Study on ferrocenes, part 9 [1] substrate selective transformations of some ferrocenylhydrazones

Article abstract of DOI:10.1016/S0022-2860(01)00429-X

Different ferrocenyl-substituted hydrazones of heteroaryl/arylhydrazines (1a-e, 3 and 10) were reacted with dimethyl fumarate and gave two epimeric pairs of pyrazolidines (t,c-2a and t,c-2c), pyrazolines (5a,c, 6, 11), pyrazoles (8, 12) and condensed tria

Full text of DOI:10.1016/S0022-2860(01)00429-X

Journal of Molecular Structure 569 .2001) 185±194
www.elsevier.nl/locate/molstruc
Study on ferrocenes, part 9 [1] substrate selective transformations
of some ferrocenylhydrazones
a
A. Abran, A. Csampai , A. Kotschy, O. Barabas, P. Sohar
Â
*
Â
Â
Â
Â
Department of General and Inorganic Chemistry, Research Group for Structural Chemistry and Spect., Hungarian Academy of Sciences
È È
 Â
Ð Eotvos Lorand University, P.O. Box 32, H-1518 Budapest 112, Hungary
Received 15 November 2000; accepted 13 December 2000
Abstract
Different ferrocenyl-substituted hydrazones of heteroaryl/arylhydrazines .1a±e, 3 and 10) were reacted with dimethyl
fumarate and gave two epimeric pairs of pyrazolidines .t,c-2a and t,c-2c), pyrazolines .5a,c, 6, 11), pyrazoles .8, 12) and
condensed triazoles .4a, 7). On treatment with dicyanodichloroquinone .DDQ) the pyrido[2,3-d]pyridazinyl-substituted hydra-
zones .1b,e) were cyclized to condensed triazoles 4b,e. The reactivity of 1a±e in the cycloaddition process and oxidation level
of the products were found to be highly dependent on the temperature and on the heteroaryl/aryl group. The signi®cant
difference was explained by theoretical calculations. The structures of products were determined by IR, 1D and 2D NMR
methods .including HMQC, HMBC and DNOE) and single crystal X-ray analysis. q 2001 Elsevier Science B.V. All rights
reserved.
Keywords: Ferrocenylhydrazones; Dipolar cycloaddition; Stereoselectivity; Semi-empirical calculations; Structure determination
1. Introduction
pharmaceutical interest, we investigated the analo-
gous cycloaddition and oxidative cyclization
reactions of the ferrocenylhydrazones of pyrido[2,3-
d]- and pyrido[3,4-d]pyridazinylhydrazines .1b±e),
thieno[3,2-c]pyridylhydrazine .3) and 4-bromo-
phenylhydrazine .10) as shown on Scheme 1.
In our previous paper [2] we reported on the 1,3-
dipolar cycloaddition reactions of ferrocenyl-hydra-
zones of 2-hydrazinopyridine and 4-hydrazinophtha-
lazinones with some dipolarophilic reagents. The
reactivity of the substrates in these reactions
proceeding by proton shift followed by the actual
cycloaddition was found to be dependent on the
heterocyclic moiety. In order to get more information
about the mechanism and to extend the group of 3-
ferrocenyl-1-heteroaryl/arylpyrazole- and polycon-
densed 1,2,4-triazole derivatives, which may be of
2. Experimental
Melting points .uncorrected) were determined with
a Boetius apparatus. The IR spectra were recorded in
KBr pellets with a BRUKER IFS 55 FT-spectrometer.
The 1H and 13C NMR spectra were recorded in CDCl3
or [D6]DMSOsolution in 5 mm tubes at RT, on a
Bruker DRX 500 spectrometer at 500 .1H) and 125
.13C) MHz, with the deuterium signal of the solvent as
the lock and TMS as internal reference. The standard
* Corresponding author. Tel.: 136-1-209-0599; fax: 136-1-209-
0602.
Â
E-mail address: sohar@para.chem.elte.hu .P. Sohar).
a
Corresponding author.
0022-2860/01/$ - see front matter q 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-2860.01)00429-X

Â
A. Abran et al. / Journal of Molecular Structure 569 92001) 185±194
186
Scheme 1.
