Facile Preparation of Fe3O4/C Nanocomposite and Its Application for Cost-Effective and Sensitive Detection of Tryptophan

In this study, we reported facile synthesis of Fe3O4/C composite and its application for the cost-effective and sensitive determination of tryptophan (Trp) in human serum samples. Fe3O4/C composites were prepared by a simple one-pot hydrothermal method followed by a mild calcination procedure, using FeCl3∙6H2O as Fe3O4 precursor, and glucose as reducing agent and carbon source simultaneously. The Fe3O4/C composite modified glassy carbon electrode (Fe3O4/C/GCE) was prepared by drop-casting method. The microstructure and morphology of Fe3O4/C composite was characterized by powder X-ray diffraction (XRD) and scanning electron microscopy (SEM), respectively. Due to large specific surface area and synergistic effect from Fe3O4 nanoparticles and carbon coating, Fe3O4/C composite showed excellent electrocatalytic activity toward the oxidation of Trp. As a result, the proposed Fe3O4/C/GCE displayed superior analytical performances toward Trp determination, with two wide detection ranges (1.0–80 μM and 80–800 μM) and a low detection limit (0.26 μM, S/N = 3). Moreover, successful detection of Trp in human serum samples further validate the practicability of the proposed sensor.


Introduction
Tryptophan (Trp), one of eight essential amino acids in human body, plays indispensable roles in maintaining human growth, metabolism, and positive nitrogen balance [1]. Trp is also a vital precursor for melatonin and serotonin. It has been reported that metabolic disorders or abnormal levels of Trp may cause several serious diseases including Alzheimer's diseases, delusions, and hallucinations [2,3]. Even worse, highly oxidized products of Trp can even lead to some horrible cancers [3]. Trp has become a most promising biomarkers for early diagnosis of some cancers, such as gastric cancer and breast carcinoma. For instance, early diagnosis of gastric cancer can be realized by detecting Trp level in gastric fluid [4]. Hence, it is particularly urgent to develop rapid and reliable techniques to detect Trp in physiological samples.
To date, various advanced nanostructures have been developed as sensitive sensing materials for the construction of electrochemical sensors [25][26][27][28][29]. It is well-known that transition metal oxide nanomaterials have obvious advantages over noble metal nanoparticles in terms of the cost and abundance [30][31][32][33]. Nanostructured transition metal oxides such as TiO 2 [15,16], SnO 2 [9], CuO [11], and Ta 2 O 5 [34] have been utilized for the detection of Trp. Despite the wide variety of transition metal oxides available, Fe 3 O 4 nanoparticles (Fe 3 O 4 NPs) are highly encouraged as sensing materials because of their low fabrication cost, environmental friendliness, and excellent electrochemical properties [35]. Moreover, the good biocompatible nature of Fe 3 O 4 NPs endows them attractive candidates for various biosensors [36]. For example, highly ordered mesoporous Fe 3 O 4 NPs with high surface area showed high catalytic activity toward dopamine, with wide linear detection range of 2-600 nM and low detection limit of 0.8 nM [37]. Magnetic Fe 3 O 4 NPs modified electrode showed excellent catalytic activity toward the detection of Sundan I, with linear detection range of 1-20 µM and low LOD of 0.001 µM [38]. However, the poor electrical conductivity of Fe 3 O 4 NPs has restricted their applications on electrochemical sensors. To address this issue, coupling of Fe 3 O 4 NPs with highly conductive carbonaceous materials is very effective to increase electrical conductivity, and thus, enhance electrochemical sensing properties [39]. More recently, coupling of Fe 3 O 4 with carbon nanofibers and multiwalled carbon nanotubes has been proven to be very feasible for improving the electrical conductivity [40,41]. In our recent studies, compositing Fe 3 O 4 with graphene nanosheets showed low charge resistant (R ct ) and superior electrochemical sensing performances toward the sensitive detection of bioactive molecules, such as dopamine and rutin [42,43]. However, the tedious route and high cost for the synthesis of carbonaceous materials (i.e., carbon nanotubes, carbon nanofibers, graphene) will undoubtedly increase the overall production cost of the sensor [44,45], which adversely affects their commercial applications. Therefore, compositing with carbonaceous materials that possess merits of facile preparation and high electrical conductivity not only significantly improves the sensitivity, but also greatly decreases the fabrication cost.
Herein, a facile one-pot route followed by a mild calcination procedure has been proposed to prepare Fe 3 O 4 /C composite for the sensitive detection of Trp, using FeCl 3 ·6H 2 O as Fe 3 O 4 precursor, and glucose as reducing agent and carbon source simultaneously. Fe 3 O 4 /C composites not only retain unique properties from individual component, but yield synergistic enhancement effect toward the oxidation of Trp. As a result, the electrochemical response of Trp were boosted greatly. The voltammetric parameters, such as solution pH, accumulation potential, and accumulation time, were systematically optimized. At last, the proposed sensor was successfully applied to detect Trp from human serum samples with good recovery.

