Anionic access to silylated and germylated binuclear heterocycles

Article abstract of DOI:10.1002/chem.201402597

A simple access to silylated and germylated binuclear heterocycles, based on an original anionic rearrangement, is described. A set of electron-rich and electron-poor silylated aromatic and heteroaromatic substrates were tested to understand the mechanism and the factors controlling this rearrangement, in particular its regioselectivity. This parameter was shown to follow the rules proposed before from a few examples. Then, the effect of the substituents borne by the silicon itself, in particular the selectivity of the ligand transfer, was studied. Additionally, this chemistry was extended to germylated substrates. A hypervalent germanium species, comparable to the putative intermediate proposed with silicon, seems to be involved. However, a pathway implicating the elimination of LiCH₂Cl was observed for the first time with this element, leading to unexpected products of the benzo-oxa (or benzo-aza) germol-type.

Full text of DOI:10.1002/chem.201402597

DOI: 10.1002/chem.201402597
Full Paper
&
Heterocycles
Anionic Access to Silylated and Germylated Binuclear
Heterocycles
Thomas Boddaert,[a, c] Cyril FranÅois,[a] Laetitia Mistico,[a] Olivier Querolle,[b]
Lieven Meerpoel,[d] Patrick Angibaud,[b] Muriel Durandetti,*[a] and Jacques Maddaluno*[a]
Abstract: A simple access to silylated and germylated binu-
clear heterocycles, based on an original anionic rearrange-
ment, is described. A set of electron-rich and electron-poor
silylated aromatic and heteroaromatic substrates were
tested to understand the mechanism and the factors con-
trolling this rearrangement, in particular its regioselectivity.
This parameter was shown to follow the rules proposed
before from a few examples. Then, the effect of the substitu-
ents borne by the silicon itself, in particular the selectivity of
the ligand transfer, was studied. Additionally, this chemistry
was extended to germylated substrates. A hypervalent ger-
manium species, comparable to the putative intermediate
proposed with silicon, seems to be involved. However,
a pathway implicating the elimination of LiCH2Cl was ob-
served for the first time with this element, leading to unex-
pected products of the benzo-oxa (or benzo-aza) germol-
type.
Introduction
ty and the rate of metabolism. Also, the lipophilicity of silylated
analogues is superior to that of the all-carbon molecule and
generally favors the penetration of the drug within the cell
membrane in vivo. These phenomena explain that the silicon
atom is generally regarded as a bioisostere of the carbon
atom.[2–3]
Swapping the carbon and silicon atoms is known to alter sig-
nificantly the physicochemical properties of organic com-
pounds and, when it comes to molecules of biological interest,
of the corresponding absorption, distribution, metabolism, ex-
cretion, and toxicity (ADME-Tox).[1] These differences are partly
related to the size of the silicon atom, which is larger than the
carbon atom (with a covalent radius of 1.17 vs. 0.77 ꢀ for
carbon) and makes longer covalent bonds with other elements
(the CꢀSi bond is 1.87 vs. 1.54 ꢀ for an average CꢀC bond).
This characteristic triggers significant differences in the molec-
ular geometry and thus modifies the efficiency and/or selectivi-
All these reasons explain that the systematic exploration of
the metalloid effect in drugs is the object of an increasing in-
terest in the recent years.[4] Furthermore, this isosterism was
also applied to other fields than in medicinal chemistry. For ex-
ample in odorants design, replacing a carbon atom by a silicon
is also a new strategy to create unique signatures in perfum-
ery.[5] More recently, the carbon atom substitution was pushed
further and the next crystallogen element, germanium, attract-
ed attention. The C/Ge swap that goes with a further increase
of the covalent radius, lipophilicity, and electropositivity was
rapidly applied to bioactive compounds[6–7] as well as perfume
ingredients or in the design of fungicides.[8]
[a] Dr. T. Boddaert, Dr. C. FranÅois, L. Mistico, Dr. M. Durandetti,
Dr. J. Maddaluno
Laboratoire COBRA, CNRS UMR 6014 & FR 3038
Universitꢀ de Rouen and INSA de Rouen, 76821
Mont St Aignan Cedex (France)
Fax: (+33)235-522-971
E-mail: muriel.durandetti@univ-rouen.fr
jmaddalu@crihan.fr
The group of Tacke has been one of the main contributors
in the field and this laboratory published among the most re-
markable examples (Figure 1). For instance, a C/Si substitution
[b] Dr. O. Querolle, Dr. P. Angibaud
Janssen Research & Development, a division of Janssen-Cilag
Chaussꢀe du Vexin, BP615, 27106 Val de Reuil (France)
[c] Dr. T. Boddaert
Current address:
Organic Chemistry P3A Group
ICMMO, CNRS UMR 8182, Universitꢀ Paris-Sud
15 rue Georges Clemenceau, 91405 Orsay cedex (France)
[d] Dr. L. Meerpoel
Janssen Research & Development
a Division of Janssen Pharmaceutica N.V.
Turnhoutseweg 30, 2340 Beerse (Belgium)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201402597.
Figure 1. A few examples of C/Si or C/Ge switch in drugs or fragrance.
Chem. Eur. J. 2014, 20, 10131 – 10139
10131
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
in venlafaxin, an antidepressant marketed for several years (Ef-
fexor, Efexor, and Trevilor), affords (R)-sila-venlafaxin A, which
exhibits antiemetic properties.[9] The case of the sila- and
germa-analogous derivatives of the muscarinic antagonist cy-
crimines B and C, respectively, which display a higher affinity
cellent, except in the case of the 2-aminopyridine derivatives
12 and 13 for which purification problems were encountered
(entries 12 and 13, Table 1). The a/b regioselectivity follows the
trends underlined in our original paper, that is, the electron-de-
ficient aromatic moieties favor the a isomers whereas electron-
rich substituents lead to the b isomers. The pyridine deriva-
tives follow this rule particularly well: in these cases, the single
isomers a are obtained (entries 8–13, Table 1). In contrast, the
than the corresponding alcohols, is also worth noting.[10]
A
spectacular example is given with the silylated analogue of
haloperidol: A drug prescribed since the 1950s for the treat-
ment of neuropsychiatric disorders such as schizophrenia, that
is, sila-haloperidol D. This compound is not only as powerful as
its carbon analogue (with an inhibition constant of 0.85 nm for
the hD2 receptors against 4.0 nm in the case of haloperidol),
but has also a very different metabolic fate in vivo.[11] Note that
silicon-containing odorants like sila-rhubafuran E and disila-gal-
axolide F have been synthetized very recently in the same lab-
oratory, introducing enthusing possibilities of development in
the field.[12]
o-bromothiophenol derivative
5
leads only to isomer
b (entry 5, Table 1), even with derivative 6 featuring a fluorine
atom in position 4 (entry 6, Table 1).
On the other hand, a strong ortho-substituent effect is ob-
served in the case of the 2-bromo-3-methyl-aniline derivative
4, which provides exclusively the a isomer (entry 4, Table 1),
whereas the 2-bromo-5-methyl analogue led to a 37:63 ratio
between the a and b isomers.[14] Note that the construction of
a [3.4.0] silylated bicyclic skeleton is also possible by this route,
as demonstrated in the case of thiophene 7 (entry 7, Table 1).
The striking selectivity observed with the pyridine deriva-
tives led us to check that our mechanistic hypothesis, that is,
the formation of a pentacoordinated silicon intermediate
(Scheme 1), still applied in these cases. We thus prepared the
azasylolpyridine 27 by adding 2 equiv of n-BuLi on protected
2-bromo-3-amino-pyridine and quenching the resulting dia-
nion with Me2(CH2Cl)SiCl. Compound 27 features a silylated
five-membered ring which, after reaction with the lithium car-
benoid generated from chloroiodomethane, provided the dihy-
dropyrido-azasiline 21 in 56% yield, most probably through
the same siliconate intermediate (Scheme 2).
Undoubtedly, the relatively limited number of methodologi-
cal tools available for the synthesis of heterocycles incorporat-
ing elements such as Si or Ge, still hampers these researches.[13]
We have recently described a new rearrangement (termed sila-
Matteson) that gives access to two regioisomeric silylated het-
erocycles of the dihydrobenzooxa- (or aza-) silines family. The
electronic character of the fused aromatic ring was shown to
govern the ratio between the regioisomers and a mechanism
involving
a
pentaorganosilicate species was evidenced
(Scheme 1).[14] This preliminary study was conducted on di-
methyl-substituted silicon tethers borne by several phenol and
aniline derivatives. We describe in this paper the generalization
of this chemistry to a series of new silylated substrates, as well
as its extension toward the germylated ones.
