Stable 3-imino-2,3-dihydroindazol-1-yl radicals

Article abstract of DOI:10.1002/ejoc.201001659

Oxidation of 2-tert-butyl-3-(tert-butylamino)-2H-indazoles with lead(IV) oxide yields red, stable radicals, which can be distilled under high vacuum and recrystallized. They revert to their precursors on Pd-catalysed hydrogenation. Configuration and atomi

Full text of DOI:10.1002/ejoc.201001659

FULL PAPER
DOI: 10.1002/ejoc.201001659
Stable 3-Imino-2,3-dihydroindazol-1-yl Radicals
Helmut Quast,*[a][‡] Gottfried Philipp,[a] Karl-Heinz Ross,[a] and Ulrich M. Doht[a]
Dedicated to Professor Siegfried Hünig on the occasion of his 90th birthday
Keywords: EPR spectroscopy / Hydrogenation / Nitrogen heterocycles / Oxidation / Radicals
Oxidation of 2-tert-butyl-3-(tert-butylamino)-2H-indazoles
with lead(IV) oxide yields red, stable radicals, which can be
distilled under high vacuum and recrystallized. They revert
to their precursors on Pd-catalysed hydrogenation. Configu-
ration and atomic distances were obtained by X-ray diffrac-
tion analysis. EPR hfc constants were assigned with the help
of 15N labeling and exchange of a phenyl hydrogen atom for
a tert-butyl group. The X-ray and EPR results show that the
novel radicals are to be considered as cyclic hydrazyl radi-
cals, modified by delocalization of the unpaired electron in-
volving the exocyclic imino group and the benzene ring.
Introduction
Since more than one century, stable radicals play impor-
tant roles in may areas of organic chemistry.[1–3] So far, all
stable N-radicals devoid of heteroatoms other that nitrogen
belong to the families[3,4] of aminyl (R2N·)[5] and hydrazyl
radicals (R2N–N·–R) (both types being acyclic or having
the radical moiety built into a cyclic skeleton like phen-
azinyl, 1,2,4-triazolinyl, and 1,4-dihydrobenzo-1,2,4-triazin-
yl radicals), and symmetric, resonance-delocalized hydraz-
idinyl radicals, viz., tetraazapentenyl (R2N–N·–CR=N–
NR2), tetrazolinyl, and verdazyl radicals.[6] In the general
framework of Hünig’s two-step reversible redox systems,[7]
the intermediate neutral radical R2N–(CH=CH)n–N·–R be-
tween the corresponding reduced and oxidized forms is ex-
emplified by hydrazyl (n = 0) and several heterocyclic N-
radicals, e.g., phenazinyl (n = 1) and 1,4-dihydrobenzo-
1,2,4-triazinyl radicals (n = 1). A number of N-radicals are
known whose stabilization by one or more structural fea-
tures, viz., steric protection, spin delocalization, nitrogen
atoms as carriers of spin densities, and charge distribution
by electronic effects of substituents,[3] permitted characteri-
zation by X-ray crystallography.[8] Here we report on a
novel type of such N-radicals, the (E)-3-imino-2,3-dihy-
droindazol-1-yl radicals 1.
Results and Discussion
A host of 3-amino-2H-indazoles are known.[9] However,
none of them has been considered as source of a radical.
In reactions conceived as syntheses of aryl-substituted 3-
iminodiaziridines, we obtained, surprisingly, blue fluores-
cent 3-amino-2H-indazoles 7 as main products besides
small amounts of isomeric 3-(arylimino)diaziridines 3 and
2-aminobenzimidazoles 6.[10] 1,3-Elimination of sulfuric
acid from N-hydroxyguanidine O-sulfonic acids 2 under the
influence of strong bases afforded red mixtures of isomers
3, 6, 7. Likewise, chlorination/1,3-dehydrochlorination at
low temperature of guanidine 5 gave a red mixture con-
sisting mainly of 3a and 7a (Scheme 1). The same red color
was observed on heating of solutions of 3a under rigorous
exclusion of molecular oxygen.[11] As 3-iminodiaziridines
are weak oxidants,[12] we suspected that the red color was
due to oxidation of small amounts of one of the products
6a or 7a. Indeed, oxidation of 7a in benzene solution with
lead(IV) oxide or silver oxide[5,6] gave a red, under high vac-
uum distillable oil (75% yield), which slowly crystallized to
afford deep-red crystals (1a), m.p. 39–40 °C. In a similar
experiment, 7b furnished a 78% yield of black-red crystals
(1b), m.p. 102 °C after recrystallization from acetonitrile.