Synthesis of Fe 3 O 4 /C Composites
Fe 3 O 4 /C composite material was synthesized via a facile one-pot hydrothermal method. Typically, 3.6 g of glucose, 0.5 g of FeCl 3 ·6H 2 O, and 11.6 g of urea were dissolved in 80 mL of ultrapure water under stirring for 10 min. Then, the mixture solution was transferred to a 100 mL Teflon-lined stainless-steel autoclave and reacted at 180 • C for 14 h. After it was cooled down to room temperature, the product was separated by external magnetic field. Then, the product was washed by ultrapure water/ethanol for several times and dried in vacuum oven for 12 h. Finally, the Fe 3 O 4 /C was obtained after a mild calcination at 600 • C for 4 h under Ar atmosphere protection. For comparison, the pure Fe 3 O 4 NPs were prepared by the same method in absence of glucose.

Fabrication of Fe 3 O 4 /C Modified Electrode
At first, the glassy carbon electrodes (GCEs) were polished by α-Al 2 O 3 slurry with different fine sizes (0.3 µm and 0.05 µm). Then, they were alternately washed by ethyl alcohol and ultrapure water under ultrasonication for 2~3 times (each for 1 min), respectively. Five microliters of Fe 3 O 4 /C dispersion (1 mg/mL) was drop-casted onto the surface of polished GCE, then dried under an infrared lamp to obtain Fe 3 O 4 /C composites modified GCE (Fe 3 O 4 /C/GCE). For comparison, pure Fe 3 O 4 NPs modified GCE (Fe 3 O 4 /GCE) was also prepared by the similar method.

Electrochemical Experiments
The electrochemical behaviors of as-prepared Fe 3 O 4 /C/GCE were investigated in a freshly prepared 0.1 M PBS containing 10 µM Trp by cyclic voltammetry (CV). The quantitative analysis of Trp were performed by second derivative linear scan voltammetry (SDLSV). CV and SDLSV were recorded on an electrochemical workstation (CHI 760E, Shanghai Chenhua Inc., Shanghai, China) and a Polarographic Analyzer (JP-303E, Chengdu Instrument Company, Chengdu, China), respectively. Both CVs and SDLSVs were measured using a conventional three-electrode assemble, in which bare or modified GCEs worked as working electrode, while platinum wire electrode and saturated calomel electrode (SCE) acted as counter electrode and reference electrode, respectively. Before all the electrochemical measurements for Trp, a suitable accumulation was performed under stirring at 500 rpm to increase the electrochemical responses. After a 5 s rest, both the CVs and SDLSVs were measured at a scanning rate of 0.1 V/s. The potential was scanned from −0.2 to 1.2 V for both CV and SDLSV.

Detection of Trp in Human Serum
Under the optimal voltammetric conditions, Trp in human serum was detected using SDLSV technique. Typically, 1 mL human serum sample was diluted to 10 mL with 0.1 M PBS (pH 6.5). Standard addition method was further used to validate the reliability of the proposed sensor. . The corresponding peaks of magnetite in the Fe 3 O 4 /C sample are also apparent, indicating that a good crystallization is observed. Moreover, the crystallization is not influenced by carbon coating. Besides, a wide peak located at 20 •~3 0 • is observed, which could be ascribed to diffraction peak of amorphous carbon layer. These results confirm that the Fe 3 O 4 /C nanocomposite was synthesized successfully. Under the optimal voltammetric conditions, Trp in human serum was detected using SDLSV technique. Typically, 1 mL human serum sample was diluted to 10 mL with 0.1 M PBS (pH 6.5). Standard addition method was further used to validate the reliability of the proposed sensor.

Microstructure and Morphology Characterization of Fe3O4/C Composites
The microstructures of pure Fe3O4 NPs and Fe3O4/C composite were characterized by power Xray diffraction (XRD) technique. As presented in Figure 1, curve a and curve b are the XRD patterns of pure Fe3O4 NPs and Fe3O4/C composite, respectively. The emerged various sharp peaks indicating that these as-prepared samples show good crystallization performance. Moreover, the diffraction peaks located at 30.063°, 35.451°, 43.037°, 53.546°, 57.166°, and 62.726° are corresponding to crystal facets of (220), (311), (400), (422), (511), and (440), respectively. This result is in good accordance with standard PDF card of magnetite (JSPDS 01-1111, α = 8.374Å). The corresponding peaks of magnetite in the Fe3O4/C sample are also apparent, indicating that a good crystallization is observed. Moreover, the crystallization is not influenced by carbon coating. Besides, a wide peak located at 20°~30° is observed, which could be ascribed to diffraction peak of amorphous carbon layer. These results confirm that the Fe3O4/C nanocomposite was synthesized successfully. The surface morphologies of as-prepared pure Fe3O4 NPs and Fe3O4/C nanocomposites were further characterized by scanning electron microscopy (SEM). As presented in Figure 2a,b, the irregular shapes of Fe3O4 NPs are observed with excellent dispersity, and the size of these Fe3O4 NPs is roughly estimated as 200~300 nm. However, in the Fe3O4/C composite sample, the particle size is larger than that of pure Fe3O4 NPs (Figure 2c,d). Because of the carbon coating shell, the nanoparticles are aggregated with each other. Interestingly, the contrast of carbon shell and Fe3O4 core is manifested in Figure 2b, because the electron bombardment under high voltage propels the electron transmission. The light spots could be considered as Fe3O4 nanoparticles because of heavy atom Fe, and other dark parts are identified as carbon shell. It is a side evidence that the core-shell Fe3O4/C composite nanoparticles are formed successfully. The surface morphologies of as-prepared pure Fe 3 O 4 NPs and Fe 3 O 4 /C nanocomposites were further characterized by scanning electron microscopy (SEM). As presented in Figure 2a,b, the irregular shapes of Fe 3 O 4 NPs are observed with excellent dispersity, and the size of these Fe 3 O 4 NPs is roughly estimated as 200~300 nm. However, in the Fe 3 O 4 /C composite sample, the particle size is larger than that of pure Fe 3 O 4 NPs (Figure 2c,d). Because of the carbon coating shell, the nanoparticles are aggregated with each other. Interestingly, the contrast of carbon shell and Fe 3 O 4 core is manifested in Figure 2b, because the electron bombardment under high voltage propels the electron transmission. The light spots could be considered as Fe 3 O 4 nanoparticles because of heavy atom Fe, and other dark parts are identified as carbon shell. It is a side evidence that the core-shell Fe 3 O 4 /C composite nanoparticles are formed successfully.

Electrochemical Active Area
The CVs of bare GCE, Fe3O4/GCE and Fe3O4/C/GCE were recorded in 0.1 M PBS containing 0.5 mM K3Fe(CN)6 ( Figure 3). A pair of quasi-reversible redox peaks (ipa = 13.69 μA; ipc = 14.71 μA) are observed at the bare GCE, with peak-peak separation (∆Ep) of 0.153 V. At the Fe3O4/GCE, the redox peak currents (ipa = 12.03 μA; ipc = 12.61 μA) are inferior to those of bare GCE, probably because of the poor electrical conductivity of Fe3O4 NPs. Moreover, the peak-peak separation (∆Ep) decrease to 0.145 V, suggesting Fe3O4 NPs effectively reduced the active polarization. Among these three electrodes, Fe3O4/C/GCE shows the largest response peak currents (ipa = 19.04 μA; ipc = 25.29 μA) with lowest peak-peak potential separation (∆Ep = 0.079 V), due to the synergistic effect from high electrocatalytic activity of Fe3O4 NPs and excellent electrical conductivity of carbon shells. The cathodic peak currents (ipc) of bare GCE, Fe3O4/GCE, and Fe3O4/C/GCE are 14.71 μA, 12.61 μA, and 25.29 μA, respectively. Referring to the Randles-Sevcik equation [46][47][48]: where ipc represents the cathodic peak currents of K3[Fe(CN)6], n represents the transferred electron number during the redox reaction of K3[Fe(CN)6], D is the diffusion coefficient of K3[Fe(CN)6] (7.6 × 10 −6 cm 2 s −1 [49]), v is the scanning rate (V s −1 ), A is the electrochemical active area (cm 2 ), and C is the concentration of K3[Fe(CN)6] (mol cm −3 ). The electrochemical active areas of bare GCE, Fe3O4/GCE and Fe3O4/C/GCE are calculated as 0.01255 cm 2 , 0.01075 cm 2 , and 0.02209 cm 2 , respectively. The electrochemical active area of Fe3O4/C/GCE is about two-fold that of bare GCE, indicting that the Fe3O4/C composite modification could enhance the effective electroactive surface area, which can facilate the adsorption of target analytes on the electrode surface and offer more catalytic active sites, and thus accelerate the redox process of target analytes.

Electrochemical Active Area
The where i pc represents the cathodic peak currents of K 3 [Fe(CN) 6 ], n represents the transferred electron number during the redox reaction of

Electrochemical Responses of Trp on Various Modified Electrodes
The electrochemical responses of 10 μM Trp on bare GCE, Fe3O4/GCE, and Fe3O4/C/GCE electrodes were investigated by CV (Figure 4a). Clearly, cathodic peaks are not observed at all electrodes, implying the oxidation of Trp is an irreversible process. At bare GCE, a relatively poor anodic peak (ipa = 4.603 μA) is observed at 0.929 V. After the modification of GCE with Fe3O4 NPs, the ipa of Trp increases to 7.414 μA, suggesting Fe3O4 NPs have high electrocatalytic activity toward Trp oxidation. However, the anodic peak shifts positively to 0.963 V, due to the poor electrical conductivity of Fe3O4 NPs. When decorated with Fe3O4/C nanocomposites, a sharp anodic peak appears with highest response peak current (13.13 μA) and lowest working potential (0.750 V). The ipa of 10 μM Trp at Fe3O4/C/GCE is about two times higher than that at bare GCE. The enhanced anodic peak current and decreased anodic peak potential is mainly due to the synergistic effect of Fe3O4 NPs and carbon coating. Specifically, Fe3O4 NPs offer favorable catalytic active sites for the Trp oxidation; The carbon coating improves the electrical conductivity and decreases the charge transfer resistance (Rct). Besides, the large electrochemical active area of Fe3O4/C composite also contributed to the enhancement on the ipa of Trp. The SDLSV responses of 10 μM Trp at various electrodes are shown in Figure 4b. On bare GCE, a weak anodic peak (ipa = 9.852 μA) occurs at 0.852 V. When modified with Fe3O4 NPs, the anodic peak shifts to 0.901 V and the anodic peak current increases to 16.42 μA. After the modification of GCE with Fe3O4/C composites, a well-shaped anodic peak was observed with largest anodic peak current of 46.55 μA and lowest anodic peak potential of 0.746. The SDLSV results are highly consistent with the CV results. Obviously, SDLSV is more sensitive than CV for Trp detection. So, SDLSV was performed for the detection of Trp.

Electrochemical Responses of Trp on Various Modified Electrodes
The electrochemical responses of 10 µM Trp on bare GCE, Fe 3 O 4 /GCE, and Fe 3 O 4 /C/GCE electrodes were investigated by CV (Figure 4a). Clearly, cathodic peaks are not observed at all electrodes, implying the oxidation of Trp is an irreversible process. At bare GCE, a relatively poor anodic peak (i pa = 4.603 µA) is observed at 0.929 V. After the modification of GCE with Fe 3 O 4 NPs, the i pa of Trp increases to 7.414 µA, suggesting Fe 3 O 4 NPs have high electrocatalytic activity toward Trp oxidation. However, the anodic peak shifts positively to 0.963 V, due to the poor electrical conductivity of Fe 3 O 4 NPs. When decorated with Fe 3 O 4 /C nanocomposites, a sharp anodic peak appears with highest response peak current (13.13 µA) and lowest working potential (0.750 V). The i pa of 10 µM Trp at Fe 3 O 4 /C/GCE is about two times higher than that at bare GCE. The enhanced anodic peak current and decreased anodic peak potential is mainly due to the synergistic effect of Fe 3 O 4 NPs and carbon coating. Specifically, Fe 3 O 4 NPs offer favorable catalytic active sites for the Trp oxidation; The carbon coating improves the electrical conductivity and decreases the charge transfer resistance (R ct ). Besides, the large electrochemical active area of Fe 3 O 4 /C composite also contributed to the enhancement on the i pa of Trp. The SDLSV responses of 10 µM Trp at various electrodes are shown in Figure 4b. On bare GCE, a weak anodic peak (i pa = 9.852 µA) occurs at 0.852 V. When modified with Fe 3 O 4 NPs, the anodic peak shifts to 0.901 V and the anodic peak current increases to 16.42 µA. After the modification of GCE with Fe 3 O 4 /C composites, a well-shaped anodic peak was observed with largest anodic peak current of 46.55 µA and lowest anodic peak potential of 0.746. The SDLSV results are highly consistent with the CV results. Obviously, SDLSV is more sensitive than CV for Trp detection. So, SDLSV was performed for the detection of Trp.

Effect of Solution pH
In order to investigate the influence of solution pH, the CVs of 10 μM Trp were measured in various pH PBS (Figure 5a). As plotted in Figure 5b, the ipa increases with pH varying from 5.5 to 6.5, then gradually declines with pH further increasing. So, pH 6.5 was selected for the subsequent measurements. Moreover, the anodic peak potential (Epa) is highly linear to solution pH, and the

Effect of Scanning Rate
The CV curves of 10 μM Trp were recorded at different scanning rates (0.03~0.30 V s −1 ) in 0.1 M PBS (pH 6.5), aiming to reveal the electrooxidation mechanism of Trp on Fe3O4/C/GCE. As can be seen from Figure 6a, the ipa of Trp increases with the speeding of scanning rates, but the background currents also increase with the increase of the scanning rates. More importantly, a good linear

Effect of Solution pH
In order to investigate the influence of solution pH, the CVs of 10 µM Trp were measured in various pH PBS (Figure 5a). As plotted in Figure 5b, the i pa increases with pH varying from 5.5 to 6.5, then gradually declines with pH further increasing. So, pH 6.5 was selected for the subsequent measurements. Moreover, the anodic peak potential (E pa ) is highly linear to solution pH, and the corresponding linear regression equation is E pa (V) = −0.061pH + 1.234 with correlation coefficient (R 2 ) of 0.994. The slope (−61mV/pH) is very close to theoretical value from Nernst equation (−59 mV/pH), suggesting the equal number of protons (H + ) and electrons (e − ) are involved in the electrochemical oxidation of Trp [50].

Effect of Solution pH
In order to investigate the influence of solution pH, the CVs of 10 μM Trp were measured in various pH PBS (Figure 5a). As plotted in Figure 5b, the ipa increases with pH varying from 5.5 to 6.5, then gradually declines with pH further increasing. So, pH 6.5 was selected for the subsequent measurements. Moreover, the anodic peak potential (Epa) is highly linear to solution pH, and the corresponding linear regression equation is Epa(V) = −0.061pH + 1.234 with correlation coefficient (R 2 ) of 0.994. The slope (−61mV/pH) is very close to theoretical value from Nernst equation (−59 mV/pH), suggesting the equal number of protons (H + ) and electrons (e − ) are involved in the electrochemical oxidation of Trp [50].

Effect of Scanning Rate
The CV curves of 10 μM Trp were recorded at different scanning rates (0.03~0.30 V s −1 ) in 0.1 M PBS (pH 6.5), aiming to reveal the electrooxidation mechanism of Trp on Fe3O4/C/GCE. As can be seen from Figure 6a, the ipa of Trp increases with the speeding of scanning rates, but the background currents also increase with the increase of the scanning rates. More importantly, a good linear

Effect of Scanning Rate
The CV curves of 10 µM Trp were recorded at different scanning rates (0.03~0.30 V s −1 ) in 0.1 M PBS (pH 6.5), aiming to reveal the electrooxidation mechanism of Trp on Fe 3 O 4 /C/GCE. As can be seen from Figure 6a, the i pa of Trp increases with the speeding of scanning rates, but the background currents also increase with the increase of the scanning rates. More importantly, a good linear relationship between anodic peak currents (i pa ) and scanning rates (lnv) is found in Figure 6b. The linear equation can be expressed as i pa (µA) = 6.479v (V/s) + 2.582 (R 2 = 0.982). It demonstrates the electrooxidation of Trp on Fe 3 O 4 /C/GCE is an adsorption-controlled process [30,31,33]. As shown in Figure 6c, the anodic peaks are shifted toward the positive potential direct. The anodic peak potentials linearly increase with Napierian logarithm of scanning rates (lnv), and the linear equation can be expressed as E pa (V) = 0.03593lnv + 0.9500 (R 2 = 0.990). As for an adsorption-controlled and totally irreversible electrode process, it should follow the Lavrion's theory as follows [51]: where E p denotes the peak potential, E 0 denotes the formal potential (V), R denotes the gas constant (8.314 J K −1 mol −1 ), T denotes the temperature (298.15 K), α denotes the charge transfer coefficient, n represents the electron transferred number, F is the Faraday constant (96, 485 C mol −1 ), k 0 represents standard heterogeneous transfer rate (s −1 ), and v is the scanning rate (V/s), respectively. Referring to Laviron's equation, the slope of the linear equation is equivalent to RT/αnF (where α is assumed to be 0.5 for a totally irreversible electrode process [52]). Hence, n is calculated to be around 2. The dependence of solution pH on the anodic peak potential (mentioned in Section 3.4.1) confirms that the equal number of protons (H + ) and electrons (e − ) participate in the electrochemical oxidation of Trp. So, the electrooxidation of Trp at Fe 3 O 4 /C/GCE is a two-electrons and two-protons process (2e − , 2H + ). The electrochemical mechanism of Trp oxidation is described in Scheme 1, which is in good agreement with previous reports [13,53,54].
Biomolecules 2019, 9, x FOR PEER REVIEW 8 of 15 relationship between anodic peak currents (ipa) and scanning rates (lnv) is found in Figure 6b. The linear equation can be expressed as ipa (μA) = 6.479v (V/s) + 2.582 (R 2 = 0.982). It demonstrates the electrooxidation of Trp on Fe3O4/C/GCE is an adsorption-controlled process [30,31,33]. As shown in Figure 6c, the anodic peaks are shifted toward the positive potential direct. The anodic peak potentials linearly increase with Napierian logarithm of scanning rates (lnv), and the linear equation can be expressed as Epa(V) = 0.03593lnv + 0.9500 (R 2 = 0.990). As for an adsorption-controlled and totally irreversible electrode process, it should follow the Lavrion's theory as follows [51]: where Ep denotes the peak potential, E 0 denotes the formal potential (V), R denotes the gas constant (8.314 J K −1 mol −1 ), T denotes the temperature (298.15 K), α denotes the charge transfer coefficient, n represents the electron transferred number, F is the Faraday constant (96, 485 C mol −1 ), k 0 represents standard heterogeneous transfer rate (s −1 ), and v is the scanning rate (V/s), respectively. Referring to Laviron's equation, the slope of the linear equation is equivalent to RT/αnF (where α is assumed to be 0.5 for a totally irreversible electrode process [52]). Hence, n is calculated to be around 2. The dependence of solution pH on the anodic peak potential (mentioned in Section 3.4.1) confirms that the equal number of protons (H + ) and electrons (e − ) participate in the electrochemical oxidation of Trp. So, the electrooxidation of Trp at Fe3O4/C/GCE is a two-electrons and two-protons process (2e − , 2H + ). The electrochemical mechanism of Trp oxidation is described in Scheme 1, which is in good agreement with previous reports [13,53,54].

Effect of Accumulation Parameters
Accumulation is very feasible to improve the sensitivity of the proposed sensor, since the electrochemical oxidation of Trp on the Fe3O4/C/GCE is an adsorption-controlled process. The Trp was accumulated on the surface of Fe3O4/C/GCE at various potentials for 240 s, then their ipa of were measured by SDLSV. As presented in Figure 7a, the maximum ipa is achieved at 0.2 V. Hence, 0.2 V was recommended as the optimal accumulation potential. Moreover, the influence of accumulation time on the ipa were also investigated at fixed accumulation potential of 0.2 V (Figure 7b). As the relationship between anodic peak currents (ipa) and scanning rates (lnv) is found in Figure 6b. The linear equation can be expressed as ipa (μA) = 6.479v (V/s) + 2.582 (R 2 = 0.982). It demonstrates the electrooxidation of Trp on Fe3O4/C/GCE is an adsorption-controlled process [30,31,33]. As shown in Figure 6c, the anodic peaks are shifted toward the positive potential direct. The anodic peak potentials linearly increase with Napierian logarithm of scanning rates (lnv), and the linear equation can be expressed as Epa(V) = 0.03593lnv + 0.9500 (R 2 = 0.990). As for an adsorption-controlled and totally irreversible electrode process, it should follow the Lavrion's theory as follows [51]: where Ep denotes the peak potential, E 0 denotes the formal potential (V), R denotes the gas constant (8.314 J K −1 mol −1 ), T denotes the temperature (298.15 K), α denotes the charge transfer coefficient, n represents the electron transferred number, F is the Faraday constant (96, 485 C mol −1 ), k 0 represents standard heterogeneous transfer rate (s −1 ), and v is the scanning rate (V/s), respectively. Referring to Laviron's equation, the slope of the linear equation is equivalent to RT/αnF (where α is assumed to be 0.5 for a totally irreversible electrode process [52]). Hence, n is calculated to be around 2. The dependence of solution pH on the anodic peak potential (mentioned in Section 3.4.1) confirms that the equal number of protons (H + ) and electrons (e − ) participate in the electrochemical oxidation of Trp. So, the electrooxidation of Trp at Fe3O4/C/GCE is a two-electrons and two-protons process (2e − , 2H + ). The electrochemical mechanism of Trp oxidation is described in Scheme 1, which is in good agreement with previous reports [13,53,54].

Effect of Accumulation Parameters
Accumulation is very feasible to improve the sensitivity of the proposed sensor, since the electrochemical oxidation of Trp on the Fe3O4/C/GCE is an adsorption-controlled process. The Trp was accumulated on the surface of Fe3O4/C/GCE at various potentials for 240 s, then their ipa of were measured by SDLSV. As presented in Figure 7a, the maximum ipa is achieved at 0.2 V. Hence, 0.2 V was recommended as the optimal accumulation potential. Moreover, the influence of accumulation time on the ipa were also investigated at fixed accumulation potential of 0.2 V (Figure 7b). As the Scheme 1. Schematic diagram of the electrochemical mechanism of Trp oxidation.

Effect of Accumulation Parameters
Accumulation is very feasible to improve the sensitivity of the proposed sensor, since the electrochemical oxidation of Trp on the Fe 3 O 4 /C/GCE is an adsorption-controlled process. The Trp was accumulated on the surface of Fe 3 O 4 /C/GCE at various potentials for 240 s, then their i pa of were measured by SDLSV. As presented in Figure 7a, the maximum i pa is achieved at 0.2 V. Hence, 0.2 V was recommended as the optimal accumulation potential. Moreover, the influence of accumulation time on the i pa were also investigated at fixed accumulation potential of 0.2 V (Figure 7b). As the accumulation time prolongs, the i pa gradually increase during the first 210 s, then remain stable with further increasing accumulation time, suggesting the saturation adsorption of Trp. Hence, the accumulation was performed at 0.2 V for 210 s in the following measurements.
Biomolecules 2019, 9, x FOR PEER REVIEW 9 of 15 accumulation time prolongs, the ipa gradually increase during the first 210 s, then remain stable with further increasing accumulation time, suggesting the saturation adsorption of Trp. Hence, the accumulation was performed at 0.2 V for 210 s in the following measurements.  The superb sensing performance is mainly due to synergistic effect from high catalytic activity of Fe3O4 NPs and excellent conductive carbon coating. A comparison of analytical performances between our proposed sensor and the existing electrochemical sensors is summarized in Table 1. To our surprise, the sensing properties of our proposed sensor are at least comparable to or even superior than previous reports [9][10][11][12][13][14][15][16]21,23,24,34]. Moreover, Fe3O4/C/GCE has outstanding advantages including facile preparation and low costs, which is very important for the practical applications.  Table 1. To our surprise, the sensing properties of our proposed sensor are at least comparable to or even superior than previous reports [9][10][11][12][13][14][15][16]21,23,24,34]. Moreover, Fe 3 O 4 /C/GCE has outstanding advantages including facile preparation and low costs, which is very important for the practical applications.

