About the intermediacy of 1,2-dihydroquinazolinium salts in the Friedl?nder-Borsche synthesis of quinolinium salts in acidic medium

Article abstract of DOI:10.1016/j.tetlet.2011.09.053

Spontaneously or under various heating conditions, 2-alkyl- or 2-aryl-(iminoalkyl)benzenamines react with ketones and triflic acid (1:1:1) to give quinolinium salts. When working under milder thermal conditions, intermediate 1,2-dihydroquinazolinium derivatives can be isolated or detected in solution but decompose upon standing or heating to give the corresponding quinolinium salts.

Full text of DOI:10.1016/j.tetlet.2011.09.053

Tetrahedron Letters 52 (2011) 6298–6302
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
About the intermediacy of 1,2-dihydroquinazolinium salts in the
Friedländer–Borsche synthesis of quinolinium salts in acidic medium
⇑
José Vicente , María Teresa Chicote, Antonio Jesús Martínez-Martínez
Grupo de Química Organometálica, Departamento de Química Inorgánica, Universidad de Murcia, Apartado 4021, 30071 Murcia, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Spontaneously or under various heating conditions, 2-alkyl- or 2-aryl-(iminoalkyl)benzenamines react
with ketones and triflic acid (1:1:1) to give quinolinium salts. When working under milder thermal con-
ditions, intermediate 1,2-dihydroquinazolinium derivatives can be isolated or detected in solution but
decompose upon standing or heating to give the corresponding quinolinium salts.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 1 August 2011
Accepted 12 September 2011
Available online 17 September 2011
Keywords:
Quinolinium salts
1,2-Dihydroquinazolinium salts
Friedländer–Borsche reaction
Nitrogen heterocycles
Ketones
1. Introduction
A few reactions have been described in which quinazoline it-
self10 or some oxo-11 or dihydro-derivatives12 react with active
Compounds based on the quinoline ring structure have found
important applications related to their biological activity,1 their
ability to form polymers with excellent electronic or optical prop-
erties2,3 and, recently, as efficient corrosion inhibitors for steel.4
This explains the ongoing interest in finding new and efficient
routes for their syntheses.2,5,6 The Friedländer reaction, which in-
volves the condensation of 2-(aminoaryl)carbonyl compounds
a
-methylene compounds or with enamines or ynamines,13 to give
quinoline derivatives, and some alkylquinazolinium salts react
with quaternary heterocyclic salts yielding substituted heteroaryl
quinolines.14
R4
O
R4
with aldehydes or ketones containing a-methylene groups, is one
N
NH2
R3
of the most simple, efficient and straightforward methods for the
synthesis of quinolines.7 The limitation arising from the tendency
of 2-(aminoaryl)aldehydes to undergo self-condensation, can be
overcome by using 2-(aminoaryl)imines instead (Friedländer–
Borsche reaction). Two alternative reaction pathways6–8 for the
mechanism of the Friedländer reaction have been proposed. The
Borsche version of these two ways is shown in Scheme 1.
In pathway I the first step, rate determining, consists of the for-
mation of a Schiff base (A). An intramolecular aldol-like reaction
follows, to give a 3,4-dihydro-4-NHR2-quinoline (B) that leads to
quinoline (C) upon loss of R2NH2. Pathway II proposes that the
intermolecular aldol-like condensation occurs first to give (D), fol-
lowed by a dehydration/cyclization process that gives (C) through
the intermediacy of (B). Although most of the evidence is in favour
of the first reaction pathway, recent experimental works support
pathway II.6,8,9
R3
NR2 - H2O
NR2
1
A
R1
R1
R4
R3
I
O
II
R4
NH
N
2O
R4
- H2O
R3
NHR2
R3
NHR2
R1
R1
D
B
- R2NH2
N
R4
C
R3
R1
⇑
Corresponding author. Tel.: +34 868 884143; fax: +34 868 887785.
E-mail address: jvs1@um.es (J. Vicente).
Scheme 1. Proposed reaction pathways for the Friedländer–Borsche synthesis of
quinolines.
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.09.053

J. Vicente et al. / Tetrahedron Letters 52 (2011) 6298–6302
6299
NH2
NR2
isolated, analysed and studied by 1H NMR but its 13C{1H} NMR
showed some impurities formed during the acquisition time, indi-
cating its limited stability in solution.15 The NMR spectra also
showed the transformation of the DHQS 2 derived from ketones,
spontaneous or induced by soft heating, into the corresponding
quinolinium triflates 3 upon loss of R2NH2. The exception was
the reaction of 1d with methyl vinyl ketone (MVK) and TfOH
(CDCl3 at 25 °C), which afforded 2d5 (entry 12) as the only species
present in solution. Its concentration did not change with time at
25 °C but, when rising the temperature to 45 °C, the reaction pro-
ceeded slowly and no more changes were observed after 10 h. At
this point the NMR showed a complex mixture in which we could
not assess or deny the presence of the expected quinolinium salt or
the 4-iminium-1,2,3,4-tetrahydroquinoline compound resulting
after an isomerization process, as we have observed for other
DHQS bearing a Me group at the 4-position and electron withdraw-
ing substituents on C(2).15 In all other cases, complete conversion
of 2 into 3 + R2NH2 occurred in less than 10 h at 50 °C and no fur-
ther transformation of the quinolinium salts was observed.
R3CHO
TfOH
R1
1
1/n R3(CHO)n
TfOH
H
N
H
R3
+
NR2
TfO-
H
N
R1
E
H
R3(CHO)m
1/n
+
NR2
n-m
(TfO-)n-m
R1
F
Scheme 2. Synthesis of 1,2-dihydroquinazolinium salts from 2-alkyl or 2-aryl(imi-
noalkyl)benzenamines, aldehydes and TfOH. The counterion of cationic species is
TfOꢀ.
We have recently reported15 the reactions taking place between
2-alkyl- or 2-aryl(iminoalkyl)benzenamines H2NC6H4C(R1)@NR2-2
(1, Scheme 2), aldehydes R3CHO and triflic acid (TfOH) to give 1,2-
dihydroquinazolinium triflate salts (DHQS, E, Scheme 2). The
reaction worked also using the appropriate amounts of di- and tri-
aldehydes to afford the corresponding mono-, bis- or tris-DHQS
(F),15 but failed with most ketones; it also worked when the iminoa-
cyl Pd(II) complex trans-[PdI{C(@NXy)C6H4NH2-2}(CNXy)2] (i.e., a
compound of type 1 with R1 = trans-PdI(CNXy)2 and R2 = C6H3Me2-
2,6 = Xy) was treated with TfOH and with a variety of carbonyl
compounds, including aldehydes and ketones, to afford the corre-
sponding 1,2-dihydroquinazolinium-4-yl palladium complexes.16,17
In this paper, we study the reaction between 2-amino substi-
The instability of all these DHQS derived from ketones contrasts
with the great stability of their 1,2-dihydroquinazolinium-4-yl pal-
ladium homologues,16 or with that of DHQS derived from alde-
Table 1
Friedländer–Borsche synthesis of quinolinium salts 3 mediated by 1,2-dihydroqui-
nazolinium salts in acidic medium
R4
R4
H
N
H
N
R3
- R2NH2
NH2
NR2
R4
R3
O
R3
NR2
H+
3
R1
R1
1
2
R1
tuted arylimines, various ketones bearing at least one a-methylene
Entry
R1, R2 (1)
R3, R4
DHQS (2)
Quinolinium salt (3)
group (MeC(O)R (R = Me, Et, i-Pr, Bz, Ph) or Et2C(O)) and TfOH
(Table 1). At low temperature, DHQS form but, spontaneously or
after soft heating, they convert into quinolinium salts upon loss
of R2NH2. This suggests that DHQS could be intermediates in the
Friedländer–Borsche synthesis of quinolinium salts in an acidic
medium. This transformation has not been reported in the
literature.
1
2
H, Xy (1a)
H, Cy (1b)
H, Me
H, Me
2a
2b
3a
3a
3a
H
Et
N
3
H, Tol (1c)
H, Me
2c1
H
H
+
3c2
2. Results and discussion
H
Me
N
When we reported the syntheses of DHQS E and F (Scheme 2)
with R1 different from [Pd],15 we mentioned that the reactions
using ketones (Me2C(O) or MeC(O)Et) showed differences with re-
spect to those with aldehydes, regarding both the stability and the
nature of the resulting products. In this Letter, we report the re-
sults obtained when decided to study these decomposition pro-
cesses in depth, following the reactions by variable temperature
1H NMR spectroscopy, and including in the study various other ke-
4
H, Tol (1c)
H, Et
2c2
Me
H
3c2'
5
6
7
8
9
H, Tol (1c)
H, Tol (1c)
H, Tol (1c)
Me, Cy (1d)
Me, Cy (1d)
H, Ph
Me, Et
H, H
H, Me
H, i-Pr
2c3
2c4
2c5
2d1
2d2
3c3
3c4
a
3d1
3d2
3d3
tones bearing at least an
a-methylene group which we reacted
H
N
Bz
with compounds 1a–d and TfOH (Table 1). The reactions were car-
ried out by dissolving the appropriate compound 1 and ketone in
CDCl3 at ꢀ30 °C and adding the equimolar amount of TfOH. The
temperature was then raised (1 °C/min) and one spectrum was
measured every 10 min. The spectra showed that the reaction did
not start until TfOH was added and that DHQS (2, Table 1), started
to form immediately or after a few minutes, even at low tempera-
ture. In the case of 1d, a protonated species (G in Scheme 3) was
also detected after mixing the reagents.15 The stability in solution
of these DHQS varies from 2a (entry 1) or 2d5 (entry 12) which are
stable up to room temperature and can be isolated and fully char-
acterized, to 2c1–4 (entries 3–6) which are stable up to 10–20 °C
for a short period of time, or 2d1–4 (entries 8–11) which are iden-
tified in the interval of ꢀ30 to ꢀ10 °C, and decompose gradually
upon raising the temperature. In turn, 2b (entry 2) could be
10
Me, Cy (1d)
Me, Et
2d3
H
Me
+
3d4
H
Me
N
11
12
Me, Cy (1d)
Me, Cy (1d)
H, Bz
2d4
2d5
Ph
Me
3d4'
b
H, CH@CH2
a
The DHQS is stable after heating at 65 °C.
The DHQS is stable at room temperature but decomposes at T >45 °C to give a
b
mixture of unknown compounds. Xy = xylyl, Cy = cyclohexyl, Tol = 4-tolyl, i-
Pr = isopropyl. The counterion of cationic species is TfOꢀ.

6300
J. Vicente et al. / Tetrahedron Letters 52 (2011) 6298–6302
H
N
The influence of the R1 and R2 substituents seems not to be as
R4
2 R3
H
N
R4
R3
NH2
NR2
important as that of CH2R3 or R4. For instance, although the electron
withdrawing ability of both R1 and R2 would favour the K ? L step,
their actual influence in the process seems to be scarce in view of the
stability of 2a (R1 = H, R2 = Xy) being greater than that of 2b (R1 = H,
R2 = Cy) and much greater than that of 2d1 (R1 = Me, R2 = Cy).
In the reactions where CH2R3 = R4 or when only one of these
- R2NH2
(i)
H+
NR
R1
R1
H+
1
2
3
R1
- R2NH2
substituents bears an
salt is obtained. However, when CH2R3 is different from R4 and
both bear an -methylene group, both substituents are available
a-methylene group, only one quinolinium
R4
R3
NR2
H
N
H
N
R4
R3
NH2
NHR2
a
for deprotonation by the NR2 group in J giving rise to two isomeric
quinolinium salts. This is the case of the reaction of 1c with TfOH in
MeC(O)Et producing a mixture containing TolNH2 (Tol = 4-Tolyl),
3c2 and 3c20 in a 2:1:1 molar ratio (entry 4). Recrystallization of
the isolated mixture allowed us to enrich it in one of the isomers
(3c2:3c20 = 7:1, see Experimental Section) thus facilitating the
NMR assignments. When the reaction was repeated at room tem-
perature for 30 min and Et2O was added to the concentrated reac-
tion mixture, an orange solid insoluble in CDCl3 was obtained. In its
1H NMR spectrum in CD3CN, we detected some of the resonances
that could be assigned to the intermediate DHQS 2c2. Similarly,
the reaction of 1d with MeC(O)Bz and TfOH produced a mixture
containing CyNH2, 3d4 and 3d40 in 10:7:3 molar ratio (entry 11).
The easier deprotonation of the Me group compared to Bz could
be attributed to steric reasons or, again, to the phenyl group acting
as a net electron releasing group.
H
G
H
R1
NHR2
R1
L
J
R1
(i)
R4
R4
H
N
N
R3
R3
NHR2
NHR2
R1
K
R1
Scheme 3. Proposed reaction pathway for the synthesis of 1,2-dihydroquinazolin-
ium salts 2 and their decomposition into quinolinium salts 3. Dashed arrows are
proposed reaction transformations and solid arrows, observed reactions. The
counterion of cationic species is TfOꢀ. (i) = R3CH2C(O)R4.
hydes (E and F, Scheme 2), including those derived from acetalde-
In summary, all the DHQS of type 2 shown in Table 1 decom-
posed in solution, spontaneously or after soft heating to give the
corresponding quinolinium derivatives 3 (with the exception of
the DHQS obtained from MVK). This contrasts with the great stabil-
ity of their 1,2-dihydroquinazolinium-4-yl palladium homo-
logues16 and the DHQS derived from aldehydes.15
hyde and therefore having an a-methylene group. In fact DHQS E
with R1 = Me, R2 = cyclohexyl = Cy or CH(Me)Ph, R3 = Me (Scheme
2)15 or that with R1 = H, R2 = Tol, R3 = R4 = H (2c5, entry 7) were
recovered quantitatively after heating them at 65 °C.
Trying to account for these observations, we propose in Scheme
3 a plausible reaction pathway for the syntheses of quinolinium
salts 3 from the reactions of 2-(amino)benzenimines (1), the
appropriate ketone, and TfOH (a Friedländer–Borsche synthesis of
quinolinium salts in an acidic medium). The DHQS 2 would likely
form through the reaction pathway 1 ? G ? H ? J ? 2 as con-
firmed by the isolation of some intermediates (e.g., of type G)15
and a DFT study (to be published). We propose that the decompo-
sition of 2 into quinolinium salts 3 occurs through the reaction
pathway 2 ? J ? K ? L ? 3. The 2 ? J transformation, which
seems to be the key step, would be facilitated by CH2R3 and R4 hav-
ing electron donor ability (+I or +M effect). This would favour the
cleavage of the C(2)–N(3) bond displacing the equilibrium towards
J. From here, the positive charge on N(1) in J would increase the
3. Conclusion
When working in an acidic medium, DHQS can be intermediates
in the Friedländer–Borsche synthesis of quinolinium salts involv-
ing 2-(aminoaryl)imines and ketones containing an
-methylene functionality.
a-methyl or
a
4. Experimental section
The syntheses of compounds 2a, 2b and 2c1 were reported by
us.15 The following DHQS salts were identified in the reaction
mixture by some of their resonances (others were obscured by
the resonances corresponding to other components in the mix-
ture):Compound 2c1 (400 MHz, CDCl3, 25 °C, TMS) d 1.75 (s, 6H,
acidic character of the a-methylene protons facilitating the proton-
ation of the NR2 nitrogen (K) and the attack of the resulting meth-
ylene carbon to the partly positive iminium carbon producing a
new heterocycle (L) with an exocyclic NHR2 fragment. Abstraction
3
Me2C), 2.46 (s, 3H, MeTol), 6.85 (‘‘dt’’, 1H, JHH = 7.6 Hz,
4JHH = 0.8 Hz), 7.14 (m, 2H), 7.23–7.44 (m obscured by the reso-
nances of other components, 4H, ortho-Tol + meta-Tol), 7.58 (ddd,
of one proton of the a
-methylene group by the NHR2 fragment in L
would give quinolinium salt 3, with concomitant loss of R2NH2.
This decomposition way of 2 through J, is related to pathway I pro-
posed for the Friedländer–Borsche reaction (Scheme 1) in which A
and B protonated at N(1) correspond to intermediates J and L in
Scheme 3.
3
3
4
1H, JHH = 8.4 Hz, JHH = 7.2 Hz, JHH = 1.6 Hz), 8.07 (s, 1H, H4),
8.34 (s br, 1H, NH). Compound 2c2: 1H NMR (200 MHz, CD3CN,
Et
3
25 °C, TMS): 0.91 (t, 3H, CH3 , JHH = 7.4 Hz), 1.93 (s, 3H, Me), 2.41
(q, 2H, CHE2t, JHH = 7.2 Hz). Compound 2c3: 1H NMR (400 MHz,
3
CDCl3, 45 °C, TMS): d 2.28 (s, 3H, MeTol), 2.33 (s, 3H, Me), 8.86 (s,
The successful synthesis of 3c3 from 1c, acetophenone and
1H, CH@NTol). Compound 2c4: 1H NMR (400 MHz, CDCl3, 10 °C,
TfOH (entry 5) suggests that the
p donor ability (+M effect) of
3
TMS): d 2.37 (s, 3H, Me), 7.92 (d, 1H, JHH = 8.7 Hz, Ar), 8.67 (s,
the Ph group in 2c3, likely enhanced by the positive charge at
N(3), is crucial in this reaction given that the homologous reaction
between 1c, acetaldehyde and TfOH, produces the stable DHQS 2-
methyl-3-tolyl-1,2-dihydroquinazolinium triflate (2c5) (entry 7)
and not even traces of the corresponding quinolinum salt were de-
tected. In view of the similar electronic effects of phenyl and vinyl
groups, the failed attempt to decompose 2d5 into a quinolinium
salt could be attributed to steric reasons or to alternative decom-
position pathways competing with that leading to the quinoline.
1H, CH@NTol). Compound 2d1: 1H NMR (200 MHz, CDCl3,
ꢀ30 °C, TMS): d 1.75 (s, CMe2), 2.92 (s, MeC@N), 4.05 (m, CH, Cy).
Compound 2d2: 1H NMR (600 MHz, CDCl3, ꢀ10 °C, TMS): d 1.21–
2.18 (several m, 10H, Cy), 2.72 (s, 3H, MeC@N), 3.68 (m, CHCy),
3
3
6.80 (m, 2H, JHH = 7.8 Hz, Ar), 6.93 (d, 1H, JHH = 7.8 Hz). Com-
pound 2d3: (1H NMR (600 MHz, CDCl3, ꢀ10 °C, TMS): d 2.66 (s,
3H, MeC@N), 3.66 (m, CHCy) 6.92 (d, 1H, JHH = 7.8 Hz, Ar), 7.31
3
(t, 1H, JHH = 7.6 Hz, Ar). Compound 2d4 at 1H NMR (300 MHz,
3

J. Vicente et al. / Tetrahedron Letters 52 (2011) 6298–6302
6301
CDCl3, 55 °C, TMS): d 1.21–2.16 (several m, 10H, Cy), 2.54 (s, 3H,
MeC@N), 3.62 (m, CHCy), 6.63 (d, 1H, 3JHH = 8.1 Hz, Ar). We describe
below the synthesis and characterization of the isolated DHQS 2c5
and 2d5.
Et2O was added (20 mL, 0 °C). A suspension formed, which was
stirred at 0 °C for 15 min and filtered. The solid collected was
washed with Et2O (3 ꢁ 5 mL, 0 °C) and dried under a N2 stream
to give 3a as a light brown solid. Yield: 82 mg, 0.28 mmol, 65%.
Mp: 105 °C. HRMS (ESI) m/z calcd for C10H10N [M]+, 144.0808;
found 144.0812. 1H NMR (400 MHz, CDCl3, 25 °C, TMS): d 3.11 (s,
Chloride salts of the quinolinium cations present in 3a,18 3c2,18
19
3c20 and 3d120 have been reported. The NMR data of 3a, 3c2,18
3
3
are almost identical to those in our triflates (see below), while
the NMR data of the chlorides of 3c20,19 and 3d120 were not re-
ported. The quinolinium salt 3c318 has been identified by its previ-
ously reported 1H NMR spectrum. The remaining quinolinium salts
(3c4, 3d2, 3d4 and 3d40) were characterized upon treating a CDCl3
solution with one drop of concentrated aqueous NaOH to afford the
corresponding quinoline (3 + NaOH ? Q3 + NaTfO). The mixture
was briefly shaken until the yellow colour vanished and filtered
through a short column of anhydrous MgSO4. The quinolines were
identified by their previously reported 1H NMR (Q3c4,21 Q3d2,22
Q3d4)23 or UV–Vis (Q3d40)24 spectra. As the quinolinium salt 3d3
and its corresponding quinoline were unknown, we report below
its synthesis and characterization.
3H, Me), 7.78 (d, 1H, JHH = 8.7 Hz), 7.87 (t, 1H, JHH = 7.6 Hz),
3
3
8.05 (t, 1H, JHH = 7.8 Hz), 8.14 (d, 1H, JHH = 8.4 Hz), 8.46 (d, 1H,
3JHH = 8.4 Hz), 8.83 (d, 1H, 3JHH = 8.7 Hz). 13C{1H} NMR
(100.8 MHz, CDCl3, 25 °C, TMS): d 20.8, 120.3 (q, JCF = 319 Hz),
1
120.6, 123.2, 127.0, 128.7, 129.8, 135.1, 137.8, 146.4, 157.6. IR (Nu-
jol, cmꢀ1):
m(NH) 3302, m(C@N + C@C) 1653, 1607, 1544.
Alternatively, 3a formed quantitatively, along with the equimo-
lar amount of XyNH2 or CyNH2, when a solution of 2a or 2b in
CDCl3 was heated for 6 h at 45 or 60 °C, respectively, as shown
by 1H NMR.
4.4. Synthesis of 3c2 + 3c20 from 1c, MVK and TfOH
To a solution of 1c (150 mg, 0.71 mmol) in methyl ethyl ketone
(20 mL) was added TfOH (65 lL, 0.74 mmol). The reaction mixture
4.1. Synthesis of 2c5 from 1c, acetaldehyde and TfOH
was stirred at room temperature for 1 h and then heated at 80 °C
for 24 h. The reaction mixture was concentrated under vacuum
to dryness to give an oily residue, which was shown by 1H NMR
to contain only 3c2, 3c20 and TolNH2, in a 1:1:2 molar ratio. The
reaction crude was washed with a mixture of CH2Cl2/n-pentane
(1:20, 21 mL) and vacuum dried for 1 h to give a pale brown solid
consisting of a 1:1 mixture of isomers 3c2 and 3c20. Yield: 147 mg,
0.48 mmol, 67%. HRMS (ESI) m/z calcd for C11H12N [M]+, 158.0964;
found 158.0966. The isolated 1:1 mixture of isomers was enriched
in 3c20 (3c20:3c2 = 7:1) by recrystallizing it twice from CH2Cl2/Et2O
(1:20, 2 ꢁ 21 mL, 0 °C). Yield: 93 mg, 0.30 mmol, 42%. 1H NMR
(400 MHz, CDCl3, 25 °C, TMS): 3c2: 1H NMR (400 MHz, CDCl3,
To a solution of 1c (150 mg, 0.71 mmol) in CHCl3 (5 mL, 0 °C)
were successively added acetaldehyde (70
lL, 1.24 mmol) and
TfOH (70 L, 0.80 mmol) and the reaction mixture was stirred at
l
room temperature for 30 min. The resulting orange solution was
concentrated under vacuum to dryness to give an oily material
which was dissolved in CH2Cl2 (5 mL), filtered through a short
pad of anhydrous MgSO4, and concentrated to dryness. The residue
was washed three times with a 1:20 mixture of CH2Cl2 and Et2O
(21 mL), and vacuum dried to give 2c5 as a dark orange solid. Yield:
164.9 mg, 0.4268 mmol, 60%. Mp: 69 °C. HRMS (ESI) m/z calcd for
C
16H17N2 [M]+, 237.1386; found 237.1388. 1H NMR (400 MHz,
3
3
25 °C, TMS): d 1.52 (t, 3H, Me, JHH = 7.8 Hz, Me), 3.39 (q, 2H,
CDCl3, 25 °C, TMS): d 1.58 (d, 2H, JHH = 6.4 Hz, MeCH), 2.39 (s,
3
3
3H, MeTol), 6.08 (q, 1H, JHH = 6.0 Hz, MeCH), 6.84 (‘‘dt’’, 1H,
CH2, JHH = 7.8 Hz), 7.79–7.88 (t + d obscured by the resonances
of the 3c20 isomer, 2H), 8.05 (t obscured by the resonances of
4
3
3JHH = 7.6 Hz, JHH = 0.8 Hz), 6.99 (d, 1H, JHH = 8.0 Hz), 7.33 (d,
3
3
3c20, 2H), 8.13 (d, 1H, JHH = 8.4 Hz), 8.52 (d, 1H, JHH = 8.7 Hz),
3
3
2H, JHH = 8.4 Hz, Tol), 7.44 (d, 2H, JHH = 8.4 Hz, Tol), 7.52 (ddd,
8.85 (d, 1H, JHH = 8.7 Hz). 13C{1H} NMR (100.8 MHz, CDCl3, 25 °C,
TMS): d 13.4, 27.9, 120.3 (q, JCF = 319 Hz), 121.0, 121.6, 127.2,
3
3
3
4
1H, JHH = 8.4 Hz, JHH = 6.8 Hz, JHH = 1.2 Hz), (dd overlapped v br
1
s, 1H each, 3JHH = 8.0 Hz, 4JHH = 1.2 Hz), 8.64 (s, 1H, CH@N). IR (Nu-
128.5, 129.7, 135.2, 138.0, 146.5, 162.6. Compound 3c20: d 2.63
(s, 3H, Me), 3.03 (s, 3H, Me), 7.79–7.88 (various m obscured by
the resonances of the 3c2 isomer, 1H), 7.96 (‘‘dt’’, 1H, 3JHH = 8.4 Hz,
4JHH = 1.2), 8.05 (d, obscured by the resonances of 3c2, 1H), 8.41 (d,
jol, cmꢀ1):
m(NH) 3272, m(C@N + C@C) 1630, 1601, 1566.
4.2. Synthesis of 2d5 from 1d, MVK and TfOH
3
1H, JHH = 8.4 Hz), 8.62 (s, 1H). 13C{1H} NMR (100.8 MHz, CDCl3,
To a solution of 1d (150 mg, 0.69 mmol) in CH2Cl2 (5 mL, 0 °C)
1
25 °C, TMS): d 18.5, 19.3, 120.3 (q, JCF = 319 Hz), 120.6, 127.3,
were successively added MVK (60
lL, 0.73 mmol) and TfOH
127.7, 129.8, 132.0, 134.0, 136.7, 144.9, 157.4.
(70 L, 0.80 mmol). After 45 min of stirring at 0 °C, the yellow solu-
l
tion was concentrated under vacuum to dryness to give an oily res-
idue which was converted into a suspension by stirring it with a
CH2Cl2/Et2O mixture (1:15, 16 mL, 0 °C, 20 min). The suspension
was filtered under a nitrogen stream, and the yellow solid col-
lected, 2d5, was dried by suction. Yield: 187.3 mg, 0.4476 mmol,
4.5. Synthesis of 3d1 from 1d, acetone and TfOH
To a solution of 1d (90 mg, 0.42 mmol) in CHCl3 (5 mL) were
successively added acetone (30 lL, 0.43 mmol) and TfOH (38 lL,
65%. Mp: 133 °C. HRMS (ESI) m/z calcd for
C
18H25N2 [M]+,
0.44 mmol). The reaction mixture was heated to 55 °C for 1 h.
The resulting orange solution was filtered through a short pad of
Celite, concentrated under vacuum (1 mL) and a mixture Et2O/n-
pentane (1:2, 30 mL, 0 °C) was added. The suspension was filtered
and the solid collected was dried by suction to give 3d1 as a pale
yellow solid. Yield: 94.6 mg, 0.31 mmol, 71%. Mp: 151 °C. HRMS
(ESI) m/z calcd for C11H12N [M]+, 158.0964; found 158.0958. 1H
NMR (400 MHz, CDCl3, 25 °C, TMS): d 2.96 (s, 3H, Me), 3.04 (s,
269.2012; found 269.2015. 1H NMR (300 MHz, CDCl3, 25 °C,
TMS): d 1.11–1.92 (several m, 10H, CH2Cy), 2.79 (s, 3H, Me), 2.81
(s, 3H, Me), 3.99 (m, 1H, CHCy), 6.78–6.83 (m, 2H), 6.88 (m, 1H),
3
4
6.94 (dd, 1H, JHH = 8.1 Hz, JHH = 1.5 Hz), 7.31 (m, 2H), 7.38 (dd,
1H, JHH = 8.5 Hz, JHH = 1.5 Hz). IR (Nujol, cmꢀ1):
m(NH) 3270,
3
4
m
(C@N + C@C) 1633, 1615, 1567.
3
3H, Me), 7.58 (s, 1H, CH@C), 7.87 (t, 1H, JHH = 8.0 Hz), 8.02 (‘‘dt’’,
3
3
4
4.3. Synthesis of 3a from 1c, acetone and TfOH
1H, JHH = 8.2 Hz, JHH = 8.2 Hz, JHH = 1.2 Hz), 8.19 (d, 1H,
3JHH = 8.8 Hz), 8.44 (d, 1H JHH = 8.4 Hz), 15.17, (v br s, 1H, NH).
3
To a solution of 1c (90 mg, 0.43 mmol) in acetone (20 mL) was
added TfOH (40 L, 0.46 mmol). The reaction mixture was stirred
13C{1H} NMR (100.8 MHz, CDCl3, 25 °C, TMS): d 19.9, 20.6, 120.3
1
l
(q, JCF = 320 Hz), 121.6, 123.6, 124.7, 126.6, 129.5, 134.7, 137.5,
at room temperature for 1 h and then refluxed for 6 h. The result-
ing orange solution was concentrated under vacuum (1 mL) and
156.3, 157.3. IR (Nujol, cmꢀ1):
1603, 1525.
m(NH) 3303, m(C@N + C@C) 1651,

6302
J. Vicente et al. / Tetrahedron Letters 52 (2011) 6298–6302
4.6. Synthesis of 3d3 from 1d, Et2C(O) and TfOH
Acknowledgments
To a solution of 1d (25 mg, 0.1155 mmol) in CHCl3 (0.5 mL)
We thank Ministerio de Educación y Ciencia (MEC), FEDER
(CTQ2007-60808/BQU) and Fundación Séneca (04539/GERM/06)
for financial support and A.J.M.M. to MEC for a FPU grant.
were successively added Et2C(O) (13
lL, 0.1229 mmol) and TfOH
(12 L, 0.1375 mmol). The reaction mixture was heated to 55 °C
l
for 10 h. The resulting yellow solution was filtered through a short
pad of Celite and Et2O (15 mL, 0 °C) was added. The resulting sus-
pension was filtered and the solid collected was dried by suction to
give 3d3 as a yellow solid. Yield: 27.1 mg, 0.0808 mmol, 70%. Mp:
103 °C. HRMS (ESI) m/z calcd for C13H16N [M]+, 186.1277; found
186.1276. 1H NMR (400 MHz, CDCl3, 25 °C, TMS): d 1.42 (t, 3H,
3JHH = 7.6 Hz, MeEt), 2.54 (s, 3H, Me), 2.81 (s, 3H, Me), 3.27 (q, 2H,
References and notes
1. Michael, J. P. Nat. Prod. Rep. 2008, 25, 166.
2. Palimkar, S. S.; Siddiqui, S. A.; Daniel, T.; Lahoti, R. J.; Srinivasan, K. V. J. Org.
Chem. 2003, 68, 9371.
3. Jyothish, K.; Avirah, R. R.; Ramaiah, D. Org. Lett. 2006, 8, 111.
4. Ebenso, E. E.; Obot, I. B.; Murulana, L. C. Int. J. Electrochem. Sci. 2010, 5, 1574.
5. (a) Kozlov, N. G.; Gusak, K. N.; Kadutskii, A. P. Chem. Heterocycl. Compd. 2010,
46, 505; (b) Patteux, C.; Levacher, V.; Dupas, G. Org. Lett 2003, 5, 3061; (c)
McNaughton, B. R.; Miller, B. L. Org. Lett. 2003, 5, 4257; (d) Selvam, K.;
Swaminathan, M. Tetrahedron Lett. 2010, 4911; (e) Martínez, R.; Ramón, D. J.;
Yus, M. J. Org. Chem. 2008, 73, 9778.
6. Jiang, B.; Dong, J. J.; Jin, Y.; Du, X.-L.; Xu, M. Eur. J. Org. Chem. 2008, 2693.
7. Marco-Contelles, J.; Pérez-Mayoral, E.; Abdelouahid, S.; Carreiras, M. C.;
Soriano, E. Chem. Rev. 2009, 109, 2652.
3
4
CH2Et), 7.73 (‘‘dt’’, 1H, JHH = 8.2 Hz, JHH = 1.2 Hz), 7.86 (‘‘dt’’, 1H,
4
3
3JHH = 8.4 Hz, JHH = 1.2 Hz), 8.14 (d, 1H, JHH = 8.4 Hz), 8.34 (d,
1H, JHH = 8.4 Hz); NH resonance not observed. IR (Nujol, cmꢀ1):
3
m(NH) 3354, m(C@N + C@C) 1637, 1611, 1535.
4.7. 1H NMR resonances for the unreported quinolinium salts
3c4, 3d2, 3d4 and 3d40
8. Muchowski, J. M.; Maddox, M. L. Can. J. Chem. 2004, 82, 461; Yang, D.; Jiang, K.;
Li, J.; Xu, F. Tetrahedron 2007, 63, 7654.
9. Sridharan, V.; Ribelles, P.; Ramos, M. T.; Menéndez, C. J. J. Org. Chem. 2009, 74,
5715.
Compound 3c4: 1H NMR (400 MHz, CDCl3, 25 °C, TMS): d 1.48 (t,
3
3
3H, JHH = 7.6 Hz, MeEt), 2.69 (s, 3H, Me), 3.36 (q, 2H, JHH = 7.6 Hz,
10. Higashino, T.; Ito, H.; Hayashi, E. Chem. Pharm. Bull. 1972, 20, 1544.
11. Higashino, T.; Nagano, Y.; Hayashi, E. Chem. Pharm. Bull. 1975, 23, 746.
12. Higashino, T.; Suzuki, K.; Hayashi, E. Chem. Pharm. Bull. 1978, 26, 3485.
13. Miyashita, A.; Taido, N.; Sato, S.; Yamamoto, K.; Ishida, H.; Higashino, T. Chem.
Pharm. Bull. 1991, 39, 282.
14. Gromov, S. P.; Razinkin, M. A.; Drach, V. S.; Sergeev, S. A. Russ. Chem. Bull. 1998,
47, 1179.
15. Vicente, J.; Chicote, M. T.; Martínez-Martínez, A. J.; Bautista, D.; Jones, P. G. Org.
Biomol. Chem. 2011, 9, 2279.
16. Vicente, J.; Chicote, M. T.; Martínez-Martínez, A. J.; Bautista, D. Organometallics
2009, 28, 5915.
17. Vicente, J.; Chicote, M. T.; Martínez-Martínez, A. J.; Bautista, D. Organometallics
2011, 2425.
18. Tadaoka, H.; Cartigny, D.; Nagano, T.; Gosavi, T.; Ayad, T.; Genêt, J.-P.; Ohshima,
T.; Ratovelomanana-Vidal, V.; Mashima, K. Chem. Eur. J. 2009, 15, 9990.
19. Gagan, J. M. F.; Lloyd, D. Chem. Commun. 1967, 1043.
20. Campbell, K. N.; Schaffner, I. J. J. Am. Chem. Soc. 1945, 67, 86.
21. Martínez, R.; Ramón, D. J.; Yus, M. Tetrahedron 2006, 62, 8988.
22. Molander, G. A.; Colombel, V.; Braz, V. A. Org. Lett. 2011, 13, 1852.
23. Brown, T. M.; Cooksey, C. J.; Dronsfield, A. T.; Wilkinson, A. S. Appl. Organomet.
Chem. 1996, 10, 415.
3
CH2Et), 7.82 (t, 1H, JHH = 8.4 Hz), 8.00 (m, 2H), 8.49 (d, 1H,
3JHH = 8.8 Hz), 8.62 (s, 1H); the NH resonance was not observed.
Compound 3d2: 1H NMR (400 MHz, CDCl3, 25 °C, TMS): d 1.49 (s,
6H, 3JHH = 6.8 Hz, Mei-Pr), 2.93 (d, 3H, 3JHH = 0.8 Hz, Me), 3.66 (sept,
3
3
1H, JHH = 6.8 Hz, CHi-Pr), 7.54 (d, 1H, JHH = 0.8 Hz), 7.78 (‘‘dt’’, 1H,
3JHH = 8.0 Hz, 4JHH = 1.2 Hz), 7.97 (‘‘dt’’, 1H, 3JHH = 8.2 Hz,
4JHH = 1.2 Hz), 8.14 (dd, 1H, JHH = 7.8 Hz, JHH = 0.8 Hz), 8.14 (d,
3
4
3
1H, JHH = 8.4 Hz); the NH resonance was not observed. Com-
pounds 3d4 + 3d40 (1:0.4): 1H NMR (400 MHz, CDCl3, 25 °C,
TMS), d 2.59 (s, 3H, Me), 2.67 (s, 3H, Me), 2.71 (s, 3H, Me), 4.45
(s, 2H, CH2), 7.18–7.35 (m, 6H, Ph + CH), 7.52–7.57 (m, 3H, Ph),
3
4
7.69 (dd, 2H, JHH = 8.2 Hz, JHH = 1.2 Hz, Ph), 7.79 (ddd, 1H,
3JHH = 8.0 Hz, JHH = 6.8 Hz, JHH = 0.8 Hz), 7.85 (ddd, 1H, JHH
=
3
4
3
3
4
3
8.4 Hz, JHH = 6.8 Hz, JHH = 1.2 Hz), 7.95 (ddd, 1H, JHH = 8.4 Hz,
3JHH = 7.2 Hz, JHH = 1.2 Hz), 8.03 (d, 1H, JHH = 8.4 Hz), 8.17 (dd,
4
3
3
4
3
24. Kempter, G.; Möbius, G. J. Prak. Chem. 1966, 34, 298.
1H, JHH = 8.0 Hz, JHH = 0.4 Hz), 8.31 (d, 1H, JHH = 8.4 Hz), 8.41
3
(d, 1H, JHH = 8.4 Hz); the NH resonance was not observed.