Permethylated salts and radicals derived from azo peri-naphthalenes

Article abstract of DOI:10.1002/cplu.201200021

By careful choice of reaction conditions 1-methylnaphtho[1,8-de]-1,2,3- triazine and 1,3-dimethylnaphtho[1,8-de]-1,2,3-triazinium triflate are readily prepared by the alkylation of 1Hnaphtho[1,8-de]-1,2,3-triazine. The electronic spectra for these bathochromically shifted dyes are reported for a range of solvents as are their vibrational spectra and theoretical structures from density functional theory, B3LYP/6-311++G**. The 15N substitution into the middle nitrogen atom allows for the identification of strong azine bands in the IR and Raman spectra as well as in the 15N NMR spectrum. X-ray crystallography is used to characterize the solid-state structures of two of these new triazine derivatives and these structures are used to benchmark the theoretical and spectroscopic discussions. The corresponding 1,3-dimethyltriazinium radical is prepared and characterized by cyclic voltammetry, spectroelectrochemistry, and EPR spectroscopy. Unlike the related transient 1,2,4-triazene radicals, the 1,3-dimethyl-1,2,3-naphthotriazyl radical is markedly more stable.

Full text of DOI:10.1002/cplu.201200021

DOI: 10.1002/cplu.201200021
Permethylated Salts and Radicals Derived from
Azo peri-Naphthalenes
D. Scott Bohle,*[a] Zhijie Chua,[a] Timothy Johnstone,[a] Andrey G. Moiseev,[a, b]
Inna Perepichka,[a] and Kristopher A. Rosadiuk[a]
By careful choice of reaction conditions 1-methylnaphtho[1,8-
de]-1,2,3-triazine and 1,3-dimethylnaphtho[1,8-de]-1,2,3-triazini-
um triflate are readily prepared by the alkylation of 1H-
naphtho[1,8-de]-1,2,3-triazine. The electronic spectra for these
bathochromically shifted dyes are reported for a range of sol-
vents as are their vibrational spectra and theoretical structures
from density functional theory, B3LYP/6-311+ +G**. The
15N substitution into the middle nitrogen atom allows for the
identification of strong azine bands in the IR and Raman spec-
tra as well as in the 15N NMR spectrum. X-ray crystallography is
used to characterize the solid-state structures of two of these
new triazine derivatives and these structures are used to
benchmark the theoretical and spectroscopic discussions. The
corresponding 1,3-dimethyltriazinium radical is prepared and
characterized by cyclic voltammetry, spectroelectrochemistry,
and EPR spectroscopy. Unlike the related transient 1,2,4-tria-
zene radicals, the 1,3-dimethyl-1,2,3-naphthotriazyl radical is
markedly more stable.
Introduction
The 1,3-substituted triazoles and triazines, A and B (Scheme 1),
are potential nitrogen analogues of N-heterocyclic carbenes.
Although many elemental analogues where the central nitro-
gen atom has been replaced with carbon, phosphorus, or ar-
senic atoms are known,[1–3] the chemistry of the cationic triazo-
nium and dicationic triazinium derivatives remains compara-
tively undeveloped.[4–8] This scarcity is surprising in light of the
burgeoning triazole chemistry following the popularization of
Huisgen’s 1,3-dipolar cycloaddition of azides to alkynes. Closely
related to A and B are the ring annealed peri-naphthotriazines
C, which are particularly attractive owing the single positive
charge and small HOMO–LUMO gap.[9,10] Greater solubility and
variable colors are the expected beneficial features of C. Of the
class of compounds with a core structure C, one species, 1,3-
dimethyl-naphtho[1,8-de]-1,2,3,-triazinium, has been alluded to
twice in the literature.[11,12] In one of these papers, a brief com-
munication, the referenced details of its preparation have not
been followed up in subsequent papers in the primary litera-
ture. A dibenzyl analogue of C was also outlined and charac-
tential utility as bathochromically shifted dyestuffs.[13–15] This
suggests that not only is there a small HOMO–LUMO gap in
these compounds but that the low-lying LUMO in B or C make
these useful analogues to N-heterocyclic carbenes. Given the
promising, but limited, physical information for the peri-
naphtho[1,8-de]triazines we sought to characterize these nitro-
gen analogues of the N-heterocyclic carbenes. To this end we
report: 1) synthetic and characteristic data on four of the
protio and methylated peri-naphtho[1,8-de]triazines, 2) two
new single-crystal X-ray structures of these derivatives, 3) vibra-
tional and electronic spectroscopy assignments for the low
energy transitions, 4) DFT calculations to support these analy-
ses, 5) the reversible reduction of the 1,3-dimethyl(peri-
naphtho[1,8-de]triazine) as found by cyclic voltammetry meas-
urements, and 6) characterization of the 1,3-dimethyl(peri-
naphtho[1,8-de]triazinyl) radical by spectroelectrochemistry
and EPR spectroscopy.
1
terized as a red oil by H and 13C NMR spectroscopy.[12] Taken
together, this class of compounds is poorly characterized.
An important and aesthetically pleasing aspect of the
chemistry of peri-naphtho[1,8-de]triazines is their intense and
variable colors and there has been some interest in their po-
[a] Dr. D. S. Bohle, Z. Chua, T. Johnstone, Dr. A. G. Moiseev, Dr. I. Perepichka,
K. A. Rosadiuk
Department of Chemistry, McGill University
801 Sherbrooke St. West
Montreal, Quebec, H3A 0B8 (Canada)
Fax: (+1)514-398-3797
E-mail: Scott.Bohle@Mcgill.ca
[b] Dr. A. G. Moiseev
Center for Self-Assembly Chemical Structures (CSACS), McGill University
Otto Maass Bldng Rm. 414, 801 Sherbrooke St. West
Montreal, Quebec, H3A 0B8 (Canada)
Scheme 1. General scheme of cationic triazonium and dicationic triazinium
compounds.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cplu.201200021.
ChemPlusChem 2012, 77, 387 – 395
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
387

D. S. Bohle et al.
Results and Discussion
A useful synthetic approach to 1H-naphtho[1,8-de]-
1,2,3,-triazine, 1, is from a facile diazotization of 1,8-
diaminonaphthalene with nitrous acid.[16,17] The brick
red, crude solid precipitate of 1 was readily purified
by recrystallization from 1,4-dioxane and was ob-
Scheme 4. Methylation of 1-methylnaphtho[1,8-de]-1,2,3,-triazine using dimethylsulfate.
tained in good yields of approximately 96% if freshly distilled
1,8-diaminonaphthalene was used[18] or 55–78% if the com-
mercial reagent was employed. It is well established that when
the conjugate base of 1H-naphtho[1,8-de]-1,2,3,-triazine is alky-
and toluene. Significantly, no alkylation at the second position
is found in the products of these dimethylation reactions nor
have we found it possible to methylate 3 to give the 1,3- or
1,2-dimethyltriazinium salt. Waterman noted the alkylating abil-
ity of 5 is comparable with diazomethane.[12] We have utilized
this observation to convert the product mixture from the reac-
tion shown in Scheme 3, which is initially a mixture of the 1-
methyl isomer 2 and the 1,3-dimethyl isomer 4, into 2 exclu-
sively by treating it with base to give an overall yield
of 60%. The characteristic data for 1–5 collected in
the experimental section is in accord with the limited
and scattered information available in the literature.
lated with dimethyl sulfate
a
mixture of 1- and 2-
and results
methylnaphtho[1,8-de]-1,2,3,-triazine,
2
3
(Scheme 2).[18–22] Although the return of heterocycles from this
methylation is good (70%), the regioselectivity is poor and the
As a first step in developing the chemistry of 1–5,
and to ensure that our theoretical models match the
observed ground state geometries, we have structur-
ally characterized 2 and 5 by single-crystal X-ray anal-
ysis. The asymmetric unit of 2 has two independent
molecules, 2A and 2B. Representations of the final
models are shown in Figure 1, with key experimental and theo-
retical metric parameters for 2 and 4 collected in Table 1 and
contrasted with the known data for 3.[23,24] Complete crystallo-
graphic details are found in Tables S1–S12 in the Supporting
Scheme 2. Synthesis of 1- and 2-methylnaphtho[1,8-de]-1,2,3,-triazines.
mixture is close to 1:1 in the two isomers. The two isomers 2
and 3 are however readily separated and purified by column
chromatography on silica gel. A statistical distribution would
of course give a 2:1 ratio of 2 and 3 but the observed, almost
identical distribution, of methylation products suggests that
devising a preparation for exclusive methylation at the first po-
sition might be challenging. We have found that methylation
of 1 with methyltriflate in the absence of base gave a product
devoid of 3 (Scheme 3). Instead of alkylation at the middle ni-
trogen atom to give 3, treatment with methyltriflate gave
Scheme 3. Methylation of 1H-naphtho[1,8-de]-1,2,3,-triazine using methyltri-
flate.
products corresponding to single methylation at the first posi-
tion, to give 2, and to the product of its subsequent methyla-
tion, 1,3-dimethyl-1H-naphtho[1,8-de]-1,2,3,-triazinium, 4, which
was isolated as its triflate salt. A related triazenium salt, 5, is re-
ported to form when 2 is alkylated with dimethylsulfate in
benzene at reflux.[11] An alternative preparation has been de-
scribed by Waterman, in dichloromethane at room tempera-
ture that gives a mixture of salts corresponding to 5 and 6
(Scheme 4).[12] Regardless of their anion, the 1,3-dimethyl salts
4, 5, and 6 are darkly colored, air-stable solids that give rise to
a distinctly purple solution and have good solubility in alcohols
and polar halosolvents but have limited solubility in benzene
Figure 1. Representations of single-crystal X-ray diffraction results for 2, 3,
and the cation in 5. On the left the two independent molecules (A and B) in
2 are shown, and on the right the known[23,24] structure of 3 is added for
contrast. The structure of the cationic 1,3-dimethyl derivative 5 corresponds
to its methylsulfate salt, is shown below. Displacement ellipsoids are drawn
at the 40% probability level and hydrogen atoms are shown as spheres with
a radius of 0.05 ꢁ.
388
www.chempluschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 77, 387 – 395

Triazene Radicals
Information. A key feature of all of these structures, both ex-
perimental and theoretical, is the planarity of the peri-triazine
rings. Of particular note are the trends in the NꢀN and CꢀN
bond lengths within these rings, and this trend is graphically
displayed in Figure S7, where NꢀN and CꢀN for each ring-
bound nitrogen atom is contrasted. There is a modest inverse
trend in bond length where short NꢀN interaction correlates
with longer CꢀN bond lengths. Caution needs to be used in in-
terpreting some of this data as, for example, no estimated
standard deviation (esd) values were listed in the original pub-
lications[23,24] for the metric parameters of 3, and for the pur-
poses of Figure S7, a large esd value of 0.005 ꢁ is assigned.
Nevertheless, Figure S7 clearly supports delocalized bonding in
2, 3, and 5 and we note that the two NꢀN bond lengths in
2A, 1.351(4) and 1.256(4) ꢁ are consistent with considerable
p bonding in the azine ring. These trends are largely well mod-
eled theoretically at the density functional level, with the
single exception being the unexpectedly long CꢀN bond
lengths calculated for the cation in 5, even though the calcu-
lated NꢀN bond length is quite similar to those found experi-
mentally. Some of this variation may be due to the calculated
structure for 4 that corresponds to a gas phase structure in
shown in Figure S2 the anion packs over the triazine ring and
juxtaposes two of the sulfate oxygen atoms in a 1,3-fashion
over N(1) and N(2). The resulting packing is far from symmetric
with the two closest nitrogen–oxygen separations differing by
0.374(5) ꢁ. These solid-state interactions suggest that these
colorful cations may be potentially useful for applications in
spectrophotometric anion sensing of anions in solutions.
Finally, the 132.58 N-N-N bond angle in 3 is significantly en-
larged compared to the other related structures. For example,
the structurally characterized carbene D, with R=iPr, has
longer CꢀN bond lengths, by 0.08 ꢁ, to the carbene carbon
atom than the related NꢀN bonds in 5 but similar CnaphthaleneꢀN
bond lengths and a similar, albeit 48 smaller, N-C-N angle.[25]
Similar trends in shorter NꢀE bond
lengths and planarity are shared
with the related phosphorus and
arsenic peri-naphthalene deriva-
tives described by Richeson and
co-workers.[1]
One of the particularly striking
features of the electronic spectra
of 1–3 is the width and the vibra-
ꢀ
the absence of a counter anion. In 5, the MeOSO3 counter ion
has only weak interactions with the cation, with the shortest
contacts being between N(2) and O(3) of 2.987(2)(4) ꢁ. As
tional progression in their visible bands. An image of these col-
orful species is shown in Figure S1. To aid in determining the
origin of these bands we have calculated the gas phase
normal vibrational modes for 1,
2, 3, and 4 (Tables S13–S24). For
all of these triazine compounds
Table 1. Experimental and theoretical bond lengths/bond angles.[a]
many of the strongly active IR
Bonds
Bond lengths
for 2 A/B[b] [ꢁ]
Bond lengths
for 3 [ꢁ]
Bond lengths
for 5 [ꢁ]
bands are associated with the
triazine ring. For example, the
symmetrical 2-methyl isomer 3 is
predicted to have a pair of in-
tense bands between 1390 and
1360 cmꢀ1, with an asymmetric
NꢀNꢀN stretch at 1387.4 cmꢀ1
and a symmetric CꢀN stretch at
1363.8 cmꢀ1. Additional strong
bands are calculated for 3 at
circa 1099.3 (asymmetric ring
breathing), 1061.9 (symmetric
ring breathing), 826.7 (out of
plane ring deformation), 775.4
(out of plane CꢀH wag), and
755.7 cmꢀ1 (symmetric ring com-
pression). These are of course
only a part of the 63 normal
modes—the majority of which
are primarily associated with the
hydrocarbon backbone. To help
assign these triazine-dependent
bands we prepared the (2N) 15N-
substituted isotopomer of 1 by
treating 1,8-diaminonaphthalene
with H15NO2. Methylation using
the conditions outlined in
Scheme 2 and Scheme 3 and
N(1)ꢀN(2)/N(4)ꢀN(5)
1.351(4)/1.341(3)
1.36550
1.256(4)/1.267(4)
1.25954
1.444(4)/1.444(4)
1.45253
1.313
1.30021
1.317
1.288(2)
1.28999
1.283(2)
N(2)ꢀN(3)/N(5)ꢀN(6)
N(1)ꢀC(11)/N(4)ꢀC(22)/N(3)ꢀC(12)
1.494
1.47700
1.470(3)
1.466(2)
1.47251
1.410(3)
1.42469
1.409(2)
N(1)ꢀC(1)/N(4)ꢀC(12)
N(3)ꢀC(9)/N(6)ꢀC(20)
1.393(3)/1.398(3)
1.39606
1.409(4)/1.403(4)
1.40734
1.395
1.39113
1.400
Angles
Bond angles
for 2 A/B[b] [8]
Bond angles
for 3 [8]
Bond angles
for 5 [8]
N(1)-N(2)-N(3)/N(4)-N(5)-N(6)
C(1)-N(1)-N(2)/C(12)-N(4)-N(5)
C(9)-N(3)-N(2)/C(20)-N(6)-N(5)
N(1)-C(1)-C(10)/N(4)-C(12)-C(21)
N(3)-C(9)-C(10)/N(6)-C(20)-C(21)
C(1)-C(10)-C(9)/C(20)-C(21)-C(12)
121.2(3)/121.4(2)
121.25788
124.1(3)/123.8(3)
123.99985
120.6(3)/120.1(3)
120.71232
114.8(3)/115.2(3)
115.26151
119.5(3)/120.4(3)
120.35724
119.8(3)/119.1(3)
118.41122
132.5
131.35442
114.3
115.71686
114.2
119.56(16)
120.38765
124.45(16)
124.16764
124.74(15)
120.1
119.57788
120.6
115.68(16)
115.53536
115.50(16)
118.1
118.05585
120.04(18)
120.20640
[a] Experimental values determined by X-ray diffraction as reported here for 2 and 5 and reported in the litera-
ture for 3.[23,24] The theoretical results shown in italics are from density functional theory with triple zeta-level
basis sets, B3LYP/6-311+ +G**. [b] Both values for the independent molecules in 2 are listed together, with
that for molecule 2 A listed first. No esd values for the metrical parameters for the structures of 3 are avail-
able.[23,24]
ChemPlusChem 2012, 77, 387 – 395
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
389

D. S. Bohle et al.
separation gave (2N) 15N-substituted derivatives of 2, 3, and 4.
Their IR spectra were measured and subtracted from the
parent 14N derivatives, to give difference spectra such as that
shown in Figure 2 for 1. The results of similar difference spec-
tra are collected in Table 2 for 1, 2, 3, and 5 and contrasted
with the theoretical values predicted from the DFT calculations.
heterogeneity in the labeling experiment. In some cases, long-
distance H–15N coupling is observed in the H NMR spectrum
and these coupling constants are collected in the Experimental
Section.
1
1
In Table 3 we contrast the solvatochromism for 1, 2, 3, and 5
and include, in the last line, is the average peak to peak sepa-
rations for their vibrational progressions. Of the neutral species
the 1-methyl derivative 2 shows the largest solvent depend-
Table 3. Solution phase lmax values (nm) for the peri-methylnaphthaltria-
zines.
Solvent
1
2
3
5
acetic acid
ethanol
DMSO
acetonitrile
acetone
diethyl ether
pyridine
CH2Cl2
chloroform
ethyl acetate
455
451
446
441
439
insoluble
446
441
441
438
453
451
446
441
439
433
439
441
442
436
559
558
555
554
554
557
557
558
562
556
485
501
496
501
496
insoluble
508
503
501
insoluble
Solvent
CH2Cl2
Average peak to peak vibronic separations D [cmꢀ1
1358 1357 1416 none
]
Figure 2. Difference in IR spectra (KBr) for 1 with the middle nitrogen atom
labeled with 15N. Peaks shifted from the 14N sample are down and those that
grow in for the 15N are shown as upward deviations.
15N NMR spectroscopy is a powerful method for the charac-
terization of nitrogen-containing heterocycles and the pres-
ence of 15N at the middle position allowed for the determina-
tion of their chemical shifts, which encompass a surprisingly
large range for 2, 3, and 4 of +28.47, ꢀ92.9, and +10.99 ppm,
respectively. These are given in ppm with respect to
[15N]nitromethane as the reference. In each case only a single
sharp peak is observed in the 15N NMR spectra and there is no
indication of nitrogen scrambling in these spectra or any other
ence with large bathochromic shifts in acid and polar solvents.
For the more symmetrical 2-methyl derivative 3 the depend-
ence is smallest with the lowest energy band being in chloro-
form. Based on photoelectron spectroscopy Bartetzko and
!
Gleiter[26] assigned this band in 3 as a p* p transition. From
the observed vibronic progression our results support this as-
signment as the peak to peak energy separations correspond
well to the calculated and observed n(N3)as bands listed in
Table 2.
In addition to the marked sol-
vatochromism, the strong, vibra-
tionally split, and solvent depen-
Table 2. Triazine IR active modes in the peri-napthalene derivatives.[a]
dent spectra of 1 are also pH de-
pendent (Scheme 5). The spectra
Assigned mode(s)
IR bands
for 1 [cmꢀ1
IR bands
for 2 [cmꢀ1
IR bands
for 3 [cmꢀ1
IR bands
for 4 [cmꢀ1
]
of 1 in acid and base, shown in
Figure S4 and S5, are character-
ized by bathochromic shifts for
both the protonated and depro-
tonated derivatives, each with
a distinctive vibronic progres-
sion, Table 4. Significantly, the vi-
bronic progression for 7 is more
distinctive and well resolved
than for the corresponding di-
methyl cation for which the
peak resembles a single Gaussi-
an curve. When performed in
[D6]ethanol and monitored by
1H NMR both 7 and 8 are more
symmetric, Figure S5, which sup-
]
]
]
n(NH)/n(N3)as
–
–
–
1537.3 (9)
1562.66 (8.42)
1519.64 (2.70)
1461 (3), 1488 (14)[b]
1486.13 (4.35)
1590.21 (15.83)
1569.13 (8.26)
n(N3)as
1356.6 (11.2)
1397.60 (4.85)
1393.5 (8.4)
1394.60 (1.00)
1360 (19)
1396.10 (9.59)
1387.41 (11.17)
1355.91 (9.82)
1162 (13)
1155.57 (3.92)
819 (9)
1010.51 (2.30)
647 (5)
659.66 (4.20)
1436.85 (4.25)
n(N3)s
1124.9 (4.9)
1077.52 (4.69)
892.2 (8.7)
893.30 (11.32)
657.5 (2.6)
1298.9 (2.7)
1325.40 (1.20)
926.8 (9.7)
926.40 (11.04)
766.6 (0.1)
1393.0 (7.0)
1393.53 (5.80)
–
941.56 (4.85)
702 (4.0)
p(N3)
1w(CHN3)
768.83 (10.10)
769.40 (0.50)
602.62 (2.15)
[a] Values in bold type are experimental values determined in KBr with the isotopic shift (2N 15N) is given in
cmꢀ1 within the brackets. Corresponding theoretical values are listed in italics underneath along with their as-
signed mode(s) in the left hand column. In cases where there are several closely spaced strong bands all are
listed. [b] Strong peak overlap prevents precise determination of specific isotope shifts.
390
www.chempluschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 77, 387 – 395

Triazene Radicals
Scheme 5. Protonation and deprotonation of 1H-naphtho[1,8-de]-1,2,3,-tria-
zine.
Table 4. Electronic spectra for protonated (7) and deprotonated (8) deriv-
atives of 1.
Vibronically split visible band
7
1
8
lmax [nm]
480
1326
451
1358
475
1364
Peak to peak separation [cmꢀ1 ꢂ5 cmꢀ1
]
Figure 3. HOMO–LUMO orbitals (B3LYP/6-311+ +G**) for 2, 3, and the
cation in 4–6.
ports the proposed 1,3-diprotonation for 7 and the naked tria-
zine structure for 8, but both assignments are only within the
time-averaged exchange limit of the H NMR experiment.
bonding for 3–6. In the case of 2 the HOMO is partially N=N
bonding and NꢀN antibonding. Otherwise the localization and
symmetry of the HOMOs for all three species is very similar.
Filled orbitals with in-plane s symmetry and strong contribu-
tion from lone pair electron on the nitrogen atom are relatively
low lying.
1
Electrophilic addition and ring closure to 1,8-diaminonaph-
thalene allows for facile access to a range of unusual electron-
rich frameworks with a well-defined and rigid geometry confer-
red by the peri-naphthalene substitution. The nitration of this
framework led to the early isolation of 1H-naphtho[1,8-
de]1,2,3-triazine as one of the first 1,2,3-triazines to be pre-
pared.[27,28] In general, these triazines have good water and
oxygen stability and are readily derivatized with a range of
electrophiles.[29] Although poor solubility is a potential practical
pitfall with this type of system, for the derivatives here, even
the salts, it is not an issue. All of the compounds, 1–6, exhibit
good solubility in a range of solvents. Given these general
characteristics, their ease of synthesis, and their remarkable
electronic properties, it is surprising how sparse the naphtho-
triazine literature is.
For the LUMO orbitals the differences between the Cs and
C2v point groups is particularly striking in the polarization of
the LUMO onto the naphthalene ring opposed to the methyl
group. The nodal distribution or the four planes in the LUMO
confers NꢀN antibonding but CꢀN bonding character to this
orbital, resulting in an arrangement that resembles a dog’s
paw in the case of 3–6. Overall there is a strong contribution
from the triazine ring to the LUMO and this conclusion also fits
with the photoelectron spectrum of 3.[30] The visible transition
!
in these compounds is then assigned to a p* p band and
one perspective on the origin of this transition and the elec-
tronic structure of the 1,8-naphthotriazines is that they can be
viewed as a dialkylated azide annulated to the naphthalene
ring with perhaps minimal p interaction. Here the critical
energy gap is from the naphthalene ring as HOMO to the p*
of the RN3 or [RNNNR]+ fragments. Certainly few other tria-
zines have these intense low-energy transitions so it is not an
inherent property of the 1,2,3-triazine heterocycle.
Perhaps the most remarkable spectroscopic feature of these
triazines is the extended vibronic structure associated with the
intense visible band of the chromophore. The energy of this
transition is very sensitive to the nitrogen atom that is substi-
tuted, and less so to the nature of the substituent itself. Gener-
ation of a cationic triazine by dialkylation or diprotonation
leads to small shifts in the energy of these transitions, with
perturbation of their vibronic progression in the case of dialky-
lation. The use of 15N incorporation into the middle position of
the triazine ring has allowed for the assignment of some of
the vibrational bands associated with this chromophore. The
general weak solvatochromism, even in strong hydrogen-
bonding solvents, suggests little involvement of nitrogen lone
Given the rich electronic structures of the methylated naph-
thotriazines their electrochemistry was explored using cyclic
voltammetry. Of particular note is that the reduction of the
cation in 4 is reversible (ipc/ipa ꢁ1) down to scan rates of
50 mVsꢀ1, Table S25. Moreover, this lineshape (Figure 4) corre-
sponds to a Nerstian one-electron process in which current
flow has a linear dependence upon the square root of the
scan rate, Figure S8 and Table S25. The reversible one-electron
reduction of the cations 4 and 5 (Scheme 6), is expected to
give a 1,2,3-triazinyl radical 9, the closest reported relative of
which are the transient 1,2,4-triazinyl radicals related to pyrimi-
dine-base reduction in DNA.[31–33] Only irreversible oxidation is
observed in the CV electrochemistry of the monomethylderiva-
tives 2 and 3, Figure S9. Nitrogen-based radicals that have an-
nealed aromatic rings are often persistent and exist in facile
!
pair electrons in these transitions that are p* p. In part, the
basis of this assignment is from the frontier orbitals for 2–6
that are depicted in Figure 3. The point group symmetry of the
molecules, ignoring the geometry of the methyl hydrogen
atoms, is either Cs or C2v and all sets of frontier orbitals have
off-plane p symmetry. In addition, for the HOMO the common
motif has three nodal planes orthogonal to the molecular
plane with this orbital being NꢀN nonbonding and CꢀN anti-
ChemPlusChem 2012, 77, 387 – 395
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
391

D. S. Bohle et al.
Figure 4. Cyclic voltammogram of 4 in 0.1m nBu4NPF6 in acetonitrile at
a scan rate of 50 mVsꢀ1
.
Figure 5. UV/Vis spectra during electrochemical reduction of 4 in 0.1m
nBu4NPF6 in acetonitrile.
Scheme 6. One-electron reduction of compound 4.
equilibrium with their NꢀN dimers.[34] Although pyridyl and
pyrazolyl radicals are well-precedented, triazinyl radicals such
as 9 have not been characterized to any extent in the litera-
ture. Given the reversibility of the observed cyclic voltammetry
we sought to further spectroscopically characterize this unusu-
al species.
Scheme 7. Calculated (UB3LYP) structure of radical 9 with Barone’s EPR-II
(double-zeta) and EPR-III (triple-zeta) key metric parameters shown in
normal and italic font, respectively.
When the radical 9 is generated at a platinum gauze spec-
troelectochemistry cell by the reduction of the cation 4 in ace-
tonitrile the strong p!p* bands are rapidly replaced by less
intense and slightly lower energy bands centered at 465 nm
(Figure 5). As part of our characterization of the radical we
have optimized its structure using density functional theory,
and calculated its EPR spin coupling parameters as well its
electronic spectra. The optimized structure of the radical is
shown in Scheme 7, and the corresponding EPR and UV/Vis
data are collected in Table 5 where the experimental data are
contrasted with those calculated at three levels of theory.
Reduction of 4 with sodium/potassium amalgam at ambient
conditions leads to formation of paramagnetic transient spe-
cies (Scheme 6). Based on its EPR spectrum (ge ꢁ2.00602) it is
assigned as the nitrogen-centered radical 9. The lifetime of
radical 9 is several minutes at room temperature, but its spec-
trum gradually decreases in intensity with time. This lifetime is
however much longer than the nitrene radicals reported by
Robert et al. whose half-lives are in the microsecond realm.[35]
To increase the lifetime of radical 9, EPR experiments were
run at low temperature (203 K). The hyperfine coupling pattern
of 9 (Figure 6) consists of one 14N1 nucleus (I=1) and the two
equivalent b 14N2-nuclei (I=1) with intensity distribution about
1:1:1 spaced by 15 G and about 1:2:3:2:1 spaced by 2.7 G, re-
spectively. Splitting by six b-protons (I=1/2, presumably
ꢁ0.5 G) of the two methyl groups could not be resolved
under these conditions. Splitting of 14N1 (15 G) in 9 is close to
the value reported for similar splitting in 1-pyrrolidinil
(14.3 G).[36] The simulated EPR spectrum of 9 is close to the ex-
perimental spectrum seen in Figure S10.
Conclusion
The facile synthesis, stability, solubility, and chromophoric
properties of the azo peri-naphthalenes make them promising
candidates for a range of material applications. Among the
possibilities are their uses as new spectrophotometric anion
sensors, photochromic or di-
chromic dyes, and possibly for
Table 5. Theoretical[a] and experimental results for 1,3-dimethyl-1,2,3-triazenyl radical 9.
new donor/acceptor ensembles.
Some of the important new re-
sults described here are the im-
proved alkylation conditions for
1H-naphtho[1-8-de]triazine with
methyltriflate to give either 1-
methylnaphthotriazine or 1,3-di-
Isotropic Fermi contact
Transition energy for first three bands
Method/basis
an[2] [gauss]
an[1,3] [gauss]
1st [nm][b]
2nd [nm][b]
3rd [nm][b]
B3LYP/EPR-II[37]
B3LYP/EPR-III
B3LYP/6-311+ +G**
Experiment
10.25
4.08
8.35
15.0
1.768
0.37
1.30
2.7
824.7 (0.0000)
615.7 (0.0011)
571.6 (0.0012)
470
572.0 (0.0198)
496.7 (0.003)
478.0 (0.0000)
–
483.7 (0.0000)
481.5 (0.0000)
467.4 (0.0002)
–
[a] Unrestricted wavefunctions, UB3LYP, used in each case. [b] Oscillator strengthen is given within the brack-
ets.
methylnaphthotriazinium
de-
pending upon the reaction con-
392
www.chempluschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 77, 387 – 395

Triazene Radicals
1H), 7.25–7.27 (m, 2H); 13C NMR (75 MHz, [D6]DMSO): d=99.1,
114.5, 118.7, 118.8, 123.9, 129.5, 129.7, 133.4, 133.9, 139.5; IR (KBr):
3246w, 3214w, 3173w, 3137w, 3053w, 3011w, 2967w, 2936w,
2683w, 2485w, 2457w, 2368w, 1646w, 1604w, 1575w, 1356s,
1284s, 1186s, 1158m, 1125s, 1087m, 892m, 816s, 758s cmꢀ1
;
Raman (solid, only strong bands are listed): 1356, 1275, 1065, 627,
493 cmꢀ1; UV (DMSO, lmax (e)): 445 nm (375 mꢀ1 cmꢀ1).
Methylation of 1H-naphtho[1,8-de]-1,2,3,-triazine with di-
methyl sulfate
1-methylnaphtho[1,8-de]-1,2,3,-triazine (2) and 2-methyl-
naphtho[1,8-de]-1,2,3,-triazine (3): Slight modifications to the syn-
thesis described by Perkins[22] were made. Sodium hydroxide
(0.564 g, 14 mmol) was dissolved in warm MeOH (10 mL). 1H-
naphtho[1,8-de]triazine (1; 1.40 g, 8.4 mmol) was then dissolved in
this alcoholic alkaline solution, to give a deep red color. Dimethyl
sulfate (1.27 g, 10 mmol) was added dropwise to the solution over
10 min with vigorous stirring after which the solution was heated
at reflux for an additional 10 min. The reaction mixture was cooled
to RT, extracted with CH2Cl2 (100 mL) and dried over MgSO4. After
removal of the solvent, the crude product was purified by column
chromatography on silica gel with petroleum ether/ CH2Cl2 (1:1) as
the eluent.
Figure 6. EPR spectrum of 9 prepared by the reduction of 4 with sodium/
potassium amalgam in THF at T=203 K.
ditions. In addition, the contrast of the derivatives with the
known properties of 2-methylnaphthotrizine allows for the as-
!
signment of a p* p transition to the visible bands in these
colorful bathochromically shifted dyes. A key feature of these
derivatives is possibly the direct annealing of the donor and
acceptor rings in the same tricyclic system that seems to allow
for the combination of a low energy gap and a high extinction
coefficient. One-electron reduction of the 1,3-dimethylnaph-
thotriazinium cation with sodium/potassium amalgam allows
for the preparation of the corresponding neutral 1,3-dimethyl-
naphthotriazyl radical that has the EPR hyperfine coupling con-
sistent with localization of the electron on the nitrogen atom.
Taken together, these results illustrate the promise and poten-
tial for the azo peri-naphthalenes in many applications.
The first fraction yielded dark blue crystals (418 mg, 33%) which
were determined to be 2-methylnaphtho[1,8-de]triazine (3).
1
m.p. 131–1338C; H NMR (500 MHz, CDCl3): d=3.76 (s, 3H), 6.2 (dd,
J=7.5, 0.5 Hz, 2H), 6.84 (dd, J=8.5, 0.5 Hz, 2H), 6.95 ppm (dd, J=
8.5, 7.5 Hz, 2H); 13C NMR (125 MHz, CDCl3): d=54.9, 108.2, 120.6,
121.0, 129.3, 135.3, 146.5 ppm; IR (KBr): 3053w, 3008w, 1901w,
1631m, 1620m, 1578m, 1360s, 1162m, 1045m, 819s, 752s,
647m cmꢀ1; Raman (solid): 1581, 1436, 1398, 1367, 1344, 1298,
1256, 1199, 1153, 1100, 1073, 1039, 1001, 924, 848, 821, 787, 760,
612, 550, 489, 447 cmꢀ1
(238 mꢀ1 cmꢀ1).
;
UV (DMSO, lmax (e)): 596 nm
Experimental Section
General methods
The second fraction yielded burnt-orange crystals (452 mg, 35%)
which were determined to be 1-methylnaphtho[1,8-de]triazine (2).
m.p. 85–878C; 1H NMR (500 MHz, CDCl3): d=3.54 (s, 3H), 5.96 (d,
J=7.5 Hz, 1H), 7.07–7.16 (m, 3H), 7.25–7.32 ppm (m, 2H); 13C NMR
(125 MHz, CDCl3): d=39.1, 97.8, 115.7, 118.7, 119.4, 123.9, 128.4,
128.9, 133.7, 134.1, 138.6 ppm; IR (KBr): 3081w, 3054w, 3008w,
2928w, 1942w, 1925w, 1624s, 1578s, 1522m, 1473m, 1461m,
1438m, 1430m, 1393m, 1369s, 1298m, 1152m, 1040s, 1003s,
926s, 845m, 818 s, 766s, 659m, 612m, 448m, 414m cmꢀ1; Raman
(solid): 1581, 1512, 1428, 1371, 1336, 1317, 1271, 1241, 1211, 1176,
1134, 1042, 985, 902, 840, 798, 589, 463, 409, 341 cmꢀ1; UV (DMSO,
lmax (e)): 443 nm (340 mꢀ1 cmꢀ1). Crystals of X-ray quality were
grown from a saturated solution in ethanol, adding a drop of
chloroform, and allowing hexanes to diffuse into the solution at
48C for 4 days.
All the reagents and solvents were used as supplied commercially,
except for CH3CN, which was distilled over CaH2. Details of the in-
strumentation used in this research are described in a recent
paper.[38] The NMR spectra were measured on Varian Mercury Spec-
1
trometers with either 300, 400 or 500 MHz fields for H. Chemical
shifts are reported as d in ppm. Elemental analyses were per-
formed by The Elemental Analysis Laboratory at University of Mon-
treal.
1H-naphtho[1,8-de]-1,2,3,-triazine (1): The synthetic procedure as
described by Beecken et al.[18] was followed. In a boiling mixture of
H2O (316 mL) and glacial acetic acid (2 mL), crushed 1,8-diamino-
naphthalene (2.0 g, 12.6 mmol) was dissolved. A scoop of activated
charcoal was added once dissolution was complete to adsorb any
possible impurities. The mixture was filtered hot through a Bꢂchner
funnel and the red-brown filtrate was cooled on an ice bath to
a temperature of 258C. A solution of sodium nitrite (0.94 g,
13.6 mmol) in H2O (3.5 mL) was added dropwise over a period of
25 min. Over the course of the addition, a red precipitate formed
in suspension. The mixture was stirred for 10 min after the addition
and then filtered through fritted glass. The red slurry was collected
and recrystallized from 1,4-dioxane to give red crystals (1.18 g,
75% yield). m.p. 2458C (decomp, by differential scanning calorime-
try (DSC)); 1H NMR (400 MHz, [D6]DMSO): d=6.11 (d, J=7.6 Hz,
1H), 6.86–6.88 (m, 1H), 7.02 (d, J=8.0 Hz, 1H), 7.12 (t, J=8.0 Hz,
1,3-dimethylnaphtho[1,8-de]-1,2,3,-triazinium methylsulfate (5):
A
deep orange solution was obtained by dissolving 1-
methylnaphtho[1,8-de]-1,2,3,-triazine (100 mg, 0.55 mmol) in ben-
zene (30 mL). Dimethyl sulfate (82 mg, 0.55 mmol) was added to
the stirred solution over 10 min. The mixture was then heated at
reflux overnight and a brown precipitate gradually formed. The
mixture was then filtered and the crystals collected and washed
with diethyl ether until the washes were clear. The precipitate was
recrystallized from ethanol to give violet crystals of 1,3-
dimethylnaphtho[1,8-de]triazinium methylsulfate (5; 112 mg, 72%).
m.p. 2628C (decomp); 1H NMR (500 MHz, [D6]DMSO): d=3.34 (s,
ChemPlusChem 2012, 77, 387 – 395
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
393

D. S. Bohle et al.
3H), 3.87 (s, 6H), 7.14 (d, J=7.5 Hz, 2H), 7.55 (dd, J=8.0, 8.0 Hz,
2H), 7.71 ppm (d, J=8.5 Hz, 2H,); 13C NMR (125 MHz, [D6]DMSO):
d=44.3, 53.2, 108.3, 122.3, 126.1, 129.4, 131.7, 133.3 ppm; IR (KBr):
3117w, 3083w, 3053w, 3013w, 2984w, 2956w, 2916w, 2839w,
1641m, 1600m, 1582w, 1529m, 1483m, 1472m, 1459m, 1387m,
1344m, 1248s, 1230s, 1201w, 1166m, 1157m, 1058m, 1011s,
1-methylnaphtho[1,8-de]-2-[15N]-1,2,3,-triazine: 1H NMR (500 MHz,
CDCl3): d=3.54 (d, 3H, J=2.0 Hz) 5.96 (d, 1H, J=7.5 Hz), 7.07–7.17
(m, 3H), 7.25–7.32 ppm (m, 2H); 13C NMR (125 MHz, CDCl3): d=
39.17 (d, 2JCN =7.9 Hz), 97.8, 115.8 (d, J=3.25 Hz), 118.7, 119.4,
123.9, 128.4, 128.9, 133.7, 134.2, 138.7 ppm; 15N NMR (50 MHz,
CDCl3, ref. [15N]nitromethane): d=28.47 ppm; IR (KBr): 3080, 3052,
3012, 1627, 1578, 1518, 1469, 1385, 1368, 1296, 1253, 1195, 1153,
828s, 774s, 737s, 700m, 608m, 577m, 551w, 434w, 414m cmꢀ1
;
Raman (solid): 1584, 1478, 1459, 1424, 1390, 1367, 1348, 1260,
1085, 1065, 779, 695, 569, 493, 413 cmꢀ1; UV (EtOH, lmax (e)):
482 nm (665 mꢀ1 cmꢀ1). Crystals of X-ray quality were grown from
a solution in ethanol (5 mg of the purple crystals) and allowing
hexanes to diffuse into the solution at RT for 5 days.
1038, 999, 917, 818, 766, 610 cmꢀ1
.
2-methylnaphtho[1,8-de]-2-[15N]-1,2,3,-triazine: 1H NMR (500 MHz,
CDCl3): d=3.76 (d, 3H, J=2.0 Hz), 6.2 (d, 2H, J=7.0 Hz), 6.84 (d,
2H, J=8.5 Hz), 6.95 ppm (t, 2H, J=7.8 Hz); 13C NMR: (125 MHz,
2
CDCl3): d=54.9 (d, JCN =1.7 Hz), 108.25 (d, J=3.3 Hz), 120.6, 121.0,
129.3, 135.3, 146.53 ppm (d, J=2.4 Hz); 15N NMR (50 MHz, CDCl3,
ref. [15N]nitromethane): d=ꢀ92.9 ppm; IR (KBr): 1630s, 1619m,
1578m, 1366m, 1343s, 1331s, 1261m, 1162m, 1080m, 1042m,
Methylation of 1H-naphtho[1,8-de]-1,2,3,-triazine with
methyl trifluoromethanesulfonate
819s, 752s, 642m, 603m cmꢀ1
1,3-dimethylnaphtho[1,8-de]-2-[15N]-1,2,3,-triazinium
.
1-methylnaphtho[1,8-de]-1,2,3,-triazine (2): Methyl trifluorome-
thanesulfonate (0.29 g, 1.77 mmol) was added dropwise to a sus-
pension of 1H-naphtho[1,8-de]triazine (1; 0.10 g, 0.590 mmol) in
dry CH3CN (10 mL) at 08C and under nitrogen. The reaction mix-
ture was kept cold for 1 h and stirred for an addition 2 h at ambi-
ent temperature. The reaction progress was monitored by TLC on
silica with hexanes/ethyl acetate (3:1, v/v) as the eluent. Aqueous
sodium hydroxide solution (15%, 20 mL) was then added to the
dark violet reaction mixture and stirred for an additional 30 min.
The product was extracted with chloroform, washed with H2O,
brine, and dried over MgSO4. The solvent was evaporated to fur-
nish 1-methylnaphtho[1,8-de]triazine (2) as a burnt-orange solid
(0.065 g, 62%).
trifluoro-
methanesulfonate: 1H NMR (500 MHz, [D6]acetone): d=4.09–4.10
(m, 6H), 7.24 (d, 2H, J=8.0 Hz), 7.60 (t, 2H, J=7.8 Hz), 7.75 ppm
(d, 2H, J=8.5 Hz); 13C NMR: (125 MHz, [D6]acetone): d=43.86 (d,
2JCN =8.25 Hz), 108.1, 122.33 (d, J=1.375 Hz), 126.21 (d, J=
2.88 Hz), 129.1, 131.44 (d, J=2.25 Hz), 133.67 ppm; 15N NMR
(50 MHz, [D6]acetone, ref. [15N]nitromethane): d=10.99 ppm; 19F
(470 MHz, [D6]acetone): d=ꢀ78.86 ppm; IR (KBr): 3125, 3085,
1652w, 1646m, 1605m, 1588w, 152m, 1474m, 1458m, 1386m,
1353m, 1267s, 1223m, 1152s, 1032s, 988m, 929w, 819s, 766s,
698s, 637 s, 504m, 422s cmꢀ1
.
UV/Visible spectroscopy and titrations of 1H-naphtho[1,8-
1,3-dimethylnaphtho[1,8-de]triazinium
trifluoromethanesulfo-
de]-1,2,3,-triazine
nate (4): Methyl trifluoromethanesulfonate (0.48 g, 2.96 mmol) was
added dropwise to a suspension of 1H-naphtho[1,8-de]triazine (1;
0.10 g, 0.590 mmol) in dry CH3CN (10 mL) at RT and under nitrogen.
After 1 h an additional amount of methyl trifluoromethanesulfo-
nate (0.14 g, 0.854 mmol) was added dropwise and the reaction
mixture was stirred for a further 2 h. The reaction progress was
monitored by TLC on silica with hexanes/ethyl acetate (3:1, v/v) as
the eluent. After evaporation of the solvent, the reaction mixture
was redissolved in a minimum amount of CH2Cl2 and treated with
sodium bicarbonate (0.1 g) and stirred for 30 min. After filtration
the residue was purified by column on silica gel (a short plug) to
give 1,3-dimethylnaphtho[1,8-de]-1,2,3,-triazinium trifluorometha-
nesulfonate (4) as a dark violet solid (0.162 g, 80%). Rf =0.3
(CH2Cl2/MeOH 3:0.3); m.p. 2158C ( by DSC); 1H NMR (500 MHz,
[D6]acetone): d=4.10 (s, 6H), 7.24 (d, J=8 Hz, 2H), 7.60 (t, J=
8.0 Hz, 2H), 7.75 ppm (d, J=8 Hz, 2H); 13C NMR (125 MHz,
[D6]DMSO): d=44.3, 108.3, 121 (q, J=320 Hz, CF3), 122.3, 124.9,
126.1, 129.4, 131.7, 133.3 ppm; 19F NMR (470 MHz, [D6]acetone) d=
ꢀ78.86 ppm; IR (KBr): 3125m, 3096w, 3060w, 1648m, 1604m,
1589m, 1534m, 1488m, 1461m, 1393m, 1308s, 1264s, 1233s,
1151s, 1053s, 1034s, 986m, 822s, 767s, 702s, 651s, 640s, 583m,
A quartz cuvette (1 cm path length) was filled with 3 mL of
0.26 mm solution of 1H-naphtho[1,8-de]triazine in ethanol. Concen-
trated hydrochloric acid was diluted tenfold with ethanol and ti-
trated dropwise into the cuvette, with mixing and acquisition of
a UV/Visible absorption spectrum after each addition. A similar ti-
tration was carried out using sodium ethoxide.
Solvatochromism of the naphtho[1,8-de]-1,2,3,-triazine de-
rivatives
Solutions of the crystals were prepared in a variety of solvents of
differing polarities. The solutions were prepared so as to give ab-
sorptions of approximately 1 AU. The wavelength of maximum ab-
sorption was tracked, and the peak to peak separations were de-
termined by calculating the second derivatives of representative
spectra.
Computational studies
520s cmꢀ1
; UV/Vis (H2O, lmax): 206, 223, 240sh, 300bump,
All of the calculations described above were performed using
Gaussian 98.[39] Computations were carried out at the restricted
Hartree–Fock (RHF),[40] and density functional theory (DFT) levels.
The DFT calculations used the hybrid B3LYP functional and triple
zeta 6–311+ +G** basis sets.[41,42] The calculated molecular geo-
metries were fully optimized and correspond to minima on the po-
tential-energy surface as confirmed by the absence of imaginary vi-
brational frequencies. For the closed-shell diamagnetic species all
calculations were performed using the RB3LPY default wavefunc-
tions, but for radical 9 the unrestricted open-shell wavefunctions,
UB3LYP, were employed.
315bump, 331 nm; MS (ESI): m/z calcd for C13H12N3O3SF3 [M]: 347;
and calcd for [MꢀSO3CF3]+: 198; found: 198 (100%); elemental
analysis calcd (%) for C13H12N3O3SF3 (347): C 44.91, H 3.46, N 12.09,
S 9.23; found: C 44.91, H 3.54, N 12.04, S 9.50.
Characteristic data for naphtho[1,8-de]-2-[15N],1-2,3,-triazine
derivatives
The 15N-derivatives were prepared as above using [15N]sodium ni-
trite.
394
www.chempluschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 77, 387 – 395

Triazene Radicals
[7] P. G. Gassman, Acc. Chem. Res. 1970, 3, 26–33.
EPR spectroscopy
[8] F. Grandinetti, Recent Res. Dev. Mol. Struct. 2002, 1, 23–40.
[9] I. Novak, L. J. Harrison, W. Li, B. Kovac, J. Phys. Chem. A 2007, 111, 2619–
2624.
[10] V. Mezheritskii, Adv. Heterocycl. Chem. 2008, 95, 1–25.
[11] H. Beecken, Angew. Chem. 1967, 79, 316–317; Angew. Chem. Int. Ed.
Engl. 1967, 6, 360.
[12] M. T. Waterman, PhD Thesis, Ohio State University (Columbus), 1997.
[13] J. Fabian, Z. Chem. 1987, 27, 28–29.
[14] J. Fabian, Dyes Pigm. 2010, 84, 36–53.
Continuous wave EPR spectra were obtained by using a Bruker
Elexsys E580 spectrometer (X-band). Simulation was done with the
Pest Winsim[45]. The EPR samples were prepared by treatment of
starting material with potassium/sodium amalgam in dry THF at
RT, the solution is then placed in an EPR quartz tube (4 mm) and
frozen with liquid nitrogen. The EPR spectra were taken after plac-
ing the quartz tube in the cavity and allowing it to warm up to
203 K for a few minutes.
[15] J. Fabian, R. Zahradnik, Angew. Chem. 1989, 101, 693–710; Angew.
Chem. Int. Ed. Engl. 1989, 28, 677–694.
[16] A. G. Ekstrand, Chem. Ber. 1887, 20, 219–226.
Electrochemistry
[17] H. Erdmann, J. v. Liebig, Justus Liebigs Ann. Chem. 1888, 247, 305–326.
[18] H. Beecken, P. Tavs, H. Sieper, Liebigs Ann. Chem. 1967, 704, 166–171.
[19] H. Beecken, P. Tavs, Liebigs Ann. Chem. 1967, 704, 172–175.
[20] H. Sieper, P. Tavs, Liebigs Ann. Chem. 1967, 704, 161–165.
[21] P. Tavs, H. Sieper, H. Beecken, Liebigs Ann. Chem. 1967, 704, 150–160.
[22] M. J. Perkins, J. Chem. Soc. 1964, 3005–3008.
[23] A. Gieren, V. Lamm, Acta Crystallogr. Sect. B 1980, 36, 3142–3144.
[24] A. Hazell, R. Mariezcurrena, Acta Crystallogr. Sect. B 1980, 36, 3140–
3142.
[25] P. Bazinet, G. P. A. Yap, D. S. Richeson, J. Am. Chem. Soc. 2003, 125,
13314–13315.
[26] R. Bartetzko, R. Gleiter, Angew. Chem. 1978, 90, 481–482; Angew. Chem.
Int. Ed. Engl. 1978, 17, 468–469.
The cyclic voltammetry experiments were performed with
a BAS CV-50W electrochemical analyzer in anhydrous acetonitrile,
using 0.1m Bu4NPF6 in CH3CN as a supporting electrolyte with scan
rates of 50–1000 mVsꢀ1. Platinum wire, platinum disk, and Ag/
AgNO3 were used as counter, working, and reference electrodes,
respectively. The Fc/Fc+redox couple was used as an internal ref-
erence. Spectroelectrochemistry experiments were performed by
using a 011240 SEC-C Thin Layer Quartz Glass Spectroelectrochemi-
cal cell Kit (Bas Inc.). UV/visible spectra were measured by using
a HP 8453 Diode Array spectrophotometer from Hewlett Packard.
[27] A. de Aguiar, Chem. Ber. 1874, 7, 306–319.
[28] G. T. Morgan, F. M. G. Micklethwait, J. Chem. Soc, Trans., 1909, 97, 2557–
2564.
X-ray crystallography
[29] R. J. Kobylecki, A. McKillop, Adv. Heterocycl. Chem. 1976, 19, 215–278.
[30] J. Mathew, C. H. Suresh, Inorg. Chem. 2010, 49, 4665–4669.
[31] M. D. Sevilla, S. Swarts, J. Phys. Chem. 1982, 86, 1751–1755.
[32] R. G. Shulman, R. O. Rahn, J. Chem. Phys. 1966, 45, 2940–2948.
[33] F. A. Neugebauer, G. Rimmler, Magn. Reson. Chem. 1988, 26, 595–600.
[34] R. G. Hicks in Stable Radicals, (Ed., R. G. Hicks), Wiley, New York, 2010,
pp. 245–279.
Single-crystal X-ray analysis were carried out with a BRUKER SMART
CCD diffractometer using graphite-monochromated MoKa radiation
(l=0.71073 ꢁ). The structures were solved by direct methods and
refined by full-matrix least squares on F2 of all data, using SHELXTL
software.[43] Key crystallographic data are summarized in Tables S1–
S12.[44]
[35] M. Robert, A. Neudeck, G. Boche, C. Willeke, K. S. Rangappa, P. Andrews,
New J. Chem. 1998, 22, 1437–1444.
Acknowledgements
[36] M. Shiotani, L. Sjçquist, A. Lund, S. Lunell, L. Eriksson, M.-B. Huang, J.
Phys. Chem. 1990, 94, 8081–8085.
[37] V. Barone in Recent Advances in Density Functional Methods, Part 1, (Ed.,
D. P. Chong), World Scientific Publ. Co., Singapore, 1996.
[38] D. S. Bohle, I. Perepichka, J. Org. Chem. 2009, 74, 1621–1626.
[39] M. J. Frisch et al., Gaussian03, Gaussian Inc., Pittsburgh, PA, 2003.
[40] C. C. Roothaan, J. Rev. Mod. Phys. 1951, 23, 69–89.
[41] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[42] W. C. Y. Lee, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[43] SHELXTL, SHELXTL 5.10 ed., Bruker AXS, Inc., Madison, Wisconsin, 1997.
[44] CCDC 797882 (2) and 797883 (4) contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via www.ccdc.cam.a-
c.uk/data_request/cif.
We gratefully acknowledge the CRC, CFI, and NSERC for support
in the forms of a CRC, CFI, and discovery grants to D.S.B.
Keywords: dyes/pigments
solvatochromisim · triazenes
· peri-naphthalenes · radicals ·
[1] H. A. Spinney, I. Korobkov, G. DiLabio, G. P. A. Yap, D. S. Richeson, Orga-
nometallics 2007, 26, 4972–4982.
[2] H. A. Spinney, G. P. A. Yap, I. Korobkov, G. DiLabio, D. S. Richeson, Orga-
nometallics 2006, 25, 3541–3543.
[45] Pest Winsim: http://www.niehs.nih.gov/research/resources/software/
tools/index.cfm.
[3] H. M. Tuononen, R. Roesler, J. L. Dutton, P. J. Ragogna, Inorg. Chem.
2007, 46, 10693–10706.
[4] R. A. Abramovitch, R. Jeyaraman in Azides Nitrenes, (Eds., E. F. V. Scriven),
Academic, Orlando, 1984, pp. 297–357.
[5] O. E. Edwards, Uniw. Adama Mic. Pozn., Wydz. Mater., Fiz. Chem., [Pr.],
Ser. Chem. 1975, 18, 9–26.
Received: January 27, 2012
[6] D. E. Falvey in Reactive Intermediate Chemistry, (Eds., R. A. Moss, M. S.
Platz, M, Jones Jr.), Wiley, New York, 2004, pp. 593–648.
Published online on March 13, 2012
ChemPlusChem 2012, 77, 387 – 395
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
395