Structural, magnetic, and computational correlations of some imidazolo-fused 1,2,4-benzotriazinyl radicals

Article abstract of DOI:10.1002/chem.201304538

1,3,7,8-Tetraphenyl-4,8-dihydro-1H-imidazolo[4,5g][1,2,4]benzotriazin-4-yl (5), 8-(4-bromophenyl)-1,3,7-triphenyl-4,8-dihydro-1H-imidazolo[4,5g][1,2,4] benzotriazin-4-yl (6), and 8-(4-methoxyphenyl)-1,3,7-triphenyl-4,8-dihydro-1H- imidazolo[4,5g][1,2,4]benzotriazin-4-yl (7) were characterized by using X-ray diffraction crystallography, variable-temperature magnetic susceptibility studies, and DFT calculations. Radicals 5-7 pack in 1 D π stacks made of radical pairs with alternate short and long interplanar distances. The magnetic susceptibility (χ vs. T) of radicals 5 and 6 exhibit broad maxima at (50±2) and (50±4) K, respectively, and are interpreted in terms of an alternating antiferromagnetic Heisenberg linear chain model with average exchange-interaction values of J=-31.3 and -35.4 cm^-1 (g _solid=2.0030 and 2.0028) and an alternation parameter a=0.15 and 0.38 for 5 and 6, respectively. However, radical 7 forms 1 D columns of radical pairs with alternating distances; one of the interplanar distances is significantly longer than the other, which decreases the magnetic dimensionality and leads to discrete dimers with a ferromagnetic exchange interaction between the radicals (2J=23.6 cm^-1, 2zJ′=-2.8 cm^-1, g _solid=2.0028). Magnetic exchange-coupling interactions in 1,2,4-benzotriazinyl radicals are sensitive to the degree of slippage and inter-radical separation, and such subtle changes in structure alter the fine balance between ferro- and antiferromagnetic interactions. Bigger is stronger: π-Extension of 1,2,4-benzotriazin-4-yl radicals leads to an enhanced overlap between SOMO orbitals and, therefore, to a strong effective exchange coupling within alternating 1 D chains of imidazolo-benzotriazinyls. However, subtle changes in interplanar distances and slippage angles switch the observed magnetic properties (see figure).

Full text of DOI:10.1002/chem.201304538

DOI: 10.1002/chem.201304538
Full Paper
&
Magnetic Properties
Structural, Magnetic, and Computational Correlations of Some
Imidazolo-Fused 1,2,4-Benzotriazinyl Radicals
Christos P. Constantinides,[a] Andrey A. Berezin,[a] Maria Manoli,[a] Gregory M. Leitus,[b]
Georgia A. Zissimou,[a] Michael Bendikov†,[b] Jeremy M. Rawson,[c] and
Panayiotis A. Koutentis*[a]
Dedicated to the memory of Professor Michael Bendikov
Abstract: 1,3,7,8-Tetraphenyl-4,8-dihydro-1H-imidazolo[4,5g]-
[1,2,4]benzotriazin-4-yl (5), 8-(4-bromophenyl)-1,3,7-triphen-
yl-4,8-dihydro-1H-imidazolo[4,5g][1,2,4]benzotriazin-4-yl (6),
values of J=ꢁ31.3 and ꢁ35.4 cmꢁ1 (gsolid =2.0030 and
2.0028) and an alternation parameter a=0.15 and 0.38 for 5
and 6, respectively. However, radical 7 forms 1D columns of
radical pairs with alternating distances; one of the interpla-
nar distances is significantly longer than the other, which de-
creases the magnetic dimensionality and leads to discrete
dimers with a ferromagnetic exchange interaction between
the radicals (2J=23.6 cmꢁ1, 2zJ’=ꢁ2.8 cmꢁ1, gsolid =2.0028).
Magnetic exchange-coupling interactions in 1,2,4-benzotria-
zinyl radicals are sensitive to the degree of slippage and
inter-radical separation, and such subtle changes in structure
alter the fine balance between ferro- and antiferromagnetic
interactions.
and
8-(4-methoxyphenyl)-1,3,7-triphenyl-4,8-dihydro-1H-
imidazolo[4,5g][1,2,4]benzotriazin-4-yl (7) were characterized
by using X-ray diffraction crystallography, variable-tempera-
ture magnetic susceptibility studies, and DFT calculations.
Radicals 5–7 pack in 1D p stacks made of radical pairs with
alternate short and long interplanar distances. The magnetic
susceptibility (c vs. T) of radicals 5 and 6 exhibit broad
maxima at (50ꢀ2) and (50ꢀ4) K, respectively, and are inter-
preted in terms of an alternating antiferromagnetic Heisen-
berg linear chain model with average exchange-interaction
Introduction
open-shell organic molecules hold promise as multifunctional
materials because of their unique physical properties.[2] The
presence of unpaired electrons can give rise to materials with
new functionalities that combine magnetic, optical, and trans-
port properties. However, the observed physical properties
greatly depend on the solid-state packing, which is difficult to
predict and control. Crystal engineering of organic radicals is
challenging and breakthroughs achieved to date have been
serendipitous.[3] Interesting properties, such as bulk ferromag-
netism or metallic conductivity are a rarity for organic radicals
because intermolecular forces are weak and short range, which
leads to low-dimensional magnetic exchange interactions.[4]
Nevertheless, systematic structure–property studies can identi-
fy design rules for the preparation of materials with useful
properties.
Multifunctional materials are under intensive study as compo-
nents in electronic devices owing to their potential to reduce
the size, weight, cost, and power consumption of these devi-
ces, while simultaneously improving their efficiency and up-
grading their capabilities.[1] Organic molecules have an impor-
tant role as building blocks for such materials because they
can be tailor-made to exhibit desired properties. Persistent
[a] Dr. C. P. Constantinides, Dr. A. A. Berezin, Dr. M. Manoli, G.A. Zissimou,
Dr. P. A. Koutentis
Department of Chemistry
University of Cyprus, P.O. Box 20537
1678 Nicosia (Cyprus)
Fax: (+357)22892809
Numerous families of persistent organic radicals have been
extensively studied.[2] The 1,2,4-benzotriazin-4-yls (e.g., parent
radical 1, Figure 1), which were first prepared in the late 1960s
by Blatter et al.,[5] have recently attracted interest owing to
their air and moisture stability.[6] Although persistent organic p
radicals have a tendency to dimerize in the solid state, to date
there has been no evidence for dimerization in 1,2,4-benzotria-
zinyls. In addition, a range of low-dimensional magnetic prop-
erties have been reported.[6]
E-mail: koutenti@ucy.ac.cy
[b] Dr. G. M. Leitus, Prof. M. Bendikov
Department of Organic Chemistry
Weizmann Institute of Science
76100 Rehovot (Israel)
[c] Prof. J. M. Rawson
Department of Chemistry & Biochemistry
University of Windsor
401 Sunset Avenue, Windsor, ON, N9B 3P4 (Canada)
[†] Deceased, 2 July 2013
Increasing the dimensionality of the magnetic interactions
requires strong exchange-coupling interactions that propagate
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201304538.
