Paper test strip for silver ions detection in drinking water samples based on combined fluorometric and colorimetric methods

2.1 Characterization methods

BP crystals were first disrupted by ultrasound using an ultrasonic cell disruptor (BILON92-IIL, Shanghai Bilang Instrument Co., ltd., China) with ice bath and then heated in an electric blast drying oven (GZX-9140MBE, Shanghai Boxun Industrial Co., ltd., Medical equipment factory). The resulting BPQDs were suspended on a copper grid, and the transmission electron microscopy (TEM) images were obtained by a JEM 2100 transmission electron microscope (JEOL Co., ltd., Japan) at an accelerating voltage of 200 kV. Scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDS) measurements were performed on a Thermo Fisher Scientific FIB-SEM GX4 instrument (Thermo Scientific ltd., USA). Ultraviolet–visible (UV–vis) absorption and fluorescence spectra of BPQDs were recorded by LUH4150 UV–Vis and F-7100 spectrophotometers (Hitachi, Japan), respectively. Photographs were taken with a Coolpix 4500 digital camera (Nikon, Japan). Fourier transform infrared (FT-IR) spectra were recorded on a Nicolet 6700 spectrometer (Thermo Scientific ltd., USA). Chemical bonding states of BPQDs elements were determined by K-Alpha X-ray photoelectron spectroscopy (XPS) (Thermo Scientific ltd., USA). X-ray diffraction (XRD) patterns of BPQDs products were obtained by an AD/Max-2500 diffractometer (Rigaku, Japan). The viscosity of BPQDs solution was measured using a microscopic visual rheometer (Antonpa Group, Austria).

2.2 Chemicals and materials

BP crystal powder was purchased from Nanjing XFNANO, Inc. (Nanjing, China). AgNO₃, Ca(NO₃)₂, Fe(NO₃)₃, Co(NO₃)₂, ZnSO₄, Mg(NO₃)₂, BaCl₂, Ni(NO₃)₂, HgCl₂, Pb(NO₃)₂, CrCl₃, Al₂(SO₄)₃, ZrCl₄, Ba(NO₃)₂, HNO₃, and NaOH were ordered from Aladdin Reagent Co., ltd. (Shanghai, China). Quinine sulfate (QS) was obtained from Fluka (Buchs, Switzerland). All chemicals procured from commercial sources were analytical reagents, which were used directly without specific illustration. A Milli-Q water purification system was utilized to prepare ultrapure water for all solutions.

2.3 Preparation of BPQDs

A method for BPQDs synthesis reported in the literature was used for reference (Fan et al., 2022) with some modifications. Briefly, bulk BP (9 mg) was added in dimethylformamide (DMF) (25 mL) and ultrasonically broken in an ice bath at a power of 65 % for 2 h. The obtained suspension was transferred to an autoclave and heated at 200 °C for 12 h. After cooling to room temperature, the solution was centrifuged at 6000 rpm for 10 min. The supernatant was separated with a 0.22-μm filter. BPQDs products were dispersed in DMF. To obtain the concentration of BPQDs, the reduced pressure distillation was carried out at 80 °C. Then BPQDs solid products were obtained, which were stored at 4 °C.

2.4 Preparation of paper test strips

Paper test strips were prepared with filter paper on glass plates. Firstly, the filter paper was cut into 1 × 1 cm² small pieces. Secondly, 20 μL BPQDs was added dropwise to each piece. Then the filter paper pieces were allowed to dry in a vacuum drying oven at 40 °C for 1 h. After drying, the obtained BPQDs@filter paper was cut into strips with appropriate size for later use, which was stored at 4 °C after sealing.

2.5 Fluorometric and colorimetric detection of Ag⁺

Before the determination of Ag⁺ the experimental conditions were optimized. Herein, the colorimetric system was chosen as the representative for the condition optimization. First, the acidity of system was investigated using 0.1 M NaOH and 0.1 M HNO₃ solutions to adjust pH. Briefly, after adjusting pH, 200 μL BPQDs (0.34 mg/mL) solution was mixed with 2 μL Ag⁺ (6.37 mg/mL) solution and diluted to 500 μL. After the solutions with pH 5–13 were incubated for 6 min, the absorbance was measured, respectively. Second, the incubation time for Ag⁺ detection was also tested. After titration of Ag⁺ (13.59 μg/mL) in BPQDs solution, the absorbance of BPQDs is recorded every 2 min until it reaches a plateau.

