Introduction

Calcium ions (Ca2+) are widely found in disparate fields, such as industry and agriculture, [1] and have been recognized as a highly important element for life [2, 3]. Advanced detection techniques for Ca2+ such as capillary electrophoresis, [4] inductively coupled plasma atomic emission spectrometry [5] and mass spectrometry [6] have been developed. In addition, the method of titration has been widely used for Ca2+ determination in the past decades [7,8,9,10]. At first, Ca2+ separated from the magnesium was determined by ethylene diamine tetraacetic acid (EDTA) titration. Then, the influence of masking agents and indicators on the chelatometric determination of Ca2+ was investigated [11]. This well-known technology, however, is often limited by the low selectivity and complexity of the detection process based on the intrinsic complexation response [12, 13]. Traditionally, the endpoint was indicated with ion-selective electrodes by Eriochrome Black T for calcium titration at pH 10 [14,15,16]. Griko studied the correlation titration reaction between Ca2+ and EDTA, and obtained the strong dependence of binding thermodynamics on reaction in the buffer [17]. Therefore, it is of great importance to develop a unique and facile system for the detection of low-concentration of Ca2+.

Oxytetracycline (OTC), a compound belongs to the tetracycline family, is known to be a basic medicinal antibiotic substance in life [18,19,20]. As a broad-spectrum antibiotic, it has been used in clinical application for the treatment or prevention of infections in bovine, pig, poultry, and fish productions [21,22,23]. And it also has been widely used in agriculture and animal husbandry as a growth promoter or bacterial infection inhibitor against a wide range of Gram-positive and Gram-negative bacteria based on the inhibition of bacterial protein synthesis [24, 25]. OTC is an amphoteric compound with unique structure of the presence of both Lewis base and Lewis acid functional groups. The location of donor group allows tetracycline being an applicable ligand to chelate with metal ions [26].

Taking advantage of the specific performance of OTC, we developed a new convenient and sensitive system for the detection of Ca2+ using a turn on probe. The probe OTC has extremely weak measurable fluorescence in the organic solvent of N,N-dimethylformamide (DMF), a polar inert solvent, which is a very versatile solvent with wide industrial applications [27]. While maintaining a stable molecular state, the reaction product of OTC and Ca2+ also has good solubility and chemical stability. And the addition of Ca2+ leads to the OTC to produce an enhanced fluorescence response, which shows a stable and rapid response in a narrow range of calcium concentration.

Experimental Section

Materials and Reagents

Oxytetracycline (OTC) dihydrate (C22H24N2O9·2H2O), N,N-Dimethylformamide (DMF), dimethyl sulfoxide (DMSO), acetonitrile, dichloromethane, acetone, isopropanol, L-ascorbic acid (Vc), L-histidine (His) were purchased from Aladdin Chemistry Co. Ltd. (Shanghai, China). Absolute ethanol and methanol were got from Beijing Chemical Works. Calcium chloride (CaCl2), sodium chloride (NaCl), potassium chloride (KCl), barium chloride (BaCl2), sodium nitrate (NaNO3), sodium bicarbonate (NaHCO3), sodium sulfate (Na2SO4) and iron chloride (FeCl3), mercuric chloride (HgCl2), manganese chloride (MnCl2), lead nitrate (Pb(NO3)2), nickel chloride (NiCl2), cupric chloride (CuCl2) and ammonium chloride (NH4Cl) were purchased from Sinopharm Chemical Reagent Co. Ltd. All reagents and solvents selected were used directly without further purification. Ultrapure water (18.2 MΩ·cm) was taken from the Millipore water purification system. The fluorescent cuvette was washed twice with ethanol and then dried before use.

