Illuminating the mechanism of the borane-catalyzed hydrosilylation of imines with both an axially chiral borane and silane

Article abstract of DOI:10.1002/chem.201202693

The reduction of C=O groups with silanes catalyzed by electron-deficient boranes follows a counterintuitive mechanism in which the Si-H bond is activated by the boron Lewis acid prior to nucleophilic attack of the carbonyl oxygen atom at the silicon atom. The borohydride thus formed is the actual reductant. These steps were elucidated by using a silicon-stereogenic silane, but applying the same technique to the related reduction of C=N groups was inconclusive due to racemization of the silicon atom. The present investigation now proves by the deliberate combination of our axially chiral borane catalyst and axially chiral silane reagents (in both enantiomeric forms) that the mechanisms of these hydrosilylations are essentially identical. Unmistakable stereochemical outcomes for the borane/silane pairs show that both participate in the enantioselectivity-determining hydride-transfer step. These experiments became possible after the discovery that our axially chiral C₆F ₅-substituted borane induces appreciable levels of enantioinduction in the imine hydrosilylation. Copyright

Full text of DOI:10.1002/chem.201202693

FULL PAPER
DOI: 10.1002/chem.201202693
Illuminating the Mechanism of the Borane-Catalyzed Hydrosilylation of
Imines with Both an Axially Chiral Borane and Silane
Marius Mewald and Martin Oestreich*[a]
Abstract: The reduction of C=O
groups with silanes catalyzed by elec-
tron-deficient boranes follows a coun-
terintuitive mechanism in which the
tion of C=N groups was inconclusive
due to racemization of the silicon
atom. The present investigation now
proves by the deliberate combination
of our axially chiral borane catalyst
and axially chiral silane reagents (in
both enantiomeric forms) that the
mechanisms of these hydrosilylations
are essentially identical. Unmistakable
stereochemical outcomes for the
borane/silane pairs show that both par-
ticipate in the enantioselectivity-deter-
mining hydride-transfer step. These ex-
periments became possible after the
discovery that our axially chiral C6F5-
substituted borane induces appreciable
levels of enantioinduction in the imine
hydrosilylation.
ꢀ
Si H bond is activated by the boron
Lewis acid prior to nucleophilic attack
of the carbonyl oxygen atom at the sili-
con atom. The borohydride thus
formed is the actual reductant. These
steps were elucidated by using a sili-
con-stereogenic silane, but applying the
same technique to the related reduc-
Keywords: boron · hydrosilylation ·
Lewis acids · reaction mechanisms ·
silicon
Introduction
Small-molecule activation by main-group elements, the new
chemistry of frustrated Lewis pairs (FLPs) in particular, is
currently garnering tremendous attention.[1,2] The deliberate
combination of a strong (usually boron-based) Lewis acid
and various Lewis bases enables the heterolytic splitting of
[3]
ꢀ
ꢀ
Si H and even H H bonds. This intriguing activation
mode was initially found to be operative in the BACHTUNRGTNEUNG(C6F5)3-cat-
alyzed hydrosilylation of C=O and C=N groups,[4] and a re-
lated imine hydrogenation was later developed.[5] The Piers
group showed by careful analysis that BACHTNURGTNEUNG(C6F5)3 (1) activates
ꢀ
the Si H bond of I rather than the C=X group of III (I!II;
Scheme 1),[6] and our laboratory clarified the mechanism of
the actual activation step (for X=O) with the aid of a sili-
con-stereogenic silane (II!IV°!V; Scheme 1).[7] The cata-
lytic cycle closes with a conventional borohydride reduction
of the intermediate silylcarboxonium ion (V!VI;
Scheme 1), thereby releasing the electron-deficient borane
1.
Scheme 1. Mechanism of the BACHTUNRGTNE(UNG C6F5)3-catalyzed hydrosilylation of ke-
tones (X=O, reported) and imines (X=NR, proposed).
efforts to render these metal-free reductions enantioselec-
tive, the unclear mechanism represents a significant gap.
Just a few months ago, a single example was reported by the
Klankermayer group.[8] The chiral borane used in this work
indeed induces decent enantioselectivity but only in the
presence of a bulky phosphane. No asymmetric induction is
seen with the chiral borane alone. It is assumed that a
boron/phosphorus frustrated Lewis pair (FLP) forms, then
The counterintuitive mechanism of the BACTHUNRGTNEUNG(C6F5)3-catalyzed
ꢀ
Si H bond activation with subsequent carbonyl reduction
(X=O) is therefore sufficiently understood.[6,7] Conversely,
solid evidence for the cognate imine reduction (X=NR) to
follow the same steps is still pending. With regards to recent
ꢀ
cleaving the Si H bond without the involvement of the
imine.[9] The mechanism of the asymmetric reduction step as
well as the role of the phosphane (or silylphosphonium ion)
remains unclear. We disclose here conclusive insight into the
borane-catalyzed imine reduction by using the axially chiral,
electron-deficient borane (S)-2·THF recently introduced by
us (Figure 1, left).[10] Our investigation was triggered by the
observation that (S)-2·THF alone yields significant levels of
enantioselection and displays distinct matched/mismatched
[a] M. Mewald, Prof. Dr. M. Oestreich
Institut fꢀr Chemie, Technische Universitꢁt Berlin
Strasse des 17. Juni 115, 10623 Berlin (Germany)
Fax : (+49)30-314-28829
E-mail: martin.oestreich@tu-berlin.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201202693.
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

To gain more insight into the above experiment, the con-
figurational stability of an asymmetrically substituted silicon
atom attached to an imine nitrogen atom will be crucial (see
V with X=NR in Scheme 1). Moreover, the stereochemical
ꢀ
course of the Si N bond cleavage must be determined to
assign the absolute configuration at the silicon atom. It
turned out that little is known about the chemistry of
amino-substituted silicon-stereogenic silanes.[11,14] We there-
fore had to elaborate a protocol for the stereospecific cleav-
Figure 1. Axially chiral borane (S)-2·THF as well as silanes (S)-3 and (R)-
3 as stereochemical probes.