BPQDs can be used to detect Ag⁺ at pH 9. First, a series of Ag⁺ standard solutions were prepared. 200 μL BPQDs solution (0.34 mg/mL) was mixed with 2 μL Ag⁺ solution and diluted to 500 μL. The solutions were tested after 6 min incubation. The fluorescence intensity or absorbance of BPQDs was recorded in the presence and absence of Ag⁺.

During the detection of Ag⁺ with the portable test strip, BPQDs@filter paper was picked up by tweezers and immersed briefly into a solution containing Ag⁺. The paper test strip was taken out from solution and kept in air for 1 min.

2.6 QY measurements

To obtain the QY of BPQDs, QS was used as a reference reagent during measurements. Absorbance (A) (365 nm) and fluorescence intensity (λ_ex/λ_em = 365/434 nm) of QS and BPQDs were measured with A values less than 0.05. The following equation was utilized for QY calculation: Q_BPQDs = Q_QS(m_BPQDs/m_QS)(η_BPQDs/η_QS)² (1), where Q, m, and η represented QY, the slope, and the refractive index, respectively. According to the equation, the QY of BPQDs (Q_BPQDs) can be calculated with the QY of QS at 0.54.

2.7 Calculation method

First-principle calculations in this study were performed by a pseudo-potential plane-wave method using the Vienna Ab initio Simulation Package (VASP) (Kresse and Furthmüller 1996). To investigate Ag adsorption, BPQDs were modeled using a 3 × 3 supercell with a single layer containing a total of 36 P atoms. A 3 × 2 × 1 k-point sampling based on Gamma-centered Monkhorst–Pack scheme was conducted, and a vacuum layer with a thickness of 20 Å was added along the z-direction to avoid interactions between periodically repeated structures. The generalized gradient approximation (GGA) with the Perdew − Burke − Ernzerhof (PBE) functional (John et al., 1996) was employed to describe electron exchange–correlation interactions. A conjugate gradient algorithm was utilized for geometry optimizations with a force convergence criterion of 0.02 eV/Å.

2.8 Preparation of real samples

Real samples were tested by measuring Ag⁺ contents in mineral water and fruit water. The drinking water samples used in this experiment were purchased from local supermarket. They included Chun Yue ⋅ Fruit Water (Coca-Cola China) and Changbai Snow ⋅ Mineral Water (Nongfu Spring). All measurements were performed directly without pretreatment.

3.1 Characterization of BPQDs

The particle size distribution of BPQDs was confirmed by TEM. As shown in Fig. 1a, BPQDs are uniformly dispersed without aggregation, and their average particle size is 1.7 ± 0.4 nm, which is larger than those of BPQDs synthesized by other methods (Ren et al., 2020, Fan et al., 2022). Fig. 1b displays the TEM image of BPQDs + Ag⁺ with apparent aggregates, suggesting that BPQDs and Ag⁺ form new complexes.

Close

Fig. 1

TEM image of BPQDs (inset: particle size distribution) (a) and BPQDs + Ag⁺ (b), SEM images of BPQDs@filter paper (c) and BPQDs@filter paper + Ag⁺ (d), EDS images of BPQDs@filter paper + Ag⁺ (e–h).

Export to PPT

The SEM observations of BPQDs@filter paper were performed with and without Ag⁺. The microscopic surface images of BPQDs@filter paper and BPQDs@filter paper + Ag⁺ are presented in Fig. 1c and d, respectively. The EDS analysis of BPQDs was also conducted. It can be observed that BPQDs are evenly distributed on the filter paper (Fig. 1e–h). The EDS images of BPQDs@filter paper + Ag⁺ show grape-like particles due to the reaction between BPQDs and Ag⁺.

The obtained FT–IR spectra (Fig. 2a) clearly identify the chemical bonds of BPQDs with and without Ag⁺. In general, most characteristic peaks of BPQDs and BPQDs + Ag⁺ are identical. The absorption peaks at 3425 and 1470 cm⁻¹ are attributed to N—H bonds and those at 2926 and 2851 cm⁻¹ are caused by –CH₃ species. The peaks at 1717 and 1660 cm⁻¹ originate from C⚌O bonds and those at 1470 and 1391 cm⁻¹ represent –CH₃ and –CH₂– functional groups, respectively. Therefore, in the synthesis process of BPQDs, DMF acts as not only a solvent, but also a carbon source. The absorption peak at 1170 cm⁻¹ is likely caused by P⚌O species, and the peak at 1061 cm⁻¹ is from (δ) P—H bonds (Xu et al., 2020). After Ag⁺ addition, three new absorption peaks appear including (m) P—H phosphine at 2362 cm⁻¹, (υ) P–OH from O⚌P–OH at 2332 cm⁻¹ (Xu et al., 2020), and the P–Ag bond at 676 cm⁻¹(Jiang et al., 2020, Xu et al., 2020).

Close

Fig. 2

FT–IR (a) profiles of BPQDs + Ag⁺ and BPQDs, XRD (b) spectrum of BPQDs, and XPS (c, d) profiles of BPQDs + Ag⁺ and BPQDs.

Export to PPT

According to the XRD spectra (Fig. 2b), there is no characteristic peaks inducing that BPQDs possess amorphous structure. To confirm the elemental composition and surface states, XPS studies were conducted (Fig. 2c and d). Fig. 2c shows the general survey scans for BPQDs and BPQDs + Ag⁺. The atomic contents of BPQDs are O 1 s: 11.92 %, N 1 s: 2.64 %, C 1 s: 48.53 %, and P 2p: 1.96 %, respectively. The atomic contents of BPQDs + Ag⁺ are O 1 s: 20.23 %, N 1 s: 2.54 %, C 1 s: 49.97 %, P 2p: 3.84 %, and Ag 3d: 0.14 %, respectively. The atomic ratio of O 1 s and P 2p increases by a factor of two for BPQDs + Ag⁺ due to the formation of P_xO_y species catalyzed by Ag⁺ (see the characteristic peak at 133.3 eV in Fig. 2d) (Wang et al., 2020). A new peak for BPQDs + Ag⁺ at 134.2 eV is ascribed from P–Ag bond (Fig. 2d).

The UV–Vis absorption and fluorescence spectra of BPQDs are shown in Fig. 3a. There are two peaks at 260 and 350 nm in the absorption spectrum. The maximum excitation and emission wavelength are 365 nm and 434 nm, respectively. The inset of Fig. 3a shows the photograph of BPQDs under indoor and UV light. BPQDs display colorless in the day light, while they exhibit blue fluorescence under the UV lamp (365 nm). In addition, the emission spectra of BPQDs can be obtained with different excitation wavelengths. With an increase in the excitation wavelength, the position of the emission peak undergoes a slight red shift, while its intensity first gradually increases and then gradually decreases (Fig. 3b). As a result, the maximum excitation wavelength is 365 nm. The results of QY measurement are presented in Fig. 3c, and the relative QY was calculated according to equation (1) at 8.82 %. At room temperature (25 °C), the steady-state viscosity of BPQDs solution was measured. It can be seen from Fig. S1 (in Supplementary Information) that when the shear rate is infinitely close to 0, the viscosity of BPQDs solution is about 2.5 mPa·s and is similar to that of water (2.98 mPa·s). In addition, the obtained BPQDs solid products were brown and showed bright blue fluorescence under ultraviolet light.

Close

Fig. 3

Absorption, excitation, and emission spectra of BPQDs (inset: photograph of BPQDs obtained under sunlight and UV light) (a), effect of excitation wavelength on the fluorescence spectra of BPQDs in solution (b), QY values of BPQDs obtained using a QS standard (c).

Export to PPT

In general, the fluorescence spectrum and intensity of one fluorophore can vary in different solvents. The effect of solvent on the fluorescence of BPQDs was investigated. As shown in Fig. S2a (in Supplementary Information), the maximum λ_ex of BPQDs in different solvents is between 361 nm and 371 nm, and the maximum λ_em is between 440 nm and 455 nm. When the λ_ex is fixed at 365 nm, the fluorescence intensity of BPQDs is highest in DMF and lowest in DMSO (Fig. S2b, in Supplementary Information). In the mixture of DMF with one other solvent (volume ratio, 1:1), BPQDs show similar fluorescence (Fig. S2c, in Supplementary Information).

