EPR studies of nitrogen-centred free radicals. Part 53. Isolation, EPR spectra and magnetic characterization of N-(arylthio)-2,4-diaryl-6-cyanophenylaminyls

Article abstract of DOI:10.1039/b100276g

N-(Arylthio)-2,4-diaryl-6-ethoxycarbonylphenylaminyls (1), N-[(2,4-dichlorophenyl)thio]-2,4-diphenyl-6-acetylphenylaminyl (2), N-(arylthio)-2,4-diaryl-6-cyanophenylaminyls (3), N-[(2,4-dichlorophenyl)thio]-2,4-bis(4-chlorophenyl)-6-fluorophenylaminyl (4) and N-[(4-nitrophenyl)thio]-2,4-diphenylphenylaminyl (5) were generated by oxidation of the corresponding N-(arylthio)anilines. Although 4 and 5 were short-lived and decayed in 30 min, 1-3 were quite persistent and 3 could be isolated as radical crystals. EPR spectra were measured for all radicals generated and the spin density distribution was evaluated. Ab initio MO calculations (the UHF Becke 3LYP/STO 6-31G) were performed, and a quantitative discussion on the spin density distribution was made. Magnetic susceptibility measurements were performed for three isolated radicals with a SQUID magnetometer. One radical was found to couple ferromagnetically, and analysis with the one-dimensional regular Heisenberg model gave 2J/k_B = 11.2 K.

Full text of DOI:10.1039/b100276g

View Article Online / Journal Homepage / Table of Contents for this issue
EPR studies of nitrogen-centred free radicals. Part 53.1 Isolation,
EPR spectra and magnetic characterization of N-(arylthio)-2,4-
diaryl-6-cyanophenylaminyls
2
Masaaki Nakatsuji,a Yozo Miura*a and Yoshio Tekib
a Department of Applied Chemistry, Faculty of Engineering, Osaka City University,
Sumiyoshi-ku, Osaka 558-8585, Japan
b Department of Material Science, Graduate School of Science, Osaka City University,
Sumiyoshi-ku, Osaka 558-8585, Japan
Received (in Cambridge, UK) 2nd January 2001, Accepted 9th March 2001
First published as an Advance Article on the web 9th April 2001
N-(Arylthio)-2,4-diaryl-6-ethoxycarbonylphenylaminyls (1), N-[(2,4-dichlorophenyl)thio]-2,4-diphenyl-6-acetyl-
phenylaminyl (2), N-(arylthio)-2,4-diaryl-6-cyanophenylaminyls (3), N-[(2,4-dichlorophenyl)thio]-2,4-bis(4-chloro-
phenyl)-6-fluorophenylaminyl (4) and N-[(4-nitrophenyl)thio]-2,4-diphenylphenylaminyl (5) were generated by
oxidation of the corresponding N-(arylthio)anilines. Although 4 and 5 were short-lived and decayed in 30 min,
1–3 were quite persistent and 3 could be isolated as radical crystals. EPR spectra were measured for all radicals
generated and the spin density distribution was evaluated. Ab initio MO calculations (the UHF Becke 3LYP/STO
6-31G) were performed, and a quantitative discussion on the spin density distribution was made. Magnetic
susceptibility measurements were performed for three isolated radicals with a SQUID magnetometer. One
radical was found to couple ferromagnetically, and analysis with the one-dimensional regular Heisenberg model
gave 2J/kB = 11.2 K.
Thioaminyl radicals are interesting as isolable nitrogen-centred
free radicals.2 This family of radicals, as well as cyclic NS rad-
icals,3 are attractive spin sources for organic magnetism. Since
thioaminyls have an extensively delocalized π-spin system, their
magnetic interactions are expected to be strong compared with
aminoxyl or imine N-oxide-aminoxyl radicals which have local-
ized π-spin systems.4–6 In recent years we have designed isolable
thioaminyl radicals. The isolated thioaminyls were subjected to
magnetic measurements with a SQUID magnetometer, and it
was found that six thioaminyls couple ferromagnetically.1,7–9
The magnetic interactions are 3.6–28 K (2J/kB), which are much
stronger than those of aminoxyl or imine N-oxide-aminoxyl
radicals. In a continued study we investigated o-ethoxycarbonyl
(1), acetyl (2), cyano (3) and fluoro (4) substituted thioaminyl
radicals 1–4, along with ortho-unsubstituted thioaminyl 5.
These thioaminyls are structurally interesting because we
expected that these functional groups may strengthen the
magnetic interactions between the radical molecules through
the electrostatic interaction of functional groups. How-
ever, these radicals might not be sufficiently persistent to be
isolated because the protection of the ortho position of the
anilino benzene ring by the EtOCO, MeCO, or CN group
seemed to be insufficient. Although radicals 4 and 5 did
not persist for a long period, 1–3 were very long lived, and 3
could be isolated as radical crystals. Herein we report gener-
ation, isolation, EPR spectra, spin density calculations and
magnetic properties of 3, along with EPR spectra of 1, 2, 4 and
5. Furthermore we report the decomposition mechanism of 4
and 5.
dibromo-6-R-anilines (R = OCOEt, COMe, CN, F) in 89–93%
yields. The Pd-catalyzed cross-coupling reaction of 2,4-
dibromo-6-R-anilines with arylboronic acid yielded 2,4-diaryl-
6-R-anilines in 76–90% yields.11 Preparation of 2,4-diaryl-
aniline was previously reported by us.12 Reaction of 2,6-diaryl-
6-R-anilines or 2,4-diarylaniline with arenesulfenyl chlorides
gave 6 in 40–48%, 7 in 42%, 8 in 36–52%, 9 in 52% and 10 in
48% yields, respectively. Precursors 6–8 and 10 were obtained as
crystals, while 9 was a semi-solid.
