Pyrogallol Detection Based on the Cobalt Metal–Organic Framework of Nanomaterial-Enhanced Chemiluminescence

Abstract

The cobalt metal–organic framework (Co-MOF) is a kind of crystalline porous material within a periodic network structure, which is formed via the self-assembly of a Co metal center and a bridged organic ligand. In this paper, a Co-MOF was facilely synthesized via an ultrasonic method and applied to enhance the chemiluminescence (CL) emission of the NaIO₄-H₂O₂ system. The synthesized Co-MOF was nanosheet-like in nature and stacked in 2–3-micrometer flower shapes. Compared to the NaIO₄-H₂O₂ system without a Co-MOF, the CL intensity of the Co-MOF-NaIO₄-H₂O₂ system was enhanced about 70 times. This CL mechanism was determined to be a result of the synergistic effects of chemiluminescence resonance energy transfer (CRET) and electron–hole annihilation (EHA). The Co-MOF not only acted as a catalyst to accelerate the generation of reactive oxygen species in the CL reaction, but also worked as an emitter to further enhance the CL. Based on the Co-MOF-NaIO₄-H₂O₂ system, a highly sensitive CL analysis method was established for pyrogallol (PG) detection. Addition of PG into the CL system generated ¹O₂*, which could transfer energy to the Co-MOF and further enhance the CL response. The enhanced CL was linear with the PG concentration. The CL analysis method exhibited a linear range of 1 × 10⁻⁴ M to 1 × 10⁻⁷ M, as well as having a linear correlation coefficient of 0.9995 and a limit of detection of (S/N = 3) of 34 nM.

1. Introduction

Pyrogallol (PG) belongs to the phenol family and is inherently toxic. However, as it is a chemical reagent and chemical raw material, PG is widely used in pharmaceutical synthesis, cosmetics, dyeing materials, and electronic products [1]. In the biomedical field, the interactions between PG analogs and biomacromolecules (e.g., protein, DNA/RNA, and polymers) can be used for the design of useful biomedical materials [2]. Due to the widespread use of PG, it is consequentially discharged into industrial effluents and, thus, adds hazards into aquatic environments. PG can cause oxidative stress, which may lead to biological poisoning or even death [3,4,5]. The abundance of PG in rivers is about 4 μM [6]. For As4.1 juxtaglomerular cells, the IC₅₀ of PG is about 60 μM for 48 h, as determined using an MTT assay [7]. Many analytical methods have been designed for PG detection [8,9,10,11,12], but they show the disadvantages of high-cost, complicated and time-consuming procedures, and low sensitivity. Therefore, it is necessary to establish a new cost-effective technique for PG detection that is characterized by high sensitivity, ease of operation, and rapid analysis.

Chemiluminescence (CL) analysis is a characteristic method used in analytical chemistry, as well as in environmental monitoring, drug analysis, food testing, biological analysis, and clinical detection, because of its simple equipment, convenient operation, rapid analysis, and easy automation [13,14,15,16,17]. However, it is undeniable that the CL analysis method still has limitations, such as easy interference, poor selectivity, and low signal response [18,19]. Selecting appropriate sensitizers and adding them into CL systems is an effective route toward enhancing CL intensity and improving both sensitivity and accuracy. The development of nanomaterials sheds light on ways to devise novel CL sensitizers [20]. Nanomaterials can enhance CL intensity in two ways: chemiluminescence resonance energy transfer (CRET) and electron–hole annihilation (EHA). CRET is a process through which energy transfers from a CL reagent to an energy acceptor during the CL reaction; EHA is a process through which the nanomaterials serve as CL emitters in order to produce enhanced CL from the combination of hole- and electron-injected nanomaterials [19,21]. Metal–organic frameworks (MOFs) represent a new class of nanomaterials, and they are composed of metal nodes and organic ligands and have become one of the most attractive materials due to their tunable structure and function [22,23]. Due to their porous structure and low energy band, many researchers have begun to pay attention to their application. For example, Singh et al. [24] prepared a two-fold, interpenetrated, and mixed-ligand Cd(II)-organic framework with great photoluminescence (PL) with which to detect Fe³⁺ and Cr (IV). Liu et al. [25] designed a Zn-MOF with a new organic linker, which contained a pyridine group, was able to absorb CO₂ due to its large specific surface area and porous structure, and could also detect Cu²⁺ due to its excellent PL properties. Therefore, it is an appealing research direction regarding the application of MOF materials with specific energy levels and porous molecular structures to sensitize the CL process [26]. However, MOFs used in aqueous solutions have poor stability and dispersity, meaning that their applications in liquid-phase CL systems face serious challenges. Additionally, MOFs are always synthesized through hydro- and solvo-thermal reactions, which involve energy- and time-consuming steps and impede their CL applications [27]. Therefore, novel MOFs with higher stability and better dispersity need to be designed and synthesized through facile and economical approaches in order to enhance CL intensity for sensitive detection.

In this work, we synthesized a cobalt metal–organic framework (Co-MOF) using an ultrasonic method that used 2-aminoterephthalic acid (NH₂-H₂BDC) as a ligand and applied it to sensitize the ultraweak CL system of NaIO₄-H₂O₂. The Co-MOF dramatically enhanced the CL response of the NaIO₄-H₂O₂ system. Optimum CL signals were obtained by optimizing experimental conditions. The CL process was systematically studied and demonstrated to occur via two synergistic mechanisms: CRET and EHA. The CL signals showed a linear relationship with the concentration of PG within a specific range. As a result, the Co MOF-NaIO₄-H₂O₂ system was developed as a CL analysis method for PG detection, having the advantages of high sensitivity, good reproducibility, and fast analysis speed.

2. Materials and Methods

2.1. Reagents and Materials

Sulfuric acid (H₂SO₄, 98%) was purchased from Beijing Chemical Reagent Co. (Beijing, China). Hydrogen peroxide (H₂O₂, 30%), ethanol (99.7%) and isopropyl alcohol (98.5%) were supplied by Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Cobalt chloride hexahydrate (CoCl₂∙6H₂O, 99%), N,N-dimethylformamide (DMF, C₃H₇NO, 99.9%), triethylamine (C₆H₁₅N, 99.5%), NH₂-H₂BDC (C₈H₇NO₄, ≥98%) and PG (C₆H₆O₃, 99%) were obtained from Beijing InnoChem Science and Technology Co., Ltd. (Beijing, China). Sodium periodate (NaIO₄, ≥99.5%), p-benzoquinone (BQ, C₆H₄O₂, 99.7%) and thiourea (99%) were purchased from Aladdin Reagent Co., Ltd. (Shanghai, China). Ascorbic acid (AA, C₆H₈O₆, ≥99%), 5,5-dimethyl-1-pyrroline-N-oxide (DMPO, C₆H₁₁NO) and 2,2,6,6-tetramethylpiperidine (TEMP, C₉H₁₉N) were bought from Sigma-Aldrich Chemical Co. (St. Louis, MO, USA). In a typical experiment, none of the chemicals were further purified.

2.2. Apparatus

The scanning electron microscope (SEM) images were obtained via a JSM-7900 scanning electron microscope (JEOL, Tokyo, Japan) to study the micro-morphologies of the samples. The transmission electron microscopy (TEM) image and selected area electron diffraction (SAED) patterns were determined via a JEOL-1400 transmission electron microscope (JEOL, Tokyo, Japan). Energy-dispersive X-ray spectroscopy (EDS) was recorded on a JED-2300 T transmission electron microscope (JEOL, Tokyo, Japan). The elemental composition of the surface of the material was measured via the X-ray photoelectron spectrometry (XPS) tool created by AXIS SUPRA, Shimadzu, Kyoto, Japan. Fourier transform infrared (FT-IR) spectra were used to determine functional groups and structures depending on PerkinElmer Frontier FT-IR spectrometer. The PL spectra were determined via an Agilent Cary Eclipse spectrofluorometer (F-7000, Hitachi, Japan) with a 5-nanometer silt width. UV-vis absorption spectra were collected via a PerkinElmer Lambda 950 spectrophotometer. The batch experiment was performed via a BPCL ultraweak CL analyzer (BPCL-GP21Q-TGC, Guangzhou microlight technology Co., Ltd., Guangzhou, Guangdong, China), and the voltage of the photomultiplier tube (PMT) was controlled at 700 V. CL spectra were collected via an FLS-1000 steady-state/transient fluorescence spectrometer (Edinburgh Instruments Ltd., Livingstone, Edinburgh, UK). Electron paramagnetic resonance (EPR) spectra were determined via a Bruker E500 spectrometer to make sure that the correct free radicals were generated. Brunauet–Emmett–Teller (BET) surface areas and pore size distribution were detected via the autosorb iQ analyzer (Quantachrome Instruments Co., Ltd., Boynton Beach, Florida, USA).

2.3. Synthesis of Co-MOF

Co-MOF was synthesized via the ultrasonic method illustrated in previous literature [28]. Firstly, 32 mL DMF, 2 mL C₂H₅OH and 2 mL H₂O were mixed in a 100-milliliter flask. Subsequently, 0.75 mmol of NH₂-H₂BDC was added into the above flask and dissolved using an ultrasonic wave. Next, 0.75 mmol of CoCl₂·6H₂O was added and ultrasonically treated until completely dissolved. Finally, 0.8 mL of triethylamine was quickly poured into the flask and vigorously stirred for 5 min. The flask was kept under ultrasonic treatment for 8 h at a power of 90 kHz under ambient conditions. The resulting product was then centrifuged at a speed of 10,000 rpm for 5 min to remove the supernatant, washed twice with C₂H₅OH and dispersed in 30 mL of H₂O. The washed product was freeze-dried under a vacuum and then collected for further use. The ultrasonic time for Co-MOF synthesis was further optimized to obtain the best CL enhancement.

2.4. CL Emission Study of the Co-MOF-NaIO₄-H₂O₂ System

All of the CL signals were collected via the BPCL luminescence analyzer at a time interval of 0.01 s. The PMT was operated at a voltage of 1000 V. In total, 1 mL of 0.08 M NaIO₄ solution and 100 μL of 4 mg∙mL⁻¹ Co-MOF were mixed into the CL reaction cell. Next, 1 mL of 0.12 M H₂O₂ solution (acidified by 0.006 M H₂SO₄) was rapidly injected, and the CL reaction was triggered. The CL intensity was measured, and the kinetic curves were collected. A control experiment was conducted without Co-MOF to demonstrate the effect of Co-MOF on the CL process of NaIO₄-H₂O₂ system. In order to obtain the optimum CL emission, the concentrations of all of the substances in the CL system were optimized using the univariate method. The CL intensities were measured at a series of concentrations of Co-MOF (1 mg∙mL⁻¹, 2 mg∙mL⁻¹, 3 mg∙mL⁻¹, 4 mg∙mL⁻¹, 5 mg∙mL⁻¹, and 6 mg∙mL⁻¹) and H₂O₂ (0.04 M, 0.08 M, 0.10 M, 0.12 M, 0.14 M and 0.16 M), NaIO₄ (0.01 M, 0.02 M, 0.03 M, 0.04 M, 0.05 M, 0.06 M, 0.07 M, 0.08 M, 0.09 M, and 0.10 M), as well as different concentration of H₂SO₄ (0.002 M, 0.003 M, 0.004 M, 0.005 M, 0.006 M, 0.007 M, 0.008 M, and 0.009 M) to adjust pH values. The following experiments were carried out under optimal conditions.

2.5. CL Mechanism Study of the Co-MOF-NaIO₄-H₂O₂ System

EPR and free radical quenching experiments were the two main methods used to explore the CL mechanism of the Co-MOF-NaIO₄-H₂O₂ system. The EPR spectrum was collected at the X-band frequency of 9.85 GHz using a 300 W xenon lamp (300 nm < λ < 1100 nm) as the light source. The output radiation was focused on the sample in the cavity through an optical fiber (50 cm in length, 0.3 cm in diameter). TEMP and DMPO were used as the specific detection reagents for ¹O₂, ∙OH, and ∙O₂⁻. In the free-radical-quenching experiments, specific free-radical quenchers were used to detect the free-radical intermediates of the reaction system. Furthermore, the role of dissolved oxygen in the system was analyzed through the bubbling experiment of oxygen and nitrogen. CL kinetic curves were collected, with all of the solutions pre-treated with nitrogen. In the control experiment, the solutions were first treated with nitrogen and then treated with oxygen. Moreover, CL kinetic curves were collected via various injection sequences of reagents. Three different injection sequences were investigated. The first way was to add acidified H₂O₂ into the mixture of NaIO₄ and Co-MOF, the second way was to inject NaIO₄ into the mixture of Co-MOF and acidified H₂O₂, and the third way was to infuse Co-MOF into the mixed solution of NaIO₄ and acidified H₂O₂. CL spectrum and UV-Vis spectrum tests were also conducted to verify the CL mechanism.

