Catalysis with Chalcogen Bonds

Article abstract of DOI:10.1002/anie.201611019

Herein, we introduce catalysts that operate with chalcogen bonds. Compared to conventional hydrogen bonds, chalcogen bonds are similar in strength but more directional and hydrophobic, thus ideal for precision catalysis in apolar solvents. For the transfer hydrogenation of quinolines and imines, rate enhancements well beyond a factor of 1000 are obtained with chalcogen bonds. Better activities with deeper σ holes and wider bite angles, chloride inhibition and correlation with computed anion binding energies are consistent with operational chalcogen bonds. Comparable to classics, such as 2,2′-bipyrroles or 2,2′-bipyridines, dithieno[3,2-b;2′,3′-d]thiophenes (DTTs), particularly their diimides, but also wide-angle cyclopentadithiazole-4-ones are identified as privileged motifs to stabilize transition states in the focal point of the σ holes on their two co-facial endocyclic sulfur atoms.

Full text of DOI:10.1002/anie.201611019

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201611019
German Edition: DOI: 10.1002/ange.201611019
Homogeneous Catalysis Very Important Paper
Catalysis with Chalcogen Bonds
Sebastian Benz, Javier Lꢀpez-Andarias, Jiri Mareda, Naomi Sakai, and Stefan Matile*
Abstract: Herein, we introduce catalysts that operate with
chalcogen bonds. Compared to conventional hydrogen bonds,
chalcogen bonds are similar in strength but more directional
and hydrophobic, thus ideal for precision catalysis in apolar
solvents. For the transfer hydrogenation of quinolines and
imines, rate enhancements well beyond a factor of 1000 are
obtained with chalcogen bonds. Better activities with deeper
s holes and wider bite angles, chloride inhibition and correla-
tion with computed anion binding energies are consistent with
operational chalcogen bonds. Comparable to classics, such as
2,2’-bipyrroles or 2,2’-bipyridines, dithieno[3,2-b;2’,3’-d]thio-
phenes (DTTs), particularly their diimides, but also wide-angle
cyclopentadithiazole-4-ones are identified as privileged motifs
to stabilize transition states in the focal point of the s holes on
their two co-facial endocyclic sulfur atoms.
C
halcogen bonds originate from the “s holes” in the
s* orbitals of the covalent bonds of electron-deficient
sulfur, selenium, or tellurium but not oxygen atoms (Fig-
ure 1a,b).[1–7] Sulfur atoms as chalcogen-bond donors are thus
À
similar to hydrogen-bond donors, such as N H (rather than
Figure 1. a) Location of s holes (blue circles, F1 ꢀ1808, F2 ꢀ708) on
hydrogen-bond acceptors, such as oxygen atoms). Compared
to conventional hydrogen-bond donors, chalcogen-bond
donors are more directional and more hydrophobic, charac-
teristics which make them appear perfect for catalysis in
apolar media. However, contrary to the related halogen
bonds,[1,2,8,9] the s holes are somewhat hidden on the side of
the chalcogen atom, next to the second covalent bond
(Figure 1a,b). Perhaps because of this peculiar location,
their use, particularly in medicinal chemistry and materials
sciences, has so far mainly focused on intramolecular
conformational control and solid-state crystal engineering.[1–3]
Experimental evidence for anion binding in solution is rare
and recent,[4] and anion transport with chalcogen bonds has
just been realized.[5] Moreover, pioneering examples for
intramolecular conformational control in covalent catalysis
have been reported.[6] Herein, we introduce chalcogen bonds
to intermolecular, non-covalent catalysis.
electron-deficient chalcogen (Ch) atoms (EWG: electron-withdrawing
group). b) electrostatic potential surface (EPS) of SF2 (top) and
SeF2 (bottom, blue: electron poor, red: electron rich).[2] c) General
concept for substrate activation in the focal point of the two cofacial
chalcogen-bond donors in DTTs (dithieno[3,2-b;2’,3’-d]thiophenes).
d) Computed structure of the complex of DTT 1’ (see Figure 2, 1 with
methyl instead of isobutyl substituents) with pyridine and e) formalde-
hyde. f) EPS of catalyst 1’ (MP2/6-311+ +G**//M062X/6-311G**,
isosurface: 0.009 au; red: À0.012 au, blue: 0.07 au).[5]
central single-atom bridge D between the peripheral rings is
conceived as important to increase the bite angle for optimal
bidentate substrate binding in the focal point of the s holes on
the cofacial endocyclic chalcogens Ch (Figure 1c).[5] In DTT
1, this bridge D is a sulfone acceptor, and cyano acceptors are
added in ortho position to further deepen the s holes on the
co-facial sulfurs (Figures 1c–f, Figure 2a). In density func-
tional theory (DFT) calculations at the M062X/6-311G**
level of theory,[5] carbonyl lone pairs were poorly recognized
by these deep s holes (Figure 1e). Presumably because of
mismatched geometries, only one lone pair seems to be bound
in the tilted global minimum. The computed interaction
energies were correspondingly weak (e.g. Eint = À5.8 kcal
molÀ1 for formaldehyde binding to DTT 1). Better results
were obtained for the single lone pair on endocyclic nitrogen
atoms of aromatic heterocycles (e.g. Eint = À8.1 kcalmolÀ1 for
pyridine binding to DTT 1, Figure 1d). Based on these
predictions from theory, transfer hydrogenation of quino-
lines[9] was selected to initially explore the possibility to
catalyze organic transformations with chalcogen bonds (Fig-
ure 2b).
Dithieno[3,2-b;2’,3’-d]thiophenes (DTTs)[7] and related
structures were considered as general motif for chalcogen
bonding in functional systems (Figure 1c) that complements
classics, such as 2,2’-bipyrroles or 2,2’-bipyridines.[5] The
[*] S. Benz, Dr. J. Lꢀpez-Andarias, Dr. J. Mareda, Dr. N. Sakai,
Prof. S. Matile
Department of Organic Chemistry
University of Geneva
Geneva (Switzerland)
E-mail: stefan.matile@unige.ch
Homepage: http://www.unige.ch/sciences/chiorg/matile/
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under http://dx.doi.org/10.
1002/anie.201611019.
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Figure 2. a) Structure of catalyst candidates 1–8 explored in this study to b) accelerate the transfer hydrogenation of quinolines 9 and imine 12.
Catalysts 1–8 (Figure 2a) were envisioned to elaborate on
chalcogen-bond catalysis. They were accessible in a few
synthetic steps from commercially available starting materials
following published methods.[5,10–12] Initial studies focused on
DTTs 1–3 because of their proven ability to bind and
transport anions in solution.[5] In the presence of 30 mol%
1, the reduction of 2-phenyl quinoline 9a with Hantzsch ester
10 afforded tetrahydroquinoline 11a within 12 days at room
temperature in 96% yield (Figure 3a, Table 1, entry 1). No
Addition of TBACl to the reaction mixture caused the
complete inhibition of catalyst 1 (Figure 3b). The IC50 = 27 Æ
1 mm obtained from the dose response curves was in the range
of the previously found chloride binding in THF (Table S1).[5]
The good correlation between the catalytic activity and
computed anion binding energies as well as inhibition with
chloride provided experimental support for operational
Table 1: Characteristics of chalcogen-bond catalysts.[a]
Entry Cat[b] Eint [kcal
molÀ1 [c]
S[d] h [%][e] kcat
/
DDGTS
[kJmolÀ1 [g]
]
[f]
]
kuncat
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
–
1
–
1
1
1
2
3
4
5
6
7
8
–
À34.6
À27.4
À32.7
À24.8
À26.9
À23.5
À28.3
À36.1
–
9a 96
9a 81
9a 62
490
67
100
125
–
–
500
1290
–
4.0
–
À15.3
À10.4
À11.4
À11.9
–
9a 82
[h]
9a
9a
–
–
[h]
–
9a 78
À15.3
À17.7
–
9a 94
[h]
9
9a –
10
11
12
13
14
15
16
17
18
19
20
21
22
À34.6
9b 97
À3.1
[h]
–
9b
9c
9d
–
–
–
–
Figure 3. a) Formation of product 11a with time in the presence of
[h]
[h]
À34.6
À34.6
À34.6
À27.4
À32.7
À24.8
À26.9
À23.5
À28.3
À36.1
–
–
–
–
–
^
substrates 9a and 10 and 0 mol% catalyst ( ) or 30 mol% catalysts
*
*
3 (&), 1 ( ), and 8 ( ). b) Initial velocity of the formation of product
11a in the presence of 9a, 10, 30 mol% catalyst 1 and increasing
concentrations of TBACl, with fit to Hill equation (IC50 =27Æ1 mm).
12 95
12 85
12 72
50
3.9
6.6
5.6
2.1
1.8
190
335
–
À9.6
À3.4
À4.6
À4.3
À1.8
À1.4
À13.0
À14.3
–
12 99
[h]
12
12
–
–
turnover was observed under the same conditions without
catalyst (Figure 3a, Table 1, entry 9). Only at higher substrate
concentrations and after prolonged reaction time, trace
amounts of product 11a were formed. From the initial
velocity of product formation, a kuncat = 3.9 ꢀ 10À5 mÀ1 hÀ1 was
obtained for transfer hydrogenation without catalyst. Com-
parison with the obtained kcat = 1.9 ꢀ 10À2 mÀ1 hÀ1 revealed
a rate enhancement of kcat/kuncat = 490 for catalyst 1 (Table 1,
entry 1). This rate enhancement suggested that activation of
the hydride acceptor by chalcogen bonding to catalyst
1 stabilized the transition state (and/or destabilized the
substrate) by DDGTS = À15.3 kJmolÀ1 (Table 1, entry 1).
Replacement of the cyano acceptors by sulfones in 3 and
removal of one cyano acceptor in 2 reduced the catalytic
activity 5 to 7 times, respectively (Table 1, entries 1–3).
[h]
12 85
12 95
[h]
12
–
[a] Reactions were conducted in CD2Cl2 at 208C with 10 (entries 1–13:
281 mm; entries 14–22: 375 mm), catalyst 1–8 (entries 1–8, 10, 12 and
13: 30 mol%; entries 14–21: 10 mol%) and substrates 9 (entries 1–13:
128 mm) or 12 (entries 14–22: 250 mm), and monitored by 1H NMR
spectroscopy. [b] Catalysts, see Figure 2a. [c] Binding energies with
chloride in the gas phase, calculated with M062X/6–311G**, data for 1–3
are from Ref. [5]. [d] Substrates, see Figure 2b. [e] Yield of the reduced
product, determined by 1H NMR integration. [f] Rate enhancement,
kuncat (9a)=3.9ꢁ10À5 mÀ1 hÀ1, kuncat (9b)=0.19mÀ1 hÀ1,kuncat (12)=
1.5ꢁ10À3 mÀ1 hÀ1. [g] Transition-state stabilization (or ground-state
destabilization), kcat/kuncat =exp (ÀDDGTS/RT). [h] Not determined, very
slow conversion.
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
chalcogen bonds in intermolecular catalysis. Quinolines 9b–
p catalysis.[13] This interpretation was supported by the higher
9d were reduced in the presence of catalyst 1 as well (Table 1,
entries 10–13). Quinoline 9b was the most reactive substrate
and yielded the product 10b in 97% yield within 8 h, whereas
the reduction of quinolines 9c and 9d was much slower.
In the presence of 10 mol% catalyst 1, the reduction of
the more reactive imine 12 afforded amine 13 in 95% yield
(Table 1, entry 14, Figure 2b). Compared to the uncatalyzed
background reaction (Table 1, entry 22), the rate enhance-
ment calculated to kcat/kuncat = 50. In support of operational
chalcogen bonds, decreasing depth of the s holes in catalysts 3
activity of the expanded DDI 7 to reduce the planar imine 12
(more active than 1) than the twisted 2-phenyl quinoline 9a
(as active as 1).
Taken together, these results provide compelling direct
experimental evidence for existence and significance of
intermolecular chalcogen bonds in catalysis. Dithieno[3,2-
b;2’,3’-d]thiophenes (DTTs), particularly their diimides, but
also wide-angle cyclopentadithiazole-4-ones are thus con-
firmed as privileged motifs to integrate chalcogen bonds into
functional systems in the broadest sense.
and 2 reduced the rate enhancement to kcat/kuncat = 6.6 and kcat
/
k
uncat = 3.9, respectively (Table 1, entries 15 and 16). Com-
pared to quinoline reduction, transition-state stabilizations
were smaller for this transformation, presumably because the
nitrogen lone pair in imine 12 is less accessible and less basic.
The weakest catalyst of the series was 5,5’-dicyano-2,2’-bi-
Acknowledgements
We thank M. Macchione and Q. Verolet for contributions to
synthesis, the NMR and the Sciences Mass Spectrometry
(SMS) platforms for services, and the University of Geneva,
the Swiss National Centre of Competence in Research
(NCCR) Molecular Systems Engineering, the Swiss NCCR
Chemical Biology, and the Swiss NSF for financial support.
1,3-thiazole 6 (Figure 2). A rate enhancement of kcat/kuncat
=
1.8 was found for the reduction of imine 12 (Table 1,
entry 19). The structural motif was of interest also because
it is present in the light emitting oxyluciferin that accounts for
the bioluminescence of fireflies, and in several macrocyclic
natural products.[1] The negligible catalytic activity found for 6
reflected potent self-inhibition expected from intramolecular
Conflict of interest
1,4 S N chalcogen bonds (Figure 2a).[1,2] Attempts to
À
increase catalytic activity by additional cation binding to
lock the two nitrogen atoms on one side were so far not
successful. However, almost as poor activity found for 5,5’-
bithiazole 5 indicated that a correct bite angle for transition-
state stabilization in the focal point of the two s holes is
presumably even more important (kcat/kuncat = 2.1, Table 1,
entry 18; Figure 1d, Figure 2). Indeed, the ultralarge bite
angle with a tight carbonyl bridge in the cyclopenta[2,1-b;3,4-
b’]dithiazole-4-one 4 increased catalytic activity with regard
to imine reduction significantly, despite the absence of
strongly withdrawing cyano acceptors (kcat/kuncat = 5.6,
Table 1, entry 17). Quinoline reduction with the carbonyl-
bridged catalyst 4 was with kcat/kuncat = 125 even better than
with the DTT catalysts 2 and 3 but remained about 4-times
inferior to the most powerful catalyst 1 (Table 1, entries 1–4).
Intermolecular chalcogen bonding between sulfur and oxygen
has been observed previously for this intriguing wide-angle
motif 4 in the solid state.[10]
The authors declare no conflict of interest.
Keywords: chalcogen bonds · dithienothiophenes ·
homogeneous catalysis · transfer hydrogenation
[1] B. R. Beno, K.-S. Yeung, M. D. Bartberger, L. D. Pennington,
N. A. Meanwell, J. Med. Chem. 2015, 58, 4383 – 4438.
[2] A. Bauzꢁ, T. J. Mooibroek, A. Frontera, ChemPhysChem 2015,
16, 2496 – 2517.
[3] a) A. Kremer, A. Fermi, N. Biot, J. Wouters, D. Bonifazi, Chem.
Eur. J. 2016, 22, 5665 – 5675; b) P. C. Ho, P. Szydlowski, J.
Sinclair, P. J. W. Elder, J. Kubel, C. Gendy, L. M. Lee, H. Jenkins,
J. F. Britten, D. R. Morim, I. Vargas-Baca, Nat. Commun. 2016,
7, 11299.
[4] a) G. E. Garrett, E. I. Carrera, D. S. Seferos, M. S. Taylor, Chem.
Commun. 2016, 52, 9881 – 9884; b) N. A. Semenov, A. V. Lon-
chakov, N. A. Pushkarevsky, E. A. Suturina, V. V. Korolev, E.
Lork, V. G. Vasiliev, S. N. Konchenko, J. Beckmann, N. P.
Gritsan, A. V. Zibarev, Organometallics 2014, 33, 4302 – 4314.
[5] S. Benz, M. Macchione, Q. Verolet, J. Mareda, N. Sakai, S.
Matile, J. Am. Chem. Soc. 2016, 138, 9093 – 9096.
[6] a) V. B. Birman, X. Li, Org. Lett. 2006, 8, 1351 – 1354; b) S.
Fukumoto, T. Nakashima, T. Kawai, Angew. Chem. Int. Ed. 2011,
50, 1565 – 1568; Angew. Chem. 2011, 123, 1603 – 1606; c) C. A.
Leverett, V. C. Purohit, D. Romo, Angew. Chem. Int. Ed. 2010,
49, 9479 – 9483; Angew. Chem. 2010, 122, 9669 – 9673; d) E. R. T.
Robinson, D. M. Walden, C. Fallan, M. D. Greenhalgh, P. H.-Y.
Cheong, A. D. Smith, Chem. Sci. 2016, 7, 6919 – 6927.
[7] a) M. E. Cinar, T. Ozturk, Chem. Rev. 2015, 115, 3036 – 3140;
b) G. Barbarella, F. Di Maria, Acc. Chem. Res. 2015, 48, 2230 –
2241.
DTT diimides or “DDIs” 7 and 8 were synthesized[5,11,12]
to increase catalytic activity beyond DTT 1 (Figure 2).
Consistent with computational predictions in the gas phase
(Eint = À36.1 kcalmolÀ1, Table 1) and fluorescence measure-
ments for 1:1 chloride binding in solution (KD = 950 Æ 50 mm,
Figure S5, Table S1), DDI 8 accelerated the transfer hydro-
genation of quinoline 9a more than a thousand times (kcat
/
k
k
uncat = 1290, Table 1 entry 8). The rate enhancement kcat
uncat = 335 for the reduction of imine 12 identified DDI 8 was
/
more than 6 times above the previous best, that is, dicyano
DTT 1 (Table 1, entries 21, 14).
The less-activated DDI 7 with “sulfide” instead of
“sulfone” bridges between the two thiophenes remained
[8] a) G. Cavallo, P. Metrangolo, R. Milani, T. Pilati, A. Priimagi, G.
Resnati, G. Terraneo, Chem. Rev. 2016, 116, 2478 – 2601; b) S. H.
Jungbauer, S. M. Huber, J. Am. Chem. Soc. 2015, 137, 12110 –
12120.
remarkably active with regard to both quinoline (kcat/kuncat
=
500) and imine reduction (kcat/kuncat = 190, Table 1, entries 7,
20), possibly because of additional contributions from anion-
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
[9] a) A. Bruckmann, M. A. Pena, C. Bolm, Synlett 2008, 900 – 902;
b) W. He, Y.-C. Ge, C.-H. Tan, Org. Lett. 2014, 16, 3244 – 3247.
[10] Y. A. Getmanenko, C. Risko, P. Tongwa, E. Kim, H. Li, B.
Sandhu, T. Timofeeva, J. Bredas, S. R. Marder, J. Org. Chem.
2011, 76, 2660 – 2671.
[13] a) L. Liu, Y. Cotelle, A.-J. Avestro, N. Sakai, S. Matile, J. Am.
Chem. Soc. 2016, 138, 7876 – 7879; b) Y. Cotelle, V. Lebrun, N.
Sakai, T. R. Ward, S. Matile, ACS Cent. Sci. 2016, 2, 388 – 393.
[11] W. Hong, H. Yuan, H. Li, X. Yang, X. Gao, D. Zhu, Org. Lett.
2011, 13, 1410 – 1413.
Manuscript received: November 10, 2016
[12] See Supporting Information.
Final Article published: && &&, &&&&
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Homogeneous Catalysis
S. Benz, J. Lꢁpez-Andarias, J. Mareda,
N. Sakai, S. Matile*
&&&& — &&&&
Catalysis with Chalcogen Bonds
In the focal point: The integration of
chalcogen bonds into non-covalent cata-
lytic systems is realized with deep s holes
and wide bite angles between cofacial
endocyclic sulfur atoms of dithieno[3,2-
b;2’,3’-d]thiophenes (DTTs), ready to
activate hydride acceptors for transfer
hydrogenation of quinolines and imines
by lone-pair recognition in the transition
state.
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!