Results and discussion
Syntheses of precursors
The corresponding precursors 6–10 were prepared according to
Scheme 1. Thus, treatment of 2-substituted anilines with benzyl-
trimethylammonium tribromide (BTMA Br3)10 in CH2Cl2–
MeOH in the presence of CaCO3 gave the corresponding 2,4-
738
J. Chem. Soc., Perkin Trans. 2, 2001, 738–744
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b100276g

View Article Online
were performed. Precursors 9 and 10 were oxidized with PbO2
in benzene, and the mixtures were filtrated. The green-coloured
solutions were allowed to stand until the green colour com-
pletely disappeared. After concentration, column chroma-
tography of the solutions gave 13 and 14 in 41 and 38% yields,
respectively. The structures of 13 and 14 were determined by
IR, 1H NMR, mass spectra and elemental analyses.
Scheme 1 Reactions and conditions: i, BTMA Br3, CaCO3, CH2Cl2–
MeOH, room temp; ii, Ar1B(OH)2, (PPh3)4Pd, K2CO3, benzene–EtOH–
H2O, reflux; iii, Ar2SCl, Et3N, dry ether or THF, 0 ЊC; iv, PbO2,
benzene, room temp.
A plausible mechanism for the decomposition of 4 and 5 is
depicted in Scheme 2.8,14–16 The first step is the N–C coupling
Generation and isolation of radicals
Oxidation of precursors 6–10 was carried out in benzene using
PbO2 as oxidant. When PbO2 was added to stirred solutions of
precursors, the solutions immediately turned to dark green. In
the cases of 1–3 the characteristic green colours were almost
constant for one day under the atmospheric conditions.
However, when the solutions were left for a month at room
temperature, the green colours completely disappeared (the half
lifetimes, 1–3 weeks). In the case of 4 the dark green colour
immediately disappeared and the solution turned dark brown.
Thus, for measurements of EPR spectra, 4 was generated by the
reaction of 9 and di-tert-butyl diperoxyoxalate. The diperoxy
ester decomposes at a moderate rate at room temperature
in hydrocarbon solvents such as benzene to give tert-butoxyl
radicals.13 In the case of 5 the dark green colour was kept for
30 min, and the solution finally turned dark brown.
Isolation of 1–3 was tried by the following procedures. The
corresponding precursors 6–8 were oxidized with PbO2 and the
reaction mixtures filtered. The solvent was removed by freeze-
drying, and the resultant crystalline powder was crystallized.
In the case of 3 all radicals examined (3a–d) were successfully
isolated in 37–47% yields. On the other hand, isolation of 1 and
2 was unsuccessful in spite of much effort and only sulfinamides
11 and 12 were formed in 43 and 38% yields, respectively, in the
oxidation of 6a and 7. Usually, sulfinamide is not formed in
oxidation of precursors with PbO2 and it is not clear why sulf-
inamide was formed in such a high yield in these cases. Removal
of the sulfinamides by recrystallization or column chrom-
atography was unsuccessful. Recrystallization gave a mixture of
radical and sulfinamide, and column chromatography resulted
in complete decomposition of radical. To suppress the form-
ation of sulfinamides, oxidation was carried out in degassed
benzene. However, formation of 11 or 12 was still observed,
suggesting that the sulfinamide oxygen comes not from the
atmospheric oxygen, but from PbO2. Oxidation with Ag2O was
also attempted, but formation of 11 or 12 was still observed.
Scheme 2
between two thioaminyl radicals to give 15 or 16. Elimination
of Ar S ϩ F from 15 yields 13, and elimination of Ar S ϩ H
from 16 yields 14. If R is bulky enough to prevent the coupling
reaction or if the C–R bond of the formed C–N coupled dimer
is sufficiently strong, the radicals will be long lived like 1–3.
Thus, it is likely that the stabilities of this type of radical
depend on the bulkiness of R and the bond strength of C–R.
2
2
ؒ
ؒ
ؒ
ؒ
EPR spectra
EPR spectra of 1–5 were recorded at room temperature using
benzene as solvent. A typical EPR spectrum is shown in Fig. 1
and EPR parameters are summarized in Table 1. Fig. 1 shows
that the EPR spectrum of 3a consists of a simple 1 : 1 : 1 triplet.
Similar 1 : 1 : 1 triplet EPR spectra were observed for 1, 2 and
3b–d. The absence of proton hyperfine splittings is attributed to
the presence of many protons with unresolved small hyperfine
couplings (hfc). The peak-to-peak line widths (∆Hpp) are
therefore relatively large (0.283–0.316 mT). In the case of 4 an
EPR spectrum (Fig. 2) was split into a 1 : 2 : 2 : 1 quartet
by the interaction with a nitrogen atom and a fluorine atom.
Computer simulation of this spectrum gave aN = 0.873 and
The mechanism for decomposition of 4 and 5
In contrast to 1–3, 4 and 5 were short-lived. The characteristic
green colours disappeared immediately (4) or within 30 min (5).
To clarify their decomposition mechanism, product analyses
J. Chem. Soc., Perkin Trans. 2, 2001, 738–744
739

View Article Online
Table 1 EPR parameters for 1–5 in benzene at 20 ЊCa
b
c
Radical
aN
aother
g
1a
1b
2
3a
3b
3c
0.908
0.894
0.907
0.851
0.847
0.844
0.853
0.853
0.873
0.898
2.0059
2.0058
2.0059
2.0058
2.0058
2.0060
2.0059
2.0059
2.0058
2.0061
3d
3b-dd
4d
0.130 (2H),e 0.085 (1H) f
0.980 (1F)
5d
0.390 (1H)g
a Hyperfine coupling (hfc) constants are given in mT. b The values for
the NS nitrogen. c The number in parentheses refer to the number
of equivalent H or F. d Hfc constants are determined by computer
simulation. e The anilino meta-protons. f The arylthiyl ortho-proton.
g The anilino ortho-proton.
Fig. 3 EPR spectrum of 3b-d in benzene at 20 ЊC. (a) Experimental;
(b) simulation.
Fig. 1 The experimental EPR spectrum of 3a in benzene at 20 ЊC.
