Preparation and stereochemistry of dioxatetraazaperhydroanthracenes and -perylenes from the reaction of 2-hydrazinoethanols with aldehydes and glutaraldehyde

Article abstract of DOI:10.1021/jo951878p

Hydrazinoethanols 1 were reacted with aldehydes 2 and 6 and glutaraldehyde (14) in aqueous solution to give dioxatetraazaperhydroanthracenes 3, 7, 12, and 13 and -perylenes 15 in yields of 19-88 and 42-72%, respectively. Compounds 3, 7, 12, and 15 were deduced by ¹³C-NMR spectra to have two C₂ symmetry axes, while compound 12 was shown to have a symmetry axis by X-ray crystallography. The most favorable stereoisomers were consistent with predictions obtained by the semiempirical molecular orbital method AMI. The structure of compound 15 was confirmed by X-ray crystallography.

Full text of DOI:10.1021/jo951878p

J . Org. Chem. 1996, 61, 4125-4129
4125
P r ep a r a tion a n d Ster eoch em istr y of
Dioxa tetr a a za p er h yd r oa n th r a cen es a n d -p er ylen es fr om th e
Rea ction of 2-Hyd r a zin oeth a n ols w ith Ald eh yd es a n d
Glu ta r a ld eh yd e
Tadashi Okawara,* Shuji Ehara, Hideaki Kagotani, Yoshinari Okamoto, Masashi Eto,
Kazunobu Harano, Tetsuo Yamasaki, and Mitsuru Furukawa
Faculty of Pharmaceutical Sciences, Kumamoto University, Oe-hon-machi, Kumamoto 862, J apan
Received October 19, 1995X
Hydrazinoethanols 1 were reacted with aldehydes 2 and 6 and glutaraldehyde (14) in aqueous
solution to give dioxatetraazaperhydroanthracenes 3, 7, 12, and 13 and -perylenes 15 in yields of
9-88 and 42-72%, respectively. Compounds 3, 7, 12, and 15 were deduced by 1 C-NMR spectra
3
1
2
to have two C symmetry axes, while compound 12 was shown to have a symmetry axis by X-ray
crystallography. The most favorable stereoisomers were consistent with predictions obtained by
the semiempirical molecular orbital method AM1. The structure of compound 15 was confirmed
by X-ray crystallography.
In tr od u ction
a few stereoisomers are chosen from among many pos-
sible isomers and accordingly attempted to prepare tri-
and pentaheterocycles with four nitrogens in a single
ring. In this paper, we report the reaction of hydrazi-
noethanols 1 with aldehydes 2 and 6 and glutaraldehyde
Recent developments concerning polyazapolycycles,1
including X-ray crystallographic data, have demonstrated
that the formation of these compounds is controlled by
steric and lone pair-lone pair interactions. Nielsen et
al., Willer et al., and Ferguson et al. emphasized the
importance of the relationship between 1,3-steric and
peri-electronic interactions of polyazapolycycles. In ad-
dition, Crabb and Katritzky reviewed conformational
equilibria and lone-pair interactions in nitrogen-con-
taining saturated six-membered rings.10
-9
(14) to provide dioxatetraazaperhydro-anthracenes 3, 7,
4
6
9
1
2, and 13 and -perylenes 15 and include a prediction of
the most likely stereoisomer by AM1. One of these
predictions was confirmed by X-ray crystallography.
Resu lts a n d Discu ssion
We attempted to react hydrazinoethanols 1 with alde-
hyde 2 and 6 and glutaraldehyde (14). The reaction of
1a -d with 2 equiv of formaldehyde (2) was carried out
in EtOH at room temperature for 24 h to give dioxa-
tetraazaperhydroanthracenes 3a -d in 29-88% yield
(Scheme 1). The mass spectrum of 3a showed a molec-
ular ion peak at m/ z 200, which represented the loss of
four molecules of water from 2 equiv of 1a and 4 equiv
of 2. No absorptions for hydroxyl, amino, or CdN double
The reaction of hydrazine with formaldehyde and
hydrogen peroxide was first reported in 1921.
1
1
The
structure originally reported was modified12 and was
finally determined by X-ray structure analysis to be a
tricyclic compound with one six- and two seven-mem-
bered rings.13 Katritzky elucidated why this structure
was produced by comparing heats of formation for
isomers using a molecular orbital method (AM1).1 These
previous studies seemed relevant to our own work with
polyheterocycles15 which are formed from the reaction of
polyfunctionalized compounds with carbonyls. We be-
came interested in the factors which determined why only
4
1
3
bonds were observed in the IR spectrum. The C-NMR
spectrum showed one-half of the total number of expected
carbon signals (8 carbons) at δ 41.82, 63.11, 66.35, and
82.40, which is consistent with a C symmetry axis.
2
Another plausible structure (4) which would have
1
3
X
required five carbon signals in the C-NMR spectrum
Abstract published in Advance ACS Abstracts, May 15, 1996.
