Fluorescence alarming ON–OFF–ON switch derived from biocompatible carbon nanoparticle–hemoglobin–H₂O₂ interaction

Abstract

The present study comprises a deeper comprehension on the applicability of environmentally benign water soluble fluorescent Carbon Nanoparticles (CNs) synthesised from jamun (Syzygium cumini). We have synthesised two types of CNs i.e., surface non-passivated and surface passivated. The morphological investigation of either type of CN shows a narrow particle size distribution with an average size of 40 nm and 42 nm respectively. These CNs show potential candidacy as cell imaging agents due to cell viability up to a high concentration of 30 μg mL⁻¹ as revealed by cytotoxicity analysis. Furthermore, we employed hemoglobin as a target to explore the biosensing ability of CNs based on a turn ON–OFF mechanism. Moreover, we have made a reversible fluorescent turn ON–OFF–ON switch taking the CNs–Hb composite system in conjugation with a large excess H₂O₂. This unbelievable fluorescence enhancement of the CNs–Hb composite system on exposure to excessive H₂O₂ is due to a decrement in particle size that conversely increases the surface area to volume (S/V) ratio which in turn intensifies the PL spectra. Using the above intriguing changes, a nanoparticle based implication logic operation (IMP) was created with Hb and H₂O₂ as inputs.

---

1. Introduction

At first blush the broad absorption, narrow emission range and outstanding quantum efficiency of semiconductor quantum dots make them a powerful contender for detection and labelling.^1–3 However heavy metal toxicity^4,5 and low chemical stability of these semiconductor chalcogenides towards photo-oxidation⁶ force them to behave as a double edge sword. In addition to this, poor stability of organic dyes prevent their long term analysis. Making a move towards carbon nanoparticles, they permit a pertinent avenue in order to get rid of the above dilemma. Owing to their intriguing properties like high water solubility, excellent luminescence,⁷ good photostability, easy functionalization, robust chemical inertness, low toxicity and good biocompatibility,^8,9 due to which they exhibit manifold applications in bioimaging,^10,11 biolabelling,¹² and in drug delivery.¹³ Furthermore, they show great opportunities for photovoltaic devices,^14–16 energy conversion efficiency¹⁷ and related applications due to their photoinduced electron transfer (PET) property.¹⁸ The synthesis of CNs is mainly classified into two broad categories, namely, bottom-up and top-down methods.¹⁹ The synthesis of CNs through the bottom-up method involves the subjection of small molecules like fructose and glucose to external energy such as microwave pyrolysis, hydrothermal treatment and ultrasonication. On the other side, the top-down method involves the treatment of bulky molecules like multi-walled carbon nanotubes (MWCNTs) and graphite powder to harsh chemical or physical conditions.¹⁹ The vast majority of methods involving the synthesis of CNs require prolonged synthesis, post synthetic treatment, low yield and other more severe conditions.^20,21 Here, in this article we aim to produce fluorescent carbon nanoparticles (CNs) via an environmentally benign, less time consuming and cost effective hydrothermal method²² as a superior alternative for mass production of CNs (Scheme 1) using jamun (Syzygium cumini) extract as a oxygenated carbon source.

Scheme 1 Synthesis of CNs from anthocyanin pigments of Syzygium cumini through a hydrothermal process.

2. Experimental section

2.1 Material

Syzygium cumini (from a market), ethylamine, sodium chloride, manganese chloride, lithium chloride, potassium chloride, ferric chloride, cupric chloride, nickel chloride, calcium chloride and hemoglobin were purchased from Sigma Aldrich. H₂O₂, Sodium phosphate monobasic anhydrous and sodium phosphate (dibasic) were purchased from Alfa Aesar. All solutions were prepared using deionised water.

2.2 Synthesis of CNs

CNs-A. A mixture of Syzygium cumini extract (100 mg) was dissolved in 15 mL of distilled water. The mixture was then transferred into a 50 mL Teflon-lined stainless steel autoclave and heated at a constant temperature of 180 °C for 2 h. After completion, cooling was done under ambient conditions. The resultant brown solution was dissolved in 20 mL of distilled water then ultracentrifuged at 10 000 rpm for 1 h. The supernatant liquid was filtered through a 0.22 μm syringe filter and then the filtrate was stored at 4 °C for further characterization (Scheme SI1†).

CNs-B. A mixture of Syzygium cumini extract (100 mg) was dissolved in 200 μL of ethylamine and 15 mL of distilled water. The successive preparatory steps are similar to CNs-A.

2.3 Material characterization

A Shimadzu UV-vis 2450 spectrophotometer was used for recording UV-vis absorption spectra in the range of 200–600 nm. Photoluminescence spectra were taken by a Horiba scientific Fluoromax-4C spectrophotometer. A quartz cuvette with a 10 mm path length and volume of 3.5 mL was used for collecting the spectra. Infra red spectra (IR) of CNs were recorded using a Thermo scientific Nicolet 6700. The use of spectral subtraction provided reliable and reproducible results. An Invia Renishaw Raman spectrophotometer was used for recording the Raman spectra of CNs. Confocal laser scanning microscopy (CLSM) (Ziess LSM 780) was used for cell imaging. An Elementar vario Micro instrument was used for the CHNS analysis of CNs. The TEM study was carried out using a TEM TECHNAI G2 20 S-TWIN. A drop of highly diluted CN solution was placed on a carbon coated copper grid. Again a drop was added before drying it. Afterwards drying was carried out at ambient temperature.

3. Results and discussion

Here we have synthesized two types of CNs, one which was devoid of surface passivation designated as CNs-A and another that involves surface treatment using a surface passivating agent²³ like ethylamine, designated as CNs-B. The forthcoming section describes optical studies of CNs along with an IR study for surface state analysis of both types of CNs followed by size conformation using TEM analysis. Furthermore, we have performed a cytotoxicity test using synthesized CNs. Moreover, we have discovered a reversible turn ON–OFF–ON switch based on the CNs–Hb–H₂O₂ interaction.

3.1 Characterization

The surface functional groups were studied by Fourier transform infrared spectroscopy (FTIR) (Fig. 1a). A broad stretching vibration observed from 3100–3600 cm⁻¹ for CNs-A and B were assigned to the ν_(O–H) group. Stretching vibrations at 1649 cm⁻¹ and 1648 cm⁻¹, for CNs-A and B were assigned to the presence of the ν_(C=O) group. The occurrence of these functional groups improves the hydrophilicity. A bending vibration from 1350–1430 cm⁻¹ was assigned to the presence of the ν_(CH2) group. The C=C stretch with high sp² characteristics is observed at 1400–1600 cm⁻¹.

Fig. 1 (a) FT-IR spectra of the CNs. (b) Raman spectra of the CNs-A. (c) The optical absorption spectra in aqueous solution. (d and e) Photoluminescence emission spectra of CNs-A and B excited from 275–445 nm. (f) Photoluminescence lifetime decay curves of the CNs.

The Raman spectra of the CNs-A display two broad peaks at around 1386 cm⁻¹ and 1605 cm⁻¹, which are attributed to the D band (sp³-hybridized) and G band (sp²-hybridized) respectively. The D band is associated with vibrations of carbon atoms in the termination plane and related to sp³ defects while the G band corresponds to the vibration of sp²-hybridized carbon atoms. The ratio of the intensity of the disordered D band and crystalline G band I_D/I_G was 1.22, indicative of the nanocrystalline nature of CNs (Fig. 1b). This was further confirmed by TEM analysis. Elemental analysis revealed the composition of the CNs to be C 53.60 wt%, H 4.31 wt% and N 0.0 wt%, for CNs-A and C 56.22 wt%, H 4.77 wt% and N 2.18 wt% for CNs-B.

The UV-vis absorption spectra display a broad absorption band in the range of 200–600 nm for both the CNs-A and CNs-B. The CNs-A show relatively more prominent absorption bands located at 250 nm for the π–π* transition with a lower intensity absorption shoulder (compared to the π–π* transition) at around 288 nm for n–π* transition. The former π–π* transition is strongly coupled with surface electronic states. Compared to CNs-A, CNs-B involving surface passivation show a less featured absorption band at around 253 nm for the π–π* transition with an absorption shoulder located at 325 nm for the n–π* transition shown in Fig. 1c. It may be assumed that it was the surface amino-groups that passivate the traps on the surface of CNs-B and lead to a broad transition in the region of 290–350 nm with the shoulder at 325 nm (Fig. 1c). Moreover, the n–π* transition associated with surface passivated CNs exhibits a bathochromic shift by 37 nm relative to CNs-A. Unlike the n–π* transition, the π–π* transition is steady for both CNs-A and B. The absorption spectra of both the CNs-A and B portray high energy tailing in the visible region which was ascribed to Mie scattering caused by nanoparticles.²⁴

Moreover, Fig. 1d and e shows the photoluminescence behavior of CNs-A and CNs-B respectively. The shift in emission peak positions with different excitation wavelengths arose not only from different sizes of CNs but also from different emissive sites on CNs.²⁵ CNs-A exhibit excitation-wavelength-dependent photoluminescence (PL) properties with emission peaks ranging from 300 nm to 650 nm on excitation from 275 nm to 445 nm. The strongest fluorescence emission band located at 426 nm, was observed under 335 nm excitation. In CNs-A, there was a clear bathochromic shift with the increase of excitation wavelength, which explains multiple transition modes with different probabilities (Fig. 1d). However, CNs-B exhibit excitation-wavelength-independent photoluminescence properties with emission peaks ranging from 300 nm to 650 nm centered at 500 nm on excitation from 275 nm to 445 nm. The related fluorescence spectrum shows the presence of a single transition mode. Moreover, the emission peak of CNs-B was red shifted by a 74 nm contrast to the strongest intensity peak in CNs-A centered at 426 nm. On the basis of the above interpretation it is evident that the photoluminescence behavior is not centered on the core shell but is also due to surface functionality.

