Introduction

Carbon dots (CDs) are a new class of carbon nanomaterials and exhibit the fluorescence properties that trigger extensive studies since they were discovered in 2004 [1]. The fluorescence of CDs efficiently responds to the presence of analytes, which has been utilized in sensing applications [2]. As a fluorescent probe, CDs possess the attractive advantages of the flexibility in surface modification, high solubility in water, good photostability, nontoxicity, and excellent biocompatibility [3]. Typically, the strategies for design of CD-based probes include coordination of CDs with analytes, post-functionalization of CDs with specific receptors, and integration of CDs with others (e.g., fluorophores, quenchers, substrates) [4]. Surface modification and post-functionalization of CDs expand their potential for sensing applications [5]; however, construction of the multi-component detection through CDs without tedious surface modification is always a challenging task for the fluorescent probes. For example, CDs serve as an “on-off-on” fluorescent probe for detection of more than one analyte if metal ions (e.g., chromium(VI) ion, iron(III) ion, mercury(II) ion) as a quencher are removed from the surface of CDs by other analytes [6,7,8,9,10,11,12].

Copper(II) ion (Cu2+) is one of the essential elements in the human body. A lack of Cu2+ or too much Cu2+ intake will lead to serious health problems [13]. Therefore, it is of considerable importance to develop some probes for detection of Cu2+ [14]. In particular, CD-based fluorescent probes for detection of Cu2+ have been rapidly developed in the past few years [15]. The fluorescence of CDs can be quenched by Cu2+ and then recovered after adding other analytes (e.g., thiamine [16], glutathione [17], sulfide ion [18]), which provides a chance to develop the multi-component detection. It is well-known that Cu2+ can be reduced by ascorbic acid (AA) or sodium ascorbate to Cu+, which is commonly found in copper-catalyzed azide-alkyne cycloaddition (CuAAC) [19]. Achadu et al. prepared fluorescent tannic acid (TA)-modified graphitic carbon nitride dots (gC3N4 dots) as an “on-off-on” fluorescent probe for detection of AA via Cu2+ mediation [20]. The fluorescence of gC3N4 dots was quenched by Cu2+ and then recovered after adding AA. To the best of our knowledge, however, CDs without tedious surface modification have not been reported yet as an “on-off-on” fluorescent probe for the multi-component detection of Cu2+ and AA.

In this work, a CD-based “on-off-on” fluorescent probe for detection of Cu2+, AA, and acid phosphatase (ACP) is developed for the first time (Scheme 1). The surface carboxyl groups of CDs coordinate with Cu2+, leading to the fluorescence quenching [21]. Cu2+ is then reduced by AA, resulting in the fluorescence recovery [20]. The mixture of Cu2+ and AA producing Cu+ in situ plays an important catalytic role in CuAAC [19]. AA may originate from real samples including VC tablets, orange juice, and fresh orange. In addition, enzymatic hydrolysis of ascorbic acid phosphate (AAP) in the presence of acid phosphatase (ACP) also produces AA [22]. As a result, this is a convenient and simple fluorescent probe for the multi-component detection of Cu2+, AA (in real samples), and ACP.

Scheme 1
scheme 1

Schematic illustration of a CD-based “on-off-on” fluorescent probe for detection of Cu2+, ascorbic acid (AA), and acid phosphatase (ACP). The fluorescence of CDs is quenched by Cu2+ and then recovered with the addition of AA (a). Cu2+ can be reduced by AA to Cu+, which is usually used in copper-catalyzed azide-alkyne cycloaddition (CuAAC) (b). AA is commonly found in real samples such as VC tablets, orange juice, and fresh orange as well as enzymatic hydrolysate of ascorbic acid-phosphate (AAP) (c).

Full size image

Experimental

Chemicals and materials

Activated carbon (100-mesh particle size, powder) was purchased from Dengke Chemical Reagent Co. Ltd. (Tianjin, China). Sulfuric acid (H2SO4, 95.0–98.9%), nitric acid (HNO3, 65.0–68.0%), sodium hydroxide (NaOH, 96.0%), and hydrochloric acid (HCl, 36.0–38.0%) were purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China). All metal salts (NaCl, KCl, Sr(NO3)2, MgCl2, CaCl2, AlCl3, AgNO3, ZnCl2, Cd(NO3)2, BaCl2, Mn(NO3)2, Pb(NO3)2, Hg(NO3)2, Cr(NO3)2, Na2SO4, Na2SO3, Na2CO3, NaNO3, CuSO4, FeCl3, FeCl2), sucrose (Suc), glucose (Glu), d-fructose (Fru), lysine (Lys), tryptophan (Trp), glycine (Gly), cholesterol (Chol), tartaric acid (TA), citric acid (CA), l-ascorbic acid (AA), and glutathione (GSH) were purchased from Aladdin Chemical Reagent Co. Ltd. (Shanghai, China). Acid phosphatase (ACP) from potato, horseradish peroxidase (HRP), glucose oxidase (GOx), furin, and l-ascorbic acid 2-phosphate (AAP) trisodium salt were purchased from Sigma-Aldrich (Shanghai, China). Vitamin C (ascorbic acid) assay kit (colorimetric method) was purchased from Jiancheng Bioengineering Institute (Nanjing, China). Vitamin C tablets were purchased from a local pharmacy. Orange juice and fresh orange were purchased from a local supermarket. Deionized water was used throughout the experiment. Other chemicals were of analytical grade and directly used without any further purification.

Synthesis of carbon dots

Activated carbon (2.0 g) was added into a mixture of concentrated sulfuric acid (180 mL) and nitric acid (60 mL). After stirring at 80 °C for 5 h, the mixture was cooled and diluted with deionized water (800 mL). The dark solution was neutralized with sodium hydroxide and dialyzed in a dialysis bag (MWCO 1000 Da) for 3 days to remove large non-fluorescent materials. Finally, the solution was filtered through a 0.22-μm polyether sulfone (PES) membrane and the filtrate was stored at 4 °C for further use.

Characterization of CDs

The morphology of CDs was characterized by JEM-2010 FEF transmission electron microscope (JEOL, Japan). The absorption spectrum of CDs was performed on a UV 2600 UV-vis spectrophotometer (Shimadzu, Japan). The fluorescence spectra of CDs were recorded using a 10-mm path length quartz cell on an FS5 spectrofluorometer (Edinburgh, UK). The slit widths of excitation and emission were both 5 nm. Zeta potential of CDs was determined using a Zetasizer Nano ZS90 (Malvern, UK).

Detection of Cu2+, AA, and ACP

For detection of Cu2+, CDs solution and Cu2+ solution were mixed in the tubes. The mixture was diluted with deionized water to 2 mL and shaken thoroughly for 30 s. The fluorescence spectra were recorded. The concentration of Cu2+ ranged from 0 to 500 μM.

For detection of AA, CD solution and Cu2+ solution were mixed in the tubes. AA solution was then added into the tubes. The mixture was diluted with deionized water to 2 mL and shaken thoroughly for 30 s. The fluorescence spectra were recorded. The concentration of Cu2+ was 150 μM. The concentration of AA ranged from 0 to 3000 μM.

For detection of ACP, CD solution and Cu2+ solution were mixed in the tubes. ACP solution was mixed with AAP (2 mM, 20 μL) followed by incubation at 37 °C for 1 h. Then, the reaction solution was added into the tubes. The mixture was diluted with deionized water to 2 mL and shaken thoroughly for 30 s. The fluorescence spectra were recorded. The concentrations of Cu2+ and AAP were 150 μM and 20 μM, respectively. The concentration of ACP ranged from 0 to 7.5 mU mL−1.

Detection of AA in real samples

VC tablet, orange juice, and fresh orange, three kinds of real samples containing AA, were chosen to evaluate analytical performance of the probe. VC tablets were crushed and then dispersed in deionized water (50 mg mL−1). The mixture was centrifuged at 4000 rpm for 5 min to remove undissolved materials. Orange juice was diluted with deionized water and centrifuged at 4000 rpm for 5 min. Fresh orange was pretreated according to the literature [23]. Briefly, the edible part of fresh orange was taken and precisely weighed. Then, oxalic acid (0.2 mg mL−1, 50 mL) was mixed with processed orange. The mixture was treated by mashing and centrifugation at 6000 rpm for 20 min in a refrigerated centrifuge at 4 °C. The supernatant was filtered through a 0.45-μm PES membrane. The samples of VC tablets, orange juice, and fresh orange were diluted to 66, 15, and 50 times, respectively. The procedure for detection of AA in real samples was the same as specified above.

Determination of copper after dialysis

The samples of CDs + Cu2+, CDs + Cu2+ + AA, and Cu2++AA were prepared. The concentrations of CDs, Cu2+, and AA were 0.28 mg mL−1, 150 μM, and 2800 μM, respectively. Then, these samples were dialyzed in a dialysis bag (MWCO 3500 Da) for 24 h. The solution was transferred into porcelain boat followed by digestion in a muffle furnace at 600 °C for 2 h. The residue was dissolved in dilute HCl (1 mM, 2 mL). The concentration of copper was determined by Optima 5300 DV inductively coupled plasma optical emission spectrometer (ICP-OES) (PerkinElmer, USA).

Results and discussion

Synthesis and characterization of CDs

CDs were synthesized through oxidation of activated carbon using concentrated sulfuric acid and nitric acid [24,25,26]. A lot of hydroxyl and carboxyl groups exist on the surface of CDs. Therefore, the as-prepared CDs exhibit remarkable solubility and good stability in water. The CD aqueous solution can be stored at 4 °C for several months. CDs are nanoparticles in the range of 1–3 nm characterized by TEM (Fig. 1a) and have zeta potential of − 25.23 ± 0.32 mV (Fig. 1b). UV-vis absorption of CDs shows a single absorption band at 223 nm and the fluorescence emission of CDs displays a remarkable emission peak at 580 nm (Fig. 1c). With the increase of excitation wavelength from 340 to 420 nm, the fluorescence intensity of CDs decreases, showing excitation-independent behavior (Fig. 1d). The fluorescence emission of CDs excited at 360 nm is relatively strong; thus, the excitation at 360 nm is used throughout the experiment. CD aqueous solution by irradiation at 365 nm shows a bright yellow fluorescence (Fig. 1d, inset).

Fig. 1
figure 1

TEM image (a), zeta potential (b), UV-vis absorption (blue line) and fluorescence emission (black line, excitation wavelength of 360 nm) (c) of CDs and fluorescence emission spectra of CDs under different excitation wavelengths (from 340 to 420 nm) (d). The inset shows the photograph of CDs under natural light (left) and UV light at 365 nm (right)

Full size image

The effect of solution conditions on the fluorescence stability of CDs was investigated. No significant change in the fluorescence intensity of CD aqueous solution with pH values ranging from 2 to 7 is observed (see Electronic Supplementary Material (ESM) Fig. S1). The fluorescence intensity of CDs maintains strong and stable at different temperatures (see ESM Fig. S2). The change in the fluorescence intensity at different NaCl concentrations is negligible (see ESM Fig. S3). The results show that CDs possess outstanding fluorescence stability when pH, temperature, and NaCl concentration of CD aqueous solution change in a certain range.

Detection of Cu2+

The fluorescence emission spectra of CDs after adding Cu2+ were recorded. With the increase of Cu2+, the fluorescence intensity at 580 nm gradually decreases (Fig. 2a). There is a good linear relationship between the fluorescence intensity and the concentration of Cu2+ in the range of 5–100 μM (Fig. 2b). The limit of detection (LOD) for Cu2+ is 2.4 μM. The fluorescence quenching is rapid and the quenching completion time is approximately 1 min (see ESM Fig. S4). CDs synthesized by chemical oxidation of activated carbon powder using concentrated acids possess abundant carboxyl groups and can adsorb Cu2+ owing to the chelation [21]. The favorable energy level of the d-orbitals of Cu2+ with respect to the lowest unoccupied molecular orbitals of CDs produces non-radiative photoinduced electron transfer and consequently fluorescence quenching [27]. In a word, the fluorescence quenching of CDs by Cu2+ is attributed to Cu2+ having high thermodynamic affinity and fast chelating with carboxyl groups, leading to strong complex interaction between CDs and Cu2+ [28].

