Unsymmetrical 1α3-1,2,4,6-thiatriazinyls with aryl and trifluoromethyl substituents: Synthesis, crystal structures, EPR spectroscopy, and voltammetry

Article abstract of DOI:10.1021/ic2003996

A general synthetic route to 3-trifluoromethyl-5-aryl-1α³- 1,2,4,6-thiatriazinyl radicals was developed. X-ray structures were obtained for all five neutral radicals and show that they exist in the solid state as cofacial dimers linked by S...S contacts. X-ray structures were also obtained for two of the precursor chlorothiatriazines along with several aryl N-imidoylamidines, p-methoxybenzamidine, and N-chlorosulfonyl- N,N'-benzamidine. Cyclic voltammetric studies were performed on the [R₂C ₂N₃S]. radicals in CH₃CN and CH ₂Cl₂ with [ⁿBu₄N][PF₆] as the supporting electrolyte under vacuum conditions in an all-glass electrochemical cell. The results provide quasi-reversible formal potentials for the [R₂C₂N₃S]-/0 process in the range of -0.61 to -0.47 V, irreversible peak potentials for the [R₂C ₂N₃S]0/+ process from 0.59 to 0.91 V at lower concentrations, and the appearance of a second, reversible oxidation process from 0.69 to 0.94 V at higher concentrations (versus the Fc0/+ couple; Fc = ferrocene). This behavior was indicative of monomer-dimer equilibrium in solution, as ascertained from digital models of the voltammograms. There is a small but measurable trend in both the oxidation and reduction potentials with varying remote aryl substituents. EPR spectra were obtained for all five neutral radicals in CH₂Cl₂ solutions, which confirm the concentration of the unpaired electron density on the heterocyclic core. Trends were also seen in the hyperfine splitting constants aN with varying remote aryl substituents. Calculations were performed for all three oxidation states of the [R₂C₂N₃S]-/./+ monomeric rings; the resulting theoretical redox energies correlate well with solution phase voltammetric data.

Full text of DOI:10.1021/ic2003996

ARTICLE
pubs.acs.org/IC
Unsymmetrical 1λ3-1,2,4,6-Thiatriazinyls with Aryl and
Trifluoromethyl Substituents: Synthesis, Crystal Structures,
EPR Spectroscopy, and Voltammetry
Renꢀe T. Boerꢀe,* Tracey L. Roemmele, and Xin Yu
Department of Chemistry and Biochemistry, University of Lethbridge, Lethbridge, AB Canada T1K 3M4
S Supporting Information
b
ABSTRACT: A general synthetic route to 3-trifluoromethyl-5-
aryl-1λ3-1,2,4,6-thiatriazinyl radicals was developed. X-ray
structures were obtained for all five neutral radicals and show
that they exist in the solid state as cofacial dimers linked by
S
S contacts. X-ray structures were also obtained for two of
3 3 3
the precursor chlorothiatriazines along with several aryl N-imi-
doylamidines, p-methoxybenzamidine, and N-chlorosulfonyl-
N,N0-benzamidine. Cyclic voltammetric studies were per-
formed on the [R2C2N3S]• radicals in CH3CN and CH2Cl2
with [nBu4N][PF6] as the supporting electrolyte under vacuum
conditions in an all-glass electrochemical cell. The results
provide quasi-reversible formal potentials for the [R2C2N3S]ꢀ/0 process in the range of ꢀ0.61 to ꢀ0.47 V, irreversible peak
potentials for the [R2C2N3S]0/þ process from 0.59 to 0.91 V at lower concentrations, and the appearance of a second, reversible
oxidation process from 0.69 to 0.94 V at higher concentrations (versus the Fc0/þ couple; Fc = ferrocene). This behavior was
indicative of monomerꢀdimer equilibrium in solution, as ascertained from digital models of the voltammograms. There is a small
but measurable trend in both the oxidation and reduction potentials with varying remote aryl substituents. EPR spectra were
obtained for all five neutral radicals in CH2Cl2 solutions, which confirm the concentration of the unpaired electron density on the
heterocyclic core. Trends were also seen in the hyperfine splitting constants aN with varying remote aryl substituents. Calculations
were performed for all three oxidation states of the [R2C2N3S]ꢀ/•/þ monomeric rings; the resulting theoretical redox energies
correlate well with solution phase voltammetric data.
’ INTRODUCTION
’ RESULTS AND DISCUSSION
There is strong continuing interest in the synthesis and
structural chemistry of sulfurꢀnitrogenꢀcarbon heterocyclic
free radicals, with an emphasis on applications to the design of
molecular conductors and molecular magnetism.1 This interest
extends to transition-metal coordination complexes that incor-
porate such radicals as ligands—including spin-active ligands.2
We have contributed to the electrochemical characterization of the
redox properties of this whole class of compounds and, in
particular, have characterized by solution electrochemistry the
influence of aryl ring substituents on 5-aryl-1,2,3,4-dithiadiazolyls.3
In order to extend our approach to a new class of C,N,S
radicals, we have developed a general synthetic route to the
asymmetrically substituted 3-trifluoromethyl-5-aryl-1λ3-1,2,4,6-
thiatriazinyls (Chart 1). We recently communicated the utility of
such thiatriazinyl radicals as novel π-donor ligands to the
organometallic moiety CpCr(CO)2;4 here, we elaborate the
chemistry of these heterocyclic free radicals, including structural
characterization of novel intermediates and several final products
in the solid state by X-ray crystallography, solution EPR spec-
troscopy, and a detailed electrochemical investigation inter-
preted in light of DFT calculations.
A primary goal of this work was the development of a more
general route to 1,2,4,6-thiatriazinyl radicals which would allow
for modification of the exocyclic substituents (Scheme 1). Var-
ious methods have been attempted to prepare both symmetric
and asymmetrically substituted thiatriazines.5 Amidines have
been reported to react with S3N3Cl3 to form 1-chloro-1,2,4,6-
thiatriazines.5e,f In this work, we first converted amidines into
N-imidoylamidines and found that these, when passivated as
hydrochlorides, afford 1-chloro-1,2,4,6-thiatriazines by direct
reaction with sulfur dichloride in high yield.5b,6 An established
literature procedure was used to prepare the para-substituted
benzamidine hydrochlorides,7 but the free bases 1aꢀe have not
previously been reported. Full characterization of these useful
intermediates is provided in the Experimental Section, and the
crystal structure of 1a is presented in Figure S1 (Supporting
Information).
Aryl N-Imidoylamidines. In contrast to amidine chemistry,
aryl N-imidoylamidines have received less attention, although
Received: February 25, 2011
Published: May 03, 2011
r
2011 American Chemical Society
5123
dx.doi.org/10.1021/ic2003996 Inorg. Chem. 2011, 50, 5123–5136
|

Inorganic Chemistry
ARTICLE
Chart 1. The 3-Trifluoromethyl-5-aryl-1λ3-1,2,4,6-
thiatriazinyls
Scheme 1. Synthetic Route to the 3-Trifluoromethyl-5-aryl-
1,2,4,6-thiatriazines
Figure 1. Thermal ellipsoid (30%) plot showing the two independent
hydrogen-bonded molecules of 2c as found in the crystal at ꢀ100(2) °C
along with an additional H4B0 atom to indicate how the chain propagates
in the lattice. The H1BꢀN5 and H4B0ꢀN2 bonds can be viewed as
incipient tautomerism leading to the diimine isomer. Only the principal
components of the disordered CF3 groups are shown for clarity (Figures
S3 and S4, Supporting Information, are plots of 2a and 2e.).
nitrogen atom N5, which, if exchange occurred, would inter-
convert it to the diimine. In the 13C NMR spectra, seven distinct
carbon peaks (Table S2) were observed, which are attributable to
the seven different types of carbon atoms in 2aꢀe.
X-ray structures of imidoylamidines are rare. In fact, no
structures of unsubstituted imidoylamidines have been reported
in the Cambridge Crystal Structure Database (CSD, version 5.31,
updated Nov. 2009), and only one structure of an imidoylami-
dinium salt is known (with [Se2Cl10]2ꢀ as the counterion).10
However, crystal structures of imidoylamidinate anions coordi-
nated to various metals are known,11 and the structure of the
related parent biguanide has been reported.12 Colorless crystal-
line plates of 2a, 2c, and 2e were grown by sublimation in a three-
zone tube furnace under a dynamic vacuum, and their structures
were determined at low temperatures by X-ray diffraction. All
three crystal structures contain two crystallographically indepen-
dent molecules which display very similar inter- and intramole-
cular hydrogen bonding,13 as shown by dashed lines in Figure 1
for compound 2c as a representative example. The atom
numbering scheme is the same for all three structures to facilitate
comparison; the H-bonding data for 2a, 2c, and 2e are discussed
in the Supporting Information (Table S3).
The average bond lengths and angles determined from all six
crystallographically independent molecules of 2a, 2c, and 2e are
shown in Figure 2. This clearly shows that C8ꢀN3 is short,
characteristic of an imine, while C7ꢀN1 and C7ꢀN2 are
identical within esd and of intermediate length, characteristic
of a highly delocalized amidine. Aryl/CF3 substitution therefore
leads to a rather unsymmetrical imidoyl amidine geometry,
approximating an imino-substituted amidine.
several preparative routes have been reported.8 In this work, it
was found that direct reaction of trifluoroacetonitrile with the
free-base amidines gave the desired imidoylamidines in high
yields in accordance with the report by Schaeffer.9 The colorless
sublimed products were found to be pure by 1H and 13C NMR
spectroscopy, MS, and elemental analysis. The expected signals
1
for hydrogen atoms attached to carbon were observed in H
NMR spectra recorded in CDCl3 (Table S1, Supporting In-
formation). Three separate NH peaks are also observed (Figure
S2, Supporting Information), indicating nonequivalency of the
hydrogen atom environments. The signals resonating around 11
and 6.7 ppm are noticeably broadened compared to the one near
9 ppm, indicating a higher rate of exchange between the former.
Conversion to N-Imidoylamidine Hydrochlorides. Direct
reaction between 2c and sulfur dichloride to form 4c resulted in
low yields of the product as an intractable oil that proved hard to
purify. However, pacifying the reactive nitrogen base with HCl
increased the yield of 4c significantly and afforded a low-melting
solid. Therefore, all of the imidoylamidines were converted with
HCl(g) in dry diethyl ether to the corresponding hydrochloride
salts 3aꢀe, which precipitated as white, insoluble solids that were
characterized by infrared spectroscopy. Each adduct has a unique
fingerprint region but similar broad bands in the 3300 cmꢀ1 range
characteristic for NH stretches involved in hydrogen bonding.
These results are consistent with either of the above tauto-
mers, i.e., either with hydrogen attached to three different N
atoms as on the left or, as on the right, with one imino and two
amino hydrogen atoms in which the latter are nonequivalent due
to strong intramolecular hydrogen bonding. The solid-state
structure (Figure 1) shows the dominance of the iminoꢀamino
tautomer but also shows H-bonding of H1B to the backbone
5124
dx.doi.org/10.1021/ic2003996 |Inorg. Chem. 2011, 50, 5123–5136

Inorganic Chemistry
ARTICLE
Figure 2. Average bond lengths (Å, top) and angles (deg, bottom) from
crystal structures of the six independent imidoylamidine molecules
found in 2a, 2c, and 2e. Errors are standard deviations.
Figure 4. Average bond lengths (Å, top) and angles (deg, bottom) with
standard deviations from crystal structures of the 1-chlorothiatriazines
4b and 4e.
and 4e from acetonitrile solutions at ꢀ35 °C, and the crystal
structures were determined at low temperatures. These struc-
tures confirm the formation of the six-membered ring with an
almost perpendicular SꢀCl bond, as shown in Figure 3. The
C2N3 atoms are coplanar (maximum deviation 0.047(1) Å), with
the sulfur atoms tipped slightly out of the plane (0.288(2) Å, 4b;
0.323(2) Å, 4e). The interatomic distances in 4b and 4e are quite
comparable. Averaged bond lengths and angles from the two
structures are presented in Figure 4. The phenyl rings in both
structures have average CꢀC bond lengths of 1.389(10) Å and
are essentially coplanar with the heterocyclic core. The packing
of 4e shows regular stacks aligned with the crystallographic c axis
without any significant short contacts, but 4b has typical S-
(δþ)ꢀN(δꢀ) short contacts between pairs of rings (Figure S5,
Supporting Information).1d
Figure 3. Thermal ellipsoid (30%) plots with atom numbering schemes
showing the molecular structures of (a) 4b and (b) 4e as they are found
within the crystal lattices at ꢀ100(2) °C. Only the principal components
of the disordered CF3 groups are shown.
1-Chloro-3-trifluoromethyl-1,2,4,6-thiatriazines. An excess
of SCl2 was used to ensure the completion of the reaction of the
imidoylamidine hydrochloride salts 3aꢀe to the 1-chloro-thia-
triazines 4aꢀe. Crude yields were typically high (>80%), and
these materials were used without further purification in the
synthesis of the radicals. 1H NMR spectra (Table S4, Supporting
Information) confirmed ring formation through the absence of
NH signals, and only the phenyl and para-substituted hydrogen
atoms appear in the spectra between 7.0 and 8.6 ppm. The signals
of H1, meta to the thiatriazine core, are weakly affected by
heterocycle formation, shifting only ∼0.10 ppm downfield com-
pared to the imidoylamidines. However, the signals of H2, ortho
to the heterocyclic core, are shifted downfield by ∼0.60 ppm.
This is diagnostic for the formation of a thiazyl ring; for example,
the chemical shifts of the ortho hydrogen atoms in comparably
substituted 1,5-dithia-2,4,6,8-tetrazocine heterocycles are very
similar to those of 4aꢀe.14 13C NMR was not obtained due to
instability of the compounds to hydrolysis. However, extremely
moisture-sensitive X-ray-quality crystals could be grown for 4b
The sensitivity of 4 to moisture is highlighted by the hydrolysis
of 4c from adventitious H2O in CH3CN to give 6 as colorless
blocks suitable for X-ray analysis (Figures S6ꢀS8, Supporting
Information). The hydrolysis product 6 surprisingly retains the
SꢀCl bond (unlike {PhCN}2{NH}SdO, 7),15 but a CF3CHd
NH moiety is eliminated while the sulfur is converted to
oxidation state VI.
5-Aryl-3-trifluoromethyl-1,2,4,6-thiatriazinyls. Several re-
ducing agents have been used previously to effect the reductive
elimination of the chloride ion from 1-chloro-1,2,4,6-thiatria-
zines. In early work, sodium verdazyl was often used.5 Here, we
employed triphenylantimony because of its general utility in the
5125
dx.doi.org/10.1021/ic2003996 |Inorg. Chem. 2011, 50, 5123–5136

Inorganic Chemistry
ARTICLE
Figure 5. Thermal ellipsoid (30%) plot and atom numbering scheme
showing the solid-state face-to-face and head-to-head dimerization of 5c,
which comprises the asymmetric unit in the lattice at ꢀ100(2) °C. The
dashed line indicates the close contact between sulfur atoms, for which
the S10 S20 distance is 2.625(1) Å. The structure of 5a (Figure S9,
3 3 3
Supporting Information) is very similar, with an S10 S20 distance of
3 3 3
2.6370(3) Å.
reduction of thiazyl halides.5e,f A half-molar ratio of solid
triphenylantimony was added by a solids-addition bulb to
acetonitrile solutions of 4aꢀe after careful removal of oxygen
via freezeꢀthawꢀdegassing. Reduction to the radical was im-
mediate in all cases, as the solution turned from a clear dark red to
a dark purple (almost black) solution with the precipitation of
some solid. Precipitated crude products were purified by sub-
limation in a three zone tube furnace and produced X-ray-quality
plates of the radicals 5aꢀe. Diffraction data were obtained
at ꢀ100 ( 2 °C and the structures solved and refined at this
temperature. However, in the case of 5b, persistent high R factors
were encountered, and hence the structure was redetermined at
RT with better results. Formation of a superlattice is suspected at
the lower temperature. In each case, the solid-state structures
display the same face-to-face and head-to-head (presumably
diamagnetic) dimers obtained previously for bis(3,5-diphenyl-
1,2,4,6-thiatriazinyl) (8),5e despite the presence of the bulky CF3
groups. For both 5a and 5c, one kind of crystallographically
independent dimer is observed in their crystal lattices (Figure 5).
For 5b and 5d, two sets of wedge-shaped dimers are found
stacked back-to-front in the lattice (Figure 6).
Figure 6. Thermal ellipsoid (30%) plot with atom numbering scheme
showing the molecular structure of 5b found within the crystal lattice at
23(2) °C. Four crystallographically independent molecules form two
distinct dimers. The S10 S20 distance is 2.684(1) Å, and the
3 3 3
S30 S40 distance is 2.6515(8) Å. A very similar arrangement is found
3 3 3
in the latticefor 5d (Figure S10, Supporting Information) with S10 S20
3 3 3
and S30 S40 distances of 2.659(1) Å and 2.635(1) Å, respectively.
3 3 3
The shortest contact between C2N3S rings is always through
the sulfur atoms, with an average distance of 2.643(21) Å. This is
longer than a normal disulfide linkage but within range of other
known thiatriazinyls such as 8,5e whose shortest contact is
2.666(3) Å, and 3,5-bis(dimethylamino)-1,2,4,6-thiatriazinyl
(9), whose shortest contact is again through the sulfur atoms
at 2.5412(8) Å.16 Average bond lengths and angles were calcu-
lated for the entire series 5aꢀe, and the results are presented in
Figure 7. The average SꢀN bond length over both bonds
(SX0ꢀNX0 and SX0ꢀNX2, X = 1ꢀ4) is 1.633(10) Å, which
is shorter than the SꢀN single bond in S4N4H4 (1.665 Å) and
longer than the N- ꢀ -S bond in S4N4 (1.616 Å). The average
CꢀN bond length in the ring is 1.332(16) Å, longer than an
average CdN double bond (∼1.270(15) Å)17 and shorter than
Figure 7. Average bond lengths (Å, top) and angles (deg, bottom) with
standard deviations in 1,2,4,6-thiatriazinyls determined from five crystal
structures. A common numbering scheme was used among crystal-
lographically independent monomers (X = 1ꢀ4). The average intradi-
mer short S S contact distance is 2.643(21) Å.
3 3 3
an average CꢀN single bond (1.472(5) Å). A typical heterocyclic
C- ꢀ -S bond length is approximately 1.352(5) Å.17 The asymmetric
substitution pattern seen in 5aꢀe allows for a comparison of the
bond lengths and angles with the symmetric diphenyl and
dimethylamino analogues 8 and 9. The CꢀN bond lengths are
very similar, as are the SꢀN bond lengths with only the SꢀN
bond closest to the CF3 group averaging slightly longer than in 8.
5126
dx.doi.org/10.1021/ic2003996 |Inorg. Chem. 2011, 50, 5123–5136

Inorganic Chemistry
ARTICLE
Figure 8. Thermal ellipsoid (30%) plot with atom numbering scheme
showing the molecular structure of 5e found within the crystal lattice at
ꢀ100(2) °C. Four crystallographically independent molecules form into
two distinct dimers with intradimer S10 S20 and S30 S40 dis-
3 3 3
3 3 3
tances of 2.6237(9) Å and 2.6255(8) Å, respectively. These dimers
associate into a single tetrameric set through S(δþ)ꢀN(δꢀ) interac-
tions (shown in red). The S20 N30 and S30 N20 distances are
Figure 9. (a) Experimental and (b) simulated EPR spectra of 5d in
CH2Cl2 at 18 °C, modulation amplitude 0.01 mT, modulation fre-
quency 100 kHz. Simulations were performed with WinSim (version
0.98, 2002)19 software using a 100% Lorentzian line shape.
3 3 3
3 3 3
3.169(2) Å and 3.199(2) Å, while the S40 N20 and S10 N30
3 3 3
3 3 3
distances are 3.082(2) Å and 3.034(2) Å, respectively. The two dimers
are out of register with each other.
reported, but, although eight radicals are thereby associated,
the interactions do not extend throughout the lattice.
The bond angles within the heterocyclic core are significantly
less symmetrical with asymmetric substitution on the ring.
Thus, the internal angles at the carbon of attachment of the
CF3 group are noticeably larger than in 8.5e Correspondingly,
the average internal angles at sulfur are noticeably smaller than
those in 8.
In all five structures, the many independent C2N3S rings are
almost planar (deviations in the range of 0.003ꢀ0.054 Å). The
mean planes of each set of dimers are tipped toward each other
EPR Spectroscopy. The 1,2,4,6-thiatriazinyls are members of
a larger class of unsaturated C,N,S compounds for which stable
neutral free radicals can be generated.1a In this work, the
trifluoromethyl thiatriazinyls 5aꢀe were generated by dissolving
high purity sublimed crystals in dichloromethane inside vacuum
sealed EPR tubes. In all cases, well-defined EPR spectra with high
signal-to-noise ratios were obtained, displaying complex splitting
patterns due to coupling to the three nonequivalent nitrogen
nuclei in the thiatriazinyl ring along with three equivalent fluorine
atoms of the directly attached CF3 groups, which cause each
subpeak to split into small quartets. Excellent agreement was
obtained between the experimental and simulated EPR spectra
for all five compounds (Figure 9 shows a typical example; see also
Figures S17ꢀS20, Supporting Information). However, the as-
signment of the hyperfine splitting (hfs) to the three nitrogen
atoms required input from quantum calculations. The electronic
structures of 5aꢀe were determined from UB3LYP/6-31G(d)
hybrid-DFT calculations (using Gaussian 98)18 for the mono-
meric radicals in the gas phase. Full geometry optimization was
undertaken for each structure, and frequency calculations con-
firmed the geometries to be minima. The experimental and
calculated hfs constants are compiled in Table 1.
The unpaired electron occupies a π-SOMO which is deloca-
lized over the heterocyclic core, a representation of which is
shown in Figure 10 using 5b as a typical example. The largest
coefficient is on sulfur, but there is considerable unpaired electron
density on each of the three nitrogen atoms and to a lesser extent
on the carbon atoms in the ring. There are small coefficients on the
fluorine atoms of the directly bound CF3 group, as well as on the
phenyl carbon atoms. There is, however, no experimental evidence
for hyperfine coupling to the aryl ring hydrogen atoms, and hfs
from 19F is likely to result mostly from spin polarization.
such that the shortest contacts are always S S, and the planes
3 3 3
intersect with angles between normals in the range of 5ꢀ17°.
Three distinct packing modes are found in the solid state among
these five thiatriazinyl structures. In 5a and 5c, there is a simple
2 þ 2 mode, wherein two dimeric units (Figure 5) are centro-
symmetrically associated through sideways S(δþ) N(δꢀ)
3 3 3
short contacts. However, the pairs of dimers are essentially
coplanar in the case of 5a (Figure S11, Supporting Information),
while in 5c they are almost exactly out of register (Figure S12,
Supporting Information). These tetrameric units are isolated
from other such units in the crystal lattice. Sideways S
N
3 3 3
contacts range from 2.941(1) to 3.341(1) Å. In 5b and 5d, there
are four independent molecules that form two sets of typical
dimers. However, the two sets of dimers are each associated with
each other through S(δþ) N(δꢀ) short contacts (Figures
3 3 3
S13 and S14, Supporting Information). Interestingly, in each
structure, one such tetrameric unit is closer to the coplanarity
that is observed in 5a, while a second is out of register as in 5c.
The variation in sideways S N contacts observed in these two
3 3 3
structures is larger, from 3.025(2) to 3.551(2) Å. The final
example is 5e (Figure 8), for which four independent thiatriazi-
nyls form one tetramer in the asymmetric unit. The sideways
S(δþ) N(δꢀ) interactions for this structure range from
3 3 3
3.034(2) to 3.199(2) Å, and the association is out of register as
in 5c. This structure is particularly interesting because there is
a further interaction between neighboring rings, leading to a
partial overlap of two adjacent tetramers (Figures S15 and
S16, Supporting Information). The shortest atomic contacts
In each case, there are three distinct aN values, except for 5a,
where two of the values are nearly identical. (Such accidental
degeneracy of symmetry-nonequivalent nitrogen atoms has been
reported for 3,5-bis(4-methoxyphenyl)-1,2,4,6-thiatriazinyl, for
which all three nitrogen atoms appear to have identical hfs of
0.3927 mT.)20 The trends in the hfs with the remote para R
are S40
C410 at 3.447(2) Å. This structure contains the
3 3 3
most extensive set of interactions between thiatriazinyls ever
5127
dx.doi.org/10.1021/ic2003996 |Inorg. Chem. 2011, 50, 5123–5136

