Carbon nitride quantum dot-based chemiluminescence resonance energy transfer for iodide ion sensing

Abstract

In this study, a dramatically enhanced chemiluminescence (CL) was observed in Ce(IV) and sulfite system in the presence of graphitic carbon nitride quantum dots (g-CNQDs). On the basis of CL spectra, UV-vis absorption, fluorescence (FL), Fourier transform infrared spectroscopy (FTIR), X-ray photoelectron spectroscopy (XPS) and electron paramagnetic resonance (EPR) and the effect of various free radical scavengers, a possible CL mechanism of radiative recombination of the holes-injected and electrons-injected g-CNQDs was suggested to account for the surprising g-CNQD CL behavior. Meanwhile, the excited sulfur dioxide molecules , produced from the interaction between Ce(IV) and sulfite under acidic conditions, could transfer energy to g-CNQDs and further enhance the CL emission. In addition, the CL was dependent on the FL quantum yields of g-CNQDs with different surface states, since the chemiluminescence resonance energy transfer (CRET) efficiency was affected by the FL quantum yields of the g-CNQDs. The designed CL system was successfully applied to determine I⁻ in urine samples with good recoveries. It is anticipated that g-CNQDs could be a new class of CRET receptors for fabricating CL sensors. These findings would provide new insight into the optical properties of g-CNQDs and further broaden their applications in CL fields.

---

Introduction

Graphitic carbon nitride (g-C₃N₄), the most stable allotrope among various carbon nitrides under ambient conditions, consists of ordered tris-triazine moieties covalently connected by trigonal nitrogen and stacked in a graphitic fashion.¹ Nanoscale g-C₃N₄, including g-C₃N₄ nanosheets^2,3 and graphitic carbon nitride quantum dots (g-CNQDs),⁴ exhibit easy preparation, low cost, high quantum yield, desirable photostability and good biocompatibility. These unique properties make them a promising candidate for fluorescence sensors,^5–7 bioimaging,^8,9 persistent luminescence,¹⁰ and invisible security inks,¹¹ etc.

Chemiluminescence (CL) is the production of electromagnetic radiation by a chemical reaction between at least two reagents in which an electronically excited intermediate or product is obtained and subsequently relaxes to the ground state with emission of light.^12–14 Due to its advantages of excellent sensitivity, wide dynamic range, simple instrumentation, convenient automatic operation, and free of background scattering light interference, CL turns out to be a powerful analysis technique for rapid weak analysis.^15–17 Generally, the sensitivity of a CL system usually depends on the quantum yield of luminophors during the luminescence reaction, thus intense researches focus on exploring the new CL systems and obtaining new insight into the interaction between CL reagents. Nanomaterial-based CL emerges recently, in which the nanomaterials serve as catalysts, luminophors or energy acceptor and are used to increase luminescence yield in CL systems. Especially, CL phenomena based on nanoscale g-C₃N₄ have been reported in the recent two years. g-C₃N₄ exhibits favorable electrochemiluminescence properties and show great potential in sensing and immunoassay.^18–20 Direct CL of g-CNQDs in the presence of NaClO²¹ and K₃[Fe(CN)₆]²² were observed by our group successively. Afterwards, nanoscale g-C₃N₄ have been found to enhance the luminol–H₂O₂ CL system.²³ What's more, some metal ions have been found to enhance g-C₃N₄ nanosheets–NaHSO₃ CL system, which was distinctly different from the phenomenon that metal ions can quench the fluorescence of g-C₃N₄ nanosheets as reported before.⁶ However, the g-C₃N₄ nanosheets hereinto served as catalysts in CL analysis.²⁴ In the above-mentioned works, there are two possible mechanisms explaining nanoscale g-C₃N₄-based CL: (1) g-C₃N₄ acted as emitting species after direct oxidation;^21,22 (2) g-C₃N₄ acted as catalysts of a reaction involving others luminophors.^23,24

Generally, chemiluminescence resonance energy transfer (CRET) involves a nonradiative dipole–dipole transfer of energy from a chemiluminescent donor to a suitable acceptor molecule. Owing to the absence of photo-bleaching and auto-fluorescence, CRET proves to be a promising approach for high selectivity and sensitivity determination.^25–27 It is well-known that the oxidation of sulfite in acidic solutions using cerium(IV) is an important CL reaction, but the CL emission is quite weak. Fortunately, introducing energy transfer into the redox reaction, some fluorescent compounds, such as fluoroquinolone,²⁸ Tb³⁺–fluoroquinolone complexes²⁹ and luminescent QDs,^30,31 could greatly enhance the CL intensity of the weak CL system.

In the present work, g-CNQDs were first found to enhance the weak CL emission of Ce(IV)–sulfite CL system. The preliminary investigations demonstrated that the radiative recombination of the hole-injected and electrons-injected g-CNQDs possibly contributed to the surprising g-CNQDs CL behavior, and the excited sulfur dioxide molecules , produced from the interaction between Ce(IV) and sulfite under acidic condition, could transfer energy to g-CNQDs and further enhance the CL emission. In addition, the addition of I⁻ led to a significant CL quenching of Ce(IV)–SO₃²⁻–g-CNQDs CL system. Based on this, a simple, selective, and sensitive CL sensing approach for I⁻ was established. This work further extends the scope of g-CNQDs in CL field.

Experimental section

Synthesis of g-CNQDs

The g-CNQDs were prepared based on our reported paper through thermopolymerization of melamine and EDTA.¹⁶ The TEM, FTIR, XRD and XPS of the g-CNQDs were depicted in Fig. S2 and S3.†

Chemical treatment of g-CNQDs

Briefly, 0.1 M NaBH₄ was dropwise added to the g-CNQDs solution under continuous stirring for 2 h at room temperature, and then transferred to ultrasound cleaner (100 W) for 10 h. The r-g-CNQDs were obtained by dialyzed against ultra-pure water for 2 days to remove the excess reductant. Oxidized g-CNQDs (o-g-CNQDs) were achieved by dialyzed the g-CNQDs–Ce(IV)–SO₃²⁻ mixing solution. The obtained solution of r-g-CNQDs and o-g-CNQDs were freeze dried and weighed to same weight, then dispersed in ultra-pure water and stored at 4 °C.

Ce(IV)–SO₃²⁻–g-CNQDs CL process

CL kinetic curves were acquired by batch experiments, which were operated in a 2 mL quartz cuvette. 100 μL g-CNQDs and 100 μL SO₃²⁻ were premixed, and then 100 μL Ce(IV) were injected by a microliter syringe from upper injection port. The addition orders of the solutions were changed to investigate the interaction of the reagents and design the flow injection analysis (FIA) system. The CL profiles were displayed and integrated for a 0.1 s interval at −800 V.

According to the above batch CL signals, the Ce(IV)–SO₃²⁻–g-CNQDs CL test was constructed as a flow injection system which consisted of three flow lines. SO₃²⁻ solution was carried by water and injected through a six-valve injector. g-CNQDs and Ce(IV) in H₂SO₄ (0.1 M) were pumped into the mixing coil installed in front of the PMT as shown in Fig. S1.† The flow rates were 12 rpm for both g-CNQDs and Ce(IV) solution. The CL signals were recorded with the BPCL luminescence analyzer.

Urine sample pretreatment

Human urine samples were originally collected from healthy adults volunteers, then the samples were centrifuged at 8000 rpm for 10 min. The obtained supernatants were diluted 100-fold with deionized water (DIW) prior to analysis.

Results and discussion

Ce(IV)–SO₃²⁻–g-CNQDs CL system and the CL mechanism

The oxidation of SO₃²⁻ by Ce(IV) in acidic solution produced weak CL emission, which was attributed to the light emission during relaxation of the excited state SO₂ to the ground state.³⁰ Interestingly, the CL intensity was greatly enhanced by about 60-fold in the presence of g-CNQDs (Fig. 1a). The g-CNQDs could enhance the CL of other oxidants and SO₃²⁻, in addition, the different forms of sulfite had little effect on the CL as shown in Fig. S4.† In order to explore the effect of g-CNQDs on the Ce(IV)–SO₃²⁻ system, batch CL experiments were used to evaluate the performance of solutions with different injection orders. As showed in Fig. 1b, the highest CL was monitored with the addition of the Ce(IV) into the mixture of SO₃²⁻ and g-CNQDs because no reaction occurred before the injection. The CL reached to the peak rapidly and remained for 10 s before being utterly quenched, suggesting the enhancement process happened instantaneously.

Fig. 1 (a) CL of the Ce(IV)–SO₃²⁻ and Ce(IV)–SO₃²⁻–g-CNQDs; (b) CL kinetic curves of the Ce(IV)–SO₃²⁻–g-CNQDs system with different reagent mixing orders (legend: (1) injecting Ce(IV) into SO₃²⁻ and g-CNQDs, (2) injecting SO₃²⁻ into Ce(IV) and g-CNQDs, (3) injecting g-CNQDs into Ce(IV) and SO₃²⁻).

Based on the review of quantum dots-enhanced CL mechanism,¹⁷ we speculated that there were mainly three possible mechanisms explaining the enhancing effect of g-CNQDs on Ce(IV)–SO₃²⁻ CL system: (1) g-CNQDs acted as the catalysts in the Ce(IV)–SO₃²⁻ CL reaction like reported literature,¹⁷ which facilitated the radical generation and the formation of . (2) transferred the energy to the fluorescent g-CNQDs, which played as emitter in the enhanced CL reaction system. (3) The Ce(IV) and ˙HSO₃⁻ which was generated in the Ce(IV)–SO₃²⁻ reaction could respectively reacted with g-CNQDs to produce positively (g-CNQDs˙⁺) and negatively (g-CNQDs˙⁻) charged g-CNQDs, then excited-state g-CNQDs (g-CNQDs*) were formed through the electron-transfer annihilation of g-CNQDs˙⁺ and g-CNQDs˙⁻, which was same as reported article.³²

Study of the emitting species

In order to ascertain the emission species of Ce(IV)–SO₃²⁻–g-CNQDs CL system, the CL spectrum was attained by using F-7000 fluorescence spectrometer with the xenon lamp turned off and BPCL with the cut off filter (400–640 nm). A peak centered at 550 nm within the range of 450–600 nm was observed in Ce(IV)–SO₃²⁻ CL (Fig. 3a), which was in well agreement with the CL spectrum of the emitter .^17,33 While the CL spectrum shifted to the maximum emission peak at 475 nm in the presence of g-CNQDs (Fig. 3a), which resembled to the fluorescent emission of the g-CNQDs (Fig. 3b). The results demonstrated that CL emission species could be attributed to g-CNQDs* instead of , which excluded the possible mechanism that g-CNQDs acted as the catalyst in the Ce(IV)–SO₃²⁻ CL reaction.

The UV-vis absorption spectra and FL spectra of g-CNQDs before and after CL reaction under the same conditions were measured to verify what happened on g-CNQDs during the process. The UV-vis absorption spectra of g-CNQDs given in Fig. 2a exhibited a broad absorption between 200 and 500 nm, which barely change after mixing with SO₃²⁻. Ce(IV) in H₂SO₄ solution exerted bright yellow with a typical absorption around 320 nm, and the peak intensity decreased with the addition of the SO₃²⁻, resulting from the reduction of Ce(IV) to Ce(III) by SO₃²⁻. Nevertheless, the absorption spectra of Ce(IV) or Ce(IV)–SO₃²⁻ dropped to some degree with the addition of g-CNQDs. In addition, the maximum FL intensity of g-CNQDs decreased when Ce(IV) or Ce(IV)–SO₃²⁻ was injected in (in Fig. 2a). Changes in the surface structure of the g-CNQDs were thought to be the cause of the decreasing of the UV-vis absorption spectra and FL spectra.

Fig. 2 (a) UV-vis absorption of the reagents in the CL reaction. (b) The FTIR of the g-CNQDs and the g-CNQDs treated with Ce(IV) and SO₃²⁻. C 1s spectra of (c) the g-CNQDs and (d) the g-CNQDs treated with Ce(IV) and SO₃²⁻.

The pristine g-CNQDs and Ce(IV)–SO₃²⁻ treated g-CNQDs were the g-CNQDs experienced during the CL reaction. The FTIR spectra in Fig. 2b elucidated that functional groups such as –OH, –C=O, – further characterized by FTIR and XPS to find out what changes of COO⁻ were decorated on the surface of the g-CNQDs. The high-intensity peak at 1611 cm⁻¹ corresponding to the asymmetric stretching vibrations of the carboxylate anions increased after mixed with Ce(IV)–SO₃²⁻, which revealed that the surface hydroxyl groups of g-CNQDs were probably partly oxidized to carbonyl groups. Remarkably, C 1s peak in the range of 286–289 eV increased, and the peak gradually shifted towards the high binding-energy side, implying a rise in the percentage of oxygen-containing groups.²⁹ The peak of carbonyl groups increased at 289 eV and C–O peak characteristic of hydroxyl groups reduced at 286 eV as shown in Fig. 2c and d, the quantitative analysis also suggested that the oxygen content increased from 0.28 to 0.32.

EPR characterization was applied to investigate and quantify the ground-state properties of luminescent g-CNQDs. The EPR signal in Fig. 3c at 2.04959 indicated the existence of singly occupied orbital in ground-state g-CNQDs, suggesting that g-CNQDs could be served as electron donors or acceptors during the reaction.³⁴ And the EPR spectra of g-CNQDs after reacted with Ce(IV) or Ce(IV)–SO₃²⁻ were presented in Fig. 3c. The g-value for g-CNQDs decreased from 2.04959 to 1.98362 after mixed with Ce(IV) and scarcely changed when SO₃²⁻ was further added. The results revealed a change on the singly occupied orbital in the g-CNQDs when mixing with Ce(IV) and Ce(IV)–SO₃²⁻, and during which process an electron transfer reaction occurred. In addition, the peak intensity and area decreased suggesting the g-CNQDs were possible electron receptors and Ce(IV) could inject hole into the g-CNQDs in the reaction, which resembled with the CL of CDs and Ce(IV).^35,36

Fig. 3 (a) CL spectra of the Ce(IV)–SO₃²⁻ (blank line) and Ce(IV)–SO₃²⁻–g-CNQDs (red line). (b) UV-vis absorption spectra and FL spectra for g-CNQDs excited from 320 to 460 nm. (c) EPR spectra of the g-CNQDs and the g-CNQDs treated with Ce(IV) and Ce(IV)–SO₃²⁻. (d) ˙HSO₃ radicals generated via the reaction of 5,5-dimethyl-1-pyrroline N-oxide (DMPO) probe in the Ce(IV)–SO₃²⁻ and Ce(IV)–SO₃²⁻–g-CNQDs system.

EPR was also used to investigate whether the free radical intermediates, like ˙HSO₃, were being in the CL system, by employing DMPO as a specific target molecule of hydroxyl radical (˙OH) and ˙HSO₃.^33,37–40 The results shown in Fig. 3d obviously confirmed the existence of ˙HSO₃ in Ce(IV)–SO₃²⁻ and Ce(IV)–SO₃²⁻–g-CNQDs. However, it should be noted that with the presence of g-CNQDs the content of ˙HSO₃ free radicals was reduced, which might be attributed to their consumption during the reaction process with g-CNQDs as depicted in Fig. 3d. The ˙HSO₃ produced from Ce(IV)–SO₃²⁻ could act as electron donator to provide electron to g-CNQDs, which induced the reducing of the ˙HSO₃.

As well known, thiourea and tert-butyl alcohol are good ˙HSO₃ radical scavengers^41,42 both exerted inhibition of the CL system. In our experiment, it could be found that the original CL intensities were respectively quenched by AA, thiourea or tert-butyl alcohol. 0.1 mM thiourea could quench 90% of the original CL intensity. The results on the other hand demonstrated that ˙HSO₃ was responsible for the increased CL in the Ce(IV)–SO₃²⁻–g-CNQDs system (Fig. S6a–S6c†). Additionally, the CL peak located at 475 nm (Fig. 3a) might attributed to the emission from g-CNQDs, which was in agreement with the fluorescence spectra of g-CNQDs excited at 450 nm (Fig. 3b). In view of the above results, we speculated that g-CNQDs should play a role as the emitting species, rather than the catalysts in the CL system.

Based on the discussion above, a possible CL enhancement mechanism of the Ce(IV)–SO₃²⁻–g-CNQDs system can be summarized in Scheme 1. In acid medium, ˙HSO₃ was formed from the reaction of HSO₃⁻ and Ce(IV) [1], then two ˙HSO₃ radicals combined to produce S₂O₆²⁻, and it was unstable in solution, then converted to sulfate and [2 and 3]. When returned to its ground state, weak CL was generated [4]. Then with the addition of g-CNQDs into the solution, on one hand, Ce(IV) could act as the hole injector and convert g-CNQDs to g-CNQDs˙⁺ [5], and the electron donated from ˙HSO₃ to g-CNQDs produced g-CNQDs˙⁻ [6]. Finally, the electron–hole annihilation in or between the g-CNQDs˙⁺ and g-CNQDs˙⁻ resulted in excited-state g-CNQDs (g-CNQDs*) [7], which was unstable and decomposes to g-CNQDs by releasing energy to generate light [9].

Ce(IV) + HSO₃⁻ → ˙HSO₃ + Ce(III) (1)

2˙HSO₃ → S₂O₆²⁻ + 2H⁺ (2)

Ce(IV) + g-CNQDs → Ce(III) + g-CNQDs˙⁺ (5)

˙HSO₃ + g-CNQDs → SO₃ + H⁺ + g-CNQDs˙⁻ (6)

g-CNQDs˙⁺ + g-CNQDs˙⁻ → g-CNQDs* (7)

Scheme 1 Schematic illustration of the CL mechanism of g-CNQDs–Ce(IV)–SO₃²⁻ system.

On the other hand, a CRET might occur between (donors) and g-CNQDs (acceptors) [8], since the wide emission spectra range of (450–600 nm) overlapped the absorption spectra of the g-CNQDs. Then reaction [9] occurred and luminescence at 475 nm was emitted. It could be deduced according to the EPR results that the residual ˙HSO₃ could still produce the to trigger the reaction [8], despite of the reduction of ˙HSO₃ when g-CNQDs was added.

In order to further confirm the CRET process, the g-C₃N₄ nanosheets obtained from ultrasound exfoliation, g-CNQDs, r-g-CNQDs and o-g-CNQDs were respectively injected into Ce(IV)–SO₃²⁻ CL system to validate the universal enhancement properties, which had some different enhancement degrees to the Ce(IV)–SO₃²⁻ CL as depicted in Fig. 4a. The g-C₃N₄ nanosheets provided less enhancement than the pristine g-CNQDs owing to its lower FL efficiency. The o-g-CNQDs had more carboxyl groups but less FL efficiency and produced relatively lower CL intensity, while the r-g-CNQDs with higher FL efficiency generated relatively greater enhancement effect than g-CNQDs. The FL, FTIR and XPS of the nanoscale g-C₃N₄ were shown in Fig. 4b–d. The results revealed that g-CNQDs were indeed the CL emitting species and the CRET was happened in the CL reactions. It turned out that the g-CNQDs with high fluorescence efficiency were suitable for accepting energy from Ce(IV)–SO₃²⁻ CL system, in other words, the g-CNQDs with higher fluorescence efficiency had greater CRET efficiency and then produced greater enhanced CL intensity.

g-CNQDs* → g-CNQDs + hν (475 nm) (9)

Fig. 4 (a) Enhancement activities of g-CNQDs, o-g-CNQDs, r-g-CNQDs. (b) Fluorescence emission spectra of different g-CNQDs, excitation wavelength was 360 nm. (c) FTIR spectra and (d) XPS of different g-CNQDs.

Analytical application of the CL system

Iodine, as an essential micronutrient for normal human, serves vital functions in neurological development and thyroid gland function.⁴³ Generally, abnormal levels of iodine, deficiency or overabundance have been considered as an indicator for many diseases including thyroid enlargement, hypothyroidism and hyperthyroidism. Therefore, the adequate detection and quantification of iodine in biological fluids is of consistently significance in clinical analysis. To our excitement, an interesting phenomenon was observed from the batch CL experiments that I⁻ could obviously inhibit the CL intensity. The mixing solutions turned blue after the addition of starch solutions, which indicated the I₂ was formed in the reaction of Ce(IV) and I⁻. The inhibition of I⁻ was owing to the competitive reaction with Ce(IV) which induced the consumption of Ce(IV) and thus reducing the CL intensity.

To establish the optimal conditions for detection of I⁻, the effects of the concentration of Ce(IV), Na₂SO₃, H₂SO₄, and g-CNQDs on the CL analysis were investigated. As shown in Fig. S5a,† the CL intensity in three different concentrations of I⁻ increased with the concentration of Ce(IV) reached maximum and kept stable at the highest concentrations. Therefore, an optimal concentration of 1 × 10⁻⁴ M was selected for subsequent investigations to provide strong CL response and favorable determination. Additionally, 1 × 10⁻⁴ M g-CNQDs provided the highest CL intensity in Fig. S5b,† and higher concentration of g-CNQDs were not desired, considering the enhanced light quenching effect resulting from higher chance of molecule collision of g-CNQDs. Notably, H₂SO₄ was herein chosen as the acid medium, since Ce(IV) possessed extremely stability and highly solubility in H₂SO₄ solution, while it could react with HCl and HNO₃ to cause denaturation. However, the oxidability of the Ce(IV) and the CL reaction process were significantly influenced by the concentration of H₂SO₄. Through Fig. S5c,† it could be seen that the CL signals decreased rapidly when the concentration of the H₂SO₄ was higher than 0.1 M. Hence, 1 × 10⁻⁴ M Ce(IV), 1 × 10⁻⁴ M g-CNQDs and 0.1 M H₂SO₄ was used for subsequent experiments to ensure stable and strong signal.

In addition, the flow rate of carrier strongly affected the residence time of the reactants in the spiral CL detection cell and then the CL intensities. As a matter of a fact, low flow rate resulted in a prolonged residence time but a broad and weak CL signal; while high flow rate a short residence time but ever-growing CL responses. It might be explained that the reaction rate was rather fast and then higher flow rates produced higher CL intensities. However, the higher flow rate also had drawbacks, such as increasing the pressure in the flow line and producing irresponsible signals. As shown in Fig. S5d,† the flow rate intensely influenced the identification of the water and the two concentrations of I⁻. Taking account of recognition effect, the CL intensity and consumption of the solution, the 12 rpm was chosen as the optimal flow rate for I⁻ determination.

Analytical performance and applications

Under the optimal experimental conditions, analytical figures of merit were investigated for different I⁻ concentrations as depicted in Fig. 5a. There was a preferable linearity between CL intensity and the concentration of I⁻ in the range from 3.0 × 10⁻⁷ to 3.0 × 10⁻⁵ M with a correlation coefficient of 0.9956 as shown in Fig. 5b. The CL intensity versus the concentration of I⁻ could be fitted into a regression equation as lg I₀ − lg I = 0.033[I⁻] (μM) + 0.063. The limit of the detection was estimated to be as low as 5.8 × 10⁻⁸ M.

Fig. 5 (a) Flow injection signals and (b) standard curve for the determination of I⁻ with its concentration in the range from 3 × 10⁻⁷ to 3 × 10⁻⁵ M. (c) selectivity of the CL sensor for I⁻ over other common biological molecules and cations at the concentration of 1 × 10⁻⁵ M. Experimental conditions: 1.0 × 10⁻⁵ M Ce(IV) in 0.1 M H₂SO₄, 1.0 × 10⁻⁵ M SO₃²⁻, 1.0 × 10⁻⁴ M g-CNQDs, 12 rpm flow rate, and a 100 μL sample injector.

Under the optimal conditions, a series of coexisting small biological molecules and cations in human urine were demonstrated under similar conditions for quantitation of I⁻. As shown in Fig. 5c, only AA, DA, Cys and His could reduce the CL to some extent. Based on the above results, the practical applicability of this proposed method was investigated by determination of the recovery of the spiked I⁻ in human urine samples from two volunteers. Informed consent was obtained from them. All experimental procedures were conducted in conformity with International Ethical Guidelines for Biomedical Research Involving Human Subjects, and protocols were approved by Medical ethics committee of Sichuan University. As shown in Table 1, the results showed the proposed CL method were in good agreement with ICP-MS, and the recoveries for sample determination were in the range of 94.0–105.4%, indicating the measurements were comparable and acceptable.

Table 1 Determination of I⁻ in urine samples

Sample	Spiked (μM)	ICP-MS^a (μM)	Proposed CL sensor^a (μM)	Recovery^a (%)
a Mean ± SD, n = 3.
1	3	2.91 ± 0.12	2.82 ± 0.16	94.0 ± 6.5
	5	5.36 ± 0.08	5.47 ± 0.23	109.4 ± 6.4
	8	7.83 ± 0.17	7.75 ± 0.34	96.8 ± 3.6
2	3	3.13 ± 0.21	3.02 ± 0.15	100.6 ± 5.0
	5	5.31 ± 0.14	5.27 ± 0.23	105.4 ± 4.6
	8	8.07 ± 0.17	7.96 ± 0.37	99.5 ± 4.6

---

Conclusions

In this study, the interaction of Ce(IV) and sulfite in the presence of g-CNQDs was accompanied by very strong CL. The enhanced CL mechanism was speculated to be attributed to radiative recombination of the hole-injected and electrons-injected g-CNQDs with studies by CL spectrum, UV-vis absorption, fluorescence, FTIR, XPS and EPR and the effect of various free radical scavengers. The produced excited sulfur dioxide molecules from reaction between Ce(IV) and SO₃²⁻, could transfer energy to g-CNQDs and further enhance the CL emission. What's more, the CRET efficiency depended on the FL quantum yields of the g-CNQDs with different surface states. Since the established CL signals could be effectively quenched by I⁻, which was successfully applied to determine I⁻ in urine samples with good recoveries and high reproducibility. These studies provided new insight into the optical properties of g-CNQDs and further broaden their applications in CL fields.

Acknowledgements

Authors gratefully acknowledged financial support for this project from the National Natural Science Foundation of China [No. 21375089] and Science & Technology Department of Sichuan Province of China (2015JY0272).

Notes and references

1. S. W. Cao, J. X. Low, J. G. Yu and M. Jaroniec, Adv. Mater., 2015, 27, 2150–2176.
2. M. C. Rong, L. P. Lin, X. H. Song, T. T. Zhao, Y. X. Zhong, J. W. Yan, Y. R. Wang and X. Chen, Anal. Chem., 2015, 87, 1288–1296.
3. Q. B. Wang, W. Wang, J. P. Lei, N. Xu, F. L. Gao and H. X. Ju, Anal. Chem., 2013, 85, 12182–12188.
4. J. Zhou, Y. Yang and C. Y. Zhang, Chem. Commun., 2013, 49, 8605–8607.
5. M. C. Rong, L. P. Lin, X. H. Song, Y. R. Wang, Y. X. Zhong, J. W. Yan, Y. F. Feng, X. Y. Zeng and X. Chen, Biosens. Bioelectron., 2015, 68, 210–217.
6. J. Q. Tian, Q. Liu, A. M. Asiri, A. O. Al-Youbi and X. P. Sun, Anal. Chem., 2013, 85, 5595–5599.
7. S. Barman and M. Sadhukhan, J. Mater. Chem., 2012, 22, 21832–21837.
8. X. D. Zhang, X. Xie, H. Wang, J. J. Zhang, B. C. Pan and Y. Xie, J. Am. Chem. Soc., 2013, 135, 18–21.
9. X. D. Zhang, H. X. Wang, H. Wang, Q. Zhang, J. F. Xie, Y. P. Tian, J. Wang and Y. Xie, Adv. Mater., 2014, 26, 4438–4443.
10. Y. R. Tang, H. J. Song, Y. Y. Su and Y. Lv, Anal. Chem., 2013, 85, 11876–11884.
11. Z. P. Song, T. R. Lin, L. H. Lin, S. Lin, F. F. Fu, X. C. Wang and L. Q. Guo, Angew. Chem., Int. Ed., 2016, 128, 2773–2777.
12. Y. Y. Su and Y. Lv, RSC Adv., 2014, 4, 29324–29339.
13. Y. Q. Dong, C. Q. Chen, J. P. Lin, N. N. Zhou, Y. W. Chi and G. N. Chen, Carbon, 2013, 56, 12–17.
14. Y. Q. Dong, J. H. Cai, X. You and Y. W. Chi, Analyst, 2015, 140, 7468–7486.
15. Z. H. Chen, Y. Tan, K. F. Xu, L. Zhang, B. Qiu, L. H. Guo, Z. Y. Lin and G. N. Chen, Biosens. Bioelectron., 2016, 75, 8–14.
16. M. Zhao, H. F. Li, W. Liu, Y. M. Guo and W. R. Chu, Biosens. Bioelectron., 2016, 79, 581–588.
17. H. Chen, L. Lin, H. F. Li and J. M. Lin, Coord. Chem. Rev., 2014, 263–264, 86–100.
18. L. C. Chen, D. J. Huang, S. Y. Ren, T. Q. Dong, Y. W. Chi and G. N. Chen, Nanoscale, 2013, 5, 225–230.
19. L. C. Chen, X. T. Zeng, P. Si, Y. M. Chen, Y. W. Chi, D. H. Kim and G. N. Chen, Anal. Chem., 2014, 86, 4188–4195.
20. C. Zhou, Y. M. Chen, P. X. Shang and Y. W. Chi, Analyst, 2016, 141, 3379–3388.
21. Y. R. Tang, Y. Y. Su, N. Yang, L. C. Zhang and Y. Lv, Anal. Chem., 2014, 86, 4528–4535.
22. X. Q. Fan, Y. Feng, Y. Y. Su, L. C. Zhang and Y. Lv, RSC Adv., 2015, 5, 55158–55164.
23. H. L. Yu, Y. He, W. H. Li and T. Duan, Sens. Actuators, B, 2015, 220, 516–521.
24. Y. Z. Zheng, X. N. Dou, H. F. Li and J. M. Lin, Nanoscale, 2016, 8, 4933–4937.
25. X. Y. Huang, L. Li, H. F. Qian, C. Q. Dong and J. C. Ren, Angew. Chem., Int. Ed., 2006, 45, 5140–5143.
26. R. Freeman, X. Q. Liu and I. Willner, J. Am. Chem. Soc., 2011, 133, 11597–11604.
27. E. Golub, A. Niazov, R. Freeman, M. Zatsepin and I. Willner, J. Phys. Chem. C, 2012, 116, 13827–13834.
28. Y. Rao, Y. Tong, X. R. Zhang, G. A. Luo and W. R. G. Baeyens, Anal. Lett., 2000, 33, 1117–1129.
29. N. Lian, H. C. Zhao, C. Y. Sun, S. L. Chen, Y. Lu and L. P. Jin, Microchem. J., 2003, 74, 223–230.
30. C. Y. Sun, B. Liu and J. H. Li, Talanta, 2008, 75, 447–454.
31. L. Y. Zheng, Y. W. Chi, Y. Q. Dong, J. P. Lin and B. B. Wang, J. Am. Chem. Soc., 2009, 131, 4564–4565.
32. L. Bao, C. Liu, Z. L. Zhan and D. W. Pang, Adv. Mater., 2015, 27, 1663–1667.
33. W. Xue, Z. Lin, H. Chen, C. Lu and J. M. Lin, J. Phys. Chem. C, 2011, 115, 21707–21714.
34. X. Wang, L. Cao, F. S. Lu, M. J. Meziani, H. T. Li, G. Qi, B. Zhou, B. A. Harruff, F. Kermarrec and Y. P. Sun, Chem. Commun., 2009, 3774–3776.
35. M. Amjadi, J. L. Manzoori, T. Hallaj and M. H. Sorouraddin, Microchim. Acta, 2014, 181, 671–677.
36. Z. Lin, W. Xue, H. Chen and J. M. Lin, Chem. Commun., 2012, 48, 1051–1053.
37. M. Amjadi and T. Hallaj, J. Lumin., 2016, 171, 202–207.
38. X. L. Shi, J. Inorg. Biochem., 1994, 56, 155–165.
39. T. Ozawa and A. Hanaki, Biochem. Biophys. Res. Commun., 1987, 142, 410–416.
40. Y. H. Li and J. R. Lu, Anal. Chim. Acta, 2006, 577, 107–110.
41. I. M. Psarellis, N. T. Deftereos, E. G. Sarantonis and A. C. Calokerinos, Anal. Chim. Acta, 1994, 294, 27–34.
42. J. M. Lin and T. Hobo, Anal. Chim. Acta, 1996, 323, 69–74.
43. V. R. Preedy, G. N. Burrow and R. R. Watson, Comprehensive Handbook of Iodine: Nutritional, Biochemical, Pathological, and Therapeutic Aspects, Academic Press, New York, 2009.

---

Footnote

† Electronic supplementary information (ESI) available: Details of characterization and influencing factor of CL system were included here. See DOI: 10.1039/c6ra15509j

---