Retro-Brook Rearrangement of Ferrocene-Derived Silyl Ethers

Article abstract of DOI:10.1002/chem.201501711

An intramolecular Li-Si exchange was observed on various lithiated ferrocenylbenzyl silyl ethers. The thermodynamically more stable C-silylated isomers were isolated in good yields and fully characterized. The reaction mechanism of the [1,4] retro-Brook rearrangement was investigated by DFT calculations. Two distinct reaction routes were proposed and a possible stabilization effect of the ferrocenyl fragment on the C-silylated isomers was described. The diastereoselective rearrangement of the trimethylsilyl group to the ortho position of the ferrocenyl cyclopentadienyl ring was also accomplished and the absolute configuration of the product was determined. Rearrange it: The lithiation of 2-bromophenyl-substituted ferrocenyl silyl ethers initiates an effective [1,4]-retro-Brook rearrangement (see scheme, R=methyl, tBu, or iPr). This rearrangement proceeds also if a carbanionic intermediate is generated by diastereoselective ortho-lithiation of the ferrocene moiety.

Full text of DOI:10.1002/chem.201501711

DOI: 10.1002/chem.201501711
Full Paper
&
Brook Rearrangement
Retro-Brook Rearrangement of Ferrocene-Derived Silyl Ethers
[a]
ˇ
Vladimír Bariak, Andrea Malastovµ, Ambroz Almµssy,* and Radovan Sebesta*
Abstract: An intramolecular Li–Si exchange was observed
on various lithiated ferrocenylbenzyl silyl ethers. The thermo-
dynamically more stable C-silylated isomers were isolated in
good yields and fully characterized. The reaction mechanism
of the [1,4] retro-Brook rearrangement was investigated by
DFT calculations. Two distinct reaction routes were proposed
and a possible stabilization effect of the ferrocenyl fragment
on the C-silylated isomers was described. The diastereoselec-
tive rearrangement of the trimethylsilyl group to the ortho
position of the ferrocenyl cyclopentadienyl ring was also
accomplished and the absolute configuration of the product
was determined.
suitable substrates for the Brook rearrangement too.[11] Acyl-
silanes in the presence of copper alkoxides undergo [1,2]-silyl
migration and afford 1-siloxy-1-alkenylcopper species.[12] In-
stead of lithium anions, a zinc-promoted Brook rearrangement
has also recently been described by Marek and co-workers.[13]
The reversed process, that is, the transfer of a silyl group from
an oxygen atom to a carbon atom, is also known but much
less studied.[14] Rawal and co-workers prepared silyl-substituted
salen ligands with the help of a retro-Brook rearrangement.[15]
Other Br–Li–Si exchanges were also described.[16] The afore-
mentioned geminal bis(silanes) can also be formed through re-
versed Brook rearrangement from 3-silyl allyloxysilanes.[17] The
carbanion, which is required for the retro-Brook rearrange-
ment, can be also formed by ortho or lateral deprotonation of
O-silylated phenols.[18] Silyloxymethyllithiums also undergo this
rearrangement.[19] The reversed Brook rearrangement was de-
scribed in only a few examples, possibly because of an unfav-
orable thermodynamic situation of this rearrangement. Trans-
fer of the silyl group from an oxygen atom to a carbon atom
can usually succeed for one of three reasons. It may be kineti-
cally viable, that is, much faster than the reaction from
a carbon atom to an oxygen atom. This type of retro-Brook re-
arrangement can be followed by subsequent trapping of the
transient intermediate.[18a] In the retro-Brook rearrangement,
stoichiometric amount of lithium base is required, therefore
formation of a stable lithium alkoxide can compensate for the
unfavorable change in the BDE due to a OÀSi versus CÀSi ex-
change.[20] It can also succeed if the molecule resulting from
the retro-Brook rearrangement is more stable for other reasons
than just a comparison of the OÀSi/CÀSi bond energies.
Introduction
Functionalized organosilane derivatives are useful as reagents
in organic synthesis and catalysis. However, introduction of
a silyl group at a desired position in the molecule is not always
routine. Therefore, the development of selective methods for
the formation of CÀSi bonds is highly desirable. Intramolecular
substitutions involving attack of an alkoxide on a silicon atom
result in the transfer of a silyl group from a carbon atom to an
oxygen atom.[1] This reaction has been for the first time
observed by Brook and is nowadays known as the Brook rear-
rangement.[2] It is usually driven by higher thermodynamic sta-
bility of an O-silylated derivative in comparison to a C-silylated
compound. The typical OÀSi bond dissociation energy (BDE) in
simple organosiloxanes is 510–560 kJmolÀ1, whereas the BDE
of a typical CÀSi bond is considerably lower, usually around
400 kJmolÀ1 or less.[3] The Brook rearrangement has been used
in various ways in the organic synthesis. The concept of anion
relay chemistry often uses a Brook rearrangement reaction as
a key step.[4] Geminal bis(silanes) are also useful substrates for
Brook rearrangements.[5] Highly functionalized compounds
such as siloxyallenes,[6] aminocyclopropanols,[7] or silyl dienol
ethers[8] can be obtained through the Brook rearrangement.
Carbanions, resulting from the Brook rearrangement of a-
siloxy silanes can be advantageously carboxylated with carbon
dioxide, thus leading to hydroxy acid derivatives.[9] On the
other hand, little is known about Brook rearrangements on or-
ganometallic compounds. Stueger and co-workers described
photoinduced Brook-type rearrangement reactions on acyl-
cyclopolysilanes.[10] Arene chromium tricarbonyl complexes are
In this context, we describe herein the retro-Brook re-
arrangement on O-silylated ferrocenyl alcohols initiated by a
Br–Li exchange as well as through diastereoselective ortho-lith-
iation of the ferrocene moiety. To the best of our knowledge,
there is no report of Brook-type rearrangement reactions on
ferrocene derivatives. These transformations constitute useful
addition to the synthetic toolbox, which enable the prepara-
tion of silylated ferrocenyl derivatives. The resulting C-silylated
ˇ
[a] V. Bariak, Dr. A. Malastovµ, Dr. A. Almµssy, Prof. R. Sebesta
Comenius University in Bratislava
Faculty of Natural Sciences, Department of Organic Chemistry
Mlynska dolina, Ilkovicova 6, CH-2, 842 15 Bratislava (Slovakia)
ˇ
E-mail: ambroz.almassy@fns.uniba.sk
radovan.sebesta@fns.uniba.sk
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201501711.
Chem. Eur. J. 2015, 21, 13445 – 13453
13445
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ferrocenyl alcohols are valuable starting materials for the
synthesis of various ferrocenyl compounds, including chiral
ligands.
Results and Discussion
Retro-Brook rearrangement initiated by Br–Li exchange
During our investigation of new ferrocenyl phosphanes we
attempted to prepare phosphane alcohol from the corre-
sponding silyl ether 1a through lithiation and subsequent
reaction with a chlorophosphane. However, no phosphane in-
corporation was observed. Instead, the seemingly unexpected
product of a [1,4] retro-Brook rearrangement 2a was isolated.
Furthermore, compound 2a was obtained in preparatively
highly interesting 79% yield with only small amount of the
debrominated alcohol 3b (Scheme 1).
Scheme 2. Synthesis of the silyl ethers 1a–1d and 6 and retro-Brook
rearrangement of the silyl ethers 1a–1d. HMDS=hexamethyldisilazane.
rial (16%) and the rearranged product (79%) with only trace
amount (5%) of the debrominated silyl ether (determined from
1H NMR spectroscopy of the raw mixture). As expected, the
retro-Brook rearrangement did not proceed on the substrate 6,
which carried the bromine in the para position of the phenyl
ring. Complete conversion of the starting materials together
with the presence of the debrominated silyl ethers isolated as
byproducts indicates that the initial fast Br–Li exchange is
followed by the slower intramolecular rearrangement step. The
concentration of the reactants affects the rate of the
bimolecular lithiation more significantly than the rate of the
intramolecular rearrangement. The experiment on substrate 6
also suggests that the retro-Brook rearrangement on com-
pounds 1 is an intramolecular process.
Scheme 1. Initial observation of the retro-Brook rearrangement of the silyl
ether 1a.
To explore the rearrangement of the silyl groups in more
detail, a series of racemic ferrocenyl(2-bromobenzyl)silyl ethers
was prepared from the corresponding alcohol rac-4 following
a procedure by Corey and Venkateswarlu (see the Experimental
Section, Method B).[21] The silyl ether 1b was obtained in 90%
yield upon treatment of the alcohol 4 with tert-butyldimethyl-
silyl (TBDMS) chloride in DMF with imidazole as a base. Com-
pound 1c was obtained in a similar way in 71% yield. Instead
of a silyl chloride, triisopropylsilyl (TIPS) triflate was employed.