Apparatus and Methods

Fluorescence measurements were carried out on a F-7100 Fluorescence spectrometer (HITACHI, Japan) with excitation and emission slit widths of 10 mm. UV-vis absorption was determined at a Shimadzu UV-2550 spectrometer at room temperature. Fourier transformed infrared (FTIR) spectra of OTC and CaCl2 were performed on the FTIR spectrophotometer. The model of the electronic balance used in the experiment was FA2204N (Shanghai Jinghai Instrument Co.Ltd.). Freeze dryer (FD-1A-50, Beijing BIOCOOL Experimental Instrument Co.Ltd) was prepared for the preparation of composite solid samples.

Fluorescence Responses to Ca2+

When DMF solution was used as the detection environment, the strong polarity of the DMF not only makes the OTC have good solubility, but also makes the response of OTC and Ca2+ more stable. The experimental error was reduced by formulating the OTC probe into a 1 mM stock solution in methanol, [28] while weighing the anhydrous calcium chloride to obtain a 1 mM stock solution in water for use, and then diluting the different stock solutions for later use. Typically, 6 μL of the OTC solution was injected into 1994 μL of DMF in a quartz fluorescence cuvette to give a final concentration of 3 μM. The response was recorded after Ca2+ was added to the probe solution under the same conditions. The final concentrations of Ca2+ were 0, 0.5, 1, 1.5, 2, 2.5, 3, 4, and 5 μM, respectively. All fluorescence spectra were chosen to be excited at 365 nm. The scan rate is the default value of 1200 nm min −1.

Fluorescence Responses to Biological Reagents and Metal Ions

As is known to us, heavy metal ions or other biological reagents are abundant in domestic tap water environments with a small concentration, while the hydroxyl functional groups on OTC and the position of hydrogen ions can rapidly coordinate with many metal ions, so it is necessary to study the selectivity and anti-interference ability of OTC to other metal ions. Among them, Na+, K+, Ca2+, Ba2+ and Fe3+ were selected as typical ions. The effect of other not typical ions such as Hg2+, Mn2+, Pb2+, Ni2+, Cu2+ and NH4+ were also evaluated. In addition, it includes anionic carbonate (CO32−), bicarbonate (HCO3−), sulfate (SO42−), nitrate (NO3−) and chloride (Cl−), which are often present in water. In addition, several amino acids were investigated as representative interferences, such as L-ascorbic acid (Vc) and L-histidine (His).

Results and Discussion

Optimal Conditions of Solution Environment

The performance of OTC in the following solution environments, including ultrapure water, Tris buffer (10 mM), ethanol solution, DMF, DMSO, acetonitrile, dichloromethane, acetone, and isopropanol were investigated. As shown in the Fig. 1a, OTC exhibited a low fluorescence response in ultrapure water and Tris buffer solution (pH = 7), in which there was no fluorescence-enhanced response with the addition of Ca2+. However, the fluorescence of OTC can be enhanced significantly when it dissolved in organic solvent DMF or ethanol, which can be attributed to its instability in ultrapure water and Tris buffer solution. It also exhibited good response to calcium ions in other organic solvents including DMSO and isopropanol. In the solution system of acetonitrile, methylene chloride, and acetone, no significant fluorescence enhancement was observed. The final concentration of OTC and calcium ions was 3 μM, and the conditions remained consistent. It is possible that the hydroxyl group forms hydrogen bonds with water in aqueous solution which damaged the sensitive response of OTC to Ca2+, while for the organic solvent DMF, it maintained the unhydrolyzed structure of OTC. By evaluating the experimental conditions, the solvent with high polarity and density was found to enable the complex molecules to maintain stable fluorescence performance, and thus the best experimental conditions were obtained.