Repeatability, Reproducibility and Stability Assays
The repeatability and reproducibility assays were also carried out to verify the practicability. The repeatability was examined by ten successive measurements of the ipa of 10 μM Trp (Figure 9). The relative standard deviation (RSD) of 5.65% (n = 10), demonstrating good repeatability. The reproducibility was also checked by paralleled measurements of 10 μM Trp, using five Fe3O4/C/GCEs

Repeatability, Reproducibility and Stability Assays
The repeatability and reproducibility assays were also carried out to verify the practicability. The repeatability was examined by ten successive measurements of the i pa of 10 µM Trp (Figure 9). The relative standard deviation (RSD) of 5.65% (n = 10), demonstrating good repeatability. The reproducibility was also checked by paralleled measurements of 10 µM Trp, using five Fe 3 O 4 /C/GCEs fabricated by the same procedures. The RSD (4.78%, n = 5) is accepted, suggesting highly reproducible for electrode preparation. Finally, the stability was assessed by monitoring the anodic peak currents of 10 µM Trp at a same Fe 3 O 4 /C/GCE during a week. When not in use, Fe 3 O 4 /C/GCE was immersed in Trp solution and stored at a 4 • C refrigerator. After a week, the anodic peak current retains 90.4% of the original response. fabricated by the same procedures. The RSD (4.78%, n = 5) is accepted, suggesting highly reproducible for electrode preparation. Finally, the stability was assessed by monitoring the anodic peak currents of 10 μM Trp at a same Fe3O4/C/GCE during a week. When not in use, Fe3O4/C/GCE was immersed in Trp solution and stored at a 4 °C refrigerator. After a week, the anodic peak current retains 90.4% of the original response.

Detection Trp in Human Serum Samples
SDLSV was adopted for quantitative analysis of Trp due to its superior sensitivity and high resolution [30,31,33]. The detection results were listed in Table 2. The Trp concentration in human serum sample is 20.13 μM with RSD of 1.5% (n = 3). Moreover, the good recoveries (105-107%) shows the proposed Fe3O4/C/GCE has great application prospects in the detection of Trp from various real samples.

Conclusions
In this study, we reported facile one-pot synthesis of Fe3O4/C composite and its application for cost-effective and sensitive detection of tryptophan (Trp) in human serum samples. The Fe3O4/C/GCE was prepared by drop-casting method. Owing to the large electrochemical active area and synergistic effect from Fe3O4 NPs and carbon coating, the response anodic peak current enhanced greatly. As a result, the proposed Fe3O4/C/GCE showed two wide linear detection ranges (1.0-80 μM and 80-800 μM) and a low LOD of 0.26 μM. Finally, the proposed Fe3O4/C/GCE successfully realized the detection of Trp in human serum samples with satisfactory recoveries (105-107%). Together with facile preparation and low cost, Fe3O4/C/GCE has shown tremendous application prospects on the detection of Trp from various actual samples.

Detection Trp in Human Serum Samples
SDLSV was adopted for quantitative analysis of Trp due to its superior sensitivity and high resolution [30,31,33]. The detection results were listed in Table 2. The Trp concentration in human serum sample is 20.13 µM with RSD of 1.5% (n = 3). Moreover, the good recoveries (105-107%) shows the proposed Fe 3 O 4 /C/GCE has great application prospects in the detection of Trp from various real samples.

Conclusions
In this study, we reported facile one-pot synthesis of Fe 3 O 4 /C composite and its application for cost-effective and sensitive detection of tryptophan (Trp) in human serum samples. The Fe 3 O 4 /C/GCE was prepared by drop-casting method. Owing to the large electrochemical active area and synergistic effect from Fe 3 O 4 NPs and carbon coating, the response anodic peak current enhanced greatly. As a result, the proposed Fe 3 O 4 /C/GCE showed two wide linear detection ranges (1.0-80 µM and 80-800 µM) and a low LOD of 0.26 µM. Finally, the proposed Fe 3 O 4 /C/GCE successfully realized the detection of Trp in human serum samples with satisfactory recoveries (105-107%). Together with facile preparation and low cost, Fe 3 O 4 /C/GCE has shown tremendous application prospects on the detection of Trp from various actual samples.