2.6. PG Detection

The constructed Co-MOF-NaIO₄-H₂O₂ system was developed as a CL sensor for PG detection. Calibration curve was recorded by detecting the relationship between CL intensities and PG concentration series under the optimal experimental conditions. Next, 100 μL PG (dissolved in H₂O), 1 mL 0.08 M NaIO₄ solution, and 100 μL 4 mg∙mL⁻¹ Co-MOF were mixed into the CL reaction cell. Finally, 1 mL of 0.12 M H₂O₂ solution (acidified by 0.006 M H₂SO₄) was injected to detect the CL signals.

3. Results and Discussion

3.1. Characterization of Co-MOF

The ultrasonic synthesized Co-MOF displayed a flower-shaped morphology with a diameter of approximately 2~3 μm (Figure 1a). This morphology was formed by stacking several pieces of nanosheets, which were transparent and ultrathin (Figure 1b). The large surface and numerous active sites of Co-MOF nanosheets were conducive to the CL reactions. The SAED image in Figure 1b’s inset image demonstrates the single-crystal diffraction properties of the Co-MOF. The XRD pattern confirmed the high crystallinity of Co-MOF and coincided with the simulated pattern of cobalt-based ultrathin MOFs nanosheets (Figure 1h) [28,29]. The EDS-mapping diagram showed the composition of the Co-MOF, which was composed of Co, C, N, and O uniformly distributed across the nanosheets (Figure 1c). XPS was utilized to corroborate the chemical composition and provide further information on the surface electronic state of the Co-MOF (Figure 1d). As shown in Figure 1e, the characteristic signals at 781.6 eV and 797.6 eV were attributed to Co 2p_3/2 and Co 2p_1/2, respectively [30]. Satellite peaks accompanying them at 786.4 eV and 802.7 eV revealed the formation of antibonding orbitals between the Co and O elements [31]. The characteristic signal of O 1s at 531.8 eV and its fitting peaks at 531.6 eV, 532.1 eV, and 533.4 eV were attributed to Co-O, C-O, and C=O, respectively (Figure 1f) [29]. The signature signal of C is located at 283.6 eV and 287.4 eV, and its fitting peaks located at 283.4 eV, 284.8 eV, and 287.4 eV were attributed to C-C, C-O-C, and O-C=O, respectively (Figure 1g) [28]. All of the above results confirmed the successful coordination between Co²⁺ and NH₂·H₂BDC. The coordination between Co²⁺ and NH₂·H₂BDC was further verified via the FT-IR spectra (Figure 1i). Apparent differences were observed via the FT-IR spectra of Co-MOF and NH₂·H₂BDC. The –OH in the –COOH group of NH₂·H₂BDC had a broad-stretching vibrational absorption peak in the region of 3000 cm⁻¹ to 2500 cm⁻¹. In contrast, the peak of Co-MOF disappeared within the FT-IR spectra due to the deprotonation of NH₂·H₂BDC during the coordination process [29]. Compared to the FT-IR spectra of NH₂·H₂BDC, three new peaks appeared at the low wavenumbers of 708 cm⁻¹, 450 cm⁻¹, and 423 cm⁻¹, which were attributed to the stretching vibration of Co-O [29,32,33]. In addition, the stretching vibration peak of the H₂O molecule at 3087 cm⁻¹ indicated that Co²⁺ also coordinated with H₂O in Co-MOF. Asymmetric and symmetric stretching vibration absorptions in the -COO- group in the ligand NH₂·H₂BDC resulted in the two peaks located at 1558 cm⁻¹ and 1381 cm⁻¹, respectively. The peaks at 3473 cm⁻¹ and 3342 cm⁻¹ were derived from the asymmetric and symmetric stretching vibrations in the -NH₂ group in the ligand NH₂·H₂BDC. The intensities of the two peaks in the FT-IR spectra of Co-MOF were significantly weaker than those of the FT-IR spectra of NH₂·H₂BDC, further indicating the coordination effect of NH₂·H₂BDC [29,33,34]. As shown in Figure 1j’s inset image, the prepared Co-MOF exhibited a bright blue color under the UV lamp (λ_ex = 365 nm). Co-MOF gave an emission centered at 420-nanometer excitation due to lights with different wavelengths varying from 320 nm to 400 nm (Figure 1j).

The hydrothermal synthesized Co-MOF displayed a morphology completely different to that of the ultrasonic synthesized Co-MOF. As shown in Figure 2a,b, the Co-MOF synthesized via the hydrothermal method exhibited a regular circular shape, rather than layer stacked flower shape, and its size was approximately 500 nm and 1 μm, which was slightly smaller than the size of Co-MOF synthesized via ultrasound.

3.2. CL Emission of the Co-MOF-NaIO₄-H₂O₂ System

As shown in Figure 3a, the ultraweak CL emission of the NaIO₄-H₂O₂ system was dramatically enhanced using the ultrasonic synthesized Co-MOF. Free Co²⁺ ions were also tested by adding CoCl₂ solution into the NaIO₄-H₂O₂ system, while its CL enhancement was negligible, indicating that the CL enhancement was rooted from Co-MOF. In comparison, the CL enhancement of hydrothermally synthesized Co-MOF was also tested and demonstrated to be weaker than that of ultrasonically synthesized Co-MOF (Figure 3b). Ultrasonic waves proved able to have a certain effect on the synthesized Co-MOF and the CL emission performed via the Co-MOF-NaIO₄-H₂O₂ system. The N₂ adsorption and desorption isotherms were carried out to examine the differences between the specific surface areas and pore sizes of the hydrothermally synthesized and ultrasonically synthesized Co-MOFs to explore the effect of ultrasonic waves (Figure 4). All of the isotherms (red line in Figure 4) displayed a closed hysteresis loop of type H3, demonstrating that a type IV classification (BDDT classification) was available for them, and the samples exhibited mesoporous structures. According to the multi-point BET calculations, the specific surface area of the Co-MOF synthesized via the ultrasonication method was 27.42 m²/g, while that of the Co-MOF synthesized via the hydrothermal method was 10.88 m²/g. As shown in the inset of Figure 4a and b, the main pore sizes of ultrasonically synthesized and hydrothermally synthesized Co-MOF were identified via the Barrett–Joyner–Halenda (BJH) model to be 29.83 and 3.80 nm, respectively. Larger pores in ultrasonically synthesized Co-MOF demonstrated more structural defects that were caused by ultrasonic waves. The larger specific surface areas and greater structural defects of the ultrasonically synthesized Co-MOF was promising for CL enhancement.

Subject to the optimal CL response, the experimental parameters of the Co-MOF-NaIO₄-H₂O₂ system were systematically optimized via the univariate method. The effects of the ultrasonic time required for Co-MOF synthesis and the concentrations of Co-MOF, H₂SO₄ (to adjust the optimal pH), H₂O₂, and NaIO₄ on the CL intensity were separately investigated. The effect of the Co-MOF synthesis time on the CL of the Co-MOF-NaIO₄-H₂O₂ system is shown in Figure 5a. The Co-MOF had the best enhancement effect on the CL intensity of NaIO₄-H₂O₂ system after 8 h of ultrasonic synthesis. The CL intensity of the Co-MOF-NaIO₄-H₂O₂ system increased with the increase in Co-MOF amount until it reached 4 mg·mL⁻¹ (Figure 5b). According to preliminary experiments, the Co-MOF-NaIO₄-H₂O₂ system has excellent CL performance under H₂SO₄ acidification conditions. When 0.12 M H₂O₂ was acidified with 0.006 M H₂SO₄, the CL intensity of the system was highest (Figure 5c,d). The CL signal of the system was optimal when the concentration of NaIO₄ was 0.08 M (Figure 5e). Unless otherwise specified, subsequent experiments were carried out under optimal experimental conditions.

Moreover, different injection sequences of reagents in the system are essential factors that affect CL intensity. Therefore, the influence of these sequences on the CL intensity of the system needs to be investigated. Under the optimal experimental conditions, the most robust CL kinetic curves were obtained when NaIO₄ solution was injected into the mixture of acidic H₂O₂ and Co-MOF (Figure 5f). The reason for this outcome is that Co-MOF acts as a catalyst that accelerates the reaction of H₂O₂ decomposition to generate ·OH, which is beneficial for CL process. However, since the system underwent two CL reactions, the CL kinetic curve showed a shoulder peak that was not favorable for detection. The subsequent experiments were conducted by injecting acidic H₂O₂ solution into the mixture containing NaIO₄ and Co-MOF.

3.3. CL Mechanism of the Co-MOF-NaIO₄-H₂O₂ System

Nanomaterials introduced into liquid-phase CL systems generally act as catalysts [35] and luminophores [36,37]. To elucidate the role of Co-MOF and the main light emission species of the system, the CL spectra of the Co-MOF-NaIO₄-H₂O₂ system were collected using a transient fluorescence spectrometer. The results showed that the system had a maximum CL emission of 470 nm and the luminescence range was 350–600 nm (Figure 6a), which overlapped with the emission band of dimol singlet oxygen ¹(O₂)₂ [38,39,40] and Co-MOF. Therefore, (¹O₂)₂* and Co-MOF* were speculated to be the main emitters present in the Co-MOF-NaIO₄-H₂O₂ system. Figure 6b shows the UV-Vis spectra in which the NaIO₄, H₂O₂, NaIO₄-H₂O₂, Co-MOF-NaIO₄, and Co-MOF-NaIO₄-H₂O₂ systems had no characteristic absorption peaks. In contrast, Co-MOF had a prominent strong characteristic absorption peak at 300–400 nm. After the addition of H₂O₂, the absorption peak decreased, and its position was redshifted, demonstrating that Co-MOF catalyzed H₂O₂ to give ·OH and reacted with ·OH. As a result, H₂O₂, which is a strong oxidant, also produced a weak and rapidly quenched CL signal when mixed with Co-MOF (Figure 6c, black line). The disappeared absorption peak of Co-MOF at 300–400 nm in the Co-MOF-NaIO₄ system further proved that Co-MOF not only played the role of catalyst in this system, but also participated in the reaction [41]. This phenomenon was certified based on the weak CL signals noted in Figure 6c (blue line). In addition, no other absorption peaks were observed; thus, it was clear that substances with additional UV-vis absorption were not produced via the CL system.

(¹O₂)₂* derived from reactive oxygen species (ROS) (¹O₂, ·O₂⁻ and ·OH) was proved to be one of the luminophor molecules in the Co-MOF-NaIO₄-H₂O₂ system via the CL spectra (Figure 6a). Different radical scavengers and EPR spectra were used to detect the three kinds of ROS in the CL system and explore the roles that ROS played in the CL reaction process. The generation of ROS was verified by detecting the changes in the CL intensity of the Co-MOF-NaIO₄-H₂O₂ system before and after the addition of the specific radical scavengers. AA is an effective reactive oxygen radical quencher [42]. The CL intensity gradually decreased with increasing concentration of AA (Figure 7a). Therefore, it could be concluded that reactive oxygen radicals were produced via the Co-MOF-NaIO₄-H₂O₂ system and played crucial roles in the CL process. H₂O₂ could be oxidized using IO₄⁻ and yield ·O₂⁻ [35,43]. As a free radical scavenger of ·O₂⁻, BQ significantly quenched the CL intensity of the Co-MOF-NaIO₄-H₂O₂ system, indicating the presence of and important role played by ·O₂⁻ in the CL system (Figure 7b). Thiourea and isopropanol are common ·OH radical scavengers and were used to detect ·OH via this CL reaction system. The addition of different concentrations of thiourea into the system was found to have an inhibitory effect on the CL intensity (Figure 7c). The same phenomenon was observed when isopropanol was added into the CL system. In total, 10 μL of isopropanol quenched more than 95% of the CL in the Co-MOF-NaIO₄-H₂O₂ system (Figure 7d). The reason for this outcome was that isopropanol quickly captured ·OH to generate acetone carbonyl radicals, which then produced acetone [44]. Since isopropanol preferentially reacted with ·OH, the chances of ·OH participating in the CL reactions were reduced, resulting in the decreased CL intensity.

