Iodine(III)-catalyzed rearrangements of imides: A versatile route to α,α-dialkylated α -hydroxy carboxylamides

Article abstract of DOI:10.1002/chem.201405888

A tertiary hydroxy group a to a carboxyl moiety comprises a key structural motif in many bioactive substances. With the herein presented metal-free rearrangement of imides triggered by hypervalent λ³-iodane, an easy and selective way to gain access to such a compound class, namely α,α-disubstituted-α-hydroxy carboxylamides, was established. Their additional methylene bromide side chain constitutes a useful handle for rapid diversification, as demonstrated by a series of further functionalizations. Moreover, the in situ formation of an iodine(III) species under the reaction conditions was proven. Our findings clearly corroborate that hypervalent λ³-benziodoxolones are involved in these organocatalytic reactions.

Full text of DOI:10.1002/chem.201405888

DOI: 10.1002/chem.201405888
Communication
&
Synthetic Methods
Iodine(III)-Catalyzed Rearrangements of Imides: A Versatile Route
to a,a-Dialkylated a-Hydroxy Carboxylamides
Anna Ulmer,[a] Maciej Stodulski,[b] Stefanie V. Kohlhepp,[a] Christoph Patzelt,[a]
Alexander Pçthig,[a] Wolfgang Bettray,[b] and Tanja Gulder*[a, b]
substituted-a-hydroxy carboxylamides 7, which can easily be
further modified to a wide range of important products. In ad-
Abstract: A tertiary hydroxy group a to a carboxyl moiety
comprises a key structural motif in many bioactive sub-
dition, experimental evidence for the formation of chemically
stances. With the herein presented metal-free rearrange-
ment of imides triggered by hypervalent l3-iodane, an
sensitive bromo-imine benziodoxolones 18 as active species is
provided for the first time.
easy and selective way to gain access to such a compound
As part of our research program aiming at the development
class, namely a,a-disubstituted-a-hydroxy carboxylamides,
of new catalytic halogenation methods, we showed that aryl
was established. Their additional methylene bromide side
amides 2 are converted easily into either the dibrominated
product 1[4] or the oxoindole 4,[5] depending on the structure
of the iodobenzamide catalyst (5 and 6; Scheme 1). Encour-
chain constitutes a useful handle for rapid diversification,
as demonstrated by a series of further functionalizations.
Moreover, the in situ formation of an iodine(III) species
under the reaction conditions was proven. Our findings
clearly corroborate that hypervalent l3-benziodoxolones
are involved in these organocatalytic reactions.
Hypervalent l3-iodanes have received huge attention over the
last years and proofed themselves as valuable and reliable or-
ganic alternatives to common transition-metal compounds
owing to their low toxicity and mild reactivity.[1] Besides their
application in simple electron-transfer reactions, iodine(III)
compounds have recently also entered other fields of organic
chemistry, in particular atom-transfer[1b–l] and rearrangement re-
actions.[2] In contrast to their broad synthetic potential, the use
of iodine(III) species, especially in large-scale processes, is still
rare. This discrepancy can be attributed to the low atom econ-
omy of iodine(III) reagents and their often low chemical stabili-
ty. The necessity to overcome this bottleneck of reagent-based
iodine(III) chemistry has fueled intense research in the develop-
ment of the catalytic utilization of these versatile com-
pounds.[1j,3] However, the search for selective external oxidants
turned out to be difficult and thus often limits the success of
iodine-catalyzed reactions.[3b] Herein, we describe the rear-
rangement of imides 3 catalyzed by o-iodobenzamides 6 and
N-bromosuccinimide (NBS) as the oxidant. This synthetic path-
way gives access to very useful building blocks, that is, a,a-di-
Scheme 1. Reaction scope of our iodine(III)-catalyzed transformations of aryl
amides. [a] A ratio of ꢀ1:3 (1/4) was observed.
aged by these results, we turned our focus towards the explo-
ration of the reaction scope of this catalytic system. We started
by replacing the N-alkyl moiety in 2 by an additional carboxyl
functionality, thus significantly changing the conformation of
the starting material.[6] To our surprise, imide 3a was converted
more rapidly (<5 min) than 2 under these conditions. Instead
of the expected oxoindole, the open a,a-disubstituted-a-hy-
droxy carboxyl amide 7a was isolated.
As these compounds comprise an intriguing but synthetical-
ly challenging scaffold, which can be found in many bioactive
natural products and medically important agents, such as the
anticancer drug bicalutamide (8), we set out to evaluate the
optimum reaction conditions by using imide 3b as a model
substrate (Table 1). By employing the standard conditions for
halocarbocyclization,[5,7] amide 7b was isolated in a very good
yield of 79% (entry 1). Control experiments without catalyst 6
[a] A. Ulmer, S. V. Kohlhepp, C. Patzelt, Dr. A. Pçthig, Dr. T. Gulder
Department Chemie and Catalysis Research Center (CRC)
Technische Universitꢀt Mꢁnchen
Lichtenbergstrasse 4, 85747 Garching (Germany)
E-mail: tanja.gulder@tum.de
Homepage: http://www.halogene.ch.tum.de
[b] Dr. M. Stodulski, Dr. W. Bettray, Dr. T. Gulder
Institute of Organic Chemistry, RWTH Aachen University
Landoltweg 1, 52056 Aachen (Germany)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201405888.
Chem. Eur. J. 2015, 21, 1444 – 1448
1444
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
mation. Reactions conducted with the simpler and cheaper
iodobenzene (10) as catalyst generally showed extended reac-
tion times (entries 14–16). In addition, the conversion of sub-
strates equipped with trisubstituted double bonds, such as
starting materials derived from tiglic, angelic, and a-methyl
cinnamic acid (!7m–7q, compare Scheme 2), with 10 mol%
of iodobenzene (10) as the catalyst occurred to be very slow
and did not reach full conversion, even after 16 h (not shown).
In all entries, oxazoline-4-one hydrate 9 was detectable in the
crude mixture in varying amounts (see the Supporting Infor-
mation), a reaction intermediate that was fully converted into
7b during workup.
Having the optimum conditions evaluated, we tested the
scope and generality of the developed method. The rearrange-
ment to 7 occurred with complete regioselectivity in very
good yields (67%–98%), proceeding exclusively via a 5-exo-tet
cyclization mode (17 examples, Scheme 2), which is unusual
for 1,5 dienes.[12]
Table 1. Optimization of the reaction conditions for the iodine(III)-cata-
lyzed rearrangement of imide 3b.
Entry
Catalyst
Solvent
Br+
Additive
t [min]
Conv. [%][a]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
6a[b]
–
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH3CN
CH3CN
HFIP
HFIP
HFIP
HFIP
HFIP
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
11
NH4Cl
–
NH4Cl
TFA
TFA
–
NH4Cl
–
5
60
60
5
20
60
5
120
5
5
160
240
120
30
300
600
120
>99 (79)[c]
–
<10
>99
>99
80
>99
>99
>99
>99 (85)[c]
>99
>99
–
>99
>99
>99
72
–
6a[b]
6b[b]
6a[b]
6a[b]
6b[b]
6a[b]
6a[b]
6a[d]
6a[e]
–
NH4Cl
–
–
–
–
–
–
–
–
10[b]
10[d]
10[e]
–
HFIP
HFIP
HFIP
HFIP
For details on the reaction conditions, see the Supporting Information.
[a] Determined by 1H NMR spectroscopy of the crude mixture using an in-
ternal standard. [b] 10 mol% catalyst was used. [c] Isolated yield.
[d] 1 mol% catalyst was used. [e] 0.1 mol% catalyst was used. NBS=N-
bromosuccinimide, HFIP=1,1,1,3,3,3-hexafluoro-2-propanol.
showed no or only trace amounts of 7b, even after prolonged
reaction times (entries 2 and 3).[8] Brønsted acids as additives,
such as trifluoroacetic acid (TFA), also triggered a rapid conver-
sion of 3b if the glycine derivative 6a (see Scheme 1) was
used as catalyst (entry 4). When applying the ester analog 6b,
a slightly prolonged reaction time was necessary (entry 5),
a trend that was observed likewise in other solvents (entries 7
and 8; see the Supporting Information). In contrast to the bro-
mocarbocyclization,[5] no additive was needed to achieve high
conversion rates during the rearrangement of imide 3b
(entry 6). In the subsequent solvent screening it was found
that the yields were higher in polar, non-nucleophilic solvents,
giving the best results when hexafluoro-2-propanol (HFIP) was
used (entries 9 and 10), a solvent that is known to be superior
in hypervalent-iodine catalysis.[9]
Scheme 2. Substrate scope for the rearrangement of imides 3 with NBS. For
details on the reaction conditions, see the Supporting Information. [a] Isolat-
1
ed yield. [b] Determined by H NMR spectroscopy from the crude mixture.
[c] R1 =OtBu. [d] Derived from the angelic acid derivative of 3. [e] Derived
from the tiglic acid derivative of 3.
As the transformation in HFIP is very fast (<5 min), a lower
catalyst loading was examined. Reducing the amount of cata-
lyst 6a down to 0.1 mol% resulted in significantly longer reac-
tion times of up to 240 min to reach full conversion (entries 11
and 12). With 1.1 equivalent bromo benziodoxole 11[10] instead
of NBS together with 6 as the halogen-transferring species, ex-
tended reaction times were needed to form 7b (entry 17),
which is in line with the structure-dependent reactivity ob-
served earlier.[4,11] However, this result shows that bromo l3-io-
danes are, in principle, able to perform the described transfor-
It is worth noting that no aromatic bromination of the sub-
strates was observed. Aryl, benzyl, as well as alkyl substituents
on the nitrogen (7a–7g) and the carboxyl moiety (R1) 7j were
equally well tolerated. When R2 =H or R1 is a good leaving
group, such as OtBu, the described method gave rise to the
cyclic, synthetically versatile building blocks 7h and 7i in 67
and 85% yield, respectively. In all transformations, only the car-
Chem. Eur. J. 2015, 21, 1444 – 1448
www.chemeurj.org
1445
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
boxyl oxygen acted as the intramolecular nucleophile, leaving
the aromatic residues even in 7k and 7l untouched. a,b-Sub-
stituted imides were readily transferred into the tertiary esters
7m–7q with excellent diastereoselectivities (up to>95:5) in
80–83% yield. The relative configuration of the C-2 and C-3
methyl unit in 7p was assigned as trans by X-ray analysis, as
was expected for the rearrangement involving an intermediate
halonium ion. To demonstrate the practicability, the experi-
ments were also performed on a gram scale, giving com-
pounds 7a, 7b, 7e–7j, 7o, and 7p in similar yields (63–
98%).[13]
With a successful and easily to conduct synthesis of these
a,a-disubstituted a-hydroxy amides 7 by using NBS as the
electrophilic halogenation reagent in combination with pre-
catalyst 6a in hand, we turned our focus to further improve
the economy of our method. As the succinimide is accumulat-
ed in equimolar amounts as a waste product during the rear-
rangement, we were eager to replace NBS by a more atom-
economic and cheaper Br+ source. In this context, an in-situ
generation of the electrophilic bromonium species from readily
available bulk chemicals is highly desirable. Based on our expe-
rience of using the inorganic bromide salt KBr together with
oxone as external oxidant[14] as a NBS substitute in iodine(III)-
catalyzed reactions,[4,5] we subjected a variety of imides 3 to
these conditions (Scheme 3).
Scheme 4. Selected transformations of a-alkyl-a-hydroxy amide 7b.
an initial survey of subsequent reactions was exemplarily un-
dertaken for substrate 7b (Scheme 4). Removal of the ester
functionality to afford 12 occurred smoothly by using Lewis
acids, such as BCl3. Replacement of the bromine atom by a hy-
droxy function or a chloride atom can also be achieved to give
the diol 13 and the chloro derivative 14, respectively, each in
63% yield. The dimethyl congener 15 was obtained by hydro-
genation of 7b in excellent 93% yield. To our delight, a-hy-
droxy-b-amino acid 16 and b-lactam 17 were accessible upon
treatment of 7b with base in very good yields of 91%–94%.
To shed light on the role of the iodobenzamide 6 in the cat-
alytic rearrangement of 3, first investigations on the actual cat-
alytic species were performed. As the proposed bromo benzio-
doxolones 18 are very reactive and instable compounds, thus
precluding their isolation,[15] we commenced in situ NMR stud-
ies. Immediately after the addition of NBS to a mixture of 6a
and TFA in CD2Cl2, a new set of signals appeared in the
1H NMR spectra (Figure 1, labelled with *), with all chemical
Scheme 3. Substrate scope for the rearrangement of imides 3 by using KBr/
oxone. For details on the reaction conditions, see the Supporting Informa-
1
tion. [a] Isolated yield. [b] Determined by H NMR spectroscopy from the
crude mixture. [c] R1 =OtBu. [d] Derived from the tiglic acid derivative of 3.
All substrates 3 were converted, giving the corresponding
amides 7 in similar yields as if NBS was applied. In general,
however, a hampered reactivity for the rearrangement to 7
was observed, which resulted in significantly extended reaction
times, also when other solvents, such as acetonitrile or HFIP,
were used. In the case of the C-3-substituted products 7p and
7q, the transformation was still incomplete after 18 h: imides
3p and 3q were still detectable in the reaction mixture at this
time and the amides 7p and 7q were isolated only in low
yields of 54 and 43%, respectively.
1
Figure 1. Comparison of the H NMR spectra of iodobenzene 6a and the in
situ formed hypervalent iodine species 18a (labelled with *).
shifts being significantly moved downfield.[16] In particular, the
aromatic protons ortho and para to the amide (C-3 and C-5)
were downfield shifted by ca. d=0.8 ppm compared to 6a, an
observation that was already made in earlier studies on io-
danes.[17] The signal multiplicity of the methylene protons on
With this straightforward access to a-alkyl-a-hydroxy amides
7 available, their synthetic utility was probed next. Therefore,
Chem. Eur. J. 2015, 21, 1444 – 1448
www.chemeurj.org
1446
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
pound 19. Nucleophilic ring opening by bromide forms the
bromonium ion 20. Thereby, precatalyst 6 is regenerated, thus
closing the catalytic cycle. Compound 20 is then opened by an
intramolecular, nucleophilic attack of the carboxyl group,
giving rise solely to the five-membered cyclic iminium 21. Al-
ternatively, the latter, intramolecular process can also take
place first, affording oxazolonium 22. Substitution of the iodi-
ne(III) moiety in 22 by bromide then gives key intermediate
21. The existence of 21 was demonstrated indirectly by the
isolation and characterization of its hydrate 9, which yields the
desired product 7b by opening of the oxazolidin-4-one ring.
No formation of the corresponding six-membered intermediate
was detectable.
Figure 2. Comparison of the 13C NMR spectra of iodobenzene 6a and the in
situ formed hypervalent iodine species 18a (labelled with *).
In conclusion, we showed that a,a-disubstituted-a-hydroxy
carboxylamides 7 are easily accessible from imides 3 in
C-2’ changed from doublet to singlet, hinting at a cyclic imine
benziodoxole structure 18a.[18] The oxidation level of the
iodine atom in 18a was unambiguously assigned as (III) by
13C NMR spectroscopy. Here, a strong deshielding (ca. d=
25 ppm) of C-1 was evident in contrast to the signal for C-1 in
a
metal-free, hypervalent-iodine-catalyzed rearrangement,
which opens a selective and efficient route to these highly
useful building blocks. The methylene bromide functionality at
the tetrasubstituted a-carbon atom in 3 gives the opportunity
to introduce structural diversity and complexity into the carbo-
cyclic framework. Besides this, the formation of a hypervalent
iodine(III) compound in such reactions was verified, giving, for
the first time, analytical evidence for the existence of the
chemically sensitive and thus very unstable bromo-imine ben-
ziodoxolone species 18, which is selectively generated from
precatalyst 6 upon oxidation with NBS. These results give
a first impression of the processes occurring in catalytic io-
dine(III)-triggered halogenations and, therefore, will help to
design more effective and selective catalysts for this distinct re-
action type in the future.
6a (Figure 2), which is diagnostic for
a tricoordinated
iodane.[19] When conducting these studies with the methyl
ester derivative 6b, a similar set of new signals was generated,
accounting for the bromo-imine benziodoxole 18b. To further
evaluate the structure of the in situ formed iodine(III) com-
pounds 18, ESI-MS experiments of a mixture of 6b with NBS
were performed. In these investigations, the newly formed spe-
cies resembled the molecular formula expected for 18b (see
the Supporting Information).
Taken together, we propose that the rearrangement of
imides 3 proceeds through an electrophilic mechanism.[20] The
transformation commences with the oxidation of iodo-
benzamide 6 by NBS (Scheme 5), producing the actual, more
reactive bromo reagent 18. Next, an electrophilic addition of
the l3-iodane 18 to the electron-poor and thus deactivated
olefin moiety in 3b is facilitated, producing the iodine(III) com-
Acknowledgements
We thank L. Weigand, S. Hoch, and F. Mayr for technical assis-
tance. This work was funded by the “Fonds der Chemischen In-
dustrie” (Liebig fellowship to T.G. and fellowship to C.P.) and
the Emmy–Noether program of the DFG (GU 1134/3-1).
Keywords: a-hydroxy
carboxylamides
·
hypervalent
compounds · imides · iodine · rearrangements
[1] a) M. Uyanik, K. Ishihara, Chem. Commun. 2009, 2086–2099; b) P. J.
Stang, V. V. Zhdankin, Chem. Rev. 1996, 96, 1123–1178; c) T. Kitamura, Y.
Fujiwara, Org. Prep. Proced. Int. 1997, 29, 409–458; d) A. Varvoglis, Tetra-
hedron 1997, 53, 1179–1255; e) V. V. Zhdankin, P. J. Stang, Chem. Rev.
2002, 102, 2523–2584; f) M. Ochiai, H. Tohma, T. Wirth, G. F. Koser, A.
Varvoglis, Y. Kita, V. V. Zhdankin, Top. Curr. Chem. 2003, 224, 1; g) T.
Wirth, Angew. Chem. Int. Ed. 2005, 44, 3656–3665; Angew. Chem. 2005,
117, 3722–3731; h) R. M. Moriarty, J. Org. Chem. 2005, 70, 2893–2903;
i) V. V. Zhdankin, P. J. Stang, Chem. Rev. 2008, 108, 5299–5358; j) T. Dohi,
Y. Kita, Chem. Commun. 2009, 2073–2085; k) E. A. Merritt, B. Olofsson,
Synthesis 2011, 517–538; l) J. P. Brand, D. F. Gonzꢁlez, S. Nicolai, J.
Waser, Chem. Commun. 2011, 47, 102–115; m) A. Parra, S. Reboredo,
Chem. Eur. J. 2013, 19, 17244–17260; n) V. V. Zhdankin, Hypervalent
Iodine Chemistry: Preparation, Structure, and Synthetic Applications of
Polyvalent Iodine Compounds, Wiley, Chichester, 2013.
[2] F. V. Singh, T. Wirth, Synthesis 2013, 45, 2499–2511.
[3] a) R. D. Richardson, T. Wirth, Angew. Chem. Int. Ed. 2006, 45, 4402–4404;
Angew. Chem. 2006, 118, 4510–4512; b) F. V. Singh, T. Wirth, Chem.
Asian J. 2014, 9, 950–971, and references cited therein.
Scheme 5. Proposed reaction mechanism.
Chem. Eur. J. 2015, 21, 1444 – 1448
www.chemeurj.org
1447
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
[4] M. Stodulski, A. Goetzinger, S. V. Kohlhepp, T. Gulder, Chem. Commun.
2014, 50, 3435–3438.
[5] D. C. Fabry, M. Stodulski, S. Hoerner, T. Gulder, Chem. Eur. J. 2012, 18,
10834–10838.
[6] While N-alkyl amide 2 exhibits a s-cis amide conformation, imide 3a
comprises a s-trans amide functionality. For details, see A. Itai, Y. Toriu-
mi, S. Saito, H. Kagechika, K. Shudo, J. Am. Chem. Soc. 1992, 114,
10649–10650.
Snyder, D. S. Treitler, Angew. Chem. Int. Ed. 2009, 48, 7899–7903;
Angew. Chem. 2009, 121, 8039–8043; c) S. A. Snyder, D. S. Treitler, A. P.
Brucks, J. Am. Chem. Soc. 2010, 132, 14303–14314.
[13] Ring-opening of 9 on a gram scale was achieved completely upon
treatment of the reaction mixture with SiO2.
[14] The oxidation of 6a to the corresponding iodane by oxone followed by
the reaction with bromide to yield 17 is also possible and cannot be
excluded.
[7] When catalyst 5 was used instead of 6, a mixture of products was ob-
served in the 1H NMR spectrum of the crude mixture. Here, besides
other compounds, 7b along with the dibrominated product was identi-
fied.
[15] The isolation of 18 has not been successful yet, owing to the high in-
stability of bromo-imine benziodoxoles.
[16] The same downfield-shifted 1H NMR signals appeared in the reaction
mixture consisting of 3b, NBS, TFA, and 6a in CD2Cl2.
[8] When dehalogenated analogs of 6 were used as catalysts, no product
was formed.
[17] E. V. Vinogradova, P. Mueller, S. L. Buchwald, Angew. Chem. Int. Ed. 2014,
53, 3125–3128; Angew. Chem. 2014, 126, 3189–3192.
[9] T. Dohi, N. Yamaoka, Y. Kita, Tetrahedron 2010, 66, 5775–5785; for se-
lected, most recent examples employing HFIP as a solvent in iodine(III)
reactions, see: a) S. Desjardins, G. Maertens, S. Canesi, Org. Lett. 2014,
16, 4928–4931; b) S. K. Alla, P. Sadhu, T. Punniyamurthy, J. Org. Chem.
2014, 79, 7502–7511; c) I. Sokolovs, D. Lubriks, E. Suna, J. Am. Chem.
Soc. 2014, 136, 6920–6928; d) M. Ito, H. Kubo, I. Itani, K. Morimoto, T.
Dohi, Y. Kita, J. Am. Chem. Soc. 2013, 135, 14078–14081; e) A. Ahmad, P.
Scarassati, N. Jalalian, B. Olofsson, L. F. Silva, Tetrahedron Lett. 2013, 54,
5818–5820; f) M. Uyanik, T. Yasui, K. Ishihara, Angew. Chem. Int. Ed.
2013, 52, 9215–9218; Angew. Chem. 2013, 125, 9385–9388.
[10] a) R. L. Amey, J. C. Martin, J. Am. Chem. Soc. 1979, 101, 3060–3065;
b) G. C. Geary, E. G. Hope, K. Singh, A. M. Stuart, Chem. Commun. 2013,
49, 9263–9265.
[18] Due to the acidic conditions, the hypervalent iodine exists more likely
in its imine benziodoxole 18 than in the corresponding benziodazolone
form. For details, see a) V. V. Zhdankin, R. M. Arbit, B. J. Lynch, P. Kiprof,
V. G. Young Jr., J. Org. Chem. 1998, 63, 6590–6596; b) V. V. Zhdankin,
R. M. Arbit, M. McSherry, B. Mismash, V. G. Young Jr., J. Am. Chem. Soc.
1997, 119, 7408–7409.
[19] a) T. M. Balthazor, D. E. Godar, B. R. Stults, J. Org. Chem. 1979, 44, 1447–
1449; b) A. R. Katritzky, B. L. Duell, J. K. Gallos, H. D. Durst, Magn. Reson.
Chem. 1989, 27, 1007–1011; c) A. R. Katritzky, J. K. Gallos, H. D. Durst,
Magn. Reson. Chem. 1989, 27, 815–822.
[20] The reaction proceeding through a radical mechanism was ruled out by
control experiments.
[11] D. C. Braddock, G. Cansell, S. A. Hermitage, Chem. Commun. 2006,
2483–2485.
Received: October 30, 2014
[12] a) J. Barluenga, M. Trincado, E. Rubio, J. M. Gonzꢁlez, Angew. Chem. Int.
Ed. 2003, 42, 2406–2409; Angew. Chem. 2003, 115, 2508–2511; b) S. A.
Published online on December 2, 2014
Chem. Eur. J. 2015, 21, 1444 – 1448
www.chemeurj.org
1448
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim