Protonation behaviour of 2-phenyl-1,3-diazaazulene derivatives

Article abstract of DOI:10.1016/j.tet.2017.12.051

Three 2-phenyl-1,3-diazaazulene derivatives were synthesized and their protonation behaviours were investigated systematically via UV–vis absorption titration and ¹H NMR titration, as well as theoretical calculations. One of them exhibited a monoprotonation process while the others displayed prominent halochromic diprotonation responses. Interestingly, upon protonation of 2-phenyl-1,3-diazaazulene derivatives, the coplanarity and conjugation of the 16-π-conjugated backbones were well kept, while the electronic structures were controllably adjusted. The response mechanism of 1,3-diazaazulene derivatives towards acid is through the attachment of acid proton to the nitrogen atom in the diazaazulene ring, resulting in the change of the hybridization of protonated-N from sp² to sp³, which differed from that of the well-known azulene (analogue of 1,3-diazaazulene, protonation at carbon atom). This work would provide a new insight into the protonation research of the organic functional molecules.

Full text of DOI:10.1016/j.tet.2017.12.051

Accepted Manuscript
Protonation behaviour of 2-phenyl-1,3-diazaazulene derivatives
Peili Sun, Danling Rao, Pu Zhang, Yujun Qin, Zhi-Xin Guo
PII:
S0040-4020(17)31335-2
10.1016/j.tet.2017.12.051
TET 29207
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 25 September 2017
Revised Date: 16 December 2017
Accepted Date: 25 December 2017
Please cite this article as: Sun P, Rao D, Zhang P, Qin Y, Guo Z-X, Protonation behaviour of 2-
phenyl-1,3-diazaazulene derivatives, Tetrahedron (2018), doi: 10.1016/j.tet.2017.12.051.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Protonation behaviour of 2-phenyl-1,3-
diazaazulene derivatives
Leave this area blank for abstract info.
Peili Sun, Danling Rao, Pu Zhang , Yujun Qin and Zhi-Xin Guo
Department of Chemistry, Renmin University of China, Beijing 100872, P. R. China

1
ACCEPTED MANUSCRIPT
Tetrahedron
journal homepage: www.elsevier.com
Protonation behaviour of 2-phenyl-1,3-diazaazulene derivatives
Peili Sun, Danling Rao, Pu Zhang , Yujun Qin and Zhi-Xin Guo
Department of Chemistry, Renmin University of China, Beijing 100872, P. R. China
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
Three 2-phenyl-1,3-diazaazulene derivatives were synthesized and their protonation behaviours were
investigated systematically via UV-vis absorption titration and 1H NMR titration, as well as theoretical
calculations. One of them exhibited a monoprotonation process while the others displayed prominent
halochromic diprotonation responses. Interestingly, upon protonation of 2-phenyl-1,3-diazaazulene
derivatives, the coplanarity and conjugation of the 16-π-conjugated backbones were well kept, while
the electronic structures were controllably adjusted. The response mechanism of 1,3-diazaazulene
derivatives towards acid is through the attachment of acid proton to the nitrogen atom in the
diazaazulene ring, resulting in the change of the hybridization of protonated-N from sp2 to sp3, which
differed from that of the well-known azulene (analogue of 1,3-diazaazulene, protonation at carbon
atom). This work would provide a new insight into the protonation research of the organic functional
molecules.
Available online
Keywords:
1,3-diazaazulene derivatives
protonation
response mechanism
π-conjugation
2017 Elsevier Ltd. All rights reserved
.
1. Introduction
Among various organic chromophores, azulene derivatives
were reported to display fine-tuned photophysical and optical
behaviours owing to their unique polarized charge distribution on
the fused seven- and five-membered rings,14 thus showing great
advantages as optical material candidates.15 Researches towards
azulene derivatives have also involved the halochromic responses
owing to their specific structures.16-22 It was reported that, upon
protonation, the dipole moments of azulene derivatives with
special substituents could be largely enhanced,16 resulting in the
adjustment of the electronic structures and the changes of
corresponding physical properties, such as the typical
halochromic response.23,13,24 Studies have indicated that the
protonation of azulene-based derivatives always occurs on the C1
(or C3) atom in the five-membered ring,25,21,13,26 with the
formation of tropylium cation.27 Based on the protonation
response mechanism, the protonation of azulene derivatives
would undoubtedly convert the hybridization of the protonated
carbon atom from sp2 to sp3, leading to the decrease of the
molecular conjugation, which might break the planarity of
azulenes and induce dramatic geometric variations.
Functional organic chromophores are drawing more and more
attention due to their potential applications in optoelectronic
fields,1 including light-emitting diodes,2 solar cells,3,4 nonlinear
optical materials,5 and sensors.6 The protonation of organic
chromophores, which is considered as a facile and effective way
to tune the transition gaps, alter the molecular interactions and
control the aggregation morphology of the organic
chromophores, has been one of the hottest research topics due to
its potential utilities in drug delivery,7 chemical sensors8 and
fluorescence switching devices.9 For example, perylene bisimides
were reported to exhibit near-infrared absorptions beyond 1100
nm due to the great bathochromic shift of their charge transfer
bands resulted from the acid-base stimuli.10 Dihydro-
teraazaacene diimides containing 6 or 7 laterally fused six-
membered rings show obvious colour changes upon protonation-
deprotonation process.11
Generally, the response mechanism is based on the acid-base
reactions between the organic chromophores and extra acid
stimulus, which results in the chromophore-based cation with
proton attached to a special site. Upon the transformation of the
chromophore from neutral state to a positively charged cation,
some properties including molecular geometry12 and orbital
energy gaps13 could exhibit certain alterations, and some
compounds even demonstrate dramatic halochromic behaviours.
Usually, the molecular geometry would be greatly influenced
upon protonation.
Scheme 1. Chemical structures for 1, 2, and 3.
———
Corresponding authors: Tel.: +0-106-251-2822; fax: +0-106-251-6444;
e-mail: zhangpu@ruc.edu.cn (P. Zhang), gzhixin@ruc.edu.cn (Z.X. Guo).

2
Tetrahedron
ACCEPTED MANUSCRIPT
1,3-diazaazulene (DAA) is a unique derivative of azulene in
which two carbon atoms are replaced by nitrogen atoms, with a
rather large dipole moment locating within the heteroaromatic
core.28,29 Our group and collaborators have previously reported a
series of donor-π-acceptor typed DAA derivatives which show
relatively small ground dipole moment (ꢀg) and large first-order
hyperpolarizability (β).30-32 The above mentioned DAA
derivatives all contain a basic conjugated skeleton of 2-phenyl-
1,3-diazaazulene (1, as shown in Scheme 1), which should be
responsible for their superior performances in nonlinear optics.
As is proved, the simple molecule 1 is coplanar, indicating that
the introduction of phenyl group to 2 C atom of DAA would
enlarge the conjugation system to form a 16-π-conjugated
construction.30 Meanwhile, considering the two nitrogen atoms in
the five-membered ring, which would probably facilitate the
protonation behaviour, it is supposed that structures based on 1
would exhibit rather novel performances compared with azulene
derivatives as promising acid-stimuli responsive materials.
Fig. 1. UV-vis absorption spectra of 1 (20 ꢀM), 2 (15 ꢀM), and 3 (15 ꢀM) in
dichloromethane.
The typical UV-vis absorption spectra of 1, 2, and 3 measured
in dichloromethane are shown in Fig. 1. 1 exhibited a major
absorption band at 356 nm, similar to that of DAA.34 Compared
with 1, a pronounced red-shift of the peak to 455 nm was
observed for 2. As for 3, a peak with rather larger red-shift to 580
nm was obtained, with a ꢁλ value of 224 nm compared with that
of 1. These red shifts should be caused by the introduction of
electron-donating group (N,N-dimethylamino) and/or electron-
withdrawing group (nitro) to 1, which induces marked
intramolecular charge transfer and tunes the electron transitions
of 2 and 3.
Thus in this paper, based on the three typical DAA derivatives
2-phenyl-1,3-diazaazulene (1), 2-(p-N,N-dimethylaminophenyl)-
1,3-diazaazulene
(2),
and
6-nitro-2-(p-N,N-
dimethylaminophenyl)-1,3-diazaazulene (3) (Scheme 1), the
protonation behaviours were detailedly investigated through UV-
vis absorption titration and 1H NMR titration experiments.
Moreover, the protonation mechanisms of the chromophores
were proposed combined with the theoretical calculations. It was
found that 1, 2, and 3 well kept their 16-π-conjugated and
coplanar structures during protonation while the molecular
HOMO-LUMO energy gaps were greatly changed, indicating
that the electronic structures of the 1-based derivatives could be
well tuned via protonation without damaging the conjugation and
planarity.
2.3. UV-vis absorption titration studies
2. Results and discussion
2.1. Synthetic route
Scheme 2. Synthetic route for 1, 2 and 3.
Synthetic procedures for 1, 2 and 3 were shown in Scheme 2.
They were generally obtained through the condensation reactions
between 2-methoxy tropone (or 2-methoxy-5-nitro tropone) and
corresponding benzamidinium hydrochlorides under reflux
condition, with potassium tert-butoxide as base and ethanol as
solvent. Detailed procedures and characterizations were
presented in the experimental section. 3 was synthesized
according to our previous report.33,30,32 1 was obtained as white
powder, 2 as red powder and 3 as light violet powder. They were
all soluble in many organic solvents, such as dichloromethane,
chloroform, ethanol, ethylacetate and so on.
2.2. UV-vis absorption studies
Fig. 2. UV-vis titration spectra of 1 (20 ꢀM) upon increasing addition of TFA
(a) 0.0 to 6.0 equivalents; (b) large excess equivalents, and the gray line in (b)
was recorded upon the subsequent addition of TEA The inset of (a) shows:
the absorption values at 392 nm versus equivalents of TFA.

3
UV-vis absorption titration,
a
common method for
ACCEPTED MANUSCRIPT
investigation of the halochromic behaviours of organic
chromophores,35,36 was used to monitor the protonation processes
of 1, 2, and 3 in dichloromethane upon the addition of
trifluoroacetic acid (TFA) as the proton donor.
With the increasing addition of TFA, the intensity of original
peak of 1 at 356 nm was slowly decreased with the emerging of a
new peak around 392 nm (Fig. 2a). The isobestic point at 367 nm
further proved the protonation process of 1. The absorption value
at 392 nm was plotted versus the equivalents of TFA (inset of
Fig. 2a), which exhibited a linear relationship up to 1.4
equivalent of TFA, implying that about 1.4 equivalent of TFA
was needed for the one protonation of 1. The excess addition of
TFA afterwards did not induce any other new absorption peak
(Fig. 2b). Subsequent addition of triethylamine (TEA) could
neutralize the system back to the pristine state (the gray line in
Fig. 2b). In this case, the protonation of 1 probably only need one
proton, with slight solution colour change.
Fig. 4. UV-vis titration spectra of 3 (15 ꢀM upon increasing addition of TFA
from (a) 0.0 to 6.0 equivalents s; (b) large excess equivalents, and the gray
line in (b) was recorded upon the subsequent addition of TEA; and (c) the
solution colour changes of 3 under different equivalents of TFA. The insets
show: (a) the absorption values at 627 nm versus equivalents of TFA and (b)
the absorption values at 400 nm versus equivalents of TFA.
bleaching of the primary absorptions at 455 nm and 290 nm,
respectively. The isobestic point at 477 nm indicated the spectral
changes of two different species. As the equivalent of TFA was
increased from 2 to 6, the peak at 515 nm was slightly red-shifted
towards 520 nm, with the shape unchanged. The absorption value
at 515 nm was plotted versus the equivalents of TFA (inset of
Fig. 3a), which exhibited a linear relationship up to 1.1
equivalent of TFA, implying that about 1.1 equivalent of TFA
was needed for one protonation of 2. In addition, the solution
colour gradually changed from light yellow to dark pink (Fig.
3c). However, when TFA was dramatically increased to hundreds
of equivalents, another two new absorption bands near 330 nm
and 380 nm arose (Fig. 3b), together with the colour changing
from dark pink to colourlessness (Fig. 3c). Meanwhile, the peak
at 515 nm decreased and gradually red-shifted to a more
broadened band. Likewise, the isobestic point at 395 nm proved
the coexistence of the initially formed species and the newly
formed one. There was also a linear increase as the TFA’s
equivalent was increased from 1000 to 7000 (inset of Fig. 3b).
Upon the continuing addition of TEA, the resulting solution also
could be recovered to neutral state (the gray line shown in Fig.
3b). Thus, different from 1, there should be a subsequent
protonation of 2 which is responsive to large equivalents of acid.
In this context, 2 is a double-protonated stimuli response: the first
protonation is responsive to small equivalents of acid, the
subsequent one is responsive to large equivalents of acid.
Fig. 3. UV-vis titration spectra of 2 (15 ꢀM) upon increasing addition of TFA
from (a) 0.0 to 6.0 equivalents; (b) large excess equivalents, and the gray line
in (b) was recorded upon the subsequent addition of TEA; and (c) the solution
colour changes of 2 under different equivalents of TFA. The insets show: (a)
the absorption values at 515 nm versus equivalents of TFA and (b) the
absorption values at 383 nm versus equivalents of TFA.
Fig. 3a shows the absorption spectra evolution for 2 when
TFA was increased from 0 to 6 equivalents. Upon the gradual
addition of TFA, a new absorption was formed at around 515 nm.
The increasing of this absorption was accompanied with the

4
Tetrahedron
ACCEPTED MANUSCRIPT
We also studied the UV-vis absorption properties of 2 under
different addition of trifluoromethanesulfonic acid (Fig. S4). It
shows that maximum absorption peaks of 2 under 1.0 and 5.0
equivalents trifluoromethanesulfonic acid are in good agreement
with the results of 2 in the presence of 1.5 and thousands of
equivalents of TFA, respectively.
UV-vis titration results for 3 are displayed in Fig. 4, which
show similar results when compared with those of 2. There was a
new absorption at 627 nm and an isobestic point at 590 nm while
the TFA was increased from 0 to 6 equivalents (Fig. 4a). The
absorption value at 627 nm was plotted versus the equivalents of
TFA (inset of Fig. 4a), which exhibited a linear relationship up to
2 equivalents of TFA, implying that about 2 equivalents of TFA
was needed for one protonation of 3. As addition of TFA was up
to thousands of equivalents, the main peak around 630 nm
gradually decreased and shifted towards 680 nm, together with
the emerging of a new peak at 400 nm (isobestic point at 432 nm,
shown in Fig. 4b). Another linear relationship from 6000 to
40000 equivalents of TFA for the absorption values at 400 nm
was shown in the inset of Fig. 4b. Besides, the addition of TEA
could induce the regaining of the neutral 3 (the gray line shown
in Fig. 4b), indicating the reversible protonation-deprotonation
processes for 3. Different from 2, the solution colour changed
from light violet to blue and finally to colourlessness (Fig. 4c).
Thus, the influence caused by protons on 3 is also a double-
protonation course: the first protonation is responsive to small
equivalents of acid, the subsequent one is responsive to large
equivalents of acid.
2.4. 1H NMR titration studies
1
To further confirm the results from UV-vis titration, H NMR
titration, another useful way for the research of protonation,37,38
was also carried out to study the protonation behaviours of 1, 2,
and 3. For both 2 and 3, the 1H signals associated with the methyl
moieties of N,N-dimethylamino phenyl groups were almost
unchanged, indicating that the protons from TFA did not bind to
the nitrogen atoms of the animo group, implying the binding of
the protons with the nitrogen atoms in the DAA rings. Fig. 5
1
shows the partial H NMR titration spectra of 1, 2, and 3 upon
different addition of TFA from 0 to 6 equivalents. Compared
1
with the original signals, the H NMR spectra of the protonated
ones all corresponded to symmetrical species (Fig. 5), which
suggested a fast exchange of the proton binding between the two
diazaazulene nitrogen atoms.39
As for 1, the large shifts in the spectra were observed until the
addition of 1.5 equivalents of TFA as shown in Fig. 5a. The
signals of protons in the seven-membered ring exhibited similar
downfield shifts, and the signals of protons in the phenyl ring
showed only slight shifts. For the signal of proton H(d), an
upfield shift was observed, while for the signals of H(e) and H(f),
downfield shifts were observed. However, when TFA
concentration was above 1.5 equivalents, the signals of the
protons H(a), H(b), H(e) and H(f) displayed a slight downfield
shift, and the signals of the protons H(c) behaved a slight upfield
shift. Still, H(d) presented a more influenced shift than others,
indicating that signal of H(d) for the protonated species was
much easier to be affected by the solvent polarity. The chemical
shifts of all the protons were plotted along with the equivalents of
TFA (Fig. S1). The linear relationships were observed when the
concentrations of TFA were from 0 to 1.5 equivalents, suggesting
that about 1.5 equivalents of TFA were needed to realize the
protonation of 1, which agreed well with the UV-vis titration
results.
1
Fig. 5. Partial H NMR titration spectra upon addition of TFA from 0 to 6
equivalents for (a) 1, (b) 2, and (c) 3, respectively (CDCl3 as solvent).
1
Equally, H NMR protonation results of 2 are shown in Fig.
5b. Upon protonation, the signals of protons H(a) and H(b) in
seven-membered ring showed similar downfield shifts, whereas
that of proton H(d) exhibited an upfield shift. In addition, signal
of proton H(c), which was in a smaller distance from the nitrogen
atom of the conjugated ring, displayed a downfield-shift to
upfield-shift behaviour, while signal of proton H(e) remained
nearly unchanged. The chemical shifts of labeled protons in 2

5
were plotted versus TFA equivalent (Fig. S2), which exhibited
linear relationships under the TFA equivalent range of 0-1.2, in
agreement with the first protonation course of the 2 obtained
from UV-vis protonation titration results.
were calculated, and the results are listed in Table 1 (2), Table
ACCEPTED MANUSCRIPT
S1 (1), and Table S2 (3), respectively. 2 is representative and
selected for the detailed analysis.
Table 1
The 1H NMR results of 2 under addition of 1.0 and 5.0
equivalents trifluoromethanesulfonic acid were also studied (Fig.
6). It can be seen that, signals of·2 in 1.0 equivalent
trifluoromethanesulfonic acid are similar to those of 2 in 1.5
equivalent trifluoroacetic acid, while signals of 2 performed
downwards shifting in 5.0 equivalents trifluoromethanesulfonic
acid, indicating that the second protonation should occur. The
signals of protons in 5.0 equivalents trifluoromethanesulfonic
The corrected zero-point vibrational energy (ZPE) and proton affinity (PA)
values calculated using G09 at different sites of 2 and monoprotonated 2
H+·2).
Sitesa
2
H+·2
Scaled ZPE
(au)
PA
Scaled ZPE
(au)
PA
(kcal/mol)
247.61
247.61
223.79
(kcal/mol)
--
-
acid are broad due to the poor solubility of 2H+·2·CF3SO3 in
N1
N3
N5'
0.3036
0.3036
0.3038
--
chloroform.
0.3166
0.3174
158.64
161.00
a Sites according to Scheme 1.
For 1, its initial protonation should only occur on one of the
two nitrogen atoms. Accordingly, the subsequent protonation on
the other nitrogen atom is very difficult since the PA value is
rather small (Table S1). It means that only monoprotonation
would occur in one of the two nitrogen atoms of 1. This is in
good accordance with the UV-vis titration as well as the 1H NMR
titration results.
Fig. 6. Partial 1H NMR results of compound 2 in neutral and protonated
As for 2 and 3, the possibility of the first protonation on the
amino group must be considered. However, since the PA value of
the heterocyclic nitrogen atom (N1 or N3) is rather large than
that of amino nitrogen atom (N5'), it’s quite clear that the first
protonation should occur on the one nitrogen atom of the
diazaazulene ring (Table 1, Table S2). Meanwhile, since the PA
value of N5' is larger than that of N3 (or N1) for the subsequent
protonation, the second protonation should occur on the amino
nitrogen atom (N5') for these two molecules (Table 1, Table S2).
states. (TFMS: trifluoromethanesulfonic acid, CDCl3 is used as the solvent)
For 3, signals for each proton behaved relatively less chemical
shifts compared with those of 1 and 2 (Fig. 5c). The addition of
TFA initially induced downfield shifting for signals of H(b),
H(c), and H(e), respectively. Later, signals for protons H(c) and
H(d) exhibited an upfield shifting, while H(b) continued a
downfield shifting and H(e) shifted little. The chemical shifts of
labeled protons in 3 were also plotted versus equivalents of TFA
(Fig. S3). Although there was no obvious linear relationship
observed, when the equivalents of TFA were less than 2.0, the
chemical shifts for each proton presented relatively regular
changes, implying the completion for the first protonation of 3,
which was also in coincidence with the UV-vis titration results.
2.5. DFT studies
To further confirm the protonation sites during the protonation
processes for 1, 2, and 3, a theoretical study of possible
protonation selection, as well as resulted protonated species, is
particularly critical. In this context, DFT (density function
theory) and TD-DFT (time-dependent density function theory)
calculations were carried out using the Gaussian 09 software
suite at the B3LYP/6-31G* level.40
Usually, to evaluate the protonation sites for organic
compounds, proton affinity (PA), which is assumed as the
negative of the enthalpy (ꢁH) for the B + H+ → BH+ process (B
represents organic compound in general),41 is essentially critical.
In the meantime, the variation in zero-point vibration energies
(ZPE) should be also considered. Thus, PA and ZPE were all
calculated and analyzed based on the optimized geometries.
Since nitrogen is demonstrated to be the most susceptible site
upon proton attack in the aromatic nitrogen-containing
heterocyclic compounds,41 only the two nitrogen atoms in the
five-membered ring were considered in the calculation. As the
symmetry of 1 is C2v, it is reasonable to only consider one
nitrogen atom. While the nitrogen atom of the nitro group in 3
will not be considered for protonation, the nitrogen atom of the
amino group in both 2 and 3 should not be neglected. Hence,
only the PA and ZPE values of the two types of nitrogen atoms
Fig. 7. ESP maps for neutral 2 and its monoprotonated species.
The above conclusion can be observed directly from the
electrostatic potential (ESP) maps as well as their first protonated
species (Fig. 7, Fig. S5 and Fig. S6). Clearly, each of the nitrogen
atoms in the diazaazulene ring is negatively charged for the
neutral 1, 2, and 3, indicating that the first proton should be
prioritized to bind to one of the diazaazulene nitrogen atoms. On
the other hand, as can be seen from the ESP maps of the
monoprotonated 1, 2, and 3, since all of the three geometries are
coplanar, the positive charge in the diazaazulene nitrogen atom is
nearly delocalized over the whole molecule, and the positive
potentials decrease along with the distance from the protonated
nitrogen atom, leading to the further protonation at site N5' rather
than at N3 (or N1). These results are in well accordance with
Tang’s report.25

6
Tetrahedron
ACCEPTED MANUSCRIPT
Table 2
Summarization of HOMO, LUMO energies and corresponding maximum absorption bands of 1, 2, 3 and their protonated forms.
Parameters
1
H+·1
2
H+·2
2H+·2
3
H+·3
2H+·3
a
EHOMO
-6.05
-2.39
3.66
359
356
-10.03
-6.69
3.34
391
-5.00
-2.03
2.97
441
455
-8.41
-6.02
2.39
516
-13.82
-9.13
4.69
334
-5.40
-3.05
2.35
528
590
-8.78
-6.67
2.11
571
-13.40
-9.69
3.71
358
a
ELUMO
ꢁEb
λmax/nmc
λmax/nmd
390
515
383
627
400
a Calculated energy values using G09.
b ꢁE = ELUMO − EHOMO
.
c Calculated values.
d Experimental values.
(oscillator strength 0.3748, Table S10). These calculated results
roughly agree with the experimental results and have the same
changing tendency.
It was reported that similar to azulene and its derivatives,
DAA is negatively charged in the five-membered ring.28
Considering that the molecules investigated are conjugated, when
the first protonation occurs on the nitrogen atom of the five-
membered ring, the negative charge will be delocalized, and the
energy gap between HOMO and LUMO will be lowered. This
will lead to the red shifting of the maximum absorption peak
compared with that of the corresponding neutral state. However,
for 2 or 3, the subsequent protonation takes place on the nitrogen
atom of amino group, which is an electron donor in the
conjugated molecule. Clearly, this binding of proton will largely
reduce the electron donating effect, resulting the blue shifting of
the maximum absorption of either 2 or 3 compared with that of
the corresponding neutral species.
Fig. 8. Optimized geometries and molecular frontier orbitals of 2 and its
protonated species.
Molecular frontier orbitals based on the optimized geometries
of 1, 2, and 3 and their corresponding protonated species were
calculated (Fig. 8, Fig. S7-S24, and Table S3 to Table S11).
Several selected parameters are summarized in Table 2. As can
be seen, both HOMO and LUMO energy levels for H+·1
(monoprotonated 1), H+·2 (monoprotonated 2) and H+·3
(monoprotonated 3) are lowered. As for H+·1, the energy gap is
decreased from 3.66 eV (neutral 1) to 3.34 eV. The largest
oscillator strength 0.420 of H+·1 is attributed to the transition of
S0 → S2, with a maximum wavelength of 391 nm (Table S6).
This agrees well with the UV-vis absorption experimental results.
2.6. Charge distribution analysis
The NBO (natural bond orbital) value of an atom in a
molecule is representative of the charge density.42 The calculated
NBO values are listed in Table 3. The NBO value of N1 (or N3)
is in the order of 2 ≥ 1 > 3. This result agrees well with the
1
experimental results of the UV-vis and H NMR titration, which
shows the increasing need of 1.2, 1.4, and 2.0 equivalents of TFA
for the first protonation of 2, 1, and 3, respectively. It seems that
the electron-donating group of amino in the phenyl ring can
enhance the electronegativity of the five-membered ring in 2,
even though the enhancement is not strong enough to make a big
difference compared with that of 1. However, the electron-
accepting group of nitro in the seven-membered ring can really
reduce a great deal of the electronegativity of the five-membered
ring in 3, making it the hardest for the first protonation among
the three molecules. Similarly, the NBO value of N5' of 2 is
larger than that of 3, also in good accordance with the UV-vis
titration results. Again, the electron-accepting group of nitro in
the seven-membered ring has big influence in reducing the
electronegativity of the five-membered ring.
In view of the optimized geometries and electronic structures
of H+·2 and 2H+·2 (diprotonated 2), monoprotonation of 2
significantly narrows the energy gap from 2.97 eV (neutral 2) to
2.39 eV (Table 2), corresponding to the maximum absorption
wavelength of 516 nm (oscillator strength 0.4575, Table S7).
This is partially resulted from the LUMO energy distribution,
which is better delocalized over the whole conjugated system.
However, diprotonation of 2 obviously leads to a more lowered
HOMO energy level, resulting in an increased energy gap of 4.69
eV, corresponding to a maximum absorption wavelength of 334
nm (oscillator strength 0.4287, Table S8). The calculated results
fit well with the UV-vis absorption experimental results, which
are mainly attributed to the decreased intramolecular charge
transfer as can be seen from the HOMO and LUMO charge
distribution (Fig. 8).
Table 3
The calculated NBO values of nitrogen atoms for 1, 2, and 3.
Sitesa
N1
1
2
3
Similarly, the calculation results of the optimized geometries
and electronic structures of H+·3 indicate that the energy gap is
narrowed from 2.35 eV (neutral 3) to 2.11 eV (Table 2),
corresponding to the maximum absorption wavelength of 571 nm
(oscillator strength 0.5988, Table S9). The corresponding results
of 2H+·3 (diprotonated 3, Fig. S8) show a more lowered HOMO
energy level, resulting in an increased energy gap of 3.71 eV,
corresponding to a maximum absorption wavelength of 358 nm
−0.512
−0.512
--
−0.514
−0.514
−0.430
−0.507
−0.507
−0.419
N3
N5'
a Sites according to Scheme 1.
2.7. Molecular geometries

7
Table 4
of protonated 1 and 3 were not obtained. It is possible that the
changes in molecular polarity lead to
crystallization.
ACCEPTED MANUSCRIPT
a difference in
Optimized geometry parameters for 2 and protonated 2.
Molecules
Bond length
(angstrom)
Angle
(degree)
Cα-
Dihedral
(degree)
2.9. Protonation mechanism
N1-C2
C2-C1'
N1-C2- N1-C2-
Cα-N1
N3
C1'-C2'
N1-C2
103.9
109.2
108.4
5.3
2
H+·2
2H+·2
ꢁ1a
ꢁ2b
1.34
1.37
1.37
0.03
0.03
1.36
1.39
1.38
0.03
0.02
1.45
1.43
1.47
-0.02
0.02
116.0
110.1
111.5
-5.9
-0.09
-0.03
-0.05
0.06
4.5
-4.5
0.04
Scheme 3. Proposed protonation mechanisms of 1, 2, and 3.
a ꢁ1= H+·2 − 2.
b ꢁ2=2H+·2 − 2.
1
According to the UV-vis titration, H NMR titration, and the
theoretical calculation results, the protonation mechanisms of 1, 2,
and 3 could be proposed as shown in Scheme 3. 1 kept its
coplanar and 16-π-conjugated structure during its sole
monoprotonation. 2 and 3 both went a two-step protonation with
their coplanarity and 16-π-conjugated structure sustained. Hence,
it could be summarized that 2-phenyl-1,3-diazaazulene is a
special backbone, which is planar and π conjugated over the
whole skeleton, and it could always keep its coplanarity and
conjugated skeleton during protonation, different from many
reported structures.43,17
The selected optimized geometry parameters of 1, 2, 3 and
their protonated derivatives are listed in Table 4 (2), Table S11
(1), and Table S12 (3), respectively. 2 is also selected for the
explanation. For 2 and its protonated ones, the dihedral angles of
N1-C2-C1'-C2' which determine the planarity of the diazaazulene
ring and the phenyl ring are between 0.06 to -0.09°, indicating
the diazaazulene part and the phenyl part is coplanar for either
the neutral 2 or the two protonated 2. Meanwhile, the bond
lengths of Cα-N1, N1-C2, as well as C2-C1' changed less than
0.03 angstrom, and the angles of Cα-N1-C2 and N1-C2-N3
exhibited ꢁ values less than 5.9°. Very similar results were also
obtained for those of 1 and 3 (Table S11-S12). Thus, it can be
concluded that either monoprotonation or diprotonation of 1, 2,
and 3 has little influence on their molecular skeleton geometries,
indicating that the coplanar and 16-π-conjugated structure
changed little during protonation. It is mainly because that during
the protonation, the hybridization of the protonated diazaazulene-
nitrogen atom changed from sp2 to sp3, inducing a smaller twist
between rings and keeping its coplanarity as well as an
unaffected conjugation structure.
3. Conclusion
In summary, three DAA derivatives 1, 2, and 3 were
synthesized and their protonation behaviours and mechanisms
towards TFA were investigated systematically via UV-vis
titration, and 1H NMR titration experiments, as well as theoretical
calculations. As supposed, 1, 2, and 3 all displayed an initial
nitrogen-protonation (nitrogen of the diazaazulene ring) response
to the TFA stimulus, resulting in a less tendency to weaken the
conjugation degree of the molecular skeleton compared with their
azulene-based analogues. Besides, both 2 and 3 performed
further subsequent protonation responses at the nitrogen atom of
the amino group to the large amount of TFA up to thousands of
equivalents. During the initial protonation process, the HOMO-
LUMO energy gaps were narrowed. However, the subsequent
protonation of either 2 or 3 induced great increase of HOMO-
LUMO energy gaps, conducing to the bleaching of the coloured
solutions. Notably, upon protonation, the molecular electronic
structures of 1, 2, and 3 were finely tuned, while the molecular
coplanarity and 16-π conjugation were sustained, which is
different from many reported chromophores. Further research is
undertaken for the response properties of films based on these
structures, which might show promising prospects in chemical
sensors and optoelectronic materials.
2.8. Single crystal structure of monoprotonated 2
4. Experimental section
Fig. 9. Single crystal molecular structure of monoprotonated 2.
4.1. Materials
The molecular structure of H+·2·CF3COO- is illustrated as
Oak Ridge Thermal Elliposoid Plot (ORTEP) diagram in Fig. 9.
The crystal data of H+·2·CF3COO- are listed in the supporting
information (Table S13 and Table S14). Single crystal X-ray
result indicates that one proton is attached to the nitrogen atom of
the diazaazulene ring, forming a hydrogen bond (N2—H2···O2,
with a distance of 2.683 Å). The molecular structure of
H+·2·CF3COO- shows that the diazaazulene ring and the benzene
ring are almost coplanar with a dihedral angle of 1.4 (2)° (the
torsion angle for N2—C8—C9—C14), implying that 2 kept its
16-π-conjugated and coplanar structure during protonation.
However, although a lot of solvents have been tried, the crystals
All reagents were purchased from J&K (China) and used as
received unless other mentioned. Trifluoroacetic acid (99.5%)
was purchased from Sigma-Aldrich. Dichloromethane (spectral
grade) was purchased from J&K (China) for UV-vis experiments.
All reagents were weighed and handled in room temperature.
Flash column chromatography was performed over silica gel
200-300. All the compounds were purified by flash column
chromatography and characterized by H NMR, 13C NMR, and
1
HRMS, respectively (Please see Supporting Information).
4.2. Instrumentation

8
Tetrahedron
ACCEPTED MANUSCRIPT
1H NMR spectra were recorded on Bruker 400 or 600 MHz
× 1 cm quartz cuvette. After each addition, the solution was
and the chemical shifts were reported in parts per million (δ)
relative to the internal solvent signals (7.26 ppm for CDCl3, and
2.50 ppm for DMSO-d6). 13C NMR spectra were obtained at
Bruker 100 or 150 MHz and referenced to the internal solvent
signals (central peak 77.0 ppm for CDCl3). The peak patterns are
indicated as follows: s, singlet; d, doublet; dd, doublet of doublet;
t, triplet; q, quartet; m, multiplet. The coupling constants are
reported in Hertz (Hz). APEX II (Bruker Inc.) was used for ESI-
MS.
mixed well and then the UV-vis absorption spectrum was
measured. (c) 5, 8, 12, 20, 30, and 60 ꢀL of TFA (15000 M in
dichloromethane) was added successively to 3 mL of 3 (15 ꢀM in
dichloromethane) in a 1 cm × 1 cm quartz cuvette. After each
addition, the solution was mixed well and then the UV-vis
absorption spectrum was measured.
4.6. Solution preparation for 1H NMR titration studies
Titration of 1 and 2: 0, 2.5, 5.0, 7.5, 10, 15, 20, 25, 30, 40, 50,
and 60 ꢀL of TFA (5000 M in CDCl3) was added successively to
600 ꢀL of 1 or 2 (50.0 mM in CDCl3) in a NMR test tube. After
each addition, the solution was mixed well and then the 1H NMR
spectrum was measured.
UV-vis absorption spectra were recorded on Varian Cary 50
UV-vis spectrophotometer with dichloromethane as solvent.
1H NMR protonation titration spectra were recorded on
Bruker 400 MHz with CDCl3 as solvent.
Titration of 3: 0, 2.5, 5.0, 7.5, 10, 15, 20, 25, 30, 40, 50, and
60 ꢀL of TFA (2000 M in CDCl3) was added successively to 600
ꢀL of 3 (20.0 mM in CDCl3) in a NMR test tube. After each
addition, the solution was mixed well and then the H NMR
spectrum was measured.
4.3. 2-Phenyl-1,3-diazaazulene (1)
Benzamidine hydrochloride (235 mg, 1.5 mmol) was
dissolved in 40 mL of anhydrous ethanol and stirred for 5
minutes. Then 2-methoxy tropone (136 mg, 1.0 mmol) was added
in, followed by the addition of t-BuOK (168 mg, 1.5 mmol). The
resulting mixture was refluxed for 3 hours. The solvent was
removed, and a white solid was obtained which was subjected to
flash column chromatography on silica gel with CH3OH: CH2Cl2/
1:50 as eluents. After removing of the solvent, 161 mg of white
powder was obtained. Yield: 78.1%. 1H NMR (600 MHz,
CDCl3): δ 8.82 (d, J = 10.0 Hz, 2H), 8.67-8.62 (m, 2H), 8.07 (t, J
= 9.8 Hz, 2H), 8.01 (t, J = 9.7 Hz, 1H), 7.57-7.52 (m, 3H); 13C
NMR (150 MHz, CDCl3): δ 176.84, 164.00, 137.46, 134.22,
133.08, 131.56, 129.47, 128.83, 77.21, 77.00, 76.79; HRMS
(ESI): calcd for C14H11N2, m/z 207.0917 [M+H+]; found, m/z
207.0915.
1
4.7. DFT calculation
All calculations were carried out using the Gaussian 09
quantum chemistry program package.40 DFT calculations on the
geometries optimization and TD-DFT calculations on electronic
structures were both performed at B3LYP/6-31G* level for the
singlet ground states (gas phase). The zero-point vibrational
energies (ZPE) were scaled according to Wong (0.9804).44
Acknowledgments
This work was supported by the Research Funds of Renmin
University of China (15XNLQ04).
4.4. 2-(P-N,N-dimethylaminophenyl)-1,3-diazaazulene (2)
References and notes
4-N,N-dimethylaminno-benzamidine hydrochloride (298 mg,
1.5 mmol) was dissolved in 40 mL of anhydrous ethanol and
stirred for 5 minutes. Then 2-methoxy tropone (136 mg, 1.0
mmol) was added in, followed by the addition of t-BuOK (168
mg, 1.5 mmol). The resulting mixture was refluxed for 3 hours.
The crude product was subjected to flash column
chromatography on silica gel with n-hexane: ethyl acetate/ 5:1 as
eluents. After removing of the solvent, 174 mg of red powder
1. Ostroverkhova, O. Chem. Rev. 2016, 116, 13279-13412.
2. Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.;
Mackay, K.; Friend, R. H.; Burn, P. L.; Holmes, A. B. Nature 1990, 347,
539-541.
3. Mathew, S.; Yella, A.; Gao, P.; Humphry-Baker, R.; Curchod, B. F. E.;
Ashari-Astani, N.; Tavernelli, I.; Rothlisberger, U.; Nazeeruddin, M. K.;
Graetzel, M. Nat. Chem. 2014, 6, 242-247.
4. Zeng, W.; Cao, Y.; Bai, Y.; Wang, Y.; Shi, Y.; Zhang, M.; Wang, F.; Pan,
C.; Wang, P. Chem. Mater. 2010, 22, 1915-1925.
1
was obtained. Yield: 70.0%. H NMR (400 MHz, DMSO-d6): δ
8.62 (d, J = 9.7 Hz, 2H), 8.38 (d, J = 8.9 Hz, 2H), 8.09 (t, J = 9.5
Hz, 2H), 8.00 (t, J = 9.8 Hz, 1H), 6.86 (d, J = 9.0 Hz, 2H), 3.06
(s, 6H); 13C NMR (150 MHz, CDCl3): δ 177.32, 164.40, 152.79,
135.22, 134.09, 131.64, 131.18, 120.42, 111.79, 77.28, 77.06,
76.85, 40.11; HRMS (ESI): calcd for C16H16N3, m/z 250.1339
[M+H+]; found, m/z 250.1326.
5. Verbiest, T.; Houbrechts, S.; Kauranen, M.; Clays, K.; Persoons, A. J.
Mater. Chem. 1997, 7, 2175-2189.
6. Bansal, A. K.; Hou, S.; Kulyk, O.; Bowman, E. M.; Samuel, I. D. W.
Adv. Mater. 2015, 27, 7638-7644.
7. Wang, T.; Sun, G.; Wang, M.; Zhou, B.; Fu, J. ACS Appl. Mater.
Interfaces 2015, 7, 21295-21304.
8. Yao, Z.; Hu, X.; Huang, B.; Zhang, L.; Liu, L.; Zhao, Y.; Wu, H.-C. ACS
Appl. Mater. Interfaces 2013, 5, 5783-5787.
4.5. Solution preparation for UV-vis titration studies
9. Kwon, M. S.; Gierschner, J.; Seo, J.; Park, S. Y. J. Mater. Chem. C 2014,
2, 2552-2557.
Titration by 0 – 6 equivalents of TFA: 0, 2, 4, 6, 8, 10, 12, 14,
16, 18, 20, 25, 30, 35, 40, 50, and 60 ꢀL of TFA (2.0 M for 1, 1.5
M for 2 and 3, all in dichloromethane) was added successively to
3 mL of each compound (20 M for 1, 15 ꢀM for 2 and 3, all in
dichloromethane) in a 1 cm × 1 cm quartz cuvette. After each
addition, the solution was mixed well and then the UV-vis
absorption spectrum was measured.
10. Lin, M.-J.; Fimmel, B.; Radacki, K.; Wuerthner, F. Angew. Chem. Int.
Ed. 2011, 50, 10847-10850.
11. Cai, K.; Yan, Q.; Zhao, D. Chem. Sci. 2012, 3, 3175-3182.
12. Zakavi, S.; Omidyan, R.; Talebzadeh, S. RSC Adv. 2016, 6, 82219-
82226.
13. Zadeh, E. H. G.; Tang, S.; Woodward, A. W.; Liu, T.; Bondar, M. V.;
Belfield, K. D. J. Mater. Chem. C 2015, 3, 8495-8503.
14. Kurita, Y.; Kubo, M. J. Am. Chem. Soc. 1957, 79, 5460-5463.
15. Muranaka, A.; Yonehara, M.; Uchiyama, M. J. Am. Chem. Soc. 2010,
132, 7844-7845.
16. Murai, M.; Ku, S.-Y.; Treat, N. D.; Robb, M. J.; Chabinyc, M. L.;
Hawker, C. J. Chem. Sci. 2014, 5, 3753-3760.
17. Amir, E.; Murai, M.; Amir, R. J.; Cowart, J. S., Jr.; Chabinyc, M. L.;
Hawker, C. J. Chem. Sci. 2014, 5, 4483-4489.
18. Tang, T.; Lin, T. T.; Wang, F. K.; He, C. B. J. Phys. Chem. B 2015, 119,
8176-8183.
Titration by large equivalents, of TFA: (a) 5, 10, and 50 ꢀL of
TFA (2000 M in dichloromethane) was added successively to 3
mL of 1 (20 ꢀM in dichloromethane) in a 1 cm × 1 cm quartz
cuvette. After each addition, the solution was mixed well and
then the UV-vis absorption spectrum was measured. (b) 2, 3, 4, 6,
10, and 20 ꢀL of TFA (15000 M in dichloromethane) was added
successively to 3 mL of 2 (15 ꢀM in dichloromethane) in a 1 cm

9
19. Koch, M.; Blacque, O.; Venkatesan, K. J. Mater. Chem. C 2013, 1,
33. Rao, D.-L.; Sun, P.-L.; Qin, Y.; Zhang, P.; Guo, Z.-X. Mater. Lett.
2017, 205, 182-185.
ACCEPTED MANUSCRIPT
7400-7408.
20. Birzan, L.; Cristea, M.; Draghici, C. C.; Tecuceanu, V.; Maganu, M.;
Hanganu, A.; Razus, A. C.; Buica, G. O.; Ungureanu, E. M. Dyes
Pigment. 2016, 131, 246-255.
21. Gai, L.; Chen, J.; Zhao, Y.; Mack, J.; Lu, H.; Shen, Z. RSC Adv. 2016, 6,
32124-32129.
22. Tang, T.; Chi, H.; Lin, T. T.; Wang, F. K.; He, C. B. Phys. Chem. Chem.
Phys. 2014, 16, 20221-20227.
23. Wang, F.; Lin, T. T.; He, C.; Chi, H.; Tang, T.; Lai, Y.-H. J. Mater. Chem.
2012, 22, 10448-10451.
34. Jinguji, M.; Ashizawa, M.; Nakazawa, T.; Tobita, S.; Hikida, T.; Mori, Y.
Chem. Phys. Lett. 1985, 121, 400-404.
35. Feraud, G.; Esteves-Lopez, N.; Dedonder-Lardeux, C.; Jouvet, C. Phys.
Chem. Chem. Phys. 2015, 17, 25755-25760.
36. Zucchero, A. J.; Tolosa, J.; Tolbert, L. M.; Bunz, U. H. F. Chem. Eur. J.
2009, 15, 13075-13081.
37. Kim, J.; Yennawar, H. P.; Lear, B. J. Dalton Trans. 2013, 42, 15656-
15662.
38. Freire, F.; Cuesta, I.; Corzana, F.; Revuelta, J.; Gonzalez, C.; Hricovini,
M.; Bastida, A.; Jimenez-Barbero, J.; Asensio, J. L. Chem. Commun.
2007, 43, 174-176.
24. Tang, T.; Lin, T. T.; Wang, F. K.; He, C. B. Polym. Chem. 2014, 5, 2980-
2989.
25. Tang, T.; Lin, T.; Wang, F.; He, C. Phys. Chem. Chem. Phys. 2016, 18,
18758-18766.
39. Dueggeli, M.; Christen, T.; von Zelewsky, A. Chem. Eur. J. 2005, 11,
185-194.
26. Murai, M.; Ku, S.-Y.; Treat, N. D.; Robb, M. J.; Chabinyc, M. L.;
Hawker, C. J. Chem. Sci. 2014, 5, 3753-3760.
27. Lewis, J. W.; Nauman, R. V. Can. J. Chem. 1985, 63, 2081-2085.
28. Braun, J. R.; Lin, T.-S.; Burke, F. P.; Small, G. J. J. Chem. Phys. 1973,
59, 3595-3599.
29. Nozoe, T.; Mukai, T.; Murata, I. J. Am. Chem. Soc. 1954, 76, 3352-3353.
30. Ma, X.-H.; Fu, L.-M.; Zhao, Y.; Ai, X.-C.; Zhang, J.-P.; Han, Y.; Guo,
Z.-X. J. Polym. Sci., Part B: Polym. Phys. 2011, 49, 649-656.
31. Bravaya, K. B.; Grigorenko, B. L.; Nemukhin, A. V.; Zhu, Y.-J.; Zhang,
J.-P. THEOCHEM 2008, 855, 40-44.
40. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; et al. Gaussian 09,
Revision A.02. Wallingford CT, Gaussian, Inc., 2009.
41. Russo, N.; Toscano, M.; Grand, A.; Mineva, T. J. Phys. Chem. A 2000,
104, 4017-4021.
42. Glendening, E. D.; Landis, C. R.; Weinhold, F. Wiley. Interdiscip. Rev.
Comput. Mol. Sci. 2012, 2, 1-42.
43. Yamaguchi, K.; Murai, T.; Guo, J.-D.; Sasamori, T.; Tokitoh, N.
ChemistryOpen 2016, 5, 434-438.
44. Wong, M. W. Chem. Phys. Lett. 1996, 256, 391-399.
32. Zhu, Y.-J.; Qin, A.-J.; Fu, L.-M.; Ai, X.-C.; Guo, Z.-X.; Zhang, J.-P.; Ye,
C. J. Mater. Chem. 2007, 17, 2101-2106.