A New Amphiphilic Br?nsted Acid as Catalyst for the Friedel–Crafts Reactions of Indoles in Water

Article abstract of DOI:10.1002/ejoc.201801612

A new amphiphile 1a featuring with two -NHSO₂C₄F₉ sites is designed and synthesized as Br?nsted diacid. The amphiphilicity arose from nano-aggregations of hydrophobic fluorocarbon chains and extension of hydrophilic NH groups, confirmed by Tyndall effect and TEM. The acidity of 1a rivals to concentrated H₂SO₄, examined by the Hammett acidity function, ³¹P NMR and conductance titration. Compound 1a demonstrated excellent catalytic performance in the Friedel–Crafts alkylation of indoles in water. Reactive substrates, such as β-monosubstituted vinyl ketones in 1,4-addition and aldehydes in condensation, were quantitatively converted, and the products were easily isolated by filtration or extraction in 85–96 % yields. Additionally, there was no observation of effect of the acidity weakening of 1a on catalysis when poor-reactive substrates were used, such as β,β-disubstituted vinyl ketones in 1,4-addition and ketones in condensation. The two kinds of reactions were practised very smoothly. The formed nano-aggregations and the strong acidity of 1a were responsible for the highly efficient aqueous catalysis. Notably, 1a was recycled at least 3 times thanks to its amphiphilic structure. It was demonstrated the nano-aggregations played a key role in the aqueous synthesis.

Full text of DOI:10.1002/ejoc.201801612

DOI: 10.1002/ejoc.201801612
Full Paper
Acid Catalysis | Very Important Paper |
A New Amphiphilic Brønsted Acid as Catalyst for the Friedel–
Crafts Reactions of Indoles in Water
Yuan Cheng,[a] Xiongyu Ou,[a] Jimei Ma,[a] Linhao Sun,[a] and Zhong-Hua Ma*[a]
Abstract: A new amphiphile 1a featuring with two -NHSO2C4F9 were easily isolated by filtration or extraction in 85–96 % yields.
sites is designed and synthesized as Brønsted diacid. The Additionally, there was no observation of effect of the acidity
amphiphilicity arose from nano-aggregations of hydrophobic weakening of 1a on catalysis when poor-reactive substrates
fluorocarbon chains and extension of hydrophilic NH groups, were used, such as ꢀ,ꢀ-disubstituted vinyl ketones in 1,4-addi-
confirmed by Tyndall effect and TEM. The acidity of 1a rivals to tion and ketones in condensation. The two kinds of reactions
concentrated H2SO4, examined by the Hammett acidity func- were practised very smoothly. The formed nano-aggregations
tion, 31P NMR and conductance titration. Compound 1a dem- and the strong acidity of 1a were responsible for the highly
onstrated excellent catalytic performance in the Friedel–Crafts efficient aqueous catalysis. Notably, 1a was recycled at least 3
alkylation of indoles in water. Reactive substrates, such as ꢀ- times thanks to its amphiphilic structure. It was demonstrated
monosubstituted vinyl ketones in 1,4-addition and aldehydes in the nano-aggregations played a key role in the aqueous synthe-
condensation, were quantitatively converted, and the products sis.
Introduction
factant-combined catalysts and their applications in aqueous
systems. The surfactant segments assemble into micelles as
microreactor, whose hydrophobic cores trap the labile sub-
strates and exclude water molecules,[3b,5] while the micelle exte-
rior remains hydrated, and allows the permeation of protons for
substrate activation.[6] This surfactant-combined strategy has
also been extended to solid Brønsted acids.
Using water as a reaction media offers advantages for cost,
safety, and environment, and is therefore considered green and
ideal for laboratory and industry processes. Hence many efforts
have been devoted to the development of aqueous catalytic
reactions.[1] In aqueous suspension, some reactions involving
organic substrates are accelerated, which has been defined as
“on water” reactions by Sharpless.[1b] Several approaches have
been aimed to strengthen the interaction of organic substrates
under “on water” conditions, including the use of surfactants as
amphiphilic additives,[1d,1e] the use of phase-transfer catalysts,[2]
and the specifically functionalized design of the catalyst struc-
ture.[1e–1g] Water also brings some benefits for product isola-
tion. Various types of reactions forming C-C and C-heteroatom
bond have been carried out in water with enhanced reactivity
and selectivity, such as the nucleophilic attack of a carbanion
species in conjugate addition reaction, known as Michael-type
reactions.[1a,1c] For example, a Michael addition of ꢀ,ꢀ-disubsti-
tuted nitroalkenes with malonate derivatives was reported by
Song, in which the otherwise unreactive substrate systems were
demonstrated as on-water catalytic reaction.[1a]
The hydrophobic assembly is probably caused by fluorocar-
bon chain except hydrocarbon chain. Perfluoroalkyl chains have
a stronger hydrophobic character than hydrocarbon chains due
to their greater molecular cross-sectional area.[7] A classic exam-
ple is -C8F17, which is applied in fluorous biphasic media for
fluorous catalyst recycling.[8] Even relatively short -C4F9 chains
exhibit significant hydrophobicity. Our previous research dem-
onstrated that -C4F9 hindered access of water molecules to ad-
jacent SO2NHSO2 acid sites,[9] and hydrogen was impeded to
bind with the negative conjugate base SO2N–SO2.[10] Similarly,
in Wang's reports, pyrrolidine perfluorobutyl sulfonamide was
used in aqueous catalysis.[11] On the other hand, perfluoroalkyl
sulfonylimides, (RFSO2)2NH, known as nitrogen acids, is widely
applied in catalytic field instead of SO3H analogues to practice
challenging tasks because of their strong acidity.[3c,3d,9] Gener-
ally, compounds with high fluorine content (> 60 %) strongly
prefer to go into fluorous phase in fluorocarbon–hydrocarbon
systems.[12] These distinct properties could provide water-
shielded microenvironments for aqueous reactions. However,
too long fluorocarbon chain will inevitably lead to low miscibil-
ity of organic and fluorous substrates, weaken the interaction
of organic substrates, and increas the cost.