However, it is likely that the nature of the silylating agent does
not influence the O-silylation. Introduction of the sterically
more demanding tert-butyldiphenylsilyl (TBDPS) group, by
using the corresponding TBDPS chloride led to the product 1d
in lower yield (30%) (Scheme 2). Also the 4-bromo trimethyl-
silyl (TMS) ether 6 was prepared by the silylation of the corre-
sponding alcohol 5 in 70% yield (see the Experimental Section,
Method A).[22] Compound 6 represents the type of substrate,
which is unable to undergo an intramolecular rearrangement
because of the long distance between the reacting centers.
Treatment of the aforementioned bromo silyl ethers 1a–1d
with nBuLi in THF at À788C for 2 h resulted in the clean
transfer of the silyl group from the oxygen atom to the ortho
position of the adjacent phenyl ring (Scheme 2). The corres-
ponding C-silyl alcohols 2a–2d were obtained in good yields
(53–79%) in all of these cases. Performing the reaction on silyl
ether 1a at a concentration of 160 mm resulted in the forma-
tion of 79% of the retro-Brook rearrangement product 2a in
2 h. However, the reaction of substrate 1a at higher dilution
(c=40 mm) afforded a mixture consisting of the starting mate-
We have also attempted to perform a “classical” Brook
rearrangement reaction on the product of the retro-Brook rear-
rangement, that is, the C-silylated alcohol 2a. However, expo-
sure of a solution of compound 2a in THF to BuLi overnight
led after workup only to the isolation of the starting material
2a. Brook rearrangement of compound 2a proceeded to
a small extent (8% of the corresponding product 1a was ob-
1
served by H NMR spectroscopy) when HMPA was used as an
additive. This experiment showed that on substrates 1, silyl
transfer from an oxygen atom to a carbon atom is more pref-
erable than the more common silyl transfer from a carbon
atom to an oxygen atom.
Retro-Brook rearrangement initiated by ortho-lithiation
On ferrocene derivatives, a carbanion can be advantageously
introduced by directed ortho-metalation of the cyclopentadi-
enyl (Cp) ring. The possibility to transfer a silyl group diastereo-
selectively to one of the ortho positions of the substituted Cp
ring could provide a useful method for the preparation of silyl-
Chem. Eur. J. 2015, 21, 13445 – 13453
www.chemeurj.org
13446
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
protected ferrocenyl alcohols possessing the element of planar
chirality. Configuration of the stereogenic plane is typically in-
duced in the diastereoselective-directed ortho-metalation and
is controlled by the configuration of the a-carbon chiral center.
Observed rearrangement may be, therefore, used for the ste-
reoselective protection of the preferred ortho position of the
Cp ring. Starting from the 1-ferrocenylethanol 3a and racemic
ferrocenylbenzylalcohol 3b, the OTMS ethers 7a and 7b were
prepared in good yields by silylation with hexamethyldisilazane
(HMDS) in nitromethane (Scheme 3).
Figure 1. Tentative transition state models explaining the diastereoselectivity
of the ortho-lithiation-initiated retro-Brook rearrangement.
ortho position of the Cp ring. However, the product 8a was
also accompanied with the desilylated alcohol 3a (Scheme 3).
The diastereoselectivity of the ortho-lithiation can be
explained by the tentative transition state model depicted in
Figure 1. Preference for the lithiation to the pro-Rp position
(pro-Rp denotes the position, from which the (Rp)-silylation
product results) is probably caused by minimization of the
steric repulsion between the silyl group and the a-methyl
group as well as the silyl group and the ferrocenyl moiety.
Undesirable desilylation can be prevented by the use of
bulkier silyl groups. The TBDMS ether 9a was completely re-
sistant towards desilylation side reaction. The rearrangement
of the tert-butyldimethylsilyl group provided two diastereo-
isomeric ortho-tert-butyldimethylsilylated derivatives 10a and
small amount of the product of the rearrangement reaction on
the non-substituted Cp ring 11 a. This alternative route was
even more pronounced in the case of the TBDPS ether 9b,
where the corresponding product 11 b was obtained as the
major compound (Scheme 4). It seems likely that the products
Scheme 3. Retro-Brook rearrangement initiated by ortho-lithiation.
Initial attempts to perform an ortho-lithiation-initiated retro-
Brook rearrangement by using nBuLi failed. We presumed that
the lithiation in the ortho position of the substituted Cp ring
requires a stronger base. However, upon treatment of the silyl
ethers 7a and 7b with tBuLi the rearrangement product 8a
was observed only in traces (<5%, determined by 1H NMR
spectroscopy) and the corresponding phenyl analogue 8b was
isolated in 9% yield. The main products of both of these reac-
tions were the corresponding desilylated alcohols 3a (isolated
in 70% yield) and 3b (isolated in 32% yield), respectively. How-
ever, only one of the possible diastereoisomers was detected
in both compounds 8a and 8b. These results show that ortho-
lithiation of ferrocenyl silyl ethers, followed by mutual ex-
change of lithium for a silyl group can proceed diastereoselec-
tively. To avoid the formation of possible byproducts due to
ortho-lithiation of the phenyl ring in compound 7b, we
decided to optimize the reaction conditions on silyl ether 3a.
The influence of additives on ortho-lithiation/retro-Brook re-
arrangement was explored. Various additives were described in
the literature, which enhance the reactivity or stereoselectivity
of ortho-lithiation reactions of ferrocene derivatives.[23] The ad-
dition of LiCl or N,N,N’,N’-tetramethylethylenediamine (TMEDA)
did not improve the yield of the desired product. On the other
hand, employment of tBuOK as additive improved the yield of
alcohol 8a significantly. It also allowed the rearrangement of
the less reactive substrates 9a and 9b, with bigger silyl
groups. The best results were achieved by using an excess of
the Schlosser base (4 molequiv nBuLi/tBuOK 1:1) in THF at
temperatures below À608C.[24]
11 a and 11 b result from
a long-range intramolecular
Scheme 4. Ortho-lithiation-initiated retro-Brook rearrangement on the
sterically demanding silyl ethers 9a and 9b. Determination of the absolute
configuration by chemical correlation (Ac=acetyl, py=pyridine).
The ortho-lithiation/rearrangement of the TMS ether 7a
proceeded regioselectively and diastereoselectively at the
Chem. Eur. J. 2015, 21, 13445 – 13453
www.chemeurj.org
13447
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
rearrangement. The distance in space between the C(Li) of the
non-substituted Cp ring and the Si atom is favorable for the
reaction.
is anti oriented to the FeÀCp moiety. Starting from the
homochiral TBDMS ether (R)-9a and considering the described
interactions, we have determined the absolute configuration
of the retro-Brook product (R,Rp)-10a for the major isomer and
(R,Sp)-10a for the minor isomer.
The homochiral O-silylated derivatives (R)-7a and (R)-9a
were prepared by previously described procedures (see the Ex-
perimental Section, Method A for compound 7a and Method B
for compound 9a) starting from alcohol (R)-3a, obtained by
enantioselective Corey–Bakshi–Shibata (CBS) reduction of ace-
tylferrocene.[25] Retro-Brook rearrangement was carried out by
using nBuLi/tBuOK as the base and the C2-silylated products
8a and 10a, respectively, were characterized by specific optical
rotation and CD spectroscopy.
Computational investigation of the reaction mechanism
The encouraging experimental results in the observed retro-
Brook rearrangement of the ferrocenyl silyl ethers 1 prompted
us to investigate the reaction mechanism by computational
methods. The main question that we posed was about the
possible reasons why this rearrangement works well on these
compounds. In other words, what effects govern the thermo-
dynamic viability of the reversed Brook rearrangement on this
type of substrates? Therefore, we decided to compare the
thermodynamic stability of the C-lithiated intermediate 13
with the C-silylated intermediate 14. We also would like to
locate a transition state of the rearrangement and thus find an
explanation for the formation of the CÀSi bond.
To determine the absolute configuration of the obtained
structures, alcohol 8a was converted to the corresponding
amine 12 and compared with compound (R,Sp)-12, described
earlier by Ugi and co-workers.[26a] Alcohol 8a was O-acetylated
with AcCl/py and the acetoxy group was substituted on treat-
ment with dimethylamine to yield compound 12 (Scheme 4) in
excellent overall yield (92%). It has been documented that nu-
cleophilic substitutions on the a-carbon atom in ferrocenyl
compounds proceed with retention of the configuration of the
The first step of the reaction sequence is an electrophilic
substitution of bromine with lithium upon treatment of com-
pound 1a with BuLi. This Br–Li exchange is fast and exother-
mic. The reaction produces BuBr and the C-lithiated intermedi-
ate 13 represented by two stable conformers, that is, syn-13
and anti-13 (syn and anti denote the orientation of the lithium
atom with respect to the iron atom in the ferrocene moiety).
