Fluorescent Carbon Dots Derived from Vehicle Exhaust Soot and Sensing of Tartrazine in Soft Drinks.

Recycling of waste into valuable products plays a significant role in sustainable development. Herein, we report the conversion of vehicle exhaust waste soot into water-soluble fluorescent carbon dots via a simple acid refluxion method. The obtained carbon dots were characterized using microscopic and spectroscopic techniques. Microscopic techniques reveal that the prepared carbon material is spherical in shape with an average particle size of ∼4 nm. Spectroscopic studies exhibited that the carbon dots are emissive in nature, and the emission is excitation-dependent. Further, the prepared carbon dots were successfully utilized as a fluorescent probe for the detection of tartrazine with a limit of detection of 26 nM. The sensitivity of carbon dots has also been realized by the detection of trace amounts of tartrazine in commercial soft drinks. Overall, this work demonstrates the conversion air pollutant soot into value-added fluorescent nanomaterials toward sensing applications.

Introduction

Conversion of waste into a value-added material is a trend in research; owing to this leads sustainable development via recycling of used and unused scraps, especially soot waste.[1] Industries and vehicles exhaust pollutants, such as carbon monoxide, nitrogen oxide, particulate matter, ammonia, and sulfur dioxide, which impose many adverse effects on a living being.[2] In the present decade, we cannot completely avoid the usage of fuel in vehicles owing to the various hurdles on the 100% utilization of renewable energy resources. Among the various air pollutants, a vehicle exhaust soot emission consists of higher percentage of carbon with other elements like N and S from the source of fuel.[3] On the other side, researchers are looking for an invention of a novel, functionalized, and biocompatible nanomaterial for a wide range of applications.[4] The growing research areas such as bioimaging, biosensing, light-emitting diodes, etc. are consistently searching of a nanomaterial, which fits for the mentioned fields; especially in the past one decade, researchers are attracted by carbon-based nanomaterials owing to their specific properties of water solubility, fluorescence nature, photo and storage stability, biocompatibility, etc.[5−7] These urge the focused research on improving the efficiency of a fuel and conversion of pollutants into value-added products to maintain a sustainable environment.

Carbon dots (CDs) are widely used as an alternative for organic fluorophores and heavy metal-containing semiconductor quantum dots in biomedical, photovoltaic, and sensing applications as well as anti-icing/bioadhesor.[8−11] CDs are a new type of nanomaterial that consists of a graphitic core and an amorphous shell with an average particle size of less than 10 nm, which shows good water solubility, low toxicity, fluorescence property, and photostability.[12] It can be generated by various techniques such as electrochemical deposition,[13] microwave,[14] hydrothermal and solvothermal methods,[15] pyrolysis,[16] etc. Their surface is viable for modification by carbonyl, carboxylic, and amine functional groups, significantly making them as recognizable by enzymes in the organism. However, most of the reported CDs were derived from carbon-rich organic molecules like glucose, cellulose, and phenolic compounds[17−19] and biomass such as algal blooms,[20] coriander leaves,[21] lotus root,[22] radish,[23] ginger,[24] grass,[25] coffee bean shells,[26] chitosan[27] etc. Conversion of pollutants into value-added products like CDs using ultrasonication of food waste has been reported by Lee et al.[28] Similarly, conversion of soot collected from a burning candle,[29] tire soot,[30] and natural gas[31] into CDs and their biological applications have been reported.[32,33] Recently, we have also reported the fuel waste into fluorescent CDs for sensing applications.[34]

Food additives are used to improve the taste, quality, appearance, and other commercial requirements. Few types of food additives[35] are coloring agents (tartrazine and sunset yellow), flavor enhancers (monosodium glutamate commercially known as Ajinomoto), artificial sweetener (aspartame and acesulfame K), etc., which are commonly used in food and beverages. However, most human beings are allergic to such food additives, which cause many adverse health issues like hives or diarrhea, asthma, rhinitis and sinusitis, itching, rashes, and swelling.[36,37] For instance, a food colorant like tartrazine is an azo group dye molecule, whose limit of existence in the food content is fixed by the World Health Organization and the Food and Agricultural Organization, is 50 mg/kg. However, in most of the commercial food beverages as well as in the domestic preparations, the level of the coloring agent is not measured according to the WHO report, which may cause[38] few of the abovementioned negative effects. Hence, the detection of tartrazine has gained significance, and a few related reports could be found in the literature.[39] Aloe-derived carbon dots were used to detect tartrazine in a food sample.[40] Yuan et al. detected a well-known food colorant, sunset yellow, from the analysis of soft drinks with a satisfactory result by means of fluorescent CDs.[41] Even though a number of methods and materials are available for the detection of food additives, the high cost, tedious procedures, difficulty in synthesis, etc. motivate us to bring a simple, sensitive, and cost-effective material for the same matter.

In this work, the conversion of pollutant (exhaust soot derived from a diesel-fueled lorry) into a value-added product (CDs) is reported via the facile one-step acid refluxion method. The fluorescence nature of the CDs has been effectively used as a probe for the detection of tartrazine. The detection was extended in commercially available soft drinks.

Results and Discussion

Synthesis of Carbon Dots

Carbon dots were synthesized from a vehicle exhaust soot by a one-pot acid refluxion method using nitric acid.[42] In brief, the soot was collected from a diesel engine lorry exhaust pipeline. The collected soot (1 g) was mixed with 200 mL of 5 M nitric acid, and the mixture was refluxed at 100 °C for 12 h. Consequently, the reflexed mixture was kept aside without disturbance to cool down to room temperature naturally. It is observed that the reaction mixture was turned to a reddish brown color with black precipitate deposition at the bottom. Furthermore, the solution mixture was centrifuged at 5000 rpm for 30 min to separate the bulkier carbon particles and unreacted soot material, which yields clear reddish brown color carbon dots. The brown fluorescent supernatant was collected, neutralized with sodium carbonate, and filtered via a 0.22 μm membrane filter (cellulose) to remove more than micrometer-sized particles and unreacted materials. Following that, the solution was dialyzed using a dialysis membrane bag (1000 MWCO) for 32 h. This process is essential to remove the less than nanometer-sized particles, small molecules, and removal of excess salt to derive the pure carbon dots (dialysis water is changed every 6 h). The clear dark brown solution was collected, and the volume was reduced via vacuum distillation and dried in a vacuum oven. Finally, dried fluorescent carbon dots were collected (55% product yield) and stored at room temperature.

Surface Characterization

The SEM images of collected soot and the derived CDs are given in Figure a,b, respectively, and the elemental composition details with the EDAX spectrum is shown in Figure S1. The SEM analysis clearly illustrates that the waste soot particles are rich in carbon (85.45%) and oxygen (14.54%), and the CDs derived from the soot after acid refluxion consist of aggregates of carbon nanoparticles, having nearly spherical morphology with relatively regular size and homogeneity. The EDAX data analysis exposes that CDs mainly consists of C (26.9%) and O (42.7%), which demonstrates that CDs are made up of an oxygenous carbon structure. Other elements also present in trace amounts may originate from the precursor soot. The powder XRD pattern of carbon dots is shown in Figure , which exhibits a sharp diffraction peak at 2θ = 29° and also many small diffraction peaks at 2θ = 32, 35, 38, 42, and 48°. It indicates that the carbon atoms are arranged in a considerably unsystematic manner as well as an amorphous nature induced by more oxygen-containing groups on the surface of CDs. Figure a shows the TEM image and histogram of CDs, and it reveals that the particles are quasi-spherical in shape with a nanometer size. The average particle size of CDs is in the range of 3–4 nm, and the size distribution is displayed in the histogram plot (Figure b). Figure shows the AFM image of the derived CDs, and it exposes that most of the particles are spherical-shaped and well dispersed. The height profile (Figure S2) indicates the maximum of 6 nm of size from the surface, and the surface looks quite smooth, which confirms the uniform distribution of the particles on the surface. The particles size derived from TEM and AFM are consistent with each other, and both characterizations confirm that the prepared CDs are nanometer-sized particles.

Figure 1

HR-SEM images of (a) soot and (b) CDs.

Figure 2

Powder XRD pattern of CDs.

Figure 3

(a) HR-TEM image (50 nm scale) and (b) histogram plot of CDs.

Figure 4

Atomic force microscopy image of CDs in a 10 μm scale (right-hand side) and the same in a three-dimensional view (left-hand side).

The FT-IR analysis of the prepared CDs is shown in Figure a, which portrays that the presence of peak positions at 1795 and 3410 cm–1 were associated with the stretching vibrations of carbonyl (C=O) and hydroxyl (−OH) groups, respectively. The characteristic peaks at 1330 and 829 cm–1 were due to the stretching vibrations of sp2 and sp3 C–H, and the peak located at 1593 cm–1 corresponds to the C=C stretching vibration. The 2437 cm–1 peak is assigned for NO3– stretching vibration. The FT-IR data revealed the formation of unsaturated carbon during the carbonization process and the presence of rich oxygen-containing groups (carboxyl, carbonyl, and hydroxyl) on the surface of CDs. Raman analysis (Figure S3) exposed that the appearance of D and G bands of the prepared CDs were at 1351 and 1552 cm–1, respectively. The existence of D and G bands illustrate the confirmation of sp3-hybridized amorphous carbon particles and the ε2g mode of sp2-hybridized graphitic carbon atoms, respectively. Further, the ratio of ID/IG of 1.01 clearly represents the prepared CDs consisting of a nanocrystalline graphitic core surrounded by a disorder/amorphous shell. The proton NMR analysis of CDs is shown in Figure S4, which reveals that the sp3 C–H proton is present at the region 1–2 ppm, carbonyl, hydroxyl, and ether protons are present at 3–4 pm, and the aromatic sp2 and aldehyde proton is an occurrence in the region 7–9 ppm. The solid-state 13C NMR spectrum is shown in Figure S5, and the signal at 190 ppm was assigned to carboxyl carbon (COO−), while the signal at 125 ppm was designated to aromatic carbons. Further, the signal around 97 ppm is for carbonyl carbons (C–O), and the signal in the range of 6–50 ppm was assigned to carbon bonded with hydrogen or other carbons. Zeta potential measurement exposed that CDs are negatively charged with a value of −0.2 mV, which specifies the presence of a negatively charged functional group on the surface of CDs and the surface is the hydrophilic nature. This imparts the high solubility of CDs in water. Further, the nature of bonding on the surface of CDs is characterized by the X-ray photoelectron spectroscopy technique and is shown in Figure S6a. The survey spectrum represented the presence of C, O, N, and Na (285, 530, 407, and 497 eV) elements. Nitrogen and sodium ions originated from the oxidative acid treatment and neutralization process. Hence, our interest is to analyze the carbon and oxygen bonding nature on the surface of CDs. So, the deconvoluted spectra of C 1s and O 1s were thoroughly examined and presented in Figure S6b,c. In the C 1s deconvoluted spectrum, the derivative peaks at 282.9, 284.5, and 287.9 eV were assigned to −C–C–, −C=C/C–N, and C=O, respectively. Similarly, the peaks located in the O 1s deconvoluted spectrum at 531.3, 533.3, and 536.1 eV were designated to the bonding of −C=O, −C–OH, and −C–O–C–, respectively. From the results of NMR, FT-IR, and XPS measurements, we have visualized the structure of the prepared material (CDs), which consists of various functional moieties and bonding nature on its surface.

Figure 5

FT-IR spectrum of CDs.

Photophysical Characterization and Fluorescence Stability Studies

The UV–visible absorption spectrum of CDs is shown in Figure a, and it exhibits a sharp peak at 295 nm, which corresponds to π–π* transitions in the aromatic C=C bond in the carbon core. A broad shoulder in the region of 330–450 nm is due to n–π* transitions in the C=O bond, which originates from the surface functional groups of CDs. As similar to the reported carbon dots, the prepared CDs also exhibit the excitation-dependent emissive property (Figure b). The emission maximum is centered at 425 nm at an excitation of 360 nm. The origin of the fluorescence property of the derived CDs may arise from the various defects present on the surface of CDs.[34] Further, the fluorescence decay of carbon dots shown in Figure c is fitted by a multiexponential function [F(t) = A1 exp(−t/τ1) + A2 exp(−t/τ2) + A3 exp(−t/τ3)] with the lifetimes of τ1 = 0.56 ns (18%), τ2 = 1.78 ns (35%), and τ3 = 5.21 ns (47%), and the average lifetime is calculated as τavg = 3.17 ns. The multiple lifetimes of CDs are due to a wide range of chemical environments on the surface of CDs.[43] The quantum yield (φ) of CDs is determined using quinine sulfate as a reference.[44] For the measurement of φ, the optical density of CDs in water (η = 1.33) was fixed to 0.1 at a wavelength of 366 nm. Fluorescence quantum yield (ϕF) for CDs was calculated by using eq where the subscript “S” refers to the samples, the subscript “R” refers to quinine sulfate, A is the absorbance at the excitation wavelength, I is the integrated emission area, and η is the solvent refraction index. The quantum yield for CDs is found to be 3%, which is good enough for various sensing applications.

Figure 6

(a) UV–visible absorption, (b) corrected emission spectra at different excitation wavelengths, and (c) time-resolved fluorescence decay (λexi = 360 nm) of CDs.

Further, it is necessary to analyze fluorescence stability of CDs for versatile applications. The stability is investigated under four different parameters such as storage time, pH, light irradiation, and ionic strength. The detailed results are depicted in Figure S7, which explores that the fluorescence property of CDs remains stable for up to 90 days. It shows a higher fluorescence nature at the neutral pH, and there was a pH-dependent fluorescence behavior. Optimum fluorescence behavior in the pH range of 4–9 and lower of higher of that leads to diminishing in fluorescence ability. Photostability of CDs is checked by UV light irradiation for a period of 2 h, and the quite stable nature of CDs is found. The effect of ionic strength is checked with addition of maximum of 1 M NaCl in CDs, and it is observed that there is not much variation in the fluorescence property in the presence of NaCl. So, the prepared CDs exhibit stability in different environments, which may extend its utilization in biological fields.

Sensing of Tartrazine

Tartrazine is one of the most widely used food coloring agents, and its excess limit in food may cause allergic symptoms to human beings. So, it is essential to detect tartrazine in food samples. For this purpose, we have adopted a highly sensitive and simple analytical methodology, namely, the fluorescence quenching technique. The fluorescence quenching measurements of CDs in the presence of various concentrations of tartrazine (0.1–6 μM) were conducted, and the spectrum is displayed in Figure a. Upon increasing the concentration of tartrazine, the fluorescence intensity of CDs is found to regularly decrease. The sensitivity of CDs is estimated from the slope of the plot (Figure b) between the relative change (ΔF = F0 – F) in fluorescence intensity and tartrazine concentration. For tartrazine concentration in the range from 0.1 to 0.5 μM, the plot showed an excellent linear trend with a slope of 1.56 × 108 M–1 and the LOD was calculated to be 26 nM. The mechanism of fluorescence quenching of CDs by tartrazine may either be due to the excited state energy transfer or inner filter effect. Since both the CDs and tartrazine exhibit absorption in the same region, the most possible quenching mechanism is the inner filter effect. A similar type of sensing mechanism was proposed in the literature.[45] The time-resolved fluorescence technique is employed to understand the sensing mechanism. Figure S8 shows the fluorescence decay of CDs and with tartrazine. As expected, the fluorescence decay is not altered even in the presence of the highest concentration of tartrazine; thus, the quenching type is static with an inner filter effect.

Figure 7

(a) Fluorescence quenching study of CDs in the absence and presence of various concentrations of tartrazine (0.1–6 μM) in water. (b) Detection limit plot.

Since the possibility of occurrence of few metal ions in food products along with coloring agents is high, it is essential to study the interference of metal ions at the same time as sensing of tartrazine. To understand the role of metal ions on the fluorescence quenching of CDs by tartrazine, we have conducted the interference studies with various metal ions. CDs were exposed to various metal ions (1 μM), and fluorescence spectra were recorded. Surprisingly, the Fe3+ ion quenches the fluorescence of CDs predominantly. Figure shows the selectivity plot of CDs with metal ions. Further, to quantify Fe3+ ion sensing, the fluorescence spectrum of CDs was recorded as a function of Fe3+ ion, as shown. The fluorescence spectrum of CDs was gradually decreased with the increasing concentration of Fe3+ ion (Figure S9a). The limit of detection of the Fe3+ ion was calculated using the three (σ/slope) relations. The linearity was observed for the plot of fluorescence intensity change against the lower concentration of Fe3+ ion (Figure S9b). The lowest detection limit was determined to be 13 nM. A combined XPS and FT-IR result indicated that there are carboxyl, hydroxyl, and amine functional groups on the surface of the as-synthesized CDs. Therefore, it could be inferred that CDs can form complex with Fe3+ ions through such functional groups, resulting in fluorescence quenching with high selectivity/sensitivity. Similar reports are available in the literature.[46−48]

Figure 8

Selectivity plot of CDs with various metal ions (1 μM).

Intriguingly, both tartrazine and Fe3+ ion quench the CD fluorescence with LOD values of 26 and 13 nM, respectively. In this higher sensitivity, it is very difficult to distinguish the one that quenches the fluorescence of CDs. To get rid of the mentioned issue, we are giving remedy for sensitive detection of tartrazine even if Fe3+ ions persist in the solution. Based on our experimental observations, the selective sensing of tartrazine could be achieved by the addition of 1 mM potassium thiocyanate to solution mixtures. Since Fe3+ can readily complex with potassium thiocyanate and it is not free to quench the fluorescence of CDs, in this way, one can qualitatively detect the tartrazine in the mixture. The effect of interference of other food additives, such as monosodium glutamate (flavor enhancer) and acesulfame K (artificial sugar) has also been investigated, and the results are shown in Figure S10. The fluorescence of CDs is not quenched by both monosodium glutamate and acesulfame K. However, in the presence of tartrazine, the strong quenching was observed. Hence, the prepared CDs are prospective materials for the detection of tartrazine even with other food additives.

Detection in Soft Drinks

The prepared fluorescent CDs were employed to detect the trace level of tartrazine in various soft drink samples. The results of the pretreated soft drinks spiked with known concentrations of tartrazine are shown in Table . The excellent recoveries ranging from 87 to 100.6% were attained. These results signify the accuracy and reliability of the prepared CDs toward the determination of tartrazine in commercial soft drinks.

Table 1

Recovery of Tartrazine Spiked with Different Soft Drinks by CDs

soft drinks	added (μM)	found (μM)	recovery (%)
sample I	0.1	0.0991	99.18
	0.2	0.2012	100.61
	0.4	0.3994	99.85
sample II	0.1	0.1008	100.87
	0.2	0.1980	99.02
	0.4	0.4012	100.3
sample III	0.1	0.0870	87
	0.2	0.1997	99.87
	0.4	0.3922	98.06

Conclusions

In summary, successful conversion of waste soot into fluorescent CDs by a simple one-pot method is demonstrated. The fluorescence property of the derived CDs was effectively used to detect tartrazine through a turn-off fluorescence method. Further, this method is successfully employed to detect tartrazine in soft drinks. Thus, the prepared CDs could be used for selective and sensitive determination of tartrazine in foodstuff products. Overall, this work demonstrated the conversion of a vehicle soot to a value-added product and the same has been successfully utilized for sensing applications.

Experimental Section

Materials and Methods

Waste soot was collected from a diesel engine lorry exhaust pipeline. Ultrapure water was used for the synthesis of carbon dots and all measurements. A dialysis membrane bag was purchased from HiMedia, and a dialysis tube Float-A-Lyzer was purchased from Synergy Scientific Services. Tartrazine was purchased from Sigma-Aldrich and used as such without further purification. Monosodium glutamate and acesulfame K were procured from TCL chemicals. The commercial coloring agent (Kesari powder) was purchased from a nearby provisional store.

The particle size and morphology of the synthesized CDs were investigated by high-resolution transmission electron microscopy (HRTEM; JEOL JEM-2100; with an accelerating voltage of 120 kV). The TEM specimen was prepared by dropping a diluted CD solution onto a carbon-coated copper grid followed by the evaporation of the water solvent. SEM measurement and EDAX analysis were done by using Nova NANO SEM 600 from FEI Company, Netherlands. The FT-IR spectrum was recorded by using a JASCO FT-IR ATR 6300 spectrometer at room temperature in the range of 4000–400 cm–1. 1H NMR (500 MHz) and 13C NMR (125, 100 MHz) spectra of CDs were recorded on a Bruker NMR spectrometer in D2O with tetramethylsilane (TMS) as an internal standard. Topographic analysis is done with Park Systems (XE-100 AFM) for an area of 10 × 10 μm scale with a resolution of <1.5 nm. Absorption spectra were recorded using a PerkinElmer Lamda 25 UV–visible spectrophotometer. The fluorescence quenching measurements were carried out with an Hitachi-made fluorescence spectrometer (model: F-4500). Time-resolved fluorescence decays were obtained by the time-correlated single photon counting (TCSPC) technique, exciting the sample at a 360 nm LED source. Data analysis was carried out by the software provided by IBH (DAS-6), which is based on deconvolution techniques using the nonlinear least squares method, and the quality of the fit is ascertained with a value of χ2 < 1.1.

Sensing of Tartrazine in Soft Drinks

The soft drink samples were procured from a local superstore and used as is. In a typical assay, 0.3 mL of CDs is added to 2.7 mL of soft drinks in a cuvette and fluorescence was recorded. Further, the tartrazine solution (up to 30 μL; the maximum concentration of tartrazine is 6 μM) was gradually added into the CD solution and the solution was mixed well prior to the fluorescence spectrum being recorded.