Table 1. Optimization of the organocatalytic triazole synthesis.[a]
Table 2. Scope of arylazides in the organocatalytic triazole synthesis.[a]
Azide
Triazole
D Yield [%][b,d]
MW Yield
[%][c,d]
1
2
3
4
5
6
7
8
9
4-(MeO)PhN3 (2a)
PhN3 (2b)
4-(Br)PhN3 (2c)
4-(Cl)PhN3 (2d)
4-(Me)PhN3 (2e)
4-(NO2)PhN3 (2 f)
3aa
3ab
3ac
3ad
3ae
3af
3ag
3ah
3ai
90[e]
85
58
83
85
60
73
80
–
90
68
70
Catalyst
Heating
t [h]
Yield [%][b]
88
1
2
4
4
4
4
5
6
7
8
D
D
92
90
100
1
1
1
1
1
1
100
67
[f]
[f]
–
75[c]
3,4-di
N
89
70
3[e]
4
D
92[c] (90)[d]
85[c] (83)[d]
3-(MeO)PhN3 (2h)
2-(MeO)PhN3 (2i)
2-(CH2OH)PhN3 (2j)
MW
MW
MW
MW
MW
MW
D
75
71
[f]
5
6
7
8
9
10
–
10
3aj
39[g]
48[g]
60
63
[a] Ketone 1a (2 mmol), arylazide (1 mmol) and proline (20 mol%) in
CH2Cl2 (1.5 mL). [b] Heating for 5–6 d. [c] Heating for 1 h. [d] Isolated
yield. [e] 10 mol% of proline were used. [f] Starting materials recovered.
[g] Decomposition of the azide 2j was observed under the reaction condi-
tions.
[f]
–
–
–
[f]
9
[f]
none
[a] Ketone 1a (1 or 2 mmol), arylazide 2a (1 mmol) and catalyst
(20 mol%) in CH2Cl2 (1.5 mL). [b] NMR yield using mesitylene as inter-
nal reference. [c] Ratio ketone/azide=2:1. [d] Isolated yield. [e] Per-
formed with 10 mol% of proline. [f] Starting materials recovered.
were also obtained with halogen-substituted arylazides
owing to decomposition into the corresponding anilines by
loss of nitrogen (Table 2, entries 3 and 4).[17] The position of
the substituent on the aromatic ring appeared to only have a
minor impact on the reaction (Table 2, entries 1, 8 and 9).
The reaction could even be performed with arylazides con-
taining a free hydroxyl group, such as 2j (Table 2, entry 10).
In addition, the reactivity of several ketones was evaluat-
ed (Table 3). Interestingly, among the various ketones exam-
ined (Table 3, entries 1–7), cyclic ketones appeared as the
substrates of choice, as each ring size, with the exception of
cyclopentanone,[18] gave good to excellent yields under both
heating conditions, that is thermal and microwave irradia-
tion. The best yield, however, was obtained with cycloocta-
none (1c), which afforded the corresponding 1,2,3-triazole
in 90% isolated yield (Table 3, entry 3). It is worth noting
that dissymmetrical cyclic ketones afforded high levels of re-
gioselectivity. Hence, 3,3-dimethylcyclohexanone (1e) led to
a single regioisomer 3ea (Table 3, entry 5), in which the het-
erocycle is furthest from the gem-dimethyl group for steric
reasons. On the other hand, 4,4-dimethylcyclohexanone (1 f)
furnished triazole 3 fa, which is an isomer of 3ea. b-Tetra-
lone (1g) also led to the formation of a single regioisomer
3gb (Table 3, entry 7), the structure of which was confirmed
by X-ray analysis[19] (Figure 1). This regioselectivity can be
afforded similar levels of conversion after only one hour
(Table 1, entry 4).
Next, several secondary amines were tested under these
optimized conditions to evaluate their catalytic activity and
determine the structural features of the catalyst required to
promote optimum triazole formation. Hence, we were able
to show that a catalytic amount of L-pipecolinic acid (5),
which bears a larger ring size than proline, did not catalyze
the reaction (Table 1, entry 5). Interestingly, (S)-5-pyrroline-
2-yl-1H-tetrazole (6),[15] having an acid function with a pKa
close (tetrazole pKa =4.86 in H2O) to the pKa of proline
(pKa =4.75 in H2O), was able to promote the transformation
under these conditions, however, in lower yield (60%,
Table 1, entry 6). Surprisingly, pyrrolidine (7) catalyzed the
reaction, whereas morpholine (8) did not, highlighting the
effect of the ring size of the catalyst (Table 1, entries 7 and
8). Moreover, pyrrolidine resulted in lower yields than pro-
line (63 vs. 85%), showing the beneficial effect of the acid
moiety in proline. Finally, an acyclic secondary amine, such
as diethylamine (9), did not catalyze the reaction (Table 1,
entry 9). As expected, no 1,2,3-triazole was observed, when
the reaction was run in the absence of a catalyst (Table 1,
entry 10). Consequently, proline proved to be the best cata-
lyst for this reaction by combining a secondary amine of the
adequate ring size and an acidic function.
With these two sets of optimized conditions in hands, we
then examined the influence of the arylazide.[16,2d] The re-
sults are depicted in Table 2. Excellent results were ob-
served with arylazides functionalized with electron-donating
groups (Table 2, entries 1, 5, 7–9), whereas electron-with-
drawing groups, such as the nitro group, tended to deacti-
vate the arylazide partner (Table 2, entry 6). Lower yields
Figure 1. ORTEP view of the crystal structure of 3gb.
12918
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 12917 – 12921