Fig. 2
figure 2

Fluorescence emission spectra of CDs with the addition of Cu2+ (0, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 120, 150, 200, 250, 300, 400, and 500 μM from top to bottom) (a). The linear relationship of the fluorescence intensity at 580 nm versus the concentration of Cu2+ in the range of 5–100 μM (b). The selectivity of “on-off” fluorescence strategy towards different cations and anions (c). F0 and F represent the fluorescence intensities at 580 nm of CDs and CDs + analyte, respectively. The concentration of analyte is 100 μM. The excitation wavelength is 360 nm. All experiments were repeated three times.

Full size image

To test the selectivity of “on-off” fluorescence strategy, analytes such as cations (Na+, K+, Sr2+, Mg2+, Ca2+, Al3+, Ag+, Zn2+, Cd2+, Ba2+, Mn2+, Pb2+, Hg2+, Cr2+, Fe3+, and Fe2+) and anions (SO42−, SO32−, CO32−, NO3−, and Cl−) were chosen. Compared with Na+, K+, Sr2+, Mg2+, Ca2+, Al3+, Ag+, Zn2+, Cd2+, Ba2+, and Mn2+, Cu2+ effectively quenches the fluorescence of CDs (Fig. 2c). Pb2+, Hg2+, Cr3+, and Fe2+, especially, Fe3+, have some interference as they can also be used as the quencher of CDs (see ESM Fig. S5). Fe2+, the product of Fe3+ getting an electron, has a relatively weak quenching ability. This phenomenon has been utilized to detect Fe3+ by a CD-based “on-off-on” fluorescent probe [29,30,31]. It also provides a thought to detect the reducing agents that can reduce Cu2+ to Cu+.

Detection of AA

For detection of AA, the fluorescence of CDs was first quenched by Cu2+ and was then recovered due to the reduction of Cu2+ in the presence of AA. According to Fig. 2a, 150 μM of Cu2+ was chosen, because the fluorescence of CDs was quenched by approximately 79% and changed slightly with further increasing the concentration of Cu2+. The fluorescence emission spectra of CDs + Cu2+ after adding AA were recorded. The fluorescence intensity at 580 nm of CDs + Cu2++AA gradually increases with the increase of AA (Fig. 3a). The relationship between the fluorescence intensity and the concentration of AA in the range of 100–2800 μM is linear (Fig. 3b). LOD for AA is 60 μM. The fluorescence recovery is rapid (see ESM Fig. S6).

Fig. 3
figure 3

Fluorescence emission spectra of CDs + Cu2+ with the addition of AA (0, 100, 200, 400, 600, 800, 1000, 1200, 1400, 1600, 1800, 2000, 2400, 2600, 2800, 3000, 3500, and 4000 μM from bottom to top) (a). The linear relationship of the fluorescence intensity at 580 nm versus the concentration of AA in the range of 100–2800 μM (b). The selectivity of “off-on” fluorescence strategy towards different analytes (c). F0 and F represent the fluorescence intensities at 580 nm of CDs + Cu2+ and CDs + Cu2++analyte, respectively. The concentration of analyte is 2800 μM. The excitation wavelength is 360 nm. All experiments were repeated three times. Chemical formulas of TA, CA, and GSH (d)

Full size image

The interferences (Na+, K+, Ca2+, Suc, Glu, Fru, Lys, Trp, Gly, Chol, TA, CA, and GSH) were chosen to evaluate the selectivity of “off-on” fluorescence strategy. The results show that AA could recover the fluorescence of CDs + Cu2+ because of the disaggregation of CDs along with the reduction of Cu2+ by AA (Fig. 3c) [20]. Cations (e.g., Na+, K+, Ca2+), carbohydrates (e.g., Suc, Glu, Fru), amino acids (e.g., Lys, Trp), and Chol have ignorable effect on the fluorescence recovery of CDs + Cu2+. Hydroxy acids (e.g., TA, CA) have some interference as they are usually used for the reducing agents. In particular, GSH, a well-known reducing agent, has better the fluorescence recovery ability than AA. This phenomenon has been employed to design CDs + Cu2++GSH “on-off-on” fluorescent probe for detection of GSH [32,33,34]. The chemical formulas of TA, CA, and GSH show that they contain the reducing groups such as carboxyl and thiol (Fig. 3d).

