Facile synthesis of copper doped carbon dots and their application as a “turn-off” fluorescent probe in the detection of Fe³⁺ ions

Abstract

Heteroatom doped carbon dots, due to their excellent fluorescent properties and related applications, have attracted great attention from different fields. Herein, we reported for the first time a facile and economic approach to synthesize copper doped carbon dots (Cu-CDs) via a one-step hydrothermal approach. The chemical and fluorescent properties of these Cu-CDs as well as the mechanism involved for the enhancement of the fluorescent were investigated through various microscopic and spectroscopic analyses, such as transmission electron microscopy (TEM), fluorescence spectra, X-ray photoelectron spectroscopy (XPS) and Fourier transform infrared spectroscopy (FTIR). It is observed that the Cu-CDs with the average size of 3.76 nm are spherical and monodisperse. Additionally, the as-prepared Cu-CDs exhibited a strong emission at 440 nm when excited at 350 nm (λ_ex) and the photoluminescence quantum yields (PLQY) reached 9.81% after the optimization of synthesis conditions. The excellent fluorescent properties of Cu-CDs make them great candidates for the selective and fast detection of Fe³⁺ in range of 0.001–200 μM, while the limit of detection (LOD) is 1 nM, which can be further applied to detect Fe³⁺ in human blood serum and other biomedical applications.

---

Introduction

As a new member of the functional nanocarbon family, carbon dots (CDs) were first reported in 2004 by Xu et al.¹ and then the zero-dimensional (0D) carbon nanostructure attracted significant research curiosity.² Owing to the unique properties of CDs including chemical inertness, appreciable solvent dispersibility, excellent optical absorptivity, and low toxicity,^7–11 CDs have attracted a large amount of research, resulting in the synthesis of a variety of carbon-based 0D structures from diverse carbon source with distinct intrinsic structures, such as carbon dots (CDs), graphene quantum dots (GQDs),³ and carbon nanodots (CNDs).⁴ Due to the considerable ease of synthesis, functionalization, low cytotoxicity, and feasibility for facile large scale production, CDs with stable fluorescence are employed in a wide variety of applications including bio-imaging, medical diagnosis, optoelectronic devices, and fluorescent probes.^5,6 However, the unfavorably low fluorescence quantum yield of CDs inhibit their widespread use. To solve this issue, different pathways such as surface passivation, functionalization, and heteroatom doping are proposed to derive CDs with enhanced photoluminescence features. Among all these strategies, heteroatom-doping gained significant research attention owing to its facile and easy manipulating nature. For example, Yang et al.¹² reported a facile method to prepare nitrogen-doped CDs (N-CDs) using citric acid and ethylenediamine as precursors under hydrothermal condition, which are applied in sensors and bioimaging. Recently, Xu et al.⁵ reported sulfur-doped CDs (S-CDs) with 67% photoluminescence quantum yield (PLQY) generating from sodium citrate and sodium thiosulfate and the S-CDs show great performance on sensitive metal ion detection. Further, Dong and co-workers¹³ developed a facile, one-step, and high-yield hydrothermal method to synthesize nitrogen and sulfur co-doped CDs (N, S-CDs) using citric acid and L-cysteine. In addition, Wu et al.¹⁴ reported the preparation of Cu–N-doped CDs and investigated both the doping-promoted electron transfer and the performance of the CDs in photo oxidation reactions. However, most of the studies focused on development of non-metal co-doped CDs. So far, to best of our knowledge, there is no report pertaining to low cost and facile method of doping CDs with heavy metal solely without presence of any nonmetallic co-dopant. In this study, copper doped CDs (Cu-CDs) were prepared for the first time using a hydrothermal method, simply by using sodium citrate and cuprous chloride as precursors with an optimum molar ratio of 1 : 0.25 (sodium citrate: cuprous chloride). The properties of the as-synthesized Cu-CDs were investigated using a suite series of advanced analytic techniques, such as high-resolution transmission electron microscopy (TEM), fluorescence spectroscopy, X-ray photoelectron spectroscopy (XPS) and Fourier transform infrared spectroscopy (FTIR). Furthermore, the novel Cu-CDs were applied as fluorescent probes for selective, sensitive detection of Fe³⁺ ion. The obtained functional Cu-CDs displayed enhanced sensitive response towards Fe³⁺ in the concentration range of 0.001–200 μM in a response time of 60 seconds. Futuristically, the system could readily be adopted for various applications including bioanalysis of Fe³⁺ in human blood serum.^15,16

Experimental section

Materials

Sodium citrate and cuprous chloride were purchased from Tianjin Guangfu Technology Development Co., Ltd. Ferric trichloride was procured from Sinopharm Chemical Reagent Co., Ltd. The solutions were prepared using deionized water from Dongguanshi Qianjing Environmental Equipment Co., Ltd. Reagent grade Ce(NO₃)₃, MgCl₂, CuSO₄, MnCl₂, NiCl, ZnCl₂, KCl, NaCl, and FeCl₃, were obtained Tianjin Guangfu Technology Development Co., Ltd.

Synthesis of Cu-CDs

In this study, the Cu-CDs were synthesized by hydrothermal method. First, a mixture (25 mL) of sodium citrate (0.1 M) and cuprous chloride (0.025 M) was added in a 50 mL Teflon-lined stainless steel autoclave. The autoclave was maintained at a fixed temperature (190, 195, 200, 205 or 210 °C) for 6 h. The product was filtered using a cylindrical filtration membrane filter (0.22 μm) and purified sample was kept for further use.

Characterization

The morphology of the as-prepared Cu-CDs was characterized by high-resolution transmission electron microscopy (HRTEM; Model JEM-2100 and JEM-2100F, JEOL). The fluorescence measurements were performed with a fluorescence spectrophotometer (FS5 from Techcomp (China) Ltd). The samples were excited at 340 nm, and the emission spectrum in the range of 300 to 550 nm was measured. The slit width was fixed at 3 nm and 3.5 nm for of excitation and emission, respectively. X-ray Photoelectron Spectroscopy (XPS) analysis was carried out using a ESCALAB 250 spectrometer with a monochromatic X-ray source Al Kα excitation (1486.6 eV). Binding energies were calibrated based on C_1s at 284.6 eV. The quantum yield of the samples were recorded using FLS980 Fluorescence Spectrometer from Techcomp (China) Ltd.

Detection of the of Fe³⁺

For this, 10 μL Cu-CDs were diluted into 1000 μL using deionized water (termed as blank). At room temperature, the fluorescence emission intensity at 440 nm of the blank sample excited by light at 340 nm was measured and marked as F₀. Then the solution of Fe³⁺ with different concentration was added (10 μL, concentration range, 0–200 μM) into the above solution, mixed well and allowed to equilibrate for 1 min. After 1 min, the fluorescent intensity (F₁) of the samples were measured. The change of fluorescent intensity is defined as ΔF (ΔF = F₀ − F₁).

Results and discussion

Optimization of synthesis conditions

Synthesis of CDs can be classified into two main pathway: broken off from carbon resources and built up from carbon-containing precursor. In this study, sodium citrate and cuprous chloride, which are essential for the synthesis of Cu-CDs (Fig. S1†), were selected as the carbon precursor and copper precursor, respectively. Fig. 1(a) schematically illustrates the Cu-CDs synthesis process. Several factors, including the concentration of copper source in combination with sodium citrate, the temperature, and the time of the reaction, can affect the photoluminescence of Cu-CDs. Therefore, optimization of experimental conditions is essential to achieve Cu-CDs with excellent fluorescence properties. Initially, the ratio of cuprous chloride to sodium citrate varied from 0.15 to 0.75, maintaining other parameters such as reaction time (6 h) and temperature (200 °C) constantly. As can be observed in Fig. 1(b), at the optimum ratio of 0.25 between cuprous chloride and sodium citrate, the product displayed highest quantum yield. Lower or higher source ratios produced Cu-CDs with decreased fluorescence intensity. The decreased fluorescence at higher ratios could be possibly due to the blockage of the passivated surface defects by excess Cu in the ensemble.¹⁷ The effect of reaction temperature on the PLQY of as-synthesized Cu-CDs was also investigated. The reaction temperature increased in a step of 5 from 190 °C to 210 °C. The results indicated that the as-prepared Cu-CDs exhibited higher PLQY when synthesized at 200 °C (Fig. 1(c)). Similarly, the duration of the reaction was also varied to estimate the optimum reaction time. As illustrated in Fig. 1(d), Cu-CDs synthesized with 6 h reaction exhibiting highest fluorescence implying that 6 h is the optimum experimental condition for this reaction. Under these condition, a facile carbonization and efficient Cu doping occur to generate Cu-CDs with appreciable PLQY (9.81%). To our knowledge, this is the first product of single heavy metal doped CDs with stable blue fluorescence and high PLQY.¹⁸

Fig. 1 (a) Scheme of the synthesis of Cu-CDs with blue luminescence. Quantum yield (QY) of Cu-CDs as a function of (b) ratio of copper source to carbon source; (c) temperature of the hydrothermal method; (d) time of reaction.