1) Chaykovsky, M.; Koppes, W. M.; Russell, T. P.; Gilardi, R.;
George, C.; Flippen-Anderson, J . L. J . Org. Chem. 1992, 57, 4295.
2) Nielsen, A. T.; Nissan, R. A.; Chafin, A. P.; Gilardi, R. D.; George,
C. F. J . Org. Chem. 1992, 57, 6756.
3) Craig, D. C.; Kassiou, M.; Read, R. W. J . Org. Chem. 1992, 57,
901.
4) Nielsen, A. T.; Nissan, R. A.; Vanderah, D. J .; Coon, C. L. J .
Org. Chem. 1990, 55, 1459.
5) Black, D. S.; Craig, D. C.; Giitsidis, O.; Read, R. W.; Salek, A.;
Sefton, M. A. J . Org. Chem. 1989, 54, 4771.
6) Willer, R. L.; Moore, D. W.; Lowe-Ma, C. K.; Vanderah, D. L. J .
Org. Chem. 1985, 50, 2368.
7) Willer, R. L.; Moore, D. W.; Vanderah, D. L. J . Org. Chem. 1985,
0, 2365.
1
(
was therefore eliminated. In the H-NMR spectrum, non-
equivalent methylene hydrogens at positions 1, 4, and 9
are indicated at δ 4.28 and 4.62, 2.25 and 4.19, and 3.22
and 4.00, respectively, with assignment based on H-
(
1
(
3
13
C NMR COSY.
(
Compound 3a has three possible stereoisomers, 3a a ,
a b, and 3a c. To determine which of these is the most
(
3
f
stable, their heats of formation (∆H ) were calculated by
(
a molecular orbital method (AM1).16 The results are
summarized in Figure 1.
(
5
2
Compounds 3a c and 3a a had the lowest (12.13 kcal/
mol) and highest (23.66 kcal/mol) heats of formation,
respectively. This suggests that the trans-trans fused
form 3a a results in repulsion between the lone pairs on
nitrogens and oxygens, whereas comparison of the cis-
cis fused forms 3a b and 3a c indicates that lone pair-
(8) Weisman, G. R.; Ho, S. C. H.; J ohnson, V. Tetrahedron Lett. 1980,
1, 335.
(
9) Ferguson, I. J .; Katritzky, A. R.; Patel, R. J . Chem. Soc. Perkin
Trans. 2 1976, 1564.
10) Crabb, T. A.; Katritzky, A. R. Adv. Heterocyc. Chem. 1984, 36,
. Katritzky, A. R. Topics in Heterocyclic Chemistry; Castle, R., Ed.;
Interscience: New York, 1964; p 35.
(
1
(
(
(
11) von Girsewald, C.; Siegens, H. Chem. Ber. 1921, 54, 492.
12) Schmitz, E. Liebigs Ann. Chem. 1960, 635, 73.
13) Whittleton, S. N.; Seiler, P.; Dunitz, J . D. Helv. Chim. Acta.
(15) Okawara, T.; Ehara, S.; Matsumoto, S.; Okamoto, Y.; Fu-
rukawa, M. J . Chem. Soc., Perkin Trans. 1 1990, 2160; Okawara, T.;
Uchiyama, K.; Okamoto, Y.; Yamasaki, T.; Furukawa, M. J . Chem.
Soc., Chem. Commun. 1990, 1385; J . Chem. Res. 1992, 264.
1
981, 64, 2614.
14) Katritzky, A. R.; Luce, H. H.; Anders, E.; Sullivan, J . M. Chem.
Ber. 1989, 122, 1989.
(
S0022-3263(95)01878-0 CCC: $12.00 © 1996 American Chemical Society

4
126 J . Org. Chem., Vol. 61, No. 12, 1996
Okawara et al.
F igu r e 1.
Sch em e 1
Sch em e 2
Sch em e 3
lone pair repulsion in 3a c is less than that in 3a b.
Therefore, the stereochemistry of 3a is most likely given
by 3a c. To elucidate the reaction pathway further, the
reaction was performed in EtOH at -78 °C to give 1,3,4-
perhydrooxadiazines (5a -d ) in 22-29% yields. Struc-
tures 5a -d were confirmed by showing that the molec-
ular ion peak resulted from the loss of a molecule of water
from 1 and 2. In addition, while amino absorption was
Sch em e 4
-
1
seen at 3200 cm , no CdN double bond absorption was
1
3
seen in the IR spectra, and the C-NMR spectrum
showed three methylene carbons at δ 57.82, 60.60, and
7
1.20 in the case of 5a . Compound 5 was allowed to react
with 2 to give 3 in quantitative yield. Therefore, the
deduced reaction pathway to 3 is as follows: Compound
5
, which is derived from 1 and 2, is presumed to react
with a second molecule of 2, thus avoiding lone pair-
lone pair repulsion to produce cis-cis fusion.
Next, the reaction of 5a and 5b with aldehyde 6 was
attempted, to give 1,5-disubstituted dioxatetraaza-per-
hydroanthracenes 7a -d in 24-41% yields, but none of
the expected 9,10-disubstituted compound 8 (Scheme 2).
In the 13C-NMR spectrum of 7a , the methine carbon (N-
C-O) at positions 1 and 5 appeared at δ 87.44 below the
expected value in 8. To ascertain the structure of 7a ,
the H- C-NMR long-range correlation was observed,
indicating three-bond coupling between the carbon at
position 4 and the hydrogen at position 10.
is converted into another intermediate 10, which then
intermolecularly condenses to give 7 (Scheme 3).
Next, the reaction of hydrazinocyclohexanol (11) with
2 equiv of 2 was carried out in EtOH to give dioxatet-
raazaperhydrodibenzoanthracenes 12 and 13 in yields of
33 and 16%, respectively (Scheme 4).
1
13
The 13C-NMR spectrum of compound 12 showed eight
Structure assignment was also supported by another
route to 7 which used the reaction of 1 with 6 followed
by treatment with 2. Therefore, the reaction pathway
to 7 is as follows: Intermediate 9 derived from 5 and 6
2
carbons with a C symmetry axis, while compound 13
showed 16 carbon signals with no symmetry axis. Com-
pound 12 has three possible stereoisomers, 12a , 12b, and
12c. Their heats of formation by AM1 are summarized
in Figure 2. The trans-cis-cis-trans forms 12b and 12c
are more stable than the all-trans fused 12a . Further-
more, 12c is presumed to be the most favored stereoiso-
mer due to the lack of lone pair-lone pair repulsion on
nitrogen and oxygen.
(
16) Stewart, J . J . P. QCPE Bull. 1983, 3, 43. The MOPAC program
(
QCPE455) was used for AM1 calculations. Initial geometrical data
were constructed using ANCHOR II (Fujitsu Ltd.) and fully optimized
using the Pulay self-consistent convergence procedure and the PRE-
CISE option. Symmetry constraints were not applied.

Dioxatetraazaperhydroanthracenes and -perylenes
J . Org. Chem., Vol. 61, No. 12, 1996 4127
F igu r e 2.
appeared as a doublet of doublets at δ 2.50 and 4.03. This
shift difference could be attributed to the fact that the
hydrogens which are proximal to the lone pairs on oxygen
and nitrogen are shifted to lower fields than the hydro-
gens in the opposite direction.17 All of the hydrogens and
1
1
1
carbons of 15a could be assigned from H- H- and H-
1
3
C-NMR COSY.
Although several stereoisomers can be formed from the
reaction of 1a with 14, we estimated that the most
favorable stereoisomers were those with an axis of
symmetry, giving either the all-trans fused form 15a a
or a form which contained two cis-two trans fused rings,
1
1
5a b and 15a c. To differentiate among 15a a , 15a b, and
5a c, an NOE difference spectrum of 15a was measured.
Irradiation of the methine proton H-6a (H-12a) at δ
.58 enhanced the signals of the equatorial protons H-6a
H-12a) and H-1 (H-7) at δ 1.75 and 3.75, and of the
F igu r e 3. Stereo view of 13.
4
(
Sch em e 5
axial protons H-5 (H-11) and H-3a (H-9a) at δ 1.65
and 4.89 ppm, respectively. However, these results were
not unambiguous and we could not clearly discriminate
among the three stereoisomers 15a a , 15a b, and 15a c.
Accordingly, the heats of formation were calculated by
AM1 as shown in Figure 4. Stereoisomer 15a b is more
stable than 15a a and 15a c by 13.55 and 5.68 kcal/mol,
respectively. These results show that the lone pairs of
stereoisomers 15a a and 15a c are repulsed more than
those of 15a b. The lone pairs of isomer 15a b are
arranged so as to minimize this repulsion, as shown in
Figure 4. Consequently, isomer 15a b is assumed to be
formed preferentially over 15a a and 15a c. Since other
structures could also result from this reaction, the
structure was unambiguously established as 15a by
X-ray crystallography, as shown in Figure 5.
The stereochemistry of compound 13 was not eluci-
1
13
dated by H- C-NMR COSY. However, the structure
of 13 was determined by X-ray crystallography to be the
trans-cis-trans-trans fused form 13a (Figure 3).
As an extension of this method, we attempted to react
a -c with glutaraldehyde (14). A mixture of 1 and 14
in THF was stirred for 15 h at room temperature.
Removal of solvent gave an oily residue, which was
treated with ether to give the desired products 15
A comparison of the bond lengths, bond angles, and
dihedral angles of the final coordinates obtained from
AM1 and X-ray analysis is shown in Table 1.