3.2 Optimization of sensing conditions

Acidity of the solution and reaction time are two important parameters for the detection of Ag⁺. Following the steps of Section 2.5, the absorbance can be obtained under different pH values (Fig. S3a, in Supplementary Information). The absorbance gets to the maximum with pH 8–9 and it decreases in an acidic or strong base environment. The possible mechanism may be that when pH is greater than 9, Ag⁺ first reacts with excess OH^– leading to the amount of free Ag⁺ decrease and inducing the decrease of absorbance for BPQDs + Ag⁺. To eliminate the solvent effect, the pH of DMF solution was adjusted after the same treatment with the results presented in Fig. S3a (in Supplementary Information). Regardless of the pH adjustment, the treated DMF was incapable of detecting Ag⁺. Thus, it is not possible to directly detect Ag⁺ in DMF solution. The results of response time for Ag⁺ are illustrated in Fig. S3b (in Supplementary Information). There is almost no change in the absorbance value observed after 6 min of reaction, indicating that the incubation time of BPQDs and Ag⁺ is 6 min.

3.3 Selectivity

Before the selectivity study, different systems were tested for Ag⁺ detection. The fluorescence and absorption response to different species are shown in Fig. S4 (in Supplementary Information). There are five different systems including Ag⁺, DMF, DMF + Ag⁺, BPQDs and BPQDs + Ag⁺ tested, respectively. As expected, there are no fluorescence and absorption peaks for Ag⁺ and there is no obvious change of fluorescence and absorption spectra for DMF and DMF + Ag⁺. The fluorescence intensity of BPQDs decreases significantly in the presence of Ag⁺ (Fig. S4a, in Supplementary Information). Furthermore, a new and distinct absorption peak appears for BPQDs + Ag⁺ (Fig. S4b, in Supplementary Information). It can be deduced that BPQDs can be used for assay of Ag⁺ based on a dual-signal output of fluorescence and absorption methods.

In addition, to broaden the application scope of the present system, water content was tested. Both the solution and paper test strip systems were used to detect water in DMF (Fig. S5, in Supplementary Information). It can be observed that there is negligible change for fluorescence intensity, absorbance and color of paper for BPQDs in the presence and absence of water. Therefore, the proposed system can not be used for water detection.

To confirm the specificity of BPQDs for Ag⁺ determination, 15 metal ions with the same concentration (13.59 μg/mL) were tested. The fluorescence and absorption spectra and corresponding intensity and absorbance are shown in Fig. 4. As shown in Fig. 4a, the fluorescence intensity of BPQDs is significantly reduced by Ag⁺, which is not observed for other metal ions. Fig. 4c shows that after the addition of Ag⁺ to the BPQDs solution, a new peak appears at 410 nm, and only Ag⁺ has this phenomenon. The solution of BPQDs + Ag⁺ is yellowish-brown, whereas others remain colorless (see the inset of Fig. 4d). It can be concluded that a chemical reaction occurs possibly in the BPQDs + Ag⁺ system to form new species. Therefore, BPQDs can be used simultaneously to detect Ag⁺ by monitoring changes in the fluorescence intensity and UV–Vis absorbance.

Close

Fig. 4

Effect of various metal ions on the fluorescence intensity (a, c) and absorbance (b, d) of BPQDs (the inset of Fig. 4d: the photographs of BPQD solutions with different metal ions), (condition: 0.34 mg/mL BPQDs, pH 9, time 6 min), respectively.

Export to PPT

To examine the interferences of other metal ions mixing with Ag⁺, the fluorescence intensity and UV–Vis absorbance were obtained for 14 metal ions mixing with Ag⁺ with the same concentration, respectively. Fig. S6a and S6b (in Supplementary Information) clearly show that most other metal ions produce negligible effect on the detection process.

To facilitate the detection of Ag⁺, BPQDs were transferred to filter paper to fabricate the BPQDs@filter paper test strips. In this process, 20 μL BPQDs (0, 0.068, 0.136, 0.204, 0.272, and 0.340 mg/mL) was added to filter paper substrate (Fig. 5a). The filter paper strips were subsequently dried and used for the detection of 25.48 μg/mL Ag⁺. It can be observed that the color of strips deepens with increasing concentration of BPQDs. As Fig. 5b illustrates, the test strip with Ag⁺ turns yellowish-brown, whereas others don't change color. It can also confirm the selectivity of BPQDs@filter paper for detecting Ag⁺.