Bruker microprogram NOEMULT.AU to generate
NOE was used with a selective preirradiation time.
DEPT spectra were run in a standard manner, using
only the Q ˆ 1358 pulse to separate CH/CH3 and CH2
lines phased ªupº and ªdownº, respectively. The 2D
HMQC and 2D HMBC spectra were obtained by
using the standard Bruker pulse programs INV4GS
and INV4GSLPLRND, respectively.
Hydrazones 1a±e, 3 and 10. Preparation and char-
acterization of 1a,b,e are reported in Ref. [3].

Â
A. Abran et al. / Journal of Molecular Structure 569 92001) 185±194
187
Table 1
The calculated heats of formation .kJ/mol)/HOMO energies .eV) for the neutral .1pa±e and 3p) and relative heats of formation .kJ/mol) for the
dipolar forms .I±III) .AM1, MOPAC 6.0)
DHf .1p, 3p in row f)
DDHf .I±1p, 3p in row f)
DDHf .II±1p)
DDHf .III±1p, 3p in row f)
a
b
c
d
e
f
338.53/28.87
389.40/28.55
385.39/29.03
382.88/29.16
399.82/29.03
399.22/28.35
62.26/27.77
69.08/27.77
60.04/27.93
60.96/27.94
62.34/27.87
56.76/27.58
±
19.33/28.85
28.66/28.92
19.20/29.02
20.58/29.07
21.59/28.96
24.62/27.88
158.57/26.70
182.13/26.64
177.40/26.44
163.72/26.69
±
According to the described procedure formyl-ferro-
cene and the appropriate hydrazine component .8-
hydrazino-6-methylpyrido[3,4-d]pyridazin-5.6H)-one
.Ac) for 1c, 5-hydrazino-7-methylpyrido[3,4-d]pyri-
dazin-8.7H)-one .Ad) for 1d, 4-hydrazinothieno[3,2-
c]pyridine [7] for 3 and 4-bromo-phenylhydrazine for
10) were condensed to give novel hydrazones: 8-[.E)-
ferrocenylmethylidene-hydrazino]-6-methylpyrido
[3,4-d]pyridazin-5.6H)-one .1c) .89%); 5-[.E)-
ferrocenylmethylidene-hydrazino]-7-methylpyrido
[3,4-d]pyridazin-8.7H)-one .1d) .76%); 4-[.E)-
ferrocenylmethylidene-hydrazin0o]thieno[3,2-c]pyridine
.3) .92%); N-.4-bromophenyl)-N -ferrocenylmethylidene-
hydrazine .10) .91%). The two-step preparation of
Ac,d via the appropriate chloro compound .Bc,d) is
described below.
.1H, s), 9.01 .1H, d, J ˆ 5.4), 7.68 .1H, d, J ˆ 5.4),
3.77 .3H, s); 13C NMR .CDCl3) 158.3, 153.7, 151.3,
135.9, 134.4, 122.4, 117.6, 40.0; Anal. of Bc,d calcd.
for C8H6ClN3O.195.6) C 49.12, H 3.09, Cl 18.12, N
21.48; Found for Bc/Bd C 49.10/49.21, H 3.12/3.12,
Cl 18.08/18.12, N 21.50/21.43.
Preparation of Ac,d. A mixture of hydrazine
hydrate .98%, 22 ml), EtOH .3 ml) and the corre-
sponding chloro compound .Bc,d) .3.91 g, 0.02 mol)
was re¯uxed for 1 h then diluted with water .100 ml).