Conformer anti-13 is about 5 kJmolÀ1 thermodynamically more
stable than conformer syn-13 (calculated at the RI-SCS-MP2
level by using def2-TZVP basis set)[38,39] (Figures 3 and 4). The
modeling of the reaction mechanism showed that despite the
stability of conformer anti-13 its conversion to the rearrange-
ment product anti-14 through the transition state TS-1 pro-
ceeds through a higher activation barrier (Ea =56.8 kJmolÀ1,
DFT, RI-SCS-MP2/def2-TZVP). On the other hand, the less stable
intermediate syn-13 requires a lower activation energy (Ea =
39.0 kJmolÀ1) to transform to the product syn-14 through tran-
sition state TS-2. The conformer syn-14 is calculated to be
more stable than anti-14 by approximately 8 kJmolÀ1. These
calculations confirmed that the C-silylated isomer 14 is ener-
getically more favorable than the O-silylated isomer 13. There-
fore, the equilibrium should be strongly shifted on the side of
the C-silylated isomer (>99.95%). This notion is based on
a Boltzmann distribution of the isomers syn-13 and syn-14
(DE=18.9 kJmolÀ1). Indeed, virtually no product 7a of a Brook
rearrangement was observed upon treatment of compound
2a with nBuLi for several hours at room temperature. Only
small amount of the product of a Brook rearrangement (8%)
was observed, when HMPA was used as additive.
1
stereogenic center.[27] The analysis of the H NMR spectra re-
vealed that our diastereoisomer 12 [d=0.29 (TMS), 1.18 (CH3),
2.36 (NMe2), 3.34 (CaH), 4.12 ppm (Cp)] differs from compound
(R,Sp)-12 [d=0.26 (TMS), 1.22 (CH3), 2.04 (NMe2), 3.82 (CaH),
4.07 ppm (Cp)] reported in the literature.[26b] Signs and values
of the optical rotation match literature data (see the Experi-
mental Section for details). The CD spectra also confirm that
amine 12 obtained from the retro-Brook rearrangement and
the one obtained by lithiation of the Ugi amine are diastereo-
isomers (see the Supporting information). Therefore, the
absolute configuration of the product of the retro-Brook
rearrangement was assigned as (R,Rp)-8a.
The relative configuration of the isolated diastereoisomers
10a was determined by NOESY NMR experiments on both
isomers (Figure2 ). For both diastereoisomers, in the most
Figure 2. Determination of the relative configuration of compound 10a by
NOESY NMR spectroscopy.
preferred conformer the tert-butyl group is located anti to the
FeÀCp moiety. Repulsion of the Ca methyl group with the
TBDMS group affects the conformation of the chiral center on
the a-carbon atom resulting in a positive NOE between the Ca
methyl group and the corresponding C5ÀH in both diastereo-
isomers. Whereas in the major isomer the OH group is syn ori-
ented with respect to the FeÀCp moiety, in the minor isomer
the Ca methyl group as well as the Ca proton display interac-
tions with the unsubstituted Cp ring. Therefore, the OH group
The energy of the C-silylated isomer 14 is lower in both
conformations, although during the rearrangement, an
energetically less stable bond (i.e., a CÀSi bond) is formed.
The stabilizing electronic effects were studied by population
analysis of all electrons condensed on the atoms (see the Sup-
porting Information, important overlaps are highlighted). This
analysis showed that the rearrangement of isomer 13 to com-
pound 14 is accompanied with an increase of the positive
Chem. Eur. J. 2015, 21, 13445 – 13453
www.chemeurj.org
13448
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Although, quantum chemical
calculations suggest a stabilizing
effect of ferrocene, the reaction
proceeds without ferrocene too.
Control experiments on two sub-
strates also provided the corre-
sponding retro-Brook rearrange-
ment products. The reaction on
(1-(2-bromophenyl)ethoxy)trime-
thylsilane proceeded cleanly, but
reaction on ((2-bromophenyl)(-
phenyl)methoxy)trimethylsilane
afforded
a mixture of com-
pounds consisting of the starting
material, the retro-Brook rear-
rangement product, and the de-
brominated compound (see the
Supporting Information for de-
tails).
Figure 3. Reaction coordinate for the retro-Brook rearrangement reaction of compound 1a.
Figure 4. Calculated transition states for the retro-Brook rearrangement
reaction of compound 1a.
Scheme 5. Mechanism of the retro-Brook rearrangement reaction on
ferrocenyl silyl ethers.
overlap of the oxygen-bonded Li atom with the adjacent aro-
matic system of the ferrocenyl moiety. Although the conformer
anti-13 is completely lacking this interaction, the correspond-
ing O-lithiated conformer anti-14 is stabilized by interaction of
the lithium atom with the carbon atoms C1 and C2 of the sub-
stituted Cp ring. A different situation occurs in compounds
syn-13 and syn-14, where the Li is located in between the cy-
clopentadienyl rings. In the C-lithiated isomer syn-13 a positive
overlap is observed between the lithium atom and the nearest
carbon atoms C1 and C17 of both cyclopentadienyl rings and
the Fe atom. A similar Li–Fe interaction was computationally
predicted in other derivatives too.[28] The shift of the Li atom to
an oxygen atom results in shortening of the LiÀFe distance
and thus the overlap of both atoms and the stabilizing interac-
tion increases. The proposed mechanism of the retro-Brook re-
arrangement consists of two processes. The first process is the
attraction of the lithium cation towards the aromatic system of
ferrocene. This interaction is then followed by the transfer of
the silyl group. In this step the SiÀO bond is broken and a new
CÀSi bond is formed. It can proceed from both conformers
syn-13 and anti-13 (Scheme 5).
Conclusion
Ferrocenylbenzylic silyl ethers undergo effective intramolecular
silyl group transfer to the ortho position of the phenyl ring,
which was initiated by a Br–Li exchange upon treatment with
nBuLi. The influence of the adjacent ferrocenyl fragment on
the course of the retro-Brook rearrangement was explained by
DFT calculations and a plausible reaction mechanism was pro-
posed. Similar rearrangement to the ortho position of the fer-
rocenyl Cp ring was also observed. This reaction was initiated
by diastereoselective ortho-lithiation of the ferrocene moiety.
In case of a TMS group, the resulting ortho-silylated alcohols
were obtained in moderate yield (35%). Interestingly, retro-
Brook rearrangement of the TMS group prefers the opposite
diastereotopic site compared to classical diastereoselective
ortho-lithiation of an Ugi amine. In the case of a TBDPS
derivative, the rearrangement occurred predominantly to the
unsubstituted Cp ring.
Chem. Eur. J. 2015, 21, 13445 – 13453
www.chemeurj.org
13449
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(dq, J=6.4, 4.6 Hz, 1H; -CH(OH); 13C NMR (75 MHz, CDCl3): d=23.9
(-CH3), 65.8 (-CH-OH), 66.3 (Fc), 66.4 (Fc), 68.1 (Fc), 68.1 (Fc), 68.5
(C5H5), 95.0 ppm(Fc).
Experimental Section
Starting materials
1
2-Bromobenzoylferrocene,[25] 4-bromobenzoylferrocene,[28] and
benzoylferrocene[29] were prepared by Friedel–Craft acylation ac-
cording to literature procedures. Compound (R,Sp)-12 was prepared
according to the procedure of Ugi and co-workers.[26]
Ferrocenyl(phenyl)methanol (rac-3b): Yield: 87%; H NMR (300 MHz,
CDCl3): d=2.49 (d, J=3.2 Hz, 1H; -OH), 4.13–4.23 (m, 4H; Fc), 4.21
(s, 5H; C5H5), 5.45 (d, J=3.1 Hz, 1H; -CH(OH)), 7.20–7.27 (m, 1H;
Ph), 7.27–7.35 (m, 2H; Ph), 7.35–7.41 ppm (m, 2H; Ph); 13C NMR
(75 MHz, CDCl3): d=66.0 (Fc), 67.4 (Fc), 68.1 (Fc), 68.1 (Fc), 68.5
(C5H5), 72.0 (-CH-OH), 94.2 (Fc), 126.2 (2 ” Ph), 127.4 (Ph), 128.2 (2”
Ph), 143.2 ppm (Ph).
4-Bromophenyl(ferrocenyl)methanol (rac-5): Yield: 67%;1H NMR
(300 MHz, CDCl3): d=2.47 (d, J=2.9 Hz, 1H; -OH), 4.14–4.25 (m,
4H; Fc), 4.22 (s, 5H; C5H5), 5.40 (d, J=2.9 Hz, 1H; -CH(OH)), 7.22–
7.28 (m, 2H; Ph), 7.40–7.47 ppm (m, 2H; Ph); 13C NMR (75 MHz,
CDCl3): d=65.7 (Fc), 67.4 (Fc), 68.3 (Fc), 68.3 (Fc), 68.5 (C5H5), 71.3
(-CH-OH), 94.0 (Fc), 121.2 (Ph), 127.9 (2”Ph), 131.2 (2”Ph),
142.2 ppm (Ph).
NMR spectra were recorded on Varian NMR System intruments at
1
238C at frequency 300 or 600 MHz for H and 75 or 151 MHz for
13C, respectively. Chemical shifts (d) are given in ppm relative to
tetramethylsilane as an internal standard. IR spectra were
measured on Thermo Scientific Nikolet iS10 spectrometer with
Smart iTR technology with diamond ATR. Elemental analyses were
performed on varioMICRO cube (Elementar) instrument for simul-
taneous C,H,N,S analysis with options for O. High-resolution mass
spectra were measured on
a mass spectrometer with H-ESI
Orbitrap ionization in positive mode.
(a-Ferrocenyl-2-bromobenzyloxy)trimethylsilane (rac-1a): Yield: 86%;
1
m.p. 58–608C; H NMR (300 MHz, CDCl3): d=0.09 (s, 9H; -Si(CH3)3),
General procedures
4.03–4.09 (m, 2H; Fc), 4.13 (s, 5H; C5H5), 4.15–4.18 (m, 2H; Fc), 5.97
(s, 1H; -CH-OTMS), 7.04–7.12 (m, 1H; Ph), 7.26–7.32 (m, 1H; Ph),
7.48 (dd, J=8.0, 1.2 Hz, 1H; Ph), 7.6 ppm (dd, J=7.8, 1.8 Hz, 1H;
Ph); 13C NMR (75 MHz, CDCl3): d=0.20 (-Si(CH3)3), 66.1 (Fc), 67.0
(Fc), 67.1 (Fc), 67.7 (Fc), 68.7 (C5H5), 71.5 (-CH-OTMS), 93.3 (Fc),
121.6 (Ph), 127.5 (Ph), 128.6 (Ph), 129.2 (Ph), 132.1 (Ph), 144.4 ppm
(Ph); IR (neat): n˜ =818, 841, 878, 1019, 1080, 1117, 1250, 2899,
2960, 3053, 3101 cmÀ1; elemental analysis calcd for C20H23BrFeOSi:
C 54.2, H 5.23; found: C 54.06, H 5.25.
Reduction of ketones:
A solution of the starting material
(1.4 mmol, 1.0 equiv) in dry THF (10 mL) was added dropwise to
a solution of LiAlH4 (3.5 mmol, 2.5 equiv) in dry THF (10 mL) at 08C.
The progress of the reaction was monitored by TLC (SiO2, hexane/
Et2O 4:1). After completion the reaction was quenched with ice
cold water. After filtration of the resulting mixture the organic
phase was washed with water and brine, dried over Na2SO4, and
evaporated under vacuum. The product was separated by column
chromatography (SiO2, hexane/EtOAc 8:1).
tert-Butyl(a-ferrocenyl-2-bromobenzyloxy)dimethylsilane
(rac-1b):
1
O-Silylation of alcohols: Method A: HMDS (1.1 mmol, 1.1 equiv)
was added to a solution of the alcohol (1.0 mmol, 1.0 equiv) in
MeNO2 (1 mL) at RT. The progress of the reaction was monitored
by TLC (SiO2, hexane/Et2O 4:1). After completion the reaction mix-
ture was concentrated under vacuum. The product was separated
by column chromatography (SiO2, hexane/EtOAc 8:1).
Yield: 90%; m.p. 75–778C. H NMR (300 MHz, CDCl3): d=À0.18 (s,
3H; -SiCH3), 0.16 (s, 3H; -SiCH3), 0.96 (s, 9H; -SiC(CH3)3), 4.01–4.10
(m, 3H; Fc), 4.14 (s, 5H; C5H5), 4.11–4.16 (m, 1H; Fc), 5.98 (s, 1H;
-CH-OTBDMS), 7.06–7.13 (m, 1H; Ph), 7.27–7.34 (m, 1H; Ph), 7.49
(dd, J=8.0, 1.2 Hz, 1H; Ph), 7.60 ppm (dd, J=7.8, 1.5 Hz, 1H; Ph);
13C NMR (75 MHz, CDCl3): d=À4.9 (-SiCH3), À4.7 (-SiCH3), 18.1
(-SiC(CH3)3), 25.8 (-SiC(CH3)3), 65.6 (Fc), 66.7 (Fc), 67.3 (Fc), 67.7 (Fc),
68.6 (C5H5), 71.4 [-CH(OTBDMS)], 93.6 (Fc), 121.6 (Ph), 127.4 (Ph),
128.6 (Ph), 129.2 (Ph), 132.0 (Ph), 144.4 ppm (Ph); IR (neat): n˜ =810,
870, 1022, 1105, 1260, 1463, 1712, 2854, 2924, 2949, 3093 cmÀ1; el-
emental analysis calcd for C23H29BrFeOSi: C 56.92, H 6.02; found: C
57.22, H 6.07.
Method B: To a mixture of the alcohol (0.67 mmol, 1.0 equiv), imida-
zole (1.68 mmol, 2.5 equiv), and DMF or dichloromethane (2 mL)
the appropriate silyl reagent (1.68 mmol, 2.5 equiv) was added at
RT. The mixture was stirred at RT overnight. The reaction mixture
was concentrated in vacuum and the product was separated by
column chromatography (SiO2, hexane/EtOAc 8:1).
Lithiation and retro-Brook-rearrangement: A Solution of the silyl
ether (0.32 mmol, 1 equiv) in dry THF (2 mL) was cooled to À788C.
A solution of nBuLi in hexane (1.6m, 0.38 mmol, 1.2 equiv) was
added dropwise. The mixture was stirred at À788C for 2 h and
then the temperature was allowed to rise to RT. The reaction was
quenched by addition of water (1 mL). The layers were separated
and the aqueous layer was extracted with Et2O (2”5 mL). The com-
bined organic phases were dried (Na2SO4) and concentrated under
vacuum. The product was separated by column chromatography
(SiO2, hexane/EtOAc 8:1).
2-Bromophenyl(ferrocenyl)methanol (rac-4): Yield: 87%; 1H NMR
(300 MHz, CDCl3): d=2.64 (d, J=3.6 Hz, 1H; -OH), 4.12–4.18 (m,
2H; Fc), 4.19–4.22 (m, 1H; Fc), 4.26 (s, 5H; C5H5), 4.39–4.42 (m, 1H;
Fc), 5.79 (d, J=3.6 Hz, 1H; -CH(OH)), 7.15–7.07 (m, 1H; Ph), 7.35–
7.28 (m, 1H; Ph), 7.49 (dd, J=8.0, 1.1 Hz, 1H; Ph), 7.61 ppm (dd,
J=7.8, 1.8 Hz, 1H; Ph); 13C NMR (75 MHz, CDCl3): d=66.3 (Fc), 67.6
(Fc), 67.9 (Fc), 68.2 (Fc), 68.5 (C5H5), 70.2 (-CH-OH), 93.7 (Fc), 122.4
(Ph), 127.5 (Ph), 127.8 (Ph), 128.9 (Ph), 132.6 (Ph), 142.4 ppm (Ph).
(a-Ferrocenyl-2-bromobenzyloxy)triisopropylsilane (rac-1c): Yield:
71%; m.p. 59–638C; 1H NMR (300 MHz, CDCl3): d=0.99 (d, J=
6.2 Hz, 9H; -SiCH(CH3)2), 1.01–1.07 (m, 12H; -SiCH(CH3)2), 3.94–4.00
(m, 1H; Fc), 4.01–4.06 (m, 1H; Fc), 4.04 (s, 5H; C5H5), 4.08–4.12 (m,
1H; Fc), 4.15–4.20 (m, 1H; Fc), 6.04 (s, 1H; -CH-OTIPS), 7.10–7.18
(m, 1H; Ph), 7.33–7.41 (m, 1H; Ph), 7.52 (dd, J=8.0, 1.2 Hz, 1H;
Ph), 7.77 ppm (dd, J=7.8, 1.5 Hz, 1H; Ph); 13C NMR (75 MHz,
CDCl3): d=11.4, 16.9, 17.0, 65.0 (Fc), 65.7 (Fc), 66.4 (Fc), 67.4 (Fc),
67.6 (C5H5), 70.4(-CH-OTIPS), 92.8 (Fc), 121.0 (Ph), 126.2 (Ph), 127.6
(Ph), 128.4 (Ph), 130.9 (Ph), 143.8 ppm (Ph); IR (neat): n˜ =812, 882,
1020, 1037, 1052, 1084, 1105, 1260, 1462, 2861, 2938, 3094 cmÀ1
;
elemental analysis calcd for C26H35BrFeOSi: C 59.21, H 6.69; found:
C 59.14, H 6.83.
