Regioselectivity in nucleophilic ring-opening of aziridinones

Article abstract of DOI:10.1039/a800564h

The proportions of products derived from competing modes of ring-opening of 1,3-di-tert-butylaziridinone and similar aziridinones by a variety of nitrogen, oxygen, sulfur and halogen nucleophiles do not agree with simple guidelines postulated in the liter

Full text of DOI:10.1039/a800564h

View Article Online / Journal Homepage / Table of Contents for this issue
Regioselectivity in nucleophilic ring-opening of aziridinones
Erach R. Talaty*† and Mashitah M. Yusoff
Department of Chemistry, Wichita State University, Wichita, Kansas 67260-0051, USA
The proportions of products derived from competing modes
of ring-opening of 1,3-di-tert-butylaziridinone and similar
aziridinones by a variety of nitrogen, oxygen, sulfur and
halogen nucleophiles do not agree with simple guidelines
postulated in the literature for these types of aziridinones.
alkyl–nitrogen bond. However, there are exceptions to even this
rough rule (entries 5, 18 and 19). The aprotic oxygen and sulfur
nucleophiles also exhibit considerable variation in their se-
lectivity, the alcohols being most consistent in favoring
cleavage of the alkyl–nitrogen bond. The protic ionic nucleo-
philes also do not all afford products derived from acyl–
nitrogen cleavage, as suggested in the literature.1 In fact, the
sharp difference between bromides and iodides on one hand and
alkoxides on the other hand suggests the intervention of yet
another factor, namely, hardness or softness of the nucleophile.
Iodide, a soft nucleophile, attacks the soft alkyl carbon of 1,
whereas alkoxide, a hard nucleophile, prefers to attack the
harder acyl carbon.
,2
An important aspect of the chemistry of aziridinones is their
mode of ring-opening by nucleophiles and the factors that
O
H
2
3
1
N
R2
A priori, one can envisage the following two schemes that
may determine the selectivity in ring-opening of the azir-
R1
1
1
1
1
= R = But
2
N N
idinones. Scheme 1 resembles the S 1/S 2 type dichotomy
1
2
3
4
R
R
R
R
2
encountered in nucleophilic aliphatic substitution [path (a)
being unimolecular and path (b) being bimolecular], whereas
Scheme 2 (competing bimolecular pathways) was apparently
the basis of the guidelines given in the literature.1 If Scheme 1
were to be the exclusive one prevailing, then the selectivity
could be altered proportionately by changing the concentration
of a nucleophile that tends to give both types of products. In our
case, no such alteration or, at best, minor alterations could be
effected in some cases that we examined. Scheme 1 would also
indicate a strong dependence on nucleophilicity, path (b) being
favored by powerful nucleophiles. However, as observed above,
an excellent nucleophile such as iodide affords just the opposite
type of product. It thus appears that, at least in the case of
aziridinones 1–4, no one scheme can satisfactorily explain the
competing modes of ring-opening by nucleophiles, a conclusion
that does not follow from a study of other types of azir-
= R = 1-adamantyl
= 1-adamantyl, R = But
2
t
2
= Bu , R = 1-adamantyl
,2
govern the outcome. It has been reported1 that ionic, aprotic
nucleophiles (Z ) cause rupture exclusively of the acyl–
,2
2
nitrogen bond (1,2-bond), whereas non-ionic protic nucleo-
philes (HZ) afford products derived solely or mainly from
scission of the alkyl–nitrogen bond (1,3-bond). Our previous
publications3 clearly contradict this sample rule, albeit with a
limited set of nucleophiles. We have embarked on a broader
study of nucleophilic ring-opening of aziridinones.
,4
An examination of the nitrogen nucleophiles (aprotic) in
Table 1 reveals immediately that all of them do not cleave solely
9
idinones.
1,2
the alkyl–nitrogen bond of 1 as alleged in the literature. In
fact, the only ones that exhibit this pattern are all of the aromatic
amines (entries 10–17), none of which was used previously in
conjunction with aziridinones 1–4. On the other hand, the
simple aliphatic primary amines rupture exclusively the acyl–
nitrogen bond of 1. The secondary amines exhibit varying
degrees of ring-opening (entries 3 and 5 giving exactly opposite
results) that also seem to depend on the nature of the aziridinone
We have extended the study of nitrogen nucleophiles to
include those that might be synthetically useful as a route to
larger heterocycles, as illustrated by the three new examples
shown in Scheme 3.10
Note that (i) these are the only heterocyclic products isolated
in each case; (ii) PhNHCN and PhNH (mentioned in Table 1)
2
give products derived from opposite modes of cleavage; (iii)
specificity in the substitution pattern of these five-membered
heterocyclic compounds can be controlled by the method of
synthesis; and (iv) the structures of these three heterocyclic
compounds could not have been predicted by previous guide-
lines.
(
e.g. compare entries 2 and 32), indicating a subtle dependence
on steric factors. Roughly speaking, stronger, unhindered
nitrogen nucleophiles tend to cleave the acyl–nitrogen bond,
whereas sterically hindered, weaker ones favor scission of the
O
O
H
R1
H
R1
N
R2
+
HNu:
N
R2
+
HNu:
(
a)
(b)
(a)
(b)
slow
O
O
H
N
O
O
H
N
O
Nu
fast
–
NHR2
+
R2
H
N
R2
Nu
NHR2
R2
Nu
HNu:
R1
R1
R1
H
R1
H
H
R1
Nu
H
Scheme 1
Scheme 2
Chem. Commun., 1998
985

