Bistabilities in 1,3,2-dithiazolyl radicals

Article abstract of DOI:10.1021/ja048618m

New synthetic methods for heterocyclic 1,3,2-dithiazolyl (DTA) radicals have been developed, and trends in the molecular spin distributions and electrochemical properties of a series of DTA radicals are reported. The crystal structures of [1,2,5]thiadiazolo[3,4-f[1,3,2]benzodithiazol-2-yl (TBDTA) and [1,3,2]-pyrazinodithiazol-2-yl (PDTA) have been determined. The structure of TBDTA (at 293 and 95 K) contains two molecules in the asymmetric unit, each of which generates π-stacked arrays, one consisting of antiparallel chains of centrosymmetrically associated dimers, the other comprising parallel chains of unassociated radicals. The structure of PDTA (at 293 and 95 K) is simpler, consisting of slipped stacks of π-dimers. Variable-temperature magnetic susceptibility (χ_P) measurements on TBDTA indicate essentially paramagnetic behavior for the unassociated radical π-stacks over the range 5-400 K. By contrast PDTA is diamagnetic at all temperatures below 300 K, but between 300 and 350 K the value of χ_P follows a sharp and well-defined hysteresis loop, with Tc↓ = 297 K and Tc↑ = 343 K. These features are symptomatic of a regime of bistability involving the observed low temperature π-dimer structure and a putative high-temperature radical π-stack. A mechanism for the interconversion of the two phases of PDTA and related structures is proposed in which hysteretic behavior arises from cooperative effects associated with the breaking and making of a lattice-wide network of intermolecular S- - -N′ and/or S- - -S′ interactions.

Full text of DOI:10.1021/ja048618m

Published on Web 06/12/2004
Bistabilities in 1,3,2-Dithiazolyl Radicals
Jaclyn L. Brusso,1a Owen P. Clements,1a Robert C. Haddon,1b Mikhail E. Itkis,1b
Alicea A. Leitch,1a Richard T. Oakley,*,1a Robert W. Reed,1a and
John F. Richardson1c
Contribution from the Department of Chemistry, UniVersity of Waterloo, Waterloo,
Ontario N2L 3G1, Canada, Department of Chemistry and Center for Nanoscale Science and
Engineering, UniVersity of California, RiVerside, California 92521-0403, and Department of
Chemistry, UniVersity of LouisVille, LouisVille, Kentucky 40292
Received March 10, 2004; E-mail: oakley@sciborg.uwaterloo.ca
Abstract: New synthetic methods for heterocyclic 1,3,2-dithiazolyl (DTA) radicals have been developed,
and trends in the molecular spin distributions and electrochemical properties of a series of DTA radicals
are reported. The crystal structures of [1,2,5]thiadiazolo[3,4-f][1,3,2]benzodithiazol-2-yl (TBDTA) and [1,3,2]-
pyrazinodithiazol-2-yl (PDTA) have been determined. The structure of TBDTA (at 293 and 95 K) contains
two molecules in the asymmetric unit, each of which generates π-stacked arrays, one consisting of
antiparallel chains of centrosymmetrically associated dimers, the other comprising parallel chains of
unassociated radicals. The structure of PDTA (at 293 and 95 K) is simpler, consisting of slipped stacks of
π-dimers. Variable-temperature magnetic susceptibility (øP) measurements on TBDTA indicate essentially
paramagnetic behavior for the unassociated radical π-stacks over the range 5-400 K. By contrast PDTA
is diamagnetic at all temperatures below 300 K, but between 300 and 350 K the value of øP follows a sharp
and well-defined hysteresis loop, with TCV ) 297 K and TCv ) 343 K. These features are symptomatic of
a regime of bistability involving the observed low temperature π-dimer structure and a putative high-
temperature radical π-stack. A mechanism for the interconversion of the two phases of PDTA and related
structures is proposed in which hysteretic behavior arises from cooperative effects associated with the
breaking and making of a lattice-wide network of intermolecular S- - -N′ and/or S- - -S′ interactions.
Chart 1
Introduction
The potential applications of organic radicals in molecular
spin-electronic or “spintronic” devices has catalyzed efforts to
develop new stable and magnetically active molecules.2 Within
this context the solid-state structures and magnetic properties
of 1,3,2-dithiazolyl (DTA) radicals (Chart 1) have been a focus
of recent research. However, despite the near-zero dimerization
enthalpy of the dithiazolyl core,3 the monocyclic derivatives 1
5
(R1 ) R2 ) CN,4 CF3 ) crystallize as diamagnetic dimers (at
room temperature) aligned cofacially and linked by weak
interannular S- - -S bonds; the dimers are then closely associated
(pairwise) into tetrameric clusters. In condensed-ring dithiazolyls
clustering of radicals is not observed, but spin-quenching is still
possible, as in the benzene-based derivative BDTA 2,3 which
(1) (a) University of Waterloo. (b) University of California. (c) University of
Louisville.
displays a rare centrosymmetric mode of association. By
contrast, the naphthalene compound NDTA 3 is paramagnetic
(at room temperature). Its structure consists of discrete radicals
locked into herringbone arrays,6 with neighboring rings oriented
in the “tilted-T” arrangement which favors structure-making
CH- - -aromatic ring interactions.7,8 Replacement of CH groups
(2) (a) Miller, J. S.; Drillon, M. Magnetism: Molecules to Materials II,
Molecule-Based Materials; Wiley-VCH: New York, 2001. (b) Gregg, J.;
Allen, W.; Viart, N.; Kirschman, R.; Sirisathitkul, C.; Schille, J. P.; Gester,
M.; Thompson, S.; Sparks, P.; Da Costa, V.; Ounadjela, K.; Skvarla, M. J.
Magn. Magn. Mater. 1997, 175, 1. (c) Coey, J. M. D. J. Magn. Magn.
Mater. 1999, 197, 1.
(3) (a) Awere, E. G.; Burford, N.; Haddon, R. C.; Parsons, S. Passmore, J.
Waszczak, J. V.; White, P. S. Inorg. Chem. 1990, 29, 4821. (b) Awere, E.
G.; Burford, N.; Mailer, C.; Passmore, J.; Schriver, M. J.; White, P. S.
Banister, A. J.; Oberhammer, H.; Sutcliffe, L. H. J. Chem. Soc., Chem.
Commun. 1987, 66.
(4) Wolmersha¨user, G.; Kraft, G. Chem. Ber. 1990, 123, 881.
(5) Du, H.; Haddon, R. C.; Krossing, I.; Passmore, J.; Rawson, J. M.; Schriver,
M. J. Chem. Commun. 2002, 1836.
(6) Barclay, T. M.; Cordes, A. W.; George, N. A.; Haddon, R. C.; Oakley, R.
T.; Palstra, T. T. M.; Patenaude, G. W.; Reed, R. W.; Richardson, J. F.;
Zhang, J. J. Chem. Soc., Chem. Commun. 1997, 873.
(7) Hobza, P.; Selzle, H. L.; and Schlag, E. W. J. Am. Chem. Soc. 1994, 116,
3500
9
8256
J. AM. CHEM. SOC. 2004, 126, 8256-8265
10.1021/ja048618m CCC: $27.50 © 2004 American Chemical Society

Bistabilities in 1,3,2-Dithiazolyl Radicals
A R T I C L E S
by nitrogen, as in the pyrazine9 and quinoxaline6 derivatives
PDTA 4 and QDTA 5, militates against the “tilted-T” motif,
and both these radicals adopt π-stacked structures. In PDTA
the radicals were reported to be associated at room temperature,
while in QDTA they remain undimerized, although on cooling
the unpaired spins are quenched, presumably as a result of
dimerization.9
Scheme 1
In the thiadiazole- and thiadiazolopyrazine-based compounds,
TDTA 610,11 and TPDTA 7,12 there are no peripheral hydrogens,
and consistently, slipped π-stack arrays are observed in the solid
state. As an added feature both materials exhibit a form of
bistability13 in which a paramagnetic phase (undimerized
π-stacks) coexists over a specific temperature range with a
diamagnetic phase (π-dimer stacks).14 The structural intercon-
versions are characterized by a wide hysteresis loop in the
temperature dependence of the magnetic susceptibility.
As a continuation of our magneto-structural investigations
of condensed-ring 1,3,2-dithiazolyls, we have prepared and
characterized the thiadiazolobenzene derivative TBDTA 8, in
which the pyrazine residue of TPDTA 7 has been replaced by
a benzene ring. Our initial belief was that this minor molecular
modification would not alter the slipped stack packing found
for TPDTA and would allow a comparison of the degree of
magneto-thermal hysteresis in the two compounds. We have
also reexamined the pyrazine derivative PDTA 4, in the hope
of identifying a bistability similar to that found in TDTA 6.
Contrary to expectation, TBDTA adopts a crystal structure with
subtleties hitherto unobserved for 1,3,2-DTA radicals and
displays no magneto-thermal hysteresis. Satisfyingly, however,
the magnetic susceptibility of PDTA exhibits a well-defined
high-temperature hysteresis heralding the coexistence of two
polymorphs. A mechanism for the phase interconversion of
TDTA and PDTA is proposed to account for the bistability in
these systems.
Scheme 2
chloride [8][Cl], were quite straightforward. The latter com-
pound could be metathesized to a hexafluoroantimonate salt [8]-
[SbF6] for characterization purposes or reduced directly with
triphenylantimony to the desired radical TBDTA 8. This latter
step gave poor yields (<15%) regardless of the choice of solvent
(acetonitrile or liquid SO2), reductant (Ag, Zn, Ph3Sb), or starting
salt (chloride or hexafluoroantimonate). Once isolated, however,
TBDTA was relatively air stable in the solid state and was easily
purified by vacuum sublimation.
Results
Synthesis. Preparation of the initial target TBDTA 8 (Scheme
1) began with the oxidative thiocyanation of o-phenylenediamine
9 with bromine and potassium thiocyanate in methanol at -78
°C, to afford the bis-thiocyanate 10. Although this reaction was
reported over forty years ago,15 a number of modifications, e.g.,
the use of low temperatures, slow addition rates, pH control
during workup, were required to achieve consistent results.
Closure of the thiadiazole ring with thionyl chloride, and
reductive cleavage of the thiocyanate 11 with aqueous sulfide
to yield the dithiol 12, followed well-established procedures.
Likewise oxidation of 12 to the bis(sulfenyl chloride) 13, and
cyclization of 13 with trimethylsilyl azide to the dithiazolylium
The pyrazine radical PDTA 4 was synthesized using a
modification (Scheme 2) of the procedure originally described
by Kraft and Wolmersha¨user.16 Accordingly, the dithiouronium
salt 15, prepared from dichloropyrazine 14, was hydrolyzed in
basic solution and the resulting dithiolate quenched with acetic
anhydride (rather than acetic acid) to afford the bis-thioacetate
16. After purification, the latter was converted into the bis-
(sulfenyl chloride) 17, which was condensed with trimethylsilyl
azide to give the crude dithiazolylium chloride salt [4][Cl]. As
above, PDTA 4 was liberated by reduction of the crude salt
with triphenylantimony and purified by vacuum sublimation.
(8) (a) Gavezzotti, A.; Desiraju, G. R. Acta Crystallogr. 1988, B44, 427. (b)
Bernstein, J.; Sarma, J. A. R. P.; Gavezzotti, A. Chem. Phys. Lett. 1990,
174, 361.
(9) Heckman, G.; Johann, R.; Kraft, G.; Wolmersha¨user, G. Synth. Met. 1991,
41-43, 3287.
(10) Fujita, W.; Awaga, K. Science 1999, 286, 261.
(11) McManus, G. D.; Rawson, J. M.; Feeder, N.; van Duijn, J.; McInnes, E. J.
L.; Novoa, J. J.; Burriel, R.; Palacio, F.; Oliete, P. J. Mater. Chem. 2001,
11, 1992.
(12) Barclay, T. M.; Cordes, A. W.; George, N. A.; Haddon, R. C.; Itkis, M.
E.; Mashuta, M. S.; Oakley, R. T.; Patenaude, G. W.; Reed, R. W.;
Richardson, J. F.; Zhang, H. J. Am. Chem. Soc. 1998, 120, 352.
(13) (a) Kahn, O.; Martinez, C. J. Science 1998, 279, 44. (b) Kahn, O. Chem.
Br. 1999, 35, 24.
(14) Itkis, M. E.; Chi, X.; Cordes, A. W.; Haddon, R. C. Science 2002, 296,
1443.
(15) Raby, C. Ann. Chim. 1961, 6, 481.
EPR Spectra and Electrochemistry We have characterized
the spin distribution of TBDTA 8 in solution by EPR spectros-
copy. To place the results in context we have revisited the EPR
spectra of compounds 2-7. While hyperfine coupling constants
have been reported for all of these materials,3,12,17 the data refer
(16) Wolmersha¨user, G.; Kraft, G. Chem. Ber. 1989, 122, 385.
9
J. AM. CHEM. SOC. VOL. 126, NO. 26, 2004 8257

A R T I C L E S
Brusso et al.
Figure 2. CV scan (100 mV/S) on TBDTA 8 in MeCN, [n-Bu4N][PF6]
supporting electrolyte.
As might be expected, the spin distributions of TPDTA and
TBDTA are similar, although the pyrazine residue in the former
is a substantially more effective electron sink than a benzene
ring.
To generate a quantitative comparison19 of the redox proper-
ties of the heterocyclic 1,3,2-DTAs 2-8, we have performed
cyclic voltammetry measurements on the entire set. As an
example the CV scan from -1.0 to +1.0 V (vs SCE) for
TBDTA 8 is shown in Figure 2. The resulting half-wave
potentials for oxidation and reduction for all the compounds
are listed in Table 1. As can be seen from these data, variations
in the nature of heterocyclic substituent have a major effect on
Figure 1. X-band EPR spectra of 1,3,2-DTA radicals, at 293 K in CH2-
Cl2, SW ) 4.0 mT.
Table 1. Isotropic Hyperfine g-Values, Coupling Constants,a and
Half-Waveb Potentials
radical
g
aN
aH
E (-1/0) E (0/+1)
2 2
(0/+1)
BDTA, 2
2.0069 1.127
(1.130)
2.0067 1.140
(1.128)
2.0071 1.126, 0.070
(1.118, 0.060)
2.0065 1.089, 0.129
(1.099, 0.101)
0.057, 0.045
(-0.066, -0.047)
0.097
(-0.110)
0.120
-1.20c 0.15
-1.08c 0.27
-0.88c 0.53
-0.73c 0.62
0.65
the redox characteristics of the radical, e.g., the value of E1/2
changes by almost 1 V. All the radicals show a reversible wave
for 0/+1 couple, but only for TPDTA and TBDTA are the waves
for the -1/0 couple reversible. The new radical TBDTA can
be considered a hybrid of several of the previously reported
materials, but electrochemically the nearest match is with PDTA.
From a chemical perspective, the relative ease of aerial oxidation
of TBDTA, compared to the complete stability of TDTA and
TPDTA toward molecular oxygen, is easily rationalized.