ꢀ
age of an Si N bond (see the Supporting Information for
ꢀ
selectivities with axially chiral silanes (S)-3 and (R)-3
(Figure 1, right).
details) and found that NH4Cl and MeOH transform an Si
ꢀ
N into an Si O bond with complete inversion of the config-
uration. This setup was applied to the amino-substituted
silane intermediate 7a after BACHTUNRTGNEUNG(C6F5)3-mediated imine hydro-
Results and Discussion
silylation to afford the methoxy-substituted silane in racemic
form [(SiR)-5!rac-8; Scheme 3, top]. As shown by Piers
As mentioned above, we had verified the SN2-Si mechanism
ꢀ
of the Si H bond activation with silicon-stereogenic silane
(SiR)-5 (see IV° with X=O in Scheme 1).[7] Subsequent
ꢀ
cleavage of the Si O bond with retention of the configura-
tion[11] liberated (SiS)-5, which corresponds to a Walden in-
ꢀ
version at the silicon atom in the heterolytic Si H split-
ting.[7] Importantly, the alcohol formed in the reductive Si
ꢀ
O cleavage was enantiomerically enriched (38% ee with
(SiR)-5 (90% ee) for R1 =Ph and R2 =Me, not shown), thus
demonstrating considerable reagent control of the chiral
silane through single-point binding.[7,12] An attempt to prove
the mechanism of the imine reduction by the same strategy
failed. Reduction of a representative imine with (SiR)-5 was
stoichiometric in 1 to afford the amine in racemic form
(4a!rac-6a; Scheme 2).[12]
Scheme 3. Probing the stereochemical course at the silicon atom (upper)
and potential racemization process (lower).
et al., the borohydride reduction becomes rate-determining
in reactions with sterically demanding reactants.[4b] The slow
borohydride reduction might account for dissociation of the
silyliminium intermediate 9a, thus resulting in a transient
planar (i.e., achiral) silylium ion 10 (Scheme 3, bottom).
These results clearly show that, unlike the C=O reduction,[7]
silicon-stereogenic silanes cannot be used as stereochemical
probes in the C=N reduction.[12]
We have recently reported the preparation and detailed
characterization of the axially chiral, electron-deficient
borane (S)-2 that is conveniently isolated as its THF adduct
(S)-2·THF (Figure 1, left).[10] It was designed as a chiral ana-
Scheme 2. B
silane.
ACHTUNGTRENNUNG(C6F5)3-mediated imine reduction with a silicon-stereogenic
This unexpected outcome brought us to consider another
reduction pathway. As one equivalent of the borane Lewis
acid is present, the formation of its imine adduct is not un-
likely.[13] Reduction of the
BACHTNGUTERNNU(G C6F5)3-activated imine
(Figure 2, path a) rather than silyliminium ion (Figure 2,
path b) by the borohydride is, therefore, at least conceivable.
The former path would proceed without any involvement of
the chiral silicon moiety.
logue of BACHTUNRTGNEUNG(C6F5)3 (1), and we were delighted to see that (S)-
ꢀ
2·THF was indeed able to activate Si H bonds but there
was either no asymmetric induction in carbonyl hydrosilyla-
tion[10] or numbers were low.[15] In turn, imine hydrosilyla-
tion catalyzed by (S)-2·THF without the addition of another
Lewis base produced promising enantioselectivities, suffi-
ciently high for mechanistic investigations. We note here
that, compared to Klankermayerꢃs observation,[8] the pres-
ence of bulky phosphanes such as Mes3P had no effect.
Figure 2. Possible hydride-transfer scenarios.
&
2
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Hydrosilylation of Imines
FULL PAPER
We focused on imines with various aryl substituents at the
nitrogen atom in our systematic screening (Table 1). An at-
tempt in toluene, the commonly used solvent for BACHTUNRGTNEUNG(C6F5)3-
mentioning is that asymmetric induction is independent of
conversion and catalyst loading. Hence, thermal E/Z isomer-
ization of imines in the presence of the borane catalyst[13]
seems not to interfere in the reduction of imines 4a–4d cat-
alyzed by (S)-2·THF.
The electronic and steric effects exerted by the aryl nitro-
gen substituent on the reaction rate were further verified in
two competition experiments (Scheme 4).[6] We subjected an
Table 1. Enantioselective imine hydrosilylation catalyzed by (S)-2.[a]
Ar
Solvent[b] t [h] Conv. [%][c] Yield [%][d] ee [%][e,f]
1
2
3
4
5
6
7
8
Ph (4a)
Ph (4a)
Ph (4a)
Ph (4a)
toluene
neat[g]
CH2Cl2
74
45
74
70
80
66
84
55
quant
quant
47
53
62
50
78
52
94
97
n.d.
2
9
33
33
30
36
41
7
1,2-C6H4F2 66
4-anisyl (4b) 1,2-C6H4F2 68
2-tolyl (4c) 1,2-C6H4F2 17
1-Np (4d) 1,2-C6H4F2 18
Scheme 4. Competition experiments between imines 4a and 4b as well as
4a and 4c (ratios based on conversion of imines determined by GLC
analysis with tetracosane as internal standard after >97% conversion of
Me2PhSiH).
1-Np (4d)
toluene
120
[a] 0.1 or 0.2 mmol scale. Np=naphthyl, n.d.=not determined, quant=
quantitative. [b] 1.0 m solutions. [c] Determined by GLC analysis using
tetracosane as internal standard. [d] Isolated yield after flash column
chromatography. [e] Determined by HPLC analysis using chiral station-
ary phases. [f] Absolute configurations of 6c and 6d were assigned by
comparison with 6a and 6b. [g] Me2PhSiH (2.0 equiv) used.
equimolar mixture of imine 4a and electron-rich imine 4b
to borane catalysis. Remarkably, more nucleophilic 4b
reacts faster than 4a, whereas 4b is reduced at a slower rate
than 4a when reacted separately (Table 1, entries 4 and 5).