Inorganic Chemistry
ARTICLE
Table 1. Experimental and Calculateda EPR Spectroscopic Data for 5aꢀe
aN2
aN4
aN6
aC3,5(avg)
aS1
aF(avg)
exptc (mT) calc (mT)
cmpd expt (mT) calc (mT) expt (mT) calc (mT) expt (mT) calc (mT) expt (mT) calc (mT)
calcb (mT)
5a
5b
5c
5d
5e
0.310
0.320
0.325
0.324
0.332
0.297
0.309
0.313
0.313
0.319
0.435
0.444
0.445
0.446
0.454
0.409
0.418
0.422
0.422
0.428
0.436
0.429
0.425
0.419
0.404
0.460
0.443
0.437
0.433
0.423
0.705
0.545
0.675
0.572
0.666
ꢀ0.518
ꢀ0.531
ꢀ0.536
ꢀ0.534
ꢀ0.542
0.555
0.555
0.555
0.558
0.558
0.032
0.036
0.039
0.040
0.044
ꢀ0.063
ꢀ0.068
ꢀ0.069
ꢀ0.068
ꢀ0.070
a From UB3LYP calculations, multiplied by a scaling factor of 0.81. b Sulfur hfs was not observable due to the very small natural abundance of this isotope.
c hfs is to the three fluorine atoms in the directly bound CF3 group.
Figure 10. KohnꢀSham isosurface of the π-SOMO of 5b from a
UB3LYP/6-31G(d) calculation.
Figure 11. Overlapping CVs of the redox couples of 5e in CH3CN
solution (0.1 M [nBu4N][PF6]) at (a) dilute concentration (blue), (b)
moderate concentration (red), and (c) high concentration (black).
substituents on the phenyl ring are quite convincing. As controls,
we can cite the powerfully π-electron donating NMe2 groups
directly bound in 3,5-bis(dimethylamino)-1,2,4,6-thiatriazinyl
are highly susceptible to decomposition from oxygen or hydro-
(9),20 in which aN2 = aN6 is twice as large as aN4. Conversely,
lysis by adventitious moisture. An all-glass electrochemical cell
with two powerfully electron withdrawing CF3 and CCl3 groups
sealed under vacuum conditions which has been described
directly attached in 3,5-bis(trifluoromethyl)-1,2,4,6-thiatria-
previously was used in this work to avoid interference.21 Cyclic
voltammetric studies were performed in both CH2Cl2 and
zinyl24 and 3,5-bis(trichloromethyl)-1,2,4,6-thiatriazinyl,5d,20b
aN4 is found to be larger by about 40% than aN2 and aN6. The
CH3CN solutions at temperatures of 20 ( 2 °C and scan rates
of ν = 0.1ꢀ10 V sꢀ1. The voltammetric results at full concentra-
tion are summarized in Table 2.
hyperfine splitting constants of 5aꢀe are intermediate between
these extremes, as expected. Nevertheless, as the remote para
substituents become more electron donating (from e to a), the
The 1,2,4,6-thiatriazinyls are very soluble in dichloromethane
and moderately soluble in acetonitrile. In all cases, the electro-
chemical response changed with analyte concentration, with the
appearance of a new redox couple at more positive potentials as
the solution became more concentrated (Figure 11). This change
correlated with the color of the solution (from pale yellow to
deep brown) as more analyte was added. From the current
response of the CV data, the active concentrations from voltam-
metry taken when the solutions were pale yellow are estimated to
be 10ꢀ20 times lower than the nominal concentration values
listed in Table 2. At very low concentrations (evidenced by a pale
yellow color in solution), the peaks associated with a reduction of
neutral material (Epc1/a1) have a large return wave over a range of
scan rates, while that for oxidation (Epa2) shows no return wave at
a
N6 value increases due to greater spin density on the NdSdN
region of the ring. The spin density on the remaining nitrogen
atoms decreases correspondingly, such that the sums of aN values
are found to be 1.190(5) for all five exemplars (standard
deviation). Moreover, the variation in aN values smoothly follows
Hammet σ(p) coefficients (Figure S21, Supporting Infor-
mation). Thus, EPR spectroscopy proves to be a very sensitive
probe of the unpaired spin density on thiatriazinyl rings. The
directly bound CF3 group causes further splitting of the EPR
signals into quartets, implying rotational averaging of the three
fluorine nuclei, but the aF values are small. Similar aF values have
been reported for 3,5-bis(trifluoromethyl)-1,2,4,6-thiatriazinyl.20
Voltammetry. The study of redox active heterocycles requires
handling of low concentrations of free radicals in solution which
scan rates up to 10 V sꢀ1
.
5128
dx.doi.org/10.1021/ic2003996 |Inorg. Chem. 2011, 50, 5123–5136

Inorganic Chemistry
ARTICLE
Table 2. Voltammetry Data for Compounds 5aꢀe in CH3CN and CH2Cl2 at Full Concentrationa
cmpd
conc. (mM)
Epc1 (V)
Epa1 (V)
Em (V)b
Epa2 (V)c
Epc3 (V)
Epa3 (V)
En (V)d
Ecell (V)e
CH3CN solns with 0.1 M [nBu4N][PF6] supporting electrolyte
5af
5b
5cg
5dh
5ei
7.3
6.3
ꢀ0.64
ꢀ0.60
ꢀ0.60
ꢀ0.55
ꢀ0.51
ꢀ0.51
ꢀ0.51
ꢀ0.48
ꢀ0.45
ꢀ0.42
ꢀ0.58
ꢀ0.56
ꢀ0.54
ꢀ0.50
ꢀ0.47
0.59
0.67
0.73
0.74
0.84
0.62
0.71
0.74
0.78
0.85
0.76
0.81
0.89
0.90
0.96
0.69
0.76
0.82
0.84
0.91
1.27
1.32
1.36
1.34
1.38
10.0
11.0
6.7
CH2Cl2 solns with 0.4 M [nBu4N][PF6] supporting electrolyte
5aj
5b
5ck
5dl
5em
10.0
8.7
ꢀ0.66
ꢀ0.66
ꢀ0.64
ꢀ0.63
ꢀ0.55
ꢀ0.55
ꢀ0.54
ꢀ0.52
ꢀ0.47
ꢀ0.44
ꢀ0.61
ꢀ0.60
ꢀ0.58
ꢀ0.55
ꢀ0.50
0.66
0.71
0.76
0.81
0.91
0.66
0.74
0.78
0.79
0.87
0.78
0.86
0.92
0.98
1.01
0.72
0.80
0.85
0.89
0.94
1.33
1.40
1.43
1.44
1.44
11.0
9.0
6.7
a Reported vs Fc0/þ on a Pt working electrode, ν = 0.2 V sꢀ1, T = 20 ( 2 °C. b Em = [Ep þ Epc1]/2 ≈ E0/. c Irreversible wave observed. d En =
a1
a3
e
f
[Ep þ Epc3]/2 ≈ E0/. Ecell = En ꢀ Em. IRR oxidation wave at 1.75 V. g IRR reduction wave at ꢀ1.87 V. h IRR reduction wave at ꢀ1.80 V.
i IRR reduction wave at ꢀ2.07 V. j IRR oxidation wave at 1.84 V. k IRR reduction wave at ꢀ1.78 V. l IRR reduction wave at ꢀ1.76 V. m IRR reduction
wave at ꢀ1.65 V.
0.69 to 0.91 V vs Fc0/þ in CH3CN and 0.72 to 0.94 V vs Fc0/þ in
CH2Cl2 and has an overall change of 0.22 V. The average Ecell
Table 3. Voltammetry Data for Compounds 5aꢀe in CH3CN
and CH2Cl2 at Low Concentrationa
values (Ecell = En ꢀ Em) are 1.33 ( 0.04 V in CH3CN and 1.41 (
cmpd
Epc1 (V)
Epa1 (V)
Em (V)b
Epa2 (V)c
Ecell (V)d
0.05 V in CH2Cl2, with a small but notable increase (0.11 V in
both solvents) as the substituent R moves from electron donating
to electron withdrawing. Values for 5d are slightly lower than
expected in both solvents, although the individual Em and En
values for 5d still follow the expected trend.
CH3CN solns with 0.1 M [nBu4N][PF6] supporting electrolyte
5a
5b
5c
5d
5e
ꢀ0.60
ꢀ0.59
ꢀ0.57
ꢀ0.53
ꢀ0.51
ꢀ0.54
ꢀ0.52
ꢀ0.50
ꢀ0.46
ꢀ0.41
ꢀ0.57
ꢀ0.56
ꢀ0.54
ꢀ0.50
ꢀ0.46
0.59
0.67
0.73
0.74
0.84
1.13
1.19
1.23
1.20
1.25
At lower concentrations (Table 3), the potential range for the
“reduction” process is from ꢀ0.57 to ꢀ0.46 V vs Fc0/þ in
CH3CN (change of 0.11 V) and ꢀ0.51 to ꢀ0.42 V vs Fc0/þ in
CH2Cl2 (change of 0.09 V). The potential range for the
irreversible “oxidation” process is from 0.59 to 0.84 V vs Fc0/þ
in CH3CN and 0.66 to 0.91 V vs Fc0/þ in CH2Cl2 for an overall
change of 0.25 V in both solvents. Average Ecell values (where
CH2Cl2 solns with 0.4 M [nBu4N][PF6] supporting electrolyte
5a
5b
5c
5d
5e
ꢀ0.54
ꢀ0.53
ꢀ0.50
ꢀ0.49
ꢀ0.45
ꢀ0.47
ꢀ0.47
ꢀ0.45
ꢀ0.42
ꢀ0.38
ꢀ0.51
ꢀ0.50
ꢀ0.48
ꢀ0.46
ꢀ0.42
0.66
0.71
0.76
0.81
0.91
1.13
1.18
1.21
1.23
1.29
a2
Ecell = Ep ꢀ Epa1) are 1.20 ( 0.05 V in CH3CN and 1.21 (
0.06 V in CH2Cl2, which are essentially identical for both
solvents. This contrasts with the small but notable change in
Ecell values (∼100 mV) at higher concentrations between the two
solvents. For a monomeric neutral thiatriazinyl radical, the redox
process is assumed to involve removal from or addition to the π-
SOMO and can be represented as
a Reported vs Fc0/þ on a Pt working electrode, ν = 0.2 V sꢀ1, T = 20 (
2 °C. b Em = [Ep þ Epc1]/2 ≈ E0/. c Irreversible wave observed. d Ecell
=
a1
a2
a1
Ep ꢀ Ep
.
As the concentration increased (evidenced by development of
a brown color), a second oxidation process (Epa3/c3) appeared at
a more positive potential, and in the limit, this process displayed
return waves of comparable peak current height (Ipc3/Ipa3 = 0.95
(5a), 0.96 (5b), 0.89 (5c), 0.93 (5d), 0.93 (5e)) under all of the
conditions used in this study. Figure 11 shows overlapping traces
at three different concentrations of 5e. At the highest concentra-
tion, process Epa2 is just a residual shoulder on the reversible wave
at more anodic potential. Much the same thing occurred for all
five compounds in both solvents. In each case, there was only a
single potential associated with reduction of neutral material at all
concentrations studied.
When the solutions are at the full indicated concentration
(Table 2), the potential range for the “reduction” process is
from ꢀ0.58 to ꢀ0.47 V vs Fc0/þ in CH3CN and ꢀ0.61 to ꢀ0.50 V
vs Fc0/þ in CH2Cl2 and has an overall change of 0.11 V in both
solvents. The potential range for the “oxidation” process is from
Since the same redox orbital is involved in the electrochemical
oxidation and reduction reactions, one expects the substituents R
to influence both redox processes by the same amount, so long as
the influence is purely inductive. Exactly this has been observed
in electrochemical investigations of more than 20 1,2,3,5-
dithiadiazolyls.3 However, unlike the dithiadiazolyls, the redox
orbital of the thiatriazinyls (the SOMO, Figure 10) does not have
a node at the substituent-bearing carbon atoms. It is to be
expected, therefore, that resonance effects can have a significant
influence. These are expected to affect the oxidation and reduc-
tion processes differently.
Another factor that cannot be discounted is the effect of dimer
formation on the redox processes since thiatriazinyls form strong
5129
dx.doi.org/10.1021/ic2003996 |Inorg. Chem. 2011, 50, 5123–5136

Inorganic Chemistry
ARTICLE
Figure 12. The in-phase and out-of-phase combinations of the 3b1
SOMO of model 3,5-dihydro-1,2,4,6-thiatriazinyls.
Figure 13. Calculated CVs for both (a) monomeric (blue line) and (b)
dimeric (black line) solutions of 5e resulting from the input of kinetic
parameters listed in the text and following the “ladder” scheme shown in
dimers (d(S S = 2.643(21) Å) compared to 1,2,3,5-dithiadia-
3 3 3
zolyls (mean d(S S = 3.03ꢀ3.16 Å).3c,22 In a dimeric thia-
3 3 3
Scheme 2. These CVs follow the specific case of ν = 0.2 V sꢀ1, Keq 1
=
triazine, the redox process can be represented as
400, kf1 = 1 Mꢀ1 sꢀ1, Keq 2 = 0.0766, kf2 = 0.01 Mꢀ1 sꢀ1, Keq 3 = 271, kf3 =
1 Mꢀ1 sꢀ1, Keq 4 = 106, kf4 = 5 sꢀ1 (first-order decay of [TTA]þ), ks1
=
ks2 = ks3 = ks4 = 0.03 cm sꢀ1, conc. dimer =0.01 mol Lꢀ1, conc. monomer
0.001 mol Lꢀ1. The potential was first swept in the cathodic direction
starting from 0.4 V.
The formation of thiatriazinyl dimers has been convincingly
interpreted as the result of diffuse side-on overlap of the 3b1
π orbital of the individual thiatriazinyl rings (Figure 12) in a face-
to-face dimeric arrangement similar to that found in the solid
state for dithionite salts.5e,23 This overlap represents the only net
“bonding” interaction between the rings; that it is centered on a
single sulfur atom which has the largest share of the unpaired spin
density leads to relatively strong bonding. The dimer is a
diamagnetic species, and hence oxidation occurs from a different
orbital (the in-phase combination) than reduction (the out-of-
phase combination). It is therefore to be expected that the
substituents influence these distinct orbitals by different amounts.
Digital Modeling of CVs of the 1,2,4,6-Thiatriazinyls. In
order to further probe the effect of monomerꢀdimer equilibrium
on the voltammetry, we turned to digital modeling of the
voltammograms at both low and high concentrations using
DigiElch software.24 Starting with lower concentrations of the
radical in solution (∼1/10 the values listed in Table 2), we estab-
lished a plausible model for the monomer undergoing both a
chemically irreversible oxidation and a reversible reduction. The
resultant CV is shown as the blue trace in Figure 13. In this case,
the model assumes that following a one-electron oxidation of the
thiatriazinyl radical (abbreviated as [TTA]• in Scheme 2) there is
a chemically irreversible first-order decay step (Keq 4= 106, kf4 = 5
Scheme 2. Proposed “Ladder” Scheme Interrelating the
Voltammetric Behavior of Thiatriazinyls at Lower and Higher
Concentrations
have been undertaken using absorption spectroscopy.2e The self-
consistent fit of these data to the “ladder” scheme is thus satisfying
and chemically reasonable in view of the large association constants
expected for thiatriazinyls. That the dication favors monomers fits
well with removal of the very electrons involved in the association.
The only surprise is the apparent “decomposition” of the mono-
meric cation under inert conditions where these 6π H€uckel species
are expected to be thermally stable.5i We propose that the
irreversibility of this electrode process may involve electron self-
exchange.26 Thus, a rapid pseudo-first-order cascade reaction:
s
ꢀ1). No chemical step is associated with the one-electron
reduction. Then, at higher concentrations, the model which fit
most closely to the experimental CVs follows the “ladder”
scheme shown in Scheme 2.25 The favorable formation of the
dimer [TTA2] at higher concentrations (Keq 1 = 400ꢀ800, kf1 = 1
þ
þ
3
3
½TTAꢁ þ ½TTAꢁ T ½TTAꢁ þ ½TTAꢁ
ð3Þ
M
ꢀ1 sꢀ1) and the unfavorable formation of the dication dimer
ꢀ1
[TTA2]2þ (Keq 2 < 1, kf2 = 0.01 M
s
ꢀ1) results in very
could remove the cation from the electrode surface when the
dominant species in solution is [TTA]•. On the other hand, when
[TTA2] is the dominant species, the driving force for the self-
exchange reaction weakens, and this reaction may becomesluggish.
While we believe that this mechanism is a plausible proposal,
detailedtestinginwhich actualvoltammogramsarefit to theorywill
be required to obtain precise values for the kinetic parameters. Such
an undertaking is beyond the scope of the current project.
Correlation of Redox Potentials with Gas Phase Calcula-
tions. Previous studies have shown that quantum calculations,
when used judiciously, can provide corroboration of electroche-
mical results.3 Fully geometry optimized calculations at the
(U)B3LYP/6-31G(d) level of theory were performed for the
satisfactory simulations (they are a good visual match to experi-
mental data, see black trace in Figure 13) when the oxidation of
the [TTA2] dimer is a two-electron process. Assuming a one-
electron oxidation for this step does not result in a good match. The
formation of the monoanionic dimer [TTA2] •ꢀ from [TTA]• þ
[TTA]ꢀ was also favored (Keq 3 = 271, kf3 < 10 M ꢀ1 sꢀ1).
The value for Keq 1 of 400ꢀ800 is in reasonable agreement
with previously derived equilibrium constants for the association
of diphenyl thiatriazinyls and selenatriazinyls from Oakley et al.
from EPR experiments (Keq = 33 for sulfur and Keq = 2000 for
selenium).6 Similar investigations into the monomerꢀdimer equi-
librium for 4-(20-pyridyl)-1,2,3,5-dithiadiazolyl (Keq = 0.05 at RT)
5130
dx.doi.org/10.1021/ic2003996 |Inorg. Chem. 2011, 50, 5123–5136

Inorganic Chemistry
ARTICLE
Table 4. Calculated and Experimental Redox Potentials of
5aꢀe for the Monomeric Radical
ꢀ1/0 process
0/þ1 process
Em (V)
Em (V)
Epa1 (V) Epa1 (V)
cmpd calc (eV) CH3CN
CH2Cl2 calc (eV) CH3CN
CH2Cl2
5a
5b ꢀ1.94
5c
ꢀ1.99
ꢀ1.89
ꢀ0.57
ꢀ0.56
ꢀ0.54
ꢀ0.50
ꢀ0.46
0.11
ꢀ0.51
ꢀ0.50
ꢀ0.48
ꢀ0.46
ꢀ0.42
0.09
7.29
7.50
7.65
7.72
7.90
0.61
0.59
0.67
0.73
0.74
0.84
0.25
0.66
0.71
0.76
0.81
0.91
0.25
5d ꢀ2.16
5e
ꢀ2.25
Δ
0.36
monomeric cation, anion, and neutral thiatriazine derivatives. No
attempt was made to model the weakly bound dimeric molecules.
The energies of these optimized molecules were then used to
define the theoretical redox reactions:
0=þ 1 process : Ecation ꢀ Eradical ¼ Eoxidation
ð4Þ
ꢀ 1=0 process : Eanion ꢀ Eradical ¼ Ereduction
ð5Þ
Table 4 lists the calculated and experimental redox potentials
for 5aꢀe from the experimental CV data at low concentrations
taken from Table 3, and correlation curves are presented in
Figure 14. The results are graphed by process type, and each
graph includes results for both solvents.
Excellent correlation is obtained between the calculated gas
phase and experimental solution phase oxidation and reduction
potentials of the monomeric radicals 5aꢀe, which is particularly
noteworthy for the 0/þ1 process, which uses peak potentials
from an irreversible process in both solvents.3b The reliability of
the electrochemical data for this series obtained from CV is
confirmed by this correlation. The solution oxidation and reduc-
tion potentials also correlate well with Hammett σp substituent
constants (Figures S22 and S23, Supporting Information).27
Figure 14. Plot of the calculated vs experimental monomer reduction
potentials (top) and oxidation potentials (bottom) of 5aꢀe as measured
by cyclic voltammetry in CH2Cl2 (black line, R = 0.975, top; R = 0.975,
bottom), and (b) CH3CN (red line, R = 0.993, top; R = 0.994, bottom)
solution. Error bars express an estimated error in measured potential
of (0.01 V.
’ EXPERIMENTAL SECTION
General Procedures. Trifluoroacetonitrile (PCR Inc.) was ob-
tained commercially and used as received. SCl2 (Aldrich) was distilled
from PCl3 with CaCl2 moisture protection. Triphenylantimony
(Aldrich) and HCl(g) (Praxair) were obtained commercially and used
as received. The substituted aryl amidine hydrochlorides were prepared
according to the literature method.7 Unless otherwise indicated, all
procedures were performed under an atmosphere of purified N2 using a
drybox, Schlenkware, and vacuum-line techniques. Solvents used were
reagent-grade or better. Acetonitrile (HPLC grade) was double-distilled
from P2O5 and CaH2. Dichloromethane was distilled from CaH2.
n-Heptane was dried by distillation over LiAlH4 and toluene over sodium.
Anhydrous diethyl ether was dried by distillation from sodium wire.
Spectroscopic Methods. Infrared spectra were obtained as KBr
plates and were recorded on a Bomem MB102 Fourier transform
spectrometer. Melting points were determined on an electrothermal
melting point apparatus (capillaries) and are uncorrected. EPR spectra
(X band) were recorded on a Bruker EMX-113/12 spectrometer as
solutions in dichloromethane in 4 mm Pyrex glass tubes sealed under
vacuum conditions. NMR spectra were acquiredat 250.13 (1H) and 62.90
(13C) MHz on a Bruker/Tecmag AC250 spectrometer using CDCl3 as
the solvent and thus a reference. Mass spectra were recorded by the Mass
Spectrometry Center, University of Alberta, Edmonton. Microanalyses
were performed by MꢀHꢀW Laboratories in Phoenix, Arizona.
’ CONCLUSIONS
This paper reports a comprehensive investigation of a com-
plete series of unsymmetrical aryl/trifluoromethyl-substituted
thiatriazinyls. The synthetic work has elaborated new and
versatile imidoylamidine reagents and demonstrated their ability
to form the ring compounds through direct condensation with
sulfur halides. Conversion to the neutral thiatriazinyls was
successful for all five exemplars. All of the synthetic results are
supported by thorough structural studies that provide extremely
reliable mean parameters from high-quality, low-temperature
diffraction data. The complex EPR spectral data are shown to
fit to well-defined substituent trends even though the variation in
hfs values is small. Finally, complex voltammetric behavior
involving the first reliably detected monomerꢀdimer equilibria
in the history of the electrochemistry of thiazyl rings has been
modeled using sophisticated digital software to provide a inter-
pretation that is consistent with the predicted chemical nature
of the various redox products. Further work on thiatriazinyls
including detailed fitting of monomerꢀdimer equilibria to volta-
mmetric data is in progress in our laboratory and will be reported
at a future time.
Electrochemical Methods and Procedures. Crystals of high
purity 5aꢀe were collected, weighed, and loaded into individual break-
seal tubes and attached to the electrochemical cell by fusing the glass
along with a known mass of ferrocene in a separate break-seal.25 The
electrodes are all platinum and were sealed into the glass and cleaned in
5131
dx.doi.org/10.1021/ic2003996 |Inorg. Chem. 2011, 50, 5123–5136

Inorganic Chemistry
ARTICLE
between experiments using concentrated HNO3. Cyclic voltammo-
grams (CVs) were obtained at temperatures of 21 ( 2 °C in CH3CN
and CH2Cl2 solutions containing 0.1 and 0.4 M [nBu4N][PF6], respectively,
as the supporting electrolyte. The electrolyte was loaded into the cell and
dried under vacuum conditions in an oil bath set at 80 °C overnight prior
to transfer of the solvent. Both CH2Cl2 and CH3CN were distilled and
stored over molecular sieves prior to use. The solvents were freezeꢀ
thawꢀdegassed at least five times prior to their transfer into the cell over a
vacuum line. For CVs, all measurements were recorded at room tempera-
A 1:1 mixture of ethanol and concentrated HCl (30 mL) was added slowly
through a dropping funnel to give a pink precipitate. The solid was recovered
by filtration and dried in the air. It was then redissolved in distilled water and
treated with excess KOH. CH2Cl2 (2 ꢂ 200 mL) was used to extract the
product. The solvent was rotary evaporated, and the solution was dried using
andrydrous sodium sulfate to give a yellow solid. The solid was sublimed at
70 °C under vacuum conditions to give 1.51 g (21.8% yield) of pure 1b. 1H
NMR (CDCl3):δ2.40 (s, 3H, CH3), 4.83 (s, 3H, NH), 7.23 (d, 2H, JHꢀH
8.1 Hz, CArH), and 7.51 (d, 2H, JHꢀH = 8.4 Hz, CArH).
ClC6H4C(NH)NH2, 1d. Amidine 1d was prepared by the same
method reported for 1a using 4-chlorobenzonitrile (7.69 g, 0.0559 mol)
=
ture (22 ( 2 °C), and the scan rates were varied between 0.1 and 10 V sꢀ1
.
In a typical experiment, the electrolyte solution is scanned to obtain a
background current and determine the effective electrochemical window.
Next, the seal to the analyte is broken, and a small amount is dissolved
until a signal with satisfactory S/N is obtained. The concentration of the
analyte is incrementally increased while obtaining CV traces. Finally, all of
the analyte is thoroughly mixed with the electrolyte solution to establish a
known “full” concentration. After all CV data are satisfactorily recorded,
the seal to the reference is broken, and a sufficient quantity of ferrocene is
added to the cell to obtain an accurate reference for the voltage scale.
X-Ray Crystallography. Crystals were selected and mounted in
Paratone on the ends of thin glass capillaries and cooled on the
goniometer head to ꢀ100 °C with the Bruker low-temperature acces-
sory attached to the APPEX-II diffractometer. Mo KR radiation (λ =
0.71073 Å) was used throughout. Multiscan absorption corrections were
applied to all of the data sets, and refinement was conducted with full-
matrix least-squares on F2 using SHELXTL 6.14.28 In the case of 5b, data
sets collected at ꢀ100 °C solve and refine but are stuck at R1 = 0.14 due
to a suspected superlattice formation. Therefore, another crystal was
selected, covered in epoxy cement, and glued to the end of a glass fiber.
Diffraction data collected at 23 °C refined to a low R factor and provided
a fully acceptable structure. Details of the refinements and disorder
models for typical rotational disorder of CF3 groups are provided in the
electronic crystallographic information file. Crystal data and refinement
parameters are also summarized in Table 5. Structures of 2c and 5c were
determined at room temperature and at ꢀ100 °C.
and Et2O LiN(SiMe3)2 (12.7 g, 0.0524 mol) in 200 mL of distilled
3
diethyl ether. The pale pink-colored solid was sublimed at 91 °C under
1
vacuum conditions to give 4.33 g (53.7% yield) of pure 1d. H NMR
(CDCl3): δ 3.88 (s, 3H, NH), 7.41 (d, 2H, JHꢀH = 8.4 Hz, CArH), and
7.75 (d, 2H, JHꢀH = 8.7 Hz, CArH).
CF3C6H4C(NH)NH2, 1e. Amidine 1e was prepared by the same
method reported for 1a using R,R,R-trifluoro-p-tolunitrile (8.94 g,
52.2 mmol) and Et2O LiN(SiMe3)2 (12.4 g, 51.5 mmol) in 200 mL
3
of distilled diethyl ether. After the solid was collected by filtration and
air-dried, it was sublimed at 85 °C under vacuum conditions to give 5.61
g (57.9% yield) of pure 1e. 1H NMR (CDCl3): δ 4.90 (s, 3H, NH), 7.75
(d, 2H, JHꢀH = 8.4 Hz, CArH), 7.70 (d, 2H, JHꢀH = 8.4 Hz, CArH). 19
NMR (CDCl3): δ ꢀ62.02 (CF3).
F
Synthesis of Trifluoromethyl Imidoylamidines 2aꢀe.
C6H5C3N3H3F3, 2c. Trifluoroacetonitrile (2.06 g, 21.7 mmol) was
measured on a vacuum line by PVT methods and added to 2.53 g
(21.1mmol) of benzamidine 1c in 20 mLofacetonitrileina 300ꢂ 25mm
heavy-wall Pyrex reactor fitted with a Rotaflow stopcock equipped with a
small magnetic stirring bar at ꢀ196 °C. The mixture was warmed to RT,
heated to 60 °C with stirring for 10 min, and then cooled to RT and
evaporated, leaving a clear oil. Upon refrigeration, this solidified to leave
4.37 g of a white solid (20.3 mmol, 96% yield), mp 42ꢀ45 °C. 1H NMR
(δ, CDCl3): 6.8 (s, NH), 7.38ꢀ7.54 (m, 3H, phenyl), 7.83ꢀ7.88 (m,
2H, CArH), 9.1 (s, NH), 11.0 (s, NH). 13C NMR (δ, CDCl3): 117.8 (q,
281 Hz), 127.4 (s), 129.0 (s), 132.2 (s), 135.2 (s), 163.5 (q, 33 Hz),
165.9 (s). Mass spectrum (m/e): 214 (PhC2N3H2CF3þ, 100%), 169
(PhC2N2H2F2þ, 4%), 146 (PhC2N3H3þ, 12%), 129 (PhC2N2þ, 6%),
104 (PhCNHþ, 48%). Analysis calculated (found): 50.24 (50.35)% C,
3.75 (4.01)% H, 19.53 (19.37)% N.
Computational Details. DFT calculations undertaken for this
study employed geometry optimization at the (U)B3LYP/6-31G(d) level.
Harmonic vibrational frequencies were calculated for all optimized geo-
metries to confirm that they are stationary points. Calculations of energies
and EPR hyperfine splitting (hfs) constants used the same level of theory.
All calculations were performed using the Gaussian 98 suite of programs.18
Synthesis of Amidines 1aꢀe. C6H4C(NH)NH2, 1c. Benzami-
dine hydrochloride hydrate (26.3 g, 0.173 mol) and KOH (32.4 g,
0.576 mol) were added to 175 mL of distilled water. CH2Cl2 (2 ꢂ
200 mL) was used to extract the product. The solvent was rotary
evaporated, leaving a pale yellow oil which solidified overnight in the
freezer. The solid was sublimed at 80 °C under vacuum conditions to
give 12.79 g (61.9% yield) of pure white 1c. 1H NMR (CDCl3): δ 5.28
(s, 3H, NH), 7.42ꢀ7.62 (m, 5H, phenyl).
4-CH3OC6H5C2N3H3CF3, 2a. Imidoylamidine 2a was prepared by
the same method as 2c using CF3CN (1.09 g, 11.5 mmol) and 1a (1.58 g,
10.5 mmol) in 10 mL of acetonitrile. A white solid remained, which was
sublimed in vacuo at 46 °C to leave 1.43 g of pure white crystals
(5.8 mmol, 56% yield), mp 41ꢀ45 °C. 1HNMR(δ,CDCl3):3.9(s,OCH3),
6.7 (s, NH), 6.9 (d, 2H, CArH),7.9(d, 2H, CArH), 9.0 (s, NH), 11.0 (s, NH).
13C NMR (δ, CDCl3): 55.7 (s), 114.3 (s), 117.9 (q, 281 Hz), 127.3 (s),
129.3 (s), 163.1 (s), 163.4 (q, 33 Hz), 165.3 (s). Mass spectrum (m/e):
244 (CH3OPhC2N3H2CF3þ, 42%), 199 (CH3OPhC2N2H2F2þ, 4%),
176 (CH3OPhC2N3H3þ, 20%), 159 (CH3OPhC2N2þ, 13%), 134
(CH3OPhCNHþ, 100%). Analysis calculated (found): 48.98 (48.85)%
C, 4.11 (4.22)% H, 17.14 (17.32)% N.
4-CH3C6H5C2N3H3CF3, 2b. Imidoylamidine 2b was prepared by the
same method as 2c using CF3CN (1.34 g, 14.1 mmol) and 1b (1.96 g,
14.6 mmol) in 15 mL of acetonitrile. A pink solid remained, which was
sublimed in vacuo at 48 °C to leave 2.48 g of white crystals (10.8 mmol,
77% yield), mp 54ꢀ57 °C. 1H NMR (δ, CDCl3): 2.4 (s, CH3), 6.7 (s,
NH), 7.3 (d, 2H, CArH), 7.8 (d, 2H, CArH), 9.1 (s, NH), 11.0 (s, NH).
13C NMR (δ, CDCl3): 21.7 (s), 117.8 (q, 281 Hz), 127.6 (s), 129.7 (s),
132.4 (s), 142.8 (s), 163.6 (q, 33 Hz), 165.8 (s). Mass spectrum (m/e):
228 (CH3PhC2N3H2CF3þ, 100%), 183 (CH3PhC2N2H2F2þ, 2%), 160
(CH3PhC2N3H3þ, 7%), 143 (CH3PhC2N2þ, 4%), 118 (CH3PhCNHþ,
30%). Analysis calculated (found): 52.40 (52.34)% C, 4.40 (4.60)% H,
18.33 (18.50)% N.
CH3OC6H4C(NH)NH2, 1a. 4-Methoxybenzonitrile (15.08 g,
0.1133 mol) was added to Et2O LiN(SiMe3)2 (27.26 g, 0.1129 mol)
3
in 200 mL of distilled diethyl ether and stirred for 20 h with a drying tube
attached. A 1:1 mixture of ethanol and concentrated HCl (30 mL) was
added slowly through a dropping funnel, leaving a white precipitate. The
solid was recovered by filtration and dried in the air. It was then
redissolved in distilled water and treated with excess NaOH solution,
affording a pale purple solid. This was sublimed at 91 °C under vacuum
conditions to give 3.72 g (22.2% yield) of pure 1a. Plate-shaped X-ray-
quality crystals were collected on a coldfinger following vacuum sublima-
tion. 1H NMR (CDCl3): δ 3.85 (s, 3H, CH3), 4.99 (s, 3H, NH), 6.93 (d,
2H, JHꢀH = 9.0 Hz, CArH), 7.57 (d, 2H, JHꢀH = 8.4 Hz, CArH).
CH3C6H4C(NH)NH2, 1b. p-Tolunitrile (6.356 g, 0.05425 mol) was
added to Et2O LiN(SiMe3)2 (12.47 g, 0.05165 mol) in 200 mL of
3
distilled diethyl ether and stirred for 22 h with a drying tube attached.
5132
dx.doi.org/10.1021/ic2003996 |Inorg. Chem. 2011, 50, 5123–5136

Inorganic Chemistry
ARTICLE
Table 5. Crystallographic Data and Refinement Parameters for Compounds 1ꢀ6
complex
1a
2a
2c
2e
4b
4e
empirical formula
Mr
C8H10N2O
150.18
C10H10F3N3O
245.21
C9H8F3N3
C10H7 F6N3
283.19
C10H7ClF3N3S
293.70
C10H4ClF6N3S
347.67
215.18
cryst size (mm3)
cryst syst
space group
a (Å)
0.25 ꢂ 0.15 ꢂ 0.12
triclinic
P1
0.45 ꢂ 0.37 ꢂ 0.20
triclinic
0.401 ꢂ 0.208 ꢂ 0.116
monoclinic
P21/n
0.652 ꢂ 0.429 ꢂ 0.310
monoclinic
P21/n
0.31 ꢂ 0.16 ꢂ 0.16
orthorhombic
P21212
0.17 ꢂ 0.14 ꢂ 0.14
monoclinic
P21/c
P1
5.005(10)
10.15(2)
15.54(3)
86.58(2)
87.40(2)
79.77(2)
775(3)
9.2967(19)
10.538(2)
11.787(3)
93.783(2)
105.567(2)
97.335(2)
1097.2(4)
4
8.8135(17)
16.356(3)
14.069(3)
90.00
11.2481(8)
14.6690(10)
14.3328(10)
90.00
14.507(3)
17.395(3)
4.7852(8)
90.00
14.5920(14)
11.8343(11)
7.4331(7)
90.00
b (Å)
c (Å)
R (deg)
β (deg)
γ (deg)
V (Å3)
101.503(2)
90.00
93.7750(10)
90.00
90.00
94.7720(10)
90.00
90.00
1987.3(7)
8
2359.8(3)
8
1207.6(3)
4
1279.1(2)
4
Z
4
Dc (Mg mꢀ3
)
1.287
1.484
1.438
1.594
1.615
1.805
3
μ (mmꢀ1
)
0.088
0.134
0.130
0.165
0.511
0.530
θ range
2.04ꢀ27.89
8956
1.80ꢀ27.00
12270
1.93ꢀ27.42
28147
1.99ꢀ26.36
24664
1.83ꢀ29.59
12307
2.22ꢀ28.83
14808
coll. reflns
obsd reflns
3580
4750
4532
4826
2987
3137
no. of params
F(000)
219
333
345
404
192
217
320
504
880
1136
592
688
T (K)
173(2)
173 (2)
173(2)
173(2)
173(2)
173(2)
R1, wR2 [I > σ2(I)]
R1, wR2 [all data]
GOF
0.0410, 0.0959
0.0695, 0.1053
0.978
0.0338, 0.0842
0.0420, 0.0896
1.038
0.0370, 0.0859
0.0496, 0.0934
1.041
0.0468, 0.1217
0.0538, 0.1298
1.038
0.0379, 0.0965
0.0550, 0.1026
1.031
0.0327, 0.0843
0.0414, 0.0895
1.056
complex
5a
5b
5c
5d
5e
6
empirical formula
Mr
C10H7F3N3OS
274.25
C10H7F3N3S
258.25
C9H5F3N3S
244.22
C9H4ClF3N3S
278.66
C10H4F6N3S
312.22
C7H7ClN2O2S
218.66
Cryst size (mm3)
cryst syst
space group
a (Å)
0.476 ꢂ 0.192 ꢂ 0.118 0.185 ꢂ 0.171 ꢂ 0.153 0.287 ꢂ 0.215 ꢂ 0.086 0.29 ꢂ 0.22 ꢂ 0.14 0.311 ꢂ 0.155 ꢂ 0.06 0.13 ꢂ 0.12 ꢂ 0.12
triclinic
triclinic
triclinic
triclinic
triclinic
tetragonal
P41212
9.4879(9)
9.4879(9)
21.584(4)
90.00
P1
P1
P1
P1
P1
7.1075(8)
11.5130(13)
13.3832(15)
88.0000(10)
86.2810(10)
80.4830(10)
1077.5(2)
4
11.5126(12)
14.7539(15)
15.0498(15)
111.2160(10)
99.9130(10)
107.6770(10)
2153.0(4)
8
9.136(4)
10.728(4)
11.016(5)
84.652(4)
69.663(4)
68.715(3)
942.6(7)
4
11.1851(15)
14.799(2)
14.806(2)
110.9450(10)
103.210(2)
106.5720(10)
2038.8(5)
8
10.7943(6)
13.3162(8)
16.7910(10)
94.0420(10)
107.4760(10)
96.1190(10)
2275.6(2)
8
b (Å)
c (Å)
R (deg)
β (deg)
γ (deg)
V (Å3)
90.00
90.00
1943.0(5)
8
Z
Dc (Mg mꢀ3
)
1.691
1.593
1.721
1.816
1.823
1.495
3
μ (mmꢀ1
)
0.333
0.321
0.362
0.600
0.358
0.576
θ range
1.79ꢀ27.42
15562
1.96ꢀ26.59
23211
1.97ꢀ27.45
13575
1.61ꢀ28.49
22885
1.90ꢀ28.73
26627
2.34ꢀ26.86
21297
coll. reflns
obsd reflns
4872
8841
4267
9467
10697
2076
no. of params
F(000)
327
729
577
629
777
124
556
1048
492
1112
1240
896
T (K)
173(2)
296(2)
173(2)
0.0368, 0.0976
0.0444, 0.1036
1.047
173(2)
173(2)
173(2)
0.0421, 0.1102
0.0610, 0.1166
0.984
R1, wR2 [I > σ2(I)]a,b 0.0273, 0.0732
R1, wR2 [all data]a,b
GOFc
0.0379, 0.1035
0.0572, 0.1127
1.026
0.0426, 0.0893
0.0825, 0.1054
1.022
0.0440, 0.1137
0.0689, 0.1256
1.036
0.0302, 0.0757
1.032
a wR2 = [∑{w(Fo ꢀ Fc2)2}/∑w (Fo2)2]1/2; R1 = ∑||Fo| ꢀ |Fc||/∑|Fo|. b Fo > 4σ(Fo). c GOF = [∑{w (Fo ꢀ Fc2)2}/(n ꢀ p)]1/2, where n = number of
2
2
reflections and p = number of refined parameters.
4-ClC6H5C2N3H3CF3, 2d. Imidoylamidine 2c was prepared by the
same method as 2c using trifluoroacetonitrile (2.07 g, 21.8 mmol) and
1d (3.25 g, 21.0 mmol) in 20 mL of acetonitrile. A white solid remained
which was sublimed in vacuo at 49 °C to leave 4.72 g of pure white
crystals (18.9 mmol, 90% yield), mp 45ꢀ48 °C. 1H NMR (δ, CDCl3):
6.7 (s, NH), 7.4 (d, 2H, CArH), 7.8 (d, 2H, CArH), 9.2 (s, NH), 11.0
5133
dx.doi.org/10.1021/ic2003996 |Inorg. Chem. 2011, 50, 5123–5136

Inorganic Chemistry
ARTICLE
(s, NH). 13C NMR (δ, CDCl3): 117.7 (q, 281 Hz), 128.8 (s), 129.3 (s),
133.6 (s), 138.6 (s), 163.3 (q, 34 Hz), 164.8 (s). Mass spectrum (m/e):
248 (ClPhC2N3H2CF3þ, 42%), 180 (ClPhC2N3H3þ, 10%), 163
(ClPhC2N2þ, 13%), 138 (ClPhCNHþ, 100%). Analysis calculated
(found): 43.31 (43.50)% C, 2.83 (3.00)% H, 16.83 (17.06)% N.
4-CF3C6H5C2N3H3CF3, 2e. Imidoylamidine 2e was prepared by the
same method as 2c using trifluoroacetonitrile (1.02 g, 10.7 mmol) and
1e (1.98 g, 10.5 mmol) in 10 mL of acetonitrile. A white solid remained
which was sublimed in vacuo at 60 °C to leave 2.52 g of white crystals
recrystallized with hot heptane to leave an orange solid (2.78 g, 99%
yield). 1H NMR (δ, CDCl3): 7.53 (d, 2H, CArH), 8.41 (d, 2H, CArH).
4-CF3C6H5C2N3SClCF3, 4e. Compound 4e was prepared using the
same method as 4c using SCl2 (2.2 mL, 34.6 mmol) in 10 mL of
acetonitrile and 3e (2.13 g, 6.7 mmol) in 40 mL of acetonitrile. The
solvent was removed in vacuo, leaving a dark orange oil (2.50 g, 98%
yield). 1H NMR (δ, CDCl3): 7.82 (d, 2H, CArH), 8.59 (d, 2H, CArH).
Synthesis of 3-Trifluoromethyl-5-aryl-1,2,4,6-thiatriazi-
nyls 5aꢀe. C6H5C2N3SCF3 5c. A pear-shaped solids addition funnel
was loaded with triphenylantimony (1.42 g, 4.0 mmol) and attached to a
side arm flask containing 4c (2.23 g, 8.0 mmol) and 25 mL of
acetonitrile. After degassing the solvent with three freezeꢀthaw cycles
and warming to RT, while still under a vacuum, the Ph3Sb was added
with stirring. A fine dark purple precipitate formed, and the crystals were
vacuum filtered under nitrogen to give 1.14 g of crude product
(4.7 mmol, 59% yield), mp. (dec.) 110ꢀ3 °C. Purification was achieved
by vacuum sublimation in a three-zone tube furnace. Mass spectruþm
1
(8.9 mmol, 85% yield), mp 80ꢀ83 °C. H NMR (δ, CDCl3): 6.8 (s,
NH), 7.7 (d, 2H, CArH), 8.0 (d, 2H, CArH), 9.3 (s, NH), 11.1 (s, NH).
13C NMR (δ, CDCl3): 117.7 (q, 281 Hz), 123.9 (q, 273 Hz), 126.0 (q, 4
Hz), 127.9 (s), 133.9 (q, 33 Hz), 138.6 (s), 163.2 (q, 34 Hz), 164.7 (s).
Mass spectrum (m/e): 282 (CF3PhC2N3H2CF3þ, 100%), 237 (CFþ3-
PhC2N2H2F2þ, 5%), 214 (CF3PhC2N3H3þ, 13%), 197 (CF3PhC2N2
,
3%), 172 (CF3PhCNHþ, 45%). Analysis calculated (found): 42.42
(42.32)% C, 2.49 (2.39)% H, 14.84 (14.88)% N.
Synthesis of Trifluoromethyl Imidoylamidine Hydrochlor-
(m/e): 245 (PhC3N3SF3þ, 43%), 149 (PhCN2Sþ, 6%), 129 (PhC2N2
,
3%), 104 (PhCNþ, 100%). Analysis calculated (found): 44.26 (44.12) %
C, 2.06 (2.24) %H, 17.21 (17.13) %N.
ides 3aꢀe. C6H5C3N3H3F3 HCl, 3c. Anhydrous HCl was bubbled
3
through a suspension of 2c (2.04 g, 9.5 mmol) in diethyl ether. The
white precipitate was filtered in the air and dried to leave 2.34 g of crude
solid (9.3 mmol, 98% yield), mp (dec.) 174ꢀ175 °C. For FTIR data, see
the Supporting Information.
4-CH3OC6H5C2N3SCF3, 5a. Compound 5a was prepared using the
same method as 5c from 4a (1.25 g, 4.0 mmol) and Ph3Sb (0.79 g,
2.2 mmol) in 20 mL of acetonitrile to give 0.41 g of dark purple solids
(1.5 mmol, 37% yield), mp (dec.) 135ꢀ40 °C. Purification was achieved
by vacuum sublimation in a three-zone tube furnace. Mass spectrum
(m/e): 274 (CH3OPhC3N3SF3þ, 65%), 255 (CH3OPhC3N3SF2þ, 4%),
179 (CH3OPhCN2Sþ, 3%), 117 (CH3OPhCNþ, 100%). Analysis calcu-
lated (found): 43.80 (43.90)% C, 2.57 (2.71)% H, 15.32 (15.18)% N.
4-CH3C6H5C2N3SCF3, 5b. Compound 5b was prepared using the
same method as 5c from 4b (1.05 g, 3.6 mmol) and Ph3Sb (0.64 g,
1.8 mmol) in 15 mL of acetonitrile to give 0.46 g of dark purple solids
(1.8 mmol, 50% yield), mp (dec.) 140ꢀ2 °C. Purification was achieved
by vacuum sublimation in a three-zone tube furnace. Mass spectrum
(m/e): 258 (CH3PhC3N3SF3þ, 100%), 239 (CH3PhC3N3SF2þ, 4%),
163 (CH3PhCN2Sþ, 24%), 117 (CH3PhCNþ, 45%). Analysis calcu-
lated (found): 46.51 (46.67)% C, 2.73 (2.96)% H, 16.27 (16.41)% N.
4-ClC6H5C2N3SCF3, 5d. Compound 5d was prepared using the
same method as 5c from 4d (2.78 g, 8.9 mmol) and Ph3Sb (1.71 g,
4.8 mmol) in 30 mL of acetonitrile to give 1.36 g of dark purple solids
(4.9 mmol, 55% yield), mp (dec.) 131ꢀ4 °C. Purification was achieved
by vacuum sublimation in a three-zone tube furnace. Mass spectrum
(m/e): 278 (ClPhC3N3SF3þ, 100%), 259 (ClPhC3N3SF2þ, 5%), 183
(ClPhCN2Sþ, 18%), 137 (ClPhCNþ, 47%). Analysis calculated
(found): 38.79 (38.72)% C, 1.45 (1.41)% H, 15.08 (15.08)% N.
4-CF3C6H5C2N3SCF3, 5e. Compound 5e was prepared using the
same method as 5c from 4e (2.32 g, 6.7 mmol) and Ph3Sb (1.31 g,
3.7 mmol) in 25 mL of acetonitrile to give 0.45 g of a dark purple product
(1.4 mmol, 20% yield), mp (dec.) 101ꢀ5 °C. Purification was achieved
by vacuum sublimation in a three-zone tube furnace. Mass spectrum
(m/e): 312 (CF3PhC3N3SF3þ, 100%), 197 (CH3PhC2N2þ, 24%), 171
(CF3PhCNþ, 30%). Analysis calculated (found): 38.47 (38.52)% C,
1.29 (1.53)% H, 13.46 (13.34)% N.
4-CH3OC6H4C3N3H3F3 HCl, 3a. Imidoylamidine hydrochloride
3
3a was prepared using the same method as 3c from 2a (1.01 g,
4.1 mmol), providing 1.09 g of a crude white solid (3.9 mmol, 94%
yield), mp (dec.) 213ꢀ215 °C.
4-CH3C6H4C3N3H3F3 HCl, 3b. Imidoylamidine hydrochloride 3b
3
was prepared using the same method as 3c from 2b (1.59 g, 6.9 mmol),
providing 1.50 g of solid 3b (5.7 mmol, 82% yield), mp (dec.) 198ꢀ200 °C.
4-ClC6H4C3N3H3F3 HCl, 3d. Imidoylamidine hydrochloride 3d
3
was prepared using the same method as 3c from 2d (2.40 g, 9.6 mmol),
providing 2.65 g of a white solid (9.3 mmol, 97% yield), mp (dec.) >
244 °C.
4-CF3C6H4C3N3H3F3 HCl, 3e. Imidoylamidine hydrochloride 3e
3
was prepared using the same method as 3c from 2e (2.37 g, 8.4 mmol),
providing 2.65 g of a white solid (8.3 mmol, 99% yield), mp (dec.)
181ꢀ183 °C.
Synthesis of 1-Chloro-5-aryl-3-trifluoromethyl-1,2,4,6-
thiatriazines 4aꢀe. C6H5C2N3SClCF3, 4c. Freshly distilled SCl2
(2.8 mL, 44.1 mmol) in 10 mL of acetonitrile was added dropwise to a
suspension of imidoylamidine HCl, 3c (2.24 g, 8.9 mmol), in 30 mL of
3
acetonitrile. The yellow mixture was heated to reflux for 1.5 h, giving an
orange-red color. After filtering to remove any solids, the solvent was
removed in vacuo to leave an orange oil (2.23 g, 8.0 mmol, 90% yield). 1H
NMR (δ, CDCl3): 7.52ꢀ7.58 (m, 2H, CArH), 7.67ꢀ7.74 (m, H, CArH),
8.45ꢀ8.48 (m, 2H, CArH).
4-CH3OC6H5C2N3SClCF3, 4a. Compound 4a was prepared using
the same method as 4c using SCl2 (1.6 mL, 25.1 mmol) in 10 mL of
acetonitrile added dropwise to 3a (1.45 g, 5.1 mmol) in 25 mL of
acetonitrile. The solvent was removed in vacuo, leaving a dark orange oil.
Recrystallization with hot toluene followed by hot acetonitrile left an
orange oil (1.29 g, 81% yield). 1H NMR (δ, CDCl3): 3.94 (s, CH3O),
7.02 (d, 2H, CArH), 8.45 (d, 2H, CArH).
Isolation of N-Chlorosulfonyl-N,N0-benzamidine, 6. Attem-
pts to grow crystals of 4 by slow cooling (ꢀ30 °C) in anhydrous CH3CN
were successful only for 4b and 4e; in the case of 4c, colorless crystals of a
hydrolysis product (6) were obtained, which could be identified as
N-chlorosulfonyl-N,N0-benzamidine by a single-crystal X-ray diffraction
structure determination. Further characterization of this material was
not undertaken.
4-CH3C6H5C2N3SClCF3, 4b. Compound 4b was prepared using the
same method as 4c using SCl2 (1.8 mL, 28.3 mmol) in 10 mL of
acetonitrile and 3b (1.47 g, 5.5 mmol) in 25 mL of acetonitrile. The
solvent was removed in vacuo, leaving an orange-red solid (1.77 g, 97%
yield). 1H NMR (δ, CDCl3): 2.49 (s, CH3), 7.35 (d, 2H, CArH), 8.37 (d,
2H, CArH).
4-ClC6H5C2N3SClCF3, 4d. Compound 4d was prepared using the
same method as 4c using SCl2 (2.2 mL, 34.6 mmol) in 10 mL of
acetonitrile and 3d (2.54 g, 8.9 mmol) in 40 mL of acetonitrile. The
solvent was removed in vacuo, leaving a dark orange oil. This was
’ ASSOCIATED CONTENT
S
Supporting Information. Supplemental crystal structure
b
information, a representative NMR spectrum of imidoylamidine,
5134
dx.doi.org/10.1021/ic2003996 |Inorg. Chem. 2011, 50, 5123–5136

Inorganic Chemistry
ARTICLE
infrared spectral data, tables of hydrogen bonds, graphs showing
correlation of hfs and solution redox potentials with Hammett
parameters, additional EPR spectra, data from Gaussian calculations,
and electronic crystallographic data in CIF format. This material is
available free of charge via the Internet at http://pubs.acs.org.
(3) (a) Boerꢀe, R. T.; Roemmele, T. L. Coord. Chem. Rev. 2000,
210, 369–445. (b) Boerꢀe, R. T.; Moock, K. H. J. Am. Chem. Soc. 1995,
117, 4755–4760. (c) Boerꢀe, R. T.; Moock, K. H.; Parvez, M. Z. Anorg.
Allg. Chem. 1994, 620, 1589–1598.
(4) Ang, C. Y.; Boerꢀe, R. T.; Goh, L. Y.; Koh, L. L.; Kuan, S. L.; Tan,
G. K.; Yu, X. Chem. Commun. 2006, 4735–4737.
(5) (a) Geevers, J.; Hackmann, T.; Trompen, W. P. J. Chem. Soc. C
1970, 875–878. (b) Kornuta, P. P.; Derii, L. I.; Romanenko, E. A. Chem.
Heterocycl. Compd. 1978, 273, 226. (c) Kornuta, P. P.; Derii, L. I.;
Markovskii, L. N. Zh. Org. Khim. 1980, 16, 1308–1313. (d) Markovskii,
L. N.; Kornuta, P. P.; Katchkovskaya, L. S.; Polumbrik, P. M. Sulfur Lett.
1983, 1, 143–145. (e) Hayes, P. J.; Oakley, R. T.; Cordes, A. W.;
Pennington, W. T. J. Am. Chem. Soc. 1985, 107, 1346–1351. (f) Cordes,
A. W.; Hayes, P. J.; Josephy, P. D.; Koenig, H.; Oakley, R. T.;
Pennington, W. T. J. Chem. Soc., Chem. Commun. 1984, 1021–1022.
(g) Cordes, A. W.; Craig, S. L.; Condren, M. S.; Oakley, R. T.; Reed,
R. W. Acta Crystallogr. 1986, C42, 922–923. (h) Ramakrishna, T. V. V.;
Elia, A. J.; Vij, A. Inorg. Chem. 1999, 38, 3022–3026. (i) Boerꢀe, R. T.;
Cordes, A. W.; Hayes, P. J.; Oakley, R. T.; Reed, R. W.; Pennington,
W. T. Inorg. Chem. 1986, 25, 2445–2450. (j) Knapp, C.; Watson, P. G.;
Lork, E.; Friese, D. H.; Mews, R.; Decken, A. Inorg. Chem. 2008,
47, 10618–10625. (k) Chen, S.-J.; Behrens, U.; Fischer, E.; Mews, R.;
Pauer, F.; Sheldrick, G. M.; Stalke, D.; Stohrer, W.-D. Chem. Ber. 1993,
126, 2601–2607. (l) Cordes, A. W.; Oakley, R. T. Acta Crystallogr. 1987,
C43, 1645–1646. (m) Farrar, J. M.; Patel, M. K.; Kaszynski, P.; Young,
V. G., Jr. J. Org. Chem. 2000, 65, 931–940. (n) Kaszynski, P. J. Phys.
Chem. A 2001, 105, 7615–7625. (o) Kaszynski, P. J. Phys. Chem. A 2001,
105, 7626–7633.
’ AUTHOR INFORMATION
Corresponding Author
*Tel.: (403) 329-2045. Fax.: (403)329-2057. E-mail: boere@
uleth.ca.
’ ACKNOWLEDGMENT
We thank the University of Lethbridge and the Natural
Sciences and Engineering Research Council of Canada for finan-
cial support of this work. Some preliminary groundwork for this
project was undertaken by Russell J. Goodman. We thank Meg
O’Shea for performing DFT calculations and WestGrid for access
to computational facilities and Dr. Gotthelf Wolmersh€auser for
determining RT X-ray structures of 2c and 5c.
’ REFERENCES
(1) (a) Chivers, T. A Guide to Chalcogen-Nitrogen Chemistry; World
Scientific Publishing Co.: Singapore, 2005. (b) Hicks, R. G. Stable
Radicals: Fundamentals and Applied Aspects of Odd-Electron Compounds;
Wiley: New York, 2010. (c) Torroba, T. J. Prakt. Chem. 1999,
341, 99–113. (d) Lekin, K.; Winter, S. M.; Downie, L. E.; Bao, X.; Tse,
J. S.; Desgreniers, S.; Secco, R. A.; Doube, P. A.; Oakley, R. T. J. Am. Chem.
Soc. 2010, 132, 16212–16224. (e) Winter, S. M.; Cvrkalj, K.; Dube, P. A.;
Robertson, C. M.; Probert, M. R.; Howard, J. A. K.; Oakley, R. T. Chem.
Commun. 2009, 7306–7308. (f) Fujita, W.; Awaga, K. Science 1999,
286, 261–262. (g) Leitch, A. A.; Oakley, R. T.; Reed, R. W.; Thompson,
L. K. Inorg. Chem. 2007, 46, 6261–6270. (h) Gilroy, J. B.; Joe, B.; Lemaire,
M. T.; Patrick, B. O.; Hicks, R. G. Cryst. Eng. Comm. 2009, 11,
2180–2184. (i) Decken, A.; Mailman, A.; Passmore, J. Chem. Commun.
2009, 6077–6079. (j) Vasilieva, N. V.; Irtegova, I. G.; Gritsan, N. P.;
Lonchakov, A. V.; Makarov, A. Y.; Shundrin, L. A.; Zibarev, A. V.
J. Phys. Org. Chem. 2010, 23, 536–543. (k) Semenov, N. A.; Pushkarevsky,
N. A.; Lonchakov, A. V.; Bogomyakov, A. S.; Pritchina, E. A.; Suturina,
E. A.; Gritsan, N. P.; Konchenko, S. N.; Mews, R.; Ovcharenko, V. I.;
Zibarev, A. V. Inorg. Chem. 2010, 49, 7558–7564. (l) Kanai, K.; Yoshida,
H.; Noda, Y.; Iwasaki, A.; Suizu, R.; Tsutumi, J.; Imabayashi, H.; Ouchi, Y.;
Sato, N.; Seki, K.; Awaga, K. Phys. Chem. Chem. Phys. 2009,
11, 11432–11436. (m) Cordes, A. W.; Haddon, R. C.; Oakley, R. T.
Adv. Mater. 1994, 6, 798–802. (n) Cordes, A. W.; Haddon, R. C.; Oakley,
R. T. Phosphorus, Sulfur, Silicon Relat. Elem. 2004, 179, 673–684. (o)
Barclay, T. M.; Cordes, A. W.; Haddon, R. C.; Itkis, M. E.; Oakley, R. T.;
Reed, R. W.; Zhang, H. J. Am. Chem. Soc. 1999, 121, 969–976. (p) Brusso,
J. L.; Clements, O. P.; Haddon, R. C.; Itkis, M. E.; Leitch, A. A.; Oakley,
R. T.; Reed, R. W.; Richardson, J. F. J. Am. Chem. Soc. 2004,
126, 8256–8265. (q) Alberola, A.; Less, R. J.; Pask, C. M.; Rawson,
J. M.; Palacio, F.; Oliete, P.; Paulsen, C.; Yamaguchi, A.; Farley, R. D.;
Murphy, D. M. Angew. Chem., Int. Ed. 2003, 42, 4782–4785. (r) Rawson,
J. M.; Alberola, A.; Whalley, A. J. Mater. Chem. 2006, 16, 2560–2565.
(s) Benin, V.; Kaszynski, P. J. Org. Chem. 2000, 65, 8086–8088.
(6) Oakley, R. T.; Reed, R. W.; Cordes, A. W.; Craig, S. L.; Graham,
J. B. J. Am. Chem. Soc. 1987, 109, 7745–7749.
(7) Boerꢀe, R. T.; Oakley, R. T.; Reed, R. W. J. Organomet. Chem.
1987, 331, 161–167.
(8) (a) Peak, D. A. J. Chem. Soc. 1952, 215–226.(b) Oto, K.;
Ichikawa, E. Patent Japan JP 48003811, 1973; Chem. Abstr. 1973, 79,
18443. (c) Liebscher, J.; Knoll, A.; Berger, A.; Krenzke, A. Patent East
Ger. DD 219,479, 1985; Chem. Abstr. 1985, 103, 195834. (d) Brown,
H. C.; Schuman, P. D. J. Org. Chem. 1963, 28, 1122–1127. (e) Schaeffer,
F. C.; Hechenbleikner, I.; Peters, G. A.; Wystrach, V. P. J. Am. Chem. Soc.
1958, 81, 1466–1470.
(9) Peters, G. A.; Schaeffer, F. C. Patent: Ger 1,175,684; Chem.
Abstr. 1964, 61, 14694c.
(10) Privett, A. J.; Craig, S. L.; Jeter, D. Y.; Cordes, A. W.; Oakley,
R. T.; Reed, R. W. Acta Crystallogr. 1987, C43, 2023–2025.
(11) (a) Kopylovich, M. N.; Pombeiro, A. J. L.; Fischer, A.; Kloo, L.;
Kukushkin, V. Y. Inorg. Chem. 2003, 42, 7239–7248. (b) Gushchin, P. V.;
Tyan, M. R.; Bokach, N. A.; Revenco, M. D.; Haukka, M.; Wang, M.-J.;
Lai, C.-H.; Chou, P.-T.; Kukushkin, V. Y. Inorg. Chem. 2008, 47, 11487–
11500. (c) Norrestam, R. Acta Crystallogr. 1984, C40, 955–957.
(d) Robinson, V.; Taylor, G. E.; Woodward, P.; Bruce, M. I.; Wallis,
R. C. J. Chem. Soc., Dalton Trans. 1981, 1169–1173. (e) Barker, J.; Kilner,
M.; Mahmoud, M. M.; Wallwork, S. C. J. Chem. Soc., Dalton Trans.
1989, 837–841. (f) Hursthouse, M. B.; Mazid, M. M.; Robinson, S. D.;
Sahajpal, A. J. Chem. Soc., Dalton Trans. 1994, 3615–3620. (i) Aris, D. R.;
Barker, J.; Phillips, P. R.; Alcock, N. W.; Wallbridge, M. G. H. J. Chem.
Soc., Dalton Trans. 1997, 909–910.
(12) Pinkerton, A. A.; Schwarzenbach, D. J. Chem. Soc., Dalton Trans.
1978, 989–996.
(13) Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements, 2nd
ed.; Reed Educational and Professional Publishing Ltd.: Oxford, U. K.,
1997; p 60.
(14) (a) Boerꢀe, R. T.; Moock, K. H.; Derrick, S.; Hoogerdijk, W.;
Preuss, K.; Yip, J. Can. J. Chem. 1993, 71, 473–486. (b) Boerꢀe, R. T.; Fait,
J.; Larsen, K.; Yip, J. Inorg. Chem. 1992, 31, 1417–1423.
(15) Cordes, A. W.; Oakley, R. T. Acta Crystallogr., Sect. C 1987,
C43, 1645–1646.
(2) (a) Preuss, K. E. Dalton Trans. 2007, 2357–2369. (b) Hearns,
N. G. R.; Preuss, K. E.; Richardson, J. F.; Bin-Salamon, S. J. Am. Chem.
Soc. 2004, 126, 9942–9943. (c) Awaga, K.; Tanaka, T.; Shirai, T.;
Fujimori, M.; Suzuki, Y.; Yoshikawa, H.; Fujita, W. Bull. Chem. Soc.
Jpn. 2006, 79, 25–34. (d) Jennings, M.; Preuss, K. E.; Wu, J. Chem.
Commun. 2006, 341–343. (e) Britten, J.; Hearns, N. G. R.; Preuss, K. E.;
Richardson, J. F.; Bin-Salamon, S. Inorg. Chem. 2007, 46, 3934–3945.
(f) Hearns, N. G. R.; Clꢀerac, R.; Jennings, M.; Preuss, K. E. Dalton Trans.
2009, 3193–3203. (g) Hearns, N. G. R.; Fatila, E. M.; Clꢀerac, R.;
Jennings, M.; Preuss, K. E. Inorg. Chem. 2008, 47, 10330–10431.
(16) Boerꢀe, R. T. Unpublished data.
(17) CRC Handbook of Chemistry and Physics, 70th ed.; Weast, R. C.,
Ed.; CRC Press: Boca Raton, FL, 1989; p Fꢀ188.
5135
dx.doi.org/10.1021/ic2003996 |Inorg. Chem. 2011, 50, 5123–5136

Inorganic Chemistry
ARTICLE
(18) 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.; Barone, 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.; Foresman, J. B.;
Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.;
Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle,
E. S.; Pople, J. A. Gaussian 98, Revision A.7; Gaussian Inc.: Pittsburgh,
PA, 1998.
(19) Duling, D. R. J. Magn. Reson., Ser. B 1994, 104, 105–110.
(20) (a) Boerꢀe, R. T.; Oakley, R. T.; Reed, R. W.; Westwood, N. P. C.
J. Am. Chem. Soc. 1989, 111, 1180–1185. (b) Boerꢀe, R. T.; Roemmele,
T. L. Phosphorus, Sulfur, Silicon 2004, 179, 875–882.
(21) Moock, K. H.; Rock, M. H. J. Chem. Soc., Dalton Trans. 1993,
2459–2463.
(22) (a) Lau, H. F.; Ang, P. C. Y.; Ng, V. W. L.; Kuan, S. L.; Goh,
L. Y.; Borisov, A. S.; Hazendonk, P.; Roemmele, T. L.; Boerꢀe, R. T.;
Webster, R. D. Inorg. Chem. 2008, 47, 632–644. (b) Beekman, R. A.;
Boerꢀe, R. T.; Moock, K. H.; Parvez, M. Can. J. Chem. 1998, 76, 85–93.
(23) Oakley, R. T. Prog. Inorg. Chem. 1988, 36, 299–391.
(24) (a) Rudolph, M. J. Electroanal. Chem. 2003, 543, 23.
(b) Rudolph, M. J. Electroanal. Chem. 2004, 571, 289. (c) Rudolph,
M. J. Electroanal. Chem. 2003, 558, 171. (d) Rudolph, M. J. Comput.
Chem. 2005, 26, 619. (e) Rudolph, M. J. Comput. Chem. 2005, 26, 1193.
(25) (a) Evans, D. H. Chem. Rev. 1990, 90, 739–751. (b) Evans,
D. H. Chem. Rev. 2008, 108, 2113–2144.
(26) Izutsu, K. Electrochemistry in Nonaqueous Solutions, 2nd ed.;
Wiley-VCH: Weinheim, Germany, 2009; p 100ff.
(27) Boerꢀe, R. T.; Bond, A. M.; Chivers, T.; Feldberg, S. W.;
Roemmele, T. L. Inorg. Chem. 2007, 46, 5569–5607.
(28) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, A64, 112–122.
5136
dx.doi.org/10.1021/ic2003996 |Inorg. Chem. 2011, 50, 5123–5136