Fig. 1f shows the fluorescence decay of CNs-A and CNs-B in aqueous solution on excitation at 375 nm using the Time Correlated Single Photon Counting (TCSPC) technique. This shows tri-exponential and bi-exponential behavior with an average lifetime τ_(average) of 10.23 ns and 20.53 ns for CNs-A and CNs-B respectively (Table 1). This measured lifetime is related to the radiative and non-radiative electron–hole recombination processes. The radiative rate, k_r, is correlated to the fluorescence quantum yield, ϕ, as k_r = ϕ/τ_(average). The fluorescence radiative rate constant was calculated by taking an average lifetime of 10.23 ns and 20.53 ns for CNs-A and CNs-B respectively. It was found to be in the order of 10⁵ s⁻¹ and 10⁶ s⁻¹ for CNs-A and CNs-B respectively. It is noteworthy that k_r for CNs-B is 10 times more than that of CNs-A which suggests that the radiative recombination in CNs-B was faster than that of CNs-A. Moreover, k_r of this order suggests that emission in these CNs originates from an optically allowed transition. Whilst the non-radiative rate constant of CNs was calculated by using ϕ = k_r/(k_r + k_nr), which was found to be in the order of 10⁷ s⁻¹ for either type of CN, it is also very fast. These observed recombination processes having a PL lifetime in the nanosecond range result from the strong coupling between the extended π electron system of the core with that of surface functionalities.

Table 1 Lifetime decay of CNs-A and B

Sample	Excitation (nm)	τ1 (ns)	τ2 (ns)	τ3 (ns)	Average τ (ns)	CHISQ
CNs-A	375	0.43(14.84%)	2.75(57.78%)	13.60(27.38%)	10.23	1.26
CNs-B	375	6.46(50.72%)	24.37(49.29%)	—	20.53	1.26

---

The morphology of the CNs was characterized using transmission electron microscopy (TEM) which shows the high degree of monodispersity with a narrow size distribution range having an average size of 40 and 42 nm for CNs-A and B respectively (Fig. 2a and c). The presence of lattice fringes in High Resolution Transmission Electron Microscopy (HRTEM) shows the crystalline nature of CNs with lattice spacing of 0.35 and 0.27 nm for CNs-A and B respectively (Fig. 2a (inset) and d). Selected Area Electron Diffraction (SAED) shows a concentric ring pattern along with a bright spot which is again significant proof for the crystallinity of CNs-A (Fig. 2b). Results obtained from the TEM study are consistent with the Raman analysis.

Fig. 2 (a) TEM image of CNs-A with scale bar 200 nm (inset: HRTEM image showing the existence of crystalline parts) (b) SAED pattern of the CNs-A. (c) TEM image of CNs-B with scale bar 50 nm (d) HRTEM image of CNs-B.

3.2 Application in bioimaging

A possible application of as synthesized carbon nanoparticles is explored in cell-imaging. The inherent cytotoxicity was measured using C3H10T1/2 cells through the MTT assay. Cells were treated with CNs at final concentrations of 15 and 30 μg mL⁻¹. After 24 h of treatment the old culture medium was replaced with fresh medium and 5 mg mL⁻¹ of MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) dye was added to each well in triplicate. MTT added cells were incubated at 37 °C for 4 h. After completion of the incubation, the MTT containing medium was aspirated and then 100 μL of dimethyl sulphoxide (DMSO) was added to dissolve the formazone crystal formed by live cells. As depicted in Fig. 3c we did not observed significant cell death with CNs-A and B at a 15 μg mL⁻¹ concentration of CNs (p < 0.05). Furthermore, we wanted to observe whether increasing the concentration of CNs causes more severe effects on cell viability or not. The C3H10T1/2 cells were treated with 30 μg mL⁻¹ of CNs which causes no significant cell damage. Overall, from this experiment we concluded that a concentration of 15 μg mL⁻¹ can be safe for other studies. CNs-A and B show fluorescence in living cells suggesting that these may be associated with the efficient delivery of nanoparticles inside the cellular environment leading to more intense signals (Fig. 3a and b).

Fig. 3 (a) CLSM images showing fluorescence signals of C3H10T1/2 cells imaged by CNs-A and B under excitation 493 nm. (b) Cytotoxicity tests for CNs-A and B.

3.3 Application in biosensing

Now, our aim is to explore the biosensing application of synthesized CNs by perceiving its biological interaction with hemoglobin (Hb) through photoluminescence spectroscopy. In order to perform a study of CNs with hemoglobin, we employed ferric ion as a marker to investigate whether the quenching in photoluminescence is due to an iron moiety in heme or from the globin part.

A literature survey figures out that CNs are efficient nanosensors owing to their narrow range of selectivity.^26–30 Herein, the Fe³⁺ ion imparts a drastic change in PL quenching as compared to other metal salts (Fig. 4a and b). This is attributed to non-radiative electron transfer that involves partial transfer of an electron from the excited state of CNs to the d orbital of the Fe³⁺ ion.³¹ Although similar responses were obtained for both CNs A and B, CNs-B show a comparatively higher sensitivity toward the Fe³⁺ ion due to amine induced modulation of the chemical and electronic transition.^32,33 Successive addition of the Fe³⁺ ion leads to a gradual decrease in PL intensity at 435 nm revealing that the intensity of the CNs-A was sensitive to Fe³⁺ ion concentration (Fig. SI1†) (a similar study was performed with CNs-B and the Stern–Volmer plot for quenching in CNs A and B is provided in Fig. SI2 and Table SI1†).

Fig. 4 (a) Fluorescence response of CNs-A in the presence of different cations. (b) Performance of CNs-A comparing fluorescence intensities in the presence of different metal ions.

To rule out the possibility of the presence of any fluorescence quenching caused by the globin moiety, a parallel experiment using bovine albumin serum protein has been carried out to ensure the cause of quenching. It has no emission in the range of 250–650 nm (Fig. 5e). The overall fluorescence from the CNs-protein composite system is due to the CN moiety itself, and there is no other fluorescence contribution from the protein part. Therefore the aforementioned analysis concluded that, in particular, the heme moiety in haemoglobin accounts for the quenching phenomenon.

Fig. 5 (a) Optical response of CNs-A in the presence of different concentrations of Hb (a–g): 0–5 (μM). (b) Typical PL quenching of CNs-A in different concentrations of Hb (a–g): 0–5 (μM). (c) Linear relationship between I₀/I and Hb concentration in the range of 0–5 (μM). (d) UV-vis spectrum of CNs-A (1) and Hb (2) and the emission spectrum of CNs-A excited at 335 nm (3). (e) Fluorescence response of CNs in the absence and presence of protein (bovin albumin serum). (f) Photoluminescence lifetime decay curves of the CNs (CNs-A, CNs-A + Hb, CNs-A + H₂O₂, CNs-A + Hb + H₂O₂).

This overwhelming effect Fe³⁺ over other metal salts laid the foundation of further study that covers the biosensing application of CNs using haemoglobin. The absorption spectrum of hemoglobin shows two peaks, one of which is weak and less prominent at 270 nm whereas another is strong and centred at 405 nm (Fig. 5a).

3.4 Inner filter effect (IFE)

Scrutinizing the absorption spectra of CNs-A, having two bands at 250 nm and 288 nm and their emission centred around 426 nm on excitation with 335 nm, along with the absorption spectrum of hemoglobin, indicate the presence of an overlapping absorption spectrum of haemoglobin with that of the absorption and emission spectra of CNs (Fig. 5d). Here, the absorbance enhancement of Hb in the presence of CNs (Fig. 5a) was directly converted into the fluorescence quenching of CNs in the photoluminescence spectra (Fig. 5b). This overlapping of spectra ensures the possibility of the inner filter effect (IFE) occurring in a very efficient manner.^26–30,34 The fluorescent assay based on the IFE is an easy and sensitive way because the change in absorbance of a fluorophore transforms exponentially into a fluorescence intensity change.²⁷ Primarily the IFE is treated as an error in spectrofluorimetry. Recent studies enlighted by Shang and Dong and Xu et al. showed the IFE of AuNPs on the model fluorophore MDMO-PPV and quantum dots respectively.^27a,b

Moreover, Zhang et al. demonstrate the IFE of Cr(VI) using a hydrophilic ionic chemosensor.^27c In addition to this Huang et al.³⁵ and Barati et al.³⁶ showed the IFE of Hemoglobin on the fluorescence quenching of CNs. The high molar extinction coefficient of Hemoglobin and the efficient overlap of its absorption spectrum with the absorption and emission spectrum of CNs make it a suitable absorber for the IFE. Furthermore for a detailed explanation whether the absorption wavelength and/or emission light were/was absorbed by the hemoglobin in the system based on the IFE, we have performed two sets of experiments. The first set of experiments proves that the fluorescence decrease was excitation wavelength-dependent. The maximum fluorescence decrease occurs at 405 nm (irrespective to Hb concentration), which was treated as the absorption maximum of the hemoglobin peak associated with a higher molar extinction coefficient Fig. 6a. Besides this, there was also a fluorescence decrease at 275 nm which was related to the absorption maximum of another hemoglobin peak having a lower molar extinction coefficient. Furthermore the second set of experiment shows the extent of fluorescence quenching at different emission wavelengths with excitation at 335 nm. Herein, the maximum decrease in fluorescence occurs at a wavelength of 400 nm which was near to the absorption maximum of hemoglobin. Consequently, Hb absorbs emission light at 380 nm, 400 nm and 420 nm more intensively than longer wavelengths Fig. 6b. This investigation shows that hemoglobin absorbs the emission light of CNs. These two experiments finally suggested that the fluorescence of CNs mainly decreased by the IFE of Hb rather than other processes. Further proof for the IFE comes from the lifetime measurement before and after the addition of hemoglobin discussed in the upcoming section. Here successive addition of Hb increases the absorbance of Hb, which would shield absorption and excitation light from the fluorophore. Thus there was less light available for the fluorophore and, as a result, the absorption and emission intensity of CNs decreased.

Fig. 6 (a) The relative fluorescence intensity at different wavelengths versus the equivalent of Hb. (b) The relative fluorescence intensity at different emission wavelengths versus the equivalent of Hb. The excitation wavelength was 335 nm.

Further successive addition of haemoglobin in CNs results in considerable quenching of fluorescence intensity along with a gradual bathochromic shift thus CNs can used as fluorescent probes for haemoglobin detection (Fig. 5b).^35,36 Variation of fluorescence intensity of CNs is proportional to Hb concentration, which is well explained by considering the Stern–Volmer relationship:

F₀/F = 1 + k_qτ₀[Q] = 1 + k_sv[Q] (1)

where F₀ and F are designated as fluorescence intensities in the absence and presence of a quencher respectively, k_q is the bimolecular quenching constant, where τ₀ is designated as the lifetime of CNs in the absence of hemoglobin, k_sv is the Stern–Volmer quenching constant (obtained from the slope of the curve) and [Q] is the quencher concentration (Hb). This gives a linear regression equation with k_sv 2.99 × 10⁴ L mol⁻¹ and 2.18 × 10⁴ L mol⁻¹ along with a correlation coefficient of 0.98 and 0.99 for CNs-A and CNs-B respectively. Furthermore, we have evaluated k_q that shows the accessibility of the CNs to the quencher or the efficiency of quenching. The value of k_q is found to be 2.9 × 10¹² M⁻¹ s⁻¹ and 1.06 × 10¹² M⁻¹ s⁻¹ for CNs-A and CNs-B (Fig. SI3 and Table SI2†) respectively. However, in the case of diffusion-controlled quenching, the value of k_q is near to 1 × 10¹⁰ M⁻¹ s⁻¹ while a smaller value of k_q shows steric shielding of the fluorophore thus a low quenching efficiency. Here, the apparent value of k_q is larger than the diffusion-controlled limit which shows a binding interaction between CNs and Hb. This may be regarded as significant proof for the formation of the ground state complex between CNs and Hb. Moreover, on the basis of the k_q value, it is concluded that the binding interaction of CNs-A is slightly higher than that of the CNs-B. This bimolecular quenching constant is directly related to collisional frequency and quenching efficiency.

In order to gain deeper insight into the quenching phenomenon, we have further elaborated our discussion using k_sv. It is important to recognize that the obtained linear Stern–Volmer plot does not prove whether the quenching is dynamic or static since both can give a linear Stern–Volmer plot.³⁷ These two main quenching processes can be differentiated by temperature variation, viscosity or a time related single photon counting measurement (TCSPC). Therefore we have recorded the fluorescence life time decay curve of the CNs-A and CNs-A–Hb system (Fig. 5f) using a TCSPC experiment in order to observe the change in fluorescence lifetime of the fluorophore before and after addition of the quencher. Whilst static quenching involves a constant lifetime decay of the fluorophore both in the presence and absence of the quencher, dynamic quenching involves the change in fluorescence lifetime of the fluorophore after addition of the quencher.³⁷

We obtained well fitted triexponential decay components in the absence and presence of Hb which give an average lifetime of 10.22 and 10.09 ns (Table 2). A constant value of the average fluorescence lifetime before and after the addition of the quencher clearly reveals the presence of static quenching in the CNs–Hb system. The unaltered lifetime obtained after addition of Hb involves the formation of a non-emissive ground state complex. This non emissive complex does not decrease the decay time of uncomplexed CNs. The entire fluorescence is only because of remaining uncomplexed CNs.³⁷ Moreover the constant lifetime before and after the addition of the quencher indicates no complex formation in the excited state or energy transfer between Hb and CNs. On the basis of this fact it is concluded that the quenching is due the IFE rather than other processes.

Table 2 Life time decay of CNs-A with different inputs

Samples	Excitation (nm)	τ1 (ns)	τ2 (ns)	τ3 (ns)	Average τ (ns)
CNs-A	375	3.97(39.86%)	0.70(13.3%)	12.13(46.82%)	10.22
CNs-A + Hb	375	3.77(39.73%)	11.89(47.84%)	0.61(12.43)	10.09
CNs A + H2O2	375	4.00(39.57%)	0.66(12.48%)	11.81(47.94%)	9.99
CNs-A + Hb + H2O2	375	4.00(39.66%)	0.65(13.31%)	12.02(47.03%)	10.14

---

Another piece of evidence for the formation of the ground state complex between CNs and Hb comes from the absorption spectrum of the CNs, Hb and CNs–Hb composite system.³⁶ The non-superimposition of the absorption spectrum of the CNs–Hb composite and the sum of the individual absorption spectrum of CNs and Hb within the experimental error confirm the formation of the ground state complex. The enhancement of the effective size of the ground state complex (CNs–Hb composite) is observable by a bathochromic shift of the absorption spectra of the composite as compared to the sum of the individual absorption spectrum of the CNs and Hb system (Fig. SI4†).