EPR is the most direct method used to identify radical intermediates in chemical reactions [45]. DMPO was used as the scavenger agent for ·OH detection in the Co-MOF-NaIO₄-H₂O₂, NaIO₄-H₂O₂, and Co-MOF-NaIO₄ systems. ·OH reacts with DMPO to form radical adduct DMPO-·OH, which can generate a quadruple peak of EPR signals with an intensity ratio of 1:2:2:1 [46]. The characteristic EPR signals of DMPO-·OH were collected via the Co-MOF-NaIO₄-H₂O₂ system (Figure 7e, red line), which confirmed the formation of ·OH. In the absence of Co-MOF, the characteristic EPR signals of DMPO-·OH were also collected via the NaIO₄-H₂O₂ system (Figure 7e, blue line, square symbols). EPR signals of DMPO-·OH collected via the NaIO₄-H₂O₂ system decreased sharply compared to those signals collected via the Co-MOF-NaIO₄-H₂O₂ system, suggesting that the addition of Co-MOF promoted the decomposition of H₂O₂ to produce ·OH. There was no obvious EPR signals of DMPO-·OH in the Co-MOF-NaIO₄ system (Figure 7e, black line), verifying that ·OH was derived from the decomposition of H₂O₂. The triplet EPR signals with the intensity ratio of 1:1:1 in Figure 7e (blue line, asterisk symbols) were assigned to 5,5-dimethyl-2-oxopyrroline-1-oxyl (DMPOX), which was derived from the oxidation reaction between ¹O₂ and DMPO [47]. DMPO is also the specific detection reagent for ·O₂⁻ when CH₃OH is used as the solvent. The characteristic 1:1:1:1 quartette EPR signals of DMPO-·O₂⁻ in Figure 7f confirmed the presence of ·O₂⁻ via the Co-MOF-NaIO₄-H₂O₂, NaIO₄-H₂O₂, and Co-MOF-NaIO₄ systems. IO₄⁻ reacted with H₂O₂, generating ·O₂⁻ and Co-MOF that accelerated this process [45]. Moreover, ·OH derived from H₂O₂ consumed ·O₂⁻ and then promoted the production of ·O₂⁻. This process was the reason that the EPR signals collected via the Co-MOF-NaIO₄-H₂O₂ and NaIO₄-H₂O₂ systems were stronger than those of Co-MOF-NaIO₄ system. TEMP is a specific scavenger for ¹O₂ and reacts with ¹O₂ to generate TEMPO, which is a stable nitrogen oxide radical with a characteristic EPR spectrum [48]. In Figure 7g, the characteristic triplet peaks of ¹O₂ with an intensity ratio of 1:1:1 were observed via the Co-MOF-NaIO₄-H₂O₂, NaIO₄-H₂O₂, and Co-MOF-NaIO₄ systems. ¹O₂ was the product of the reaction between ·O₂⁻ and ·OH. EPR signals collected via the ¹O₂ in Co-MOF-NaIO₄-H₂O₂ system (red line) were significantly weaker than those collected via the NaIO₄-H₂O₂ system (blue line) because Co-MOF reacted with (¹O₂)₂*, which was derived from ¹O₂. In order to confirm whether dissolved oxygen participated in the CL process of the Co-MOF-NaIO₄-H₂O₂ system, nitrogen and oxygen were alternately infused into the system. The CL intensity collected via the system was significantly reduced when nitrogen was infused for 15 min (Figure 7h, blue line). The CL intensity was restored when the system was subsequently infused with O₂ for 15 min (Figure 7h, red line). It could be concluded that dissolved oxygen played an essential role in the system and reacted with IO₄^- to generate ·O₂⁻ and ·OH [43]. This reaction process explained why EPR signals of ¹O₂ were also observed via the Co-MOF-NaIO₄ system, even though there was no H₂O₂.

Based on the above results, the possible CL mechanism of the Co-MOF-NaIO₄-H₂O₂ system was inferred and shown in Figure 8. The specific reactions were described in Equations (1)–(11). Firstly, Co-MOF acted as a catalyst to decompose NaIO₄ and H₂O₂ to produce radical intermediates (·OH and ·O₂⁻) (Equations (1) and (2)) [40,49,50]. The catalytic behavior of Co-MOF was beneficial for radical intermediate generation. IO₄⁻ could be oxidized using dissolved O₂ to create ·OH and ·O₂⁻ (Equation (3)) [43]. The unstable ·O₂⁻ could trigger a reaction to produce ¹O₂ (Equation (4)) [51]. (¹O₂)₂* with higher energy was generated using ¹O₂ (Equation (5)) [40] and worked as one of the critical luminophores in the Co-MOF-NaIO₄-H₂O₂ system. Oxidative radicals ·OH and IO₄⁻ acted as electron acceptors for Co-MOF and injected holes into Co-MOF (Equations (6) and (7)). The reductive ·O₂⁻ acted as an electron donor to inject electrons into the Co-MOF (Equation (8)) [52]. Due to the CRET effect, the excited state (¹O₂)₂* transferred its energy to Co-MOF to generate Co-MOF* (Equation (9)) [53] to produce a strong CL response. The EHA effect between hole-injected Co-MOF and electron-injected Co-MOF jointly promoted the enhanced CL collected via the Co-MOF-NaIO₄-H₂O₂ system (Equations (10) and (11)). The CRET and EHA effects together caused the strong CL release.

$${\mathrm{H}}_2{\mathrm{O}}_2\stackrel{\mathrm{Co}-\mathrm{MOF}}{\to }2\cdot \mathrm{OH}$$

(1)

$$\mathrm{I}{\mathrm{O}}_4^{-}+{\mathrm{H}}_2{\mathrm{O}}_2\stackrel{\hspace{1em}\mathrm{Co}-\mathrm{MOF}\hspace{1em}}{\to }\mathrm{I}{\mathrm{O}}_3^{-}+\cdot {\mathrm{O}}_2^{-}+{\mathrm{H}}_2\mathrm{O}$$

(2)

IO₄⁻ + O₂ + 3H₂O → 2·O₂⁻ + 2·OH+ IO₃⁻ + 4H⁺

(3)

H⁺ + ·O₂⁻ + OH → ¹O₂ + H₂O

(4)

¹O₂ + ¹O₂ → (¹O₂)₂* → hv (λ = 478 nm)

(5)

·OH + Co-MOF + H⁺ → Co-MOF^∙+ + H₂O

(6)

IO_4⁻ + Co-MOF + 2H⁺ → Co-MOF^∙+ + IO_3⁻ + H₂O

(7)

·O₂⁻ + Co-MOF → Co-MOF^∙− + O₂

(8)

(¹O₂)₂* + Co-MOF → Co-MOF* + 2O₂

(9)

Co-MOF^∙+ + Co-MOF^∙− → Co-MOF* + Co-MOF

(10)

Co-MOF* → Co-MOF + hv

(11)

3.4. PG Detection

The Co-MOF-NaIO₄-H₂O₂ CL system constructed in this experiment was successfully used to detect PG because it could enhance the CL response. PG can be oxidized using ·O₂⁻ to generate o-quinone and hydroquinone, which slowly react to yield quinone condensation products and Diels–Alder dimers [54]. These complex oxidation reactions involve forming multiple products from quinone intermediates through different reaction pathways controlled via kinetics. The oxidation process of PG is shown in Figure 9a. During this oxidation process, the excited state of singly linear oxygen (¹O₂*) was formed, and it enhanced the CL through intramolecular energy transfer. As shown in Figure 9b, the CL intensity showed a linear relationship with PG concentration in the range of 1 × 10⁻⁴ M to 1 × 10⁻⁷ M, having a linear correlation coefficient of 0.9995 and a limit of detection (LOD) (S/N = 3) of 34 nM. Compared to the existing CL methods used for PG detection, the detection method described in this article has a wider linear range and lower LOD. For example, Zhuang et al. [55] developed a KIO₄-isothiocyanatoisoluminol CL system to investigate the concentration of PG within the liner range of 1.75 × 10⁻⁸–2.5 × 10⁻⁶ M. Shah et al. [40] synthesized N-CDs to enhance the CL of KIO₄-H₂O₂ systems within the liner range of 1.0 × 10⁻⁴–1.0 × 10⁻⁷ M and LOD of 46 nM. Mostafa et al. [10] constructed a lucigenin-PG system to detect PG, and the results showed that the LOD was 940 nM, which was higher than that of our method. Therefore, the CL detection method that we developed has a promising future in the application of PG detection.

4. Conclusions

In conclusion, the Co-MOF prepared via the ultrasonic method in this work effectively enhanced the CL signal of the NaIO₄-H₂O₂ system. The Co-MOF-mediated CL enhancement was attributed to the synergistic effects of CRET and EHA. A novel CL analysis method was established for PG detection based on the quantitative relationship between the PG concentrations and the CL signal of the Co-MOF-NaIO₄-H₂O₂ system. The research content and application fields of MOF materials were enriched through this work. In future work, CL analysis methods based on MOF nanomaterials are expected to be used in multiple fields, such as clinical diagnosis, food safety, and environmental pollution monitoring.