View Article Online
Table 1 Relative modes of ring opening of aziridinones by various nucleophiles (0.06 mm each) in boiling toluene
Major product
H 3
d (CDCl )
1
,2 : 1,3
a
C-But
N-But
Entry
Nucleophile
Aziridinone Yield (%)
cleavage (%)
Mp/°C
a-CH
1
2
3
4
5
6
7
8
9
PrNH
BnNH
Et NH
Bu NH
BnNHMe
Pyrrolidine
Morpholine
Piperidine
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
3
3
3
2
2
88
92
32
54
54
65
48
52
42
95
94
90
96
74
85
89
12
94
65
84
80
62
71
72
80
68
93
86
86
87
87
89
42
68
40
70
100 : 0
100 : 0
14 : 86
< 22 : > 78
86 : 14
6 : 94
< 2 : > 98
16 : 84
9 : 91
76–77
81.5–82.5
oil
58–59
oil
106–107
129–131
94–95
0.97
0.97
1.02
1.03
0.95
1.02
1.03
1.03
1.02
1.09
1.08
1.08
1.09
0.98
1.12
1.11
1.10
1.03
0.95
0.96
0.89
1.04
0.96
1.02
1.02
0.98
1.03
1.17
1.17
1.14
0.91
—
1.04
0.99
1.35
1.35
0.99
1.33
1.37
1.36
1.35
1.29
1.29
1.29
1.29
1.28
1.31
1.31
1.24
1.41
1.26
1.37
1.00
1.12
1.36
1.13
1.13
1.06
1.41
1.35
1.35
1.36
1.01
0.99
1.37
1.26
2.89
2.92
2.51
2.51
3.27
2.36
2.27
2.34
2.32
3.22
3.17
3.18
3.15
2.74
3.40
3.28
4.23
3.54
3.48
3.08
2.80
3.17
3.13
3.20
3.20
3.12
3.54
4.05
4.05
4.00
2.90
2.80
2.05
3.37
2
2
2
Piperazine
227
b
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
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
PhNH
p-MeC
p-BrC
p-MeOC
2
0 : 100
0 : 100
0 : 100
0 : 100
0 : 100
0 : 100
0 : 100
0 : 100
100 : 0
100 : 0
< 7 : > 93
100 : 0
71 : 29
5 : 95
55 : 45
86 : 14
> 93 : < 7
55 : 45
0 : 100
0 : 100
0 : 100
100 : 0
83 : 17
8 : 92
172–173
157–157.5
191–192
146–147
105–106
212–214
229–230
oil
258 (decomp.)
85–86
57–58
oil
oil
6
H
4
4
NH
NH
NH
2
6
H
2
6
H
4
2
1-Aminonaphthalene
4-Aminoazobenzene
4-Aminobiphenyl
2
Ph NH
c
2
H NCN
c
BnNHCN
MeOH
MeONa
d
PhOH
t
Bu CH
PhSH
PhSLi
2
OH
oil
52–53
52–53
52–53
HSCH
NCSNH
NaI/acetone
MgI (anhydrous)
MgBr
Mg(OEt)
BnNH
Morpholine
2
CO
2
Et
2
e
H
2
258 (decomp.)
168–169
168–169
157–158
oil
120–121
171–172
204–205
215–216
121–122
f
2
.
2
OEt
2
2
2
—
—
1-Aminonaphthalene
Morpholine
1-Aminonaphthalene
12 : 88
11 : 89
7 : 93
a
Combined yield of the two cleavage products after purification. Structure confirmed by using 15N-labelled aniline, which exhibited a doublet (J = 8.05
b
1
3
15
t
c
Hz) at d 69.22 in its C (proton-decoupled) NMR spectrum (Ph NHCHBu CO). The products are imidazolidinones, obtained by initial acyl–nitrogen
cleavage followed by cyclization involving intramolecular nucleophilic attack on the nitrile (ref. 5). See also ref. 6, where reaction was conducted in
methanolic solution. The major product is identical with that from entry 18 (net loss of H S) (ref. 7). Ref. 8.
2
d
e
f
S
O
3
E. R. Talaty, A. E. Dupuy, Jr. and C. M. Utermohlen, J. Chem. Soc.,
Chem. Commun., 1971, 16.
R1
H
H2N
NHMe
4 M. M. Yusoff and E. R. Talaty, Tetrahedron Lett., 1996, 48, 8695.
5
N
E. R. Talaty, M. M. Yusoff, S. A. Ismail, J. A. Gomez, C. E. Keller and
J. M. Younger, Synlett, 1997, 6, 683.
N
R2
NHMe
6
7
J. C. Sheehan and J. H. Beeson, J. Am. Chem. Soc., 1967, 89, 362.
E. R. Talaty, J. A. Gomez, J. A. Dillon, E. Palomino, M. O. Agho, B. C.
Batt, K. J. Gleason, S. Park, J. R. Hernandez, M. M. Yusoff, G. A. Rupp,
F. C. Malone, M. F. Brummett, C. L. Finch, S. A. Ismail, N. Williams
and M. Aghakhani, Synth. Commun., 1987, 1063.
O
R1
H
Ph
PhNHCN
N
1
N
8 H. Quast and H. Leybach, Chem. Ber., 1991, 124, 2105, also studied this
R2
NH
reaction.
9
R. V. Hoffman, N. K. Nayyar and W. Chen, J. Org. Chem., 1995, 60,
121, have relied exclusively on Scheme 1 to explain products arising
4
1
O
2
from an aziridinone involved as an intermediate (R = Ph or CO Et,
R1
R = Me) and have correlated products with nucleophilic constants.
2
(
1) H2NCN
N
However, in no case was the aziridinone actually isolated. H. E.
Baumgarten, J. F. Fuerholzer, R. D. Clark and R. D. Thompson, J. Am.
Chem. Soc., 1963, 85, 3303; S. Sarel, B. A. Weissman and Y. Stein,
H
t
(2) BuBr, Bu OK
N
R2
NHBu
t
Tetrahedron Lett., 1971, 373, studied the reaction of KOBu , MeOH or
Scheme 3
t
Bu OH with 1-tert-butyl-3-phenylaziridinone and 1-tert-butyl-3-bis-
norcholanylaziridinone, respectively, the results being in accord with
the simple guidelines (refs. 1 and 2).
Notes and References
†
E-mail: talaty@twsuvm.uc.twsu.edu
10 Second and third reactions: work done by S.A. Ismail in our
laboratory.
1
2
I. Lengyel and J. C. Sheehan, Angew. Chem., Int. Ed. Engl., 1968, 25.
G. L’abbe, Angew. Chem., Int. Ed. Engl., 1980, 276.
Received in Corvallis, OR, USA, 20th January 1998; 8/00564H
986
Chem. Commun., 1998