Control of the dual reactivity (iminium-dienamine) of β-arylmethyl α,β-unsaturated aldehydes in organocatalytic 1,3-dipolar cycloadditions with N-benzoyl C,N-cyclic azomethine imines

Article abstract of DOI:10.1021/jo5018519

1,3-Dipolar cycloadditions of C,N-cyclic azomethine imines with α,β-unsaturated aldehydes can be performed with complete control of the regio-, exo-, and enantioselectivity under aminocatalytic conditions. The so far never studied competence of the iminium-dienamine reactivity inherent to β-alkyl α,β-unsaturated aldehydes was studied, which was possible by allowing achievement of complete control of the chemoselectivity in reactions of the β-arylmethyl derivatives with azomethine imines by using different additives and organocatalysts, whose role has been rationalized by DFT calculations and chemical proofs. Thus, it has been possible to selectively obtain the pyrazolidines resulting from both the attack to the C2-C3 (via iminium) and the C3-C4 (via dienamine) bonds at the starting enals, which can be used as precursors of interesting tetrahydroisoquinolinic compounds.

Full text of DOI:10.1021/jo5018519

Subscriber access provided by WESTERN KENTUCKY UNIV
Article
Control of the Dual Reactivity (Iminium-Dienamine) of beta-Arylmethyl
alpha,beta-Unsaturated Aldehydes in Organocatalytic 1,3-Dipolar
Cycloadditions with N-Benzoyl C,N-Cyclic Azomethine Imines
Cristina Izquierdo, Francisco Esteban, Alejandro Parra, Ricardo
Alfaro, José Alemán, Alberto Fraile, and José Luis García Ruano
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/jo5018519 • Publication Date (Web): 08 Oct 2014
Downloaded from http://pubs.acs.org on October 17, 2014
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Organic Chemistry is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Control of the Dual Reactivity (Iminiumꢀ
Dienamine) of βꢀArylmethyl α,βꢀUnsaturated
Aldehydes in Organocatalytic 1,3ꢀDipolar
Cycloadditions with NꢀBenzoyl C,NꢀCyclic
Azomethine Imines
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cristina Izquierdo, Francisco Esteban, Alejandro Parra, Ricardo Alfaro, José
Alemán,* Alberto Fraile* and José Luis García Ruano
Departamento de Química Orgánica (Módulo 1), Facultad de Ciencias, Universidad
Autónoma de Madrid. 28049ꢀMadrid, Spain
Eꢀmail: jose.aleman@uam.es, alberto.fraile@uam.es
RECEIVED DATE (to be automatically inserted after your manuscript is accepted
if required according to the journal that you are submitting your paper to)
Abstract ꢀ 1,3ꢀDipolar cycloadditions of C,Nꢀcyclic azomethine imines with α,βꢀ
unsaturated aldehydes can be performed with a complete control of the regioꢀ, exoꢀ, and
enantioselectivity under aminocatalytic conditions. The so far never studied competence
of the iminiumꢀdienamine reactivity inherent to βꢀalkyl α,βꢀunsaturated aldehydes was
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
studied, being possible by allowing to achieve the complete control of the
chemoselectivity in reactions of the βꢀarylmethyl derivatives with azomethine imines by
using different additives and organocatalysts, whose role has been rationalized by DFT
calculations and chemical proofs. Thus, it has been possible to selectively obtain the
pyrazolidines resulting from both the attack to the C2ꢀC3 (via iminium) and the C3ꢀC4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(via dienamine) bonds at the starting enals, which can be used as precursors of
interesting tetrahydroisoquinolinic compounds.
Introduction
1
One of the most studied substrates in organocatalysis are α,βꢀUnsaturated
aldehydes due to the versatility derived from their easy transformation into electrophilic
(iminium) or nucleophilic (dienamine) species with pyrrolidine type catalysts, both of
them offering two reactive positions. Iminium species are able to be attacked by
nucleophiles at Cꢀ1 and Cꢀ3 whereas dienamines react with electrophiles at Cꢀ2 and Cꢀ4
(Scheme 1). As the result of this dual reactivity and the ambivalent character of the
2
3
4
formed species, it has been possible to get the α−, β−, and γꢀfunctionalization of the
5
enals, as well as the attack to their carbonyl carbons (Cꢀ1). The regioselectivity control
difficulties derived from the ambivalent character of the iminium species have been
3
,5
overcome by the wellꢀdefined tendencies of the nucleophiles to attack on Cꢀ3 or Cꢀ1,
whereas for dienamines this problem was solved modulating the steric hindrance at Cꢀ2
6
and Cꢀ4. On the other hand, chemoselectivity control difficulties derived from the dual
reactivity (iminium vs dienamine) were obviated by using reagents without dual
reactivity (nucleophiles or electrophiles).
2
ACS Paragon Plus Environment

Page 3 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Different reactivies of α,βꢀunsaturated aldehydes under aminocatalysis.
1,3ꢀDipolar cycloadditions (1,3ꢀDC) have a preeminent role in the construction
of heterocyclic systems and heterofunctionalized acyclic molecules resulting from their
7
reductive ring opening. The simultaneous formation of iminium and dienamine species
from β−alkyl α,β−unsaturated aldehydes (R = CH R' at Scheme 1) in the presence of
2
8
pyrrolidine type organocatalysts suggests that their 1,3ꢀDC with dipoles, presumably
also with ambivalent reactivity, would be an appropriate scenario for studying the
chemoselectivy derived from the dual reactivity (iminiumꢀdienamine) of the enals.
9
However, the so far studied dipoles have only exhibited reactivity with the iminium
species, and therefore these chemoselectivity studies (dienamineꢀiminium) could not be
performed.
Organocatalytic reactions of dipoles with iminium species, considered as
10
11
normalꢀelectronꢀdemand (NED) cycloadditions, are well known. By contrast, the
1
0
enantioselective organocatalytic inverseꢀelectronꢀdemand (IED) 1,3ꢀDCs are, to our
knowledge, scarcely known and mainly restricted to the use of vinylethers catalyzed by
12
Brønsted acids. Interestingly, the dienamine has not been applied in IED 1,3ꢀDCs
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
reactions, thus remaining as a challenge. At this point we decided to investigate these
1
3
reactions using the C,Nꢀcyclic azomethine imines as dipoles, recently incorporated in
12a,13a,b
enantioselective [2+3]ꢀdipolar cycloadditions.
These dipoles can be easily
prepared and give access to pharmaceuticaly attractive chiral substituted
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
14
tetrahydroisoquinoline skeletons. Moreover, species I might exhibit the desired dual
reactivity (NED and IED) that is not found with other dipoles. In this work we describe
the 1,3ꢀDC of the Nꢀbenzoyl C,Nꢀcyclic azomethine imines with βꢀaryl and βꢀalkyl α,βꢀ
unsaturated aldehydes under different organocatalytic conditions which have
demonstrated their efficiency for obtaining cycloadducts on the fragments C(2)ꢀC(3)
(NED processes via iminium I) and C(3)ꢀC(4) (IED processes via dienamine II) in a
completely chemoselective way, simultaneously controlling the regioꢀ, enantioꢀ, and
endo/exo- selectivities (Scheme 2). The breaking of the NꢀN bonds at the cycloadducts
resulting by both routes provided interesting tetrahydroisoquinoline derivatives.
N
Q
R
CHO
k1
via Iminium ion
N
Bz
NED 1,3ꢀDC
N
H
I
OHC
R
R
—
Enantioselectivity control
N
Bz
K
N
— Endo/exo selectivity control
—
—
Regioselectivity control
Chemoselectivity control
(Dienamine vs iminium ion)
N
Q
R
CHO
k2
N
Bz
via Dienamine
IED 1,3ꢀDC
N
H
R
R
CHO
II
Scheme 2. Dual Reactivity (IminiumꢀDienamine) of βꢀArylmethyl α,βꢀunsaturated
aldehydes with NꢀBenzoyl C,NꢀCyclic Azomethine Imines.
4
ACS Paragon Plus Environment

Page 5 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
During the preparation of this manuscript, the IED 1,3ꢀDCs of the dipole 1a with
15
some of the dienamines studied in this paper has been published. As the results
reported in the aforementioned publication showed serious discrepancies with our
results, we have added some comments and experiences concerning it (see addendum at
the end of the results discussion section).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Results and Discussion
Screening conditions and first experiments.
We started our study with βꢀaryl (or heteroaryl) α,βꢀunsaturated aldehydes,
unable to form dienamine species, with the aim of knowing the possibilities of the C,Nꢀ
cyclic azomethine imines in the asymmetric 1,3ꢀDC under organocatalytic conditions,
despite the synthetic problem has been elegantly solved by Maruoka by using titaniumꢀ
13a
BINOLate complexes as catalysts. After optimizing the conditions in the reaction of
dipole 1a with the aldehyde 2a (see S.I.), we found that the use of the Jørgensenꢀ
Hayashi’s catalyst 3A (5 mol%) in toluene at room temperature afforded the expected
aldehyde that was in situ transformed into the alcohol 4a (84% yield, two steps) for
16
determining its enantiomeric excess (98% ee, entry 1 at Table 1). Similar conditions
were applied to other α,βꢀunsaturated aldehydes and the results are collected in Table 1.
The presence of electronꢀdonating (2b and 2c, entries 2ꢀ3) or electronꢀwithdrawing
groups (2d, 2e, and 2f, entries 4ꢀ6) did not significantly modify the efficiency of these
reactions. Similar behavior was observed for the αꢀfuryl derivative 2g (entry 7), but the
pyridyl derivative 2h (entry 8) exhibited a lower reactivity, being necessary longer
reaction times (80% conversion after 5 days) and a higher catalytic loading (20 mol%).
Finally, reactions of the βꢀalkyl enals 2i and 2j were also studied (entries 10 and 12).
17
They gave less satisfactory results with the catalyst 3A but we found that the use of
3
B cleanly provided 4i and 4j in high yields after longer reaction times (3 days). The
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
higher efficiency of 3B was also checked for the reaction of the β−pyridyl enal 2h
(compare entries 8 and 9), which was completely transformed into the aldehyde in only
16 h. As 4i was obtained in 85% ee under conditions of entry 10, we repeat the reaction
at 0 ºC increasing the amount of catalyst to 20 mol%, being thus able to obtain 4i
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
enantiomerically pure (>99% ee) after 12 h in 61% yield (entry 11).
Table 1. Results obtained in reactions of the dipole 1a with the aldehydes 2aꢀ2j
catalyzed by 3.
Entry
Aldehyde (R)
Catalyst Time (h) Product (ee, %)
Yield (%)
84
b
1
2a (Ph)
3A
3A
3A
3A
3A
3A
3A
3A
3B
3B
3B
24
24
48
36
36
30
24
120
16
72
12
4a (98)
4b (96)
4c (98)
4d (>99)
4e (98)
4f (96)
4g (98)
4h (96)
4h (98)
4i (85)
b
2
2b (4ꢀMeOC
2c (2ꢀMeOC
2d (4ꢀClC
2e (4ꢀMeO CC H )
6
H
4
)
85
b
3
6
H
4
)
88
b
4
6
H
4
)
78
b
5
91
2
6
4
b
6
2f (4ꢀCF
3
C
6
H
4
)
78
b
7
2g (2ꢀFuryl)
2h (2ꢀPy)
2h (2ꢀPy)
2i (Me)
91
b,c,d
8
62
b,c
9
77
e
10
76
c,e,f
11
2i (Me)
4i (>99)
61
6
ACS Paragon Plus Environment

Page 7 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
e
12
2j (Et)
3B
72
4j (98)
78
a
b
Determined by supercritical fluid chromatography (SFC). Conditions: 0.2 mmol of 2
0.4 mL of toluene), 0.3 mmol of 1a. Catalytic loading 20 mol%. Conversion 80%.
Conditions: 0.4 mmol of 2 (0.4 mL of CH Cl ), 0.2 mmol of 1a. 0 ºC.
2 2
c
d
e
(
f
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Remarkably, both regioselectivity and endo/exo selectivity of all these reactions
were complete, only yielding the exoꢀadducts 4, and the enantiomeric excesses of the
obtained alcohols were excellent (96 to 99% ee, Table 1). Configurational assignment of
compounds 4 indicated in Table 1 was performed by comparison of the spectroscopic
parameters and specific rotations of 4a and 4i with those of the same compounds
13a
18
described by Maruoka (see S. I.).
As conclusion, we can state that asymmetric [3+2]ꢀcycloaddition of the dipole
13a
1a with conjugated enals, previously reported under titaniumꢀBINOLate complexes,
can be alternatively performed with similar efficiency under organocatalytic conditions
using a low catalytic loading (5 mol%). The reaction can be performed with aldehydes
containing coordinating substituents, which presumably could present some
incompatibilities with metallic catalysts. Moreover, the absence of cycloadducts
resulting in the 1,3ꢀDC of the dipole 1a to the dienamine species (βꢀalkyl enals) was
also interesting, whose formation would be possible for these substrates.
βꢀArylmethyl α,β−unsaturated aldehydes (2kꢀ2q) were chosen as dipolarophiles
to study the 1,3ꢀdipolar reactions with the species II (Scheme 2), because the formation
4
e,f
of the dienamine would be favored by aryl groups as it was previously postulated.
The reaction of the dipole 1a with 2k in toluene, catalyzed by 3A (20 mol%) at rt,
afforded a 30:70 mixture of two compounds 5k and 6k (Scheme 3), that were identified
as the adducts resulting in the 1,3ꢀDC of 1a to the fragments C(2)ꢀC(3) (via iminium ion
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
I) and C(3)ꢀC(4) (via dienamine II). These results suggested the existence of the two
species in equilibrium, dienamine II and iminium ion I, with both being able to react
under the organocatalytic conditions used. Remarkably, the exclusive formation of 6k
as the only adduct resulting via dienamine indicated that the 1,3ꢀDC of II is completely
regioꢀ and exo-selective.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Despite these promising results, both chemoselectivity (5k/6k ratio) and
enantioselectivity (75% ee for 6k) were moderate and should be improved. It prompted
us to investigate the influence of the reaction parameters on both types of selectivities,
with the aim of finding conditions allowing the exclusive evolution through the iminium
or dienamine species, with the highest possible ee. A selection of the most significant
results obtained in this optimization with 2k was collected in Table 2 (see S.I. for full
screening).
Scheme 3. Reactions of 1a with 2k catalyzed by 3A.
2 2
Initially we explore the influence of different catalysts (3Aꢀ3D) in CH Cl by
using 20 mol% of catalytic loading (entries 6ꢀ9). Mixtures of two products 5k and 6k
were obtained in all the cases. The use of the catalysts 3A and 3B provided the best
results, with the first one giving the highest amount of 6k (entry 9), resulting in the
evolution via dienamine, whereas 3B mainly yielding the product via iminium 5k (entry
6). As it occurred in reactions at Table 1, the enantiomeric excesses could not be
8
ACS Paragon Plus Environment

Page 9 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
1
9
established directly by supercritical fluid chromatography (SFC). The influence of the
solvent on the course of these reactions was moderate or low, being only significant the
improvement of the yield observed in toluene for reactions catalyzed by 3B (compare
entries 5 and 6). The use of benzoic acid (20 mol%) as additives increased the
proportion of 6k, providing almost equimolecular mixtures of 5k and 6k in reactions
catalyzed by 3B (entry 4), which were scarcely dependent of the pKa of the used acid
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
see the influence of other acids in S.I). Similarly the addition of oꢀfluorobenzoic acid to
reactions catalyzed by 3A also increased slightly the ratio 5k:6k up to 23:77 (entry 10).
Much more important was the influence of the tetrabutylamonium bromide (TBAB) as
additive in reactions catalyzed by 3B, which produced a significant improvement of the
proportion of 5k that was exclusively obtained in very high enantiomeric excess (entry
3
). These reactions were additionally optimized by decreasing the catalytic loading to
0 mol% (entry 2) and the temperature to 0 ºC (entry 1) and these conditions were used
1
in the scope. Less successful were the trials to optimize chemoselectivity of the
reactions modifying the reaction parameters in order to favor the evolution through
dienamines. In this sense, moreover the small effect of the acidic additives (entry 10), it
was only remarkable the slight improvement in the proportion of 6k observed by
decreasing the temperature to 0 ºC (compare entries 9 and 11).
a
Table 2. Chemoselectivity optimization for the 1,3ꢀDC of 1a and 2k.
9
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
b
c
d
Entry
Cat.
Solvent
Toluene
Toluene
Toluene
Toluene
Toluene
Additive
T
5k:6k
Yield (%) ee (%)
e
e
g
1
2
(S)ꢀ3B
(S)ꢀ3B
TBAB
0 ºC >98 :2
70
70
73
98
96
98
g
TBAB
Rt
Rt
>98 :2
>98 :2
50 : 50
71:29
72:28
66:34
27:73
26:74
23:77
20:80
2:>98
g
3
(S)ꢀ3B
(S)ꢀ3B
(S)ꢀ3B
(S)ꢀ3B
(S)ꢀ3D
(S)ꢀ3C
(S)ꢀ3A
(S)ꢀ3A
(S)ꢀ3A
(S)ꢀ3A
TBAB
h
f
f
f
4
BzOH
Rt
(ꢀ)
(96)
(85)
(66)
(51)
55
ꢀ /ꢀ
5
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Rt
92/90
92/88
5/2
6
CH
2
2
Cl
2
2
2
2
2
2
2
Rt
7
CH
Cl
Rt
8
CH Cl
Rt
68/51
2
f
9
CH
CH
CH
CH
2
2
2
2
Cl
Cl
Cl
Cl
Rt
90/ꢀ
i
f
10
OFBA
Rt
58
90/ꢀ
f
11
ꢀ
ꢀ
0 ºC
0 ºC
58
90/ꢀ
j
12
68
94
a
Conditions: 0.2 mmol of 2 (0.2 mL of the indicated solvent), 0.1mmol of 1a and 20
mol% of catalyst. Determined by H NMR analysis. Major isomer after derivatization
b
1
c
d
(
see text) and flash chromatography. Combined yields in brackets. Determined by
e
supercritical fluid chromatography (SFC) after derivatization. 10 mol% of the catalyst
f
g
h
i
3
B was used. Not determined. 200 mol% of additive. Benzoic acid (20 mol%). oꢀ
j
Fluorobenzoic acid (20 mol%). Hydrated dipole 1a´ was used.
10
ACS Paragon Plus Environment

Page 11 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
At this point, some erratic results were obtained by repeating the experiments
under conditions of the entry 9, which provided slightly different 5k:6k ratios working
under identical conditions. After performing an extensive experimentation we realized
that these differences were due to the amount of water contained in the highly
hygroscopic dipole, which was related to the time from its preparation and to the storing
conditions. The preparation of dipoles 1, according to the procedure reported by
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
3a
Maruoka, involved a final step of drying over Na
2 4
SO before evaporating the solvent
(
see S.I.). The composition of the so obtained fresh sample (1a) differed from that
resulting before this step (1a’) in one molecule of H O, as it was established by
2
thermogravimetric studies (see S.I.). After crystallization of 1a’, it was unequivocally
20
demonstrated, by Xꢀray diffraction studies, that it was the hemiaminal corresponding
to the incorporation of a molecule of water to the dipole 1a (Scheme 4). According to
this, the IR spectra of both samples mainly differ in the absorption of the OH group in
1
1
a’ (see S.I.). The HꢀNMR spectra of 1a and 1a’ were identical, but the second one had
an additional signal (δ= 1.6 ppm) which would belong to a water molecule. This
experiment suggested that the hemiaminal 1a’ was highly unstable in solution and was
quickly transformed into 1a (only detected by NMR), losing a water molecule.
Nevertheless this transformation is likely not to be so simple (see later) because the use
of the hemiaminal 1a’ (instead of the dipole 1a) in these 1,3ꢀDC had a large influence
on their chemoselectivity, increasing substantially the proportion of 6k, which becomes
exclusively obtained when the reactions were conducted under the conditions of the
2
entry 12 (3A, CH Cl2, and 0 ºC). Thus, conditions at entries 1 and 12 were respectively
considered as the optimal ones for obtaining the best results concerning
chemoselectivity (in both cases a very high enantioselectivity was observed) and they
were used for evaluating the scope of the 1,3ꢀDC of the dipole 1 to the iminium I and
dienamine II species generated from unsaturated aldehydes.
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 4. Structural differences between 1a and 1a’.
Scope of the 1,3ꢀDC via Iminum and Dienamine Species
We first studied the reactions of the hemiaminal 1a’ with different βꢀarylmethyl
α,β−unsaturated aldehydes under conditions of the entry 12 at Table 2. Resulting
aldehydes could be isolated but, in order to determine their enantiomeric purity by chiral
supercritical fluid chromatography (SFC), their in situ NaBH reduction into the
4
alcohols 7 were performed. Results are depicted in Table 3.
Table 3. Results obtained in reactions of the aldehydes 2 with the hemiaminals 1a’ꢀ1c’
under conditions of the entry 12 at Table 2.
a
Entry
Aldehyde (Ar)
2k (C
2l (pꢀMeC
ProꢀDipole (R)
1a’ (H)
Product (ee, %)
Yield (%)
1
2
6
H
5
)
7k (94)
68
75
6
H
4
)
1a’ (H)
7l (90)
12
ACS Paragon Plus Environment

Page 13 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
3
4
2m (oꢀMeC H )
1a’ (H)
1a’ (H)
7m (87)
40
73
6
4
2n (pꢀOMeC H )
7n (87)
7o (89)
6
4
5
1a’ (H)
76
2o(
)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
7
8
9
2p (pꢀFC
6
H
4
)
1a’ (H)
1a’ (H)
7p (92)
7q (92)
7r (88)
7s (92)
63
61
69
51
3 6 4
2q (pꢀCF C H )
2k (C
2k (C
6
H
H
5
)
1b’ (F)
6
5
)
1c’ (OMe)
a
Determined by supercritical fluid chromatography (SFC) after derivatization.
Only one alcohol 7 was obtained (>98:2 in all the cases), demonstrating that the
chemoselectivity control was almost complete under these conditions. We have
evaluated the influence on the reaction course exerted by the presence of electronꢀ
donating (2l, 2n, and 2o, entries 2, 4, and 5) and electronꢀwithdrawing groups (2p and
2
q, entries 6 and 7) at the para position of the aryl group. The enantioselectivity was
not significatively altered by the electronic effects of the substituents, with the ee of the
obtained alcohols 7 ranging between 87 and 92% (Table 3). The yields were good,
ranging between 61 and 76%. The presence of an orthoꢀsubstituent at the aryl group
(
2m, entry 3) significantly decreased the yield of 7m (40%). The hemiaminals 1b’ and
1c’ derived from other C,Nꢀcyclic azomethine imines, bearing substituents at Cꢀ6 of the
tetrahydroisoquinoline ring, also reacted in a completely chemoselective way, yielding
7
r and 7s in high enantiomeric excesses, respectively (entries 8 and 9). We have also
studied the reactions of other βꢀalkyl α,βꢀunsaturated aldehydes, like 2i and 2j (see
Table 1), under conditions of the Table 3. Unfortunately, complex mixtures were found
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
by HꢀNMR analysis which contained the signals of compounds 4i and 4j, whereas
those of 7i and 7j were not easily recognizable.
On the other hand, we have studied the reactions of the dipoles 1 (freshly dried)
with 2kꢀ2q under conditions of the entry 1 at Table 2 (see Table 4). Derivatization of
the resulting aldehydes 5kꢀ5s was now performed by Wittig reaction with
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
19
3 2
Ph PCHCO Et, exclusively yielding esters 8. The chemoselectivity control was also
complete under these conditions (>98:2), which determined the exclusive evolution via
iminium species. The introduction of electronꢀdonating (2l, 2n, and 2o, entries 2, 4, and
5
) and electronꢀwithdrawing groups (2p and 2q, entries 6 and 7) did not significantly
modify the obtained results and the ee of compounds 8 ranged between 94ꢀ99%. The
aldehyde 2m, with a methyl group at the ortho position, exhibited a similar behavior
(
entry 3). Other dipoles, derived from C,Nꢀcyclic azomethine imines with substituents at
the tetrahydroisoquinoline ring (1b and 1c), also evolved in a completely
chemoselective way, exclusively yielding 8r and 8s in high ee (entries 8 and 9). The
yield was good for 8r (78%) but only moderate for 8s (41%), which evidenced a
negative influence of the electron donating group at dipole, also detectable in the
formation of 7s (51%, entry 9 at Table 3).
Table 4. Results obtained in reactions of the aldehydes 2 with the dipoles 1aꢀc under
conditions of the entry 1 at Table 2.
a
Entry
Aldehyde (Ar)
Dipole (R)
Product (ee, %)
Yield (%)
14
ACS Paragon Plus Environment

Page 15 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
1
2
3
4
2k (C
2l (pꢀMeC
2m (oꢀMeC H )
6
H
5
)
1a (H)
1a (H)
1a (H)
1a (H)
8k (98)
8l (94)
70
75
71
85
6
4
H )
8m (99)
8n (98)
6
4
2n (pꢀOMeC
o(
2p (pꢀFC
2q (pꢀCF
2k (C
2k (C H )
6 4
H )
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
1a (H)
8o (98)
78
2
)
6
7
8
9
6
H
4
)
1a (H)
1a (H)
8p (99)
8q (96)
8r (>99)
8s (96)
72
75
78
41
3
C
6
H )
4
6
H
5
)
1b (F)
1c (OMe)
6
5
a
Determined by supercritical fluid chromatography (SFC) after derivatization.
The absolute configurations of the adducts resulting in the cycloaddition via
dienamine (Table 3) or iminium (Table 4) species, were unequivocally assigned by Xꢀ
ray analysis, from the hydrochlorides 10·HCl (1S, 2S, 10bS) and 9·HCl (1R, 2R, 10bR),
obtained in the reductive amination of the corresponding aldehydes (see Figure 1 and
21
S.I.). As expected, the configuration assigned to compounds of Tables 1 and 4 (both
resulting in the evolution of the reactions via iminium species) were identical, as it
could be anticipated by assuming a similar stereochemical course.
15
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
N
Bz
N
H
NH Cl
2
=
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Br
9
—HCl
N
H
N Bz
F3C
H N
=
2
Cl
Br
1
0—HCl
Figure 1. Determining the absolute configuration of 9 and 10.
Synthetic applications
Compounds at Tables 3 and 4 can be used as intermediates for preparing
interesting enantiomerically pure products containing tetrahydroisoquinoline skeletons.
As the key step of many of these transformations is the breaking of their NꢀN bonds, we
have studied these reactions on the cycloadducts 6q (92% ee) and 5k (98% ee) as
representatives of compounds from Tables 3 and 4 (see Scheme 5). The cleavage of the
2
NꢀN bond could be easily performed with SmI . However, this reagent affects to the
CHO group and therefore it could not be directly used on aldehydes 5k and 6q, unless
they were previously modified. The simple formation of the acetal from 6q allowed to
solve the problem and the hydrogenolysis of the NꢀN bond provided enantiopure 2ꢀ
substituted tetrahydroisoquinolines that can be used as intermediates of many other
16
ACS Paragon Plus Environment

Page 17 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
1
4
transformations. On the other hand, the breaking of the NꢀN after homologation of the
CHO group into the (CH ) ꢀCO Me from 5k opened the access to benzoquinolizidine
2
2
2
derivatives 14.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 5. Chemical transformations on the cycloadducts.
Mechanistic Considerations: Chemoselectivity
When the experimental conditions had been found, allowing the exclusive
formation of the isomeric aldehydes 5 and 6, we tried to understand the influence of the
different factors on the chemoselectivity of these reactions. There were three variables
provoking strong changes in the chemoselectivity (dienamine:iminium, 6:5 ratio):
catalyst, dipole, and additive. Their influence is summarized in Scheme 6. Other factors
like solvent and the electronic density of the aryl ring at 2 had a less marked influence
(see above).
The experimental results indicated that the reaction of the dipole 1a with the
aldehyde 2k, catalyzed by 3A, afforded a 74:26 mixture of 6k and 5k. This ratio was
completely inverted to 29:71, when 3B was used as the catalyst. The use of the
hemiaminal 1a’ (instead of dipole 1a) produced an increase of the proportion of 6k that
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
was completely chemoselective in the presence of 3A (left, Scheme 6). Finally, when
reactions were conducted in the presence of tetrabutylammonium bromide (TBAB) a
significant increase in the proportion of 5k was observed with both catalysts (the
influence of other additives was much lower). Thus, the reaction catalyzed by 3B
became completely chemoselective, yielding iminium product 5k (rightꢀScheme 6). We
will try to understand the role of the three factors (catalyst, nature of the dipole, and
presence of TBAB) on the chemoselectivity control.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Catalyst 3
(
20 mol%)
N
Bz
N
Bz
Ph
N
N
O
+
H
H
1a or 1a'
2k
CHO
OHC
Ph
Ph
5k
6k
Catalyst Factor
Additive Factor
(with dipole 1a)
Dipole Factor
(with dipole 1a)
Ph
3A
with dipole 1a´
Ph
with TBAB
N
H
OTMS
>
98 : 2
74 : 26
6k : 5k
30 : 70
6k : 5k
6
k : 5k
Ar
3B
Ar
OTMS
Ar = 3,5ꢀ(CF ) ꢀC H
6 3
N
with dipole 1a´
H
with TBAB
3
2
3
0 : 70
k : 5k
29 : 71
6k : 5k
2 : >98
6k : 5k
6
Scheme 6. Different factors affecting the chemoselectivity (iminium vs dienamine) of
the 1,3ꢀDC of dipole 1a (or 1a’) with 2k.
a) Influence of the Catalyst: The strong dependence of the ratio of cycloadducts 5
(generated via iminium ion) and 6 (generated via dienamine) with the nature of the
catalyst used (see middle, Scheme 6) must be explained taking into account both the
concentration of the intermediates I and II formed from each catalyst (K in Scheme 2)
18
ACS Paragon Plus Environment

Page 19 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
and, mainly, their relative reactivity with the dipole (K and K in Scheme 2). The
3
4
1
relative concentration of dienamine II was studied by HꢀNMR, by mixing 2k (1 equiv.)
with the catalysts 3A or 3B (0.5 equiv). According to our results (Scheme 7), the
22
formation of IIA was faster and its proportion higher than those of IIB. As the
formation of dienamines involved deprotonation of the iminium species, the lower
basicity of 3B, bearing aromatic rings with stronger ꢀI effects, could explain the slower
formation of IIB. On the other hand, the higher stability of the dienamines IIA had been
estimated by theoretical calculations (see S.I.).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
Scheme 7. HꢀNMR detection of the dienamines IIA and IIB.
As expected, the signals corresponding to the unstable iminium species I were
1
not visible in the HꢀNMR spectra, indicating they exist in a very low concentration
(much lower than that of the dienamines II). Despite this, significant amounts of
compounds 5 are formed in these reactions, thus suggesting a higher reactivity of the
iminium species.
Concerning the influence of the catalysts on the reactivity of the different
species (I and II) involved in our reaction, we analyzed, by DFT calculations, the
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
energetic content of the frontier molecular orbitals for dipole 1a, iminium ions (IA and
IB), and dienamines (IIA and IIB) (see Figure 2). We took into account the corrections
2 2
of the solvents used in the reactions (CH Cl for reactions of the dienamines and toluene
for those of the iminium species). According to a SustmannꢀHuisgen type analysis, the
reactions of dipole 1 with the dienamines II should be mainly governed by the
interaction HOMOdienamineꢀLUMOdipole, (ꢀE value of the HOMOdipoleꢀLUMOdienamine
interaction is too large for taking it into consideration), which agrees with the IED
character expected for the 1,3ꢀdipolar cycloaddition of dienamines. By contrast, the
interactions HOMOdipoleꢀLUMOiminium should be the most important in the reactions of
the dipole with the iminium ions, thus confirming their NED character.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Differences in energy induced by the catalysts in the HOMO orbital of the
dienamines (top, Figure 2), determined in CH
2 2
Cl , (ꢀEIIB (2.52)–ꢀEIIA (2.36) = 0.16 eV
~
3.7 kcal/mol) allow us to predict that IIA should react with 1a more easily than IIB.
Thus, taking into account that reactivity and concentration (see above) of IIA were both
higher than those of IIB, the experimentally observed higher amount of compounds 6
obtained in reactions catalyzed by 3A is completely justified. From the energetic
content of the frontier molecular orbitals for dipole 1a and the iminium ions IA and IB
(bottom, Figure 2), calculated in toluene, we can conclude the higher reactivity of the
later ones (ꢀEIA (2.05)–ꢀEIB (1.78) = 0.27 eV ~ 6.2 kcal/mol), thus explaining the
higher efficiency of the catalyst 3B observed in reactions at Table 1.
On the other hand, on the basis of this analysis, iminium ions seem to be more
reactive than the dienamines with both catalyst, but this tendency is more marked in
reactions catalyzed by 3B, where the energetic difference of the interacting orbitals for
2
3
each species is larger (ꢀEIIB (2.52)–ꢀEIB (1.78) = 0.74 eV ~ 17.1 kcal/mol).
20
ACS Paragon Plus Environment

Page 21 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Energies of the FMO calculated for dipole 1a, and the dienamines (IIA and
IIB) and iminium ions (IA and IB) formed by 2k with the catalysts 3A and 3B, taking
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
into account the correction of the solvents indicated in each case. DFT calculations were
carried out at B3LYP/6ꢀ311+g(d,p) level of theory.
The fact that reactions via dienamine, observed in the case of 1a, has not been
detected in organocatalytic reactions of azomethine ylides with βꢀalkyl α,βꢀunsaturated
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
9
aldehydes is intriguing. In order to obtain some explanation, we calculated the energy
associated to the frontier orbitals of these dipoles (EHOMO = ꢀ5.0 eV, ELUMO = ꢀ1.88 eV,
see S. I.), and found they are substantially higher than those obtained for the Nꢀbenzoyl
C,Nꢀcyclic azomethine Imines shown in Figure 2. From these values, it can be deduced
that reactions of azomethine ylides with the iminium species (IA or IB) should be
substantially easier and those with the enamines (IIA and IIB), more difficult, than
those observed with our Nꢀbenzoyl C,Nꢀcyclic azomethine Imines, which could
satisfactorily explain (see. Figure S1 in S. I.) the observed differences in the behavior of
24
both dipoles.
b) Influence of the Dipole Structure. As we have previously indicated, the way to
prepare the dipole had important consequences in the chemoselectivity. As the iminium
and dienamine species were not able to react directly with the hemiaminal 1a’, this
should be considered as the precursor of the real reactive species. In order to explain the
experimental results, we propose that the conversion of 1a’ into dipole 1a takes place in
+
two steps, with formation of the protonated species 1a(H ) as intermediate. This species
could only react as electrophile with the dienamine species II, resulting in the formation
of compounds 6, according to a nonꢀconcerted twoꢀstep process involving the formation
of the intermediates IV (Scheme 8). By contrast, the dipole 1a could react with both,
iminium ion (I) and dienamine (II) species, respectively affording the isomers 5 and 6
according to concerted or twoꢀstep processes. From Scheme 8 it can be deduced that the
22
ACS Paragon Plus Environment

Page 23 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
+
hydration of 1a into 1a’ would take also place in two steps (protonation to form 1a(H )
ꢀ
and subsequent attack of the OH ). It would explain the erratic results obtained in
different experiences performed when dipole was not freshly prepared, because it would
25
contain a variable amount of 1a’.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
H+
H+
HOꢀ
O
O
O
N
N
N
N
Ph
N
Ph
N
Ph
HOꢀ
OH
H
H
a´
1
1
a(H+)
1a
I or II
II
no reaction
II
I
N
NHCOPh
Ar
N
N
NCOPh
CHO
NCOPh
Ar
OTMS
Ar
Ar
OHC
N
Ar
6
5
IV
Scheme 8. Influence of the dipole source on the chemoselectivity.
c) Influence of the Additives: The positive influence of the quaternary ammonium salts
as efficient coꢀcatalysts on organocatalytic Michael additions to α,βꢀunsaturated
26
aldehydes catalyzed by 3 had been previously described. Moreover, in our reactions, a
double role of TBAB can be postulated. The first one consists in the dehydration of the
ꢀ
hemiaminal 1a’ by assisting the elimination of the OH in the first step, and the
deprotonation in the second one (Scheme 9). It would allow the quicker transformation
+
of 1a’ into the dipole 1a, thus minimizing the presence of the species 1a(H ) and
therefore reducing the amount of compounds 6 formed. On the other hand, we have
1
investigated by HꢀNMR the influence of the TBAB on the concentration of the
23
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
7
dienamine II generated by mixing compound 2k with the catalysts 3A and 3B. In both
cases we observed a substantial decreasing in the proportion of dienamine II, which
becoming undetectable in the presence of 3B (see S. I.). It could be partially responsible
of the total absence of compounds 6 working under conditions of the Table 4. The
influence of TBAB on the proportion iminium ion I : dienamine II at the equilibrium
shown at Scheme 2, could be a consequence of the stabilization of I provided by the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
8
bromide ion.
Brꢀ
O
N
N
Ph
1
a(H+)
H
Brꢀ
N
O
O
+
ꢀ
Bu N OH
N
4
N
Ph
N
+
Ph
H O
2
+
NBu
OH
H
4
Bu N
1a´
4
ꢀ
Br
+NBu4
1a
Scheme 9. Possible role of TBAB in reactions of 1a with 2.
d) exo-Selectivity and enantioselectivity. The complete control of the
diastereoselectivity (only exoꢀadducts were obtained) in reactions of cyclic azomethine
13a
imines with the iminium species derived from enals 2 had been reported by Maruoka.
Similarly, the very high enantioselectivity of many reactions of nucleophiles and
dipoles with these enals catalyzed by prolinol organocatalysts like 3A and 3B has been
2
associated to the steric hindrance exerted by the group CAr OTMS at the iminium
2
9
species intermediate. If we assume that reactions of species II with dipole 1 were
concerted processes, both enantioselectivity and exoꢀselectivity can be explained as
indicated in scheme 10. The first one would be consistent with the steric hindrance
produced by the CAr
2
OTMS group on the face it occupies in the presumably most
24
ACS Paragon Plus Environment

Page 25 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
stable conformation (based on theoretical calculations, see S. I.) of the dienamine
species II, shown in topꢀScheme 10. The higher ee obtained in Table 4 with respect to
those at Table 3, could be a consequence of the longer distance existing between the
2
CAr OTMS group and the reactive double bond in the dienamines. On the other hand,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
the exoꢀselectivity could be explained by the strong steric interactions of the aryl ring at
3
II with the proton H at dipole 1a, by assuming it adopts the conformation indicated in
Scheme 10, which is the most stable one according to theoretical calculations (see S.I.)
25
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Dipole (1)
Dienamine IIA (2k)
Diastereo and Enantiocontrol in Dienamine Catalysis
Ar
Ar
TMSO
TMSO
N
Ar
C
C
H2
H2
N
H
H4
N
H1
H3
H3
N
N
Bz
Bz
H1
N
H
H
Exoꢀapproach
Endoꢀapproach
N
N
Bz
H1
Bz
H1
N
N
H2
Ph
H2
Ph
CHO
CHO
Observed
nonꢀObserved
Scheme 10. exo-Selectivity and enantioselectivity.
26
ACS Paragon Plus Environment

Page 27 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Addendum
This section has been added to the manuscript after the publication by Du, Wang
15
et al. concerning the synthesis of alcohols 7 by the in situ reduction of the aldehydes
resulting in the organocatalytic reaction of the dipole 1 with enals 2 in the presence of
different acids as the additives. Under the optimal conditions (catalyst 3B (20 mol%),
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
2 2
,4ꢀDNBA (20 mol%), CH Cl , T = ꢀ20 ºC), they described the exclusive formation of
compounds 7 in very good yields (87% for 7k) after 12 hours. We were surprised with
these results because during our optimization process we had checked the influence of
the different benzoic acids as additives (see above and S.I) but we were not able to get
the complete control of the chemoselectivity in any case. The best 6k:5k ratio we had
obtained after 24 h was 77:23 under catalysis of 3A and oꢀfluorobenzoic acid as additive
(20 mol%) in CH Cl at rt (Table 2 entry 10). Worse chemoselectivity was obtained
2 2
with 3B, which afforded almost equimolecular mixtures of both compounds in the
presence of benzoic acid (entry 4, Table 2) and other acidic additives (see S.I.).
Obviously, the results reported in reference 15 were very different to those presented in
this paper and, therefore, they were reinvestigated. In this sense, we reproduced twice
the reactions of 2k (1.5 equiv) with dipoles 1a and 1a’ under the conditions indicated in
entries 14 and 15 of Table 1 at the reference 15. Results we obtained are shown in
Scheme 11 and detailed in S. I. After 12 h, the conversions we observed ranged between
30
5
0 and 63% (at this time the aldehyde 2k had completely disappeared ) which is not
compatible with the high yields reported in reference 15. Moreover, in all the cases we
obtained mixtures of compounds 5k and 6k, where the latter one (derived from the
dienamine) was never the major component in these reactions catalyzed by 3B. Even
31
starting from dipole 1a’, which was more prone to the evolution via dienamine, and
using the most acidic additive (2,4ꢀdinitrobenzoic acid), the reaction mixture only
27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 60
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
contains 51% of 6k. Some other significant differences concern the specific rotation of
3
2
compounds 7k, 7l, 7m, 7p and 7s measured under similar conditions, and the
33
configuration assigned to compounds 4i and 4j. As the results reported in reference 15
were not reproducible in our hands, we proposed the use of the dipole 1a’ and the
catalyst 3A to get the complete control of the stereoselectivity for exclusively obtaining
adducts resulting in the evolution of the dienamine species.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 11. Wang´s reaction conditions applied by us.
Conclusions
Reactions of C,Nꢀcyclic azomethine imines with α,βꢀunsaturated aldehydes,
catalyzed by 3A and 3B evolve with high yields and a complete control of the
stereoselectivity. For βꢀarylmethyl enals, mixtures of isomers resulting in the evolution
via dienamine and iminium ion are observed, which allow us for the first time to study
the chemoselectivity dienamineꢀiminium control. After finding the proper conditions for
getting the complete selectivity in both senses, the role of the main factors controlling it
has been rationalized by chemical procedures and theoretical calculations. Through
28
ACS Paragon Plus Environment

Page 29 of 60
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
these processes, highly functionalized optically active tetrahydroisoquinoline
derivatives can be obtained from readily available starting materials.
Experimental section.
General Methods and Materials. NMR spectra were acquired on a 300 spectrometer,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
13
running at 300 and 75 MHz for H and C, respectively. High temperature NMR
1
spectra were acquired on a 500 spectrometer, running at 500 and 126 MHz for H and
13
C, respectively. Chemical shifts (δ) are reported in ppm relative to residual solvent
1
13
signals (CDCl , 7.26 ppm for H NMR and 77.0 ppm for C NMR or C D Cl , 5.91
3
2
2
4
1
13
13
ppm for H NMR and 74.2 ppm for C NMR). C NMR spectra were acquired on a
broadband decoupled mode. Analytical thin layer chromatography (TLC) was
performed using preꢀcoated aluminiumꢀbacked plates and visualized by ultraviolet
irradiation or by phosphomolybdic acid or potassium permanganate stain. Purification
of reaction products was carried out by flash chromatography (FC) using silica gel.
20
Optical rotations were measured at room temperature and [α]
D
values are given in
deg•cm •g •dm ; concentration c is listed in g•(100 mL) . The enantiomeric excess
ee) of the products was determined by supercritical fluid chromatography (SFC) using
3
ꢀ1
ꢀ1
ꢀ1
(
mixtures of supercritical CO and methanol and Chiralpack IA, IB or ID columns as
2
chiral stationary phases. High Resolution Mass Spectra (HRMS) was recorded using
Electron Impact (E.I.) at 70 eV (TOF analyzer), Fast Atom Bombardment (FAB+)
(
Magnetic Sector Analyzer), or electrospray (ESI+) (QꢀTOF analyzer) techniques. The
IR spectra frequencies are given in cm−1. Catalysts and solvents are commercially
34
13a
available and were used without previous purification. Starting enals, dipoles 1aꢀc,
3
5
and the 0.1M solution of SmI2 were obtained following the standard procedure
described in the literature.
29
ACS Paragon Plus Environment