In conclusion, in the polyheterocycles derived from
1
2
-hydrazinoethanols, the cis-fused isomer is assumed to
be formed preferentially to avoid 1,2 and 1,3 lone pair
interactions. Heats of formation calculated from AM1
and X-ray structure analyses strongly support these
results.
(
Scheme 5). The structures of these products were
assigned as 3,9-dioxa-3b,6b,9b,12b-tetraazaperhydro-
perylenes 15 on the basis of the following observations.
Product 15a , derived from 1a and 14, showed a molecular
ion peak at m/ z 280, which was consistent with the sum
of two molecules each of 1a and 14 after eliminating four
molecules of water. The IR spectrum revealed the
absence of amino, hydroxy, and carbonyl groups. The
DEPT 13C-NMR spectrum indicated five methylene car-
bons at δ 17.76, 28.17, 30.97, 36.16, and 66.48 and two
methine carbons at δ 66.21 and 87.86, which supports a
Exp er im en ta l Section
Gen er a l Meth od s. Melting points are uncorrected. 1H
1
3
NMR spectra were recorded at 60 and 400 MHz, and C NMR
spectra were measured at 100 MHz. All spectra were recorded
3 2 6
in CDCl , D O, and DMSO-d as solvent, and chemical shifts
are reported in δ relative to TMS. Hydrazinoethanols 1b-d
18
were prepared by Gever’s method.
Gen er a l P r oced u r e for 3,7-Disu bstitu ted 2,6-Dioxa -4a ,
8
a ,9a ,10a -tetr a a za p er h yd r oa n th r a cen e (3). A solution of
1
1 (10 mmol) and 2 (2.3 mL, 30 mmol, 37% w/v in water) in
symmetrical structure. The H-NMR spectrum showed
C-4, C-5, and C-6 methylene groups at δ 1.22-1.79 and
(
17) Armarego, W. L. F. Stereochemistry of Heterocyclic Compounds,
C-3a and C-6a methine protons as a doublet of doublets
Part 1; Wiley-Interscience: New York, 1977. Weisman, G. R.; J ohnson,
V.; Fiala, R. E. Tetrahedron Lett. 1980, 21, 3635.
(18) Gever, G. J . Am. Chem. Soc. 1954, 76, 1283.
1
and a doublet at δ 4.58 and 4.90, respectively. In H-
1
3
C NMR COSY, the methylene protons at C-2 each

4
128 J . Org. Chem., Vol. 61, No. 12, 1996
Okawara et al.
F igu r e 4.
F igu r e 5. Stereo view of compound 15a b.
Ta ble 1. Bon d Len gth s, Bon d An gles, a n d Dih ed r a l
An gles of 15a As Obta in ed fr om AM1 a n d X-r a y An a lysis
2H), 3.35 (d, J ) 11.7 Hz, 2H), 4.08 (dd, J ) 12.1, 11.0 Hz,
2H), 4.48 (d, J ) 10.6 Hz, 2H), 4.74 (dd, J ) 11.0, 2.9 Hz, 2H),
+
5
.34 (d, J ) 11.7 Hz, 2H); MS-EI m/ z 352 (M ). Anal. Calcd
AM1
X-ray
for C20
24 4 2
H N O : C, 68.16; H, 6.86; N, 15.90. Found: C, 68.40;
bond length (Å)
N3b-N12b
C3a-N3b
H, 7.04; N, 15.76.
1.400
1.450
Gen er al P r ocedu r e for 6-Su bstitu ted 1,3,4-P er h ydr oox-
a d ia zin e (5). To a solution of 1 (50 mmol) in EtOH (100 mL)
was added dropwise 2 (3.8 mL, 50 mmol, 37% w/v in water)
at -78 °C. The reaction mixture was then stirred at -78 °C
for 4 h and allowed to stand until it reached room temperature.
The EtOH was evaporated under reduced pressure, and the
residue was recrystallized from solvent.
1.486
1.443
1.510
1.431
1.444
1.483
O3-C3a
C6a-N3b
bond angle (deg)
N3b-C6a-N6b
O3-C3a-N3b
113.688
112.979
113.688
111.289
109.724
113.158
111.400
108.208
C6a-N3b-N12b
C3a-N3b-N12b
5a : yield 1.3 g (29%); mp 137-139 °C (recryst from dioxane);
1
H NMR (D O) δ 2.73 (s, 2H), 3.70-3.73 (m, 4H); IR (KBr)
2
dihedral angle (deg)
-
1
+
N6b-C6a-N3b-N12b
O3-C3a-N3b-N12b
C6-C6a-N3b-N12b
N12b-C12a-N10b-C10a
C4-C3a-N3b-N12b
-51.554
-55.988
-177.998
178.483
-60.517
-59.184
176.328
-175.262
-179.000
3180 cm ; MS-CI m/ z 89 (M + 1 ). Anal. Calcd for
O: C, 40.89; H, 9.15; N, 31.79. Found: C, 40.61; H,
8.99; N, 31.51.
3 8 2
C H N
5
b: yield 1.1 g (22%); mp 106-108 °C (recryst from THF);
-177.259
1
H NMR (CDCl ) δ 1.09 (d, J ) 6.3 Hz, 3H), 2.28 (br, 2H),
.99 (br, 1H), 3.65 (s, 2H), 4.02 (m, 1H), 4.69 (br, 1H); IR (KBr)
3220 cm ; MS-CI m/ z 103 (M + 1 ). Anal. Calcd for
C H N O: C, 47.04; H, 9.87; N, 27.43. Found: C, 46.92; H,
3
2
-
1
+
EtOH (30 mL) was stirred for 24 h. The EtOH was evaporated
under reduced pressure. The resulting solid was triturated
with isopropyl ether, filtered, and recrystallized.
4
10
2
9.81; N, 27.46.
3
a : yield 5.3 g (88%); mp 215-216 °C (recryst from EtOH);
5
c: yield 1.3 g (23%); mp 123-125 °C (recryst from AcOEt);
1
H NMR (CDCl ) δ 2.25 (d, J ) 11.6 Hz, 2H), 3.22 (d, J ) 11.9
3
1
H NMR (CDCl ) δ 0.95 (t, J ) 7.6 Hz, 3H), 1.41-1.53 (m,
3
Hz, 2H), 3.88 (dt, J ) 11.4, 3.3 Hz, 2H), 4.00 (dd, J ) 11.4, 3.3
Hz, 2H), 4.19 (dd, J ) 11.6, 3.3 Hz, 2H), 4.28 (d, J ) 10.6 Hz,
2
1
1
H), 2.39 (br, 2H), 3.08 (br, 1H), 3.71 (s, 2H), 3.77-3.86 (m,
-1
H), 4.78 (br, 1H); IR (KBr) 3220 cm ; MS-CI m/ z 117 (M +
2
H), 4.62 (d, J ) 10.6 Hz, 2H), 5.20 (d, J ) 11.9 Hz, 2H); MS-
+
5 12 2
). Anal. Calcd for C H N O: C, 51.69; H, 10.41; N, 24.11.
+
EI m/ z 200 (M ). Anal. Calcd for C
8
H
16
N
4
O
2
: C, 47.99; H,
.05; N, 27.98. Found: C, 48.01; H, 8.14; N, 27.91.
b: yield 4.0 g (58%); mp 212-213 °C (recryst from AcOEt);
H NMR (CDCl ) δ 1.26 (d, J ) 5.9 Hz, 6H), 2.32 (d, J ) 11.0
Found: C, 51.60; H, 10.52; N, 24.37.
8
5
d : yield 2.2 g (27%); mp 193-194 °C (recryst from
3
1
1
dioxane); H NMR (DMSO-d
6
) δ 2.59 (br, 2H), 3.68 (s, 2H), 4.48
1
3
-
(
t, J ) 5.6 Hz, 1H), 5.38 (br, 1H); IR (KBr) 3180 cm ; MS-CI
Hz, 2H), 3.20 (d, J ) 12.1 Hz, 2H), 3.72-3.83 (m, 2H), 3.76 (d,
J ) 11.0 Hz, 2H), 4.31 (d, J ) 10.6 Hz, 2H), 4.61 (d, J ) 10.6
+
9 12 2
m/ z 165 (M + 1 ). Anal. Calcd for C H N O: C, 65.83; H,
+
7.37; N, 17.06. Found: C, 65.56; H, 7.22; N, 17.13.
Hz, 2H), 5.13 (d, J ) 12.1 Hz, 2H); MS-EI m/ z 228 (M ). Anal.
Gen er a l P r oced u r e for 1,3,5,7-Tet r a su b st it u t ed 2,6-
Dioxa -4a ,8a ,9a ,10a -t et r a a za p er h yd r o-a n t h r a cen e (7).
Meth od A. A solution of 5a ,b (10 mmol) and aldehydes 6a ,b
(10 mmol) in EtOH (20 mL) was stirred for 24 h at room
temperature. The EtOH was evaporated under reduced pres-
sure. The solid was obtained by trituration with Et O, filtered
2
by washing with isopropyl ether, and recrystallized from
solvent.
Calcd for C10
20 4 2
H N O : C, 52.61; H, 8.83; N, 24.54. Found: C,
5
2.61; H, 8.87; N, 24.39.
c: yield 2.0 g (29%); mp 138-139 °C (recryst from AcOEt);
H NMR (CDCl ) δ 0.96 (t, J ) 7.26 Hz, 6H), 1.41-1.68 (m,
3
1
3
4
3
1
0
H), 2.34 (dd, J ) 11.2, 2.6 Hz, 2H), 3.21 (d, J ) 11.6 Hz, 2H),
.21 (d, J ) 11.6 Hz, 2H), 3.55-3.64 (m, 2H), 3.77 (dd, J )
1.2, 10.9 Hz, 2H), 4.33 (d, J ) 10.6 Hz, 2H), 4.60 (d, J ) 1
+
.6 Hz, 2H), 5.13 (d, J ) 11.6 Hz, 2H); MS-EI m/ z 228 (M ).
Anal. Calcd for C10
H
20
N
4
O
2
:
C, 56.22; H, 9.44; N, 21.86.
Meth od B. A solution of 1a ,b (10 mmol) and aldehydes
6a ,b (10 mmol) in EtOH (20 mL) was stirred for 24 h at room
temperature. To the reaction mixture was added 2 (0.75 mL,
10 mmol, 37% w/v in water), and the solution was stirred for
Found: C, 56.18; H, 9.59; N, 21.95.
d : yield 4.6 g (44%); mp 259-260 °C (recryst from
3
1
benzene); H NMR (DMSO-d ) δ 2.56 (dd, J ) 12.1, 2.9 Hz,
6

Dioxatetraazaperhydroanthracenes and -perylenes
J . Org. Chem., Vol. 61, No. 12, 1996 4129
15a : yield 1.00 g (72%); mp 225 °C; 1H NMR (CDCl
) δ
2
4 h at room temperature. The solution was then treated as
described in method A.
a : method A yield 0.93 g (41%), method B 0.89 g (38%);
3
1.43-1.75 (m, 6H), 2.37 (dd, J ) 2.5, 11.7 Hz, 2H), 3.75 (td, J
) 3.7, 11.7 Hz, 2H), 3.90 (td, J ) 2.9, 11.7 Hz, 2H), 4.03 (dd,
J ) 2.9, 11.7 Hz, 2H), 4.58 (d, J ) 2.2 Hz, 2H), 4.89 (dd, J )
7
1
mp 127-128 °C (recryst from AcOEt); H NMR (CDCl
3
) δ 1.19
3.7, 11.0 Hz, 2H); 13C NMR (CDCl
) δ 17.76, 28.17, 30.97,
36.16, 66.21, 66.48, 87.86; IR (KBr) 2950, 1142, 1070, 950, 903,
(
)
4
d, J ) 6.3 Hz, 6H), 2.17 (dd, J ) 11.7, 2.0 Hz, 2H), 3.35 (d, J
3
11.9 Hz, 2H), 3.87-3.98 (m, 4H), 4.08 (d, J ) 11.9 Hz, 2H),
-1
+
.63 (q, J ) 6.3 Hz, 2H), 4.90 (d, J ) 11.9 Hz, 2H); MS-EI
838 cm ; MS-EI m/ z 280 (M ), 141, 110, 82, 55. Anal. Calcd
+
m/ z 228 (M ). Anal. Calcd for C10
N, 24.54. Found: C, 53.10; H, 8.97; N, 24.81.
b: method A yield 1.22 g (48%), method B 0.50 g (22%);
H
20
N
4
O
2
: C, 52.61; H, 8.83;
for C H O N : C, 59.88; H, 8.63; N, 19.98. Found: C, 59.81;
1
4
24
2
4
H, 8.58; N, 19.80.
5b: yield 0.65 g (42%); mp 234-235 °C; H NMR (CDCl
δ 1.20 (d, J ) 6.0 Hz, 6H), 1.47-1.79 (m, 12H), 2.29-2.53 (m,
H), 3.09-3.98 (m, 4H), 4.46-4.62 (m, 2H), 4.81 (m, 2H); IR
7
1
1
3
)
1
mp 138-139 °C (recryst from EtOH); H NMR (CDCl ) δ 0.93
3
(
t, J ) 7.6 Hz, 6H), 1.38-1.59 (m, 4H), 2.18 (dd, J )12.2, 2.3
2
(
(
Hz, 2H), 3.35 (d, J ) 11.9 Hz, 2H), 3.87-3.98 (m, 2H), 3.28 (d,
-1
KBr) 2950, 1235, 1165, 1072, 1000, 820 cm ; MS-EI m/ z 308
J ) 11.9 Hz, 2H), 3.87-4.05 (m, 4H), 4.35 (t, J ) 6.6 Hz, 2H),
+
M ), 155, 139, 82, 55. Anal. Calcd for C16
28 2 4
H O N : C, 62.31;
+
4
.92 (d, J ) 11.9 Hz, 2H); MS-EI m/ z 256 (M ). Anal. Calcd
H, 9.15; N, 18.17. Found: C, 62.11; H, 9.10; N, 17.74.
for C12
H, 9.39; N, 21.74.
c: method A yield 0.61 g (24%), Method B 0.36 g (14%);
24 4 2
H N O : C, 56.22; H, 9.44; N, 21.86. Found: C, 55.16;
1
1
5c: yield 0.77 g (46%); mp 183 °C; H NMR (CDCl
3
) δ 0.95
(
3
t, J ) 2.0 Hz, 6H), 1.33-1.83 (m, 16H), 2.28-2.56 (m, 2H),
7
1
3
.11-3.97 (m, 4H), 4.48-4.68 (m, 2H), 4.69-5.03 (m, 2H);
C
1
mp 155-157 °C (recryst from EtOH); H NMR (CDCl
3
) δ 1.18
NMR (CDCl
6.29, 87.28; IR (KBr) 2900, 1238, 1170, 1038, 940, 862 cm
3
) δ 9.29, 17.97, 27.02, 28.26, 30.97, 46.71, 66.24,
(
1
2
d, J ) 5.6 Hz, 6H), 1.20 (d, J ) 5.6 Hz, 6H), 2.22 (dd, J )
-
1
7
;
1.9, 3.0 Hz, 2H), 3.32 (d, J ) 11.9 Hz, 2H), 3.66-3.83 (m,
+
MS-EI m/ z 336 (M ), 169, 153, 82, 55. Anal. Calcd for
18 32 2 4
C H O N : C, 64.25; H, 9.59; N, 16.65. Found: C, 64.29; H,
H), 4.60 (q, J ) 5.9 Hz, 2H), 4.81 (d, J )11.9 Hz, 2H); MS-EI
+
m/ z 256 (M ). Anal. Calcd for C12
24 4 2
H N O : C, 56.22; H, 9.44;
9
.43; N, 16.30.
N, 21.86. Found: C, 55.92; H, 9.49; N, 21.80.
X-r a y An a lysis19 of 13 a n d 15a . Single crystals of
C H O N4 and C H O N4 corresponding to 13 and 15a ,
16 28 2 14 24 2
respectively, were prepared by allowing an ethanol solution
of the compound to stand for about 3 weeks. Their crystal-
lographic data are shown in Table 1.
7
d : method A yield 0.97 g (34%), Method B 0.65 g (23%);
1
mp 117-119 °C (recryst from AcOEt); H NMR (CDCl
3
) δ 0.95
(
4
3
)
(
t, J ) 7.6 Hz, 6H), 1.21 (d, J ) 5.9 Hz, 6H), 1.41-1.62 (m,
H), 2.50 (dd, J ) 11.6, 2.6 Hz, 2H), 3.27 (d, J ) 11.9 Hz, 2H),
.69 (dd, J ) 11.6, 10.6 Hz, 2H), 3.79-3.88 (m, 2H), 4.30 (t, J
6.60 Hz, 2H), 4.85 (d, J ) 11.9 Hz, 2H); MS-EI m/ z 284
Cr ysta l Da ta . The cell constants were determined from a
least-squares procedure using the Bragg angles of 20 reflec-
tions measured on a RIGAKU AFC-6 four-circle autodiffrac-
tometer equipped with a graphite monochromatic Mo KR
source, which was interfaced to a PANAFACOM U-1200
minicomputer. The space groups were selected from system-
atic absences and the number of molecules per unit cell and
were later confirmed in subsequent structure refinement.
Intensity data were collected in the range of 2θ < 55° using
the ω-2θ scan technique. A variable scan rate was adopted.
Two reflections were monitored after every measurement of
+
28 4 2
M ). Anal. Calcd for C14H N O : C, 59.13; H, 9.92; N, 19.70.
Found: C, 59.06; H, 9.85; N, 19.78.
Hyd r a zin ocycloh exa n ol (11). Cyclohexene oxide (10.1
mL, 100 mmol) was slowly added to hydrazine hydrate (25 mL,
5
00 mmol) that had been preheated to 60-70 °C. The reaction
mixture was then heated at 100 °C for 2 h. Hydrazine was
removed under reduced pressure. The residue was triturated
with Et
mp 66-68 °C; IR (KBr) 3200 cm
,6-Dioxa -4a ,8a ,9a ,10a -tetr a a za p er h yd r od iben z[c,j]a n -
2
O and cooled under ice and water: yield 9.3 g (95%);
-
1
.
2
th r a cen e (12) a n d 2,6-Dioxa -4a ,8a ,9a ,10a -tetr a a za p er h y-
d r od iben z[c,h ]a n th r a cen e (13). A solution of hydrazinocy-
clohexanol (11) (3.9 g, 30 mmol) and 2 (4.5 mL, 60 mmol, 37%
w/v in water) in EtOH (30 mL) was stirred for 24 h at rt. The
EtOH was then removed under reduced pressure. The residue
1
00 reflections. Of the 3105 and 1548 independent reflections
for 13 and 15a , 1564 and 742 were treated as observed (F
.0σF and F > 4.0σF), respectively. The intensities were
o
>
3
o
corrected for Lorentz and polarization effects, but no correction
was applied for absorption.
3
was subjected to silica gel column chromatography (CHCl :
Str u ctu r e Solu tion a n d Refin em en t. An overall tem-
perature factor obtained from a Wilson plot gave the correct
solution. The structure was determined by the direct method
using the MULTAN78 programs.20 An E map calculated using
the signed E’s (E >1.2) revealed the positions of all of the
expected non-hydrogen atoms. Refinements were carried out
by the block-diagonal least-squares method. Six cycles of
isotropic refinement and six cycles of anisotropic refinement
led to an R index. All of the hydrogens were located at the
calculated positions. After adding the hydrogens, but keeping
their positional and thermal parameters fixed [B(H) )
B(C)+1.0], and refining, we obtained a final R. All of the
MeOH ) 19:1). The solids 12 and 13 obtained from each
fraction were recrystallized from benzene and ethyl acetate,
respectively.
1
2: yield 3.05 g (33%); mp 280-285 °C; MS-EI m/ z 308
+
1
(
6
3
M ); H NMR (CDCl
3
) δ 1.03-1.12 (m, 2H), 1.35-1.50 (m,
H), 1.76-1.86 (m, 8H), 3.26-3.32 (m, 2H), 3.61-3.68 (m, 2H),
.62 (d, J ) 12.5 Hz, 2H), 4.30 (d, J ) 10.6 Hz, 2H), 4.69 (d,
1
3
J ) 10.6 Hz, 2H), 4.99 (d, J ) 12.5 Hz, 2H); C NMR (CDCl
δ 24.47, 24.65, 27.08, 30.63, 58.17, 82.51 (CH ), 52.91, 80.57
CH). Anal. Calcd for C16 4: C, 62.31; H, 9.15; N, 18.17.
Found: C, 62.58; H, 9.28; N, 18.20.
3: Yield 1.48 g (16%); mp 231-233°C; MS-EI m/ z 308
3
)
2
(
28 2
H O N
1
2
1
+
1
structure-solving programs and the drawing program (ORTEP)
(
6
M ); H NMR (CDCl
H), 1.75-1.90 (m, 8H), 2.10 (t, J ) 8.1 Hz, 1H), 3.28-3.34
m, 2H), 3.33 (d, J ) 9.5 Hz, 1H), 3.51-3.54 (m, 1H), 3.53 (d,
J ) 11.4 Hz, 1H), 4.12 (d, J ) 7.7 Hz, 1H), 4.18-4.24 (m, 1H),
3
) δ 1.04-1.38 (m, 2H), 1.40-1.44 (m,
were provided by the Computer Center of Kumamoto Univer-
sity, along with the Universal Crystallographic Computation
Program System (UNICS III).22
(
4
1
2
8
.23 (d, J ) 9.5 Hz, 1H), 4.35 (d, J ) 7.7 Hz, 1H), 4.42 (d, J )
0.6 Hz, 1H), 4.81 (d, J ) 10.6 Hz, 1H); 13C NMR (CDCl
) δ
4.07, 24.23, 24.38, 24.65, 26.20, 26.69, 30.57, 61.35, 66.39,
J O951878P
3
2.48, 84.03 (CH
2
), 53.50, 66.39, 80.42, 82.48 (CH). Anal.
: C, 62.31; H, 9.15; N, 18.17. Found: C,
(19) The atomic coordinates for this structure have been deposited
with the Cambridge Crystallographic Data Centre. The coordinates
can be obtained, on request, from the Director, Cambridge Crystal-
lographic Data Center, 12 Union Road, Cambridge, CB21EZ, U.K.
Calcd for C16
6
H
28
O
2
N
4
1.74; H, 8.97; N, 17.80.
Gen er a l P r oced u r e for 2,8-Disu bstitu ted 3,9-Dioxa -
b,6b,9b,12b-tetr a a za p er h yd r op er ylen es 15. To a vigor-
(20) Main, P.; Hull, S. E.; Lessinger, L.; Germain, G.; Declercq, J .
3
P.; Wooltson, M. M. MULTAN 78, A System of Computer Programs
for the Automatic Solution of Crystal Structure from X-Ray Diffraction
Data, University of York, England, 1978.
ously stirred solution of 1 (5 mmol) in THF (15 mL) was
gradually added 2 (1 mL, 5 mmol, 50% w/v in water) at room
temperature. After the reaction mixture was stirred for 15 h,
the solvent was evaporated to dryness under reduced pressure.
After the residue was treated with ether (15 mL), the resulting
precipitate was collected and recrystallized from EtOH.
(21) J ohnson, C. K. ORTEP, Report ORNL-3794; Oak Ridge Na-
tional Laboratory: Oak Ridge, TN, 1965. The program was locally
modified for use of X-Y plotters devised for personal computers.
(22) Sakurai, T.; Kobayashi, K. Rigaku Kenkyusho Houkoku 1974,
55, 69. Kawano, S. Koho Comp. Center Kyushu Univ. 1983, 16, 113.