Determination of Cyanide in Water and Food Samples Using an Efficient Naphthalene-Based Ratiometric Fluorescent Probe.

Monitoring cyanide levels in water and food samples is crucial. Herein, we rationally developed a simple and efficient fluorescent probe for cyanide determination. The probe displayed selective ratiometric fluorescent response to cyanide. In addition, after treatment with cyanide, the fluorescence ratios (I 509/I 466) exhibited a good linearity with cyanide concentration in the range of 0-60 μM, and the detection limit was determined to be 0.23 μM (S/N = 3). Significantly, the practical application demonstrated that the probe was able to quantitatively detect cyanide concentration in natural water samples. Monitoring of endogenous cyanide in cherry nut by the probe was also successfully conducted. Notably, upon fabrication of test strips, the probe could be conveniently utilized for field measurement of cyanide in bitter almond without relying on sophistical instruments. Furthermore, the cyanide in potato tissues was determined for the first time by means of fluorescence imaging.

Introduction

Cyanide ion (CN–) is well known for its extreme toxicity to human health because of its close interaction with ferric iron atoms in metalloenzymes, which can inhibit important biological processes associated with oxidative metabolism and cellular respiration.[1] However, despite their toxicity, cyanides are widely used in a variety of industries, including mining, electroplating, metal cleaning, pharmaceuticals, plastics, etc.[2] The cyanide-contaminated industrial waste may pollute water resources, thereby posing great threat to human health. Moreover, cyanogenesis is widespread in plant species, including a large number of important food crop species, such as sorghum, cassava, almonds, bamboo shoot, white clover, sprouting potato, etc.[3−5] In cyanogenic plants, cyanide compounds are naturally generated via enzymatic hydrolysis of cyanogenic glycosides when the plant cells are ruptured.[6] Cyanide exposures to human beings frequently occur by consumption of raw or improperly processed cyanogenic food and plants,[7−9] which often leads to death or permanent neurologic deficits such as the paralytic disease recognized as konzo.[10,11] Thereby, the development of effective methods for detecting cyanide levels in water and food samples is of high demand.

Various strategies for detecting cyanide have been developed, including mass spectrometry,[12] electrochemistry,[13] ion chromatography,[14] Raman spectrometry,[15] flow injection,[16] colorimetry,[17] and fluorescent probe. Among these strategies, fluorescent probe seems to be most promising owing to its distinguished features of high selectivity and sensitivity, fast response time, and technical simplicity.[18] Recently, many fluorescent probes for sensing cyanide have been constructed by virtue of the high binding ability of cyanide to metal ions (Co2+, Cu2+, Fe3+, Hg2+, Zn2+),[19−23] H+,[24] and boronic acid derivatives.[25] Additionally, cyanide fluorescent probes have been developed on the basis of the specific nucleophilic reactivity of cyanide to electrophilic double bonds, such as C=C,[26] C=O,[27] C=N,[28] C=S,[29] or C=N+.[30] Nevertheless, most of the reported fluorescent probes for determining cyanide in water samples and food samples only displayed fluorescence turn-on response.[31−37] A major limitation of the fluorescence turn-on probe is that the signal output is vulnerable to factors such as excitation intensity, probe concentration, and instrumental efficiency. By contrast, a ratiometric fluorescent probe can effectively eliminate the above-mentioned limitations and consequently provide more accurate analysis.[38,39] In addition, the preparation of the reported probes generally needs sophisticated synthesis procedures. From a wide application perspective, fluorescent probes with simple synthetical procedures are favorable. Therefore, the development of a simple and easy-to-prepare ratiometric fluorescent probe for determining cyanide in water and food samples would be of great significance.

To this end, we rationally designed and synthesized a novel fluorescent probe, compound 1, for sensing cyanide in water samples and food samples. In probe 1, a naphthalene fluorescence dye was employed as fluorophore, while a dicyanovinyl group functioned as the specific sensing reactive unit. Remarkably, probe 1 displayed ratiometric fluorescent response to the cyanide with high selectivity and sensitivity, making probe 1 usable to precisely sense cyanide. Furthermore, probe 1 was easily prepared from the commercially available materials in only one step with high yield. The simple synthetic procedure is conducive to the wide application of probe 1. Importantly, the practical application experiments established that probe 1 was successfully applied for quantitatively determining cyanide in various water and food samples.