The precipitated product was recrystallized from
water. Ac Yield 75%; mp 214±178C; IR .cm21
)
1
3330, 3281, 1658, 1609; H NMR .[D6]DMSO) 9.52
.1H, s), 8.90 .1H, d, J ˆ 5:5), 8.29 .1H, d, J ˆ 5:5),
8.21 .1H, s), 4.14 .2H, br s), 3.65 .3H, s); 13C NMR
.[D6]DMSO) 157.6, 151.8, 148.1, 144.4 132.1 120.6,
Preparation
of
8-chloro-6-methylpyrido[3,4-
119.0, 39.6. Ad Yield 88%; mp 227±298C; IR .cm21
)
1
d]pyridazin-596H)-one 9Bc) and 5-chloro-7-methyl-
pyrido[3,4-d]pyridazin-897H)-one 9Bd). Sodium
.1.38 g, 0.06 mol) and, subsequently, 8-chloro-
pyrido[3,4-d]pyridazin-5.6H)-one [8]/5-chloropyrido
[3,4-d]pyridazin-8.7H)-one [8] were dissolved in
methanol .200 ml) with moderate heating. The
mixture was stirred at 35±408C until a clear solution
was obtained .20±25 min). Following the addition of
CH3I .8.52 g, 0.06 mol) the solution was re¯uxed for
1 h then cooled, ®ltered and evaporated. The solid
residue was extracted with CH2Cl2 .100 ml). The
solution was washed with water .2 £ 50 ml), dried
.MgSO4) and evaporated to obtain the pure product.
Bc Yield 67%; mp 153±558C; IR .cm21) 1674, 1595,
3322, 3283, 1650, 1611; H NMR .[D6]DMSO) 9.43
.1H, s), 9.00 .1H, d, J ˆ 5:5), 8.27 .1H, s), 7.96 .1H,
d, J ˆ 5:5), 4.14 .2H, br s), 3.62 .3H, s); 13C NMR
.[D6]DMSO) 156.8, 153.0, 150.6, 146.2, 130.0, 122.4,
116.2, 39.2; Anal. of Ac,d calcd. for C8H9N5O.191.2)
C 50.26, H 4.75, N 36.63; Found for Ac/Ad C 50.32/
50.44, H, 4.73/4.77, N 36.61/36.70.
Reactions of hydrazones 1a±e, 3 and 10 with
DMFM, 9general procedure for Methods A and B).
A
reaction mixture containing the hydrazone
.2.5 mmol) and DMFM .0.86 g, 6 mmol) was stirred
and heated at 708C .Method A) or re¯uxed .Method B)
in 30 ml of freshly distilled MeCN in the presence of
Ê
0.5 g of molecular sieves .3 A) under argon atmo-
1
1573; H NMR .CDCl3) 9.12 .1H, s), 8.95 .1H, d,
sphere. The TLC-monitored reactions were conducted
for the following periods of time: 8 h .Method A) and
15 h .Method B) for 1a; 40 h .Methods A and B) for
1b±e; 30 min .Method A) and 2 h .Method B) for 3;
3 h .Method A) and 6 h .Method B) for 10. The solvent
J ˆ 5.4), 8.22 .1H, d, J ˆ 5.4), 3.76 .3H, s); 13C
NMR .CDCl3) 159.1, 152.8, 150.6, 134.7, 133.7,
121.4, 119.4, 40.2; Bd Yield 71%; mp 162±648C;
1
IR .cm21) 1670, 1596, 1569; H NMR .CDCl3) 9.62

Â
A. Abran et al. / Journal of Molecular Structure 569 92001) 185±194
188
d
n
n
n
n

Â
A. Abran et al. / Journal of Molecular Structure 569 92001) 185±194
189
d

Â
A. Abran et al. / Journal of Molecular Structure 569 92001) 185±194
190
Table 4
Results of DNOE experiments with compounds 1±3, 6, 7 and 12 .Interacting pairs showing only trivial effects .NOE between geminal or vicinal
hydrogens) are not included in this Table. Responses relevant for stereostructures are exclusively given.)
Responding signals
Saturated signal
H-8
H-40
H-2,5a
H-30
H-3
OMe
H-6
NH
1d, t-2c
t-2c
3
3
H-2,5b
OMe
H-5
t-2a,c, 6
t-2a
7
c-2a
c-2a
c-2a
12
t-2a
a
Substituted cyclopentadiene ring .Fc).
was evaporated in vacuo to dryness. The residue was
contaminated by paramagnetic substances in each
case. The separation by chromatography on silica
.Kieselgel type 9385; eluent: CHCl3) and subsequent
recrystallization from ethanol gave the pure products
.see Tables 2±5): 4-[.3Rp,4Sp,5Sp)-4,5-bis-carbo-
methoxy-3-ferrocenyl-2,3,4,5-tetrahydropyrazol-1-yl]-
2-methylphthalazin-1.2H)-one .t-2a) .40% by Method
A); 4-[.3Rp,4Rp,5Rp)-4,5-bis-carbo-methoxy-3-ferro-
cenyl-2,3,4,5-tetrahydropyrazol-1-yl]-2-methylphtha-
lazin-1.2H)-one .c-2a) .33% by Method A); 8-
[.3Rp,4Sp,5Sp)-4,5-bis-carbomethoxy-3-ferrocenyl-
2,3,4,5-tetrahydropyrazol-1-yl]-6-methylpyrido[3,4
-d]pyridazin-5.6H)-one .t-2c) .38% by Method A
and 44% by Method B); 8-[.3Rp,4Rp,5Rp)-4,5-bis-
carbomethoxy-3-ferrocenyl-2,3,4,5-tetrahydropyrazol-
1-yl]-6-methylpyrido[3,4-d]pyridazin-5.6H)-one .c-
2c) .4% by Method A and 10% by Method B); 4-
[.3Rpp,4Rp,5Sp)-4,5-bis-carbomethoxy-3-ferrocenyl-
4,5-dihydropyrazol-1-yl]thieno[3,2-c]pyridine .6)
.75% by Method A); 3-ferrocenyl-[1,2,4]-tria-
zolo[4,3-a]thieno[3,2-c]pyridine .7); .14% by
Method A and 69% by Method B) 4-.4,5-bis-carbo-
methoxy-3-ferrocenylpyrazol-1-yl)-thieno[3,2-c]
pyridine .8) .22% by Method B); .3Rpp,4Rp,5Sp)-1-
.4-bromophenyl)-4,5-bis-carbomethoxy-3-ferro-
cenyl-4,5-dihydropyrazole .11) .82% by Method
A); 1-.4-bromophenyl)-4,5-bis-carbomethoxy-3-
ferrocenyl-pyrazole .12) .69% by Method B). Hydra-
zones 1b,d,e were recovered unchanged in good yields
.93, 88 and 95%) by chromatography.
Oxidation of hydrazones 1b,e, pyrazolidines t,c-2c
and pyrazolines 6 and 11 by Method C. The mixture
of the substrate .1 mmol) and DDQ .0.227 g; 1 mmol)
was dissolved in CH2Cl2 .5 ml). The reaction mixture
was stirred at RT for 6 h then evaporated to dryness.
The solid residue was recrystallized from ethanol .5±
10 ml) to obtain the corresponding product .see
Tables 2±5): 3-ferrocenyl-5-methyl-[1,2,4]-tria-
zolo[4,3-b]pyrido[2,3-d]pyridazin-6.5H)-one
.4b)
.67%); 3-ferrocenyl-5-methyl-[1,2,4]-triazolo[4,3-b]-
pyrido[3,2-d]pyridazin-6.5H)-one .4e) .71%); 5c
.83% from t-2c and 62% from c-2c); 8 .77%); 12
.85%). Employing the same procedure for hydrazones
1a,c,d and pyrazolidines t,c-2a in each case a deep
dark solution with unidentifyable paramagnetic
substances was formed.
Partial hydrolysis and decarboxylation of pyrazole
8 by Method D. The substrate .0.501 g; 1 mmol) was
dissolved in 100 ml of MeCN±H2O.97-3). The solu-
tion was boiled for 15 min then evaporated in vacuo.
The solid residue was recrystallized from EtOH .4 ml)
to obtain 4-.4-carbomethoxy-3-ferrocenylpyrazol-1-
yl)-thieno[3,2-c]pyridine .9) .0.332 g; 75%) as yellow
needles .see Tables 2±5).
3. Results and discussion
The reaction of 1a with dimethylfumarate .DMFM)
in boiling MeCN .Method B), which resulted in the
mixture of condensed triazole 4a and pyrazoline 5a,
has been reported [2]. When the conversion was
conducted at lower temperature .708C, Method A), a
5.5:4.5 mixture of epimeric pyrazolidines .t-2a and c-
2a) could be isolated in good yield .73%). Neither 4a
nor 5a were isolated after this reaction. Now, we
Oxidation of pyrazolidines t,c-2a and pyra-
zolines 6 and 11 carried out under the conditions
of Method B .reaction time 6 h) followed by the
puri®cation described above yielded 76±88% of
5a .from t,c-2a), 8 .from 6) and 12 .from 11).

Â
A. Abran et al. / Journal of Molecular Structure 569 92001) 185±194
191
Table 5
Physical and analytical data of compounds 1c,d, t-2a,c, c-2a,c 3, 4b,e, 5c and 6±12
Appearance
Mp 8C
Formula
Calcd%/Found%
C
H
N
S/Bra
1c
Orange powder
Orange powder
Yellow needles
Yellow needles
Yellow powder
Yellow powder
Brown powder
Red powder
194±197
203±205
135±137
140±143
139±141
134±137
204±206
168±170
176±177
170±171
160±162
150±152
153±154
179±180
157±160
131±133
120±121
C19H17FeN5O58.93/59.00
C19H17FeN5O58.93/58.95
4.42/4.39
4.42/4.37
18,09/18.14
18,09/18.10
±
±
1d
t±2a
c±2a
t±2c
c±2c
3
C26H26FeN4O5
C25H25FeN5O5
C26H26FeN4O5
C25H25FeN5O5
C18H15FeN3S
58.88/58.79
56.51/56.57
4.94/4.98
4.74/4.71
4.94/4.88
4.74/4.70
4.19/4.18
10.56/10.56
13.18/13.20
10.56/10.58
13.18/13.21
11.63/11.68
±
±
±
±
58.88/58.89
56.51/56.60
59.85/59.78
8.86/8.90
4b
4e
C19H15FeN5O59.24/59.29
C19H15FeN5O59.24/59.20
3.92/3.90
3.92/3.96
18.18/18.11
18.18/18.17
±
±
Red powder
5c
6
Orange powder
Orange needles
Orange powder
Orange powder
Orange powder
Red needles
C25H23FeN5O5
C24H21FeN3O4S
C18H13FeN3S
56.73/56.81
57.27/57.28
4.38/4.40
4.21/4.16
3.65/3.69
3.82/3.75
3.87/3.87
3.94/3.88
4.03/4.11
3.66/3.58
13.23/13.19
8.35/8.36
11.70/11.73
8.38/8.40
9.48/9.55
7.31/7.38
5.33/5.35
5.35/5.34
±
6.37/6.40
8.93/8.92
6.40/6.41
7.23/7.20
20.86/20.91
15.21/15.19
15.27/15.22
7
60.18/60.14
57.50/57.49
59.61/59.56
53.30/53.33
52.60/52.49
52.80/52.77
8
C24H19FeN3O4S
C22H17FeN3O2S
C17H15BrFeN2
C23H21BrFeN2O4
C23H19BrFeN2O4
9
10
11
12
Orange needles
Orange cubes
a
S% for compounds 3, 6, 7, 8 and 9 and Br% for compounds 10, 11 and 12.
intended to compare the reactivity of 1b±e [3] with
that of 1a towards DMFM.
signi®cant effect on the redox properties both of the
hydrazone and the primarily formed pyrazolidine, as
supported by the following observations. While the
reaction of 1a by Method B gave directly pyrazoline
5a and condensed triazole 4a, [2] under the same
conditions 1c produced again t-2c and c-2c with a
slightly lower selectivity .8.2:1.8, yield: 54%), but
formation neither of pyrazoline 5c nor a condensed
triazole analogous to 4a was detected even in traces.
In control experiments both t-2a and c-2a could be
oxidized to 5a in re¯uxing MeCN in the presence of
DMFM under an argon atmosphere .Method B).
