Article pubs.acs.

org/ac

Graphene Quantum Dot as a Green and Facile Sensor for Free Chlorine in Drinking Water
Yongqiang Dong, Geli Li, Nana Zhou, Ruixue Wang, Yuwu Chi,* and Guonan Chen
Ministry of Education Key Laboratory of Analysis and Detection for Food Safety, Fujian Provincial Key Laboratory of Analysis and Detection for Food Safety, and Department of Chemistry, Fuzhou University, Fujian 350108, China ABSTRACT: Free chlorine was found to be able to destroy the passivated surface of the graphene quantum dots (GQDs) obtained by pyrolyzing citric acid, resulting in signicant quenching of their uorescence (FL) signal. After optimizing some experimental conditions (including response time, concentration of GQDs, and pH value of solution), a green and facile sensing system has been developed for the detection of free residual chlorine in water based on FL quenching of GQDs. The sensing system exhibits many advantages, such as short response time, excellent selectivity, wide linear response range, and high sensitivity. The linear response range of free chlorine (R2 = 0.992) was from 0.05 to 10 M. The detection limit (S/N = 3) was as low as 0.05 M, which is much lower than that of the most widely used N-N-diethyl-p-phenylenediamine (DPD) colorimetric method. This sensing system was nally used to detect free residual chlorine in local tap water samples. The result agreed well with that by the DPD colorimetric method, suggesting the potential application of this new, green, sensitive, and facile sensing system in drinking water quality monitoring.

hlorine (Cl2), hypochlorous acid (HClO), and hypochlorite (ClO) are strong oxidizing agents and have been extensively used as disinfectants in treating water, including drinking water, swimming pool water, wastewater for nonpotable reuse, and others. The sum of dissolved Cl2, HClO, and ClO in water is dened as free residual chlorine. In water treatment, the concentration of free residual chlorine should be strictly controlled, namely neither too low nor too high. Free residual chlorine with too low of a level cannot kill pathogenic bacteria and viruses in water eectively to keep water clear and clean. However, free residual chlorine with too high of a level may also be harmful, because the excess free chlorine would react with organic materials existing in the water to produce a large number of undesirable byproducts, especially trihalomethanes (THMs).13 THMs, including chloroform (CHCl3), dichlorobromemethane (CHCl2Br), chlorodibro-memethane (CHClBr2), and bromoform (CHBr3), have been reported to be potentially harmful to human beings and animals; for example, they may lead to cancer,46 harm of the immune system, cardiovascular system, and respiratory tract,7 reproductive failure,8,9 and skin irritation.1012 Therefore, it is necessary to monitor the concentration of free residual chlorine in water. Continuing interest is focused on the development of sensitive, selective, rapid, simple, and green analytical methods for the determination of free residual chlorine in water. Up to now, many methods have been reported, such as spectrophotometric methods,1317 which includes the most widely used colorimetric method based on the reaction of free chlorine with N-N-diethyl-p-phenylenediamine (DPD),13 chemiluminescence methods,18,19 electrochemical methods,20,21 liquid chromatographic methods,22,23 and so on. However, these methods suer
2012 American Chemical Society

from their respective disadvantages, such as the requirement for many types of reagents that may produce strong toxicity, low detection sensitivity, poor selectivity, and complicated performance. As a result, developing a facile and green detection method for free residual chlorine with higher detection sensitivity, better selectivity, and lower-cost is still of signicance. Graphene quantum dots (GQDs), as recently emerging carbon-based materials, are graphene sheets smaller than 100 nm.24 GQDs have attracted more and more attention due to their special advantages,2429 such as low toxicity, high uorescent activity, robust chemical inertness, and excellent photostability. However, to the best of our knowledge, most attention is focused on their preparation and property study; little attention has been paid to the analytical application of the GQDs. We recently prepared GQDs by pyrolyzing citric acid.30 The GQDs are surface-passivated by the incompletely carbonized citric acid and thus have good uorescence (FL) activity. The strong oxidative free chlorine can destroy the surface passivation layer of the GQDs, leading to signicant FL quenching. Herein, we developed a sensitive, selective, rapid, facile, and especially green method to detect free chlorine in drinking water based on the FL quenching of GQDs.
Received: July 11, 2012 Accepted: September 10, 2012 Published: September 10, 2012
8378
dx.doi.org/10.1021/ac301945z | Anal. Chem. 2012, 84, 83788382

Analytical Chemistry

EXPERIMENTAL SECTION Materials. Citric acid (99+%) was purchased from Alfa Aesar. A 10% sodium hypochlorite standardized by iodometric titration was used for daily preparation of diluted working standards. Other reagents were of analytical grade and used as received. Doubly distilled water was used throughout the experiment. Preparation of GQDs. The GQDs were synthesized by pyrolyzing citric acid as described elsewhere.30 Briey, 2 g of citric acid was put into a 5 mL beaker and heated to 200 C by a heating mantle for about 30 min, until the citric acid changed to an orange liquid. Then, the liquid was dissolved into 100 mL of 10 mg mL1 NaOH solution drop by drop, with continuous, vigorous stirring. The obtained GQD solution was adjusted to pH 8 with 10 mg mL1 NaOH solution and stored in the refrigerator. Analysis of Real Sample. A local tap water sample was collected and used in this study. Aliquots (500 L) of this tap water were spiked with standard sodium hypochlorite solutions (10 L; nal concentration, 08 M). The spiked samples were then diluted to 1000 L with PBS (0.1 M, pH 8.0) containing GQDs (nal concentration, 0.14 mg mL1) and then analyzed using the developed sensing technique. Methods. An atomic force microscopy (AFM) image was obtained by tapping-mode on a Nanoscope IIIa (Digital Instruments) with NSC15 tips (silicon cantilever, MikroMasch). UVvis absorption spectra were characterized by a UV/vis/NIR spectrophotometer (Lambda 750). All uorescent spectra were obtained by a FL spectrophotometer (Cary Eclipse Varian).

Article

The maximum excitation wavelength and maximum emission wavelength of the GQDs are 362 and 460 nm, respectively (see Figure 1). The GQDs exhibit an apparent UVvis absorption band centered around 362 nm, which corresponds well to the excitation spectra. FL Response of the GQDs to the Free Chlorine and the Corresponding Mechanism. As shown in Figure 2, the

Figure 2. UV and FL spectrum of 0.14 mg mL1 GQD solution in the absence and presence of 20 M free chlorine. The inset shows the uorescent photos of 0.7 mg mL1 GQD solution in the absence (right) and presence (left) of 100 M free chlorine illuminated by an UV beam of 365 nm.

The GQDs used in this study are nanosheets of 1.4 nm in height (the inset of Figure 1). The GQD aqueous solution shows bright blue emission under excitation of 365 nm UV light (8.9% quantum yield, selecting quinine sulfate as the standard). The emission wavelength of the GQDs is excitation independent.

RESULTS AND DISCUSSION

Figure 1. UV and FL spectrum of the obtained GQD solution (EX presents excitation spectra, EM presents emission spectra). The inset showed the AFM image of GQDs on mica substrates with the height prole along the line in the image.
8379

colorless GQD solution emits bright blue FL under the excitation at 365 nm, but the blue FL was signicantly quenched by the free chlorine (the inset of Figure 2). Their corresponding emission spectra (see Figure 2) show that the FL intensity at 460 nm has decreased by 80%. However, unlike the obvious changes in FL, nearly no change in the characteristic UV absorption of GQDs at 362 nm can be observed. At the same time, the zeta potential of the GQDs decreased from 20.9 to 41.2 mV as 20 M free chlorine was added into 0.14 mg mL1 GQD solution, suggesting that the surface states of the GQDs have been changed obviously after the addition of free chlorine. Although the exact mechanism of carbon-based quantum dots is still an open question, more and more evidence indicate that the FL properties of carbon-based quantum dots (including the GQDs) are aected obviously by their surface states.26,3133 For example, surface passivation26 and reduction treatment3133 both enhance the FL intensity of carbon-based quantum dots. Therefore, it is reasonable to believe that the free chlorine quenches the FL of GQDs by aecting their surface states. It has been discussed elsewhere that the GQDs used here should be surface-passivated, probably by the incompletely carbonized citric acid.30 The strongly oxidizing free chlorine can destroy the surface passivation of the GQDs, resulting in the quenching of the bright FL emission. To further prove our hypothesis, eects of other commonly used oxidants on the FL of the obtained GQDs are also investigated. Experimental results show those oxidants with stronger oxidation capability than free chlorine (such as KMnO4, K2CrO4, and OH) can also quench the FL of the GQDs, while those oxidants with weaker oxidation capability than free chlorine (such as O2, MnCl2, I2, and Br2) and some other reactive oxygen species except for OH (such as H2O2 and O2) have nearly no eect on the FL signal of the GQDs. These experimental results further verify that the
dx.doi.org/10.1021/ac301945z | Anal. Chem. 2012, 84, 83788382

Analytical Chemistry oxidation mechanism plays the key role in the FL quenching of GODs. Establishment of FL Sensing for Free Chlorine. To understand the response rate of the FL signal of the GQDs to the free chlorine, the time-dependent FL changes upon addition of 10 M hypochlorite were rst monitored. As shown in Figure 3, the FL intensity of GQDs at 460 nm was

Article

higher than 0.14 mg mL1, the FL intensity of GQDs will be insensitive to the addition of free chlorine (which equals the decrease of the GQDs concentration). Accordingly, 0.14 mg mL1 GQD solution was used for the later FL measurements. Another key factor for the sensing system is the pH value of solution. On one hand, the pH value may aect the FL activity of GQDs. As shown in Figure 5 (blue columns), in acidic

Figure 3. Time-dependent FL response of the 0.14 mg mL1 GQDs to 10 M free chlorine in pH 8 PBS.

quenched by 65% as soon as the hypochlorite was added into the solution and kept stable during the following 1 h observation. This result suggests that the quenching of free chlorine to the FL of GQD is quite rapid, implying a promising application in fast, stable, and convenient sensing. The concentration of GQDs is a key factor for this sensing system. To choose a suitable concentration of GQDs, the relationship between FL intensity of GQD solution and concentration of GQDs was studied (Figure 4). Typically, the

Figure 5. FL response of 0.14 mg mL1 GQDs in the absence and presence of 10 M free chlorine at dierent pH values.

Figure 4. Concentration-dependent FL response of GQDs in pH 8 PBS. The inset shows the linear response range of GQDs.

solutions, FL intensity of GQDs is relatively low and obviously pH dependent, which is unfavorable for developing a sensitive and stable sensing method. In contrast, the FL intensity of GQDs reaches a maximum and keeps stable at pH higher than 7, indicating that neutral and alkaline conditions might be suitable for this sensing system. On the other hand, pH value also may aect the forms of free chlorine in water.34 Free chlorine exists mainly as hypochloric acid (HClO) in neutral and weakly alkaline solutions (pH 39) and as hypochlorite anion (ClO) in strong alkaline solutions (pH > 9). The ClO exhibits weaker oxidation capability than the HClO, which is a disadvantage in destroying the surface passivation of the GQDs and quenching the FL intensity. As shown in Figure 5, the quenching eect of free chlorine on the FL intensity of GQDs decreases with increasing pH value in the strong alkaline range (pH > 9). Then, neutral and weakly alkaline conditions (pH 7 9) are suitable for this sensing system. Typically, a weakly alkaline condition (pH = 8) was chosen for this sensing system. Sensitivity of the Sensing System. Under the optimized conditions discussed above, the linear response range of the sensing system was measured. As shown in Figure 6, the FL intensity of GQDs at 460 nm is highly sensitive to free chlorine and decreases as the concentration of free chlorine is increased. There is a good linear relationship between the quenching ratio and the concentration of free chlorine in the range from 0.05 to 10 M (see inset of Figure 6 and eq 1):
(F0 F )/F0 = 5.92 104C + 0.0334 (R2 = 0.992)
(1)

FL intensity is directly proportional to the concentration of GQDs in the range from 0.02 to 0.14 mg mL1. However, the increasing tendency slows down when the concentration of GQDs is higher than 0.14 mg mL1. That is to say, the GQDs FL intensity is relatively insensitive to the change of concentration of GQDs at concentrations higher than 0.14 mg mL1. As mentioned above, the free chlorine quenches the FL signal of GQDs through changing the high FL GQDs into poor FL GQDs, namely, reducing the concentration of high FL GQDs. In other words, when the concentration of GQDs is
8380

where F0 and F are the FL intensity of GQDs in the absence and presence of free chorine, respectively, and C is the concentration of free chlorine. The theoretical detection limit, DL,35 was calculated to be 0.05 M (or equivalently 1.78 g L1), using the equation DL = 3/m, where is the relative standard deviation and m is the slope of the calibration graph (n = 12, condence level of 99.7%).The DL of the present GODdx.doi.org/10.1021/ac301945z | Anal. Chem. 2012, 84, 83788382

Analytical Chemistry

Article

Application. On the basis of the above results, the applicability of this proposed method for detection of free chlorine in water was demonstrated. We applied a standard addition method to measure the concentration of free chlorine in a local tap water sample (see Figure 8). The results show that

