3.1. Photophysical and AIPE Properties
The UV-Vis absorption spectra and normalized emission spectra at room temperature of complexes
1 and
2 in THF solution are shown in
Figure S3. The photophysical data of both complexes can be found in
Table S1. Similar to most Ir(III) complexes, both
1 and
2 exhibit strong absorption bands below 300 nm, which belong to the typical ligand-centered (
1LC(
1π-π*)) transitions. The lower energy absorption between 300-430 nm is attributed to a combination of metal-to-ligand charge transitions (
1MLCT/
3MLCT), ligand-to-ligand charge transitions (
1LLCT/
3LLCT), and ligand-centered (
3LC(
3π-π*)) leaps. The normalized emission spectra indicate that the maximum emission wavelength and peak shape of
2 remain largely unchanged compared to those of
1. The phosphorescence lifetimes of
1 and
2 are in the range of 1.77-2.18 μs at room temperature (
Figure S4). The phosphorescence quantum yields of
1 and
2 were measured in degassed CH
2Cl
2 using Ir(ppy)
2(acac) (0.34) as a standard [
32]. The results indicate that phosphorescence quantum yields of 0.18 for
1 and 0.27 for
2, respectively. Incorporating an electron-donating group (5-CH
3) onto the pyridine ring of the complex
2 resulted in an increased phosphorescence quantum yield.
Due to the large conjugation plane of carbazolyl group, molecular motion may be greatly restricted when close stacking occurs. As a result, there is a high probability that
1 and
2 will exhibit AIPE property. To confirm this conjecture, suspensions of
1 and
2 in THF/H
2O with various water fractions were prepared, and their emission spectra and UV-vis absorption spectra were recorded as shown in
Figure S5.
Figure 2a,b show that the emission intensity of
1 or
2 increases gradually as the water fraction increases from 0 to 60%. Further increasing the water fraction resulted in the maximum emission intensity of
1 and
2 at 70% and 90%, respectively, indicating a significant AIPE phenomenon. The ratio plot (
Figure 2c) shows that
2 exhibits a higher AIPE property than
1 under the same experimental conditions, which is similar to the findings reported by Yang et al. in 2011 [
33]. The introduction of a methyl group into the complex is believed to change the arrangement of
2 in the aggregated state, resulting in tighter intermolecular interaction to restrict intramolecular movement efficiently. This leads to enhanced luminescence of
2 at high water fractions.
3.2. Detection of PA
2 exhibits impressive AIPE property in THF/H
2O, suggesting its potential for detection of PA in aqueous media. We further conducted the luminescence quenching experiments of
2. As shown in
Figure 3a,b, the emission intensity of
2 decreased progressively as the concentration of PA increased. The quenching efficiency of
2 was 28% when the concentration of PA reached 10 μM. The quenching efficiency of
2 increased significantly to 98% after the addition of PA solution with a concentration of 300 μM (30 equiv.).
The quenching constant (
KSV) is usually used as an index to the sensitivity of a sensor to detect PA. To obtain the
KSV of
2 for PA, a Stern-Volmer plot of
2 was constructed using the ratio of the luminescence intensity without PA (
I0) to the luminescence intensity with PA (
I)
vs. the concentration of PA (
Figure 4). The Stern-Volmer plot displays excellent linearity within the concentration of PA range from 0-10 μM, which is indicative of static quenching [
34,
35]. Within the concentration of PA range from 0-500 μM, there is a clear non-linear relationship due to the combination of static and dynamic quenching [
36,
37]. The sensitivity of
2 for PA is assessed utilizing the Stern-Volmer equation:
I0/
I =
KSV[Q] + 1 [
38]. In the concentration range from 0-10 μM, a linear fit is used to calculate the
KSV of
2 to be 37320 M
-1. This value is higher than those of previously reported neutral Ir(III) complexes with carbazolyl groups for PA detection [
15,
16]. In addition, the LOD of
2 for PA can be calculated using the formula LOD = 3
σ/
K, where
σ represents the standard deviation of 11 blank samples of
2 (
Table 1), and
K represents the slope of the linear relationship between the luminescence intensity
I and PA concentration [
39].
Selectivity is deemed a key factor influencing the probe performance. Therefore, the luminescence response studies on several common analytes were conducted, including NB, 1,3-DNB, NM,
p-Cresol,
m-Cresol, Phenol, and MEHQ. The emission spectra can be seen in
Figure 5a.
Figure 5c shows that when PA was added, the luminescence quenching efficiency of
2 was higher than that of the other analytes, demonstrating excellent selectivity.
In order to further explore selectivity and anti-interference capacity of
2 and to expand scope of its application for the detection of PA, various ionic compounds were used for the emission spectroscopic studies, including CaCl
2, MnCl
2, FeCl
2, NiCl
2, ZnCl
2, CoCO
3, NaHCO
3, MgSO
4, CuSO
4, KF, KBr, and CH
3COONa. Adding solutions of various compounds did not lead to noticeable alteration in the emission spectra or emission intensity of
2 (
Figure 5b). This suggests that the addition of ions has small impact on the luminescent property of
2 in THF/H
2O.
Competition experiments were carried out by introducing 30 equiv. of PA solution into suspensions containing various analytes or ionic compounds. As shown in
Figure 5c,d, the luminescence quenching efficiency of
2 by PA was found to be almost unaffected in the presence of various competing compounds. Therefore,
2 shows outstanding anti-interference capacity when used to detect PA.
Water resources represent a crucial component of the ecological environment and are utilized in a multitude of social production and life processes. PA is readily soluble in hot water and has the potential to pollute water sources. Consequently, it is of paramount importance to detect the presence of PA in common water samples. The emission spectra of
2 before and after the addition of PA in various water samples as shown in
Figure 6a. In comparison to the luminescence observed within deionised water, the emission spectra of
2 in four common water samples exhibit similar shapes, yet distinct emission intensity. The quenching efficiency of
2 for detecting PA in different water samples is shown in
Figure 6b, and the results did not differ significantly. The spectral analysis indicates that efficiency of luminescence quenching of
2 by PA is considerably higher than that of the impurities present in four common water samples.
3.1. Sensing Mechanism
To investigate the sensing mechanism of
2 for detecting PA, we calculated the LUMOs and highest occupied molecular orbitals (HOMOs) of
2, adduct (
2 + PA), and PA.
Figure 7a shows that the LUMO energy of PA is -3.49 eV, which is lower than that of
2. This implies that
2 and PA are capable of undergoing PET. The excited-state electrons in the LUMO of
2 are easily transferred to the LUMO of PA rather than returning directly to its own HOMO, resulting in the quenching of the luminescence of
2. The energy gap of
2 is higher than that of adduct (
2 + PA), indicating the adduct is much more stable. Förster Resonance Energy Transfer (FRET) does not occur during sensing because there is no overlap between the emission spectrum of
2 and the absorption spectrum of PA (
Figure 7b). Additionally, the normalized emission spectra of
2 remain almost unchanged after the addition of PA at different concentrations (
Figure S10), proving that no other emitting species are formed during the quenching process [
15]. Thus, the sensing mechanism of
2 for PA is proposed to be PET.