The EPR D-parameter of 1,3-diarylcyclopentane-1,3-diyl triplet diradicals as a probe for steric substituent effects in benzyl-type radicals

Article abstract of DOI:10.1039/a905773k

The zero-field splitting D parameter of a set of ortho-substituted-1,3-diarylcyclopentane-1,3-diyl triplet diradicals 3 has been determined by EPR spectroscopy in a 2-methyltetrahydrofuran glass matrix at 77 K. While for meta- and para-substituted triplet

Full text of DOI:10.1039/a905773k

View Article Online / Journal Homepage / Table of Contents for this issue
The EPR D-parameter of 1,3-diarylcyclopentane-1,3-diyl triplet
diradicals as a probe for steric substituent effects in benzyl-type
radicals†
Waldemar Adam,*a Claus van Barneveld,*a Jan-Stefan Gerkeb and Frank Gerrit Klärnerb
a Institute of Organic Chemistry, University of Würzburg, Am Hubland, D-97074 Würzburg,
Germany. E-mail: adam@chemie.uni-wuerzburg.de; Fax: 0931-888-4756;
Internet: http://www-organik.chemie.uni-wuerzburg.de
b Fachbereich 8—Chemie, University of Essen, Universitätsstraße 5, D-45117 Essen, Germany
Received (in Cambridge, UK) 19th July 1999, Accepted 22nd September 1999
The zero-field splitting D parameter of a set of ortho-substituted 1,3-diarylcyclopentane-1,3-diyl triplet diradicals 3
has been determined by EPR spectroscopy in a 2-methyltetrahydrofuran glass matrix at 77 K. While for meta- and
para-substituted triplet diradicals 3 the D parameter is only a function of the electronic nature (spin donor versus
spin acceptor) of the substituent, for the ortho-substituted derivatives the steric influence must also be considered.
We show herein that the steric effect of the ortho substituent diminishes delocalization of the cumyl spin density
into the aromatic ring through twisting of the π system out of conjugation with the cumyl radical center; indeed,
the steric effect may outweigh the electronic influence of the substituent. The computed (PM3) spin density of the
minimum-energy conformer reproduces well the experimentally observed steric effect on the D parameter of the
triplet diradicals 3. These results demonstrate that the EPR-spectroscopic D parameter serves as a sensitive probe
for steric as well as electronic substituent effects.
Introduction
The zero-field splitting parameter D of the localized 1,3-
diaryl-substituted triplet diradicals 3, which are accurately
2
3µ0g2µB ρAρ
determined by EPR spectroscopy under matrix isolation (2-
B = const. ρA
(1)
D =
3
methyltetrahydrofuran (MTHF), 77 K), provide important
information on the electronic properties of such high-spin
systems.1 The D value, which is ca. 0.05 cmϪ1 for these triplet
species, derives from the dipole–dipole interaction between the
two unpaired spins and reflects the electronic nature of the
diradical. It depends on the spin densities ρB and ρA at the two
radical sites and the distance dAB between the two radical cen-
ters,2 which for the cyclopentanediyl triplet diradicals 3 is ca.
238 pm [eqn. (1)].3 The spin densities ρB and ρA vary with the
spin-delocalizing properties of the substituted aryl groups, such
that the change of the D parameter in the triplet diradicals 3
reflects the nature and efficacy of the substituent on the aryl
group in interacting with the cumyl spin moiety. While for meta
and para substituents only electronic effects operate, for ortho
substituents additional steric effects are expected. Depending
on the spin-accepting (SA) or spin-donating (SD) ability of the
substituent, electronic effects either lower (SA) or raise (SD)
16π
dAB
the D value;4 for the steric influence, D should always increase.
This is due to twisting of the cumyl spin center out of con-
jugation with the aromatic π system, thereby hindering the
delocalization of spin; in other words, the spin is more effect-
ively localized at the cumyl site.
In this work we have examined a selected set of ortho-
substituted triplet diradicals 3 derived from azoalkanes 1 by
photodenitrogenation in MTHF matrix at 77 K to assess the
efficacy of steric effects of ortho substitution in the triplet
diradicals 3. As displayed by the results herein, such steric
effects may outweigh electronic ones.
Results
Syntheses
The substituted 4H-pyrazoles 6 were prepared in analogy to
reported procedures (Scheme 1).5 Due to the steric hindrance of
the ortho substituents, the usual synthetic route to the azo-
alkanes 1 failed, namely the acid-catalyzed cycloaddition of the
dienophile to the 4H-pyrazoles 6. High pressure (12.5 kbar) and
elevated temperature (у 100 ЊC) were necessary to force the
cycloaddition with cyclopentene as the dienophile. The yields
† Computational details such as heats of formation of the diradicals 3
and the spin densities of the cumyl position of the radicals 2 as a
function of the torsional angle θ are available as supplementary data.
For direct electronic access see http://www.rsc.org/suppdata/p2/1999/
2723, otherwise available from BLDSC (SUPPL. NO. 57657, pp. 2) or the
RSC Library. See Instructions for Authors available via the RSC web
page (http://www.rsc.org/authors).
J. Chem. Soc., Perkin Trans. 2, 1999, 2723–2728
This journal is © The Royal Society of Chemistry 1999
2723

View Article Online
Table 1 D values and torsional angles (θ) of ortho-substituted diradi-
cals 3
a
b
c
d
e
Diradical
Substituent
Dortho
θA
θB
∆θB
Dpara
3a
3b
3c
3d
3e
3f
H
F
Cl
Br
CH3
OCH3
5.06
5.14
5.07
5.13
0
—
3
14
15
—
50
51
55
71
73
—
0
1
5
21
23
—
5.06
5.14
5.01
5.03
5.05
5.08
5.12
f
a |D/hc|ؒ10Ϫ2 in cmϪ1, measured in a 2-MTHF matrix at 77 K,
error 0.00002 cmϪ1; |E/hc| < 0.002 cmϪ1 b Torsional angle calculated
.
according to method A (see text). c Torsion angle calculated according
to method B (see text). d Chance of the torsional angle θB relative to the
parent system 3a. e Ref. 4b. f No EPR signal detected.
Fig. 1 Calculated (PM3) cumyl spin density (ρ) of the radical 2a versus
the torsional angle (θ).
Fig. 2 Possible steric interactions in the cumyl radical 2 and the triplet
diradical 3.
tion of the θ angle. For the coplanar conformation (θ = 0Њ), the
cumyl spin density is lowest (maximum delocalization), while
for the perpendicular conformation (θ = 90Њ) it is highest (min-
imum delocalization). Since according to eqn. (1) the D value of
the triplet diradical 3 is an experimental measure of the spin
density for the cumyl radical fragment 2,3 the torsional angle θ
is readily estimated from Fig. 1 by extrapolation. These values
are given as θA (method A) in Table 1 for the ortho-substituted
triplet diradicals 3. As expected, the torsional angle θ increases
with the effective size9 of the ortho groups in the order
H ~ F < Cl < Br ~ CH3, except for the ortho-fluoro substituent
(cf. Discussion).
Scheme 1 Synthesis of azoalkanes 1. (i) NaNH2, THF, ca. 20 ЊC,
1–2 d; (ii) K2CO3, (n-Bu)4NBr, toluene, ca. 80 ЊC, 2 d; (iii) N2H4ؒH2O,
CHCl3, 60 ЊC, 1 d; (iv) cyclopentene, toluene, 110 ЊC, 12.5 kbar, 2 d.
of the azoalkanes 1 were low because at ca. 100 ЊC much
denitrogenation to the corresponding housane took place, while
at ca. 80 ЊC no cycloaddition occurred.
EPR Spectroscopy
An alternative, but computationally more elaborate task to
estimate the torsional angle (θ) is to calculate the minimum-
energy conformation of the aryl substituent at the radical
center. It will not suffice to consider just the cumyl-radical
fragment 2 for this purpose, as becomes apparent when its
steric interactions are compared with those in the actual triplet
diradicals 3 (Fig. 2). In addition to the steric repulsion between
the ortho substituent and the cumyl hydrogen atoms (inter-
action a in 2), there exist effective repulsions b in the diradical 3,
whose compromise will result in the final minimum-energy
conformation. Hence, the heats of formation for the set of
ortho-substituted triplet diradicals 3 need to be calculated as a
function of the torsional angle to assess the minimum-energy
conformation. For this purpose, first the energy change along
the reaction coordinate for the rotation of the substituted
phenyl ring was calculated by the PM3-AHUF/CI method as
described above to acquire the minimum-energy conformer,
which was further optimized by an additional UHF/6-31G*
calculation.10 These torsional angles θB (method B) of the
ortho-substituted diradicals are given in Table 1.
Except for the derivative 1f, the photolysis of the azoalkanes
1 in 2-methyltetrahydrofuran (MTHF) glass at 77 K with the
364-nm line of an argon-ion laser led to the persistent triplet
diradicals 3. For the ortho-methoxy-substituted case no EPR
signal was observed. Analysis of the Z signals in the EPR
spectra (for a typical one, cf. Fig. 1 in ref. 6) afforded the D
parameter as one-half of the distance between the low- and the
high-field peaks; the E parameters were expectedly very small
(≤0.002 cmϪ1).
The EPR data are summarized in Table 1. As can be seen, all
ortho-substituted triplet diradicals 3 possess higher D values
compared to the parent system 3a. This implies that a higher
spin density is located at the cumyl position for the ortho-
substituted than for the parent triplet diradical and, hence,
there is less delocalization of spin into the aromatic ring. The
highest D value (0.0514) was obtained for the fluoro substituent
in 3b and the lowest one (0.0507) for the chloro substituent in
3c.
Determination of the torsional angle
The efficacy of spin delocalization by the aryl groups from the
cumyl position in the triplet diradicals 3 depends on the tor-
sional angle (θ) about the C–C bond that connects the two
structural units, as displayed in Fig. 1 for the parent system 2a.
The spin densities of the cumyl-radical fragments were calcu-
lated by the PM3-AUHF/CI semiempirical method7,8 as a func-
Discussion
Already for the parent diradical 3a, the torsional angle of 50Њ
indicates substantial twisting of the phenyl groups out of the
plane of the spin-carrying cyclopentanediyl ring. For the parent
cumyl radical 2a (Fig. 2, X = H), the lowest-energy conform-
2724
J. Chem. Soc., Perkin Trans. 2, 1999, 2723–2728

View Article Online
It has been demonstrated in this work that the steric demand
of an ortho substituent has a significant effect on the cumyl spin
density in cyclopentanediyl-type triplet diradicals 3 and, con-
sequently, also on the D parameter. When the steric size of the
substituent is below the threshold value (ca. 170 pm), e.g., H
and F, no steric effect is observed and the electronic influence of
such substituents on the D parameter is the same in the ortho
and para positions. For substituents with a steric size above this
threshold, i.e. Cl, Br and Me, the spin density at the cumyl site
is disproportionally increased through twisting out of conju-
gation and this steric effect surpasses the electronic one. These
ortho-substituted triplet diradicals show not only D values
higher than those of the para-substituted ones, but also higher
than the parent system 3a. For meta and para substituents no
additional steric effects operate in the triplet diradicals and,
thus, the cumyl radical fragment is a good model to account for
the observed electronic substituent effects on the cumyl spin
density; however, for ortho substituents, additional steric effects
apply, which makes it necessary to compute the spin densities
for the complete triplet diradical species to account for the
substituent effects on the experimental D parameter.
Fig. 3 Calculated torsional angles (∆θB) in diradicals 3 versus the
van der Waals radii of the ortho substituents.
ation is planar, i.e., the steric repulsion a between the methyl
and ortho hydrogens is insignificant and maximum delocaliz-
ation of the spin at the cumyl position into the phenyl ring is
possible. In contrast, for the rigid cyclopentanediyl triplet
diradical 3a (Fig. 2, X = H), this a interaction is significant and
the phenyl ring rotates out of planarity to relax this steric repul-
sion. However, during this rotation, the other ortho hydrogen
atom of the phenyl group encounters the geminal methyl
hydrogen atoms at the 2 position of the cyclopentanediyl ring
and this steric interaction b prevents further rotation. Thus, the
50Њ value for the θB angle (Table 1) of the parent cyclo-
pentanediyl triplet diradical 3a (Fig. 2, X = H) is the best com-
promise between the a and b repulsions. Relative to the parent
diradical 3a, the ∆θB values for the other derivatives 3b–e are
also given in Table 1 and allow comparison with the previously
acquired θA torsional angles for the cumyl radicals 2a–e.
