T. Eom and A. Khan
Dyes and Pigments 188 (2021) 109197
with synthetic ease. In considering azo compounds with a blue
appearance (absorption in the red region), our design differs from
Wooley’s design [22] in terms of structural symmetry and from com-
mercial azo blue/trypan blue in terms of compactness of the structure
and underlines the importance of multiple donor/acceptors moieties and
their relative arrangement on the aromatic scaffold.
To further red-shift the absorption spectrum, we hypothesized that
protonation of the azo bond under acidic conditions would be a useful
strategy [23]. However, we observed that the addition of a drop of hy-
drochloric acid (solution pH = 1–2) shifted the absorption spectrum to
the blue region (Fig. 1). The shift could be reversed by neutralization of
the acid in the solution. This meant that we were observing a rare
negative halochromism property in the monoazobenzene scaffolds [24].
The extent of the blue-shift depended upon the molecular structure.
Compound 1 exhibited a minimum shift of 18 nm while compound 6
displayed a maximum shift of 114 nm.
Scheme 1. Synthesis of 1–4.
synthetic steps beginning with commercially available starting materials
(Figs. S1-S6). Compound 4 [21] with only one donor group could be
prepared in one synthetic step from commercially available N-ethyl-
anilinoethanol in 68% isolated yield (Figs. S7-S8). The hydroxyl group
was included in the molecular structure to aid water-solubility of the
compounds. Despite this, use of dimethylsulfoxide (DMSO) was neces-
Next, changes in the absorption spectra as a function of pH was
studied with the help of buffer solutions (Fig. 2 and Figs. S17-S21). The
pH was varied from near neutral (7.5) to acidic (2.5). This study pro-
duced complex results. For example, compounds 1 and 2 displayed a
complete transition to the shorter wavelengths. However, in the case of
1, the transition was very fast and a change of 1 pH unit (7.5–6.5) was
enough to completely shift the spectra. Compound 2 had a gradual
change and each unit of the pH change produced a significant change in
the absorption spectrum and a complete shift was observed only at pH 4.
Compound 3, 5, and 6 exhibited incomplete and slow transition and
resisted any shift until the pH became lower than 4. Interestingly,
compound 4 displayed no change in its absorption spectrum. Except for
compound 4 (Fig. S19), a well-defined isosbestic point was observed in
all cases indicating that no secondary reactions occurred under acidic
conditions.
sary as
a co-solvent to prepare aqueous solutions for UV–Vis
spectroscopy.
The UV–Vis absorbance spectra revealed many interesting features
(Fig. 1). Compound 1 having the nitro group at the ortho and the cyano
group at the para positions to the azo bond absorbed at 484 nm. How-
ever, compound 2 in which their positions were switched showed a red-
shift of 57 nm. This meant that the electron withdrawing capabilities of
the cyano and the nitro groups were optimum at the ortho and para
positions, respectively. Interestingly, however, compound 3 with only
one methoxy group showed the maximum shift of 71 nm. Compound 4
with no methoxy group showed a shift of 58 nm.
The fact that compound 3 exhibited the maximum red-shift indicated
that the second methoxy group was not required at the ortho position.
Therefore, inspired by the studies of Woolley and coworkers [22],
compound 5 and 6 were prepared (Scheme 2 and Figs. S9-S16). In these
compounds one of the methoxy group is located at the meta position to
the azo bond. The difference between 5 and 6 is that in 6 the amine
donor is part of a five-membered ring. From Hallas studies [18], a
red-shift is expected with the five-membered cyclic donor. Indeed, the
λmax in compounds 5 and 6 are located at 584 and 604 nm, respectively.
These results indicated that by properly placing the acceptors and the
donors, red absorbing azo compounds could be successfully accessed
The stability of the produced azo compounds against a biologically
reducing agent was studied next (Figs. S22-S27). The compounds were
exposed to 10 mM glutathione concentration as is typically found in
biological systems. Here, compound 1 is reduced completely. Com-
pounds 2 and 5 showed a small reduction in the absorption intensity
indicating that a small portion of the molecules reduced under these
conditions. Compound 3 and 4 remained unchanged and indicated their
stability towards 10 mM glutathione. The solution of compound 6 was
observed to form a fine precipitate upon the addition of glutathione.
Therefore, the reduction in absorption intensity in 6 can be presumed to
be due to their reduction in solubility rather than the azo cleavage
reaction.
To further examine sensitivity to glutathione, a much higher con-
centration of 100 mM was used in the next set of experiments. This study
indicated that compound 2 could not withstand the higher concentra-
tion of glutathione and was reduced completely. Compounds 3–5
showed a small reduction in absorption intensity indicating some mo-
lecular decomposition. Compound 6 once again showed a much greater
reduction in the absorption intensity. However, as mentioned before,
this compound experienced poor solubility in the presence of gluta-
thione. Compounds 3–4, therefore, seem to be relatively robust towards
biological reducing conditions created by glutathione. From these re-
sults, it appears that an ortho nitro substituent (as in compound 1) helps
in the reduction of the azo bond presumably by stabilizing the hydra-
zobenzene intermediate (Fig. 3) with the oxygen atoms. To examine this
hypothesis, a new compound, 7, was prepared (Fig. 4 and Figs. S28-
S29). Compound 7 cleaved with 10 mM glutathione concentration to a
larger extent than 2 but to a lower extent than 1 (Fig. S30). This result
indicated that the ortho nitro substituent does indeed help to cleave the
azo bond, however, the second acceptor is also an important parameter.
The best sensitivity is obtained when the molecule contains three donors
and two acceptors and the ortho acceptor is a nitro substituent. A com-
parison with a previously known compound 8 which is observed to be
less sensitive to 1, 2, and 7 and more sensitive to 3 and 4 reinforces this
conclusion (Fig. S31) [16].
Fig. 1. UV–Vis spectra of compounds 1–4 in 2:8 DMSO:Water mixture and 5–6
in 3:7 DMSO:Water mixture before (black line) and after (red line) acidification
with hydrochloric acid.
2