A similar situation is seen in the competition experiment
with phenyl-substituted 4a and sterically hindered 4c. Less-
hindered 4a reacts faster than 4c in its presence, whereas 4c
is the fast-reacting imine in the separate experiment
(Table 1, entries 4 and 6). These results are interpreted as a
reflection of the different borane–imine adduct strengths
and the ability of the imine to attack the silicon atom from
catalyzed reactions, with phenyl-substituted imine 4a afford-
ed disappointing 2% ee at low reaction rate (Table 1,
entry 1). Minor improvement was seen when the reaction
was run without solvent (Table 1, entry 2). The solvent po-
larity had a strong effect, and enantiomeric excesses in-
creased significantly in polar aprotic solvents such as CH2Cl2
and 1,2-C6H4F2 (Table 1, entries 3 and 4). An enantiomeric
excess of 2% (Table 1, entry 1) versus 33% (Table 1,
entry 4) is an unprecedented solvent effect in borane-cata-
lyzed hydrosilylation reactions. We next turned to electronic
and steric effects of the aryl protecting group.[4b] The reac-
tion rate for 4-anisyl-substituted imine 4b was slower with
almost identical enantiomeric excess (Table 1, entry 5). The
decreased rate is likely to be due to the formation of a
stronger borane–imine adduct with an electron-richer nitro-
gen atom in 4b than in 4a, thereby making less free borane
available for the silane activation.[4b] Conversely, more steric
bulk around the nitrogen donor as in 4c (with 2-tolyl) or 4d
(with 1-naphthyl) substantially increased the reaction rate,
thereby resulting in full conversion in less than one day
(Table 1, entries 6 and 7). Moreover, slightly enhanced enan-
tioinduction is also obtained (36 and 41% ee).[16] Similar ob-
servations in the racemic series were made by Piers et al. for
ꢀ
the backside in the Si H bond-activation step. The more nu-
cleophilic (as for 4a/4b) and less-hindered imines (as for 4a/
4c) win in competitive experiments but lose in the individu-
al setups. These results support a mechanism in which the
ꢀ
Si H bond is activated by the borane followed by the pre-
ferred backside attack of a nucleophilic and unhindered
imine at the silicon atom.
Doubts remained about the stereochemistry-determining
hydride-transfer step (see Figure 2). We exclude path c, as
Lewis acid activation of the imine was disproven. The signif-
icant enantioselectivities obtained in reduction with the
chiral borane (S)-2 (Table 1) make path d, that is, hydride
transfer from another equivalent of the silane, seem more
than unlikely. If path d were the actual hydride-transfer
mode, asymmetric induction would arise from a chiral boro-
hydride counteranion. Polar solvents have been shown to be
detrimental to enantioinduction in asymmetric counteranion
catalysis,[17] and enantioselectivity even increases from non-
polar (toluene) to polar (CH2Cl2 and 1,2-C6H4F2) solvents in
our case. To finally distinguish between path a and path b,
or, in other words, to prove the participation of both the
borane and the silicon reactant in the hydride transfer, we
prepared both enantiomers of the new axially chiral silane
3[18] (Figure 1, right). Combination with (S)-2 ought to form
matched/mismatched pairs for reduction path b, and the out-
BACHTUNGTRENNUNG
(C6F5)3 (1),[4b] and these strongly support the Si H bond-
ꢀ
activation mode and exclude a Lewis acid activation mecha-
nism (see Figure 2, path c). We do have, however, no ex-
planation yet for the pronounced solvent effect on asymmet-
ric induction. The effect on the reaction rate is rationalized
by better stabilization of V (Scheme 1) in more polar sol-
vents. For example, imine 4d also reacts slowly in toluene,
again with little enantioselectivity (7 versus 41% ee; Table 1,
entries 8 and 7, respectively). Another observation worth
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

M. Oestreich and M. Mewald
come would not be blurred by racemization of the chiral
silane.
only on enantioinduction but also reaction rate. The enan-
tiomeric excesses were sufficiently high for mechanistic in-
vestigations, and combinations of the borane (S)-2 and the
new axially chiral silanes 3 [(S)-2/(R)-3 and (S)-2/(S)-3
pairs] displayed distinct matched/mismatched selectivities
(Table 2). By this, we now confirmed the tentative mecha-
nism for this imine reduction, and our data are in accord-
ance with the mechanistic picture originally proposed by the
Piers group and refined by us (Scheme 1). We regard the
present study as a useful basis for the development of im-
proved Lewis acid catalyzed asymmetric hydrosilylation re-
actions.
We performed the catalyses with a slightly increased load-
ing of (S)-2 (5 mol%) and tested the influence of the config-
uration of 3 with 4a and 4d (unhindered and hindered
imines) in 1,2-C6H4F2 and toluene (polar and nonpolar sol-
vents).[19] As observed before (see Table 1), reactions of 4d
were generally faster than those of 4a (Table 2). We were
Table 2. Matched/mismatched scenarios in the imine hydrosilylation with
either enantiomer of axially chiral silane 3 catalyzed by (S)-2.[a]
Experimental Section
Ar
3
Solvent[b]
t [h]
Conv [%][c]
ee [%][d,e]
All reactions were performed in flame-dried glassware using an M.
Braun glovebox (O2 <0.5 ppm, H2O<0.5 ppm) or conventional Schlenk
techniques under a static pressure of nitrogen. Liquids and solutions
were transferred with syringes. Solvents (THF, Et2O, toluene, 1,2-C6H4F2,
n-hexane, and CH2Cl2) were purified and dried following standard proce-
dures. Technical-grade solvents for extraction or chromatography (cyclo-
hexane, tert-butyl methyl ether, Et2O, and ethyl acetate) were distilled
prior to use. CCl4, Cl2, MeOH, nBuLi (in hexanes), TfOH, LiAlH4, imi-
dazole, DMAP, NH4Cl, and Ph(H)SiCl2 were purchased from commercial
suppliers and used without further purification. N,N,N’,N’-Tetramethyl-
1
2
3
4
5
6
7
8
Ph (4a)
Ph (4a)
1-Np (4d)
1-Np (4d)
Ph (4a)
Ph (4a)
1-Np (4d)
1-Np (4d)
R
S
R
S
R
S
R
S
1,2-C6H4F2
1,2-C6H4F2
1,2-C6H4F2
1,2-C6H4F2
toluene
toluene
toluene
toluene
96
96
22
57
60
92
90
26
16
53
29
22 (S)
2 (R)
13 (S)
16 (R)
18 (S)
16 (R)
29 (S)
18 (R)
23
160
160
69
69
[a] 0.05 or 0.1 mmol scale. [b] 0.5 m solutions. [c] Determined by GLC
analysis using tetracosane as internal standard. [d] Determined by HPLC
analysis using chiral stationary phases. [e] Major enantiomer of 6a and
6d, respectively, in parentheses.
AHCTUNGTREGeNNUN nediamine (TMEDA) and dibenzylamine were dried over sodium and
distilled prior to use. MePh2SiH and Ph2SiH2 were dried over LiAlH4 and
distilled prior to use. Imines 4a–4d were prepared according to known
procedures and spectroscopic data agreed with reported data. Borane
(S)-2·THF was prepared according to our previously reported proce-
dure.[10]
BACHTUNGTERNNU(G C6F5)3 was prepared according to the procedure of Erker
delighted to find unmistakable stereochemical outcomes for
both (S)-2/(R)-3 and (S)-2/(S)-3 pairs with all imine/solvent
combinations. We consider those cases matched when the
same absolute configuration is seen as with achiral
Me2PhSiH. The matched/mismatched cases are also clear
from the reaction rates with toluene but not 1,2-C6H4F2 as
solvent. This observation might be again interpreted by a
better stabilization of ion pair V (Scheme 1) in polar sol-
vents, thus making the matched/mismatched effect more
pronounced in a nonpolar solvent with V being less stabi-
lized. Gratifyingly, (R)-3 invariably produced (S)-6 (match-
ed), whereas (S)-3 always afforded (R)-6 (mismatched).
Hence, axially chiral 3 determines the absolute configura-
tion of the resulting amine. This considerable reagent con-
et al.[20] (SiR)-1-Isopropyl-1,2,3,4-tetrahydro-1-silanaphthalene [(SiR)-5]
was prepared according to our previously reported procedure.[21] Analyti-
cal thin-layer chromatography (TLC) was performed on silica gel 60
F254 glass plates by Merck. Flash column chromatography was per-
formed on silica gel 60 (40–63 mm, 230–400 mesh, ASTM) by Merck
1
using the indicated solvents. H, 13C, and 29Si NMR spectra were recorded
in CDCl3 using Bruker AV400 and Bruker AV500 instruments. Chemical
shifts are reported in parts per million (ppm) and are referenced to the
residual solvent resonance as the internal standard (CDCl3: d=7.26 and
77.2 ppm). Data are reported as follows: chemical shift, multiplicity (s=
singlet, d=doublet, t=triplet, q=quartet, sept=septet, mc =centrosym-
metric multiplet, m=multiplet, br=broad), coupling constants (Hz), and
integration. Gas liquid chromatography (GLC) was performed using an
Agilent 7820A gas chromatograph equipped with an HP-5 capillary
column (30 mꢄ0.32 mm, 0.25 mm film thickness) by Agilent Technologies
using the following program: N2 carrier gas, injection temperature 2508C,
detector temperature 3008C; temperature program: start temperature
408C, heating rate 108Cminꢀ1, end temperature 2808C for 10 or 30 min.
Enantiomeric excesses were determined by analytical high-pressure
liquid chromatography (HPLC) analysis using an Agilent 1200 Series in-
strument with a chiral stationary phase using a Daicel Chiralcel OJ-RH
trol is absolutely remarkable because BACTHUNRGTNEUNG(C6F5)3-catalyzed
imine reductions that employed chiral 3 exclusively yielded
racemic amines. Without ambiguity, these data prove that
reduction path b is the predominant hydride-transfer mecha-
nism (Figure 2, left).
column (MeCN/H2O mixtures as solvent),
a Daicel Chiralcel OJ-H
column (n-heptane/iPrOH mixtures as solvent), or a Daicel Chiralcel
OD-H column (n-heptane/iPrOH mixtures as solvent). Optical rotations
were measured using a Schmidt and Haensch Polartronic H 532 polari-
Conclusion
AHCTUNGERTGmNNUN eter. Mass spectrometry (MS) was obtained from the Analytical Facility
at the Institut fꢀr Chemie, Technische Universitꢁt Berlin.
Preparation of (S)-4,5-dihydro-3H-dinaphtho[2,1-c:1’,2’-e]silepine ((S)-3):
To recap, we demonstrated the potential of our borane (S)-2
as a chiral analogue of BACHTUNRGTNE(UNG C6F5)3 (1) to induce 33% ee for a
phenyl-substituted imine (Table 1, entry 4) and even 62% ee
for a benzyl-substituted imine.[16] These values were ach-
ieved due to the discovery of a strong solvent effect not
A
solution of n-butyllithium (1.5m in hexanes, 8.19 mL, 12.3 mmol,
2.50 equiv) was concentrated under vacuum to an oil and diluted in Et2O
(10 mL). solution of (S)-2,2’-dimethyl-1,1’-binaphthalene (1.39 g,
A
4.92 mmol, 1.00 equiv) in Et2O (10 mL) was added to this dropwise at
08C followed by TMEDA (1.87 mL, 12.6 mmol, 2.55 equiv). The mixture
&
4
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Hydrosilylation of Imines
FULL PAPER
was maintained at room temperature for 20 h to result in a deep red sus-
pension. The liquid was removed by filtration, and the residue was dried
under full vacuum to furnish a red/brown solid (1.25 g, 2.37 mmol). The
solid was dissolved in THF (15 mL) and cooled to ꢀ788C. Ph(H)SiCl2
(0.52 mL, 3.56 mmol, 1.50 equiv based on metalated intermediate) was
added in one portion, and a color change from deep red to yellow was
observed. The solution was slowly warmed to room temperature and
quenched by the addition of water. The aqueous layer was extracted with
tert-butyl methyl ether, and the combined organic layers were washed
with brine. The organic layer was dried over MgSO4, and the solvents
were removed under reduced pressure. The residue was purified by flash
column chromatography on silica gel (cyclohexane) to yield (S)-4-phenyl-
4,5-dihydro-3H-dinaphtho[2,1-c:1’,2’-e]silepine as a white solid. Yield:
660 mg (1.71 mmol, 35% overall yield; 72% based on metalated inter-
mediate). M.p. 151–1528C (cyclohexane); Rf =0.14 (cyclohexane); GLC
(HP-5): tR =44.1 min; [a]2D0 =+71.0 (c=0.58, CHCl3); 1H NMR
(500 MHz, CDCl3): d=2.25 (d, AB spin system, 2J=13.6 Hz, 1H), 2.28
(dd, AB spin system, 2J=13.3 Hz, 3J=7.1 Hz, 1H), 2.31 (dd, AB spin
system, 2J=13.6 Hz, 3J=2.3 Hz, 1H), 2.36 (dd, AB spin system, 2J=
13.3 Hz, 3J=2.1 Hz, 1H), 4.69 (ddd, 3J=7.1 Hz, 3J=2.3 Hz, 3J=2.1 Hz,
silane), the vial was taken out of the glovebox, and the reaction mixture
was diluted with tert-butyl methyl ether (1 mL). The solution was filtered
over a small pad of silica gel, and all solvents of the filtrate were re-
moved under reduced pressure. GLC-MS analysis of the crude product
showed a peak that corresponded to the aminosilane intermediate 7a.
Crude 7a was then dissolved in THF (1 mL), and MeOH (20.5 mg,
0.640 mmol, 3.81 equiv) and NH4Cl (10.5 mg, 0.196 mmol, 1.17 equiv)
were added. After 40 h, GLC analysis indicated full conversion of amino-
silane 7a. The reaction mixture was filtered over a short sand/cotton pad
to remove the solids, and the solvents were removed under reduced pres-
sure. Flash column chromatography on silica gel (cyclohexane/ethyl ace-
tate 99:1) afforded amine rac-6a (15 mg, 76 mmol, 45%) and silyl ether
rac-8 (5.9 mg, 27 mmol, 16%) both in racemic form. For the analytical
data of 6a, see below. For the analytical data of 8, see the Supporting In-
formation.
General procedure for borane-catalyzed imine hydrosilylation: In
a
AHCTUNGTREGgNNUN lovebox, a 2 mL vial equipped with a magnetic stir bar was charged
with (S)-2·THF (3–5 mol%). A solution of the indicated imine 4 (0.050–
0.20 mmol, 1.0 equiv), the silane (1.0 equiv), and tetracosane (internal
standard for GLC analysis, 0.5–2.0 mg) in the indicated solvent (0.5–1m)
was added to the borane in one portion, and the vial was closed. The
mixture was stirred for the indicated time (conversion monitored by
GLC analysis) and subsequently subjected to flash column chromatogra-
phy on silica gel (cyclohexane/ethyl acetate mixtures) to afford amines 6
as solids or oils.
3
3
1H), 7.13 (dm, J=8.5 Hz, 1H), 7.17 (dm, J=8.7 Hz, 1H), 7.18–7.24 (m,
3
2H), 7.30–7.35 (m, 3H), 7.37–7.43 (m, 5H), 7.55 (d, J=8.4 Hz, 1H), 7.84
(d, 3J=8.2 Hz, 1H), 7.89–7.94 ppm (m, 3H); 13C NMR (126 MHz,
CDCl3): d=20.37, 20.42, 124.6, 126.0, 126.1, 126.48, 126.53, 127.7, 128.1
(2C), 128.19, 128.21, 128.3 (2C), 130.1, 132.1, 132.2, 132.6, 132.7, 132.8,
132.9, 134.0, 134.7, 135.6, 136.0 ppm; 29Si NMR (99 MHz, CDCl3): d=
ꢀ2.0 ppm; IR (ATR): n˜ =3056 (m), 2130 (s), 1618 (m), 1592 (m), 1507
(s), 1427 (m), 1404 (m), 1355 (m), 1327 (m), 1242 (m), 1141 (s), 1112 (s),
1024 (m), 964 (s), 925 (m), 836 (s), 815 (s), 737 (s), 693 cmꢀ1 (s); HRMS
(EI): m/z: calcd for C28H22Si [M+]: 386.14853; found: 386.14918. Accord-
ing to the protodesilylation reported by Hayashi and Shintani et al.,[22]
N-(1-Phenylethyl)aniline (6a): Prepared according to the general proce-
dure. Analytical data: isolated as colorless oil. Rf =0.70 (cyclohexane/
ethyl acetate 4:1); GLC (HP-5): tR =17.4 min; HPLC (Daicel Chiralcel
OD-H, 208C, n-heptane/iPrOH 90:10, flow rate 0.7 mLminꢀ1
, l=
230 nm): tR =8.8 min ((S)-6a), 10.0 min ((R)-6a); 1H NMR (400 MHz,
CDCl3): d=1.53 (d, 3J=6.7 Hz, 3H), 4.13 (brs, 1H), 4.50 (q, 3J=6.7 Hz,
1H), 6.53 (dd, 3J=8.6 Hz, 4J=1.0 Hz, 2H), 6.66 (tt, 3J=7.3 Hz, 4J=
1.0 Hz, 1H), 7.10 (mc, 2H), 7.20–7.27 (m, 1H), 7.29–7.42 ppm (m, 4H);
13C NMR (101 MHz, CDCl3): d=25.2, 53.6, 113.4, 117.4, 126.0, 127.0,
128.8, 129.2, 145.3, 147.4 ppm; IR (ATR): n˜ =3410 (s), 3053 (m), 3024
(m), 2967 (m), 2924 (m), 2868 (m), 1601 (s), 1505 (s), 1449 (m), 1428 (w),
1373 (w), 1353 (w), 1318 (m), 1258 (m), 1206 (w), 1180 (w), 1140 (w), 750
(s), 701 cmꢀ1 (s); HRMS (ESI): m/z: calcd for C14H15NH [M+H]+:
198.1277; found: 198.1277.
TfOH (0.17 mL, 1.9 mmol, 3.5 equiv) was added to
a solution
of (S)-4-phenyl-4,5-dihydro-3H-dinaphtho[2,1-c:1’,2’-e]silepine (210 mg,
0.543 mmol, 1.00 equiv) in CH2Cl2 (15 mL), and the mixture was stirred
at room temperature. After 1 h, all volatiles were removed under full
vacuum, and the residue was dissolved in CH2Cl2 (5 mL) and again sub-
jected to full vacuum for 30 min. The foamy residue was dissolved in
Et2O (8 mL), and LiAlH4 (102 mg, 2.69 mmol, 4.95 equiv) was carefully
added in three portions. The suspension was stirred at room temperature
for 2 h and was then carefully quenched with water. After extraction of
the aqueous layer with tert-butyl methyl ether, the combined organic
layers were dried over MgSO4, and the solvents were removed under re-
duced pressure. Purification by flash column chromatography on silica
gel (cyclohexane) followed by recrystallization from cyclohexane at 48C
yielded the title compound (S)-3 as colorless crystals. Yield: 101 mg
(0.33 mmol, 60%). M.p. 182–1838C (cyclohexane); Rf =0.29 (cyclohex-
ane); GLC (HP-5): tR =27.7 min; [a]2D0 =+413.7 (c=0.38, CHCl3);
1H NMR (400 MHz, CDCl3): d=2.08 (dt, AB spin system, 2J=13.5 Hz,
3J=5.0 Hz, 2H), 2.15 (dt, AB spin system, 2J=12.8 Hz, 3J=1.1 Hz, 2H),
4.09 (tt, 3J=4.5 Hz, 3J=1.0 Hz, 2H), 7.09 (dm, 3J=8.5 Hz, 2H), 7.18
(ddd, 3J=8.2 Hz, 3J=6.7 Hz, 4J=1.3 Hz, 2H), 7.38 (ddd, 3J=8.1 Hz, 3J=
6.7 Hz, 4J=1.3 Hz, 2H), 7.46 (d, 3J=8.4 Hz, 2H), 7.87 (d, 3J=8.1 Hz,
2H), 7.89 ppm (dm, 3J=8.2 Hz, 2H); 13C NMR (126 MHz, CDCl3): d=
16.9, 124.6, 125.9, 126.3, 127.3, 128.1, 128.3, 132.0, 132.5, 132.7,
136.0 ppm; 29Si NMR (99 MHz, CDCl3): d=ꢀ22.6 ppm; IR (ATR): n˜ =
3040 (w), 2130 (s), 1616 (w), 1590(m), 1506 (m), 1411 (m), 1352 (m), 1328
(m), 1241 (m), 1221 (w), 1135 (m), 1055 (w), 971 (m), 920 (s), 846 (m),
823 cmꢀ1 (s); HRMS (EI): m/z: calcd for C22H18Si [M+]: 310.11723;
found: 310.11648.
4-Methoxy-N-(1-phenylethyl)aniline (6b): Prepared according to the gen-
eral procedure. Analytical data: isolated as yellow oil. Rf =0.20 (cyclo-
hexane/ethyl acetate 20:1); GLC (HP-5): tR =20.1 min; HPLC (Daicel
Chiralcel OD-H, 208C, n-heptane/iPrOH 98:2, flow rate 0.7 mLminꢀ1
,
l=230 nm): tR =25.1 min ((R)-6b), 28.3 min ((S)-6b); 1H NMR
(400 MHz, CDCl3): d=1.51 (d, 3J=6.7 Hz, 3H), 3.70 (s, 3H), 4.20 (brs,
3
1H), 4.42 (q, J=6.7 Hz, 1H), 6.46–6.51 (m, 2H), 6.68–6.73 (m, 2H), 7.23
(tt, 3J=7.1 Hz, 4J=1.5 Hz, 1H), 7.30–7.40 ppm (m, 4H); 13C NMR
(101 MHz, CDCl3): d=25.3, 54.4, 55.9, 114.7, 114.9, 126.0, 127.0, 128.7,
141.7, 145.6, 152 ppm; IR (ATR): n˜ =3402 (s), 2963 (m), 2832 (m), 1617
(w), 1508 (s), 1449 (m), 1406 (w), 1372 (w), 1294 (w), 1231 (s), 1177 (m),
1139 (m), 1035 (s), 816 (s), 752 (s), 699 cmꢀ1 (s); HRMS (ESI): m/z:
calcd for C15H17NOH [M+H]+: 228.1380; found: 228.1383.
2-Methyl-N-(1-phenylethyl)aniline (6c): Prepared according to the gener-
al procedure. Analytical data: isolated as colorless oil. Rf =0.48 (cyclo-
hexane/ethyl acetate 20:1); GLC (HP-5): tR =18.1 min; HPLC (Daicel
Chiralcel OJ-H, 208C, n-heptane/iPrOH 97:3, flow rate 0.5 mLminꢀ1, l=
230 nm): tR =21.5 min ((S)-6c), 26.1 min ((R)-6c); 1H NMR (400 MHz,
3
CDCl3): d=1.58 (d, J=6.7 Hz, 3H), 2.25 (s, 3H), 3.87 (brs, 1H), 4.56 (q,
3J=6.7 Hz, 1H), 6.39 (dd, 3J=8.1 Hz, 4J=1.1 Hz, 1H), 6.62 (ddd, 3J=
7.4 Hz, 3J=7.4 Hz, 4J=1.2 Hz, 1H), 6.98 (ddd, 3J=8.0 Hz, 3J=7.4 Hz,
4J=1.6 Hz, 1H), 7.07 (ddd, 3J=7.3 Hz, 4J=1.8 Hz, 5J=0.9 Hz, 1H), 7.24
(tt, 3J=7.1 Hz, 4J=1.5 Hz, 1H), 7.31–7.41 ppm (m, 4H); 13C NMR
(101 MHz, CDCl3): d=17.8, 25.4, 53.4, 111.2, 117.0, 121.7, 125.9, 127.0,
127.1, 128.8, 130.1, 145.3, 145.4 ppm; IR (ATR): n˜ =3433 (m), 3023 (w),
2965 (m), 2923 (w), 2866 (w), 1604 (s), 1508 (s), 1446 (s), 1372 (m), 1313
(s), 1259 (s), 1212 (m), 1144 (m), 1050 (w), 1017 (w), 984 (w), 920 (w),
838 (w), 743 (s), 698 cmꢀ1 (s); HRMS (ESI): m/z: calcd for C15H17NH
[M+H]+: 212.1434; found: 212.1432.
The enantiomer of the title compound (R)-3 was prepared from (R)-2,2’-
dimethyl-1,1’-binaphthalene and analytical data matched those of (S)-3.
[a]2D0 ((R)-3)=ꢀ417.9 (c=0.28, CHCl3).
BACHTUNGTRENNUNG(C6F5)3-catalyzed imine hydrosilylation with silicon-stereogenic silane
Si
ꢀ
( R)-5 followed by stereospecific Si N bond cleavage: In a glovebox, a
2 mL vial equipped with a magnetic stir bar was charged with B(C6F5)3
solution of imine 4a (32.8 mg,
AHCTUNGTRENNUNG
(86.0 mg, 0.168 mmol, 1.00 equiv).
A
0.168 mmol, 1.00 equiv) and silane (SiR)-5 (>99% ee, 32.0 mg,
0.168 mmol, 1.00 equiv) in toluene (1.5 mL) was added to the borane in
one portion and the vial was closed. After 46 h (full conversion of
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

M. Oestreich and M. Mewald
N-(1-Phenylethyl)naphthalen-1-amine (6d): Prepared according to the
general procedure. Analytical data: isolated as white solid. M.p. 95–968C
(cyclohexane); Rf =0.47 (cyclohexane/ethyl acetate 20:1); GLC (HP-5):
tR =23.3 min; HPLC (Daicel Chiralcel OD-H, 208C, n-heptane/iPrOH
98:2, flow rate 0.7 mLminꢀ1, l=230 nm): tR =22.9 min ((S)-6d), 36.8 min
((R)-6d); 1H NMR (400 MHz, CDCl3): d=1.69 (d, 3J=6.7 Hz, 3H), 4.71
(q, 3J=6.7 Hz, 1H), 4.76 (brs, 1H), 6.41 (mc, 1H), 7.19–7.23 (m, 2H),
7.26 (tt, 3J=7.3 Hz, 4J=1.3 Hz, 1H), 7.32–7.37 (m, 2H), 7.43–7.52 (m,
4H), 7.79–7.84 (m, 1H), 7.93–7.98 ppm (m, 1H); 13C NMR (101 MHz,
CDCl3): d=25.4, 53.7, 106.1, 117.4, 119.9, 123.4, 124.8, 125.8, 125.9, 126.7,
127.1, 128.8, 128.9, 134.4, 142.2, 145.0 ppm; IR (ATR): n˜ =3434 (m), 2973
(m), 2866 (w), 1575(s), 1523 (s), 1476 (s), 1446 (m), 1408 (s), 1370 (m),
1345 (m), 1279 (s), 1220 (m), 1139 (s), 1102 (w), 1026 (w), 765 (s),
703 cmꢀ1 (s); HRMS (ESI): m/z: calcd for C18H17NH [M+H]+: 248.1434;
found: 248.1431.
[6] D. J. Parks, J. M. Blackwell, W. E. Piers, J. Org. Chem. 2000, 65,
3090–3098.
[7] S. Rendler, M. Oestreich, Angew. Chem. 2008, 120, 6086–6089;
Angew. Chem. Int. Ed. 2008, 47, 5997–6000.
[8] a) D. Chen, V. Leich, F. Pan, J. Klankermayer, Chem. Eur. J. 2012,
18, 5184–5187; for related reports on enantioselective hydrogena-
tion of imines with FLPs, see: b) D. Chen, Y. Wang, J. Klankermay-
er, Angew. Chem. 2010, 122, 9665–9668; Angew. Chem. Int. Ed.
2010, 49, 9475–9478; c) G. Ghattas, D. Chen, F. Pan, J. Klankermay-
er, Dalton Trans. 2012, 41, 9026–9028; for a related report on dia-
stereoselective hydrogenation of imines with FLPs, see: Z. M.
Heiden, D. W. Stephan, Chem. Commun. 2011, 47, 5729–5731.
ꢀ
[9] For the reversible heterolytic Si H bond activation in non-hindered
silanes with a tethered boron/phosphorus FLP, see: W. Nie, H. F. T.
Klare, M. Oestreich, R. Frçhlich, G. Kehr, G. Erker, Z. Natur-
forsch., DOI: 10.5560/ZNB.2012-0181.
N-Benzyl-1-phenylethanamine:[16] Prepared according to the general pro-
cedure. Analytical data: isolated as colorless oil. Rf =0.25 (cyclohexane/
ethyl acetate 4:1); GLC (HP-5): tR =17.9 min; HPLC (Daicel Chiralcel
[10] M. Mewald, R. Frçhlich, M. Oestreich, Chem. Eur. J. 2011, 17,
9406–9414.
OD-H, 208C, n-heptane/iPrOH 99:1, flow rate 0.75 mLminꢀ1
, l=
[11] a) A. Weickgenannt, M. Oestreich in Asymmetric Synthesis: More
Methods and Applications (Eds.: M. Christmann, S. Brꢁse), Wiley-
VCH, Weinheim, 2012, pp. 35–42; b) M. Oestreich, Synlett 2007,
1629–1643; c) R. J. P. Corriu, C. Guerin, J. J. E. Moreau in Topics in
Stereochemistry, Vol. 15 (Ed.: E. L. Eliel), Wiley, New York, 1984,
pp. 355–374; d) L. H. Sommer, Stereochemistry, Mechanism, and Sil-
icon, McGraw-Hill, New York, 1965; e) L. H. Sommer, Intra-Sci.
Chem. Rep. 1973, 7, 1–44.
230 nm): tR =8.6 min (R), 9.4 min (S); 1H NMR (400 MHz, CDCl3): d=
1.38 (d, 3J=6.6 Hz, 3H), 1.93 (brs, 1H), 3.60 (d, 2J=13.2 Hz, 1H), 3.67
(d, 2J=13.2 Hz, 1H), 3.82 (q, 3J=6.6 Hz, 1H), 7.22–7.42 ppm (m, 10H);
13C NMR (101 MHz, CDCl3): d=24.4, 51.5, 57.4, 126.7, 126.9, 126.9,
128.2, 128.3, 128.5, 140.4, 145.3 ppm; IR (ATR): n˜ =3062 (w), 3026 (m),
2963 (m), 2925 (w), 1493 (m), 1452 (s), 1369 (w), 1201 (w), 1124 (s),
700 cmꢀ1 (s); HRMS (ESI): m/z: calcd for C15H17NNa [M+Na]+:
212.1434; found: 212.1440.
