Highly selective fluorescent chemosensor for detection of Fe3+ based on Fe3O4@ZnO

The combination of fluorescent nanoparticles and specific molecular probes appears to be a promising strategy for developing fluorescent nanoprobes. In this work, L-cysteine (L-Cys) capped Fe3O4@ZnO core-shell nanoparticles were synthesized for the highly selective detection of Fe3+. The proposed nanoprobe shows excellent fluorescent property and high selectivity for Fe3+ due to the binding affinity of L-Cys with Fe3+. The binding of Fe3+ to the nanoprobe induces an apparent decrease of the fluorescence. Thus a highly selective fluorescent chemosensor for Fe3+ was proposed based on Fe3O4@ZnO nanoprobe. The magnetism of the nanoprobe enables the facile separation of bound Fe3+ from the sample solution with an external magnetic field, which effectively reduces the interference of matrix. The detection limit was 3 nmol L−1 with a rapid response time of less than 1 min. The proposed method was applied to detect Fe3+ in both serum and wastewater samples with acceptable performance. All above features indicated that the proposed fluorescent probe as sensing platform held great potential in applications of biological and analytical field.

manipulability and good biocompatibility. Also it has widespread applications in magnetic bioseparation 33 , drug delivery 34 and magnetic resonance imaging 35 . In this work, we develop an L-cysteine capped magnetic Fe 3 O 4 @ZnO nanosensor (Fe 3 O 4 @ZnO@L-Cys) for detection and removal of Fe 3+ (Fig. 1). The results showed that Fe 3 O 4 @ZnO@L-Cys quantificationally detected Fe 3+ with high sensitivity and selectivity under a pH range (pH 4.98-7.39) and could remove Fe 3+ from the water sample. Moreover, the fabricated magnetic fluorescent probe could be removed by external magnetic field, and the potential secondary pollution was avoided.

Experimental
Regents and apparatus. Fe 3 O 4 nanoparticles were purchased from Aladdin Chemical Co., Ltd. Zinc acetate (Zn(Ac) 2 ) was purchased from Tianjin Hongyan Chemical Reagent Factory. Triethanolamine was purchased from Guangcheng Chemical Reagent Co., Ltd. (Tianjin). L-Cys was purchased from Yunxiang Chemical Industry Co., Ltd. Absolute ethyl alcohol was purchased from Fuyu Chemical Reagent Factory. All other reagents used in this study were analytical grade, and ultrapure water was used in the preparation of all solutions.

Preparation of Fe 3 O 4 @ZnO@L-Cys.
The Fe 3 O 4 @ZnO was prepared according to the published procedure 36 . 60 mg of L-Cys was dispersed into 20 mL of ethanol solution by sonication for 20 min in 100 mL conical flask. Then, 10 mg of Fe 3 O 4 @ZnO was added into the conical flask. The flask was wrapped with aluminum foil and vigorous stirring for 6 h. The L-Cys was linked on the surface of Fe 3 O 4 @ZnO by thiol groups of L-Cys 37 . The product was magnetically collected and washed with ultrapure water and ethanol for four times, respectively. The sample of Fe 3 O 4 @ZnO@L-Cys was re-dispersed into 50 mL ethanol solution (Fe 3 O 4 @ZnO@L-Cys stocking solution).

Effect of pH values and ionic strength.
The effect of pH values was studied as follows: 300 μL of Fe 3 O 4 @ ZnO@L-Cys stocking solution was suspended in 2.7 mL of phosphate buffered saline (PBS) (20 mmol L −1 ) aqueous solution in colorimetric cylinder at different pH values (4.98, 5.83, 6.30, 7.02, 7.39, 7.95 and 8.35, respectively). The suspension was laid aside for 5 min and the emission spectra of the suspension were measured. Then, 200 μL of Fe 3+ (2 mmol L −1 ) was added respectively. The suspension was laid aside for another 5 min and the emission spectra of the suspension were measured.
To test the influence of ionic strength on the fluorescence of Fe 3 O 4 @ZnO@L-Cys before and after the addition of Fe 3+ , a series of  of PBS (20 mmol L −1 , pH 7.02) aqueous solution. Then the fluorescence intensity was tested. Subsequently, 300 μL of Fe 3+ was added into the above solution. The fluorescence intensity was tested again every other 30 s for 10 min.
To evaluate the stability of Fe 3 O 4 @ZnO@L-Cys, the emission spectra was measured every other 10 d.