Close

Fig. 5

Photographs of BPQDs@filter paper test strips with different BPQDs concentrations (from left to right: 0, 0.068, 0.136, 0.204, 0.272, and 0.340 mg/mL) for the detection of Ag⁺ with the same concentration (a), photographs of BPQDs@filter paper test strips with various metal ions (b).

Export to PPT

Close

Fig. 6

Fluorescence spectra of BPQDs obtained at different Ag⁺ concentrations (a) and the corresponding linear relationship between the fluorescence decrease and Ag⁺ concentration (b), UV–Vis absorption spectra of BPQDs obtained at different Ag⁺ concentrations (inset: the photographs of BPQD solutions with Ag⁺) (c) and the corresponding linear relationship between absorbance and Ag⁺ concentration (d), (condition: Ag⁺ in Fig. 6a and 6c 0, 1.70, 3.40, 6.79, 10.9, 13.59, 16.99, 21.23, 25.48 μg/mL, 0.34 mg/mL BPQDs, pH 9, time 6 min).

Export to PPT

3.4 Sensitivity

Fluorescence and absorption spectra of BPQDs were measured with and without Ag⁺ in different concentrations (Fig. 6a and c). The relationship between the fluorescence intensity at 434 nm (Fig. 6b) or absorbance at 410 nm (Fig. 6d) and the concentration of Ag⁺ were plotted. It can be observed that there is a linear relationship between the fluorescence decrease (F₀-F) and Ag⁺ concentration (Fig. 6b) with the LOD at 1.56 μg/mL. There is also a linear relationship between absorbance and Ag⁺ concentration (Fig. 6d) with the LOD at 0.19 μg/mL. In the inset of Fig. 6c, the color of the BPQDs solution changes from colorless to yellowish-brown with increasing Ag⁺ concentration.

Table S1 (in Supplementary Information) summarizes the results for Ag⁺ detection using various fluorescence and colorimetric methods. By comparison with these methods using different materials, it can be found that BPQDs are firstly used to detect Ag⁺. The proposed method has a relatively wide detection range, and the LOD is at a medium level. BPQDs can realize the dual-channel detection of Ag⁺ by fluorometry and colorimetry. Furthermore, other methods listed here are performed in solution, and only our BPQDs system uses paper test strip for Ag⁺ detection, enabling the portable detection.

3.5 Stability

Stability is an important parameter to assess the performance of a sensor. The stability of the BPQDs sensor was investigated by two ways.

First, the performance of BPQDs in detecting Ag⁺ was evaluated by the absorption method after the storage of BPQDs in a refrigerator (about 4 °C) and at room temperature (about 26 °C) (Kuswandi et al., 2008), respectively. The absorbance change of BPQDs + Ag⁺ over time is obtained, and A₀ and A are the absorbance before and after storage of BPQDs, respectively (Fig. S7a, in Supplementary Information). It can be clearly seen that when BPQDs solution is stored at room temperature, A/A₀ value reaches below 80 % within one day. When it is stored at 4 °C, however, the value of A/A₀ is still over 90 % after six days.

Second, the fluorescence properties of BPQDs and QS, as well as carbon dots (CDs) and gold nanoclusters (Au-NCs) were compared under 300 W Xe lamp illumination (Long et al., 2018). After the irradiation of Xe lamp for a period, the fluorescence intensity changes for the four fluorescers were recorded with F₀ and F representing intensity before and after illumination, respectively (Fig. S7b, in Supplementary Information). The corresponding photographs of the above species under the UV lamp are collected (Fig. S7c-g, in Supplementary Information). The results show that after 30 min of Xe lamp irradiation the fluorescence intensity of QS and Au-NCs decreases to below 50 %, while that of CDs decreases to 87.2 %. However, for BPQDs the fluorescence intensity changes less than 10 % after irradiation of 90 min. The corresponding photographs also display the similar phenomenon. It can be deduced that the fluorescence stability of BPQDs is higher than that of QS, CDs and Au-NCs.