[12] D. T. Hog, M. Oestreich, Eur. J. Org. Chem. 2009, 5047–5056.
[13] J. M. Blackwell, W. E. Piers, M. Parvez, R. McDonald, Organometal-
lics 2002, 21, 1400–1407.
[14] We are only aware of reports by Terunuma et al. on the stereospecif-
ꢀ
ic Si N bond hydrolysis under acidic conditions: a) D. Terunuma, K.
Acknowledgements
Murakami, M. Kokubo, K. Senda, H. Nohira, Bull. Chem. Soc. Jpn.
1980, 53, 789–794; b) D. Terunuma, K. Senda, M. Sanazawa, H.
Nohira, Bull. Chem. Soc. Jpn. 1982, 55, 924–927.
This research was supported by the Deutsche Forschungsgemeinschaft
(Oe 249/6-1). M.O. is indebted to the Einstein Foundation (Berlin) for an
endowed professorship. We thank Professor Warren Piers (University of
Calgary) for suggesting to us that racemization of the silicon-stereogenic
silane might explain our previous preliminary observations.
[15] M. Mewald, M. Oestreich, unpublished material, 2012.
[16] We would like to mention here that acetophenone-derived imine
with a benzyl protection group is reduced by (S)-2·THF with prom-
ising 62% ee. Optimization of the enantiomeric excess is not the
purpose of the present investigation but ongoing efforts in our labo-
ratory are directed toward the modification of the borane catalyst.
[17] For a counteranion-directed asymmetric imine hydrogenation in-
cluding matched/mismatched combinations with a chiral iridium cat-
alyst, see: a) C. Li, C. Wang, B. Villa-Marcos, J. Xiao, J. Am. Chem.
Soc. 2008, 130, 14450–14451; for the seminal contribution in coun-
teranion-directed catalysis, see: b) S. Mayer, B. List, Angew. Chem.
2006, 118, 4299–4301; Angew. Chem. Int. Ed. 2006, 45, 4193–4195.
[18] For details of its preparation, see the Experimental Section. A relat-
ed silepine structure was reported before: M. E. Jung, K. T. Hogan,
Tetrahedron Lett. 1988, 29, 6199–6202.
[1] a) G. C. Welch, R. R. San Juan, J. D. Masuda, D. W. Stephan, Science
2006, 314, 1124–1126; b) G. C. Welch, D. W. Stephan, J. Am. Chem.
Soc. 2007, 129, 1880–1881.
[2] For authoritative reviews, see: a) D. W. Stephan, Org. Biomol.
Chem. 2012, 10, 5740–5746; b) D. W. Stephan, G. Erker, Angew.
Chem. 2010, 122, 50–81; Angew. Chem. Int. Ed. 2010, 49, 46–76;
c) D. W. Stephan, Dalton Trans. 2009, 3129–3136; d) D. W. Stephan,
Org. Biomol. Chem. 2008, 6, 1535–1539; e) A. L. Kenward, W. E.
Piers, Angew. Chem. 2008, 120, 38–42; Angew. Chem. Int. Ed. 2008,
47, 38–41.
[3] For an excellent review that bridges the areas of Si H and H H ac-
tivation, see: W. E. Piers, A. J. V. Marwitz, L. G. Mercier, Inorg.
Chem. 2011, 50, 12252–12262.
[4] a) D. J. Parks, W. E. Piers, J. Am. Chem. Soc. 1996, 118, 9440–9441;
b) J. M. Blackwell, E. R. Sonmor, T. Scoccitti, W. E. Piers, Org. Lett.
2000, 2, 3921–3923.
[19] Dihydrosilane 3 is assumed to react as a monohydride, thereby
avoiding more complicated selectivity effects due to another hydride
transfer. Experiments with Ph2SiH2 showed that one equivalent is
completely consumed under the reaction conditions of Table 1
(26% ee with imine 4a in 1,2-C6H4F2).
ꢀ
ꢀ
[20] C. Wang, G. Erker, G. Kehr, K. Wedeking, R. Frçhlich, Organome-
tallics 2005, 24, 4760–4773.
[5] a) P. A. Chase, T. Jurca, D. W. Stephan, Chem. Commun. 2008,
1701–1703; b) D. Chen, J. Klankermayer, Chem. Commun. 2008,
2130–2131; c) P. Spies, S. Schwendemann, S. Lange, G. Kehr, R.
Frçhlich, G. Erker, Angew. Chem. 2008, 120, 7654–7657; Angew.
Chem. Int. Ed. 2008, 47, 7543–7546; d) V. Sumerin, F. Schulz, M.
Atsumi, C. Wang, M. Nieger, M. Leskelꢁ, T. Repo, P. Pyykkç, B.
Rieger, J. Am. Chem. Soc. 2008, 130, 14117–14119; e) K. Cherni-
chenko, M. Nieger, M. Leskelꢁ, T. Repo, Dalton Trans. 2012, 41,
9029–9032.
[21] a) M. Oestreich, U. K. Schmid, G. Auer, M. Keller, Synthesis 2003,
2725–2739; b) S. Rendler, M. Oestreich, C. P. Butts, G. C. Lloyd-
Jones, J. Am. Chem. Soc. 2007, 129, 502–503.
[22] R. Shintani, K. Moriya, T. Hayashi, J. Am. Chem. Soc. 2011, 133,
16440–16443.
Received: July 27, 2012
Published online: && &&, 0000
&
6
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Hydrosilylation of Imines
FULL PAPER
Chiral detectives: Borane-catalyzed
imine hydrosilylation is investigated
with the aid of an axially chiral borane
(catalyst) and silane (reactant). Their
uses in separate experiments as well as
(mis)matched combinations offer con-
clusive evidence of a mechanism iden-
Reaction Mechanisms
M. Mewald, M. Oestreich* . &&&& — &&&&
Illuminating the Mechanism of the
Borane-Catalyzed Hydrosilylation of
Imines with Both an Axially Chiral
Borane and Silane
tical to that of related BACHTNUTRGNE(UNG C6F5)3-cata-
lyzed carbonyl hydrosilylation. Both
the borane and silane participate in
the selectivity-determining hydride-
transfer step (see figure).
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!