Characterization

The structure and morphology of the Cu-CDs were analyzed using high-resolution transmission electron microscopy (HRTEM). According to Fig. 2(a), the Cu-CDs show a spherical morphology and a narrow size distribution in the range from 2 to 6 nm with an average size of around 3.76 nm (Fig. 2(b)). The effect of extreme environment such as highly oxidizing solutions (for example hydrogen peroxide) on Cu-CDs was also tested and the Cu-CDs manifested high stability and steady fluorescence against highly oxidizing H₂O₂ (0.1 mM) (Fig. 2(c)), which reveals the as-prepared Cu-CDs are promising material candidates for a wide range of applications for their robust fluorescence. What is more, it is observed that the fluorescent intensity of Cu-CDs is stable in a wide range of pH values (3–10) (Fig. S2†), which makes them a promising material used in biological field. Fig. 2(d) demonstrates the emission spectra of Cu-CDs excited using the light of wavelength range from 270 to 400 nm. Fig. 2(e) and (f) show the 3D map and contour plot of corresponding photoluminescence spectrum of Cu-CDs. Careful analysis of Fig. 2(d–f) can reveal that the as-prepared Cu-CDs have the strongest emission at 440 nm when excited at 350 nm (λ_ex). It could be also observed that Cu-CDs exhibit excitation independent emission properties, which could be attributed to the homogeneous surface structure, and monodispersity of Cu-CDs.² The UV/Vis absorption spectra recorded from an aqueous solution of Cu-CDs have only one prominent feature centered on 273 nm (Fig. S3†). Additionally, Raman spectroscopy was utilized to characterize the intrinsic structure of the C-dots (Fig. S4†) with different source ratio, the peak around 1600 cm⁻¹ could be assigned as G band which is correlated with sp² carbon networks and the peak around 1400 cm⁻¹ could be assigned as D band reflecting the disorder or defects in the graphitized structure.

Fig. 2 (a) Large area TEM image of Cu-CDs showing the monodispersed nature; (b) the diameter distribution of Cu-CDs calculated from TEM image; (c) effect of H₂O₂ on the fluorescence intensity of Cu-CDs; (d) photoluminescence spectrum of Cu-CDs; (e) the 3D map plot of the excitation spectrum and emission spectra to the Cu-CDs; (f) contour plot of photoluminescence spectrum of Cu-CDs.

The elemental compositions of the Cu-CDs were analyzed using X-ray photoelectron spectroscopy (XPS). Fig. S5† was the survey scan spectra of Cu-CDs, which showed the presence of C, Cu, O. According to the C_1s XPS spectrum (Fig. 3(a)), the carbon exists in two different states: O–C=O, and C–C. The significant portion of capping oxygen ions on the surfaces of Cu-CDs implies the high level of passivation in Cu-CDs. The O_1s spectrum (Fig. 3(b)) is consistent with C_1s results and confirms the existence of oxidized carbon in Cu-CDs. The Cu_2p spectrum (Fig. 3(c)) has a comparatively weaker signal pointing to the ultra-low doping in the structure. However, a clear signature of Cu confirms the doping. The presence of Cu_2p3/2 around binding energy 933 eV and Cu_2p1/2 around 956.7 ev indicates that the doped Cu also exists in the oxidized Cu⁺ state. The absence of satellite peak at 962 eV and 943 eV, characteristic of Cu²⁺ further confirms that the doped Cu in carbon dot is in Cu⁺ state. This could be plausibly due to the comparatively higher reactivity of Cu⁺ (compared to Cu²⁺).^17,18 The samples were analyzed using Fourier transform infrared spectroscopy (FTIR) spectroscopy as well (Fig. 3(d)). It is worth to note that in addition to characteristic aromatic stretching of carbon dots, there is the prominent signature of O–H stretch in Cu-CDs. This is consistent with the XPS observation which shows oxidized carbon presumably in the form of carbon connected to OH groups. The presence of polar OH group on Cu-CD surface is expected to render the CDs excellent water solubility. This will be expected to be extremely beneficial for employing the Cu-CDs for sensing application in an aqueous environment, such as biosensing and bio-imaging.^11,15

Fig. 3 (a) High-resolution C_1s XPS spectra for Cu-CDs; (b) high-resolution O_1s XPS spectra for Cu-CDs; (c) high-resolution Cu_2p XPS spectra for Cu-CDs; (d) FTIR spectra for Cu-CDs; (e) high-resolution C_1s XPS spectra for Cu-CDs with different reaction ratio; (f) XRD results for Cu-CDs.

The carbon dots with different reaction ratio, as mentioned in Fig. 1(b), are further tested with XPS to invest the mechanism of high quantum yield. According the carbon XPS spectrum (Fig. 3(e)), the content of O–C=O exhibited a maximum value at the ratio of 0.25, which is in agreement with highest quantum yield at the ratio of 0.25 (Fig. 1(b)). This result can be explained by the high passivation of carbon due to oxidation.²⁵ The high O% at the ratio of 0.25 (Table 1) provides a consistent evidence to prove the high carbon oxidation state. It is also noted that although Cu% between 0.1 and 0.5 is different (Table 1), both quantum yield is similar. This may be due to the similar C% and O% at both ratios, which resulting similar passivation. Fig. 3(f) presents the XRD patterns of Cu-CDs. It was observed that the carbon dots include Cu₂O, CuO and trace Cu, in which Cu₂O is a major chemical component for copper.

Table 1 Elemental analysis (right) for Cu-CDs with different ratio

Ratio	C%	O%	Cu%
0.1	64.82	34.93	0.24
0.25	62.25	37.37	0.38
0.5	64.66	34.95	0.39
0.75	68.49	31.35	0.16

---

Detection of the concentration of Fe³⁺

Fe³⁺ is one of the essential metal ions in human body, for example, iron stores in the body are regulated by intestinal absorption. Abnormal concentrations of Fe³⁺ linked to several diseases, such as anemia, intelligence decline, heart failure, cancer¹⁹ and so on. Thus, it is imperative to detect and monitor Fe³⁺ content in the human blood serum. In this research, the high water solubility, and stable luminescence of Cu-CDs were adopted to develop a novel fluorescence-based nanoprobe. The as-synthesized Cu-CDs demonstrated appreciable fluorescence quenching with fast response time. Fig. 4(a) shows the time-dependent fluorescence quenching of Cu-CDs in the presence of Fe³⁺ (100 μM). There was no apparent change in intensity pointing to the quick response time after 1 min, hence it could be understood that the fluorescence response is fast. Fig. 4(b) illustrates the change in fluorescence of Cu-CDs after the addition of different concentration of Fe³⁺ (0–200 μM) and equilibrated for 1 min. It is reasonable to conclude that more Cu-CDs show a concurrent decrease of fluorescence intensity with increasing concentration of Fe³⁺. As can be observed in Fig. 4(c), change in fluorescence intensity of the solution of the Cu-CDs (ΔF) is linearly dependent on concentrations of Fe³⁺, and linear regression can be defined as ΔF = 0.042 + 0.0028C (R² = 9967) with limit of detection of 1 nm. For practical utilization, a sensor in addition to being ultra-sensitive should be highly specific toward target as well. The selectivity of Cu-CDs towards different metal ions was also evaluated. For this a mixture of various ions including Na⁺, Mg²⁺, Ba²⁺, Cu²⁺, Ge³⁺, K⁺, Fe³⁺, Zn²⁺, Ni⁺ with 100 μM concentration were introduced into the Cu-CD solution and the fluorescence quenching at 440 nm was examined. As shown in Fig. 4(d), Fe³⁺ alone showed significant quenching on Cu-CDs, while other ions tested have no appreciable change. This categorically proves that the prepared Cu-CDs are ideal candidates for selective, sensitive detection of Fe³⁺, which with appropriate modification can be easily adapted for important sensing application in biology and various other fields. The photoluminescence of Cu-CDs quenched by Fe³⁺ is related to the electron transfer process between Cu-CDs and Fe³⁺, which was considered to be able to absorb on the surface of Cu-CDs, with the result that Fe³⁺ was coordinate with phenolic hydroxyl group on the surface of Cu-CDs.¹² This coordination interaction would lead to nonradiative electron/hole recombination, and results in the fluorescence quenching.

Fig. 4 (a) Time-dependent fluorescence changes of Cu-CDs in the presence of Fe³⁺ (100 mM); (b) emission spectra of the C-dots solution with different concentrations of Fe³⁺ (0, 0.5, 10, 50, 70, 100, and 200 μM); (c) the change of fluorescence intensity of Cu-CDs solution versus the concentration of Fe³⁺; error bars in (c) represent the standard deviations of three independent measurements; (d) the change of fluorescence intensity at 440 nm for Cu-CDs in the presence of various metal ions.

In order to explore the possibility of the practical application for the proposed sensor, it is applied for the detection of Fe³⁺ in the tap water. It can be oberserved from Fig. S6(a),† the fluorescence intensity of Cu-CDs decreased with the increase of the concentration of Fe³⁺ in the tap water, and the ΔF shown a linear relationship with the concentration of Fe³⁺ (ΔF = 0.045 + 0.00136C, R² = 9900). The results demonstrated that the proposed sensor was capable of detecting Fe³⁺ in the tap water. In consequence, the as-prepared Cu-CDs could be adopted as fluorescence probe for the detection H₂O₂ based on Fenton reaction as reported,^26,27 which is of great importance in environmental, pharmaceutical and clinical fields. Following that, a versatile fluorescent biosensing platform could be constructed for the detection of H₂O₂-related enzymes and substrates.

For further understanding the mechanism of copper doped CDs application on detection Fe³⁺, we have performed ab initio calculation based on two well-established density functional theoretical VASP^20,21 and OLCAO codes.^22–24 The structures relaxed by VASP and the properties calculated using OLCAO. The calculated effective charge of Cu is 10.554, which indicated the valence value of Cu is +1 in Cu-CDs, which is consistent with previous XPS analysis results, and the responding atomic orbital arrangement of Cu atom is 1s²2s²2p⁶3s²3p⁶3d¹⁰. The calculated density of states (DOS) of pure CDs and Cu-CDs are presented Fig. 5. As shown in Fig. 5, there are a few impurities levels in the forbidden band, which will results into the increases of the optical transition process,²³ namely enhanced fluorescence intensity somehow. The mentioned simulation results match well with our previous experimental analysis. As far as we know, it is the first simulation work involved single metal doped CDs, and the related electronic structure characterization on Cu-CDs. The calculated density of states of Cu-CDs indicates there are continuous impurity level inducing by Cu ions adsorption. The continuous intermediate levels avoid the photon energy repetitive absorption by the same electronic, and extend impurity level life resulting in the photon energy absorption by another photon. Moreover, the continuous impurity level serve as recombination and trapping centre leads to nonradiative recombination, which enhances the fluorescence efficiency of CDs (Fig. S7†).

Fig. 5 The total density of states (a) pure CDs; (b) Cu-CDs.

Conclusions

In summary, water-soluble, luminescent CDs doped with a single heavy metal atom were synthesized for the first time via hydrothermal method. The facile process employed standard, inexpensive precursors (sodium citrate and cuprous chloride) to generate blue luminescence with appreciable fluorescence quantum yield (9.81%). The obtained Cu-CDs showed spherical morphology with an average size of 3.76 nm. The monodispersity and highly surface passivated chemical framework of Cu-CDs, and their resultant excitation independent fluorescence was employed for selective and sensitive detection of Fe³⁺. The as-prepared CDs showed excellent detection capability towards Fe³⁺ over a wide concentration range of 0.001–200 μM. It is believed that the fast and selective detection capability of Cu-CDs for Fe³⁺, with intelligent modification, will plausibly lead to nimble strategies to diagnose various deceases including anemia, intelligence decline, heart failure, and cancer and will opening up a new phase-space for next-gen smart diagnostic systems.

Acknowledgements

We thank National Nature Science Foundation of China (No. 51505501), Science Foundation of China University of Petroleum Beijing (No. 2462014YJRC011), State Key Laboratory of Metal Matrix Composite, Shanghai Jiao Tong University (No. mmckf-14-11), and State Key Laboratory of Silicate Materials for Architectures, Wuhan University of Technology (No. 47152005, SYSJJ2015-10 and SYSJJ2016-05), Natural Science Foundation of Hubei Province with No. 2015CFB227, the Fundamental Research Funds for the Central Universities with No. 20410686 for the support.

Notes and references

1. X. Xu, R. Ray, Y. Gu, H. J. Ploehn, L. Gearheart, K. Raker and W. A. Scrivens, J. Am. Chem. Soc., 2004, 126, 12736.
2. Q. Xu, J. Zhao, Y. Liu, P. Pu, X. Wang, Y. Chen, C. Gao, J. Chen and H. Zhou, J. Mater. Sci., 2015, 50, 2571.
3. H. Li, X. He, Z. Kang, H. Huang, Y. Liu, J. Liu, S. Lian, C. H. A. Tsang, X. Yang and S. T. Lee, Angew. Chem., Int. Ed., 2010, 49, 4430.
4. M. J. Krysmann, A. Kelarakis, P. Dallas and E. P. Giannelis, J. Am. Chem. Soc., 2012, 134, 747.
5. Q. Xu, P. Pu, J. Zhao, Y. Liu, C. Dong, C. Gao, Y. Chen, J. Chen and H. Zhou, J. Mater. Chem. A, 2015, 3, 542.
6. S. Y. Lim, W. Shen and Z. Gao, Chem. Soc. Rev., 2015, 44, 362.
7. L. Tang, R. Ji, X. Cao, J. Lin, H. Jiang, X. Li, K. S. Teng, C. M. Luk, S. Zeng, J. Hao and S. P. Lau, ACS Nano, 2012, 6, 5102.
8. Z. Yang, M. Xu, Y. Liu, F. He, F. Gao, Y. Su, H. Wei and Y. Zhang, Nanoscale, 2014, 6, 1890.
9. S. Zhu, J. Zhang, S. Tang, C. Qiao, L. Wang, H. Wang, X. Liu, B. Li, Y. Li, W. Yu, X. Wang, H. Sun and B. Yang, Adv. Funct. Mater., 2012, 22, 4732.
10. J. Shen, Y. Zhu, C. Chen, X. Yang and C. Li, Chem. Commun., 2011, 47, 2580.
11. Z. Yang, M. Xu, Y. Liu, F. He, F. Gao, Y. Su, H. Wei and Y. Zhang, Nanoscale, 2014, 6, 1890.
12. S. Zhu, Q. Meng, L. Wang, J. Zhang, Y. Song, H. Jin, K. Zhang, H. Sun, H. Wang and B. Yang, Angew. Chem., Int. Ed., 2013, 52, 3953.
13. Y. Dong, H. Pang, H. B. Yang, C. Guo, J. Shao, Y. Chi, C. M. Li and T. Yu, Angew. Chem., Int. Ed., 2013, 52, 7800.
14. W. Wu, L. Zhan, W. Fan, J. Song, X. Li, Z. Li, R. Wang, J. Zhang, J. Zheng, M. Wu and H. Zeng, Angew. Chem., Int. Ed., 2015, 54, 6540.
15. Q. Zhao, S. Chen, L. Zhang and H. Huang, Anal. Methods, 2015, 7, 296.
16. L. Long, L. Zhou, L. Wang, S. Meng, A. Gong and C. Zhang, Anal. Chim. Acta, 2014, 812, 145.
17. J. Liu and Y. Lu, J. Am. Chem. Soc., 2007, 129, 9838.
18. S. Yang, J. Sun, X. Li, W. Zhou, Z. Wang, P. He, G. Ding, X. Xie, Z. Kang and M. Jiang, J. Mater. Chem. A., 2014, 2, 8660.
19. N. Narayanaswamy and T. Govindaraju, Sens. Actuators, B, 2012, 161, 304.
20. G. Kresse and J. Furthmller, Comput. Mater. Sci., 1996, 6, 15–50.
21. G. Kresse and J. Furthmüller, Phys. Rev. B: Condens. Matter, 1996, 54, 11169–11186.
22. W. Y. Ching and P. Rulis, J. Phys.: Condens. Matter, 2009, 21, 104–202.
23. W. Y. Ching, J. Am. Ceram. Soc., 1990, 73, 3135–3160.
24. W. Y. Ching and P. Rulis, Electronic Structure Methods for Complex Materials: The Orthogonalized Linear Combination of Atomic Orbitals, OUP Oxford, 2012.
25. Q. Xu, Y. Liu, C. Gao, J. Wei, H. Zhou, Y. Chen, C. Dong, S. T. Sreeprasad, N. Li and Z. Xia, J. Mater. Chem. C, 2015, 3, 9885–9893.
26. J. Wei, L. Qiang, J. Ren, X. Ren, F. Tang and X. Meng, Anal. Methods, 2014, 6, 1922–1927.
27. J. Wei, J. Ren, J. Liu, X. Meng, X. Ren, Z. Chen and F. Tang, Biosens. Bioelectron., 2014, 52, 304–309.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra27658f

‡ Contribute equally to this work.

---