As from the above discussion it is clear that hemoglobin itself has an extremely high potential for fluorescence quenching of CNs.^20,21 Now the forthcoming section involves building of a reversible turn ON–OFF–ON switch based on the interaction of excessive H₂O₂ with the preceding CNs–Hb composite system. A literature survey conceded that further application of H₂O₂ leads to more pronounced fluorescence intensity quenching based on hydroxyl radical (˙OH) formation which was accomplished by the Fenton reaction (Scheme 3).^38–42 This ˙OH has a very high reduction potential of 2.74 V that makes it a very efficient oxidising agent. This effectively oxidises the surface emissive group of CNs, which leads to the quenching in fluorescence intensity of CNs.

Scheme 2 Mechanism of the turn ON–OFF–ON system in CNs under different input conditions of Hb and H₂O₂.

Scheme 3 Interaction of the CNs–Hb composite with H₂O₂.

To the best of our knowledge, it is the first report which shows an enormous fluorescence enhancement on excessive addition of H₂O₂ to the CNs–Hb composite (Fig. 8a). We have made four systems to understand the interaction of excessive H₂O₂ and the CNs–Hb composite. The first system has CNs-A alone, the second system has CNs-A + Hb, the third system has CNs-A + H₂O₂ and the fourth system has CNs-A + Hb + H₂O₂. The photoluminescence spectra involving the interaction of H₂O₂ with CNs show high fluorescence enhancement as compared to CNs alone. The TEM study affirmed a noticeable reduction in particle size from 40 nm to 4 nm on addition of a large excess of H₂O₂ (Fig. SI7†). This decrement in particle size conversely increases the surface area to volume ratio which is strong evidence for enhanced photoluminescence in the CNs-A–H₂O₂ system.

There is also an increment in photoluminescence intensity of the CNs–Hb system on addition of H₂O₂ (Fig. 8a). However, the peak intensity is less as compared to the CNs A–H₂O₂ system. It is proposed that there are two process working simultaneously in the CNs A–Hb–H₂O₂ system, viz. elevation in the surface area to volume ratio of CNs on addition of a large excess of H₂O₂ that accounts for the credit point of the PL enhancement in Fig. 8a. Secondly, generation of ˙OH by the Fenton reaction occurring in the reaction medium, regarded as the debit point for the PL enhancement. It is a well known fact that H₂O₂ causes degradation of haemoglobin thereby releasing the Fe²⁺ ion from the complexed CNs A–Hb composite. This released Fe²⁺ (derived according to Scheme 3) and H₂O₂ act as scavengers for ˙OH and convert it into ⁻OH and HO₂˙ respectively (Scheme 3 proposed mechanism in a large excess of H₂O₂).³⁸ In this way the majority of the ˙OH radical is consumed readily before imparting its effect on the emissive group present on the surface of CNs. However the remaining few unconsumed ˙OH radicals present in the medium impart their role on CNs. Therefore the quenching effect of the ˙OH radical was overshadowed through its overconsumption by H₂O₂. Hence besides having ˙OH radicals in the system, we obtained enhanced fluorescence (Fig. 8a). A constant lifetime after addition of excessive H₂O₂ in either system i.e., CNs A and the CNs A–Hb composite system are direct evidence for the absence of any type of energy transfer or electron transfer process. The entire phenomenon of fluorescence turn ON–OFF–ON is solely governed by tuning in the particle size of CNs (Table 2).

From the above discussion it is apparent that the overall PL enhancement of the CNs A–Hb–H₂O₂ system under the influence of H₂O₂ is due to mitigation in the particle size (Scheme 2). We have also performed similar experiments with a phosphate buffer of pH 7.4 to normalize the system in a biological medium. Analogous results were obtained regarding the quenching trend (Fig. 7a) and PL enhancement on adding an excess of peroxide (Fig. SI5†). Variation in PL intensity of CNs with hemoglobin concentration maintains linearity up to a very high concentration of 10 μM with k_sv, the correlation coefficients are shown in Table SI3.† A similar study for CNs-B under control pH 7.4 is shown in Fig. SI6 and Table SI3.†

Fig. 7 (a) Typical PL quenching of CNs-A in different concentrations of Hb 0–10 μM (a–k). (b) Linear relationship between I₀/I and Hb concentration in the range of 0–10 (μM) under controlled pH 7.4.

This reversible fluorescence ON–OFF–ON switch influences the construction of a logic operation for nanoparticles as shown in Fig. 8a. IMPLICATION logic gate is obtained using Hb and H₂O₂ as inputs. The system is considered to be in the ON state when it is above the threshold value and designated as 1. It remains OFF when below the threshold value and denoted as 0. Emission intensity at 421 nm was assigned as output 1 (Fig. 8a). In the case of output 1 a fluorescence of 7.5 × 10⁴ has been taken as a threshold value. In the presence of Hb the fluorescence intensity of the solution is below the threshold value (0). In all other circumstances it remained in the ON state (1). This logic operation resembles the IMPLICATION (Fig. 8b) logic gate and is represented in Fig. 8c as a combinatorial logic scheme.