To expand again the scope of this transformation, we decid-
ed to vary the silicon substituents and check if we could fine-
tune the selective transfer of this rearrangement. In addition,
the phenyl appendage on the silicon constitutes a supplemen-
tary asset since the SiꢀPh link can be easily replaced by a varie-
ty of other elements. We thus prepared a series of substrates
including phenol, aniline, and 4-aminopyridine derivatives
bearing a dimethyl-, methyl/phenyl-, or diphenylchloromethyl-
silyl tether (Table 2). The precursors 28–35 were easily ob-
tained from the corresponding aromatics and bis(chlorome-
thyl)dimethylsilane, bis(chloromethyl)methylphenylsilane, or
bis-(chloromethyl)diphenylsilane, which are either commercial-
ly available or were synthesized from bis(chloromethyl)-
(methyl)chlorosilane and dichlorodiphenylsilane in almost
quantitative yields.[16] Next, the halogen–lithium exchange fol-
lowed by the rearrangement was performed in THF at ꢀ408C
for 1 h using n-BuLi.
Scheme 1. Sila-Matteson rearrangement through a hypervalent-silicon
species.
In these standard conditions, the isomers a and/or b were
recovered in good yields for the SiMe2 and SiMePh derivatives
(Table 2). Interestingly, substituting a methyl by a phenyl group
favored the formation of the b isomer in the case of phenyl
ethers 28 and 29 and afforded exclusively the b isomer in the
case of the aniline derivative 32 (compare entries 1/2 and 4/5,
Table 2). These results suggest that, in addition to the electron-
ic control exerted by the aromatic nucleus, steric effects im-
posed by the other silicon substituents influence the fate of
the pentavalent intermediate.[17] When it comes to the 4-ami-
nopyridine derivatives 8 and 34, this effect is not observed
Results and Discussion
We first decided to check the robustness of our methodology
toward various aromatic derivatives. In our original communi-
cation, we only varied the substituents R (Scheme 1) borne by
the phenyl nucleus. To extend the scope of this transformation,
we tried to apply it to new substrates, in particular heterocyclic
ones (Table 1). The halogen–lithium exchange and subsequent
nucleophilic substitution was performed at ꢀ408C using n-bu-
tyllithium in tetrahydrofuran.[15] The results of this methodolog-
ical widening are satisfactory since yields remain good-to-ex-
Chem. Eur. J. 2014, 20, 10131 – 10139
www.chemeurj.org
10132
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
anymore, probably because the
prime electron-deficient charac-
ter of the pyridine moiety drives
the rearrangement toward the
a isomer, as noted before with
Table 1. Extension of the sila-Matteson rearrangement to various substrates.
Entry
Starting material
Yield
[%][a]
Product(s)
a/b[b]
other
electron-withdrawing
14a/14b
50:50
1
1
2
94[c]
phenyl substituents.[14] Note that
the results in entries 7 and 8
(Table 2) demonstrate that the
sila-Matteson rearrangement ap-
plies equally well to pyridinic
substrates, extending substan-
tially its interest in heterocyclic
chemistry.
15a/15b
10:90
2
3
4
70[c]
In the case of the SiPh2 deriva-
tives 30, 33, and 35, we ob-
served for the first time the for-
mation of a five-membered sily-
16a/16b
3
4
85
75:25
lated
heterocycle
bearing
a phenyl and a benzyl group on
the silicon atom. This isomer re-
sults probably from the migra-
tion of a phenyl from the penta-
valent species toward the chlor-
omethyl appendage, further
supporting the hypothesis of
the formation of this hyperva-
lent intermediate (Scheme 3).
Isomer c is the major product in
the case of the aniline 33 and
pyridine 35. It is known that the
migratory aptitude of a phenyl
group is superior to that of
a methyl;[18] nevertheless, this re-
arrangement was not observed
in the case of the SiMePh sub-
strates 29, 32, and 34. Interpret-
ing these results requires cau-
tion since the pentavalent inter-
mediate is generally believed to
17a/17b
87[c]
–
100:0d
18a/18b
0:100
5
6
5
6
99
–
–
19a/19b
0:100
87[c]
20a/20b
45:55
7
8
9
7
8
9
67
21a/21b
100:0
78[c]
–
–
adopt
a
trigonal-bipyramidal
22a/22b
100:0
43
conformation subjected to the
Berry pseudorotation.[19] It is
known, at least on comparable
trigonal-bipyramidal
tetraoxy-
phosphorans,[20] or pentaorgano-
silicates[17] that the phenyl
groups tend to occupy the
apical position (apicophilicity).
We can thus suggest that 30,
33, and 35 evolve through a pen-
tavalent species similar to that
pictured on Scheme 3 in which
the two phenyls are axial and
the chloromethyl in the equato-
rial plane.[21] This situation seems
ideal for the Ph!Cl substitution
23a/23b
10
11
12
10
11
12
99[c]
99[c]
18[c,e]
–
–
100:0
24a/24b
100:0
25a/25b
–
100:0
Chem. Eur. J. 2014, 20, 10131 – 10139
www.chemeurj.org
10133
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
to take place, at least if the
pseudorotation is the kinetically
determining step.[22]
Table 1. (Continued)
Entry
Starting material
Yield
[%][a]
Product(s)
a/b[b]
A more drastic extension of
this methodology consisted in
considering the germylated ana-
logues. The starting materials
were prepared by treating the
26a/26b
–
13
13
39[c,e]
100:0
corresponding
(thio)phenols,
[a] Total yield for the two isomers after purification. [b] a/b determined in the crude reaction mixture by
1H NMR spectroscopy. [c] tBuLi was employed instead of n-BuLi (see Ref. [15]). [d] A small amount (7%) of
isomer d (see below and in the Supporting Information) was obtained in this case. [e] The reaction was
quenched with acetic acid to avoid a sila-ring fission into silanol.
anilines and pyridines with bis(-
chloromethyl)dimethylgermane.
They were then subjected to the
above conditions of cyclization
(Table 3).
The results show that the same chemistry applies to the ger-
mylated substrates and follow similar patterns, leading to the
formation of binuclear heterocycles incorporating a Ge nu-
cleus. In most cases, the two [4:4:0] bicyclic a and b isomers
are obtained, suggesting that the reaction goes through a pen-
tacoordinated germanium entity probably comparable to the
putative intermediate proposed in the case of silicon. Note
that in entries 1, 3, 4, and 10 (Table 3) a new family of deriva-
tives d is now observed. Their structure, based on a [4:3:0] bi-
cyclic skeleton, is of the benzo-oxa (or benzo-aza) germol-type.
These products can result from the release of a LiCH2Cl carbe-
noid of the same pentavalent intermediate (Scheme 4). This re-
action, which affords compounds 54d, 56d, 57d, was not ob-
served with silicon. The difference between the radii of Si
(111 pm) and Ge (125 pm) can help to rationalize this observa-
tion: The elimination rate of LiCH2Cl would increase for the
largest element with respect to that of the intramolecular sub-
stitution. Interestingly, compound 53 leads exclusively to 63d,
suggesting that the ortho-methyl group is also likely to slow
down the chlorine substitution. This effect, which was
glimpsed in the silylated series (see Table 1, entry 4, footnot-
e [d]) is synergetic to the Ge influence and can explain that
only the d product was obtained in this case.
In fine accord with the results obtained for the silicon deriv-
atives 5 and 6, germanyl thioether 45 provides exclusively the
b isomer, as with all electron-rich aromatic substrates. Indeed,
the rearrangement of 49 led exclusively to isomer a (entry 6,
Table 3), whereas the silicon analogue allowed the formation
of a 86:14 ratio between the a and b isomers. In contrast,
a 22:78 mixture of a and b was obtained when reacting 48
(entry 5, Table 3), whereas the cyclization of the silicon ana-
logue provided the sole isomer b. It is tempting to correlate
these results to the relative electronegativity of silicon (1.90)
and germanium (2.01). Indeed, a germylated appendage de-
creases the electron density on the aromatic ring more than its
silylated analogue, favoring the exclusive heterolytic cleavage
of the ArꢀGe bond of the pentavalent species and thus the se-
lective formation of isomer a (entry 6, Table 3). Similarly, the
electron-deficient pyridine rings trigger a rearrangement in
favor of the sole a isomer, regardless of the nature of the sub-
stituent (Si/Ge), the position of the nitrogen or the heteroele-
ment in the fused saturated heterocycle (entries 7 to 9,
Table 3).
Scheme 2. Ring expansion supporting a mechanism involving a hypervalent
siliconate.
Table 2. Effect of the silicon substituents on the selectivity of the cycliza-
tion.
Entry
Starting
material
Z[c]
Y
R
R’
Product
(a/b/c)[a]
Yield
[%][b]
36a/36b/36c
60:40:0
37a/37b/37c
24:76:0
38a/38b/38c
20:50:30
39a/39b/39c
45:55:0
40a/40b/40c
0:100:0
41a/41b/41c
0:35:65
21a/21b/21c
100:0:0
42a/42b/42c
100:0:0
1
2
3
4
5
6
7
8
9
28
29
30
31
32
33
8
O
CH
CH
CH
CH
CH
CH
N
Me
Ph
Ph
Me
Ph
Ph
Me
Ph
Ph
Me
Me
Ph
85
70
74
84
80
78
63
77
51
O
O
N-Boc
N-Boc
N-Boc
N-Boc
N-Boc
N-Boc
Me
Me
Ph
Me
Me
Ph
34
35
N
43a/43b/43c
25:0:75
N
[a] Ratios a/b/c were determined in the crude reaction mixture by
1H NMR spectroscopy. [b] Total yield for the three isomers after purifica-
tion. [c] Boc=tert-butoxycarbonyl.
Scheme 3. Putative mechanism for the formation of isomer c.
Chem. Eur. J. 2014, 20, 10131 – 10139
www.chemeurj.org
10134
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Conclusion
Table 3. Extension of the rearrangement to germylated substrates using n- or tBuLi.
This manuscript reports an
access to a variety of silylated
and germylated binuclear het-
erocycles, based on an original
anionic rearrangement. We show
that the direct correlation be-
tween the electron density on
the aromatic ring and the pro-
portion of the a and b isomers
we had noticed in a recent note
applies to a large number of
substrates, including heterocyclic
ones. This latter family of sub-
strates affords a series of hetero-
aromatic and silylated binuclear
scaffolds of significant interest in
medicinal chemistry but also po-
tentially important for the dis-
covery of new analogues of clas-
Entry Starting
material
Yield
[%][a]
Product(s)
a/b/d[b]
54a/54b/
54d
44 82
1
2
25:6:69
55a/55b/
55d
45 99
46 57
–
–
–
0:100:0
56a/56b/
56d
3
4
5
6
7
0:54:46
sical fragrances. In
a second
57a/57b/
57d
47 89[c]
48 62
49 55
50 75
–
phase of the study, we demon-
strate that replacing the two
methyl borne by the silicon
tether with one or two phenyl
leads to different mixtures of the
known a and b isomers, and to
a never observed isomer c when
a SiPh2 link is employed. It is
very likely that isomer c results
from the migration of one of the
two phenyl nuclei. The forma-
tion of this new family of com-
pounds buttresses the hypothe-
sis of a common pentavalent sili-
con precursor for all the isomers,
the chemoselectivity of the rear-
rangement obviously depending
on the nature of the substituents
of the silicon. Pleasingly enough,
this chemistry extends efficiently
to the germylated analogues.
The rearrangement follows a pat-
tern similar to that of silicon,
suggesting that a comparable
anionic hypervalent germanium
species is involved. Moreover,
benzo-oxa and benzo-azager-
mol-type products (product d)
were also recovered in several
cases, suggesting that an addi-
tional pathway implicating the
elimination of LiCH2Cl occurs
with Ge, a route marginally de-
tected in the case of Si. This Siꢀ
Ge swap effect is reinforced by
0:83:17
58a/58b/
58d
–
–
–
15:85:0
59a/59b/
59d
–
–
100:0:0
60a/60b/
60d
100:0:0
61a/61b/
61d
51 93[c]
52 84[c]
53 92[c]
–
–
–
–
–
8
100:0:0
62a/62b/
62d
9
100:0:0
63a/63b/
63d
–
10
0:0:100
[a] Total yield for the three isomers after purification. [b] Ratios a/b/d were determined in the crude reaction
1
mixture by H NMR spectroscopy. [c] tBuLi was employed instead of n-BuLi (see Ref. [15]).
Chem. Eur. J. 2014, 20, 10131 – 10139
www.chemeurj.org
10135
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
chromatography on silica gel using the appropriate mixture of elu-
ents.
1
Compound 14a: H NMR (300 MHz, CDCl3): d=6.63 (brs, 1H), 6.56
(s, 1H), 5.90 (s, 2H), 3.00 (brs, 2H), 1.82 (s, 2H), 1.43 (s, 9H),
0.10 ppm (s, 6H); 13C NMR (75 MHz, CDCl3): d=155.0, 145.3, 144.9,
134.2, 127.4, 109.0, 108.2, 100.9, 79.9, 35.1, 28.4 (3C), 19.0,
ꢀ2.8 ppm (2C); MS (CI, isobutane): m/z (%): 322 (20) [M+H+], 266,
(100) [M+H+ꢀtBu], 222 (20) [M+H+ꢀBoc].
Scheme 4. Putative mechanism for the formation of product d.
1
Compound 14b: H NMR (300 MHz, CDCl3): d=6.84 (brs, 1H), 6.78
an ortho substituent, a phenomenon that could be employed
to control the chemoselectivity using an easily removable
ortho substituent.
(s, 1H), 5.92 (s, 2H), 3.76–3.72 (m, 2H), 1.47 (s, 9H), 1.13–1.09 (m,
2H), 0.22 ppm (s, 6H); 13C NMR (75 MHz, CDCl3): d=154.2, 148.1,
144.9, 143.3, 123.0, 111.0, 108.0, 100.9, 80.2, 45.4, 28.4 (3C), 13.6,
ꢀ1.1 ppm (2C).
tert-Butyl 7-(dibenzylamino)-3,3-dimethyl-3,4-dihydrobenzo[e]-
[1,3]azasiline-1(2H)-carboxylate (15a): 1H NMR (300 MHz, CDCl3):
d=7.37–7.19 (m, 11H), 6.89 (d, J=2.2 Hz, 1H), 6.52 (dd, J=8.1,
2.2 Hz, 1H), 4.60 (s, 4H), 3.03 (brs, 2H), 1.82 (s, 2H), 1.35 (s, 9H),
0.12 ppm (s, 6H); 13C NMR (50 MHz, CDCl3): d=154.9, 147.4, 141.6,
138.7 (2C), 130.3, 128.6 (4C), 126.8 (2C), 126.7 (4C), 122.1, 111.9,
110.8, 79.5, 54.2 (2C), 35.5, 28.4 (3C), 17.4, ꢀ2.6 ppm (2C); IR (neat):
n˜ =2934, 1688, 1597, 1148, 727 cmꢀ1; HRMS (ESI+): m/z calcd for
C29H37N2O2Si: 473.2624; found: 473.2623.
In conclusion, this paper opens an access to new binuclear
heterocycles incorporating a silicon or germanium atom in one
of the rings. A stimulating next step will consist in including
these scaffolds into more elaborated bioactive molecules and
measure their impact on the reactivity, stability, and ADME pro-
files. Thus, an analogue of Motesanib (Figure 2), a drug-candi-
date acting as antagonist of a vascular endothelial growth
factor receptors (VEGFRs), platelet-derived growth factor
(PDGFR), and dedicated to the treatment of solid tumors, is
currently targeted. This work is in progress and the results will
be reported in due course.
tert-Butyl 7-(dibenzylamino)-4,4-dimethyl-3,4-dihydrobenzo[b]-
[1,4]azasiline-1(2H)-carboxylate (15b): M.p.: 1248C; 1H NMR
(300 MHz, CDCl3): d=7.36–7.19 (m, 11H), 6.77 (d, J=1.9 Hz, 1H),
6.59 (dd, J=8.4, 1.9 Hz, 1H), 4.65 (s, 4H), 3.79–3.75 (m, 2H), 1.41 (s,
9H), 1.13–1.09 (m, 2H), 0.22 ppm (s, 6H); 13C NMR (75 MHz, CDCl3):
d=154.0, 150.0, 149.9, 138.3 (2C), 134.3, 128.6 (4C), 126.8 (2C),
126.5 (4C), 116.8, 109.9, 109.5, 79.8, 53.7 (2C), 45.2, 28.3 (3C), 13.7,
ꢀ0.9 ppm (2C); IR (neat) : n˜ =2928, 1691, 1378, 1139, 700 cmꢀ1
;
HRMS (ESI+): m/z calcd for C29H37N2O2Si: 473.2624; found:
473.2617.
6-(1,3-Dioxolan-2-yl)-8-methoxy-3,3-dimethyl-3,4-dihydro-2H-
1
benzo[e][1,3]oxasiline (16a): H NMR (300 MHz, CDCl3): d=6.84 (d,
J=1.8 Hz, 1H), 6.82 (d, J=1.8 Hz, 1H), 5.70 (s, 1H), 4.17–4.13 (m,
2H), 4.05–4.00 (m, 2H), 3.86 (s, 3H), 3.85 (s, 2H), 2.00 (s, 2H),
0.16 ppm (s, 6H).
Figure 2. Structure of Motesanib and of a silylated/germylated analogue
target.
8-Methoxy-3,3-dimethyl-3,4-dihydro-2H-benzo[e][1,3]oxasiline-6-
carbaldehyde: (after acetal deprotection): 1H NMR (300 MHz,
CDCl3): d=9.80 (s, 1H), 7.22 (d, J=1.8 Hz, 1H), 7.21 (d, J=1.8 Hz,
1H), 3.95 (s, 2H), 3.89 (s, 3H), 2.08 (s, 2H), 0.19 ppm (s, 6H);
13C NMR (75 MHz, CDCl3): d=191.4, 152.6, 151.3, 130.3, 129.5,
127.4, 107.5, 63.7, 56.2, 16.1, ꢀ3.7 ppm (2C); MS (EI, 70 eV): m/z (%)
: 236 (100) [M+], 221, (64) [M+ꢀMe), 205 (36) [MꢀOMe].
6-(1,3-Dioxolan-2-yl)-8-methoxy-4,4-dimethyl-3,4-dihydro-2H-
benzo[b][1,4]oxasiline (16b): 1H NMR (300 MHz, CDCl3): d=7.02
(d, J=1.8 Hz, 1H), 6.98 (d, J=1.8 Hz, 1H), 5.73 (s, 1H), 4.42–4.38
(m, 2H), 4.17–4.13 (m, 2H), 4.05–4.00 (m, 2H), 3.87 (s, 3H), 1.16–
1.12 (m, 2H), 0.28 ppm (s, 6H).
Experimental Section
General procedures
GC analysis was carried out using a 24 m HP-methyl silicon capilla-
ry column. Mass spectra were recorded with a quadrupolar MS in-
strument coupled to a gas chromatograph. Elemental analyses
were performed in house on standard equipment. Column chro-
matographies were run on standard silica gel (230–400 mesh).
1H NMR spectra were recorded in CDCl3 or C6D6 at 300 MHz, and
13C NMR spectra were recorded at 75 MHz; chemical shift (d) are
given in parts per million (ppm) and the coupling constants (J) in
hertz. IR spectra were recorded by transmission on an IRFT spec-
trometer.
8-Methoxy-4,4-dimethyl-3,4-dihydro-2H-ben-zo[b][1,4]oxasiline-
6-carbaldehyde: (after acetal deprotection): 1H NMR (300 MHz,
CDCl3): d=9.84 (s, 1H), 7.45 (d, J=1.8 Hz, 1H), 7.35 (d, J=1.8 Hz,
1H), 4.46–4.50 (m, 2H), 3.91 (s, 3H), 1.20–1.16 (m, 2H), 0.33 ppm (s,
6H); 13C NMR (75 MHz, CDCl3): d=191.3, 158.9, 149.4, 131.9, 130.2,
121.8, 109.8, 67.6, 56.0, 12.9, ꢀ1.7 ppm (2C); MS (EI, 70 eV): m/z
(%): 236 (12) [M+], 208 (100) [M+ꢀCO].
General procedure for the anionic rearrangement: A solution of
n-BuLi in hexane (1.6m, 1.2 equiv) or at ꢀ788C a solution of tBuLi
in hexane (1.7m, 2.4 equiv) was added dropwise at ꢀ408C into
a solution of halogenoaryl precursor (1.0 equiv) in anhydrous THF
(10 mLmmolꢀ1). After 1 h, the reaction mixture was quenched at
ꢀ408C or at ꢀ788C with a saturated aqueous solution of ammoni-
um chloride, allowed to warm up to room temperature and ex-
tracted with diethyl ether. The resulting organic phase was then
dried over anhydrous sodium sulfate and concentrated under re-
duced pressure. The crude mixture was finally purified by flash
tert-Butyl
3,3,5-trimethyl-3,4-dihydrobenzo[e][1,3]azasiline-
1(2H)-carboxylate (17a): 1H NMR (300 MHz, CDCl3): d=7.05–6.99
(m, 3H), 3.08 (brs, 2H), 2.31 (s, 3H), 1.93 (brs, 2H), 1.47 (s, 9H),
0.13 ppm (s, 6H); 13C NMR (75 MHz, CDCl3): d=154.9, 140.8, 138.8,
132.6, 127.0, 124.9, 124.1, 79.4, 34.5, 28.2 (3C), 20.4, 14.3, ꢀ3.0 ppm
Chem. Eur. J. 2014, 20, 10131 – 10139
www.chemeurj.org
10136
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(2C); n˜ =2952, 2925, 2865, 1672, 1143, 843 cmꢀ1; HRMS (ESI+): m/z
calcd for C16H26NO2Si: 292.1733; found: 292.1735.
13C NMR (75 MHz, CDCl3): d=155.8, 141.4, 136.1, 132.2, 125.3, 62.1,
15.6, ꢀ1.5 ppm (2C); HRMS (ESI+): m/z calcd for C9H14NOSi:
179.0766; found: 179.0742.
4,4-Dimethyl-3,4-dihydro-2H-benzo[b][1,4]thiasiline
(18b):
1H NMR (300 MHz, CDCl3): d=7.43–7.40 (m, 1H), 7.25–7.22 (m, 1H),
7.18 (tap.d, J=7.8 Hz, 1.5 Hz, 1H), 7.11 (tap.d, J=7.2 Hz, 1.5 Hz, 1H),
3.08–3.03 (m, 2H), 1.34–1.29 (m, 2H), 0.28 ppm (s, 6H); 13C NMR
(75 MHz, CDCl3): d=145.1, 135.1, 135.0, 129.1, 128.1, 124.8, 28.6,
15.4, ꢀ1.0 ppm (2C); IR (neat): n˜ =2953; 2922, 2886, 1418, 12498,
743 cmꢀ1; MS (EI, 70 eV): m/z (%): 194 (60) [M+], 166 (100), 151
(84).
tert-Butyl 3,3,6-trimethyl-3,4-dihydropyrido[3,2-e][1,3]azasiline-
1(2H)-carboxylate (25a): M.p.: 988C; 1H NMR (300 MHz, CDCl3): d=
8.09 (d, J=1.8 Hz, 1H), 7.24 (d, J=1.8 Hz, 1H), 3.09 (s, 2H), 2.27 (s,
3H), 1.89 (s, 2H), 1.44 (s, 9H), 0.11 ppm (s, 6H); 13C NMR (75 MHz,
CDCl3): d=155.1, 151.5, 146.0, 139.3, 131.2, 128.7, 80.2, 34.9, 28.5
(3C), 18.7, 18.0, ꢀ2.6 ppm (2C); IR (neat): n˜ =2959, 1693, 1361,
1158, 837 cmꢀ1; HRMS (ESI+): m/z calcd for C15H25N2O2Si: 293.1685;
found: 293.1675.
6-Fluoro-4,4-dimethyl-3,4-dihydro-2H-benzo[b][1,4]thiasiline
(19b): 1H NMR (300 MHz, CDCl3): d=7.21 (dd, J=5.1, 8.6 Hz, 1H),
7.12 (dd, J=2.9, 8.5 Hz, 1H), 6.89 (ddd, J=8.6, 8.5, 2.9 Hz, 1H),
3.05–3.00 (m, 2H), 1.34–1.29 (m, 2H), 0.29 ppm (s, 6H); 13C NMR
(75 MHz, CDCl3): d=160.7 (d, J=246 Hz), 139.7 (d, J=3 Hz), 137.9
(d, J=3 Hz), 129.8 (d, J=7 Hz), 120.8 (d, J=19 Hz), 116.2 (d, J=
22 Hz), 28.9, 15.0, ꢀ1.3 ppm (2C); IR (neat): n˜ =2952, 1452, 1243,
7-Chloro-3,3-dimethyl-1,2,3,4-tetrahydropyrido[3,2-e]-
[1,3]azasiline (26a): M.p.: 1288C 1H NMR (300 MHz, CDCl3): d=
7.09 (d, J=7.3 Hz, 1H), 6.50 (d, J=7.3 Hz, 1H), 4.84 (brs, 1H), 2.72
(d, J=3.0 Hz, 2H), 1.90 (s, 2H), 0.14 ppm (s, 6H); 13C NMR (75 MHz,
CDCl3): d=158.3, 146.1, 142.5, 114.6, 112.5, 31.7, 16.5, ꢀ3.6 ppm
(2C); IR (neat): n˜ =3284, 2957, 2895, 1589, 1247 cmꢀ1
.
1201, 813, 784 cmꢀ1
tert-Butyl
.
3-Methyl-3-phenyl-3,4-dihydro-2H-benzo-1,3-oxasiline
(37a):
1H NMR (300 MHz, CDCl3): d=7.51–7.48 (m, 2H), 7.41–7.33 (m, 3H),
7.14–7.10 (m, 2H), 6.98–6.94 (m, 2H), 4.03 (d, J=14.7 Hz, 1H), 3.94
(d, J=14.7 Hz, 1H), 2.33 (d, J=15.6 Hz, 1H), 2.19 (d, J=15.6 Hz,
1H), 0.47 ppm (s, 3H); 13C NMR (75 MHz, CDCl3): d=158.4, 135.6,
134.0 (2C), 132.1, 130.0, 128.2 (2C), 127.7, 126.9, 123.0, 119.8, 63.6,
15.6, ꢀ5.0 ppm ; MS (CI, isobutane): m/z (%): 241 (58) [M+H+],
163 (100) [MꢀPh].
3,3-dimethyl-3,4-dihydrothieno[3,2-e][1,3]azasiline-
1
1(2H)-carboxylate (20a): H NMR (300 MHz, CDCl3): d=6.81 (d, J=
5.7 Hz, 1H), 6.59 (d, J=5.7 Hz, 1H), 3.10 (s, 2H), 1.87 (s, 2H), 1.53
(s, 9H), 0.13 ppm (s, 6H); 13C NMR (75 MHz, CDCl3): d=154.3, 138.3,
128.0, 119.0, 117.9, 81.5, 36.5, 28.4 (3C), 13.1, ꢀ3.2 ppm (2C); IR
(neat): n˜ =2974, 2887, 1692, 1331, 1154, 824, 698, 636 cmꢀ1
.
tert-Butyl
1,1-dimethyl-2,3-dihydrothieno[2,3-b][1,4]azasiline-
1
4-Methyl-4-phenyl-3,4-dihydro-2H-benzo-1,4-oxasiline
(37b):
4(1H)-carboxylate 20b. H NMR (300 MHz, CDCl3): d=6.92 (d, J=
5.7 Hz, 1H), 6.84 (d, J=5.7 Hz, 1H), 4.03–3.99 (m, 2H), 1.56 (s, 9H),
1.01–0.97 (m, 2H), 0.24 ppm (s, 6H); 13C NMR (75 MHz, CDCl3) d=
152.8, 151.3, 127.5, 121.1, 118.8, 82.0, 45.5, 28.3 (3C), 12.2,
ꢀ1.5 ppm (2C); IR (neat): n˜ =2974, 1695, 1365, 1287, 1149, 828,
1H NMR (300 MHz, CDCl3): d=7.56–7.53 (m, 2H), 7.40–7.27 (m, 5H),
6.98–6.89 (m, 2H), 4.45–4.30 (m, 2H), 1.38–1.31 (m, 2H), 0.59 ppm
(s, 3H); 13C NMR (75 MHz, CDCl3): d=165.3, 136.9, 135.6, 134.5 (2C),
131.3, 129.7, 128.1 (2C), 121.2, 119.0, 117.8, 66.7, 13.7, ꢀ3.0 ppm; IR
(neat): n˜ =3056, 2956, 2861, 1429, 1210, 759, 696, 422 cmꢀ1
3-Methyl-3-phenyl-3,4-dihydro-2H-benzo-1,3-oxasiline
.
694 cmꢀ1
tert-Butyl
.
(38a):
3,3-dimethyl-3,4-dihydropyrido[3,4-e][1,3]azasiline-
1H NMR (300 MHz, CDCl3): d=7.53–7.50 (m, 4H), 7.43–7.33 (m, 6H),
7.18–7.09 (m, 2H), 6.99–6.95 (m, 2H), 4.27 (s, 2H), 2.55 ppm (s, 2H);
13C NMR (75 MHz, CDCl3): d=158.4, 134.8 (4C), 133.4, 132.0, 130.0
(2C), 128.1 (4C), 127.1, 126.9, 123.0, 119.7 (2C), 62.7, 14.4 ppm.
1(2H)-carboxylate (21a): M.p.: 1348C; 1H NMR (200 MHz, CDCl3):
d=8.31 (s, 1H), 8.30 (d, J=8.0 Hz, 1H), 7.15 (d, J=8.0 Hz, 1H), 3.01
(s, 2H), 1.92 (s, 2H), 1.46 (s, 9H), 0.13 ppm (s, 6H); 13C NMR
(75 MHz, CDCl3): d=154.6, 151.7, 148.3, 147.0, 129.0, 121.3, 81.2,
34.7, 28.4 (3C), 15.6, ꢀ2.7 ppm (2C); IR (neat): n˜ =2965, 2878, 1693,
1363, 1149, 825 cmꢀ1; MS (EI, 70 eV): m/z (%): 278 (20) [M+], 222
(92) [M+ꢀtBu], 178 (92) [M+ꢀBoc], 163 (100) [M+ꢀBocꢀMe];
HRMS (ESI+): m/z calcd for C14H23N2O2Si: 279.1529; found:
279.1517.
4-Methyl-4-phenyl-3,4-dihydro-2H-benzo-1,4-oxasiline
(38b):
1H NMR (300 MHz, CDCl3): d=7.61–7.58 (m, 4H), 7.43–7.33 (m, 8H),
6.99–6.92 (m, 2H), 4.45–4.41 (m, 2H), 1.65–1.61 ppm (m, 2H);
13C NMR (75 MHz, CDCl3): d=165.5, 136.5, 135.4 (4C), 134.4, 131.5,
129.8 (2C), 128.0 (4C), 121.1, 117.9 (2C), 116.9, 66.5, 15.5 ppm.
tert-Butyl
5-chloro-3,3-dimethyl-3,4-dihydropyrido[3,4-e]-
3-Benzyl-3-methyl-2,3-dihydrobenzo[d][1,3]oxasilole
(38c):
[1,3]azasiline-1(2H)-car-boxylate (22a): M.p.: 898C 1H NMR
(300 MHz, CDCl3): d=8.08 (d, J=5.4 Hz, 1H), 7.11 (d, J=5.4 Hz,
1H), 3.02 (s, 2H), 2.15 (s, 2H), 1.46 (s, 9H), 0.17 ppm (s, 6H);
13C NMR (75 MHz, CDCl3) d=154.2, 152.1, 150.5, 145.1, 128.5, 121.3,
81.6, 34.4, 28.3 (3C), 15.3, ꢀ2.9 ppm (2C); IR (neat): n˜ =2962, 2874,
1707, 1578, 1373, 1314, 1144, 822 cmꢀ1; MS (CI, isobutane): m/z
(%): 315/313 (100) [M+H+], 163 (80) [M+H+ꢀtBu]; HRMS (ESI+):
m/z calcd for C14H22ClN2O2Si: 313.1139; found: 313.1141.
1H NMR (300 MHz, CDCl3): d=7.53–7.50 (m, 2H), 7.40–7.33 (m, 5H),
7.19–7.08 (m, 3H), 6.99–6.84 (m, 4H), 4.04 (s, 2H), 2.82 (A of AB
quartet, J=15.0 Hz, 1H), 2.72 ppm (B of AB quartet, J=15.0 Hz,
1H); 13C NMR (75 MHz, CDCl3): d=169.3, 137.1, 134.7 (2C), 133.7,
133.2, 132.3, 130.2, 128.6 (2C), 128.4 (2C), 128.1 (2C), 124.9, 120.5,
117.0, 112.9, 60.4, 22.3 ppm; HRMS (ESI+): m/z calcd for C20H18OSi:
302.1127; found: 302.1120.
tert-Butyl 4-methyl-4-phenyl-3,4-dihydrobenzo[b][1,4]azasiline-
1(2H)-carboxylate (40b): 1H NMR (300 MHz, CDCl3): d=7.53–7.47
(m, 2H), 7.41–7.28 (m, 6H), 7.13 (ddd, J=7.3, 7.2, 1.5 Hz, 1H), 4.12–
3.99 (m, 1H), 3.87–3.76 (m, 1H), 1.55–1.43 (m, 1H), 1.47 (s, 9H),
1.37–1.25 (m, 1H), 0.59 ppm (s, 3H); 13C NMR (75 MHz, CDCl3): d=
154.0, 149.3, 137.1, 134.3, 134.1 (2C), 129.4, 129.3, 128.6, 127.9 (2C),
126.2, 124.8, 80.3, 45.1, 28.3 (3C), 13.9, ꢀ3.2 ppm; HRMS (ESI+): m/
z calcd for C20H26NO2Si: 340.1733; found: 340.1736.
tert-Butyl
3,3-dimethyl-3,4-dihydropyrido[4,3-e][1,3]azasiline-
1
1(2H)-carboxylate (23a): H NMR (300 MHz, CDCl3): d=8.33 (d, J=
4.9 Hz, 1H), 8.30 (s, 1H), 7.13 (d, J=4.9 Hz, 1H), 2.94 (A of AB quar-
tet, J=15.2 Hz, 1H), 2.73 (B of AB quartet, J=15.2 Hz, 1H), 2.24 (s,
2H), 1.32 (s, 9H), 0.21 (s, 3H), 0.17 ppm (s, 3H); 13C NMR (75 MHz,
CDCl3): d=155.6, 148.5, 147.4, 144.3, 140.5, 125.4, 80.7, 43.9, 28.1
(3C), 17.2, ꢀ0.61 ppm (2C); IR (neat): n˜ =2965, 1689, 1364,
1155 cmꢀ1; HRMS (ESI+): m/z calcd for C14H23N2O2Si: 279.1529;
found: 279.1531.
tert-Butyl 4,4-diphenyl-3,4-dihydrobenzo[b][1,4]azasiline-1(2H)-
carboxylate (41b): 1H NMR (300 MHz, CDCl3): d=7.54–7.51 (m,
4H), 7.41–7.36 (m, 9H), 7.13 (dd, J=7.0, 1.6 Hz, 1H), 4.03–3.99 (m,
2H), 1.71–1.67 (m, 2H), 1.37 ppm (s, 9H); 13C NMR (75 MHz, CDCl3):
d=154.0, 149.8, 135.4, 135.3 (4C), 134.7 (2C), 129.7 (2C), 129.6,
3,3-Dimethyl-3,4-dihydro-2H-[1,3]oxasilino[6,5-c]pyridine (24a):
1H NMR (300 MHz, CDCl3): d=8.22 (s, 1H), 8.09 (d, J=4.5 Hz, 1H),
7.02 (d, J=4.5 Hz, 1H), 3.67 (s, 2H), 2.20 (s, 2H), 0.27 ppm (s, 6H);
Chem. Eur. J. 2014, 20, 10131 – 10139
www.chemeurj.org
10137
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
128.0 (4C), 127.0, 126.4, 124.9, 80.3, 45.1, 28.2 (3C), 13.2 ppm;
HRMS (ESI+): m/z calcd for C25H28NO2Si: 402.1889; found:
402.1894.
(75 MHz, CDCl3): d=154.3, 148.7, 133.5, 129.1, 128.5, 126.4, 125.4,
80.2, 46.0, 28.5 (3C), 16.1, ꢀ1.7 ppm (2C); IR (neat) : n˜ =2971, 2929,
2901,1694, 1421, 802, 744 cmꢀ1; MS (CI, isobutane) m/z (%): 323
(11) [M+H+], 306 (11) [M+H+ꢀMe], 268 (100) [M+H+ꢀtBu], 222
(22) [MꢀBoc]; HRMS (ESI+): m/z calcd for C17H26N2O2GeNa:
387.1114 [M+Na+MeCN]; found: 387.1104.
tert-Butyl
3-benzyl-3-phenyl-2,3-dihydro-1H-benzo[d]-
1
[1,3]azasilole-1-carboxylate (41c) : H NMR (300 MHz, CDCl3): d=
8.24 (d, J=8.6 Hz, 1H), 7.56–7.45 (m, 2H), 7.43–7.30 (m, 5H), 7.18–
7.03 (m, 3H), 7.02–6.89 (m, 3H), 3.29 (s, 2H), 2.79 (A of AB quartet,
J=14.1 Hz, 1H), 2.68 (B of AB quartet, J=14.1 Hz, 1H), 1.54 ppm
(s, 9H); 13C NMR (75 MHz, CDCl3): d=154.3, 152.6, 137.1, 135.3,
134.7 (2C), 133.6, 131.5, 130.2, 128.6 (2C), 128.4 (2C), 128.1 (2C),
127.9, 124.9, 122.0, 118.1, 80.7, 35.6, 28.4 (3C), 22.6 ppm; MS (CI,
isobutane): m/z (%): 402 (11) [M+H+], 346 (100) [M+H+ꢀtBu],
302 (55) [M+H+ꢀBoc].
tert-Butyl 3,3-dimethyl-2,3-dihydro-1H-benzo[d][1,3]azagermole-
1-carboxylate (56d): 1H NMR (300 MHz, CDCl3): d=8.24 (d, J=
8.4 Hz, 1H), 7.45 (dd, J=7.5, 1.5 Hz, 1H), 7.31 (ddd, J=8.4, 7.5,
1.5 Hz, 1H), 6.99 (tap. , J=7.5 Hz, 1H), 3.27 (s, 2H), 1.57 (s, 9H),
0.53 ppm (s, 6H); 13C NMR (75 MHz, CDCl3): d=155.0, 149.3, 132.5,
130.1, 129.6, 122.2, 118.6, 80.7, 37.8, 28.6 (3C), ꢀ1.7 ppm (2C); MS
(CI, isobutane) m/z (%): 309 (11) [M+H+], 292 (9) [MꢀMe], 254
(100) [M+H+ꢀtBu], 210 (37) [M+H+ꢀBoc].
tert-Butyl
3-methyl-3-phenyl-3,4-dihydropyrido[3,4-e]-
[1,3]azasiline-1(2H)-carboxylate (42a): 1H NMR (300 MHz, CDCl3):
d=8.35–8.33 (m, 2H), 7.44–7.31 (m, 5H), 7.22 (d, J=5.7 Hz, 1H),
3.38 (d, J=15.0 Hz, 1H), 3.20 (d, J=15.0 Hz, 1H), 2.22 (d, J=
15.0 Hz, 1H), 2.11 (d, J=15.0 Hz, 1H), 1.48 (s, 9H), 0.42 ppm (s,
3H); 13C NMR (75 MHz, CDCl3): d=154.3, 151.7, 148.4, 147.0, 134.6,
133.6 (2C), 130.1, 128.5, 128.1 (2C), 121.2, 81.2, 34.2, 28.2 (3C), 14.8,
ꢀ4.5 ppm.
tert-Butyl
6-(dibenzylamino)-3,3-dimethyl-2,3-dihydro-1H-
benzo[d][1,3]azagermole-1-carboxylate (57b): 1H NMR (300 MHz,
CDCl3): d=7.37–7.20 (m, 10H), 7.15 (d, J=8.2 Hz, 1H), 6.70 (brs,
1H), 6.60 (dd, J=8.2, 2.5 Hz, 1H), 4.63 (s, 4H), 3.80–3.72 (m, 2H),
1.37 (s, 9H), 1.39–1.29 (m, 2H), 0.34 ppm (s, 6H); 13C NMR (75 MHz,
CDCl3): d=154.2, 149.8, 149.6, 138.5 (2C), 133.9, 128.6 (4C), 126.9
(2C), 126.6 (4C), 119.7, 110.4, 110.2, 79.7, 53.8 (2C), 46.1, 28.3 (3C),
15.8, ꢀ1.6 ppm (2C); IR (neat) : n˜ =3296, 2928, 1689, 1595, 1363,
tert-Butyl
3,3-diphenyl-3,4-dihydropyrido[3,4-e][1,3]azasiline-
1150, 1014, 730, 695 cmꢀ1
;
HRMS (ESI+): m/z calcd for
1
1(2H)-carboxylate (43a): H NMR (300 MHz, CDCl3): d=8.06 (d, J=
5.4 Hz, 2H), 7.52–7.36 (m, 11H), 3.62 (s, 2H), 2.47 (s, 2H), 1.48 ppm
(s, 9H).
C29H37N2O272Ge: 517.2076; found: 517.2094.
tert-Butyl 7-(dibenzylamino)-4,4-dimethyl-3,4-dihydrobenzo[b]-
[1,4]azagermine-1(2H)-carboxylate (57d): 1H NMR (300 MHz,
CDCl3): d=7.82 (brs, 1H), 7.38–7.19 (m, 11H), 6.41 (dd, J=8.1,
2.2 Hz, 1H), 4.66 (s, 4H), 3.26 (s, 2H), 1.45 (s, 9H), 0.48 ppm (s, 6H);
13C NMR (50 MHz, CDCl3): d=154.8, 151.1, 150.8, 138.5 (2C), 132.9,
128.5 (4C), 126.9 (2C), 126.8 (4C), 116.2, 107.2, 102.7, 80.4, 53.7 (2C),
38.7, 28.3 (3C), ꢀ1.7 ppm (2C); IR (neat) : n˜ =3308, 2973, 1694,
1592, 1326, 1154, 727, 694 cmꢀ1; HRMS (ESI+): m/z calcd for
C28H35N2O272Ge: 503.1919; found: 503.1930.
tert-Butyl
3-benzyl-3-phenyl-2,3-dihydro-1H-[1,3]azasilolo[4,5-
c]pyridine-1-carboxylate (43c): 1H NMR (300 MHz, CDCl3): d=
8.42–8.40 (m, 2H), 7.47–7.28 (m, 4H), 7.19–7.01 (m, 5H), 6.86–6.85
(m, 2H), 3.25 (A of AB quartet, J=15.3 Hz, 1H), 3.15 (B of AB quar-
tet, J=15.3 Hz, 1H), 2.50 (A of AB quartet, J=13.8 Hz, 1H), 2.43 (B
of AB quartet, J=13.8 Hz, 1H), 1.33 ppm (s, 9H); 13C NMR (75 MHz,
CDCl3): d=154.7, 151.4, 150.1, 147.0, 137.7, 135.1, 133.8 (2C), 129.8
(2C), 128.6 (2C), 128.4(2C), 127.7, 124.7, 119.74, 119.68, 82.3, 41.1,
28.0 (3C), 25.0 ppm.
