Synthesis, X-ray studies, and catalytic allylic amination reactions with ruthenium(IV) allyl carbonate complexes

Article abstract of DOI:10.1021/om050749o

Results from selected ruthenium-catalyzed allylic amination reactions, using a carbonate Ru-Cp* (or Cp) allyl cationic precursor, are reported. With a phenyl-substituted allyl substrate as the starting material, the rates of the amination reactions, as well as the regioselectivity, are shown to depend on the structure of the allyl substrate, the amine nucleophile, and the solvent. Two new Ru-allyl carbonate complexes are reported, as well as the solid-state structure for the new carbonate salt [Ru(Cp)(O₂C{OBu ^t})(η³-PhCHCHCH₂)](PF₆). A number of aniline- and substrate-related Cp* and Cp η⁶- arene complexes of Ru(II) are described.

Full text of DOI:10.1021/om050749o

Organometallics 2006, 25, 323-330
323
Articles
Synthesis, X-ray Studies, and Catalytic Allylic Amination Reactions
with Ruthenium(IV) Allyl Carbonate Complexes
Ignacio Ferna´ndez, Rene´ Hermatschweiler, and Paul S. Pregosin*
Laboratory of Inorganic Chemistry, ETH HCI, Ho¨nggerberg CH-8093 Zu¨rich, Switzerland
Alberto Albinati* and Silvia Rizzato
Department of Structural Chemistry (DCSSI), UniVersity of Milan, 20133 Milan, Italy
ReceiVed September 1, 2005
Results from selected ruthenium-catalyzed allylic amination reactions, using a carbonate Ru-Cp* (or
Cp) allyl cationic precursor, are reported. With a phenyl-substituted allyl substrate as the starting material,
the rates of the amination reactions, as well as the regioselectivity, are shown to depend on the structure
of the allyl substrate, the amine nucleophile, and the solvent. Two new Ru-allyl carbonate complexes
are reported, as well as the solid-state structure for the new carbonate salt [Ru(Cp)(O2C{OBut})(η3-
PhCHCHCH2)](PF6). A number of aniline- and substrate-related Cp* and Cp η6-arene complexes of
Ru(II) are described.
Chart 1
Introduction
The allylic alkylation reaction can be catalyzed by a number
of transition-metal complexes. The most commonly used metal
is palladium;1 however, molybdenum,2a iridium,2b and ruthe-
nium3 are all currently in use. The interest in Ru(II) complexes
is based on the observed regioselectivity3 in that, for unsym-
metrical allyl substrates, the preferred site of attack leads to a
branched, rather than a linear, product.
The most commonly used ruthenium catalyst precursor
contains a Cp or Cp* ligand. Trost and co-workers3 have
reported that [Ru(Cp* or Cp)(CH3CN)3](PF6) (1a,b, respec-
tively) are excellent catalysts for this reaction and, specifically,
that with the Cp* complex, 1a, reaction of the allyl substrate
PhCHdCHCH2X (X ) carbonate (2a), chloride (2b)) with a
carbon nucleophile, Nu-, occurs preferentially at the more
substituted position (see eq 1).
PhCHdCHCH2X + Nu- [Ru(Cp*)(CH3CN)3](PF6)8
DMF or acetone
however, although their catalysts are selective (and some are
enantioselective7), they are relatively slow. Several reports8-10
suggest that Ru(1,5-COD) complexes such as 4,10 or even the
relatively simple chloro derivative 5,9 are also useful catalysts
for this allylation chemistry.
PhCH(Nu)CHdCH2 + X- (1)
Bruneau4-7 and co-workers have used chelating nitrogen
complexes of Ru(II) for this reaction (e.g. 3; see Chart 1);
We have recently reported11 that the source of the observed
high branched-to-linear regioselectivity, for X ) Cl, has an
(1) Trost, B. M.; van Vranken, D. Chem. ReV. 1996, 96, 395-422.
(2) (a) Krska, S. W.; Hughes, D. L.; Reamer, R. A.; Mathre, D. J.;
Palucki, M.; Yasuda, N.; Sun, Y.; Trost, B. M. Pure Appl. Chem. 2004,
76, 625-633. (b) Leitner, A.; Shu, C. T.; Hartwig, J. F. Proc. Natl. Acad.
Sci. U.S.A. 2004, 101, 5830-5833.
(3) Trost, B. M.; Fraisse, P. L.; Ball, Z. T. Angew. Chem., Int. Ed. 2003,
41, 1059-1061.
(4) (a) Mbaye, M. D.; Demerseman, B.; Renaud, J.-L.; Bruneau, C. J.
Organomet. Chem. 2005, 690, 2149-2158. (b) Mbaye, M. D.; Demerseman,
B.; Renaud, J. L.; Toupet, L.; Bruneau, C. Angew. Chem., Int. Ed. 2003,
42, 5066-5068.
(7) Renaud, J. L.; Bruneau, C.; Demerseman, B. Synlett 2003, 408-
410.
(8) Kondo, H.; Kageyama, A.; Yamaguchi, Y.; Haga, M.; Kirchner, K.;
Nagashima, H. Bull. Chem. Soc. Jpn. 2001, 74, 1927-1937. Morisaki, Y.;
Kondo, T.; Mitsudo, T. A. Organometallics 1999, 18, 4742-4746. Kondo,
T.; Morisaki, Y.; Uenoyama, S.; Wada, K.; Mitsudo, T. J. Am. Chem. Soc.
1999, 121, 8657-8658. Zhang, S.; Mitsudo, T.; Kondo, T.; Watanabe, Y.
J. Organomet. Chem. 1993, 450, 197-207.
(9) Burger, E. C.; Tunge, J. A. Org. Lett. 2004, 6, 2603-2605.
(10) Yamamoto, Y.; Nakagai, Y.; Itoh, K. Chem. Eur. J. 2004, 10, 231-
236.
(5) Mbaye, M. D.; Renaud, J. L.; Demerseman, B.; Bruneau, C. Chem.
Commun. 2004, 1870-1871.
(6) Mbaye, M. D.; Demerseman, B.; Renaud, J. L.; Toupet, L.; Bruneau,
C. Angew. Chem., Int. Ed. 2003, 42, 5066-5068.
10.1021/om050749o CCC: $33.50 © 2006 American Chemical Society
Publication on Web 12/09/2005

324 Organometallics, Vol. 25, No. 2, 2006
Ferna´ndez et al.
electronic origin11 on the basis of X-ray and NMR data, together
with DFT calculations. These conclusions were based on studies
of the Ru(IV) allyl complex [Ru(Cp*)Cl(CH3CN)(η3-PhCHCH-
CH2)](PF6) (6), which was prepared via the reaction of the allyl
substrate 2b with 1a (eq 2). We have also synthesized12 the
revealed that the phenylallyl ligands in these cationic complexes
are distorted, such that the bond distance from the metal to the
allyl carbon adjacent to the phenyl group is much longer than
the corresponding distance to the methylene allyl carbon.
Subsequently, we have used the allyl carbonate complex 7a
as a catalyst12 for the allylic alkylation reaction using the
dimethyl malonate anion as the nucleophile and find relatively
rapid formation of the organic products. Specifically, a com-
parison of the rate of complete conversion to organic products
(followed by NMR) for 7a vs 1a, using the branched carbonate
substrate, revealed that complex 7a was 25 times faster than
the tris(nitrile) 1a. Using the linear substrate, catalyst 7a is only
a factor of 2 faster than 1a.
We report here new synthetic and structural studies on Ru(IV)
Cp and Cp* carbonate complexes, e.g. the Cp analogue 7b, as
novel Ru(IV) carbonate complex [Ru(Cp*){OC(OBut)O}(η3-
PhCHCHCH2)](PF6) (7b) from the branched tert-butyl carbonate
plus 1a in DMF at ambient temperature (eq 3). A number of
well the results of a series of Ru-catalyzed allylic amination
reactions. These new catalytic results suggest an important role
for the solvent in the catalysis and raise some questions with
respect to the generality of the regioselectivity of ruthenium in
this amination reaction. There are not many examples of allylic
amination using Ru(II).4,8,19
Results and Discussion
Catalytic Allylic Amination. The results from Ru-catalyzed
allylic amination reactions (eq 4) using (a) the branched and
linear substrates 9 and 10, respectively, (b) a number of different
catalysts, including 1, and (c) several amine nucleophiles are
given in Tables 1-5.
Ru-allyl complexes are known:4,6,13-18 e.g., 8 in Chart 1.
The solid-state structure12 of this unexpected carbonate, 7a,
reveals that the tert-butyl carbonate ligand is coordinated in a
bidentate fashion via two oxygen atoms, with the allyl ligand
showing an endo configuration such that the phenyl substituent
is remote from the bulky Cp* ligand. Obviously, the oxidative
addition affords the monocationic carbonate and not the bis-
(acetonitrile) dicationic complex. It is often thought that the
carbonate leaving group rapidly decomposes to alkoxide and
CO2.3 DFT calculations11,12 carried out on model complexes
Several important points are immediately obvious.
1. In acetonitrile solution, both the branched/linear product
ratio and the reaction rate depend markedly on the structure of
the allyl carbonate, with the branched isomer reacting quickly
and the linear analogue slowly or not at all (see Table 1). In
some, but not all, cases the branched/linear ratio is good to
excellent: e.g., for the aniline nucleophiles. The use of aliphatic
amines affords mixed results. The reaction can be either faster
or slower than that for aniline, and the branched/linear ratio
can be good (morpholine, entry 3, 91:1) or poor (diethylamine,
entry 6, 33:67; triethylamine, entry 7, 0:100).
2. The reactions with the CpRu carbonate catalyst 7b (instead
of the Cp* catalyst; see Table 2) were slower but afforded
similar (but not identical) branched/linear product ratios. Entries
2-4 are based on 9 mol % instead of 3 mol % catalyst, so that
(11) Hermatschweiler, R.; Fernandez, I.; Pregosin, P. S.; Watson, E. J.;
Albinati, A.; Rizzato, S.; Veiros, L. F.; Calhorda, M. J. Organometallics
2005, 24, 1809-1812.
(12) Hermatschweiler, R.; Fernandez, I.; Breher, F.; Pregosin, P. S.;
Veiros, L. F.; Calhorda, M. J. Angew. Chem., Int. Ed. 2005, 44, 4397-
4400.
(13) Becker, E.; Slugovc, C.; Ruba, E.; Standfest-Hauser, C.; Mereiter,
K.; Schmid, R.; Kirchner, K. J. Organomet. Chem. 2002, 649, 55-63.
(14) Slugovc, C.; Ruba, E.; Schmid, R.; Kirchner, K. Organometallics
1999, 18, 4230-4233.
(15) Ruba, E.; Simanko, W.; Mauthner, K.; Soldouzi, K. M.; Slugovc,
C.; Mereiter, K.; Schmid, R.; Kirchner, K. Organometallics 1999, 18, 3843-
3850.
(16) Gemel, C.; Kalt, D.; Mereiter, K.; Sapunov, V. N.; Schmid, R.;
Kirchner, K. Organometallics 1997, 16, 427-433.
(17) Cadierno, V.; Diez, J.; Garcia-Garrido, S. E.; Gimeno, J. Chem.
Commun. 2004, 2716-2717. Cadierno, V.; Garcia-Garrido, S. E.; Gimeno,
J. Chem. Commun. 2004, 232-233.
(18) Kondo, H.; Yamaguchi, Y.; Nagashima, H. Chem. Commun. 2000,
1075-1076.
(19) Matsushima, Y.; Onitsuka, K.; Kondo, T.; Mitsudo, T.; Takahashi,
S. J. Am. Chem. Soc. 2001, 123, 10405-10406.

Ruthenium(IV) Allyl Carbonate Complexes
Organometallics, Vol. 25, No. 2, 2006 325
Table 1. Selected Cp* Ruthenium Catalyzed Allylic
Amination Reactions Using the Cp* Carbonate Complex 7a
in Acetonitrile Solutiona
Table 4. Comparison of 1a and 1b in the
Ruthenium-Catalyzed Allylic Amination Reactions in
Acetonitrilea
entry
substrate
amine
t (min)
branched/linearb
entry
substrate
[Ru]
amine
t (min)
branched/linearb
Branched Carbonate
tert-butylamine
morpholine
1
2
3
4
9
9
9
9
1a
morpholine
aniline
morpholine
aniline
28
15
65
e
91:9c
93:7
91:9
1
2
3
5
6
7
8
9
10
11
9
9
9
9
9
9
9
9
9
9
110
27
30
62
6
60:40c
91:9
91:9
83:17
33:67
0:100e
93:7
95:5
93:7
85:15
1a
1bd
1bd
morpholined
piperidine
diethylamine
a
Conditions: 0.07 mmol of the branched carbonate substrate 9, 0.21
mmol of amine, 0.002 mmol of catalyst (3 mol %), in 0.5 mL of acetonitrile,
room temperature. b All reactions led to 100% of conversion. The branched/
linear ratio was determined by 1H NMR spectroscopy. c 3% of isomerized
linear carbonate 10 was also detected. d 0.006 mmol of catalyst (9 mol %).
e No conversion after 24 h.
triethylamine
aniline
1320
16
4-methoxyaniline
4-fluoroaniline
4-nitroaniline
12
15
96
Linear Carbonate
morpholine
aniline
Table 5. Selected Ruthenium-Catalyzed Allylic Amination
Reactions Using a Substrate with an n-Propyl (Instead of
Phenyl) Side Chaina
12
13
10
10
1920
68:32
f
f
a Conditions: 0.07 mmol of the carbonate substrate, 0.21 mmol of amine,
0.002 mmol of catalyst (3 mol %), in 0.5 mL of acetonitrile, room
temperature. b All reactions led to 100% conversion. The branched/linear
ratio was determined by 1H NMR spectroscopy. c 6% of isomerized linear
carbonate 10 was also detected. d 0.08 mmol of amine was employed. e We
believe this to be a hydroxide salt. f No conversion after 24 h.
sub-
branched/
linearb
entry strate [Ru]
amine
solvent
t (min)20
Table 2. Selected Ruthenium-Catalyzed Allylic Amination
Reactions Using the Cp Carbonate Complex 7b in
Acetonitrile Solutiona
1
2
3
4
5
5
11
11
11
12
12
12
7a
7a
7a
7a
7a
7a
morpholine MeCN
morpholine acetone
360
1
c
45:55
29:71
c
21:79
10:90
c
aniline
MeCN
morpholine MeCN
morpholine acetone
1200 (20 h)
entry
substrate
amine
t (min)
branched/linearb
1
c
1
9
9
9
9
morpholine
morpholine
aniline
210
70
d
89:11
89:11
aniline
MeCN
2c
3c
4c
a Conditions: 0.07 mmol of the carbonate substrate, 0.21 mmol of amine,
0.002 mmol of catalyst (3 mol %), in 0.5 mL of acetonitrile, room
temperature. The entry “1” for time, in the table, indicates only that the
reaction was finished by the time the sample could be measured in the
NMR spectrometer. b All reactions led to 100% of conversion. The
branched/linear ratio was determined by 1H NMR spectroscopy. c No
conversion after 24 h.
4-fluoroaniline
840 (14 h)
86:14
a Conditions: 0.07 mmol of the branched carbonate substrate 9, 0.21
mmol of amine, 0.002 mmol of catalyst (3 mol %), in 0.5 mL of acetonitrile,
room temperature. b All reactions led to 100% conversion. The branched/
linear ratio was determined by 1H NMR spectroscopy. c 0.006 mmol of
catalyst (9 mol %). d No conversion after 36 h.
1a is much faster than 1b (9 mol % was used for 1b instead of
3 mol % with 1a).
5. The amination reactions with the two n-propyl substrates
11 and 12 (see Table 5), gave poor branched/linear ratios and
were very slow in acetonitrile solution. Again, the catalyses were
Table 3. Selected Ruthenium-Catalyzed Allylic Amination
Reactions with Morpholine in Acetonea with Different
Catalysts
entry
substrate
[Ru]
amine
t (min)a
branched/linearb
1
2
3
4
5
9
10
9
9
9
7a
7a
morpholine
morpholine
morpholine
morpholine
morpholine
1
1
1
1
1
29:71
11:89
80:20d
61:39
82:18
7bc
1a
1bc
a Conditions: 0.07 mmol of the carbonate substrate, 0.21 mmol of amine,
0.002 mmol of catalyst (3 mol %), in 0.5 mL of acetone, room temperature.
The entry “1” for time, in the table, indicates only that the reaction was
finished by the time the sample could be measured in the NMR spectrometer.
b All reactions led to 100% conversion. The branched/linear ratio was
determined by 1H NMR spectroscopy. c 0.006 mmol of catalyst (9 mol %).
d 2% of isomerized linear carbonate 10 was also detected.
rapid in acetone. Obviously, the presence of a phenyl group in
the substrate plays a role, as does the solvent. However, the
substitution of an n-propyl group for a phenyl group does not
interfere with the oxidative addition reaction. We have been
able to prepare the Ru(IV) carbonate analogue of 7a, complex
13, in good yield, via reaction with substrate 11 (see Experi-
mental Section).
the times shown (for 100% conversion) are much slower than
for the data in Table 1. The lack of conversion for aniline will
be discussed below.
3. In acetone solution (instead of acetonitrile; see Table 3)
with morpholine as the nitrogen nucleophile, the reaction is
relatively fast, i.e., ca. 100% conversion in about 1 min;
however, the branched/linear ratio is poor. Interestingly, the best
branched/linear ratio is found for 1b, Trost’s Cp complex. This
complex is reported3 to give a poorer branched/linear ratio,
relative to Cp*, in the alkylation reaction.
4. With the Trost catalysts 1a and 1b (see Table 4), the
regioselectivity is as good as or better than that with 7a. Catalyst
The reaction with triethylamine, the only tertiary amine tried,
afforded a surprising result: 100% of the cationic linear

326 Organometallics, Vol. 25, No. 2, 2006
Ferna´ndez et al.
Chart 2. Monocationic Arene Complexes of Ru(II) as PF6
Salts
Figure 1. Isomerization of the branched tert-butyl carbonate to
the linear isomer in CD3CN using 3 mol % of both [RuCp*(CH3-
CN)3](PF6) (black) and [RuCp(CH3CN)3](PF6) (red) (0.0021 mmol).
The salt and the branched tert-butyl carbonate (0.07 mmol) were
dissolved in CD3CN, and the isomerization was monitored by H
NMR spectroscopy.
reported.23 Indeed, reaction of a small excess of linear substrate
10 with 1a or 1b affords the η6-arene complexes 15 and 16 in
good yield (see Chart 2). These complexes show the ex-
pected25,26 low-frequency proton and carbon resonances associ-
ated with a complexed aromatic moiety. The anisidine complex
17a was tried as the catalyst for the reaction of anisidine with
the branched carbonate 9. There is no conversion at all after 16
h.
1
compound 14, which was not isolated20 but readily identified
1
via its H and 13C NMR spectra. Apart from the two olefinic
With no excess of linear substrate, i.e., in the stoichiometric
1:1 reaction, we do not find complete formation of the arene
complex 15. Interestingly, the analogous reaction with the
branched substrate 9 immediately affords the Ru(IV) allyl 7a,
via an oxidative addition reaction, in high yield. We have already
shown via DFT calculations12 that there should be a ca. 3 kcal/
mol energy difference between the two arene complexes, with
the branched isomer being higher in energy. Moreover, the
leaving group is much further away from the ruthenium atom
in 15 than it would be in the isomeric arene complex from the
branched isomer. Consequently we suggest that the observed
differences in rate between these two substrates, in the catalytic
amination, are related to the time necessary for the η6-arene to
dissociate, thus allowing the oxidative addition to proceed.
Continuing, given the lack of reactivity of aniline with
substrate 10 when 7b is the catalyst (Table 1) as well as the
failure of aniline to react with the n-propyl substrates 11 and
12 (see Table 5), we considered the possibility of intermediates
such as the η6-arene complexes 17 and 18, also shown in Chart
2. In fact, 17a-c are readily obtained via reaction of [RuCp*-
(CH3CN)3](PF6) and aniline in acetone solution at room
temperature. These complexes, as purple powders, can be
precipitated from acetone in excellent yield by addition of
diethyl ether. The rate of formation of the η6-anisidine complex,
in CD3CN solution, starting from the Trost complexes 1a,b (see
Figure 2), can be monitored using 1H NMR spectroscopy.
Clearly, the Cp* analogue forms the arene complex at a faster
rate. We conclude that arene complexes are likely to exist in
acetonitrile solution. Indeed, in an in situ experiment using 5
equiv of substrate 10, 5 equiv of anisidine, and 1 equiv of
catalyst 7a, a small amount of the η6-aniline derivative can be
3
resonances (δ 6.31 and 7.00, J(H,H) ) 15.7 Hz), and the
methylene protons (δ 3.96), the three equivalent ethyl groups
are readily visible. We are not certain as to the anion.
The various branched/linear ratios (Table 1) might arise,
partially, from isomerization processes. Indeed, one can show
that the branched substrate 9 can isomerize to 10 (see Figure
1); however, this seems to be a rather slow reaction when
compared to the catalytic amination. An alternate isomerization
pathway, suggested by a reviewer, concerns isomerization of
the product rather than the starting material. In a second repeat
experiment, we have allowed anisidine to react with 9 in the
presence of our catalyst 7a (see Table 1, entry 9). After 14 min
we find a ca. 94:6 branched/linear ratio. We then let this reaction
mixture stand for 16 h and found the branched/linear ratio to
be ca. 88:12. Once again there is slow isomerization. We
conclude that our observed branched/linear ratios in Table 1
may indeed be affected by isomerization reactions;21 however,
isomerization is not likely to be important for the fast reactions.
Arene Complexes of Ru(II). Given the differences observed
in the rates of the catalytic experiments, especially those found
between the branched and linear substrates, 9 and 10 in Table
2, it seemed likely that η6-arene complexes of Ru(II) might be
formed:22-24 e.g., the cation RuCp*(η6-C6H5F)+ has been
(20) In the paper by Matsushima et al. (Matsushima, Y.; Onitsuka, K.;
Takahashi, S. Organometallics 2004, 23, 3763-3765), a Ru(II) compound
containing the olefin-complexed cation Et3NCHdCH2 is suggested as
“possible”. The authors do not give either NMR or solid-state data for this
structure.
(21) In a report by Cadierno et al. (Cadierno, V.; Garcia-Garrido, S. E.;
Gimeno, J.; Nebra, N. Chem. Commun. 2005, 4086-4088), the authors
report oxidative addition of the C-N bond of an allylamine using ruthenium.
(22) This type of Cp-arene complex is well-known; see: Elschenbroich,
C.; Salzer, A. Organometallics, 2nd ed.; VCH: Weinheim, Germany, 1992;
pp 356-357.
(23) Aneetha, H.; Jimenez-Tenorio, M.; Puerta, M. C.; Valerga, P.;
Sapunov, V. N.; Schmid, R.; Kirchner, K.; Mereiter, K. Organometallics
2002, 21, 5334-5346.
(24) Liu, S. H.; Huang, X.; Ng, W. S.; Wen, T. B.; Lo, M. F.; Zhou, Z.
Y.; Williams, I. D.; Lin, Z. Y.; Jia, G. C. Organometallics 2003, 22, 904-
906.
(25) Mann, B. E.; Taylor, B. F. 13C NMR Data for Organometallic
Compounds; Academic Press: London, 1981.
(26) den Reijer, C. J.; Dotta, P.; Pregosin, P. S.; Albinati, A. Can. J.
Chem. 2001, 79, 693-704. den Reijer, C. J.; Drago, D.; Pregosin, P. S.
Organometallics 2001, 20, 2982-2989. Geldbach, T. J.; den Reijer, C. J.;
Worle, M.; Pregosin, P. S. Inorg. Chim. Acta 2002, 330, 155-160.

Ruthenium(IV) Allyl Carbonate Complexes
Organometallics, Vol. 25, No. 2, 2006 327
Table 7. 1H and 13C NMR Data for η6-Carbonate
Complexes 16a and 16b in CD3CN at 299 K
Figure 2. Formation of the η6-aniline complexes in CD3CN with
[RuCp*(CH3CN)3](PF6) (black) and [RuCp(CH3CN)3](PF6) (red).
The salt, e.g., [RuCp*(CH3CN)3](PF6) (0.012 mmol), and anisidine
(0.012 mmol) were dissolved in CD3CN, and the η6-aniline complex
1
formation was monitored by H NMR spectroscopy.
Table 8. 1H and 13C NMR Data for the Cp* Aniline
Complexes 17a-c in CD3CN at 299 K
Table 6. 1H and 13C NMR Data for Cp* η6-Carbonate
Complexes 15a and 15b in DMF-d7 at 299 K
Table 9. Bond Lengths (Å) and Angles (deg) for 7b and 7a
1
observed via H NMR. Presumably, in acetone solution, these
7b
7a
η6-aniline derivatives are much more labile such that more
reactive organometallic complexes are formed. Selected NMR
data for these arene complexes are given in Tables 6-8.
Solid-State Studies. Reaction of the Cp salt 1b with the
branched allyl precursor 9, in acetone solution, gave the new
carbonate complex 7b in good yield. The 13C resonances for
the allyl ligand, δ 62.3 (dCH2), 97.6 (dHC), and 99.9
(dHCPh), are consistent with our previous findings.11,12
X-ray Study of 7b. The solid-state structure of 7b was
determined by X-ray diffraction methods, and selected bond
angles and bond distances are given in Table 9, along with some
comparison data for the Cp* cation 7a. This is only the second
reported structure of such an allyl carbonate complex. A view
of the cation is given in Figure 3.
The immediate coordination sphere for this coordinatively
saturated Ru(IV) cation consists of an Ru atom surrounded by
the Cp ligand, the η3-allyl ligand, and the tert-butyl carbonate
ligand. The last group is clearly coordinated in a bidentate
fashion via the two oxygen atoms. The O-Ru-O bite angle is
small, ca. 62°, and the two Ru-O separations, 2.116(2) and
Ru-O(1)
Ru-O(2)
2.116(2)
2.124(3)
Ru1-O
Ru1-O
2.148(3)
2.116(3)
Ru-C(1)
Ru-C(2)
Ru-C(3)
2.190(4)
2.150(3)
2.294(4)
Ru1-C
Ru1-C
Ru1-C
2.162(5)
2.137(5)
2.303(5)
Ru-C(1P)
Ru-C(2P)
Ru-C(3P)
Ru-C(4P)
Ru-C(5P)
2.177(4)
2.236(4)
2.236(4)
2.184(4)
2.168(4)
Ru-C(10)
Ru-C(20)
Ru-C(30)
Ru-C(40)
Ru-C(50)
2.192(3)
2.206(3)
2.214(3)
2.257(3)
2.239(3)
O(1)-Ru-O(2)
62.1(1)
2.124(3) Å, are not significantly different and correspond to
literature expectations.27 However, the three Ru-C(allyl) dis-
tances, Ru-C(1) ) 2.190(4) Å, Ru-C(2) ) 2.150(3) Å, and
Ru-C(3) ) 2.294(4) Å, are all different, with the Ru-C(H)Ph
separation being much larger than the other two. Similar
distortions have been observed previously,11,12 and the pattern
is similar to that found for the Cp* cation (see Table 9).
(27) Kuznetsov, V. F.; Jefferson, G. R.; Yap, G. P. A.; Alper, H.
Organometallics 2002, 21, 4241-4248.

328 Organometallics, Vol. 25, No. 2, 2006
Ferna´ndez et al.
Figure 3. ORTEP view of the Ru(II) cationic complex 7b. The PF6 anion is omitted for clarity.
Kondo et al.18 have reported the structure of 19 and find Ru-
C(allyl) separations of 2.193(5) and 2.206(3) Å, for the two
terminal carbons, and 2.132(3) Å for the central allyl carbon.
Clearly, the Ru-C(3) value of 2.294(4) Å is relatively long.
were dried and distilled by using standard procedures and stored
under nitrogen. NMR spectra were recorded with Bruker DPX 300,
400, and 500 MHz spectrometers at room temperature. Chemical
shifts are given in ppm and coupling constants (J) in hertz.
Elemental analyses and mass spectroscopic studies were performed
at the ETHZ.
Crystallography. Air-stable orange crystals of 7b suitable for
X-ray diffraction were obtained by layering pentane in a CH2Cl2
solution of the isolated complex. A crystal of 7b was mounted on
a Bruker SMART diffractometer, equipped with a CCD detector,
and cooled, using a cold nitrogen stream, to 150(2) K for the data
collection. The space group was determined from the systematic
absences, while the cell constants were refined at the end of the
data collection with the data reduction software SAINT.28 The
experimental conditions for the data collection and crystallographic
and other relevant data are given in the Supporting Information.
The collected intensities were corrected for Lorentz and polarization
factors28 and empirically for absorption using the SADABS
program.29
The average Ru-C(Cp) distance in 7b is 2.212 (4) Å, only
marginally different (4σ) from that found in the Cp* cation 7a,
at ca. 2.221 Å.
Conclusions. For our Ru(II) catalysts, it would seem that
there are a number of variables which are important with respect
to the nature of the amine product formed in our reactions. The
allyl substrate, the structure of the nucleophile, and certainly
the reaction solvent all play important roles in this allylic
amination chemistry. There are indications that the differing
observed reaction rates are, partly, related to formation of η6-
arene complexes. Moreover, while it is possible to greatly
accelerate these catalytic reactions via the use of acetone as
solvent, the desired regioselectivity is lost. Obviously, there are
subtle changes in the organometallic chemistry which are not
yet understood and these can lead to catalytic results that are
not readily explainable. Specifically, from Table 3, one notes
that changing from the Cp* catalyst 7a (entry 1) to the Cp
catalyst 7b (entry 3) results in a substantial change of the
preferred regioselectivity. Clearly, further studies in this area
will be necessary.
The structure was solved by direct and Fourier methods and
refined by full-matrix least squares,30 minimizing the function
∑w(Fo2 - (1/k)Fc2)2 and using anisotropic displacement parameters
for all atoms, except the hydrogens and those affected by disorder
(see below).
The difference Fourier maps clearly showed severe disorder of
the fluorine atoms in the equatorial plane of the PF6 octahedron,
even at low temperature. A model was constructed allowing 12
different positions for the equatorial fluorine atoms. During the
refinement the sum of the site occupancy factors was constrained
to obtain the correct stoichiometry. This model clearly shows that
(28) BrukerAXS. SAINT Integration Software; Bruker Analytical X-ray
Systems, Madison, WI, 1995.
(29) Sheldrick, G. M. SADABS, Program for Absorption Correction;
University of Go¨ttingen, Go¨ttingen, Germany, 1996.
(30) Sheldrick, G. M. SHELX-97. Structure Solution and Refinement
Package; University of Go¨ttingen, Go¨ttingen, Germany, 1997.
Experimental Section
All reactions and manipulations were performed under a N2
atmosphere using standard Schlenk techniques. Solvents and amines

Ruthenium(IV) Allyl Carbonate Complexes
Organometallics, Vol. 25, No. 2, 2006 329
there is an almost continuous scattering density in the equatorial
plane and that this anion behaves in a manner resembling a spinning
top.
1JNC ) 2.8 Hz, C-1), 115.4 (HCd, C-2), 142.0 (HCd, C-3), 127.6
(HCAr, C-5), 129.2 (CAr, C-6), 129.1 (CAr, C-7), 153.7 (Cipso
,
C-4). We are assuming hydroxide is the anion, although this has
not been proven.
No extinction correction was deemed necessary. Upon conver-
gence the final Fourier difference map showed no significant peaks.
The contribution of the hydrogen atoms in their calculated positions
was included in the refinement using a riding model (B(H) )
a[B(Cbonded)] (Å2), where a ) 1.5 for the hydrogen atoms of the
methyl groups and a ) 1.2 for the others). The scattering factors
used, corrected for the real and imaginary parts of the anomalous
dispersion, were taken from the literature.31 The standard deviations
on intensities were calculated in terms of statistics alone. All
calculations were carried out by using the PC version of the
programs WINGX,32 SHELX-97,30 and ORTEP.33
Catalytic Experiments. In a typical experiment, a 0.07 mmol
sample of allylic carbonate 9 or 10 was added to a mixture
consisting of acetonitrile (0.5 mL) and the Ru catalyst (0.002 mmol,
3% mol) in an oven-dried 5 mm NMR tube. The amine derivative
(0.21 mmol) was added, and the mixture was monitored by 1H NMR
spectroscopy at room temperature. Modifications to these experi-
mental conditions are reported in the tables.
[RuCp(O2C{OBut})(η3-phenylallyl)]PF6 (7b). [RuCp(CH3-
CN)3]PF6 (100 mg, 0.230 mmol) and branched phenyl tert-butyl
carbonate (57.0 mg, 0.242 mmol) were stirred in acetone (2 mL)
for 15 min at room temperature. The solution volume was reduced
under vacuum, and Et2O was added, precipitating an orange-brown
powder. The solid was washed with diethyl ether (Et2O) and dried
under vacuum to yield 114 mg (91%) of crude product. Crystals
suitable for X-ray study were obtained by layering pentane in a
CH2Cl2 solution of the isolated complex. NMR (Me2CO-d6, 299
K): 1H, δ 1.46 (9H), 4.80 (1H, J ) 11.0 Hz), 4.81 (1H, J ) 6.5
Hz), 5.95 (1H, J ) 11.3, 6.5 Hz), 6.32 (5H), 6.45 (1H, J ) 11.3
Hz), 7.43 (2H, J ) 7.7, 7.3 Hz), 7.62 (1H, J ) 7.7), 7.77 (2H, J )
7.3); 13C, δ 27.9 (CH3), 62.3 (dCH2), 95.8 (C), 97.6 (CH), 99.9
(CH), 129.4 (HCAr), 131.2 (HCAr), 131.3 (HCAr), 134.5 (Cipso),
164.1 (CO3). Anal. Calcd for C19H23O3F6PRu: C, 41.84; H, 4.25;
Found: C, 40.80; H, 4.20. ESI MS: m/z 401.1 (M+), 301.1 (M+
- OC(OBut)O + H2O), 283.1 (M+ - OC(OBut)O).
[RuCp*{η6-(PhCHdCHOCO2But)}]PF6 (15a). [Cp*Ru(CH3-
CN)3]PF6 (100 mg, 0.198 mmol) was added to a stirred solution of
tert-butyl cinnamyl carbonate (140 mg, 0.595 mmol, 3 equiv) in 2
mL of acetone, and the brown solution was stirred at room
temperature for 1 h. The solvent was removed under vacuum, and
the product was precipitated twice from acetone/pentane and washed
with Et2O. Crystallization from acetone/Et2O (diffusion) afforded
a yellow-brown crystalline powder. Yield: 115 mg (94%). NMR
(DMF-d7, 299 K): 1H, δ 1.51 (9H), 1.98 (15H), 4.85 (2H, J ) 5.7,
1.6 Hz), 6.11 (1ArH, J ) 5.7, 1.5 Hz), 6.16 (1ArH, J ) 6.0, 5.7
Hz), 6.34 (1ArH, J ) 6.0 Hz), 6.49 (1H, J ) 16.0, 1.5 Hz), 6.71
(1H, J ) 16.0, 5.7 Hz); 13C, δ 10.0 (CH3), 27.5 (CH3), 66.6 (CH2),
82.3 (C), 85.8 (HCAr), 88.2 (HCAr), 88.3 (HCAr), 96.8 (Cipso),
97.4 (C), 127.4 (HCd) 131.4 (HCd), 153.7 (CO3). Anal. Calcd
for C24H33O3RuPF6: C, 46.83; H, 5.40. Found: C, 46.75; H, 5.57.
MS (ESI): m/z 471.2 (M+), 401, 371.2 (M+ - OBut), 371.2 (M+
- CO2But), 355.3 (M+ - OCO2But), 315.3 (M+ - C3H4OCO2-
But).
[RuCp*{η6-(p-OMe-C6H4CHdCHOCO2But)}]PF6 (15b). [Cp*-
Ru(CH3CN)3]PF6 (100 mg, 0.198 mmol) was added to a stirred
solution of tert-butyl para-methoxycinnamyl carbonate (157 mg,
0.595 mmol, 3 equiv) in 2 mL of acetone, and the brown solution
was stirred at room temperature for 1 h. The solvent was removed
under vacuum, and the product was precipitated twice from acetone/
pentane and washed with Et2O. Crystallization from acetone/Et2O
(diffusion) afforded a yellow-brown crystalline powder. Yield: 122
mg (95%). NMR (DMF-d7, 299 K): 1H, δ 1.51 (9H), 1.95 (15H),
3.92 (3H), 4.83 (2H, J ) 5.7, 1.5 Hz), 6.22 (1ArH, J ) 6.7 Hz),
6.29 (1ArH, J ) 6.7 Hz), 6.46 (1H, J ) 15.7, 1.5 Hz), 6.67 (1H,
J ) 15.7, 5.7 Hz); 13C, δ 9.9 (CH3), 27.5 (CH3), 57.2 (OMe), 66.6
(CH2), 76.5 (HCAr), 82.3 (C), 84.7 (HCAr), 95.2 (Cipso), 96.1 (C),
126.8 (HCd), 131.0 (HCd), 153.7 (CO3). Anal. Calcd for C25H35O4-
RuPF6: C, 46.51; H, 5.46. Found: C, 46.30; H, 5.47. MS (ESI):
m/z 501.2 (M+), 442.2 (M+ - But), 385.2 (M+ - OCO2But), 315.3
(M+ - C3H4OCO2But - MeO).