tert-Butyl(a-ferrocenyl-2-bromobenzyloxy)diphenylsilane
(rac-1d):
1
Yield: 30%; m.p. 102–1068C; H NMR (300 MHz, CDCl3): d=1.05 (s,
9H; -SiC(CH3)3), 3.91–3.97 (m, 1H; Fc), 3.94 (s, 5H; C5H5), 3.98–4.02
(m, 2H; Fc), 4.06–4.02 (m, 1H; Fc), 6.05 (s, 1H; -CH-OTBDPS), 7.03–
7.12 (m, 1H; Ph), 7.18–7.26 (m, 2H; Ph), 7.28–7.49 (m, 8H; Ph),
7.67–7.74 (m, 2H; Ph), 7.78 ppm (dd, J=7.8, 1.6 Hz, 1H; Ph);
13C NMR (75 MHz, CDCl3): d=19.4 (-SiC(CH3)3), 27.0 (-SiC(CH3)3), 66.2
(Fc), 66.9 (Fc), 67.8 (Fc), 68.7 (Fc), 68.9 (C5H5), 72.4 (-CH-OTBDPS),
1-Ferrocenylethanol [(R)-3a]: Yield: 89%; 1H NMR (300 MHz, CDCl3):
d=1.44 (d, J=6.4 Hz, 3H; -CH3), 1.88 (d, J=4.6 Hz, 1H; -OH), 4.14–
4.17 (m, 2H; Fc), 4.18–4.23 (m, 2H; Fc), 4.19 (s, 5H; C5H5), 4.54 ppm
Chem. Eur. J. 2015, 21, 13445 – 13453
www.chemeurj.org
13450
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
93.1 (Fc), 122.1 (Ph), 127.2 (Ph), 127.4 (2 ” Ph), 127.5 (2”Ph), 128.5
(Ph), 129.5 (2”Ph), 129.6 (Ph), 131.8 (Ph), 133.3 (Ph), 133.8 (Ph),
136.0 (2”Ph), 136.0 (2”Ph), 144.1 ppm (Ph); IR (neat): n˜ =819,
1020, 1064, 1105, 1260, 1469, 2856, 2928, 2963, 3064 cmÀ1; elemen-
tal analysis calcd for C33H33BrFeOSi: C 65.03, H 5.46; found: C 64.73,
H 5.78.
J=7.5, 1.2 Hz, 1H; Ph), 7.58 ppm (dd, J=7.8, 1.1 Hz, 1H; Ph);
13C NMR (75 MHz, CDCl3): d=13.1, 19.2, 19.3, 66. 9 (Fc), 67.2 (Fc),
67.8 (Fc), 68.5 (C5H5), 68.6 (Fc), 71.2 (-CH(OH)), 94.9 (Fc), 126.8 (Ph),
127.6 (Ph), 129.0 (Ph), 134.2 (Ph), 136.1 (Ph), 149.5 ppm (Ph); IR
(neat): n˜ =803, 881, 1011, 1104, 1260, 1462, 2864, 2947, 3553 cmÀ1
;
elemental analysis calcd for C26H36FeOSi: C 69.63, H 8.09; found: C
69.58, H 8.17.
(a-Ferrocenyl-4-bromobenzyloxy)trimethylsilane (rac-6): Yield: 70%;
1
m.p. 84–858C; H NMR (300 MHz, CDCl3): d=0.07 (s, 9H; -Si(CH3)3),
2-(tert-Butyldiphenylsilyl)phenyl(ferrocenyl)methanol (rac-2d): Yield:
62%; m.p. 50–588C; 1H NMR (300 MHz, CDCl3): d=1.19 (s, 9H;
-SiC(CH3)3), 1.25 (d, J=3.2 Hz, 1H; -OH), 3.19–3.25 (m, 1H; Fc), 3.80
(s, 5H; C5H5), 3.83–3.88 (m, 1H; Fc), 3.95–4.00 (m, 1H; Fc), 4.25–
4.30 (m, 1H; Fc), 5.41 (d, J=3.2 Hz, 1H; -CH-OH), 7.51–7.27 (m, 9H;
Ph), 7.71–7.77 (m, 2H; Ph), 7.78–7.84 (m, 2H; Ph), 7.92–7.99 ppm
(m, 1H; Ph); 13C NMR (75 MHz, CDCl3): d=18.8 (-SiC(CH3)3), 28.3
(-SiC(CH3)3), 66.3 (Fc), 66.5 (Fc), 66.6 (Fc), 67.9 (Fc), 68.5 (C5H5), 71.0
(-CH(OH)), 92.2 (Fc), 126.4 (Ph), 127.9 (Ph), 128.1 (Ph), 129.1 (Ph),
129.4 (Ph), 129.7 (Ph), 130.4 (Ph), 132.7 (Ph), 135.6 (Ph), 135.8 (Ph),
136.2 (Ph), 136.5 (Ph), 151.3 ppm (Ph); IR (neat): n˜ =817, 1007,
1103, 1427, 1588, 2857, 2927, 3069, 3564 cmÀ1; elemental analysis
calcd for C33H34BrFeOSi: C 74.71, H 6.46; found: C 74.8, H 6.77.
Phenyl(2-trimethylsilylferrocenyl)methanol (rac-8b):[31] Yield: 9%;
1H NMR (300 MHz, CDCl3): d=0.34 (s, 9H; -Si(CH3)3), 1.97 (d, 1H;
J=3.8 Hz, 1H; -CH-OH), 3.92–3.95 (m, 1H; Fc), 4.03 (s, 5H; C5H5),
4.13–4.16 (m, 1H; Fc), 4.24–4.27 (m, 1H; Fc), 5.69 (d, J=3.8 Hz, 1H;
-CH-OH), 7.28–7.36 (m, 1H; Ph), 7.36–7.44 (m, 2H; Ph), 7.46–
7.52 ppm (m, 2H; Ph).
3.86–3.90 (m 1H; Fc), 4.05–4.11 (m, 2H; Fc), 4.08 (s, 5H; C5H5) 4.12–
4.16 (m, 1H; Fc), 5.50 (s, 1H; -CH-OTMS), 7.26–7.32 (m, 2H; Ph),
7.41–7.50 ppm (m, 2H; Ph); 13C NMR (75 MHz, CDCl3): d=0.3
(-Si(CH3)3), 66.5 (Fc), 67.3 (Fc), 67.7 (Fc), 67.8 (Fc), 68.7 (C5H5), 72.9
(-CH-OTMS), 93.1 (Fc), 120.9 (Ph), 128.3 (2”Ph), 131.1 (2”Ph),
143.5 ppm (Ph); IR (neat): n˜ =815, 839, 883, 1010, 1086, 1103, 1249,
1486, 2889, 2957, 3093 cmÀ1
; elemental analysis calcd for
C20H23BrFeOSi: C 54.2, H 5.23; found: C 54.4, H 5.18.
1-Ferrocenylethyloxytrimethylsilane [(R)-7a]:[30] Yield: 86%; [a]2D0
=
1
À10.6 (c=0.50, CHCl3); H NMR (300 MHz, CDCl3): d=0.13 (s, 9H;
-Si(CH3)3), 1.50 (d, J=6.4 Hz, 3H; -CHCH3), 4.08–4.17 (m, 3H; Fc),
4.14 (s, 5H; C5H5), 4.18–4.22 (m, 1H; Fc), 4.72 ppm (q, J=6.4 Hz,
1H; -CH-OTMS); 13C NMR (75 MHz, CDCl3): d=0.5 (-Si(CH3)3), 23.9
(-CHCH3), 65.7(Fc), 66.5 (-CH-OTMS), 67.5 (2”Fc), 67.8 (Fc), 68.5
(C5H5), 93.1 ppm (Fc).
a-Ferrocenylbenzyloxytrimethylsilane (rac-7b): Yield: 82%; m.p. 58–
598C; 1H NMR (300 MHz, CDCl3): d=0.06 (s, 9H; -Si(CH3)3), 3.90–
3.94 (m, 1H; Fc), 4.03–4.10 (m, 2H; Fc), 4.07 (s, 5H; C5H5), 4.17–4.20
(m, 1H; Fc), 5.54 (s, 1H; -CH-OTMS), 7.20–7.28 (m, 1H; Ph), 7.28–
7.36 (m, 2H; Ph), 7.37–7.44 ppm (m, 2H; Ph); 13C NMR (75 MHz,
CDCl3): d 0.3 (-Si(CH3)3), 66.7 (Fc), 67.4 (Fc), 67.5 (Fc), 67.7 (Fc), 68.6
(C5H5), 73.5 (-CH-OTMS), 93.6 (Fc), 126.6 (2”Ph), 127.1 (Ph), 127.9
(2”Ph), 144.4 ppm (Ph);IR (neat): n˜ =813, 838, 883, 1069, 1102,
1-Ferrocenylethyloxy(tert-butyl)dimethylsilane [(R)-9a]: Yield: 68%;
[a]2D0 = +15.6 (c=0.50, CHCl3); 1H NMR (300 MHz, CDCl3): d=0.07
(s, 3H; -SiCH3), 0.09 (s, 3H; -SiCH3), 0.93 (s, 9H; -SiC(CH3)3), 1.46 (d,
J=6.3 Hz, 3H; -CHCH3), 4.05–4.08 (m, 1H; Fc), 4.09–4.12 (m, 2H;
Fc), 4.14–4.18 (m, 1H; Fc), 4.15 (m, 6H; Fc, C5H5), 4.67 ppm (q, J=
6.3 Hz, 1H; -CHCH3); 13C NMR (75 MHz, CDCl3): d=À4.6 (-SiCH3),
À4.2 (-SiCH3), 18.1 (-SiC(CH3)3), 24.1 (-CHCH3), 25.9 (-SiC(CH3)3), 65.7,
66.1, 67.2, 67.3, 67.4, 68.5 (C5H5), 94.3 ppm (Fc); HRMS calcd for
C18H28FeOSi: 367.115652 [M+Na]+; found: 367.11473.
1248, 2886, 2955, 3081 cmÀ1
; elemental analysis calcd for
C20H24FeOSi: C 65.93, H 6.64; found: C 65.77, H 6.63.
Ferrocenyl(2-(trimethylsilyl)phenyl)methanol (rac-2a): Yield: 79%;
m.p. 97–1018C; 1H NMR (300 MHz, CDCl3): d=0.37 (s, 9H;
-Si(CH3)3), 2.40 (d, J=3.7 Hz, 1H; -OH), 4.14–4.22 (m, 4H; Fc), 4.26
(s, 5H; C5H5), 5.73 (d, J=3.7 Hz, 1H; -CH(OH)), 7.20–7.28 (m, 1H;
Ph), 7.30–7.38 (m, 1H; Ph), 7.43–7.52 ppm (m, 2H; Ph); 13C NMR
(75 MHz, CDCl3): d=0.9 (-Si(CH3)3), 67.0 (Fc), 67.3 (Fc), 67.4 (Fc),
68.6 (C5H5), 71.1 (-CH(OH)), 94.5 (Fc), 126.7 (Ph), 127.0 (Ph), 129.4
(Ph), 134.3 (Ph), 137.9 (Ph), 148.7 ppm (Ph); IR (neat): n˜ =836, 999,
1122, 1250, 2926, 2958, 3045, 3089, 3534, 3585 cmÀ1; elemental
analysis calcd for C20H24FeOSi: C 65.93, H 6.64; found: C 65.72, H
6.65.
1-Ferrocenylethyloxy(tert-butyl)diphenylsilane (rac-9b): Yield: 98%;
1H NMR (300 MHz, CDCl3): d=1.07 (s, 9H; -SiC(CH3)3), 1.40 (d, J=
6.3 Hz, 3H; -CHCH3), 3.87–3.93 (m, 1H; Fc), 3.98–4.03 (m, 1H; Fc),
4.06–4.11 (m, 1H; Fc), 4.08 (s, 5H; C5H5), 4.21–4.25 (m, 1H; Fc), 4.71
(q, J=6.3 Hz, 1H; -CHCH3), 7.30–7.48 (m, 6H; Ph), 7.62–7.69 (m,
2H; Ph), 7.69–7.76 ppm (m, 2H; Ph); 13C NMR (75 MHz, CDCl3): d=
19.3 (-SiC(CH3)3), 23.8 (-CHCH3), 27.1 (-SiC(CH3)3), 65.6, 67.2, 67.4,
67.5, 67.8, 68.5 (C5H5), 93.6 (Fc), 127.4 (Ph), 129.5 (Ph), 134.1 (Ph),
134.7 (Ph), 136.0 ppm (Ph); HRMS calcd for C28H32FeOSi:
468.157182 [M]+; found: 468.15648.
2-(tert-Butyldimethylsilyl)phenyl(ferrocenyl)methanol (rac-2b): Yield:
68%; m.p. 122–1248C; 1H NMR (300 MHz, CDCl3): d=0.32 (s, 3H;
-SiCH3), 0.43 (s, 3H; -SiCH3), 0.94 (s, 9H; -SiC(CH3)3), 2.35 (d, J=
4.0 Hz, 1H; -OH), 4.14–4.22 (m, 4H; Fc), 4.24 (s, 5H; C5H5), 5.70 (d,
J=4.0 Hz, 1H; -CH-OH), 7.19–7.26 (m, 1H; Ph), 7.29–7.37 (m, 1H;
Ph), 7.44–7.52 ppm (m, 2H; Ph); 13C NMR (75 MHz, CDCl3): d=À3.0
(-SiCH3), À2.3 (-SiCH3), 17.6 (-SiC(CH3)3), 27.0 (-SiC(CH3)3), 66.8 (Fc),
67.2 (Fc), 67.5 (Fc), 68.5 (C5H5), 68.6 (Fc), 71.5 (-CH(OH)), 94.6 (Fc),
126.6 (Ph), 126.9 (Ph), 129.2 (Ph), 135.4 (Ph), 135.9 (Ph), 149.2 ppm
(Ph); IR (neat): n˜ =820, 1005, 1294, 1359, 2853, 2922, 2947,
3550 cmÀ1; elemental analysis calcd for C23H30FeOSi: C 67.97, H
7.44; found: C 67.99, H 7.53.
Lithiation and retro-Brook rearrangement to the ortho position
of the ferrocenyl Cp ring: nBuLi (1.6m solution in hexane,
4.0 equiv) was added dropwise to a solution of 1-ferrocenylethoxy-
silane (1.0 equiv) and tBuOK (4.0 equiv) in anhydrous THF (6 mL
per mmol of substrate) at À658C. The mixture was stirred at
À658C for 24 h. A saturated aqueous solution of NH4Cl (2 mL) was
added and the mixture was stirred until the temperature increased
to RT. The layers were separated and the aqueous part was extract-
ed with Et2O (3”5 mL). The combined organic parts were dried
over Na2SO4 and evaporated. The products were isolated by
column chromatography (SiO2, hexane/EtOAc 8:1 + 1% Et3N).
Ferrocenyl(2-triisopropylsilylphenyl)methanol (rac-2c): Yield: 53%;
m.p. 74–788C; H NMR (300 MHz, CDCl3): d=1.11 (d, J=7.5 Hz, 9H;
3”-SiCHCH3), 1.17 (d, J=7.5 Hz, 9H; 3”-SiCHCH3), 1.49 (sp, J=7.5,
3H; 3”-CH(CH3)2), 2.40 (d, J=4.1, 1H; -OH), 4.14–4.20 (m, 3H; Fc),
4.21–4.25 (m, 1H; Fc), 4.27 (s, 5H; C5H5), 5.57 (d, J=4.1 Hz, 1H;
-CH(OH)), 7.19–7.28 (m, 1H; Ph), 7.29–7.38 (m, 1H; Ph), 7.52 (dd,
1-(2-Trimethylsilylferrocenyl)ethanol [(R,Rp)-8a]: Yield: 35% and 8%
of the desilylated alcohol 3a; [a]2D0 =À44.7 (c=0.83, CHCl3);
1H NMR (300 MHz, CDCl3): d=0.27 (s, 9H; -Si(CH3)3), 1.32 (d, J=
6.3 Hz, 3H; -CHCH3), 2.06 (s, 1H; -OH), 4.08–4.12 (dd, J=2.3, 1.3 Hz,
1H; Fc), 4.20 (s, 5H; C5H5), 4.29–4.33 (m, 1H; Fc), 4.44–4.49 (m, 1H;
Fc), 4.58 ppm [dq, J=6.3, 1.2 Hz, 1H; -CH(OH)]; 13C NMR (75 MHz,
1
Chem. Eur. J. 2015, 21, 13445 – 13453
www.chemeurj.org
13451
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
CDCl3): d=1.0 (-Si(CH3)3), 25.0 (CHCH3), 65.4 (-CH-OH), 67.2 (Fc),
68.3 (C5H5), 70.0 (Fc), 70.5 (CFc-Si), 74.0 (Fc), 101.4 ppm (Fc); HRMS
calcd for C15H22FeOSi: 302.078932 [M]+; found: 302.07789.
1-(1’-tert-Butyldiphenylsilylferrocenyl)ethanol (rac-11b): Yield: 28%;
1H NMR (300 MHz, CDCl3): d=1.07 (s, 9H; -SiC(CH3)3), 1.28 (d, J=
6.4 Hz, 3H; -CHCH3), 3.80–3.88 (m, 2H; Fc), 3.93–3.97 (m, 1H; Fc),
3.98–4.03 (m, 1H; Fc), 4.14–4.19 (m, 2H; Fc), 4.26 (q, J=6.4 Hz, 1H;
-CHCH3), 4.37–4.44 (m, 2H; Fc), 7.35–7.49 (m, 6H; Ph), 7.70–
7.80 ppm (m, 4H; Ph); 13C NMR (75 MHz, CDCl3): d=18.7
(-SiC(CH3)3), 23.7 (-CHCH3), 28.2 (-SiC(CH3)3), 65.5, 66.1, 66.4, 67.5,
68.3, 69.0, 69.0, 71.2, 75.4, 75.7, 94.5, 127.4 (Ph), 129.1 (Ph),135.5
(Ph), 136.4 ppm (Ph): HRMS calcd for C28H32FeOSi: 468.157182 [M]+;
found: 468.15619.
