Quantification of the β-stabilizing effect of the dicarbonyl(η 5-cyclopentadienyl)iron group

Article abstract of DOI:10.1002/hlca.200590137

Kinetic investigations of the reactions of (prop-2-enyl) dicarbonyl(cyclopentadienyl)iron complexes 1 with benzhydrylium ions 3, and of dicarbonyl(cyclopentadienyl)[(1,2-η)propene]iron(II) tetrafluoroborate (9 · BF₄) with π-nucleophiles have be

Full text of DOI:10.1002/hlca.200590137

1754
Helvetica Chimica Acta Vol. 88 (2005)
Quantification of the b-Stabilizing Effect of the
Dicarbonyl(h5-cyclopentadienyl)iron Group
by Felix Dulich1), Karl-Heinz M¸ller2), Armin R. Ofial, and Herbert Mayr*
Department Chemie und Biochemie der Ludwig-Maximilians-Universit‰t M¸nchen,
Butenandtstra˚e 5 13 (Haus F), D-81377 M¸nchen, Germany
(fax: ‡ 49-89-2180-77717; e-mail: herbert.mayr@cup.uni-muenchen.de)
Dedicated to Professor Rolf Huisgen on the occasion of his 85th birthday
Kinetic investigations of the reactions of (prop-2-enyl)dicarbonyl(cyclopentadienyl)iron complexes 1 with
benzhydrylium ions 3, and of dicarbonyl(cyclopentadienyl)[(1,2-h)propene]iron(II) tetrafluoroborate (9 ¥ BF4)
with p-nucleophiles have been performed to elucidate the magnitude of the b-effect of the [(CO)2FeCp] group
(Fp group). Introduction of the Fp group into the allylic position of propene and 2-methylpropene increases the
nucleophilicity of the p-bonds by nine and six orders of magnitude, respectively, with the result that the allyl-Fp
complexes 1a (N ˆ 6.78) and 1b (N ˆ 8.45) are among the strongest neutral p-nucleophiles. Replacement of one
b-H-atom in the isopropyl cation by the Fp group reduces the electrophilicity by more than 20 orders of
‡
magnitude, so that 9 ranks among the weakest cationic C-electrophiles (E ˆ À11.2).
1. Introduction. The nucleophilic reactivity of CˆC bonds can be enhanced by
electropositive elements in allylic position. In previous work [1][2], we have shown that
the reactivities of CH2ˆCHÀCH2R towards C-electrophiles increases considerably in
the series R ˆ H < Pr< Me3Si < Bu3Sn, with relative reaction rates (krel) toward
‡
[Ar2CH] of 1.0 (H), 1.5 (Pr), 1.6Â 104 (Me3Si), and 7.4 Â 107 (Bu3Sn) in CH2Cl2 at 208.
The comparison of compounds differing by eight orders of magnitude in reactivity
was achieved by employing benzhydrylium ions of variable electrophilicity as reference
electrophiles [1][3]. We have now employed the same method for the characterization
of the nucleophilic reactivities of allyl compounds that carry a dicarbonyl(h5-cyclo-
pentadienyl)iron (Fp) moiety, and we will show that the b-activating effect of the Fp
group even exceeds that of the Bu3Sn group by more than one order of magnitude.
While the reactions of electrophiles with allyl silanes, stannanes, and other
metalated compounds are usually accompanied by successive demetalation, the
reactions of allyl-Fp complexes [4] with electrophiles have been reported to give
persistent h2-Fp cations [5 8], as shown for the reaction of 1a with tropylium
tetrafluoroborate (Scheme 1) [9]. When 1a is combined with a neutral, electron-
deficient CˆC bond, a 1,5-dipole is formed, which may cyclize to a five-membered ring,
in which the Fp group has moved across the CˆC bond, as exemplified by the reaction
of 1a with dimethyl methylidenemalonate (2; Scheme 2) [4].
The literature on many analogous formations of carbocyclic and heterocyclic five-
membered-ring compounds [10 13] has been reviewed [14][15]. These reactions
indicate a high nucleophilicity of allyl-Fp complexes such as 1a [16][17] and 1b [18],
1
)
)
Present address: GSF-Forschungszentrum f¸r Umwelt und Gesundheit, Neuherberg, Germany.
Present address: CIBA Spezialit‰tenchemie, Lampertheim, Germany
2
¹ 2005 Verlag Helvetica Chimica Acta AG, Z¸rich

Helvetica Chimica Acta Vol. 88 (2005)
1755
Scheme 1
Scheme 2
‡
which prompted us to employ the weak electrophiles 3 listed in Table 1 for kinetic
investigations.
2. Results and Discussion. 2.1. Synthesis of Allyl-Fp Complexes. The allyl-Fp
complexes 1a and 1b were prepared by nucleophilic substitution of the corresponding
allyl chlorides (Scheme 3) following literature procedures [16 18]. For the reduction
of the commercially available bimetallic complex di-m-carbonyldicarbonylbis(h5-cyclo-
pentadienyl)diiron (4), a variety of methods have been employed so far, as summarized
by K¸hn and R¸ck-Braun [19][20]. Because all of these procedures suffer from certain
deficits, we synthesized K[FeCp(CO)2] (5) by treatment of 4 with potassium-graphite
laminate, C8K [21], which was obtained according to the procedure described by
Rabinovitz and co-workers [22].
Scheme 3
2.2. Electrophilic Addition to Allyl-Fp Complexes. Treatment of 1a with the
benzhydrylium salts 3d ¥ BF4 and 3e ¥ PF6 gave the p-complexes 6a ¥ BF4 and 6b ¥ PF6,
respectively, which were recrystallized from CH2Cl2/Et2O (Scheme 4). The difference
1
in hybridization of the allyl fragments in 1a and 6a was reflected by their H-NMR
spectra. While the former indicated a typical vinyl group, with terminal H-atom

1756
Helvetica Chimica Acta Vol. 88 (2005)
‡
Table 1. Structures and ElectrophilicityParameters (E) of Benzhydrylium Ions 3 Used in Reactions with
Compounds 1a and 1b. The E values were taken from [3].
Compound
Formula
E
‡
3a
À 4.72
‡
3b
À 5.53
À 5.89
‡
3c
‡
3d
À 7.02
À 7.69
‡
3e
‡
3f
À 8.76
resonances at d(H) 4.54 and 4.81, the corresponding vinyl H-atoms of 6a were more
shielded, resonating at d(H) 3.42 and 3.81, respectively. The conversion of the h1-
complex 1a into the cationic h2-complex 6a is accompanied by a reduction in electron
density at the Fe-atom, which results in a shift of the carbonyl IR frequencies from 1948
and 2010 cmÀ1 in 1a [16] to 2035 and 2075 cmÀ1 in 6a ¥ BF4.

Helvetica Chimica Acta Vol. 88 (2005)
1757
Scheme 4
We were not able to purify the analogous addition products obtained from the 2-
methylpropenyl complex 1b. Therefore, the course of the reaction with the benzhy-
drylium salts was elucidated by treatment of the crude reaction mixtures with NaI in
acetone, which resulted in the metal-free compounds 7a c (Scheme 4).
2.3. Kinetics. Kinetic experiments were performed in CH2Cl2 under pseudo-first-
order conditions, with at least 10 mol-equiv. of the allyl-Fp complexes 1a,b relative to
the benzhydrylium salts 3. Under these conditions, exponential decays of the UV/VIS
absorption maxima of the benzhydrylium ions at 600 640 nm were observed, from
which the pseudo-first-order rate constants kobs were derived. Second-order rate
constants (k2; Table 2) were derived from the relation kobs ˆ k2 [Nu].
‡
Table 2. Second-Order Rate Constants (k2) for the Reaction of 1 with Benzhydrylium Ions 3 . In CH2Cl2 at 208.
The E values were taken from [3].
Nucleophile
Electrophile
E
k2 [mÀ1 sÀ1
]
‡
1a
3a
À 4.72
À 5.53
À 5.89
À 7.02
À 7.69
À 5.89
À 7.02
À 7.69
À 8.76
98.9a)
11.9
7.26
0.694
0.128
145
15.8
3.43
0.631
‡
3b
‡
3c
‡
3d
‡
3e
3c
‡
1b
‡
3d
‡
3e
3f
‡
a) Extrapolated with the Eyring activation parameters DH≥ ˆ 28.35 Æ 0.59 kJ molÀ1 and DS≥ ˆ À109.9 Æ 2.5 J
molÀ1 KÀ1 obtained from measurements in the temp. range À 50 to À 208.
Plots of log k2 for these reactions vs. the empirical electrophilicity parameters E
showed linear correlations (Fig. 1), implying that these reactions follow Eqn. 1, where
electrophiles are characterized by the electrophilicity parameter E, and nucleophiles

1758
Helvetica Chimica Acta Vol. 88 (2005)
are characterized by the nucleophilicity parameters N and s, the temperature being
fixed at 208.
log k ˆ s (N ‡ E)
(1)
The similarity of the slope parameters s for compounds 1 and for other compounds
with terminal CˆC bonds [1] implies that the relative reactivities are almost
independent of the electrophile. As a consequence, the comparison of nucleophilic
reactivities can be based on the nucleophilicity parameters N, as illustrated in Fig. 2.
Fig. 1. Correlation between the Empirical ElectrophilicityParameter E and the Second-Order Rate Constants k2
‡
(at 208 in CH2Cl2) for the Reactions of the Fp Complexes 1a,b with Different Benzhydrylium Ions 3
Obviously, the introduction of an Fp moiety into the allylic position of propene
increases N by almost nine logarithmic units, while the reactivity of 2-methylpropene
grows only by seven logarithmic units (see Fig. 2, bottom vs. top). The different
behavior of both series can be rationalized by the reduced electron demand of the
tertiary carbenium center arising from 2-methylpropene, compared with the secondary
carbenium center arising from propene. In both series, the allyl compounds (Fig. 2, left
row) and their 2-methyl analogues (right row), the b-activating effect of the Fp group
exceeds that of the corresponding Sn compounds considerably, and is responsible for
the fact that compounds 1a,b rank among the strongest neutral p-nucleophiles,
comparable to silylated −ketene acetals× [1][23], only exceeded by enamines [24].
2.4. The Electrophilicityof Alkene-Fp Cations. An alternative approach to the
quantification of b-effects is the comparison of the electrophilicities of the b-substituted
‡
carbocations 8 . Quantum-chemical calculations show that increasing hyperconjuga-
‡
tive stabilization of the positive charge in 8 by the CÀR s-bond is associated with a
‡
decrease of the C ÀCÀR angle a [25]. For R ˆ Fp, bending of the hyperconjugating
‡
bond is so strong that the structure is generally represented by formula 9 .

Helvetica Chimica Acta Vol. 88 (2005)
1759
Fig. 2. Comparison of the Empirical NucleophilicityParameters N of the Fp Complexes 1a,b with those of
Analogous Allyl Stannanes, Silanes, and Germanes
‡
Several methods for the generation of the propene-Fp complex 9 have been
reported [10][16][26][27], including treatment of propene with FpBr/AlCl3, hydride
abstraction from propyl-Fp with triphenylmethylium (−tritylium×) ions, protonation of
1a, exchange of the olefin ligand in the corresponding cationic 2-methylpropene-Fp
complex, among others. In this work, the protonation of 1a has been optimized
(Scheme 5). Exposure to HBF4 in a mixture of CH2Cl2, Et2O, and propanoic anhydride
yielded a crystalline precipitate of pure 9 ¥ BF4 in almost quantitative yield.

1760
Helvetica Chimica Acta Vol. 88 (2005)
Scheme 5
Whereas the reaction between 9 ¥ BF4 and cyclohex-1-enylpyrrolidine was reported
to afford a 1:1-mixture of two regioisomers [28], 9 ¥ BF4 and the higher ring-homologue
10a reacted in CH2Cl2 at 08 regioselectively to 11 ¥ BF4, which was not isolated, but
demetalated and hydrolyzed to 2-(isopropyl)cyclohexanone (12; Scheme 6,a). The
−ketene acetal× 10b reacted analogously. Again, the sensitive initial product 13 was not
isolated, but demetalated either with HCl or CeIV in CO-saturated MeOH [29] to afford
the mono or diesters 14 and 15, respectively (Scheme 6,b).
Scheme 6
The rates of the reactions of 9 ¥ BF4 with the nucleophiles 10a e (Table 3) were
studied under pseudo-first-order conditions (>10 equiv. of Nu) by means of UV/VIS
spectroscopy in the case of 10a (signal increase at 305 340 nm), or by conductometry
in the case of 10b e. Second-order rate constants k2 were either calculated from kobs
k2 [Nu] (for nucleophiles 10a c), or derived from plots of pseudo-first-order rate
constants kobs vs. [Nu] (for 10d,e).
ˆ

Helvetica Chimica Acta Vol. 88 (2005)
1761
Table 3. Second-Order Rate Constants and Eyring Activation Parameters (see text) for the Reactions of 9 ¥ BF4
with the Nucleophiles 10. All reactions were performed in CH2Cl2. The N and s parameters were taken from
[1][23].
Nucleophile
N
s
k2 (at Tˆ 208)
DH≥
[kJ molÀ1
DS≥
[J molÀ1 KÀ1
Temp. range
[8]
[MÀ1 sÀ1
]
]
]
10a
10b
10c
10d
10e
13.36
9.00
12.56
10.61
10.21
0.81
0.98
0.70
0.86
0.82
4.50a)
0.0516a)
2.32a)
0.373
41.20
50.49
47.08
À 91.74
À 97.18
À 77.21
À 28 to ‡ 20
À 32 to ‡ 25
À 48 to ‡ 1
0.651
a) Calculated with the Eyring activation parameters.
The reactions of benzhydrylium tetrafluoroborates with silylated enol ethers or
silylated −ketene acetals× have been demonstrated to proceed with rate-determining
CÀC bond formation, followed by fast desilylation [23][30]. Hence, a similar course
can be expected for the reactions of 9 ¥ BF4 with the nucleophiles 10b e, as shown in
Scheme 7. All rate constants listed in Table 3 can, therefore, be assumed to correspond
to the CÀC bond-forming step, and used for the calculation of the electrophilicity
‡
parameter E, which, for 9 , amounts to À 11.2 Æ 0.9, by minimizing the error D2 ˆ S(log
k2 À s(N ‡ E))2. The standard deviation of Æ 1 unit of E is typical for non-benzhydryl
type electrophiles [1].
Scheme 7
‡
The kinetics of the addition reaction of 9 with phosphanes has previously been
reported by Kane-Maguire and co-workers [31], though it is not fully clear whether,
under the conditions of these experiments, also partial olefin-phosphane ligand
exchange occurs [32]. We can now use Eqn. 1 to calculate the rates for the reactions of
‡
‡
9
with phosphanes from the electrophilicity parameter E(9 ) and the previously
published N and s parameters of phosphanes [33]. In Table 4, the calculated rate
constants are compared with the experimental ones. The two sets of values −match×
within two orders of magnitude, which is remarkable in view of the large structural
varieties of electrophiles and nucleophiles treated by Eqn. 1.

1762
Helvetica Chimica Acta Vol. 88 (2005)
Table 4. Comparison of Calculated vs. Experimental Rate Constants for the Reaction of 9 ¥ BF4 with Different
Phosphanes
Nucleophile
Na)
sa)
kexp [mÀ1 sÀ1]b)
kcalc [mÀ1 sÀ1]c)
Bu3P
Ph3P
15.49
14.33
0.69
0.65
140
3.8
912
108
c
a) From [33]. b) In acetone at 208; data from [31]. ) Calculated according to Eqn. 1, with E(9 ) ˆ À11.2.
‡
Because the half-life t of bimolecular reactions with equal initial concentrations of
reactants (c0) is given by t ˆ 1/(kc0), the reactivity parameters can also be employed to
estimate reaction times of electrophile/nucleophile combinations [34]. For example, by
‡
substituting the parameters E of 9 , and N and s of 1a in Eqn. 1, a second-order rate
constant k2 of 6.3 Â 10À5 mÀ1 sÀ1 (at 208) is calculated, which corresponds to 50%
conversion of 1m solutions of the reactants within ca. 4 h. In accord with this estimate,
Rosenblum and co-workers [5b] reported that the reaction of 9 ¥ BF4 (1m solution in
MeNO2) with 1 equiv. of 1a yields 38% of a 1:1 mixture of regioisomers after 4 h at
room temperature (Scheme 8).
Scheme 8. Reaction of 1a with 9 ¥ BF4, as Reported by Rosenblum and Co-workers [5b]
The electrophilicity parameter E for the isopropyl cation is not precisely known, but
can be estimated to be ca. ‡ 11 [35], which is four orders of magnitude more
electrophilic than tertiary alkyl cations [36]. One, therefore, can conclude that
substitution of a b-H-atom in the isopropyl cation by the Fp group, which is associated
with bridging, reduces the C-electrophilicity by more than 20 orders of magnitude. The
‡
propene-Fp cation 9 , therefore, is among the weakest cationic C-electrophiles, and has
been reported to react only with strong nucleophiles such as carbanions [28], enamines
[28], amines, and phosphanes [32].
The nucleophilicity parameters N of 1a,b, and the electrophilicity parameter E of
(determined in this work) can be used to estimate the feasibility of the reaction
‡
9
cascade described in Scheme 9. While strong and weak electrophiles (E >À 11) can be
expected to attack 1a,b at a sufficient rate, only strong nucleophiles (N > 7) can be
considered as potential reaction partners for the second step of Scheme 9.
Experimental Part
General. IR Spectra: Perkin-Elmer Paragon-1000 PC FT-IR spectrometer; in cmÀ1 1H- and 13C-NMR
.
Spectra: Bruker ARX-300 (300/75.5 MHz, resp.) and Varian VXR-400S (400/100 MHz, resp.) spectrometers; in
CDCl3, CD2Cl2, or (D6)acetone soln., referencing to residual solvent signals; d in ppm, J in Hz. MS: Varian
MAT-311; in m/z (rel. %). HR-MS: Jeol MStation JMS-700.

Helvetica Chimica Acta Vol. 88 (2005)
1763
Scheme 9
Potassium Dicarbonyl(h5-cyclopentadienyl)ferrate (5). Potassium-graphite laminate (C8K), prepared from
graphite (6.00 g, 500 mmol) and K (2.00 g, 51.2 mmol) according to [21], was suspended in THF (180 ml) [22].
At r.t., the bimetallic complex 4 (6.00 g, 17.0 mmol) was added, and the mixture was stirred for 3 h. After
dilution with pentane (180 ml), the graphite was separated by filtration under Ar atmosphere to leave an orange
filtrate, which was evaporated under vacuum to yield 5 (5.94 g, 81%). Orange powder, identified by comparison
of the spectral data with literature reports [20].
(h1-Prop-2-enyl)dicarbonyl(h5-cyclopentadienyl)iron(II) (1a). As described above, a soln. of 5 in THF
(180 ml) was prepared from the bimetallic complex 4 (6.00 g, 17.0 mmol) and C8K, obtained from K (2.00 g,
51.2 mmol) and graphite (6.00 g, 500 mmol). After removal of graphite by filtration, the orange filtrate was
cooled to 08. Then, a soln. of 3-chloropropene (4.20 ml, 52.0 mmol) in THF (30 ml) was added over a period of
20 min. The mixture was allowed to warm to r.t., stirred for another 2 h, diluted with hexane (200 ml), and
filtered through Celite. After removal of the solvent in vacuum, the crude product was purified by Kugelrohr
distillation (528/1  10À3 mbar) to yield 1a (3.63 g, 49%). Brown oil . 1H-NMR (300 MHz, CD2Cl2): 2.11 (d, J ˆ
8.2, CH2(3)); 4.54 (d, J ˆ 8.6, 1 H of CH2(1)); 4.72 (s, Cp); 4.81 (d, J ˆ 16.8, 1 H of CH2(1)); 6.04 (ddt, J ˆ 16.8,
8.6, 8.2, HÀC(2)). 13C-NMR (75.5 MHz, CD2Cl2): 4.7 (t, C(1)); 86.1 (d, Cp); 105.9 (t, C(3)); 148.7 (d, C(2));
217.3 (s, CO). The chemical shifts were in accord with the NMR data reported earlier [37].
(h1-2-Methylprop-2-enyl)dicarbonyl(h5-cyclopentadienyl)iron(II) (1b) [18]. As described above for the
formation of 1a, a soln. of 5 (devoid of graphite) in THF (180 ml) was prepared, and treated with a soln. of 2-
methylpropenyl chloride (5.00 ml, 51.0 mmol) in THF (30 ml). The crude product was purified by Kugelrohr
distillation (618/1 Â 10À3 mbar) to yield 1b (4.2 g, 54%). Brown oil . 1H-NMR (300 MHz, (D6)acetone): 1.75 (s,
2-Me); 2.11 (s, CH2(3)); 4.42 (m, 1 H of CH2(1)); 4.61 (m, 1 H of CH2(1)); 4.82 (s, Cp). 13C-NMR (75.5 MHz,
(D6)acetone): 8.1 (t, C(3)); 23.6 (q, 2-Me); 87.3 (d, Cp); 105.6 (t, C(1)); 155.6 (s, C(2)); 218.2 (s, CO). The
chemical shifts were in accord with the NMR data reported in [37].
Reactions of Fp Complexes 1 with Benzhydrylium Ions 3. General Procedure A (GP A). Under exclusion of
H2O, O2, and light, to a soln. of the appropriate salt 3 in CH2Cl2 was slowly added a CH2Cl2 soln. of 1 at 08.
Stirring was continued at this temp. for 4 h. The mixture was then allowed to warm up to r.t., and stirring was
continued for another 12 h. For workup, see below.
Dicarbonyl(h5-cyclopentadienyl)[(1,2-h)-4,4-bis{4-(dimethylamino)phenyl}but-1-ene]iron(II) Tetrafluoro-
borate (6a ¥ BF4). Prepared according to GP A, from 3d ¥ BF4 (330 mg, 0.970 mmol) and 1a (235 mg, 1.08 mmol)
in CH2Cl2 (15 ml). After removal of the solvent in vacuum, the crude product was dissolved in CH2Cl2 (7 ml),
and cooled to À 308. During 10 min, Et2O was added to this soln. by condensation. Further cooling to À 328 led
to precipitation of a solid, which was isolated by filtration, and dried in vacuum to afford 6a (290 mg, 54%).
Dark green powder. IR: 3120, 2931, 2889, 2856, 2804, 2075, 2035, 1612, 1521, 1481, 1449, 1431, 1352, 1082, 1057,
945, 875, 822, 595, 565, 536, 521. 1H-NMR (300 MHz, (D6)acetone): 2.15 (m, 1 H of CH2(3)); 2.91 (s, 2 Me2N);
3.26 (m, 1 H of CH2(3)); 3.42 (d, J ˆ 14.6, 1 H of CH2(1)); 3.81 (d, J ˆ 8.3, 1 H of CH2(1)); 4.04 (dd, J ˆ 8.0, 6.4,
HÀC(4)); 5.03 (m, HÀC(2)); 5.81 (s, Cp); 6.76 (m, 4 arom. H); 7.15, 7.25 (2d, AA'BB', J ˆ 8.5 each, 2  2 arom.
H). 13C-NMR (75.5 MHz, (D6)acetone): 41.8 (q, (Me2N); 43.9 (t, C(3)); 53.8 (d, C(4)); 56.9 (t, C(1)); 87.5 (d,
C(2)); 90.6 (d, Cp); 114.7 (d); 129.6 (d), 130.1 (d); 134.4 (s), 134.6 (s); 150.0 (s); 211.9 (s, CO), 212.0 (s, CO).
Dicarbonyl(h5-cyclopentadienyl)[(1,2-h)-4,4-bis(4-pyrrolidinophenyl)but-1-ene]iron(II) Hexafluorophos-
phate (6b ¥ PF6). Prepared according to GP A, from 3e ¥ PF6 (1.25 g, 2.77 mmol) and 1a (675 mg, 3.10 mmol)
in CH2Cl2 (20 ml). Workup as described for 6a afforded 6b ¥ PF6 (1.26 g, 68%). Green powder. IR: 3120, 2920,

1764
Helvetica Chimica Acta Vol. 88 (2005)
1
2075, 2033, 1612, 1519, 1488, 1372, 844, 595, 557, 534. H-NMR (300 MHz, (D6)acetone)3): 1.97 (br. s, 1 H of
HÀC(3), 4 CH2); 3.21 (br. s, 4 CH2); 3.42 4.05 (m, 5 H); 4.95 (mc, HÀC(2)); 5.84 (s, Cp); 6.49 (br. s, 4 H);
7.08 7.15 (m, 4 H). 13C-NMR (75.5 MHz, (D6)-acetone)3): 25.9 (t); 43.8 (t, C(3)); 48.3 (t); 53.5 (d, C(4)); 56.1 (t,
C(1)); 87.5 (d, C(2)); 90.1 (d, Cp); 112.5 (d); 112.6 (d); 129.0 (d); 129.4 (d); 131.6 (s); 131.7 (s); 147.2 (s); 147.5
(s); 209.1 (s, CO); 211.5 (s, CO).
4,4-Bis[4-(diphenylamino)phenyl]-2-methylbut-1-ene4) (7a) [3]. Compound 3a ¥ BF4 (500 mg, 0.850 mmol)
was reacted with 1b (260 mg, 1.12 mmol) in CH2Cl2 (12 ml) according to GP A. Removal of the solvent in
vacuum afforded the crude metal complex, which, however, could not be purified. For decomplexation, the
compound was dissolved in acetone (6 ml), and treated with NaI (0.26 g, 1.7 mmol). After 13 h at r.t., H2O
(8 ml) was added, the mixture was extracted with CHCl3 (3 Â 25 ml), the combined org. layers were dried
(MgSO4), and evaporated in vacuum. The resulting crude product was purified by column chromatography
(CC) (SiO2; hexane/AcOEt 50 :1) to afford 7a (210 mg, 44 %). Colorless solid. IR: 3060, 3034, 2964, 2927, 1589,
1
1506, 1492, 1328, 1314, 1278, 1177, 1155, 1076, 1029, 1016, 892, 820, 754, 696, 622, 556, 536, 513, 499. H-NMR
(400 MHz, CDCl3): 1.62 (s, 3 H); 2.64 (d, J ˆ 7.9, 2 H); 3.98 (t, J ˆ 7.9, 1 H); 4.53 (s, 1 H); 4.64 (s, 1 H); 6.89 7.17
(m, 28 H). 13C-NMR (100 MHz, CDCl3): 22.7 (q); 44.3 (t); 48.3 (d); 112.5 (t); 122.5 (d); 123.9 (d); 124.0 (d);
128.6 (d); 129.1 (d); 139.2 (s); 143.5 (s); 145.6 (s); 147.9 (s). The chemical shifts were in accord with the NMR
‡
‡
data reported in [3]. HR-MS: 556.2876 (M , C41H36N2 ; calc. 556.2878).
4,4-Bis[4-(dimethylamino)phenyl]-2-methyl-but-1-ene4) (7b) [3]. Prepared in analogy to 7a, from 3d ¥ BF4
(805 mg, 2.37 mmol) and 1b (638 mg, 2.75 mmol) in CH2Cl2 (15 ml), followed by treatment with NaI (1.50 g) in
acetone (12 ml), extraction with CHCl3, and purification by CC (SiO2; hexane/AcOEt 15 :1). Yield of 7b:
1
380 mg, 54%. Colorless crystals. H-NMR (400 MHz, CDCl3): 1.71 (s, 2-Me); 2.73 (d, J ˆ 7.7, CH2(3)); 2.90 (s,
2 Me2N); 4.04 (t, J ˆ 7.7, H ÀC(4)); 4.65 (br. s, 1 H of HÀC(1)); 4.71 (br. s, 1 H of HÀC(1)); 6.68 (d, J ˆ 8.4, 4
arom. H); 7.15 (d, J ˆ 8.4, 4 arom. H); the chemical shifts were in accord with the NMR data reported in [3].
13C-NMR (100 MHz, CDCl3): 22.9 (q, 2-Me); 41.1 (q, Me2N); 44.6 (t, C(3)); 47.6 (d, C(4)); 112.5 (t, C(1)); 112.9
(d); 128.6 (d); 134.1 (d); 144.5 (s, C(2)); 149.1 (s).
4,4-Bis(4-pyrrolidinophenyl)-2-methylbut-1-ene4) (7c) [3]. Prepared in analogy to 7a, from 3e ¥ BF4
(600 mg, 1.33 mmol) and 1b (410 mg, 1.76 mmol) in CH2Cl2 (12 ml), followed by treatment with NaI
(750 mg) in acetone (10 ml), extraction with CH2Cl2 (3 Â 40 ml), and purification by CC (SiO2; hexane/Et2O
1
10 :1). Yield of 7c: 285 mg, 60%. H-NMR (300 MHz, CDCl3): 1.68 (s, 2-Me); 1.96 (mc, 8 H); 2.69 (d, J ˆ 7.9,
CH2(3)); 3.23 (mc, 8 H); 4.00 (t, J ˆ 7.9, HÀC(4)); 4.62, 4.67 (2s, CH2(1)); 6.45 6.50 (m, 4 H); 7.02 7.05 (m,
4 H). 13C-NMR (75.5 MHz, CDCl3): 22.4 (q, 2-Me); 25.3 (t); 44.2 (t, C(3)); 47.0 (d, C(4)); 47.4 (t); 111.7 (t,
C(1)); 112.4 (d); 128.2 (d); 132.3 (s); 144.1 (s); 149.9 (s). The chemical shifts were in accord with the NMR data
reported in [3].
Dicarbonyl(h5-cyclopentadienyl)[(1,2-h)propene]iron(II) Tetrafluoroborate (9 ¥ BF4). A soln. of 1a
(600 mg, 2.75 mmol) in CH2Cl2/Et2O/propanoic anhydride 5 :10 :2 (10 ml) was cooled to 08. Addition of
HBF4 ¥ OEt2 (52% in Et2O; 2.7 g, 7.32 mmol) under vigorous stirring caused the precipitation of a yellow solid.
Stirring at 08 was continued for 1 h. Then, the mixture was allowed to warm to r.t. The solid was collected by
filtration under N2 atmosphere, washed with Et2O (3 Â 5 ml), and dried in vacuum to afford 9 ¥ BF4 (750 mg,
89%). Yellow solid. 1H-NMR (400 MHz, (D6)acetone): 1.90 (d, J ˆ 6.1, Me); 3.65 (d, Jtrans ˆ 14.6, 1 H of
HÀC(1)); 4.05 (d, Jcis ˆ 8.2, 1 H of HÀC(1)); 5.34 (m, HÀC(2)); 5.86 (s, Cp). 13C-NMR (100 MHz,
(D6)acetone): 21.8 (q, C(3)); 56.2 (t, C(1)); 86.1 (d, C(2)); 90.2 (d, Cp); 209.5 (s, CO); 211.4 (s, CO). The
chemical shifts were in accord with the NMR data reported in [38].
Reactions of 9 ¥ BF4 with Nucleophiles 10. General Procedure B (GP B). In the dark, to a suspension of 9 ¥
BF4 in CH2Cl2 was added the appropriate nucleophile (10a or 10b) at 08. The mixture was stirred at 08 for 4 h.,
and then at r.t. for 12 h. For workup, see below.
2-Isopropylcyclohexanone (12). Prepared according to GP B, from 10a (230 mg, 1.39 mmol) and 9 ¥ BF4
(290 mg, 0.948 mmol) in CH2Cl2 (8 ml). For the decomplexation, the mixture was cooled to 08, and saturated
with HCl gas for 90 min. After removal of the solvent by distillation at 208/15 mbar, the residue was dissolved in
a small amount of acetone, and purified by CC (SiO2; CH2Cl2) and Kugelrohr distillation (758/5 Â 10À3 mbar) to
1
afford 12 (76 mg, 57%). Colorless oil. H-NMR (300 MHz, CDCl3): 0.86, 0.88 (2d, J ˆ 6.9 each, 2 Me); 1.56
2.41 (m, 10 H). 13C-NMR (75.5 MHz, CDCl3): 19.1, 21.2 (2q, 2 Me); 24.3 (t); 26.4 (d); 28.1 (t); 29.3 (t); 42.1 (t);
57.3 (d); 213.7 (s). The chemical shifts were in accord with the NMR data reported in [39].
1
1H,1H- and H,13C-COSY experiments were used to assign the NMR signals.
3
4
)
)
Not fully systematic name to reflect the analogy to related compounds.

Helvetica Chimica Acta Vol. 88 (2005)
1765
Methyl 2,2,3-Trimethylbutanoate (14). Prepared according to GP B, from 10b (215 mg, 1.23 mmol) and 9 ¥
BF4 (250 mg, 0.818 mmol) in CH2Cl2 (8 ml). The decomplexation was performed as described for 12, the
Kugelrohr distillation being performed at 608/8 Â 10À3 mbar, to afford 14 (77 mg, 65 %). Colorless oil. 1H-NMR
(300 MHz, CDCl3): 0.71 (d, J ˆ 6.9, 2  3-Me); 0.97 (s, 2  2-Me); 1.84 (sept., J ˆ 6.9, HÀC(3)); 3.53 (s, MeO).
13C-NMR (75.5 MHz, CDCl3): 17.7 (q, 3-Me); 21.6 (q, 2-Me); 35.0 (d, C(3)); 45.8 (s, C(2)); 51.6 (q, MeO); 178.9
(s, C(1)). The chemical shifts were in accord with the NMR data reported in [40].
Dimethyl 2,2,3-Trimethylpentanedioate (15). Prepared according to GP B, from 10b (430 mg, 2.47 mmol)
and 9 ¥ BF4 (650 mg, 2.13 mmol) in CH2Cl2 (15 ml). The solvent was evaporated in vacuum to leave a solid
residue, which was dissolved in MeOH (20 ml) and saturated with CO at À 788. While the mixture was kept
saturated by a permanent flow of CO, a soln. of Ce(NH4)2(NO3)6 in MeOH (15 ml) was added over a period of
60 min. Stirring under a CO atmosphere at À 788 was continued for 1 h. Then, the mixture was allowed to warm
to r.t., the solvent was distilled off (308C/15 mbar), the residue was dissolved in a small amount of acetone, and
purified by CC (SiO2; CH2Cl2) and Kugelrohr distillation (708/6 Â 10À4 mbar) to afford 15 (201 mg, 53%).
Colorless oil. 1H-NMR (300 MHz, CDCl3): 0.81 (d, J ˆ 6.9, 3-Me); 1.05, 1.06 (2s, 2  2-Me); 1.97, 2.29 (2mc,
HÀC(3), CH2(4)); 3.59 (s, 2 MeO). 13C-NMR (75.5 MHz, CDCl3): 15.2 (q, 3-Me); 21.8, 22.0 (2q, 2-Me); 36.8 (d,
C(3)), 36.9 (t, C(4)); 45.5 (s, C(2)); 51.7, 51.9 (2q, 2 MeO); 173.7 (s, CˆO), 177.9 (s, CˆO). The chemical shifts
were in accord with the NMR data reported in [41].
Kinetics. The temp. of the solns. used for kinetic studies was maintained constant within Æ 0.28 by means of
a circulating-bath cryostat (water/glycol bath for T >À 108, EtOH bath for T<À 108), and monitored with a
thermocouple probe inserted into the reaction mixture. Eyring activation parameters DH≥ and DS≥ of the
electrophile/nucleophile combinations (see Table 3) were calculated from second-order rate constants
measured at different temperatures (DT> 30 K).
As the reactions of the colored benzhydrylium ions 3 with allyl-Fp complexes 1a,b gave rise to colorless
products, the rates of the reactions were determined photometrically. The kinetic experiments were performed
in Schlenk glassware under a N2 atmosphere and exclusion of H2O and O2. UV/VIS Spectra in the range of
600 640 nm were collected as a function of time by use of a fiber-optics system (SchˆllyKGS III ), described in
[2a], that was combined with a J&M TIDAS diode-array spectrophotometer [42]. Typically, the nucleophile
(Nu) concentrations (0.1 3 mm) were considerably higher than those of the cations [0.01 0.03 mm), which
resulted in pseudo-first-order kinetics, with an exponential decay of the benzhydrylium ion concentration. The
data are summarized in Tables 5 and 6.
The kinetics of the reaction of enamine 10a with the Fp complex 9 ¥ BF4 (Table 7) were determined
photometrically [2a] by following the product formation in the UV/VIS spectrum (Zeiss MCS-220 diode-array
spectrometer, resembling a similar experimental setup as reported in [42]). By using the increase of the signal
integral in the range of 305 340 nm and the relation tkobs ˆ ln(At À Aend) À ln(A0 À Aend), the rate constants kobs
were obtained from a plot of ln (At À Aend) vs. time (t). The second-order rate constants were then calculated
from kobs ˆ k2 [Nu].
‡
Because of the formation of neutral products, the reactions of the cationic Fp complex 9 with the silylated
−ketene acetals× 10b e (Table 7) were followed by conductometry (Tacussel CD810 conductometer, Pt plate
electrodes WTW LTA1), as described in [2a]. At different temperatures, calibration curves for the correlation
‡
‡
between conductivity and [9 ] were determined by measuring the conductivity of solns. of 9 as a function of
‡
concentration. For the reactions of 9 with 10b,c, the second-order rate constants were calculated from kobs ˆ k2
‡
[Nu]. However, for the reactions of 9 with 10d,e, k2 values (Table 3) were obtained from the slopes of plots of
kobs vs. [Nu].
We thank the Deutsche Forschungsgemeinschaft (Ma 673-19) for support of this work.

1766
Helvetica Chimica Acta Vol. 88 (2005)
‡
Table 5. Kinetics of the Reactions of the Fp Complex 1a with Benzhydrylium Ions 3 . In CH2Cl2; the
electrophiles were analyzed spectrophotometrically. All concentrations expressed in mol/l.
‡
‡
T [8]
[3 ]0
[1a]
[1a]0/[3 ]0
Conversion [%]
k [mÀ1 sÀ1
]
3e ¥ BF4 (l ˆ 630 nm):
20.0
20.0
20.0
20.0
1.94 Â 10À5
1.40 Â 10À3
1.63 Â 10À3
1.88 Â 10À3
2.51 Â 10À3
72
99
121
149
45
53
64
55
1.05 Â 10À1
1.19 Â 10À1
1.23 Â 10À1
1.63 Â 10À1
1.64 Â 10À5
1.55 Â 10À5
1.68 Â 10À5
3d ¥ BF4 (l ˆ 600 nm):
20.0
20.0
20.0
2.05 Â 10À5
1.26 Â 10À3
1.85 Â 10À3
2.91 Â 10À3
61
102
152
86
68
54
5.95 Â 10À1
6.71 Â 10À1
8.15 Â 10À1
1.82 Â 10À5
1.91 Â 10À5
3c ¥ BF4 (l ˆ 630 nm):
20.0
20.0
20.0
20.0
1.40 Â 10À5
4.17 Â 10À4
8.03 Â 10À4
1.46 Â 10À3
1.94 Â 10À3
30
53
86
80
74
69
53
6.48
7.94
7.82
6.80
1.52 Â 10À5
1.69 Â 10À5
1.46 Â 10À5
133
3b ¥ BF4 (l ˆ 630 nm):
20.0
20.0
20.0
20.0
1.91 Â 10À5
1.92 Â 10À4
5.82 Â 10À4
8.94 Â 10À4
1.11 Â 10À3
10
30
50
65
92
95
89
82
1.20 Â 101
1.17 Â 101
1.19 Â 101
1.21 Â 101
1.93 Â 10À5
1.78 Â 10À5
1.70 Â 10À5
3a ¥ BF4 (l ˆ 640 nm):
À 20.0
À 30.0
À 40.0
À 50.0
2.12 Â 10À5
4.55 Â 10À4
7.71 Â 10À4
1.02 Â 10À3
2.37 Â 10À3
21
34
43
97
72
79
82
70
1.37 Â 101
7.31
4.05
2.24 Â 10À5
2.37 Â 10À5
2.45 Â 10À5
1.94
‡
Table 6. Kinetics of the Reactions of the Fp Complex 1b with Benzhydrylium Ions 3 . In CH2Cl2; the
electrophiles were analyzed spectrophotometrically. All concentrations expressed in mol/l.
‡
‡
T [8]
[3 ]0
[1b]0
[1b]0/[3 ]0
Conversion [%]
k [mÀ1 sÀ1
]
3f ¥ BF4 (l ˆ 630 nm)
20.0
20.0
20.0
20.0
1.45 Â 10À5
1.99 Â 10À3
1.26 Â 10À3
1.55 Â 10À3
7.33 Â 10À4
137
120
101
52
55
75
62
59
7.21 Â 10À1
5.71 Â 10À1
6.23 Â 10À1
6.10 Â 10À1
1.05 Â 10À5
1.54 Â 10À5
1.42 Â 10À5
3e ¥ BF4 (l ˆ 630 nm)
20.0
1.66 Â 10À5
1.65 Â 10À3
99
95
3.43
3d ¥ BF4 (l ˆ 600 nm)
20.0
20.0
20.0
20.0
20.0
1.79 Â 10À5
2.97 Â 10À4
4.63 Â 10À4
8.15 Â 10À4
8.76 Â 10À4
9.58 Â 10À4
17
33
53
53
66
75
94
84
88
86
1.57 Â 101
1.57 Â 101
1.67 Â 101
1.52 Â 101
1.54 Â 101
1.40 Â 10À5
1.54 Â 10À5
1.66 Â 10À5
1.45 Â 10À5
3c ¥ BF4 (l ˆ 630 nm)
20.0
20.0
20.0
20.0
20.0
1.34 Â 10À5
1.47 Â 10À4
2.75 Â 10À4
4.03 Â 10À4
5.42 Â 10À4
6.06 Â 10À4
11
21
31
41
52
85
83
81
82
80
1.54 Â 102
1.46 Â 102
1.43 Â 102
1.45 Â 102
1.45 Â 102
1.34 Â 10À5
1.28 Â 10À5
1.32 Â 10À5
1.16 Â 10À5

Helvetica Chimica Acta Vol. 88 (2005)
1767
Table 7. Kinetics of the Reactions of 9 ¥ BF4 with Nucleophiles 10. In CH2Cl2 solution. All concentrations are
expressed in mol/l. The kinetics were determined by conductometry, except for 10a, where UV/VIS
spectroscopy was used (l ˆ 305 340 nm).
‡
‡
T [8]
[9 ]0
[10]0
[9 ]0/[10]0
Conversion [%]
k2 [mÀ1 sÀ1
]
10a:
20.0
20.0
20.0
7.0
À 0.4
À 10.2
À 27.5
5.76 Â 10À4
5.36 Â 10À4
9.19 Â 10À4
7.99 Â 10À4
1.03 Â 10À3
9.84 Â 10À4
7.93 Â 10À4
2.20 Â 10À2
8.37 Â 10À3
2.19 Â 10À2
1.87 Â 10À2
1.05 Â 10À2
1.47 Â 10À2
1.72 Â 10À2
38
15
24
23
10
15
22
94
98
90
96
98
99
95
4.90
5.01
5.81
1.84
1.16
4.61 Â 10À1
1.68 Â 10À1
10b:
25.0
20.0
20.0
3.19 Â 10À4
2.23 Â 10À4
3.19 Â 10À4
2.59 Â 10À4
3.08 Â 10À4
3.34 Â 10À4
3.83 Â 10Àv4
1.35 Â 10À2
9.88 Â 10À3
1.88 Â 10À2
7.55 Â 10À3
1.72 Â 10À2
2.11 Â 10À2
1.61 Â 10À2
42
44
59
29
56
63
42
85
88
92
80
69
76
27
7.04 Â 10À2
5.22 Â 10À2
5.36 Â 10À2
5.18 Â 10À2
3.90 Â 10À3
1.43 Â 10À3
4.62 Â 10À4
20.0
À 12.0
À 22.2
À 31.7
10c:
1.0
À 9.0
À 9.0
À 18.8
À 48.2
7.14 Â 10À4
8.97 Â 10À4
7.74 Â 10À4
8.76 Â 10À4
8.42 Â 10À4
2.76 Â 10À2
3.31 Â 10À2
4.75 Â 10À2
5.38 Â 10À2
5.78 Â 10À2
39
37
61
61
69
59
58
50
50
56
5.55 Â 10À1
2.63 Â 10À1
2.99 Â 10À1
9.48 Â 10À2
5.19 Â 10À3
10d:
20.0
20.0
20.0
20.0
8.03 Â 10À4
6.94 Â 10À4
7.36 Â 10À4
7.42 Â 10À4
9.78 Â 10À3
1.55 Â 10À2
3.31 Â 10À2
7.23 Â 10À2
12
22
45
97
74
89
84
64
2.40 Â 10À3 a
4.90 Â 10À3 a
1.19 Â 10À2 a
2.59 Â 10À2 a
)
)
)
)
10e:
20.0
20.0
20.0
20.0
20.0
3.54 Â 10À4
3.37 Â 10À4
3.43 Â 10À4
3.34 Â 10À4
3.20 Â 10À4
3.39 Â 10À2
4.53 Â 10À2
2.76 Â 10À2
1.43 Â 10À2
2.06 Â 10À2
96
134
81
43
64
94
76
72
76
88
1.99 Â 10À2 a
2.58 Â 10À2 a
1.30 Â 10À2 a
6.90 Â 10À3 a
8.70 Â 10À3 a
)
)
)
)
)
a) kobs [sÀ1]; k2 was obtained from the slopes of plots of kobs vs. [Nu].
REFERENCES
[1] H. Mayr, B. Kempf, A. R. Ofial, Acc. Chem. Res. 2003, 36, 66 77.
[2] a) H. Mayr, R. Schneider, C. Schade, J. Bartl, R. Bederke, J. Am. Chem. Soc. 1990, 112, 4446; b) H. Mayr,
R. Schneider, B. Irrgang, C. Schade, J. Am. Chem. Soc. 1990, 112, 4454; c) H. Mayr, R. Schneider, U.
Grabis, J. Am. Chem. Soc. 1990, 112, 4460; d) G. Hagen, H. Mayr, J. Am. Chem. Soc. 1991, 113, 4954.
[3] H. Mayr, T. Bug, M. F. Gotta, N. Hering, B. Irrgang, B. Janker, B. Kempf, R. Loos, A. R. Ofial, G.
Remennikov, H. Schimmel, J. Am. Chem. Soc. 2001, 123, 9500.
[4] M. Rosenblum, Acc. Chem. Res. 1974, 7, 122.
[5] a) A. Rosan, M. Rosenblum, J. Tancrede, J. Am. Chem. Soc. 1973, 95, 3062; b) P. J. Lennon, A. Rosan, M.
Rosenblum, J. Tancrede, P. Waterman, J. Am. Chem. Soc. 1980, 102, 7033.
[6] T. W. Bodnar, A. R. Cutler, Organometallics 1985, 4, 1558.

1768
Helvetica Chimica Acta Vol. 88 (2005)
[7] S. R. Berryhill, J. G. Breitenbucher, Tetrahedron Lett. 1990, 31, 1671.
[8] G. E. Agoston, M. P. Cabal, E. Turos, Tetrahedron Lett. 1991, 32, 3001; S. Jiang, E. Turos, Tetrahedron Lett.
1991, 32, 4639.
[9] A. Cutler, D. Ehntholt, P. Lennon, K. Nicholas, D. F. Marten, M. Madhavarao, S. Raghu, A. Rosan, M.
Rosenblum, J. Am. Chem. Soc. 1975, 97, 3149.
[10] A. Cutler, D. Ehntholt, W. P. Giering, P. Lennon, S. Raghu, A. Rosan, M. Rosenblum, J. Tancrede, D. Wells,
J. Am. Chem. Soc. 1976, 98, 3495.
[11] T. S. Abram, R. Baker, C. M. Exon, V. B. Rao, J. Chem. Soc., Perkin Trans. 1 1982, 285; R. Baker, C. M.
Exon, V. B. Rao, R. W. Turner, J. Chem. Soc., Perkin Trans. 1 1982, 295; T. S. Abram, R. Baker, C. M. Exon,
V. B. Rao, R. W. Turner, J. Chem. Soc., Perkin Trans. 1 1982, 301.
[12] W. Malisch, J. Zˆller, M. Schwarz, V. J‰ger, A. M. Arif, Chem. Ber. 1994, 127, 1243.
[13] S. Jiang, E. Turos, Organometallics 1993, 12, 4280.
[14] M. Rosenblum, J. Organomet. Chem. 1986, 300, 191.
[15] M. E. Welker, Chem. Rev. 1992, 92, 97.
[16] M. L. H. Green, P. L. I. Nagy, J. Chem. Soc. 1963, 189.
[17] T. L. Underiner, in −Encyclopedia of Reagents for Organic Synthesis×, Ed. L. A. Paquette, J. Wiley & Sons,
Chichester, 1995, Vol. 1, p. 100 101.
[18] W. P. Giering, M. Rosenblum, J. Organomet. Chem. 1970, 25, C71.
[19] J. K¸hn, K. R¸ck-Braun, J. Prakt. Chem. 1997, 339, 675; G. R. Stephenson, M. Perseghini, A. Togni, in
−Science of Synthesis×, Ed. M. Lautens, Thieme Verlag, Stuttgart, 2001, Vol. 1, Chapt. 1.7, p. 745 888
[20] T. Bodnar, E. Coman, K. Menard, A. Cutler, Inorg. Chem. 1982, 21, 1275; M. Brookhart, W. B. Studabaker,
G. R. Husk, Organometallics 1987, 6, 1141.
[21] A. F¸rstner, in −Encyclopedia of Reagents for Organic Synthesis×, Ed. L. A. Paquette, J. Wiley & Sons,
Chichester, 1995, Vol. 6, p. 4226 4230.
[22] D. Tamarkin, Y. Cohen, M. Rabinovitz, Synthesis 1987, 196.
[23] T. Tokuyasu, H. Mayr, Eur. J. Org. Chem. 2004, 2791.
[24] B. Kempf, N. Hampel, A. R. Ofial, H. Mayr, Chem. Eur. J. 2003, 9, 2209.
[25] Y. Apeloig, in −The Chemistry of Silicon Compounds×, Eds. S. Patai, Z. Rappoport, J. Wiley & Sons, New
York, 1993, p. 57 226; S. G. Wierschke, J. Chandrasekhar, W. L. Jorgensen, J. Am. Chem. Soc. 1985, 107,
1496.
[26] W. P. Giering, M. Rosenblum, J. Chem. Soc., Chem. Commun. 1971, 441.
[27] D. E. Laycock, J. Hartgerink, M. C. Baird, J. Org. Chem. 1980, 45, 291.
[28] P. Lennon, A. M. Rosan, M. Rosenblum, J. Am. Chem. Soc. 1977, 99, 8426.
[29] S. Jiang, T. Chen, E. Turos, Organometallics 1995, 14, 4710.
[30] J. Burfeindt, M. Patz, M. M¸ller, H. Mayr, J. Am. Chem. Soc. 1998, 120, 3629.
[31] L. Cosslett, L. A. P. Kane-Maguire, J. Organomet. Chem. 1979, 178, C17; L. A. P. Kane-Maguire, E. D.
Honig, D. A. Sweigart, Chem. Rev. 1984, 84, 525.
[32] P. Lennon, M. Madhavarao, A. M. Rosan, M. Rosenblum, J. Organomet. Chem. 1976, 108, 93.
[33] B. Kempf, H. Mayr, Chem. Eur. J. 2005, 11, 917.
[34] H. Mayr, A. R. Ofial, in −Carbocation Chemistry×, Eds. G. A. Olah, G. K. S. Prakash, J. Wiley & Sons,
Hoboken, N.J., 2004, Chapt. 13, p. 331.
[35] J. P. Pezacki, D. Shukla, J. Lusztyk, J. Warkentin, J. Am. Chem. Soc. 1999, 121, 6589.
[36] M. Roth, H. Mayr, Angew. Chem., Int. Ed. 1995, 34, 2250.
[37] C. Siebert, in −Gmelin Handbook of Inorganic Chemistry×, 8th edn., 1990, B13, p. 100.
[38] U. Busch, P. H¸bener, N. Kuhn, H. Schumann, C. Weber, in −Gmelin Handbook of Inorganic Chemistry−,
8th edn., 1990, B17, p. 6.
[39] M. Miyashita, S. Hara, A. Yoshikoshi, J. Org. Chem. 1987, 52, 2602; D. H. Marr, J. B. Stothers, Can. J. Chem.
1967, 45, 225.
[40] K. H. Ingold, D. C. Nonhebel, J. C. Walton, J. Phys. Chem. 1986, 90, 2859.
[41] M. Kawai, M. Onaka, Y. Isumi, Bull. Chem. Soc. Jpn. 1988, 61, 2157.
[42] A. D. Dilman, S. L. Ioffe, H. Mayr, J. Org. Chem. 2001, 66, 3196; H. Mayr, A. R. Ofial, Einsichten
Forschung an der LMU M¸nchen 2001, 20(2), 30.
Received March 15, 2005