Ring construction via pseudo-intramolecular hydrazonation using bifunctional δ-keto nitrile

Article abstract of DOI:10.1016/j.tetlet.2011.10.159

An α-nitro-δ-keto nitrile efficiently reacted with hydrazines at room temperature, even in the absence of a catalyst, to afford the corresponding hydrazones; the reactions proceeded through a pseudo- intramolecular process. The hydrazone derived from hydr

Full text of DOI:10.1016/j.tetlet.2011.10.159

Tetrahedron Letters 53 (2012) 82–85
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Ring construction via pseudo-intramolecular hydrazonation
using bifunctional d-keto nitrile
⇑
Shotaro Hirao, Kazuya Kobiro, Jun Sawayama, Kazuhiko Saigo, Nagatoshi Nishiwaki
School of Environmental Science and Engineering, Kochi University of Technology, Tosayamada, Kami, Kochi 782-8502, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
An a-nitro-d-keto nitrile efficiently reacted with hydrazines at room temperature, even in the absence of
Received 1 October 2011
Revised 26 October 2011
Accepted 31 October 2011
Available online 7 November 2011
a catalyst, to afford the corresponding hydrazones; the reactions proceeded through a pseudo-intramo-
lecular process. The hydrazone derived from hydrazine monohydrate underwent water-assisted cycliza-
tion, which yielded the corresponding diazepine. The hydrazones derived from 4-nitrophenylhydrazine
and 2,4-dinitrophenylhydrazine were converted into pyridazines upon being heated in DMSO.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Cyclization
Heterocycles
Hydrazones
Imines
Pseudo-intramolecular hydrazonation
The development of highly efficient synthetic methods for vari-
ous organic compounds is crucial for the efficient utilization of car-
bon resources and for low environmental impact. Reactions not
requiring protecting groups or special reagents facilitate the simpli-
fication of manipulations from the viewpoint of atom economy.1 In
order to improve reaction efficiency, the collision frequency of reac-
tants should be increased. Generally, an intramolecular process
proceeds faster than an intermolecular process because the spatial
proximity increases the collision frequency of reaction sites. With
regard to intermolecular processes, the use of reaction fields such
as capsules,2 cages,3 bowls,4 and micelles5 has been recognized as
being effective because reactive substrates located close to each
other make the reaction highly efficient.
Indeed, vicinally functionalized 1,4-dihydropyridines were synthe-
sized upon treatment of the keto nitrile 1 with amines.9 Further-
more, the reaction of 1 with a diamine entailed tandem cyclization
to afford a 1,7-diazabicyclo[4.3.0]nonane derivative, in which pseu-
do-intramolecular imination was
a
key step.10 These results
prompted us to study the pseudo-intramolecular reaction of 1 by
usinga hydrazineinsteadof adiamineasa dinucleophile (Scheme1).
Initially, the keto nitrile 1 was allowed to react with unsubsti-
tuted hydrazine (2a) at room temperature in CH3CN. After stirring
for 48 h, the formation of a spot different from those of 1 and 2a
was confirmed on TLC. Structural analysis on the basis of spectral
and analytical data clarified the formation of diazepine 7a (Scheme
2).11 The 1H NMR signal of the methyne proton of 1 (Ha) at
6.24 ppm was considerably shifted to 4.88 ppm (Hg). Moreover,
the signals of the carbonyl and cyano carbons shifted from 207
to 138 ppm and from 112 to 172 ppm, respectively, in the 13C
NMR. These spectral changes indicate that 2a reacted with both
functional groups of 1. The formation of 7a presumably proceeded
via pseudo-intramolecular hydrazonation followed by nucleophilic
attack on the cyano group. To confirm this hypothesis, the forma-
tion process of 7a was monitored by NMR spectroscopy. Just after
the addition of 2a to a solution of 1 in CD3CN, small signals of the
hydrazinium salt 3a and the diazepine 7a were observed in the 1H
NMR, in addition to the signals of 1.12 All the signals of 1 and 3a
gradually disappeared, converting to the signals ascribed to 7a.
On the other hand, methylhydrazine (2b) exhibited different
reactivity under the same reaction conditions (Scheme 1). The
reaction was considerably more complicated, giving an intractable
mixture. Similar complications were also observed in the reactions
In this context, we have demonstrated pseudo-intramolecular
processes that proceed under very mild conditions. Indeed,
a-ary-
lated and
a
-nitrated b-keto esters underwent transacylation6,7
with amines very smoothly and rapidly to afford the corresponding
amides without any detectable by-products. In these reactions, the
highly acidic keto ester reacted to form the corresponding ammo-
nium salt with the amine immediately. Then, under equilibrium,
the electrophilic keto ester and the nucleophilic amine were regen-
erated simultaneously and in close proximity. As a result, the reac-
tion proceeded efficiently like an intramolecular process.
The a
-nitro-d-keto nitrile 18 also satisfies two criteria of a sub-
strate that is to be used in a pseudo-intramolecular reaction: it has
an acidic hydrogen and a reactive functional group in the molecule.
⇑
Corresponding author. Tel.: +81 887 57 2517; fax: +81 887 57 2520.
E-mail address: nishiwaki.nagatoshi@kochi-tech.ac.jp (N. Nishiwaki).
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.10.159