[RuCp{η6-(PhCHdCHOCO2But)}]PF6 (16a). tert-Butyl cin-
namyl carbonate (135 mg, 0.57 mmol, ca. 5 equiv) was added to a
stirred solution of [CpRu(CH3CN)3]PF6 (50 mg, 0.115 mmol) in 2
mL of acetone, and the brown solution was stirred at room
temperature for 2 h. The solvent was removed under vacuum,
affording a brownish oil, which was precipitated from CH2Cl2/
hexane at 4 °C and washed with Et2O. Yield: 56 mg (89%). NMR
(CD3CN, 299 K): 1H, δ 1.50 (9H), 4.69 (2H, J ) 5.0, 1.2 Hz),
5.31 (5H, Cp), 6.10 (1ArH, J ) 5.8, 1.1 Hz), 6.17 (2ArH, J )
5.8.7 Hz), 6.33 (2ArH, J ) 5.7 Hz), 6.43 (1H, J ) 15.9 Hz), 6.51
(1H, J ) 15.9, 4.9 Hz); 13C, δ 27.3 (CH3), 66.2 (CH2), 81.2 (C),
81.4 (CH), 84.4 (HCAr), 85.9 (HCAr), 86.1 (HCAr), 99.3 (Cipso),
127.3 (HCd), 131.6 (HCd), 153.5 (CO3). Anal. Calcd for C19H23O3-
RuPF6: C, 41.84; H, 4.25. Found: C, 41.95; H, 4.23. MS (ESI):
m/z 401.1 (M+), 345.0 (M+ - But), 301.0 (M+ - CO2But).
[RuCp*(O2C{OBut})(η3-n-propylallyl)]PF6 (13). [RuCp*(CH3-
CN)3]PF6 (50 mg, 0.099 mmol) and branched n-propyl tert-butyl
carbonate (19.8 mg, 0.099 mmol) were stirred in acetone (1.5 mL)
for 30 min at room temperature. The solution volume was reduced
under vacuum, and Et2O was added, precipitating a yellow-brown
powder. The solid was washed with Et2O and dried under vacuum
to yield 54.7 mg (95%) of crude product. NMR (DMF-d7, 299 K):
1H, δ 1.69 (9H), 1.71-2.02 (4H), 1.93 (15H), 3.48 (1H, J ) 10.0
Hz), 4.26 (1H, J ) 10.0, 10.0, 3.5 Hz), 4.66 (1H, J ) 6.5 Hz),
5.54 (1H, J ) 10.0, 10.0, 6.5 Hz), 7; 13C, δ 8.8 (CH3), 13.93 (CH3),
23.3 (CH2), 28.2 (CH3), 33.6 (CH2), 67.4 (dCH2), 86.2 (C), 92.4
(dCH), 106.1 (dCH), 107.4 (C), 164.5 (CO3). Anal. Calcd for
C21H35O3F6PRu: C, 43.37; H, 6.07. Found: C, 42.63; H, 5.76. ESI
MS: m/z 437.1 (M+), 319.2 (M+ - OC(OBut)O).
N-Triethyl-3-phenylprop-2-ene Ammonium Salt (14). NMR
[RuCp{η6-(p-OMe-C6H4CHdCHOCO2But)}]PF6 (16b). tert-
Butyl para-methoxycinnamyl carbonate (91 mg, 0.35 mmol, ca. 5
equiv) was added to a stirred solution of [CpRu(CH3CN)3]PF6 (30
mg, 0.069 mmol) in 1.5 mL of acetone, and the brown solution
was stirred at room temperature for 2 h. The solvent was removed
under vacuum, affording a deep brown oil, which was precipitated
twice from CH2Cl2/hexane at 4 °C and washed with Et2O. Yield:
35 mg (87%). NMR (CD3CN, 299 K): 1H, δ 1.49 (s, 9H), 3.76 (s,
3H), 4.66 (d, 2H, J ) 4.8, 0.9 Hz), 5.29 (s, 5H, Cp), 6.20 (2ArH,
J ) 6.9 Hz), 6.26 (2ArH, J ) 6.9 Hz), 6.37 (1H, J ) 16.1 Hz),
6.43 (1H, J ) 16.1, 4.7 Hz); 13C, δ 26.9 (CH3), 57.1 (OCH3), 65.8
(CH2), 74.0 (HCAr), 80.2 (C), 80.6 (CH), 82.2 (HCAr), 96.2 (Cipso),
126.4 (HCd), 130.7 (HCd), 134.4 (Cipso), 153.1 (CO3). Anal. Calcd
3
(CD3CN-d7, 299 K, 400.13 MHz): 1H, δ 1.32 (3H, J ) 7.1, JNH
) 1.7, H-9), 3.28 (2H, J ) 7.2 Hz, H-8), 3.96 (2H, J ) 7.6 Hz,
H-1), 6.31 (1H, J ) 15.7, 7.6 Hz, H-2), 7.00 (1H, J ) 15.7 Hz,
H-3), 7.32-7.43 (3H, H-5, 6), 7.59 (2H, J ) 6.6, 1.1 Hz, H-5);
13C, δ 7.3 (CH3, C-9), 53.0 (CH2, 1JNC ) 2.9 Hz, C-8), 59.4 (CH2,
(31) International Tables for X-ray Crystallography; Wilson, A. J. C.,
Ed.; Kluwer Academic: Dordrecht, The Netherlands. 1992; Vol. C.
(32) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837.
(33) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.

330 Organometallics, Vol. 25, No. 2, 2006
Ferna´ndez et al.
for C20H25O4RuPF6: C, 41.74; H, 4.38. Found: C, 41.97; H, 4.41.
MS (ESI): m/z 431.0 (M+), 375.0 (M+ - But), 331.1 (M+ - CO2-
But).
precipitated with Et2O. The purple solid was washed with Et2O to
yield 33 mg (82%) of product. NMR (CD3CN, 299 K): 1H, δ 3.66
(s, 3H), 4.49 (s, 2H, NH2), 5.21 (s, 5H, Cp), 5.71 (2ArH), 5.92
(2ArH); 13C, δ 57.0 (OCH3), 68.3 (HCAr), 71.9 (HCAr), 79.3 (CH),
122.5 (Cipso), 130.2 (Cipso). Anal. Calcd for C12H14NOF6PRu: C,
33.19; H, 3.25; N, 3.23; Found: C, 34.34; H, 3.40; N, 3.40. ESI
MS: m/z 290.1 (M+).
[RuCp(η6-p-toluidine)]PF6 (18b). [RuCp(CH3CN)3]PF6 (30 mg,
0.069 mmol) and p-toluidine (11 mg, 0.103 mmol) were stirred in
acetone (1.5 mL) for 30 min at room temperature. The solvent was
concentrated under vacuum, and the product was precipitated with
Et2O. The light purple solid was washed with Et2O to yield 21 mg
(73%) of product. NMR (CD3CN, 299 K): 1H, δ 2.18 (s, 3H), 4.53
(s, 2H, NH2), 5.15 (s, 5H, Cp), 5.75 (2ArH), 5.87 (2ArH); 13C, δ
18.9 (CH3), 70.1 (HCAr), 79.5 (CH), 84.7 (HCAr), 96.6 (Cipso),
124.2 (Cipso). Anal. Calcd for C12H14NF6PRu: C, 34.46; H, 3.37;
N, 3.35. Found: C, 35.67; H, 3.78; N, 3.45. ESI MS: m/z 274.0
(M+).
[RuCp(η6-p-chloroaniline)]PF6 (18c). [RuCp(CH3CN)3]PF6 (40
mg, 0.092 mmol) and p-chloroaniline (46 mg, 0.361 mmol) were
stirred in acetone (1.5 mL) for 30 min at room temperature. The
solvent was concentrated under vacuum, and the product was
precipitated with Et2O. The purple solid was washed with Et2O to
yield 20 mg (50%) of product. NMR (CD3CN, 299 K): 1H, δ 4.78
(s, 2H, NH2), 5.28 (s, 5H, Cp), 5.85 (2ArH), 6.28 (2ArH); 13C, δ
69.7 (HCAr), 81.3 (CH), 85.1 (HCAr), 121.2 (Cipso), 125.4 (Cipso).
Anal. Calcd for C11H11NF6PClRu; C, 30.12; H, 3.53; N, 3.19.
Found: C, 31.47; H, 2.81; N, 3.47. ESI MS: m/z 294.0 (M+).
[RuCp*(η6-anisidine)]PF6 (17a). RuCp*(CH3CN)3]PF6 (50 mg,
0.099 mmol) and anisidine (12.2 mg, 0.099 mmol) were stirred in
acetone (1.5 mL) for 20 min at room temperature. The solution
volume was reduced under vacuum, and Et2O was added, precipi-
tating a purple powder. The solid was washed with Et2O and dried
under vacuum to yield 46 mg (92%) of crude product. NMR (CD3-
CN, 299 K): 1H, δ 1.91 (15H), 3.68 (3H), 4.31 (2H), 5.22 (2H, J
) 6.5 Hz), 5.57 (2H, J ) 6.5 Hz); 13C, δ 9.7 (CH3), 56.8 (OCH3),
71.7 (HCAr), 73.9 (HCAr), 94.6 (C), 120.4 (Cipso), 129.0 (Cipso).
Anal. Calcd for C17H24NOF6PRu: C, 40.48; H, 4.80; N, 2.78.
Found: C, 41.11; H, 4.88; N, 3.10. ESI MS: m/z 360.2 (M+).
[RuCp*(η6-p-toluidine)]PF6 (17b). [RuCp*(CH3CN)3]PF6 (45
mg, 0.079 mmol) and p-toluidine (8.6 mg, 0.08 mmol) were stirred
in acetone (1.5 mL) for 40 min at room temperature. The solution
volume was reduced under vacuum, and Et2O was added, precipi-
tating a pale purple powder. The solid was washed with Et2O and
dried under vacuum to yield 38 mg (98%) of crude product. NMR
(CD3CN-d7, 299 K, 500.23 MHz): 1H, δ 1.99 (15H), 2.18 (3H),
4.51 (2H), 5.36 (2H, J ) 6.0 Hz), 5.56 (2H, J ) 6.0 Hz); 13C, δ
9.5 (CH3), 17.2 (CH3), 73.4 (HCAr), 86.7 (HCAr), 94.3 (C), 95.5
(Cipso), 122.0 (Cipso). Anal. Calcd for C17H24NF6PRu: C, 41.81; H,
4.95; N, 2.87. Found: C, 41.96; H, 5.00; N, 2.96. ESI MS: m/z
344.2 (M+).
[RuCp*(η6-p-fluoroaniline)]PF6 (17c). [RuCp*(CH3CN)3]PF6
(45 mg, 0.079 mmol) and p-fluoroaniline (7.7 µl, 0.08 mmol) were
stirred in acetone (1.5 mL) for 60 min at room temperature. The
solution volume was reduced under vacuum, and Et2O was added,
precipitating a pale purple powder. The solid was washed with Et2O
and dried under vacuum to yield 35 mg (80%) of crude product.
NMR (CD3CN-d7, 299 K, 500.23 MHz): 1H, δ 1.94 (15H), 4.53
(2H), 5.32 (2H, J ) 6.5, JFH 1.5 Hz), 5.88 (2H, J ) 6.5, JFH 3.0
Hz); 13C, δ 9.6 (CH3), 71.5 (HCAr, JCF ) 6.7 Hz), 76.7 (HCAr,
JCF ) 23.1 Hz), 96.0 (C), 122.1 (Cipso), 132.4 (Cipso, JCF ) 267.3
Hz). Anal. Calcd for C16H21NF7PRu: C, 39.03; H, 4.30; N, 2.84.
Found: C, 39.20; H, 4.30; N, 2.96. ESI MS: m/z 348.1 (M+).
[RuCp(η6-anisidine)]PF6 (18a). [RuCp(CH3CN)3]PF6 (40 mg,
0.092 mmol) and p-methoxyaniline (15 mg, 0.122 mmol) were
stirred in acetone (1.5 mL) for 30 min at room temperature. The
solvent was concentrated under vacuum, and the product was
Acknowledgment. The support and sponsorship provided
by COST Action D24 “Sustainable Chemical Processes: Ste-
reoselective Transition Metal-Catalyzed Reactions” are kindly
acknowledged. Further, P.S.P. thanks the Swiss National Science
Foundation and the ETHZ for financial support. A.A. thanks
MURST for a grant (PRIN 2002). I.F. thanks the Junta de
Andalucia for a research contract. We also thank Johnson
Matthey for the loan of precious metals.
Supporting Information Available: A CIF file giving crystal-
lographic data for compound 7b. This material is available free of
charge via the Internet at http://pubs.acs.org.
OM050749O