Fig. 4 Optimized structure of 3b calculated by the MNDO/AM1
method.
lower than those of the others (0.873–0.908 mT). If this
decrease in aN is explained in terms of an enhancement of
canonical form B (Scheme 3) due to an electron-withdrawing
Scheme 3
CN group, an increase in g must also be observed. However,
such an increase in g is not observed for any radicals 3. There-
fore, an enhancement of canonical form B can be ruled out. An
alternative explanation is a delocalization of the unpaired elec-
tron spin onto CN. The ab initio calculations predict a con-
siderable delocalization of the unpaired electron spin onto CN
as described below. Since the aN values for 1 and 2 are very
similar to that of 5, we assume that the EtOCO and COMe
groups are significantly twisted from the anilino benzene ring.
Fig. 2 EPR spectrum of 4 in benzene at 20 ЊC. (a) Experimental; (b)
simulation.
aF = 0.980 mT. An EPR spectrum of 5 was split into doublets of
a triplet by the interaction with a nitrogen and an anilino ortho-
proton. Computer simulation of this spectrum gave aN = 0.898
and aH = 0.390 mT.
The 2- and 4-phenyl groups of 3b were deuterated to detect
hyperfine splittings due to protons. Such deuteration reduces
the ∆Hpp of EPR spectra and this has often allowed the
observation of hyperfine splittings due to protons. The pre-
cursor 8b-d was prepared according to the same procedure as
for the corresponding nondeuterated compound. An EPR
spectrum of 3b-d is shown in Fig. 3, along with a computer
simulation. The spectrum is split into a 1 : 1 : 1 triplet, and each
component of the triplet is further split into an incomplete
1 : 3 : 3 : 1 quartet. Computer simulation of the spectrum gave
aN = 0.853 and aH = 0.130 (2H) and 0.085 mT (1H). The
protons giving the hyperfine coupling constant of 0.130 mT are
assigned to the anilino meta-protons and the proton giving the
hfs constant of 0.085 mT to the phenylthiyl ortho-proton.
When the hfc constants are compared to each other, one may
note that the aN values of 3 (0.844–0.853 mT) are ~0.05 mT
Spin density calculations
The spin density distribution in 3b was calculated by the ab
initio molecular orbital (MO) calculations based on density
functional theory. The electronic structure of the radical
molecule was calculated using Gaussian 98.17 The molecular
structure was optimized by the semi-empirical MO calculations
(MNDO/AM1 method). The ab initio calculations were carried
out using the STO 6-31G basis sets and the UHF Becke 3LYP
hybrid method. The optimized geometry is shown in Fig. 4 and
the calculated spin densities are shown in Fig. 5. In Table 2, the
calculated and experimental hfc constants are compared.
Fig. 4 shows that the ArN–SAr π-framework is almost
planar, and the benzene ring at the para position makes a
twisted angle of 32Њ with the anilino benzene ring. On the other
hand, the benzene ring at the ortho position makes a twisted
740
J. Chem. Soc., Perkin Trans. 2, 2001, 738–744

View Article Online
Table 2 Experimental and calculated hyperfine coupling constants of
3d-d
Atom
Experimental |ax|/mT
Calculationa,b ax/mT
Nc
Ha
Hb
Hc
0.853
0.130
0.130
0.085
1.088
0.247
0.274
Ϫ0.086
Fig. 6 χmolT vs. T plots of 3a. The solid line was calculated by the one-
dimensional Heisenberg model.
a The values calculated for 3b. b The calculations were performed by the
ab initio method at the UHF B3LYP/STO 6-31G level. c The value for
the NS nitrogen.
Fig. 7 χmolT vs. T plots of 3b. The solid line was calculated by the one-
dimensional alternating model.
it is obvious that the intrachain interaction is ferromagnetic and
the interchain interaction is antiferromagnetic, we could not
reconstruct the experimental data by eqn. (1).
Fig. 5 Total atomic spin densities for 3b calculated by the UHF Becke
3LYP hybrid method using the STO 6-31G basis set.
angle of 89Њ due to the steric congestion around the radical
centre. Such an optimized structure is common for thioaminyl
radicals, as clarified by X-ray crystallographic analyses.1,8,9
Since it was previously confirmed that the structures of thio-
aminyls optimized by the above calculation method are in
excellent agreement with those determined by X-ray crystallo-
graphy,1 we believe that the present calculations yield a
reliable structure.
Although the calculations give somewhat large hfc constants
for N, as well as the anilino meta-protons, good agreements
between the observed hfc constants and the calculations are
found (see Table 2). Fig. 5 shows that the unpaired electron
spin is extensively delocalized over the whole of the radical
molecule, with the particularly high spin densities on the NS
and anilino benzene ring. We also note that there is some spin
density on the cyano group. This is in accordance with the
above explanation for a decrease in aN of 3.
(1)
Radical 3b showed a strong antiferromagnetic interaction. As
found in Fig. 7, the χmolT value at 300 K is already lower than
the Curie constant of 0.376 emu K molϪ1 for the S = ½ systems,
and it was drastically decreased with decreasing temperature
and became nearly 0 emu K molϪ1 at 2 K. Such a magnetic
behaviour indicates a strong antiferromagnetic interaction
between the radical molecules, and this suggests that there is a
large SOMO–SOMO overlap between the neighbouring radical
molecules. The best fit with the experimental data with the
alternating linear chain model19 [eqn. (2)] gave 2J/kB = Ϫ200 K
and α (alternating parameter) = 0.87.
H = Ϫ2JΣ(S2i Ϫ 1S2i + αS2i S2i + 1
)
(2)
Magnetic susceptibility measurements
Antiferromagnetic interaction was also observed for 3d. The
χmolT vs. T plots were analyzed with the dimer model, and the
intra-dimer interaction was determined to be 2J/kB = Ϫ56.8 K.
In the case of 3c, the magnetic interaction was very weak and
the presence of a large number of lattice defects was indicated
in the SQUID magnetometer. Thus, a reliable analysis of the
experimental data was impossible.