Chem. Eur. J. 2014, 20, 1 – 10
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Thiazolo[5’,4’:4,5]benzo[1,2-e][1,2,4]triazin-4-yls 4 were pre-
pared in three steps by C6 amination of benzotriazinone 8 fol-
lowed by acylation and aroylation to give N-(benzotriazin-6-yl)-
carboxamides that undergo ring closure on treatment with
P2S5 to give the thiazolo-fused benzotriazinyls.[7a] Following
a similar synthetic strategy, reaction of benzotriazinone 8 with
N’-arylbenzamidines in PhMe at approximately 1008C in the
presence of N,N-diisopropylethylamine (Hꢁnig’s base) gave 6-
benzimidamidobenzotriazines 9 in 79 to 95% yield, which cy-
clodehydrate upon treatment with neat AcOH at approximate-
ly 1208C to give 8-substituted 1,3,7-triphenyl-4,8-dihydro-1H-
imidazo[4,5g][1,2,4]benzotriazin-4-yls 2 in 70 to 81% yield
(Scheme 1).[7b] In this manner, radicals 5, 6, and 7 were pre-
pared in overall yields of 77, 55, and 61%, respectively.[7b]
Figure 1. 1,3-Diphenyl-1,4-dihydro-1,2,4-benzotriazin-4-yl (1) and structures
of imidazolo-, oxazolo-, and thiazolo-fused benzotriazinyls 2–7 with IUPAC
numbering of the ring systems.
throughout the bulk of the material. Thus, we designed and
prepared a range of imidazolo-, oxazolo-, and thiazolo-fused
1,2,4-benzotriazinyls (2, 3, and 4 in Figure 1).[7] We anticipated
that the extension of the p-acene core of these radicals could
provide an effective pathway for stronger exchange interac-
tions because 1,2,4-benzotriazinyls tend to form p-slipped
stacked columns. Moreover, the introduction of substituents
that can promote intermolecular contacts could increase the
dimensionality of the magnetic interactions.
Cyclic voltammetry and EPR spectroscopy
The redox behavior of imidazolo-fused radicals 5–7 was previ-
ously shown to differ from the typical redox behavior of non-
fused 1,2,4-benzotriazinyls[6] in that their reductions were
quasi-reversible. However, we have now attributed this quasi-
reversible behavior to the wet electrolyte. With dry electrolyte
both the oxidation (0/+1) and reduction (ꢁ1/0) processes of
radicals 5–7 were fully reversible (Figures S1–S3 in the Sup-
porting Information). The half-way oxidation potentials are
E1/20/+1 =0.09, 0.07, 0.05 V and half-way reduction potentials are
E1/2ꢁ1/0 =ꢁ1.03, ꢁ0.96, ꢁ0.98 V for 5, 6, and 7, respectively,
versus the ferrocene/ferrocinium (Fc/Fc+) couple as +0.352 V
(1 mm in CH2Cl2 with nBu4NBF4 as the supporting electrolyte).
Solid-state EPR spectra of radicals 5–7 (Figures S4–S6 in the
Supporting Information) are essentially isotropic singlets
(gsolid =2.0030, 2.0028, 2.0028 for 5, 6, and 7, respectively) that
are consistent with organic radicals with little spin-orbit cou-
pling. The solution EPR spectra (Figures S7–S9 in the Support-
ing Information) exhibit typical 1,2,4-benzotriazinyls seven-line
multiplets, which are consistent with the unpaired electron
coupling with the three similar but slightly nonequivalent 14N
Herein we present the solid-state characterization of 1,3,7,8-
tetraphenyl-4,8-dihydro-1H-imidazolo[4,5g][1,2,4]benzotriazin-
4-yl
(5),
8-(4-bromophenyl)-1,3,7-triphenyl-4,8-dihydro-1H-
imidazolo[4,5g][1,2,4]benzotriazin-4-yl (6), and 8-(4-methoxy-
phenyl)-1,3,7-triphenyl-4,8-dihydro-1H-imidazolo[4,5g]-
[1,2,4]benzotriazin-4-yl (7) and provide magnetostructural cor-
relations based on variable-temperature (VT) magnetic suscept-
ibility data, single-crystal X-ray diffraction studies, and DFT cal-
culations.
Results
Synthesis and characterization
Within the context of our ongoing studies on this family of
radicals, we have developed high-yielding synthetic routes to
several 7-substituted benzotriazinyls and investigated their
chemistry.[8] Interestingly, 1,3-diphenylbenzo[e][1,2,4]triazin-
7(1H)-one (8), the oxidation product of Blatter radical 1, was
a useful scaffold for the preparation of imidazolo-, oxazolo-,
and thiazolo-fused benzotriazinyls.[7] Benzotriazinone 8 under-
goes regiospecific nucleophilic substitution at C6, electrophilic
substitution at C8, and a range of cyclization reactions to give
linear and angular fused benzotriazinones, some of which
were unusual zwitterionic compounds.[8]
nuclei.[6,9] Previous EPR and ENDOR studies by Neugebauer on
[9a]
15N-labeled benzotriazinyls showed that aN(1) @aN(4) >aN(2)
.
Simulation of the first-derivative mode gave gsoln =2.0025,
2.0031, 2.0028 for 5, 6, and 7, respectively. The hyperfine cou-
pling constants (hfcc) were calculated to be aN(1) =7.04, aN(2)
=
4.66, and aN(4) =4.89 G for radical 5, aN(1) =7.06, aN(2) =4.73, and
a
N(4) =4.77 G for radical 6, and aN(1) =7.04, aN(2) =4.92, and
N(4) =4.94 G for radical 7. Spin densities estimated based on
a
hfcc by using McConnell’s equation[10,11] are 1N1 =0.282, 1N2
=
0.220, and 1N4 =0.231 for radical 5, 1N1 =0.282, 1N2 =0.223, and
1N4 =0.225 for radical 6, and
1N1 =0.282, 1N2 =0.232, and
1N4 =0.233 for radical 7.
X-Ray diffraction studies
The crystallographic rather than
the IUPAC numbering system is
Scheme 1. Synthetic route to 1,3,7-triphenylimidazolobenzotriazinyls 2. Radicals 5 (2: R1 =R2 =Ph), 6 (2: R1 =Ph,
R2 =4-BrC6H4), and 7 (2: R1 =Ph, R2 =4-MeOC6H4) were prepared in overall yields of 77, 55, and 61%, respectively.
Reagents and conditions: i) iPr2EtN, PhMe, 1008C, 1–4 d, 79–95%; ii) AcOH, 1208C, 1 h, 70–81%.
used in discussion only for this
section. Suitable single crystals
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
of radicals 5–7 for X-ray diffraction studies were obtained by
slow diffusion of n-pentane into a solution of each radical in
benzene. The radicals crystallized in the monoclinic space
N1ꢁPh was variable [36.3(8)–45.1(3)8], in line with previous ob-
servations on nonfused benzotriazinyls in which this phenyl
substituent participated in p–p face-to-face and edge-to-face
interactions.[6c]
¯
group P1 with one molecule in the asymmetric unit. The intra-
molecular geometrical parameters (Table 1) of the triazinyl ring
The C1ꢁPh substituent consistently adopts a coplanar orien-
tation with the triazinyl ring [9.4(2), 7.8(9), and 0.2(4)8 for 5, 6,
and 7, respectively] owing to the absence of any steric interac-
tions and possibly two weak intramolecular contacts between
the ortho-H of the Ph and the N2 and N3 atoms of the amidra-
zonyl moiety.
Table 1. Selected experimental (X-ray) bond lengths and angles of
radicals 5–7.
Similar intramolecular geometries, unit cell dimensions
¯
(Table T1 in the Supporting Information), and space group (P1)
for radicals 5–7 give rise to solid-state packing arrangements
that are comparable for the three radicals. The main feature of
the crystal packing is the formation of 1D p-slipped stacked
columns. To facilitate our analysis the discussion below is divid-
ed into 1) the intra-stack interactions along the p-stacking di-
rection and 2) the in-plane interactions perpendicular to the
stacking direction.