An effective promotion for nucleophilic addition is via the
use of Brønsted or Lewis acids to activate electrophiles.[3] How-
ever, the activating ability of Brønsted acids is commonly weak-
ened in presence of water. A few studies,[4] initiated by Kobaya-
shi's pioneering work,[3e] were developed on Brønsted acid-sur-
[a] Department of Chemistry, College of Sciences,
Huazhong Agricultural University
No.1, Shizishan Street, Hongshan District, Wuhan, 430070, China
E-mail: mazonghua@mail.hzau.edu.cn
Supporting information and ORCID(s) from the author(s) for this article are
available on the WWW under https://doi.org/10.1002/ejoc.201801612.
Herein, in the structure of designed sulfonimide catalyst 1a/
1b, fluorine content and acid density are well balanced by split-
ting long perfluoroalkyl chain into two relatively short perfluo-
Eur. J. Org. Chem. 0000, 0–0
1
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
roalkyl chains, attached to respective imide Brønsted acid sites -CONH-, which is possibly decomposed by the strong acidity,
(Scheme 1). This method adjusts fluorine content at relative low but stabilized by -C4F9 in 1a. It is inferred that the nonpolar
level with elevated acid density per molecule. The acidity is segments, including -C4F9 groups and phenyl ring, aggregate
examined by using the Hammett acidity function, the 31P NMR to protect -CONH- group. Especially in the presence of water,
shift of Et3P=O probe, and conductance titration. The binary due to the hydrophobic hydration effect, the aggregation splits
acid catalyst has been applied in water to promote the nucleo- each other's H-bond association of the water molecules near
philic addition (the Friedel–Crafts reactions). The substrates hydrophobic segments, and the bulk water thus hardly en-
used herein were indoles, a widely distributed core structure in croaches -CONH- group.[6a,16] The similar aggregation much less
nature, and there is a large number of applications of synthetic efficiently happens to the shorter -CF3 of 1b. 1a is thus a stable
indoles as pharmaceuticals and agrochemicals.[3a,13] With the amphiphile.
use of 1a, Friedel–Crafts alkylation of indoles in water was easily
performed, via 1,4-addition with ꢀ-monosubstituted vinyl ket-
ones and condensation with aldehydes, respectively. The prod-
ucts are isolated without column chromatography or recrystalli-
zation. The aqueous solution containing 1a is recyclable at least
three runs. Furthermore, typically poor-reactive ꢀ,ꢀ-disubsti-
tuted vinyl ketones were also activated by the strong acidity of
the catalyst, giving medium to good product yields.
A simple Tyndall effect trial was conducted to confirm the
formation of the hydrophobic aggregation (Figure 1a). The Tyn-
dall effect is often used as a measurement of a colloid, and the
intensity of the effect is proportional to the mean volume of
the particles. In our trial, obvious Tyndall phenomenon hap-
pened to 1a solution, showing the nano-level aggregation ex-
isted. As a comparison, no similar phenomenon was observed
for CF3SO3H solution. Furthermore, TEM images of 1a aqueous
solution is shown in Figure 1(b). A large amount of nano-size
aggregates was observed, with a size of 15–20 nm. Their rela-
tively small diameters implied a stacked hydrophobic segment
among molecules.
Figure 1. Tyndall effect (a) and TEM images of 1a. 10 mmol/L 1a and
12.5 mmol/L CF3SO3H aqueous solutions.
Scheme 1. The synthesis route and structure of acid catalyst 1.
The Acidity Analysis
Results and Discussion
The acidity of 1a was examined by the Hammett acidity func-
tion (H0), Gutmann method, and conductance titration. The re-
sults are summarized in Table 1. H0 was obtained by UV/Vis
spectrophotometry method, which is a semiquantitative meas-
ure of acidity used for concentrated solutions of strong acid.[10]
As shown, the H0 value of 1a was –0.52, definitely comparable
with tested –0.33 of concentrated H2SO4. The acidity was fur-
ther determined by the 31P NMR shift of the Et3P=O probe
according to the classical Gutmann method.[10,17] The probe
formed Et3P=O···H H-bond with acids. The stronger acidity, the
stronger hydrogen bond, and the lower electron density at the
phosphorus, and the larger 31P chemical shift. The chemical
shift difference (Δδ31P) of 1a/Et3P=O adducts to the reference
(C6H5)2ClP=O was 46.44, very close to 46.40 of H2SO4/Et3P=O
adducts.[10] These results are also supported by the initial con-
ductances of 1a and H2SO4 with the same concentrations,
Catalyst Design
Compounds 1a/1b comprise two N-H groups, respectively at-
tached to perfluoroalkyl groups (Scheme 1). Perfluoroalkyl
groups bring some advantages of stabilizing the anionic charge
of the conjugate bases due to its electron-withdrawing effect.
Furthermore, there is an extensive delocalization system over
N, O and phenyl ring, which conspire to increase the propensity
of dissociation of N–H bonds, largely improving the intrinsic
acidity and decreases the dependence on external solvation.[14]
An example is the analogue (CF3SO2)2NH, whose pKa values
were 2.8, 2.7 and 2.1, respectively in H2O, MeOH, and DMSO.
The values demonstrate the effect of the two factors on high
acidity.[15] Compounds 1a/1b is thus reasoned to serve as
strong acid.
1b was given by acidification of 1b·2Na, but decomposed which are both 0.136 ms/cm, meaning the close ionization abil-
when handled at ca. 55–65 °C to remove the remained solvent; ity of 1a and H2SO4. The values are obviously higher than
whereas 1a was obtained. The result suggests that a certain 0.111 ms/cm of oxalic acid. Also, the acid content of 1a is calcu-
linkage in 1b is more susceptible to the intrinsic acidity at lated as 2.49 mmol/g (vs. theoretic 2.62 mmol/g) based on the
higher temperature than that in 1a. The most likely linkage is titration curve, suggesting at least 95 % purity of 1a. These re-
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
2
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
sults demonstrate 1a can rival the acidity of concentrated with reduced 2.5 mol-% of 1a (Entry 10). Furthermore, 91 %
H2SO4.
yield occurred with the equimolar 2a and 3e (Entry 11).
With the optimized reaction conditions in hand, the general-
ity of the catalytic system was explored, and the results are
listed in Scheme 2. To our delight, the high catalytic perform-
ance of 1a was extended to various indoles with high func-
tional group tolerance. ꢀ-Monosubstituted vinyl ketones gave
86–98 % yields (4e–4p); while ꢀ, ꢀ-Disubstituted vinyl ketone
exhibited 78–85 % yields (4d, 4r–4t) and moderate 42 % yield
of 4q.
Table 1. The Hammett function, 31P NMR chemical shifts, and initial conduc-
tivity of acids.
Acid[a]
H0
δ31P [ppm] Initial conductivity [ms/cm]
1a
–0.52
–0.33
46.44
46.40
0.136
0.136
conc. H2SO4
Catalytic Michael Addition of Indoles with α,ꢀ-Unsaturated
Ketones in Water
Acid-catalyzed Michael addition reactions of indoles with α,ꢀ-
unsaturated carbonyl compounds construct 3-substituted ind-
oles, a type of important precursors in the synthesis of biologi-
cally active compounds and natural products.[18] Our initial at-
tempt involved the use of differently active substrates to react
with indole 2a, such as vinyl aldehyde (3a), 3-methyl-2-butenal
(3b), methyl acrylate (3c), 4-methyl-3-penten-2-one (3d), and 3-
penten-2-one (3e), in the presence of 1a. The results are shown
in Table 2.
Table 2. Reaction optimization of Michael addition of indole with various α,ꢀ-
unsaturated carbonyl compounds.[a]
Entry
3
2a/3
Solvent
Cat.
Temp.
Yield of 4
[mol/mol]
[mol-%] [°C]
[%]
[b]
1
2
3
4
5
6
7
8
3a
3b
3c
1:1
1:1
1:3
1:3
1:3
1:3
1:3
1:3
1:3
1:3
1:1
THF
THF
THF
THF
MeOH
EtOH
MeCN
H2O
H2O
H2O
H2O
2.5
2.5
5.0
5.0
5.0
5.0
5.0
5.0
5.0
2.5
2.5
0
0
–
–
[b]
50
50
50
50
50
50
50
30
30
NR[c]
80
49
48
61
82
6
3d
3d
3d
3d
3d
3e
3e
3e
Scheme 2. Examination of the substrate scope for Michael addition of α, ꢀ-
unsaturated ketone with substituted indoles. Reaction conditions: 2 mL of
aqueous solution of 1a (2.5 mol-%), 2 (0.50 mmol), 3e/3f (0.50 mmol) or 3d
(1.50 mmol), 30 °C for 4e–4p or 50 °C for 4d, 4q–4t, TLC analysis with petro-
leum ether/EtOAc.
9
10
11
62
91
[a] Reaction conditions: 2 mL of solution of 1a, indole (0.50 mmol), TLC analy-
sis with petroleum ether/EtOAc, 5 h for 3d and 3 h for 3e. [b] No purification.
[c] No reaction noted.
Various Brønsted acid catalysts were used to catalyze the re-
actions to get a mechanistic insight, and the results are summa-
rized in Table 3. Keeping loading the same mol of H+ into the
Active 3a and 3b were easily consumed, but no purification system, H2SO4 gave only 8 % and 22 % yields of 4k, in 3 h and
was conducted for products 4a and 4b, due to the complex 24 h, respectively; while CF3SO3H gave 26 % yield in 3 h (Entries
product system based on TLC analysis; while the reaction of 2–4). Interestingly, the micelle-type acid, o-C12H25C6H4SO3 H,
poor-reactive ester 3c was sluggish using up to 5.0 mol-% 1a gave 91 % yield in 2 h (Entry 5). The difference was insignifi-
(Entries 1–3). The reaction results of ꢀ,ꢀ-disubstituted vinyl ket- cantly relative to acid strength for ꢀ-monosubstituted vinyl
one 3d, which lacks reactivity attributed to the steric and elec- ketone. The acid strength, based on the H0 values, has the fol-
tronic nature of ꢀ-carbon, are given in entries 4–8. Solvent lowing order: CF3SO3OH (–3.52) > H2SO4 (–0.33) ≈ 1a (–0.52) >
screening demonstrated THF and H2O afforded comparable o-C12H25C6H4SO3H (ca. 0.44–0.58).[10] It was thus reasoned that
80 % and 78 % yields, superior to other solvents (48–61 %). Wa- the hydrophobic segments -C12H25 of o-C12H25C6H4SO3H
ter was thus preferred from the perspective of green chemis- formed micelles, locating the organic substrates and facilitating
try.[1a,1d] The relatively reactive vinyl ketone 3e furnished com- the reaction. Contrarily, H2SO4 and CF3SO3H lacked hydro-
plex products under the same conditions, and only 6 % yield of phobic segment, giving poor catalytic efficiency. The similar hy-
4e was given (Entry 9); whereas, it went up to 62 % at 30 °C drophobic nano-aggregations were formed in 1a for its amphi-
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
3
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
philic structure, leading to 98 % yield of 4k (Entry 1), slightly
higher than 91 % by o-C12H25C6H4SO3H. This probably sug-
gested less product residual in the nano-aggregations. The
nano-aggregations were confirmed by Tyndall effect and TEM
image (Figure 1). The fluorine content of 1a comes to 45 %, and
the lipophobicity of fluorocarbon chain promoted the separa-
tion of the product from the catalyst. Moreover, C4F9- groups
are adjacent to N-H sites, and the hydrophobic aggregations
can control the local polarity around active sites. The controlling
allows modulating the activity of catalysts.[1g,9] A homemade
acid, Cl(CH2)3SO2NHSO2C4F9, was used to support the idea.[10]
The acid gave slightly reduced yield (91 %), in which
ClCH2CH2CH2- instead of C4F9- led to the fluorine content down
to 39 %. Moreover, the product isolation is simple. The aqueous
system with 1a allowed obvious oily product floated on the
face (Figure 2, inset). A simple extraction using EtOAc was
enough to offer pure products, without chromatography and
recrystallization, which reduced the waste generation.
Table 3. Catalytic effect of some commercial and homemade acids on Michael
addition of α, ꢀ-unsaturated ketone with indole.[a]
Figure 2. Proposed enrichment and activation of α,ꢀ-unsaturated ketone by
hydrophobic segments of 1a in water. The inset is the experimental system.
Benefiting from its bulkiness, the aggregations stack into a
large-size nuclear hydrophobic zone; while the attached -NH
sites hang on the outer zone. -NH groups form H-bond with
the carbonyl group of enol ketone or competitive bulky water.
However, the fluorocarbon chains, along with phenyl ring, help
to resist bulk water, and the organic substrates was thus acti-
vated. Moreover, the ethylene linkage of enol ketones possibly
get accessible to the phenyl rings. The organic substrate is thus
enriched around 1a, facilitating the further reaction.
The recycling performance of 1a was investigated using 2a
and 3f to give 4k, and the results are in Table 4. CCl4 was used
to extract 4k, and the aqueous phase was fed with fresh 2a
and 3f for the next run. As shown, the catalyst could be reused
at least three runs with > 92 % yields. The extension of reaction
time was possibly owed to slight catalyst loss. In the following
cycle, the yield was down to 87 % in 12 h.
Entry
Cat.
3
Cat. Loading Time
Yield
[%]
[mol-%]
[h]
1
2
3
4
5
6
7
8
9
1a
H2SO4
H2SO4
CF3SO3H
o-C12H25C6H4SO3H
Cl(CH2)3SO2NHSO2C4F9
1a
3f
3f
3f
3f
3f
3f
3d
3d
3d
2.5
2.5
2.5
5.0
5.0
5.0
5.0
10.0
10.0
3
3
24
3
2
3
5
5
5
98
8
22
26
91
91
82
36
61
[b]
C12H25C6H4SO3H
[b]
Cl(CH2)3SO2NHSO2C4F9
[a] Reaction conditions: 2 mL of aqueous solution of 1a, 2a (0.50 mmol), TLC
analysis with petroleum ether/EtOAc. [b] Homemade acid.[10]
On the other hand, 1a-catalysed addition of ꢀ,ꢀ-disubsti-
tuted vinyl ketone 3d gave 82 % yield, much higher than 36 %
by C12H25C6H4SO3H and 61 % by Cl(CH2)3SO2NHSO2C4F9. 1a
and Cl(CH2)3SO2NHSO2C4F9, with more excellent acidity, acti-
vated 3d more efficiently than C12H25C6H4SO3H. 1a thus exhib-
ited the best catalytic performance.
In our experiments, the use of water as medium improved
the reaction yield and controlled the product solubility, which
simplified the product purification process in the reaction of
active ꢀ-monosubstituted substrate. The strong acidity of 1a
expanded the substrate scope to the ꢀ,ꢀ-disubstituted ketones,
to which o-C12H25C6H4SO3H was not effective due to lower
acidity. Furthermore, 1a is a dual Brønsted acid, and the catalyst
dosage was less (2.5 mol-% vs. 5 mol-%).
Table 4. Recycling results of 1a-catalyzed Michael addition.[a]
Run
Time [h]
Yield [%]
1
2
3
4
2.5
3
6
97
93
92
87
12
[a] Reaction conditions: 6 mL of aqueous solution of 1a (2.5 mol-%), 2a
(1.50 mmol), 3f (1.50 mmol), 30 °C, TLC analysis with petroleum ether/EtOAc.
ca. 1 mL of CCl4 was used to extracted the product.
Catalytic Condensation of Indoles with Aldehyde in Water
Based on the observations and literatures, a tentative enrich- Furthermore, the catalytic generality of 1a in water was investi-
ment and activation model of α,ꢀ-unsaturated ketone is pro- gated in condensation of indole with aldehyde. The product 3,
posed, as depicted in Figure 2. The fluorocarbon chains of 1a 3′-bis(indolyl)alkanes (BIAs), which were widely isolated from
tend to regroup and segregate from both the hydrophilic -NH various natural sources, exhibit a range of biological activi-
and lipophilic phenyl ring, and aggregate in a separate zone.[7b] ties.[13b] The investigations started with 4-anisaldehyde (5a) and
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
4
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
2a in water at 30 °C (Table 5). 2a was consumed completely in
15 minutes in the presence of 2.5 mol-% 1a, but with complex
products (Entry 1). A reduced amount of 1a (1 mol-%) produced
unsolvable sticky product (Entry 2), but contaminated by resid-
ual 2a and 5a. Probably, 1a worked highly efficiently, and BIA
precipitated quickly so that the reagents were wrapped in. A
trace of organic solvents was thus added (Entries 3, 4). As men-
tioned above, the hydrophobic aggregations were loose, and
thus water remains in contact with the hydrophobic chain to
some degree.[6a] The added organic solvents filled the inter-
space, and induced the hydrophobic zone becoming denser. A
little CHCl3 or EtOAc (5 %, v: v) was as expected, and provided
94 % and 91 % yields, respectively. It suggested most of the
remained substrate went into the hydrophobic aggregation.
EtOH was also used, only producing sticky product very quickly
(Entry 5). Yield of 79 % was obtained by column chromatogra-
phy in extended 4 h (Entry 6). This was reasonable considering
EtOH dispersed over the system because of the water-solubility.
Table 5. Optimization of the medium between 4-anisaldehyde and indole.[a]
Scheme 3. Examination of the substrate scope for Friedel–Crafts alkylation of
indoles with aldehydes in water. Reaction conditions: 2 mL of aqueous solu-
tion of 1a (1.0 mol-%), H2O: EtOAc = 95: 5 (v: v), 2 (0.50 mmol), 5 (1.00 mmol),
30 °C, TLC analysis with petroleum ether/EtOAc.
Entry
Cat. Loading
[mol-%]
Solvent
(v/v)
Time
[h]
Yield
[%]
[b]
1
2
3
4
5
6
2.5
1.0
1.0
1.0
1.0
0.5
H2O
H2O
0.2
0.5
1
0.5
0.2
4
–
–
[b]
H2O/CHCl3 = 95:5
H2O/EtOAc = 95:5
H2O/EtOH = 95:5
H2O/EtOH = 95:5
94
91
[b]
–
79
[a] Reaction conditions: 1a in 2 mL of H2O, 4-anisaldehyde (0.50 mmol), indole
(1.00 mmol), 30 °C, TLC analysis with petroleum ether/EtOAc. [b] No isolation.
Less toxic EtOAc was used in the exploration experiments of
substrate scope. The results are shown in Scheme 3. Simple
filtration for powder products or extraction for sticky products
were carried out with > 95 % purity, confirmed by NMR spectro-
scopy. Excellent yields of 87–99 % were given.
Similarly to above Michael addition, the strong acidity of 1a
expanded the catalytic spectrum to the unactived ketones in
the condensation reactions (Scheme 4). Acetone and butanone
gave satisfactory 74–94 % yields (8a–8d, 8f–8i). The strong
acidity of 1a successfully activated the aliphatic ketones. Some-
what surprising, only moderate yield (42 %) of 4e was obtained.
Various Brønsted acid catalysts were also used to catalyze
the reactions, shown in Table 6. Under the similar conditions,
1a gave 91 % yield of 6a, 54 % yield for o-C12H25C6H4SO3H and
81 % yield for Cl(CH2)3SO2NHSO2C4F9, but no detectable yield
for H2SO4 (Entries 1–4). The formation of hydrophobic location
of the catalysts played a crucial role. Additionally, 1a exhibited
more excellent catalytic ability than the other two due to its
acidity superiority both in the reactions of anisaldehyde and
acetone (Entries 5–7).
Scheme 4. Examination of the substrate scope for Friedel–Crafts alkylation of
indoles with ketones in water. Reaction conditions: 0.6 mL of H2O, 1a 4.0 mol-
%, 2 (1.00 mmol), 7 (2.00 mmol), 30 °C, TLC analysis with petroleum ether/
EtOAc.
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
5
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
LC-MSD-Trap-XCT instrument. IR spectra were recorded on a FT-IR
Bruker (VERTEX 70) using liquid film technology. Conductometric
titration was performed on a DDS-11A (Shanghai Hongyi). TEM was
performed with a JEM-2100PLUS with an acceleration voltage of
200 kV.
Table 6. Catalytic effect of some commercial and homemade acids on Friedel–
Crafts alkylation of anisaldehyde/acetone with indole.[a]
Synthesis of N-[(Perfluorobutyl)sulfonyl]-3-{N-[(perfluorobutyl)-
sulfonyl]sulfamoyl}benzamide (1a): 3-(chlorosulfonyl)benzoyl
chloride was synthesized according to
a
previously de-
scribed method.[19] (Trichloromethyl) benzene (20 mL, 140 mmol)
was added dropwise into stirred ClSO3H (184 mL, 280 mmol) over
3 hours at 120 °C under nitrogen atmosphere. After stirring addi-
tional 4 hours, the remained ClSO3H was removed by vacuum distil-
lation (1 Torr, 30 °C), and then the product was collected at 120 °C
(1 Torr) as bright yellow liquid in 72 % yield (24.14 g). 1H NMR
(CDCl3, 600 MHz): δ = 8.76 (t, 1 H, Ar-H, 4J = 1.8 Hz), 8.50–8.48
Entry
Cat.
6a/8a Cat. Loading Time
Yield
[%]
[mol-%]
[h]
3
4
1
2
3
4
5
6
7
1a
H2SO4
6a
6a
6a
6a
8a
8a
8a
1
1
2
2
4
8
8
0.5
7
1
0.5
1
6
91
< 1 %
54
81
80
(dddd, 1 H, Ar-H, J = 7.8, J = 1.2 and 1.8 Hz), 8.37–8.35 (dddd, 1
3
4
3
H, J = 7.8, J = 1.2 and 1.8 Hz), 7.86 (t, 1 H, Ar-H, J = 7. 8 Hz). IR
(liquid film [cm–1]): 1753 (cc=o), 1385, 1173 (cs=o).
o-C12H25C6H4SO3H
Cl(CH2)3SO2NHSO2C4F9
1a
o-C12H25C6H4SO3H
Cl(CH2)3SO2NHSO2C4F9
Et3N (12.12 g, 120 mmol) was added dropwise into the stirred mix-
ture of 3-(chlorosulfonyl)benzoyl chloride (5.74 g, 24 mmol) and
C4F9SO2NH2 (14.71 g, 49 mmol) in 120 mL of dichloroethane. After
stirring additional 48 h at 70 °C, filtration was conducted in order
to remove the produced Et3N·HCl. To the filtrate was added 0.5 mol/
L H2SO4 to neutralize excess Et3N. Extracted with dichloroeth-
ane(3 × 10 mL), the combined organic extracts were washed with
5 % Na2CO3 solution and brine, dried with anhydrous Na2SO4, fil-
tered and concentrated. The residue was purified by chromatogra-
phy on silica gel (PE/EtOAc = 1:3, and then EtOAc/acetone = 95:5).
The salt 1a·2Et3N, was obtained as a yellow sticky liquid in 74 %
yield (17.09 g). 1H NMR (600 MHz, DMSO): δ = 8.83 (br. s, 2 H, 2NH),
8.34 (m, 1 H, Ar-H), 8.03–8.02 (m, 1 H, Ar-H), 7.85–7.83 (m, 1 H, Ar-
H), 7.50 (t, 1 H, Ar-H, 3J = 7.8 Hz), 3.12–3.08 (m, 12 H, NCH2), 1.17
(t, 18 H, CH2CH3, 3J = 7.2 Hz). 19F NMR ([D6]DMSO, 564 MHz): δ =
–80.37 (t, 3 F, CF3, 3J = 9.6 Hz), –80.43 (t, 3 F, CF3, 3J = 9.6 Hz),
–113.05 (t, 2 F, COCF2, 3J = 13.5 Hz), –113.54 (t, 2 F, SO2CF2, 3J =
13.5 Hz), –120.99- –121.03 (m, 2 F, CF2), –121.04- –121.06 (m, 2 F,
CF2), –125.57- –125.62 (m, 2 F, CF2), –125.70- –125.75 (m, 2 F, CF2).
13C NMR ([D6]DMSO, 100 MHz): δ = 169.2 (s, C=O), 145.5, 138.2,
131.5, 129.0, 128.4, 126.9 (all s, aromatic C), 122.1–105.7 (m, C-F, due
62
52
4
[a] Reaction conditions: 2 mL of aqueous solution, H2O: EtOAc = 95: 5 (v: v),
2a (0.50 mmol), 5a (1.00 mmol); or 0.6 mL of H2O, 2a (1.00 mmol), 7a
(2.00 mmol), 30 °C, TLC analysis with petroleum ether/EtOAc.
In our catalytic reactions, the careful purification by column
chromatography and recrystallization were not obligated. This
is significant, because synthetic chemistry encourages to reduce
the reaction waste. The advantage was attributed to the local
hydrophobic effect of the catalyst 1a. The hydrocarbon chain
helped to enrich the substrates, and the fluorocarbon chain re-
pelled away the excess water molecules and product molecules
due to its hydrophobicity/lipophobicity. The waste generated
from silica and toxic solvents was thus reduced. This is signifi-
cant for searching environmentally benign reactions and reduce
the E factor.[1d]
1
to the complicated coupling of C4F9 group, cannot calculate JC,F
Conclusions
2
and JC,F correctly), 46.2 (s, NCH2), 8.9 (s, CH3). HR MS [ESI(+)] calcd.
for C27H37F18N4O7S3 [M + H] 967.1531, found 967.1537.
In summary, we demonstrated a designed Brønsted acid 1a en-
ables Friedel–Crafts alkylation of indoles in water. The hydro-
phobic aggregations of fluorocarbon chains, along with phenyl
ring, of 1a protected acid sites from bulk water and enriched
the substrates. The amphiphilic structure has been proven cru-
cial for the high catalytic efficiency. The catalyst reasonably re-
duces the employment of the fluorocarbon number by splitting
the longer fluorocarbon chain into shorter, with strong acidity
and hydrophobicity maintained. The strong acidity of 1a
enabled the reactions of some unreactive substrates in water.
Anhydrous Na2CO3 (0.74 g, 7.0 mmol) was added to a solution of
1a·2Et3N (2.30 g, 2.4 mmol) in CH3OH (6 mL). The suspension was
stirred at 35 °C for 24 h. The excess of Na2CO3 was removed by
centrifugation, and the supernate was evaporated. After dried, salt
1a·2Na was obtained as off white powder in 96 % yield (1.85 g).[20]
Salt 1a·2Na (3.4 g, 4.2 mmol) was completely dissolved in 50 mL of
CH3CN for ionic exchange, with ca. 30 g of NKC-9 resin at the flow
rate of 1d/2s. The acidity of the eluent was detected by pH paper.
The solution was evaporated to give yellow 1a in 80 % yield (2.2 g).
1H NMR ([D6]DMSO, 400 MHz, δ): 11.76 (br. s, 2 H, NH with H2O),
3
8.37 (m, 1 H, Ar-H), 8.06 (d, 1 H, Ar-H, J = 7.6 Hz), 7.89 (d,1 H, Ar-
3
3
H, J = 8.0 Hz), 7.53 (t, 1 H, Ar-H, J = 7.6 Hz). 19F NMR ([D6]DMSO,
Experimental Section
376 MHz): δ = –80.51- 80.63 (m, 6 F, 2CF3), –113.14 (t, 2 F, COCF2,
General Remarks: (Trichloromethyl) benzene (99 %) was bought 3J = 13.5 Hz), –113.57 (t, 2 F, SO2CF2, 3J = 13.5 Hz), –120.99- –121.03
from Aladdin. [D6]DMSO, and CDCl3 were purchased from Alfa Aesar
Chemical Company. NMR spectra were recorded on Bruker AV-
400/600 spectrometers at room temperature in [D6]DMSO or CDCl3.
(m, 2 F, CF2), –121.04- –121.06 (m, 2 F, CF2), –121.13- –121.15 (m, 4
F, 2CF2), –125.72- –125.91 (m, 4 F, 2CF2). 13C NMR ([D6]DMSO,
100 MHz): δ = 169.3 (s, C=O), 145.7, 138.3, 131.8, 129.3, 128.8, 127.2
(all s, aromatic C), 122.4–106.1 (m, C-F, due to the complicated cou-
pling of C4F9 group, cannot calculate JC,F and JC,F correctly). HR
MS [ESI(–)] calcd. for C15H15F18N2O7S3 [M – H] 762.8977, found
1
Tetramethylsilane (TMS) served as internal standard for H and 13C
NMR; CFCl3 as internal standard for 19F NMR; Ph2ClP=O as internal
standard for 31P NMR spectroscopy. MS was recorded on an Agilent
1
2
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
6
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
762.8984; calcd. for C15H14F18N2O7S3 [M-2H] 380.9452, found
380.9480.
Keywords: Amphiphiles · Homogeneous catalysis ·
Perfluoroalkyl-sulfonylimide · Electrophilic substitution ·
Indole
Synthesis of N-Triflyl-3-(N-triflyl Sulfamoyl)benzamide (1b):
Please see supporting information.
UV/Vis Acidity Determination of 1a: The Hammett acidity func-
tion determination was based on the reported UV/Vis methods.[21]
The indicators, o-nitroaniline (400 nm, pKa = –0.29), was used in
the experiment. 1a and the indicator were dissolved in distilled
acetonitrile at concentrations of 0.02 mol/L and 2–30 × 10–5 mol/L,
respectively. The mixture was shaken overnight at 150 r/min. Then
their UV/Vis spectra were recorded on an UV-2450 spectrophotome-
ter. The calibration curves with the linearity higher than 0.999 in
the unprotonated form were constructed using distilled acetonitrile
solutions.
31P NMR Determination of Adducts of Et3P=O/1a: The 31P chemi-
cal shift were recorded using 5-mm coaxial NMR tubes with 1a/
Et3P=O 1:1 adduct in the external tube, and (C6H5)2ClP=O in the
inner tube as external reference. Et3P=O (0.01 g) was dissolved with
1 mL of CD3CN, and 0.5 mL of solution was taken. All the samples
were prepared in a glove box under nitrogen atmosphere. The data
were collected under static and 2.5 mm magic-angle spinning
(MAS) conditions at 25 °C. The proton decoupling method was used
to ensure accuracy and sensitivity, and the scanning number was
32.[10]
31P NMR Determination of Adducts of Et3P=O/1a: Conductomet-
ric titration was carried out by adding 4.70 × 10–3 mol/L NaOH solu-
tion to 2.0 × 10–4 mol/L solution of 1a and H2SO4. The electrode
was immersed in the solution for 1 min at room temperature, and
then the conductivity value was recorded. The inflection points sep-
arating the two distinct phases were determined as the intersec-
tions of linearly fitted conductivity data. Conclusive values were
obtained by averaging three parallel results.
[1] a) J. H. Sim, C. E. Song, Angew. Chem. Int. Ed. 2017, 129, 1861–1865; b)
S. Narayan, J. Muldoon, M. G. Finn, V. V. Fokin, H. C. Kolb, K. B. Sharpless,
Angew. Chem. Int. Ed. 2005, 44, 3275–3279; c) Y. Zhang, B.-W. Wei, H. Lin,
L. Zhang, J.-X. Liu, H.-Q. Luo, X.-L. Fan, Green Chem. 2015, 17, 3266–3270;
d) B. H. Lipshutz, F. Gallou, S. Handa, ACS Sustainable Chem. Eng. 2016,
4, 5838–5849; e) B. H. Lipshutz, S. Ghorai, Green Chem. 2014, 16, 3660–
3679; f) W. Guo, X. Liu, Y. Liu, C. Li, ACS Catal. 2018, 8, 328–341; g)
D. Singappuli-Arachchige, J. S. Manzano, L. M. Sherman, I. I. Slowing,
ChemPhysChem 2016, 17, 2982–2986; h) S. Xiao, F. Figge, G. Stirnemann,
D. Laage, J. A. McGuire, J. Am. Chem. Soc. 2016, 138, 5551–5560.
[2] H. M. Nelson, S. H. Reisberg, H. P. Shunatona, J. S. Patel, F. D. Toste, Angew.
Chem. Int. Ed. 2014, 53, 5600–5603.
[3] a) F. G. Cirujano, M. Stalpaert, D. E. D. Vos, Green Chem. 2018, 20, 2481–
2485; b) I. V. Vasilenko, F. Ganachaud, S. V. Kostjuk, Macromolecules 2016,
49, 3264–3273; c) T. Akiyama, K. Mori, Chem. Rev. 2015, 115, 9277–9306;
d) T. Akiyama, Chem. Rev. 2007, 107, 5744–5758; e) K. Manabe, S. Iimura,
X.-M. Sun, S. Kobayashi, J. Am. Chem. Soc. 2002, 124, 11971–11978.
[4] a) T. Chang, L. He, L. Bian, H. Han, M. Yuan, X. Gao, RSC Adv. 2014, 4,
727–731; b) S. Shirakawa, S. Kobayashi, Org. Lett. 2007, 9, 311–314.
[5] M. Lafantaisie, A. Mirabaud, B. Plancq, T. Ollevier, ChemCatChem 2014, 6,
2244–2247.
[6] a) J. A. Long, B. M. Rankin, D. Ben-Amotz, J. Am. Chem. Soc. 2015, 137,
10809–10815; b) B. Zhang, Z. Jiang, X. Zhou, S. Lu, J. Li, C. Li, Angew.
Chem. Int. Ed. 2012, 51, 13159–13162.
[7] a) V. H. Dalvi, P. J. Rossky, Proc. Natl. Acad. Sci. USA 2010, 107, 13603–
13607; b) M. P. Krafft, J. G. Riess, Biochimie 1998, 80, 489–514.
[8] X. Hao, O. Yamazaki, A. Yoshida, J. Nishikido, Tetrahedron Lett. 2003, 44,
4977–4980.
[9] H.-J. Zhang, Y. Cheng, H. Yuan, Y. Wang, Z.-H. Ma, Catal. Sci. Technol.
2017, 7, 4691–4699.
Catalytic Friedel–Crafts Alkylation of Indoles: Michael addition
with α,ꢀ-unsaturated ketones: To a stirred mixture of 1a (2.5 mol-
%) and α,ꢀ-unsaturated ketone (0.5 mmol) in 2 mL of H2O, indole
was added in one portion. The mixture was stirred at 30 °C and
monitored by TLC (PE/EtOAc = 5:1). Extracted with EtOAc (3 × 2 mL),
the combined organic extracts were dried with anhydrous Na2SO4,
and then rotary evaporated to afford a crude product. Column chro-
matography on silica gel (petroleum ether ethel/EtOAc = 6:1) was
needed only for the purification of the product from ꢀ,ꢀ-disubsti-
tuted α,ꢀ-unsaturated ketone. The obtained residue was identified
[10] L. You, X.-F. Ren, Y. Wang, Z.-H. Ma, Y. Gu, J.-Z. Ma, ChemCatChem 2016,
8, 3394–3401.
[11] L. Zu, H. Xie, H. Li, J. Wang, W. Wang, Org. Lett. 2008, 10, 1211–1214.
[12] a) Y. Koda, T. Terashima, M. Sawamoto, J. Am. Chem. Soc. 2014, 136,
15742–15748; b) P. Kirsch, Mordern Fluoroorganic Chemistry: Synthesis,
Reactivity, Applications, Wiley-VCH Verlag Gmbh& Co. KGaA, Weinheim,
2004.
[13] a) M. Shiri, Chem. Rev. 2012, 112, 3508–3549; b) M. Shiri, M. A. Zolfigol,
H. G. Kruger, Z. Tanbakouchian, Chem. Rev. 2010, 110, 2250–2293.
[14] a) C. D. Gheewala, B. E. Collins, T. H. Lambert, Science 2016, 351, 961–
965; b) A. Haas, C. Klare, P. Betz, J. Bruckmann, C. Krüger, Y.-H. Tsay, F.
Aubke, Inorg. Chem. 1996, 35, 1918–1925.
[15] N. N. Chipanina, I. V. Sterkhova, T. N. Aksamentova, L. V. Sherstyannikova,
V. A. Kukhareva, B. A. Shainyan, Russ. J. Gen. Chem. 2008, 78, 2363–2373
Engl. Transl.
[16] J. G. Davis, B. M. Rankin, K. P. Gierszal, D. Ben-Amotz, Nat. Chem. 2013,
5, 796.
1
with H NMR spectroscopy. The catalyst amount was based on the
mol of indole.
Condensation Reaction with Aldehydes/Ketones: The reactions
of indole (1 mmol) and aldehyde (0.5 mmol) were carried out at
30 °C with 1 mol-% 1a in 2 mL of H2O/EtOAc (95:5), monitored by
TLC (PE/EtOAc = 5:1). The product was obtained by direct filtration.
Additionally, ketones (2.0 mmol) was used with 4.0 mol-% 1a in
0.6 mL of H2O. Column chromatography on silica gel (petroleum
ether ethel/EtOAc = 6:1) was needed for the reactions of ketones.
[17] U. Mayer, V. Gutmann, W. Gerger, Monatsh. Chem. 1975, 106, 1235–1257.
[18] J. Ma, S. Ng, Y. Yong, X.-Z. Luo, X. Wang, X.-W. Liu, Chem. Asian J. 2010,
5, 778–782.
[19] D. W. Edward, N. Y. Lewiston, J. L. Robert, S. N. J. Pitman, United States
Patent Office 1966, US 3290370.
1
All the compounds were identified with H NMR spectroscopy. The
[20] Z.-H. Ma, H.-B. Han, Z.-B. Zhou, J. Nie, J. Mol. Catal. A 2009, 311, 46–53.
[21] C. Thomazeau, H. Olivier-Bourbigou, L. Magna, S. Luts, B. Gilbert, J. Am.
Chem. Soc. 2003, 125, 5264–5265.
catalyst amount was based on the mol of indole.
Acknowledgments
We gratefully acknowledge the Fundamental Research Funds
for the Central Universities (No. 2662015PY060).
Received: October 29, 2018
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
7
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Acid Catalysis
Y. Cheng, X. Ou, J. Ma, L. Sun,
Z.-H. Ma* .............................................. 1–8
A New Amphiphilic Brønsted Acid as
Catalyst for the Friedel–Crafts Reac-
tions of Indoles in Water
A designed Brønsted acid enables Frie- bulk water and enriched the sub-
del–Crafts alkylation of indoles in wa- strates. The amphiphilic structure has
ter, whose hydrophobic aggregations been proven crucial for the high cata-
of fluorocarbon chains, along with lytic efficiency.
phenyl ring, protected acid sites from
DOI: 10.1002/ejoc.201801612
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
8
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim