S. Britto et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 120 (2014) 489–493
491
(J = 6 Hz) (8). The C3 protons in 7 and 8 were observed as a triplet
around d 3.79 and 3.23 ppm (J = 6 Hz) respectively. The C4 protons
of 7 showed a quintet in the upfield at d 2.41 ppm. The multiplets
appeared like singlets for C3, C4, C5 and C6 protons in 9 may be
due to the relaxation effect. The same kind of effect could be seen
for all the ring protons in 15. The merged multiplet appeared for C4
and C5 protons of 8. The aromatic protons of 7–9 also appeared as
merged multiplets. The salt 12 is obtained as a mixture of rota-
mers, therefore, all the eight aromatic protons appeared as a
merged mutiplet. Two singlets for C1 protons adjacent to the pos-
itively charged nitrogen atom and two triplets for C3 protons were
appeared. One triplet for C4 protons of two rotamers appeared, this
may be due to the same environment possessed by all the four pro-
tons. For the two methyl groups, four singlets appeared like two
tively. The amines 27–29 showed C„C stretch in the range of
1598–1597 cmÀ1
.
The merged multiplet was observed for aromatic protons of 16–
29 in the range of d 7.45–7.11 ppm. A sharp singlet appeared in the
range of d 3.73–3.42 ppm for benzylic protons of 16–26 and the
same was shifted to d 3.99 ppm for 27–29. This peak appeared as
broad singlet for 28. The ring protons of these amines appeared
in the range of d 1.69–1.23 ppm and the same was shifted to d
2.42–2.21 ppm for 25 and 27. This is due to the electron withdraw-
ing effect of phenyl rings as well as the triple bonds. The character-
istic triplet for CH2 protons adjacent to the ring nitrogen appeared
in the range of d 2.68–2.26 ppm for 16–20, 28 and 29. These
triplets are looking like singlets due to the relaxation effect in
21–23. In all the amines, methylene protons in the alkyl groups
showed merged peaks with ring protons.
The ipso carbon of the compounds 16–19 and 20–23 was
observed in the range d 141.72–141.55 ppm, for 27 and 28, in the
range d 140.02–138.62 ppm, and for 7 membered amines 24–26,
in the range d 143.25–143.08. This may be due to the substituent
effect. 13C NMR spectra of 16, 25–29 showed a signal for the qua-
ternary carbon in the region of d 60.26–60.02 ppm. The same was
observed in the region of d 64.82–64.51 ppm for 17–19 and d
57.65–57.25 ppm for 21, 22, and 27. The six membered amines
20 and 28 showed the quaternary carbon at d 53.24 ppm and d
58.23 ppm respectively. The seven membered amine 24 showed
the same carbon at d 55.98 ppm. The signal for alkyne carbons in
27 were showed at d 87.82 and 83.69 ppm. The same in 28 were
observed at d 88.31 and 84.40 ppm. These carbons in 29 presented
at d 90.72 and 82.05 ppm. The difference in the chemical shifts may
be due to the different ring size.
doublets. The appearance of
a signal around at d 193.69–
187.52 ppm in 13C NMR and mass peaks in HRMS confirmed the
presence of iminium moiety in all the thioiminium ions.
Synthesis of bis-alkylated and alkynylated amines
We report that thioiminium salts thus prepared from lactams
and amides, can be converted into 2,2-disubstitued amines by
reaction with simple nucleophile i.e. organocerium reagents. Treat-
ment of the thioiminium salt 7 with allylmagnesium bromide
afforded the desired gem-diallylated pyrrolidine [1]. The best re-
sults are obtained with 3 equiv of allylmagnesium bromide in
THF at room temperature.
In contrast to Klaver et al. [6] neither the solvent nor the tem-
perature influences the product distribution. For instance, when
the reaction of 7 was performed with only one equivalent of allyl-
magnesium bromide in dichloromethane at À78 °C, no monoallyla-
tion was observed. Instead, the bisallylated amine was obtained in
40% yield together with lactam 7 arising from aqueous workup.
Introduction of geminal dialkyl group proved not to be feasible
by using either alkylmagnesium halides or alkyllithium deriva-
tives, despite the isolated example reported by Klaver et al. [6].
These reagents are presumably too basic and deprotonate the
thioiminium salts rather than undergoing the desired addition.
Organocerium derivatives are often used to avoid undesired depro-
tonation during addition of organolithium and organomagnesium
reagents to ketones [7]. Alkylcerium dichlorides, easily prepared
from the commercially available organolithium derivatives are
tested with thioiminium iodides (7–9, 12 and 15). These reagents
reacted cleanly with the thioiminium iodides 7–9, 12 and afforded
the gem-dialkyl derivatives 16–19, 20–23, 24–26 and 41–44 and
gem-dialkynyl amines 27–29 (Scheme 2) in good yields over the
three steps from the parent lactams. The progress of the reaction
was followed by TLC analysis using cyclohexane : TBME as eluent.
Interestingly, the reaction in the presence of 1 equiv of n-butylce-
rium with 7 affords 18 as single isolated product in 20% yield.
According to the results of the organomagnesium allylation, no
product resulting from the monoaddition was detected [8].
The reaction was also tested with acetamide 15, but was not
successful. Then, we prepared N-tert-butyltetrahydroisoquinoline
41 from acetamide 10 via the thioiminium ion 12. The acetyla-
tion-gem-dimethylation process represents a useful method for
the conversion of secondary amines into tert-butyl tertiary amines.
Secondary and tertiary alkylcerium reagents do not add to the
thioiminium salts and after aqueous work up, the parent lactams
are recovered [1]. It is unclear whether this result is due to a lack
of reactivity or a competitive deprotonation. Amines in Schemes
2, 3 and 5 are not stable for a long time at room temperature.
Hence, they are immediately converted to their hydrochlorides.
The compounds 16–19 and 20–26 showed their characteristic
CAH (aromatic), C@C ring stretch (aromatic) and C-N stretch in
the range of 3000–2996, 1605–1466 and 1258–1026 cmÀ1 respec-
The multiplets for aromatic protons in the region of d 7.43–
7.03 ppm were observed for 41–44. The C2 protons of these amines
showed a sharp singlet in the region of d 3.80–3.69 ppm and C3
and C4 protons appeared as multiplets in the region of d 2.83–
2.70 ppm.
Synthesis of amine hydrochlorides
The preparation of hydrochloride salts of above amines is
shown in Scheme 4 and 6. The characteristic N-H stretch of 30–
32, 36 and 38–40 could be seen between 2597–2534 cmÀ1
,
whereas, 33 and 37 showed the same at 2857 cmÀ1
. The
compounds 34 and 35 showed the N-H stretch between 2489–
2477 cmÀ1. This variation may be due to the difference in ring size.
A broad singlet was observed in the range d 12.10–11.03 and
10.96–10.38 ppm for 30–40. This difference may due to the varia-
tion in ring size. The aromatic protons peaks are well separated.
They show the coupling of protons clearly and explain the split-
ting pattern of benzylic as well as ring protons satisfactorily. The
spin coupling of NAH protons and geminal coupling of benzylic
protons resulted in the doublet of doublets appeared for benzylic
protons. In Fig. 1, the splitting of all the signals in 1H NMR of the
free amine 16 by NAH proton in the corresponding amine hydro-
chloride 30 has been shown. The COSY of 30 proves this spatial
arrangement of NAH very well (Fig. 2). The interaction of NAH
protons with the ring protons made the peaks for the ring protons
to be multiplets. This interaction is found to be weak in the cases of
34, 35 and 38.
The spectra of 30 and 35 showed the quaternary carbon in the
range of d 69.03–69.02 ppm, whereas, 31–33 show in the range d
75.87–75.17 ppm. The hydrochloride 34 showed the same at d
62.74 ppm, whereas, 36 and 37 showed in the range of d 68.39–
68.31 ppm. The seven membered amine hydrochloride 38 showed
the quaternary carbon at d 67.31 ppm, whereas, others 39 and 40
show in the range of d 73.78–73.10 ppm. This variation may be
due to the increase in number of a, ß, c and d carbons.