Crystal Structure of TBDTA 8. Blocklike crystals of
TBDTA, grown by dynamic vacuum sublimation at 10-3 Torr
along a temperature gradient of 100-45 °C, belong to the
monoclinic space group P21/c. Single-crystal X-ray data ob-
tained at 293 and 95 K are summarized in Table 2, and listings
of pertinent intra- and intermolecular contacts are provided in
Table 3. The most notable feature of this structure, which sets
it apart from all other reported dithiazolyl radicals, is the
presence of two environmentally distinct molecules (A and B)
in the asymmetric unit, i.e., Z ) 8. ORTEP drawings of the
two molecules are shown in Figure 3. One of these (A)
constitutes half of a centrosymmetric dimer, with an intradimer
S1- - -S2′ distance of 3.252(2) Å at 293 K, while the other (B)
remains unassociated. The S-N and S-C distances (Table 3)
within the DTA rings of the two molecules are, however,
remarkably similar, as expected from the fact that the enthalpy
of dimerization of DTA radicals is negligible.3 Other than a
general shortening of both the unit cell parameters and the
intermolecular distances, there is very little structural difference
between the low- and high-temperature data.
NDTA, 3
PDTA, 4
QDTA, 5
TDTA, 6
(-0.104)
0.065, 0.023
(-0.050, -0.024)
-0.66c
2.0061 1.115, 0.084
(1.111, 0.082)
TPDTA, 7 2.0072 0.959, 0.209, 0.028
(1.032, 0.164, 0.038)
TBDTA, 8 2.0072 1.098, 0.020
(1.025, 0.036)
-0.06 1.06
-0.76 0.46
0.16
(-0.176)
a In mT, X-band spectra at 293 K, in CH2Cl2; computed (B3LYP/6-
31G**) values are in parentheses. b Volts vs SCE, in MeCN. c Irreversible
wave, Epc cited.
to the use of different temperatures, solvents, and even isotope
content, and these variations hinder direct comparisons within
the series. In addition we have in several cases been able to
resolve several previously unresolved couplings. As may be seen
in Figure 1, all the spectra (in CH2Cl2, 293 K) are characterized
by a three-line pattern arising from strong hyperfine coupling
to the dithiazolyl nitrogen. Delocalization of spin density from
the DTA ring onto the adjacent ring leads to a slight decrease
in value of aN, and hyperfine coupling to the peripheral benzene
protons, or pyrazine/dithiazole nitrogens, is observed. Additional
but smaller coupling to the protons (or nitrogens) of the third
ring can also be extracted by spectral simulation. Full analyses
of the experimental spectra along with calculated (B3LYP/6-
31G**) coupling constants are presented in Table 1. As noted
elsewhere, the predictive ability of this level of theory is good.18
The A and B molecules pack in slipped π-stack arrangements
running along the x- and z-directions respectively, as illustrated
in Figure 4. When the full unit cell is viewed along z (Figure
5), the ribbonlike nature of the molecular arrays becomes
(17) (a) Preston, K. F.; Sandall, J. P. B.; Sutcliffe, L. H. Magn. Reson. Chem.
1988, 26, 755. (b) Chung, Y.-L.; Sandall, J. P. B.; Sutcliffe, L. H.; Joly,
H.; Preston, K. F.; Johann, R.; Wolmersha¨user, G. Magn. Res. Chem. 1991,
31, 625. (c) Chung, Y.-L.; Fairhurst, S. A.; Gullies, D. G.; Kraft, G.;
Krebber, A. M. L.; Preston, K. F.; Sutcliffe, L. H.; Wolmersha¨user, G.
Magn. Reson. Chem. 1992, 30, 774.
(19) (a) Kaszynski, P. J. Phys. Chem. A 2001, 105, 7626. (b) Boere´, R. T.;
Roemmele, T. L. Coord. Chem. ReV. 2000, 210, 369.
(18) Kaszynski, P. J. Phys. Chem. A 2001, 105, 7615.
9
8258 J. AM. CHEM. SOC. VOL. 126, NO. 26, 2004

Bistabilities in 1,3,2-Dithiazolyl Radicals
A R T I C L E S
Table 2. Crystallographic Data
compound
TBDTA, 8
LT-PDTA, 4
formula
fw
C6H2N3S3
212.29
C6H2N3S3
212.29
C4H2N3S2
156.21
C4H2N3S2
156.21
a, Å
b, Å
c, Å
9.938(3)
22.373(6)
7.411(2)
-
9.8847(6)
22.153(1)
7.2905(5)
-
6.9963(7)
7.8859(7)
11.4386(11)
110.257(1)
95.155(2)
103.625(2)
565.14(9)
1.836
6.8564(8)
7.8466(9)
11.3107(13)
109.965(2)
94.944(2)
103.483(2)
546.95(11)
1.897
R, deg
â, deg
111.761(4)
-
111.549(1)
-
γ, deg
V, Å3
1530.5(7)
1.843
P21/c
1484.8(2)
1.899
P21/c
F(calcd), g cm-3
space group
Z
P1h
4
P1h
4
8
8
temp, K
293(2)
0.902
0.71073
2851/0/233
direct methods
0.0423, 0.0915
95(2)
0.929
0.71073
3121/0/233
direct methods
0.0303, 0.0691
293(2)
0.828
0.71073
2100/0/179
direct methods
0.0341, 0.0879
95(2)
0.856
0.71073
1965/ 0/179
direct methods
0.0339, 0.0801
µ, mm-1
λ, Å
data/restraints/parameters
solution method
R, Rw (on F2)a
2
2 2
4
a R ) [∑|Fo| - |Fc|]/[∑|Fo|] for I > 2σ(I); Rw ) {[∑w|Fo| - |Fc| | ]/[∑(w|Fo| )]}2.
Table 3. Summary of Intra- and Intermolecular Distances (Å)a in
TBDTA 8
molecule A
293 K 95 K
molecule B
293 K 95 K
S1-N1
S2-N1
S1-C1
S2-C2
S1- - -S2′
S3- - -S2′
1.640(2) 1.644(2)
1.648(2) 1.648(2)
1.749(2) 1.749(2)
1.750(2) 1.746(2)
3.252(2) 3.201(2)b
S4-N4
S5-N4
S4-C7
S5-C8
-
1.647(2) 1.652(2)
1.645(2) 1.648(2)
1.746(2) 1.747(2)
1.744(2) 1.748(2)
3.579(2) 3.514(2)c S6- - -S4′
3.565(2) 3.501(2)c
S3- - -N1′ 3.243(2) 3.174(2)c S6- - -N4′ 3.190(2) 3.109(2)c
a ESDs are in parentheses. b Intradimer contacts. c Lateral contacts.
Figure 4. Packing of TBDTA 8, viewed down y, showing the crossed
π-stack arrays. The dashed lines connect the sulfurs of the centrosymmetric
dimers of radical A.
Figure 3. ORTEP drawing (50% probability ellipsoids) of the asymmetric
unit of TBDTA 8 (at 293 K), showing atom numbering.
evident. Within the B stacks consecutive molecules along z are
slightly offset to either side of the glide plane (at y ) 1/4, 3/4),
while in the A stacks neighboring molecules are related by
centers of symmetry. In both stacks neighboring molecules along
the ribbons are connected by close (shorter than van der Waals20)
S- - -N′ and S- - -S′ contacts (Table 3). A comparison of the
layers within the two π-stacks (Figure 6) illustrates the anti-
parallel nature of neighboring A stack ribbons, which sets up
the centrosymmetric dimerization mode noted above. By
contrast, all the molecules within the B stacks run parallel.