Effect of solution conditions on the fluorescence quenching and recovery

The fluorescence emission spectra of CDs as well as after adding analytes are summarized in Fig. 4a. With the addition of Cu2+, the fluorescence of CDs is quenched (red line). When AA is then added, the fluorescence recovery is observed (blue line). AA has no significant influence on the fluorescence of CDs (magenta line). Cu2+ and AA are mixed first and the mixture is added into CDs; the result shows the fluorescence recovery (olive line). It is reasonable to deduce that the surface carboxyl groups of CDs coordinate with Cu2+, inducing aggregation of CDs, thus causing the fluorescence quenching [21]. In the presence of AA, Cu2+ is reduced to Cu+, decreasing coordination interaction of the COO−–Cu2+ complex and aggregation of CDs [20].

Fig. 4
figure 4

Fluorescence emission spectra of CDs, CDs + Cu2+, CDs + Cu2++AA, CDs + AA, and CDs + (Cu2++AA) (a). The inset shows the photograph of CDs + Cu2+, CDs + Cu2++AA, CDs + AA, and CDs + (Cu2+ + AA) under UV light at 365 nm. Zeta potential of CDs, CDs + Cu2+, and CDs + Cu2++AA at different solutions (b). The effect of pH on the fluorescent intensity of CDs + Cu2+ (c). The effect of pH on the fluorescent intensity of CDs + Cu2++AA (d). All experiments were repeated three times

Full size image

We also characterized zeta potential of CDs, CDs + Cu2+, and CDs + Cu2+ + AA at different solutions such as deionized water, phosphate-buffered saline (PBS), and Tris-HCl buffer solution (Fig. 4b). Anions and amides have significant effect on the surface charge of CDs and coordination interaction of the COO−–Cu2+ complex. The results show that deionized water is the most suitable solution for CD-based “on-off-on” fluorescent probe. In addition, neutral aqueous solution yields better results of the fluorescence quenching (Fig. 4c) and recovery (Fig. 4d) than acidic or alkaline solution.

Confirmation of “on-off-on” fluorescence strategy

To further confirm “on-off-on” fluorescence strategy, we prepared three samples (CDs + Cu2+, CDs + Cu2+ + AA, and Cu2++AA) and collected the residual copper ions (e.g., Cu2+, Cu+) after dialysis and digestion (Fig. 5a). The concentration of copper was determined by ICP-OES and the copper content ratio was calculated with the copper content in feed as 100% (Fig. 5b). As the above deduction, the COO−–Cu2+ complex in CDs + Cu2+ through coordination interaction causes that some copper ions are bound, while copper ions in other samples freely diffuse. The change in zeta potential also confirms aggregation and disaggregation of CDs as depicted in Fig. 5c. In addition, we also determined the residue content of other metal ions after dialysis. We selected Pb2+, Hg2+, Cr3+, and Fe3+ as they can also induce the fluorescence quenching of CDs. The result shows that the residue contents of Pb2+, Hg2+, and Cr3+ are lower than that of Cu2+ (see ESM Fig. S7). However, Fe3+ is detected in high content, which is consistent with the result of the fluorescence quenching of CDs.

Fig. 5
figure 5

Three samples treated by dialysis for determination of copper through ICP-OES (a). The inset table lists the concentration of copper in three samples. The copper content ratio in three samples calculated from ICP-OES data (b). Zeta potential ratio of CDs, CDs + Cu2+, and CDs + Cu2+ + AA (c). All experiments were repeated three times

Full size image

Detection of AA in real samples

Practical application of the probe for detection of AA in real samples (VC tablets, orange juice, and fresh orange) was carried out with VC assay kit for comparison. The samples must be pretreated to reduce the influence of impurities. The florescence intensity at 580 nm of CDs + Cu2+ after adding the pretreated samples was recorded (Fig. 6a). The concentration of AA was then calculated (Fig. 6b). The results show that the data detected using the probe are closer to those calculated from instructions than the data detected using VC assay kit, indicating that the probe presents acceptable precision and accuracy for detection of AA in real samples.

Fig. 6
figure 6

Fluorescence intensity at 580 nm of CDs + Cu2+ with the addition of real samples such as VC tablets, orange juice, and fresh orange containing AA (a). The concentration of AA calculated from instructions, detected using our probe, and detected using VC assay kit (b). All experiments were repeated three times

Full size image

Detection of ACP

Indeed, buffers such as PBS and Tris-HCl are usually used to control solution pH for enzyme catalysis. As shown in Fig. 4b, however, we found that the anions (e.g., PO43−) and amides (e.g., Tris) have significant effect on the surface charge of CDs and coordination interaction of the COO−–Cu2+ complex [35]. Additionally, the solution at a pH of 7.4 results in the best performance of the fluorescence quenching (Fig. 4c) and recovery (Fig. 4d). Therefore, aqueous solution was chosen in the test of detection of ACP. The probe can also detect ACP, a type of enzyme, which catalyzes hydrolysis of AAP to produce AA (Fig. 7a). The fluorescence intensity of CDs + Cu2+ + AAP gradually increases with the increase of ACP (Fig. 7b). A good linear relationship between the fluorescence intensity and the concentration of ACP in the range of 0.5–6 mU mL−1 is obtained (Fig. 7c). LOD for ACP is 0.056 mU. Other enzymes (e.g., Furin, HRP, GOx) were chosen to evaluate the selectivity of the probe. The results show that the probe exhibits high selectivity for detection of ACP (Fig. 7d). In fact, AAP or ACP has no significant influence on the fluorescence of CDs + Cu2+ (Fig. 7e). The co-existence of AAP and ACP produces AA, leading to the fluorescence recovery of CDs.

Fig. 7
figure 7

AA produced by enzymatic hydrolysis of AAP in the presence of ACP (a). Fluorescence emission spectra of CDs + Cu2+ + AAP with the addition of ACP (0, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6. 6.5, 7, and 7.5 mU mL−1 from bottom to top) (b). The concentrations of Cu2+ and AAP are 150 μM and 20 μM, respectively. The linear relationship of the fluorescence intensity at 580 nm versus the concentration of ACP in the range of 0.5–6 mU mL−1 (c). The selectivity of “off-on” fluorescence strategy towards different enzymes (d). F0 and F represent the fluorescence intensities at 580 nm of CDs + Cu2++AAP and CDs + Cu2+ + AAP + enzyme, respectively. The concentration of enzyme is 6 mU mL−1. The excitation wavelength is 360 nm. Fluorescence emission spectra of CDs + Cu2+, CDs + Cu2+ + AAP, CDs + Cu2+ + ACP, and CDs + Cu2+ + AAP + ACP (e). All experiments were repeated three times

Full size image

Conclusions

A CD-based “on-off-on” fluorescent probe is developed for detection of copper(II) ion (Cu2+), ascorbic acid (AA), and acid phosphatase (ACP). CDs synthesized through strong acid oxidation of activated carbon possess yellow fluorescence, remarkable solubility, and good stability in water. The fluorescence of CDs can be quenched by Cu2+ and then recovered by AA on basis of the reduction of Cu2+ to Cu+. This phenomenon is applied for detection of Cu2+ and AA. For real samples such as VC tablets, orange juice, and fresh orange, the probe shows acceptable precision and accuracy for detection of AA, compared with commercial VC assay kit. Besides, the probe can also detect ACP that catalyzes the hydrolysis of AAP to produce AA. The probe, which utilizes the environmentally friendly carbon nanomaterials and “on-off-on” fluorescence strategy, has several distinct advantages such as nontoxicity, low cost, high sensitivity, and acceptable selectivity in comparison with colorimetric method. This work provided a facile method for the multi-component detection as well as AA in real samples.