Figure 6. The FL response of 0.14 mg mL1 GQDs upon addition of various concentrations of free chlorine in a pH 8 solution (from top: 0, 0.05, 0.1, 0.3, 0.5, 0.7, 1, 2, 4, 6, 8, and 10 M). Inset: SternVolmer plot of FL quenching of the GQDs by free chlorine.

based FL sensing method is much lower than the detection limit of the most widely used DPD colorimetric methods (1.40 M or equivalently 49.70 g L1).13 Specicity of the Sensing System. The selectivity of this sensing method for free chlorine was evaluated before its application in a real sample. As mentioned above, only oxidants with stronger oxidation capability than free chlorine would quench the FL of GQDs. However, most natural water samples would not contain such strong oxidants, suggesting that this sensing system would not have interference by oxidants. Eects of some common ions (ClO4, Co3+, Br, NO3, CH3CO2, Cl, SO42, S2, SO32, CO32, PO43, NH4+, Cu2+, Ca2+, Mg2+, Zn2+, and Fe3+) and some heavy metal ions (Cd2+, Cr3+, Ag+, Pb2+, and Hg2+) were also investigated. As shown in Figure 7, these ions have nearly no interference to the FL response of GQDs even at a rather high concentration (common ions were 1 mM; heavy metal ions were 100 M; Hg2+ was 10 M). Apparently, the present sensing system exhibits excellent selectivity for the detection of free chlorine.

Figure 8. The FL response of 0.14 mg mL1 GQDs upon addition of 50% tap water and various concentrations of free chlorine in a pH 8 solution.

the concentration of free residual chlorine in the tap water sample was 5.66 0.06 M (n = 10, condence level of 95%). The value is consistent with the result obtained by the DPD colorimetric method, i.e., 5.52 0.36 M (n = 10, condence level of 95%). However, in ten times of measurements, the present FL sensing method showed a much lower relative standard deviation (1.43%) than the DPD method (8.73%), suggesting that the proposed method may be more excellent than the DPD method in the detection of free chlorine in water, especially for those samples with low concentration of free chlorine.

CONCLUSIONS In summary, the GQDs obtained from citric acid were used to detect free chlorine in water. The strong oxidative free chlorine can destroy the passivated surface of GQDs and quench the FL signal obviously. This method has several advantages: (1) The detection method is highly sensitive. The detection limit, 0.05 M, is the lowest among the reported methods. (2) This is a very green and simple method. Except for the nontoxic GQDs, no other reagent is used. (3) The FL method is fast. The response time is less than 1 min. (4) The linear response range is very wide, i.e., from 0.05 to 10 M. (5) The detection method exhibits excellent selectivity. This proposed sensing method has been successfully applied in the detection of free chlorine in tap water and is envisioned to have promising applications in drinking water quality monitoring.

Figure 7. Selectivity of the GQD-based sensor for free chlorine over other ions in pH 8 solution: concentrations of GQDs and free chlorine were 1.4 mg mL1 and 10 M respectively; concentrations of other common ions were all 1 mM; concentrations of heavy metal ions were 100 M (except for the Hg2+); concentration of Hg2+ was 10 M.
8381

AUTHOR INFORMATION

Corresponding Author

*E-mail: y.w.chi@fzu.edu.cn. Fax/Tel: +86-591-22866137.

Notes

The authors declare no competing nancial interest.

ACKNOWLEDGMENTS This study was nancially supported by National Natural Science Foundation of China (21075018), Program for New
dx.doi.org/10.1021/ac301945z | Anal. Chem. 2012, 84, 83788382

