Aggregation-Induced Emission-Active Iridium(III) Complexes for Sensing Picric Acid in Water

Abstract

Herein, two new iridium(III) complexes, namely Ir2 and Ir3, with a phenyl or triphenylamine (TPA) moiety at the 4-position of the phenyl ring at 2-phenylbenzothiazole, have been synthesized, and their emission properties have been studied systematically compared with the non-substituted complex Ir1. These three complexes exhibit aggregation-induced emission (AIE) in H₂O/CH₃CN. The TPA-substituted complex Ir3 shows the highest AIE activity. All complexes can be used as sensors to detect picric acid (PA) in water. The Stern–Volmer constant (K_SV) of Ir3 for the detection of PA was determined to be 1.96 × 10⁶ M⁻¹, with a low limit of detection of 2.52 nM. Proton nuclear magnetic resonance spectra, high-resolution mass spectrometry analysis, and density function theory calculations confirm that the emission quenching mechanism of Ir3 is caused by photo-induced electron transfer. Furthermore, the efficient detection of PA in natural water proves that Ir1–Ir3 can be used as promising sensors in the natural environment. These results suggest that the AIE-active iridium(III) complexes can be used to detect PA under environment-friendly conditions.

1. Introduction

Among nitroaromatics, picric acid (PA) containing three nitro groups shows superior explosive power and is extensively used to manufacture high explosives, which are a severe threat to society, human life, and property [1,2,3]. Additionally, PA plays an essential role in the leather, fuel manufacturing, pesticide, chemical fiber, pharmaceutical, and dye industries [4,5]. At the same time, a large amount of wastewater containing PA produced during its manufacturing, carriage, and utilization may cause water and soil pollution [6]. Therefore, it is highly desirable to develop a sensitive and selective detection method for PA.

Various analytical methods have been developed for explosives detection, including mass spectrometry [7], electrochemical methods [8], quartz crystal microbalance [9], and surface-enhanced Raman scattering [10]. However, these techniques usually need complex sample pretreatment and expensive and sophisticated instrumentation. In contrast, photoluminescence has been demonstrated to be an ideal analytical tool for trace explosives detection owing to its high sensitivity, operational simplicity, and real-time monitoring capability [11,12,13,14]. However, common luminescent materials frequently suffer from weak luminescence at the solid or aggregated states owing to aggregation-caused quenching (ACQ), which limits the detection of PA in aqueous media. Notably, the aggregation-induced emission (AIE) concept proposed by Ben Zhong Tang and coworkers effectively solves the ACQ problem [15]. Subsequently, many compounds with AIE activities have been designed and synthesized for PA detection [16,17,18,19,20]. Cyclometalated Ir(III) complexes have numerous advantages, such as large Stokes shifts, high photoluminescence quantum yields, long lifetimes, and adequate stability [21,22,23]. Therefore, the development of AIE-active Ir(III) complexes for the detection of PA has become a research focus. However, most Ir(III) complexes reported for the detection of PA are based on 2-phenylpyridine (ppy), 1-phenylpyrazole(ppz), and their derivatives as cyclometalating ligands [24]. More importantly, the relationship between the AIE activity of an Ir(III) complex and its detection efficiency of PA is still elusive.

Although the AIE effect effectively settles the problems of weak emission or emission quenching of the luminescent materials [25] in the water medium caused by the ACQ influence and has made great progress rapidly, it is still a great challenge to control the AIE activity of the luminescent materials. In AIE systems, the most extensively accepted mechanism is the restriction of intramolecular rotations (RIR), which can block the non-radiative pathway and open up the radiative channel. Generally, propeller-like or rotor-like substitutes are often used to activate the AIE property of the luminophores [26]. The nonplanar structure of triphenylamine (TPA) makes it an ideal building block for AIE-active luminogens. In addition, TPA exhibits strong electron-donating properties. Luminescent materials containing a TPA moiety facilitate interaction with electron-withdrawing nitroaromatic compounds [27,28,29]. Very recently, we found that the introduction of a diphenylamino group into the cyclometalating ligand significantly improved the detection efficiency of PA of the corresponding Ir(III) complex [30]. In order to systematically understand the effect of the molecular structure on AIE activity, in this work, two novel AIE-active cationic Ir(III) complexes, namely Ir2 and Ir3 with 2-phenylbenzothiazole derivatives as cyclometalating ligands, have been designed and synthesized (Scheme 1). This substantially broadens the structural diversity of the cyclometalating ligands that are not restricted to ppy or ppz-based ligands. By altering the substituent (phenyl or TPA) at the cyclometalating ligand, the regulation and enhancement of the AIE activities of the Ir(III) complexes have been realized compared with a non-substituted Ir1. Notably, the interaction of Ir(III) complexes with PA can effectively quench their luminescence, facilitating the eminently selective and sensitive detection of PA in a water medium. Moreover, the AIE activities of the Ir(III) complexes are positively correlated with the detection efficiencies of PA. These results provide a new idea for designing AIE-active Ir(III) complexes for PA detection.

2. Materials and Methods

2.1. Materials and Measurements

Unless otherwise stated, all the raw materials were commercially bought and have not been further purified. The solvents were handled according to requirements before use. Cyclometalating ligands (L2 and L3) were prepared following the reported methods [31]. Proton nuclear magnetic resonance (¹H NMR) spectra were recorded using a 400 MHz Varian Unity Inova spectrophotometer. Carbon-13 nuclear magnetic resonance (¹³C NMR) spectra were recorded using a 500 MHz Bruker AVANCE III spectrophotometer. High-resolution mass spectrometry (HRMS) was recorded using an Agilent-G6224 or an LTQ Orbitrap XL spectrometer equipped with an ESI source in a positive model. Fourier-transform infrared (FT-IR) spectra were recorded by Nicolet 6700 spectrometer using the KBr pellet method. The Ultraviolet-visible (UV-vis) absorption spectra were recorded using a Lambda 750s spectrophotometer. The emission spectra were recorded using a HITACHI F-7000 fluorescence spectrophotometer. Photoluminescence quantum yields (Φ_PL) were calculated compared with [Ir(ppy)₂(acac)] (Φ_PL = 0.34 in CH₂Cl₂, under deoxygenated conditions). Phosphorescence lifetimes were measured using an Edinburgh FLS920 spectrometer. Density function theory (DFT) calculations were performed by PBE0/genecp. The LanL2DZ basis set was applied for iridium atoms, while the 6-311++G** basis set was used to handle all other atoms. The polarizable continuum model (PCM) model was adopted to consider the solvent effects. All of these calculations were carried out with the Gaussian 16 C.01 procedures [32].

2.2. Synthesis of the Ir(III) Complexes

IrCl₃·3H₂O (0.200 mmol, 70.5 mg) was reacted with 2.50 equiv. cyclometalating ligands (0.500 mmol, L1 106 mg, L2 144 mg, and L3 227 mg) in a mixture of 2-ethoxyethanol and water (6.00 mL/2.00 mL) at 110 °C for 24 h under an N₂ atmosphere to obtain a cyclometalated iridium bridged-chloride dimer. In the case of no further depuration, the dimer and 1,10-phenanthroline (0.600 mmol, 108 mg) were added to 2-ethoxyethanol at 120 °C under nitrogen for 24 h. As it cooled to room temperature, a 10-fold excess of saturated KPF₆ solution was poured, and the reaction mixture was stirred for 3 h. After adding 15.0 mL of water, the mixture was extracted with CH₂Cl₂ (3 × 15.0 mL). The combined organic layers were dried over Na₂SO₄. The solvent was removed by rotary evaporation, and the crude products were subjected to depuration by column chromatography (methanol/ dichloromethane = 1:100, v/v).

Ir1. Yield: 74%; a yellow solid. ¹H NMR (400 MHz, DMSO-d₆) δ 8.97-8.94 (m, 2H), 8.42-8.41 (m, 2H), 8.35 (s, 2H), 8.17-8.13 (m, 4H), 8.06 (d, J = 7.6 Hz, 2H), 7.30 (t, J = 7.6 Hz, 2H), 7.17 (t, J = 7.6 Hz, 2H), 7.00-6.91 (m, 4H), 6.40 (d, J = 7.6 Hz, 2H), 5.74 (d, J = 8.4 Hz, 2H). ¹³C NMR (125 MHz, DMSO-d₆) δ 181.31, 151.19, 149.77, 148.51, 146.80, 140.18, 139.46, 132.89, 132.04, 131.16, 130.69, 128.35, 127.81, 127.43, 127.01, 125.86, 124.58, 123.19, 116.31. FT-IR (KBr, cm⁻¹): υ = 3057, 1581, 1469, 1447, 1407, 1298, 1267, 838, 754, 724, 557. HRMS (ESI, m/z): calcd. for C₃₈H₂₄IrN₄S₂ [M − PF₆]⁺, 793.1072; found 793.1072.

