Rapid, Acid-Free Synthesis of High-Quality Graphene Quantum Dots for Aggregation Induced Sensing of Metal Ions and Bioimaging.

Graphene quantum dots (GQDs) are zero-dimensional materials that exhibit characteristics of both graphene and quantum dots. Herein, we report a rapid, relatively green, one-pot synthesis of size-tunable GQDs from graphene oxide (GO) by a sonochemical method with intermittent microwave heating, keeping the reaction temperature constant at 90 °C. The GQDs were synthesized by oxidative cutting of GO using KMnO4 as an oxidizing agent within a short span of time (30 min) in an acid-free condition. The synthesized GQDs were of high quality and exhibited good quantum yield (23.8%), high product yield (>75%), and lower cytotoxicity (tested up to 1000 μg/mL). Furthermore, the as-synthesized GQDs were demonstrated as excellent fluorescent probes for bioimaging and label-free sensing of Fe(III) ions, with a detection limit as low as 10 × 10-6 M.

Introduction

Graphene, being a sheet of two-dimensional (2D) monolayer of sp2 bonded carbon atoms, does not show optical photoluminescence (PL) due to the absence of band gap, which limits its application in optical imaging. However, by converting this two-dimensional (2D) graphene sheet into zero-dimensional graphene quantum dots (GQDs), its band gap can be tuned due to quantum confinement and the edge effect, which has unlocked a large window of applications.[1] GQDs consist of an atomic layer of nanosized graphite that shows excellent properties of graphene, like large surface area, high carrier transport mobility, superior mechanical flexibility, and excellent thermal and chemical stability. When compared to widely explored typical semiconductor quantum dots (CdSe, CdTe, and PbTe), GQDs show superior properties, such as high photostability, aqueous dispersibility, biocompatibility, low cytotoxicity, low cost, etc.[2−5] The unique properties of GQDs find applications in bioimaging,[6−8] optical sensing,[9−13] photovoltaics,[14−16] light emitting diodes,[17] photocatalysis,[18] photodetectors,[19] and so forth. Irrespective of the emerging reports on GQDs over the past 5 years, their wide applicability is limited due to the usage of malicious corrosive chemicals and prolonged reaction time for synthesis.

GQDs have been synthesized from different carbon-based materials like fullerene,[20] graphene oxide (GO),[21−25] graphite rods,[26] carbon nanotube,[27] glucose,[1,28] carbon fibers,[6] coal,[29] etc. Several reports exist describing the synthesis of GQDs using various approaches, such as hydrothermal, solvothermal, sonochemical, corrosive chemical oxidation, high-resolution electron beam cutting, etc. Indeed, microwave-assisted synthetic methods offer several advantages when they are used in combination. With their extremely short reaction time, the utilization of the microwave-assisted method is quite remarkable for rapid, one-step synthesis of nanomaterials that have high energy of activation to achieve high product yield. Some of the limitations of recent reports on the various synthesis parameters, quantum yield, and product yield of GQDs are discussed here. Zhu et al. reported a one-step ultrasonic synthesis of GQDs with a relatively higher quantum yield of 27.8%, but the time period taken for synthesis was 4 h.[30] Li et al. reported microwave-assisted synthesis of greenish-yellow luminescent GQDs that involved use of strong acids and a quantum yield up to 11.7%.[31] In another report, Wang et al. prepared white-light-emitting GQDs by a two-step microwave-assisted hydrothermal method, where corrosive acids were used as the oxidizing agent with a reaction time of 14 h.[32] Shin et al. obtained GQDs with a relatively poor quantum yield of 9% synthesized from graphite using high-power microwave irradiation (600 W) in the presence of sulfuric acid.[33] In another work, Lin et al. prepared water-soluble GQDs with 9.9 wt % product yield from graphite flakes using a potassium-intercalation method.[34] Thus, a rapid, environmentally benign method for the synthesis of high-quality GQDs with better quantum yield, product yield, aqueous dispersibility, and low cytotoxicity is highly required for futuristic biomedical applications.

Herein, we report a facile, rapid, one-step acid-free synthetic route using the sonochemical method with intermittent microwave heating for the synthesis of high-quality GQDs with better properties. The GQDs were synthesized within a short span of time (30 min) and exhibited good quantum yield, high product yield, and lower cytotoxicity. Till date, not many reports exist where green synthesis of GQDs with cell viability were demonstrated for higher concentrations.[35−37] Typically, synthesis of GQDs from graphene oxide involves a tedious procedure that uses harsh chemicals (strong acids and organic solvents), elevated temperature, and prolonged reaction times. However, our synthetic approach employs KMnO4 for the oxidative cutting of GO in an aqueous medium with the simultaneous treatment of both ultrasound and microwave irradiations (Scheme ). Thereafter, the as-synthesized GQDs were well characterized for their structural and chemical properties. The cytotoxicity and amicability of GQDs as fluorescent probe in bioimaging applications were also explored. Further, the GQDs were used for the sensitive and label-free detection of metal ions (Fe(III) ions), with a detection limit as low as 10 μM. Thus, our study demonstrates the synthesized GQDs possess significant potential in metal-ion sensing and bioimaging applications.

Scheme 1

Schematic Illustration for the Synthesis of GQDs by the Oxidative Cleavage of GO Using KMnO4 by the Sonochemical Method with Intermittent Microwave Heating, Keeping the Reaction Temperature Constant

Results and Discussion

Synthesis and Characterization of GQDs

GQDs were synthesized by oxidative cutting of GO using KMnO4 as an oxidizing agent via a one-step sonochemical method, with intermittent microwave heating, in 30 min at different microwave powers, keeping the reaction temperature constant at 90 °C. Samples synthesized using different microwave powers of 100, 200, 300, and 400 W were labeled as GQD 1, GQD 2, GQD 3, and GQD 4, respectively. The product yield of GQD 1–4 was calculated and obtained in the range of 75–81%. The morphology and nanostructure of GQDs were characterized by transmission electron microscopy (TEM) and atomic force microscopy (AFM). Figure a–d showed the TEM images of as-synthesized GQDs with relatively uniform shape and size distribution. The corresponding histograms shown in the inset of Figure a–d revealed that GQD 1, GQD 2, GQD 3, and GQD 4 have an average lateral diameter of 5, 4, 3, and 2 nm, respectively. The high-resolution transmission electron microscopy (HR-TEM) images (Figure e–h) were also found to be in agreement with the corresponding histograms. The fast Fourier transform (FFT) patterns of GQD samples were presented as inset figures of corresponding HR-TEM images. From HR-TEM images of single GQD (Figure i–l), two types of lattice parameters, 0.210 and 0.242 nm, which correspond to the hexagonal lattice plane spacing of d1100 and d1120, respectively, were obtained. The AFM image (Figure S3) showed the typical topographic morphology of GQDs, and the section profile revealed an average height of 2.5 nm that corresponds to approximately 2–3 layers of graphene.

Figure 1

TEM images of (a) GQD 1, (b) GQD 2, (c) GQD 3, and (d) GQD 4 samples over large area (scale bar 20 nm) and the inset histograms show size distribution of corresponding GQDs. HR-TEM images of (e) GQD 1, (f) GQD 2, (g) GQD 3, and (h) GQD 4 samples (scale bar 5 nm) and corresponding FFT patterns of GQDs were shown as inset images. HR-TEM images of single GQD show the lattice fringes of (i) GQD 1, (j) GQD 2, (k) GQD 3, and (l) GQD 4 samples (scale bar 2 nm).

Further, the surface functionalities of GO and GQDs were studied using X-ray photoelectron spectroscopy (XPS) and Fourier transform infrared (FTIR) spectroscopy. From the FTIR spectra (Figure a), it could be inferred that all of the GQDs exhibited the characteristic peak of the carboxyl group at 1386 cm–1, carbonyl group at 1700 cm–1, and a broad absorption peak at 3431 cm–1 due to bending vibrations of O–H bonds. Peaks at 2920 and 2850 cm–1 are associated with the stretching vibrations of C–H bonds and the peak at 1615 cm–1 is due to C=C bonds of benzene ring vibrations. The peaks at 1258 and 1035 cm–1 are attributed to the vibrational absorption bands of C–O–C and C–O stretching vibrations in epoxides, respectively. It was observed that the presence of carboxylic, hydroxyl, and carbonyl groups render GQDs easily dispersible in water with good stability.

Figure 2

(a) FTIR spectra of GO and GQD samples, (b) table shows atom % compositions of C and O elements of GO and GQD 4, as obtained from XPS survey spectra, XPS high-resolution spectra of C 1s analysis of (c) GO and (d) GQD 4 samples. (e) Raman spectra of GO and GQD 4 samples. (f) XRD patterns of GO and GQD 4 samples.

Among the GQDs synthesized, GQD 4 was studied in detail due to its smaller size and higher quantum yield. XPS measurements were carried out to investigate the chemical compositions of GO and GQD 4 samples. The survey spectra (Figure S4) clearly showed C 1s and O 1s peaks at ∼284.6 and ∼533 eV, respectively, indicating mainly carbon and oxygen elements are present in GO and GQD 4 samples. Apart from this, insignificant amounts of N 1s and Na 1s peaks also appeared in the survey spectra of GO at ∼399 and ∼1069 eV, which may be originated from NaNO3 and NaOH used in the synthesis of GO. Further, elemental compositions were estimated from survey spectra (Figure b). A comparison of the high-resolution spectra of C 1s revealed an obvious change in the carbon chemical environments from GO to GQDs. It is known that GQDs are merited with excellent stability and hydrophilicity due to the abundant hydroxyl and carboxylic groups on their surface and edges. This was evident by C 1s spectra, which confirmed that the synthesized GQDs were decorated with hydroxyl, carbonyl, and carboxylic acid functionalities (Figure d). Moreover, when compared to C 1s high-resolution spectra of GO (Figure c), the composition of oxygen chemical environments in C 1s high-resolution spectra of the GQD sample was found to be increased, which indicates that strong oxidation has occurred.

Figure e shows the Raman spectra of GO and GQD samples. In the case of GQD 4, the D band (1360 cm–1) and G band (1580 cm–1) were recorded with the intensity ratio (ID/IG) of 0.90. The integrated intensity ratio of the disorder D band to the crystalline G band (ID/IG) for GO was 0.78, which increased to 0.90 after the oxidative cutting to form GQDs. The increased disorder may be due to the introduction of defects to the graphene basal planes and edges. Additionally, the –1 in Raman spectra indicate the formation of high-quality GQDs.[38] The typical XRD patterns of GO sheets and GQD 4 are shown in Figure f. The XRD pattern of GO showed a strong diffraction peak at 2θ of 10.4° (002), whereas the GQD 4 sample showed a broader (002) peak centered at 21.4°. Further, GO showed a d-spacing value of 0.849 nm due to the introduction of functional groups. The d-spacing of GQD 4 was calculated to be 0.409 nm, which is smaller than that of GO, and this is because GQDs are oxidized only at the edges due to their very small size.

The optical properties of fluorescent GQDs were studied using UV–vis and PL spectra. Figure a shows UV–vis spectra of GQD samples of different sizes synthesized namely, GQD 1–4. A distinct broad absorption peak at 350 nm and a strong absorption below 300 nm were observed from the spectra, which are attributed to the n−π* transition and π–π* transition of aromatic sp2 domains, respectively. Additionally, a slight shift toward the blue region from 360 to 330 nm was observed as the size of GQDs decreased. This result indicates that the size of GQDs can significantly affect the absorption properties of GQDs. The inset of Figure a shows the photographs of GQDs of different sizes irradiated under 365 nm UV light and GQD 4 sample under day light.

Figure 3

(a) UV–vis absorption spectra of GQD 1–4 samples. The inset of panel (a) shows photographs of the corresponding GQDs irradiated under 365 nm UV light and GQD 4 under day light (b) PL spectra of GQD samples excited at 350 nm. (c) PL spectra of the GQD 4 sample at different excitation wavelengths from 350 to 420 nm.

Figure b shows the corresponding PL spectra of the as-synthesized GQD samples. The different emission color may indicate the size-dependent nature of GQDs. The emission maxima of GQD 1, GQD 2, GQD 3, and GQD 4 solutions were at 510, 480, 460, and 430 nm, respectively. In addition, GQD samples displayed excitation-dependent emission and when the excitation wavelength is varied from 340 to 420 nm, a redshift was observed with a remarkable decrease in PL intensity, as shown in Figure c. GQDs usually have a quantum confinement effect, i.e., they have a size-dependent effect on their PL properties. Here, the quantum confinement effect was confirmed by the PL spectra (Figure b) and HR-TEM images (Figure e–h), i.e., smaller sized GQD samples lead to blueshift in the emission. Further, the quantum yield was estimated using quinine sulfate in 0.05 M sulfuric acid solution as the reference (Table ). It is remarkable that the maximum quantum yield of 23.8% was obtained for GQD 4 sample.

Table 1

Estimated Quantum Yield of GQD Samples

sample	peak area	Abs at 350 nm wavelength	refractive index of the solvent	quantum yield
quinine sulfate	2.23 × 109	0.1	1.33	0.546
GQD 1	3.74 × 108	0.1	1.33	0.092
GQD 2	4.67 × 108	0.1	1.33	0.113
GQD 3	7.05 × 108	0.1	1.33	0.172
GQD 4	8.95 × 108	0.1	1.33	0.238

Sensing of Fe3+ Ions

GQDs could be promising nanomaterials as fluorescent sensors due to their small size, unique optical properties, and high photostability. In the present study, we explored the as-synthesized GQD 4 sample for sensitive, selective, and label-free detection of Fe3+ metal ions. It is well known that Fe3+ ions play an important role in the biochemical process in living systems by complexation with various regulatory proteins. On the other hand, excess Fe3+ ions can lead to overproduction of free radicals and hence cytotoxicity. The high Fe3+ concentration in neurons is also a key marker for Parkinson’s disease.[32] Therefore, the detection of Fe3+ ions in biological systems as well as their environmental monitoring is important. To study the selectivity of a synthesized fluorescent probe, the effect of various metal ions (Fe3+, Na+, K+, Co2+, Mn2+, Zn2+, Ca2+, Ba2+, Mg2+, Pb2+, Ni2+, etc.) of same concentration (100 μM) on the fluorescent intensity of the GQD 4 sample (0.05 mg/mL) was studied by recording their PL spectra upon the excitation wavelength of 350 nm. As shown in Figure a, there is an apparent quenching of fluorescence intensity in the presence of Fe3+ when compared with other metal ions. From previous reports, it is clear that the hydroxyl groups showed a good binding affinity toward Fe3+ ions. The fluorescence quenching in the case of Fe3+ ion may be due to the complexation of Fe3+ ions and phenolic hydroxyl groups of GQDs, which resulted in the electron transfer to d-orbital of Fe3+ ions.[39,40] Thus, fluorescence quenching of GQDs are highly sensitive and selective to Fe3+ ions, whereas other metal ions are not able to exert significant quenching and hence GQDs can be used for the selective sensing of Fe3+ ions.

Figure 4

(a) Effect of different metal ions (concentration, 100 μM) on the fluorescent intensity of GQDs (0.05 mg/mL), (b) photographs of GQDs with different concentrations of Fe3+ ions (0, 20, 600, and 800 μM) under 365 nm UV light (c) PL emission spectra of GQDs in the presence of varying concentrations of Fe3+ ions (0–800 μM), and (d) linear regression plot performed in the low concentration range of 10–120 μM.

Figure b shows the photographs of GQDs in the presence of different concentrations of Fe3+ ions (0, 20, 600, and 800 μM) under 365 nm UV light irradiation. From Figure c, it could be observed that fluorescence quenching of Fe3+ ion is concentration dependent and therefore experiments were carried out at various concentrations of Fe3+ ions (0–800 μM). It was observed that the fluorescence intensity of GQDs gradually decreases upon increasing Fe3+-ion concentration. The quenching efficiency showed a good linear relationship with the linear regression value, R2, of 0.9926 for a concentration of a range of 10–120 μM of Fe3+ ions (Figure d). Thus, GQDs were used for sensitive and label-free detection of metal ions [Fe(III) ions], with a detection limit as low as 10 μM.

To investigate the fluorescence response between GQDs and Fe3+, TEM imaging (Figure a,b) was done where the morphology change is observed, which shows that size of the GQD–Fe3+ complex increases when compared to that of GQD 4 (Figure d). This provides convincing evidence for the aggregation of GQDs in the presence of Fe3+. Also, the FTIR spectrum of the GQD–Fe3+ complex shows redshift and weakened absorbance of characteristic peaks when compared to that of GQD (Figure S5). This indicates that there is a change in the chemical behavior of surface functionalities of GQDs due to the strong affinity between Fe3+ and phenolic hydroxyl groups of GQDs. Furthermore, the fluorescence lifetime of GQD and the GQD–Fe3+ complex was studied using time-correlated single photon counting (TCSPC). The fluorescence lifetime of GQDs is 2.57 ns and has three components: 2.5 ns (ca. 46%), 0.6 ns (ca. 12%), and 7.4 ns (ca. 42%). On addition of Fe3+ ions, the average decay time of the GQD–Fe3+ complex is decreased to 0.037 ns, which confirms the formation of the GQD–Fe3+ complex (Figure c). Thus, the changes observed in morphology, fluorescence lifetime, and FTIR spectra confirm the formation of the aggregate complex between GQDs and Fe3+ ions, resulting in the fluorescence quenching of GQDs.

Figure 5

(a) TEM image of GQD–Fe3+ (scale bar 50 nm) (b) HR-TEM image of GQD–Fe3+ (scale bar 10 nm) (c) time-correlated single photon counting (TCSPC) spectra of GQDs (red) and GQD–Fe3+ (blue) (330 nm excitation and delay time at 427 nm emission).

Cytotoxicity Assay of GQDs

The biocompatibility of the synthesized GQDs were assessed against human cervical cancer cell line HeLa cells using the [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] (MTT) assay. GQD 4 samples were tested in this study because of their smaller size and high quantum yield. HeLa cells were incubated at six different concentrations of GQD 4 (25, 50, 100, 250, 500, and 1000 μg/mL) for 24 h. The potential toxicity of the nanoparticles is determined by a colorimetric technique based on the ability of live cells to reduce yellow MTT to purple formazan crystals.[41]Figure a shows that the GQD sample has low cytotoxicity, with a cell viability ≥90% for concentrations ranging from 25 to 1000 μg/mL. It is noteworthy that GQDs even at a relatively higher concentration of 1000 μg/mL have not showed cytotoxicity with a cell viability of >92%. Moreover, the morphology of the cells incubated with GQDs appeared normal as that of the untreated control cells. Thus, the MTT assay as well as visual observation of the cells ascertained the excellent biocompatibility for GQDs. Thus, GQDs could be used as efficient biocompatible nanoprobes for bioimaging as well as for theranostic applications. Considering the excellent biocompatibility and photostability of the GQDs, they were further utilized as fluorescent nanoprobes for bioimaging applications. The concentration at which the maximum viability was observed was taken as the optimum concentration for cellular uptake studies.

Figure 6

(a) Cytotoxicity analysis of various concentrations of GQD 4 (25–1000 μg/mL) on HeLa cells incubated for a 24 h MTT assay, (2) confocal fluorescence images of HeLa cells incubated for 4 h in the presence and absence of GQD 4, with an excitation of 350 nm, (b) and (c) showed the fluorescent image and bright field image of the HeLa cells with GQD 4 (25 μg/mL) and (d) and (e) showed the fluorescent and bright field images of the HeLa cells without GQDs (scale bar 20 μm).

In Vitro Bioimaging Studies

One of the major challenges in cellular imaging using semiconductor quantum dots is its intrinsic toxicity that has limited its application. The GQDs synthesized in this work showed several advantages over those of existing ones in terms of ease of size tunability, green synthetic approach, surface functionality, physiological stability, photostability, size-tuned emission, and excellent biocompatibility, making it suitable for bioimaging applications. The in vitro cellular imaging studies of the GQD 4 sample were carried on HeLa cells. The optimum concentration chosen for cellular uptake studies was 25 μg/mL. The GQD sample was incubated with HeLa cells for a period of 4 h. This was visualized by the intrinsic fluorescence property of the GQDs when excited using the A360/10 excitation filter. The blue fluorescence observed was uniformly distributed throughout the cell. The transmitted light images of the cells and their corresponding fluorescence images have been shown in Figure b,c. The bright field images of the cells incubated with GQDs showed normal morphology of HeLa cells (Figure c), confirming the biocompatibility of GQDs. The cells incubated with GQDs exhibited bright blue fluorescence, and this was clearly observed in the fluorescent image when excited at a wavelength of 350 nm (Figure b). Figure d,e shows the fluorescent and bright field images of the HeLa cells without GQDs. It could be noted that there is no fluorescence in control cells, confirming that the emission is from the GQDs in Figure b. The GQDs were found to be cell membrane-permeable and found to generate fluorescence inside the cell. It could be observed that GQDs owing to their size, biocompatibility, and surface functionality can be easily internalized through the cell membrane. It is well documented that carboxylated GQDs exhibit good biocompatibility and great potential for in vitro as well as in vivo bioimaging applications.[42] Thus, these GQDs serve as a suitable nanoprobe that has immense potential in live-cell imaging as well as biomedical applications.

Conclusions

In summary, we reported a rapid, acid-free, one-step sonochemical strategy with intermittent microwave heating for the synthesis of size-tunable high-quality GQDs by the oxidative cutting of GO using KMnO4 in half an hour. The synthesized GQDs are highly crystalline and have uniform size distribution. These GQDs exhibited good quantum yield up to 23.8%, high product yield (75–81%), and stable size-dependent photoluminescence in aqueous solutions. Our studies indicate that the GQDs possess very good biocompatibility, good aqueous dispersibility, and high photostability and hence can be applied as excellent fluorescent probes for live-cell imaging. Fluorogenic probes are of particular interest in the area of cancer detection or bioimaging. Our findings on the biocompatibility and fluorogenic property of the GQDs observed in cancer cells highlight their significance in clinical diagnostic applications. The GQDs synthesized were also demonstrated for the detection of physiologically relevant metal ions and particularly for the sensing of Fe3+ ions in aqueous media.

Experimental Section

Materials

Graphite (<150 μm), sodium nitrate (NaNO3), hydrogen peroxide (H2O2), quinine sulfate, and sulfuric acid (H2SO4) were purchased from Sigma-Aldrich. Potassium permanganate (KMnO4) was purchased from SD Fine-Chemicals Limited. Ultrapure deionized water (18.2 MΩ cm, 25 °C, Milli-Q D3, Merck, Germany) was used in all of the experiments.

Synthesis of Graphene Oxide

Graphene oxide (GO) was synthesized by the Hummers method (Figure S1). Two grams of natural graphite was added to a cold (0 °C) concentrated solution of H2SO4 and NaNO3 (4 g) in a 500 mL flask with vigorous stirring by keeping the temperature of the mixture below 10 °C. The reaction mixture was stirred at 35 °C for 2 h until it becomes pasty brown and was then diluted with 100 mL deionized water in an ice bath. Again, it is stirred for 30 min and H2O2 (20 mL, 30 wt %) was added slowly to the mixture to reduce the residual KMnO4, after which the color of the mixture changes to brilliant yellow. The sample was then filtered and washed with 800 mL of 5% HCl to remove the metal ions, followed by the addition of excess of deionized water to remove the acid. Later, the solution was centrifuged, dried at 60 °C, and used for further characterizations.

Synthesis of GQDs

GQDs were prepared from GO using KMnO4 as an oxidizing agent in 30 min by the sonochemical method with intermittent microwave heating, keeping the reaction temperature constant at 90 °C. Briefly, 1 mg/mL GO and 1 M KMnO4 were mixed in a 1:1 ratio in a RB flask to form a homogeneous mixture. This mixture was treated under microwave irradiation along with ultrasonication in a microwave reactor (SienoUWave 1000) at 90 °C, operating at different microwave powers of 100, 200, 300, and 400 W for 30 min. After this treatment, the mixture was centrifuged at 3000 rpm for 10 min to remove the unreacted GO. The supernatant containing GQD was collected after the centrifugation. The supernatant solution was then filtered through a 0.45 μm poly(tetrafluoroethylene) membrane, and the filtrate was dialyzed in a 1000 Da dialysis bag. After purification, the solvent was evaporated to obtain solid GQDs. TEM images of GQDs synthesized using the sonochemical method without microwave heating showed the formation of GQDs with large size (Figure S2).

Characterizations

Scanning electron microscope micrographs were acquired using an EVO 18 Special Edition scanning electron microscope (Carl Zeiss, Germany) operated at 20 kV acceleration voltage. The size and morphology of GQDs were observed through a transmission electron microscope (TEM) operated at an accelerating voltage of 300 kV. High-resolution TEM (HR-TEM) images and energy-dispersive X-ray analysis were performed on an FEI Tecnai 30 G2S-TWIN transmission electron microscope. The HR-TEM images were further analyzed with Gatan Digital Micrograph software. Atomic force microscopy (AFM) imaging was performed using a MultiMode 8 AFM equipped with NanoScope V controller (Bruker, Santa Barbara, CA) in air at ambient temperature (22 ± 2 °C). Si cantilevers (NSG 01, NT-MDT) with a typical radius of curvature of approximately 10 nm were used. The force constants of the AFM probe were in the range of 2.5–10 N/m and with a resonance frequency in the range of 120–180 kHz. The scan rate used was 1 Hz. Raw data were processed offline using Bruker’s NanoScope Analysis software. The FTIR spectra were measured with a Perkin Elmer Series Spectrum Two FTIR spectrometer over the wavenumber range 4000–500 cm–1. The sample was directly mixed and pelletized with KBr. Wide-angle X-ray scattering measurements were performed with the XEUSS SAXS/WAXS system using the Genisxmicro source from Xenocs operated at 50 kV and 0.6 mA. The Cu Kα radiation (λ = 1.54 Å) was collimated with an FOX2D mirror and two pairs of scattering less slits from Xenocs. The 2D patterns were recorded on a Mar345 image plate and processed using Fit2D software. The UV–visible (UV–vis) absorption spectra of the GQDs were obtained using a spectrophotometer (SHIMADZU UV-2401 PC; Shimadzu, Japan), employing a 1 cm path length quartz cell at room temperature. The photoluminescence (PL) spectra of GQDs were recorded on a Spex-Fluorolog FL22 spectrofluorimeter equipped with a double grating 0.22 m Spex 1680 monochromator and a 450 W Xe lamp as the excitation source. Raman spectra were obtained using the WITec Raman microscope (Witec Inc. Germany, alpha 300R), with a laser beam directed to the sample through a 60× water immersion objective and a Peltier cooled CCD detector. Samples were excited with a 632.8 nm excitation wavelength laser, and Stokes-shifted Raman spectra were collected in the range of 0–3000 cm–1 with 1 cm–1 resolution. X-ray photoelectron spectroscopy (XPS) was performed using a Kratos Axis UltraDLD instrument equipped with a monochromated Al Kα X-ray source (hν = 1486.6 eV) operating at 15 kV and 10 mA (150 W). The survey scans were obtained over the range 0–1200 eV binding energy (BE) at pass energy of 160 eV and used to estimate the surface elemental composition. High-resolution spectra were recorded for C 1s with a pass energy of 20 eV. The Kratos charge neutralizer system was used on all samples. Sample charging effects on the measured BE positions were corrected by setting the lowest BE component of the C 1s spectral envelope to 284.7 eV, corresponding to the C–C/C–H species. The deconvolution of peaks was performed by using the CasaXPS software. Quantum yield was measured using quinine sulfate in 0.05 M sulfuric acid solution as a standard. Fluorescence lifetime experiments were conducted using a Delta Flex modular time-correlated single photon counting (TCSPC) spectrometer system, employing the 330 nm nanoLED as the excitation source and PPD 850 detector. Decay in the fluorescence intensity (I) with time (t) was fitted by three exponential functions. The quality of the fits was checked by examining the χ2 value. All experiments were run in triplicate, with similar results obtained.

Sensing of Fe3+ Ions

The sensitive detection of Fe3+ ions was done in aqueous medium at room temperature. To study the sensitivity toward Fe3+, fluorescence emission spectrum of GQDs (0.05 mg/mL) was recorded upon the excitation wavelength of 350 nm. Then, a series of different concentrations of Fe3+ (0–800 μM) was freshly prepared and added into the aqueous solution containing the same amount of GQDs (0.05 mg/mL) and the corresponding fluorescence spectra were recorded under the same excitation wavelength. The selectivity of Fe3+ sensing was evaluated by adding other common metal ions (Na+, K+, Co2+, Mn2+, Zn2+, Ca2+, Ba2+, Mg2+, Pb2+, and Ni2+) of same concentration (100 μM) to GQD solution (0.05 mg/mL), and the PL emission spectra were recorded under identical conditions.

Maintenance of Cell Lines

The cell line used in the present study is human cervical cancer cell line, HeLa cells, obtained from National Centre for Cell Science, Pune, India. For maintenance of cell lines, Dulbecco’s modified Eagle’s medium (Sigma-Aldrich) containing 10% fetal bovine serum (Gibco), antibiotics (100 U/mL penicillin and 100 μg/mL streptomycin), and amphotericin (0.25 μg/mL) (HiMedia) were employed. The cells were maintained in CO2 incubators at 37 °C with 5% CO2 in air and 99% humidity. The passaging of cells when confluent was carried out using 0.25% trypsin and 0.02% ethylenediaminetetraacetic acid (HiMedia) in phosphate-buffered saline. Experiments were carried out after 36 h of seeding the cells at appropriate density in suitable well plates.

Assessment of Cell Viability

The cell viability after incubating HeLa cells with GQDs was determined by the [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] (MTT) assay. It is a colorimetric assay based on the ability of live but not dead cells to reduce the tetrazolium component of MTT into purple formazan crystals. The cells were spread in 96-well plates at 5 × 103 cells/well. After 36 h of seeding, the cells were incubated with different concentrations of GQDs (25, 50, 100, 250, 500, and 1000 μg/mL) for 24 h. Subsequently, the cells were exposed to MTT at a concentration of 50 μg/well for 2.5–3 h at 37 °C in CO2 incubator. The working solution of MTT was prepared in Hanks balanced salt solution (HBSS). After viewing formazan crystals under the microscope, crystals were solubilized by treating the cells with dimethyl sulfoxide/isopropanol solvent mixture at a ratio of 1:1 for 20 min, at 37 °C. The percentage of cell viability was determined by recording the optical absorbance at 570 nm using a microplate reader (Synergy-4 Multimode reader, Biotek, Winooski, VT) relative to the nontreated cells. The cell viability was calculated using the following equationwhere IntGQDs is the optical density (OD) value of the cells incubated with different concentrations of GQDs and Intcontrol is the OD value of the cells incubated with the culture medium alone. Cell viability of control cells were kept as 100%.

Cellular Uptake of GQDs and in Vitro Bioimaging

The cellular uptake studies of the GQDs were executed by fluorescence imaging of HeLa cells. The cells were seeded at a density of 5 × 103 cells/well of 96-well black plates (BD Biosciences) for the purpose. After 36 h of seeding, the cells were incubated with GQDs (25 μg/mL) in HBSS for 4 h. Twenty five micrograms per milliliter was chosen for cellular uptake studies as at this concentration maximum cell viability was observed. Subsequently, cells were washed thrice with HBSS to remove the unbound particles. Images of the cells were collected by a high-content spinning disk facility (BD Pathway 855; BD Biosciences) using AttoVision 1.5.3 software. The images were taken at 40× magnification, using A360/10 excitation filter and 435 LP emission filter.