However, in the absence of DMFM no oxidation
occurred in the boiling solvent suggesting that the
oxidizing agent is the reagent itself. Accordingly,
neither t-2c nor c-2c could be oxidized to 5c with
DMFM. However, DDQ dissolved in CH2Cl2 .Method
C) proved to be a suitable reagent not only to perform
conversions t; c-2c ! 5c, but also oxidative cycliza-
tion of pyrido[2,3-d]pyridazinylhydrazones ꢀ1b; e !
4b; e† at room temperature. Using this reagent the
hydrazones 1a,c,d and the epimers t,c-2a underwent
decomposition to unidentifyable substances. Pyrazo-
lines 5a,c resisted aromatization by Method C.
Under the same conditions .Method A) the two
pyrido[2,3-d]pyridazinylhydrazones .1b,e) and one
of the pyrido[3,4-d]pyridazinylhydrazone .1d) did
not react at all. After prolonged treatment .40 h)
at re¯ux temperature .Method B) 1b,d,e were
recovered unchanged from the reaction mixtures.
Pyrido[3,4-d]pyridazinylhydrazone .1c) was the
only isomer which underwent
a very slow
cycloaddition even at 708C affording a 9:1 mixture
of diastereomeric pyrazolidines t-2c and c-2c
.reaction time: 40 h; yield: 42%). In accord with
our expectations the t:c ratios for 2a,c .5.5:4.5 and
9:1, respectively) clearly indicate that the cycload-
dition involving the intermediate formed by proton
shift [2] from 1c takes place with much higher
diastereoselectivity than that involving the dipol
formed from 1a of more pronounced donor char-
acter. On the other hand, the importance of the
acceptor character of the dipolarophilic component
is also indicated by the fact that similar reactions
of 1a±e could not be carried out with cinnamic
acid ethyl ester containing a less activated CyC
double bond than does DMFM.
Reactions of thieno[3,2-c]pyridyl- and 4-bromo-
phenylhydrazones .3 and 10) also demonstrate the
The electronaf®nity of the N-substituent has also a

Â
A. Abran et al. / Journal of Molecular Structure 569 92001) 185±194
192
was effected by boiling MeCN±H2O.97-3) . Method
D). Analogous conversion of pyrazole 12 could not be
carried out by Method D.
The different reactivity pattern of closely related
hydrazones .i.e. only 1a and 1c reacted with
DMFM) is quite unprecedented in the literature. In
order to rationalize this experimental ®nding we
carried out semi-empirical calculations for the
possible ªactiveº forms of the hydrazones, using the
p
model systems 1pa±e and 3p .Me±CHyN±NHQ,
indicates that the Fc is replaced by a methyl group
as shown on Scheme 2). Earlier results suggest [2]
that the generation of the active species proceeds via
proton shift inside the hydrazone moiety. However,
except for 10, this proton transfer can lead to more
different intermediates .Scheme 2). The proton can
move onto the adjacent hydrazone nitrogen .Ia±f),
in form II onto the nitrogen of the distant heterocycle
B .unavailable at 1pa and 3p), meanwhile in form III
onto the proximal nitrogen of the pyridazine/pyridine
ring A.
In the course of the calculations the geometry of the
models 1p and 3p and all the isomeric forms I±III was
optimised on semi-empirical .AM1) level and the
heats of formation as well as the relevant frontier
orbital .HOMO) energies .with reasonable 2pz atomic
orbital coef®cients on both the carbon and the distant
nitrogen of the hydrazone moiety) were calculated
.MOPAC 6.0) and listed in Table 1. The calculated
data suggest that the formation of the zwitterionic
forms I and particularly II are unfavorable, and their
formation .even in equilibrium) might practically be
ruled out.
On the other hand, formation of the isomers III is
not only feasible because of their reasonably lower
heats of formation, but also re¯ects the tendency
that in case of the reactive condensed pyridazines
.1pa,c) their formation is more favorable than that of
the completely resistant isomers .1pb,d,e). Our
hypothesis is in very good accord with the case of
thienopyridyl derivative 3, the far most reactive inves-
tigated substrate, of which conversion by Method A
was completed within 0.5 h. Calculations on its model
analogue .3p) furnished a strong argument for its high
reactivity as the formation of tautomer IIIf is particu-
larly favorable and its HOMO energy is also signi®-
cantly higher than those for IIIa±e. Consequently, it
seems plausible that the key step in the investigated
Scheme 2.
dependence of the reactivity of ferrocenylhydrazones
and the oxidation level of products on the N-substi-
tuent. On treatment by Method A both 3 and 10 gave
directly pyrazolines .6 and 11, respectively). Prob-
ably, due to their fast oxidation the primarily formed
pyrazolidines could not be isolated in either cases.
Besides pyrazoline 6 .75%) some condensed triazole
7 .14%) was also isolated. The reaction of 3 by
Method B afforded 7 as the major product .69%)
and a smaller amount of 8 .22%). Under the same
conditions bromophenyl derivative 10 was converted
to pyrazole 12. Aromatization of the isolated pyrazo-
lines .6 ! 8 and 11 ! 12, respectively) could be
achieved either by Method B or C. Finally, the
enhanced electron density of the pyridine ring due
to the presence of the fused electron-releasing thio-
phene is re¯ected by the facile conversion of 8 ! 9
probably including hydrolysis assisted by the pyri-
dine-nitrogen and subsequent decarboxylation which

Â
A. Abran et al. / Journal of Molecular Structure 569 92001) 185±194
193
reactions of 1a±e and 3 is the formation of the reac-
tive intermediate III which then undergoes a fast
cycloaddition associated with 1,4-proton migration
.Scheme 2).
ture of the cis isomers is manifested in the up®eld shift
.steric compression shift [5,6]) of the C-40 and one of
C-2,5.Fc) lines by 2.2 and 1.5 ppm for c/t-2a and by
1.6 and 0.8 ppm for the c-t pair of 2c. This effect is
observable also for C-1.Fc) line of c/t-2a pair
.2,7 ppm). Due to bad solubility, the shift of C-1.Fc)
signal of c-2c is detectable only with limited accuracy,
thus the similar effect for the pair c/t-2c was not deter-
minable. The anisotropic shielding effect of the ester
carbonyl group also results in an up®eld shift of the H-
2,5 and H-3,4 signals in the cis-2a,c isomers, while the
NH-signal shows an opposite and larger down®eld
shift relative to the trans-counterparts .by 0.36 and
0.54 ppm for the t-c pairs of 2a,c), due to stronger
association in accord with the less hindered steric
position of the NH-group in c-2a,c.
1
The characteristic IR-bands and the H and 13C
NMR data of the new compounds are given in Tables
2 and 3.
The spectral data con®rm the structures unambigu-
ously without any explanation. .The single crystal X-
ray analysis of pyrazole 12 was also carried out and
provided as supplementary proof.) It is necessary to
make the following remarks: In the pyrazolines .5c, 6
and 11) and pyrazolidines .t/c-2a,c) the rotation of the
ferrocenyl moiety is hindered. As a consequence, the
signals of H/C-2,5 and H/C-3,4 of the substituted
cylopentadienyl ring are separated due to the chemical
non-equivalance of the related atom pairs in contrast
to the same signals of the other compounds containing
a pyrazole ring, in which these atom pairs have iden-
tical shifts and signals.
4. Conclusions
The possible tautomeric structure of the hydrazones
.1) containing the NH-group in the ring was excluded
for 1d by DNOE experiment .Table 4) due to the NOE
observed between the exocyclic-NH and H-8.
Similarly, the trans position of the two carbo-
methoxy groups was con®rmed by mutual NOE's
stated for the methoxy hydrogens and the vicinal H-
40 or H-50 atoms in the pyrazoline/dine ring .e.g. t-2a,
see Table 4).
The cis or trans orientation of H-30 and H-40 .the
ferrocenyl moiety and the vicinal carbomethoxy
group) follows from the signi®cantly different H-30
and H-50 chemical shifts of the isomers. The H-3
atom is more shielded in the trans isomers .by 0.36
and 0.54 ppm for the c-t pairs of 2a,c), due to the
anisotropic shielding effect of the ester group. [4]
The similar shielding of the Fc-moiety was observed
on H-50 in the cis-isomers .0.45 and 0.49 ppm fo0r the
c-t pairs of 2a and 2c). The t- or c-position of H-3 and
H-40 is also re¯ected from the chemical shifts of the
methoxy±hydrogens: while both signals have similar
shifts in the trans-isomers .about 3.75 ppm), in their
cis-counterparts one line is signi®cantly up®eld
shifted .by ca 0.4 ppm±3.36 ppm) again due to the
anisotropic shielding effect of the vicinal Fc-group.
The assignment of the up®eld shifted OMe-signal to
the group in Pos. 4 was proved by DNOE experiment
.cf. Table 4) for c-2a. The more crowded steric struc-
The reactivity of ferrocenylhydrazones of the inves-
tigated heteroarylhydrazines .1a±e, 3) in 1,3-dipolar
cycloaddition reaction with DMFM in MeCN highly
depends on the N-heteroaryl group and the employed
temperature. Besides the thieno[3,2-c]pyrid-4-yl deri-
vative .3) of outstanding reactivity only the phtha-
lazon-4-yl- and the pyrido[3,4-d]pyridazon-8-yl
derivatives .1a,c) can be converted into the corre-
sponding 3-ferrocenyl-1-heteroaryl-pyrazole with
different oxidation levels .t-2a,c, c-2a,c, 5a,c, 6, 8).
According to comparative semi-empirical calcula-
tions .AM1), which give plausible reason for the
observed substrate selectivity, the reactive inter-
mediate is formed by a tautomerization step followed
by the actual stereoselective cycloaddition .t-2a/c-2a
and t-2c/c-2c @ 1) associated with 1,4-proton migra-
tion. The transformation of N-.4-bromo-phenyl)hy-
drazone 10 with DMFM takes place in the expected
way resulting in pyrazoline .11) and/or pyrazole .12)
derivatives.
Under the conditions employed for cycloadditions
the highly reactive N-thienopyridylhydrazone 3 partly
undergoes oxidative cyclization in competitive
manner to yield condensed 1,2,4-triazole 7, too.
Similar transformation of the pyrido[2,3-d]pyridazo-
nylhydrazones 2b,e which are completely resistant to
cycloaddition, can be effected by DDQ in CH2Cl2 at

Â
A. Abran et al. / Journal of Molecular Structure 569 92001) 185±194
194
Â
Vanko, R.E. Bozak, J. Mol. Struct. 527 .2000) 297±304 .Part
room temperature to obtain condensed triazoles 4b,e
in good yields.
8).
Â
[2] A. Abran, A. Csampai, Zs. Bocskei, P. Sohar, Tetrahedron 55
Â
Â
Â
È
.1999) 441±448 .Part 7 and references therein).
Â
Â
Â
Â
[3] A. Abran, A. Csampai, V. Harmath, P. Sohar, Models in Chem-
istry, Acta Chim. Hung. 135 .1998) 439±447 .The preparation
of 1a,b,e) .
Acknowledgements
The authors are indebted to Dr H. Medzihradszky-
Â
Â
[4] P. Sohar, Nuclear Magnetic Resonance Spectroscopy, 1, CRC
Schweiger for analyses, Mr S. Palagyi for technical
Press, Boca Raton, 1983 .pp. 31±34).
Â
[5] P. Sohar, Nuclear Magnetic Resonance Spectroscopy, 2, CRC
assistance and Mrs M. Barna for help in preparation of
the manuscript. The ®nancial support .OTKA T-
29651) is gratefully acknowledged.
Press, Boca Raton, 1983 .pp. 154±155).
[6] D.M. Grant, B.V. Cheney, J. Am. Chem. Soc. 89 .1967) 5315±
5319.
Â
[7] J.S. Nyu, W.L. Christopher, J.P. Yevich, R. Butler, R.F.
Schlemmer, J. Med. Chem. 32 .6) .1989) 1147±1156.
Â
[8] K. Kormendy, T. Kovacs, F. Ruff, I. Kovesdi, Acta Chim.
References
È
È
Hung. 112 .4) .1983) 487±499.
  Â
È
[1] P. Sohar, P. Perjesi, K.W. Tornroos, S. Husebye, A. Vertes, Gy.