4,4'-azobis(halopyridinium) derivatives: Strong multidentate halogen-bond donors with a redox-active core

Article abstract of DOI:10.1002/chem.201103071

In summary, an easily accessible class of highly potent bis(pyridinium)-based multidentate XB donors has been introduced, which, in addition to their inherent XB acidity, also include a built-in redox option. In the previously established benchmark reaction to test the carbon-bromine bond-activation potential of XB donors, these compounds (2a, 2a', and 2b) are slightly more effective than bis(imidazolium)- based XB donors. Initially, the analysis of the overall process was complicated by the oxidation of bromide to elemental bromine by the azobis(halopyridinium) compounds during the reaction. The bromine thus formed was part of an additional activation process, which could, however, be suppressed by addition of cyclohexene. Two directions will be pursued in future work: since we demonstrated the activation capabilities of pyridinium-based XB donors towards carbon-bromine bonds, we plan to develop other types of activating reagents with redox-innocent bridging motives. Parallel to this, we strive to exploit the coupled oxidation/ XB-activation potential of the activators presented herein in suitable reactions.

Full text of DOI:10.1002/chem.201103071

DOI: 10.1002/chem.201103071
4,4’-Azobis(halopyridinium) Derivatives: Strong Multidentate Halogen-Bond
Donors with a Redox-Active Core
Florian Kniep, Sebastian M. Walter, Eberhardt Herdtweck, and Stefan M. Huber*[a]
The fact that certain terminal halogen substituents can act
as electrophilic centers has been known for almost 150
years.[1] Nevertheless, halogen bonds,[2] that is, the interac-
tions of such Lewis acids with nucleophiles, have only been
systematically investigated in the last 20 years.[3] While most
of these studies have been performed in the solid state,[4] in-
creased interest has recently been devoted to studies in solu-
tion, for example, to the design of halogen-bond-based
anion receptors.[5] Likewise, perfluorinated w-iodoalkanes
have been reported to catalyze the reduction of quinoline
derivatives by a Hantzsch ester, and the catalysis was postu-
lated to be based on a halogen bond.[6] For a reasonably
strong halogen bond (XB) to occur between a halogen-
tors with that of the imidazolium-based ones to find the best
basis for highly potent XB-based activators.
Starting from the known[13] halo-4-aminopyridine deriva-
tives 1a–1c, the symmetrical target molecules 2a–2c were
obtained by azo-homocoupling and subsequent methylation.
The asymmetrical derivative 2e was synthesized by coupling
1a with 1d (Scheme 1). For the azo coupling reactions, the
bond-donor[7] R X (X=Cl, Br, or I) and a nucleophile Nu,
À
the R group needs to be sufficiently electronegative. Al-
though polyfluorinated XB donors have often been used,
cationic backbones also provide the necessary electron with-
[8]
À
drawal to render the R X bond electrophilic. The high di-
À
rectionality of the resulting halogen bond (with R X···Nu
ꢀ1808) distinguishes this interaction from its sister noncova-
Scheme 1. Synthesis of 4,4’-azobis(pyridinium) compounds 2a–2e;
lent bond, the hydrogen bond.[9]
i) PhIL2ACTHNUGTRNEU(GN OTf)2 (1.5 equiv; OTf=triflate, L=4-dimethylaminopyridine
(DMAP)), CH2Cl2, 3days, RT; ii) MeOTf or Me3OBF4, CH2Cl2, 24 h, RT;
overall yields: 2a 72, 2a’ 65, 2b 63, 2c 54, and 2e 22%.
In the course of our investigations towards the application
of halogen bonds in organic synthesis and organocatalysis,
we recently reported the activation of the carbon–bromine
bond in benzhydryl bromide and related systems by mono-
and bidentate imidazolium-based XB donors.[10] There is
also evidence from X-ray structural analyses that halopyridi-
nium moieties form strong XBs with halides.[11] Consequent-
ly, in an effort to extend the XB-based activation of carbon–
heteroatom bonds to other multidentate XB donors, we set
out to synthesize previously unknown[12] azo-bridged bis(ha-
lopyridinium) compounds and investigate their activation
potential in the benzhydryl bromide based benchmark reac-
tion. The azo bridge was chosen as a short and easy to in-
stall linkage for the connection of two halopyridinium frag-
ments. One of our primary goals was to compare the halo-
gen-bond donor strength of the pyridinium-derived activa-
strong oxidant PhIL
by Weiss et al.,[14] was necessary in all cases, while the more
common coupling reagents NaOCl or PhI(OAc)2 gave only
2ACHTUNGERTN(NUNG OTf)2 (L=DMAP), first introduced
AHCTUNGTRENNUNG
low yields of the products under similar reaction conditions.
All compounds were obtained as pure trans isomers (see
Figure 1).[15]
The result of the X-ray structural analysis of 2a
(Figure 1)[16] shows that the pyridine moieties are orientated
almost perpendicular to each other. Each of the four iodine
substituents forms a strong XB to a triflate counteranion, as
indicated by the distances given in Figure 1, which are well
below the sum of the van der Waals radii (3.48 ꢀ).[17,18] The
O atom of the triflate, in turn, forms a second XB and thus
links two dications of XB donor 2a. In contrast, the bromine
substituents in XB donor 2b form only weak contacts with
an F atom in triflate (for the X-ray structural analysis, see
the Supporting Information).
[a] Dipl.-Chem. F. Kniep, S. M. Walter, Dr. E. Herdtweck,
Dr. S. M. Huber
Department Chemie, Technische Universitꢁt Mꢂnchen
Lichtenbergstraße 4, 85747 Garching (Germany)
E-mail: stefan.m.huber@tum.de
Subsequently, the azo compounds 2 were tested for their
potential to activate the carbon–bromine bond in benzhy-
dryl bromide (3).[10] Without any activating reagent, the
Ritter product 4 was only formed in negligible amounts
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201103071.
1306
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1306 – 1310

COMMUNICATION
Table 1. Activation of benzhydryl bromide (3) solvolysis by various acti-
vating reagents (in the presence of one equivalent of cyclohexene). See
also Table S1 in the Supporting Information.
Activating
reagent
A
Additive[b] Yield of 4
Hydrazine Hydrolysis
[%], 36 h [%], 36 h
(96 h)[d] (96 h)[e]
[%], 36 h
(equiv)[a]
A
N
ACHTUNGTRENNUNG
1
2
3
4
5
6
7
8
9
–
–
–
py
–
ꢁ5 (ꢁ5)
ꢁ5 (ꢁ5)
9 (19)
ꢁ5 (ꢁ5)
ꢁ5 (ꢁ5)
ꢂ95 (ꢂ95)
ꢁ5 (ꢁ5)
38 (49)
–
–
–
ꢁ5 (ꢁ5)
8 (11)
HOTf (1.0)[f]
2d (1.0)
2d (1.0)
Br2 (1.0)[f]
Br2 (1.0)
2b (0.2)[g]
2b (1.0)
ꢁ5 (ꢁ5)
–
31 (47)
9 (21)
–
58 (91)
py
py
py
py
py
py
-
py
py
py
py
py
py
py
–
23 (52)
ꢁ5 (ꢁ5)
ꢁ5 (ꢁ5)
ꢁ5 (ꢁ5)
ꢁ5 (ꢁ5)
ꢁ5 (ꢁ5)
ꢁ5 (ꢁ5)
ꢁ5 (ꢁ5)
16 (ꢁ5)
27 (30)
–
ꢁ5 (ꢁ5)
ꢁ5 (ꢁ5)
ꢁ5 (ꢁ5)
17 (28)
18 (26)
22 (31)
10 (20)
8 (16)
50 (50)
–
Figure 1. X-ray structural analysis of XB donor 2a (ellipsoids drawn at
50% probability; only two of the four XBs per dication are shown); se-
76 (ꢂ95)
10 2a (0.2)[g]
11 2a (1.0)
12 2a (1.0)
13 2a’ (1.0)
14 2c (1.0)
15 2c (2.0)[h]
16 2e (2.0)[h]
17 8a (1.0)
18 8b (1.0)
19 9 (1.0)[f]
40 (51)
À
À
lected bond lengths [ꢀ] and angles [8]: N1 N1 1.242(3), C5 I2 2.089(3),
ꢂ95 (ꢂ95)
93 (ꢂ95)
62 (81)
11 (32)
17 (44)
9 (10)
75 (89)
6 (14)
ꢁ5 (ꢁ5)
À
C2 I2 2.090(3), C2-I1-O1 176.37(9), C5-I2-O1 169.82(8), C5-C1-N1-N1
À58.8(3), C2-C1-N1-N1 127.2(2), C5-C1-C1-C2 À115.7(3); ꢄ =center of
inversion.
30 (34)
86 (90)
even after 96 h (Table 1, entry 1) and the same was true in
the presence of pyridine (either 10 or 20 mol%; Table 1,
entry 2; see also Table S1 in the Supporting Information).
Even one equivalent of HOTf induces only a comparatively
low yield of 4 (Table 1, entry 3 and Table S1, entry 3 in the
Supporting Information). Since we have already shown[10]
that small amounts of acid can be neutralized by pyridine,
we can rule out traces of acid as the actual activating re-
agent for all of the experiments below because 10 or
20 mol% pyridine were added in all relevant cases.[19]
ꢁ5 (ꢁ5)
–
–
11 (6)
ꢁ5 (ꢁ5)
[a] Equivalents of activating reagent relative to 3. [b] Additive: 0.1 equiv-
alents of pyridine (py; relative to 3). [c] Yield of 4 after 36 h (96 h) at
room temperature according to H NMR analysis (see the Supporting In-
formation). [d] Formation of the hydrazine corresponding to the respec-
tive azo compound (activating reagent) after 36 h (96 h) according to
1H NMR analysis, relative to 3 (see text). [e] Formation of side-products
caused by hydrolysis (benzhydrol and dibenzhydryl ether) after 36 h
(96 h) according to 1H NMR analysis; yield is given as the percentage of
1
3
that is consumed by hydrolysis. [f] Without cyclohexene additive.
In a first round of experiments, we investigated the reac-
tion of benzhydryl bromide with stoichiometric amounts of
the azo compounds 2a–2e in the presence of pyridine
(20 mol%; see Table S1 in the Supporting Information).
Under these conditions, a clear difference was observed be-
tween the nonhalogenated compound 2d (which gave a
minute yield of 4 after 36 h) and the halogenated derivatives
2a and 2b (75 and 82% yield, respectively, of 4 after 36 h,
see Table S1 in the Supporting Information). This seemed to
indicate that a (presumably XB-based) activation mecha-
nism was in place that was dependent on the halogen sub-
stituents in 2a and 2b, just as in the case of our imidazoli-
um-based XB donors.[10]
However, with the present class of XB donors, the situa-
tion is more complicated than in the case of the imidazoli-
um-based activators: without the addition of pyridine, amide
4 is formed in 69% yield in the presence of nonhalogenated
pyridinium salt 2d after 36 h (Table S1, entry 4 in the Sup-
porting Information), a yield that cannot be explained by
traces of acid alone. Further indications of a second activa-
tion mechanism were provided by single crystals obtained
from the reaction mixture of activator 2c with benzhydryl
bromide. X-ray structural analysis (Figure 2)[20] revealed
[g] With 20 mol% of cyclohexene additive. [h] With two equivalents of
cyclohexene additive.
them to be complexes of the dication of 2c with an anion of
2À [21]
À
the formal composition Br4
.
The two outer Br Br bonds
are markedly longer than the central one, and thus the
anion can be seen as the adduct of two bromide anions to
elemental bromine.[22] Stabilization of the terminal bromine
atoms is provided by a strong halogen bond with the iodo-
pyridinium moiety (sum of the van der Waals radii:
3.83 ꢀ)[17] and possibly a weak hydrogen bond with one of
the protons of a neighboring methyl group (sum of the van
der Waals radii: 3.05 ꢀ).[17]
The source of the elemenal bromine found in the crystal
became apparent when the hydrazine species corresponding
to 2c (i.e., its reduced form) was detected in the reaction
mixture by ESI MS and NMR spectroscopy. It appears that
a redox reaction between the HBr formed during the reac-
tion and the azo compound had taken place.[23] Hydrazine
species were also detected in all other benzhydryl test reac-
tions, albeit in different amounts (see Table S1 in the Sup-
porting Information).
Chem. Eur. J. 2012, 18, 1306 – 1310
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1307

S. M. Huber et al.
HBr to bromine.[25] Thus, the observed formation of product
4 in the first round of experiments was based partly on acti-
vation by the azo compounds and partly on activation by
Br2, and the extent to which both contributed to the overall
process is unknown. It is noteworthy, however, that some
observations already seemed to indicate that the activation
by 2a and 2b was largely azobis(halopyridinium) based
(e.g., the fact that virtually no hydrazine was formed in the
reaction with 2b).
Further tests established that the addition of one equiva-
lent of cyclohexene to the reaction mixture suppressed the
bromine-based activation of benzhydryl bromide (Table 1,
entry 7), providing an opportunity to detect the genuine azo-
bis(pyridinium)-based activation. Thus, in an effort to mea-
sure the activation potential of the azo compounds 2 and to
compare their XB-donor strength with our previous XB
donors,[10] we started a second round of experiments (see
Figure 3 and Table 1). This time, one equivalent of cyclohex-
Figure 2. X-ray structural analysis of a complex of the dication of 2c with
2À
(formally) Br4 (ellipsoids drawn at 50% probability); selected bond
À
À
lengths [ꢀ] and angles [8]: N1 N1’ 1.238(5), C2 I1 2.100(4), C2-C1-N1-
2À
À
N1’ 180.0, C2-C1-C1’-C2’ 180.0; exact distances within Br4 : Br2 Br2
À
2.4257(6), Br1 Br2 3.0185(6); ꢄ =centers of inversion.
To investigate this side reaction further, compound 2d
was treated with one equivalent of HBr and some additives
in CD3CN (Scheme 2). After several hours, HBr was quanti-
Figure 3. Yield versus time profile of selected reactions from Table 1 (y
axis: yield of 4 based on 1H NMR spectroscopy). Numbers for the indi-
vidual lines correspond to entries in Table 1. For further graphs, see the
Supporting Information.
ene and 10 mol% of pyridine were added to all test reac-
tions. A marked difference was again found between the ref-
erence compound 2d (which gave a negligible yield of 4
after 36 and 96 h) and the brominated or iodinated deriva-
tives 2b and 2a (76 and 93% yield, respectively, of 4 after
36 h; see Table 1, entries 9 and 12). Since 2a, 2b, and 2d
differ only at the 2 and 6 positions of the pyridine moieties,
the halogen substituents are crucial for the activation, which
is, hence, very likely caused by halogen bonds. The fact that
the brominated derivative 2b is less active than the iodinat-
ed compound 2a is also in agreement with halogen-bond
theory.[2,3]
As expected, lower yields of 4 were obtained when only
20 mol%, instead of one equivalent, of 2b or 2a was used
(Table 1, entries 8 and 10). Curiously, exchanging the triflate
counteranion for a tetrafluoroborate anion (2a’) led to a
slightly reduced yield of 4 (Table 1, entry 13), although a
higher yield of 4 might have been expected due to the pres-
ence of a less coordinating counteranion (compared with tri-
flate).[10]
Scheme 2. Test reactions to investigate the oxidation of HBr by nonhalo-
genated azo compound 2d; all reactions were performed in CD3CN and
equimolar ratios of all compounds were used; py=pyridine.
tatively oxidized by 2d, which was evident by formation of
0.5 equivalents of hydrazine 5. The presence of one equiva-
À
lent of Br2 (which should form Br3 with the bromide) did
not suppress the reaction. Addition of one equivalent of pyr-
idine does not impede the oxidation of HBr itself, but com-
pletely inhibits the same reaction in the presence of bro-
mine.[24] Interestingly, no significant protonation of 2d was
observed with HOTf.
Comparison experiments (Table 1, entry 6) showed that
elemental bromine itself was also capable of activating benz-
hydryl bromide, and hence a second activation mechanism
was present, based on the oxidation of the reaction product
1308
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1306 – 1310

Halogen-Bond Donors
COMMUNICATION
The partly iodinated azo compounds 2c and 2e were orig-
inally synthesized to test the activation mode of XB donor
2a.[26] In all cases in which these activators were used, how-
ever, nonnegligible amounts of hydrolysis of benzhydryl bro-
mide to benzhydrol (6) and dibenzhydryl ether (7) were
found (Scheme 3), although the water content of the reac-
elemental bromine by the azobis(halopyridinium) com-
pounds during the reaction. The bromine thus formed was
part of an additional activation process, which could, howev-
er, be suppressed by addition of cyclohexene. Two directions
will be pursued in future work: since we demonstrated the
activation capabilities of pyridinium-based XB donors to-
wards carbon–bromine bonds,
we plan to develop other types
of activating reagents with
redox-innocent bridging mo-
tives. Parallel to this, we strive
to exploit the coupled oxida-
tion/XB-activation potential of
the activators presented herein
in suitable reactions.
Scheme 3. Hydrolysis products benzhydrol (6), dibenzhydryl ether (7) and reference activating reagents 8 and
9.
Experimental Section
tion solution was identical to that in all other cases, particu-
larly in those with 2a and 2b (which did not show compara-
ble hydrolysis, see Table 1).
Synthesis of 2a: Under an argon atmosphere, the oxidant PhIACHTUNGTRENNUNG(DMAP)2-
AHCTUNRTGEG(NUNN OTf)2 (0.89 g, 1.18 mmol, 1.50 equiv) was suspended in dichloromethane
The cause of this remarkable difference remains unclear.
The hydrazines of 2c and 2e, formed during the reaction,
might act as a base and thus might increase the nucleophilic-
ity of the traces of water that are present. In contrast, the
hydrazines of 2a and 2b might be less basic for steric rea-
sons. Due to the unexpected hydrolytic side reaction, no de-
finitive conclusions can be drawn from the comparison of 2c
and 2e with 2a and 2b, although it is remarkable that both
2c and 2e generate considerably less amide 4 than 2a and
2b after 36 h (even if two equivalents of the activating re-
agent are used).
(15 mL). A solution of 3,5-diiodo-4-aminopyridine (0.27 g, 0.78 mmol,
1.00 equiv) in dichloromethane (5 mL) was added, the mixture was
heated at reflux for 2 h, and then stirred at room temperature for 2 days.
After filtration, the deep-red solution was concentrated in vacuo and pu-
rified by flash column chromatography (hexanes/EtOAc, 5:1), yielding
(E)-1,2-bis(3,5-diiodopyridin-4-yl)diazene (0.45 g, 0.66 mmol, 85%).
Subsequently, again under an argon atmosphere, (E)-1,2-bis(3,5-diiodo-
pyridin-4-yl)diazene (0.25 g, 0.36 mmol, 1.00 equiv) was dissolved in di-
chloromethane (10 mL). After dropwise addition of methyl trifluorome-
thane sulfonate (0.16 mL, 1.45 mmol, 4.0 equiv), the red solution was
stirred at room temperature for 2 days. The precipitate was filtered off,
washed with dichloromethane (2ꢄ20 mL) and dried in vacuo, yielding 2a
(0.31 g, 0.30 mmol, 85%) as a red-brown solid.
For comparison, compounds 8a, 8b,[10] and 9[27]
(Scheme 3) were also employed under identical conditions
as potential activators. The bis(iodoimidazolium) derivative
8a (Table 1, entry 17) was slightly less active than 2a,
whereas the bis(bromoimidazolium) compound 8b (Table 1,
entry 18) was markedly less potent than 2b. Thus, overall,
the halopyridinium-based activators proved to be more
active in this benchmark reaction than the haloimidazolium
derivatives. In stark contrast, thiourea 9 does not induce any
product formation (Table 1, entry 19). Instead, a side reac-
tion takes place that consumes approximately 40% of 3. Al-
though the product of the side reaction could not be unam-
biguously identified, in all likelihood the thiourea compound
acts as a nucleophile towards 3. As a consequence, the pres-
ent Ritter-like reaction (Table 1) cannot be achieved with
thiourea-based noncovalent activation.[28]
For further experimental procedures and spectroscopic data see the Sup-
porting Information.
Acknowledgements
Our research is funded by the Fonds der Chemischen Industrie (Liebig
scholarship to S.M.H.), the Deutsche Forschungsgemeinschaft (DFG),
and the Leonhart–Lorenz Foundation. F.K. and S.M.W. thank the TUM
Graduate School. We thank Prof. Dr. Thorsten Bach and his group for
their great support.
Keywords: azo compounds · cations · halogen bonds ·
halogens · oxidation · solvolysis
In summary, an easily accessible class of highly potent
bis(pyridinium)-based multidentate XB donors has been in-
troduced, which, in addition to their inherent XB acidity,
also include a built-in redox option. In the previously estab-
lished benchmark reaction to test the carbon–bromine
bond-activation potential of XB donors, these compounds
(2a, 2a’, and 2b) are slightly more effective than bis(imida-
zolium)-based XB donors. Initially, the analysis of the over-
all process was complicated by the oxidation of bromide to
[1] a) F. Guthrie, J. Chem. Soc. 1863, 16, 239; for early reviews, see:
b) O. Hassel, C. Rømming, Quart. Rev. Chem. Soc. 1962, 16, 1;
c) H. A. Bent, Chem. Rev. 1968, 68, 587.
[2] Structure and Bonding, Vol. 126, Halogen Bonding: Fundamentals
and Applications (Eds.: P. Metrangolo, G. Resnati), Springer, Berlin,
2008.
[3] For selected recent reviews, see: a) P. Metrangolo, H. Neukirch, T.
Pilati, G. Resnati, Acc. Chem. Res. 2005, 38, 386; b) P. Metrangolo,
F. Meyer, T. Pilati, G. Resnati, G. Terraneo, Angew. Chem. 2008,
120, 6206; Angew. Chem. Int. Ed. 2008, 47, 6114; c) M. Fourmiguꢅ,
Chem. Eur. J. 2012, 18, 1306 – 1310
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1309

S. M. Huber et al.
Curr. Opin. Solid State Mater. Sci. 2009, 13, 36; d) A. C. Legon,
Phys. Chem. Chem. Phys. 2010, 12, 7736.
[16] Single-crystal X-ray structure analysis of compound 2a: red-brown
prism, [(C12H10I4N4)2+],
2ACTHNGUTERNNUG
[(CF3O3S)À], Mr =1016.00; monoclinic,
[4] For selected reviews, see: a) P. Metrangolo, G. Resnati, Chem. Eur.
J. 2001, 7, 2511; b) K. Rissanen, CrystEngComm 2008, 10, 1107;
c) L. Brammer, G. M. Espallargas, S. Libri, CrystEngComm 2008,
10, 1712; d) R. Bertani, P. Sgarbossa, A. Venzo, F. Lelj, M. Amati,
G. Resnati, T. Pilati, P. Metrangolo, G. Terraneo, Coord. Chem. Rev.
2010, 254, 677.
[5] a) A. Mele, P. Metrangolo, H. Neukirch, T. Pilati, G. Resnati, J. Am.
Chem. Soc. 2005, 127, 14972; b) M. G. Sarwar, B. Dragisic, S. Sagoo,
M. S. Taylor, Angew. Chem. 2010, 122, 1718; Angew. Chem. Int. Ed.
space group C2/c (no. 15), a=24.4159(3), b=9.0554(1), c=
14.7108(2) ꢀ, b=122.0621(5)8, V=2756.40(6) ꢀ3, Z=4, l
(MoKa)=
0.71073 ꢀ, m=4.750 mmÀ1 1calcd =2.448 gcmÀ3
T=173(1) K, F-
(000)=1872, crystal dimensions 0.20ꢄ0.25ꢄ0.36 mm; R1=0.0158
(2508, Io >2s(Io)), wR2=0.0384 (all 2546 data), GOF=1.119, 165
parameters, D1max/min =0.79/À0.51 eꢀÀ3. For detailed information see
the Supporting Information and CCDC-849763 contains the supple-
mentary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
ACHTUNGTRENNUNG
,
,
AHCTUNGTRENNUNG
´
2010, 49, 1674; c) E. Dimitrijevic, P. Kvak, M. S. Taylor, Chem.
Commun. 2010, 46, 9025; d) A. Caballero, N. G. White, P. D. Beer,
Angew. Chem. Int. Ed. 2011, 50, 1845.
[6] A. Bruckmann, M. A. Pena, C. Bolm, Synlett 2008, 900.
[7] For congruence with the hydrogen-bond nomenclature, the electro-
philic partner of the interaction is denoted as a halogen-bond donor,
contrary to its actual role as electron acceptor.
[8] See, for example: a) N. Kuhn, T. Kratz, G. Henkel, J. Chem. Soc.
Chem. Commun. 1993, 1778; b) R. Weiss, M. Rechinger, F. Hampel,
Angew. Chem. 1994, 106, 901; Angew. Chem. Int. Ed. Engl. 1994, 33,
893; c) A. J. Arduengo III, M. Tamm, C. J. Calabrese, J. Am. Chem.
Soc. 1994, 116, 3625; d) C. J. Serpell, N. L. Kilah, P. J. Costa, V.
Fꢅlix, P. D. Beer, Angew. Chem. 2010, 122, 5450; Angew. Chem. Int.
Ed. 2010, 49, 5322; e) N. L. Kilah, M. D. Wise, C. J. Serpell, A. L.
Thompson, N. G. White, K. E. Christensen, P. D. Beer, J. Am. Chem.
Soc. 2010, 132, 11893, and references therein.
[17] A. Bondi, J. Phys. Chem. 1964, 68, 441–451.
[18] If “polar flattening” is considered (S. C. Nyburg, C. H. Faerman,
Acta Crystallogr. B 1985, 41, 274), a value of 3.30 ꢀ is obtained.
[19] The addition of 10 or 20 mol% of pyridine represents a compromise
between the need to suppress side reactions and the slow consump-
tion of benzhydryl bromide by pyridine, reducing the achievable
yield of 4.
[20] Single crystal X-ray structure analysis of a complex of the dication
of 2c with Br42À: red plate, [(C12 H12 I2 N4)2+], 2[(Br)À], (Br2), Mr =
785.66; orthorhombic, space group Cmca (no. 64), a=6.9454(2), b=
11.8626(4), c=24.2662(8) ꢀ, V=1999.30(11) ꢀ3, Z=4, l
ACHTUNGTRENNUNG
0.71073 ꢀ, m=11.141 mmÀ1
, ,
1calcd =2.610 gcmÀ3
AHCTUNGTRENNUNG
(994, Io >2s(Io)), wR2=0.0424 (all 1003 data), GOF=1.147, 67 pa-
rameters, D1max/min =0.58/À0.51 eꢀÀ3. For detailed information see
the Supporting Information and CCDC-849765 contains the supple-
mentary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif..
[9] See: G. R. Desiraju, Angew. Chem. 2011, 123, 52; Angew. Chem. Int.
Ed. 2011, 50, 52, and references therein.
[10] S. M. Walter, F. Kniep, E. Herdtweck, S. M. Huber, Angew. Chem.
2011, 123, 7325; Angew. Chem. Int. Ed. 2011, 50, 7187–7191.
[11] See, for example: a) M. Freytag, P. G. Jones, B. Ahrens, A. K. Fisch-
er, New J. Chem. 1999, 23, 1137; b) L. Brammer, G. M. Espallargas,
H. Adams, CrystEngComm 2003, 5, 343; c) T. A. Logothetis, F.
Meyer, P. Metrangolo, T. Pilati, G. Resnati, New J. Chem. 2004, 28,
760; d) G. Mꢆnguez Espallargas, L. Brammer, P. Sherwood, Angew.
Chem. 2006, 118, 449; Angew. Chem. Int. Ed. 2006, 45, 435; e) F. F.
Awwadi, R. D. Willett, B. Twamley, J. Mol. Struct. 2009, 918, 116;
f) G. Mꢆnguez Espallargas, F. Zordan, L. A. Marin, H. Adams, K.
Shankland, J. van de Streek, L. Brammer, Chem. Eur. J. 2009, 15,
7554; g) K. Raatikainen, M. Cametti, K. Rissanen, Beilstein J. Org.
Chem. 2010, 6, 4, and references therein.
2À
[21] a) For (rare) examples of the Br4 anion, see: M. C. Aragoni, M.
Arca, F. A. Devillanova, M. B. Hursthouse, S. L. Huth, F. Isaia, V.
Lippolis, A. Mancini, H. Ogilvie, Inorg. Chem. Commun. 2005, 8,
79, and references therein; b) for the related I4 anion, see: A.
Abate, M. Brischetto, G. Cavallo, M. Lahtinen, P. Metrangolo, T.
Pilati, S. Radice, G. Resnati, K. Rissanen, G. Terraneo, Chem.
Commun. 2010, 46, 2724, and references therein.
2À
À
[22] Sum of the van der Waals radii for Br Br: 3.70 ꢀ, see Ref. [17].
[23] For previous precedence of HBr oxidation by azo compounds with
strongly electron-withdrawing substituents, see, for example: a) M.
Kobayashi, J. Chem. Soc. Jpn. 1953, 74, 968; for the reduction of
azobis(pyridinium) compounds, see, for example: b) J. E. Rockley,
L. A. Summers, Aust. J. Chem. 1981, 34, 2683; c) P. R. Ashton, C. L.
Brown, J. Cao, J.-Y. Lee, S. P. Newton, F. M. Raymo, J. F. Stoddart,
A. J. P. White, D. J. Williams, Eur. J. Org. Chem. 2001, 957.
[24] As expected, the addition of only 0.2 equivalents of pyridine only
seemed to suppress a similar fraction of the overall reaction.
[25] We found no indication of bromination of product 4 in the reaction
mixture.
[12] Only four Cl-, Br-, or I-substituted 4,4’-azobisACTHNUTRGNE(NUG pyridine) derivatives
are known [see: a) Z. Talik, Rocz. Chem. 1962, 36, 1313; b) R.-A.
Fallahpour, M. Neuburger, Helv. Chim. Acta 2001, 84, 715], none of
which features halogen substituents at the 2/2’ or 6/6’ positions.
[13] See, for example: a) S. Singh, G. Das, O. V. Singh, H. Han, Tetrahe-
dron Lett. 2007, 48, 1983 (for 1a); b) V. CaÇibano, J. F. Rodrꢆguez,
M. Santos, M. A. Sanz-Tejedor, M. C. CarreÇo, G. Gonzꢇlez, J. L.
Garcꢆa-Ruano, Synthesis 2001, 14, 2175 (for 1b); c) Y. M. Choi-Sle-
deski, R. Kearney, G. Poli, H. Pauls, C. Gardner, Y. Gong, M.
Becker, R. Davis, A. Spada, G. Liang, V. Chu, K. Brown, D. Collus-
si, R. Leadley, Jr., S. Rebello, P. Moxey, S. Morgan, R. Bentley, C.
Kasiewski, S. Maignan, J.-P. Guilloteau, V. Mikol, J. Med. Chem.
2003, 46, 681 (for 1c), and references therein.
[26] Orientating DFT calculations show the feasibility of bidentate coor-
dination of 2a to a single bromide anion (see the Supporting Infor-
mation).
[27] See, for example, a) Z. Zhang, P. R. Schreiner, Chem. Soc. Rev.
2009, 38, 1187; b) Ref. [28], and references therein.
[14] a) R. Weiss, J. Seubert, Angew. Chem. 1994, 106, 900–901; Angew.
Chem. Int. Ed. Engl. 1994, 33, 891–893; b) J. Seubert, Dissertation,
University of Erlangen–Nuremberg (Germany), 1995.
[15] When the neutral (nonmethylated) precursor of 2a was irradiated at
360 nm in benzene for 3 h, a second isomer, supposedly the cis form,
was formed according to TLC. However, within a few minutes it
had completely reverted to the trans isomer and thus could not be
isolated. In contrast, when compound 2a was irradiated at 360 nm in
CH3CN for several hours, no second isomer was detected.
[28] However, an SN1-type activation of benzhydryl bromide (and its de-
rivatives) with thiourea-based catalysts has been proposed in the re-
action of said compounds with enamines (i.e., stronger nucleophiles
than acetonitrile): A. R. Brown, W.-H. Kuo, E. N. Jacobsen, J. Am.
Chem. Soc. 2010, 132, 9286.
Received: September 30, 2011
Revised: November 25, 2011
Published online: January 4, 2012
1310
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1306 – 1310