Synthesis of 3-Benzazepines by Metal-Free Oxidative C–H Bond Functionalization–Ring Expansion Tandem Reaction

Article abstract of DOI:10.1002/adsc.201600985

A metal-free synthesis of biologically important benzazepines is achieved through a single synthetic operation involving an oxidative C–H bond functionalization and ring expansion with diazomethanes as key reagent. This represents a new, strong methodology for the straightforward construction of the seven-ring N-heterocyclic structures under mild conditions using a 2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPO) oxoammonium salt as oxidant. Moderate to good yields are achieved from simple, readily available tetrahydroisoquinolines, and this methodology has been further successfully applied for the synthesis of the 3-benzazepine drug Lorcaserin. A possible mechanistic pathway for the ring expansion step, comprising the extrusion of nitrogen in a concerted asynchronic process, is proposed based on both mechanistic proof and density function theory (DFT) calculations. (Figure presented.).

Full text of DOI:10.1002/adsc.201600985

FULL PAPERS
DOI: 10.1002/adsc.201600985
Synthesis of 3-Benzazepines by Metal-Free Oxidative C–H Bond
Functionalization–Ring Expansion Tandem Reaction
a,b
+a,b
+c
a
Andrea Gini, Julia Bamberger,
Javier Luis-Barrera, Mercedes Zurro,
d
c,
a,b,
Rubꢀn Mas-Ballestꢀ, Josꢀ Alemꢁn, * and Olga Garcꢂa MancheÇo *
a
Institute for Organic Chemistry, University of Regensburg, 93053 Regensburg, Germany
Fax : (+49)-941-943-1717; e-mail: olga.garcia-mancheno@chemie.uni-regensburg.de
Straubing Center of Science for Renewable Resources, 94315 Straubing, Germany
Organic Chemistry Department, Universidad Autꢃnoma de Madrid (U.A.M.), 28049 Madrid, Spain
b
c
Fax : (+34)-91-497-3966; e-mail: jose.aleman@uam.es (homepage: www.uam.es/jose.aleman)
Inorganic Chemistry Department, Universidad Autꢃnoma de Madrid (U.A.M.), 28049 Madrid, Spain
d
+
These authors contributed equally to this work.
Received: September 6, 2016; Revised: October 6, 2016; Published online: && &&, 0000
Supporting information for this article can be found under: http://dx.doi.org/10.1002/adsc.201600985.
Abstract: A metal-free synthesis of biologically im- odology has been further successfully applied for the
portant benzazepines is achieved through a single synthesis of the 3-benzazepine drug Lorcaserin. A
synthetic operation involving an oxidative C–H bond possible mechanistic pathway for the ring expansion
functionalization and ring expansion with diazome- step, comprising the extrusion of nitrogen in a con-
thanes as key reagent. This represents a new, strong certed asynchronic process, is proposed based on
methodology for the straightforward construction of both mechanistic proof and density function theory
the seven-ring N-heterocyclic structures under mild (DFT) calculations.
conditions using a 2,2,6,6-tetramethylpiperidine 1-
oxyl (TEMPO) oxoammonium salt as oxidant. Mod-
erate to good yields are achieved from simple, readi- Keywords: benzazepines; isoquinolines; metal-free
ly available tetrahydroisoquinolines, and this meth- conditions; ring-expansion; tandem reaction
Introduction
3-Benzazepines are a unique class of compounds im-
portant in drug discovery. In particular, the tetrahy-
dro-3-benzazepine is a common skeleton in a number
[
1]
of natural and pharmaceutical products. In the last
half century, these compounds have been explored as
a
potent dopaminometic and antidopaminergic
[2]
agents. They also present other pharmacological ac-
tivities such as analgesic, antihypertensive or anticanc-
er properties. Thus, for example, (Figure 1) Lorca-
serin has serotonergic properties and acts as an ano-
rectic compound, which is currently used as a weight-
[3]
[4]
loss drug. Furthermore, Fenoldopam is employed as
[
5]
antihypertensive agent, whereas Ivabradine is a car-
diotonic agent used for the symptomatic management
of angina pectoris by inhibition of the funny chan-
[6]
nel.
The main strategy to synthesize the 3-benzazepine Figure 1. Selected examples of biologically active 3-benzaze-
core is based on a Lewis-acid mediated Friedel– pines (LG= leaving group, TM= transition metal).
Adv. Synth. Catal. 0000, 000, 0 – 0
1
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
rivatives, which will liberate a nitrogen molecule as
leaving group, would be ideal as nucleophiles.
However, under the required oxidative conditions,
important compatibility issues have to be overcome
with this type of reagents in order to avoid their in-
trinsic reactivity to form the corresponding carbonyl
[
12]
compound [Scheme 2, Eq. (1)]. Therefore, the oxi-
dative formation of the iminium species [Eq. (2), k2]
must be faster than the decomposition pathway of the
diazo compound [Eq. (1), k ] to undergo the desired
1
nucleophilic attack.
Scheme 2. Undesired competitive typical oxidation of diazo-
methanes to carbonyls.
Scheme 1. Classical and present synthetic approaches.
Crafts-type cyclization approach (Scheme 1, top).
Herein, we present a straightforward preparation of
benzazepines by an innovative, mild and functional
[
7,8]
In all these cases, the 2-arylethylamines are used as group tolerant (R and R’) metal-free oxidative C–H
starting materials, which have to be derived to an acy- bond functionalization–ring expansion process from
[
13–15]
clic key intermediate for the final cyclization through simple THIQs and diazomethanes.
several synthetic steps. Recently, N-alkyl-3-benzaze-
pines have been afforded by an osmium-catalyzed hy-
[
9]
droamination reaction. This methodology also re- Results and Discussion
quires multistep substrate synthesis and is limited to
N-alkyl derivatives. Besides these principal ap- We started our screening with N-phenyl-THIQ 1a as
proaches, other synthetic routes include Pummerer model substrate and TMSCHN as nucleophile for
2
cyclizations, electrophilic aromatic substitutions or the the optimization study of the reaction conditions
condensation of keto acids with amines, among a few (Table 1). A variety of standard oxidants in C–H func-
[8]
others. However, most of these methods required (i) tionalization were tested in dichloromethane at room
several steps, (ii) harsh conditions, (iii) prolonged re- temperature (entries 1–8). DDQ (I) and t-BuOOH
action times and (iv) do not tolerate a broad substitu- (II) led to a complex mixture or no conversion, re-
tion pattern on the nitrogen atom and the aromatic spectively. To our delight, the use of 1.2 equivalents of
+
À
[16,17]
ring (e.g., need of electron-rich groups). These issues the TEMPO oxoammonium salt III (T BF )
pro-
4
limit the preparation of diverse substituted 3-benzaze- vided for the first time the benzazepine 3a (entry 3).
pines. Therefore, there is still a need for simple, mild The initial low conversion of 43% was enhanced to
and direct synthetic methods for their synthesis. The 88% by employing 1.5 equivalents of the oxidant
common oxidative functionalization of tetrahydroiso- (entry 4). Other TEMPO salts (IV–VI) with different
quinolines (THIQs) implies the formation of the imi- counter-anions and substitution in the backbone also
nium ion intermediate and the following addition of promoted this reaction (entries 5–7). With the best
nucleophiles in the alpha position (Scheme 1, oxoammonium salt III, the temperature was increased
[10,11]
middle).
We have envisioned a new methodology to 60 and 808C, finding full conversion in both cases
for the construction of the seven-membered ring after 4 and 1 hours, respectively (entries 8 and 9).
structures by an oxidative C–H functionalization ap- However, when the amount of 2a was decreased to
[
11]
proach from readily available THIQ.
To achieve 1.5 equivalents, lower conversions were found at both
such a transformation, the nucleophile employed 60 and 808C (entries 10 and 11).
should possess a leaving group in its structure that
Under the optimized conditions (entry 9), the scope
can promote, after the first addition, a ring expansion of the reaction was investigated (Table 2). Although
process (Scheme 1, bottom). Thus, diazomethane de- we focussed on the synthesis of highly versatile 1,2-
Adv. Synth. Catal. 0000, 000, 0 – 0
2
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
[a]
Table 1. Optimization of the reaction conditions.
[b]
Ent
Oxidant (x equiv.)
2a (equiv.)
T [8C]
t [h]
Conv. (yield) [%]
[
[
c]
1
2
3
4
5
6
7
8
9
DDQ (I) (1.2)
t-BuOOH (II) (1.2)
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
1.5
1.5
r.t.
r.t.
r.t.
r.t.
r.t.
r.t.
r.t.
60
18
18
18
18
18
18
18
4
–
–
d,e]
+
À
T BF (III) (1.2)
43 (26)
88
45
88
4
+
À
T BF (III) (1.5)
4
+
À
T ClO (IV) (1.5)
4
À
+
T OTf (V) (1.5)
+
-À
AcNH-T BF (VI) (1.5)
8
4
+
À
T BF (III) (1.5)
>99 (78)
>99 (77)
75
4
+
À
T BF (III) (1.5)
80
60
80
1
4
1
4
+
À
10
1
T BF (III) (1.5)
4
+
À
1
T BF (III) (1.5)
88
4
[
[
[
[
[
a]
Conditions: 1a (0.2 mmol), oxidant and TMSCHN (2a) in DCM (0.1M) at the desired temperature in a sealed tube.
Conversion of 1a determined by H NMR. Isolated yield in brackets.
Complex mixture.
No reaction observed.
2
b]
1
c]
d]
e]
Reaction in MeCN (TEMPO=2,2,6,6-tetramethyl-piperidine 1-oxyl).
unsubstituted unsaturated 3-benzazepines, which ene bridged compound 1q was also supported without
allow a rich variety of further functionalization of the observing partial demethylenation under the oxidative
double bond, other commercially available diazome- reaction conditions and the structure of 3q was con-
[
18]
thanes were also tested to show the generality of the firmed by X-ray analysis. Substrates with deactivat-
method. Thus, the reaction of 1a with less nucleophil- ing electron-withdrawing groups such as fluoro or
ic diazoacetates such as ethyl (2b), benzyl (2c) and bromo also provided 3-benzazepines 3s and 3t.
tert-butyl (2d) led to a mixture of the corresponding
The synthetic applicability of the developed meth-
isomeric C-1 and C-2 substituted 3-benzazepines 3/3’ odology was demonstrated by the synthesis of the
in good yields (3b/3b’ 1:1.2, 63%; 3c/3c’ 1:1, 68%; 3d/ drug Lorcaserin (Scheme 3). The corresponding
3
d’ 1.6:1, 66%). Next, different substituted THIQ de- THIQ 1u was prepared in a gram scale in three steps
rivatives 1 were explored. The reaction with N-aryl- from cheap p-chlorobenzaldehyde following a litera-
THIQs bearing both electron-donating and electron- ture procedure. The enantioenriched version can be
withdrawing groups proceeded efficiently, providing also accomplished in a longer sequence to achieve
the 3-benzazepines 3e–g in 50–72% yield. Further- (S)-1u in 90% ee. Consequently, our oxidative C–H
more, the reaction tolerated different common N-pro- functionalization/ring expansion methodology was
tecting groups like Boc or Cbz (3h and 3i) or acyl employed, leading to 3-benzazepine 3u. A further
groups like pivaloyl (3j), and it could be scaled-up to one-pot enamine reduction and N-Boc deprotection
[
19]
[
20]
1
.0 mmol with no significant detriment in the final of 3u with Et SiH in trifluoroacetic acid/dichlorome-
3
yield (64 vs. 68%, 3h). Further derivatives bearing thane, followed by treatment with HCl (1M in diox-
bulky carbamates like adamantyl derivatives 3k and ane), provided the targeted derivative Lorcaserin hy-
3
l were obtained in moderate to good yields, whereas drochloride salt in a quantitative yield.
the alkylic substituted amine 3m was formed in Aiming at bringing some light into the mechanism
a lower yield. We next explored the substitution at of the reaction, several experiments were carried out.
the aromatic unit of the THIQ. Because a large Initially, a radical mechanistic pathway for the oxida-
number of 3-benzazepines contains methoxy or oxy- tion step to form the key N-acyl iminium intermediate
genated groups (see for example, Fenoldopam or was discarded by the reaction with a radical clock as
Ivabradine, Figure 1), different oxygenated deriva- indirect methodology to determine a free-radical re-
tives (1n–1r) were next studied. The reaction with the action (see the Supporting Information for more de-
[
21]
6-methoxy-substituted THIQ (3n, 74%), as well as tails).
The fact that no ring-opening by-products
with 6,8-dimethoxy (3p, 50%) and two methoxy were observed suggested a direct hydride abstraction
[
22]
groups at the 6 and 7 positions with both an N-Boc vs. a radical pathway. This different nature of the
3o, 72%) and an N-adamadyl carbamate (3r, 52%) TEMPO salt with respect to other more classical radi-
as protecting groups proceeded smoothly. The methyl- cal oxidants might also be the key for the success in
(
Adv. Synth. Catal. 0000, 000, 0 – 0
3
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
[a,b]
Table 2. Substrate scope.
Scheme 3. Synthetic application: synthesis of the drug Lor-
caserin.
particular, the reactivity of labelled 1h, 1n and 1v
with dissimilar substituents at para position to the re-
active center was studied (Scheme 4). Differences on
reactivity were found, showing a higher reactivity for
the more electron-rich derivative 1n-D2 bearing the
p-MeO substituent, whereas the p-Br substituted
THIQ 1v-D2 was the least reactive. However, only
slightly changes in the final isomeric ratio of the two
possible labelled deuterated compounds were found.
Thus, in all cases a mixture nearly to 70:30 of the re-
gioisomers, presenting the deuterium atom in the
proximal (3) and distal (3’) positions with respect to
the nitrogen moiety, was obtained.
To get a deeper understanding of such pathways,
DFT calculations using the M06 functional and 6-
3
1G(d,p) basis set were performed (see the Support-
[
23]
ing Information for details). In particular, the reac-
tion of 1h, 1n and 1v with TMSCHN was studied.
2
For each substrate, we considered the (R,R) and (S,R)
diasteroisomers, which have been specifically oriented
[
a]
+
À
Conditions:1 (0.2 mmol), T BF (1.5 equiv.) and RCHN2
4
(
1.5–2.4 equiv.) in DCM (0.1M) at 808C for 1–4 h (see
the Supporting Information for each case).
Isolated yield.
Isomeric ratio determined by H NMR.
[
[
[
[
b]
c]
d]
e]
1
1.0 mmol scale.
Concomitant formation of the corresponding N-methy-
lated salts.
this reaction with sensitive diazomethane nucleo-
philes. Additionally, considering the observed sub-
strate-dependent 3:3’ selectivity with substituted di-
azomethanes (Table 2, for example, 3b:3b’ vs. 3l as
single isomer), further experiments with representa-
tive deuterium containing N-Boc precursors were car-
ried out, indicating two possible competitive parallel
mechanisms for the second step with TMSCHN . In Scheme 4. Mechanistic studies with D-labelled THIQs 1.
2
Adv. Synth. Catal. 0000, 000, 0 – 0
4
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
À1
to allow the aromatic C-atom or the N-atom to act as (S,R) isomer, Ea=6.8 kcalmol
for C–C pathway
À1
the nucleophilic center, respectively (Figure 2). For and 8.6 kcalmol for C–N pathway, see Figure 2].
simplicity, the scenario found for the starting material
Interestingly, while the N-attack results in the for-
1h is primarily presented (for 1n and 1v see the Sup- mation of an aziridine intermediate, the cyclopropane
porting Information). First, the (S,R) diastereoisomer analogue (C-attack) has not been found as a minimum
À1
is significantly more stable (ꢀ2 kcalmol ) than the in the potential energy surface. Exhaustive explora-
(
R,R). It is also observed that the activation energies tion of the reaction coordinates indicate that, al-
of the nucleophilic attack are just slightly lower for though a saddle region corresponding to geometries
the initial C–C bond formation [from the most stable close to the cyclopropane putative intermediate is
Figure 2. DFT-calculated reaction pathways for THIQ 1h.
Adv. Synth. Catal. 0000, 000, 0 – 0
5
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
found, an energy minimum cannot be located with Conclusions
such a structure. Therefore, when the initial step con-
sists in the formation of the C–C bond, the system In conclusion, we have developed a metal-free meth-
evolves barrier-less towards the benzazepine product. odology for the direct synthesis of important 3-benz-
In contrast, when the initial step consists in the C–N
ACHTUNGTRENNUaGN zepines from readily available tetrahydroisoquino-
bond formation, an aziridine intermediate is pro- lines (THIQs). Our approach involves the oxidative
duced, which further evolves through a ring expansion formation of an iminium ion intermediate, followed
process to generate the final benzazepine product. by the addition of a diazomethane derivative RCHN2
Such a process from the aziridine intermediate has as nucleophile and subsequent rearrangement and
À1
a significant activation energy of 16–17 kcalmol . ring expansion. Moreover, we have shown two con-
The high thermodynamic stability of the aziridine II current possible operative mechanistic pathways
À1
intermediate (18 kcalmol
more stable than I), based on experimental proof and DFT calculations.
makes very improbable the backward reaction to I, Finally, the synthetic utility of this methodology was
which should overpass the activation energy of proven by the preparation of the 3-benzazepine drug
À1
ꢀ
27 kcalmol . Thus, regardless of the pathway fol- Lorcaserin.
lowed, once the first transition state is overpassed, the
reaction proceeds irreversibly towards the benzaze-
pine product. Therefore, the formation of an aziridine Experimental Section
intermediate does not affect the overall tendency of
the system to evolve through one or the other path- General Methods
way. In fact the comparable values for the initial acti-
vation energies suggest that both pathways are plausi-
ble and that the disappearance of the substrate 1h
through one or the other pathway should follow simi-
1
13
19
H, C and F NMR spectra were recorded in CDCl and
3
13
1
DMSO-d6 (reference signals:
7
vance 300, 400 or 600 MHz instrument. Chemical shifts (d)
H=7.26 ppm,
C=
7.16 ppm, CDCl ) on a Bruker ARX-300 and a Bruker Ad-
3
lar kinetics. Interestingly, while the consumption of 1h are given in ppm and spin-spin coupling constants (J) are
by means of the C–C pathway proceeds to the direct given in Hz. Analytical thin layer chromatography was per-
formation of the product, in the case of the C–N path- formed using silica gel 60 F254 and a solution of KMnO or
4
phosphomolybdic acid served as staining agent. Column
chromatography was performed on silica gel 60 (0.040–
0.063 mm). Exact masses (HR-MS) were recorded on an
Agilent Q-TOF 6540 UHD spectrometer (samples in
way the accumulation of the aziridine intermediate
could result in a slower rate for the benzazepine pro-
[24]
duction. One of the most surprising results of the
mechanistic investigations shown above is the fact
that the occurrence of one or the other pathway is in-
dependent of the electronic nature of the phenyl ring.
A priori, if the reaction pathways were triggered by
CH OH as solvent) using electrospray (ESI) or electron
3
(
EI) ionization techniques. Chiral high pressure liquid chro-
matography (HPLC) analyses were performed on an Agi-
lent 1200 series instrument. CH Cl and Et N were distilled
2
2
3
the initial nucleophilic attack of either the aromatic over CaH . THF and Et O were distilled and dried over Na.
2
2
carbon or the nitrogen center, the substitution on the
aromatic ring should direct the reactivity towards one General Procedure for the Ring Expansion Process
or the other mechanism. In particular, electron-donat-
In a flame-dried pressure Schlenk tube, the TEMPO oxoam-
monium salt T BF (72.9 mg, 0.30 mmol, 1.5 equiv.) and
ing groups should enhance the C–C pathway and elec-
tron-withdrawing groups should tip the scales towards
the C–N mechanism. However, such a tendency was
not observed experimentally. Regarding the analysis
+
À
4
the corresponding tetrahydroisoquinoline
.0 equiv.) were suspended in dry DCM (0.1M) under an
argon atmosphere. Afterwards the nucleophile RCHN (2)
1 (0.20 mmol,
1
2
of the values of natural charges and the geometries in (0.48 mmol, 2.4 equiv.) was added dropwise. Depending on
relevant atoms present in reagents and transition the substrate, the reaction mixture was stirred at 808C for
[
25]
1
h, 4 h or overnight. After evaporating the solvent, the
states, we found that the extrusion of the nitrogen
molecule is the driving force for the formation of the
crude product was purified by flash column chromatography
see the Supporting Information).
(
species III from the intermediate I. Once the N frag-
2
ment is taking place, the formation of the carbocation
and the C- or N-driven nucleophilic attack takes
place in a barrier-less concerted asynchronic process.
In consequence, the modulation of the nucleophilicity
of the aromatic carbon has a minor effect on the reac-
tion pathway that results in the formation of the 3-
benzazepine structures.
Acknowledgements
The Deutsche Forschungsgemeinschaft (DFG) is acknowl-
edged for financial support. The Deutscher Akademischer
Austauschdienst (DAAD) supported this work within the
DAAD-PPP Spanien exchange programme (project
5
7087264). J.A. thanks the MICINN for a “Ramon y Cajal”
contract and the project CTQ2015-64561 (MINECO/
Adv. Synth. Catal. 0000, 000, 0 – 0
6
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
FEDER). We gratefully acknowledge computational time
provided by the Centro de Computaciꢀn Cientꢁfica at the
Universidad Autꢀnoma de Madrid CCC-UAM.
T. Knauber, C.-J. Li, Angew. Chem. 2014, 126, 76;
Angew. Chem. Int. Ed. 2014, 53, 74; g) R. Narayan, K.
Matcha, A. P. Antonchick, Chem. Eur. J. 2015, 21,
1
4678.
11] Selected recent examples: a) C.-J. Li, Acc. Chem. Res.
009, 42, 335; b) A. G. Condie, J. C. Gonzalez-Gomez,
[
2
References
C. R. Stephenson, J. Am. Chem. Soc. 2010, 132, 1464;
c) E. Boess, D. Sureshkumar, A. Sud, C. Wirtz, C.
Fares, M. Klussmann, J. Am. Chem. Soc. 2011, 133,
[
1] Some reviews: a) J. H. Shah, R. M. Hindupur, H. N.
Pati, Curr. Bioact. Compd. 2015, 11, 170; b) J. H. Ryan,
C. Hyland, J. Just, A. G. Meyer, J. A. Smith, C. C. Wil-
liams, Heterocycl. Chem. 2013, 25, 455; c) J.-P. K.
Meigh, Science of Synthesis, (Eds.: S. M. Weinreb, E.
Schaumann), Thieme, Stuttgart, 2004, Vol. 17, p 825.
2] See for example: a) C. C. Tenn, L. P. Niles, J. Pharma-
col. Exp. Ther. 1995, 274, 84; b) D. La Torre, A. Falorni,
Ther. Clin. Risk Manag. 2007, 3, 929; Y. Yu, J. Wang,
P. H. Sun, Y. Guo, Z. Zhang, G. Jin, X. J. Zhen, Neuro-
chemistry 2008, 104, 946; c) S. Sarkar, S. M. Husain, D.
Schepmann, R. Frçhlich, B. Wꢅnsch, Tetrahedron 2012,
8
106; d) E. Boess, C. Schmitz, M. Klussmann, J. Am.
Chem. Soc. 2012, 134, 5317; e) Q.-H. Zheng, W. Meng,
G. J. Jiang, Z.-X. Yu, Org. Lett. 2013, 15, 5928; f) G.
Zhang, X. Ma, G. Cheng, D. Liu, R. Wang, Org. Lett.
2
014, 16, 656; g) C. Min, A. Sanchawala, D. Seidel,
[
Org. Lett. 2014, 16, 2756; h) H. Ueda, K. Yoshida, H.
Tokuyama, Org. Lett. 2014, 16, 4194; i) T. Wang, M.
Schrempp, A. Berndhauser, O. Schiemann, D. Menche,
Org. Lett. 2015, 17, 3982; j) Q. Chen, J. Zhou, Y. Wang,
C. Wang, X. Liu, Z. Xu, L. Lin, R. Wang, Org. Lett.
2
015, 17, 4212.
6
8, 2687.
[
3] Selected examples: a) D. P. Behan, B. M. Smith, C.
Bjenning, WO Patent WO/2006/071740, 2006; b) S. Mo-
hapatra, M. R. Hellberg, Z. Feng, U.S. Patent 0293475,
[12] a) W. Ando, H. Miyazaki, K. Ito, D. Auchi, Tetrahedron
Lett. 1982, 23, 555; b) T. Aoyama, T. Shioiri, Tetrahe-
dron Lett. 1986, 27, 2005; c) T. Nojima, K. Ishiguro, Y.
Sawaki, J. Org. Chem. 1997, 62, 6911.
2
007; c) I. Shimada, K. Maeno, Y. Kondoh, H. Kaku,
K. Sugasawa, Y. Kimura, K. Hatanaka, Y. Naitou, F.
Wanibuchi, S. Sakamoto, S. Tsukamoto, Bioorg. Med.
Chem. 2008, 16, 3309.
[13] T. Stopka, L. Marzo, M. Zurro, S. Janich, E. U.
Wꢅrthwein, C. G. Daniliuc, J. Alemꢁn, O. Garcꢂa Man-
cheÇo, Angew. Chem. 2015, 127, 5137; Angew. Chem.
Int. Ed. 2015, 54, 5049.
[
4] a) B. Smith, J. Smith, 5HT2C receptor modulators. U.S.
Patent 6,953,787, 2005; D. Nikolic, P. P. Toth, A. Ferlita,
V. D. Bartolo, G. Montalto, M. Banach, M. Rizzo, Clin.
Lipidol. 2014, 9, 179; b) Q. Zhu, J. Wang, X. Bian, L.
Zhang, P. Wei, Y. Xu, Org. Process Res. Dev. 2015, 19,
[14] For a review on expansions to 7-membered rings: E. J.
Kantorowski, M. J. Kurth, Tetrahedron 2000, 56, 4317.
[15] Ring expansion of isoquinolium salts with CH N : a) T.
2
2
Kametani, M. S. Premila, S. Hirata, H. Seto, H.
Nemoto, K. Fukumoto, Can. J. Chem. 1975, 53, 3824;
b) for a 2-step synthesis, see: T. Xiao, L. Li, G. Lin, Z.-
W. Mao, L. Zhou, Org. Lett. 2014, 16, 4232; for a ring
expansion of sulfonylimines with diazomethanes: c) A.-
J. Xia, T.-R. Kang, L. He, L.-M. Chen, W.-T. Li, J.-L.
Yang, Q.-Z. Liu, Angew. Chem. 2016, 128, 1463;
Angew. Chem. Int. Ed. 2016, 55, 1441; d) for a recent
Ag-catalyzed 3-component reaction, see: T. Xiao, P.
Peng, Y. Xie, Z.-Y. Wang, L. Zhou, Org. Lett. 2015, 17,
1
263.
[
5] See for example: a) M. B. Murphy, C. Murray, G. D. N.
Shorten, N. Engl. J. Med. 2001, 345, 1548; b) W. C.
Oliver, G. A. Nuttall, K. J. Cherry, P. A. Decker, T.
Bower, M. H. Ereth, Anesth. Analg. 2006, 103, 833.
6] See for example: J. C. Tardif, I. Ford, M. Tendera,
M. G. Bourassa, K. Fox, Eur. Heart J. 2005, 26, 2529.
7] For some recent examples based on a Friedel–Crafts
reaction, see: a) M. Bandini, M. Tragni, A. Umani-
Ronchi, Adv. Synth. Catal. 2009, 351, 2521; b) R. B.
Kargbo, Z. Sajjadi. Hashemi, S. Roy, X. Jin, R. J. Herr,
Tetrahedron Lett. 2013, 54, 2018; c) M. Mizukami, K.
Wada, G. Sato, Y. Ishii, N. Kawahara, S. Nagumo, Tet-
rahedron 2013, 69, 4120.
8] First synthesis of 3-benzazepines: a) K. Dimroth, H.
Freyschlag, Angew. Chem. 1956, 68, 518. See also:
b) M.-Y. Chang, C.-K. Chan, S.-Y. Lin, R.-T. Hsu, Tet-
rahedron 2012, 68, 10272; c) B. W. Burbaum, C. A.
Gilson, S. Aytes, S. A. Estrada, D. Sengupta, B. Smith,
M. Rey, U. Weigl, WO Patent WO 2005019179, 2005.
9] A. ꢆlvarez-Pꢀrez, C. Gonzꢁlez-Rodrꢂguez, C. Garcꢂa-
Yebra, J. A. Varela, E. OÇate, M. A. Esteruelas, C. Saꢁ,
Angew. Chem. 2015, 127, 13555; Angew. Chem. Int. Ed.
[
[
4
332.
[
[
16] For example: a) M. A. Mercadante, C. B. Kelly, J. M.
Bobbitt, L. J. Tilley, N. E. Leadbeater, Nat. Protoc.
2
013, 8, 666; b) O. Garcꢂa MancheÇo, T. Stopka, Syn-
thesis 2013, 45, 1602.
[
17] TEMPO salt oxidants in C–H functionalization: a) S.
Wetz, S. Kodama, A. Studer, Angew. Chem. 2011, 123,
1
1713; Angew. Chem. Int. Ed. 2011, 50, 11511; b) H.
Richter, R. Frçhlich, C. G. Daniliuc, O. Garcꢂa Man-
cheÇo, Angew. Chem. 2012, 124, 8784; Angew. Chem.
Int. Ed. 2012, 51, 8656; c) C. Yan, Y. Liu, Q. Wang,
RSC Adv. 2014, 4, 60075; d) S. Sun, C. Li, P. E. Florean-
cig, H. Lou, L. Liu, Org. Lett. 2015, 17, 1684.
[
[
18] CCDC 1453370 contains the supplementary crystallo-
graphic data for compound 3q in this paper. These data
can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
2
015, 54, 13357.
10] Selected reviews: a) C. J. Scheuermann, Chem. Asian J.
010, 5, 436; b) M. Klussmann, D. Sureshkumar, Syn-
[
2
thesis 2011, 353; c) C. S. Yeung, V. M. Dong, Chem.
Rev. 2011, 111, 1215; d) C. Liu, H. Zhang, W. Shi, A.
Lei, Chem. Rev. 2011, 111, 1780; e) R. Rohlmann, O.
Garcꢂa MancheÇo, Synlett 2013, 24, 6; f) S. A. Girard,
[19] C. D. Perchonock, J. A. Finkelstein, J. Org. Chem. 1980,
45, 2000.
Adv. Synth. Catal. 0000, 000, 0 – 0
7
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
[
20] See the Supporting Information for the detailed synthe-
sis of (S)-1u.
21] The formation of a mixture of THIQ, product and the
evolved unavoidably to the III,C-pathway suggesting
that, after the N-attack, a migration of the TMS group
to the benzylic carbon takes place. Thus, both pathways
end with the same energy.
[
corresponding N-methylated substances was observed
1
by H NMR. The low yield obtained with N-butyl-
[25] The values of natural charges in relevant atoms in both
the reagents and the TS were evaluated. The C or the
N atoms involved in the nucleophilic attack did not
THIQ 1m can also be attributed to N-methylation pro-
cesses.
+
[
22] T. A. Hamlin, C. B. Kelly, J. M. Ovian, R. J. Wiles, L. J.
Tilley, N. E. Leadbeater, J. Org. Chem. 2015, 80, 8150.
23] a) Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2006, 120,
present significant differences, whereas the C–N2 frag-
ment showed the main electronic rearrangement. Fur-
thermore, an increase in the electron density of atoms
in the diazo group and a decrease of the adjacent C-
atom was observed (see the Supporting Information).
[
2
15; b) A. V. Marenich, C. J. Cramer, D. G. Truhlar, J.
Phys. Chem. B 2009, 113, 6378.
[
24] A minimum in the potential surface for the 7-ring
III,N-pathway could not be found. The optimization
Adv. Synth. Catal. 0000, 000, 0 – 0
8
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
Synthesis of 3-Benzazepines by Metal-Free Oxidative C–H
Bond Functionalization–Ring Expansion Tandem Reaction
9
Adv. Synth. Catal. 2016, 358, 1 – 9
Andrea Gini, Julia Bamberger, Javier Luis-Barrera,
Mercedes Zurro, Rubꢀn Mas-Ballestꢀ, Josꢀ Alemꢁn,*
Olga Garcꢂa MancheÇo*
Adv. Synth. Catal. 0000, 000, 0 – 0
9
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!