5 (X=H)
6 (X=Br)
7 (X=MeO)
bond lengths [ꢂ]
C1ꢁN2
1.340(2)
1.331(2)
1.385(2)
1.323(2)
1.416(2)
1.345(9)
1.314(9)
1.399(4)
1.321(9)
1.41(1)
1.343(4)
1.342(4)
1.392(3)
1.310(4)
1.401(4)
C1ꢁN3
N4ꢁC6
Intra-stack interactions
N5ꢁC5
C4ꢁC6
The p-extended acene core of radicals 5–7 favors p stacking
and formation of 1D columns along the a axis. The radicals
within the columns are related with respect to each other
through a center of inversion (ꢁx, ꢁy, ꢁz) that places the
NꢁPh substituents on opposite sides to avoid the build-up of
steric congestion (Figure 2). The degree of slippage and mean
interplanar distance between the radicals alternates (Table 2)
and gives rise to two distinct centrosymmetric pairs, radicals
I–II and II–III (Figure 2). The radicals of pair I–II interact mainly
through the fused imidazolo rings. The center of inversion run-
ning through the middle of these rings gives rise to a pair of
symmetrical weak CꢁH···N contacts between the imidazolo N5
bond angles [8]
C1-N2-N1
115.9(1)
116.7(1)
106.6(1)
105.9(1)
114.7(5)
115.5(6)
106.7(5)
105.7(6)
115.5(2)
116.2(2)
106.1(2)
105.3(2)
C1-N3-C2
C5-N4-C6
C5-N5-C4
dihedral angles [8]
C26-C21-C5-N5
C28-C27-N4-C6
C14-C9-N1-N2
N2-C1-C15-C16
23.5(2)
60.1(2)
44.1(2)
9.4(2)
20.0(1)
60.8(9)
36.3(8)
7.8(9)
27.2(4)
58.1(3)
45.1(3)
0.2(4)
are similar to previously studied nonfused benzotriazinyl radi-
cals.[6] The 1,2,4-amidrazonyl moiety is essentially planar as de-
fined by the angle between the plane of N1, N2, C1, N3 atoms
and the plane of the fused benzimidazole (3.45, 5.08, 3.458 for
5, 6, and 7, respectively). The CꢁN bond lengths [C1ꢁN2,
1.340(2)–1.345(9) ꢂ; C1ꢁN3, 1.314(9)–1.342(4) ꢂ], which are in-
termediate between typical single and double CꢁN bonds, and
the C1-N2-N1 [114.7(5)–115.9(1)8) and C1-N3-C2 (115.5(6)–
116.7(1)8] angles, which are typical for sp2-hybridized nitrogen
atoms, support strong delocalization of the unpaired electron
over the amidrazonyl nitrogen atoms. The imidazole ring in
radicals 5–7 is planar with typical interatomic lengths and
angles: N4ꢁC6 [1.385(2)–1.399(4) ꢂ], N5ꢁC5 [1.310(4)–
1.323(2) ꢂ], C4ꢁC6 [1.401(4)–1.416(2) ꢂ]; C5-N4-C6 [106.1(2)–
106.7(5)8], C5-N5-C4 [105.3(2)–105.9(1)8]. The C5ꢁPh substitu-
ent is out of plane [20.0(1)–27.2(4)8] as a result of the steric in-
teraction between ortho hydrogen atoms H22 and H32. A
large torsion angle is observed for the imidazolo N4ꢁPh
[58.1(3)–60.8(9)8] as a result of steric interactions at both sides
of the phenyl, thus the para-phenyl substituents have little
effect on the electron density around the molecule; only
subtle differences are observed with the redox potentials and
the hfcc of radicals 5–7. The torsion angle of the triazinyl
and an ortho-hydrogen of the imidazolo N4ꢁPh [dC28···N5
=
3.315(3)–3.376(2) ꢂ, a CꢁH···N=140.5(2)–156.24(9)8]. The driv-
ing force for the formation of radical pair II–III is the positive
overlap between the benzotriazinyl rings. This is further sup-
ported by a pair of crystallographically equivalent CꢁH···N con-
tacts between the triazinyl N3 and an ortho-hydrogen of the
imidazolo N4ꢁPh [dC32···N3 =3.472(2)–3.492(4) ꢂ, a C32ꢁH···N3=
170.6(9)–175.9(4)8].
Table 2. Slippage angles and interplanar distances of pairs I-II and II-III
for radicals 5-7.
Radical
Pair
Longitudinal
(f1) [8][a]
Latitudinal
(f2) [8][a]
Interplanar distance
(d) [ꢂ]
5
6
7
I–II
II–III
I–II
II–III
I–II
II–III
47.3
30.0
52.6
34.5
49.1
34.2
16.6
10.7
5.6
16.2
19.7
12.0
3.697
3.685
3.692
3.578
4.124
3.528
[a] See Figures S10 and S11 in the Supporting Information for parameters
used to define the degree of longitudinal and latitudinal slippage.
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
3
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
pated that exchange interactions in pair II–III will be stronger
as the radicals overlap mainly through the spin-bearing benzo-
triazinyls. Exchange interactions in pair I–II are expected to be
weaker because the interplanar distances are longer (3.692–
4.124 ꢂ) and longitudinal angles larger (47.3–52.68).
Inter-stack interactions
The in-plane inter-stack interactions in imidazolo-fused benzo-
triazinyls 5–7 are dominated by weak hydrogen bonds. In all
three radicals, the imidazolo N5 participates in CꢁH···N interac-
tions with the para-hydrogen of NꢁPh [dC12···N5 =3.410(4)–
3.416(1) ꢂ, a C12ꢁH···N5=157.1(1)–166.0(2)8], which connect
neighboring molecules into a head-to-tail arrangement to form
chains that run parallel to the b axis (Figure S12 in the Sup-
porting Information). Hydrogen-bond acceptors at the para
position of the imidazolo N4ꢁPh in radicals 6 and 7 enable
these inter-stack interactions. The MeO group in radical 7 acts
as a hydrogen-bond acceptor from the meta hydrogen of
C5ꢁPh [dC25···O1 =3.427(4) ꢂ,
a
C25ꢁH···O1=148.3(2)8] and
helps to stabilize the formation of the chain along the b axis
(Figure S12b in the Supporting Information). In radical 6, a pair
of crystallographically equivalent weak CꢁH···Br interactions
connect antiparallel chains to form 2D sheets in the bc plane
(Figure S12c in the Supporting Information). In all three radi-
cals a tight packing with no voids is completed with edge-to-
face p contacts between the N1- and N4-phenyls and the
C1/C5-Ph substituents.
Magnetic properties
Magnetic exchange coupling interactions were probed by
using variable-temperature magnetic-susceptibility measure-
ments by using a SQUID magnetometer in the temperature
range of 5 to 300 K and in an applied field of 0.4 T. Data were
collected in both warming and cooling modes with no signifi-
cant differences in sample susceptibility. The data were correct-
ed for both sample diamagnetism (Pascal’s constants) and the
diamagnetism of the sample holder. Plots of c versus T for radi-
cals 5 and 6 and cT versus T for radical 7 are presented in
Figure 3.
Figure 2. Packing diagrams of radicals a) 5, b) 6, and c) 7 along the a axis
showing the shortest intermolecular contacts. Thermal ellipsoids are drawn
at 50% probability (hydrogen atoms are omitted for clarity). a) Pair I–II:
d
d
d
C1···C23 =3.378(2) ꢂ (ꢁx, 1ꢁy, 1ꢁz), dC2···C22 =3.304(2) ꢂ (ꢁx, 1ꢁy, 1ꢁz),
C28···N5 =3.576(2) ꢂ, a C28ꢁH···N5=156.24(9)8 (ꢁx, 1ꢁy, 1ꢁz); pair II–III:
C1···C26 =3.318(2) ꢂ (1ꢁx, 1ꢁy, 1ꢁz), dC32···N3 =3.472(2) ꢂ,
a C32ꢁH···N3=170.62(9)8 (1ꢁx, 1ꢁy, 1ꢁz). b) Pair I–II: dC2···C23 =3.375(9) ꢂ
(1ꢁx, ꢁy, 1ꢁz), dC2···C22 =3.286(9) ꢂ (1ꢁx, ꢁy, 1ꢁz), dC28···N5 =3.529(8) ꢂ,
a C28ꢁH···N5=140.5(4)8 (1ꢁx, ꢁy, 1ꢁz); pair II–III: dC1···C26 =3.381(9) ꢂ
(ꢁx, ꢁy, 1ꢁz), dC32···N3 =3.483(9) ꢂ, a C32ꢁH···N3=175.9(4)8 (ꢁx, ꢁy, 1ꢁz).
c) Pair I–II: dC8···C22 =3.392(3) ꢂ (1ꢁx, ꢁy, 1ꢁz), dC28···N5 =3.315(3) ꢂ,
a C28ꢁH···N5=140.5(2)8 (1ꢁx, ꢁy, 1ꢁz); pair II–III: dC1···C26 =3.362(4) ꢂ
(ꢁx, ꢁy, 1ꢁz), dC32···N3 =3.492(4) ꢂ, a C32ꢁH···N3=175.3(4)8 (ꢁx, ꢁy, 1ꢁz).
The g values of radicals 5–7 determined in the solid state
(gsolid =2.0030, 2.0028, and 2.0028 for 5, 6, and 7, respectively)
are very close to that of a free electron, which denotes a small
spin-orbit interaction. Isotropic magnetic exchange interactions
are, therefore, responsible for the observed magnetic proper-
ties of imidazolo-fused benzotriazinyl radicals 5–7. The extend-
ed p-based nature of these radicals coupled with their propen-
sity to form 1D p-slipped columns suggests that radicals 5–7
will exhibit strong unidimensional magnetic interactions prop-
agating parallel to the stacking direction, along which the orbi-
tal overlap between the singly-occupied molecular orbitals
(SOMOs) is maximized.
The interplanar distance between radicals I–II is significantly
longer for 8-(4-bromophenyl) radical 7 (dI–II =4.124 ꢂ) com-
pared with that of 8-phenyl- and 8-(4-methoxyphenyl)-substi-
tuted radicals 5 and 6 (dI–II =3.697 and 3.692 ꢂ, respectively),
which indicates a more pronounced separation between radi-
cals I and II in radical 7 and a drop in dimensionality from
a 1D chain to remote dimers. However, the interplanar dis-
tance of pair II–III for radical 7 is shorter (dII–III =3.528 ꢂ vs.
3.685 and 3.578 ꢂ for radicals 5 and 6, respectively), which de-
notes a stronger interaction between the radicals. It is antici-
Upon cooling from 300 K, the molar susceptibility (c) in-
creases gradually and reaches a broad maximum at Tmax
=
(50ꢀ2) and (50ꢀ4) K for radicals 5 and 6, respectively, which
indicates strong antiferromagnetic interactions (Figure 3a and
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
b). Below the maximum at 50 K, c falls off gradually down to
15 K for radical 5 and 20 K for radical 6 before it increases
again at lower temperatures. This increase in c at low tempera-
tures is attributed to uncoupled radicals at defect sites in the
lattice. The alternating interplanar distances along the stacking
direction (Table 2) indicate that the appropriate model to fit
the magnetic data is Hatfield’s expression for the magnetic
susceptibility of isotropic Heisenberg alternating antiferromag-
netic (AF) chains of S=1= spins.[12] The corresponding spin
2
Hamiltonian is given by
n=2
X
H ¼ ꢁ2J
½S2iS2iꢁ1 þ aS2iS2iþ1ꢂ
i¼1
in which 2J (=J1) and 2aJ (=J2) are the exchange integrals
along the stacking direction and a (=J2/J1) is a parameter that
indicates the degree of alternation. At the extremes, when a=
0 the model corresponds to an isolated dimer system and
when a=1 the model reduces to the linear-chain model.
The best fit to the magnetic susceptibility data was obtained
by using a set of polynomial parameters for 0ꢃaꢃ0.4. A pa-
rameter 1 was included to account for the susceptibility arising
from uncoupled radicals at the lattice defect sites. The fit re-
produces well both the position and the maximum in c when
J=ꢁ31.3 cmꢁ1, a=0.15, 1=0.94, gsolid =2.0030 for radical 5
and J=ꢁ35.4 cmꢁ1, a=0.38, 1=0.91, gsolid =2.0028 for radical
6 (Figure 3). By applying the ratio of a=J2/J1, in which J1 =2J
and J2 =2aJ, the estimated values along the chain direction
are 2J=ꢁ62.6 cmꢁ1, 2aJ=ꢁ9.4 cmꢁ1 for radical 5 and 2J=
ꢁ70.8 cmꢁ1, 2aJ=ꢁ26.9 cmꢁ1 for radical 6.
A different magnetic behavior was observed for radical 7; it
follows Curie–Weiss behavior in the temperature range of 5 to
300 K with C=0.338 emuKmolꢁ1 close to the value expected
for S=1= spin (Figure 3c, inset). The positive Weiss constant
2
(q=1.83 K) indicates the presence of local ferromagnetic inter-
actions between the radicals. When the temperature is lowered
from 300 K, cT increases gradually and reaches a maximum at
T
max =(30ꢀ4) K (Figure 3c). Below this temperature, cT falls off
sharply, which indicates the presence of antiferromagnetic in-
teractions at low temperatures. The interplanar distance be-
tween radicals I–II is significantly longer for radical
7
(dI–II =4.124 ꢂ) and leads to more discrete pairs of radicals II–III
(dII–III =3.528 ꢂ) and, therefore, the appropriate model to fit the
magnetic data is the Bleaney–Bowers expression for the mag-
netic susceptibility of interacting pairs of S=1= spins (based
2
[13]
ˆ ˆ
Figure 3. Temperature dependence of c for radicals a) 5 and b) 6; c) temper-
ature dependence of cT and 1/c (inset) for radical 7. a) The solid line repre-
on the Hamiltonian H=ꢁ2JABꢀSASB).
The Bleaney–Bowers
model was extended to cover the low-temperature down
curve of cT; inclusion of a mean field term (2zJ’) to account for
the antiferromagnetic inter-dimer interactions provided
a much improved fit with 2J=23.6 cmꢁ1, 2zJ’=ꢁ2.8 cmꢁ1
sents the best fit to the 1D alternating AF linear chain model for S=1= spin;
2
J=ꢁ31.3 cmꢁ1, a=0.15, 1=0.94, gsolid =2.0030, R=5.82ꢃ10ꢁ4 (correlation
factor R=ꢀ[cobsdꢁccalcd]2/ꢀ[cobsd]2). b) The solid line represents the best fit to
the 1D alternating AF linear chain model for S=1= spin; J=ꢁ35.4 cmꢁ1
,
2
,
a=0.38, 1=0.91, gsolid =2.0028, R=7.58ꢃ10ꢁ4 (correlation factor
gsolid =2.0028.
R=ꢀ[cobsdꢁccalcd]2/ꢀ[cobsd]2). c) The solid line represents the best fit to the
Bleaney–Bowers model for a pair of interacting S=1= spins
2
with 2J=23.6 cmꢁ1, 2zJ’=ꢁ2.8 cmꢁ1, gsolid =2.0028, R=6.55ꢃ10ꢁ4 (correla-
tion factor R=ꢀ[(cT)obsꢁ(cT)calcd]2/ꢀ[(cT)obs]2). Inset: Curie–Weiss behavior in
the 5–300 K region, C=0.338 emuKmolꢁ1 and q=1.83 K.
Computational modeling
Quantum chemistry methods can provide estimates for the
magnetic exchange coupling interactions within radical pairs.
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Although the macroscopic magnetic behavior is the cumula-
tive outcome of all existing magnetic interactions, for radicals
with low dimensional magnetic topology, for example, imida-
zolo-fused benzotriazinyls 5–7, the computed exchange inter-
actions correlate qualitatively with experimentally determined
exchange couplings.[14]
to the basis set, that is, 6-311++G(2d,p), to take in account
the heavy Br atom, flips JII–III from 18.7 to ꢁ56.0 cmꢁ1
.
Discussion
The magnetic behavior of radicals 5–7 can be rationalized by
using the molecular orbital (MO) model.[16] In the MO model,
the exchange coupling interaction (JAB =2Kij +4cijSij2) is propor-
tional to the size of the overlap between SOMO orbitals (Sij),
which depends on the distance and orientation of the interact-
ing radicals. The longitudinal and latitudinal slippage angles
and the interplanar distances between the radicals determine
the nature of the magnetic exchange along the p-stacking di-
rection. In related p systems, Oakley et al. demonstrated that
there is a fine balance between ferro- and antiferromagnetic
interactions by adjusting the degree of slippage through
chemical pressure (introduction of small perturbations by syn-
thetic means) or with application of physical pressure.[17]
The calculated spin–spin exchange interactions of radical
pairs I–II (JI–II) and II–III (JII–III) were determined by using the un-
projected equation 2JDFT =2(EBSꢁET)/2.[15] The energies of the
triplet (ET) and broken symmetry singlet (EBS) states were deter-
mined by single-point calculations on crystallographically de-
termined geometries of radical pairs I–II and II–III (Table T2 in
the Supporting Information). Although advanced and compu-
tationally expensive quantum chemical methods, such as
CASSCF and CASPT2, provide good results, DFT functionals,
such as B3LYP, perform well in the computation of exchange
interactions.[14] The latter method was employed for the calcu-
lations of the JI–II and JII–III values by using the 6-311+ +G(d,p)
basis set (Table 3).
Exchange coupling within radical pair II–III is expected to be
stronger than that of pair I–II because the radicals overlap pre-
dominantly over the spin-bearing triazinyl moieties, which
maximizes the SOMO–SOMO interaction. Conversely, the mo-
lecular arrangement in radical pair I–II places the 7-Ph substitu-
ent, which hosts little spin density, on top of the adjacent tria-
zinyl ring. Therefore, in the absence of other dominating ex-
change-coupling interactions, J1 and the weaker J2 could be as-
signed to the interactions within radical pairs II–III and I–II, re-
spectively. Calculations confirm the qualitative assignment of J1
to pair II–III and J2 to pair I–II because jJII–III j>jJI–II j and jJ1 j>j
J2 j. This is in line with the observed geometrical parameters of
pairs I–II and II–III; both the longitudinal (average II–IIIf1 =32.98)
and latitudinal (average II–IIIf2 =13.08) angles of pair II–III are
smaller than those of radical pair I–II (average I–IIf1 =49.68,
I–IIf2 =14.08). Furthermore, the interplanar distances of pair
II–III are shorter than those of pair I–II (average dII–III =3.597 ꢂ
vs. average dI–II =3.838 ꢂ).
Table 3. Calculated exchange interactions of pair I–II (JI–II) and pair II–III
(JII–III) for their X-ray determined geometries by using the UB3LYP level of
theory.
Radical
JI–II [cmꢁ1
]
JII–III [cmꢁ1
]
J2 [cmꢁ1
]
J1 [cmꢁ1
]
5
6
ꢁ4.4[a]
ꢁ38.0[a]
ꢁ45.8[b]
ꢁ3.8[a]
ꢁ18.1[a]
ꢁ9.4
ꢁ26.9
–
ꢁ2.8[c]
ꢁ62.6
ꢁ70.8
–
23.6[c]
18.7[a]
ꢁ56.0[b]
26.0[a]
7
[a] 6-311+ +G(d,p). [b] 6-311+ +G(2d,p). [c] For radical 7: J2 =2zJ’ and
J1 =2J.
The computed exchange interactions for radical pairs II–III
are stronger than those of pairs I–II. The sign of these interac-
tions (JII–III) depend mainly on the interplanar distance between
the radicals within the dimer because changes in the slippage
angles (II–IIIf1 and II–IIIf2) are only subtle (Table 2). On going
from an interplanar distance of dII–III =3.685 to 3.528 ꢂ, the ex-
The proximity of dI–II to dII–III for radicals 5 and 6 (Table 2)
leads to a 1D magnetic topology, but the subtle inequality be-
tween them results in an alternation in the exchange-coupling
interactions along the chain. Conversely, for radical 7 the sig-
nificantly larger interplanar distance between the radicals of
pair I–II (dI–II @dII–III) transforms the magnetic topology from 1D
to 0D with discrete dimers that interact with each other
through a weak antiferromagnetic exchange coupling. A care-
ful examination of the crystal packing of radical 7 did not
reveal any apparent reason for the observed change in dI–II.
The magnitude of the exchange interactions of imidazolo-
fused benzotriazinyls 5 and 6 (J1 =ꢁ9.4 to ꢁ26.9 cmꢁ1 and J2 =
ꢁ62.6 to ꢁ70.8 cmꢁ1) is considerably bigger than the ex-
change interactions of nonfused benzotriazinyls, such as the
1D regular chains of the 1,3-diphenyl-7-phenyl- and 7-(4-fluo-
rophenyl)-1,4-dihydro-1,2,4-benzotriazin-4-yls with 2J=ꢁ25.8
and ꢁ23.6 cmꢁ1.[6c] The p-extended acene core of imidazolo-
fused radicals leads to the enhanced orbital overlap and, there-
fore, to stronger exchange interactions. However, although the
selected substituents (e.g., Br and MeO in radicals 6 and 7)
give side-on interactions between 1D chains, they fail to in-
change interactions flip from antiferromagnetic (JII–III
=
ꢁ18.1 cmꢁ1 in 5) to ferromagnetic (JII–III =26.0 cmꢁ1 in 7). Con-
versely, the exchange interactions for radical pair I–II are
weaker and merely antiferromagnetic in nature, however, in
this case their magnitudes depend predominantly on the slip-
page angles. Surprisingly, a significantly larger interplanar dis-
tance for radical 7 (dI–II =4.124 ꢂ, JI–II =ꢁ3.8 cmꢁ1) does not
alter the exchange interaction (dI–II =3.697 ꢂ, JI–II =ꢁ4.4 cmꢁ1
for radical 5). However, although the longitudinal angles for
radicals 5–7 are comparable (I–IIf1 =47.3–52.68) the latitudinal
angle for radical 6 is considerably smaller (I–IIf2 =5.68) com-
pared with that of radicals 5 and 7 (I–IIf2 =16.6 and 19.78, re-
spectively), which explains the bigger exchange interaction ob-
served for pair I–II of radical 6 (JI–II =ꢁ38.0 cmꢁ1). Note that the
6-311++G(d,p) basis set predicted a ferromagnetic interaction
(JII–III =18.7 cmꢁ1) for pair II–III of radical 6 whereas experimen-
tally an antiferromagnetic interaction (J1 =ꢁ70.8 cmꢁ1) was ob-
served (Figure 3b). Nevertheless, the addition of 2d functions
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
(100); elemental analysis calcd (%) for C32H22N5: C 80.65, H 4.65, N
14.70; found: C 80.52, H 4.56, N 14.82.