Results and Discussion

Synthesis and Characterization

As shown in Scheme , probe 1 was easily synthesized in only one step. Condensation of 2-naphthalenecarboxaldehyde and propanedinitrile in the presence of zinc chloride afforded probe 1 in 84.3% yield. The simple synthetic procedure significantly lowers the cost of probe preparation, which contributes to the wide application of the probe. The product has been fully characterized by 1H NMR and 13C NMR spectroscopy, electrospray ionization (ESI) mass spectrometry, and elemental analysis.

Scheme 1

Synthetical Procedures of Probe 1 and its Reaction with Cyanide Affording Compound 1-CN

Optical Responses to Cyanide

To study the optical properties of probe 1 with cyanide, the fluorescence spectra were firstly investigated in 20 mM potassium phosphate buffer/dimethylformamide (DMF) (4:6 v/v, pH 7.4). As shown in Figure a, probe 1 (20 μM) itself exhibited intense fluorescence emission at 466 nm. Upon adding increasing amount of cyanide, the fluorescence emission at 466 nm was gradually attenuated and a new fluorescence emission centered at 509 nm was increased. Meanwhile, an isoemission point was noted at 482 nm, implying that the sensing reaction is a clear process. The red-shifted fluorescence response made probe 1 sense cyanide in ratiometric manners that can effectively reduce the potential interference factors from changes in excitation intensity, probe distribution, and instrumental efficiency. In the absence of cyanide, the fluorescence intensity ratio (I509/I466) of probe 1 was calculated to be 0.49. With addition of increasing amount of cyanide, the ratios (I509/I466) were gradually augmented and became constant when the amount of cyanide reached 80 μM (Figure b). At this point, the ratio (I509/I466) was calculated to be 1.66. As depicted in Figure S1, the ratios (I509/I466) also displayed a good linearity with cyanide concentration in the range of 0–60 μM (Y = 0.48646 + 0.01771X, R2 = 0.99782), implying that probe 1 can be potentially used for quantitatively detecting cyanide. The detection limit was found to be 0.23 μM (S/N = 3), which was below the maximum allowable cyanide concentration (1.9 μM) in drinking water set by World Health Organization (WHO).[40] In addition, after reaction with cyanide, the fluorescence color of probe 1 gave obvious changes from blue to green (Figure c). Moreover, the photostability study showed that probe 1 was stable in the buffer solution upon irradiating at 297 nm for 30 min (Figure S2). The effect of water volume fraction on probe 1 sensing cyanide was also explored. Probe 1 can be used for sensing cyanide even in the solution with 70% water (Figure S3).

Figure 1

(a) Changes in fluorescence spectra (λex = 297 nm) of probe 1 (20 μM) with various amounts of cyanide (0–140 μM); (b) changes in fluorescence emission ratios (I509/I466) of probe 1 (20 μM) with various amounts of cyanide (0–140 μM); (c) fluorescence color changes of probe 1 (20 μM) before and after addition of cyanide (140 μM); and (d) changes in UV/vis spectra of probe 1 (20 μM) with various amounts of cyanide (0–140 μM). The error bars represent the standard deviation (n = 3).

The absorption spectra of probe 1 upon titration with cyanide solution were recorded. Probe 1 displayed an intense absorption centered at 332 nm and a shoulder absorption at 279 nm (Figure d). Upon progressively adding cyanide, the absorption peaks at 332 and 279 nm gradually decreased. Meanwhile, a weak absorption at 344 nm and an absorption at 278 nm were finally formed. Correspondingly, after treatment with cyanide, the solution color of probe 1 under visible light changed from yellow to colorless (Figure S4).

Selectivity Studies

The specific nature of probe 1 to cyanide was inspected. Introducing 140 μM of various anions (CN–, F–, Cl–, Br–, I–, HCO3–, NO3–, SCN–, CH3COO–, HSO3–, ClO4–) and biological molecules (glucose, Gly, Cys) into probe 1 solution gave rise to almost no fluorescence intensity ratio (I509/I466) response (Figure ). The obvious ratio (I509/I466) changes only occurred when the cyanide was added. Moreover, naked eye detection was carried out. Under the illumination of UV light, the solution of probe 1 showed specific fluorescence color change to cyanide (Figure , inset). The sensing behaviors of probe 1 to cyanide in the presence of other species were also studied. The other species hardly intervened the response of probe 1 to cyanide (Figure S5). These evidences elaborated that probe 1 displayed highly selective response to cyanide.

Figure 2

Fluorescence emission ratio (I509/I466) response of probe 1 (20 μM) to 140 μM of various anions and biological molecules in 20 mM potassium phosphate buffer/DMF (4:6 v/v, pH 7.4): (1) blank; (2) CN–; (3) F–; (4) Cl–; (5) Br–; (6) I–; (7) HCO3–; (8) NO3–; (9) SCN–; (10) CH3COO–; (11) HSO3–; (12) ClO4–; (13) glucose; (14) Gly; and (15) Cys. The excitation wavelength was 297 nm. Inset: visual fluorescence color changes of probe 1 (20 μM) under illumination of a handheld UV lamp in the presence of various anions (140 μM): (a) blank; (b) CN–; (c) Cl–; (d) SCN–; (e) HSO3–; and (f) Cys.

Response Time and Effect of pH

The sensing rate of probe 1 with cyanide was monitored by fluorescent spectra (Figure ). After treatment with 140 μM cyanide, pronounced fluorescence intensity variations were noted at 509 and 466 nm, respectively. Notably, the intensities can reach constant within 1 min. The rapid response toward cyanide is beneficial for the detection of cyanide in real time. The fluorescence responses of probe 1 to cyanide under different pH values were measured. In the absence of cyanide, the fluorescence intensity ratio of probe 1 showed negligible variation in the pH range of 2.3–10.1 (Figure S6), denoting that probe 1 was stable in the solution under these pH values. However, in the presence of cyanide, the fluorescence intensity ratio of probe 1 drastically enhanced when the pH value of solution was higher than 6.26. Therefore, probe 1 is suitable for the detection of cyanide under neutral and alkaline pH conditions.

Figure 3

Time-dependent changes in the fluorescence intensities recorded at 466 and 509 nm observed upon treatment of probe 1 (20 μM) with cyanide (140 μM).

Sensing Reaction Mechanism

To investigate the sensing reaction mechanism of probe 1 with cyanide, the sensing reaction product of probe 1 with cyanide, 1-CN, was isolated. The fluorescence excitation spectra and fluorescence emission spectra of 1-CN were identical with cyanide-treated probe 1 (Figure S7). Therefore, 1-CN was responsible for the fluorescence properties of probe 1 with cyanide. Then, 1-CN was characterized by 1H NMR and 13C NMR spectroscopy. In the 1H NMR spectra of free probe 1, the peak at 8.28 ppm corresponds to the vinylic proton of dicyanovinyl group (Ha) (Figure ). Nevertheless, in 1-CN, the signal of vinylic proton at 8.28 ppm disappeared. Concomitantly, two new signals appeared at 4.63 ppm (Hb) and 4.32 ppm (Hc). Moreover, in the 13C NMR spectra of probe 1, the signals for the two vinylic carbons of dicyanovinyl group were located at 159 and 82 ppm, respectively. The two signals were shifted to 39 and 29 ppm correspondingly in the 13C NMR spectra of 1-CN (Figure S8). Therefore, the sensing reaction of probe 1 with cyanide is the nucleophilic addition of cyanide to the β-position of dicyanovinyl moiety in probe 1. Furthermore, the formation of cyanide adduct 1-CN was confirmed by ESI mass spectrometry, where a major peak at m/z 230.01 is assigned to [1-CN]− (Figure S9).

Figure 4

1H NMR (400 MHz) spectra of (1) probe 1 and (2) isolated product of probe 1 + cyanide in CDCl3.

Density Functional Theory (DFT) Calculations

To get insight into the optical response of probe 1 to cyanide, probe 1 and its cyanide adduct 1-CN were examined by time-dependent density functional theory (TD-DFT) calculations at the B3LYP/6-31+G** level using Gaussian 09 program. For probe 1, the electron distribution and energy levels of the frontier molecular orbitals are illustrated in Figure . In the ground-state geometry, the HOMO-1 → LUMO transition (oscillator strength f = 0.7296) and HOMO → LUMO + 1 transition (f = 0.3082) in probe 1 were allowable. And it is evident that the HOMO-1 → LUMO transition was the intramolecular charge transfer process from naphthalene moiety to the dicyanovinyl moiety. This transition was responsible for the UV/vis absorption peak of probe 1 at 332 nm (Figure d). In contrast, the HOMO → LUMO + 1 transition corresponded to the π–π* local transition in naphthalene moiety, which contributed to the UV/vis absorption peak of probe 1 at 279 nm (Figure d). In the first excited-state geometry, the LUMO → HOMO transition is allowable (f = 0.1679), which was the reason for the fluorescence of probe 1 at 466 nm (Figure a). The frontier molecular orbitals of 1-CN are displayed in Figure S10. In the ground-state geometry, it is clear that the HOMO → LUMO + 3 transition (f = 0.4488) is electron transfer from the naphthalene moiety to the 1,1,2-tricyano ethane moiety, which was responsible for UV/vis absorption peak of 1-CN at 278 nm (Figure d). In the first excited-state geometry of 1-CN, the LUMO → HOMO-1 transition (f = 0.4435) is attributed to the emission of 1-CN at 509 nm (Figure a). Therefore, the optical properties of probe 1 and 1-CN have been theoretically revealed.

Figure 5

Explanation of the absorption and fluorescence emissions of probe 1: the molecular geometry relaxation upon excitation and the frontier molecular orbitals associated with the vertical excitation (i.e., absorption, left-hand-side columns) and fluorescence emission (right-hand-side column) of probe 1. The vertical excitations were studied on the basis of optimized ground-state geometry, and the fluorescence emission was studied on the basis of the optimized geometry of the excited state. Water was employed as the solvent (PCM model). IC represents internal conversion and CT represents conformation transformation. Excitation and radiative processes are denoted as the solid arrow, and the nonradiative processes are denoted as the dotted arrow.

Practical Application

Cyanide in the industrial waste could possibly be dumped into water and pollute the water resource. Thus, probe 1 was applied for sensing cyanide in the natural water samples (tap water, Yangtze River water, and pond water). As shown in Table , cyanides in these natural water samples were not detected. When different concentrations of cyanide were further added into the natural water samples, probe 1 was able to measure the concentrations of cyanide with good recovery (Table ). These studies established that probe 1 is capable of quantitatively determining cyanide levels in the natural water samples.

Table 1

Determination of Cyanide Concentrations in Natural Water Samples

sample	CN– spiked (mol L–1)	CN– recovered (mol L–1)a	recovery (%)
tap water 1	0	not detected
tap water 2	3.00 × 10–4	(2.97 ± 0.04) × 10–4	99.0
tap water 3	2.00 × 10–3	(1.99 ± 0.02) × 10–3	99.5
Yangtze River 1	0	not detected
Yangtze River 2	3.00 × 10–4	(2.95 ± 0.03) × 10–4	98.3
Yangtze River 3	2.00 × 10–3	(1.94 ± 0.06) × 10–3	97.0
pond water 1	0	not detected
pond water 2	3.00 × 10–4	(2.99 ± 0.03) × 10–4	99.7
pond water 3	2.00 × 10–3	(2.02 ± 0.01) × 10–3	101.0

Relative standard deviations were calculated based on three times of measurement.

In cyanogenic food plants, the cyanide is generated by enzymatic hydrolysis of cyanogenic glycoside when their tissues are damaged. Thus, we further explored the potential application of probe 1 for determining cyanide in various food samples. First, probe 1 was utilized to detect cyanide in cherry,[41] a popular fruit. The cherry flesh extract samples and cherry nut extract samples were prepared from commercially available cherry. Then, 0.5 mL of the prepared extract samples were added to the probe 1 solution. As shown in Figure , no notable fluorescence variation was noted in the probe 1 solution with cherry flesh extract sample, while pronounced fluorescence response was observed in probe 1 with cherry nut extract sample. These inferred that the cyanide was mainly generated in the cherry nut but not in cherry flesh.

Figure 6

Fluorescence emission spectra of 20 μM probe 1 (black square solid), probe 1 with 0.5 mL of cherry flesh extract sample (blue circle solid), and probe 1 with 0.5 mL cherry nut extract sample (bright green triangle up solid). The excitation wavelength was 297 nm. Inset: visual fluorescence color changes of 20 μM probe 1, probe 1 with cherry flesh extract sample, and probe 1 with cherry nut extract sample under illumination of a handheld UV lamp.

Subsequently, to investigate the practical application of probe 1, test strips were prepared by immersing filter papers into DMF solution of probe 1 (0.1 M) and then drying them in air. Then, the test strips were dipped into solution with different volumes of bitter almond extract samples to determine the endogenous cyanide in bitter almond. As shown in Figure , the fluorescence colors of test strips gradually changed from blue to green with increasing volume of bitter almond extract samples (Figure c–f), denoting that probe 1 test strips were able to determine the different concentrations of endogenous cyanide in bitter almond extract samples. The present sensing strategy is appealing because it does not rely on elaborate instrumentation and can be used for field measurement.

Figure 7

Photographs of probe 1 test strip (a), probe 1 test strip dipped into 5 mL of KCN solution (100 μM) (b), probe 1 test strip dipped into 5 mL of solution with 0 mL (c), 0.1 mL (d), 0.2 mL (e), and 0.6 mL (f) of bitter almond sample. The photos were taken under illumination of a handheld UV lamp.

Fluorescence imaging of plant tissues by probe 1 is another avenue to determine the endogenous cyanide in food samples. Potato, a vegetable food, was chosen as the proof of concept. The nonsprouting and sprouting potato were cut into fresh slices at room temperature. After incubated with probe 1 solution, the slices were used for fluorescence imaging by a laser confocal scanning microscope. As shown in Figure a, strong fluorescence in the blue channel was observed in the nonsprouting potato slice, displaying that probe 1 was able to permeate into potato tissues. In addition, there was almost no fluorescence in the green channel. Thus, no cyanide was generated in the nonsprouting potato. However, in the sprouting potato slice, strong fluorescence in the green channel (Figure e) and almost no fluorescence in the blue channel was observed (Figure d), inferring that the cyanide was produced in the sprouting potato. Moreover, when the sprouting potato slice was preincubated with cyanide inhibitor AgNO3 solution,[42] and then stained with probe 1, strong fluorescence was noted in the blue channel (Figure g) and weak fluorescence was detected in the green channel (Figure h), indicating that fluorescence response in sprouting potato is a cyanide-dependent event. It should be mentioned that this is the first example of monitoring cyanide levels in potato tissues by means of fluorescence imaging.

Figure 8

Confocal fluorescence microscopy images for cyanide detection in potato slices using probe 1. (a, b) Fluorescence images of nonsprouting potato slices stained with 50 μM probe 1 in blue and green channels, respectively. (c) Bright field image of (a) or (b). (d, e) Fluorescence images of sprouting potato slices stained with 50 μM probe 1 in blue channel and green channel, respectively. (f) Bright field image of (d) or (e). (g, h) Fluorescence images of the sprouting potato slices pretreated with 100 μM AgNO3 and then stained with 50 μM probe 1 in blue and green channels, respectively. (i) Bright field image of (g) or (h). The scale bar in (a)–(i) is 12 μm.

Conclusions

In this work, we rationally constructed a novel fluorescent probe, compound 1, for determining cyanide concentration in natural water samples and food samples. In probe 1, the naphthalene dye was utilized as fluorophore and the dicyanovinyl group was employed as the specific recognition site. Notably, the probe can be readily synthesized in only one step with high yield, which is beneficial to the wide application of probe 1. Upon treatment with cyanide, probe 1 displayed ratiometric fluorescent response with high sensitivity and selectivity, and the detection limit was determined to be 0.23 μM (S/N = 3). The sensing reaction mechanism was fully investigated. The optical response of probe 1 to cyanide was rationally confirmed by TD-DFT calculation. Importantly, probe 1 was successfully applied for determining cyanide in water and food samples, such as natural water, cherry, almond, and potato. We believe that the novel and simple ratiometric fluorescent probe developed in this work will be widely used for determining cyanide levels in various water and food samples.

Experimental Section

Synthesis of Compound 1

2-Naphthalenecarboxaldehyde (0.500 g, 3.2 mmol), propanedinitrile (0.211 g, 3.2 mmol), and zinc chloride (0.435 g, 3.2 mmol) were mixed in a glass test tube. The mixture was heated to melt and stirred for 1 h. After cooling to room temperature, the crude product was washed with 5% aqueous ethanol and filtered. Then, the yellow solid was recrystallized in dichloromethane and petroleum ether to afford compound 1 (0.55 g, yield 84.3%). Mp: 136–138 °C; 1H NMR (CDCl3, 400 MHz), δ (ppm): 8.28 (s, 1H), 8.08 (dd, J = 1.6 Hz, 8.4 Hz, 1H), 7.95 (m, 3H), 7.90 (s, 1H), 7.68 (t, J = 7.2 Hz, 1H), 7.61 (t, J = 7.2 Hz, 1H); 13C NMR (CDCl3, 100 MHz), δ (ppm): 159.7, 135.9, 134.4, 132.6, 130.0, 129.7, 129.6, 128.6, 128.0, 127.8, 124.2, 114.0, 112.9, 82.2; MS (ESI) m/z 205.05 [M + H]+; elemental analysis calcd (%) for C14H8N2: C 82.33, H 3.95, N 13.72; found C 82.02, H 3.97, N 13.68.

Measurement Procedures

First, the stock solution of fluorescent probe 1 (5 × 10–4 M) was dissolved in DMF and the stock solutions of various anions or biological molecules (1 × 10–3 M) were dissolved in water. Then, the test solution of probe 1 (20 μM) with various testing species in 20 mM potassium phosphate buffer/DMF (v/v 4:6, pH 7.4) was prepared by placing 0.2 mL of the probe 1 stock solution, 2.8 mL of DMF, and an appropriate aliquot of each anion or biological molecule stock into a 5 mL volumetric flask, and then diluting the solution to 5 mL with potassium phosphate buffer solution. The resulting solution was shaken and stood at room temperature for 5 min before measuring the absorption spectra and fluorescence spectra.

Determination of Cyanide in Natural Water Samples

The natural water samples were obtained from tap water, Yangtze River water, and pond water (from the campus of Jiangsu University). The determination of cyanide concentration in the water samples was conducted by addition of 0.1 mL of water sample to a solution containing 0.2 mL of probe 1 stock solution (5 × 10–4 M) and 2.8 mL of DMF in a 5 mL volumetric flask and then diluting the solution to 5 mL with 20 mM potassium phosphate buffer (pH 7.4). After the resulting solution was incubated at room temperature for 5 min, the fluorescence emission spectra (λex = 297 nm) were recorded. On the basis of fluorescence intensity ratio (I509/I466) and the equation provided in Figure S1, the cyanide concentrations in the three natural water samples were quantitatively determined. Subsequently, different concentrations of cyanide (3.0 × 10–4 and 2.0 × 10–3 M) were added into three natural water samples. Then, the concentrations of cyanide in these natural water samples were determined by the same method.

General Procedure for Preparation of Food (Cherry Flesh, Cherry Nut, and Bitter Almond) Extract Samples

The food samples were purchased from a local market (Zhenjiang, Jiangsu Province, P.R. China). As indicated by the manufactures, the cherry was originated from Qingdao, Shandong Province, China, and the bitter almond was from Baoding, Hebei Province, China.

The procedure for the preparation of food extract samples was according to a reported procedure.[36] Food samples (10 g, cherry flesh, cherry nut, or bitter almond) were well crushed and pulverized, and then stored in a sealed flask for 60 min at room temperature to release the cyanide. After that, 10 mL of water and 50 mg of NaOH were added to the flask. The resulting mixture was vigorously stirred for 5 min. The obtained mixture was centrifuged at 10 000 rpm for 10 min. The supernatant as food extract sample was used for further analysis.

Determination of Endogenous Cyanide in Cherry Flesh Extract Sample and Cherry Nut Extract Sample

Cherry flesh extract sample or cherry nut extract sample (0.5 mL) was added into a solution of 0.2 mL of probe 1 stock solution (5 × 10–4 M) and 2.8 mL of DMF in a 5 mL volumetric flask, and then diluting the solution to 5 mL with 20 mM potassium phosphate buffer (pH 7.4). The resulting solution was shaken well and incubated at room temperature for 5 min before recording the spectra.

Determination of Endogenous Cyanide in Bitter Almond Extract Sample by Test Strips

The test strips were prepared by immersing filter papers into DMF solution of probe 1 (0.1 M) and then drying them in air. The obtained test strips were dipped into 5 mL solution of DMF with different volumes of bitter almond extract sample (0, 0.1, 0.2, 0.6 mL). After 5 min, the test strips were lifted out, dried in air, and then observed under a 365 nm UV lamp.

Determination of Endogenous Cyanide in Potato by Fluorescence Imaging

Nonsprouting and sprouting potato were washed and sliced into 0.5–1 mm thick disks. Then, the slices were incubated with probe 1 (50 μM) in DMF medium for 30 min at 37 °C. After that, the slices were washed with PBS three times. For the control experiment, the sprouting potato slices were pretreated with 100 μM AgNO3 solution for 15 min, and then the slices were further incubated with 50 μM probe 1 solution for 30 min. Finally, fluorescence imaging was conducted by a laser confocal scanning microscope. The blue channel was set at 425–475 nm with excitation at 364 nm, and the green channel was set at 500–550 nm with excitation at 364 nm.