrile (0.5 mL) and the allylic
6390 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6386 –6391

Angewandte
Chemie
carbonate substrate (0.07 mmol) in an oven-dried 5-mm NMR
tube. The arene derivative (0.21 mmol) was added, and the
mixture was monitored by 1H NMR spectroscopy at 353 K.
Crystal structure of 3: Yellow crystals of [Ru(Cp*)(h3-
CH2CHCHPh)(MeCN)2](PF6)2·acetone were obtained from an
[18] a) Y. Nishibayashi, M. Yamanashi, Y. Takagi, M. Hidai, Chem.
Commun. 1997, 859; b) G. Onodera, H. Imajima, M. Yamanashi,
Y. Nishibayashi, M. Hidai, S. Uemura, Organometallics 2004, 23,
5841, and references therein.
[19] DFT calculations were performed with the Gaussian98 software
package using the mPW1PW91 hybrid functional. The basis set
that was used included a standard SDD which was augmented
with an f-polarization function for Ru and 4-31G(d) for the other
atoms. The atomic charges resulted from a natural population
analysis (NPA). Computational details and the corresponding
list of references are given as Supporting Information.
[20] a) M. Valentini, H. Rüegger, P. S. Pregosin, Helv. Chim. Acta
2001, 84, 2833; b) P. S. Pregosin, E. Martinez-Viviente, P. G. A.
Kumar, Dalton Trans. 2003, 4007; c) P. S. Pregosin, P. G. A.
Kumar, I. Fernandez, Chem. Rev. 2005, 105, 2977; d) F. Q. Song,
S. J. Lancaster, R. D. Cannon, M. Schormann, S. M. Humphrey,
C. Zuccaccia, A. Macchioni, M. Bochmann, Organometallics
2005, 24, 1315; e) D. Zuccaccia, E. Clot, A. Macchioni, New J.
Chem. 2005, 29, 430.
acetone solution of
3
which was layered with n-hexane;
¯
C26H36F12N2OP2Ru, triclinic, space group P1; a = 9.535(1), b =
11.750(1), c = 15.879(1) Š, a = 109.395(1), b = 96.719(1), g =
100.965(1)8, V= 1615.8(2) Š3, Z = 2, 1calcd = 1.611 MgmÀ3, crys-
tal dimensions 0.48 ” 0.27 ” 0.23 mm, Bruker SMART Apex
diffractometer with CCD detector, MoKa radiation
(0.71073 Š), 200 K, 2Vmax = 56.688, 16798 reflections, 7984
independent (Rint = 0.0217), empirical absorption correction
SADABS (ver. 2.03), direct methods, refinement against full
matrix (versus F2) with SHELXTL (ver. 6.12) and SHELXL-97,
406 parameters, R1 = 0.0504 and wR2 (all data) = 0.1363, max./
min. residual electron density 1.106/À0.464 eŠÀ3. All non-
hydrogen atoms were refined anisotropically; the contribution
of the hydrogen atoms, in their calculated positions, was included
in the refinement using a riding model. Crystal structure of 4:
[21] The radii from the X-ray data for the cations in 3, 4, and 5 are 4.6,
4.8, and 4.5 Š, respectively.
[22] In a typical catalytic experiment the concentration of the catalyst
precursor is 4.2 mm. The PGSE diffusion studies were performed
at 2 mm, so that, in the catalytic solution, even more aggregation
is expected.
Red
(PF6)2·CH2Cl2·alkane were obtained from a methylene chloride
solution, which was layered with n-hexane;
crystals
of
[Ru(Cp*)(h3-CH2CHCHPh)(dmf)2]-
C28H40Cl2F12N2O2P2Ru, orthorhombic, space group Pccn; a =
21.369(1), b = 22.712(1), c = 15.580(1) Š, V= 7561.4(6) Š3, Z =
8, 1calcd = 1.579 MgmÀ3, crystal dimensions 0.44 ” 0.42 ” 0.32 mm,
Bruker CCD1k diffractometer, MoKa radiation (0.71073 Š),
200 K, 2qmax = 52.748, 58019 reflections, 7735 independent
(Rint = 0.0272), empirical absorption correction SADABS
(ver. 2.03), direct methods, refinement against full matrix
(versus F2) with SHELXTL (ver. 6.12) and SHELXL-97, 515
parameters, R1 = 0.0457 and wR2 (all data) = 0.1558, max./min.
residual electron density 1.193/À0.568 eŠÀ3. All non-hydrogen
atoms were refined anisotropically; the contribution of the
hydrogen atoms, in their calculated positions, was included in the
refinement using a riding model. One CH2Cl2 molecule could be
refined. Presumably owing to solvent loss or incomplete
incorporation, additional crystal solvent molecule(s) could only
be refined by positioning two carbon atoms (” 0.5C4), one of
which had to be split over two positions and refined using the
ISOR restraint. One of the PF6À anions was disordered. Each of
the fluorine atoms was split over two positions, which were
refined against each other (FVAR = 0.53). CCDC- 295048 and
295049 contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.
ac.uk/data_request/cif. b) A reviewer has correctly suggested
that the Friedel–Crafts chemistry might arise from a rearrange-
ment of the branched organic product PhCH(OPh)CHCH2
formed from the nucleophilic attackof phenol on the Ru IV allyl
complex. To test this we allowed this product, PhCH-
(OPh)CHCH2, to react with catalyst 3 under our standard
conditions (CH3CN at 353 K). After 12 h, that is, more than one
order of magnitude longer than necessary for 100% conversion,
we did find about 60% of the Friedel–Crafts product. However,
this reaction is much too slow to be relevant.
[14] G. Olah, J. Org. Chem. 2001, 66, 5943.
[15] A. Fürstner, D. Voigtlander, W. Schrader, D. Giebel, M. T.
Reetz, Org. Lett. 2001, 3, 417.
[16] a) I. Shimizu, T. Sakamoto, S. Kawaragi, Y. Maruyama, A.
Yamamoto, Chem. Lett. 1997, 137; b) A. V. Malkov, S. L. Davis,
I. R. Baxendale, W. L. Mitchell, P. Kocovsky, J. Org. Chem. 1999,
64, 2751, and references therein; c) J. Choudhury, S. Podder, S.
Roy, J. Am. Chem. Soc. 2005, 127, 6162.
[17] M. Bandini, A. Melloni, A. Umani-Ronchi, Angew. Chem. 2004,
116, 560; Angew. Chem. Int. Ed. 2004, 43, 550.
Angew. Chem. Int. Ed. 2006, 45, 6386 –6391ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6391