S. Hirao et al. / Tetrahedron Letters 53 (2012) 82–85
83
Scheme 1. Pseudo-intramolecular hydrazonation of the keto nitrile 1.
a plausible reaction pathway, density functional theory (DFT) cal-
culations for the simplified model compound 8 bearing no methyl
group at the b-position were performed. Comparison of the activa-
tion energies required to reach transition states showed that the
activation energy for the seven-membered ring transition state is
lower than that for the six-membered one. In the former case,
water plays an important role; a water molecule binds to both
the cyano group and the amino group to effect a water-assisted
ring construction owing to the close proximity of the groups.
Although it was suggested that the hydrazone 5a was intermedi-
ately formed as mentioned above, 5a was too reactive to be isolated.
Hence, we attempted the isolation of the hydrazones 5 through the
use of aromatic hydrazines 2e–i. In the cases of 4-meth-
ylphenylhydrazine (2e) and 4-methoxylphenylhydrazine (2f), the
reactions were also too complex. In the reaction of 1 with phen-
ylhydrazine (2g), the formation of hydrazone 5g was confirmed by
the 1H NMR of a crude product. However, the isolation of 5g was
not achieved because of its rather unstable nature. The use of elec-
tron-deficient 4-nitrophenylhydrazine and 2,4-dinitrophenylhy-
drazine (2h,i) enabled us to isolate the hydrazones 5h,i in
quantitative yields: products spontaneously precipitated within
3 h of the addition of 2h,i to a solution of 1 (Scheme 3). The structure
of 5h was determined on the basis of its spectral and analytical
data.14 Finally, the formation of the hydrazones was confirmed by
the X-ray single crystal analysis of 5i (see Supplementary data).
Hydrazone formation from the keto nitrile 1 and hydrazine 2h
(1:1) was monitored by 1H NMR spectroscopy in methanol-d4
(used as the solvent), which dissolved the hydrazone homoge-
neously. As a result, approximately half of 2h was consumed with-
in 1 h to afford 5h in a 58% yield. The hydrazonation readily
proceeded under mild conditions as a result of the pseudo-intra-
molecular nature of the reaction; in actuality, a similar reaction
of 2h with acetone without involving any functional group to pro-
mote a pseudo-intramolecular process resulted in the formation of
the corresponding hydrazone only in a 28% yield after 1 h (Fig. 3).
Scheme 2. Reaction of the keto nitrile 1 with the hydrazine 2a.
with bulky t-butylhydrazine (2c) and N,N-dimethylhydrazine (2d).
Hence, we monitored the reaction of 1 with 2b by 1H NMR to elu-
cidate the reaction behavior of the substrates.13 When 2b was
added to a solution of 1 in CD3CN, significant changes in the reso-
nance corresponding to both reagents were observed: the singlet
signal at 6.24 ppm, attributed to the methyne proton (Ha) of 1,
completely disappeared, and a couple of singlets with equivalent
intensities at around 1.2 ppm and attributed to the two enantio-
topic methyl groups (b and c) coalesced into a singlet signal at
1.23 ppm. In addition, the N-methyl group of 2b shifted downfield
by about 0.28 ppm. These observations indicate the formation of
the hydrazinium salt 3b. However, the reaction mixture became
complex within several hours, even at room temperature. When
DMSO-d6 was employed as the solvent instead of CD3CN, signals
of ring products were not observed, too.
Considering the mechanism previously proposed for the reac-
tion of 1 with 1,2-diamines,10 diazepines 7a would be formed via
the hydrazinium salt 3a and then the hydrazone 5a. Although
the 1H NMR showed no signal for 5a, the observation of the signals
of 3a supported the proposed mechanism. For alkylhydrazines
2b–d, the corresponding diazepines 6b,c were possibly formed.
However, the isolation of 6b,c failed, probably because of their
instability; the N-substituted diazepines 6b,c could not be con-
verted into a stable form such as 7a by tautomerization (Fig. 1).
In the present reaction, another ring closure affording a six-
membered ring is also possible, as shown in Figure 2. To elucidate
Figure 2. Comparison of two transition states from 8 to the seven-membered and
six-membered ring products 9 and 10, respectively.
Figure 1. Stabilization by tautomerization.

84
S. Hirao et al. / Tetrahedron Letters 53 (2012) 82–85
Scheme 3. Reaction of the keto nitrile 1 with the hydrazine 2h.
When the isolated hydrazone 5h was heated in DMSO at room
temperature for 1 day, a crystalline product was afforded; its spec-
tral data were quite different from those expected for the corre-
sponding diazepine. The product contained a cyano group and an
imino group, which were confirmed by 13C NMR and IR spectra.
Its 1H NMR spectrum was similar to that of the hydrazone 5h ex-
Scheme 4. A plausible mechanism for the formation of the pyridazines 13h,i.
cept for the disappearance of the singlet signal of the acidic a-pro-
ton (Hh). In addition, a DEPT experiment showed that the methyne
carbon (Ch) had changed to a quaternary carbon, which suggested
that the nitrogen atom of the hydrazone attacked the methyne car-
bon to form
a six-membered ring. The empirical formula
C14H15N5O3 also indicated that dehydration proceeded during the
ring construction. On the basis of these spectral and analytical
data, the product was determined to be the pyridazine 13h
(Scheme 4).15 2,4-Dinitrophenylhydrazone 5i was also transformed
to the pyridazine 13i in a similar way (at 70 °C, in 69% yield).
Thus, the reactivity of 1 with 2 varied with the hydrazine substi-
tuent used (Scheme 5). These differences are attributed to the
inherent basicities and nucleophilicities of the hydrazines 2 and
the hydrazones 5. The more basic hydrazines 2a–d (unsubstituted
and alkyl substituted) easily formed the hydrazinium salts 3a–d,
which were observed by 1H NMR. When the hydrazine was liber-
ated under equilibrium, the pseudo-intramolecular hydrazonation
proceeded and afforded 5a–d. However, the nitrogens of 5a–c were
so nucleophilic that ring closure occurred very quickly to afford the
diazepines 6a–c, among which 6a could be converted into stable 7a.
In the case of the aromatic hydrazines 2e–i, although the hydra-
zinium salts 3e–i were the most likely to be formed, they were
hardly detected by 1H NMR because the equilibrium shifted to
the intimate-pair 4 owing to the low basicity of 2e–i. Thus, the
hydrazones 5e–i were immediately afforded by a pseudo-intramo-
lecular process just after the salts 3e–i were formed. An aromatic
ring with an electron-withdrawing substituent can stabilize the
hydrazone and decrease the nucleophilicity of the neighboring
Scheme 5. Summary of results of reactions of keto nitrile 1 with hydrazines.
nitrogen, which facilitates the isolation of the hydrazones 5h,i.
When the hydrazones 5h,i were heated in a polar solvent, tauto-
merization of the nitro group occurred, affording the nitronic acids
11h,i, and the Ch-carbon became sufficiently electrophilic to be at-
tacked by the anilino nitrogen to form the pyridazine frameworks
of 12h,i. The results of DFT calculations (B3LYP/6-31++G⁄) of model
compounds 14 and 15 showed the considerably higher electrophi-
licity of the nitronic acid carbon (Ci) compared to a cyano group,
which is in good agreement with the experimental results (Fig. 4).
In summary, wesuccessfullysynthesized diazepine7a and pyrid-
azines 13h,i by treating keto nitrile 1 with hydrazines under mild
Figure 3. Comparison of the hydrazonation rate of the keto nitrile 1 (j) and
acetone (N) with the hydrazine 2h.
Figure 4. Mulliken populations determined by DFT calculations (B3LYP/6-31++G⁄).

S. Hirao et al. / Tetrahedron Letters 53 (2012) 82–85
85
6. (a) Nishiwaki, N.; Nishida, D.; Ohnishi, T.; Hidaka, F.; Shimizu, S.; Tamura, M.;
Hori, K.; Tohda, Y.; Ariga, M. J. Org. Chem. 2003, 68, 8650; (b) Nakaike, Y.; Taba,
N.; Itoh, S.; Tobe, Y.; Nishiwaki, N.; Ariga, M. Bull. Chem. Soc. Jpn. 2007, 80, 2413.
conditions, which involved pseudo-intramolecular hydrazonation
as a key step. In addition, the intermediate hydrazones 5h,i were
successfully isolated when 4-nitrophenyl- and 2,4-dini-
trophenylhydrazines 2h,i were used. Moreover, the DFT calculations
supported the experimental results reasonably. Consequently, the
present pseudo-intramolecular process will be applicable as an effi-
cient protocol in synthetic chemistry, provided suitably designed
substrates satisfying the two criteria are employed.
7. We consider Ballini’s reactions proceed also in
a pseudo-intramolecular
process; Ballini, R.; Bosica, G.; Fiorini, D. Tetrahedron 2003, 59, 1143.
8. The keto nitrile 1 was readily prepared from commercially available ethyl
nitroacetate via three steps; Nishiwaki, N.; Nogami, T.; Ariga, M. Heterocycles
2008, 75, 675.
9. Nishiwaki, N.; Kakutani, K.; Tamura, M.; Ariga, M. Chem. Lett. 2009, 38, 680.
10. Nishiwaki, N.; Hirao, S.; Sawayama, J.; Saigo, K.; Kobiro, K. Chem. Commun.
2011, 47, 4938.
11. The diazepine 7a. Yellow solid. Mp 113–115 °C. IR (Nujol) 1634, 1549, 1462,
1348 cmÀ1 1H NMR (CDCl3) d 0.91 (s, 3H), 1.07 (s, 3H), 2.17 (s, 3H), 2.17 (d,
;
Acknowledgment
J = 19.0 Hz, 1H), 2.52 (d, J = 19.0 Hz, 1H), 4.83 (s, 1H), 6.35 (br s, 2H); 13C NMR
(CDCl3) d 25.8 (CH3), 26.6 (CH3), 27.7 (CH3), 33.1(C), 41.7 (CH2), 90.7 (CH),
137.7 (C), 172.2 (C). Analytical data were not given satisfactorily because of
instability of 7a.
This work was supported by Grants-in-Aid for Scientific Research
(No. 22550043) from Japan Society for the Promotion of Science.
12. For details about the reaction toward the diazepine 7a monitored by 1H NMR,
see Supplementary data.
13. For details about the formation of the hydrazinium salt 3b monitored by 1H
NMR, see Supplementary data.
Supplementary data
14. The hydrazone 5h. Pale yellow needles. Mp 138–139 °C. IR (KBr) 1603, 1560,
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2011.10.159.
1321, 1306 cmÀ1 1H NMR (CDCl3) d 1.33 (s, 3H), 1.41 (s, 3H), 2.01 (s, 3H), 2.53
;
(d, J = 16.4 Hz, 1H), 2.66 (d, J = 16.4 Hz, 1H), 6.04 (s, 1H), 7.01 (d, J = 9.2 Hz, 2H),
7.56 (br s, 1H), 8.18 (d, J = 9.2 Hz, 2H); 13C NMR (CDCl3) d 17.2 (CH3), 24.3
(CH3), 24.8 (CH3), 39.9 (C), 47.0 (CH2), 83.7 (CH), 111.5 (C), 111.9 (CH), 126.2
(CH), 140.8 (C), 145.4 (C), 149.6 (C); MS (FAB) 320 (M++1, 100). Anal. calcd for
C14H17N5O4: C, 52.66; H, 5.37; N, 21.93. Found: C, 52.66; H, 5.52; N, 21.95.
15. The pyridazine 13h. Orange plates. Mp 146–148 °C. IR (Nujol) 2210, 1612,
References and notes
1. (a) Trost, B. M.; Shi, Y. J. Am. Chem. Soc. 1992, 114, 791; (b) Trost, B. Acc. Chem.
Res. 2002, 35, 695; (c) Young, I. S.; Baran, P. S. Nat. Chem. 2009, 1, 193.
2. (a) Koblenz, T. S.; Wassenaar, J.; Reek, J. N. H. Chem. Soc. Rev. 2008, 37, 247; (b)
Vriezema, D. M.; Aragonès, M. C.; Elemans, J. A. A. W.; Cornelissen, J. J. L. M.;
Rowan, A. E.; Nolte, R. J. M. Chem. Rev. 2005, 105, 1445.
3. Yoshizawa, M.; Fujita, M. Pure Appl. Chem. 2005, 77, 1107.
4. Yoshizawa, M.; Klosterman, J. K.; Fujita, M. Angew. Chem., Int. Ed. 2009, 48,
3418.
1537, 1348 cmÀ1 1H NMR (CDCl3) d 1.29 (s, 3H), 1.46 (s, 3H), 1.81 (s, 3H), 2.34
;
(d, J = 14.0 Hz, 1H), 2.57 (d, J = 14.0 Hz, 1H), 7.91 (d, J = 9.0 Hz, 2H), 8.37 (d,
J = 9.0 Hz, 2H); 13C NMR (DMSO-d6) d 20.3 (CH3), 26.5 (CH3), 27.1 (CH3), 39.5
(C), 45.0 (CH2), 103.4 (C), 111.3 (C), 123.5 (CH), 125.0 (CH), 126.3 (C), 149.1 (C),
153.5 (C); MS (FAB) 302 (M++1, 100). Anal. calcd for C14H15N5O3: C, 55.81; H,
5.02; N, 23.24. Found: C, 55.74; H, 4.77; N, 23.07.
5. (a) Onel, L.; Buurma, N. J. Annu. Rep. Prog. Chem. Sect. B 2009, 105, 363; (b)
Dwars, T.; Paetzold, E.; Oehme, G. Angew. Chem., Int. Ed. 2005, 44, 7174.