N,N-Dimethyl-1-(2-trimethylsilylferrocenyl)ethan-1-amine [(R,Rp)-12]:
Acetylchloride (0.22 mmol, 5.0 equiv) was added to a solution of
compound (R,Rp)-8a (0.044 mmol, 1.0 equiv) and pyridine
(0.27 mmol, 6.0 equiv) in dry dichloromethane (1 mL) and the mix-
ture was stirred at RT for 15 h. The volatile parts were evaporated
under reduced pressure and the rest was dissolved in CH3CN
(1 mL). An aqueous solution of HNMe2 (40%, 2.2 mmol, 50 equiv)
was added and the mixture was stirred at 388C for three days. The
mixture was diluted with water (5 mL) and extracted with Et2O (3”
5 mL), dried over Na2SO4 and evaporated under vacuum. The prod-
uct was separated by column chromatography (SiO2, hexane/
EtOAc 3:1+1% Et3N). Yield: 92%; [a]2D0 = +46.0 (c=1.32, CHCl3);
+45.0 (c=1.0, EtOH); lit.:[26] +42.2 (c=0.6, EtOH). 1H NMR
(600 MHz, CDCl3): d=0.29 (s, 9H; -Si(CH3)3), 1.18 (d, J=6.9 Hz, 3H;
-CHCH3), 2.36 (s, 6H; -N(CH3)2), 3.34 (q, J=6.9 Hz, 3H; -CH-OH),
4.00–4.03 (m, 1H; Fc), 4.12 (s, 5H; C5H5), 4.26–4.28 (m, 1H; Fc),
4.32–4.34 ppm (m, 1H; Fc); 13C NMR (151 MHz, CDCl3): d=1.0
(-Si(CH3)3), 18.5 (CHCH3), 43.0 (-N(CH3)2), 59.2 (-CH(OH)), 65.9 (Fc),
68.9 (Fc), 69.1 (C5H5), 70.8 (Fc), 73.5 (Fc), 99.8 ppm (Fc); HRMS calcd
for C15H22FeOSi: 285.076192 [MÀN(CH3)2]+; found: 285.07527.
N,N-Dimethyl-1-(2-trimethylsilylferrocenyl)ethan-1-amine [(R,Sp)-12]:
1H NMR (600 MHz, CDCl3): d=0.24 (s, -Si(CH3)3), 1.21 (d, J=6.8 Hz,
3H; -CHCH3), 2.03 (s, 6H; -N(CH3)2), 3.80 (q, J=6.8 Hz, 1H; -CHCH3),
4.04–4.05 (m, 1H; Fc), 4.06 (s, 5H; C5H5), 4.22–4.24 (m, 1H; Fc),
4.26–4.28 ppm (m, 1H; Fc); 13C NMR (151 MHz, CDCl3): d=0.2
(-Si(CH3)3), 9.4 (-CHCH3), 39.6 (-N(CH3)2), 58.0 (-CHCH3), 68.6 (Fc),
68.9 (C5H5), 70.0 (Fc), 72.1 (Fc), 74.6 (Fc), 96.6 ppm (Fc); [a]2D0 = +
13.9 (c=1.46, dichloromethane); +17.3 (c=1.7, EtOH); lit.:[26]
+15.5 (c=1.5, EtOH).
Computational details
The molecules were drawn in Spartan[32] and firstly optimized by
the semi-empirical method PM3.[33] The prepared molecules were
later optimized at the RI-DFT level[34] by using the BP-86 function-
al[35] in the Turbomole package.[36] The def-SVP basis set[37] was
used for pre-optimization and then the def-TZVP basis set[38] was
used for optimization of the structures. Transition states were cal-
culated by using the PES (potential energy scan) by prolonging of
the CArÀSi bond of the product (from 2.1993 to 3.0962 Š) at the RI-
DFT level by using the BP-86 functional and the def-SVP base. The
vibration spectra of all resulted conformers were calculated. The
conformer with the highest imaginary frequency was then used for
searching of the transition states following the imaginary frequen-
cies by using def-TZVP basis set. The resulted transition states
were confirmed by one negative imaginary vibration (CArÀSi vibra-
tion). The energy of the molecules and transition states were calcu-
lated on the RI-SCS-MP2 level[39] (C(OS)=1.20 and C(SS)=0.3333)
by using the def2-TZVP basis set[38] in the Turbomole package.
Acknowledgements
1-(2-tert-Butyldimethylsilylferrocenyl)ethanol (10a): Yield: 12% mix-
ture of diastereoisomers (R,Rp)-10a and (R,Sp)-10a 2:1.
Financial support from the Slovak Grant Agency VEGA, grant
no. VEGA 1/0623/12 is gratefully acknowledged. This publica-
tion is the result of the project implementation: 26240120025
Major isomer (R,Rp)-10a: [a]2D0 =À11.2 (c=1.13, CHCl3). 1H NMR
(600 MHz, CDCl3): d=0.28 (s, 3H; -SiCH3), 0.36 (s, 3H; -SiCH3), 0.83
(s, 9H; -SiC(CH3)3), 1.33 (d, J=6.3 Hz, 3H; -CHCH3), 2.18 (brs, 1H;
-OH), 4.07–4.10 (m, 1H; Fc), 4.20 (s, 5H; C5H5), 4.36–4.38 (m, 1H;
Fc), 4.47 (dq, J=6.3, 1.2 Hz, -CH-OH), 4.49–4.51 ppm (m, 1H; Fc);
13C NMR (151 MHz, CDCl3): d=À3.9 (2”SiCH3), 17.6 (-SiC(CH3)3),
24.6 (CHCH3), 26.8 (-SiC(CH3)3), 64.5 (-CH(OH)), 67.3 (Fc), 68.5 (C5H5),
68.9 (Fc), 70.6 (Fc), 74.7 (Fc), 103.1 ppm (Fc); HRMS calcd for
C18H28FeOSi: 344.125882 [M]+; found: 344.12516.
supported by the Research
Programme funded by the ERDF.
& Development Operational
Keywords: carbanions
·
ferrocene
·
organosilanes
·
rearrangement · silyl transfer
[1] a) A. G. Brook, Acc. Chem. Res. 1974, 7, 77–84; b) P. C. B. Page, S. S. Klair,
S. Rosenthal, Chem. Soc. Rev. 1990, 19, 147–195; c) W. H. Moser, Tetrahe-
dron 2001, 57, 2065–2084.
[2] A. G. Brook, J. Am. Chem. Soc. 1958, 80, 1886–1889.
[3] Y. R. Luo, Comprehensive Handbook of Chemical Bond Energies, CRC
Press, Boca Raton, 2007.
[4] a) A. B. Smith III., W. M. Wuest, Chem. Commun. 2008, 5883–5895;
b) A. B. Smith, C. M. Adams, Acc. Chem. Res. 2004, 37, 365–377; c) A. B.
Smith, R. Tong, W.-S. Kim, W. A. Maio, Angew. Chem. Int. Ed. 2011, 50,
8904–8907; Angew. Chem. 2011, 123, 9066–9069.
[5] a) L. Gao, J. Lu, Z. Song, X. Lin, Y. Xu, Z. Yin, Chem. Commun. 2013, 49,
8961–8963; b) L. Yan, X. Sun, H. Li, Z. Song, Z. Liu, Org. Lett. 2013, 15,
1104–1107; c) X. Sun, J. Lei, C. Sun, Z. Song, L. Yan, Org. Lett. 2012, 14,
1094–1097; d) L. Gao, X. Lin, J. Lei, Z. Song, Z. Lin, Org. Lett. 2012, 14,
158–161.
Minor isomer (R,Sp)-10a: [a]2D0 =À10.5 (c=0.16, CHCl3); 1H NMR
(600 MHz, CDCl3): d=0.29 (s, 3H; -SiCH3), 0.41 (s, 3H; -SiCH3), 0.87
(s, 9H; -SiC(CH3)3), 1.60 (d, J=6.3 Hz, 3H; -CHCH3), 4.05–4.08 (m,
1H; Fc), 4.10 (s, 5H; C5H5), 4.35–4.38 (m, 1H; Fc), 4.46–4.49 (m, 1H;
Fc), 4.71 ppm (dq, J=6.3, 4.7 Hz, 1H; -CH-OH); 13C NMR (75 MHz,
CDCl3): d=À4.1 (-SiCH3), À3.6 (-SiCH3), 17.6 (-SiC(CH3)3), 23.3
(-CHCH3), 27.1 (-SiC(CH3)3), 66.2 (-CHCH3), 68.5 (Fc), 68.6 (Fc), 69.2
(C5H5), 70.7 (Fc), 75.7 (Fc), 96.9 ppm (Fc); HRMS calcd for
C18H28FeOSi: 344.125882 [M]+; found: 344.12506.
