Synthesis of 7H-indolo[1,2-a]quinolinium derivatives via oxidative photocyclization of 3H-indolium salts using blue LED photoreactor

Article abstract of DOI:10.1016/j.tet.2019.06.002

Oxidative photocyclization of 1-phenyl-2-(2-arylethenyl)-3,3-dimethyl-3H-indolium perchlorates leading to 7H-indolo[1,2-a]quinolinium salts was investigated. The synthesis was performed under Vis irradiation. Blue LEDs emitting at 454.5 nm were used as th

Full text of DOI:10.1016/j.tet.2019.06.002

Tetrahedron xxx (xxxx) xxx
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Synthesis of 7H-indolo[1,2-a]quinolinium derivatives via oxidative
photocyclization of 3H-indolium salts using blue LED photoreactor
*
ꢀ
Marta J. Sawicka , Krzysztof Lubkowski, Elwira K. Wroblewska, Alicja Dzienisz,
Jacek A. Soroka
West Pomeranian University of Technology, Szczecin, Faculty of Chemical Technology and Engineering, Department of Organic and Physical Chemistry, Al.
Piastow 42, Szczecin 71-065, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Oxidative photocyclization of 1-phenyl-2-(2-arylethenyl)-3,3-dimethyl-3H-indolium perchlorates lead-
ing to 7H-indolo[1,2-a]quinolinium salts was investigated. The synthesis was performed under Vis
irradiation. Blue LEDs emitting at 454.5 nm were used as the light source. The reaction progress was
followed by UV-Vis spectroscopy. The products were characterized by IR, 1H and 13C NMR spectroscopy.
Reaction time depended on the position of the UV-Vis absorption band of both the substrates and
products. The results revealed that the application of blue LED photoreactor makes the synthesis of 7H-
indolo[1,2-a]quinolinium easier and more convenient.
Received 2 March 2019
Received in revised form
30 May 2019
Accepted 3 June 2019
Available online xxx
Keywords:
Photochemistry
© 2019 Elsevier Ltd. All rights reserved.
Vis-irradiation
Oxidations
Polycyclic aromatic hydrocarbons
UV-Vis spectroscopy
1. Introduction
photocyclization leading to intermediate dihydrophenanthrene (2)
which in the presence of an oxidant, usually oxygen or iodine,
Derivatives of 7H-indolo[1,2-a]quinolinium salts belong to the
group of polycyclic heteroaromatics containing 3-azoniahexa-1,3,5-
triene system. They are known as precursors of very good sol-
vatochromic or fluorescent dyes [1e3]. Due to the properties of
these dyes, they have found a lot of applications, i.e. as probes in the
quantitative analysis of mixed solvents [4] or surface active agents
[5], in the determination of water in organic solvents [6], as well as
in the studies of diffusion processes in polymers [7].
A convenient and efficient way to obtain polycyclic aromatic
hydrocarbons and heteroaromatics containing hexa-1,3,5-triene
system is the reaction of oxidative photocyclization. This reaction
was used for preparation of many compounds with unique chem-
ical, optical and biological properties: helicenes and helicene-like
compounds [8,9], benzo-annulated carbazoles [10], dinaph-
thothiophenes [11], phenanthrenes and related compounds
[12e17]. The most representative example of oxidative photo-
cyclization is the synthesis of phenanthrene (3) from cis-stilbene
(1) (Scheme 1). The first step of the process is cis-stilbene
undergoes dehydrogenation [12].
During oxidative photocyclization of 3H-indolium salts, being
aza analogues of the hexa-1,3,5-triene system, a new bond is
created between the aromatic ring and the vinylene group (Scheme
2) [18e20]. Irradiation of the solution of 1-phenyl-2-(2-
arylethenyl)-3,3-dimethyl-3H-indolium perchlorates (4), leads to
dihydro intermediates (5), which are not isolable. The analogous
4,6-diphenyl-2-styrylpyridinium derivative is the only example of a
compound with the 3-azoniahexa-1,3,5-triene system whose
dihydro intermediate was successfully isolated [21].
With atmospheric oxygen, dihydro intermediates (5) form 5-
aryl-7,7-dimethyl-7H-indolo[1,2-a]quinolinium perchlorates (6).
The reaction is highly regioselective and gives products with good
yields [18e20].
Oxidative photocyclization is typically carried out with the use
of mercury lamps [8e20]. Many disadvantages and limitations are
assigned to this type of light source, including large size of mercury
arc lamps, their low shock resistance, low energetic yield, as well as
the environmental danger of potential mercury release [12]. Ul-
traviolet emission accounts for the largest part of the mercury lamp
output. Hence, special precautions must be taken to protect the
operator's eyes and skin. Significant amounts of thermal energy and
* Corresponding author.
E-mail address: msawicka@zut.edu.pl (M.J. Sawicka).
https://doi.org/10.1016/j.tet.2019.06.002
0040-4020/© 2019 Elsevier Ltd. All rights reserved.
Please cite this article as: M.J. Sawicka et al., Synthesis of 7H-indolo[1,2-a]quinolinium derivatives via oxidative photocyclization of 3H-indolium
salts using blue LED photoreactor, Tetrahedron, https://doi.org/10.1016/j.tet.2019.06.002

2
M.J. Sawicka et al. / Tetrahedron xxx (xxxx) xxx
naphthalene moiety. A photosensitive part of their structure, and
simultaneously the chromogen determining the lowest excitation
energy, is the 3-azoniahexa-1,3,5-triene system. The structures of
salts 4a-l and products 6a-l of their oxidative photocyclization are
presented in Scheme 3.
Scheme 1. Oxidative photocyclization of cis-stilbene.
2.1. The choice of an appropriate light source
Preliminary tests of oxidative photocyclization of compounds
4a-l were performed in 20 cm3 vessel LED photoreactor. Inside the
vessel, a sealed glass flask was placed with 5 cm3 of the reaction
mixture. LEDs usually emit in a quite narrow emission band, in
contrast to mercury lamps whose emission spectrum is rather
broad. Hence, it is very important to find LEDs with a proper
emission wavelength. According to the Kasha's rule, in a photo-
chemical reaction of organic compounds, interacting molecules are
in their lowest excited state. The positions of the long-wavelength
absorption band of the examined compounds, corresponding to the
lowest excitation energy, are in the range of 397-471 nm (Fig. 1a).
Therefore, two different LED photoreactors were used, emitting
violet and blue light. As shown in Fig. 1b, the diodes emission bands
were at 395.4 and 454.5 nm, respectively. Since both light sources
emitted light from the Vis region, the photoreactors were made
from glass and the reaction was carried out in ordinary glass
vessels.
To check which LED is a better light source, the 3H-indolium salt
4h was chosen as a model compound. Two methanolic solutions of
13.6 mmol dmꢀ3 of salt 4h were made and irradiated in the two
reactors. Perchloric acid was added to the reaction mixture in order
to ensure pH value lower than pKa of the substrate, since dissoci-
ation of the hydroxyl group in salt 4h was accompanied with a
distinct bathochromic shift of the absorption band (from 457 nm to
548 nm). The reaction progress was monitored using UV-Vis spec-
troscopy. After 7 h of irradiation, a reduction in the intensity of the
absorption band of compound 4h, as a result of the substrate
concentration decrease, was noticed in both reactors (Fig. 2),
however, it was more pronounced in the reactor with blue LEDs.
The experiments were repeated for compounds 4d and 4l, with
the absorption band positioned at shorter and longer wavelengths,
respectively, and in both cases, better results were obtained in the
blue LED reactor. It is in compliance with the fact that the range of
the wavelengths that the substrate absorbs should overlap with
that of the used light source emission. The photoreactor equipped
with LEDs emitting at 454.5 nm proved to be a better light source in
Scheme 2. Oxidative photocyclization of 3H-indolium salts [18].
certain amounts of ozone are also generated in the air around UV
light sources. Moreover, special quartz reaction vessels transparent
to the wavelengths longer than 200 nm are needed [12]. In addi-
tion, in some cases, the absorbance of UV light is thought to be the
cause of uncontrolled product polymerization, which reduces the
efficiency of the desired reaction [22]. Therefore, light emitting
diodes (LEDs) have recently become popular, relatively cheap and
convenient light sources for various synthetic and analytical pur-
poses, as an alternative to the mercury arc lamps [23]. A small size,
a long lifetime associated with the strength of the LED, a high
electricity-to-light yield with little heating and therefore low en-
ergy consumption, operation on direct current instead of alter-
nating one, and the fact of being a “greener” mercury-free radiation
source, with almost complete recycling, are the main advantages of
LEDs application [23e25]. Typically, the use of LEDs is associated
with visible light transition metal photocatalysts [26e28], the
activation of visible light absorbing molecules for photosynthesis
[29,30], and the generation of singlet oxygen with organic dyes
[22,31,32], as well as with the removal of pollutants from air and
water [33e35].
In the previously presented studies of 3H-indolium salts
oxidative photocyclization [18e20], a tubular quartz flow reactor
equipped with a mercury lamp, an elliptic mirror, a self-evacuating
recirculation container [36] and a pump were used. The whole
equipment occupied an area of about 2 m3 and needed at least 1 dm
[3] of methanol as a solvent. Atmospheric oxygen was the only
oxidant in the dehydrogenation process [19,20]. The objective of
the currently presented experiments was the synthesis of 7H-
indolo[1,2-a]quinolinium perchlorates from 3H-indolium salts with
the use of LEDs as a light source, instead of a mercury lamp. To the
best of our knowledge, in available literature sources, there are no
examples of oxidative photocyclization, carried out using a LED
photoreactor. Furthermore, LED photoreactors used in the pre-
sented experiments were much smaller devices of simpler con-
struction. The proposed new reaction conditions allowed
elimination of all problems associated with the work with mercury
lamps and development of a new, more convenient and efficient
way for the preparation of 7H-indolo[1,2-a]quinolinium derivatives
with a wide range of application possibilities. Moreover, the results
of the studies gave a new insight into the mechanism of oxidative
photocyclization of 3H-indolium salts.
2. Results and discussion
3H-indolium salts (4a-l), used in the studies, are derivatives of
1-phenyl-2-(2-arylethenyl)-3,3-dimethyl-3H-indolium
perchlo-
rates with para substituted benzene ring or amphi substituted
Scheme 3. Structures of substrates (4a-l) and products (6a-l).
Please cite this article as: M.J. Sawicka et al., Synthesis of 7H-indolo[1,2-a]quinolinium derivatives via oxidative photocyclization of 3H-indolium
salts using blue LED photoreactor, Tetrahedron, https://doi.org/10.1016/j.tet.2019.06.002

M.J. Sawicka et al. / Tetrahedron xxx (xxxx) xxx
3
oxidative photocyclization of the examined compounds and it was
used in further experiments.
2.2. Oxidative photocyclization in the blue LED photoreactor
Since only a decrease in the intensity of the substrate absorption
band was observed in Fig. 2, which indicated a decrease in the
substrate concentration, the studies with the use of salt 4h were
continued to find out whether it is possible to obtain the product of
its oxidative photocyclization applying Vis irradiation instead of
UV. The concentration of the methanolic solution of 4h, equal to
6 mmol dmꢀ3, was higher than that in the experiment with the
mercury lamp [19,20]. However, in comparison with the pre-
liminary tests concerning the choice of the optimal light source, it
was reduced in order to shorten the reaction time. Indeed, it is well
known that in photochemical reactions the reaction time depends
on the substrates quantity. The irradiation was carried out as long
as the UV-Vis absorption spectrum of the sample did not differ from
the spectrum of the sample taken from the reaction mixture one
hour earlier (Fig. 3).
It can be seen from Fig. 3 that the irradiation of the reaction
mixture with 454.5 nm was accompanied, at the beginning of the
reaction, by a decrease in the intensity of the absorption band
assigned to compound 4h at 457 nm. It can indicate that cyclization
of salt 4h took place in the solution. The generation of a new ring
disturbed the aromatic character of the compound and shortened
the conjugated double bond system. As a consequence, the change
in the dye chromogen took place and the colourless compound,
probably dihydro intermediate, appeared in the solution. Spectral
changes in the UV region were also noticed, especially in the
spectra of the samples taken from the reaction mixture between
the 3rd and the 7th hour of irradiation. These changes concerned
the increase in the absorption intensity in the range 300e350 nm
and the hypsochromic shift of the spectrum in the range
290e292.5 nm. After 7 h of the irradiation, these changes stopped,
whereas a new band appeared at 329.2 nm and the absorption
band at 404.6 nm started to be the most intensive one in the Vis
region. These bands can be attributed to product 6h of the oxidative
dehydrogenation of the dihydro intermediate. The increase in the
long-wavelength absorption band intensity was observed until the
12th hour of irradiation. The lack of isosbestic points in Fig. 3 in-
dicates that the intermediate dihydro compound was present in the
solution in measurable amounts and its aromatization process
needed a long time.
Fig. 1. UV-Vis absorption spectra of substrates 4a-l (a) and UV-Vis emission spectra of
violet and blue LEDs (b).
In order to find the concentration profiles of the coloured
components of the reaction mixture, the standard solutions of pure
Fig. 2. UV-Vis spectra of the sample taken from the reaction mixture during irradiation
of 4h salt solution with violet (a) and blue (b) LEDs.
Fig. 3. UV-Vis spectra of the samples taken from the reaction mixture during the
oxidative photocyclization of salt 4h.
Please cite this article as: M.J. Sawicka et al., Synthesis of 7H-indolo[1,2-a]quinolinium derivatives via oxidative photocyclization of 3H-indolium
salts using blue LED photoreactor, Tetrahedron, https://doi.org/10.1016/j.tet.2019.06.002

4
M.J. Sawicka et al. / Tetrahedron xxx (xxxx) xxx
compounds 4h and 6h were prepared and their absorption spectra
were recorded. Molar excitation coefficients of both compounds
were determined according to the Lambert-Beer law at each
wavelength from the range 350e600 nm. Using the multiple
regression analysis, the concentrations of 4h and 6h in the reaction
mixture after each hour of irradiation were calculated on the basis
of the UV-Vis spectra of the samples according to the equation:
showed that the results were similar to those obtained in the
previous experiment (Fig. 4, triangles). However, the concentration
of product 6h measured after each next hour was a little higher.
After 3 h of irradiation, over 17% of the substrate was converted to
the product. The substrate 4h was detected in the solution for the
last time in the 7th hour of irradiation, but the dehydrogenation
process has finished after 12 h. The reaction yield was increased up
to 72%. It can be concluded that the increase in oxygen concen-
tration led to an increase in the reaction yield.
After the process was finished, the reaction mixture was
concentrated. The obtained precipitate was isolated by filtration
and washed with water. In 1H NMR spectra of the sample, all signals
were attributed to product 6h. The signals corresponding to the
unreacted substrate 4h and the intermediate 5 were not observed.
Al ¼ εl4hC4hl þ εl6hC6h
l
(1)
where Al is the absorbance of the reaction mixture at i-th wave-
length, εl εl the molar extinction coefficients
,
e
4h
6h
(dm3$cmꢀ1$molꢀ1) of compounds 4h and 6h, respectively, at the i-
th wavelength, c4h, c6h e the molar concentrations (mol$dmꢀ3) of
compounds 4h and 6h, respectively, and l e the path length (cm).
The absorption bands of compounds 4h and 6h partly over-
lapped. Thus, the concentrations of these compounds in the reac-
tion mixture were difficult to determine according to Lambert-Beer
law, from only one absorbance value corresponding, for example, to
the absorption band maximum. However, the multiple regression
method allowed evaluation of concentrations of both compounds
with high accuracy. The correlation coefficients exceeded 0.99 and
the standard deviations calculated in the absorbance unit were less
than 0.01. The calculated values of c4h and c6h were plotted as a
function of the irradiation time (Fig. 4, circles). It was found that
50% of the starting material underwent the photocyclization pro-
cess after 3 h, whereas 6 h were needed to obtain full conversion of
compound 4h. The product of dehydrogenation was detected in the
solution for the first time after 1 h of irradiation and its concen-
tration was increasing very slowly at the beginning. After 3 h, only
6.61% of the substrate was converted into the product. A distinct
increase in the product concentration was observed only after 5 h of
irradiation, when most of the substrate disappeared from the so-
lution. The oxidation was completed after 12 h and led to the
product 6h with a 61% yield.
2.3. Oxidative dehydrogenation e photochemical or dark reaction?
It can be concluded from Fig. 4 that dehydrogenation was the
slowest step of the process determining the synthesis time of
product 6h under these conditions. It is generally accepted that this
step of oxidative photocyclization of many compounds usually
proceeds with the involvement of atmospheric oxygen [12]. How-
ever, in our experiment perchloric acid with oxidizing properties,
was a component of the reaction mixture. Additionally, the studied
salts were perchlorates. In order to check whether HClO4 or simply
ClOꢀ4 ion can really be oxidants, salt 4h was replaced with com-
pound 4b, which does not need acidification due to the lack of
hydroxyl group. Then, the reaction was carried out with and
without the addition of perchloric acid in the reaction mixture.
Other process conditions were the same as in the case of 4h syn-
thesis. The results obtained in these trials were exactly the same.
Moreover, in the IR spectra of product 6b obtained without HClO4
addition, the absorption band positioned at 1082 cmꢀ1 was
observed, which is an evidence of ClOꢀ4 anion presence. It proved
that ClOꢀ4 anion was not consumed in the oxidation step. Based on
this observation it can be concluded that the oxidant in the reaction
was not HClO4 but rather atmospheric oxygen.
Oxidation reactions can take place with involvement of either
diradical triplet oxygen or nonradical singlet oxygen [38]. The
ground state oxygen exists in the triplet form, which is para-
magnetic and reacts readily with other radical compounds. Thus, it
plays an important role in reactions proceeding via a radical
mechanism. However, most of the organic molecules are not radi-
cals and exist in the singlet state. Therefore, they are not able to
react with oxygen in its ground state but can react with singlet
oxygen, in the excited state. Singlet oxygen can be produced from
triplet oxygen by both dark and photochemical processes [39]. In
photochemical methods, some coloured compounds used as sen-
sitizers are very useful. They can absorb energy from light and
transfer it to triplet oxygen which undergoes excitation to singlet
oxygen [38]. The studies of Bromberg and Muszkat [37] have shown
that the oxidation of the dihydro intermediate, formed during the
conversion of stilbenes into phenanthrenes under UV irradiation, is
a self-initiated reaction proceeding via a radical chain mechanism
with common triplet oxygen. Singlet oxygen was taken into ac-
count as the reactive species as well but the mechanism of the
process was unclear [40]. However, for the compounds with the 3-
azoniahexatriene system, such as derivatives of quinolizino[3,4,5,6-
def]phenanthridinium cation [21], compounds with the 11-
azoniapyrene skeleton [41], 5,6-dihydronaphto[2,1-c]quinolizi-
nium salts [42], or 2,3-diaryl-1H-isoindolium perchlorates [43] the
mechanism with the participation of singlet oxygen was proposed
(Scheme 4). The process consists in transfer of a proton, formation
of an enamine (9), cycloaddition of singlet oxygen, protonation of
The studies of Bromberg and Muszkat [37] have shown that the
rate of dihydrophenanthrene photocyclization under the UV irra-
diation increased with the oxygen concentration, and in several
cases it was proportional to this concentration. Taking into account
a small volume of the reaction mixture as well as the solubility of
oxygen in the methanolic solution, the synthesis of product 6h
from substrate 4h was repeated under the same conditions but the
reactor was run with a stream of air bubbling through the solution,
which also stirred the mixture. The UV-Vis spectra of the samples
taken from the reaction mixture after each hour of irradiation
Fig. 4. Change in the substrate and the product concentrations during the oxidative
photocyclization of compound 4h; c6h ¼ c6ht - c6h0 where c6h0 is the initial concen-
tration of 6h, c6ht is the concentration of 6h after time t; without sensitizer:
c6h0 ¼ 0 mmol dmꢀ3, with 6h as a sensitizer: c6h0 ¼ 1.4 mmol dmꢀ3
.
the formed cycloperoxide (10) and b-elimination of hydrogen
Please cite this article as: M.J. Sawicka et al., Synthesis of 7H-indolo[1,2-a]quinolinium derivatives via oxidative photocyclization of 3H-indolium
salts using blue LED photoreactor, Tetrahedron, https://doi.org/10.1016/j.tet.2019.06.002

M.J. Sawicka et al. / Tetrahedron xxx (xxxx) xxx
5
substrate 4h to product 6h was 53%. The aromatization time was
shortened to 10 h and the reaction yield increased to 88%. It can be
concluded that compound 6h is an effective singlet oxygen
sensitizer.
Scheme 4. The proposed oxidation mechanism of compounds with the 3-
azoniahexatriene system [21].
2.4. Influence of the UV-Vis absorption band position on the
reaction time
peroxide [21].
Oxidative photocyclization of other 3H-indolium salts (4a-4g
and 4i-l) was carried out under the same conditions as described
for compound 4h. The tested compounds differed in the position of
the UV-Vis absorption band, which overlapped to a greater or lesser
extent the emission band of blue LEDs. UV-Vis spectra were used to
monitor the reaction progress. The photocyclization time t1 was
defined as the time after which the substrate was not detected in
the reaction mixture, whereas the oxidation time t2 was found as
the time after which an increase in the product concentration was
no longer observed. The time t2 was accepted as the total time of
oxidative photocyclization. The obtained results are presented in
Table 1.
It appears from the data presented in Table 1 that the trans-
formation time strongly depends on the position of the absorption
band of the substrate and the product. Generally, the shorter the
wavelength of the substrate absorption band, the shorter the
cyclization time. Based on this observation it can be concluded that
the structure of both substrates and products, determining the
absorption band position and intensity, influences the time of the
oxidative photocyclization of 3H-indolium salts. The intensive ab-
sorption band of substrates in Vis region results from the low en-
ergetic charge transfer (CT) transition. The position of this band is
strongly influenced by some substituents present in the phenyl ring
or naphthalene moiety. Electron donating substituents, such as
methoxy or hydroxyl groups, reduce the excitation energy and
consequently cause the bathochromic shift of the UV-Vis spectrum.
The same effect is observed when the conjugated system of
doubled bonds between donor and acceptor parts, which takes part
in the CT transition, is increased. It can be seen from the compar-
ison of the absorption spectra of compounds 4g and 4l (Table 1).
The replacement of benzene ring with naphthalene moiety led to
the bathochromic shift of the absorption band of 24.2 nm. The
presence of electron donating or withdrawing group in ortho po-
sition in 40-hydroxy derivatives (4c-f and 4i, 4j) also affects the
electron density, and thus influences the excitation energy. For
instance, in the UV-Vis spectrum of compounds with electron
withdrawing substituents in 30 or 50 positions, such as nitro group,
the hypsochromic shift is observed, which can be reduced by the
influence of an alkoxy group. Actually, according to the basic
Grotthuss-Draper law, in the photochemical reaction, the
Earlier studies on the synthesis of 7H-indolo[1,2a]quinolinium
salts indicated that the irradiation was necessary for the first stage
of the reaction, i.e. the photocyclization step (see Scheme 2) [18]. In
order to check whether the oxidation can be carried out as a dark or
photochemical reaction, the methanolic solution of 4h was irradi-
ated 7 h until the photocyclization was almost completed and only
small amount of substrate 4h (0.24 mmol dmꢀ3) was present in the
solution. The product 6h concentration was 1.87 mmol dmꢀ3. Then,
the reaction mixture was kept at room temperature or heated at
50 ꢁC, similarly as in the experiments with LED photoreactor,
without any access to light for 5 h. The UV-Vis spectra measured
before and after the dark reaction revealed an increase in the ab-
sorption band intensity corresponding to product 6h. Actually,
without irradiation, the reaction did not stop which proved that
triplet oxygen can act as an oxidant in the synthesis of 7H-indolium
salts. However, the concentration of 6h in the reaction mixture
increased by 0.66 mmol dmꢀ3 and 0.54 mmol dmꢀ3 with and
without heating, respectively. It was much less than the product
concentration increase observed between the 7th and 12th hour of
irradiation which was 2.23 mmol dmꢀ3
. Therefore, it can be
concluded that both singlet and triplet oxygens can act as oxidants
but the former one, which is a more reactive species, leads to better
results. In the photochemical oxidation of dihydro intermediates,
both 4h and 6h compounds can act as sensitizers of singlet state
oxygen. Taking into account the concentration profile of 6h (Fig. 4,
circles or triangles), one can admit that product 6h seems to be a
more effective sensitizer. Indeed, the concentration of 6h was
slowly increasing during the initial hours of irradiation. Only after
the 5th hour, when the concentration of 6h was 0.94 mmol dmꢀ3
and higher, it started to increase significantly faster. If substrate 4h
were a sensitizer, the concentration of singlet oxygen would be the
highest at the beginning of the irradiation and then the product
concentration increase should be the fastest. However, a much
faster increase in the product 6h concentration was observed just
after the 5th hour of irradiation, when the concentration of 4h was
1.57 mmol dmꢀ3 and still decreased. Moreover, the energy transfer
from a sensitizer to common triplet state oxygen needs a triplet
excited state sensitizer, which is generated from its singlet excited
state. Similarly to stilbenes photocyclization [20], one can suppose
that the transformation of compound 4h proceeds via a singlet
electronic excited state. Therefore, the involvement of the singlet
excited state of 4h in the sensitization process should decrease the
quantum yield of the photocyclization extending the reaction time.
An additional experiment was performed in order to prove that the
oxidation product 6h acted as a sensitizer. Oxidative photo-
cyclization was repeated under the same conditions as in the pre-
vious experiment but 3.03 mg of 6h was added at the beginning of
irradiation into 5 cm3 of a methanolic solution of 4h with the
concentration of 6 mmol dmꢀ3. The initial amount of 6h was cho-
sen on the basis of the changes in the concentration of 6h during
the reaction course (Fig. 4, triangles). The addition of 6h into the
reaction mixture did not distinctly change the course of the cycli-
zation step but noticeably influenced the progress of oxidative
dehydrogenation (Fig. 4, squares). After 3.5 h of irradiation, 50% of
the starting material underwent cyclization, and the conversion of
Table 1
Positions of the UV-Vis absorption band and time of oxidative photocyclization of
3H-indolium salts.
substrate 4
l
[nm]
product 6
l
[nm]
t1 [h]
t2 [h]
a
b
c
d
e
f
g
h
i
397.6
402
358
365
1
1
2
2
2
3
6
7
9
9
5
11
3
3
3
3
3
4
8
12
9
420.2
434.6
437.8
441.5
446.8
457
463.5
464.4
468
365.2
385.5
382
384.5
397.2
404.6
411.5
408
j
k
l
9
5
11
414.8
417.6
471
Please cite this article as: M.J. Sawicka et al., Synthesis of 7H-indolo[1,2-a]quinolinium derivatives via oxidative photocyclization of 3H-indolium
salts using blue LED photoreactor, Tetrahedron, https://doi.org/10.1016/j.tet.2019.06.002

6
M.J. Sawicka et al. / Tetrahedron xxx (xxxx) xxx
absorption band of the substrate should overlap the emission band
of the light source. Consequently, the best results should be ob-
tained for a substrate with an electron donating group, such as 4h,
whose absorption band is the closest to 454.5 nm. Actually, when
using 4h, the results indicate that the synthesis of product 6h took
the longest time and the biggest difference was observed between
t1 and t2. This may be due to the fact that the strong absorptivity of
the reactant has led to the situation, in which the light penetrated
the reaction mixture to an extremely short distance. As a result, the
reaction proceeded in a very thin zone only. This can be an expla-
nation of differences in the reaction time between substrates 4a-h.
The difference between t1 and t2 in the synthesis of 6h is probably
also caused by the strong absorptivity of salt 4h, which hinders the
activity of product 6h as a singlet oxygen sensitizer and thus ex-
tends the transformation time. However, when using substrates 4i,
4j and 4l with absorption bands positioned at longer wavelengths,
this explanation seems to be insufficient. Therefore, other factors
should be also taken into account. The probable reason for the
differences in the reaction time can be related to the position of the
product absorption band. This band is hypsochromically shifted in
comparison to that of the substrate, which results from the fact that
the indoloquinolinium moiety in the product structure is plain. It
1.3% of the starting amount of 4h. At the same time 18.9% and 41.8%
of the starting 4h was detected in methanol-acetonitrile mixture
and methanol, respectively. On the other hand, in the 5th hour the
concentration of product 6h in acetonitrile and methanol-
acetonitrile mixture was 52.4% and 77.9% of the product concen-
tration in methanol. Thus, in acetonitrile the increase in salt 6h
concentration was the slowest during five hours, but after that it
started to be the fastest. As a consequence, the transformation time,
as well as the reaction yield, were very similar in each solvent. It can
be explained based on the fact that the polarities of compounds
present in the reaction mixture changes during the reaction.
Similarly to hexane, cyclohexane and benzene, which polarities
according to ET(30) scale are 129.71 kJ molꢀ1, 125.90 kJ molꢀ1 and
143.52 kJ molꢀ1 [46], respectively, one can accept that a substrate
before cyclization, which exists as a cis isomer, is slightly more
polar than dihydro intermediate. Moreover, the intermediate being
a product of the cyclization with partly saturated character should
be less polar than the product of its dehydrogenation, which pos-
sesses an aromatic moiety. Therefore, the least polar intermediate
is the best stabilized in less polar acetonitrile and this shortened the
cyclization time and hindered the oxidation. However, the disap-
pearance of substrate 4h from the reaction mixture, which strongly
absorbed the light emitted by blue LED, allowed product 6h to act
as a singlet oxygen sensitizer. Since the life time of singlet oxygen is
longer in acetonitrile, starting from the 5th hour the increase in the
product concentration observed in this solvent is faster than that
measured in methanol or in methanol-acetonitrile mixture.
allows an effective p-electron coupling which reduces the acceptor
properties of nitrogen leading to an increase in the excitation en-
ergy. However, the absorption band of products still partly over-
lapped the emission band of the LEDs. It is the most evident for
compounds 6g-6l. Apart from the substrates, these products are
able to absorb light, which diminishes the number of photons
available for the cyclization. The cyclization time of substrate 4k
additionally confirmed this explanation. Product 6k, obtained as a
result of oxidative photocyclization of 4k, appeared to be poorly
soluble in methanol, so it precipitated at the bottom of the reaction
vessel. Low concentration of 6k in the reaction mixture did not
limit the light absorption by the substrate as much as it was
observed for compounds 6i, 6j and 6l. As a consequence, the
cyclization time was shortened. On the other hand, it should be
taken into account that the energy gained during the excitation
must be enough for the reaction to take place. It can be problematic
for compounds with electron donating substituents and thus with
the absorption band positioned at longer wavelengths. However, in
the case of substrates without substituents of this type, and thus
with higher excitation energy, it can explain why the oxidative
photocyclization proceeds easier and in shorter time.
2.6. Scale-up experiments
In the next step, an attempt was made to scale-up the synthesis
of 7H-indolo[1,2-a]quinolinium salts. For this purpose, 70 W blue
LED photoreactor with the capacity of 88 cm3 was used in the study.
The experiment was carried out using 25 cm3 (instead of 5 cm3) of
the methanolic solution of substrate 4h with the same concentra-
tion as earlier, i.e. 6 mmol dmꢀ3. The course of transformation was
similar to that observed with the use of 20 cm3 blue LED photo-
reactor. However, this time the reaction was completed after 14 h
and led to product 6c with a 91% yield. Finally, 1 mmol of the
substrate 4h was used in the experiment. The LED photoreactor
capacity allowed the use of 35 cm3 of the reaction mixture, which
means that the substrate concentration was increased to 28.6 mmol
dm-3. The other reaction conditions were not changed. After 3 h of
irradiation, 10.5% of the substrate reacted, 14% of which was con-
verted to the product. After 10 h of the process, 50% of the initial
substrate concentration was observed and then the conversion to
product 6h was approximately 49%. The substrate was detected in
the reaction mixture for the last time in the 29th hour of irradiation,
whereas the oxidation was completed after 30 h. The longer reac-
tion time was a consequence of much greater amount of the sub-
strate used in the experiment. The 1H NMR spectrum of the product
isolated from the reaction mixture after this time was identical to
that obtained with the use of 20 cm3 blue LED photoreactor. The
reaction yield was 94%.
2.5. Solvent influence on the oxidative dehydrogenation.
Oxidative photocyclization of salt 4h was repeated under the
same condition but methanol was replaced with acetonitrile and
the mixture of acetonitrile and methanol (1:1). The solvents differ
from each other with polarity, which is 231.9 kJ molꢀ1 and
190.9 kJ molꢀ1 for methanol and acetonitrile, respectively, accord-
ing to the most popular ET(30) polarity scale [46]. The use of less
polar solvents was impossible due to the poor solubility of both 4h
and 6h compounds. The life time of singlet oxygen in 50 ꢁC in these
two solvents is also different and is 9.7
m
s and 47.6
m
s for methanol
The other 7H-indolo[1,2-a]quinolinium salts were synthesized
in the same conditions. All products were isolated from the reaction
mixture and identified with the use of IR, 1H NMR and 13C NMR
spectroscopy and melting point measurements. The yield of the
reaction was calculated and compared, if possible, with that ob-
tained with the use of a mercury lamp. The results are collected in
Table 2.
and acetonitrile, respectively [47]. The reaction progress was
studied also with UV-Vis measurements. The concentration profiles
of substrate 4h and product 6h in various solvents were quite
similar. However, some difference can be noticed, including the
time of the substrate cyclization, as well as the course of the con-
centration changes of the product. The shortest cyclization time
was observed in the least polar acetonitrile. In this solvent the
substrate was detected for the last time in the 5th hour and it was
It can be concluded from the results that the use of the blue LEDs
as a light source allowed obtaining the reaction yields comparable
Please cite this article as: M.J. Sawicka et al., Synthesis of 7H-indolo[1,2-a]quinolinium derivatives via oxidative photocyclization of 3H-indolium
salts using blue LED photoreactor, Tetrahedron, https://doi.org/10.1016/j.tet.2019.06.002

M.J. Sawicka et al. / Tetrahedron xxx (xxxx) xxx
7
Table 2
The yield of oxidative photocyclization of 3H-indolium salts.
a
6
yieldLED [%]
yieldmercury lampb [%]
a
b
c
d
e
f
g
h
i
80
79
80
86
76
82
82
94
80
80
85
85
ed
ed
70c
65c
ed
65c
Scheme 5. Numbering of hydrogen atoms in the structures of products.
ed
90c
ed
spectrometer (Bruker, Germany), operating at 400.1 and
100.6 MHz, respectively. In comparison to the earlier described
experiments [19,20,44], the NMR spectra were recorded in
DMSO‑d6 as a solvent and on the instrument ensuring a higher
resolution. Tetramethylsilane was used as an internal reference.
Spectra were acquired in 5 mm probes. For NMR analyses Mes-
j
k
l
ed
ed
44e
a
The isolated yield obtained with the use of the blue LED photoreactor, the re-
action scale was 1 mmol of a substrate/35 cm3 of methanol.
b
The isolated yield obtained with the use of a mercury lamp.
c
The reaction scale was 3e9 mmol of a substrate/1 dm3 of methanol [19].
tReNova (version 10.0.1) program was used. The chemical shifts (d)
d
No data.
were given in ppm and the coupling constants (J) in Hz. The stan-
dard abbreviations for multiplicities were used (s ¼ singlet,
d ¼ doublet, dd ¼ doublet of doublets, t ¼ triplet, q ¼ quartet,
m ¼ multiplet). Hydrogen atoms were numbered according to
Scheme 5.
e
The reaction scale was 0.72 mmol of a substrate/1 dm3 of methanol [20].
to or even higher than those achieved with the use of a mercury
lamp.
Infrared spectra were taken with Alfa spectrometer with ATR
adapter (Bruker, Germany), in the range of 4000-600 cmꢀ1. UV-Vis
spectra were recorded with a computer-controlled V-670 spec-
trometer (Jasco, Japan), in the range of 285e600 nm at 25 0.1 ꢁC.
Gas-tight quartz cells with 1 cm path length (Hellma, Germany),
adjustable automatic pipettes with 0.1e1 and 1e5 cm3 volume (PZ
HTL Co., Poland), and liquid 5 and 25 mm3 syringes (Hamilton
Bonaduz AG, Switzerland) were used. UV-vis spectra were
measured after the dissolution of the samples taken from the re-
action mixture (10 mm3 from the reaction mixture with the con-
centration of 6 mmol dmꢀ3 or 2 mm3 from the reaction mixture
with the concentration of 28.6 mmol dmꢀ3) in methanol (2 cm3).
The multiple regression analysis was performed using data analysis
software system Statistica (version 13.1).
3. Conclusions
Oxidative photocyclization of 1-phenyl-2-(2-arylethenyl)-3,3-
dimethyl-3H-indolium salts in a blue LED photoreactor appeared
to be a simple and convenient way for the synthesis of 7H-indolo
[1,2-a]quinolinium perchlorates, which are the precursors of useful
solvatochromic or fluorescent probes.
The experiments have shown that photocyclization proceeds
under Vis irradiation. The LEDs emitting 454.5 nm proved to be an
appropriate light source in oxidative photocyclization of the
examined compounds. The best results were obtained when the
photoreactor was run with a stream of air bubbling through the
methanolic solution of the substrate. Oxidation can be carried out
as a dark process or a photochemical one. The time of the oxidative
photocyclization of 3H-indolium salts was influenced by the
structure of both substrates and products, determining the ab-
sorption band position and intensity. The scale-up experiments
with 1 mmol of the substrate have shown that the reaction yields
obtained with the blue LED photoreactor are comparable or even
higher than those achieved with the use of a mercury lamp.
The main advantage of the proposed solution is the elimination
of all inconveniences associated with the synthesis in photo-
reactors equipped with mercury lamps. In the case of LEDs pho-
toreactor, the reaction was carried out in a significantly smaller
device with a simpler construction, using small amounts of sol-
vents, without UV irradiation and therefore at a lower temperature
and with the use of ordinary glass vessels. It makes the synthesis
easier, energy saving, more eco-friendly and more comfortable for
operators. Taking into account that there are no reports on oxida-
tive cyclization of other compounds with the hexa-1,3,5-triene
system in LED photoreactor, these new reaction conditions could
potentially be very useful in the further synthesis of important
polycyclic aromatic hydrocarbons and heteroaromatics.
4.2. LED photoreactors
Two home-made periodical 25 W blue and 40 W violet LED
photoreactors with light emission at 454.5 nm and 395.4 nm,
respectively, were used in the initial experiments. LED strips (LED
Solar System, Poland) of 40 cm length and appropriate strength
characteristics were wrapped in a tight helix around the whole
outside surface of 20 cm3 (d ¼ 1.5 cm, h ¼ 11 cm) cylindrical glass
vessels, and headed into the vessels. The scale-up experiments
were carried out in the 70 W blue LED photoreactor with 88 cm3
capacity (d ¼ 2.5 cm, h ¼ 18 cm), constructed in the same way with
the use of LED strips of 110 cm length. The LEDs were characterized
using an SV2100 UV-Vis spectrometer (Korea Materials & Analysis
Corp., Korea).
4.3. Compounds and reagents
70% perchloric acid for syntheses was purchased from Chempur
(Poland). Acetonitrile and methanol of the Uvasol™ grade for
spectroscopy were purchased from Merck KgA (Germany),
respectively. Distilled water was obtained using a Büchi glass
apparatus and used without additional purification. 1-phenyl-2-(2-
arylethenyl)-3,3-dimethyl-3H-indolium perchlorates (4a-l) were
synthesized according to the previously described procedure
[19,20,45].
4. Experimental section
4.1. General
Melting points were determined on a Boetius hot stage appa-
ratus (Franz Küstner Nachf. KG, Germany). 1H and 13C NMR mea-
surements were performed in
a DPX 400 Avance III HD
Please cite this article as: M.J. Sawicka et al., Synthesis of 7H-indolo[1,2-a]quinolinium derivatives via oxidative photocyclization of 3H-indolium
salts using blue LED photoreactor, Tetrahedron, https://doi.org/10.1016/j.tet.2019.06.002

8
M.J. Sawicka et al. / Tetrahedron xxx (xxxx) xxx
4.4. General procedure for the synthesis of 7H-indolo[1,2-a]
quinolinium salts (6a-l) via oxidative photocyclization of 3H-
indolium perchlorates (4a-l)
124.92, 126.02, 127.64, 127.92, 129.39, 129.98, 130.30, 130.61, 134.83,
135.97, 137.07, 142.28, 154.37, 156.07, 168.42. IR (ATR): ῡ (cmꢀ1
1687, 1612, 1592, 1556, 1489, 1458, 1400, 1356, 1327, 1308, 1253,
1159, 1084, 1043, 921, 830, 764, 714, 697, 621.
)
1-phenyl-2-(2-arylethenyl)-3,3-dimethyl-3H-indolium
perchlorate (4a-l) (1 mmol) and 70% perchloric acid (2 mm3,
2.33$10ꢀ2 mmol) were dissolved in methanol (35 cm3), in a 50 cm3
glass flask. The flask with the reaction mixture was placed inside
88 cm3 blue LED photoreactor, equipped with a reflux condenser
and irradiated. The reaction mixture was blown with air stream.
The progress of the reaction was monitored using UV-Vis spec-
troscopy. After the process was completed, the reaction mixture
was concentrated to approximately 5 cm3. The precipitate was
filtered off, washed with water and dried. The crude product was
crystallized from an appropriate solvent to give the yellow powders
of 7H-indolo[1,2-a]quinolinium derivative (6a-l).
4.4.4. 5-(3,5-Dibromo-4-hydroxyphenyl)-7,7-dimethyl-7H-indolo
[1,2-a]quinolinium perchlorate (6d)
Salt 6d was synthesized from 1-phenyl-2-[2-(3,5-dibromo-4-
hydroxyphenyl)ethenyl]-3,3-dimethyl-3H-indolium
perchlorate
(4d) (0.598 g, 1 mmol). The reaction was completed after 8 h of
irradiation, yielding 0.512 g (86%) of 6d. Recrystallization from
acetonitrile gave the melting point of 323e324 ꢁC (Ref. [19]
322e323 ꢁC). 1H NMR (400 MHz, (CD3)2SO, TMS): dH (ppm) (J (Hz))
1.82 (s, 6H, CH3), 7.77e7.80 (m, 2H, H-8, H-10), 7.99 (s, 2H, H-20, H-
6’), 8.05e8.07 (m, 1H, H-9), 8.11 (t, 1H, J ¼ 7.8, H-3), 8.35e8.40 (m,
2H, H-2, H-4), 8.72 (s, 1H, H-6), 8.95 (m, 1H, H-11), 9.40 (d, 1H,
J ¼ 9.0, H-1), 10.92 (s, 1H, OH); 13C NMR (100 MHz, (CD3)2SO, TMS):
dH (ppm) 25.38, 48.83, 112.83, 119.05, 119.75, 119.87, 124.95, 127.71,
129.27, 129.44, 129.99, 130.42, 130.65, 134.42, 135.98, 136.93, 141.15,
143.32, 153.68, 155.39, 168.33. IR (ATR): ῡ (cmꢀ1) 1681, 1611, 1595,
1581,1557,1545,1479,1458,1399,1381,1369,1287,1243,1195,1160,
1091, 1062, 873, 758, 720, 621.
4.4.1. 5-Phenyl-7,7-dimethyl-7H-indolo[1,2-a]quinolinium
perchlorate (6a)
Salt 6a was synthesized from 1-phenyl-2-[2-phenylethenyl]-
3,3-dimethyl-3H-indolium perchlorate (4a) (0.423 g, 1 mmol). The
reaction was completed after 5 h of irradiation, yielding 0.338 g
(80%) of 6a. Recrystallization from acetonitrile gave the melting
point of 228e230 ꢁC (Ref. [18] 228e229.5 ꢁC). 1H NMR (400 MHz,
(CD3)2SO, TMS): dH (ppm) (J (Hz)) 1.83 (s, 6H, CH3), 7.76e7.81 (m,
7H, H-8, H-10, H-2’ e H-6’), 8.05e8.11 (m, 2H, H-3, H-9) (m, 2H, H-
3, H-9), 8.33 (d, 1H, J ¼ 8.3, H-4), 8.38 (t, 1H, J ¼ 7.6, H-2), 8.70 (s, 1H,
H-6), 8.96 (m, 1H, H-11), 9.42 (d, 1H, J ¼ 9.0, H-1); 13C NMR
(100 MHz, (CD3)2SO, TMS): dH (ppm) 25.30, 48.79, 118.99, 119.74,
119.88, 124.89, 127.78, 129.45, 129.68, 129.93, 130.60, 130.96, 131.20,
135.58, 135.90, 137.03, 141.24, 143.29, 158.40, 168.54. IR (ATR): ῡ
(cmꢀ1) 1612, 1594, 1554, 1480, 1458, 1398, 1087, 880, 772, 699, 622.
4.4.5. 5-(3-Ethoxy-4-hydroxy-5-nitrophenyl)-7,7-dimethyl-7H-
indolo[1,2-a]quinolinium perchlorate (6e)
Salt 6e was synthesized from 1-phenyl-2-[2-(3-ethoxy-4-
hydroxy-5-nitrophenyl)ethenyl]-3,3-dimethyl-3H-indolium
perchlorate (4e) (0.529 g, 1 mmol). The reaction was completed
after 9 h of irradiation, yielding 0.398 g (76%) of 6e. Recrystalliza-
tion from acetonitrile gave the melting point of 209e211 ꢁC
(Ref. [45] 207e210 ꢁC). 1H NMR (400 MHz, (CD3)2SO, TMS): dH
(ppm) (J (Hz)) 1.40 (t, 3H, J ¼ 6.9, CH3), 1.83 (s, 6H, CH3), 4.22 (q, 2H,
J ¼ 7.0,OCH2), 7.62 (d, 1H, J ¼ 1.9, H-20), 7.78e7.81 (m, 2H, H-8, H-
10), 7.87 (d, 1H, J ¼ 1.9, H-6’), 8.05e8.07 (m, 2H, H-3, H-9), 8.37 (t,
1H, J ¼ 7.8, H-2), 8.48 (d, 1H, J ¼ 8.4, H-4), 8.72 (s, 1H, H-6), 8.96 (m,
1H, H-11), 9.41 (d, 1H, J ¼ 9.0, H-1); 13C NMR (100 MHz, (CD3)2SO,
TMS): dH (ppm) 14.75, 25.42, 48.72, 65.70, 114.04, 118.90, 119.77,
120.23, 121.12, 125.51, 127.31, 127.66, 129.87, 129.98, 130.36, 130.44,
137.02, 137.55, 143.29, 148.30, 149.63, 150.10, 157.29, 161.53.IR
(ATR): ῡ (cmꢀ1) 1686, 1611, 1594, 1546, 1396, 1361, 1337, 1268, 1235,
1090, 1048, 877, 768, 749, 713, 636, 622.
4.4.2. 5-(4-chlorophenyl)-7,7-dimethyl-7H-indolo[1,2-a]
quinolinium perchlorate (6b)
Salt 6b was synthesized from 1-phenyl-2-[2-(4-chlorophenyl)
ethenyl]-3,3-dimethyl-3H-indolium perchlorate (4b) (0.458 g,
1 mmol). The reaction was completed after 5 h of irradiation,
yielding 0.359 g (79%) of 6b. Recrystallization from acetonitrile gave
the melting point of 242e244 ꢁC (Ref. [18] 240e243.5 ꢁC). 1H NMR
(400 MHz, (CD3)2SO, TMS): dH (ppm) (J (Hz)) 1.82 (s, 6H, CH3),
7.79e7.85 (m, 6H, H-8, H-10, H-20, H-30, H-50, H-6’), 8.05e8.11 (m,
2H, H-3, H-9) (m, 2H, H-3, H-9), 8.31 (d, 1H, J ¼ 8.4, H-4), 8.38 (t, 1H,
J ¼ 7.9, H-2), 8.72 (s, 1H, H-6), 8.97 (m, 1H, H-11), 9.42 (d, 1H, J ¼ 9.0,
H-1); 13C NMR (100 MHz, (CD3)2SO, TMS): dH (ppm) 24.23, 47.78,
118.00, 118.74, 118.87, 123.86, 126.67, 128.26, 128.69, 128.91, 129.26,
129.60, 131.41, 133.28, 134.94, 135.94, 135.21, 135.94, 140.17, 142.25,
156.02,167.55. IR (ATR): ῡ (cmꢀ1) 1609,1592,1551,1493,1458,1448,
1403, 1383, 1359, 1327, 1082, 1014, 837, 783, 768, 749, 723, 711, 686,
653, 620.
4.4.6. 5-(4-Hydroxy-3-methoxy-5-nitrophenyl)-7,7-dimethyl-7H-
indolo[1,2-a]quinolinium perchlorate (6f)
Salt 6f was synthesized from 1-phenyl-2-[2-(4-hydroxy-3-
methoxy-5-nitrophenyl)ethenyl]-3,3-dimethyl-3H-indolium
perchlorate (4f) (0.515 g,1 mmol). The reaction was completed after
12 h of irradiation, yielding 0.421 g (82%) of 6f. Recrystallization
from acetonitrile gave the melting point of 263e265 ꢁC (Ref. [19]
264e266 ꢁC). 1H NMR (400 MHz, (CD3)2SO, TMS): dH (ppm) (J (Hz))
1.83 (s, 6H, CH3), 3.97 (s, 3H, OCH3), 7.61 (d, 1H, J ¼ 2.0, H-20),
7.79e7.81 (m, 2H, H-8, H-10), 7.87 (d, 1H, J ¼ 2.0, H-6’), 8.05e8.09
(m, 2H, H-3, H-9), 8.38 (t, 1H, J ¼ 7.4, H-2), 8.50 (d, 1H, J ¼ 8.5, H-4),
8.74 (s, 1H, H-6), 8.96 (m, 1H, H-11), 9.41 (d, 1H, J ¼ 9.0, H-1), 11.35
(s, 1H, OH); 13C NMR (100 MHz, (CD3)2SO, TMS): dH (ppm) 25.42,
48.82, 57.28, 113.06, 118.92, 119.75, 119.91, 121.31, 124.95, 125.34,
127.31, 127.73, 129.63, 129.99, 130.38, 130.65, 136.01, 137.02, 143.30,
144.91, 148.17, 150.43, 156.48, 168.33. IR (ATR): ῡ (cmꢀ1) 1610, 1592,
1533, 1456, 1398, 1363, 1325, 1260, 1245, 1087, 1050, 919, 874, 766,
712, 622.
4.4.3. 5-(4-Hydroxy-3-nitrophenyl)-7,7-dimethyl-7H-indolo[1,2-a]
quinolinium perchlorate (6c)
Salt 6c was synthesized from 1-phenyl-2-[2-(4-hydroxy-3-
nitrophenyl)ethenyl]-3,3-dimethyl-3H-indolium perchlorate (4c)
(0.485 g, 1 mmol). The reaction was completed after 7 h of irradia-
tion, yielding 0.388 g (80%) of 6c. Recrystallization from acetonitrile
gave the melting point of 316e319 ꢁC (Ref. [19] 317e318 ꢁC). 1H
NMR (400 MHz, (CD3)2SO, TMS): dH (ppm) (J (Hz)) 1.83 (s, 6H, CH3),
7.48 (d, 1H, J ¼ 8.63, H-50), 7.79e7.81 (m, 2H, H-8, H-10), 7.97 (dd,
1H, J1 ¼ 2.0, J2 ¼ 8.5, H-60), 8.05e8.08 (m, 2H, H-3, H-9), 8.35e8.44
(m, 3H, H-2, H-4, H-2’), 8.74 (s, 1H, H-6), 8.95 (m, 1H, H-11), 9.41 (d,
1H, J ¼ 8.9, H-1), 12.09 (s, 1H, OH); 13C NMR (100 MHz, (CD3)2SO,
TMS): dH (ppm) 25.39, 48.80, 118.90, 119.73, 119.91, 120.17, 120.32,
4.4.7. 5-(4-methoxyphenyl)-7,7-dimethyl-7H-indolo[1,2-a]
quinolinium perchlorate (6g)
Salt
6g
was
synthesized
from
1-phenyl-2-[2-(4-
Please cite this article as: M.J. Sawicka et al., Synthesis of 7H-indolo[1,2-a]quinolinium derivatives via oxidative photocyclization of 3H-indolium
salts using blue LED photoreactor, Tetrahedron, https://doi.org/10.1016/j.tet.2019.06.002

M.J. Sawicka et al. / Tetrahedron xxx (xxxx) xxx
9
methoxyphenyl)ethenyl]-3,3-dimethyl-3H-indolium perchlorate
(4g) (0.453 g, 1 mmol). The reaction was completed after 24 h of
irradiation, yielding 0.370 g (82%) of 6g. Recrystallization from
acetonitrile gave the melting point of 272e274 ꢁC (Ref. [18]
271.5e273 ꢁC). 1H NMR (400 MHz, (CD3)2SO, TMS): dH (ppm) (J
(Hz)) 1.82 (s, 6H, CH3), 7.32 (d, 2H, J ¼ 8.8, H-30, H-50), 7.77e7.81 (m,
4H, H-8, H-10, H-20, H-6’), 8.03e8.10 (m, 2H, H-3, H-9), 8.36 (t, 1H,
J ¼ 7.5, H-2), 8.41 (d, 1H, J ¼ 8.5, H-4), 8.64 (s, 1H, H-6), 8.93 (m, 1H,
H-11), 9.38 (d, 1H, J ¼ 9.0, H-1); 13C NMR (100 MHz, (CD3)2SO, TMS):
dH (ppm) 25.46, 48.68, 56.15, 115.34, 118.36, 119.59, 119.83, 124.88,
127.63, 129.71, 129.90, 130.02, 130.40, 132.76, 135.78, 137.14, 141.20,
143.23, 158.22, 162.09, 168.18. IR (ATR): ῡ (cmꢀ1) 1587, 1513, 1456,
1398, 1383, 1361, 1293, 1252, 1172, 1149, 1085, 1077, 1023, 837, 765,
698, 637.
TMS): dH (ppm) 14.96, 25.64, 48.74, 65.49, 114.22, 118.54, 119.62,
119.78, 121.18, 124.21, 124.91, 126.51, 127.69, 129.78, 129.95, 130.19,
130.50, 135.85, 137.04, 141.14, 143.26, 146.19, 148.59, 157.32, 168.11.
IR (ATR): ῡ (cmꢀ1) 1612, 1591, 1501, 1459, 1428, 1363, 1285, 1239,
1085, 1068, 1043, 915, 890, 865, 788, 713, 620.
4.4.11. 5-(5-Bromo-6-methoxynaphthyl-2)-7,7-dimethyl-7H-indolo
[1,2-a]quinolinium perchlorate (6k)
Salt 6k was synthesized from 1-phenyl-2-[2-(5-bromo-6-
methoxynaphthyl-2)ethenyl]-3,3-dimethyl-3H-indolium perchlo-
rate (4k) (0.583 g, 1 mmol). The reaction was completed after 15 h
of irradiation, yielding 0.494 g (85%) of 6k. Recrystallization from
acetic acid gave the melting point of 292e294 ꢁC (Ref. [44]
291e293 ꢁC). 1H NMR (400 MHz, (CD3)2SO, TMS): 1H (ppm) (J (Hz))
1.86 (s, 6H, CH3), 4.09 (s, 3H, OCH3), 7.75 (d, 1H, J ¼ 9.2, H-70),
7.80e7.82 (m, 2H, H-8, H-10), 7.99 (dd, 1H, J1 ¼8.8, J2 ¼ 1.4, H-30),
8.06e8.11 (m, 2H, H-3, H-9), 8.32 (d,1H, J ¼ 9.1, H-80), 8.37e8.44 (m,
3H, H-2, H-4, H-40), 8.47 (s, 1H, H-1’), 8.84 (s, 1H, H-6), 8.99 (m, 1H,
H-11), 9.45 (d,1H, J ¼ 9.0, H-1); 13C NMR (100 MHz, (CD3)2SO, TMS):
dH (ppm) 25.41, 48.83, 57.56, 107.22, 115.90, 119.19, 119.80, 119.96,
124.94, 126.67, 127.84, 129.37, 129.61, 129.76, 129.97, 130.31, 130.58,
131.21, 131.32, 131.54, 133.47, 135.93, 137.14, 141.33, 143.34, 155.76,
157.91, 168.47. IR (ATR): ῡ (cmꢀ1) 1609, 1587, 1547, 1482, 1459, 1440,
1361, 1279, 1250, 1183, 1080, 1060, 880, 835, 804, 780, 768, 709, 622.
4.4.8. 5-(4-hydroxyphenyl)-7,7-dimethyl-7H-indolo[1,2-a]
quinolinium perchlorate (6h)
Salt 6h was synthesized from 1-phenyl-2-[2-(4-hydroxyphenyl)
ethenyl]-3,3-dimethyl-3H-indolium perchlorate (4h) (0.440 g,
1 mmol). The reaction was completed after 30 h of irradiation,
yielding 0.413 g (94%) of 6h. Recrystallization from acetonitrile gave
the melting point of 329e330 ꢁC (Ref. [19] 328e329 ꢁC). 1H NMR
(400 MHz, (CD3)2SO, TMS): dH (ppm) (J (Hz)) 1.82 (s, 6H, CH3), 7.13
(d, 2H, J ¼ 8.6, H-30, H-50), 7.70 (d, 2H, J ¼ 8.6, H-20, H-6’), 7.74e7.78
(m, 2H, H-8, H-10), 8.02e8.09 (m, 2H, H-3, H-9), 8.34 (t, 1H, J ¼ 7.5,
H-2), 8.44 (d, 1H, J ¼ 8.5, H-4), 8.60 (s, 1H, H-6), 8.91 (m, 1H, H-11),
9.36 (d, 1H, J ¼ 9.0, H-1), 10.43 (s, 1H, OH); 13C NMR (100 MHz,
(CD3)2SO, TMS): dH (ppm) 25.51, 48.60, 116.73, 117.99, 119.51, 119.80,
122.19, 124.86, 126.00, 127.48, 129.88, 130.28, 133.02, 135.69, 137.17,
4.4.12. 5-(6-Methoxynaphthyl-2)-7,7-dimethyl-7H-indolo[1,2-a]
quinolinium perchlorate (6l)
Salt
6l
was
synthesized
from
1-phenyl-2-[2-(6-
methoxynaphthyl-2)ethenyl]-3,3-dimethyl-3H-indolium perchlo-
rate (4l) (0.504 g, 1 mmol). The reaction was completed after 34 h of
irradiation, yielding 0.427 g (85%) of 6l. Recrystallization from
acetic acid gave the melting point of 265e267 ꢁC (Ref. [20]
264e266 ꢁC). 1H NMR (400 MHz, (CD3)2SO, TMS): dH (ppm) (J (Hz))
1.85 (s, 6H, CH3), 3.97 (s, 3H, OCH3), 7.35 (dd,1H, J1 ¼8.9, J2 ¼ 2.5, H-
70), 7.56 (d, 1H, J ¼ 2.3, H-50), 7.79e7.83 (m, 2H, H-8, H-10), 7.85 (dd,
1H, J1 ¼8.5, J2 ¼ 1.7, H-30), 8.05e8.13 (m, 3H, H-3, H-80, H-9), 8.18 (d,
1H, J ¼ 8.6, H-40), 8.35e8.40 (m, 2H, H-2, H-1’), 8.45 (dd, 1H, J1 ¼8.5,
J2 ¼ 1.0, H-4), 8.79 (s, 1H, H-6), 8.97 (m, 1H, H-11), 9.43 (d, 1H,
J ¼ 9.0, H-1); 13C NMR (100 MHz, (CD3)2SO, TMS): dC (ppm) 25.44,
48.78, 55.97, 106.51, 118.98, 119.71, 119.90, 120.46, 124.92, 127.82,
128.00,128.14,128.56, 129.88,129.94,130.20, 130.50, 130.61,130.96,
131.04, 135.80, 135.84, 137.16, 141.29, 143.31, 158.64, 159.47, 168.31.
IR (ATR): ῡ (cmꢀ1) 1609, 1588, 1549, 1486, 1478, 1464, 1454, 1441,
1383, 1360, 1310, 1271, 1230, 1212, 1188, 1171, 1082, 1072, 1021, 874,
822, 760, 708, 676, 621.
141.20, 143.19, 158.54, 158.58, 160.95, 168.02. IR (ATR): ῡ (cmꢀ1
)
1607, 1586, 1556, 1516, 1457, 1397, 1364, 1275, 1217, 1174, 1097, 1049,
851, 767, 707, 701, 621.
4.3.9. 5-(3-Chloro-4-hydroxy-5-methoxyphenyl)-7,7-dimethyl-7H-
indolo[1,2-a]quinolinium perchlorate (6i)
Salt 6i was synthesized from 1-phenyl-2-[2-(3-chloro-4-
hydroxy-5-methoxyphenyl)ethenyl]-3,3-dimethyl-3H-indolium
perchlorate (4i) (0.504 g, 1 mmol). The reaction was completed af-
ter 27 h of irradiation, yielding 0.402 g (80%) of 6i. Recrystallization
from acetonitrile gave the melting point of 267e269 ꢁC (Ref. [45]
268e270 ꢁC). 1H NMR (400 MHz, (CD3)2SO, TMS): dH (ppm) (J (Hz))
1.83 (s, 6H, CH3), 3.97 (s, 3H, OCH3), 7.31 (d,1H, J ¼ 1.9, H-60), 7.44 (d,
1H, J ¼ 1.9, H-2’), 7.78e7.81 (m, 2H, H-8, H-10), 8.04e8.11 (m, 2H, H-
3, H-9), 8.36 (t, 1H, J ¼ 7.5, H-2), 8.50 (d, 1H, J ¼ 8.4, H-4), 8.68 (s, 1H,
H-6), 8.94 (m, 1H, H-11), 9.38 (d, 1H, J ¼ 9.0, H-1), 10.47 (s, 1H, OH);
13C NMR (100 MHz, (CD3)2SO, TMS): dH (ppm) 25.46, 48.74, 57.08,
113.25, 118.56, 119.64, 119.81, 121.13, 124.31, 124.92, 126.42, 127.66,
129.83, 129.94, 130.22, 130.50, 135.87, 137.04, 141.16, 143.26, 146.04,
149.41, 157.30, 168.13. IR (ATR): ῡ (cmꢀ1) 1583, 1515, 1457, 1397,
1382, 1368, 1279, 1220, 1173, 1095, 1044, 840, 763, 707, 620.
Conflicts of interest
Declarations of interest: none.
Acknowledgments
4.4.10. 5-(3-Chloro-4-hydroxy-5-ethoxyphenyl)-7,7-dimethyl-7H-
indolo[1,2-a]quinolinium perchlorate (6j)
This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
Salt 6j was synthesized from 1-phenyl-2-[2-(3-chloro-4-
hydroxy-5-ethoxyphenyl)ethenyl]-3,3-dimethyl-3H-indolium
perchlorate (4j) (0.518 g, 1 mmol). The reaction was completed after
27 h of irradiation, yielding 0.412 g (80%) of 6j. Recrystallization
from acetonitrile gave the melting point of 314e316 ꢁC (Ref. [45]
315e317 ꢁC). 1H NMR (400 MHz, (CD3)2SO, TMS): dH (ppm) (J (Hz))
1.43 (t, 1H, J ¼ 7.0, CH3), 1.83 (s, 6H, CH3), 4.23 (q, 2H, J ¼ 7.0, OCH2),
7.31 (d, 1H, J ¼ 1.9, H-60), 7.44 (d, 1H, J ¼ 1.9, H-2’), 7.78e7.80 (m, 2H,
H-8, H-10), 8.04e8.11 (m, 2H, H-3, H-9), 8.36 (t, 1H, J ¼ 7.4, H-2),
8.48 (d, 1H, J ¼ 8.5, H-4), 8.66 (s, 1H, H-6), 8.94 (m, 1H, H-11), 9.38
(d, 1H, J ¼ 9.0, H-1), 10.28 (s, 1H, OH); 13C NMR (100 MHz, (CD3)2SO,
References
[1] J.A. Soroka, K.B. Soroka, J. Phys. Org. Chem. 4 (1991) 592e604.
[2] M.J. Sawicka, J.A. Soroka, M. Ga˛siorowska, E.K. Wroblewska, J. Solut. Chem. 41
ꢀ
(2012) 25e35.
[3] J.A. Soroka, B. Komorowska, K.B. Soroka, J. Bendig, J. Photochem. Photobiol., A
118 (1998) 81e92.
[4] M.J. Sawicka, J.A. Soroka, Cent. Eur. J. Chem. 11 (2013) 1239e1247.
ꢀ
[5] E.K. Wroblewska, M. Ga˛siorowska, Tenside Surfactants Deterg. 48 (2011)
127e129.
ꢀ
[6] M.J. Sawicka, E.K. Wroblewska, Chem. Pap. 72 (2018) 741e752.
ꢀ
[7] E.K. Wroblewska, J.A. Soroka, M.J. Sawicka, Rosłaniec. Przem Chem. 95 (4)
Please cite this article as: M.J. Sawicka et al., Synthesis of 7H-indolo[1,2-a]quinolinium derivatives via oxidative photocyclization of 3H-indolium
salts using blue LED photoreactor, Tetrahedron, https://doi.org/10.1016/j.tet.2019.06.002

10
M.J. Sawicka et al. / Tetrahedron xxx (xxxx) xxx
(2016) 834e837.
Catal. 6 (1) (2016) 451e457.
[28] Q.M. Kainz, C.D. Matier, A. Bartoszewicz, S.L. Zultanski, J.C. Peters, G.C. Fu,
Science 351 (6274) (2016) 681e684.
[29] V. Mojr, E. Svobodova, K. Strakova, T. Nevesely, J. Chudoba, H. Dvorakova,
R. Cibulka, Chem. Commun. 51 (2015) 12036e12039.
ꢀ
[8] H. Guedouar, F. Aloui, S. Moussa, J. Marrot, B.B. Hassine, Tetrahedron Lett. 55
(2014) 6167e6170.
[9] Ch Li, Y. Zhang, S. Zhang, J. Shi, Y. Kan, H. Wang, Tetrahedron 70 (2014)
3909e3914.
ꢁꢀ
ꢀ
[10] Y. Li, Z. Pang, T. Zhang, J. Yang, W. Yu, Tetrahedron 71 (2015) 3351e3358.
[11] K. Sadorn, W. Sinananwanich, J. Areephong, C. Nerungsi, C. Wongma,
C. Pakawatchai, T. Thongpanchang, Tetrahedron Lett. 49 (2008) 4519e4521.
[12] F.B. Mallory, C.W. Mallory, Org. React. 30 (1984) 1e456.
[13] H. Guedouar, F. Alouia, A. Beltifa, H.B. Mansour, B.B. Hassine, C R Chim. 20
(2017) 841e849.
[30] S. Yoshioka, M. Nagatomo, M. Inoue, Org. Lett. 17 (1) (2015) 90e93.
[31] J. Carney, R. Hammer, M. Hulce, C. Lomas, D. Miyashiro, Synthesis 44 (16)
(2012) 2560e2566.
[32] F. Levesque, P.H. Seeberger, Org. Lett. 13 (19) (2011) 5008e5011.
[33] N. Doss, P. Bernhardt, T. Romero, R. Masson, V. Keller, N. Keller, Appl. Catal. B
Environ. 154e155 (2014) 301e308.
ꢀ
[14] S. Raouafi, F. Aloui, A. Raouafi, B.B. Hassine, C R Chim. 20 (2017) 697e703.
[15] W.H. Laarhoven, Pure Appl. Chem. 56 (9) (1984) 1225e1240.
[16] K.B. Jørgensen, Molecules 15 (2010) 4334e4358.
[34] K.C. Lee, H.J. Beak, K.H. Choo, Water Res. 86 (2015) 58e65.
[35] P. Wang, T.T. Lim, Water Res. 46 (2012) 1825e1837.
[36] Soroka, K. B.; Soroka, J. A. Polish patent 147726, 1989.
[37] A. Bromberg, K.A. Muszkat, J. Am. Chem. Soc. 91 (11) (1969) 2860e2866.
[38] D.B. Min, J.M. Boff, Compr. Rev. Food Sci. F 1 (2) (2002) 58e72.
[39] C.S. Foote, F.C. Shook, R.B. Abakerli, Methods Enzymol. 105 (1984) 36e47.
[40] A. Bromberg, K.A. Muszkat, E. Fischer, Isr. J. Chem. 10 (1972) 765e773.
[41] J.A. Soroka, Chem. Scr. 29 (1989) 361e365.
[17] M. Carrera, M. de la Viuda, A. Guijarro, Synlett 27 (20) (2016) 2783e2787.
[18] K.B. Soroka, J.A. Soroka, Tetrahedron Lett. 21 (48) (1980) 4631e4632.
[19] K.B. Soroka, J.A. Soroka, Chem. Scr. 29 (1989) 167e171.
ꢀ
[20] M.J. Sawicka, J.A. Soroka, E.K. Wroblewska, I.K. Zawadzka, Pol. J. Chem. 80
(2006) 1337e1351.
[21] J.A. Soroka, Chem. Scr. 29 (1989) 367e373.
[22] J.K. Howard, K.J. Rihak, C.J.T. Hyland, A.C. Bissember, J.A. Smith, Org. Biomol.
Chem. 14 (2016) 8873e8880.
[42] J.A. Soroka, J. Photochem. Photobiol., A 64 (1992) 171e182.
[43] J.A. Soroka, I.W. Bogdanska, A.R. Kosmider, K.B. Soroka, J. Photochem. Photo-
ꢀ
ꢀ
biol., A 73 (1993) 35e45.
[23] M.R. Eskandarian, M. Fazli, M.H. Rasoulifard, H. Choi, Appl. Catal. B Environ.
183 (2016) 407e416.
[24] M.H. Rasoulifard, M. Fazli, M.R. Eskandarian, J. Ind. Eng. Chem. 20 (2014)
3695e3702.
[25] M.H. Rasoulifard, M. Fazli, M.R. Eskandarian, J. Ind. Eng. Chem. 24 (2015)
121e126.
[26] E. Meggers, Chem. Commun. 51 (2015) 3290e3301.
[27] T.P. Nicholls, G.E. Constable, J.C. Robertson, M.G. Gardiner, A.C. Bissember, ACS
[44] M.J. Sawicka, J.A. Soroka, M. Ga˛siorowska, Pol. J. Chem. Technol. 12 (1) (2010)
17e22.
[45] E.K. Wroblewska, Szczecin Univ. Technol. (2002) 40e42. PhD Dissertation.
[46] C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, third ed., VCH,
Weinheim, 2003.
[47] M. Bregnhøj, M. Westberg, F. Jensen, P.R. Ogilby, Phys. Chem. Chem. Phys. 18
(2016) 22946e22961.
ꢀ
Please cite this article as: M.J. Sawicka et al., Synthesis of 7H-indolo[1,2-a]quinolinium derivatives via oxidative photocyclization of 3H-indolium
salts using blue LED photoreactor, Tetrahedron, https://doi.org/10.1016/j.tet.2019.06.002