Analytical Chemistry Century Excellent Talents in Chinese University (NCET-100019), National Basic Research Program of China (Grant 2010CB732400), and the Program for Changjiang Scholars and Innovative Research Team in University (No. IRT1116).
(1) Bellar, T. A.; Lichtenberg, J. J.; Kroner, R. C. J. Am. Water Works Assoc. 1974, 66, 703706. (2) Rook, J. J. Water Treat. Exam. 1974, 23, 234243. (3) Schreiber, J. S. J. Am. Water Works Assoc. 1981, 73, 154159. (4) Report on Carcinogenesis Bioassay of Chloroform; National Cancer Institute: Washington, DC, 1976. (5) Theiss, J. C.; Stoner, G. D.; Skimkin, M. B.; Weisburger, E. K. Cancer Res. 1977, 37, 27172720. (6) Simpson, K. L.; Hayes, K. P. Water Res. 1998, 32, 15221528. (7) Emmanuel, E.; Keck, G.; Blanchard, J. M.; Vermande, P.; Perrodin, Y. Environ. Int. 2004, 30, 891900. (8) Tofant, A.; Ostovic, M.; Wolf, S.; Ekert Kabalin, A.; Pavicic, Z.; Grizelj, J. Vet. Med. 2010, 55, 394398. (9) Hinckley, A. F.; Bachand, A. M.; Reif, J. S. Environ. Health Perspect. 2005, 113, 18081813. (10) Haag, J. R.; Gieser, R. G. J. Am. Med. Assoc. 1983, 249, 2507 2508. (11) Neering, H. Contact Dermatitis 1977, 3, 279. (12) Fisher, A. A. Cutis 1984, 33, 2024. (13) Standard Methods for the Examination of Water and Wastewater, 16th ed.; APHA, AWWA, WPCF: Washington, D.C., 1985; pp 309 314. (14) Moberg, L.; Karlberg, B. Anal. Chim. Acta 2000, 407, 127133. (15) Bauer, R.; Rupe, C. O. Anal. Chem. 1971, 43, 421425. (16) Latinen, H. A.; Boyer, K. W. Anal. Chem. 1972, 44, 920925. (17) Pinkernell, U.; Nowack, B.; Gallard, H.; Gunten, U. V. Water Res. 2000, 34, 43434350. (18) Marino, D. F.; Ingle, J. D. Anal. Chem. 1981, 53, 455458. (19) Claver, J. J.; Miron, M. C. V.; Vallvey, L. F. C. Anal. Chim. Acta 2004, 522, 267273. (20) Murata, M.; Ivandini, T. A.; Shibata, M.; Nomura, S.; Fujishima, A.; Einaga, Y. J. Electroanal. Chem. 2008, 612, 2936. (21) Kishioka, S. Y.; Kosugi, T.; Yamada, A. Electroanalysis 2005, 17, 724726. (22) Sullivan, J.; Douek, M. J. Chromatogr., A 1998, 804, 113121. (23) Watanabe, T.; Idehara, T.; Yoshimura, Y.; Nakazawa, H. J. Chromatogr., A 1998, 796, 397400. (24) Ponomarenko, L. A.; Schedin, F.; Katsnelson, M. I.; Yang, R.; Hill, E. W.; Novoselov, K. S.; Geim, A. K. Science 2008, 320, 356358. (25) Dong, Y.; Chen, Q.; Zheng, X.; Gao, L.; Cui, Z.; Yang, H.; Guo, C.; Chi, Y.; Li, C. J. Mater. Chem. 2012, 22, 87648766. (26) Shen, J.; Zhu, Y.; Yang, X.; Li, C. Chem. Commun. 2012, 48, 36863699. (27) Lu, J.; Yeo, P. S. E.; Gan, C. K.; Wu, P.; Loh, K. P. Nat. Nanotechnol. 2011, 6, 247252. (28) Liu, R.; Wu, D.; Feng, X.; Mullen, K. J. Am. Chem. Soc. 2011, 133, 1522115223. (29) Niu, Z. Q.; Chen, J.; Hng, H. H.; Ma, J.; Chen, X. D. Adv. Mater. 2012, 24, 41444150. (30) Dong, Y.; Shao, J.; Chen, C.; Li, H.; Wang, R.; Chi, Y.; Lin, X.; Chen, G. Carbon 2012, 50, 57384743. (31) Bao, L.; Zhang, Z.; Tian, Z.; Zhang, L.; Liu, C.; Lin, Y.; Qi, B.; Pang, D. Adv. Mater. 2011, 23, 58015806. (32) Zheng, H.; Wang, Q.; Long, Y.; Zhang, H.; Huang, X.; Zhu, R. Chem. Commun. 2011, 47, 1065010652. (33) Li, L.; Ji, J.; Fei, R.; Wang, C.; Lu, Q.; Zhang, J.; Jiang, L.; Zhu, J. Adv. Funct. Mater. 2012, 22, 29712979. (34) Aoki, T.; Munemorl, M. Anal. Chem. 1983, 55, 209121. (35) (a) Flora, C. J.; Nieboer, E. Anal. Chem. 1980, 52, 10131020. (b) Chi, Y.; Chen, H.; Chen, G. Anal. Chim. Acta 1997, 354, 365373.

Article

REFERENCES

8382

dx.doi.org/10.1021/ac301945z | Anal. Chem. 2012, 84, 83788382