Novel succinct routes to quinoxalines and 2-benzimidazolylquinoxalines via the Ugi reaction

Article abstract of DOI:10.1016/j.tetlet.2014.04.013

This Letter reveals a unique, user-friendly, concise two-step, one-pot protocol for the synthesis of highly substituted quinoxalines. Conducting the Ugi reaction with appropriately functionalized classical Ugi reagents with subsequent acid treatment of the Ugi adduct affords collections of diversified quinoxalines in good to excellent yields. The methodology exploits what may be viewed as a 'convertible carboxylic acid', which in addition may be captured in an intramolecular sense to generate structurally complex 2- benzimidazolylquinoxalines in a mere two steps.

Full text of DOI:10.1016/j.tetlet.2014.04.013

Tetrahedron Letters xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Novel succinct routes to quinoxalines and
2-benzimidazolylquinoxalines via the Ugi reaction
⇑
Muhammad Ayaz, Zhigang Xu, Christopher Hulme
College of Pharmacy, BIO5 Oro Valley, The University of Arizona, 1580 E. Hanley Blvd, Oro Valley, AZ 85737, USA
Department of Chemistry and Biochemistry, The University of Arizona, Tucson, AZ 85721, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
This Letter reveals a unique, user-friendly, concise two-step, one-pot protocol for the synthesis of highly
substituted quinoxalines. Conducting the Ugi reaction with appropriately functionalized classical Ugi
reagents with subsequent acid treatment of the Ugi adduct affords collections of diversified quinoxalines
in good to excellent yields. The methodology exploits what may be viewed as a ‘convertible carboxylic
acid’, which in addition may be captured in an intramolecular sense to generate structurally complex
2-benzimidazolylquinoxalines in a mere two steps.
Received 22 March 2014
Revised 3 April 2014
Accepted 4 April 2014
Available online xxxx
Keywords:
Ugi reaction
Ó 2014 Elsevier Ltd. All rights reserved.
Post-condensation modifications
Quinoxaline
Benzimidazole
Acyl transfer
Isocyanide based multi-component reactions (IMCRs) are
viewed in many circles as a premier field of study for the genera-
tion of both new chemotype diversity and preferred methodologies
to afford known heterocycles.1 Specifically, the venerable Ugi reac-
tion, Scheme 1, has proved extremely versatile in enabling access
to libraries of small molecules through a variety of strategies that
encompass post-condensation modifications of the Ugi adduct
and exploitation of the ‘diversity of nucleophiles’ able to trap out
the intermediate ‘nitrilium ion’ of the Ugi reaction in both intra-
and intermolecular modalities. Indeed, the collection of reactions
that emanate from the original version of the Ugi transformation
have been exploited for the synthesis of chemical libraries since
the early 1990s, in-line with the emergence of high-throughput
screening.2 Efforts by several groups have seen concise methodol-
ogies developed that enable access to diazepines,3 ketopipera-
zines,4 imidazolines,5 b-lactams,6 and hydantoins to name but a
few.7 As such, the synthetic routes fall into the realm of UDC meth-
odology (Ugi/Deprotect/Cyclize) where the final ring closing event
is mediated through amide bond formation.1e,f This Letter details
studies directed toward the preparation of functionalized quinox-
alines that constitute an important class of nitrogen containing
heterocycles utilized in many sectors of the chemical industry.8,9
Unfortunately, reported methods to synthesize quinoxalines suffer
from various drawbacks, most noticeably harsh reaction conditions
O
O
O
O
N
R2 NH2
R4 NC
H
R2
H
R2
H
R1
OH
N
R1
R3
OH
H
O
H
R1
R4
N
O
R3
R3
O
O
R2
R2
R3
Mumm
Rearangement
R3
R1
O
N
HN
R1
O
R2
O
R3
N
H
H
R4
N
H
R4
R4
R1
N
HN
Scheme 1. Mechanism of the four-component Ugi reaction.
and use of toxic metals.10 The most widely used protocol11 for
quinoxaline preparation is limited to the trivial condensation of
1,2-diamines with 1,2-dicarbonyl reagents or acyl halides; a strat-
egy allowing only two points of diversity. Recently we have
reported12 the synthesis of tetrazole-embedded quinoxalinones
using the TMS-azide version of the Ugi reaction, and herein, this
Letter reveals13 an alternate user-friendly methodology comprised
of concise two-step one-pot procedures for the synthesis of highly
functionalized quinoxalines 6, utilizing the classical Ugi reaction
and an embedded amino-nucleophile to generate pre-requisite
diversity elements.
Thus, a model reaction using benzoic acid 1, 3-bromophenyl
glyoxaldehyde 2, N-Boc-(4,5-dimethyl)1,2-phenylenediamine 3,
and commercially available n-butylisonitrile 4, proceeded
⇑
Corresponding author. Tel.: +1 520 626 5322.
E-mail address: hulme@pharmacy.arizona.edu (C. Hulme).
http://dx.doi.org/10.1016/j.tetlet.2014.04.013
0040-4039/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Ayaz, M.; et al. Tetrahedron Lett. (2014), http://dx.doi.org/10.1016/j.tetlet.2014.04.013

2
M. Ayaz et al. / Tetrahedron Letters xxx (2014) xxx–xxx
O
O
N
O
O
O
O
N
O
Ph
NHBu
Br
N
Ph
NHBu
20% TFA/DCE
MW, 140 °C, H2O,
20 min
Ph
OH
NHBu
Br
1
O
2
MeOH, rt,
18 h
NHBoc
O
N
NHBoc
H
O
Br
6a, 46%
5
NC
NHBoc
Br
NH2
4
3
5, 70%
Scheme 2. Four component one-pot Ugi reaction.
Ph
O
O
O
N
O
O
Ph
NHBu
N
N
smoothly in methanol at ambient temperature giving the Ugi
adduct 5 in 70% yield (Scheme 2). Intriguingly, simple TFA treat-
ment of 5 delivers 6a in good yield, representing the first reported
synthesis of quinoxalines derived in one step from the Ugi reaction,
postulated to proceed in unexpected fashion through intermedi-
ates 7 and 8 with loss of a benzoyl group.3 Interestingly, the final
product is significantly different from its Ugi precursor and as such,
the methodology provides a unique opportunity to rapidly access
such diversity space. During our studies, it was found that the
cyclization step required optimization and different concentrations
of TFA were thus investigated. The optimal yield for quinoxaline 6a
was obtained using a 20% TFA/DCE solution in the presence of a
catalytic amount of water, with microwave irradiation at 140 °C
in 20 min (Scheme 2).
NHBu
Br
- H2O
NH2
Br
8
7
Scheme 3. Conversion of Ugi adduct into quinoxaline 6a via intermediates 7 and 8.
O
R2
R3
NH2
1. MeOH, rt, 18 h
O
H
OH
O
2
2. 20% TFA/DCE, H2O
MW 140 °C, 20 min
R
N
N
NHR4
1f
NHBoc
3
R3
R1
O
R4NC
R1
6a-6i (39 to 49%)
2
4
Interested by the final loss of the benzoyl moiety from interme-
diate 8 and in search of a potentially milder and more atom-eco-
nomic transformation, a number of carboxylic acids of different
strengths and sizes were thus evaluated for effects on isolated
yield of 6a (Table 1).
Products 6a-6i:
I
O
O
N
O
N
N
N
NHBu
NHBu
H
N
N
N
6c
, 47%
6a
, 48%
6b
, 41%
F
Study of a limited set of aromatic acids (entries 1 through 5)
revealed 2-N-Boc-aminophenyl proved optimal (73%, 6a) and one
could rationalize this observation through carbonyl-activation via
internal hydrogen bonding of the adjacent aniline. Entries 3 and
5 afforded more moderate yields of 6a, whereas 2 and 4 were
deemed incompatible with the initial condensation. In search for
optimal atom-economy, formic acid was employed and encourag-
ingly for uptake in the organic community, quinoxaline 6a was iso-
lated after acid treatment in 69% yield (entry 6), proving to be our
acid of choice. To further simplify the procedure, 20% TFA/DCE was
added directly to the Ugi adduct in methanolic solution and the
resulting mixture heated via microwave irradiation at 140 °C for
20 min (entry 7, Table 1). To our delight, the one pot procedure
Br
O
O
O
O
N
N
N
N
NHBu
N
H
N
H
N
N
6d
,49%
6e, 48%
6f,39%
F
OCF3
NHBu
F
Br
O
O
N
N
N
N
N
N
NHBu
N
H
6g
6h
6i
, 42%
F
, 45%
, 44%
F
F
OCF3
Scheme 4. Study of the reaction scope of optimized methodology.
Table 1
Effect of carboxylic acids on quinoxaline yield
NC
NHBoc
O
O
O
R3
O
O
Br
NHBu
R2
9
R4
NHBoc
O
Br
O
R1
OH
2
1
R
OH
O
MeOH
rt, 18 h
R
N
O
MeOH
rt, 18 h
NHBoc
R1
O
20% TFA/DCE
MW, 140 °C,
H2O, 15 min
1a - 1f
2
4
R4
R3
NH2
BocHN
N
O
6a
N
H
NH2
O
NC
NHBoc
R2
10
NHBoc
3
3
5a - 5f
TFA/DCE,MW
140 °C
TFA/DCE
rt, 72 h
TFA/DCE
rt, 72 h
20 min
N
R1
R4
R3
N
N
Entry
R
Yields (5a–5f)
Yield (6a)
N
H
N
N
N
N
HN
O
1
2a
3
2-N-Boc aminophenyl
2-Methoxyphenyl
4-Trifuoromethylphenyl
2,4-Dimethylaminophenyl
2,6-Difluorophenyl
H
H
47% (5a)
nd (5b)
47% (5c)
nd (5d)
51% (5e)
49% (5f)
nd (5f)
73%
nd
58%
nd
57%
69%
43%
R2
O
TFA/DCE
R2
R2
11a-11c
44-47%)
MW, 140 °C
R1
N
20 min
4a
5
N
N
13
R4
R4
6
7
12a-12f
37-46%
R3
R3
a
Complex mixture during initial MCR, yield not determined.
Scheme 5. Synthesis of 2-benzimidazolylquinoxalines 11.
Please cite this article in press as: Ayaz, M.; et al. Tetrahedron Lett. (2014), http://dx.doi.org/10.1016/j.tetlet.2014.04.013

M. Ayaz et al. / Tetrahedron Letters xxx (2014) xxx–xxx
3
translated into higher overall yield (43%) when compared to that of
the two step process (34%, entry 6) (Scheme 3).
Exploring the scope of the reaction, various aldehydes, dia-
mines, and isonitriles were used to furnish a set of diversified quin-
oxalines and gratifyingly, the transformation worked equally well
for all inputs, (Scheme 4).
Seeking to expand accessible skeletal diversity using the same
reaction conditions, it was envisioned that use of isonitrile 9
(ortho-N-Boc-phenylisonitrile)14 would facilitate access to 2-benz-
imidazolylquinoxalines, 11. The Ugi reaction proceeded smoothly
and upon microwave irradiation of Ugi products 10 with 20%
TFA/DCE, 2-benzimidazolylquinoxalines 11a–11c were obtained
in good yield (Scheme 5). Considering the well documented medic-
inal utility of benzimidazoles and quinoxalines, these tethered
combinations of the two scaffolds afford new opportunities to
probe their biological activity.
N
N
N
N
O
NO2
12d
N
N
N
N
O
12f
Br
Figure 2. X-ray crystal structures of compounds 12d and 12f.14
Acknowledgment
The authors thank the NIH (P41GM08619001) for financial sup-
port and S.A. Roberts for X-ray analyses.
During these studies an interesting series of 2-benzimidazolyl-
quinoxaline products 12a–12f were observed when utilizing bulky
ortho-substituted aromatic acids which had retained the benzoyl
group through an internal acyl-transfer from the nitrogen derived
from the original Ugi amine input to the adjacent benzimdazole
(Fig. 1). To the best of our knowledge this Nitrogen to Nitrogen acyl
transfer has no literature precedent. The closest correlation if any,
would be the use of imidazole based catalysts that are used to
expedite oxygen to oxygen or oxygen to nitrogen acyl transfer
reactions.15 Interestingly this behavior was observed only when
ortho-Br, ortho-I, or ortho-NO2-benzoic acid was employed
(Scheme 5). This peculiar acyl transfer may be partially explained
by the crystal structures of 12d and 12f respectively (Fig. 2), where
the ortho-subsituted aryl group is seemingly positioned to enjoy a
References and notes
1. (a) Dömling, A. Chem. Rev. 2006, 106, 17–89; (b) Hulme, C.; Dietrich, J. Mol.
Diversity 2009, 13, 195–207; (c) El Kaim, L.; Grimaud, L. Tetrahedron 2009, 65,
2153–2171; (d) Banfi, L.; Riva, R.; Basso, A. Synlett 2010, 23–41; (e) Hulme, C.;
Gore, V. Curr. Med. Chem. 2003, 44, 1947–1950; (f) Xu, Z.; Ayaz, M.; Capelli, A.;
Hulme, C. ACS Comb. Sci. 2012, 14, 460–464; (g) Xu, Z.; De Moliner, F.; Capelli, A.
P.; Hulme, C. Angew. Chem., Int. Ed. 2012, 51, 8037–8040.
2. Armstrong, R. W., Regents of the University of California, USA. 1995, p 91.
3. (a) Hulme, C.; Peng, J.; Tang, S. Y.; Burns, C. J.; Morize, I.; Labaudiniere, R. J. Org.
Chem. 1998, 63, 8021–8023; (b) Huang, Y.; Khoury, K.; Chanas, T.; Doemling, A.
Org. Lett. 2012, 14, 5916–5919.
4. (a) Hulme, C.; Peng, J.; Louridas, B.; Menard, P.; Krolikowski, P.; Kumar, N. V.
Tetrahedron Lett. 1998, 39, 8047–8050; (b) Hulme, C.; Peng, J.; Morton, G.;
Salvino, J. M.; Herpin, T.; Labaudiniere, R. Tetrahedron Lett. 1998, 39, 7227–
7230.
5. Hulme, C.; Ma, L.; Romano, J.; Morrissette, M. Tetrahedron Lett. 1999, 40, 7925–
7928.
planar p–p
interaction with the quinoxaline moiety.16
In summary, we have devised a simple, mild, robust, and high
yielding one-pot protocol to produce quinoxalines and tethered
quinoxaline–benzimidazoles.17,18 The products have the potential
to be further decorated and constrained using additional functional
groups. Moreover, the generic nature and simplicity of the protocol
renders it highly attractive as a valuable addition to the repertoire
of synthetic transformations used by the organic and medicinal
chemist respectively.
6. Hulme, C.; Ma, L.; Cherrier, M. P.; Romano, J. J.; Morton, G.; Duquenne, C.;
Salvino, J.; Labaudiniere, R. Tetrahedron Lett. 2000, 41, 1883–1887.
7. (a) Hulme, C.; Ma, L.; Romano, J. J.; Morton, G.; Tang, S.-Y.; Cherrier, M.-P.; Choi,
S.; Salvino, J.; Labaudiniere, R. Tetrahedron Lett. 2000, 41, 1889–1893; (b)
Pakornwit, S.; Krasivan, M. Tetrahedron Lett. 2014, 55, 2299–2303.
8. (a) Husain, A.; Madhesia, D. J. Pharm. Res. 2011, 4, 924–929; (b) Noorulla, S.;
Sreenivasulu, N. Int. J. Pharm. Sci. Res. 2011, 2, 2337–2342; (c) Patidar, A. K.;
Jeyakandan, M.; Mobiya, A. K.; Selvam, G. Int. J. Pharm. Tech. Res. 2011, 3, 386–
392; (d) Singh, D. P.; Hashim, S. R.; Singhal, R. G. Int. J. Drug Dev. Res. 2010, 2,
810–815; (e) Taher, A.; Khalil, N. A.; Ahmed, E. M. Bull. Fac. Pharm. (Cairo Univ.)
2009, 47, 19–26.
9. Heravi, M. M. ARKIVOC 2006, 16, 16–22.
10. (a) Barluenga, J. A.; Liz, F.; Cabal, R.; Maria-Paz Synthesis 1985, 313–314; (b)
Petukhov, P. A.; Tkachev, A. V. Tetrahedron 1997, 53, 9761–9768; (c) Zhao, Z.;
Wisnoski, D. D.; Wolkenberg, S. E.; Leister; Wang, Y.; Lindsley, C. W.
Tetrahedron Lett. 2004, 45, 4873–4876.
11. (a) Bachhav, H. M.; Bhagat, S. B.; Telvekar, V. N. Tetrahedron Lett. 2011, 52,
5697–5701; (b) Bendale, A. R.; Kotak, D.; Damahe, D. P.; Narkhede, S. P.; Jadhav,
A. G.; Vidyasagar, G. Der. Chemica Sinica. 2011, 2, 20–24; (c) Cuenca, A.; Pérez,
S.; Yepez, A.; Paredes, L.; Montecinos, L.; Llovera, L.; Rodríguez, C. J. Heterocycl.
Chem. 2008, 45, 1199–1201; (d) Hasaninejad, A.; Zare, A.; Mohammadizadeh,
M. R.; Karami, Z. J. Iran. Chem. Soc. 2009, 6, 153–158; (e) Jafarpour, M.;
Rezaeifard, A.; Danehchin, M. Appl. Catal. A 2011, 394, 48–51; (f) Jafarpour, M.;
Rezaeifard, A.; Heidari, M. Lett. Org. Chem. 2011, 8, 202–209; (g) Remli, M.; Ayi,
A. I.; Guedj, R. J. Fluorine Chem. 1989, 44, 15–23.
O2N
I
Br
O
O
O
N
N
N
N
N
N
N
N
N
N
Ph
N
Ph
N
Ph
12b, 39%
12a, 37%
12c, 46%
12. Gunawan, S.; Nichol, G.; Hulme, C. Tetrahedron Lett. 2012, 53, 1664–1667.
13. These results were originally outlined in the following poster: Ayaz, M.; Xu, Z.;
Hulme, C. Abstracts of Papers, 241st National Meeting of the American
Chemical Society, Anaheim, CA, March 27–31, 2011, ORGN 584.
14. Gunawan, S.; Ayaz, M.; De Moliner, F.; Frett, B.; Kaiser, C.; Patrick, N.; Xu, Z.;
Hulme, C. Tetrahedron 2012, 68, 5606–5611.
15. Alan, S. A.; Spivey, C. Top. Curr. Chem. 2010, 291, 223–280.
16. CCDC 992230 (12d) and CCDC 992449 (12f) contain the supplementary
crystallographic data for this paper. This data can be obtained free of charge
from The Cambridge Crystallographic Data Center via www.ccdc.cam.ac.uk/
data_request/cif.
I
O2N
Br
O
O
N
N
O
N
N
N
N
N
N
N
N
N
N
12e, 39%
12f, 46%
12d, 37%
N
N
N
N
N
N
N
N
N
H
N
H
17. General one-pot procedure exemplified for the synthesis of 6a: The Ugi reaction
was carried out at 0.25 mmol scale. A 10 mL microwave vial, charged with a
H
N
N
magnetic
bromophenylglyoxaldehyde
bar
was
loaded
with
and
equimolar
tert-butyl
amounts
of
3-
(2-amino-4,5-
11c, 47%
11b, 44%
11a, 46%
dimethylphenyl)carbamate. Using ꢀ0.5 mL of methanol as solvent, the
reaction was stirred at room temperature for two minutes followed by the
addition of formic acid (0.25 mmol) and n-butyl isocyanide (0.25 mmol)
respectively. The progress of the reaction was monitored by TLC and LC/MS (UV
214 nm). After completion of the reaction (ꢀ18 h), solvent was removed under
R1
= 2-NO2Ph, 2-Br-Ph, 2-I-Ph
Figure 1. Array of products derived from Scheme 4.
Please cite this article in press as: Ayaz, M.; et al. Tetrahedron Lett. (2014), http://dx.doi.org/10.1016/j.tetlet.2014.04.013

4
M. Ayaz et al. / Tetrahedron Letters xxx (2014) xxx–xxx
reduced pressure and without any intermediary purification, 0.5 mL of 20%
acetate/hexane as eluent (0–100% ethyl acetate over a period of 30 min). For
the preparation of 11a: the Ugi product 10a was treated with 20% TFA/DCE
(0.5 mL) and irradiated at 140 °C for 20 min. Solvent was evaporated in vacuo
and crude material purified using a CombiFlashÒ Rf with ethyl acetate/hexane
as eluent (0–100% ethyl acetate over a period of 30 minutes). (2-(6,7-dimethyl-
3-phenylquinoxalin-2-yl)-1H-benzo[d]imidazol-1-yl)(2-
TFA/DCE solution was added to the crude Ugi product along with a single drop
of water. The reaction was subsequently heated in a microwave at 140 °C for
20 min (Biotage Initiator60™). After cooling the reaction to room temperature,
the solvent and excess TFA were evaporated in vacuo. The final product (6a)
was purified with a CombiFlashÒ Rf using ethyl acetate/hexane as eluent (0–
100% ethyl acetate over a period of 20 min).
nitrophenyl)methanone (12a). Thick brown oil, Yield 37% (46.2 mg). 1H NMR
3-(3-Bromophenyl)-N-butyl-6,7-dimethylquinoxaline-2-carboxamide (6a). White
solid, mp 136–138 °C. Yield 48% (46.1 mg). 1H NMR (400 MHz, CDCl3): d 8.14
(dd, J = 16.8, 7.5 Hz, 2H), 7.91–7.78 (m, 3H), 7.59 (d, J = 6.3 Hz, 3H), 7.33 (t,
J = 7.9 Hz, 1H), 3.47 (dd, J = 13.6, 6.7 Hz, 2H), 1.71–1.57 (m, 2H), 1.50–1.35 (m,
2H), 0.97 (t, J = 7.3 Hz, 3H). 13C NMR (100 MHz, CDCl3): d 164.31, 151.88,
144.59, 141.85, 140.40, 139.20, 131.98, 131.82, 131.64, 130.79, 129.43, 129.32,
129.10, 127.54, 122.20, 39.59, 31.55, 20.15, 13.79. HRMS (EI+) calculated for
(400 MHz, CDCl3): d 7.98 (d, J = 7.2 Hz, 1H), 7.93–7.86 (m, 3H), 7.60 (d,
J = 8.3 Hz, 2H), 7.47–7.43 (m, 2H), 7.42–7.38 (m, 2H), 7.34 (t, J = 7.4 Hz, 2H),
7.22 (d, J = 7.8 Hz, 1H), 6.82 (s, 1H), 6.51 (s, 1H), 2.54 (s, 6H). 13C NMR
(100 MHz, CDCl3): 165.06, 152.18, 151.29, 143.84, 143.06, 142.06, 140.95,
139.41, 137.97, 135.08, 133.47, 132.68, 132.32, 129.63, 129.42, 129.06, 128.73,
128.48, 128.28, 127.31, 125.83, 125.10, 121.43, 120.38, 113.36, 20.53. (EI+)
calculated for C30H22N5O3 500.1723 [M+H+] Found 500.1725.
2-(1H-Benzo[d]imidazol-2-yl)-3-phenylquinoxaline (11a).White solid, mp 281–
283 °C. Yield 46% (37.0 mg). 1H NMR (400 MHz, CDCl3): d 10.23 (s, 1H), 8.21–
8.16 (m, 2H), 7.85–7.80 (m, 2H), 7.74–7.68 (m, 3H), 7.49 (dt, J = 8.8, 4.8 Hz, 4H),
7.31 (s, 1H), 7.23 (d, J = 7.2 Hz, 1H). 13C NMR (100 MHz, CDCl3): d 153.63,
149.01, 144.29, 141.73, 140.40, 138.59, 133.42, 131.09, 130.51, 129.69, 129.44,
129.30, 129.00, 128.16, 124.51, 122.61, 121.27, 110.94. HRMS (EI+) calculated
for C21H15N4 323.1291 [M+H+] Found 323.1294.
C
19H19BrN3O 384.0706 [M+H+]. Found 384.0702.
18. General one-pot procedure for the synthesis of 11a and 12a: Equimolar amounts
of phenylglyoxaldehyde, tert-butyl (2-amino-4,5-dimethylphenyl)carbamate,
2-nitrobenzoic acid, and tert-butyl (2-isocyano-4,5-dimethylphenyl)carbamate
were stirred at room temperature for 18 h. The Ugi adduct 10a was purified,
treated with 20% TFA/DCE (0.5 mL), and allowed to stir at rt for 3 days.
Compound 12a was subsequently purified using a CombiFlashÒ Rf with ethyl
Please cite this article in press as: Ayaz, M.; et al. Tetrahedron Lett. (2014), http://dx.doi.org/10.1016/j.tetlet.2014.04.013