Whereas 1a was not completely stable but slowly formed a
blue-violet EPR-silent dimer when kept at room tempera-
[a] Institut für Organische Chemie der Universität Würzburg
Am Hubland, 97074 Würzburg, Germany
[‡] Current address: Hoetgerstrasse 10, 49080 Osnabrück,
Germany
Fax: +49-541/8141935
E-mail: hquast@uni-osnabrueck.de
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201001659.
Eur. J. Org. Chem. 2011, 1891–1895
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1891

H. Quast, G. Philipp, K.-H. Ross, U. M. Doht
FULL PAPER
ture, the crystals of 1b survived for years, obviously owing
On palladium-catalysed hydrogenation, stable N-radicals
to the enhanced steric protection by the additional tert- like 1,2,4-triazolinyl,[8d] 1,4-dihydro-1,2,4-benzotriazinyl,[14]
butyl group at the benzene ring.[5] We did not observe reac- tetraazapentenyl, and verdazyl radicals,[6] rapidly consume
tions with molecular oxygen or moisture, a reluctance not half a mol of hydrogen to afford the corresponding reduced
often found among radicals.
radical precursors. Similarly, hydrogenation of 1a consumed
the calculated amount of hydrogen and produced 7a in
quantitative yield, thus providing unequivocal structural
evidence.
The configuration of 1b in the solid state was revealed
by X-ray diffraction analysis (Figure 2).[15] As expected on
steric grounds, 1b adopts the more favorable (E) configura-
tion. More important, the atomic distances provided the
clue to the electronic structure and thus supported the as-
signment of the nitrogen hyperfine coupling (hfc) constants
(see below): the length of the bond to the exocyclic nitrogen
atom amounts to only 128.7 pm. This value is very close to
the length of a normal carbon–nitrogen double bond.[16,17]
The distances (N-1)–(C-7a) and (C-3)–(C-3a) correspond to
normal single bonds between sp2 atoms.[16] Consequently,
the new radical is best represented by the two hydrazyl reso-
nance structures 1 and 1Ј with an only small contribution
by the ortho-quinonoid 1,5-diazapentadienyl resonance
structure 1ЈЈ. The hydrazyl character of 1 is confirmed by
the length of the N–N bond, which is exactly as long as in
1,2,4-triazolinyl 8[8d] and very close to the length of the N–
N bond in Goldschmidt’s radical 2,2-diphenyl-1-picryl-
hydrazyl (133.4 pm).[8a] These N–N distances correspond to
about 50% double bond character.
Scheme 1. Synthesis of the stable N-radicals 1 and hydrogenation
of 1a. (a) 40% KOH in methanol/petroleum ether, yield 7a 74%,
7b 51%; (b) (1) tBuOK, diethyl ether, –78 °C, (2) tBuOCl, (3) Ǟ
20 °C, yield 7a 42–50%; (c) PbO2, K2CO3, benzene, 25 °C, 15 min,
yield 1a 75%, 1b 78%; (d) PdCl2/CaCO3, DMF, 25 °C, 10 min, H2
(735 Torr), yield 7a quant.
The radical structures 1 of the deeply colored oxidation
products of 7 were based on elemental analyses and spec-
troscopic evidence: mass spectra, lack of NH bands in IR
spectra, UV/Vis spectra, strong EPR signals and, of course,
the failure to record NMR spectra. The UV/Vis spectra
show a broad, structured long-wavelength absorption in the
visible region (Figure 1), which is more pronounced than in
the case of 1,2,4-triazolinyl[13] and 1,4-dihydro-1,2,4-benzo-
triazinyl[14] radicals.
Figure 2. Configuration of 1b in the solid state and selected atomic
distances [pm] as calculated from the data reported in ref.[15] The
complete set of atomic distances is given in the Supporting Infor-
mation.
Owing to significant line broadening, the EPR spectrum
recorded for a solution of 1a in benzene shows only 58 lines
instead of 432, calculated for three different nitrogen and
four different hydrogen atoms (Figure 3). The line broaden-
ing results from unresolved couplings to the 18 hydrogen
atoms of the tert-butyl groups. This is demonstrated by the
EPR spectrum of [D18]-1a (Figure 4). Perdeuteration of
tert-alkyl groups has been used already for a long time in
radical chemistry to improve the resolution of EPR spec-
tra.[18]
Evaluation of the X-band EPR spectra by using standard
techniques afforded the hfc constants compiled in Table 1.
To aid the assignment, position 1 of 7a was labeled with
15N to afford radical [15N]-1a and the EPR spectrum of
Figure 5. In addition, the tert-butyl group at the benzene
Figure 1. UV/Vis spectrum recorded for a solution of 1b in heptane. ring of 1b simplified the EPR spectrum of 1a by eliminating
1892
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1891–1895

Stable 3-Imino-2,3-dihydroindazol-1-yl Radicals
Figure 3. First derivative of the EPR spectrum recorded for a solu-
tion of 1a in benzene.
Figure 5. First derivative of the EPR spectrum recorded for a solu-
tion of [15N]-1a in benzene (bottom) and spectrum simulated by
using the hfc constants listed in Table 1 (top).
Figure 4. First derivative of the EPR spectrum recorded for a solu-
tion of [D18]-1a in benzene.
Figure 6. First derivative of the EPR spectrum recorded for a solu-
tion of 1b in benzene (g = 2.0035, bottom) and spectrum simulated
by using the hfc constants listed in Table 1 (top).
The hydrazyl character of 1, as indicated by the X-ray
diffraction results, suggested a comparison of 1 with related
heterocyclic radicals, viz., Neugebauer’s 1,2,4-triazolinyl
8[8c,13] and Blatter’s radical 9.[14,19] The size of the g factor
of 1b (2.0035) falls into the range of the g factors of these
two radicals. In both radicals, the largest hfc constants mea-
sure the coupling to the threefold coordinated nitrogen
atoms (N-1) of the hydrazyl moieties (Scheme 2). By analo-
gy, we assign the largest hfc constants in the EPR spectra
of 1 to the analogous nitrogen atoms (N-2) of the indazole
rings. Obviously, the polar resonance structure 1Ј contrib-
utes strongly to the delocalized system. The smallest nitro-
gen hfc constant is attributed to the exocyclic nitrogen
atom, in accord with the minor importance of resonance
structure 1ЈЈ as indicated by the short C=N distance. The
π-SOMO of 1 is mainly localized on the pyrazole ring and
the coupling to 5-H (Ǟ EPR spectrum of Figure 6) and
thus proved the assignment of the corresponding hfc con-
stant a(5-H).
Table 1. Hyperfine coupling constants a [Gauss] obtained from
EPR spectra recorded for solutions of 1 in benzene.
1a (Figure 3)
[15N]-1a (Figure 5)
1b (Figure 6)
a(N-1)
a(N-2)
a(=N)
6.24
6.49
3.32
0.75
2.71
0.61
2.01
8.86
6.56
3.36
0.74
2.62
0.64
2.07
6.34
6.41
3.24
0.75
–
0.63
2.03
a(4-H)[a]
a(5-H)
a(6-H)[a]
a(7-H)
[a] The assignment may be exchanged.
Eur. J. Org. Chem. 2011, 1891–1895
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1893

H. Quast, G. Philipp, K.-H. Ross, U. M. Doht
FULL PAPER
(4.28), 246 (4.24), 229 (4.10) nm. EI MS (70 eV): m/z (%) = 301
(1), 300 (7), 299 (4). C19H30N3 (300.5): calcd. C 75.95, H 10.06, N
13.98; found C 74.98, H 10.19, N 13.94.
less on the exocyclic nitrogen atom. However, the hfc con-
stants a(5-H) and a(7-H), being larger than the correspond-
ing constants for 9 [a(5-H) = –1.30 G, a(7-H) = –1.79 G],[14]
indicate delocalization of the unpaired electron into the
benzene ring.
Hydrogenation of 1a: A vigorously stirred suspension of PdCl2 on
CaCO3 (2.6%, 200 mg) in dry dimethylformamide (DMF) was hy-
drogenated in
a quantitative micro hydrogenation apparatus
(NORMAG, Hofheim am Taunus, Germany). A solution of 1a
(246 mg, 1.005 mmol) in DMF (15 mL) was added, followed by
stirring until the consumption of H2 was complete (10 min). The
volume of of H2 consumed at 735 Torr was 12.5 mL (0.495 mmol,
99%). The catalyst was removed by filtration and the solvent dis-
tilled at 10–3 Torr to afford a residue, which was identical with 7a
(1H NMR).
Scheme 2. Reported atomic distances of 8[8d] and nitrogen hfc con-
stants [Gauss] and g factors for 8[13] and 9.[14]
EPR Spectra: EPR quartz sample tubes were charged with
weighted amounts of 1 and attached to a vacuum line (10–5 Torr).
Solvents were stored over LiAlH4, degassed by several freeze-
pump-thaw cycles, and condensed into the sample tubes, which
were sealed under a vacuum of 10–5 Torr.
Conclusions
The present study uncovers the rare case of a novel type
of stable N-radicals. The bulky substituents of the two ex-
amples 1a and 1b certainly contribute to their exceptional
stability. However, numerous, perhaps somewhat less stable
N-radicals of this type may be conceived and, probably, pre-
pared from the host of known 3-amino-2H-indazoles.
Supporting Information (see footnote on the first page of this arti-
cle): General experimental and analysis of EPR spectra (1 page),
syntheses of [D18]-7a and [15N]-7a (2 pages), UV/Vis spectra of 1a
and 7b (2 pages), IR spectra of 1 and 7 (2 pages), EPR spectra of
1 in various solvents (3 pages), atomic distances of 1b (1 page).
Acknowledgments
Experimental Section
Our special thanks go to Dr. Dieter Scheutzow, University of
Würzburg, for his help in the computational analysis of the EPR
spectra. We express our gratitude to Dr. Gerda Lange for recording
the mass spectra and to Dr. Karl Peters, Max-Planck-Institut für
Festkörperforschung, Stuttgart, for calculating the atomic dis-
tances of 1b. Financial support by the Deutsche Forschungsge-
meinschaft and the Fonds der Chemischen Industrie, Frankfurt am
Main, is gratefully acknowledged.
2-(tert-Butyl)-3-(tert-butylimino)-2,3-dihydro-1H-indazol-1-yls
1. General Procedure: A mixture of 3-amino-2H-indazole 7
(10 mmol), PbO2 (24 g, 0.10 mol) and K2CO3 (24 g, 0.17 mol) in
dry benzene (250 mL) was stirred at room temp. for 15 min. The
solid material was removed by filtration, followed by distillation of
the solvent under vacuum. The remaining dark red oil was distilled
at room temp./5ϫ10–6 Torr in a sublimation apparatus, whose cold
finger was kept at –30 °C, to afford a deep-red oil, which crys-
tallized under N2 overnight.
[1] A pragmatic distinction between persistent, stabilized, and
stable radicals has been proposed: D. Griller, K. U. Ingold,
Acc. Chem. Res. 1976, 9, 13–19.
2-(tert-Butyl)-3-(tert-butylimino)-2,3-dihydro-1H-indazol-1-yl (1a):
Dark red, shining crystals (1.83 g, 75 %, m.p. 39–40 °C), which
formed a blue-violet EPR-silent dimer at room temp. during 30 d.
[2] R. G. Hicks (Ed.), Stable Radicals: Fundamentals and Applied
Aspects of Odd-Electron Compounds, Wiley-VCH, Weinheim,
2010.
IR (neat oil): ν = 1665 (w), 1595 (m) cm–1. UV/Vis (hexane): λ
˜
max
(logε) = 555 (3.18), 515 (3.20), 478 (3.14), 394 (3.84), 376 (3.83),
337 (3.52), 296 (3.33), 268 (sh, 3.85), 249 (4.20), 241 (4.22) nm. EI
MS (70 eV): m/z (%) = 245 (12) [(M + 1)+], 244 (8) [M+], 229 (1),
189 (10), 188 (26), 174 (12), 173 (16), 158 (10), 133 (85), 132 (100),
104 (18), 103 (23). C15H22N3 (244.4): calcd. C 73.73, H 9.08, N
17.20; found C 73.80, H 9.18, N 17.42.
[3] Review: R. G. Hicks, Org. Biomol. Chem. 2007, 5, 1321–1338.
[4] Ref.[2], chapter 7.
[5] Review: W. C. Danen, F. A. Neugebauer, Angew. Chem. 1975,
87, 823–830; Angew. Chem. Int. Ed. Engl. 1975, 14, 783–789.
[6] Review: F. A. Neugebauer, Angew. Chem. 1973, 85, 485–493;
Angew. Chem. Int. Ed. Engl. 1973, 12, 455–464.
[7] Reviews: a) K. Deuchert, S. Hünig, Angew. Chem. 1978, 90,
927–938; Angew. Chem. Int. Ed. Engl. 1978, 17, 875–886; b) S.
Hünig, H. Berneth, Top. Curr. Chem. 1980, 92, 1–44.
[8] For X-ray diffraction studies of stable N-radicals, see: a) D. E.
Williams, J. Am. Chem. Soc. 1967, 89, 4280–4287; b) D. E. Wil-
liams, J. Am. Chem. Soc. 1969, 91, 1243–1245; c) D. E. Wil-
liams, Acta Crystallogr., Sect. B 1973, 29, 96–102; d) F. A. Neu-
gebauer, H. Fischer, C. Krieger, Angew. Chem. 1989, 101, 486–
488; Angew. Chem. Int. Ed. Engl. 1989, 28, 491–492; e) T.
Barth, B. Kanellakopulos, C. Krieger, F. A. Neugebauer, J.
Chem. Soc., Chem. Commun. 1993, 1626–1628; f) M. R.
Gleiter, R. K. Kremer, C. Krieger, F. A. Neugebauer, Liebigs
Ann. 1996, 1867–1870; g) M. R. Gleiter, B. Kanellakopulos, C.
Krieger, F. A. Neugebauer, Liebigs Ann./Recueil 1997, 473–483;
h) R. G. Hicks, M. T. Lemaire, L. Öhrström, J. F. Richardson,
L. K. Thompson, Z. Xu, J. Am. Chem. Soc. 2001, 123, 7154–
7159; i) B. D. Koivisto, A. S. Ichimura, R. McDonald, M. T.
Lemaire, L. K. Thompson, R. G. Hicks, J. Am. Chem. Soc.
Dimer of 1a: 1a was kept in a sublimation apparatus at room temp.
for 30 d. The volatile components (small amounts of a red oil and
a white solid) were removed by sublimation at room temp./
10–5 Torr. The residue was suspended in petroleum ether to afford
dark blue-violet crystals. UV/Vis (hexane): λmax (logε) = 631 (3.46),
600 (sh, 3.42), 583 (3.44), 406 (3.65), 392 (3.58), 383 (3.62), 320 (sh,
3.92), 305 (sh, 4.01), 285 (4.16), 214 (4.61) nm. EI MS (70 eV): m/z
(%) = 488 (14) [M+], 432 (4), 431 (23), 375 (1), 319 (2), 263 (3), 251
(1), 245 (13), 189 (3), 174 (4), 133 (100).
2,5-Bis(tert-butyl)-3-(tert-butylimino)-2,3-dihydro-1H-indazol-1-yl
(1b): Black-red shining cubes (2.34 g, 78%, m.p. 102 °C, from ace-
tonitrile), which survived unchanged for years under ambient con-
ditions. IR (KBr): ν = 1596 (w) cm–1. UV/Vis (heptane): λmax (logε)
˜
= 567 (3.191), 545 (sh, 3.212), 530 (3.262), 518 (3.264), 474 (3.212),
394 (3.81), 379 (3.77), 338 (3.54), 298 (3.36), 270 (sh, 3.89), 253
1894
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1891–1895

Stable 3-Imino-2,3-dihydroindazol-1-yl Radicals
2006, 128, 690–691; j) S. V. Rosokha, J. Zhang, J. Lu, J. K. Ko-
chi, J. Phys. Org. Chem. 2010, 23, 395–399.
[9] a) A. Schmidt, A. Beutler, B. Snovydovych, Eur. J. Org. Chem.
2008, 4073–4095, and references cited therein; b) W.
Stadlbauer, Sci. Synth. 2002, 12, 227; c) H. Quast, A. Fuß, U.
Nahr, Chem. Ber. 1985, 118, 2164–2185.
[10] H. Quast, K.-H. Ross, G. Philipp, M. Hagedorn, H. Hahn, K.
Banert, Eur. J. Org. Chem. 2009, 3940–3952.
[11] H. Quast, K.-H. Ross, G. Philipp, Eur. J. Org. Chem. 2010,
2212–2217.
[16]
[17]
M. B. Smith, J. March, March’s Advanced Organic Chemistry,
6th ed., Wiley, Hoboken, New Jersey, USA, 2007, chapter 1.
See, for example, the C=N bond in N-(4-chlorobenzylidene)-1-
adamantylamine (126.6 pm): G. L. Zhao, Y.-L. Feng, Z. Kris-
tallogr. New Cryst. Struct. 2005, 220, 197–198.
[18] a) R. J. Holman, M. J. Perkins, J. Chem. Soc. C 1971, 2324–
2326; b) R. Chiarelli, A. Rassat, Tetrahedron 1973, 29, 3639–
3647; c) D. L. Haire, E. G. Janzen, Magn. Reson. Chem. 1994,
32, 151–157; d) S. R. Burks, J. Bakhshai, M. A. Makowsky, S.
Muralidharan, P. Tsai, G. M. Rosen, J. P. Y. Kao, J. Org. Chem.
2010, 75, 6463–6467.
[19] a) H. M. Blatter, H. Lukaszewski, Tetrahedron Lett. 1968, 9,
2701–2705; b) T. L. Gilchrist, C. J. Harris, C. W. Rees, J. Chem.
Soc., Chem. Commun. 1974, 485–486; c) P. A. Koutentis, D.
Lo Re, Synthesis 2010, 2075–2079.
[12] H. G. Aurich, H.-G. Scharpenberg, Chem. Ber. 1973, 106,
1881–1896.
[13] F. A. Neugebauer, H. Fischer, Tetrahedron 1995, 51, 12883–
12898.
[14] F. A. Neugebauer, I. Umminger, Chem. Ber. 1980, 113, 1205–
1225.
Received: December 10, 2010
Published Online: February 4, 2011
[15] Atomic coordinates and displacement parameters have already
been published: K. Peters, E.-M. Peters, G. Philipp, H. Quast,
Z. Kristallogr. New Cryst. Struct. 2000, 215, 385–386.
Eur. J. Org. Chem. 2011, 1891–1895
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1895