Application of Fe 3 O 4 @ZnO@L-Cys in real samples.
Fresh human blood sample was obtained from the local hospital and pretreated according to the early published procedures 38,39 . In addition, the wastewater sample was collected from the local lake. The amount of Fe 3+ was estimated using a standard addition method. For recovery studies, known concentrations of Fe 3+ solution were added to the samples and the total iron concentrations were then determined at the same condition.  Fig. S1, the main absorption band at approximately 380 nm of the Fe 3 O 4 @ZnO had a minor enhancement in the presence of 100 μmol L −1 Fe 3+ without an obvious change of the peak shape. The slight changes of absorption spectra suggested that the quencher-Fe 3+ did not affect the structure of the nanoparticles. The absorption band of Fe 3 O 4 @ZnO is usually very sensitive to the presence of adsorbed substances 41,42 . However, the presence of Fe 3+ only generated slight changes in absorption spectra of the Fe 3 O 4 @ZnO@L-Cys. Thus, we may rule out the possibility of direct binding of Fe 3+ to the Fe 3 O 4 @ZnO from the absorption spectra point of view. It could be clearly seen that the fluorescence intensity of the Fe 3 O 4 @ZnO@L-Cys was quenched dramatically with increase of Fe 3+ . So we speculated the added Fe 3+ should interact with the L-Cys. Fe 3+ ion is a well-known efficient fluorescence quencher due to its paramagnetic properties via electron or energy transfer. And L-cysteine, a common amino acid, possesses both amino and carboxyl function groups. It could be used to recognize the Fe 3+ because the Fe 3+ was known to be preferentially binding with nitrogen atom of imino group and oxygen atom of carbonyl group 20,43 . Thus we inferred the nitrogen atom of imino group and oxygen atom of carbonyl group in the L-Cys molecule might donor electrons to the Fe 3+ , as described in Fig. 1. In the same time, other interaction sites of six-coordinated Fe 3+ may be occupied by the other Fe 3 O 4 @ZnO@L-Cys. Thus the coordination interaction occurred and induced intra-particles cross links which resulted in the fluorescence quenching 44 .

Effect of pH values and ionic strength. Usually, the pH values of probes' solution have tremendous influ-
ence on the detection of target analytes. So, the Fe 3+ -sensing ability of Fe 3 O 4 @ZnO@L-Cys at different pH was also investigated. The result showed that Fe 3 O 4 @ZnO@L-Cys was stable within a pH range from 4.98 to 7.39, and its response ability toward Fe 3+ was stable within a pH range from 4.98 to 7.39 (Fig. 3a). Therefore, we choose the neutral aqueous solution (pH 7.02) as the analytical condition for the detection and removal of Fe 3+ . The ionic strength was also a parameter for the detection of target analytes. The effect of ionic strength was presented in the Fig. 3b. As can be seen from the figure, the fluorescence intensity at 337 nm was not changed obviously before (Fig. 3b, (A)) and after (Fig. 3b, (B)) the addition of Fe 3+ with the increasing concentration of NaCl solution, indicating the stability of the analytical platform at different ionic strength.
Time course of the Fe 3 O 4 @ZnO@L-Cys toward Fe 3+ . Fig. 3c  fluorescence intensity decreased minimum and then achieved a platform. Therefore, the fluorescent probe could realize the rapid analysis of Fe 3+ in the samples.
Compared with other reports (Table S1), the method we proposed can realize the real-time analysis of trace amount of Fe 3+ with sensitivity and celerity. This may be attributed to the amount of amino and carboxyl groups on the surface of Fe 3 O 4 @ZnO.