3,3-Dimethyl-1-tosyl-1,2,3,4-tetrahydrobenzo[e][1,3]azagermine
1
3,3-Dimethyl-3,4-dihydro-2H-benzo[e][1,3]oxagermine
(54a):
(58a): H NMR (300 MHz, CDCl3): d=7.41 (d, J=8.2 Hz, 2H), 7.30 (d,
1H NMR (300 MHz, CDCl3): d=7.30–7.27 (m, 1H); 7.11–7.06 (m, 2H),
6.99–6.94 (m, 1H), 3.92 (s, 2H), 2.12 (s, 2H), 0.31 ppm (s, 6H);
13C NMR (75 MHz, CDCl3): d=158.0, 131.2, 130.9, 126.6, 123.5,
120.5, 66.0, 16.5, ꢀ3.3 (2C); MS (CI, isobutane) m/z (%): 265 (8)
[M+C3H7+], 224 (100) [M+H+], 209 (30) [MꢀMe].
J=7.3 Hz, 2H), 7.18 (d, J=8.0 Hz, 2H), 7.07–7.02 (m, 2H), 3.33 (s,
2H), 2.60 (s, 2H), 2.43 (s, 3H), 0.15 ppm (s, 6H); HRMS (ESI+): m/z
calcd for C17H21NO2GeSNa: 400.0402; found: 400.0409.
4,4-Dimethyl-1-tosyl-1,2,3,4-tetrahydrobenzo[b][1,4]azagermine
1
(58b): H NMR (300 MHz, CDCl3): d=7.60–7.57 (m, 1H), 7.46 (d, J=
4,4-Dimethyl-3,4-dihydro-2H-benzo[b][1,4]oxagermine
(54b):
8.0 Hz, 2H), 7.30 (d, J=7.3 Hz, 2H), 7.27–7.21 (m, 1H), 7.18 (d, J=
8.0 Hz, 2H), 4.03–3.99 (m, 2H), 2.36 (s, 3H), 1.27–1.23 (m, 2H),
0.05 ppm (s, 6H); 13C NMR (75 MHz, CDCl3): d=146.3, 143.3, 138.6,
135.0, 134.2, 129.6 (2C), 129.2, 127.5 (2C), 127.3, 126.6, 48.9, 21.6,
14.5, ꢀ1.9 ppm (2C); HRMS (ESI+): calcd for C19H24N2O2GeSNa:
441.0668 [M+MeCN+H+]; found: 441.0679.
1H NMR (300 MHz, CDCl3): d=7.75 (dd, J=8.2, 1.5 Hz, 1H), 7.32–
7.30 (m, 1H), 6.94–6.86 (m, 1H), 6.72–6.69 (m,1H), 4.32–4.27 (m,
2H), 1.40–1.36 (m, 2H), 0.43 ppm (6H); 13C NMR (75 MHz, CDCl3):
d=164.6, 136.4, 130.1, 127.0, 121.8, 118.1, 68.2, 15.8, ꢀ1.8 ppm
(2C).
1
3,3-Dimethyl-2,3-dihydrobenzo[d][1,3]oxagermole (54d): H NMR
tert-Butyl 3,3-dimethyl-7-(trifluoromethyl)-3,4-dihydrobenzo[e]-
[1,3]azagermine-1(2H)-carboxylate (59a): 1H NMR (300 MHz,
CDCl3): d=7.39 (brs, 1H), 7.33 (d, J=7.8 Hz, 1H), 7.21 (d, J=7.8 Hz,
1H), 3.10 (brs, 2H), 2.09 (s, 2H), 1.41 (s, 9H), 0.29 ppm (6H);
13C NMR (75 MHz, CDCl3): d=154.7, 140.8, 140.4, 129.9, 127.7 (q,
J=33 Hz), 125.2 (q, J=4 Hz), 124.3 (q, J=270 Hz), 122.7 (q, J=
4 Hz), 80.5, 35.3, 28.4 (3C), 19.2, ꢀ2.9 ppm (2C); 19F NMR (CFCl3):
d=ꢀ62.25 ppm (s, 3F); IR (neat) : n˜ =2983, 2897, 1690, 1334, 1122,
(300 MHz, CDCl3): d=7.48 (dd, J=6.9, 1.5 Hz, 1H), 7.29–7.24 (m,
1H), 6.94–6.86 (m, 2H), 4.08 (s, 2H), 0.59 ppm (s, 6H); 13C NMR
(75 MHz, CDCl3): d=165.8, 134.8, 132.8, 131.0, 120.5, 113.0, 62.8,
ꢀ1.5 ppm (2C); MS (CI, isobutane): m/z (%): 211 (100) [M+H+].
4,4-Dimethyl-3,4-dihydro-2H-benzo[b][1,4]thiagermine
(55b):
1H NMR (300 MHz, CDCl3): d=7.40–7.37 (m, 1H), 7.30–7.27 (m, 1H),
7.18–7.15 (m, 2H), 3.05–3.01 (m, 2H), 1.56–1.52 (m, 2H), 0.42 ppm
(s, 6H); 13C NMR (75 MHz, CDCl3): d=145.0, 139.0, 134.5, 129.1,
128.5, 125.7, 31.0, 18.3, ꢀ1.5 ppm (2C); IR (neat) : n˜ =3050, 2966,
2903, 1421, 1150, 744 cmꢀ1; MS (CI, isobutane): m/z (%): 240 (22)
[M+H+], 213 (100), 209 (55) [Mꢀ2Me].
834, 610 cmꢀ1
.
tert-Butyl 3,3-dimethyl-3,4-dihydropyrido[3,4-e][1,3]azagermine-
1(2H)-carboxylate (60a): 1H NMR (300 MHz, CDCl3): d=8.35 (s,
1H), 8.31 (d, J=5.4 Hz, 1H), 7.08 (d, J=5.4 Hz, 1H), 3.15 (s, 2H),
2.02 (s, 2H), 1.44 (s, 9H), 0.30 ppm (6H); 13C NMR (75 MHz, CDCl3):
d=154.2, 151.1, 147.9, 146.9, 130.8, 122.0, 80.9, 34.9, 28.2 (3C),
15.0, ꢀ3.0 ppm (2C); IR (neat): n˜ =2979, 2902, 1690, 1367, 1156,
tert-Butyl
4,4-dimethyl-3,4-dihydrobenzo[b][1,4]azagermine-
1(2H)-carboxylate (56b): 1H NMR (300 MHz, CDCl3): d=7.37–7.33
(m, 1H), 7.30–7.25 (m, 2H), 7.17–7.12 (m, 1H), 3.79–3.75 (m, 2H),
1.50–1.44 (m, 9H), 1.44–1.40 (m, 2H), 0.37 ppm (s, 6H); 13C NMR
819, 611 cmꢀ1
.
Chem. Eur. J. 2014, 20, 10131 – 10139
www.chemeurj.org
10138
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[7] For a recent example in medicinal chemistry, see: S. Fujii, Y. Miyajima,
H. Masuno, H. Kagechika, J. Med. Chem. 2013, 56, 160–166.
tert-Butyl 3,3-dimethyl-3,4-dihydropyrido[4,3-e][1,3]azagermine-
1(2H)-carboxylate (61a): 1H NMR (300 MHz, CDCl3): d=8.27 (s,
1H), 8.17 (d, J=4.8 Hz, 1H), 6.99 (d, J=4.8, 1H), 3.15 (brs, 2H), 1.99
(s, 2H), 1.36 (s, 9H), 0.26 ppm (s, 6H); 13C NMR (75 MHz, CDCl3): d=
154.5, 148.7, 146.5, 144.9, 137.4, 124.1, 80.6, 30.3, 28.3 (3C), 18.7,
[8] For examples in the design of odorants and fungicides, respectively,
see: a) R. Tacke, B. Becker, D. Berg, W. Brandes, S. Dutzmann, K. Schaller,
J. Organomet. Chem. 1992, 438, 45–55; b) R. Tacke, T. Schmid, C. Bursch-
ka, M. Penka, H. Surburg, Organometallics 2002, 21, 113–120.
[9] G. A. Showell, M. J. Barnes, J. O. Daiss, J. S. Mills, J. G. Montana, R. Tacke,
J. B. H. Warneck, Bioorg. Med. Chem. Lett. 2006, 16, 2555–2558.
[10] a) R. Tacke, T. Kornek, T. Heinrich, C. Burschka, M. Penka, M. Pꢃlm, C.
Keim, E. Mutschler, G. Lambrecht, J. Organomet. Chem. 2001, 640, 140–
165; b) R. Tacke, V. I. Handmann, K. Kreutzmann, C. Keim, E. Mutschler,
G. Lambrecht, Organometallics 2002, 21, 3727–3752.
[11] a) R. Tacke, T. Heinrich, R. Bertermann, C. Burscka, A. Hamacher, M. U.
Kassack, Organometallics 2004, 23, 4468–4477; b) R. Tacke, F. Popp, B.
Mꢃller, B. Theis, C. Burschka, A. Hamacher, M. U. Kassack, D. Schepmann,
B. Wꢃnsch, U. Jurva, E. Wellner, ChemMedChem 2008, 3, 152–164; c) T.
Johansson, L. Weidolf, F. Popp, R. Tacke, U. Jurva, Drug Metabolism and
Disposition 2010, 38, 73–83.
[12] a) S. Dçrrich, J. B. Bauer, S. Lorenzen, C. Mahler, S. Schweeberg, C.
Burschka, J. A. Baus, R. Tacke, P. Kraft, Chem. Eur. J. 2013, 19, 11396–
11408; b) B. Fçrster, R. Bertermann, P. Kraft, R. Tacke, Organometallics
2014, 33, 338–346.
[13] For recent and important examples, see: a) C. Blaszykowski, C. Brancour,
A. L. Dhimane, L. Fensterbank, M. Malacria, Eur. J. Org. Chem. 2009,
1674–1678; b) S. Furukawa, J. Kobayashi, T. Kawashima, J. Am. Chem.
Soc. 2009, 131, 14192–14193; c) S. V. Kirpichenko, A. I. Albanov, J. Orga-
nomet. Chem. 2010, 695, 663–666; d) K. Ouyang, Y. Liang, Z. Xi, Org.
Lett. 2012, 14, 4572–4575; e) G. K. Min, D. Hernꢄndez, T. Skrydstrup,
Acc. Chem. Res. 2013, 46, 457–470 and references cited therein; f) A.
Giros, R. Guillot, L. Blanco, S. Deloisy, Eur. J. Org. Chem. 2013, 5809–
5813.
ꢀ2.9 ppm (2C); IR (neat) : n˜ =2976, 2908, 1365, 1154, 833 cmꢀ1
;
HRMS (ESI+): m/z calcd for C14H23GeN2O2: 325.0971; found:
325.0974.
3,3-Dimethyl-3,4-dihydro-2H-[1,3]oxagermino[5,6-c]pyridine
1
(62a): H NMR (300 MHz, CDCl3): d=8.18 (s, 1H), 8.13 (d, J=5.0 Hz,
1H), 7.04 (d, J=5.0, 1H), 3.95 (s, 2H), 2.09 (s, 2H), 0.32 ppm (s,
6H); 13C NMR (75 MHz, CDCl3): d=153.7, 144.7, 142.3, 140.3, 125.4,
60.1, 16.2, ꢀ3.2 ppm (2C).
tert-Butyl
4-methyl-3,3-dimethyl-2,3-dihydro-1H-benzo[d]-
[1,3]azagermole-1-carboxylate (63d): 1H NMR (300 MHz, CDCl3):
d=8.09 (d, J=8.4 Hz, 1H), 7.23–7.21 (m, 1H), 6.82 (d, J=7.2 Hz,
1H), 3.26 (s, 2H), 2.38 (s, 3H), 1.58 (s, 9H), 0.58 ppm (s, 6H);
13C NMR (75 MHz, CDCl3): d=155.0, 149.3, 142.1, 130.6, 129.1,
122.6, 116.0, 80.6, 37.6, 28.6 (3C), 23.1, ꢀ1.9 ppm (2C); IR (neat) :
n˜ =2971, 2903, 1699, 1319, 1125, 778 cmꢀ1; MS: m/z (%): 320/322/
324 (30) [M+H+], 305/307/309 (100) [MꢀMe], 264/266/268 (75)
[M<M->tBu]; HRMS (ESI+): m/z calcd for C15H24NO2Ge: 324.1019;
found: 324.1021.
Acknowledgements
C.F., L.M., and T.B. are grateful to Janssen and the Laboratoire
Commun de Recherche Janssen-COBRA for Ph.D. and postdoc-
toral stipends. We also acknowledge the Rꢂgion de Haute-Nor-
mandie and the CRUNCh interregional program for their con-
stant support to our research projects. The Labex SynOrg
(ANR-11-LABX-0029) supported part of this work. The authors
would also like to thank Dr. Luc Van Hijfte for his support
during the course of the project.
[14] C. FranÅois, T. Boddaert, M. Durandetti, O. Querolle, L. Van Hijfte, L.
Meerpoel, P. Angibaud, J. Maddaluno, Org. Lett. 2012, 14, 2074–2077.
[15] In some cases, tBuLi (ꢀ788C, 2.4 equiv) was employed instead of n-BuLi
(ꢀ408C, 1.2 equiv) to avoid a Wurtz–Fittig-type addition of the aryllithi-
um on the co-generated butyl halide, a phenomenon well-known in
the literature: a) W. F. Bailey, S. C. Longstaff, Chimica Oggi (Chem. Today)
2001, 19, 28–32; b) J. Clayden, in Organolithiums: Selectivity for Synthe-
sis, Pergamon Press, Oxford, 2002, pp. 111–142.
[16] a) V. I. Handmann, R. Bertermann, C. Burschka, R. Tacke, J. Organomet.
Chem. 2000, 613, 19–25; b) M. Shimizu, M. Iwakubo, Y. Nishihara, K.
Oda, T. Hiyama, ARKIVOC (Gainesville, FL, U.S.) 2007, 29–48.
[17] E. P. A. Couzijn, A. W. Ehlers, M. Schakel, K. Lammertsma, J. Am. Chem.
Soc. 2006, 128, 13634–13639.
Keywords: cyclization
· heterocycles · lithium · reaction
mechanisms · rearrangement
[18] a) R. Damrauer, V. E. Yost, S. E. Danahey, B. K. O’Connell, Organometallics
1985, 4, 1779–1784; b) M. J. Allen, S. L. Aprahamian, E. A. Sans, H.
Shechter, J. Org. Chem. 2002, 67, 3561–3574.
[19] a) S. Berry, J. Chem. Phys. 1960, 32, 933–938; b) E. P. A. Couzijn, J. C.
Slootweg, A. W. Ehlers, K. Lammertsma, J. Am. Chem. Soc. 2010, 132,
18127–18140; c) C. Moberg, Angew. Chem. 2011, 123, 10473–10475;
Angew. Chem. Int. Ed. 2011, 50, 10290–10292.
[1] O. F. Guner, J. P. Bowen, Curr. Opin. Drug. Discovery Dev. 2013, 13, 1327–
1342.
[2] For a general review on bioisostere in drug design, see: N. A. Meanwell,
J. Med. Chem. 2011, 54, 2529–2591.
[3] For reviews in more specific C/Si isosterism, see: a) W. Bains, R. Tacke,
Curr. Opin. Drug Discovery Dev. 2003, 6, 526–543; b) J. S. Mills, G. A.
Showell, Expert Opin. Invest. Drugs 2004, 13, 1149–1157; c) A. K. Franz,
Curr. Opin. Drug Discovery Dev. 2007, 10, 654–671.
[4] For a review on organosilicon molecules with medicinal applications,
see: A. K. Franz, S. O. Wilson, J. Med. Chem. 2013, 56, 388–405.
[5] For a review on C/Si switch in odorant design, see: R. Tacke, S. Metz,
Chem. Biodiversity 2008, 5, 920–941.
[20] N. V. Timosheva, T. K. Prakasha, A. Chandrasekaran, R. O. Day, R. R.
Holmes, Inorg. Chem. 1995, 34, 4525–4526.
[21] It is anyway unlikely that the two phenyls end up within the equatorial
plane since the repulsive overlap between the aromatic p-systems is
expected to destabilize this isomer: E. P. A. Couzijn, D. W. F. van den En-
gel, J. C. Slootweg, F. J. J. de Kanter, A. W. Ehlers, M. Schakel, K. Lam-
mertsma, J. Am. Chem. Soc. 2009, 131, 3741–3751.
[22] We thank one of the referees for eliciting this discussion.
[6] For a review on C/Si/Ge switch in medicinal chemistry, see: R. Tacke, T.
Heinrich, T. Kornek, M. Merget, S. A. Wagner, J. Gross, C. Keim, G. Lam-
brecht, E. Mutschler, T. Beckers, M. Bernd, T. Reissmann, Phosphorus
Sulfur Silicon Relat. Elem. 1999, 150–151, 69–87.
Received: March 13, 2014
Published online on July 3, 2014
Chem. Eur. J. 2014, 20, 10131 – 10139
www.chemeurj.org
10139
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim