Activation of Michael Acceptors by Halogen-Bond Donors

Article abstract of DOI:10.1055/s-0036-1591841

Extending earlier studies on iodine catalysis, experimental investigations show that various halogen-bond donors can also be employed to accelerate the Michael addition between trans -crotonophenone and indole. Solvent as well as counteranion effects have been analyzed, and kinetic and computational investigations provide additional insights into the mode of activation.

Full text of DOI:10.1055/s-0036-1591841

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
© Georg Thieme Verlag Stuttgart · New York
2017, 28, A–G
letter
A
en
D. von der Heiden et al.
Letter
Synlett
Activation of Michael Acceptors by Halogen-Bond Donors
Daniel von der Heiden
Eric Detmar
11 halogen-bond
donors
R–I (10 mol%),
solvent, 1–2 d
Ph
Ph
Robert Kuchta
O
+
N
Martin Breugst*
0
0
0
0
0-
0
0
3
0-
9
5
0
8-
8
5
8
N
O
• solvent effects
H
H
• counteranion effect
• kinetic investigations
• DFT calculations
Department für Chemie, Universität zu Köln, Greinstraße 4,
50939 Köln, Germany
mbreugst@uni-koeln.de
In memoriam Professor J. Peter Guthrie
Received: 08.09.2017
Accepted after revision: 03.11.2017
Published online: 01.12.2017
DOI: 10.1055/s-0036-1591841; Art ID: st-2017-b0676-l
Abstract Extending earlier studies on iodine catalysis, experimental
investigations show that various halogen-bond donors can also be em-
ployed to accelerate the Michael addition between trans-crotonophe-
none and indole. Solvent as well as counteranion effects have been ana-
lyzed, and kinetic and computational investigations provide additional
insights into the mode of activation.
Martin Breugst (left) studied chemistry at the Ludwig-Maximilians-
Universität (LMU) in Munich (Germany). He received his Ph.D. in physi-
cal-organic chemistry from the LMU Munich under the supervision of
Prof. Dr. Herbert Mayr in 2010. Afterwards, Martin moved to the Uni-
versity of California, Los Angeles as a Feodor-Lynen postdoctoral fellow
where he worked with Prof. Dr. Kendall N. Houk on different aspects of
computational organic chemistry. Since 2013, he has worked at the De-
partment of Chemistry at the University of Cologne as an independent
researcher supported by a Liebig scholarship of the Fonds der Che-
mischen Industrie and completed his habilitation in October 2017. His
research interests include noncovalent interactions and the elucidation
of reaction mechanisms. Daniel von der Heiden (right, M.Sc. from the
Westfälische Wilhelms-Universität in Münster) is currently pursuing his
Ph.D. with Martin Breugst and is funded through a scholarship of the
Fonds der Chemischen Industrie. Robert Kuchta (middle left, B.Sc. from
the University of Cologne) and Eric Detmar (middle right, B.Sc. from the
University of Cologne) are working as M.Sc. students in the research
Key words catalysis, DFT calculations, halogen bonding, kinetics, sol-
vent effects
Halogen bonding – the noncovalent interaction between
an electrophilic region of a halogen atom and a Lewis base –
has enjoyed a tremendous success over the last decade in
various fields of chemistry.1 Within the framework of or-
ganocatalysis, Bolm and coworkers were the first to employ
1-iodoperfluoroalkanes as catalysts for the reduction of 2-
substituted quinolines with a Hantzsch ester in 2008.2 Sys-
tematic investigations by Huber and coworkers revealed a
few years later that halogen-bond-based catalysts (typically
polydentate iodinated azolium ions or perfluorinated iodo-
benzene derivatives) can be employed for the activation of
carbon–halogen bonds (Scheme 1).3 Similar catalyst sys-
tems have subsequently also been used for the activation of
imines4 and carbonyl dienophiles (Scheme 1).5 Recently,
Takemoto and coworkers reported on a halogen-bond-cata-
lyzed cross-enolate coupling reaction that yields 1,4-dicar-
bonyls in high yields.6 Building on previous investigations
on the halogen-bond properties of iodoalkynes,7 Matsuzawa,
Takeuchi, and Sugita relied on electron-deficient iodo-
alkynes as effective catalysts for the activation of thioam-
ides.8
In the light of the success of various halogen-bond do-
nors in catalysis, we have recently analyzed different io-
dine-catalyzed Michael additions (e.g., Scheme 2) both ex-
perimentally and computationally.9 Based on our investiga-
tions, we were able to show that iodine acts as a halogen-
bond donor in these reactions and significantly lowers the
activation free energies of these transformations. A hidden
Brønsted acid catalysis,10 which is often used as an explana-
tion,11 could be ruled out experimentally.9b
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G

B
D. von der Heiden et al.
Letter
Synlett
in situ NMR spectroscopy and no alternate reaction prod-
ucts (e.g., double alkylation) could be detected under the
reaction conditions. As initially anticipated, the halogen-
bond donor C1–OTf is also catalytically active and affords 3
in 36 % yield after 44 h (Table 1). However, in comparison to
molecular iodine (Scheme 2, 93 % after 3 min), the halogen-
bond donor C1–OTf is significantly less reactive. Intrigued
by this unexpected low reactivity, we next tested the same
catalyst in various solvents (Table 1). Generally, almost no
background reaction (< 5 % conversion after 44 h) could be
observed in the absence of any catalyst.14 A catalytic effect
could be observed in most aprotic solvents, when 10 mol%
C1–OTf were used. However, almost no product formation
could be detected in DMSO and DMF. This effect might be
attributed to their higher Lewis basicities compared to the
other solvents.15 A similar effect had already been observed
for iodine catalysis9b and can be explained by a competing,
strong halogen bond between the catalyst and the solvent,
deactivating the catalyst. Surprisingly, methanol, a solvent
that resulted in a poor yield in the iodine-catalyzed reac-
tion, performs very well for C1–OTf (92 % yield). However,
all reactions are significantly slower than molecular io-
dine9b indicating that C1–OTf is a much weaker catalyst for
this transformation.
N
N
CF3
N
N
I
I
Oct
Oct
O
O
OTBS
OMe
(0.5 mol%)
THF, –78 °C, 12 h
O
+
OMe
Cl
OTf
C8H17
N
N
F3C
OMe
I
Ph
Ph
Ph
N
(5 mol %)
N
+
CD2Cl2, rt, 1h
O
Ph
H
OTMS
2 BArF
4
N
N
N
N
I
I
O
Oct
Oct
O
(20 mol%)
+
CD2Cl2, rt, 6 h
Scheme 1 Selected examples for halogen-bond-catalyzed reactions
(BArF4 = B[3,5-(CF3)2C6H3]4–).
previous studies (ref 9b):
I2 (5 mol%),
CH3CN, 3 min, 93 %
Table 1 Solvent Dependency of the Reaction between trans-Crotono-
phenone (1) and Indole (2) Catalyzed by the Iodobenzimidazolium Tri-
flate C1–OTfa
this work:
R–I (10 mol%),
solvent, 44 h
Ph
O
C8H17
O
+
Ph
N
N
N
H
H
I
N
2
3
1
OTf
CH3
Scheme 2 Michael addition between trans-crotonophenone (1) and
C1–OTf
(10 mol%)
indole (2) catalyzed by molecular iodine9b and different halogen-bond
Ph
O
O
donors
+
Ph
N
solvent, 23 °C, 44 h
N
H
H
1
2
3
Building on these findings, we wondered whether io-
dine can be replaced by other halogen-bond donors in these
reactions. Therefore, we have analyzed the catalytic poten-
tial of different monodentate halogen-bond donors for the
Michael reaction of Scheme 2 between trans-crotonophe-
none (1) and indole (2). We now report on the results of
these investigations and compare their catalytic activities
to that of molecular iodine.
Entry
Solvent
Yield (%)
1
Ph2O
97
97
95
93
87
80
75
75
67
36
7a
2
dioxane
EtOAc
CD3OD
C6D6
3
4
5
We started our experimental investigation with the io-
6
DCE
dinated benzimidazolium triflate C1–OTf in acetonitrile,12
a
solvent that worked best for iodine catalysis.9b We em-
ployed 10 mol% of the halogen-bond catalyst and moni-
7
CDCl3
CD2Cl2
THF
8
1
tored the reaction progress by H NMR spectroscopy with
9
respect to the internal standard SiEt4 (see the Supporting
10
11
12
CD3CN
d6-DMSO
DMF
Information for details).13
For all cases discussed below, the yields of the isolated
product at the end of the experiments (typically after 44 h)
were in very good agreement with those determined from
0a
a Yield determined by 1H NMR spectroscopy with respect to the standard SiEt4.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G

C
D. von der Heiden et al.
Letter
Synlett
In our investigations on iodine catalysis, we were able to
experimentally rule out catalysis by traces of Brønsted ac-
ids formed by decomposition of molecular iodine.9b Similar
experiments failed in this investigation, as weak bases such
as Na2CO3 are ineffective to deactivate acid traces and non-
nucleophilic bases lead to catalyst decomposition.5 Howev-
er, NMR-spectroscopic investigations reveal that the cata-
lysts are stable in the presence of the reactants (see the
Supporting Information) and cause a considerable shift in
catalyst
(10 mol%)
Ph
O
O
+
Ph
N
solvent, 23 °C, 44 h
N
H
H
1
2
3
C8H17
C8H17
CH3
CH3
N
N
N
N
I
H
I
Br
N
N
N
N
OTf
OTf
OTf
OTf
CH3
CH3
CH3
C3–OTf
CH3
1
the H NMR and 13C NMR spectra. At this point, we cannot
C1–OTf
87 %
36 %
C2–OTf
19 %
< 1 %
< 1 %
C4–OTf
not soluble
17 %
completely rule out the formation of very small amounts of
iodine by partial decomposition of the employed catalysts.
However, no change in color of the reaction mixture has
been observed which renders iodine formation unlikely.
As iodinated benzimidazolium salts are catalytically ac-
tive in the Michael addition between 1 and 2, we subse-
quently tested other halogenated structures for their activi-
ties in this reaction. Due to different solubilities of charged
species, we employed three different solvents (C6D6, CD3CN,
and CD3OD) for these experiments. The results are summa-
rized in Scheme 3 and Figure 1. In a control experiment, the
noniodinated benzimidazolium salt C2–OTf turned out to
be significantly less reactive and a slow product formation
could only be observed in C6D6. Furthermore, the employed
counteranions (tested as NaOTf, NBu4OTf, and NBu4NTf2
salts) were completely inactive in both benzene and metha-
nol. The replacement of the octyl substituent to the smaller
methyl group (C3–OTf) only slightly affects the catalytic ac-
tivity while the solubility in less polar solvents (e.g., ben-
zene) significantly decreases. As observed before,2,3g,16 the
brominated analogue C4–OTf is also catalytically active, but
the reaction is significantly slower (Figure 1). However,
there are also cases known in which the brominated struc-
tures are more reactive than their iodinated counterparts.3a
Surprisingly, the iodinated imidazolium ions C5 and C6
turned out to be almost unreactive in benzene and acetoni-
trile (< 12 % yield). Interestingly, these structures remain
catalytically active in methanol as also indicated in the ki-
netic experiments (C5–NTf2, Figure 1). When the heterocy-
clic structure is changed to a triazole system (e.g., in C7–
OTf), the reaction also takes place but is slightly slower.
Among neutral halogen-bond donors, electron-poor io-
doalkynes like C10 – excellent catalysts for the activation of
thioamides8 – are catalytically inactive in the Michael addi-
tion studied here. Similarly, the addition of perfluoroiodo-
benzene C11 only results in a very slow product formation.
In contrast, N-halogenated succinimides C8 and C9 are
highly active in this transformation and high conversions
are already obtained after several minutes (see the Sup-
porting Information for details). However, NMR spectro-
scopic investigations indicate in these cases that a signifi-
cant decomposition of the catalysts has taken place until
the end of that reaction and the dehalogenated succinimide
not soluble
20 %
89 %
93 %
64 %
C8H17
Dipp
Bn
N
N
N
N
I
I
I
N
N
N
H3C
X
X
OTf
CH3
Dipp
Ph
C7–OTf
not soluble
12 %
C5–OTf
< 1 %
–
C5–NTf2
< 5 %
12 %
C6–OTf
C6–BPh4
not soluble
< 1 %
not soluble
< 1 %
94 %
93 %
91 %
7 %
78 %
O
O
F
F
F
F
N
I
N
Br
O2N
I
F
I
O
O
C8
C9
C10
< 5 %
< 1 %
C11
86 %
91 %
80 %
75 %
91 %
81 %
< 5 %
< 1 %
< 1 %
not soluble
Scheme 3 Catalytic activities of different halogen-bond donors in the
reaction of trans-crotonophenone (1) and indole (2) in C6D6 (blue),
CD3CN (black), and CD3OD (green); Dipp = 2,6-diisopropylphenyl
might have formed. As a consequence, other pathways have
to be discussed for these species in addition to a halogen-
bond mechanism (see also below).
To get a better understanding of the underlying reaction
mechanism and mode of activation, we monitored the ki-
1
netics of the Michael addition in more detail by H NMR
spectroscopy and determined the reaction order in the hal-
ogen-bond catalyst. Therefore, a series of kinetic experi-
ments with variable catalyst concentrations in both C6D6
and CD3OD under otherwise identical conditions were
studied in these solvents. Plotting the logarithm of the ini-
tial rate against the logarithm of the catalyst concentration
resulted in linear correlations with a slope around 0.5.
While this deviation from unity could indicate the forma-
tion of dimeric species,17 another explanation might be
more likely. An unfavorable pre-equilibrium, which results
from the dissociation of the ion pair to yield the ‘free’ halo-
gen-bond donor, results in a slope of 0.5.18 When ion pair-
ing is not possible (e.g., for NIS (C8)), a reaction order close
to unity can be determined for the halogen-bond catalyst.
Therefore, the kinetic investigations indicate that one halo-
gen-bond donor is involved in the rate-limiting transition
state and support the initial assumption of a halogen-bond
activation.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G

D
D. von der Heiden et al.
Letter
Synlett
nificantly reduced the catalytic activity of C1–OTf in both
benzene and methanol solution. This finding is also in line
with the results of the kinetic analysis and indicates the for-
mation of ion pairs under the reaction conditions. This indi-
–
cates that the interaction between C1 and the BPh4 ion
could be stronger than that to the triflate ion.19
Table 2 Counteranion Effects of the Reaction between trans-Crotono-
phenone (1) and Indole (2) Catalyzed by Iodobenzimidazolium Salts
C1–X
C8H17
N
I
N
X
CH3
C1–X
(10 mol%)
Ph
O
O
+
solvent, 23 °C, 44 h
N
N
Ph
H
H
1
2
3
Entry
d6-Benzene (%) d3-CH3CN (%) d4-MeOH (%)
1
2
3
4
5
C1–OTf
C1–NTf2
C1–BPh4
C1–BF4
87
81
< 1a
63
67
37
93
90
17
89
88
< 1
< 1
< 1
< 1
C1–BArF
4
a No product formation observed in 1H NMR spectroscopy after 44 h.
Counterion effects have been observed before for other
reactions, but it seems that no general trends can be ob-
Figure 1 Conversion vs. time profiles for the halogen-bond-catalyzed
reaction between trans-crotonophenone (1) and indole (2) in CD3OD
(25 °C)
–
served: While the BF4 salt is only slightly more effective
than the OTf– salt in the activation of benzhydryl bromide,3c
the BArF – salt is significantly more active than the OTf– an-
4
Based on the results of the kinetic analysis and as differ-
ent counterions had been used in the preparation of the
azolium salts of Scheme 3, we decided to closer analyze
how the counteranion affects the catalytic properties of the
iodobenzimidazolium catalyst C1 as a model system. Be-
sides the initially employed trifluoromethylsulfonate (OTf–),
we also relied on other weakly coordinating anions like
alogue in a Diels–Alder reaction.5 In contrast, for the activa-
tion of 1-chloroisochromane, the OTf– salt is superior to
BArF – and BPh4– salts.3g
4
In addition to the experimental investigations described
above, we also calculated the influence of selected halogen-
bond donors on the activation free energies for this trans-
formation.20 For the computational investigation, we have
analyzed this reaction in the gas phase as well as in ben-
zene and methanol (using the IEFPCM continuum model).
In general, the calculated catalytic effects follow the same
trend in the gas phase and in solution. As expected from ex-
perimental studies on solvent effects of halogen bonds,17c,21
the calculated catalytic effects were typically largest in the
gas phase compared with their values in solution. This
again indicates that the solvent has an important influence
on the halogen bond, and the results obtained from gas-
phase calculations might overestimate the strength of this
noncovalent interaction. In addition, specific solvent–solute
interactions are not taken into account in continuum mod-
els. While all results are summarized in the Supporting In-
–
bis(trifluoromethylsulfonyl)amide (NTf2 ) and the borates
–
–
–
–
BF4 , BPh4 , and BArF (B[3,5-(CF3)2C6H3]4 ). The corre-
4
sponding salts were either synthesized from the corre-
sponding benzimidazole and Meerwein’s salt (Me3OBF4) or
by anion metathesis from C1–OTf. Three solvents with dif-
ferent polarities (C6D6, CD3CN, and CD3OD, Table 2) were
chosen for this analysis, and the results are summarized in
Table 2. While the reactivities of the different salts are al-
most comparable in benzene and methanol, no reactions
have been observed in acetonitrile with any anions other
–
–
than OTf–. In benzene and methanol, the NTf2 and BF4
–
counterions resulted in comparable yields, while the BPh4
salt turned out to be unreactive or significantly less reactive
in all solvents. Interestingly, the addition of BPh4 salts sig-
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G

E
D. von der Heiden et al.
Letter
Synlett
formation, for the sake of clarity, we will only discuss the
values obtained for benzene solution in the following.
These results are summarized in Table 3.
In all cases, very high activation free energies have been
calculated. The uncatalyzed reaction proceeds with an acti-
vation free energy of +37.9 kcal mol–1 in line with no back-
ground reaction being observed after prolonged reaction
times. Among the computationally employed halogen-bond
donors, F3C–I (C12), F5C6–I (C11), and the iodoalkyne C10
resulted in higher activation free energies compared to the
uncatalyzed reactions (+2.8 < ΔΔG‡ < +3.2 kcal mol–1). De-
spite the stabilizing nature of the halogen bond, the unfa-
vorable entropy term (–TΔS) compensates this favorable in-
teraction resulting in an overall unfavorable process. In line
with these findings, these compounds were also found to
be inactive in the experimental studies described above
(Scheme 3 and the Supporting Information).
In contrast, the iodinated azolium ions C3, C13, and C14
all resulted in a substantial reduction of the activation free
energies (–6.4 < ΔΔG‡ < –4.9 kcal mol–1). However, the cal-
culated activation free energies are still substantially large
indicating that the calculations might underestimate the
strength of the halogen bond.8 Qualitatively, the computed
results are in line with the long reaction times. Further-
more, the benzimidazole catalyst is also calculated to be the
most active iodinated azolium ion. By far, the most active
halogen-bond donor in the computational investigation is
N-iodosuccinimide which lowers the activation free energy
by 7.6 kcal mol–1. NIS (C8) is also the most active catalyst in
the experimental investigations. This value is also close to
the previously calculated catalytic activity of molecular io-
dine (ΔΔG‡ = 7.6 kcal mol–1) using a slightly different com-
putational method.9b However, in the case of NIS, a signifi-
cant decomposition to the nonhalogenated succinimide
was also observed at the end of the reaction. Therefore, we
additionally examined whether the transfer of an I+ atom to
the Michael acceptor 1 could be an alternate explanation
(Scheme 4). However, the dissociation of the corresponding
C–I (in the azolium salts) or N–I bond (in NIS) resulted in
highly endergonic intermediates (ΔG > 50 kcal mol–1,
Scheme 4). As a consequence, this alternative seems to be
less likely.
Table 3 Activation Free Energies (ΔG‡) for the Michael Reaction be-
tween trans-Crotonophenone (1) and Indole (2) in the Presence of Dif-
ferent Catalystsa
Ph
O
catalyst
O
+
Ph
N
H
N
H
1
2
3
Entry
1
Catalyst
ΔG‡
O2N
I
+41.1
+41.0
C10
F
F
F
F
I
2
F
C11
F3C
C12
I
3
4
+40.7
+37.9
uncatalyzed
CH3
N
I
5
6
+33.0
+32.5
N
CH3
C13
CH3
N
I
N
CH3
C14
CH3
H3C
H3C
I
N
N
N
O
O
ΔG
I
+
+
I
7
8
31.5
N
N
N
Ph
Ph
CH3
H3C
H3C
C3
1
C3, C13, or C14
C3: ΔG = +63.7 kcal mol–1
C13: ΔG = +67.8 kcal mol–1
C14: ΔG = +66.7 kcal mol–1
O
N
I
+30.3
Scheme 4 Calculated free energies for the transfer of an iodine atom
to the Michael acceptor 1 in benzene solution
O
C8
a In kcal mol–1; M06-2X-D3/aug-cc-pVTZ/IEFPCM(benzene)//M06-2X-D3/6-
311+G(d,p)/IEFPCM(benzene); aug-cc-PVTZ-PP was used for iodine.
In summary, we were able to show that molecular io-
dine can be replaced by other halogen-bond donors as cata-
lyst in the Michael reaction between trans-crotonophenone
and indole. These results show that neutral electrophiles
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G

F
D. von der Heiden et al.
Letter
Synlett
can be activated for a subsequent nucleophilic attack by
halogen-bond donors. Therefore, these findings will turn
out useful in the development of more active (and ideally
chiral) catalysts in the field of halogen bonding.
(9) (a) Breugst, M.; Detmar, E.; von der Heiden, D. ACS Catal. 2016,
6, 3203. (b) von der Heiden, D.; Bozkus, S.; Klussmann, M.;
Breugst, M. J. Org. Chem. 2017, 82, 4037.
(10) The concept of hidden Brønsted acid catalysis was originally
introduced by Hintermann: Dang, T. T.; Boeck, F.; Hintermann,
F. J. Org. Chem. 2011, 76, 9353.
(11) Togo, H.; Iida, S. Synlett 2006, 2159.
Funding Information
(12) C1–OTf was synthesized from 2-iodo-1-octlybenzimidazole
(1.0 equiv) and methyl triflate (1.2 equiv) as described in the
Supporting Information. 1H NMR (CDCl3, 300 MHz): δ = 0.86 (t,
3JHH = 6.3 Hz, 3 H), 1.26–1.43 (m, 10 H), 1.90 (quint, 3JHH = 7.5 Hz,
2 H), 4.16 (s, 3 H), 4.49 (t, 3JHH = 7.6 Hz, 2 H), 7.56–7.63(m, 2 H),
7.70–7.75 (m, 1 H), 7.82–7.87 (m, 1 H). 13C{apt} NMR (CDCl3, 75
MHz): δ = 14.1 (s), 22.6 (s), 26.7 (s), 29.1 (s), 29.2 (s), 29.3 (s),
31.7 (s), 36.9 (s), 50.6 (s), 112.0 (s), 112.7 (s), 13.4 (s), 120.6 (d,
1JCF = 320.4 Hz), 127.3 (s), 127.4 (s), 133.2 (s), 134.2 (s). IR (neat,
ATR): ṽ = 2928 (w), 2857 (w), 1479 (w), 1028 (s), 747 (m), 637
(s) cm–1. ESI-HRMS: m/z [M]+ calcd for C16H24N2I+: 371.0979;
found: 371.0976. Anal. Calcd for C17H24F3IN2O3S: C, 39.24; H,
4.65; N, 5.38. Found: C, 39.31; H, 4.91; N, 5.16.
Financial support from the Fonds der Chemischen Industrie (Liebig
scholarship to M.B. and Ph.D. scholarships to D.v.d.H.) is gratefully ac-
knowledged.
)(
Acknowledgment
We are grateful to the Regional Computing Center of the University of
Cologne for providing computing time of the DFG-funded High Per-
formance Computing (HPC) System CHEOPS as well as for their sup-
port. We thank Prof. Dr. Albrecht Berkessel and his group for support.
C5–OTf was obtained from 2-iodo-3-methyl-1-octylimidaz-
olium bromide (1.0 equiv) by anion metathesis with NaOTf
(1.5 equiv) in CH2Cl2/H2O. 1H NMR (CDCl3, 400 MHz): δ = 0.88
(t, 3JHH = 6.8 Hz, 3 H), 1.23–1.39 (m, 10 H), 1.84 (quint, 3JHH = 7.2
Hz), 3.95 (s, 3 H), 4.17 (t, 3JHH = 7.5 Hz, 2 H), 7.57 (d, 3JHH = 2.0 Hz,
1 H), 7.73 (d, 3JHH = 2.0 Hz, 1 H). 13C{apt} NMR (CDCl3, 125 MHz):
δ = 14.2 (s), 22.7 (s), 26.4 (s), 29.1 (s), 29.1 (s), 31.8 (s), 30.5 (s),
Supporting Information
Supporting information for this article is available online at
https://doi.org/10.1055/s-0036-1591841.
S
u
p
p
ortiInfogrmoaitn
S
u
p
p
ortioInfgrmoaitn
References and Notes
1
39.9 (s), 53.1 (s), 103.1 (s), 120.8 (q, JCF = 320 Hz), 124.9 (s),
126.7 (s). 19F{1H} NMR (CDCl3, 376 MHz) δ = –78.2 (s), –78.4 (d,
1JF13C = 320 MHz). IR (neat, ATR): ṽ = 3107 (w), 2926 (w), 2857
(w), 1568 (w), 1497 (w), 1456 (w), 1246 (s), 1223 (m), 1153 (s),
1028 (s), 756 (w), 665 (w), 637 (s) cm–1. ESI-LRMS: m/z (%) =
321.0 (100) [C5]+ 320.8 (20) [Na(OTf)2]–, 148.9 (100) [OTf]–. ESI-
HRMS [C5]+: m/z calcd for C12H22N2I+: 321.0822; found:
321.0821.
(1) (a) Halogen Bonding I; Metrangolo, P.; Resnati, G., Eds.; Springer:
Heidelberg, 2015. (b) Halogen Bonding II; Metrangolo, P.;
Resnati, G., Eds.; Springer: Heidelberg, 2015. (c) Cavallo, G.;
Metrangolo, P.; Milani, R.; Pilati, T.; Priimagi, A.; Resnati, G.;
Terraneo, G. Chem. Rev. 2016, 116, 2478.
(2) Bruckmann, A.; Pena, M. A.; Bolm, C. Synlett 2008, 900.
(3) For selected reviews, see: (a) Bulfield, D.; Huber, S. M. Chem. Eur.
J. 2016, 22, 14434. (b) Breugst, M.; von der Heiden, D.;
Schmauck, J. Synthesis 2017, 49, 3224. For selected examples,
see: (c) Walter, S. M.; Kniep, F.; Herdtweck, E.; Huber, S. M.
Angew. Chem. Int. Ed. 2011, 50, 7187. (d) Kniep, F.; Rout, L.;
Walter, S. M.; Bensch, H. K. V.; Jungbauer, S. H.; Herdtweck, E.;
Huber, S. M. Chem. Commun. 2012, 48, 9299. (e) Kniep, F.;
Jungbauer, S. H.; Zhang, Q.; Walter, S. M.; Schindler, S.;
Schnapperelle, I.; Herdtweck, E.; Huber, S. M. Angew. Chem. Int.
Ed. 2013, 52, 7028. (f) Castelli, R.; Schindler, S.; Walter, S. M.;
Kniep, F.; Overkleeft, H. S.; Van der Marel, G. A.; Huber, S. M.;
Codée, J. D. C. Chem. Asian J. 2014, 9, 2095. (g) Jungbauer, S. H.;
Huber, S. M. J. Am. Chem. Soc. 2015, 137, 12110.
(13) Typical Experimental Procedure
A stock solution (1.00 mL) containing the reactants 1 and 2 as
well as the internal standard SiEt4 was added to the pure cata-
lyst in an NMR tube. This resulted in approximate concentration
of 50 mM in the catalyst, 500 mM in the reactants, and 125 mM
1
in the standard. The course of the reaction was followed by H
NMR spectroscopy. If a reasonable amount of product was
observed in the final NMR spectrum, the reaction was deacti-
vated with 100 μL DMSO. Within 2 h, all volatile residues were
removed under reduced pressure, and the product was isolated
by column chromatography (SiO2, cyclohexane/EtOAc = 9:1
(v/v), Rf = 0.11) as an off-white solid.
Analytical Data for Compound 3
(4) (a) Takeda, Y.; Hisakuni, D.; Lin, C.-H.; Minakata, S. Org. Lett.
2015, 17, 318. (b) He, W.; Ge, Y.-C.; Tan, C.-H. Org. Lett. 2014, 16,
3244.
(5) Jungbauer, S. H.; Walter, S. M.; Schindler, S.; Rout, L.; Kniep, F.;
Huber, S. M. Chem. Commun. 2014, 50, 6281.
103–104 °C. 1H NMR (CD2Cl2, 300 MHz): δ = 1.42 (d, 3J = 6.9 Hz),
3.23 (dd, 3J = 8.5 Hz, 2J = 16.5 Hz, 1 H), 3.44 (dd, 3J = 5.4 Hz, 2J =
16.4 Hz, 1 H), 3.73–3.84 (m, 1 H), 6.97 (d, 3J = 2.2 Hz, 1 H), 7.04–
7.16 (m, 2 H), 7.31 (d, 3J = 8.0 Hz, 1 H), 7.41 (t, 3J = 7.5 Hz, 2 H),
7.52 (t, 3J = 7.4 Hz 1 H), 7.63 (d, 3J = 7.8 Hz, 1 H), 7.91–7.93 (m, 2
H), 8.12 (br s, 1 H). 13C{1H} NMR (CD2Cl2, 75.5 MHz): δ = 21.4 (s),
27.5 (s), 46.8 (s), 111.7 (s), 119.5 (s), 119.5 (s), 120.7 (s), 121.7
(s), 122.2 (s), 126.8 (s), 128.4 (s), 128.9 (s), 133.3 (s), 137.0 (s),
137.8 (s), 200.0 (s). IR (neat, ATR): ṽ = 3352 (m), 3057 (w), 2982
(w), 2961 (w), 2872 (w), 1667 (s), 1618 (w), 1593 (w), 1489 (w),
1445 (m), 1339 (m) 1215 (s), 999 (m), 745 (s), 60 (m) cm–1. ESI-
HRMS: m/z [M + H]+ calcd for C18H18NO+: 264.1383; found:
264.1384.
(6) Saito, M.; Kobayashi, Y.; Tsuzuki, S.; Takemoto, Y. Angew. Chem.
Int. Ed. 2017, 56, 7653.
(7) (a) Webb, J. A.; Klijn, J. E.; Hill, P. A.; Bennett, J. L.; Goroff, N. S.
J. Org. Chem. 2004, 69, 660. (b) Goroff, N. S.; Curtis, S. M.; Webb,
J. A.; Fowler, F. W.; Lauher, J. W. Org. Lett. 2005, 7, 1891.
(c) Lieffrig, J.; Jeannin, O.; Fourmigué, M. J. Am. Chem. Soc. 2013,
135, 6200. (d) Dumele, O.; Wu, D.; Trapp, N.; Goroff, N.;
Diederich, F. Org. Lett. 2014, 16, 4722.
(8) Matsuzawa, A.; Takeuchi, S.; Sugita, K. Chem. Asian J. 2016, 11,
2863.
(14) Slightly higher values (up to 10 %) could be detected for CDCl3
and CH2Cl2.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G

G
D. von der Heiden et al.
Letter
Synlett
(15) (a) Klæboe, P. Acta Chem. Scand. 1964, 18, 27. (b) Laurence, C.;
Gal, J.-F. Lewis Basicity and Affinity Scales: Data and Measure-
ment; John Wiley and Sons: Chichester, 2010. (c) Laurence, C.;
Graton, J.; Berthelot, M.; El Ghomari, M. J. Chem. Eur. J. 2011, 17,
10431.
(16) Walter, S. M.; Jungbauer, S. H.; Kniep, F.; Schindler, S.;
Herdtweck, E.; Huber, S. M. J. Fluorine Chem. 2013, 150, 14.
(17) (a) Farnham, W. B.; Dixon, D. A.; Calabrese, J. C. J. Am. Chem. Soc.
1988, 110, 8453. (b) Kuhn, N.; Abu-Rayyan, A.; Steimann, M. Z.
Anorg. Allg. Chem. 2003, 629, 2066. (c) Jungbauer, S. H.;
Schindler, S.; Herdtweck, E.; Keller, S.; Huber, S. M. Chem. Eur. J.
2015, 21, 13625.
(19) For an excellent review on weakly and noncoordinating anions
and their reactivties, see: Krossing, I.; Raabe, I. Angew. Chem. Int.
Ed. 2004, 43, 2066.
(20) Optimizations were performed with M06-2X-D3/6-311+G(d,p)
in the gas phase and solution (iefpcm for benzene and metha-
nol). Electronic energies were calculated with M06-2X-D3/aug-
cc-pVTZ either in the gas phase or in solution. The aug-cc-
pVTZ-PP pseudopotential was used for iodine. See the Support-
ing Information for details.
(21) (a) Sarwar, M. G.; Dragisic, B.; Salsberg, L. J.; Gouliaras, C.;
Taylor, M. S. J. Am. Chem. Soc. 2010, 132, 1646. (b) Walter, S. M.;
Kniep, F.; Rout, L.; Schmidtchen, F. P.; Herdtweck, E.; Huber, S.
M. J. Am. Chem. Soc. 2012, 134, 8507. (c) Robertson, C. C.; Perutz,
R. N.; Brammer, L.; Hunter, C. A. Chem. Sci. 2014, 5, 4179.
(d) Robertson, C. C.; Wright, J. S.; Carrington, E. J.; Perutz, R. N.;
Hunter, C. A.; Brammer, L. Chem. Sci. 2017, 8, 5392.
(18) (a) Rothenberg, G. Catalysis: Concepts and Green Applications;
Wiley-VCH: Weinheim, 2008. (b) Nam, E.; Alokolaro, P. E.;
Swartz, R. D.; Gleaves, M. C.; Pikul, J.; Kovacs, J. A. Inorg. Chem.
2011, 50, 1592.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G