As the data in Table 1 exhibit, the ∆θB values are slightly
higher than θA, but the same qualitative trend is obeyed, i.e.
H ~ F < Cl < Br ~ CH3, in which the steric effect of the ortho-
fluoro substituent is again as small as that of a hydrogen atom
and that of the ortho-bromo one as large as that of the ortho-
methyl group. A plot of the relative ∆θ values versus the van der
Waals radii9 of the ortho substituents (Fig. 3) reveals that at a
threshold value of ca. 170 pm, the steric effect of the ortho
substituent manifests itself by twisting the aryl moiety addi-
tionally out of planarity relative to the parent ortho-hydrogen-
substituted case 3a. This steric effect diminishes delocalization
of the spin at the cumyl position in the triplet diradicals 3.
The ortho-chloro atom is in its effective steric size just slightly
above this threshold value and, thus, diradical 3c (∆θ ca. 5Њ) is
essentially as nonplanar as the parent diradical 3a (∆θ 0Њ) and
the ortho-fluoro-substituted derivative 3b (∆θ ca. 1Њ). In con-
trast, the bromo- and methyl-substituted cases 3d and 3e are by
ca. 20Њ more twisted than the parent system 3a and thereby
significantly less delocalized. As a consequence, the more effect-
ive the steric effect of the ortho substituent, the more twisted is
the aryl group from planarity, the more localized the cumyl spin
density and, thus, the larger the D value, as expected from
eqn. (1). This trend may be experimentally assessed by com-
parison of the ortho and para substituents. For example, for the
fluoro substituent with a negligible steric effect in these
cyclopentanediyl triplet diradicals 3, only the electronic influ-
ence should operate. The same D values would be expected for
the ortho- and para-fluoro-substituted diradicals, as experi-
mentally confirmed by the identical 0.0514 cmϪ1 value (Table 1).
On the contrary, for the sterically demanding bromo and
methyl substituents, whose ortho regioisomers deviate substan-
tially more from planarity then the parent system (and also the
respective para-substituted derivatives), the D values for the
ortho isomers 3d and 3e are significantly larger than the para
ones, as experimentally substantiated (Table 1).
Experimental
General aspects
NMR spectra were recorded on a Bruker AC200 or AC250
instrument with CDCl3 as the solvent and internal standard.
J values are given in Hz. Infrared spectra were measured on a
Perkin-Elmer Infrared Ratio Recording Spectrometer 1420,
and UV spectra were run on a Hitachi U 3200 spectrometer.
Melting points were taken on a Büchi SMP-535 or B-545
apparatus, and the combustion analyses were performed by the
Microanalytical Division of the Institute of Inorganic Chem-
istry (University of Wuerzburg). Solvents and commercially
available chemicals were purified by standard procedures or
used as bought. Column chromatography was carried out on
silica gel (0.032–0.063 mm, Woelm) with an adsorbent:sub-
strate ratio of ca. 100:1. Thin layer chromatography (TLC) was
performed on a Polygram Sil G/UV254 (40 × 80 mm) instrument
from Macherey & Nagel. Irradiations were carried out with the
333-, 351-, and 364-nm UV lines (widened beam) of a CW
argon-ion laser (INNOVA 100, Coherent Co.).
Compounds 4c,11 4e,12 4f,13 5e14 and 5f15 are known in the
literature and were prepared according to the methods
described below.
Preparation of the 1,3-diarylpropane-1,3-diones (4)
Sodium amide (100 mmol) was suspended in 200 ml of dry
THF. The corresponding ortho-substituted methyl benzoates
(50 mmol) and acetophenone (50 mmol) were added under
cooling and the resulting mixture was stirred afterwards at ca.
20 ЊC for 16 h. The dark solution was poured on 100 g of
crushed ice and acidified with 85% H3PO4 (ca. 5 ml). The
product was extracted with dichloromethane (3 × 100 ml).
Drying with MgSO4 and evaporation of the solvent afforded
the crude propane-1,3-diones, which were purified by recrystal-
lisation from ethanol or distillation under reduced pressure.
1-(2Ј-Fluorophenyl)-3-phenylpropane-1,3-dione (4b). 63%,
yellow plates, mp 63–64 ЊC; νmax(KBr)/cmϪ1 3064, 1595, 1535,
1487, 1286, 1262, 1215; δH(250 MHz; CDCl3) 6.99 (s, 1H, enol
3
3
3
2-H), 7.16 (dd, JHF = 9.6, J = 8.2, 1H, 3Ј-H), 7.28 (t, J = 7.3,
1H, 5Ј-H), 7.45–7.57 (m, 4H, 4Ј-, 3Љ-, 4Љ-, 5Љ-H), 7.98–8.05
(m, 3H, 6Ј-, 2Љ-, 6Љ-H); δC(63 MHz; CDCl3) 97.8 (dd, 4JHF = 11,
2
2
C-2), 116.5 (dd, JHF = 23, C-3Ј), 123.7 (d, JHF = 11, C-1Ј),
124.5 (dd, 4JHF = 4, C-5Ј), 127.3 (d, C-2Љ), 128.6 (d, C-3Љ), 130.1
3
3
(dd, JHF = 2, C-6Ј), 132.6 (d, C-4Љ), 133.6 (dd, J = 9, C-4Ј),
1
3
135.3 (s, C-1Љ), 161.1 (d, JHF = 255, C-2Ј), 181.2 (d, JHF = 4,
J. Chem. Soc., Perkin Trans. 2, 1999, 2723–2728 2725

View Article Online
C-1), 186.4 (s, C-3). Anal. Calcd for C15H11FO2 (242.2): C,
74.38; H, 4.58. Found: C, 74.34; H, 4.73%.
cooling to ca. 20 ЊC and addition of MgSO4 (2–3 g), the sus-
pension was stirred for 10 min, the solid material removed by
filtration and the solvent was evaporated (40 ЊC, 10 Torr). The
crude product was recrystallized from benzene–cyclohexane
(1:1).
1-(2Ј-Bromophenyl)-3-phenylpropane-1,3-dione (4d). 42%,
yellow plates, mp 63–64 ЊC; νmax(KBr)/cmϪ1 3010, 2890, 1540,
1450, 1290, 1240, 1170, 1050, 760, 720; δH(200 MHz; CDCl3)
6.68 (s, 1H, 2-H), 7.26–7.71 (m, 7H, 3Ј-, 4Ј-, 5Ј-, 6Ј-, 3Љ-, 4Љ-,
5Љ-H), 7.98 (d, 3J = 8.3, 2H, 2Љ-, 6Љ-H); δC(63 MHz; CDCl3) 98.1
(s, C-2), 120.1 (s, C-2Ј), 127.2 (d, C-2Љ), 128.5 (d, C-5Ј), 128.6
(d, C-3Љ), 129.9 (d, C-6Ј), 131.6 (d, C-3Ј), 132.4 (d, C-4Љ), 133.6
(d, C-4Ј), 134.6 (s, C-1Љ), 138.3 (s, C-1Ј), 183.9 (s, C-1), 188.5 (s,
C-3). Anal. Calcd for C15H11BrO2 (303.2): C, 59.42; H, 3.66.
Found: C, 59.70; H, 3.69%.
3-(2Ј-Fluorophenyl)-4,4-dimethyl-5-phenyl-4H-pyrazole (6b).
69%, colorless powder, mp 150–151 ЊC; νmax(KBr)/cmϪ1 3040,
2980, 2940, 1590, 1520, 1460, 1420, 1310, 1220, 1160; δH(200
MHz; CDCl3) 1.55 (s, 6H, 4-CH3), 7.17–7.29 (m, 2H, 3Ј-, 5Ј-H),
7.42–7.52 (m, 4H, 4Ј-, 3Љ-, 4Љ-, 5Љ-H), 7.69 (m, 1H, 6Ј-H), 8.05
(m, 2H, 2Љ-, 6Љ-H); δC(63 MHz; CDCl3) 21.0 (q, C-4-CH3),
2
2
60.6 (s, C-4), 116.4 (dd, JCF = 23, C-3Ј), 118.9 (d, JCF = 15,
4
C-1Ј), 124.3 (dd, JCF = 4, C-5Ј), 127.9 (d, C-2Љ), 128.7 (d, C-
3Љ), 129.9 (s, C-1Љ), 130.9 (d, C-4Љ), 131.0 (dd, JCF = 4, C-6Ј),
131.9 (dd, JCF = 9, C-4Ј), 160.1 (d, JCF = 251, C-2Ј), 178.1
(d, JCF = 2, C-3), 178.8 (s, C-5). Anal. Calcd for C17H15FN2
3
Preparation of the 2,2-dimethyl-1,3-diarylpropane-1,3-diones (5)
3
1
Sodium hydride (80 mmol) was suspended in 100 ml of dry
toluene. Under cooling, a solution of the corresponding
diketone 4 (40 mmol) was added and the solution was stirred
for 30 min at ca. 20 ЊC. During a period of 1 h, methyl iodide
(120 mmol) was added and afterwards the suspension was
heated at about 80 ЊC for 24 h. After cooling to ca. 20 ЊC, the
solution was poured on 200 ml ice and extracted with CH2Cl2
(3 × 50 ml). The organic layer was washed with water (100 ml),
dried over MgSO4 and the solvent was evaporated (40 ЊC, 10
Torr). The crude product was purified by recrystallization from
cyclohexane or distillation under reduced pressure.
3
(266.3): C, 76.67; H, 5.68; N, 10.52. Found: C, 76.20; H, 6.07;
N, 10.05%.
3-(2Ј-Chlorophenyl)-4,4-dimethyl-5-phenyl-4H-pyrazole (6c).
73%, colorless powder, mp 143–144 ЊC; νmax(KBr)/cmϪ1 2981,
1594, 1518, 1491, 1458, 1343, 1070, 1034, 1002, 783; δH(200
MHz; CDCl3) 1.53 (s, 6H, 4-CH3), 7.27 (m, 1H, 5Ј-H), 7.33–
7.43 (m, 2H, 3Ј-, 4Ј-H), 7.45–7.56 (m, 4H, 6Ј-, 3Љ-, 4Љ-, 5Љ-H),
8.06 (m, 2H, 2Љ-, 6Љ-H); δC(50 MHz; CDCl3) 21.3 (q, C-4-CH3),
62.1 (s, C-4), 126.4 (d, C-5Ј), 127.9 (d, C-2Љ), 128.8 (d, C-3Љ),
130.0 (d, C-6Ј), 130.2 (s, C-1Љ), 130.3 (d, C-3Ј), 130.5 (s, C-1Ј),
130.6 (d, C-4Љ), 131.1 (d, C-4Ј), 134.8 (s, C-2Ј), 177.9 and 179.6
(s, C-3 and C-5). Anal. Calcd for C17H15ClN2 (282.8): C, 72.21;
H, 5.35; N, 9.91. Found: C, 71.85; H, 5.31; N, 9.70%.
1-(2Ј-Fluorophenyl)-2,2-dimethyl-3-phenylpropane-1,3-dione
(5b). 58%, colorless needles, mp 49–50 ЊC; νmax(KBr)/cmϪ1
3080, 2985, 2935, 1651, 1577, 1479, 1448, 1278, 1232, 717;
3
δH(200 MHz; CDCl3) 1.63 (s, 6H, 2-CH3), 6.97 (dd, JHF = 9.3,
3
3J = 8.2, 1H, 3Ј-H), 7.10 (t, J = 7.2, 1H, 5Ј-H), 7.31–7.47 (m,
3-(2Ј-Bromophenyl)-4,4-dimethyl-5-phenyl-4H-pyrazole (6d).
77%, colorless powder, mp 120–121 ЊC; νmax(KBr)/cmϪ1 3040,
2980, 2920, 1570, 1500, 1460, 1430, 1330, 1260, 1210; δH(250
MHz; CDCl3) 1.52 (s, 6H, 4-CH3), 7.24–7.42 (m, 3H, 3Ј-, 4Ј-,
5Ј-H), 7.45–7.53 (m, 3H, 3Љ-, 4Љ, 5Љ-H), 7.72 (m, 1H, 6Ј-H),
8.08 (m, 2H, 2Љ-, 6Љ-H); δC(63 MHz; CDCl3) 21.5 (q, C-4-
CH3), 61.8 (s, C-4), 122.9 (s, C-2Ј), 127.0 (d, C-5Ј), 127.8 (d,
C-2Љ), 128.9 (d, C-3Љ), 130.0 (s, C-1Љ), 130.6 (d, C-4Љ), 130.8
(d, C-6Ј), 133.0 (d, C-4Ј), 133.1 (s, C-3Ј), 134.1 (d, C-1Ј),
177.5 and 180.7 (2 × s, C-1 and C-3). Anal. Calcd for
C17H15BrN2 (327.2): C, 62.40; H, 4.62; N, 8.56. Found: C,
62.18; H, 4.68; N, 8.85%.
4H, 4Ј-, 3Љ-, 4Љ-, 5Љ-H), 7.63 (m, 1H, 6Ј-H), 7.85 (m, 2H, 2Љ-,
6Љ-H); δC(63 MHz; CDCl3) 24.3 (q, C-2-CH3), 61.0 (s, C-2),
2
4
116.5 (dd, JCF = 23, C-3Ј), 124.3 (dd, JCF = 3, C-5Ј), 124.9 (d,
2JCF = 13, C-1Ј), 128.4 (d, C-3Љ), 129.0 (d, C-2Љ), 130.4 (dd,
3JCF = 3, C-6Ј), 132.6 (d, C-4Љ), 134.0 (dd, 3JCF = 9, C-4Ј), 136.0
1
3
(s, C-1Љ), 159.8 (d, JCF = 255, C-2Ј), 198.7 (d, JCF = 3, C-1),
5
199.8 (d, JCF = 1, C-3). Anal. Calcd for C17H15FO2 (270.3): C,
75.54; H, 5.55. Found: C, 75.95; H, 5.52%.
1-(2Ј-Chlorophenyl)-2,2-dimethyl-3-phenylpropane-1,3-dione
(5c). 43%, colorless needles, mp 55–56 ЊC; νmax(KBr)/cmϪ1 3040,
2970, 2910, 1670, 1640, 1550, 1450, 1400, 1240, 1220; δH(200
MHz; CDCl3) 1.58 (s, 6H, 2-CH3), 6.95–7.22 (m, 3H, 3Ј-, 4Ј-,
5Ј-H), 7.28–7.42 (m, 4H, 6Ј-, 3Љ-, 4Љ-, 5Љ-H), 7.94 (m, 2H, 2Љ-,
6Љ-H); δC(50 MHz; CDCl3) 23.7 (q, C-2-CH3), 60.1 (s, C-2),
125.9 (d, C-5Ј), 128.3 (d, C-3Ј), 128.7 (d, C-3Љ), 128.8 (s, C-2Ј),
129.5 (d, C-2Љ) ,130.8 (d, C-6Ј), 131.1 (d, C-4Ј), 132.6 (d, C-4Љ),
135.7 (s, C-1Љ), 137.4 (s, C-1Ј), 197.8 and 201.5 (2 × s, C-1 and
C-3). Anal. Calcd for C17H15ClO2 (286.8): C, 71.21; H, 5.27.
Found: C, 71.36; H, 5.13%.
4,4-Dimethyl-3-(2Ј-methylphenyl)-5-phenyl-4H-pyrazole (6e).
81%, colorless powder, mp 110–111 ЊC; νmax(KBr)/cmϪ1 2978,
1555, 1519, 1491, 1458, 1342, 1002, 829, 785, 767; δH(200 MHz;
CDCl3) 1.53 (s, 6H, 4-CH3), 2.37 (s, 3H, 2Ј-CH3), 7.26–7.35 (m,
4H, 3Ј-, 4Ј-, 5Ј-, 6Ј-H), 7.48–7.53 (m, 3H, 3Љ-, 4Љ-, 5Љ-H), 8.03
(m, 2H, 2Љ-, 6Љ-H); δC(50 MHz; CDCl3) 20.5 (q, C-4-CH3), 21.4
(q, C-2Ј-CH3), 61.6 (s, C-4), 125.3 (d, C-5Ј), 127.8 (d, C-2Љ),
127.9 (d, C-6Ј), 128.8 (d, C-3Љ), 129.3 (d, C-3Ј), 130.2 (d, C-4Ј),
130.6 (d, C-4Љ), 130.8 (s, C-1Ј), 131.0 (s, C-1Љ), 127.7 (s, C-2Ј),
177.6 and 181.2 (s, C-3 and C-5). Anal. Calcd for C18H18N2
(262.4): C, 82.41; H, 6.92; N, 10.68. Found: C, 82.04; H, 6.80;
N, 10.69%.
1-(2Ј-Bromophenyl)-2,2-dimethyl-3-phenylpropane-1,3-dione
(5d). 38%, colorless needles, mp 60–61 ЊC; νmax(KBr)/cmϪ1
3040, 2980, 2920, 1640, 1570, 1500, 1450, 1400, 1320, 1240;
δH(200 MHz; CDCl3) 1.58 (s, 6H, 2-CH3), 7.32–7.41 (m, 2H,
4Ј-, 5Ј-H), 7.48–7.54 (m, 3H, 3Љ-, 4Љ-, 5Љ-H), 7.66 (m, 1H, 3Ј-H),
7.86–7.92 (m, 3H, 6Ј-, 2Љ-, 6Љ-H); δC(50 MHz; CDCl3) 24.1
(q, C-2-CH3), 60.8 (s, C-2), 123.6 (s, C-2Ј), 127.9 (d, C-5Ј), 128.5
(d, C-6Ј), 128.7 (d, C-3Љ), 129.1 (d, C-2Љ), 131.3 (d, C-4Ј), 131.7
(d, C-3Ј), 133.3 (d, C-4Љ), 135.4 (d, C-1Љ), 136.4 (s, C-1Ј), 197.4
and 202.4 (2 × s, C-1 and C-3). Anal. Calcd for C17H15BrO2
(331.2): C, 61.65; H, 4.56. Found: C, 61.42; H, 4.85%.
3-(2Ј-Methoxyphenyl)-4,4-dimethyl-5-phenyl-4H-pyrazole
(6f). 78%, colorless powder, mp 130–131 ЊC; νmax(KBr)/cmϪ1
3040, 2960, 2920, 1590, 1510, 1490, 1410, 1250, 1180, 1150;
δH(200 MHz; CDCl3) 1.39 (s, 6H, 4-CH3), 3.69 (s, 3H, OCH3),
6.89–6.99 (m, 2H, 4Ј-, 5Ј-H), 7.16–7.53 (m, 4H, 3Ј-, 3Љ-, 4Љ-, 5Љ-
H), 8.04 (m, 3H, 6Ј-, 2Љ-, 6Љ-H); δC(63 MHz; CDCl3) 21.3 (q,
C-4-CH3), 55.3 (q, OCH3), 61.4 (s, C-4), 111.2 (d, C-3Ј), 111.5
(s, C-1Ј), 120.3 (d, C-5Ј), 127.7 (d, C-2Љ), 128.5 (d, C-6Ј), 128.6
(d, C-3Љ), 130.4 (s, C-1Љ), 130.6 (d, C-4Ј), 130.9 (d, C-4Љ), 157.5
(s, C-2Ј), 178.1 and 180.8 (s, C-3 and C-5). Anal. Calcd for
C18H18N2O (278.4): C, 77.67; H, 6.52; N, 10.06. Found: C,
77.56; H, 6.51; N, 9.77%.
Preparation of the 4,4-dimethyl-3,5-diaryl-4H-pyrazoles (6)
The corresponding dimethylated diketone 5 (5.00 mmol) was
dissolved in 30 ml of CHCl3 and 100% hydrazine hydrate (5.50
mmol) was added. The mixture was refluxed for 24 h. After
2726
J. Chem. Soc., Perkin Trans. 2, 1999, 2723–2728

View Article Online
Preparation of the azoalkanes (1)
(m, 2H, 2Љ-, 6Љ-H); δC(50 MHz; CDCl3) 17.1 (q, C-9), 18.7 (q,
C-10), 22.7 (q, C-2Ј-CH3), 25.7 and 25.8 (t, C-5 and C-7), 28.8
(t, C-6), 47.8 (d, C-7a) 49.0 (d, C-4a), 65.9 (s, C-8) 98.4 (s, C-4),
100.7 (s, C-1), 125.2 (d, C-5Ј), 125.5 (d, C-6Ј), 127.6 (d, C-2Љ),
128.0 (d, C-4Љ), 128.0 (d, C-4Ј), 128.4 (d, C-3Љ), 128.5 (d, C-3Ј),
136.5 (s, C-2Ј), 137.2 (s, C-1Ј), 137.5 (s, C-1Љ). Anal. Calcd for
C23H26N2 (330.5): C, 83.59; H, 7.93; N, 8.48. Found: C, 83.09;
H, 7.84; N, 8.22%.
The corresponding 4H-pyrazole 6 (1.00 mmol) and cyclo-
pentene (5.00 mmol) were dissolved in toluene (1 ml). The solu-
tion was heated to 110 ЊC for 2-3 d under high pressure (12.5
kbar). After the solvent was removed (20 ЊC, 10 Torr), the crude
solid product was purified by column chromatography on silica
gel.
(1a,4a,4aꢀ,7aꢀ)-4,4a,5,6,7,7a-Hexahydro-1-(2Ј-fluorophenyl)-
8,8-dimethyl-4-phenyl-1,4-methano-1H-cyclopenta[d]pyridazine
(1b). 12%, colorless needles, mp 114–115 ЊC, dec; νmax(KBr)/
cmϪ1 3040, 2940, 2820, 1590, 1570, 1500, 1470, 1420, 1290,
1260; λmax (C6H6)/nm 327 (ε/dm3 molϪ1 cmϪ1 27), 349 (95), 362
(167); δH(250 MHz; CDCl3) 0.20 (s, 3H, 9-H), 1.03 (s, 3H,
10-H), 1.58–1.62 (m, 6H, 5-, 6-, 7-H), 3.51–3.58 (m, 2H, 4a-,
7a-H), 7.21–7.34 (m, 3H, 3Ј-, 4Ј-, 5Ј-H), 7.48–7.57 (m, 4H, 6Ј-,
3Љ-, 4Љ-, 5Љ-H), 7.76 (m, 2H, 2Љ-, 6Љ-H); δC(63 MHz; CDCl3)
17.5 (q, C-9), 18.8 (q, C-10), 25.5 and 25.6 (t, C-5 and C-7), 28.6
(t, C-6), 49.1 (2 × d, C-4a and C-7a), 66.4 (s, C-8), 97.5 (d,
(1a,4a,4aꢀ,7aꢀ)-4,4a,5,6,7,7a-Hexahydro-1-(2Ј-methoxy-
phenyl)-8,8-dimethyl-4-phenyl-1,4-methano-1H-cyclopenta[d]-
pyridazine (1f). 8%, colorless needles, mp 139–140 ЊC, dec;
νmax(KBr)/cmϪ1 2958, 2854, 1579, 1490, 1468, 1434, 1371, 1246,
1119, 1022; λmax (C6H6)/nm 350 (ε/dm3 molϪ1 cmϪ1 121), 363
(152); δH(200 MHz; CDCl3) 0.17 (s, 3H, 9-H), 0.99 (s, 3H, 10-
H), 1.58 (m, 6H, 5-, 6-, 7-H), 3.51 (m, 1H, 4a-H), 3.84 (s, 3H,
3
4
OCH3), 4.22 (m, 1H, 7a-H), 7.01 (dd, J = 8.2, J = 1.2, 1H,
3
4
3Ј-H), 7.10 (dt, J = 7.6, J = 1.2, 1H, 5Ј-H), 7.34–7.52 (m, 4H,
3
4Ј-, 3Љ-, 4Љ-, 5Љ-H), 7.75 (m, 2H, 2Љ-, 6Љ-H), 8.14 (dd, J = 7.8,
4J = 1.7, 1H, 6Ј-H); δC(50 MHz; CDCl3) 17.5 (s, C-9), 18.8 (s,
C-10), 25.5 and 25.6 (2 × t, C-5 and C-7), 28.6 (t, C-6), 46.7 (d,
C-7a), 49.1 (d, C-4a), 55.0 (q, OCH3), 66.4 (s, C-8), 97.4 (s, C-1),
98.6 (s, C-4), 111.7 (d, C-3Ј), 120.6 (d, C-5Ј), 124.3 (s, C-1Ј),
127.5 (d, C-2Љ), 127.6 (d, C-3Љ), 128.1 (d, C-4Љ), 128.9 (d, C-6Ј),
131.4 (d, C-4Ј), 136.3 (s, C-4Љ), 158.2 (s, C-2Ј). Anal. Calcd for
C23H26N2O (346.5): C, 79.73; H, 7.56; N, 8.09. Found: C, 79.37;
H, 7.16; N, 7.79%.
2
3JCF = 2, C-1), 98.6 (s, C-4), 120.4 (dd, JCF = 23, C-3Ј), 122.8
4
2
(dd, JCF = 4, C-5Ј), 123.0 (d, JCF = 15, C-1Ј), 128.3 (d, C-2Љ),
128.3 (d, C-4Љ), 129.2 (d, C-3Љ), 132.9 (dd, 3JCF = 9, C-4Ј), 135.0
3
1
(dd, JCF = 4, C-6Ј), 135.9 (s, C-1Љ), 164.2 (d, JCF = 261, C-2Ј).
Anal. Calcd for C22H23FN2 (334.4): C, 79.02; H, 6.93; N, 8.38.
Found: C, 78.96; H, 6.49; N, 8.12%.
(1a,4a,4aꢀ,7aꢀ)-4,4a,5,6,7,7a-Hexahydro-1-(2Ј-chlorophenyl)-
8,8-dimethyl-4-phenyl-1,4-methano-1H-cyclopenta[d]pyridazine
(1c). 8%, colorless needles, mp 108–109 ЊC, dec; νmax(KBr)/cmϪ1
3030, 2960, 2900, 2820, 1600, 1460, 1430, 1400, 1310, 1240; λmax
(C6H6)/nm 327 (ε/dm3 molϪ1 cmϪ1 23), 346 (88), 360 (184);
δH(250 MHz; CDCl3) 0.19 (s, 3H, 9-CH3), 1.13 (s, 3H, 10-CH3),
1.46–1.61 (m, 6H, 5-, 6-, 7-H), 3.57 (m, 1H, 4a-H), 4.72 (m, 1H,
7a-H), 7.28–7.33 (m, 3H, 3Ј-, 4Ј-, 5Ј-H), 7.35–7.53 (m, 4H, 6Ј-,
EPR Spectroscopy
A solution of the azoalkane 1 (ca. 5 µmol) in 2-methyl-
tetrahydrofuran (ca. 0.5 mL) was placed into the EPR tube,
degassed by purging with argon gas for 15 min, and sealed. The
glass matrix was formed by cooling the sample to 77 K with
liquid nitrogen. The triplet diradical 2 was generated by irradi-
ation with the 333-, 351- and 364-nm lines of an INNOVA-100
CW argon-ion laser (widened beam, 2.0 W MLUV, 2 min) at
77 K. Its EPR spectrum was recorded on a Bruker ESP-300
spectrometer (9.43 GHz, spectra accumulation with the Bruker
1620 data system, n у 5). The D value was determined by
analysis of the Z signals.
3
3Љ-, 4Љ-, 5Љ-H), 7.75 (d, J = 7.3, 2H, 2Љ-, 6Љ-H); δC(63 MHz;
CDCl3) 17.8 (q, C-9), 19.2 (q, C-10), 25.3 and 25.5 (2 × t, C-5
and C-7), 28.7 (t, C-6), 48.8 (d, C-7a), 49.6 (d, C-4a), 67.3 (s,
C-8), 98.0 (s, C-4), 99.2 (s, C-1), 126.6 (d, C-5Ј), 127.5 (d, C-2Љ),
127.6 (d, C-6Ј), 127.7 (d, C-4Љ), 128.3 (d, C-3Љ), 129.0 (d, C-4Ј),
130.1 (d, C-3Ј), 131.8 (s, C-2Ј), 133.5 (s, C-1Ј), 135.8 (s, C-1Љ).
Anal. Calcd for C22H23ClN2 (350.9): C, 75.31; H, 6.61; N, 7.98.
Found: C, 75.14; H, 6.17; N, 7.62%.
Computations
All calculations were run on an IRIS INDIGO Silicon
Graphics Workstation. Full geometry optimization of the
cumyl radicals 2 was carried out for torsional angles θ of the
aryl groups between 0 and 90Њ by using the PM37a method and
AUHF wave functions, which are provided in the VAMP5.0
program.7b The cumyl spin densities for different angles θ were
then determined with a single-point CI calculation, which
results in good spin expectation 〈S2〉 values between 0.75
and 0.78 for these radicals. For geometry optimization of the
diradicals 3, again the minimum-energy conformer was calcu-
lated by the PM3-AUHF/CI method as described above. The
acquired conformer was then further optimized by the UHF
method with the 6-31G* basis set, which is provided by the
GAUSSIAN94 program.10
(1a,4a,4aꢀ,7aꢀ)-4,4a,5,6,7,7a-Hexahydro-1-(2Ј-bromophenyl)-
8,8-dimethyl-4-phenyl-1,4-methano-1H-cyclopenta[d]pyridazine
(1d). 7%, colorless needles, mp 115–116 ЊC, dec; νmax(KBr)/cmϪ1
3020, 2960, 2940, 1890, 1580, 1560, 1460, 1450, 1400, 1280; λmax
(C6H6)/nm 327 (ε/dm3 molϪ1 cmϪ1 65), 348 (108), 360 (183);
δH(250 MHz; CDCl3) 0.17 (s, 3H, 9-H), 1.17 (s, 3H, 10-H), 1.60
(m, 6H, 5-, 6-, 7-H), 3.61 (m, 1H, 4a-H), 4.94 (m, 1H, 7a-H),
7.18–7.26 (m, 1H, 4Ј-H), 7.36–7.54 (m, 4H, 5Ј-, 3Љ-, 4Љ-, 5Љ-H),
7.68–7.81 (m, 3H, 3Ј-, 2Љ-, 6Љ-H), 8.14–8.18 (m, 1H, 6Ј-H);
δC(63 MHz; CDCl3) 17.4 (q, C-9), 19.5 (q, C-10), 25.1 and 25.2
(2 × t, C-5 and C-7), 28.6 (t, C-6), 45.5 (s, C-7a), 49.4 (s, C-4a),
67.5 (s, C-8), 98.0 (s, C-4), 99.7 (s, C-1), 120.8 (s, C-2Ј), 127.6 (d,
C-2Љ), 128.2 (d, C-4Љ), 128.8 (d, C-3Љ), 129.2 (d, C-5Ј), 129.8 (d,
C-4Ј), 132.6 (d, C-3Ј), 133.9 (d, C-6Ј), 135.7 (s, C-1Љ), 136.1 (s,
C-1Ј). Anal. Calcd for C22H23BrN2 (395.3): C, 66.84; H, 5.86; N,
7.09. Found: C, 67.24; H, 5.44; N, 7.28%.
Acknowledgements
We express our gratitude to the Deutsche Forschungsgemein-
schaft, Volkswagen-Stiftung and the Fonds der Chemischen
Industrie for financial support. We thank Birgit Beck for tech-
nical assistance.
(1a,4a,4aꢀ,7aꢀ)-4,4a,5,6,7,7a-Hexahydro-8,8-dimethyl-1-(2Ј-
methylphenyl)-4-phenyl-1,4-methano-1H-cyclopenta[d]pyrid-
azine (1e). 10%, colorless needles, mp 120–121 ЊC, dec; νmax
(KBr)/cmϪ1 3040, 2980, 2921, 1630, 1555, 1494, 1470, 1446,
1370, 797; λmax (C6H6)/nm 327 (ε/dm3 molϪ1 cmϪ1 81), 347 (99),
360 (124); δH(200 MHz; CDCl3) 0.20 (s, 3H, 9-CH3), 1.21 (s, 3H,
10-CH3), 1.51 (m, 6H, 5-, 6-, 7-H), 2.06 (s, 3H, 2Ј-CH3), 3.65
(m, 1H, 4a-H), 4.21 (m, 1H, 7a-H), 7.25 (m, 1H, 5Ј-H), 7.31–
7.46 (m, 5H, 3Ј-, 4Ј-, 3Љ-, 4Љ-, 5Љ-H), 7.55 (m, 1H, 6Ј-H), 7.77
References
1 (a) S. P. McGlynn, T. Azumi and M. Kinoshita, in Molecular
Spectroscopy of the Triplet State, Prentice-Hall, Englewood Cliffs,
1969, Chapter 9; (b) J. A. Weil, J. R. Bolton and J. E. Wertz, in
J. Chem. Soc., Perkin Trans. 2, 1999, 2723–2728
2727

View Article Online
Electron Paramagnetic Resonance Elementary Theory and Practical
Applications, John Wiley & Sons, New York, 1994, Chapter 9;
(c) E. Wassermann, L. C. Snyder and W. A. Yager, J. Chem. Phys.,
1964, 41, 1763; (d) R. McWeeny, J. Chem. Phys., 1961, 34, 399;
(e) H. M. McConnell, J. Chem. Phys., 1958, 28, 1188.
2 W. Adam, H. M. Harrer, F. Kita and W. M. Nau, Pure Appl. Chem.,
1997, 69, 91.
3 W. Adam, H. M. Harrer, T. Heidenfelder, T. Kamel, F. Kita, W. M.
Nau and C. Sahin, J. Chem. Soc., Perkin Trans. 2, 1996, 2085.
4 (a) W. Adam, L. Fröhlich, W. M. Nau, H. G. Korth and R.
Sustmann, Angew. Chem., Int. Ed. Engl., 1993, 32, 1339; (b) W.
Adam, H. M. Harrer, F. Kita and W. M. Nau, Adv. Photochem.,
1998, 24, 205.
5 (a) W. M. Nau, H. M. Harrer and W. Adam, J. Am. Chem. Soc.,
1994, 116, 10972; (b) W. Adam, H. M. Harrer, W. M. Nau and
K. Peters, J. Org. Chem., 1994, 59, 3786; (c) K. Beck and S. Hünig,
Chem. Ber., 1987, 120, 477.
Geometry optimizations were performed using the IIS SCF
converger: P. Badziag and F. Solms, Comput. Chem., 1988, 12, 233.
8 W. Adam, H. M. Harrer, F. Kita, H.-G. Korth and W. M. Nau,
J. Org. Chem., 1997, 62, 1419.
9 A. Bondi, J. Phys. Chem., 1964, 68, 441.
10 GAUSSIAN94, Revision D.4, M. J. Frisch, G. W. Trucks, H. B.
Schlegel, P. M. W. Gill, B. G. Johnson, M. A. Robb, J. R.
Cheeseman, T. Keith, G. A. Petersson, J. A. Montgomery, K.
Raghavachari, M. A. Al-Laham, V. G. Zakrzewski, J. V. Ortiz,
J. B. Foresman, J. Cioslowski, B. B. Stefanov, A. Nanayakkara,
M. Challacombe, C. Y. Peng, P. Y. Ayala, W. Chen, M. W. Wong,
J. L. Andres, E. S. Replogle, R. Gomperts, R. L. Martin, D. J. Fox,
J. S. Binkley, D. J. Defrees, J. Baker, J. P. Stewart, M. Head-Gordon,
C. Gonzalez and J. A. Pople, Gaussian, Inc., Pittsburgh, PA, 1995.
11 C. L. Bickel, J. Am. Chem. Soc., 1947, 69, 73.
12 W. Theilacker and H. Rosewicz, Chem. Ber., 1953, 86, 1263.
13 W. Bradley and R. Robinson, J. Chem. Soc., 1926, 128, 2356.
14 M. Yoshioka, K. Nishizawa, M. Arai and T. Hasegawa, J. Chem.
Soc., Perkin Trans. 1, 1991, 541.
6 W. Adam, O. Emmert and H. M. Harrer, J. Chem. Soc., Perkin
Trans. 2, 1997, 687.
7 (a) J. J. P. Stewart, J. Comput. Chem., 1989, 10, 209; (b) T. Clark and
T. Kovàr. Unpublished results. See: T. Kovàr, Diploma Thesis,
Universität Erlangen-Nürnberg, 1987. The AUHF algorithm is
included in the VAMP 5.0 package: G. Rauhut, J. Chandrasekar,
T. Clark and T. Steinke, Universität Erlangen-Nürnberg, 1993.
15 A. Choudhary and A. L. Baumstark, Synthesis, 1989, 688.
Paper 9/05773K
2728
J. Chem. Soc., Perkin Trans. 2, 1999, 2723–2728