Magnetic anisotropy parameters of matrix-isolated septet 1,3,5-trinitreno-2,4,6-trichlorobenzene

Article abstract of DOI:10.1007/s11172-012-0315-z

An ESR spectrum of high-symmetry septet 1,3,5-trinitreno-2,4,6- trichlorobenzene generated under photolysis of 1,3,5-triazido-2,4,6- trichlorobenzene in solid argon at 15 K was recorded. Computer simulation revealed that the spectrum corresponds to the se

Full text of DOI:10.1007/s11172-012-0315-z

Russian Chemical Bulletin, International Edition, Vol. 61, No. 12, pp. 2218—2224, December, 2012
2218
Magnetic anisotropy parameters
of matrixꢀisolated septet 1,3,5ꢀtrinitrenoꢀ2,4,6ꢀtrichlorobenzene
E. Ya. Misochko, A. V. Akimov, A. A. Mazitov, D. V. Korchagin, and S. V. Chapyshev
Institute of Problems of Chemical Physics, Russian Academy of Sciences,
1 prosp. Akad. Semenova, 142432 Chernogolovka, Moscow Region, Russian Federation.
Eꢀmail: misochko@icp.ac.ru
An ESR spectrum of highꢀsymmetry septet 1,3,5ꢀtrinitrenoꢀ2,4,6ꢀtrichlorobenzene generꢀ
ated under photolysis of 1,3,5ꢀtriazidoꢀ2,4,6ꢀtrichlorobenzene in solid argon at 15 K was reꢀ
corded. Computer simulation revealed that the spectrum corresponds to the septet spin state
with the fine structure parameters DS = –0.0957 0.0006 cm–1 and ES = 0 0.0004 cm–1. These
values of the magnetic anisotropy parameters DS and ES are in good agreement with the results
of UDFT calculations. The spinꢀspin (DSS) and spinꢀorbit (DSO) coupling parameters of septet
molecules with D3h symmetry are negative and mutually enhance the magnetic anisotropy of
these molecules. The contribution of the spinꢀorbit coupling to the magnetic anisotropy of
1,3,5ꢀtrinitrenoꢀ2,4,6ꢀtrichlorobenzene is higher than 11% due to the presence of three chlorꢀ
ine atoms in the molecule. This suggests the possibility of further strengthening the magnetic
properties of septet 1,3,5ꢀtrinitrenobenzenes by introducing bromine and iodine atoms into
positions 2, 4, and 6 of their benzene rings.
Key words: azides, nitrenes, photolysis, matrix isolation, ESR spectroscopy, highꢀspin states.
Highꢀspin nitrene molecules represent ferromagnetic
solutions to the magnetic spinꢀHamiltonian (1) makes it
possible to calculate the magnetic anisotropy parameter
values with high accuracy.
clusters bearing two or more unpaired electrons. Recently,
highꢀspin nitrenes are used as model systems in studies
of intramolecular magnetic interactions that play an
important role in molecular magnetism, spintronics,
and spin chemistry.1—5 These interactions are responsible
Among organic hexaradicals, septet trinitrenes are
characterized by the largest values of the magnetic anisoꢀ
tropy parameters |DS|. According to theoretical estimates,10
the energy of ferromagnetic exchange interaction between
six unpaired electrons is ~3—10 kcal mol–1 which ensures
a stable septet state (S = 3) even at room temperature.
^
for the parameters of the fine structure tensor (D) meaꢀ
sured experimentally by ESR spectroscopy. As applied
to the problems in question, investigations of highꢀ
spin molecules are of importance for progress in theoꢀ
retical chemistry which needs highꢀprecision experiꢀ
mental data for testing new quantum chemical
methods of calculations in order to establish their preꢀ
dictive power in studies of complex magnetic systems.
The first ESR spectra of highꢀspin organic molecules
were recorded in the late 1960s.6 Nevertheless, reliable
methods of analysis of ESR spectra of powder samꢀ
ples with a complex pattern of ESR transitions between
Zeeman energy levels under conditions where the paraꢀ
meters of zeroꢀfield splitting of the magnetic spinꢀHamilꢀ
tonian
^
Recently, parameters of the DS tensor were determined
using the ESR spectra of some septet trinitrenes isolated
in a solid argon matrix.11—15 According to quantum chemꢀ
ical calculations,3,4 the major contribution to the scalar
^
parameters of the DS tensor (DS and ES) comes from diꢀ
poleꢀdipole spinꢀspin interactions between two unpaired
electrons localized on the same nitrene center (soꢀcalled
oneꢀcenter interactions), whereas spinꢀspin interactions
between unpaired electrons belonging to different nitrene
atoms (twoꢀcenter interactions) cause |DS| to decrease by
^
about 30%. Parameters of the DS tensor of septet trinitrenes
are described with reasonable accuracy within the frameꢀ
work of the semiempirical model of three weakly interactꢀ
^
^
H = gHS + SDSS
(1)
ing triplet centers6,14—16
:
^
^
^
^
^
^
(H is the magnetic field strength, DS is the zeroꢀfield splitꢀ
ting tensor, S is the total spin,  is the electron Bohr
magneton, and g is the electron gꢀfactor) are comparable
with the energy of RFꢀquanta were developed only in the
last decade.7—9 An analysis of ESR spectra using exact
DS = (1/15)(DT1 + DT2 + DT3) + (1/15)Dij,
(2)
^
where DTi are the tensors of the triplet nitrene centers and
^
the Dij tensor (i, j = 1, 2, 3) characterizes the interaction
between the triplet centers.
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 12, pp. 2199—2205, December, 2012.
1066ꢀ5285/12/6112ꢀ2218 © 2012 Springer Science+Business Media, Inc.

Matrixꢀisolated septet trinitrenobenzene
Russ.Chem.Bull., Int.Ed., Vol. 61, No. 12, December, 2012
2219
Scheme 1
Molecules 1—4 with D3h symmetry represent a particꢀ
ular type of septet trinitrenes. According to relation (2), in
these molecules, the easy magnetization axis (Z axis of the
^
DS tensor) is directed normal to the molecular plane and
the DS parameter is negative, which is a key property of
^
molecular magnets.6,16 The ES parameter of the DS tensor
characterizes the nonequivalence of the other two compoꢀ
nents; it is equal to zero because of high symmetry of the
molecules.
Note. From this point on, the D and E values are given in cm–1
.
erties. Therefore, the ESR spectra and magnetic parameꢀ
ters of trinitrenes 1 and 4 are expected to be similar. Our
study was aimed to detect and analyze the ESR spectrum
of trinitrene 4, to determine the DS and ES parameters, to
carry out a theoretical analysis of the main electronic interꢀ
actions in the molecule, and to reveal factors governꢀ
ing the magnetic anisotropy in the septet aryltrinitrene
molecules.
Experimental data on the magnetic parameters of molꢀ
ecules 1—4 are scarce. An ESR spectrum of trinitrene 1
was reported;6 however, the ESR spectrum per se was
not presented and the |D| parameter was estimated as
0.055 cm–1 using the resonance magnetic fields for three
observed ESR lines and an approximate solutions to the
spinꢀHamiltonian (1). Then, a highꢀresolution ESR specꢀ
trum of septet trinitrene 2 isolated in solid nitrogen was
described.11 The |D| parameter determined using this specꢀ
trum appeared to be equal to 0.123 cm–1, being 2.2 times
higher than that of trinitrene 1. More recently,14 an ESR
spectrum of trinitrene 5 isolated in solid argon was recordꢀ
ed. Compound 5 can be treated as a close analog of
a symmetric molecule of trinitrenobenzene 3. The |D| paraꢀ
meter of 5 is equal to 0.094 cm–1 and lies between those of
trinitrenes 1 and 2. Strongly different values of the magꢀ
netic anisotropy parameters of trinitrenes 1, 2, and 5 gave
us an impetus to study paramagnetic products of photoꢀ
lysis of triazide 6 (Scheme 1) by matrix isolation ESR
spectroscopy.
Experimental
IR spectrum was recorded with a Perkin—Elmer Spectrum
100 FTꢀIR instrument, UV spectrum was measured with
a Specord Mꢀ80 spectrophotometer, and 13C NMR spectrum
was recorded with a Bruker Avance III spectrometer operating
at 125 MHz. Mass spectrum was obtained on a Kratos MSꢀ30
instrument with a direct inlet system (ionization energy 70 eV).
Elemental analysis was performed on an Elementar Vario Microꢀ
cube Elemental Analyzer CHNS/O. Commercially available
1,3,5ꢀtrifluoroꢀ2,4,6ꢀtrichlorobenzene (Sigma—Aldrich) was
used as is.
1,3,5ꢀTriazidoꢀ2,4,6ꢀtrichlorobenzene (6). A solution of
1,3,5ꢀtrifluoroꢀ2,4,6ꢀtrichlorobenzene (2.35 g, 10 mmol) and
sodium azide (2.60 g, 40 mmol) in DMSO (20 mL) was stirred
for 10 h at 60 C and the reaction mixture was cooled to ~20 C
and poured in water (500 mL). The precipitate was filtered off,
washed on the filter with water, dried in air, and recrystallized
from EtOH. The yield was 2.92 g (96%), m.p. 93–94 C.
The most probable paramagnetic products of this
photochemical reaction should include triplet nitrene 7,
quintet dinitrene 8, and septet trinitrene 4. Molecule 4
and trinitrene 1 have the same symmetry and substituents
in their benzene rings are similar in electronꢀseeking propꢀ

2220
Russ.Chem.Bull., Int.Ed., Vol. 61, No. 12, December, 2012
Misochko et al.
IR spectrum (polycrystalline powder), /cm–1: 2135, 2110 (N3),
1528, 1411 (C=C, C=N), 1394, 1322 (N3), 1254 (C—N), 1051,
955, 793, 728. Electronic absorption spectrum (EtOH), max/nm
(log ): 202 (4.30), 250 (4.55), 272 (3.89). 13C NMR spectrum
(CDCl3, ): 121.3 (C—Cl), 134.8 (C—N3). Mass spectrum, m/z
(Irel (%)): 305 [M]+ (100). Found (%): C, 23.76; N, 41.29.
C6Cl3N9. Calculated (%): C, 23.67; N, 41.40.
lysis of the sample at 15 K. Impurity paramagnetic centers present
in the sapphire rod show a number of narrow EPR lines. The EPR
spectrum of the pure sapphire rod was recorded preliminarily;
the corresponding lines are asterisked in the experimental specꢀ
trum (Fig. 1, a). The microwave power and modulation ampliꢀ
tude ensured no saturation effects in the ESR spectra recorded.
Computer simulation of ESR spectra was carried out using
the EasySpin program package17 based on the exact numerical
diagonalization of the spin Hamiltonian (1).
Matrix isolation and ESR spectroscopy. The base unit of the
experimental setup for cryogenic measurements and the proceꢀ
dure for ESR measurements were described in detail elseꢀ
where.7,12 Thin argon films were grown by vacuum coꢀcondenꢀ
sation of two gas beams (argon and triazide 6) on the substrate of
a helium cryostat cooled to 15 K. The molecular beams were coꢀ
deposited on the lower end of a flat sapphire rod from two spaꢀ
tially separated nozzles. The diazide molecules were generated
by sublimation of diazide powder placed in a quartz tube heated
to ~87 C with an external heater. The coꢀdeposition regime was
chosen preliminarily and provided an optimum deposition rate
of molecules 6 in an Ar : 6  (103—104) : 1 ratio. The samples
were at most 150 m thick. After sample preparation the lower
end of the sapphire rod was lowered to the center of the EPR
spectrometer cavity. Photolysis at 15 K was carried out through
a window in the cavity with light of a lowꢀpressure mercury lamp
(Mercury Spectral Line Lamp, LotꢀOriel) passed through an
interference filter (LotꢀOriel,  = 297 nm, FWHM ~5 nm).
ESR spectra were recorded with a Radiopan Xꢀband spectroꢀ
meter (Poland). A window in the cavity permitted onꢀsite photoꢀ
Quantum chemical calculations. The geometry of molecules
with the total spin S = 3 was optimized within the density funcꢀ
tional theory (B3LYP/6ꢀ311G(d) approximation) using the
Gaussian program.18 The nature of stationary points was deterꢀ
mined by analyzing the results of normal coordinate analysis
(Hesse force constant matrix). Parameters of magnetic interacꢀ
tions including both spinꢀspin (SS) and spinꢀorbit (SO) coupling
were obtained from singleꢀpoint calculations based on the
B3LYP/6ꢀ311G(d)ꢀcalculated geometries using the ORCA proꢀ
gram package19 with the PBE functional20 and the AhlrichsꢀDZ
basis set.21 The contributions of the SSꢀ and SOꢀcoupling to
^
parameters of the DS tensor were calculated by the McWeeny—
Mizuno22 and Pederson—Khanna23 methods, respectively, imꢀ
plemented in the ORCA program package.19 These methods
provide good agreement between experimental data and calcuꢀ
^
lated parameters of the D tensors for a number of highꢀspin
^
nitrenes.3 The DS tensor was calculated by summation of the
^
^
^
spinꢀspin and spinꢀorbit coupling tensors: DS = DSS + DSO. Scaꢀ
a
*
*
b
Z1
XY2
XY1
XY3
XY4
A
Z2
A
Z3
A
A
A
A
200
400
600
800
H/mT
Fig. 1. (a) Experimental ESR spectrum of photolysis products of triazide 6 in argon matrix at 15 K; the signals of paramagnetic centers
in the sapphire rod used as substrate are asterisked. (b) Theoretical ESR spectrum of the septet trinitrene with the magnetic anisotropy
parameters DS = 0.0957 cm–1 and ES = 0 cm–1. Arrows show positions of the ESR lines in the experimental spectrum, which were used
for optimization of the magnetic parameters of the trinitrene (see text). The microwave frequency is 9.107 GHz. Line assignment for
the theoretical spectrum is given in accordance with Figs 2 and 3; here and in Figs 2 and 3, A denotes the extra lines.

Matrixꢀisolated septet trinitrenobenzene
Russ.Chem.Bull., Int.Ed., Vol. 61, No. 12, December, 2012
2221
^
value Rmin = 0.73 mT was obtained at DS = 0.0957 cm–1
,
lar parameters of the DS tensor (DS and ES) were calculated
conventionally:
ES = 0, and g = ge = 2.0023. This value is much smaller
than the spectral line halfwidths and thus points to good
agreement between theory and experiment. Both positions
and relative intensities of resonance lines in the theoretiꢀ
cal spectrum shown in Fig. 1, b agree well with those of
the lines in the experimental spectrum. Also, the theoretiꢀ
cal spectrum (see Fig. 1, b) shows that relatively weak
lines at 277, 609, and 642 mT in the experimental specꢀ
trum also correspond to the septet trinitrene. Line assignꢀ
ment in the spectrum of septet trinitrene was based on the
D = (3/2)Dzz, E = (Dxx – Dyy)/2,
(3)
^
where Dxx, Dyy, and Dzz are the eigenvalues of the D tensor.
S
Results and Discussion
Photolysis and ESR spectroscopy. ESR spectra of the
unphotolyzed samples exhibit no lines from any paramagꢀ
netic species. Photolysis leads to the appearance of a series
of lines with halfwidths ~1—2 mT in a wide range of magꢀ
netic field values. In the initial step of photolysis (~5 min),
the ESR spectra exhibit lines appeared at resonance magꢀ
netic fields of 18, 153, 195, 238, 313, 429, 463, and 498 mT.
Prolonged photolysis leads to an increase in the intensities
of these lines and to the appearance of weak signals at
277, 300, 609, and 642 mT. Under longꢀterm photolysis
(>50 min), the line intensities reach maximum values and
then slowly decrease. The ESR spectrum recorded at maxꢀ
imum line intensities is shown in Fig. 1, a. Earlier,11—15 it
was found that the ESR lines of triplet mononitrenes apꢀ
pear in the region 660—700 mT (x2y2 ESR transitions),
which corresponds to DT  0.95—1.05 cm–1. However, we
failed to detect even weak lines of triplet mononitrene 7 in
this range of magnetic field values. The ESR spectra of the
quintet dinitrenes structurally similar to compound 8 are
characterized by two strong lines (y1 and y2 ESR transiꢀ
tions) in the region 160 and 300 mT.24—26 Typical ESR
spectra of septet trinitrenes with DS  0.1 cm–1 exhibit
a series of five or six strong lines in the region <400 mT;
one or two of them appear in very low (less than 100 mT)
magnetic fields.11—15 These lowꢀfield ESR transitions are
observed due to the fact that one zeroꢀfield splitting value
equal to |3D|  0.3 cm–1 is close to the energy of a microꢀ
wave quantum of the Xꢀband spectrometer.
calculated transitions between Zeeman energy levels for
^
the inꢀprincipalꢀaxis orientations of the D tensor relative
S
to the direction of external magnetic field. From Fig. 2 it
follows that a number of strong spectral lines do not correꢀ
E/ cm–1
a
H || X
H || Y
1
0
–1
–2
b
H || Z
2
1
0
–1
The ESR spectrum shown in Fig. 1, a is similar to such
spectra. Therefore, assuming that the eight strong spectral
lines observed correspond to the septet trinitrene, we deꢀ
termined the parameters DS and ES of the magnetic interꢀ
action tensor by comparing the experimental spectrum
with the theoretical spectra of a powder sample simulated
based on exact solutions to the secular equation for the
spinꢀHamiltonian (1) for the system with the total spin
S = 3. The starting parameters were DS = 0.1 cm–1 and
ES = 0. To obtain the optimum theoretical spectrum, we
minimized the R functional of the rootꢀmeanꢀsquare
deviations of the experimental resonance magnetic
fields from theoretical values by varying the DS and ES
parameters:
с
XY2
XY3
Z1
XY1
XY4
A
Z2
Z3
A
A
A
A
A
200
400
600 H/mT
Fig. 2. Fineꢀstructure energy levels (E) of the septet trinitrene
with the parameters DS = 0.0957 cm–1 and ES = 0 cm–1 plotted
vs. magnetic field for the inꢀprincipalꢀaxis orientations of exterꢀ
nal magnetic field (H || X), (H || Y) (a), and (H || Z) (b). Vertical
labels indicate the allowed ESR transitions; at ES = 0 cm–1, the
,
(4)
^
where n is the number of spectral lines tested (n = 8).
Positions of these lines are labeled in Fig. 1, a. The minimum
X and Y principal axes of the DS tensor are equivalent. (c) Theoꢀ
retical ESR spectrum of powder sample.

2222
Russ.Chem.Bull., Int.Ed., Vol. 61, No. 12, December, 2012
Misochko et al.
spond to transitions in the inꢀprincipalꢀaxis orientations.
These lines are called extra lines. The theoretical plot of
the angular dependence of resonance fields allows one to
assign the extra lines. Analogous dependences for the tranꢀ
sitions responsible for the extra spectral lines at resonance
fields, where (d/dH)  , are presented in Fig. 3.
The accuracy of determination of the DS and ES values
was established using the plots shown in Fig. 4. These
plots illustrate the increase in the R functional value when
the DS and ES parameters are varied independently near
their optimum values. The value Rcr = 1.2 mT at which
the differences between the positions of lines in the experꢀ
imental and theoretical spectra become larger than the
linewidth was chosen as the critical value of the R funcꢀ
tional. The errors in determination of the DS and ES paꢀ
rameters obtained using this procedure are 0.0006 and
0.0004 cm–1, respectively.
a
Y  /2
  /4
Z  0
 = /2
 = 0
  /4
A comparison of the experimental (see Fig. 1, a) and
theoretical (see Fig. 1, b) spectra shows that only a weak
line at 300 mT in the experimental spectrum does not
correspond to the septet trinitrene. This line is labeled in
spectrum 1, a by a triangle. Probably, this is a line of
quintet dinitrene 8 because it is just the region near 300 mT
where the strongest line in the ESR spectra of quintet
dinitrenes are observed.24—26 Note that photolysis of 1,3,5ꢀ
triazidoꢀ2,4,6ꢀtricyanobenzene and 2,4,6ꢀtriazidoꢀ1,3,5ꢀ
triazine also results in the septet trinitrenes 1 and 2 as
major products and that no ESR lines from intermediate
dinitrenes were detected even in the early steps of photoꢀ
lysis. Contrary to this, studies on photolysis of some triꢀ
azidopyridines12,13,15 and 2,4,6ꢀtriazidotoluene14 revealed
successive formation of triplet mononitrenes, quintet diꢀ
nitrenes, and septet trinitrenes.
X  /2
b
A
Z2
A
A
A
A
A
200
400
600 H/mT
Fig. 3. (a) Positions of resonance magnetic fields plotted vs.
angle  at  = /2 (in the zy plane) and at  =  (in the zx plane)
for various transitions, where  and  are the Euler angles beꢀ
tween the principal axes of the DS tensor and the direction of
external magnetic field. In the limiting cases ( = 0 and  = /2)
^
Quantum chemical calculations of magnetic anisotropy
parameters of septet trinitrenes. Table 1 lists the calculatꢀ
ed values of the DS parameters of trinitrenes 1—5. For
nitrenes 2—5, the experimental and theoretical values are
in good agreement. Systematic overestimation of |DS| by
^
this corresponds to the inꢀprincipalꢀaxis orientations of the DS
tensor. Calculations were carried out for the spectrum with the
parameters DS = 0.0957 cm–1 and ES = 0. (b) The calculated
ESR spectrum of powder sample.
Table 1. Theoretical and experimental values of magnetic anisotropy parameters (D/cm–1) and the spin density
values on the nitrene centers () in the septet trinitrenes 1—5
Parametera
1
2
3
4
4Brb
5
с
Dexp
–0.055 d
–0.103
–0.096
–0.126
–0.007
1.48
–0.123 e
–0.130
–0.121
–0.152
–0.009
1.68
—
–0.096 f
–0.105
–0.093
–0.139
–0.012
1.51
.—
–0.242
–0.093
.—
–0.149
.—
–0.094 g
–0.107
–0.100
–0.143
–0.007
1.54, 1.56
DS (UDFT)
–0.109
–0.102
–0.144
–0.007
1.56
DSS (UDFT)
D´SS (UDFT)
DSO (UDFT)
(N) (UDFT)
a Calculated in the PBE/DZ approximation.
b Data taken from Ref. 27 (obtained from PBE/DZ calculations).
c The sign of the DS parameter was not determined based on the ESR spectra. The "minus" sign was chosen in
accordance with theory.
d Data taken from Ref. 6. e Data taken from Ref. 11. f This work. g Data taken from Ref. 14.

Matrixꢀisolated septet trinitrenobenzene
Russ.Chem.Bull., Int.Ed., Vol. 61, No. 12, December, 2012
2223
and size of atoms in the trinitrene molecules increases,
|DS| decreases.
a
R/mT
Figure 5 illustrates the character of spin density deloꢀ
calization in trinitrene 4. The spin density on the nitrene
centers in molecule 4 is only slightly higher than that in
trinitrene 1. Therefore, both trinitrenes are characterized
by very close theoretical |DS| values.
1.4
1.2
1.0
0.8
From the results of calculations it follows that none of
the possible septet trinitrenes can compete with trinitrene 2
in magnitude of spinꢀspin interactions DSS. The second
component of the magnetic anisotropy coming from the
spinꢀorbit coupling, DSO, is very low for trinitrenes 1—3
^
^
and 5. The principal axes of the DSS and DSO tensors
coincide and both DSS and DSO values are negative. The
contribution of the spinꢀorbit coupling to the total |DS|
value is at most 7%. Trinitrene 4 containing chlorine atꢀ
oms is characterized by the maximum |DSO| value for the
septet molecules in question, which exceeds 11%. Even
greater contribution of the spinꢀorbit coupling (~60%) to
the total |DS| value was theoretically predicted for septet
1,3,5ꢀtribromoꢀ2,4,6ꢀtrinitrenobenzene (4Br).27 Density
functional calculations revealed a recordꢀhigh value of the
magnetic anisotropy parameter |DS| for trinitrene 4Br, viz.,
it is twice as large as that of trinitrene 2 (see Table 1).
Summing up, our analysis of experimental data obꢀ
tained by ESR spectroscopy and the results of quantum
chemical calculations suggests the following.