Fig. 1
figure 1

a From left to right: DMF, aqueous solution, Tris buffer (10 mM), ethanol, DMSO, acetonitrile, dichloromethane, acetone and isopropanol. I0 and I are the fluorescence intensity of OTC at 500 nm and OTC + Ca2+ at 510 nm (λex = 365 nm), respectively. b Excitation and emission spectra (the positions of the excitation wavelength are 355 nm, 360 nm, 365 nm, 370 nm, 375 nm respectively) of the mixture of OTC and Ca2+ in DMF. c Fluorescence spectra of the probe OTC (3 μM) in the presence of different concentrations of Ca2+ in DMF. d The linear relationship between the fluorescent intensity of OTC probe and Ca2+ concentration. I and I0 are the fluorescence intensity of OTC at 500 nm and OTC + Ca2+ at 510 nm (λex = 365 nm), respectively. e The visualization photograph for OTC and addition of different concentrations about Ca2+ was taken under illumination by a 365 nm UV portable lamp in the dark. f Fluorescence intensity ratio of the probe OTC (3 μM) before and after the addition of Ca2+ (0–5 μM). g The fluorescence intensity of OTC at 3 μM with and without of Ca2+ with time. The detection time is limited to 2 h

Full size image

Fluorescence Determination of Ca2+

OTC dissolved in DMF displays absorption and emission maxima at 380 and 510 nm, respectively, with a relatively large Stokes shift (130 nm). Furthermore, as the excitation wavelength varied from 355 nm to 375 nm, the emission peak of the mixture at 510 nm in the DMF solution did not change, indicating that it is not an excitation wavelength dependent emission spectroscopy system (Fig. 1b). The excitation of the adjacent wavelength at 365 nm can only change the intensity of the fluorescence, and does not affect the experimental results which previously proposed.

The fluorescence response time of the probe OTC to Ca2+ was studied before the sensitivity study to specifically determine the sensing performance of OTC. With the addition of Ca2+, the fluorescence was immediately enhanced and remained unchanged even with increased incubation time was further increased, indicating that the equilibrium between OTC and Ca2+ was rapid and stable (Fig. 1c). Clearly, the fluorescence intensity ratio of I/I0 illustrates a superior linear correlation (R = 0.998) with Ca2+ concentration in the range of 0–3 μM (Fig. 1d), which was estimated according to the definition of the three times of the blank signal deviation (3σ). And the limit of detection (LOD) was obtained at 125 nM. In addition, this fluorescence enhancement could be easily visualized with the probe solution changing in fluorescence color from darkness to intense green under a 365 nm UV lamp (Fig. 1e), which indicates a significant change in chromatic aberration. In the same period, fluorescence data was used to explore the ratio of 1:1 as support data (Fig. 1f). Also, it was also evidenced by recording the fluorescence of OTC and OTC+ Ca2+ for 2 h as shown in Fig. 1g. The UV–vis absorption spectra of OTC (30 μM) with Ca2+ (0–50 μM) in DMF (Fig. 2a) that showed a proportional coordination pattern. Clearly, the reaction of OTC with Ca2+ results in a new absorption band at 384 nm accompanied by gradually decreasing of the original absorption peak at 369 nm, leading to the formation of an isosbestic point at 366 nm, indicating the formation of a new Ca complex. Here, the chart clearly shows the saturation at 30 μM, indicating a 1:1 coordination of compound OTC with Ca2+ (Fig. 2b).

Fig. 2
figure 2

a UV–vis absorption spectra of OTC (30 μM) with Ca2+ (0–50 μM) in DMF. b Response of the signal to Ca2+ under increasing concentrations (0–50 μM) with a concentration of OTC at 30 μM

Full size image

Fluorescence Quantum Yield Measurement

Fluorescence quantum yield of OTC was calculated using the following equation. Where ФF stands for fluorescence quantum yields, Abs and ΣF represent the absorbance at the excitation wavelength and the measured integrated fluorescence intensity, and η is the refractive index of the solvent used [29,30,31]. It is known that the refractive index of ethanol is 1.36242, while the refractive index of DMF is 1.3331. Rhodamine 6G (ФF = 95%) in ethanol was selected as standards.

$$ {\upvarphi \mathrm{F}}_{\mathrm{sample}}={\upvarphi \mathrm{F}}_{\mathrm{standard}}.\frac{{\mathrm{Abs}}_{\mathrm{standard}}}{{\mathrm{Abs}}_{\mathrm{sample}}}.\frac{\sum {\mathrm{F}}_{\mathrm{sample}}}{\sum {\mathrm{F}}_{\mathrm{standard}}}.\frac{{\upeta^2}_{\mathrm{sample}}}{{\upeta^2}_{\mathrm{standard}}} $$

At the same time, it can be proved that the increase of OTC concentration will slightly increase the fluorescence. The fluorescence quantum yields of OTC in DMF was measured to be 7.34% using Rhodamine 6G as a reference.

Binding Constant

The binding constant K of the complex was calculated with a linear relationship by the Benesi-Hildebrand method. Where Fmin is the fluorescence intensity of OTC, Fmax is the intensity measured with an excess of Ca2+, F is the intensity within a linear relationship, and K is the binding constant. The value of K was obtained from a plot of 1/(Fmax - F) against 1/[Ca2+] where K is equal to the intercept / slope [32,33,34].

$$ \frac{1}{\mathrm{F}-{\mathrm{F}}_{\mathrm{min}}}=\frac{1}{\mathrm{K}\left({\mathrm{F}}_{\mathrm{max}}-{\mathrm{F}}_{\mathrm{min}}\right)\left[{\mathrm{Ca}}^{2+}\right]}+\frac{1}{{\mathrm{F}}_{\mathrm{max}}-{\mathrm{F}}_{\mathrm{min}}} $$

Here, again, the binding constant of OTC to Ca2+ ions was obtained by the actual results of fluorescence titration and the related Benesi-Hildebrand method equation. The K value is calculated from the slope by plotting and fitting the obtained data. More importantly, within this range, it was found that 1/I-I0 is liner to 1/ [Ca2 +] (R2 = 0.994), indicating that OTC is very useful in determining the concentration of Ca2+. The binding constant (K) derived from the fluorescence titration data of OTC to Ca2+ in the DMF solution was calculated to be 5.77 × 104 M−1 (Fig. 3), which suggested that OTC exhibited good binding capacity with Ca2+.

Fig. 3
figure 3

Benesi-Hildebrand plot based on a 1:1 association stoichiometry between the OTC with Ca2+

Full size image

Effects of the Probe Concentration on the Sensitivity

Fluorescence responses of OTC (3 μM) to various ions in DMF including Fe3+, K+, Na+, Ba2+, carbonate ion (NO3−), sulfate ion (SO42−), bicarbonate ion (HCO3−), L-ascorbic acid (Vc), L-histidine (His), Hg2+, Mn2+, Pb2+, Ni2+, Cu2+ and NH4+ were investigated to confirm the unique fluorescence response of Ca2+ and OTC as shown in Fig. 4a. The concentration of the added substances was 3 μM. Only the addition of Ca2+ resulted in a significant fluorescence enhancement, and no remarkable changes in the fluorescence of OTC were observed upon the addition of other substances [35, 36]. To investigate the potential interference of these coexisting substances, furthermore, we measured the fluorescence response of OTC toward Ca2+ in the presence of mixed coexisting substances with the concentration at 3 μM. As shown in Fig. 4b, all of the coexisting substances have little effects on the detection of Ca2+. These results suggested that the compound OTC is highly selective to Ca2+ over other relevant species.

Fig. 4
figure 4

a Fluorescence responses of OTC (3 μM) to various species in DMF including Fe3+, K+, Na+, Ba2+, carbonate ion (NO3−), sulfate ion (SO42−), bicarbonate ion (HCO3−), L-ascorbic acid (Vc) and L-histidine (His), Hg2+, Mn2+, Pb2+, Ni2+, Cu2+ and NH4+. The concentration of added substances was 3 μM. b Anti-interference exploration of the probe OTC (3 μM) to coexisting Fe3+, K+, Na+, Ba2+, carbonate ion (NO3−), sulfate ion (SO42−), bicarbonate ion (HCO3−), L-ascorbic acid (Vc), and L-histidine (His) in DMF. The concentration of interfering substances was 3 μM