3.6 Applications

To explore the practical applications, BPQDs were employed for the detection of Ag⁺ in Chun Yue ⋅ Fruit Water (Coca-Cola China) and Changbai Snow ⋅ Mineral Water (Nongfu Spring) samples. The drinking water samples were spiked with certain amount of Ag⁺ to determine the percent recovery of Ag⁺ (Li et al., 2019). The results obtained by UV–Vis spectroscopy are listed in Table S2 (in Supplementary Information). It shows that Ag⁺ is not detectable in the original Fruit Water and Mineral Water samples. The spiked concentration of Ag⁺ was 3.40, 6.79, and 10.19 μg/mL, respectively. The results are listed in Table S2 (in Supplementary Information). The calculated recovery varies between 97.35 % and 101.47 %. The relative standard deviation (RSD) values for three measurements are no more than 2.42 %. Atomic absorption spectrometry (AAS) was conducted as reference method for the measurement of the corresponding spiked drinking water samples (Table S2, in Supplementary Information). The results of the two methods were compared by t-test. When the confidence is 95 %, there is no significant difference between the two systems. According to Table S3 (in Supplementary Information), t_{0.05, 2} is 4.30, and the values of t in Table S2 (in Supplementary Information) are all less than 4.30. Therefore, there is no significant difference between the two methods, and BPQDs can be effectively used for the detection of Ag⁺ in drinking samples.

BPQDs@filter paper was also used to test the real samples. A paper test strip was completely immersed into Ag⁺ solution with the concentration in the range of 0–3.40 mg/mL. Fig. S8a (in Supplementary Information) shows that the color of BPQDs@filter paper changes gradually. Chun Yue ⋅ Fruit Water samples were spiked with Ag⁺ with the concentration of 0, 0.68, 1.36, 2.04, and 2.72 mg/mL, respectively, as shown Fig. S8b in Supplementary Information. By comparison with the results from the standard and spiked samples of Ag⁺, they are consistent with each other.

3.7 Mechanisms of the fluorescence and colorimetric detection methods

Stern–Volmer curves are employed to discuss the mechanism of fluorescence quenching mode for BPQDs in the presence of Ag⁺ at different temperatures (Fig. S9a, in Supplementary Information). The relationship between fluorescence intensity and Ag⁺ concentration is linear at 4, 24 and 44 °C, respectively. The slope of the curve obtained at 4 °C is greater than those of 24 °C and 44 °C. It indicates that the quenching effect of Ag⁺ on the fluorescence of BPQDs is mainly ascribed to the static quenching mode resulting from the formation of non-fluorescent compounds. The fluorescence decay curves of BPQDs and BPQDs + Ag⁺ show almost completely overlap (with life time at 3.30 and 3.45 μs) (Fig. S9b, in Supplementary Information). Hence, the observed fluorescence quenching is consistent with static quenching mode.

According to the TEM (Fig. 1b) and SEM (Fig. 1c and d) images of BPQDs@filter paper and BPQDs@filter paper + Ag⁺, BPQDs chemically react with Ag⁺ to produce new compounds on the test paper surface. The FT–IR spectrum depicted in Fig. 2a confirms the P–Ag bond.

To further explore the mechanism of the reaction of BPQDs with Ag⁺, an adsorption model of BPQDs + Ag⁺ is established by calculating the adsorption energy and charge densities at various atoms (Kresse and Furthmüller 1996), as illustrated in Fig. 7. It shows a schematic diagram (Fig. 7a) and the top view (Fig. 7c) of the Ag⁺ adsorption on the BPQDs surface as well as the charge density difference (Fig. 7b and d) at the Ag_ad atom. The electron differential density map illustrates the interaction between Ag and P atoms. The cyan and yellow isosurfaces represent the depletion and accumulation of electrons, respectively, and the isosurface value is 3 × 10⁻³ e/ų. These results indicate that Ag is adsorbed on the outer surface of BPQDs and that electrons on the Ag surface are concentrated near P atoms. Then it leads to the formation of covalent bonds between P and Ag atoms. These results consist with each other.

Close

Fig. 7

Schematic illustration of the adsorption of Ag⁺ on the BPQDs surface: side view (a) and top view (c), and the corresponding charge density difference at the Ag_ad atom (b, d).

Export to PPT