0.0952
0.0956
0.0960
D/cm–1
R/mT
b
2.0
1.6
1.2
0.8
The magnetic anisotropy parameter DS of septet triniꢀ
trenes 1—5 is determined by oneꢀcenter spinꢀspin interacꢀ
tions between the unpaired electrons localized on the niꢀ
trene centers. The highest spin density on the nitrene cenꢀ
ters and the strongest spinꢀspin interactions were found
for the septet trinitrene molecule 2. Compared to 2, septet
1,3,5ꢀtrinitrenobenzenes bear a ~30% lower spin density
on their nitrene centers and, correspondingly, are characꢀ
terized by weaker spinꢀspin interactions. Molecule 4 conꢀ
tains three chlorine atoms and is characterized by an inꢀ
crease in the contribution of the spinꢀorbit coupling to the
total magnetic anisotropy parameter |DS|. The negative
magnetic anisotropy of trinitrenobenzenes can be signifiꢀ
cantly increased by introducing bromine atoms into their
0.0002
0.0004
0.0006 E/cm–1
Fig. 4. The functionals, R(DS) (a) and R(ES) (b), of the rootꢀ
meanꢀsquare deviations of the positions of eight lines in the
calculated EPR spectrum from those in the experimental specꢀ
trum near the optimum values DS = 0.0957 cm–1 and ES = 0.
The horizontal line shows the value of the functional R, which
exceeds the linewidths in the experimental spectrum.
merely 6—10% obtained in calculations is typical of the
computational method employed.3 Note that among the
tested molecules, the experimental value |DS| = 0.055 cm–1
for trinitrene 1 seems strongly underestimated. According
to calculations, the major contribution to the magnetic
anisotropy comes from oneꢀcenter spinꢀspin interactions
in nitrene centers. The computational algorithm makes it
possible to "extract" the energy of these interactions (D´SS)
from the total energy of spinꢀspin interactions DSS (see
Table 1). The highest degree of spin density localizaꢀ
tion on the nitrene centers ( = 1.68) is achieved for triꢀ
nitrene 2 containing the smallest number of atoms in the
molecule and three small nitrogen atoms in the aromatic
ring. This trinitrene is characterized by the largest value of
the magnetic anisotropy parameter |DS|. As the number
Fig. 5. Spin density distribution in the septet trinitrene 4 obꢀ
tained from UDFT/PBE/DZ calculations.

2224
Russ.Chem.Bull., Int.Ed., Vol. 61, No. 12, December, 2012
Misochko et al.
13. S. V. Chapyshev, D. Grote, C. Finke, W. Sander, J. Org.
Chem., 2008, 73, 7045.
14. S. V. Chapyshev, E. Ya. Misochko, A. V. Akimov, V. G.
Dorokhov, P. Neuhaus, D. Grote, W. Sander, J. Org. Chem.,
2009, 74, 7238.
15. S. V. Chapyshev, P. Neuhaus, D. Grote, W. Sander, J. Phys.
Org. Chem., 2010, 23, 340.
16. K. Itoh, Pure Appl. Chem.,1978, 50, 1251.
molecules. This strategy of increasing the magnetic anisoꢀ
tropy of highꢀspin molecules seems to be more efficient
compared to the attempts to synthesize highꢀspin moleꢀ
cules with the highest spin density on the triplet centers.
Trinitrene 4 studied in this work is characterized by
close values of the spin density on the nitrene centers and
the theoretical |DS| parameter with those of trinitrene 1. It
follows that the |DS| value reported previously for 1 is
strongly underestimated and that the ESR spectrum of 1
should be very close to that of 4. Our study confirmed that
photolysis of 1,3,5ꢀtriazidobenzenes in cryogenic matriꢀ
ces results in septet trinitrenes as major products and that
only trace amounts of intermediates (triplet mononitrenes
and quintet dinitrenes) are formed, if any. High yield of
septet trinitrenes in photolysis of 1,3,5ꢀtriazidobenzenes
may appear to be of practical interest for obtaining magꢀ
netoꢀactive organic materials by lowꢀtemperature photoꢀ
lysis of crystalline triazides.
17. S. Stoll, A. Schweiger, J. Magn. Reson., 2006, 178, 42.
18. M. J. Fisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr.,
T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi,
G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji,
M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratꢀ
mann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvaꢀ
dor, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Stain, O. Farkas, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G.
Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu,
A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J.
Fox, T. Keith, M. A. AlꢀLaham, C. Y. Peng, A. Nanayakꢀ
kara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonazalez, J. A. Pople, Gaussian 03. Reviꢀ
sion D.01, Gaussian Inc., Wallingford (CT), 2004.
19. F. Neese, ORCA, An ab initio, Density Functional and Semiemꢀ
pirical Program Package. Version 2.8.0.2, University of Bonn,
Bonn, Germany, 2011; http://www.thch.uniꢀbonn.de/tc/
orca.
20. J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett., 1996,
77, 3865; J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev.
Lett., 1997, 78, 1396.
21. A. Schafer, H. Horn, R. Ahlrichs, J. Chem. Phys., 1992, 97,
2571; doi: 10.1063/1.463096.
22. R. McWeeny, Y. Mizuno, Proc. R. Soc. London, Ser. A, 1961,
259, 554.
23. M. R. Pederson, S. N. Khanna, Phys. Rev. B, 1999, 60, 9566.
24. T. A. Fukuzawa, K. Sato, A. R. Ichimura, T. Kinoshita,
T. Takui, K. Itoh, P. M. Lahti, Mol. Cryst. Liq. Cryst., 1996,
278, 253.
This work was financially supported by the Russian
Foundation for Basic Research (Project No. 10ꢀ03ꢀ00065)
and the Division of Chemistry and Materials Science of
the Russian Academy of Sciences (Program No. 1).
References
1. F. Neese, J. Chem. Phys., 2007, 127, 164112.
2. W. Sander, D. Grote, S. Kossmann, F. Neese, J. Am. Chem.
Soc., 2008, 130, 4396.
3. E. Ya. Misochko, D. V. Korchagin, K. V. Bozhenko, S. V.
Chapyshev, S. M. Aldoshin, J. Chem. Phys., 2010 133,
064101.
4. K. Sugisaki, K. Toyota, K. Sato, D. Shiomi, M. Kitagawa,
T. Takui, ChemPhysChem., 2010, 11, 3146.
5. K. Sugisaki, K. Toyota, K. Sato, D. Shiomi, M. Kitagawa,
T. Takui, Phys. Chem. Chem. Phys., 2011, 13, 6970.
6. E. Wasserman, K. Schueller, W. A. Yager, Chem. Phys. Lett.,
1968, 2, 259.
7. E. Ya. Misochko, A. V. Akimov, V. F. Lavitskii, S. V. Chapyꢀ
shev, Russ. Chem. Bull. (Int. Ed.), 2007, 56, 2354 [Izv. Akad.
Nauk, Ser. Khim., 2007, 2284].
25. S. V. Chapyshev, H. Tomioka, Bull. Chem. Soc. Jpn, 2003,
76, 2075.
8. E. Ya. Misochko, A. V. Akimov, S. V. Chapyshev, J. Chem.
Phys., 2008, 128, 124504.
26. S. V. Chapyshev, Izv. Akad. Nauk, Ser. Khim., 2006, 1085
[Russ. Chem. Bull., Int. Ed., 2006, 55, 1126].
27. E. Ya. Misochko, A. V. Akimov, A. A. Masitov, D. V. Korꢀ
chagin, I. K. Yakushchenko, S. V. Chapyshev, J. Chem. Phys.,
2012, 137, 064308.
9. T. Koto, K. Sugisaki, K. Sato, D. Shiomi, K. Toyota, K. Itoh,
E. Wassermann, P. M. Lahti, T. Takui, Appl. Magnet. Res.,
2010, 37, 703.
10. S. V. Chapyshev, R. Walton, J. A. Sanborn, P. M. Lahti,
J. Am. Chem. Soc., 2000, 122, 1580.
11. T. Sato, A. Narazaki, Y. Kawaguchi, H. Niino, G. Bucher,
D. Grote, J. J. Wolff, H. H. Wenk, W. Sander, J. Am. Chem.
Soc., 2004, 126, 7846.
12. E. Ya. Misochko, A. V. Akimov, S. V. Chapyshev, J. Chem.
Phys., 2008, 129, 174510.
Received August 22, 2012;
in revised form October 5, 2012