Highly sensitive simultaneous electrochemical determination of myricetin and rutin via solid phase extraction on a ternary Pt@r-GO@MWCNTs nanocomposite.

The simultaneous electrochemical determination of myricetin and rutin remains a challenge due to their indistinguishable potentials. To solve this problem, we constructed a ternary platinum nanoparticle, reduced graphene oxide, multi-walled carbon nanotubes (Pt@r-GO@MWCNTs) nanocomposite via a facile one-pot synthetic method. Under the optimized conditions, the ternary Pt@r-GO@MWCNTs nanocomposite exhibited good electrocatalytic activity toward myricetin and rutin via solid phase extraction and excellent performance for the simultaneous determination of myricetin and rutin. The oxidation peak current of myricetin was proportional to its concentrations in the range of 0.05-50 μM with a detection limit of 0.01 μM (S/N = 3). The linear range for rutin was 0.05-50 μM with a detection limit of 0.005 μM (S/N = 3). The ternary nanocomposite sensor also exhibited good reproducibility and stability, and was successfully used for the simultaneous determination of myricetin and rutin in real orange juice samples with recoveries ranging between 100.57% and 108.46%.

Introduction

Flavonoids have received significant attention due to their effects on human health. In particular, myricetin and rutin are two important flavonoids naturally present in plants including berries, tea, fruits, vegetables, and medicinal herbs [1]. According to previous reports, myricetin exhibits anti-oxidant, anti-cancer, and cytoprotective properties [2]. Rutin has a wide range of physiological functions, such as anti-inflammatory, anti-bacterial, anti-aging, and anti-oxidation abilities [3]. Because of the above listed pharmacological effects, health foods, drinks, and medicinal herbs usually contain myricetin and rutin. Therefore, it is important to develop a method to determine the contents of myricetin and rutin in food and medicine.

To date, some analytical methods have been developed to simultaneously detect myricetin and rutin, including capillary electrophoresis [4,5], chemiluminescence [6], high performance liquid chromatography (HPLC) [7], and spectrophotometry [8]. These methods can sensitively detect myricetin and rutin, but they have many drawbacks including the need for expensive equipment and time-consuming experimental processes, which constrain their practical application [[9], [10], [11], [12], [13], [14], [15], [16], [17], [18]]. Compared to previously developed methods, voltammetry provides a fast response, uses simple equipment, and is highly sensitive to the determination of myricetin or rutin [[19], [20], [21], [6]arene based potentiometric sensor. J. Mol. Liq.. 2014 ">22], [23], [24], [25], [26], [27], [28]]. Although many electrochemical sensors have been reported for single determinations of myricetin or rutin [[29], [30], [31], [32], [33], [34], [35]], their simultaneous determination has seldom been reported due to their analogous chemical structures and electrochemical characteristics, resulting in an indistinguishable potential [36,37]. Therefore, it is imperative to develop an electrochemical sensor that can realize the simultaneous determination of myricetin and rutin.

Multi-walled carbon nanotubes (MWCNTs) and reduced graphene oxide (r-GO) are promising conductor materials with high surface area, stable physical and chemical properties, high mechanical strength, and good absorption properties [[38], [39], [40], [41], [42], [43], [44], [45], [46], [47]]. In addition, they are easily combined with other nanomaterials due to their constituent hydroxyl (—OH), carboxyl (—COOH), and carbonyl (–CO) functional groups [[48], [49], [50], [51]]. The noble metal Pt is widely used in the field of catalysis and fuel cells [52]. Pt nanoparticles have been loaded into carbon nanotubes to form novel composite materials that show good performances as electrochemical sensors [[53], [54], [55]].

In this study, we synthesized a novel ternary nanocomposite consisting of Pt NPs, MWCNTs and r-GO for the simultaneous determination of myricetin and rutin. The ternary nanocomposite was fabricated by a facile one-pot synthesis method and showed strong ability to adsorb myricetin and rutin during solid-phase extraction. Because of this excellent property, the proposed nanocomposite sensor exhibits a distinguishable potential, high response, wide linear detection range, and low detection limit for both myricetin and rutin. In addition, this sensor exhibits good reproducibility, reversibility, and stability and was successfully used in the quantitative determination of myricetin and rutin in real orange juice samples, yielding satisfactory results.

Experimental

Chemicals and reagents

Rutin (98%, analytical grade) and chloroplatinic acid (H2PtCl6) were purchased from J&K Chemical Corporation (Beijing, China). Myricetin (97%, analytical grade) was obtained from Tokyo Chemical Industry (TCI) Co., Ltd. The MWCNTs (95%, analytical grade) were purchased from Aladdin Chemical Reagent Co., Ltd. (Shanghai, China). All other chemicals were of analytical grade. Britton-Robinson (BR) buffer was prepared by mixing acetic acid, phosphoric acid, and boric acid. Various pH values of the 0.04 M BR buffer were obtained by adding 0.2 M NaOH and HCl. Stock solutions of 1.0 mmol/dm3 myricetin and 1.0 mmol/dm3 rutin were obtained by dissolution in 50 mL of ethanol and subsequent dilution with doubly distilled water to 100 mL. All aqueous solutions were prepared with doubly distilled water.

Fabrication of the ternary Pt@r-GO@MWCNTs nanocomposite

A novel one-pot method was used to synthesize the ternary Pt@r-GO@MWCNTs nanocomposites at room conditions (room temperature is 25 °C) (Scheme 1). First, 30 mg of monolayer graphene oxide and 30 mg of MWCNTs were placed in 30 mL of double distilled water and mixed for 30 min under ultrasonication. Subsequently, under stirring conditions, 500 μL of a 60 mM H2PtCl6 solution was added to the above solution. Simultaneously, 10 mL of a 0.05 g/mL NaBH4 solution was added and the reaction mixture was stirred vigorously for 30 min. The product was washed three times with a mixed solution of ethanol and water at a ratio of 1:1. Finally, 8 mL of an ethanol and water mixture at a ratio of 1:3 was added to the product, yielding the ternary Pt@r-GO@MWCNTs nanocomposite.

Scheme 1

Fabrication of the ternary Pt@r-GO@MWCNTs nanocomposite.

Fabrication of the ternary Pt@r-GO@MWCNTs nanocomposite sensor

First, a glassy carbon electrode (GCE, 3 mm in diameter) was polished with 0.3 and 0.05 μm alumina powders, then rinsed ultrasonically with absolute alcohol and distilled water, and dried under a nitrogen stream. Subsequently, 5 μL of a homogeneous suspension of Pt@r-GO@MWCNTs was dropped onto the electrode surface and dried at room temperature (room temperature is 25 °C) to obtain the Pt@r-GO@MWCNTs nanocomposite sensor.

Instrumentation and methods

Electrochemical cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), and differential pules voltammetry (DPV) measurements were performed using a CHI 660B electrochemical workstation (Chenhua Co. Ltd. Shanghai, China). A three-electrode system consisting of a bare or modified GCE (3 mm in diameter) working electrode, platinum wire counter electrode, and saturated calomel reference electrode (SCE) was used. DPV scans from −0.1–0.8 V with an amplitude of 0.05 V, pulse width of 0.05 s, pulse period of 0.5 s, and increment of 0.004 V were performed. For CV, scan rate was 50 mV/s and sample interval was 0.001 V. EIS was obtained in a 5.0 mmol/dm3 K3[Fe(CN)6]/K4[Fe(CN)6] solution containing 0.1 M KCl under an open circuit potential at a frequency range from 0.1 Hz to 100 kHz and 5 mV amplitude. The surface morphology was characterized via field emission scanning electron microscopy (FE-SEM; Zeiss Ultra 55, Germany) and energy dispersive X-ray spectroscopy (EDS).

Results and discussion

Morphology and characterization of the nanocomposite

Fig. 1 shows the SEM images at low and high magnifications of the ternary Pt@r-GO@MWCNTs nanocomposites. The MWCNTs and r-GO were clearly observed, and the Pt NPs were uniformly and densely distributed in the MWCNTs and r-GO. The Pt@r-GO@MWCNTs composite was also analyzed by EDS (Fig. 2), showing that the main components were C, O and Pt at approximately 85.27, 12.5, and 2.23 wt%, respectively (Si was also observed because the Pt@r-GO@MWCNTs nanocomposite was dropped onto the silicon wafer for SEM sample preparation). The above results indicate the successful fabrication of the Pt@r-GO@MWCNTs nanocomposite.

Fig. 1

SEM images at (A) low and (B) high magnification of the Pt@r-GO@MWCNTs nanocomposite.

Fig. 2

EDS of Pt@r-GO@MWCNTs. Inset is the elemental contents.

Electrochemical characterization of the Pt@r-GO@MWCNTs/GCE modified electrode

CV and EIS were performed for the electrochemical characterization of the Pt@r-GO@MWCNTs/GCE electrode. Fig. 3A shows the CVs of different electrodes in a 5.0 mmol/dm3 K3[Fe(CN)6]/K4[Fe(CN)6] solution containing 0.1 M KCl at 50 mV/s. A pair of reversible redox peaks were observed at +0.26 and + 0.17 V for the bare GCE (curve a). After modification with the Pt@r-GO@MWCNTs nanocomposite (curve b), the redox peak currents increased and a larger background current was observed due to the Pt@r-GO@MWCNTs nanocomposite dramatically increasing the electrode surface area and exhibiting good electrical conductivity [41,42]. Fig. 3B shows the EIS plots of the bare GCE and Pt@r-GO@MWCNTs/GCE. The bare GCE (curve a) exhibited a small resistance, but when the Pt@r-GO@MWCNTs nanocomposite GCE was used (curve b), it displayed a significantly smaller resistance. Therefore, the CV and EIS diagrams further demonstrate the successful preparation of the Pt@r-GO@MWCNTs/GCE sensor.

Fig. 3

(A) CVs and (B) EIS of the bare GCE (a) and Pt@r-GO@MWCNTs/GCE (b) in 5.0 mmol/dm3 K3[Fe(CN)6]/K4[Fe(CN)6] containing 0.1 M KCl.

Electrochemical behavior of myricetin and rutin

The CVs of bare GCE and Pt@r-GO@MWCNTs/GCE in 50 μM myricetin and 50 μM rutin solutions are shown in Fig. 4A. Unlike the bare GCE, two pairs of redox peaks of myricetin and rutin were clearly observed when the Pt@r-GO@MWCNTs/GCE was used along with a larger background current. Thus, the Pt@r-GO@MWCNTs nanocomposite showed a good electrocatalytic effect for detection of both myricetin and rutin. In addition, the DPV results shown in Fig. 4B for the bare GCE (Ipa = 4.617 μA, Epa = 0.276 V for myricetin, and Ipa = 1.316 μA, Epa = 0.432 V for rutin) provided very small current signals. Moreover, the potential differences were insufficient to realize the simultaneous determination of myricetin and rutin. However, after modification of the GCE with the Pt@r-GO@MWCNTs nanocomposite (curve b, Ipa = 44.45 μA, Epa = 0.320 V for myricetin, and Ipa = 51.71 μA, Epa = 0.496 V for rutin), the oxidation peak current signals increased by a factors of 11 and 50 for myricetin and rutin, respectively, over that of the bare GCE. In addition, the potential differences became large enough for the simultaneous detection of myricetin and rutin. Therefore, the Pt@r-GO@MWCNTs/GCE sensor was successful used for the simultaneous determination of myricetin and rutin.

Fig. 4

(A) CVs and (B) DPVs of 50 μM myricetin and 50 μM rutin using the bare GCE (a) and Pt@r-GO@MWCNTs/GCE (b) in pH 3.0 BR buffer solutions.

Effect of scan rate

The effects of scan rate were investigated by CVs in a pH 3.0 BR buffer solution containing 50 μM myricetin and rutin. As shown in Figs. 5A and B, with an increase in scan rate, both redox currents increased linearly. Figs. 5C and D show the linear equations for myricetin, Ipa (μA) = 0.3211ν (mV/s) − 5.2034 (R2 = 0.9912) and Ipc (μA) = −0.3428ν (mV/s) + 10.9990 (R2 = 0.9902), and for rutin Ipa (μA) = 0.5782ν (mV/s) − 1.3055 (R2 = 0.9997) and Ipc (μA) = 0.5214ν (mV/s) + 4.3750 (R2 = 0.9996). These results indicate that the electrocatalytic reactions of both myricetin and rutin on the Pt@r-GO@MWCNTs/GCE are typical adsorption-controlled processes.

Fig. 5

CVs of the Pt@r-GO@MWCNTs/GCE in pH 3.0 BR buffer solution containing (A) 50 μM myricetin and (B) 50 μM rutin at different scan rates (20–300 mV/s). The relationships of oxidation and reduction peak currents of (C) myricetin and (D) rutin with scan rates are also provided.

The effect of pH

The effect of pH was investigated by DPVs in BR buffer solution containing 50 μM myricetin and rutin at different pH values (Fig. 6A). As the pH increased (Fig. 6B), oxidation peak currents of both myricetin and rutin increased until pH 3 and then decreased at higher pH values. Therefore, the optimum pH was determined to be 3. In Fig. 6C, as the pH increased, both oxidation peak potentials of myricetin and rutin negatively shifted with a good linear relationship between the oxidation peak potentials and pH. The linear equation for myricetin was Epa (V) = −0.0702 pH + 0.5398 (R2 = 0.9918) and for rutin Epa (V) = −0.0632 pH + 0.6997 (R2 = 0.9904). The slopes of the respective equations were 70.2 and 63.2 mV/pH, suggesting that both myricetin and rutin oxidation followed the Nernst equation requiring identical number of protons and electrons. Combined with previous reports [36,37], the myricetin and rutin oxidation mechanism involved the same number electron and proton transfer processes (Scheme 2).

Fig. 6

(A) DPVs of 50 μM myricetin and rutin in BR buffer at different pH values on the Pt@r-GO@MWCNTs/GCE nanocomposite sensor. (B) The relationship between pH and oxidation peak currents of myricetin and rutin. (C) The relationship between pH and oxidation peak potentials of myricetin and rutin.

Scheme 2

The reaction mechanisms of (A) myricetin and (B) rutin.

The effects of pre-concentration time and pre-concentration potential

The pre-concentration time and potential were studied in the pH 3 BR buffer solution containing 50 μM myricetin and rutin. In Fig. 7A, at the pre-concentration potential of −0.3 V, the oxidation peak current increased progressively with an increase in pre-concentration time and achieved a maximum value when the pre-concentration time was 550 s, and decreased afterwards. Therefore, the pre-concentration time of 550 s was applied in subsequent studies. Moreover, we studied the effect of pre-concentration potential on the oxidation peak currents of myricetin and rutin using the optimized pre-concentration time. In Fig. 7B, the oxidation peak current increased until −0.3 V, after which the oxidation peak currents gradually decreased. Therefore, the optimal pre-concentration potential for determination of myricetin and rutin was −0.3 V.

Fig. 7

Effect of (A) pre-concentration time and (B) pre-concentration potential on the oxidation peak currents of 50.0 μM myricetin and 50.0 μM rutin at the Pt@r-GO@MWCNTs/GCE in pH 3.0 BR buffer solution.

Quantitative determination of myricetin and rutin

Fig. 8A shows the individual quantitative determination of different concentrations of myricetin in the presence of 2 μM rutin in the pH 3 BR buffer solution. Fig. 8B shows two linear detection ranges of 0.05–10 and 10–50 μM between the oxidation peak current and concentration with linear equations of Ip (μA) = 1.9321C (μM) + 1.9984 (R2 = 0.9901) and Ip (μA) = 0.6533C (μM) + 14.8210 (R2 = 0.9947), respectively. DPVs of the individual quantitative determination of different rutin concentrations in the presence of 2 μM myricetin in the pH 3 BR buffer solution are shown in Fig. 8C. The oxidation peak current and concentration showed two linear detection ranges of 0.05–10 and 10–50 μM (Fig. 8D) with linear equations of Ip (μA) = 2.6258C (μM) + 3.1559 (R2 = 0.9900) and Ip (μA) = 0.6131C (μM) + 22.6930 (R2 = 0.9977), respectively.

Fig. 8

(A) DPVs of myricetin at different concentrations (0.05–50 μM) in BR buffer (pH = 3) in the presence of 2 μM rutin at the Pt@r-GO@MWCNTs/GCE. (B) Linear relationship between the oxidation peak current and concentration of myricetin. (C) DPVs of rutin at different concentrations (0.05–50 μM) in BR buffer (pH = 3) in the presence of 2 μM myricetin at the Pt@r-GO@MWCNTs/GCE. (D) Linear relationship between oxidation peak current and concentration of rutin. (E) DPVs of myricetin and rutin at different concentrations (0.05–50 μM) in BR buffer (pH = 3) at the Pt@r-GO@MWCNTs/GCE. The linear relationships between oxidation peak currents and concentration of (F) myricetin and (G) rutin.

The simultaneous quantitative determination of myricetin and rutin at the Pt@r-GO@MWCNTs/GCE was investigated via DPVs. From Fig. 8E, with an increase in concentration of myricetin and rutin, the corresponding oxidation peaks current gradually increased in intensity. Fig. 8F shows that the concentration of myricetin and oxidation peak current exhibited two linear detection ranges of 0.05–10 and 10–50 μM with linear equations of Ip (μA) = 2.0561C (μM) + 1.6368 (R2 = 0.9958) and Ip (μA) = 0.5643C (μM) + 17.2910 (R2 = 0.9731), respectively. From Fig. 8G, the concentration of rutin and oxidation peak current showed two linear detection ranges of 0.05–10 and 10–50 μM with equations of Ip (μA) = 2.4047C (μM) + 3.5326 (R2 = 0.9944) and Ip (μA) = 0.6247C (μM) + 21.6050 (R2 = 0.9906), respectively. The detection limits of myricetin and rutin were 10 and 5 nM (S/N = 3), respectively. These results indicated that the Pt@r-GO@MWCNTs nanocomposite modified electrode showed higher sensitivity and selectivity for both myricetin and rutin. Moreover, compared to other reported determinations of myricetin or rutin, our Pt@r-GO@MWCNTs/GCE can realize the simultaneous determination of myricetin and rutin with lower detection limits and the wider linear ranges (Table 1) [16,36,37,[56], [57], [58], [59]]. The electrode modified with the Pt@r-GO@MWCNTs nanocomposite showed improved detection sensitivity due to the pre-concentration ability and solid-phase extraction.

Table 1

Comparison of detection performances of the developed sensor and other electrochemical devices reported for detection of myricetin or rutin.

Electrode	Method	Linear range (μM)	Limit of detection (nM)	Reference
AuNPs/MWCNTs/GCE	SWV	Myricetin:0.05–40Rutin: —	12—	[16]
Polyfurfural film/GCE	DPV	Myricetin:0.05–10Rutin: 0.001–10	100.025	[36]
β-CD–Au@PTCA–SWCNHs/GCE	DPV	Myricetin:0.01–10Rutin: 0.01–10	3.84.4	[37]
MIP/GO–MWCNTs/GCE	DPV	Myricetin: —Rutin: 0.01–1	—5	[56]
Cu–CS/MWCNTs/GCE	DPV	Myricetin: —Rutin: 0.05–100	—10	[57]
PSSA/CNTs/MBT/Au/GCE	DPV	Myricetin: —Rutin: 0.8–10	—1.8	[58]
Ni-GO/GCE	DPV	Myricetin: —Rutin: 0.011–15	—3.2	[59]
Pt@r-GO@MWCNTs/GCE	DPV	Myricetin: 0.05–50Rutin: 0.05–50	105	This work

Reproducibility and stability of the sensor

We also explored the effects of potentially interfering inorganic ions and organic molecules. When 1000-fold excesses of NaNO3, NH4F, MnSO4, CaCl2, and Zn(AC)2, and 100-fold excesses of citric acid, tartaric acid, cystine, glucose, ascorbic acid, and oxalic acid were added, as showen in Table 2, no significant signal changes were observed for oxidation peaks currents of myricetin and rutin, indicating that the assay was not disturbed. These results indicate that the modified electrode exhibited good selectivity for the determination of myricetin and rutin. Furthermore, three modified electrodes were stored under refrigerated conditions (4 °C) and used regularly for myricetin and rutin analysis at 10-day intervals. After 30 days of storage, the sensor response remained quite stable (the reduction of the sensor output signal was ˂5%), demonstrating the excellent stability of Pt@r-GO@MWCNTs/GCE. The reversibility of the Pt@r-GO@MWCNTs/GCE electrode was investigated by DPVs. We modified an electrode that was stored under refrigerated conditions and subsequently used for myricetin and rutin analysis at 8 h intervals for two days. The standard deviation of the oxidation peak currents of the myricetin and rutin was (3.54 ± 0.41) %, indicating good reversibility.

Table 2

Effect of potentially interfering inorganic ions and organic molecules on the oxidation peak current signals of 50 μM myricetin and rutin in the pH 3 BR buffer solution at the Pt@r-GO@MWCNTs/GCE (n = 3).

Interferences	Concentration (mol/L)	Signal change (%)
		Myricetin	Rutin
NaNO3	0.05	4.67 ± 0.91	1.07 ± 0.29
NH4F	0.05	−1.53 ± 0.26	−0.04 ± 0.28
MgSO4	0.05	1.60 ± 0.29	−2.09 ± 0.85
CaCl2	0.05	−4.33 ± 0.38	−2.01 ± 0.92
Zn(AC)2	0.05	0.78 ± 0.33	3.36 ± 0.24
Citric acid	0.005	4.41 ± 0.78	−3.07 ± 0.36
Tartaric acid	0.005	5.62 ± 0.33	−3.84 ± 0.66
Cystine	0.005	3.98 ± 0.44	−2.20 ± 0.35
Glucose	0.005	4.66 ± 0.63	3.03 ± 0.50
Ascorbic acid	0.005	5.10 ± 0.24	2.03 ± 0.23
Oxalic acid	0.005	−4.65 ± 0.35	−5.17 ± 0.65

Simultaneous determination of myricetin and rutin in real orange juice samples

The actual analytical application of the prepared Pt@r-GO@MWCNTs/GCE sensor for the simultaneous determination of myricetin and rutin in real orange juice was achieved by standard addition. Three parallel experiments were performed for all measurements. As shown in Table 3, the real orange juice sample recovery rate ranged between 100.57% and 108.46%, and the RSD value was ˂6%. These results indicated that the prepared sensor can be successfully applied for the actual simultaneous determination of myricetin and rutin in real orange juice samples.

Table 3

Simultaneous determination of myricetin and rutin in real orange juice samples (n = 3).

Sample	Added (μM)	Founda (μM)	Recovery (%)	RSD (%)
Myricetin	30	32.38 ± 1.26	107.92	1.6
	10	10.50 ± 0.32	105.04	1.2
	0.5	0.5423 ± 0.0790	108.46	5.9
	0.3	0.3017 ± 0.0058	100.57	0.8
Rutin	30	31.77 ± 1.95	105.89	2.5
	10	10.11 ± 0.16	101.10	0.6
	0.5	0.5334 ± 0.0600	106.68	4.5
	0.3	0.3167 ± 0.0220	105.50	2.8

Conclusions

In summary, we successfully manufactured a novel electrochemical sensor based on a ternary Pt@r-GO@MWCNTs nanocomposite. The nanocomposite sensor exhibited improved detection sensitivity due to the superior pre-concentration ability of the nanocomposite for myricetin and rutin. Under the optimized conditions, the oxidation peak currents of myricetin were proportional to its concentration with linearity ranging from 0.05 to 50 μM and a detection limit of 0.01 μM (S/N = 3). For rutin the linear range was from 0.05 to 50 μM with a detection limit of 0.005 μM (S/N = 3). The sensor also exhibited good reproducibility and stability, and was successfully used for the simultaneous determination of myricetin and rutin in real orange juice samples providing satisfactory results.

Acknowledgments

This work was financially supported by the (Grant No. 21874047, 21475046, 21427809).

Conflicts of interest

The authors declare that there are no conflicts of interest.