Ir2. Yield: 52%; a yellow solid. ¹H NMR (400 MHz, DMSO-d₆) δ 8.97 (d, J = 8.0 Hz, 2H), 8.54 (s, 2H), 8.37 (s, 2H), 8.20-8.13 (m, 6H), 7.47 (d, J = 7.6 Hz, 2H), 7.34-7.26 (m, 8H), 7.18 (d, J = 5.2 Hz, 4H), 6.96 (t, J = 7.2 Hz, 2H), 6.59 (s, 2H), 5.83 (d, J = 8.4 Hz, 2H). ¹³C NMR (125 MHz, DMSO-d₆) δ 180.87, 151.52, 150.34, 148.60, 146.81, 143.08, 139.63, 139.47, 139.41, 131.24, 130.73, 130.40, 128.86, 128.38, 128.07, 127.94, 127.54, 127.51, 126.45, 125.89, 124.57, 122.17, 116.37. FT-IR (KBr, cm⁻¹): υ = 3055, 2038, 1583, 1461, 1426, 1263, 840, 757, 723, 557. HRMS (ESI, m/z): calcd. for C₅₀H₃₂IrN₄S₂ [M − PF₆]⁺ 945.1698; found 945.1701.

Ir3. Yield: 49%; a brown solid. ¹H NMR (400 MHz, DMSO-d₆) δ 8.97 (d, J = 8.4 Hz, 2H), 8.52-8.51 (m, 2H), 8.37 (s, 2H), 8.18-8.13 (m, 4H), 8.06 (d, J = 8.0 Hz, 2H), 7.43 (dd, J = 8.0, 1.2 Hz, 2H), 7.31-7.26 (m, 10H), 7.10-7.05 (m, 8H), 6.92 (dd, J = 16.0, 7.6 Hz, 10H), 6.79 (d, J = 8.8 Hz, 4H), 6.59 (d, J = 1.6 Hz, 2H), 5.79 (d, J = 8.4 Hz, 2H). ¹³C NMR (125 MHz, DMSO-d₆) δ 180.72, 151.45, 150.37, 148.60, 147.34, 146.83, 146.60, 142.24, 139.45, 138.96, 132.29, 131.13, 130.73, 129.57, 129.41, 128.38, 127.86, 127.46, 127.23, 125.75, 124.47, 123.58, 122.11, 121.37, 116.26. FT-IR (KBr, cm⁻¹): υ = 3057, 3032, 2040, 1578, 1510, 1489, 1429, 1323, 1283, 840, 755, 724, 696, 557. HRMS (ESI, m/z): calcd. for C₇₄H₅₀IrN₆S₂ [M − PF₆]⁺ 1279.3168; found 1279.3169.

2.3. Preparation of Stock Solutions for AIE and Detection of PA

Stock solutions of Ir(III) complexes were prepared at a concentration of 0.10 mM in CH₃CN. Samples for testing the emission properties were prepared by adding different fractions of deionized water to solutions of Ir(III) complexes. The aqueous suspensions of Ir(III) complexes (10 μM) in H₂O/CH₃CN with a water fraction (f_w) of 90% were placed in a volumetric flask. Each time, a 3.0 mL aqueous suspension of Ir(III) complexes (10 μM) was placed in a quartz cuvette. For sensing studies, the stock solution of PA (10 mM) was prepared in H₂O/CH₃CN (9/1, v/v); after that, a further dilution was made with H₂O/CH₃CN (9/1, v/v) to various concentrations such as 0.10 mM, 0.20 mM, 0.30 mM, 0.40 mM, 0.50 mM, 0.60 mM, 0.70 mM, 0.80 mM, 0.90 mM, 1.0 mM, 2.0 mM, 3.0 mM, 4.0 mM, 5.0 mM, 6.0 mM, 7.0 mM, as well as 8.0 mM, and PA was detected on this account. The emission properties of Ir(III) complexes were measured by adding PA (30 μL) at different concentrations to suspensions of the Ir(III) complexes (10 μM, 3.0 mL) in H₂O/CH₃CN (9/1, v/v). For the selectivity measurement, other analytes (8.0 mM, including nitrobenzene (NB), nitromethane (NM), m-dinitrobenzene (1,3-DNB), phenol, 4-methoxyphenol (MEHQ), p-cresol, and m-cresol) were prepared under the same conditions. The selectivity experiments were conducted by adding the same doses of nitroaromatic compounds and non-nitroaromatic compounds (30 μL) into the Ir(III) complexes assay aqueous suspension. The competing experiments were implemented as follows: the emission spectra were collected when various analytes were dispersed to the suspension of Ir(III) complexes, followed by the addition of the same equivalent amount of PA. To detect natural water samples, tap water, river water, rainwater, and seawater were collected from the laboratory, the playground at Dalian University of Technology, Bohai (Dalian, China), and the Lingshui River (Dalian, China), respectively. The emission spectra of the Ir(III) complexes in different water samples after adding the PA (8.0 mM, 30 μL) were recorded.

3. Results and Discussion

3.1. Photophysical Properties

The UV-Vis absorption and the normalized emission spectra of the Ir(III) complexes (Ir1–Ir3) in dilute CH₂Cl₂ are displayed in Figure 1. The intense absorption bands of these Ir(III) complexes were observed at 250–350 nm and are associated with spin-allowed ligand-centered (¹π–π*) transitions [33]. The weaker absorption bands of Ir1 and Ir2 observed within the following range of 380–500 nm are due to the mixing between the metal-to-ligand charge transfer (¹MLCT and ³MLCT) and ligand-centered ³π–π* transitions, which is facilitated by enhanced spin-orbital coupling [34]. Clearly, Ir3 exhibited intense absorption relative to Ir1 and Ir2, which was attributed to the presence of an electron-donating TPA substituent on the cyclometalating ligands. The normalized emission spectra of Ir(III) complexes Ir1–Ir3 in CH₂Cl₂ exhibited that the introduction of a phenyl or TPA substituent on the cyclometalating ligand results in a red-shift for Ir2 (34 nm) and Ir3 (74 nm) in comparison with an unsubstituted Ir1. Additionally, Ir1 and Ir2 showed obvious fine vibronic splitting features, indicating a large ligand-centered (C^N) character (³LC). However, this unstructured emission spectrum of Ir3 may be owing to the distinct properties of the emitted excited states. Ir3 involves more ³MLCT/³LLCT features than Ir1 as well as Ir2 [35]. The phosphorescence quantum yields (Φ_PL) in deoxygenated CH₂Cl₂ were 0.55 (Ir1), 0.20 (Ir2), and 0.05 (Ir3), respectively (

Table 1. Photophysical data of Ir(III) complexes Ir1–Ir3.

Complex	λabsa (nm)	λemb (nm)	ΦPLc	τ d (μs)	kre (106 s−1)	knre (106 s−1)
Ir1	274 (5.17) 315 (3.21) 322 (3.34) 410 (0.81)	522, 559	0.55	3.73	0.148	0.120
Ir2	273 (3.61) 336 (4.15) 423 (0.88)	556, 595	0.20	4.81	0.042	0.166
Ir3	275 (3.91) 297 (4.00) 423 (3.55)	596	0.05	10.36	0.005	0.092

^a Measured in CH₂Cl₂ at a concentration of 10 μM and extinction coefficients (10⁴ M⁻¹ cm⁻¹) are shown in parentheses. ^b The maximum emission value is bold. ^c The quantum yields (Φ_solution) in deoxygenated CH₂Cl₂ were measured with [Ir(ppy)₂(acac)] (Φ_PL = 0.34) as a standard. ^d In deoxygenated CH₂Cl₂ solution. ^e The radiative and nonradiative decay rates of k_r and k_nr were calculated from k_r = Φ_PL × τ⁻¹, k_nr = τ⁻¹ − k_r.

). The low phosphorescence quantum yields of Ir3 may be due to the rotational motion of the benzene ring in TPA, which helps dissipate excited state energy in the solution state. The lifetimes (τ) of Ir1–Ir3 in degassed CH₂Cl₂ were 3.73, 4.81, and 10.36 μs at room temperature (see Supplementary Material, Figure S1 for the phosphorescence decay curves of Ir1–Ir3). These results indicate that introducing a phenyl or TPA substituent on the cyclometalating ligand results in prolonged phosphorescence lifetimes for Ir2 and Ir3 compared with Ir1. The radiative and nonradiative decay rates of these Ir(III) complexes in CH₂Cl₂ were calculated. The results were as follows (Table 1). The data demonstrate that the phenyl or TPA substituent has a critical impact on the modification of the photophysical properties of these Ir(III) complexes.