Fig. 8 (a) Emission spectra of CNs-A under different input conditions of Hb and H₂O₂ with a horizontal line (dashed) that marks the threshold value. (b) The truth table. (c) The combinatorial logic scheme.

In summary, we have demonstrated an easy one step hydrothermal approach towards water soluble photoluminescent CNs from Indian jamun (Syzygium cumini) as a carbon source. The toxicity studies exhibit that the CNs display zero to very low toxicity to a biological environment. Furthermore, we employed hemoglobin as a targeting moiety to explore the biosensing ability of CNs based on the turn ON–OFF mechanism. Moreover, we have made a reversible fluorescent turn ON–OFF–ON switch taking the CNs–Hb composite system in conjugation with a large excess of H₂O₂.

Acknowledgements

We thank Prof. Pradipta Banerjee, Director, Indian Institute of Technology Roorkee, India for his constant support and encouragement. The authors would also like to thank Prof Partha Roy, Department of Biotechnology, IIT Roorkee for performing confocal microscopy studies. P. K. gratefully acknowledges the Department of Science and Technology (SB/FT/CS-135/2012), New Delhi, India for financial support. PJSR and PS acknowledge MHRD, India, for their junior research fellowship.

References

1. X. Michalet, F. F. Pinaud, L. A. Bentolila, J. M. Tsay, S. Doose, J. J. Li, G. Sundaresan, A. M. Wu, S. S. Gambhir and S. Weiss, Science, 2005, 307, 538–544.
2. I. L. Medintz, H. T. Uyeda, E. R. Goldman and H. Mattoussi, Nat. Mater., 2005, 4, 435–446.
3. X. H. Gao, Y. Y. Cui, R. M. Levenson, L. W. K. Chung and S. M. Nie, Nat. Biotechnol., 2004, 22, 969–976.
4. A. M. Derfus, W. C. W. Chan and S. N. Bhatia, Nano Lett., 2004, 4, 11–18.
5. C. Kirchner, T. Liedl, S. Kudera, T. Pellegrino, A. M. Javier, H. E. Gaub, S. Stolzle, N. Fertig and W. J. Parak, Nano Lett., 2005, 5, 331–338.
6. W. V. Sark, P. Frederix, D. J. V. D. Heuvel, H. C. Gerritsen, A. A. Bol, J. N. J. V. Lingen, C. D. Donega and A. Meijerink, J. Phys. Chem. B, 2001, 105, 8281–8284.
7. Y. P. Sun, B. Zhou, Y. Lin, W. Wang, K. A. S. Fernando, P. Pathak, M. J. Meziani, B. A. Harruff, X. Wang, H. Wang, P. G. Luo, H. Yang, M. E. Kose, B. Chen, L. M. Veca and S. Y. Xie, J. Am. Chem. Soc., 2006, 128, 7756–7757.
8. S. T. Yang, L. Cao, P. G. J. Luo, F. S. Lu, X. Wang, H. F. Wang, M. J. Meziani, Y. F. Liu, G. Qi and Y. P. Sun, J. Am. Chem. Soc., 2009, 131, 11308–11309.
9. G. Leménager, E. D. Luca, Y. P. Sun and P. P. Pompa, Nanoscale, 2014, 6, 8617–8623.
10. S. T. Yang, X. Wang, H. Wang, F. Lu, P. G. Luo, L. Cao, M. J. Meziani, J. H. Liu, Y. Liu, M. Chen, Y. Huang and Y. P. Sun, J. Phys. Chem. C, 2009, 113, 18110–18114.
11. K. Jiang, S. Sun, L. Zhang, Y. Lu, A. Wu, C. Cai and H. Lin, Angew. Chem., Int. Ed., 2015, 54, 5360–5363.
12. C. C. Fu, H. Y. Lee, K. Chen, T. S. Lim, H. Y. Wu, P. K. Lin, P. K. Wei, P. H. Tsao, H. C. Chang and W. Fann, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 727–732.
13. T. Sun, Y. S. Zhang, B. Pang, D. C. Hyun, M. Yang and Y. Xia, Angew. Chem., Int. Ed., 2014, 53, 12320.
14. (a) H. Ding, F. Du, P. Liu, Z. Chen and J. Shen, ACS Appl. Mater. Interfaces, 2015, 7, 6889–6897; (b) Y. Xu, M. Wu, Y. Liu, X. Z. Feng, X. B. Yin, X. W. He and Y. K. Zhang, Chem.–Eur. J., 2013, 19, 2276–2283.
15. Z. Yang, M. H. Xu, Y. Liu, F. J. He, F. Gao, Y. J. Su, H. Wei and Y. F. Zhang, Nanoscale, 2014, 6, 1890–1895.
16. S. Chandra, P. Patra, S. H. Pathan, S. Roy, S. Mitra, A. Layek, R. Bhar, P. Pramanik and A. Goswami, J. Mater. Chem. B, 2013, 1, 2375–2382.
17. M. Lan, Y. Di, X. Zhu, T. W. Ng, J. Xia, W. Liu, X. Meng, P. Wang, C. S. Lee and W. Zhang, Chem. Commun., 2015, 51, 15574–15577.
18. H. T. Li, X. D. He, Z. H. Kang, H. Huang, Y. Liu, J. L. Liu, S. Y. Lian, C. H. A. Tsang, X. B. Yang and S. T. Lee, Angew. Chem., Int. Ed., 2010, 49, 4430–4434.
19. L. Cao, X. Wang, M. J. Meziani, F. S. Lu, H. F. Wang, P. J. G. Luo, Y. Lin, B. A. Harruff, L. M. Veca, D. Murray, S. Y. Xie and Y. P. Sun, J. Am. Chem. Soc., 2007, 129, 11318–11319.
20. J. Lu, J. X. Yang, J. Z. Wang, A. L. Lim, S. Wang and K. P. Loh, ACS Nano, 2009, 3, 2367–2375.
21. R. L. Liu, D. Q. Wu, S. H. Liu, K. Koynov, W. Knoll and Q. Li, Angew. Chem., Int. Ed., 2009, 48, 4598–4601.
22. K. Qu, J. Wang, J. Ren and X. Qu, Chem.–Eur. J., 2013, 19, 7243–7249.
23. D. Qu, M. Zheng, P. Du, Y. Zhou, L. Zhang, D. Li, H. Tan, Z. Zhao, Z. Xie and Z. Sun, Nanoscale, 2013, 5, 12272–12277.
24. S. Qu, X. Wang, Q. Lu, X. Liu and L. Wang, Angew. Chem., Int. Ed., 2012, 51, 12215–12218.
25. L. Bao, Z. L. Zhang, Z. Q. Tian, L. Zhang, C. Liu, Y. Lin, B. Qi and D. W. Pang, Adv. Mater., 2011, 23, 5801–5806.
26. M. Zheng, Z. Xie, D. Qu, D. Li, P. Du, X. Jing and Z. Sun, ACS Appl. Mater. Interfaces, 2013, 5, 13242–13247.
27. (a) L. Shang and S. Dong, Anal. Chem., 2009, 81, 1465–1470; (b) L. Xu, B. X. Li and Y. Jin, Talanta, 2011, 84, 558–564; (c) D. Zhang, Z. Dong, X. Jiang, M. Feng, W. Li and G. Gao, Anal. Methods, 2013, 5, 1669–1675.
28. Y. Kim, G. Jang and T. S. Lee, ACS Appl. Mater. Interfaces, 2015, 7, 15649–15657.
29. R. Wei, Z. Wei, L. Sun, J. Z. Zhang, J. Liu, X. Ge and L. Shi, ACS Appl. Mater. Interfaces, 2016, 8, 400–410.
30. Z. Zhang, Y. Shi, Y. Pan, X. Cheng, L. Zhang, J. Chen, M. J. Li and C. Yi, J. Mater. Chem. B, 2014, 2, 5020–5027.
31. (a) N. Dhenadhayalan and K. C. Lin, Sci. Rep.,  DOI:10.1038/srep10012; (b) A. Bhattacharya, S. Chatterjee, V. Khorwal and T. K. Mukherjee, Phys. Chem. Chem. Phys., 2016, 18, 5148–5158.
32. D. Wang, H. Xu, B. Zheng, Y. Li, M. Liu, J. Du and D. Xiao, Anal. Methods, 2015, 7, 5311–5317.
33. Y. Li, Y. Zhao, H. Cheng, Y. Hu, G. Shi, L. Dai and L. Qu, J. Am. Chem. Soc., 2012, 134, 15–18.
34. Q. Zhang, C. Zhang, Z. Li, J. Ge, C. Li, C. Dong and S. Shuang, RSC. Adv., 2015, 5, 95054–95060.
35. S. Huang, L. Wang, C. Huang, J. Xie, W. Su, J. Sheng and Q. Xiao, Sens. Actuators, B, 2015, 221, 1215–1222.
36. A. Barati, M. Shamsipur and H. Abdollahi, Biosens. Bioelectron., 2015, 71, 470–475.
37. J. R. Lakowicz, Principles of Fluorescence Spectroscopy, Springer, New York, 3rd edn, 2006, p. 282.
38. Y. Song, S. Zhu, S. Xiang, X. Zhao, J. Zhang, H. Zhang, Y. Fu and B. Yang, Nanoscale, 2014, 6, 4676–4682.
39. A. Puppo and B. Halliwell, Biochem. J., 1988, 249, 185–190.
40. J. M. C. Gutteridge, FEBS Lett., 1986, 201, 291–295.
41. E. Nagababu and J. M. Rifkind, Biochemistry, 2000, 39, 12503–12511.
42. J. P. Kehrer, Toxicology, 2000, 149, 43–50.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra14308c

---