Selectivity and stability.
High selectivity is a matter of necessity for an excellent sensor. Therefore, the selectivity of Fe 3 O 4 @ZnO@L-Cys for Fe 3+ (200 μmol L −1 ) was investigated by screening its response to relevant analytes under the same condition. The results showed that other metal ions could enhance the fluorescence intensity of Fe 3 O 4 @ZnO@L-Cys, and the Fe 3+ could decrease the fluorescence intensity of Fe 3 O 4 @ZnO@L-Cys (Fig. 5a). To further demonstrate the ability to recognize Fe 3+ in the presence of other competitive mental ions (Al 3+ , Pb 2+ , Cr 3+ , Cd 2+ , Mg 2+ , Mn 2+ , Cu 2+ and Co 2+ ), the anti-interferential capability of the nanoparticle was also studied. When one equivalent of Fe 3+ was added into the solution of the nanoparticle in the presence of four equivalents of other metal ions, higher concentration of the other metal ions did not affect the selectivity of Fe 3 O 4 @ZnO@L-Cys toward Fe 3+ (Fig. 5b), except Cu 2+ ion. This was because L-Cys molecule contained amino, carboxylic and thiol groups and many researches reported that the Cu 2+ could bind with L-Cys 41,45,46 . Therefore, the Cu 2+ showed an influence on the detection of Fe 3+ .
The stability of Fe 3 O 4 @ZnO@L-Cys was also examined. The fluorescence intensity of Fe 3 O 4 @ZnO@L-Cys at 337 nm was tested. After 20 d, the fluorescence intensity decreased to about 98% of its initial value, indicating the stability of Fe 3 O 4 @ZnO@L-Cys. Removal of Fe 3+ from the standard solution. To investigate the removal ability of Fe 3 O 4 @ZnO@L-Cys, Fe 3+ standard solution (3 mL, 400 μmol L −1 ) was chosen as testing solution. As indicated by Fig. S2A, the solution presented light yellow before the Fe 3 O 4 -based fluorescent nanoparticle was added into the solution. Then, 300 μL Fe 3 O 4 @ZnO@L-Cys stocking solution was added. A magnet was used to separate the Fe 3+ -bound nanosensors from aqueous solution after half an hour, the solution became clear and colorless (Fig. S2B), which indicated the Fe 3 O 4 @ZnO@L-Cys could be used for the extraction of Fe 3+ from solution. Hence, the maximum adsorption amount of Fe3O4@ZnO@L-Cys toward Fe 3+ was determined. And the result obtained by calculation is 192.64 mg/g, which can be seen clearly in Fig. S4.   Table S2. The determinated iron contents were at reasonable range in according to the literature values detected with other approaches, such as the methods of fluorescent gold nanoclusters 38 , atomic absorption spectrometry 47 and inductively coupled plasma mass spectrometry 48 . The recoveries of the known amount Fe 3+ in serum samples were 92.6-108.4%, while in wastewater samples were 89.6-113.0%. The results demonstrated reliability of Fe 3 O 4 @ZnO@L-Cys for detecting iron contents in real samples.

Conclusion
In summary, a really facile detection method based on fluorescent probe Fe 3 O 4 @ZnO@L-Cys has been developed, which allowed the highly sensitive and selective determination of Fe 3+ . It is the first time to apply Fe 3 O 4 @ZnO based sensing platform for the analysis of iron contents. And the magnetic nanoparticle Fe 3 O 4 @ZnO could be prepared easily and environmental friendly. The fluorescence intensity of fluorescent probe Fe 3 O 4 @ZnO@L-Cys was quenched significantly in the presence of Fe 3+ within 1 min. Other common metal ions at four times concentrations of Fe 3+ did not cause interference. Furthermore, the proposed fluorescent probe could be applied to detect iron contents in real samples and extract the Fe 3+ from the solution which containing high concentration of Fe 3+ with the aid of external magnetic field.