3.2. AIE Activities

To evaluate the AIE activities of Ir(III) complexes Ir1–Ir3, their emissions were measured in H₂O/CH₃CN with water fractions in the range of 0–90%. As shown in Figure 2a–c, their emission intensities changed observably with various water contents in an H₂O/CH₃CN system, exhibiting a typical AIE phenomenon. Additionally, the maximum emission intensities of Ir1–Ir3 were achieved at a water fraction of 90%, which is 1.9-, 3.5-, and 5.9-fold greater compared with those obtained in CH₃CN, respectively (Figure 2d). These results clearly demonstrate that phenyl or TPA substituent efficiently influences AIE activities. The enhanced AIE activity of Ir2 relative to Ir1 may be due to the introduction of a rotatable phenyl group. In a dilute solution, the rotatable phenyl group is active and serves as a relaxation channel for the excited states to deactivate. In the aggregated state, this phenyl group is restricted due to the physical constraint. This blocks the nonradiative pathway and thus enables the excitons to decay radiatively [26]. Ir3 exhibits the highest AIE activity among the three complexes. It may be that the multiple benzene rings of Ir3 contribute to the dissipation of the excited state energy through rotational motion. While in the aggregate, the motions become constrained, giving rise to intense luminescence.

3.3. Sensing of PA

The Ir(III) complexes with AIE activities availably prevent luminescence quenching in a water medium. Thus, Ir(III) complexes Ir1–Ir3 were used as probes for PA detection in aqueous media. To explore the potential for application of these Ir(III) complexes as PA probes, the emission quenching experiments based on aqueous suspensions of Ir(III) complexes in H₂O/CH₃CN (v/v = 9:1, 10 μM) toward different concentrations of PA were carried out. As can be seen from Figure 3c, upon the successive addition of PA (0 to 8.0 equiv.) to Ir3 in H₂O/CH₃CN (v/v = 9:1, 10 μM), the luminescent intensities of Ir3 at 603 nm gradually decreased. The quenching was obviously found when 1.0 μM of PA was added. As shown in Figure 3d, after adding PA (1.0 μM), the quenching efficiency of Ir3 was close to 70%. Under the same conditions, just 6% and 37% of emission quenching efficiencies were observed for both Ir1 and Ir2, respectively. Until the concentration of PA reached 30 μM, the quenching efficiency of Ir1 was over 70%, and when the PA concentration reached 80 μM, negligible emission was identified with a high quenching efficiency of 97% for Ir3. With the aim of measuring the quenching constant of Ir3 toward PA, the Stern–Volmer plots of (I₀/I) vs. PA concentration were fit. The luminescence quenching influence was quantitatively analyzed by applying the Stern–Volmer (SV) equation: I₀/I = K_sv[A] + 1. The relative luminescent intensity (I₀/I) exhibits good linearity within the PA concentrations in the range of 0–6.0 μM (Figure 3e), indicating that Ir3 can be applied as a probe for the quantitative detection of PA. The quenching constant (K_sv) was 1.96 × 10⁶ M⁻¹ for Ir3. Similarly, the luminescence of Ir1 and Ir2 was also quenched by PA in H₂O/CH₃CN (v/v = 9:1, 10 μM), with quenching efficiencies of 94% and 95%, respectively (Figure 3a,b). The K_sv values of Ir1 and Ir2 were measured as 7.36 × 10⁴ M⁻¹ and 4.59 × 10⁵ M⁻¹, respectively. Furthermore, according to the literature reports [36,37], the limit of detection (LOD = 3σ/K, σ represents the standard deviation of the blank measurement, and K represents the slope of the linear regression (see Figures S2 and S3 and Table S1) of Ir1–Ir3 for PA, which were measured as 50.17, 4.64, and 2.52 nM. The outcomes show that the AIE activities of these Ir(III) complexes Ir1–Ir3 are positively correlated with the detection efficiencies of PA. Compared with the literature results, the sensitivities of Ir2 and Ir3 are either comparable or better than those of some reported sensors (see Table S2). It is worth noting that there is almost no change in the emission spectra in the detection process, thus demonstrating that no other emissive species formed during the quenching process.

Based on the sensitive detection properties of Ir(III) complexes Ir1–Ir3 for PA, the sensing selectivity of these Ir(III) complexes was also conducted in H₂O/CH₃CN (v/v = 9:1, 10 μM). The phosphorescent emission quenching experiments for nitroaromatic and non-nitroaromatic compounds were studied, including picric acid (PA), nitromethane (NM), nitrobenzene (NB), and m-dinitrobenzene (1,3-DNB), as well as non-nitroaromatic compounds such as phenol, m-cresol, p-cresol, as well as 4-methoxyphenol (MEHQ). Compared with PA, all other analytes exhibited less of an effect on luminescence quenching under the same conditions (Figure 4a–c and Figure S4), which can also be confirmed from the corresponding photos under a UV lamp at 365 nm (Figure 4a–c, inset). This might be because PA is a representative electron-deficient compound with three electron-withdrawing nitro groups. Ir(III) complexes Ir1–Ir3 as electron donors can cause the photo-induced electron transfer (PET) for the emission quenching of Ir1–Ir3 [38]. The results of selectivity experiments reveal that the Ir(III) complexes Ir1–Ir3 have significantly high luminescent quenching efficiencies and selectivity for PA detection.

Anti-interference ability is another essential requirement for a suitable sensing probe. Thus, the influence of other potentially competitive analytes on PA detection was evaluated. As shown in Figure 4d–f and Figure S4, in the presence of other potential competitive analytes, PA can still significantly quench the emission of Ir(III) complexes Ir1–Ir3 in H₂O/CH₃CN (v/v = 9:1, 10 μM). These results demonstrate that Ir(III) complexes Ir1–Ir3 have an excellent anti-interference ability toward PA detection over other analytes. Additionally, to assess the actual uses of Ir(III) complexes Ir1–Ir3, their luminescent responses for PA were evaluated in natural water samples made by tap water, river water, rainwater, and seawater. As shown in Figure 5a–c, the luminescence of Ir(III) complexes Ir1–Ir3 is also quenched by PA in different samples of water/acetonitrile (v/v = 9:1). Additionally, the luminescence spectra of PA detected in the actual water samples do not change shape any more compared with those in deionized water. As can be seen from Figure 5d, there is little difference in the quenching efficiency among different water samples. For example, the quenching efficiencies of Ir3 in tap water, river water, rainwater, and seawater were calculated to be 97%, 97%, 95%, and 96%, respectively, which are similar to those in deionized water. The results indicate that these Ir(III) complexes work well in natural water samples.

3.4. Mechanism for Sensing PA

To understand the emission quenching mechanism of the Ir(III) complexes for PA, Ir3 was chosen to carry out the DFT calculations. The highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) energy levels of Ir3 were calculated. As shown in Figure 6, the LUMO energy of Ir3 (−1.85 eV) is over that of PA (−2.59 eV), which contributes to electrons jumping to PA. Thus, the excited electron can transfer from the higher energy LUMO of Ir3 to the lower energy LUMO of PA, leading to the photo-induced electron transfer-caused luminescence quenching of Ir3. The representative energy diagram shows that the energy gap of the adduct (Ir3 + PA) (3.51 eV) is lower than that (4.49 eV) of Ir3, indicating that the adduct (Ir3 + PA) is stable [39]. Additionally, the ¹H NMR spectra and HRMS of Ir3 with and without PA were recorded, and they have almost identical spectra (Figure S5), demonstrating that Ir3 does not decompose in the detection of PA. Therefore, the ¹H NMR spectra, HRMS analysis, and the DFT calculations suggest that the luminescence quenching process may be caused by PET.

4. Conclusions

In conclusion, two new AIE-active cationic Ir(III) complexes, namely Ir2 and Ir3, with phenyl and TPA substituted cyclometalating ligands were synthesized and characterized, respectively. Compared with the non-substituted complex Ir1, Ir2 and Ir3 exhibit higher AIE activities. Ir(III) complexes Ir1–Ir3 have been proved to be excellent chemosensors for the sensitive and selective detection of PA in water. The AIE activities of the complexes are positively correlated with their detection efficiencies of PA. Among them, Ir3 exhibits the highest AIE activity as well as the highest quenching constant for detecting PA. The DFT calculations, ¹H NMR spectra, and HRMS analysis demonstrate that PET is responsible for the emission quenching in sensing PA. These findings provide novel insights into the design of high-performance Ir(III) complexes for the selective and sensitive detection of PA in water.