crease the dimensionality of the exchange interactions. Struc-
tural tailoring of radicals through strategic substitution to ex-
ploit favorable supramolecular interactions might lead to mate-
rials with bulk magnetic properties.
8-(4-Bromophenyl)-1,3,7-triphenyl-4,8-dihydro-1H-imidazo[4,5g]-
[1,2,4]benzotriazin-4-yl (6):[7b] Black needles recrystallized from
PhMe (79 mg, 70%). M.p. (DSC): 318.6 (onset), 323.6 (peak max),
325.8 (decomp. onset), 328.58C (peak max); Rf (Al2O3 neutral/
CH2Cl2): 0.45; UV/Vis (CH2Cl2): lmax (loge)=235 (3.49), 277 inf. (3.73),
292 (3.76), 331 (3.10), 396 inf. (3.28), 417 (3.44), 538 (2.61), 623 nm
inf. (2.29); IR (ATR, Ge): n˜max =1487 (s), 1470 (s), 1452 (m), 1437 (s),
1402 (s), 1389 (s), 1348 (m), 1288 (m), 1196 (m), 1180 (m), 1169 (m),
1070 (m), 1015 (m), 972 (m), 897 (m), 855 (m), 845 (m), 829 (m),
775 (s), 766 (m), 758 cmꢁ1 (m); MS (MALDI-TOF): m/z (%): 558 [M+
H+3]+ (23), 557 [M+H+2]+ (64), 556 [M+H+1]+ (23), 555 [M+H]+
(100); elemental analysis calcd (%) for C32H21BrN5: C 69.20, H 3.81,
N 12.61; found: C 69.20, H 3.97, N 12.52.
Conclusion
The crystal structures and magnetic properties of three organic
radicals,
1,3,7,8-tetraphenyl-4,8-dihydro-1H-imidazolo[4,5g]-
[1,2,4]benzotriazin-4-yl (5), 8-(4-bromophenyl)-1,3,7-triphenyl-
4,8-dihydro-1H-imidazolo[4,5g][1,2,4]benzotriazin-4-yl (6), and
8-(4-methoxyphenyl)-1,3,7-triphenyl-4,8-dihydro-1H-imidazolo-
[4,5g][1,2,4]benzotriazin-4-yl (7) have been investigated. The
radicals p stack in 1D columns along the a axis and are com-
prised of slipped radical pairs (I–II and II–III) with alternating
short and long interplanar distances. Magnetic susceptibility
measurements reveal the presence of antiferromagnetic inter-
actions within the 1D p stacks of radicals 5 and 6 and ferro-
magnetic interactions within the p stacks of radical 7. Spin-
density distributions determined by using EPR spectroscopy,
DFT calculations, and a magnetostructural model have allowed
us to correlate the estimated exchange-coupling interactions
to the two different pairs within the distorted p stack; J1 =
ꢁ62.6, ꢁ70.8, and 23.6 cmꢁ1 and J2 =ꢁ9.4, ꢁ26.9, and
ꢁ2.8 cmꢁ1 (for 5, 6, and 7, respectively) correspond to pairs
II–III and I–II, respectively. The slippage of radicals down the
stacking direction coupled with the changes in interplanar dis-
tance between them lead to alterations in the SOMO orbital
overlap that dictate the nature of exchange-coupling interac-
tions. Although the p-extended acene core of imidazolo-fused
benzotriazinyls 5–7 promotes strong exchange-coupling inter-
actions, the low dimensionality of the magnetic topology can
be explained by the presence of phenyl groups around the pe-
riphery that isolate the spin-bearing units and prohibit interac-
tions over the other two dimensions. If long-range ordering is
to be realized then fine tuning of the substituents’ steric ef-
fects is required to make the 1,2,4-benzotriazinyl core more
available to the surrounding environment. Further modifica-
tions to the benzotriazinyl framework are underway.
8-(4-Methoxyphenyl)-1,3,7-triphenyl-4,8-dihydro-1H-imidazo-
[4,5g][1,2,4]benzotriazin-4-yl (7):[7b] Black needles recrystallized
from PhH (73 mg, 71%). M.p. (DSC): 305.0 (onset), 307.68C (peak
max); Rf (Al2O3 neutral/CH2Cl2): 0.25; UV/Vis (CH2Cl2): lmax (loge=
233 (3.49), 260 inf. (3.59), 285 (3.76), 332 inf. (3.02), 397 inf. (3.27),
421 (3.45), 532 (2.57), 633 nm inf. (2.19); IR (ATR, Ge): n˜max 1514 (s),
1489 (m), 1470 (s), 1452 (m), 1439 (s), 1402 (m), 1389 (s), 1350 (m),
1304 (m), 1287 (m), 1254 (s), 1179 (m), 1169 (m), 1040 (m), 899 (m),
858 (m), 839 (s), 775 (s), 768 cmꢁ1 (s); MS (MALDI-TOF): m/z (%):
509 [M+H+2]+ (2), 508 [M+H+1]+ (24), 507 [M+H]+ (100); ele-
mental analysis calcd (%) for C33H24N5O: C 78.24, H 4.78, N 13.82;
found: C 78.10, H 4.84, N 13.73.
Instrumental analyses
Melting points were determined by using a TA Instruments
DSC Q1000 with samples sealed in aluminum pans under an argon
atmosphere, with heating rates of 5 Kminꢁ1. UV/Vis spectra were
obtained by using a Perkin–Elmer Lambda-25 UV/Vis spectropho-
tometer and inflections are identified by the abbreviation inf. IR
spectra were recorded by using a Shimidazu FTIR-NIR Prestige-21
spectrometer equipped with a Pike Miracle Ge ATR accessory;
strong and medium peaks are indicated by s and m, respectively.
Low-resolution (EI) mass spectra were recorded by using a Shimad-
zu Q2010 GCMS instrument with direct inlet probe. MALDI-TOF MS
were conducted by using a Bruker BIFLEX III time-of-flight (TOF)
mass spectrometer. Cyclic voltammetry (CV) measurements were
performed by using a Princeton Applied Research Potentiostat/Gal-
vanostat 263A apparatus. The concentration of the benzotriazinyl
radical used was 1 mm in CH2Cl2. A solution of tetrabutylammoni-
um tetrafluoroborate (nBu4NBF4; 0.1m) in CH2Cl2 was used as the
electrolyte. The electrolyte was dried for 4 d in the vacuum oven at
1008C prior to the use. The reference electrode was Ag/AgCl and
the scan rate was 50 mVsꢁ1. Ferrocene was used as an internal ref-
erence; the E1/2(ox) of ferrocene in this system was 0.352 V.[18] EPR
spectra were recorded by using a Bruker EMXplus X-band EPR
spectrometer at RT on solid-state samples of the imidazolo-fused
1,2,4-benzotriazinyls and on dilute solutions in CH2Cl2. For the
dilute solution spectra, the microwave power was in the region of
5–70 mW with modulation frequencies of 50 or 100 kHz and mod-
ulation amplitudes of 0.5–1.0 Gpp. Simulations of the solution spec-
tra were made by using Winsim.[19] The near-isotropic nature of
most benzotriazinyl radical samples meant that the majority of
solid-state samples could be initially modeled as an isotropic spec-
trum by using Winsim, but those revealing slight line asymmetry
were simulated by using PIP run through the GUI “PIP for Win-
dows”.[20] Magnetic properties were studied by using a Quantum
Design SQUID MPMS2 field-shielded magnetometer. The DC (direct
Experimental Section
Synthetic procedure
The synthesis of imidazolo-fused radicals 5–7 was published pre-
viously.[7b]
Spectral analyses
1,3,7,8-Tetraphenyl-4,8-dihydro-1H-imidazo[4,5g]-
[1,2,4]benzotriazin-4-yl (5):[7b] Black needles recrystallized from
PhH (78 mg, 81%). M.p. (DSC): 308.6 (decomp. onset), 312.48C
(peak max); Rf (Et2O): 0.64; UV/Vis (CH2Cl2): lmax (loge)=239 (3.30),
272 inf. (3.54), 291 (3.61), 330 inf. (3.02), 403 inf. (3.26), 419 (3.35),
465 inf. (2.63), 540 (2.50), 624 nm inf. (2.16); IR (ATR, Ge): n˜max
=
1595 (m), 1503 (m), 1489 (s), 1470 (s), 1452 (m), 1435 (s), 1402 (s),
1389 (s), 1350 (s), 1287 (m), 1180 (m), 1026 (m), 895 (m), 854 (s),
831 (m), 777 (s), 770 (s), 762 cmꢁ1 (s); MS (EI): m/z (%): 476 [M]+
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
current) magnetic moments were measured for 42.2, 36.4, and
50.3 mg samples of radicals 5–7, respectively, which were placed in
gelatin capsules held by a polyethylene straw. The magnetic sus-
ceptibilities were measured in the temperature range of 5–300 K in
an applied field of 0.4 T. Data were collected in both warming and
cooling modes with no significant differences in sample suscepti-
bility. Data were corrected by using diamagnetic contributions of
cdia =ꢁ350, ꢁ370, and ꢁ700 (two radicals per dimer)ꢃ
10ꢁ3 emumolꢁ1 for radicals 5–7, respectively. X-Ray data for radi-
cals 5–7 were collected by using an Oxford Diffraction Supernova
diffractometer equipped with a CCD area detector with CuKa radia-
tion (l=1.5418 ꢂ) for 5 and 6 and MoKa radiation (l=0.71073 ꢂ)
for 7. A suitable crystal was attached to a glass fiber by using para-
tone-N oil, transferred to a goniostat, and cooled before data col-
lection. Unit cell dimensions were determined and refined by using
3520 (3.34ꢃqꢃ72.478) reflections for 5, 3007 (5.65ꢃqꢃ68.228) re-
flections for 6, and 3054 (3.08ꢃqꢃ28.968) reflections for 7. Empiri-
cal absorption corrections (multiscan based on symmetry-related
measurements) were applied by using CrysAlis RED software.[21]
The structures were solved by direct method and refined on F2 by
using full-matrix least squares using SHELXL97.[22] Software pack-
ages used: CrysAlis CCD[21] for data collection, CrysAlis RED[21] for
cell refinement and data reduction, WINGX for geometric calcula-
tions,[23] and Mercury 3.1[24] for molecular graphics. The non-hydro-
gen atoms were treated anisotropically. The hydrogen atoms were
placed in calculated, ideal positions and refined as riding on their
respective carbon atoms.
basis set were needed to take the Br atom into account. All the cal-
culations were performed by using the Gaussian 03 suite of pro-
grams.[25]
Acknowledgements
The authors wish to thank the University of Cyprus (medium-
size grants), the Cyprus Research Promotion Foundation (grant
nos. YGEIA/BIOS/0308(BIE)/13, NEKYP/0308/02, and ANA
BAVMISH/0308/32) and the following organizations in
Cyprus for generous donations of chemicals and glassware:
the State General Laboratory, the Agricultural Research Insti-
tute, the Ministry of Agriculture, MedoChemie Ltd., and Bio-
tronics Ltd. Furthermore, we thank the A. G. Leventis Founda-
tion for helping to establish the NMR facility in the University
of Cyprus. J.M.R. would like to thank the CFI and ORF for finan-
cial support. The work in Israel was supported by the Israeli
Science Foundation.
Keywords: antiferromagnetism
· exchange interactions ·
ferromagnetism · heterocycles · radicals
[1] I. Ratera, J. Veciana, Chem. Soc. Rev. 2012, 41, 303–349.
[2] a) D. A. Haynes, CrystEngComm 2011, 13, 4793–4805; b) O. A. Rakitin,
Russ. Chem. Rev. 2011, 80, 647–659; c) R. G. Hicks, Org. Biomol. Chem.
2007, 5, 1321–1338; d) S. Nakatsuji, H. Anzai, J. Mater. Chem. 1997, 7,
2161–2174; e) J. M. Rawson, A. Alberola, A. Whalley, J. Mater. Chem.
2006, 16, 2560–2575; f) H. Iwamura, Polyhedron 2013, 66, 3–14.
[3] a) Magnetic Properties of Organic Materials (Ed.: P. M. Lahti), Marcel-
Dekker Inc., New York, 1999; b) Molecular Magnetism: New Magnetic
Materials (Eds.: K. Itoh, M. Kinoshita), Kodansha and Gordon & Breach,
Tokyo and Amsterdam, 2000.
Crystal refinement data for 5: C32H22N5; Mr =476.55; triclinic;
¯
space group P1; a=7.8924(8), b=11.3721(9), c=13.2802(12) ꢂ; a=
96.019(7), b=90.262(8), g=96.237(8)8; V=1178.20(19) ꢂ3; Z=2;
T=100(2) K; 1calcd =1.343 gcmꢁ3; 2qmax =67. Refinement of 334 pa-
rameters on 4186 independent reflections out of 7408 measured
reflections (Rint =0.0284) gave R1 =0.0452 [I>2s(I)], wR2 =0.1356
(all data), and S=1.044 with the largest difference peak and hole
of 0.220 and ꢁ0.315 eꢁ3, respectively.
[4] a) J. A. Crayston, J. N. Devine, J. C. Walton, Tetrahedron 2000, 56, 7829–
7857; b) S. J. Blundell, F. L. Pratt, J. Phys. Condens. Matter 2004, 16,
R771–R828.
Crystal refinement data for 6: C32H21N5Br; Mr =566.00; triclinic;
¯
space group P1; a=7.8374(12), b=11.7232(16), c=13.2724(10) ꢂ;
[5] H. M. Blatter, H. Lukaszewski, Tetrahedron Lett. 1968, 9, 2701–2705.
[6] a) C. P. Constantinides, P. A. Koutentis, H. Krassos, J. M. Rawson, A. J. Ta-
siopoulos, J. Org. Chem. 2011, 76, 2798–2806; b) C. P. Constantinides,
P. A. Koutentis, J. M. Rawson, Chem. Eur. J. 2012, 18, 7109–7116; c) C. P.
Constantinides, P. A. Koutenis, J. M. Rawson, Chem. Eur. J. 2012, 18,
15433–15438; d) C. P. Constantinides, E. Carter, D. M. Murphy, M.
Manoli, G. M. Leitus, M. Bendikov, J. M. Rawson, P. A. Koutentis, Chem.
Commun. 2013, 49, 8662–8664; e) C. P. Constantinides, A. A. Berezin, M.
Manoli, G. M. Leitus, M. Bendikov, J. M. Rawson, P. A. Koutentis, New J.
Chem. 2014, 38, 949–954; f) B. Yan, J. Cramer, R. McDonald, N. L. Frank,
Chem. Commun. 2011, 47, 3201–3203.
[7] a) A. A. Berezin, C. P. Constantinides, C. Drouza, M. Manoli, P. A. Kouten-
tis, Org. Lett. 2012, 14, 5586–5589; b) A. A. Berezin, C. P. Constantinides,
S. I. Mirallai, M. Manoli, L. L. Cao, J. M. Rawson, P. A. Koutentis, Org.
Biomol. Chem. 2013, 11, 6780–6795.
[8] a) P. A. Koutentis, D. Lo Re, Synthesis 2010, 2075–2079; b) C. P. Constan-
tinides, P. A. Koutentis, G. Loizou, Org. Biomol. Chem. 2011, 9, 3122–
3125; c) T. A. Ioannou, P. A. Koutentis, H. Krassos, G. Loizou, D. Lo Re,
Org. Biomol. Chem. 2012, 10, 1339–1348; d) P. A. Koutentis, H. Krassos,
D. Lo Re, Org. Biomol. Chem. 2011, 9, 5228–5237; e) P. A. Koutentis, G.
Loizou, D. Lo Re, J. Org. Chem. 2011, 76, 5793–5802; f) M. Demetriou,
A. A. Berezin, P. A. Koutentis, T. Krasia-Christoforou, Polym. Int. 2013, 63,
674–679;
a=87.875(8), b=88.185(9), g=87.875(11)8; V=1076.2(3) ꢂ3; Z=2;
T=100(2) K; 1calcd =1.515 gcmꢁ3; 2qmax =67. Refinement of 343 pa-
rameters on 4250 independent reflections out of 8137 measured
reflections (Rint =0.0944) gave R1 =0.0863 [I>2s(I)], wR2 =0.2801
(all data), and S=1.028 with the largest difference peak and hole
of 2.146 and ꢁ1.120 eꢁ3, respectively.
Crystal refinement data for 7: C33H24N5O; Mr =506.57; triclinic;
¯
space group P1; a=8.2696(10), b=11.457(3), c=13.8478(17) ꢂ; a=
83.013(15), b=87.393(10), g=81.254(14)8; V=1286.6(4) ꢂ3; Z=2;
T=100(2) K; 1calcd =1.308 gcmꢁ3, 2qmax =25. Refinement of 353 pa-
rameters on 4525 independent reflections out of 8256 measured
reflections (Rint =0.0353) gave R1 =0.0591 [I>2s(I)], wR2 =0.1720
(all data), and S=1.082 with the largest difference peak and hole
of 0.275 and ꢁ0.243 eꢁ3, respectively.
CCDC-940089 (5), -972401 (6), and -972402 (7) contain the supple-
mentary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
Computational methods
[9] a) F. A. Neugebauer, G. Rimmler, Magn. Reson. Chem. 1988, 26, 595–
600; b) F. A. Neugebauer, I. Umminger, Chem. Ber. 1980, 113, 1205–
1225; c) F. A. Neugebauer, I. Umminger, Chem. Ber. 1981, 114, 2423–
2430.
[10] aN =QN1N, in which aN is the hfcc in Gauss; QN =21.2 G (for N2 and N4),
and QN =25 G (for N1).
Single-point calculations were carried out on the X-ray crystallo-
graphically determined geometries of radical pair I–II and II–III of
5–7 at the UB3LYP/6-311++G(d,p) level of theory to calculate the
energies of the triplet (ET) and broken-symmetry singlet (EBS) states.
For radical 6, additional calculations with the 6-311++G(2d,p)
[11] H. M. McConnell, J. Chem. Phys. 1958, 28, 1188–1192.
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
8
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[12] J. W. Hall, W. E. Marsh, R. R. Weller, W. E. Hatfield, Inorg. Chem. 1981, 20,
1033–1037.
[24] C. F. Macrae, P. R. Edgington, P. McCabe, E. Pidcock, G. P. Shields, R.
Taylor, M. Towler, J. van de Streek, J. Appl. Crystallogr. 2006, 39, 453–
457.
[13] B. Bleaney, K. D. Bowers, Proc. R. Soc. London Ser. A 1952, 214, 451–465.
[14] J. J. Novoa, M. Deumal, J. Jornet-Somoza, Chem. Soc. Rev. 2011, 40,
3182–3212.
[25] Gaussian 03, Revision C.02, M. J. Frisch, 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. Nakatsu-
ji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Na-
kajima, 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. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Och-
terski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, 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. Al-Laham,
C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, and J. A. Pople, Gaussian, Inc., Walling-
ford CT, 2004.
[15] L. Noodleman, J. Chem. Phys. 1981, 74, 5737–5743.
[16] M. Deumal, J. J. Novoa, M. J. Bearpark, P. Celani, M. Olivucci, M. A. Robb,
J. Phys. Chem. A 1998, 102, 8404–8412.
[17] M. Mito, Y. Komorida, H. Tsuruda, J. S. Tse, S. Desgreniers, Y. Ohishi, A. A.
Leitch, K. Cvrkalj, C. M. Robertson, R. T. Oakley, J. Am. Chem. Soc. 2009,
131, 16012–16013.
[18] M. Dietrich, J. Heinze, J. Am. Chem. Soc. 1990, 112, 5142–5145.
[19] D. A. O’Brien, D. R. Duling, Y. C. Fann, Winsim v0.98, EPR spectral simula-
tion for MS-Windows, Public EPR Software Tools NIEHS.
[20] M. Nilges, Illinois EPR Research Centre; PIP for Windows (v.2.1), J. M.
Rawson, University of Windsor (2011).
[21] Oxford Diffraction (2008). CrysAlis CCD and CrysAlis RED, version
1.171.32.15, Oxford Diffraction Ltd, Abingdon, Oxford, England.
[22] G. M. Sheldrick, “SHELXL97 A program for the refinement of crystal
structure”, University of Gçttingen, Germany.
Received: November 19, 2013
[23] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837–838.
Published online on && &&, 0000
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
9
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
&
Magnetic Properties
C. P. Constantinides, A. A. Berezin,
M. Manoli, G. M. Leitus, G.A. Zissimou,
M. Bendikov, J. M. Rawson,
P. A. Koutentis*
&& – &&
Structural, Magnetic, and
Computational Correlations of Some
Imidazolo-Fused 1,2,4-Benzotriazinyl
Radicals
Bigger is stronger: p-Extension of 1,2,4-
benzotriazin-4-yl radicals leads to an en-
hanced overlap between SOMO orbitals
and, therefore, to a strong effective ex-
change coupling within alternating 1D
chains of imidazolo-benzotriazinyls.
However, subtle changes in interplanar
distances and slippage angles switch
the observed magnetic properties (see
figure).
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
10
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!