Full size image

Mechanism Study

Detailed characterization was performed to determine its reaction performance by UV-Vis spectra and Fourier transform infrared (FTIR) spectroscopy. Figure 5a showed the UV − Vis absorption spectra of OTC in the absence and presence of Ca2+. Clearly, the reaction of OTC with Ca2+ results in a new absorption band at 384 nm accompanied by gradually decrease of the original absorption peak at 369 nm, indicating the formation of the new Ca complex. The FTIR spectra of the compound OTC, calcium chloride and reacted mixtures were shown in Fig. 5b. The peak at about 3400 cm−1 was observed due to the stretching of the O-H groups of OTC. The C=O group stretching vibration of amid corresponded to the peak at approximately 1585 cm−1. Nevertheless, the shift of distinctive peak, such as C=O groups, suggested there was a bonding reaction here in the process [37].

Fig. 5
figure 5

a UV–vis absorption spectra of compound OTC without and with the addition of Ca2+. b FTIR spectrum of compound OTC and the mixture of OTC + Ca2+ against the background of potassium bromide

Full size image

Before as we know, tetracycline derivatives have similar structure, but its chemical properties are different. Different position of hydroxyl group of tetracycline may affect its coordination performance which making it a good ligand exhibiting rich coordination site, such as dimethylamino and phenol-diketone group [38]. The protonation state of the compound explains the inconsistency between the binding sites of tetracycline and metal ions. In a certain solution environment, the proton dissociates from the tetracycline compound, and metal ions would choose an optimum binding site to form a stable complex [39]. It was observed that the hydration of metal ions affected the state of OTC, and it was aggregated in the presence of Ca2+. In addition, the solubility of metal complex is greater than that of the OTC, suggesting that Ca2+ somehow increases the solubility of OTC in solution [40]. OTC is known for its different ionized states in different solution conditions, and its ion binding patterns are diverse. As mentioned above, the fluorescence and absorption data showed that Ca2+ binds with OTC at a 1:1 stoichiometric rate, suggesting that the probe solution depends on the ionic binding pattern. The results showed that OTC coordinated with Ca2+ to form a stable complex in DMF, which exhibited significant fluorescence enhancement accordingly [41]. Sensing process based on the complexation under the excitation wavelength of 365 nm was showed in Scheme 1.

Scheme 1
scheme 1

Sensing process based on the complexation under the excitation wavelength of 365 nm

Full size image

Test of Ca2+ in Real Water Sample

To assess the potential application of the probe OTC for Ca2+ detection in real water systems, the recovery experiment was carried out in three kinds of water samples including tap water, river water and mineral water which were obtained from local place. The actual detection is carried out in with different contents of Ca2+. After pre-treatment of sedimentation and suspension, the water sample spiked with different amounts of Ca2+ was added the same concentration of OTC, and the fluorescence of the solution was then recorded. The relative standard deviations (RSD) were obtained by repeating the experiment two times under the same conditions. As shown in Table 1, the recoveries of various known concentrations of added Ca2+ were obtained from 103% to 109% in actual samples. Briefly, the relative standard deviations (RSD) were in the range of 1.4% to 6.7% which was generally satisfactory, showing the reliability of the probe OTC for Ca2+ determination in real samples.

Table 1 Test of Ca2+ spiked in actual samples
Full size table

Conclusion

In summary, we showed that oxytetracycline can be used as a highly sensitive fluorescent probe for detecting Ca2+ by the coordination effect to form a stable complex. A good limit of detection at 125 nM was obtained, which was favourable for sensitive and selective detection of Ca2+. The method was further applied to the living water system, which was analysed to evaluate the accuracy and applied of the present method. The strategy with unique fluorescence enhancement reported here definitely facilitates the evolution of portable and real-time calcium monitoring for visual identification application.