Crystal Structure of LT-PDTA. PDTA 4 can exist in one
of two polymorphs, hereafter referred to as the low-temperature
(LT) and high-temperature (HT) phases.21 In this contribution
we describe the crystal structure of the former. Lustrous blue-
Figure 5. Unit cell of TBDTA 8, viewed down z, showing ribbons of A
(at y ) 0, 1/2, 1) and B (at y ) 1/4, 3/4).
black rods of LT-PDTA suitable for X-ray work were grown
by static (sealed tube) sublimation at 10-2 Torr along a
temperature gradient of 60-30 °C. Under these conditions it is
likely that both polymorphs, i.e., the LT and HT phases, are
produced, but upon slow cooling of the sublimate to room
temperature the HT phase slowly reverts to the LT form. Close
inspection of the harvested sublimate revealed that the majority
of crystals were heavily twinned. This observation may reflect
the fact that the HT phase is itself twinned or that twinning is
(20) Bondi, A. J. Phys. Chem. 1964, 68, 441
(21) The packing pattern of LT-PDTA was communicated some years ago by
Wolmersha¨user and co-workers (ref 9), but details of the crystal data and
structural parameters were not reported.
9
J. AM. CHEM. SOC. VOL. 126, NO. 26, 2004 8259

A R T I C L E S
Brusso et al.
Figure 8. Unit cell of LT-PDTA. Lateral S- - -S′ and 6-center S- - -N′
contacts (dashed lines) are defined in Table 4.
Figure 6. Slipped π-stacks of A and B molecules in TBDTA 8.
Intermolecular S- - -S′ and S- - -N′ contacts (dashed lines) are defined in
Table 3.
Figure 9. Slipped π-stacks of LT-PDTA dimers. Intermolecular S- - -S′
and S- - -N′ contacts (dashed lines) are defined in Table 4.
Table 4. Summary of Intra- and Intermolecular Distances (Å)a in
LT-PDTA 4
unremarkable, mirroring those seen in TBDTA and other DTA
radicals, and while there is a general shortening of intramolecular
bond lengths occasioned by cooling, there is very little structural
difference between the 293 and 95 K data. The intermolecular
distances, notably the intradimer S- - -S′ contacts (3.322(2) and
3.272(2) Å), are more interesting. These are longer than those
observed in the centrosymmetric dimers of TBDTA and BDTA,
resembling more nearly those seen in the cofacial dimers of
the LT phases of TDTA and TPDTA. On cooling from 293 to
95 K, these dimer S- - -S′ interactions are also subject to a much
larger contraction (to 3.257(2) and 3.212(2) Å) than the
intramolecular bonds. The lateral S- - -S′ interactions S1- - -
S3′ and S1- - -S1′ are also subject to a large contraction on
cooling.
293 K
95 K
293 K
95 K
S1-N2
1.645(2) 1.647(2)
1.648(2) 1.657(2)
1.647(2) 1.651(2)
1.650(2) 1.654(2)
S1-C1
S2-C2
S3-C6
S4-C5
1.737(2) 1.743(2)
1.743(2) 1.747(2)
1.736(2) 1.737(2)
1.741(2) 1.743(2)
3.272(2) 3.212(2)b
3.874(2) 3.812(2)c
3.343(2) 3.268(2)d
S2-N2
S3-N5
S4-N5
S1- - -S3′
S1- - -S3′
S1- - -S1′
3.322(2) 3.257(2)b S2- - -S4′
3.826(2) 3.768(2)c S2- - -S4′
3.407(2) 3.294(2)d S1- - -S3′
S2- - -N4′ 2.914(2) 2.841(2)d S4- - -N3′ 2.929(2) 2.875(2)d
a ESDs are in parentheses. b Intradimer contacts. c Interdimer contacts
along stack. d Lateral contacts between stacks.
Dimers of LT-PDTA form ribbonlike arrays of centrosym-
metrically locked pairs running along the z-direction. As the
unit cell drawing in Figure 8 illustrates, there are several close
lateral intermolecular S- - -S′ (S1- - -S1′ and S1- - -S3′) and S- - -
N′ (S2- - -N4′ and S4- - -N3′) contacts within and between these
pairs of dimers. Indeed, the efficiency with which neighboring
molecules couple across inversion centers is a critical feature
of this structure. Similar packing arrangements, with 6-center
S- - -N′ bridges, are observed in TDTA and TPDTA, and as
discussed below, the bistability found in all these compounds
can be related to the role these interactions play. This kind of
nesting is, of course, precluded in TBDTA by the presence of
the buffering CH units.
Figure 9 illustrates the slipped π-dimer stacks, viewed from
a direction approximately parallel to the ribbonlike molecular
arrays. The offset of the dimers along the stack, characterized
by the interdimer S- - -S′ interactions S1- - -S3′ and S2- - -S4′,
is clearly apparent, as are the lateral S- - -S′ and S- - -N′ contacts
noted above. The ramifications of these features on the magnetic
properties of the material are described below.
Figure 7. ORTEP drawing (50% probability ellipsoids) of the LT-PDTA
dimer, showing atom numbering.
introduced in the phase reversion. Either way, careful screening
eventually afforded a few untwinned crystals of the LT phase.
Single-crystal X-ray data on this phase, obtained both at 293
and 95 K, are summarized in Table 2, and listings of pertinent
intra- and intermolecular contacts are provided in Table 4.
Crystals of LT-PDTA belong to the triclinic space group P1h
and consist of centrosymmetric pairs of PDTA dimers; an
ORTEP drawing is shown in Figure 7. The intramolecular S-N
and S-C distances within both halves of the dimer are
9
8260 J. AM. CHEM. SOC. VOL. 126, NO. 26, 2004

Bistabilities in 1,3,2-Dithiazolyl Radicals
A R T I C L E S
Upon heating above room temperature no change in øP is
observed until near 340 K, whereupon a sudden increase takes
place, reaching a plateau around 350 K which, expressed in
terms of øPT, corresponds to about one-half the value expected
(0.375) if the radicals were dissociated completely into nonin-
teracting S ) 1/2 paramagnets. Cooling the sample leads to little
change in øP , but a slight and steady decrease in øPT until a
temperature of 300 K is reached, whereupon a sharp decrease
in øP and øPT takes place, returning both functions to near their
initial room-temperature values. The resulting well-defined
hysteresis loop for both øP and øPT is the magnetic signature of
the kind of bistability noted earlier for TPDTA and TDTA, i.e.,
the coexistence of a diamagnetic π-stacked dimer (LT-PDTA)
and a paramagnetic π-stacked radical (HT-PDTA). In the present
case the regime of bistability (TCV ) 297 K and TCv ) 343 K)
for the two phases spans higher temperatures than those
observed for TDTA10,11 (TCV ) 234 K and TCv ) 316 K) and
TPDTA12 (TCV ) 120 K and TCv ) 180 K).
Figure 10. Plots of øP and øPT as a function of T for TBDTA 8.
The failure of øPT to reach the expected value of 0.375 for
1
an S ) /2 system value, coupled with the steady decrease in
øPT upon cooling over the range 350-300 K, likely heralds
the presence of antiferromagnetic coupling both along and
between the radical π-stacks in the HT phase; similar features
were earlier noted for TDTA.10,11
Discussion
The existence of polymorphism in sulfur nitrogen radicals is
well documented,22,23 but bistability, involving the hysteretic
interconversion of two structurally and energetically similar
polymorphs, is rare. The earliest indication of the phenomenon
was noted over a decade ago in the biradical24 1,3-phenylene-
bis(dithiadiazolyl),25 a diamagnetic stacked π-dimer structure
which dissociated at elevated temperatures with a concomitant
increase in ø. Upon cooling the sample some paramagnetism
was retained, but the efficiency of the cycling process was poor.
The hysteretic bistability observed later in the two 1,3,2-DTA
radicals TPDTA12 and TDTA10,11 provided more dramatic and
elegant manifestations, particularly in the case of TDTA, of the
coexistence over a wide temperature range of two distinct but
related structures, one diamagnetic, the other paramagnetic. The
driving force for change from the HT phase to LT phase is a
Figure 11. Plots of øP and øPT as a function of T for PDTA 4.
Magnetic Measurements. Variable-temperature magnetic
susceptibility (ø) measurements have been performed on mi-
crocrystalline samples of both TBDTA 8 and PDTA 4. The
results, expressed in the form of plots of øP vs T and øPT vs T,
where øP ) ø + ø0 and ø0 is the calculated diamagnetic
correction based on Pascal’s constants, are presented in Figures
10 and 11.
Taking into account that 50% of the S ) 1/2 radical centers
in TBDTA exist in the form of presumably diamagnetic dimers,
the temperature dependence of the susceptibility øP over the
range 4-400 K is well described within a Curie-Weiss
framework, with C ) 0.158 and Θ ) -12 K. A plot of øPT vs
T also shows the expected features, i.e., a relatively invariant
regime from 50 to 350 K with a value near 0.15, a little less
than half that expected if all the spins uncoupled. There is a
slight increase in øPT above 350 K, perhaps indicative of some
spin unpairing of the dimers, and there is also a weak
antiferromagnetic tail below 50 K. Overall, however, the results
reveal a remarkably rigid system, with little if any structural or
magnetic variations over a wide temperature range. This, in
itself, represents a unique finding in 1,3,2-DTA structures.
The magnetic behavior of PDTA is more dramatic. At and
below room temperature the material is essentially diamagnetic,
as expected from the dimeric association observed in the 293
and 95 K crystal structures. We ascribe the residual paramagnet-
ism to the presence of lattice defects which, if expressed in terms
of the molar fraction of Curie spins, amount to less than 0.15%.
charge density wave (CDW) driven or Peierls instability,26
a
pervasive feature of many sulfur nitrogen radicals. In the case
of TPDTA the mechanism of the phase interconversion has been
(22) (a) Cordes, A. W.; Haddon, R. C.; Hicks, R. G.; Oakley, R. T.; Palstra, T.
T. M. Inorg. Chem. 1992, 31, 1802. (b) Bryan, C. D.; Cordes, A. W.;
Haddon, R. C.; Hicks, R. G.; Kennepohl, D. K.; MacKinnon, C. D.; Oakley,
R. T.; Palstra, T. T. M.; Perel, A. S.; Scott, S. R.; Schneemeyer, L. F.;
Waszczak, J. V. J. Am. Chem. Soc. 1994, 116, 1205.
(23) (a) Banister, A. J.; Bicklebank, N.; Clegg, W.; Elsegood, M. R. J.; Gregory,
C. I.; Lavender, I.; Rawson, J. M.; Tanner, B. K. J. Chem. Soc., Chem.
Commun. 1996, 679. (b) Banister, A. J.; Bicklebank, N.; Lavender, I.;
Rawson, J. M.; Tanner, B. K, Clegg, W.; Elsegood, M. R. J.; Palacio, F.
Angew. Chem., Int. Ed. Engl. 1996, 35, 2533. (c) Palacio, F.; Antorrena,
G.; Castro, M.; Burriel, R.; Rawson, J. M.; Smith, J. N. B.; Bricklebank,
N.; Novoa, J.; Ritter, C. Phys. ReV. Lett. 1997, 79, 2336. (d) Bond, A. D.;
Haynes, D. A.; Pask, C. M.; Rawson, J. M. J. Chem. Soc., Dalton Trans.
2002, 2522.
(24) Magnetic bistability has recently been observed in a nitroxide biradical.
See: Shultz, D. A.; Fico, R.; Boyle. P. D.; Kampf, J. W. J. Am. Chem.
Soc. 2001, 123, 10403.
(25) Andrews, M. P.; Cordes, A. W.; Douglass, D. C.; Fleming, R. M.; Glarum,
S. H.; Haddon, R. C.; Marsh, P.; Oakley, R. T.; Palstra, T. T. M.;
Schneemeyer, L. F.; Trucks, G. W.; Tycko, R. R.; Waszczak, J. V.; Warren,
W. W.; Young, K. M.; Zimmerman, N. M. J. Am. Chem. Soc. 1991, 113,
3559.
(26) Peierls, R. C. Quantum Theory of Solids; Oxford University Press: London,
1955; p 108.
9
J. AM. CHEM. SOC. VOL. 126, NO. 26, 2004 8261

A R T I C L E S
Brusso et al.
Figure 12. Crystal structures of HT-TDTA (above) and LT-TDTA (below).
Intermolecular (4-center and 6-center) S- - -N′ contacts are shown with
dashed lines.
described in terms of a tectonic plate slippage of layers within
an eight-layer supercell common to both lattices.12 Interestingly
in the TBDTA structure reported here one of the two sublattices
consists of a radical π-stack, but the presence of a dimer π-stack
in the other sublattice, and a consequent cell doubling, precludes
the possibility of a Peierls instability in the radical stack. This
material thus represents the first example of a DTA radical stack
in which dimerization is completely suppressed.
Figure 13. End-on views of subcells 1 and 2 of HT-TDTA (along z), and
LT-TDTA (along x). Curved arrows denote motion required for phase
interconversion.
To date there has been no attempt to probe the mechanistic
aspects of the bistability in TDTA,10,11and with the discovery
of bistability in PDTA the need to address these issues becomes
more acute. To this end we focus first on TDTA, where the
structures of both the HT and LT phases have been established
crystallographically. The crystal structures of both forms, viewed
down the stacking axes, are shown in Figure 12.27 While the
space groups (P21/c and P1h) are different, the overall appearance
of the structures are remarkably similar. Both present ribbonlike
arrays of TDTA running laterally across the cells, and linked
by a series of intermolecular S- - -N′ interactions which span
inversion centers to produce 4-center and 6-center bridges.28
There are, however, notable differences. For example, along
the stacking direction the unit cell dimension of the LT-phase
is doubled, heralding the presence of dimers.
Of greater importance, in terms of the structural intercon-
version pathway, is the fact that the unit cell length c ) 15.0630-
(13) Å in the HT form is twice that of the corresponding
dimension a ) 7.5310(5) Å in the LT form. As a result there
are two ribbonlike arrays per cell, denoted subcells 1 and 2 in
Figure 12; these ribbons are not coplanar. Whereas in the LT
form the molecular ribbons all run parallel to [0,0,1], i.e., the
xy plane, alternate ribbons along the z-direction in the HT phase
are rotated clockwise and anticlockwise about the z axis. As a
result the mean planes of the ribbons in subcell 1 are coplanar
with the [1,-1,0] plane, while those in subcell 2 are coplanar
with the [1,1,0] plane. These critical structural features are
illustrated in Figure 13, which shows the stacking of ribbons in
the two sub-cells of the HT form, and the ribbons of dimers
found in the LT form. On the basis of the above structural
description of LT- and HT-TDTA, it is clear that a simple
tectonic plate interconversion mechanism, of the type seen in
TPDTA,12 does not apply; more subtle structural changes are
required.
To effect the necessary changes, we propose a “domino”
mechanism to convert the molecular ribbons, tilted clockwise
and anticlockwise in the two subcells of the HT form, into the
horizontal ribbons found in the LT form. As a first step in this
process, segmentation of the ribbons is required. This can occur
at either of the two bridge bonds, i.e., the 4-center or 6-center
linkages illustrated in Figure 12. On the basis of the shortness
of the 6-center S- - -N′ bridges, we suggest that the 4-center
S- - -N′ links are weaker and the more likely to open. Accord-
ingly, if these bonds break, the ribbons are divided into pairs
of TDTA molecules laterally bridged by the remaining 6-center
S- - -N′ interactions. Once disconnected from their neighbors,
these supermolecules (Figure 14) are free to rotate, clockwise
or anticlockwise, about the center of symmetry on which they
hinge so as to produce, eventually, the horizontal coplanar
arrangement characteristic of the LT-modification. Lateral
4-center linkages can then reform, and dimerization takes place.
(27) The structural account on TDTA has been constructed using the crystal-
lographic data reported in ref 10.
(28) In the LT- and HT-TDTA structures (both at 298 K, ref 10) the 4-center
S- - -N bridges are 3.088(2) and 3.132(2) Å, respectively, while the 6-center
S- - -N bridges are 2.959(2) and 3.018(2) Å, respectively.
9
8262 J. AM. CHEM. SOC. VOL. 126, NO. 26, 2004

Bistabilities in 1,3,2-Dithiazolyl Radicals
A R T I C L E S
Thus, even without any direct structural information on the HT
phase of PDTA, it is reasonable to anticipate that the regime of
bistability found for PDTA should be less than that found for
TDTA, although considerably greater than that found in TPDTA,
where cooperative effects associated with tectonic plate motion
are less pronounced. Finally, in QDTA, the absence of
hysteresis/cooperativity can be directly attributed to the space
group P21, which precludes pairwise centrosymmetric intermo-
lecular linkages. While there are lateral S- - -N′ contacts in
QDTA, they are considerably longer (3.507(1) Å), and insuf-
ficient to generate any significant barrier, let alone apportion a
cooperative effect, to the association/dissociation process.
Conclusions and Summary
In the course of searching for new molecular radical materials
we have prepared and characterized the two 1,3,2-dithiazolyl
radicals, TBDTA and PDTA, comparing along the way their
spin distributions and electrochemical properties with those of
other condensed-ring 1,3,2-dithiazolyls. From a structural
perspective our intent was to design slipped π-stack structures
exhibiting magnetic bistability. Attempts, however, to design
molecular materials with specific crystal structures remains, at
best, a sporting past-time,29 and our efforts produced some
unexpected results. In the case of TBDTA, a system closely
related in terms of size, shape, and spin distribution to TDTA
and TPDTA, two radicals known to form bistable π-stacks, a
complex structure was found that consists of two very different
sublattices, one based on a radical π-stack and the other on a
π-dimer stack. Dimerization, let alone bistability, of the resulting
open-shell sublattice is not possible. By contrast, the known
radical PDTA, earlier described as a dimer π-stack, has been
found to exhibit a wide and sharply defined magnetic hysteresis
above room temperature.
To account for the dramatic hysteresis exhibited by both
TDTA and PDTA, we have developed a mechanism for their
phase interconversion, one which involves the cooperative
breaking and making of strong intermolecular S- - -N′ interac-
tions and a domino-like rotation of molecules within the
π-stacked structure. The model allows the development of at
least empirical guidelines for the pursuit of other bistable radical-
based materials. Whether these radical/dimer bistabilities will
find applications in switching devices30 remains to be seen.
Figure 14. Domino interconversion of HT-TDTA (above) and LT-TDTA
(below), with cleavage and reformation of 4-center S- - -N′ interactions
(dashed lines).
Chart 2
Minor tectonic plate slippage required to map exactly onto the
final slipped π-dimer stack can also occur.
The domino mechanism described above relies on the
presence and preservation of the inversion symmetry; the
molecular plates rotate about the 6-center bridges, and all bond-
making and -breaking processes occur at the 4-center bridges.
It leads to a halving of the unit cell in one dimension and a
doubling in another. It requires, as a rate-determining step in
both the forward (dimerization) and backward (dissociation)
directions, the rupture of the strong electrostatic S- - -N′ contacts
which characterize both phases. Most appealing, however, is
the way in which cooperativity emerges as natural consequence
of the domino mechanism. Accordingly the rate of phase
interconversion, once initiated, will accelerate as the new lattice
self-assembles and new bridge bonds are built.
Detailed analysis of the phase interconversion mechanism of
PDTA is precluded by the absence of structural information on
the high-temperature phase. Nonetheless, comparison of the
packing of the LT forms of PDTA and TDTA reveals many
common features, notably the presence of centrosymmetric pairs
(Chart 2) linked by 6-center S- - -N′ contacts (Figures 8 and
9), suggesting that similar structural issues are involved.
However, in contrast to TDTA, there is a “weak link” in the
ribbonlike arrays of PDTA molecules. This arises from the fact
that PDTA cannot form centrosymmetric 4-center S- - -N′
bridges at all, as there is no thiadiazole ring. There are weak
S- - -S′ interactions which serve to link the 6-center linked
supermolecules, but these are likely to be cleaved more easily.
Experimental Section
General Procedures and Starting Materials. The reagents o-
phenylenediamine 9, potassium thiocyanate, bromine, thionyl chloride,
pyridine, dichloropyrazine 14, thiourea, acetic anhydride, sodium sulfide
nonahydrate, nitrosonium hexafluoroantimonate, trimethylsilyl azide,
and triphenylantimony (all from Aldrich) and chlorine gas (Matheson)
were obtained commercially and used as received. Samples of radicals
2,3 3,12 5,12 6,and 712 (for EPR and CV work) were all prepared
according to literature procedures. All solvents were at least of reagent
grade; acetonitrile (MeCN) and dichloroethane (DCE) were dried by
distillation from P2O5. All reactions were performed under an atmo-
sphere of dry nitrogen. Melting points are uncorrected. Fractional
sublimations were performed in an ATS series 3210 three-zone tube
furnace, mounted horizontally, and linked to a series 1400 temperature
(29) Enkelmann, V. Angew. Chem., Int. Ed. Engl. 1991, 30, 1121.
(30) Miller, J. Angew. Chem., Int. Ed. 2003, 42, 27.
(31) (a) Kraft, G. Dissertation, Universita¨t Kaiserslautern, 1989. (b) Wolmer-
sha¨user, G.; Johann, R. Angew. Chem., Int. Ed. Engl. 1989, 28, 920.
(32) When crude 9 was used, yields were nearer 40-50%.
9
J. AM. CHEM. SOC. VOL. 126, NO. 26, 2004 8263

A R T I C L E S
Brusso et al.
control system. Infrared spectra (Nujol mulls, KBr optics) were recorded
on a Nicolet Avatar FTIR spectrometer (at 2 cm-1 resolution). 1H NMR
spectra were run on a Bruker Avance 300 MHz NMR spectrometer.
Elemental analyses were performed by MHW Laboratories, Phoenix,
AZ 85018.
filtered off, washed with 2 × 15 mL of DCE, and dried in vacuo; crude
yield 0.96 g (0.38 mmol, 91%). IR: 3054 (w), 1421 (w), 1216 (w),
1043 (m), 880 (s), 824 (w), 770 (m), 535 (m), 429 (m) cm-1. This
material was metathesized with NOSbF6 (1.24 g, 4.67 mmol mmol) in
25 mL of MeCN, to afford a dark-red solution. Filtration of the solution,
and flash distillation of the solvent from the filtrate afforded a dark-
red solid that was recrystallized from acetic acid as red/black crystals
of [8][SbF6] (0.98 g, 2.21 mmol, 52% based on 12), dec >205 °C. IR:
3112 (w), 1419 (w), 1288 (w), 1169 (w), 1050 (m), 965 (m), 892 (m),
864 (m), 850 (w), 773 (s), 664 (vs), 647(vs), 558 (m), 535 (w), 430
(m) cm-1. Anal. Calcd for C6H2N3S3F6Sb: C, 16.08; H, 0.45; N, 9.38%.
Found: C, 16.38; H, 0.53; N, 9.18%.
Preparation of 1,2-Diamino-benzene-4,5-bis(thiocyanate) 10. A
solution of bromine (29.8 g, 0.190 mol) in 100 mL of cold MeOH was
added slowly over 2-3 h to a solution of potassium thiocyanate (36.1
g, 0.370 mol) and o-phenylenediamine 9 (10.0 g, 0.0930 mol) in 500
mL of MeOH at -78 °C. The mixture was allowed to warm slowly to
room temperature and was poured into 1 L of water. The resulting
dark-red slurry was filtered and the filtrate made basic by the addition
of aqueous ammonia. The bright-yellow precipitate so formed was
filtered off, washed with water, and air-dried. The crude product 10
(9.80 g, 0.044 mol, 48%) was purified by passing a solution of it in
ethyl acetate through a bed of silica. Removal of the solvent from the
eluate and recrystallization from chlorobenzene afforded pale-yellow
feathers, dec >135 °C. IR: 3459 (w), 3423 (m), 3378 (m), 2163 (m),
1662 (m), 1615 (m), 1563 (s), 1496 (m), 1310 (m), 1230 (w), 861 (m),
Preparation of [1,2,5]Thiadiazolo[3,4-f][1,3,2]benzodithiazol-2-
yl TBDTA 8. Triphenylantimony (0.736 g, 2.06 mmol) was added to
a cold (-20 °C) slurry of crude [8][Cl] (0.937 g, 3.82 mmol) in 20
mL of degassed MeCN, and the mixture was allowed to warm slowly
to room temperature. After 30 min the red/black solid was filtered off,
washed with 2 × 10 mL of MeCN, and dried in vacuo. The crude
product 8 was purified by gradient tube sublimation over a range of
100-45 °C/10-3 Torr. Yields of sublimed material, dec >140 °C, were
generally low, never surpassing 20% (based on crude [8][Cl]). IR: 3077
(w), 1438 (w), 1413 (w), 1236 (m), 1042 (w), 912 (w), 870 (m), 863
(s), 856 (vs), 841 (s), 817 (vs), 700 (s), 695 (s), 641 (m), 526 (w), 501
(w) cm-1. Anal. Calcd for C6H2N3S3: C, 33.94; H, 0.95; N, 19.79%.
Found: C, 33.80; H, 0.75; N, 19.44%.
1
676 (w) cm-1. H NMR (δ, DMSO) 6.90 (s, 2H), 5.47 (s, br, 4H).
Anal. Calcd for C8H6N4S2: C, 43.22; H, 2.72; N, 25.20%. Found: C,
43.40; H, 2.68; N, 25.35%.
Preparation of 2,1,3-Benzenethiadiazole-5,6-bis(thiocyanate) 11.
Thionyl chloride (3.90 g, 32.7 mmol) was added to the slurry of
recrystallized 1,2-diaminobenzene-bis(thiocyanate) 10 (2.22 g, 10.0
mmol) in 75 mL of toluene. Three drops of pyridine were added, and
the mixture was set to reflux under N2 for 1 h. After cooling the resulting
dark-brown solution to room temperature, 5 mL of ethanol was added
to quench residual thionyl chloride. The mixture was filtered and the
solvent removed from the filtrate to leave a yellow solid that was
recrystallized from ethanol to give off-white flakes of 11 (1.94 g, 7.75
mmol, 78%),mp 115-117 °C. IR: 3063 (w), 2168 (m), 1357 (w), 1237
(m), 974 (m), 913 (w), 863 (m), 858 (s), 854 (s), 831 (s), 708 (w), 539
Preparation of S,S′-Pyrazine-2,3-diyl bis-thioacetate 16. A solu-
tion of crude 2,3-pyrazine-(dithiouronium chloride) 15 (1.70 g, 5.64
mmol) prepared by boiling 2,3-dichloropyrazine and thiourea in
ethanol,33 was boiled in 75 mL of water containing 2.0 g of NaOH.
The resulting yellow solution was cooled in an ice/water bath and
quenched with acetic anhydride to afford a yellow solid which was
filtered off, washed with water, and air-dried. Extraction of this solid
(1.40 g) with boiling heptane afforded, on cooling, 16 as pale-yellow
crystalline flakes, yield 0.75 g (3.3 mmol, 58%), mp 111-113 °C. IR:
1724 (s), 1346 (m), 1170 (w), 1104 (w), 1054 (m), 1004(w), 962 (m),
1
(s), 422 (s) cm-1. H NMR (δ, CDCl3) 8.58 (s, 2H). Anal. Calcd for
C8H2N4S3: C, 38.38; H, 0.81; N, 22.38%. Found: C, 38.21; H, 0.59;
N, 22.44%.
1
874 (w), 804 (w), 604 (s) cm-1. H NMR (δ, CDCl3) 8.63 (s, 2H),
Preparation of 2,1,3-Benzenethiadiazole-5,6-dithiol 12. Compound
11 (1.94 g, 7.75 mmol) was added as a solid to a solution of Na2S‚
9H2O (6.0 g, 25.0 mmol) in 100 mL of degassed water and the mixture
heated to ∼70 °C for 30 min to produce a clear, orange-red solution.
The mixture was cooled to room temperature, and 20 mL of 10% HCl
was added dropwise to afford a heavy, pale-yellow precipitate. The
precipitate was filtered off, washed with water, and air-dried. The
product 12 (1.51 g, 7.53 mmol, 97%) was purified by vacuum
sublimation at 90 °C/10-2 Torr and recrystallized from hexane as
canary-yellow needles, mp 135-136 °C. IR: 3057 (w), 2531 (w), 1436
(w), 1240 (m), 1093 (w), 998 (m), 857 (s), 710 (w), 529 (m), 489 (m)
cm-1. 1H NMR (δ, CDCl3) 8.03 (s, 2H), 3.98 (s, 2H). Anal. Calcd for
C8H4N2S3: C, 35.98; H, 2.01; N, 13.99%. Found: C, 35.95; H, 1.82;
N, 13.72%.
2.47 (s, 6H). Anal. Calcd for C8H8N2S2O2: C, 42.09; H, 3.53; N,
12.27%. Found: C, 41.89; H, 3.65; N, 12.52%.
Preparation of [1,3,2]Pyrazinodithiazolylium Chloride [4][Cl].
The bis-thioacetate 16 (1.14 g, 5.01 mmol) was dissolved in 10 mL of
DCE and chlorine gas passed over the solution for 5 min to afford a
clear-orange solution. The solvent was flash distilled to leave crude
bis(sulfenyl chloride) 17 as a yellow solid. This material was directly
redissolved in 10 mL of DCE, and to the stirred solution was added
dropwise a solution of trimethylsilyl azide (0.650 g, 5.66 mmol) in 2
mL of DCE, producing a heavy, orange precipitate. After 30 min the
product [4][Cl] was filtered off, washed with 2 × 15 mL of DCE, and
dried in vacuo; yield 0.690 g (3.61 mmol, 72% based on 16). IR: 3027
(w), 1510 (m), 1341 (w), 1081 (m), 1027 (m), 922 (w), 813 (s), 748
(s), 555(m), 490 (s), 451(w) cm-1
.
Preparation of 2,1,3-Benzenethiadiazole-bis(sulfenyl chloride) 13.
Chlorine gas was passed over a solution of 12 (0.850 g, 4.25 mmol) in
10 mL of DCE held in an ice/water bath. A gelatinous, yellow solid
(the disulfide) immediately precipitated and then redissolved after a
few minutes to form a canary-yellow solution. The solvent and residual
chlorine were flash distilled to leave 13 as a yellow solid. IR: 3088
(w), 1747 (w), 1653 (w), 1570 (m), 1487 (m), 1438 (m), 1417 (w),
1353 (w), 1233 (m), 1223 (w), 982 (m), 876 (s), 861 (s), 826 (s), 719
(m), 708 (w), 532 (s), 493 (s), 457 (s), 421 (s) cm-1. This material was
used in subsequent steps without further purification.
Preparation of [1,2,5]Thiadiazolo[3,4-f][1,3,2]benzodithiazolylium
Chloride [8][Cl] and Hexafluoroantimonate [8][SbF6]. The bis-
(sulfenyl chloride) 13 obtained as described above was dissolved in
15 mL of DCE. To this stirred solution was added dropwise a solution
of trimethylsilyl azide (0.488 g, 4.25 mmol) in 5 mL of DCE, producing
a heavy, maroon precipitate. After 30 min the product [8][Cl] was
Preparation of [1,3,2]Pyrazinothiazol-2-yl PDTA 4. Triphenylan-
timony (0.707 g, 2.00 mmol) was added to a cold (0 °C) slurry of
crude [4][Cl] (0.690 g, 3.61 mmol) in 7 mL of degassed MeCN and
the mixture allowed to warm slowly to room temperature. After 30
min the red/black solid was filtered off, washed with 2 × 10 mL of
MeCN, and dried in vacuo. The product 4 was purified by vacuum
sublimation at 10-2 Torr, yield 0.250 g (1.60 mmol, 44% based on
crude [4][Cl]), dec 115 °C. Crystals suitable for X-ray work were grown
by gradient tube sublimation over a range of 70-30 °C/10-2 Torr
(sealed tube). IR: 3054 (w), 1525 (m), 1420 (w), 1317 (m), 1219 (w),
1194 (m), 1148 (s), 1168 (m), 850 (m), 694 (s), 688 (s), 503 (w), 482
(m), 432 (s) cm-1. Anal. Calcd for C4H2N3S: C, 30.76; H, 1.29; N,
26.90%. Found: C, 31.00; H, 1.18; N, 27.14%.
(33) The procedure follows that described for the analogous quinoxaline
compound. See: Morrison, D. C.; Furst, A. J. Org. Chem. 1956, 21, 470.
9
8264 J. AM. CHEM. SOC. VOL. 126, NO. 26, 2004

Bistabilities in 1,3,2-Dithiazolyl Radicals
A R T I C L E S
EPR Spectra. The X-Band EPR spectra of all the radicals 2-8 were
recorded at 293 K, using a Bruker EMX-200 spectrometer, on samples
dissolved in degassed dichloromethane. Hyperfine coupling constants
were obtained by spectral simulation using Simfonia34 and WinSim.
Density functional theory calculations (run on PC workstations using
the B3LYP DFT method, as contained in the Gaussian 98W suite of
programs, and the 6-31G** basis set) were used to estimate the spin
densities and isotropic hyperfine coupling constants.35 Geometries were
optimized within the constraints of C2V symmetry. Full vibrational
frequency calculations on the final geometries confirmed that they were
stationary points.
No significant decay was noted. The reflection data were processed
using SAINT.37 The structures were solved by direct methods using
SHELXS-9038 and refined by least-squares methods on F2 using
SHELXL-9739 incorporated in the SHELXTL40 suite of programs. All
non-hydrogen atoms were refined anisotropically; hydrogen atoms were
located on difference maps and refined isotropically. Details of data
collection and refinement are presented in Table 3.
Magnetic Susceptibility Measurements. Magnetic susceptibilities
were measured over the temperature range 5-380 K on a George
Associates Faraday balance operating at 0.5 T.
Cyclic Voltammetry. Cyclic voltammetry was performed using a
PINE Bipotentiostat, model AFCClBP1, with scan rates of 50-100
mV s-1 on solutions (<10-3 M) of 4, 8, and [6][SbF6] in oxygen-free
MeCN (dried by distillation from CaH2) containing 0.1 M tetra-n-
butylammonium hexafluorophosphate. A sample of salt [5][SbF6] was
prepared by metathesis between NOSbF6 and [5][Cl] in MeCN, the
latter being prepared according to the literature method. Potentials were
scanned with respect to the quasi-reference electrode in a single
compartment cell fitted with Pt electrodes and referenced to the Fc/
Fc+ couple of ferrocene at 0.38 V vs SCE.36 The Epa-Epc separation
of the reversible couples were within 10% of that of the Fc/Fc+ couple.
X-ray Measurements. Samples for analysis were glued to a glass
fiber with epoxy. X-ray data were collected at 293 and 95 K with a
Bruker SMART APEX CCD-based diffractometer using ω-scans of
width 0.3° and 30 s duration at a crystal-to-detector distance of 4.908
cm. Intensity decay over the course of the data collections was evaluated
by recollecting the first 50 frames of data at the end of the experiment.
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada (NSERCC), the U.S.
Office of Basic Energy Sciences, Department of Energy (Grant
DE-FG02-97ER45668) and the Office of Naval Research
(Contract No. N00014-99-1-0392) for financial support. We also
thank the Kentucky Research Challenge Trust Fund for the
purchase of CCD X-ray equipment and the upgrade of our X-ray
facility, and NSERCC for a post-graduate scholarship to O.P.C.
Supporting Information Available: Details of X-ray crystal-
lographic data collection and structure refinement, tables of
atomic coordinates, bond distances and angles, anisotropic
thermal parameters, and hydrogen atom positions in CIF format.
This information is available free of charge via the Internet at
http://pubs.acs.org.
JA048618M
(34) WinEPR Simfonia. Bruker Instruments, Inc., Billerica, MA.
(35) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann,
R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin,
K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barons, V.; Cossi, M.; Cammi,
R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.;
Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foreman, J. B.; Cioslowski, J.; Ortiz, J.
V.; Stefanov, B. B.; Liu, G.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng,
C. Y.; Nanayakkara, A.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-
Gordon, M.; Repogle, E. S.; Pople, J. A. Gaussian 98, revision A.6;
Gaussian, Inc., Pittsburgh, PA, 1998.
(36) Boere´, R. T.; Moock, K. H.; Parvez, M. Z. Anorg. Allg. Chem. 1994, 620,
1589.
(37) SAINT, version 6.22. Bruker Advanced X-ray Solutions, Inc., Madison,
WI. 2001.
(38) Sheldrick, G. M. SHELXS-90. Acta Crystallogr., Sect. A 1990, 46, 467.
(39) Sheldrick, G. M. SHELXL-97. Program for the Refinement of Crystal
Structures, University of Gottingen, Gottingen, Germany, 1997.
(40) SHELXTL, version 6.12. Program Library for Structure Solution and
Molecular Graphics, Bruker Advanced X-ray Solutions, Inc., Madison, WI.
2001.
9
J. AM. CHEM. SOC. VOL. 126, NO. 26, 2004 8265