The magnetic susceptibility (χmol) measurements were per-
formed for the isolated four radicals using polycrystalline
samples with a SQUID magnetometer in the temperature range
1.8–300 K. The diamagnetic corrections were made based
on Pascal’s sum rule. The spin concentrations of the radical
samples used were >95% as determined by EPR.
Fig. 6 shows χmolT vs. T plots for 3a. The χmolT value is
constant (0.376 emu K molϪ1) above 100 K and increases
with decreasing temperature below 100 K. It then reaches a
maximum of 0.54 emu K molϪ1 at 6 K and decreases with
decreasing temperature. When the magnetic behaviour was
analyzed by the modified Bleaney–Bower model, no satisfac-
tory agreement with the experimental data was obtained. On
the other hand, analysis with the one-dimensional regular
Heisenberg model18 [when θ = 0 K in eqn. (1)] gave a satisfac-
tory agreement with the experimental data in the temperature
region of 50–300 K, and 2J/kB was determined to be +11.2 K.
However, when θ = Ϫ2, Ϫ5, Ϫ10 and Ϫ15 K, no satisfactory
agreement with the experimental data was obtained. Although
Conclusions
Four kinds of N-(arylthio)-2,4-diaryl-6-R-phenylaminyls (R =
EtOCO, MeCO, CN, F) and N-[(4-nitrophenyl)thio]-2,4-di-
phenylphenylaminyl were generated. EtOCO-, MeCO- and
CN-substituted radicals were long-lived, and CN-substituted
radicals were isolated as radical crystals. F-substituted and
unsubstituted radicals were short-lived and their decom-
position mechanism was elucidated based on the product
analyses. The EPR spectra and ab initio MO calculations (UHF
Becke 3LYP/STO 6-31G) showed an extensive delocalization
over the whole of the radical molecule. Magnetic susceptibility
J. Chem. Soc., Perkin Trans. 2, 2001, 738–744
741

View Article Online
measurements were performed for three isolated radicals. One
radical (3a) showed a ferromagnetic interaction and the χmol
2,4-Diphenyl-6-acetylaniline. Yellow prisms (EtOH); yield
79%; mp 68–70 ЊC; ν (KBr)/cmϪ1 3480 and 3320 (NH2), 1650
T
vs. T plots by the one-dimensional regular Heisenberg model
gave 2J/kB = +11.2 K. The other two radicals showed an anti-
ferromagnetic interaction. Analyses of the χmolT vs. T plots
with the alternating linear chain model (3b) or the dimer model
(3d) gave 2J/kB = Ϫ200 K and α = 0.87 (3b) and 2J/kB = Ϫ56.8
K (3d), respectively.
(C᎐O); δH (CDCl3) 2.70 (s, Me, 3H), 6.62 (br s, NH2, 2H), 7.30
᎐
(t, J 7.3, ArH, 1H), 7.38–7.58 (m, ArH, 10H), 7.97 (d, J 2.4,
ArH, 1H) (Found: C, 83.53; H, 5.91; N, 4.81. C20H17NO
requires: C, 83.59; H, 5.96; N, 4.87%).
2,4-Diphenyl-6-cyanoaniline. Colourless prisms (EtOH); yield
83%; mp 88–90 ЊC; ν (KBr)/cmϪ1 3470, 3440, 3370 and 3340
(NH2), 2220 (CN); δH (CDCl3) 4.57 (br s, NH2, 2H), 7.32 (t,
J 7.8, ArH, 1H), 7.34–7.55 (m, ArH, 10H), 7.65 (d, J 2.4, ArH,
1H) (Found: C, 84.26; H, 5.17; N, 10.23. C19H14N2 requires: C,
84.42; H, 5.22; N, 10.36%).
Experimental
Melting points were determined with a Yanagimoto micro
melting point apparatus and are uncorrected. IR spectra were
obtained on a JASCO FT/IR-230 spectrophotometer. UV-Vis
spectra were recorded on a Shimadzu UV-2200 spectro-
photometer. 1H NMR spectra were measured with a JEOL
α-400 spectrometer (400 MHz); chemical shifts are expressed
in ppm (δ) using tetramethylsilane as an internal standard and
J values are given in Hz. EPR spectra were measured with
a Bruker ESP-300 spectrometer equipped with a X-band
microwave unit and with a 100 kHz field modulation. The
hyperfine coupling constants and g values were determined by
measurement of Fremy’s salt in a dilute K2CO3 aqueous solu-
tion (aN =1.309 mT; g =2.0057) as the reference. Mass spectra
(FAB) were measured by a JEOL JX-700T mass spectrometer.
Magnetic susceptibility measurements were carried out with
a Quantum Design SQUID magnetometer MPMS2 in the
1.8–300 K temperature range. The diamagnetic contributions
of samples were estimated from the Pascal diamagnetic
constants. Silica gel column chromatography was carried out
on Fuji silysia BW-127ZH (Fuji-Davison chem.) and alumina
column chromatography on Merck aluminium oxide 90.
2,4-Bis(4-chlorophenyl)-6-cyanoaniline. Colourless prisms
(EtOH); yield 79%; mp 126–128 ЊC; ν (KBr)/cmϪ1 3480 and
3390 (NH2), 2220 (CN); δH (CDCl3) 4.56 (s, NH2, 2H), 7.37–
7.50 (m, ArH, 9H), 7.61 (d, J 2.4, ArH, 1H) (Found: C, 67.02;
H, 3.61; N, 8.09. C19H12Cl2N2 requires: C, 67.27; H, 3.57; N,
8.26%).
2,4-Di(phenyl-d5)-6-cyanoaniline. Colourless prisms (EtOH);
yield 76%; mp 91–93 ЊC; ν (KBr)/cmϪ1 3470, 3440, 3370 and
3330 (NH2), 2220 (CN); δH (CDCl3) 4.57 (s, NH2, 2H), 7.55 (d,
J 2.4, ArH, 1H), 7.64 (d, J 2.4, ArH, 1H) (Found: C, 81.26; H,
5.07; N, 9.85. C21H4D10N2 requires: C, 81.39; H, 5.03; N,
9.99%).
2,4-Bis(4-chlorophenyl)-6-fluoroaniline.
Brown
needles
(EtOH); yield 83%; mp 97–99 ЊC; ν (KBr)/cmϪ1 3440 and 3300
(NH2); δH (CDCl3) 3.88 (s, NH2, 2H), 7.12 (d, J 2.4, ArH, 1H),
7.35–7.49 (m, ArH, 9H) (Found: C, 65.24; H, 3.87; N, 3.99.
C18H12Cl2FN requires: C, 65.08; H, 3.64; N, 4.22%).
Bromination of 2-ethoxycarbonyl-, 2-acetyl-, 2-cyano- and
2-fluoroanilines were carried out by treating 2-substituted
anilines (40 mmol) with benzyltrimethylammonium tribrom-
ide10 (BTMA Br3) (39.0 g, 100 mmol) in CH2Cl2 (250 cm3)–
MeOH (100 cm3) at room temperature for 1 h in the presence
of CaCO3 (11.2 g). The usual workup and crystallization
(EtOH) gave 2-(ethoxycarbonyl)-4,6-dibromoaniline (colour-
less needles; mp 72–74 ЊC) in 89%, 2-acetyl-4,6-dibromoaniline
(yellow needles, mp 129–131 ЊC) in 93%, 2-cyano-4,6-
dibromoaniline (colourless needles, mp 156–158 ЊC) in 92%,
and 2-fluoro-4,6-dibromoaniline (colourless needles, mp 60–
63 ЊC) in 89% yields. (Phenyl-d5)boronic acid was obtained by
the reported method.20 2,4-Diphenylaniline was obtained by
our previous method.12 Di-tert-butyl diperoxyoxalate was
obtained by the reported method.13
General procedure for preparation of N-(arylthio)-2,4-diaryl-6-
R-anilines (R ؍ COOEt, COMe, CN and F)
To a stirred solution of 2,4-diaryl-6-R-anilines (6.30 mmol) and
Et3N (1.46 g, 14.43 mmol) in dry ether (or anhydrous THF)
(200 cm3), was added at 0 ЊC a solution of arenesulfenyl chlor-
ide (7.8 mmol) in dry ether (or anhydrous THF) (20 cm3). After
being stirred for 2 h at 0 ЊC, the mixture was filtered and con-
centrated under reduced pressure, and the residue was chroma-
tographed on alumina with 1 : 5 benzene–hexane (9), 1 : 1
benzene–hexane (6b, 7, 8b, 8d, 8b-d), or 1 : 2 hexane–benzene
(6a, 8a, 8c, 10) and crystallized from appropriate solvents.
Compound 9 was obtained as an oil.
N-[(4-Nitrophenyl)thio]-2,4-bis(4-chlorophenyl)-6-ethoxycarb-
onylaniline 6a. Light yellow needles (EtOH–benzene); yield
48%; mp 179–181 ЊC; ν (KBr)/cmϪ1 3240 (NH), 1680 (C᎐O); δ
General procedure for preparation of 2,4-diaryl-6-R-anilines
(R ؍ COOEt, COMe, CN and F)
᎐
H
(CDCl3) 1.41 (t, J 6.8, CH3, 3H), 4.39 (q, J 6.8, CH2, 2H), 7.44–
7.55 (m, ArH, 8H), 7.66 (d, J 2.4, ArH, 1H), 7.68 (d, J 8.8, ArH,
2H), 8.33 (d, J 2.0, ArH, 1H), 8.42 (d, J 8.8, ArH, 2H), 9.64 (s,
NH, 1H) (Found: C, 60.14; H, 3.79; N, 5.13. C27H20Cl2N2O4S
requires: C, 60.12; H, 3.74; N, 5.19%).
A mixture of 2-R-4,6-dibromoanilines (25 mmol), arylboronic
acid (75 mmol), (PPh3)4Pd (1.73 g, 1.50 mmol) and K2CO3 (27.6
g) in benzene (250 cm3)–EtOH (50 cm3)–H2O (100 cm3) was
refluxed for 15 h under a nitrogen atmosphere.11,12 The organic
layer was extracted, washed with brine, dried (MgSO4), and
concentrated under reduced pressure. The residue was then
chromatographed on silica gel with 1 : 5 hexane–benzene [2,4-
bis(4-chlorophenyl)-6-fluoroaniline], CH2Cl2 [2,4-bis(4-chloro-
phenyl)-2-cyanoaniline] or benzene (the other anilines) and
crystallized from appropriate solvents.
N-[(2,4-Dichlorophenyl)thio]-2,4-bis(4-chlorophenyl)-6-eth-
oxycarbonylaniline 6b. Colourless prisms (EtOH); yield 40%;
mp 89–91 ЊC; ν (KBr)/cmϪ1 3260 (NH), 1680 (C᎐O); δ (CDCl )
᎐
H
3
1.46 (t, J 7.2, CH3, 3H), 4.45 (q, J 7.2, CH2, 2H), 7.09–7.12 (m,
ArH, 3H), 7.15–7.23 (m, ArH, 3H), 7.38–7.50 (m, ArH, 6H),
8.15 (d, J 2.4, ArH, 1H), 8.85 (s, NH, 1H) (Found: C, 57.89; H,
3.32; N, 2.40. C27H19Cl4NO2S requires: C, 57.57; H, 3.40; N,
2.49%).
2,4-Bis(4-chlorophenyl)-6-ethoxycarbonylaniline. Colourless
needles (EtOH); yield 90%; mp 120–122 ЊC; ν (KBr)/cmϪ1 3480
and 3360 (NH ), 1690 (C᎐O); δ (CDCl3) 1.42 (t, J 7.0, CH3,
᎐
2
H
3H), 4.38 (q, J 7.0, CH2, 2H), 6.02 (br s, NH2, 2H), 7.35–7.50
(m, ArH, 9H), 8.14 (d, J 2.0, ArH, 1H) (Found: C, 65.25; H,
4.38; N, 3.55. C21H17Cl2NO2 requires: C, 65.30; H, 4.44; N,
3.63%).
N-[(2,4-Dichlorophenyl)thio]-2,4-diphenyl-6-acetylaniline 7.
Yellow plates (EtOH); yield 42%; mp 136–138 ЊC; ν (KBr)/cmϪ1
3240 (NH), 1650 (C᎐O); δ (CDCl ) 2.77 (s, Me, 3H), 7.04–7.57
᎐
H
3
(m, ArH, 14H), 7.97 (d, J 2.4, ArH, 1H), 9.29 (s, NH, 1H)
742
J. Chem. Soc., Perkin Trans. 2, 2001, 738–744

View Article Online
(Found: C, 67.34; H, 4.35; N, 3.06. C26H19Cl2NOS requires: C,
67.24; H, 4.12; N, 3.02%).
2220, 1570, 1520, 1340, 860, 700 (Found: C, 71.23; H, 4.00; N,
10.05. C25H16N3O2S requires: C, 71.07; H, 3.82; N, 9.95%).
N-[(4-Nitrophenyl)thio]-2,4-diphenyl-6-cyanoaniline 8a. Yel-
low needles (EtOH–benzene); yield 37%; mp 173–175 ЊC;
ν (KBr)/cmϪ1 3370 (NH), 2220 (CN); δH (CDCl3) 5.72 (s, NH,
1H), 7.32–7.62 (m, ArH, 13H), 7.82 (d, J 2.4, ArH, 1H), 8.17
(d, J 8.8, ArH, 2H) (Found: C, 70.98; H, 4.04; N, 9.91.
C25H17N3O2S requires: C, 70.90; H, 4.05; N, 9.92%).
N-[(2,4-Dichlorophenyl)thio]-2,4-diphenyl-6-cyanophenyl-
aminyl 3b. Dark green needles; yield 37%; mp 111–113 ЊC; λmax
(benzene)/nm 668 (ε/dm3 molϪ1 cmϪ1 11800), 450 (4400sh), 386
(25300εmax), 370 (20700sh); ν (KBr)/cmϪ1 2920, 2850, 2210,
1740, 1570, 1460, 760, 690 (Found: C, 67.05; H, 3.56; N, 6.13.
C25H15Cl2N2S requires: C, 67.27; H, 3.39; N, 6.28%).
N-[(2,4-Dichlorophenyl)thio]-2,4-diphenyl-6-cyanoaniline 8b.
Colourless prisms (EtOH); yield 52%; mp 127–129 ЊC; ν (KBr)/
cmϪ1 3350 (NH), 2220 (CN); δH (CDCl3) 5.62 (s, NH, 1H),
7.13–7.57 (m, ArH, 14H), 7.81 (d, J 2.5, ArH, 1H) (Found: C,
67.33; H, 3.36; N, 6.46. C25H16Cl2N2S requires: C, 67.11; H,
3.60; N, 6.26%).
N-[(4-Nitrophenyl)thio]-2,4-bis(4-chlorophenyl)-6-cyanophen-
ylaminyl 3c. Dark green needles; yield 41%; mp 151–153 ЊC;
λmax (benzene)/nm 667 (ε/dm3 molϪ1 cmϪ1 6700), 480 (4700sh),
412 (17700εmax), 375 (13700sh); ν (KBr)/cmϪ1 3100, 2920, 2200,
1580, 1520, 1340, 1100, 1010, 860, 820, 740 (Found: C, 61.25;
H, 2.72; N, 8.66. C25H14Cl2N3O2S requires: C, 61.11; H, 2.87; N,
8.55%).
N-[(4-Nitrophenyl)thio]-2,4-bis(4-chlorophenyl)-6-cyano-
aniline 8c. Yellow prisms (EtOH–benzene); yield 39%; mp 193–
195 ЊC; ν (KBr)/cmϪ1 3370 (NH), 2230 (CN); δH (CDCl3) 5.65
(s, NH, 1H), 7.30–7.50 (m, ArH, 10H), 7.53 (d, J 2.0, ArH, 1H),
7.78 (d, J 2.0, ArH, 1H), 8.17 (d, J 8.8, ArH, 2H) (Found: C,
61.11; H, 3.25; N, 8.69. C25H15Cl2N3O2S requires: C, 60.98; H,
3.07; N, 8.53%).
N-[(2,4-Dichlorophenyl)thio]-2,4-bis(4-chlorophenyl)-6-cyano-
phenylaminyl 3d. Dark green needles; yield 39%; mp 133–
135 ЊC; λmax (benzene)/nm 673 (ε/dm3 molϪ1 cmϪ1 9600), 472
(4800sh), 392 (19300εmax), 370 (16500sh), 331 (11200sh);
ν (KBr)/cmϪ1 3080, 2920, 2210, 1570, 1490, 1440, 1090, 1020,
820 (Found: C, 58.17; H, 2.57; N, 5.21. C25H13Cl4N2S requires:
C, 58.27; H, 2.54; N, 5.44%).
N-[(2,4-Dichlorophenyl)thio]-2,4-bis(4-chlorophenyl)-6-cyano-
aniline 8d. Colourless prisms (EtOH); yield 47%; mp 108–
110 ЊC; ν (KBr)/cmϪ1 3370 (NH), 2230 (CN); δH (CDCl3) 5.56
(s, NH, 1H), 7.01–7.32 (m, ArH, 4H), 7.40–7.76 (m, ArH, 7H),
7.48 (d, J 2.4, ArH, 1H), 7.75 (d, J 2.4, ArH, 1H) (Found: C,
58.17; H, 2.57; N, 5.21. C25H14Cl4N2S requires: C, 58.16; H,
2.73; N, 5.43%).
Isolation and identification of 11
Precursor 6a (100 mg, 0.180 mmol) was dissolved in 10 cm3 of
benzene with stirring. After 1.0 g of K2CO3 was added, 2.0 g
of PbO2 was added in several portions during 2.0 min and
the resulting mixture was further stirred for 0.5 min. After fil-
tration, the filtrate was evaporated and the residue was chrom-
atographed on alumina with 1 : 1 benzene–hexane to give 11
in 43% yield. Recrystallization from hexane gave colourless
needles. Mp 138 ЊC (decomp.); m/z 553 (M+ Ϫ 1, 100); ν (KBr)/
cmϪ1 3190 (NH), 1710 (C᎐O), 1240 (SO); δ (CDCl3) 1.37 (t,
N-[(2,4-Dichlorophenyl)thio]-2,4-di(phenyl-d5)-6-cyanoaniline
8b-d. Colourless needles (EtOH); yield 36%; mp 141–143 ЊC;
ν (KBr)/cmϪ1 3270 (NH), 2230 (CN); δH (CDCl3) 5.62 (s, NH,
1H), 7.21 (dd, J 8.3 and 2.0, ArH, 1H), 7.30 (d, J 2.0, ArH, 1H),
7.31 (d, J 8.3, ArH, 1H), 7.57 (d, J 2.0, ArH, 1H), 7.81 (d, J 2.0,
ArH, 1H) (Found: C, 65.88; H, 3.62; N, 6.18. C25H6D10Cl2N2S
requires: C, 65.64; H, 3.53; N, 6.12%).
᎐
H
J 7.3, CH3, 3H), 4.45 (q, J 7.3, CH2, 2H), 7.42–7.44 (m, ArH,
8H), 7.50 (d, J 8.8, ArH, 2H), 7.57 (d, J 2.0, ArH, 1H), 8.19 (d,
J 8.8, ArH, 2H), 8.22 (d, J 2.0, ArH, 1H), 9.64 (s, NH, 1H)
(Found: C, 58.37; H, 3.86; N, 4.87. C27H20Cl2N2O5S requires: C,
58.39; H, 3.63; N, 5.04%).
N-[(2,4-Dichlorophenyl)thio]-2,4-bis(4-chlorophenyl)-6-fluoro-
aniline 9. Brown semi-solid; yield 52%; ν (KBr)/cmϪ1 3300
(NH); δH (CDCl3) 5.49 (s, NH, 1H), 7.17 (dd, J 8.4 and 1.9,
ArH, 1H), 7.24 (d, J 1.9, ArH, 1H), 7.27–7.40 (m, ArH, 10H),
7.56 (d, J 8.4, ArH, 1H); m/z (FAB) 506.9598 (C24H14Cl4FNS
requires: 506.9585).
Isolation and identification of 12
Precursor 7 (200 mg, 0.43 mmol) was dissolved in 10 cm3 of
benzene with stirring. After 2.0 g of K2CO3 was added, 6.0 g
of PbO2 was added in several portions during 2.0 min and
the resulting mixture was further stirred for 1 min. After
filtration, the filtrate was evaporated and the residue was
chromatographed on alumina with benzene to give 12 in
38%. Recrystallization from hexane–benzene gave brown
prisms. Mp 58–60 ЊC; m/z 479 (M+ Ϫ 1, 100); ν (KBr)/cmϪ1
N-[(4-Nitrophenyl)thio]-2,4-diphenylaniline 10. Yellow needles
(EtOH–benzene); yield 48%; mp 130–132 ЊC; ν (KBr)/cmϪ1
3410 and 3380 (NH); δH (CDCl3) 5.49 (s, NH, 1H), 7.29–7.57
(m, ArH, 15H), 8.16 (d, J 8.8, ArH, 2H) (Found: C, 71.98; H,
4.43; N, 7.03. C24H18N2O2S requires: C, 72.34; H, 4.55; N,
7.03%).
3480 (NH), 1650 (C᎐O), 1240 (SO); δ (CDCl3) 2.71 (s, CH3,
᎐
H
3H), 7.29 (d, J 2.0, ArH, 1H), 7.39–7.49 (m, ArH, 7H), 7.55–
7.60 (m, ArH, 4H), 7.67 (d, J 2.0, ArH, 1H), 7.82 (d, J 2.0, ArH,
1H), 8.03 (d, J 2.0, ArH, 1H), 10.37 (s, NH, 1H) (Found: C,
65.21; H, 4.11; N, 3.05. C26H19Cl2NO2S requires: C, 65.00; H,
3.99; N, 2.92%).
Isolation of N-(arylthio)-2,4-diphenyl-6-cyanophenylaminyls
Precursor 8 (100 mg) was dissolved in benzene (20 cm3) and
K2CO3 (2–3 g) was added, and the resulting mixture was
vigorously stirred. PbO2 (2.0–3.0 g) was then added in several
portions during 2.0 min and stirring was continued for an
additional 0.5–1.0 min. The dark green mixture was filtered,
the solvent was removed by freeze-drying, and the resulting
dark green crystalline residue was crystallized from EtOH–
EtOAc.
Product analysis for decomposition of 4
Precursor 9 (100 mg, 0.20 mmol) was dissolved in 10 cm3 of
benzene with stirring. After 1.0 g of K2CO3 was added, 2.0 g of
PbO2 was added in several portions during 2.0 min and the
resulting mixture was further stirred for 0.5 min. After filtra-
tion, the brown filtrate was evaporated and the residue was
chromatographed on alumina with 1 : 9 benzene–hexane to give
13 as a dark brown solid in 41% yield. Recrystallization from
EtOH gave brown needles with mp >300 ЊC; m/z (FAB) 816
(M+); δH (CDCl3) 6.66 (d, J 2.0, ArH, 1H), 7.01 (d, J 2.0, ArH,
N-[(4-Nitrophenyl)thio]-2,4-diphenyl-6-cyanophenylaminyl 3a.
Dark green needles; yield 47%; mp 123–125 ЊC; λmax (benzene)/
nm 658 (ε/dm3 molϪ1 cmϪ1 12100), 485 (6400sh), 406
(19200εmax), 375 (15200sh), 338 (10700sh); ν (KBr)/cmϪ1 3300,
J. Chem. Soc., Perkin Trans. 2, 2001, 738–744
743

View Article Online
5 Magnetic Properties of Organic Materials, ed P. M. Lahti, Marcel
Dekker, New York, 1999.
6 Proceedings of the 5th International Conference on Molecule-Based
Magnets, Mol. Cryst. Liq. Cryst., 1999, 334, 1 and 335, 1.
7 Y. Teki, K. Itoh, A. Okada, H. Yamakage, T. Kobayashi, K. Amaya,
S. Kurokawa, S. Ueno and Y. Miura, Chem. Phys. Lett., 1997, 270,
573.
8 Y. Miura, M. Momoki, M. Nakatsuji and Y. Teki, J. Org. Chem.,
1998, 63, 1555.
9 Y. Miura, S. Kurokawa, M. Nakatsuji, K. Ando and Y. Teki, J. Org.
Chem., 1998, 63, 8295.
10 S. Kajigaeshi, T. Kakinami, K. Inoue, M. Kondo, H. Nakamura, M.
Fujikawa and T. Okamoto, Bull. Chem. Soc. Jpn., 1988, 61, 597.
11 N. Miyaura and A. Suzuki, Chem. Rev., 1995, 95, 2457.
12 Y. Miura, H. Oka and M. Momoki, Synthesis, 1995, 1419.
13 P. D. Bartlett, E. P. Benzing and R. E. Pincock, J. Am. Chem. Soc.,
1960, 82, 1762.
14 Y. Miura, A. Yamamoto and M. Kinoshita, Bull. Chem. Soc. Jpn.,
1981, 54, 3215.
15 L. Benati, P. C. Montevecchi and P. Spagnolo, J. Chem. Soc., Perkin
Trans. 1, 1982, 3049.
1H), 7.13 (dd, J 8.8 and 2.0, ArH, 1H), 7.28 (d, J 8.8, ArH, 2H),
7.31 (d, J 2.0, ArH, 1H), 7.36–7.50 (m, ArH, 15H), 7.59 (d,
J 8.8, ArH, 2H) (Found: C, 61.66; H, 2.71; N, 3.39. C42H23-
Cl6FN2S requires: C, 61.56; H, 2.83; N, 3.42%).
Product analysis for decomposition of 5
Precursor 10 (100 mg, 0.25 mmol) was dissolved in 10 cm3 of
benzene with stirring. After 1.0 g of K2CO3 was added, 2.0 g
of PbO2 was added in several portions during 2.0 min and
the resulting mixture was further stirred for 0.5 min. After fil-
tration, the filtrate was evaporated and the residue was chrom-
atographed on alumina with 1 : 1 benzene–hexane to give 14 as
a dark brown solid in 38% yield. Recrystallization from EtOH
gave brown needles with mp 191–193 ЊC; m/z (FAB) 640
(M+ + 1); δH (CDCl3) 6.93 (d, J 2.4, ArH, 1H), 7.11–7.13 (m,
ArH, 2H), 7.21–7.59 (m, ArH, 20H), 7.67–7.72 (m, ArH, 3H),
7.79 (d, J 2.4, ArH, 1H), 8.76 (d, J 9.3, ArH, 2H) (Found: C,
78.66; H, 4.57; N, 6.46. C42H29N3O2S requires: C, 78.85; H,
4.57; N, 6.57%).
16 C. Balboni, L. Benati, P. C. Montevecchi and P. Spagnolo, J. Chem.
Soc., Perkin Trans. 1, 1983, 2111.
17 Gaussian 98, Revision A. 7, M. J. Frisch, G. W. Trucks, H. B.
Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G.
Zakrzewski, J. A. Montgomery, Jr., R. E. Stratmann, J. C. Burant,
S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C.
Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi,
B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A.
Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D.
Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz,
A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith,
M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C. Gonzalez,
M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong,
J. L. Andres, M. Head-Gordon, E. S. Replogle and J. A. Pople,
Gaussian, Inc., Pittsburgh PA, 1998.
18 J. C. Bonner and M. E. Fischer, Phys. Rev., 1964, A135, 640.
19 (a) W. Duffy, Jr. and K. P. Barry, Phys. Rev., 1968, 165, 647; (b) J. W.
Hall, W. March, R. R. Weller and W. E. Hatfield, Inorg. Chem.,
1981, 20, 1033.
20 Y. Miura, M. Momoki, T. Fuchikami, Y. Teki, K. Itoh and
H. Mizutani, J. Org. Chem., 1996, 61, 4300.
References
1 Part 52. Y. Miura, T. Tomimura and Y. Teki, J. Org. Chem., 2000, 65,
7889.
2 Y. Miura in Recent Research Developments in Organic Chemistry,
Transworld Research Network, India, 1998, vol. 2, part II, p. 251.
3 J. M. Farrar, M. K. Patel, P. Kaszynski and V. G. Young, J. Org.
Chem, 2000, 65, 931; G. D. McManus, J. M. Rawson, N. Feeder,
F. Palacio and P. Oliete, J. Mater. Chem., 2000, 10, 2001; T. M.
Barclay, A. W. Cordes, R. C. Haddon, M. E. Itkis, R. T. Oakley,
R. W. Reed and H. Zhang, J. Am. Chem. Soc., 1999, 121, 969, and
references cited therein.
4 A. R. Forrester, J. M. Hay and H. R. Thomson, Organic Chemistry
of Stable Free Radicals, Academic Press, London and New York,
1968; E. G. Rozantsev, Free Nitroxyl Radicals, Plenum Press, New
York and London, 1970; L. B. Volodarsky, V. A. Reznikov and V. I.
Ovcharenko, Synthetic Chemistry of Stable Nitroxides, CRC Press,
Boca Raton, 1994.
744
J. Chem. Soc., Perkin Trans. 2, 2001, 738–744