1-(1’-tert-Butyldimethylsilylferrocenyl)ethanol (R)-11a: Yield: 4%;
1
[a]2D0 =À10.0 (c=0.71, CHCl3). H NMR (300 MHz, [D6]acetone): d=
0.29 (s, 6H; -Si(CH3)2), 0.80 (s, 9H; -SiC(CH3)3), 1.40 (d, J=6.7 Hz,
3H; -CHCH3), 3.51 (d, J=5.0 Hz, 1H; -OH), 4.05–4.13 (m, 4H; Fc),
4.16–4.19 (m, 1H; Fc), 4.19–4.23 (m, 1H; Fc), 4.35–4.41 (m, 2H; Fc),
4.62 ppm (dq, J=6.7, 5.0 Hz, 1H; -CHCH3); 13C NMR (75 MHz,
[D6]acetone): d=À5.4 (-Si(CH3)2), 17.6 (-SiC(CH3)3), 24.5 (-CHCH3),
26.8 (-SiC(CH3)3), 65.8, 66.5, 67.7, 68.5, 68.6, 69.9, 72.3, 72.4, 74.7,
74.8, 95.4 ppm (Fc); HRMS calcd for C18H28FeOSi: 344.125882 [M]+;
found: 344.12481.
[6] M. Sasaki, Y. Kondo, M. Kawahata, K. Yamaguchi, K. Takeda, Angew.
Chem. Int. Ed. 2011, 50, 6375–6378; Angew. Chem. 2011, 123, 6499–
6502.
[7] B. Liu, C.-D. Lu, J. Org. Chem. 2011, 76, 4205–4209.
[8] Y. Matsuya, A. Koiwai, D. Minato, K. Sugimoto, N. Toyooka, Tetrahedron
Lett. 2012, 53, 5955–5957.
[9] T. Mita, Y. Higuchi, Y. Sato, Org. Lett. 2014, 16, 14–17.
Chem. Eur. J. 2015, 21, 13445 – 13453
www.chemeurj.org
13452
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ˇ
ˇ
´
[28] A. Almµssy, A. Skvorcovµ, B. Horvµth, F. Bilcík, V. Bariak, E. Rakovsky, R.
[10] H. Stueger, B. Hasken, M. Haas, M. Rausch, R. Fischer, A. Torvisco, Orga-
nometallics 2014, 33, 231–239.
ˇ
Sebesta, Eur. J. Org. Chem. 2013, 111 –116.
[11] W. H. Moser, K. E. Endsley, J. T. Colyer, Org. Lett. 2000, 2, 717–719.
[12] A. Tsubouchi, K. Onishi, T. Takeda, J. Am. Chem. Soc. 2006, 128, 14268–
14269.
[13] P. Smirnov, J. Mathew, A. Nijs, E. Katan, M. Karni, C. Bolm, Y. Apeloig, I.
Marek, Angew. Chem. Int. Ed. 2013, 52, 13717–13721; Angew. Chem.
2013, 125, 13962–13966.
[29] M. A. Carroll, A. J. P. White, D. A. Widdowson, D. J. Williams, J. Chem. Soc.
Perkin Trans. 1 2000, 1551–1557.
[30] Y. Onishi, Y. Nishimoto, M. Yasuda, A. Baba, Org. Lett. 2011, 13, 2762–
2765.
[31] O. Riant, G. Argouarch, D. Guillaneux, O. Samuel, H. B. Kagan, J. Org.
Chem. 1998, 63, 3511–3514.
[14] R. West, G. A. Gornowicz, J. Organomet. Chem. 1971, 28, 25–35.
[15] A. N. Thadani, Y. Huang, V. H. Rawal, Org. Lett. 2007, 9, 3873–3876.
[16] M.-K. Chung, O. C. Lightbody, J. M. Stryker, Org. Lett. 2008, 10, 3825–
3828.
[17] Z. Gan, Y. Wu, L. Gao, X. Sun, J. Lei, Z. Song, L. Li, Tetrahedron 2012, 68,
6928–6934.
[18] a) Y. He, H. Hu, X. Xie, X. She, Tetrahedron 2013, 69, 559–563; b) Y. He,
B. Ma, J. Yang, X. Xie, X. She, Tetrahedron 2013, 69, 5545–5549.
[19] D. C. Kapeller, L. Brecker, F. Hammerschmidt, Chem. Eur. J. 2007, 13,
9582–9588.
[20] Y. Mori, Y. Futamura, K. Horisaki, Angew. Chem. Int. Ed. 2008, 47, 1091–
1093; Angew. Chem. 2008, 120, 1107–1109.
[32] Spartan ’08, Wavefunction, Inc., Irvine, CA, 2008.
[33] Y. Shao, L. F. Molnar, Y. Jung, J. Kussmann, C. Ochsenfeld, S. T. Brown,
A. T. B. Gilbert, L. V. Slipchenko, S. V. Levchenko, D. P. ONeill, R. C. Lo-
chan, Jr., T. Wang, G. J. O. Beran, N. A. Besley, J. M. Herbert, C. Y. Lin, T. V.
Voorhis, S. H. Chien, A. Sodt, R. P. Steele, V. A. Rassolov, P. E. Maslen, P. P.
Korambath, R. D. Adamson, B. Austin, J. Baker, E. F. C. Byrd, H. Dachsel,
R. J. Doerksen, A. Dreuw, B. D. Dunietz, A. D. Dutoi, T. R. Furlani, S. R.
Gwaltney, A. Heyden, S. Hirata, C.-P. Hsu, G. Kedziora, R. Z. Khalliulin, P.
Klunzinger, A. M. Lee, M. S. Lee, W. Liang, I. Lotan, N. Nair, B. Peters, E. I.
Proynov, P. A. Pieniazek, Y. M. Rhee, J. Ritchie, E. Rosta, C. D. Sherrill,
A. C. Simmonett, J. E. Subotnik, H. L. Woodcock III, W. Zhang, A. T. Bell,
A. K. Chakraborty, Phys. Chem. Chem. Phys. 2006, 8, 3172–3191.
[34] M. Sierka, A. Hogekamp, R. Ahlrichs, J. Chem. Phys. 2003, 118, 9136–
9148.
[21] E. J. Corey, A. Venkateswarlu, J. Am. Chem. Soc. 1972, 94, 6190–6191.
[22] S. T. Kadam, S. S. Kim, Green Chem. 2010, 12, 94–98.
[23] a) R. Peters, D. F. Fischer, Org. Lett. 2005, 7, 4137–4140; b) S. A. Herbert,
D. C. Castell, J. Clayden, G. E. Arnott, Org. Lett. 2013, 15, 3334–3337;
[35] A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100.
[36] TURBOMOLE V6.4, TURBOMOLE GmbH, Karlsruhe, 2012.
[37] K. Eichkorn, O. Treutler, H. Öhm, M. Häser, R. Ahlrichs, Chem. Phys. Lett.
1995, 242, 652–660.
[38] K. Eichkorn, F. Weigend, O. Treutler, R. Ahlrichs, Theor. Chem. Acc. 1997,
97, 119–124.
[39] a) F. Weigend, M. Häser, Theor. Chem. Acc. 1997, 97, 331–340; b) C.
Hättig, A. Hellweg, A. Kohn, Phys. Chem. Chem. Phys. 2006, 8, 1159–
1169; c) S. Grimme, J. Chem. Phys. 2003, 118, 9095–9102.
ˇ
ˇ
´
ˇ ˇ
c) A. Skvorcovµ, E. Rakovsky, J. Kozísek, R. Sebesta, J. Organomet. Chem.
2011, 696, 2600–2606; d) C. Dietz, V. Jouikov, K. Jurkschat, Organome-
tallics 2013, 32, 5906–5917.
[24] M. Schlosser, Pure Appl. Chem. 1988, 60, 1627–1634.
[25] T. Ireland, K. Tappe, G. Grossheimann, P. Knochel, Chem. Eur. J. 2002, 8,
843–852.
[26] a) D. Marquarding, H. Klusacek, G. Gokel, P. Hoffmann, I. Ugi, J. Am.
Chem. Soc. 1970, 92, 5389–5393; b) D. Broggini, A. Togni, Helv. Chim.
Acta 2002, 85, 2518–2522.
[27] G. W. Gokel, D. Marquarding, I. K. Ugi, J. Org. Chem. 1972, 37, 3052–
3058.
Received: May 1, 2015
Published online on July 27, 2015
Chem. Eur. J. 2015, 21, 13445 – 13453
www.chemeurj.org
13453
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim