Cu/Fe Catalyzed Intermolecular Oxidative Amination of Benzylic C-H Bonds

Article abstract of DOI:10.1002/chem.201600107

We report a Cu/Fe co-catalyzed Ritter-type C-H activation/amination reaction that allows efficient and selective intermolecular functionalization of benzylic C-H bonds. This new reaction is featured by simple reaction conditions, readily available reagents and general substrate scope, allowing facile synthesis of biologically interesting nitrogen containing heterocycles. The Cu and Fe salts were found to play distinct roles in this cooperative catalysis. With a little help: A Ritter-type intermolecular amination of benzylic C-H bonds with acetonitrile, co-catalyzed by Cu^II/Fe^III is reported. A wide array of biologically interesting nitrogen containing heterocycles was prepared from 2-alkyl benzoic acids and heteroaromatic carboxylic acids under operationally simple conditions. The Cu and Fe salts were found to play distinct roles in this cooperative catalysis.

Full text of DOI:10.1002/chem.201600107

DOI: 10.1002/chem.201600107
Communication
&
Synthetic Methods |Hot Paper|
Cu/Fe Catalyzed Intermolecular Oxidative Amination of Benzylic
CÀH Bonds
Cong Liu,[a, b] Qi Zhang,[a] Hongbo Li,[a] Shuangxi Guo,[a] Bin Xiao,[b] Wei Deng,[c] Lei Liu,[b] and
Wei He*[a]
Abstract: We report a Cu/Fe co-catalyzed Ritter-type CÀH
activation/amination reaction that allows efficient and se-
lective intermolecular functionalization of benzylic CÀH
bonds. This new reaction is featured by simple reaction
conditions, readily available reagents and general sub-
strate scope, allowing facile synthesis of biologically inter-
esting nitrogen containing heterocycles. The Cu and Fe
salts were found to play distinct roles in this cooperative
catalysis.
Scheme 1. Intermolecular benzylic CÀH amination reactions.
Catalytic CÀH amination provides an appealing method for the
construction of CÀN bonds.[1] Metal nitrene insertion[2] is the
most exploited strategy for this purpose. Alternative methods
(i.e., nitriles), but also expand the toolbox for benzylic CÀH
including metal-free organonitrenoid,[3] CÀH activation/amina-
functionalization.
tion[4] and even enzyme catalysis[5] are also emerging. However,
Herein we report a Cu/Fe co-catalyzed Ritter-type intermo-
these methods are confronted with selectivity issues[6] in inter-
lecular oxidative CÀH amination of benzylic CÀH bonds ortho
molecular benzylic CÀH aminations due to the existence of
to a carboxylic group using acetonitrile as the nitrogen source
more reactive C(sp2)ÀH bonds. Oxidative CÀH activation/ami-
(Scheme 1b). This simple method is shown to be applicable to
nation reaction,[4] which takes advantage of the propensity of
a wide scope of substrates, providing rapid syntheses of a di-
the benzylic CÀH bonds in H-atom abstraction (HAA), has met
verse array of nitrogen-containing heterocycles including
with some encouraging success.[7]
a common synthetic precursor of two anxiolytic drugs Pago-
Ritter-type reactions[8] proceed through the nitrile capture of
a carbon cation intermediate, thus constituting another type
of oxidative CÀH amination strategy. For example, Li and co-
workers reported an interesting Rh-catalyzed diamination of al-
kenes by acetonitrile to access imidazolines.[9] Recently, Baran
clone and Pazinaclone. Our mechanistic studies suggested that
this reaction likely proceeds through a benzylic CÀH amina-
tion/lactamization cascade. We show that in such a cascade
[Cu] is required for the oxidative CÀH amination and [Fe] is re-
sponsible for the lactamization, while Selectfluor is necessary
et al. reported a remarkable Ritter-type intermolecular C(sp3)À for both processes.
H amination (Scheme 1a).[10] Given the rare examples[11,12] of in-
termolecular amination of benzylic CÀH bonds, new Ritter-type
oxidative amination of benzylic CÀH bonds would not only
benefit from the advantage of nonspecialized nitrogen sources
Our investigation started with a serendipitous finding that
treatment of 2-methylbenzoic acid (1a) with Selectfluor (F-
TEDA-BF4) under the catalysis of CuII in acetonitrile gives an un-
expected product in a low yield, which was subsequently iden-
tified as isoindolinone 2a by X-ray diffraction[13] (Table 1 and
Table 2). We further confirmed that acetonitrile was the nitro-
gen source by employing d3-acetonitrile (see the Supporting
Information). Interestingly, a number of intermolecular benzylic
CÀH amination reactions in acetonitrile were reported, but
amination product by acetonitrile trapping was not documen-
ted.[7d,11a] We thus decided to explore this surprising reactivity
to develop a new method for intermolecular benzylic CÀH ami-
nation.
[a] C. Liu, Q. Zhang, H. Li, S. Guo, Prof. Dr. W. He
School of Pharmaceutical Sciences, Tsinghua University
Beijing 100084 (P. R. China)
E-mail: whe@mail.tsinghua.edu.cn
[b] C. Liu, B. Xiao, Prof. Dr. L. Liu
Department of Chemistry, Tsinghua University
Beijing 100084 (P. R. China)
[c] Prof. Dr. W. Deng
Reaction optimization was carried out using 1a as the
model substrate, with Selectfluor as the oxidant (Table 1 and
the Supporting Information). Initial screening showed that
20 mol% CuCl2 could give 2a in a moderate NMR yield (60%
School of Chemical and Environmental Engineering
Shanghai Institute of Technology, Shanghai 201418 (P. R. China)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201600107.
Chem. Eur. J. 2016, 22, 6208 – 6212
6208
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
85% yield. Substitutions on the phenyl ring were allowed, as
exemplified in the cases of 4b and 4c. Benzylic CÀH bonds in
fused alkyl rings were aminated into tricyclic scaffolds (4d, 4e)
in moderate yields. The structure of compound 4e was unam-
biguously established by X-ray crystallography.[13] Substrates
bearing terminal methyl-, ethyl- and isopropyl esters could also
be converted to the corresponding products at 1008C (4 f–
4h). It was noted again that the amination was site selective
as other benzylic CÀH bonds (4b, 4d) were not affected.
Table 1. Effect of Cu/Fe combinations.[a]
Entry
Equiv [Cu]
Equiv [M]
2a/B/C [%][b,c]
1
2
3
4
5
6
0.2 CuCl2
0.2 CuCl2
–
0.01 CuCl2
0.01 CuCl2
0.01 Cu(NO3)2
–
60:40:n.d.[d]
65:25:10
n.d.:n.d.:100
n.r.
80:10:10
85:8:7[e]
0.2 FeCl2
0.2 FeCl2
–
0.2 FeCl2
0.2 [Fe(acac)3]
The reaction was also successfully extended to heterocyclic
substrates. Thus, thiophene (6a–c), benzothiophene (6d,e) and
benzofuran (6 f) embedded products were rapidly prepared
under standard reaction conditions (Table 3). It was collectively
shown that carboxylic group at all three possible positions
(namely 6a, 6b and 6d) relative to the heteroatom (S) were
capable of directing the amination, indicating that the hetero-
atom is unlikely to participate in the initial CÀH activation.
[a] Reaction conditions: 2 equiv Selectfluor, 808C, 8 h. [b] Yields deter-
mined by NMR spectroscopy with an internal standard. [c] The ratio is
normalized against 2a. [d] Incomplete conversion. [e] 80% isolated yield;
n.d.=not detected by NMR spectroscopy. n.r.=no reaction; acac=acety-
lacetonate.
Table 2. Amination of primary and secondary benzylic CÀH bonds.[a,b]
selectivity, 80% conversion; Table 1, entry 1), with a large
amount of over-oxidized product B. Interestingly, addition of
a catalytic amount of FeII salt promoted the conversion
(Table 1, entry 2). In the absence of CuII, FeII alone gave exclu-
sively 2-formylbenzoic acid C, and no CÀN bond formation was
detected (Table 1, entry 3). The subtle roles of the two metals
were further underscored by the fact that 1 mol% CuII alone
showed no catalytic activity (Table 1, entry 4). This observation
suggested that CuII might be responsible for the benzylic CÀH
amination and the over-oxidation. To circumvent the over oxi-
dation, a reduced amount of CuCl2 (1 mol%) was employed. To
our delight, over-oxidation product B could be largely avoided
and the amination product 2a was produced in 80% NMR
yield (Table 1, entry 5). This success spurred us to screen
a large panel of Cu/Fe combinations, thus revealing the opti-
mal catalyst system with FeIII. Under this condition (1 mol%
Cu(NO3)2 with 20 mol% [Fe(acac)3]), the desired product 2a
was obtained in 80% isolated yield (85% NMR yield; Table 1,
entry 6).
This simple procedure was shown to be amenable to a di-
verse array of primary and secondary benzylic substrates
(Table 2). The amination was shown to be site selective, as
other benzylic CÀH bonds remained intact (2b–2d). The reac-
tion was compatible with F-, Cl-, Br- and I-substituted sub-
strates, furnishing halogenated products poised for further
elaboration (2 f–2i). Various electron-donating groups, includ-
ing phenyl (2e), trifluoromethoxy (2j), methoxy (2k), O-acetyl
(2l) and naphthalenyl (2m) were well tolerated. Substrates
with electron-withdrawing groups such as trifluoromethyl (2n)
and cyano (2o) appeared to be less reactive, so that an elevat-
ed reaction temperature was required. In the case of 2i, 2m,
2n and 2o, the reaction suffered from incomplete conversion
and diminished yields. Attempts to improve the conversion by
using higher loading of CuII catalyst turned out to be unsuc-
cessful, similar to the observation made by Baran in some of
his hydrocarbon substrates.[10]
[a] Reaction conditions: substrate
1 (1.0 equiv), Cu(NO3)2 (1 mol%),
[Fe(acac)3] (20 mol%), Selectfluor (2.0 equiv), 808C, 8 h, N2. [b] Isolated
yields. [c] An unidentified side product was isolated. [d] 1008C, 12 h. Ac=
acetyl.
Secondary benzylic CÀH bonds were also sufficiently reactive
(Table 2). 2-Ethylbenzoic acid reacted smoothly to give 4a in
Chem. Eur. J. 2016, 22, 6208 – 6212
www.chemeurj.org
6209
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
common intermediate 10 was obtained from amination prod-
uct 4 f. Pazinaclone and Pagoclone were prepared in 2 steps
and 3 steps from 10. The overall yields from commercially
available 3 f were 35.6 and 34.1%, respectively. These examples
suggested that the current method could provide a versatile
entry point for the syntheses of bioactive compounds.
Table 3. Amination of heterocyclic substrates.[a,b]
[a] Reaction conditions: substrate
3 (1.0 equiv), Cu(NO3)2 (1 mol%),
[Fe(acac)3] (20 mol%), Selectfluor (2.0 equiv), 808C, 8 h, N2. [b] Isolated
yields. [c] 1008C, 12 h; Ac=acetyl.
Given the observed compatibility with heteroatoms, this
method should be applicable to other heteroaromatic sub-
strates, to create novel skeletons as we have noted that 6b,
6d and 6 f are not reported in the literature. The structure of
compound 6c was also verified by X-ray analysis.[13] Important-
ly, the reaction could be easily scaled up, as shown in the
gram quantity syntheses of 2a, 4a, 4 f and 6a.
As the requirement of a carboxylic directing group poses
some limitations to the synthetic utility of the reaction, we
thus sought to manipulate the acylated lactam product using
2a as an example (Scheme 2). Deprotection of amine products
(e.g., sulfonimide) of the earlier CÀH amination reactions is
often not straightforward.[14] In our case, the acetyl group in
the CÀH amination products 2a can be cleanly cleaved under
simple basic conditions[15] to afford the free lactam 7a. Hydrol-
ysis of 7a under acidic conditions gave the free amine salt 8a
in a 85% yield. Importantly, benzylamine 8a can be obtained
in a higher yield direct by hydrolysis of 2a. The easy access to
both free lactam and benzylamine, promises additional appli-
cations of our method.
Scheme 3. Syntheses of Pazinaclone and Pagoclone.
Possible pathways of the reaction are shown in Scheme 4. If
the benzyl cation (I) formation can be assumed according to
Baran’s earlier mechanistic analysis,[10] this cation can be cap-
tured in three different ways: 1) by the pendant carboxylic
oxygen to generate benzalactone D; 2) by acetonitrile to gen-
erate intermediate III. Intermediate III can in turn be trapped
by the carboxylic oxygen to form oxazepin E, or hydrolyzed to
acetamide F; 3) by water to generate benzy alcohol and subse-
quently benzaldehyde C (not shown).
The synthetic utility of amination product derived from a sec-
ondary benzylic CÀH bond was showcased in the divergent
syntheses of two investigational drugs Pazinaclone[16] and Pa-
goclone[17] (Scheme 3 and the Supporting Information). Under
the conditions described in Scheme 2, gram quantity of the
We then performed a number of experiments to determine
if any of the intermediates is a viable precursor to product 2a
(Table 4 and the Supporting Information). First, the reaction
1
was monitored by H NMR spectroscopy. During the course of
the reaction, only the signals corresponding to substrate 1a
and 2a were recorded. In other words, none of intermediates
D, E or F was observed. We then carried out mimic reactions
of D and F under the reaction conditions (we attempted to
synthesize oxazepin E but failed). Benzalactone D was shown
to convert to benzaldehyde C in a good yield (60%) under the
reaction conditions with no product 2a being discovered,
which suggested that D is unlikely the precursor to product
2a. In comparison, under the reaction conditions, acetamide F
gave the desired product 2a in a 73% yield (entry 1, Table 4).
On an important note, the lactamization of F has not been
reported, which makes this process puzzling. We thus conduct-
ed various experiments to examine the specific role of [Cu]
salt, [Fe] salt and Selectfluor in this lactamization step.[18] Se-
lectfluor was found to be indispensable for this conversion
Scheme 2. Transformation of the product.
Chem. Eur. J. 2016, 22, 6208 – 6212
www.chemeurj.org
6210
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
H amination while [Fe] is only responsible for the lactamiza-
tion. We believe such a cooperative catalytic lactamization war-
rants further investigation.
In conclusion, we delineated here an efficient intermolecular
oxidative benzylic CÀH amination reaction under the coopera-
tive catalysis of Cu/Fe hinged on the surprising trapping by
acetonitrile. This Ritter-type reaction is applicable to a wide
scope of primary and secondary benzylic CÀH bonds, furnish-
ing interesting nitrogen containing scaffolds. Future studies
are focused on the development of a directing group free var-
iant as well as the optimization of the interesting lactamization
reaction.
Experimental Section
Typical procedure: Anhydrous CH3CN (1 mL) was added to
a Schlenk tube containing 0.2 mmol substrate (1 equiv), [Fe(acac)3]
(14.1 mg, 0.04 mmol, 0.2 equiv), Cu(NO3)2 (0.5 mg, 0.01 equiv) and
Selectfluor (141.7 mg, 0.4 mmol, 2.0 equiv) under nitrogen atmos-
phere. The mixture was degassed three times by Freeze–Pump–
Thaw cycles and then was stirred for 8–12 h at 80–1008C. After
cooling to room temperature, the mixture was passed through
a short silica gel column and then concentrated under reduced
pressure. The residue was purified by column chromatography on
silica gel to afford the desired product.
Scheme 4. Possible reaction pathways and intermediates.
Table 4. The role of [Cu], [Fe] and Selectfluor in the cyclization of F.[a,b]
Acknowledgements
Entry
Conditions
2a [%]
C [%]
This work is supported by funds from the National Key Basic
Research Program of China (2012CB224802), the “863” Program
of the Ministry of Science and Technology (Grant
2012AA02A700), and NSFC (21371107 and 21221062).
1
2
3
4
5
6
7
initial reaction conditions[a]
entry 1 without Selectfluor
entry 1 with only 1.0 equiv of Selectfluor
entry 1 without [Fe] (i.e., 0.01 equiv [Cu])
entry 4 with 0.2 equiv [Cu]
entry 4 with only 1.0 equiv of Selectfluor
entry 1 without [Cu] (i.e., 0.2 equiv [Fe])
73
<5
90
0
40
70
75
10
0
5
0
30
5
13
Keywords: benzylic · CÀH activation · CÀH amination · Cu
catalysis · Fe catalysis
[a] The initial reaction conditions: 0.2 equiv [Fe], 0.01 equiv [Cu] and
2.0 equiv Selectfluor. [b] For entries 1, 3, 5, 6 and 7 the acetamide F was
fully consumed in 2 h.
[1] For recent reviews, see: a) J. L. Jeffrey, R. Sarpong, Chem. Sci. 2013, 4,
4092–4106; b) T. A. Ramirez, B. Zhao, Y. Shi, Chem. Soc. Rev. 2012, 41,
931–942; c) R. T. Gephart, T. H. Warren, Organometallics 2012, 31, 7728–
7752; d) F. Collet, R. H. Dodd, P. Dauban, Chem. Commun. 2009, 5061–
5074, also see references [2–5].
(entry 2, Table 4). Interestingly, if only 1.0 equiv of Selectfluor
was used, product 2a was obtained in a higher yield (entry 3,
Table 4). This could be understood by the fact that 1.0 equiv of
Selectfluor is consumed in the generation of acetamide F. [Fe]
salt was also required (entry 4, Table 4), which is consistent
with entry 4, Table 1. When 0.2 equiv of [Cu] was used, product
2a was obtained in a 40% yield (entry 5, Table 4). The yield
could be improved to 70% with 1.0 equiv of Selectfluor
(entry 6, Table 4). The observations in entries 4 and 5 are in line
with entry 1, Table 1, where 0.2 equiv [Cu] produced product
2a in a 60% yield. Interestingly, [Cu] was not required for the
conversion of F to product 2a, as 0.2 equiv [Fe] alone is suffi-
cient for a high yield (entry 7, Table 4). Although we do not un-
derstand the detailed mechanism for the conversion of acet-
amide F to product 2a, especially why such a process needs
an oxidant, the current results are in full agreement with each
other. In essence, under the initial reaction conditions, [Cu]
and [Fe] play very distinct roles: [Cu] is only responsible for CÀ
[2] a) K. Shin, H. Kim, S. Chang, Acc. Chem. Res. 2015, 48, 1040–1052; b) G.
Dequirez, V. Pons, P. Dauban, Angew. Chem. Int. Ed. 2012, 51, 7384–
7395; Angew. Chem. 2012, 124, 7498–7510; c) J. L. Roizen, M. E. Harvey,
J. Du Bois, Acc. Chem. Res. 2012, 45, 911 –922.
[3] a) M. Ochiai, K. Miyamoto, T. Kaneaki, S. Hayashi, W. Nakanishi, Science
2011, 332, 448–451; b) H. F. Bettinger, M. Filthaus, H. Bornemann, I. M.
Oppel, Angew. Chem. Int. Ed. 2008, 47, 4744–4747; Angew. Chem. 2008,
120, 4822–4825.
[4] For a related review, see: a) M.-L. Louillat, F. W. Patureau, Chem. Soc. Rev.
2014, 43, 901–910.
[5] a) J. A. McIntosh, P. S. Coelho, C. C. Farwell, Z. J. Wang, J. C. Lewis, T. R.
Brown, F. H. Arnold, Angew. Chem. Int. Ed. 2013, 52, 9309; Angew. Chem.
2013, 125, 9479; b) R. Singh, M. Bordeaux, R. Fasan, ACS Catal. 2014, 4,
546.
[6] For a recent investigation into the selectivity, see: E. N. Bess, R. J.
DeLuca, D. J. Tindall, M. S. Oderinde, J. L. Roizen, J. Du Bois, M. S.
Sigman, J. Am. Chem. Soc. 2014, 136, 5783–5789.
[7] a) S. Wiese, Y. M. Badiei, R. T. Gephart, S. Mossin, M. S. Varonka, M. M.
Melzer, K. Meyer, T. R. Cundari, T. H. Warren, Angew. Chem. Int. Ed. 2010,
49, 8850–8855; Angew. Chem. 2010, 122, 9034–9039; b) E. R. King, E. T.
Hennessy, T. A. Betley, J. Am. Chem. Soc. 2011, 133, 4917–4923; c) Z. Ni,
Chem. Eur. J. 2016, 22, 6208 – 6212
www.chemeurj.org
6211
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Q. Zhang, T. Xiong, Y. Zheng, Y. Li, H. Zhang, J. Zhang, Q. Liu, Angew.
Chem. Int. Ed. 2012, 51, 1244–1247; Angew. Chem. 2012, 124, 1270–
1273; d) . Iglesias, R. lvarez, . de Lera, K. MuÇiz, Angew. Chem. Int.
Ed. 2012, 51, 2225–2228; Angew. Chem. 2012, 124, 2268–2271; e) R. T.
Gephart, D. L. Huang, M. J. B. Aguila, G. Schmidt, A. Shahu, T. H. Warren,
Angew. Chem. Int. Ed. 2012, 51, 6488–6492; Angew. Chem. 2012, 124,
6594–6598; f) A. McNally, B. Haffemayer, B. S. L. Collins, M. J. Gaunt,
Nature 2014, 510, 129–133.
Phelps, R. J. Scamp, N. S. Dolan, J. M. Schomaker, J. Am. Chem. Soc.
2014, 136, 16720–16723; c) D. A. Bercovici, M. Brewer, J. Am. Chem. Soc.
2012, 134, 9890–9893; d) G. He, Y. Zhao, S. Zhang, C. Lu, G. Chen, J.
Am. Chem. Soc. 2012, 134, 3–6; e) M. Ichinose, H. Suematsu, Y. Yasuto-
mi, Y. Nishioka, T. Uchida, T. Katsuki, Angew. Chem. Int. Ed. 2011, 50,
9884–9887; Angew. Chem. 2011, 123, 10058–10061.
[13] CCDC 1046276 (2a), 1046277 (4e) and 1046278 (6c) contain the sup-
plementary crystallographic data for this paper. These data are provid-
ed free of charge by The Cambridge Crystallographic Data Centre.
[14] For a nice summary, see references [12–14] in reference [10].
[15] H. Yazawa, S. Goto, Tetrahedron Lett. 1985, 26, 3703–3706.
[16] S. M. Evans, R. W. Foltin, F. R. Levin, M. W. Fischman, Behav. Pharmacol.
1995, 6, 176–186.
[17] G. Maguire, D. Franklin, N. G. Vatakis, E. Morgenshtern, T. Denko, J. S.
Yaruss, C. Spotts, L. Davis, A. Davis, P. Fox, P. Soni, M. Blomgren, A. Sil-
verman, G. Riley, J. Clin. Psychopharmacol. 2010, 30, 48–56.
[8] For an up-to-date review, see a) D. Jiang, T. He, L. Ma, Z. Wang, RSC Adv.
2014, 4, 64936–64946.
[9] G. Li, H.-X. Wei, S. H. Kim, M. D. Carducci, Angew. Chem. Int. Ed. 2001, 40,
4277–4280; Angew. Chem. 2001, 113, 4407–4410.
[10] Q. Michaudel, D. Thevenet, P. S. Baran, J. Am. Chem. Soc. 2012, 134,
2547–2550.
[11] a) R. Bhuyan, K. M. Nicholas, Org. Lett. 2007, 9, 3957–3959; b) R. Fan, W.
Li, D. Pu, L. Zhang, Org. Lett. 2009, 11, 1425–1428; c) H. J. Kim, J. Kim,
S. H. Cho, S. Chang, J. Am. Chem. Soc. 2011, 133, 16382–16385; d) Q.
Xue, J. Xie, H. Li, Y. Cheng, C. Zhu, Chem. Commun. 2013, 49, 3700–
3702; e) T. Kang, Y. Kim, D. Lee, Z. Wang, S. Chang, J. Am. Chem. Soc.
2014, 136, 4141–4144; also see references [7c,d].
[18] We thank an anonymous reviewer for the helpful suggestion to exam-
ine the roles of Cu and Fe salts.
[12] Intramolecular benzylic CÀH aminations were more successful, for ex-
amples see: a) C. Zhu, Y. Liang, X. Hong, H. Sun, W.-Y. Sun, K. N. Houk, Z.
Shi, J. Am. Chem. Soc. 2015, 137, 7564–7567; b) J. M. Alderson, A. M.
Received: January 11, 2016
Published online on March 17, 2016
Chem. Eur. J. 2016, 22, 6208 – 6212
www.chemeurj.org
6212
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim