Iron(II)-catalyzed asymmetric hydrosilylation of acetophenone

Article abstract of DOI:10.1002/ejoc.201100550

Four new ligands containing a pentasubstituted cyclopentadiene tethered to a stereogenic diamine unit have been prepared and used in the iron-catalyzed enantioselective hydrosilylation of acetophenone. Catalytically active species have been generated in s

Full text of DOI:10.1002/ejoc.201100550

FULL PAPER
DOI: 10.1002/ejoc.201100550
Iron(II)-Catalyzed Asymmetric Hydrosilylation of Acetophenone
Michelle Flückiger[a] and Antonio Togni*[a]
Keywords: Asymmetric catalysis / Iron / Amines / Hydrosilylation / Cyclopentadienyl ligands / Carbonyl compounds
Four new ligands containing a pentasubstituted cyclopenta-
diene tethered to a stereogenic diamine unit have been pre-
or no catalytic activity. Quantitative conversions were ob-
tained working with 4 mol-% Fe at room temp. and phenyl-
pared and used in the iron-catalyzed enantioselective hydro- silane as the reducing agent. Under these conditions, ligand
silylation of acetophenone. Catalytically active species have
been generated in situ starting from various sources of iron.
Fe(acac)2 was the catalyst precursor of choice, whereas other
simple FeII or FeIII compounds resulted in significantly lower
4 afforded an enantioselectivity of 37 % ee. Attempts to iso-
late the single-component Fe complexes containing ligands
3–6 have failed so far.
milder and synthetically convenient asymmetric hydro-
silylation.[6]
Introduction
Selective, efficient, and sustainable organic synthesis is
one of the primary goals of ongoing chemical research.[1]
Catalytic approaches have already been applied to a wide
variety of synthetic problems. In addition, organometallic
catalysts have played a major role in the production of
many fine and bulk chemicals. However, many of the highly
active catalysts in use today[2] utilize the transition metals
palladium, rhodium, iridium, or ruthenium, each of which
have significant drawbacks such as limited availability, high
cost, and toxicity. Thus, recent industrial and academic re-
search has focused on more economical and environmen-
tally friendly catalytic systems that operate without signifi-
cant loss of efficiency or selectivity. First row transition
metals such as iron, copper, zinc, and manganese may be
the solution to the above-mentioned shortcomings. What
distinguishes iron[3] from the other first row transition met-
als is its high availability. As the second most common
metal in the lithosphere, iron is cheap, and a wide variety
of its salts are available on a technical scale.
Among the uses of newly developed catalytic systems
based on iron, the preparation of enantiomerically pure
secondary alcohols through an asymmetric reduction has
attracted a great deal of attention. Industrial demand for
these alcohols is quite high, because of their value as build-
ing blocks in the synthesis of pharmaceuticals, agrochem-
icals, and advanced materials.[4] To obtain enantiopure
alcohols, most synthetic routes start with prochiral ketones
which are either asymmetrically hydrogenated[5] or reduced
in a mild asymmetric hydrogen-transfer reaction or an even
Initial work on iron-catalyzed hydrogenation and hydro-
silylation of C=C double bonds or C=O double bonds was
carried out by Markó et al.[7] in the early 1980s and by
Nesmeyanov[8] in the 1960s. More recently, several Fe cata-
lysts have been reported, including systems with chiral li-
gands affording products in high yields and acceptable en-
antiomeric excesses.[9]
To develop a new Fe-based catalyst for reduction reac-
tions, we sought inspiration from Noyori’s ruthenium trans-
fer hydrogenation catalyst[10] (Figure 1). By virtue of the
chiral nitrogen ligands containing an NH group, complex 1
exemplifies the bifunctional catalytic system which led to a
significant breakthrough in asymmetric transfer hydrogena-
tion.[10,11] The mechanism involves a concerted hydrogen
transfer without prior coordination of the substrate to the
metal. Earlier catalysts contained chiral phosphane ligands
which were later found to facilitate the racemization of the
enantio-enriched alcohols produced during the reaction.[12]
However, catalyst 1 suffers much less from this undesired
reaction.[10] Thus, even after more than ten years, Noyori’s
catalyst is still one of the most effective and most selective
for asymmetric transfer hydrogenation.
[a] Department of Chemistry and Applied Biosciences, Swiss
Federal Institute of Technology, ETH Zürich,
8093 Zürich, Switzerland
E-mail: togni@inorg.chem.ethz.ch
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201100550.
Figure 1. Noyori’s ruthenium catalyst 1 and generalized structure
of envisioned iron catalyst 2.
Eur. J. Org. Chem. 2011, 4353–4360
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4353

M. Flückiger, A. Togni
FULL PAPER
spiro compound rather than the desired substitution prod-
uct (Scheme 1).
Results and Discussion
To reach this goal, a series of cyclopentadienyl-bearing
(Cp-) chiral diamine ligands were envisaged (Figure 2).
Such a ligand design is necessary because iron is easily oxid-
ized, and the organic environment provided by the ligand
should lend stability to the complex without restricting
catalytic activity. The tethered ligand systems used by
Wills[13] to achieve high rigidity in rhodium, ruthenium,
and iridium complexes and the Cp-stabilized piano-stool-
type iron complexes reported by Astruc[14] both provide
precedence for a ligand system composed of alkylated cy-
clopentadienyl derivatives bearing chiral diamine moieties
for use in a new iron-based catalyst for asymmetric re-
ductions.
The syntheses of the tetramethylcyclopentadienyl ana-
logs of ligands 5 and 6 by lithiation of 2,3,4,5-tetrameth-
ylcyclopentadiene followed by alkylation were unsuccessful.
This finding is in line with the quantum-chemical calcula-
tions performed by Krut’ko,[16] predicting the formation of
comparable amounts of three isomeric products for the al-
kylation of 2,3,4,5-tetramethylcyclopentadiene, including
gem-dialkyl-substituted cyclopentadienes and 1,2,3,4-tet-
ramethyl-5-alkylcyclopentadiene in an approximately 1:1:1
ratio (Scheme 2). Since separation of the isomeric mixture
proved unfeasible, another synthetic route was explored.
The 2,3,4,5-tetraethylcyclopentadienyl moiety was formed
by Ni-catalyzed coupling of 3-hexyne with (Z)-3-bromopro-
penoate[17] (Scheme 3). Applying the same reaction condi-
tions to 2-butyne and (Z)-3-bromopropenoate failed to
yield the targeted pentasubstituted cyclopentadiene and led
instead to the formation of ethyl (2,3,4,5,6,7-hexamethylcy-
cloheptatrien-1-yl)acetate (identified by its molecular
weight of 262.2 g/mol). Therefore, the substituted 2,3,4,5-
tetraethylcyclopentadienyl derivative was used as the Cp
core for the syntheses of the desired ligands.
Figure 2. The four new ligands synthesized in this study, each con-
taining a chiral diamine moiety and a cyclopentadiene.
The general synthetic strategy was to start with an alkyl-
ated Cp core and then add the chiral amine moiety. Along
these lines, ligand 4 was readily synthesized by modified
literature procedures,[15] and the critical step of the straight-
forward synthesis of ligand 3 is shown in Scheme 1. Unfor-
tunately, the reaction time for the synthesis of 3 was long,
and the reaction required 2 equiv. of the chiral amine. How-
ever, it did result in a 67% yield of the desired chiral Cp-
amine ligand. Attempts were made to utilize inorganic
bases such as K2CO3, K3PO4, or Cs2CO3 to trap the acid
and consume only 1 equiv. of the chiral amine, but these
were unsuccessful. Lastly, changing the leaving group from
a bromide to a tosylate failed to increase the rate of the
substitution reaction. Adding triethylamine under these
conditions led to an intramolecular reaction, yielding a
Scheme 2. Product mixture derived from the alkylation of 1,2,3,4-
tetramethylcyclopentadiene.
Scheme 3. Synthesis of substituted cyclopentadienes by nickel-cata-
lyzed coupling reactions of alkynes.[17]
The most significant difficulty encountered with using
the 2,3,4,5-tetraethylcyclopentadienyl-1-methyl derivative as
the Cp core was avoiding isomerization to the correspond-
ing thermodynamically favored, fully conjugated, exo-
double-bond derivative (Scheme 4). In all steps during and
following the coupling reaction toward ligands 5 and 6, re-
action time and temperature needed to be carefully con-
trolled to obtain the desired product and avoid unnecessary
loss of material, as no procedure is known to isomerize the
thermodynamically favored product back to the desired iso-
mer.
Scheme 1. Time-consuming substitution reaction to yield 67% of
3-C(O)CF3.
4354
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 4353–4360

Iron(II)-Catalyzed Asymmetric Hydrosilylation of Acetophenone
Scheme 4. Formation of isomeric tetraethylcyclopentadienyl deriv-
atives under kinetic/thermodynamic control.
Despite careful temperature control during the coupling
reaction, the formation of a second unexpected and unde-
sired exo-double-bond derivative was identified, based on a
Scheme 6. Reductive amination.
1
multiplet in the H NMR spectrum at δ = 5.25 ppm and
the corresponding signal in the 13C NMR spectrum at δ =
Because ligands 4–6 were obtained as a mixture of iso-
109–110 ppm. At a maximum of 13% of the overall yield, mers, all with very similar NMR spectroscopic data, unam-
this product could not be separated from the three desired biguous assignments of the corresponding signals were im-
isomers (Scheme 4) which were directly reduced to the cor- possible. Thus, for characterization we relied on the absence
responding aldehydes in a single step (Scheme 5). During of signals attributed to unwanted products as well as on
the reaction, the temperature was maintained below –70 °C mass spectrometric data.
to avoid further reduction to the corresponding alcohols.
Investigations were carried out on the catalytic potential
After workup, the mixture of aldehydes could be stored for of ligands 3–6 in combination with an iron(II) precursor,
several weeks in the refrigerator without isomerization to forming the catalyst in situ, followed by the asymmetric
the thermodynamically favored exo-double-bond derivative. transfer hydrogenation protocol reported by Beller[18] using
acetophenone (7) as a substrate (Scheme 7). In contrast to
previously published findings, we obtained high conversions
of 7 to 1-phenylethanol (8), albeit with no enantiomeric ex-
cess with or without the presence of iron salts in the reac-
tion mixture. We also found that triphenylphosphane, pre-
viously claimed to be an important additive for Fe-cata-
lyzed transfer hydrogenation, was unnecessary to obtain
high yields.
Scheme 5. Synthesis of a cyclopentadienyl aldehyde.
Scheme 7. Conditions for catalytic hydrogen transfer from litera-
ture.[18] Neither FeII, L*, or triphenylphosphane are necessary to
obtain the conversions reported.
The final step involved the introduction of the chiral
amine moiety by reductive amination. The imine formed
These findings have recently been verified independently
prior to reduction was, once again, prone to isomerize. To by Ouali and Taillefer.[19] Although the mechanism of the
avoid any further isomerization, the reducing agent iron-free transformation is not completely clear, it may be
NaBH(OAc)3 needed to be present at the beginning of the a variation of the well-known Meerwein–Ponndorf–Verley
reaction to drive the reduction step forward following for- reduction.[20]
mation of the imine (Scheme 6). Finally, removal of the tri-
Application of transfer hydrogenations protocols using
fluoroacetyl protecting group was achieved by treating the only catalytic amounts of base or changing the hydrogen
amine with NaBH4 in ethanol, followed by chromato- source from 2-propanol to formic acid or triethyl ammo-
graphic purification.
nium formate (TEAF)[21] failed to yield a functioning cata-
Eur. J. Org. Chem. 2011, 4353–4360
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4355

M. Flückiger, A. Togni
FULL PAPER
lytic system. Under the above-mentioned conditions, the duced chiral induction (Table 1, Entries 8 and 9). Iron(II)
protic environment necessary for transfer hydrogenation tetrafluoroborate showed moderate activity, although re-
may inhibit the in situ formation of the desired catalyst. production of the results proved difficult (Table 1, Entries 5
Instead, under the basic conditions used in all of these pro- and 6).
tocols, Fe(OH)2 was likely to form. This hypothesis was fur-
Ligands 3–6 gave similar results yielding low enantio-
ther supported by successful asymmetric hydrosilylation ex- meric excesses between 24 and 37% (Table 1, Entries 11–
periments, not requiring a protic environment. In addition, 15). The bulkier chiral amine moiety of ligand 6 increased
hydrosilylation required only stoichiometric amounts of the the ee value significantly compared to ligand 5. The effect
reducing agent, as opposed to the large excesses of hydro- of the alkyl moiety on the secondary amine was less clear.
gen donor necessary for transfer hydrogenation. After com- Ligand 4 yielded the product with the highest enantiomeric
pletion of the reaction under aprotic conditions, the silyl- excess, whereas the less bulky ligand 5 showed the lowest
ether could easily be transformed to the desired secondary performance with regard to chiral induction. Ligand 3 led
alcohol by aqueous workup.
to a somewhat lower overall conversion of 7 to 8, as com-
Using acetophenone (7) as a standard substrate and pared to 4–6, probably due to its bulky and rigid fluorene
phenylsilane as the reducing agent, a simple screening of moiety.
various iron compounds as catalyst precursors was carried
Initially, 2 equiv. of ligand relative to the catalyst precur-
out with ligands 3 and 6. The results of these experiments sor were used with the aim of ensuring complete complex
showed that using Fe(acac)2 as the iron source led to the formation. Subsequently, it was found that the reaction time
highest conversion of 7 to 8 (Table 1, Entries 11–17). was prolonged by adding 2 equiv. of ligand. If the ratio of
Iron(II) acetate was also an effective iron source, but re- ligand to Fe(acac)2 was 1:1, 24 h were required for complete
quired elevated reaction temperatures which resulted in re- conversion, whereas with a ratio of 2:1, 40 h were necessary.
Although this is the case, the enantiomeric excess and con-
version remained unaffected. (Table 1, Entries 14 and 15).
Table 1. Influence of different iron and ligand sources on the iron-
THF was the best solvent with regards to conversion as
well as enantiomeric excess (Table 2, Entries 1 and 2). It
seemed that not only the coordinating ability of the solvent
was important, but also the activation temperature for the
catalyst formation. Using weakly coordinating solvents,
such as toluene and dichloromethane, a large drop in enan-
tiomeric excess was observed, however only in the case of
catalyzed asymmetric hydrosilylation of 7.[a]
Table 2. Influence of different solvents and temperature in the Fe-
Entry Iron source
Ligand
Conv. [%][b]
ee [%][c]
catalyzed asymmetric hydrosilylation of 7.[a]
0[d,e]
0[f]
1[d]
3
0
2[g,h]
FeBr2
3[g,h,i] FeBr2
0[f]
6
6
0.4[f]
rac[j]
4
FeCl2
0
5[d,e]
6[d,e,i]
7[e,g]
8[g,i]
9
Fe(BF4)2·6H2O
Fe(BF4)2·6H2O
Fe(OAc)2
Fe(OAc)2
Fe(OAc)2
Fe(acac)2
Fe(acac)2
Fe(acac)2
Fe(acac)2
Fe(acac)2
Fe(acac)2
Fe(acac)2
Fe(acac)2
22.5 (Ϯ7)[f]
rac[j]
[f]
3
52.8 (Ϯ7)
5 (S)[j]
rac[j]
7[f]
6
3
98[f]
rac[j]
Entry Ligand Solvent Reaction temp. [°C] Conv. [%][b] ee [%][c]
42.8 (Ϯ41)
36.6
70.5
quant.
quant.
quant.
quant.
quant.
quant.
12 (Ϯ10)
rac
10
1
4
6
6
3
3
3
4
4
4
THF
THF
THF
THF
THF
THF
room temp.
room temp.
65
room temp.
58
63
quant.
quant.[f]
quant.[f]
70.5
73.5
80
48
95
70
37 (S)
36 (R)[g]
33 (R)[g]
27 (S)
24 (S)
25 (S)
16 (S)
19 (S)
26 (S)
11[d]
12
3
4
5
6
6
27 (S)
37 (S)
24 (S)
32 (R)
37 (R)[j]
11.4 (S)
18 (S)
2[d,e]
3[d,h]
4[e]
5[h]
6[i]
13
14
15[d,i]
16
9[k]
10[l]
7[j]
CH2Cl2 room temp.
toluene room temp.
Et2O
17
8[k]
9[l]
room temp.
[a] Reagents and conditions: in situ catalyst (0.02 mmol), L*
(0.02 mmol), tetrahydrofuran = THF (1.5 mL), 7 (0.5 mmol) for [a] Reagents and conditions: Fe(acac)2 (0.02 mmol), L*
10 min at 60 °C, then addition of phenylsilane (1.1 mmol), 24 h at
room temp. [b] Conversion was determined by NMR spectroscopy
with benzaldehyde as internal standard (conversion is equivalent
to yield). [c] Enantiomeric excess was determined by chiral HPLC.
[d] Reaction time 40 h. [e] Diethoxymethylsilane (0.55 mmol).
[f] Conversion was determined by chiral GC. [g] Reaction tempera-
ture was 65 °C, reaction time was 13.5 h. [h] Diethoxymethylsilane
(1.1 mmol). [i] L* (0.04 mmol). [j] Enantiomeric excess was deter-
mined by chiral GC. [k] 9 = (1R,2R)-cyclohexyldiamine. [l] 10 =
(1R,2R)-1,2-diphenylethane-1,2-diamine.
(0.02 mmol), solvent (1.5 mL), 7 (0.5 mmol) for 10 min at 60 °C,
then addition of phenylsilane (1.1 mmol), 24 h at reaction tempera-
ture. [b] Conversion was determined by NMR spectroscopy with
benzaldehyde as internal standard (conversion is equivalent to
yield). [c] Enantiomeric excess was determined by chiral HPLC.
[d] L* (0.04 mmol). [e] Reaction time 40 h. [f] Conversion was de-
termined by chiral GC. [g] Enantiomeric excess was determined by
chiral GC. [h] Reaction time 14 h. [i] Reaction time 23 h. [j] Forma-
tion of in situ catalyst at 35 °C. [k] Formation of in situ catalyst at
60 °C. [l] Formation of in situ catalyst at room temp.
4356
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 4353–4360

Iron(II)-Catalyzed Asymmetric Hydrosilylation of Acetophenone
dichloromethane did the yield decrease markedly (Table 2, Table 3. Influence of additional bases on the Fe-catalyzed asym-
metric hydrosilylation of 7.[a]
Entries 7 and 8). In the case of ethyl ether, the drop in the
enantiomeric excess was less severe than in the case of tolu-
ene, though conversion was significantly lower (Table 2, En-
tries 8 and 9). Heating the reaction mixture also increased
the reproducibility of the catalytic results, as in the case of
ligand 3. A shorter reaction time was required, however the
enantiomeric excess did decrease slightly (Table 2, Entries 4
and 6).
Entry
Ligand Base
Conv. [%][b] ee [%][c]
Isolation and characterization of iron complexes contain-
ing any of the ligands depicted in Figure 2 have been unsuc-
cessful so far. Given that Ru and Rh catalysts containing
multidentate ligands comprised of a Cp unit usually afford
products with a much higher ee,[13] we assume that ligands
3–6 are coordinating to iron through the amine groups only.
Using chiral 1,2-cyclohexyldiamine and 1,2-diphenyl-
ethane-1,2-diamine as the ligands (Table 1, Entries 16 and
17) gave significant lower ee values than the ones achieved
with ligands 3–6. This corroborates with the assumption
that the cyclopentadienyl moiety is uncoordinated, and thus
1
2
3
4
6
quant.
32 (R)
24 (R)
21 (R)
37 (S)
5 (S)
6
6
4
4
3
3
NaH in THF/15-crown-5 quant.
DBU[d]
Cs2CO3
DABCO
43
quant.
quant.
70.5
4
5
6[e]
7
27 (S)
n.d.[f]
[a] Reagents and conditions: Fe(acac)2 (0.02 mmol), L*
(0.02 mmol), THF (1.5 mL), base (0.02 mmol), 7 (0.5 mmol) for
10 min at 60 °C, then addition of phenylsilane (1.1 mmol), 24 h at
room temp. [b] Conversion was determined by NMR spectroscopy
with benzaldehyde as internal standard, (conversion is equivalent
fulfilling merely a steric function. Additionally, it seems un- to yield). [c] Enantiomeric excess was determined by chiral HPLC.
[d] DBU = 1,8-diazabicyclo[5.4.0]undec-7-ene. [e] Reaction time:
40 h. [f] n.d.: not determined.
likely that the Cp moiety is coordinated to iron, as the only
base in the system is the acetylacetonate ligand from the
catalyst precursor. Acetylacetonate is not sufficiently basic
Table 4. Screening of possible impurities in iron(II) acetylacetonate
to shift the deprotonation equilibrium of the HCp moiety
for catalytic activity in the asymmetric hydrosilylation of 7.[a]
to any significant extent to the right [cf. the pKa values of
13.3,[22] 26.1,[23] and 22.6[24] for acetylacetone, 1,2,3,4,5-
pentamethylcyclopentadiene, and 9-methylfluorene, respec-
tively, all measured in dimethyl sulfoxide (DMSO)]. There-
fore, deprotonation of the HCp moiety was attempted with
additional base. Addition of sodium hydride with 15-crown-
5 did not interfere with the conversion, but led to a strong
decrease in enantiomeric excess (Table 3, Entry 2). Addition
of cesium carbonate resulted in an even poorer enantio-
meric excess value, giving nearly racemic 8 (Table 3, En-
try 5). Lastly, using DABCO (1,4-diazabicyclo[2.2.2]octane)
as a base decreased the conversion to trace amounts of
product (Table 3, Entry 7).
Interestingly, significant differences in catalytic activity
were observed depending on the source of iron(II) acetyl-
acetonate. This suggests that the composition of the starting
material may be the reason for such discrepancies. As sev-
eral Fe-catalyzed reactions, reported in the literature, have
been found to be catalyzed by copper, a common impurity
in iron salts,[25] hydrosilylation of 7 to 8 with copper(II)
chloride or copper(II) acetate was attempted. However, nei-
ther copper salt showed any relevant catalytic activity
Entry Catalyst precursor
Additive
Ag2O
Conv. [%][b]
ee [%][c]
1
Fe(acac)2
quant.
32 (R)
22 (R)
n.d.
2
Fe(acac)2
Fe(acac)2 oxidized
Fe(acac)3
32.8
0.7[e]
0
0
5
3[d]
4
5
6
Cu(acac)2
CuCl2
n.d.
[a] Reagents and conditions: in situ catalyst (0.02 mmol),
6
(0.02 mmol), THF (1.5 mL), 7 (0.5 mmol) for 10 min at 60 °C, then
addition of phenylsilane (1.1 mmol), 24 h at room temp. [b] Con-
version was determined by NMR spectroscopy with benzaldehyde
as internal standard, (conversion is equivalent to yield). [c] Enan-
tiomeric excess was determined by chiral HPLC. [d] 6 (0.04 mmol),
reaction time 40 h. [e] Conversion was determined by chiral GC.
(Table 4, Entries 5 and 6). In addition, a check of the dif- salts was examined. This study revealed that the iron(III)
ferent sources of Fe(acac)2 by LA-ICP-MS (laser ablation content of the precursor salts may be responsible for the
inductively coupled plasma mass spectrometry) showed an differences in the observed catalytic activity. For example,
elevated percentage of silver in the material used to obtain sample 1 was synthesized in the laboratory starting from
the best catalytic results. Therefore, a test reaction was run Fe(CO)5, whereas the starting material for sample 2 was
with additional silver oxide. As a result, the conversion FeCl2. Sample 3 was purchased from Aldrich and sublimed
dropped dramatically as did the observed enantiomeric ex- before use. The best results (quantitative conversions) were
cess. (Table 4, Entry 2).
obtained when sample 1 was used as the iron source. Sam-
As neither impurities in copper or silver explain the de- ple 2 resulted in conversions of approximately 50%, and
pendent nature of the iron source on the catalytic activity sample 3 only gave a 15% conversion. ESR measurements
of the system, the oxidation state of the iron in the iron showed that the conversion was inversely proportional to
Eur. J. Org. Chem. 2011, 4353–4360
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4357

M. Flückiger, A. Togni
FULL PAPER
Instead, the measured spectra can be found in the Supporting In-
formation.
the increasing concentration of iron(III) in the catalyst pre-
cursors. An additional test with commercially available
iron(III) acetylacetonate and another with active iron(II)
acetylacetonate from sample 1 exposed to air confirmed the
assumption that iron(III) is not catalytically active in the
asymmetric hydrosilylation of 7 to 8 (Table 4, Entries 3 and
4).
Monotrifluoroacetyl-Protected Chiral Diamines
N-[(1R,2R)-2-Aminocyclohexyl]-2,2,2-trifluoroacetamide: To a solu-
tion of (1R,2R)-diaminocyclohexane (5 g, 43.8 mmol) in dry meth-
anol (53 mL) at 0 °C was slowly added ethyl trifluoroacetate
(5.73 mL, 48.2 mmol). After stirring for 15 h at 0 °C, the mixture
was concentrated in vacuo, and the title compound was purified
by flash chromatography on silica (ethyl acetate/methanol, 9:1; 1%
triethylamine) to yield a yellowish solid (7.01 g, 76%). 1H NMR
(400.13 MHz, CDCl3): δ = 1.16–1.45 (m, 6 H), 1.75 (m, 2 H), 1.99
(m, 1 H), 2.16 (m, 1 H), 2.50 (m, 1 H), 3.47 (m, 1 H), 6.54 (br. s,
1 H) ppm. 13C NMR (100.62 MHz, CDCl3): δ = 24.80, 25.04,
31.73, 36.31, 54.92, 57.05, 116.07 (q, J = 287.77 Hz), 157.57 (q, J
= 36.7 Hz) ppm. 19F NMR (188.29 MHz, CDCl3): δ = –75.76 ppm.
HRMS: calcd. for C8H13F3N2O [M + H]+ 211.1053; found
211.1052 (δ = 0.6 ppm error). [α]2D0 = –32.19 (c = 1, CH2Cl2).
Conclusions
Four new chiral C1 symmetric diamine ligands have been
prepared, and an efficient iron(II) system for the asymmet-
ric hydrosilylation of acetophenone with phenylsilane has
been developed. Iron(II) acetylacetonate was the best cata-
lyst precursor. Its purity and molar ratio to the ligand were
crucial for catalytic activity. However, the exact composi-
tion of the active catalyst and how the ligands coordinate
is currently still unclear. We are in the process of studying
our new system by EPR spectroscopy, and the correspond-
ing findings will be reported in due course.
N-[(1S,2S)-2-Amino-1,2-diphenylethyl]-2,2,2-trifluoroacetamide:
The same procedure as for N-[(1R,2R)-2-aminocyclohexyl]-2,2,2-
trifluoroacetamide was followed. Purification by flash chromatog-
raphy on silica (cyclohexane/ethyl acetate, 1:1; 1% triethylamine)
yielded N-[(1S,2S)-2-amino-1,2-diphenylethyl]-2,2,2-trifluoroaceta-
mide as a colorless solid (86.6%). 1H NMR (400.13 MHz, CDCl3):
δ = 1.29 (s, 2 H), 4.36 (d, J = 2.8 Hz, 1 H), 4.95 (m, 1 H), 7.00–
7.30 (m, 10 H), 7.75 (br. d, J = 6.4 Hz, 1 H) ppm. 13C NMR
(100.6 MHz, CDCl3): δ = 58.93, 59.29, 116.07 (q, J = 288.38 Hz),
126.12, 126.31, 128.14, 128.25, 128.94, 129.14, 138.85, 141.25,
156.85 (q, J = 36.93 Hz) ppm. 19F NMR (282.38 MHz, CDCl3): δ
Experimental Section
General Methods: NMR spectra were recorded in CDCl3 at ambi-
ent temperature without spinning with Bruker AC-200 (19F: 188.31)
DPX –250 (1H: 250.13), DPX-300 (1H: 300.13, 19F: 282.40, 13C:
75.47), DPX-400 (1H: 400.13, 13C: 100.61), or DPX-500 (1H:
500.23, 13C: 125.78) instruments, operating at the denoted spec-
trometer frequency given in megahertz (MHz) for the specified nu-
cleus. HPLC analyses were performed with an Agilent Series 1100
instrument (detector DAD) using an OD-H column (length =
25 cm; inner diameter = 4.6 mm; particle size = 5 μm; flow rate of
the eluent = 0.5 mL/min; hexane/iPrOH, 9:1; sample injection vol-
ume = 5 μL; and sample concentration is approximately 1 mg/mL).
=
–75.76 ppm. HRMS: calcd. for C16H15F3N2O [M +
H]+
309.1209; found 309.1209 (δ = 0.2 ppm error). [α]2D0 = –0.39 (c = 1,
MeOH).
Ligand 3: To a solution of 9-(2-bromoethyl)-fluorene (3.26 g,
12 mmol) in dry dioxane (90 mL) in a Young Schlenk was added
(1R,2R)-N-(2-aminocyclohexyl)-2,2,2-trifluoroacetamide (5.04 g,
24 mmol). After stirring at 120 °C for 7 d, the mixture was filtered
to separate the hydrobromide salt of N-[(1R,2R)-2-aminocyclo-
Retention times (tR) for (R)- and (S)-1-phenylethanol are 12.2 min hexyl]-2,2,2-trifluoroacetamide. The filtrate was washed with HCl
and 13.5 min, respectively. GC analyses were performed with a
Hewlett Packard HP 6890 Series GC system equipped with a EPC
split splitless injector using H2 as the carrier gas, a Macherey &
(2 m solution) and extracted with dichloromethane. The organic
phase was dried with MgSO4, and the solvents were removed. The
crude product was purified by flash chromatography on silica (ethyl
Nagel Lipodex E column (25 mϫ0.32 mmϫ0.25 μm), temperature acetate, 1% triethylamine) to yield N-{(1R,2R)-2-[2-(9H-fluoren-9-
= 1 min at 70 °C then 1 °C/min to 110 °C, and H2 pressure = yl)ethylamino]cyclohexyl}-2,2,2-trifluoroacetamide [3-C(O)CF3] as
0.5 bar. Retention times (tR) for (S)- and (R)-1-phenylethanol are an off-white solid (3.24 g, 67.1%). 1H NMR (250.13 MHz, CDCl3):
31.0 min and 31.7 min, respectively. High resolution mass spectra
were measured by the MS-Service des Labors für Organische
Chemie, ETH Zürich. Optical rotations were measured with a Per-
kin–Elmer 341 polarimeter equipped with a 10 cm cell at a concen-
tration of 1 g of substance per 100 mL (c = 1.0) in the given solvent.
All reactions were carried out under argon in oven dried glassware
with magnetic stirring or in a dry box under a dinitrogen atmo-
sphere unless explicitly indicated otherwise. All solvents were dried
with sodium/benzophenone or calcium hydride and distilled under
Argon prior to use. Unless explicitly indicated otherwise, reagents
were obtained from commercial sources and used as received. 9-(2-
Bromoethyl)fluorene,[26] 2-(2,3,4,5-tetramethylcyclopentadienyl)-
benzaldehyde,[15] 2-(2,3,4,5-tetraethylcyclopentadienyl)ethyl acet-
ate,[17] and iron(II) acetylacetonate[27] were synthesized by literature
δ = 0.88–2.6 (m, 15 H), 3.29 (m, 1 H), 4.07 (t, J = 5.25 Hz, 1 H),
6.71 (br. s, 1 H), 7.28–7.40 (m, 4 H), 7.49 (d, J = 6.75 Hz, 2 H),
7.76 (d, J = 7.25 Hz) ppm. 13C NMR (100.6 MHz, CDCl3): δ =
24.41, 24.76, 31.33, 33.48, 42.49, 45.66, 54.62, 60.19, 116.01 (q, J =
288.32 Hz), 120.07, 124.34, 124.41, 127.17, 127.30, 127.31, 141.13,
141.19, 146.78, 146.86, 157.44 (q, J = 36.52 Hz) ppm. 19F NMR
(282.38 MHz, CDCl3):
δ = –75.89 ppm. HRMS: calcd. for
C23H25F3N2O [M + H]+ 403.1992; found 403.1984 (δ = 1.9 ppm
error). [α]2D0 = –17.11 (c = 1, CH2Cl2). To a solution of 3–C(O)CF3
(3.24 g, 8 mmol) in dry ethanol (125 mL) was added in portions at
ambient temperature sodium borohydride (1.81 g, 48 mmol). After
stirring for 1 h at ambient temperature, the mixture was heated to
reflux for 30 min. The solvent was then removed under reduced
pressure, and the residue was treated with 1 m aqueous NaHCO3
procedures. Silica gel 60 (230–400 mesh) was purchased from and extracted with dichloromethane. The combined organic ex-
Fluka. Because ligands 4–6 were isolated as inseparable mixtures
of isomers (5 and 6 also have the minor side product with an exo
double bond), which were very similar NMR spectroscopically, we
do not list the corresponding resonances for 1H and 13C NMR.
tracts were dried with MgSO4, followed by removal of the solvents.
The resulting yellow oil was purified by flash chromatography over
silica (ethyl acetate/methanol, 9:1, 1% triethylamine) to yield 3 (2 g,
81.6%) as a light yellow oil, which turned brown upon storage at
4358
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 4353–4360

Iron(II)-Catalyzed Asymmetric Hydrosilylation of Acetophenone
ambient temperature. 1H NMR (400.13 MHz, CDCl3): δ = 0.88–
+ H]+ 415.2931; found 415.2925 (δ = 1.4 ppm error). [α]2D0 = –32.28
0.94 (m, 1 H), 1.07–1.26 (m, 4 H), 1.63 (m, 2 H), 1.85–1.97 (m, 3 (c = 1, CH2Cl2). Deprotected 5 was obtained following the same
H), 2.17–2.40 (m, 6 H), 2.69 (m, 1 H), 4.10 (t, J = 5.6 Hz, 1 H), procedure as for 3. Purification by flash chromatography over silica
7.29–7.38 (m, 4 H), 7.53 (d, J = 7.2 Hz, 2 H), 7.75 (d, J = 7.2 Hz, (ethyl acetate/methanol, 9:1; 1% triethylamine) yielded 5 as a yel-
2 H) ppm. 13C NMR (125.8 MHz, CDCl3): δ = 25.12, 25.15, 30.89, low oil (70%). 1H NMR and 13C NMR spectra can be found in
33.23, 35.41, 43.40, 45.78, 54.98, 63.59, 120.00, 124.55, 127.11,
127.14, 127.20, 141.16, 147.00, 147.04 ppm. HRMS: calcd. for
C21H26N2 [M + H]+ 307.2169; found 307.2170 (δ = 0.3 ppm error).
[α]2D0 = –42.07 (c = 1, CH2Cl2).
the Supporting Information. HRMS: calcd. for C21H38N2
[M + H]+ 319.3108; found 319.3109 (δ = 0.3 ppm error). C21H38N2
(318.54): calcd. C 79.18, H 12.02, N 8.79; found C 78.9, H 11.99,
N 8.54. [α]2D0 = –51.76 (c = 1, CH2Cl2).
Ligand 6: Following the same procedure as for 5-C(O)CF3, purifi-
cation by flash chromatography over silica (cyclohexane/ethyl acet-
ate, 10:1; 1% triethylamine] yielded 6-C(O)CF3 (41%). 1H NMR
and 13C NMR can be found in the Supporting Information. 19F
NMR (282.38 MHz, CDCl3): δ = –66.74 (main peak), –66.83 to
–66.94 (minor peaks) ppm. HRMS: calcd. for C31H39F3N2O
Ligand 4: To a solution of 2-(2,3,4,5-tetramethylcyclopentadienyl)-
benzaldehyde (1 g, 4.42 mmol) in dry methanol (35 mL) was added
N-[(1R,2R)-2-aminocyclohexyl]-2,2,2-trifluoroacetamide (0.91 g,
4.34 mmol), molecular sieves 4 Å (1.25 g), and glacial acetic acid
(2 drops, catalytic) at ambient temperature. After complete forma-
tion of the imine (checked by TLC), sodium borohydride (1 g,
26.5 mmol) was added, and the reaction mixture was stirred over-
night at ambient temperature. Heating to reflux for 1 h followed.
The molecular sieves were then filtered off, and the solvents were
removed under reduced pressure. The residue was dissolved in ethyl
acetate and then washed with saturated aqueous NaHCO3 and
brine. The organic phase was dried with MgSO4, and the solvents
were removed under reduced pressure. The crude brown oil (1.64 g,
88.2%) of 2,2,2-trifluoro-N-{(1R,2R)-2-[2-(2,3,4,5-tetramethylcy-
[MH+] = 513.3087 found 513.3088 (δ = 0.1 ppm error). [α]2D0
–72.29 (c = 1, CH2Cl2).
=
The deprotected ligand 6 was obtained following the same pro-
cedure as for 3. Purification by short flash chromatography on sil-
ica [hexane/ethyl acetate (3:2), 1% triethylamine] yielded 6 (70%).
1H NMR and 13C NMR can be found in the Supporting Infor-
mation. HRMS: calcd. for C29H40N2 [M + H]+ 417.3264; found
417.3264 (δ = 0.1 ppm error). [α]2D0 = –9.86 (c = 1, CH2Cl2).
clopentadienyl)benzylamino]cyclohexyl}acetamide
[4-C(O)CF3]
Experimental Procedure for Catalytic Asymmetric Hydrosilylation:
In a glove box, to Fe(acac)2 (5.08 mg, 0.02 mmol) in an oven dried
10 mL Schlenk was added a ligand (0.02 mmol) in dry and de-
gassed THF (1.5 mL), followed by the addition of acetophenone
(60 μL, 0.5 mmol). The Schlenk was removed from the glove box,
and then the reaction mixture was stirred under Argon and heated
to 60 °C for 10 min. After removing the oil bath, the silane
(1.1 mmol) was added, and the reaction mixture was stirred for
24 h at ambient temperature. Aqueous workup by the addition of
HCl (1 n solution, 1 mL) at 0 °C was followed by stirring for 1 h.
The mixture was extracted with ethyl acetate, washed with satu-
rated NaHCO3 and brine, and then dried with MgSO4. Filtration
over 1 cm silica to remove the remaining traces of the catalyst fol-
lowed by removal of the solvent provided a yellow oil which was
analyzed by NMR (acetaldehyde as internal standard) and chiral
HPLC without further purification.
was used without purification. HRMS: calcd. for C24H31F3N2O [M
+ H]+ 421.2461; found 421.2467 (δ =1.4 ppm error). Ligand 4 was
obtained following the same procedure as for 3. Purification by
flash chromatography on silica (ethyl acetate/methanol, 9:1; 1% tri-
ethylamine) yielded 4 as a light yellow oil (47.4% over 2 steps).
1H NMR and 13C NMR spectra can be found in the Supporting
Information. HRMS: calcd. for C22H32N2 [M + H]+ 325.2638;
found 325.2639 (δ = 0.2 ppm error). [α]2D0 = –60.35 (c = 1, CH2Cl2).
Ligand 5: To a solution of 2-(2,3,4,5-tetraethylcyclopentadienyl)-
ethyl acetate (2.64 g, 10 mmol) in dry hexane (20 mL) was added
diisobutylaluminium hydride (50 mL, 1 m in hexane) dropwise
(drop rate: 1 drop/4 s) over 3 h at –78 °C. After stirring the reaction
mixture for another hour at –78 °C, a mixture of methanol and 6 m
HCl (1:1, 50 mL) was added dropwise over 3 h. After completion,
the cooling bath was removed, and as soon as the aqueous phase
was clear, the reaction mixture was washed with deionized water to
obtain a neutral pH. The organic phase was dried with MgSO4,
and the solvents were removed to yield 2-(2,3,4,5-tetraethylcy-
clopentadienyl)acetaldehyde (1.98 g, 90%) as a yellow oil, used di-
rectly without further purification {The product can be stored in
the refrigerator for up to a month. Longer storing times or storing
at ambient temperature can lead to the 2-[2,3-(2,3,4,5-tetraethylcy-
clopent-2-en-1-ylidene)acetaldehyde isomer].} HRMS (EI): calcd.
for C15H24O [M]+ 220.1822; found 220.1823 (δ = 0.45 ppm error).
To a solution of (1R,2R)-N-(2-aminocyclohexyl)-2,2,2-trifluoroace-
tamide (2.65 g, 12.64 mmol) in dry dichloromethane (30 mL) was
added 2-(2,3,4,5-tetraethylcyclopentadienyl)acetaldehyde (2.53 g,
11.5 mmol) at 0 °C in dry dichloromethane (85 mL), followed by
the portionwise addition of sodium triacetoxyborohydride (16.1 g,
75.8 mmol). After stirring for 7 h at 0 °C, NaOH (1 m solution) was
added in portions, and the reaction mixture was extracted with
dichloromethane. The organic phase was dried with MgSO4, and
the solvents were removed under reduced pressure. Purification by
gradient flash chromatography over silica (hexane; hexane/ethyl
acetate, 10:1; hexane/ethyl acetate, 3:1; 1% triethylamine] yielded
Supporting Information (see footnote on the first page of this arti-
1
cle): H NMR and 13C NMR spectra for new compounds.
Acknowledgments
Support from the Commission for Technology and Innovation,
CTI and Solvias AG, Basel, Switzerland is gratefully acknowl-
edged.
[1] S. Boschen, L. Dieter, M. Scheringer, Naturwissenschaften
2003, 90, 93.
[2] For a recent selection of reviews, see: a) F. Kakiuchi, T. Kochi,
Synthesis 2008, 19, 3013; b) B. C. Enger, R. Loedeng, A. Hol-
men, Appl. Cat. A: General 2008, 346, 1; c) M. Jeganmohan,
C.-H. Cheng, Chem. Commun. 2008, 27, 3101; d) J. Terao, H.
Todo, H. Watabe, A. Ikumi, Y. Shinohara, N. Kambe, Pure
Appl. Chem. 2008, 80, 941; e) F. Alonso, I. P. Beletskaya, M.
Yus, Tetrahedron 2008, 64, 3047.
[3] For iron, see: a) S. Enthaler, K. Junge, M. Beller, Angew. Chem.
2008, 120, 3363; Angew. Chem. Int. Ed. 2008, 47, 3317; b) E. B.
Bauer, Curr. Org. Chem. 2008, 12, 1341; c) C. Bolm, J. Legros,
J. Le Paih, L. Zani, Chem. Rev. 2004, 104, 6217.
1
5-C(O)CF3 (3.22 g, 70%) as a yellow oil. H NMR and 13C NMR
spectra can be found in the Supporting Information. 19F NMR
(282.38 MHz, CDCl3): δ = –75.96 (main peak), –75.98, –76.00,
–76.01 (3 minor peaks) ppm. HRMS: calcd. for C23H37F3N2O [M
[4] a) E. N. Jacobsen, A. Pfaltz, H. Yamamoto (Eds.), Comprehen-
sive Asymmetric Catalysis, Springer, Berlin, 1999; b) R. Noyori,
Eur. J. Org. Chem. 2011, 4353–4360
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4359

M. Flückiger, A. Togni
FULL PAPER
Asymmetric Catalysis in Organic Synthesis, Wiley, New York,
1994; c) R. Noyori, T. Ohkuma, Angew. Chem. 2001, 113, 40;
Angew. Chem. Int. Ed. 2001, 40, 40; d) I. C. Lennon, J. A.
Ramsden, Org. Process Res. Dev. 2005, 9, 110.
[11] S. Hashiguchi, A. Fujii, J. Takehara, T. Ikariya, R. Noyori, J.
Am. Chem. Soc. 1995, 117, 7562.
[12] J. P. Genet, V. Ratovelomanana-Vidal, C. Pinel, Synlett 1993,
7, 478.
[5] a) T. Ohkuma, M. Kitamura, R. Noyori in Catalytic Asymmet-
ric Synthesis (Ed.: I. Ojima), Wiley-VCH, New York, 2000, ch.
1; b) M. Beller, C. Bolm (Eds.), Transition Metals for Organic
Synthesis, 2nd ed., Wiley-VCH, Weinheim, 2004.
[6] a) H. Nishiyama in Comprehensive Asymmetric Catalysis (Eds.:
E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer, New York,
1999, ch. 6.3; b) H. Brunner, H. Nishiyama, K. Itoh, in: Cata-
lytic Asymmetric Synthesis (Ed.: I. Ojima), Wiley-VCH, New
York, 1993, chapter 6.
[13] D. S. Matharu, J. E. D. Martins, M. Wills, Chem. Asian J. 2008,
3, 1374.
[14] D. Catheline, D. Astruc, Organometallics 1984, 3, 1094.
[15] D. S. Matharu, D. J. Morris, A. M. Kawamoto, G. J. Clarkson,
M. Wills, Org. Lett. 2005, 7, 5489.
[16] D. P. Krut’ko, M. V. Borzov, E. N. Veksler, E. N. Myshakin,
D. A. Lemenovskii, Russ. Chem. Bull. 1998, 47, 956.
[17] M. Kotora, M. Ishikawa, F.-Y. Tsai, T. Takahashi, Tetrahedron
1999, 55, 4969.
[7] L. Markó, M. A. Radhi, I. Ötvös, J. Organomet. Chem. 1981,
218, 369.
[8] a) R. Kh. Freidlina, E. C. Chukovskaya, J. Tsao, A. N. Nesme-
yanov, Dokl. Akad. Nauk SSSR 1960, 132, 374; b) A. N. Nes-
meyanov, R. Kh. Freidlina, E. C. Chukovskaya, R. G. Petrova,
A. B. Belyavsky, Tetrahedron 1962, 17, 61.
[18] S. Enthaler, B. Hagemann, G. Erre, K. Junge, M. Beller, Chem.
Asian J. 2006, 1, 598.
[19] A. Ouali, J.-P. Majoral, A.-M. Caminade, M. Taillefer, Chem-
CatChem 2009, 1, 504.
[20] G. Zassinovich, G. Mestroni, S. Gladiali, Chem. Rev. 1992, 92,
1051.
[9] a) J.-S. Chen, L.-L. Chen, Y. Xing, G. Chen, W.-Y. Shen, Z.-
R. Dong, Y.-Y. Li, J.-X. Gao, Huaxue Xuebao 2004, 62, 1745;
b) H. Brunner, R. Eder, B. Hammer, U. Klement, J. Organomet.
Chem. 1990, 394, 555; c) H. Nishiyama, A. Furuta, Chem.
Commun. 2007, 760; d) N. Shaikh, S. Enthaler, K. Junge, M.
Beller, Angew. Chem. 2008, 120, 2531; Angew. Chem. Int. Ed.
2008, 47, 2497; e) J. Yang, T. Don Tilley, Angew. Chem. 2010,
122, 10384; Angew. Chem. Int. Ed. 2010, 49, 10186; f) J. Y. Wu,
[21] a) U. Matteoli, P. Frediani, M. Bianchi, C. Botteghi, S. Gladi-
ali, J. Mol. Catal. 1981, 12, 265; b) H. Brunner, M. Kunz,
Chem. Ber. 1986, 119, 2868; c) J. M. Brown, H. Brunner, W.
Leitner, M. Rose, Tetrahedron: Asymmetry 1991, 2, 331; d) X.
Wu, J. Xiao, Chem. Commun. 2007, 2449.
[22] F. G. Bordwell, Acc. Chem. Res. 1988, 21, 456.
[23] F. G. Bordwell, M. J. Bausch, J. Am. Chem. Soc. 1983, 105,
6188.
B. N. Stanzl, T. Ritter, J. Am. Chem. Soc. 2010, 132, 13214; g) [24] W. S. Matthews, J. E. Bares, J. E. Bartmess, F. G. Bordwell, F. J.
S. Zhou, S. Fleischer, K. Junge, S. Das, D. Addis, M. Beller,
Angew. Chem. 2010, 122, 8298; Angew. Chem. Int. Ed. 2010,
49, 8121; h) P. O. Lagaditis, A. J. Lough, R. H. Morris, Inorg.
Chem. 2010, 49, 10057; i) V. V. K. M. Kandepi, J. M. S. Car-
doso, E. Peris, B. Royo, Organometallics 2010, 29, 2777.
[10] K.-J. Haack, S. Hashiguchi, A. Fujii, T. Ikariya, R. Noyori,
Angew. Chem. 1997, 109, 297; Angew. Chem. Int. Ed. Engl.
1997, 36, 285. See also: a) R. Noyori, Adv. Synth. Catal. 2003,
345, 15; b) C. Perez, S. Perez, G. A. Fuentes, A. Corma, J. Mol.
Catal. A 2003, 197, 275; c) J. E. D. Martins, G. J. Clarkson, M.
Wills, Org. Lett. 2009, 11, 847.
Cornforth, G. E. Drucker, Z. Margolin, R. J. McCallum, G. J.
McCollum, N. R. Vanier, J. Am. Chem. Soc. 1975, 97, 7006.
[25] S. L. Buchwald, C. Bolm, Angew. Chem. 2009, 121, 5694; An-
gew. Chem. Int. Ed. 2009, 48, 5586.
[26] Schlögl, B. Rieger, Z. Naturforsch. Teil B 2004, 59, 233.
[27] For sample 1, see: J. W. Fitch III, J. J. Lagowski, Inorg. Chem.
1965, 4, 910. Sublimed at 160 °C under vacuum. For sample 2,
see: J. M. Manríquez, E. E. Bunel, B. Oelckers, E. Román, In-
org. Synth. 1997, 31, 267.
Received: April 19, 2011
Published Online: June 10, 2011
4360
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 4353–4360