Introduction

Dopamine [4-(2-aminoethyl) benzene-1, 2-diol] (DA), the most significant neurotransmitter among the class of catecholamines, plays an important role in the function of renal, human metabolism, central nervous, cardiovascular, and hormonal systems [1, 2]. The normal level of DA is from 0.01 to 1 μmol L−1 in serum [3, 4]. Insufficient DA concentration due to the loss of DA producing cells may lead to a disease called Parkinson’s and Alzheimer’s disease [5, 6]. Therefore, sensitive and selective determination of DA has an important value in clinical disease diagnosis.

Doxorubicin (DOX) is a cytotoxic anthracycline antibiotic that has been widely used in the treatment of several types of cancer, such as acute lymphoblastic leukemia, thyroid, Hodgkin’s and non-Hodgkin’s lymphomas, breast, soft tissue and osteogenic sarcomas, genitourinary and bronchogenic carcinomas [7, 8]. This drug interacts with double helix of DNA and the anthracycline moiety in cancer cells and inhibits the replication and transcription of DNA [9]. However, the clinical use of doxorubicin is limited by accumulative dose-dependent irreversible chronic cardiomyopathy; this can lead to congestive heart failure, with an ultimate mortality rate of 20–40 % [10]. Therefore, sensitive determination of DOX in human fluids such as blood serum and urine is essential.

The effectiveness of many anticancer drugs, like DOX, is limited by their toxicity on cells of the hematopoietic system resulting in peripheral blood leucopenia, neutropenia and thrombocytopenia. Against this background, DA can be used as a protective agent against the hematotoxic effects of DOX [11]. Sarkar et al. [12], showed that dopamine significantly enhances the efficacies of commonly used anticancer drugs (like DOX) and also indicates that an inexpensive drug like dopamine, which is extensively used in the clinics, can acts as an antiangiogenic agent for the treatment of breast and colon cancer [12]. So, it is necessary to develop the analytical tools for the determination of DOX and DA together. A number of analytical methods have been developed for the determination of DA or DOX including liquid chromatography [13, 14], capillary electrophoresis [15, 16], spectrofluorometry [17, 18], and electroanalytical techniques [19–23]. An appropriate alternative to the above-mentioned techniques are electroanalytical methods. These methods are low cost, fast, portable, highly sensitive and able to make direct measurements in different analytical samples. However, one major problem is that the oxidation peaks of DA and DOX are too close at unmodified electrodes, which results in overlapping voltammetric responses and making their simultaneous detection highly difficult. To overcome this problem, it is necessary to make further efforts for the fabrication of electrochemically modified electrodes that can be used for simultaneously determination of these compounds.

Over the past few years, various modified electrodes have been designed for simultaneous determination of more than one compound [24–26]. Carbon nanotubes (CNTs), in forms of single-walled CNTs (SWCNTs) and multiwalled CNTs (MWCNTs), have specific properties. MWCNTs consisting of several layers of graphene [27] have also been widely used in the study of electrochemical sensors due to its good electrochemical properties, wide electrochemical potential window, and interesting compatibility with biological molecules [28–30]. Chemical oxidation of MWCNTs with reagents of different oxidation power (acidic and basic) was considered by Datsyuk et al. [31]. They have shown that the treatment of CNTs with strong oxidizing agents (acidic) causes severe etching of the graphitic surface, leading to shorter tubes. But, basic treatment is not destructive for the tubes and simultaneously, highly purified material is produced.

The chemical and electrochemical pretreatment of GCE shows significant changes in physical and electrochemical properties of it. Electrochemical pretreatment of GCE causes the formation of a carbon oxide film on the surface which provides sensitivity of the GCE in electrochemical sensing [32, 33]. Usually, electrochemical treatment of GCE was carried out in acidic or neutral medium at over a wide potential range [24, 34, 35]. However, electrochemical pretreatment of GCE and MWCNTs can be achieved in basic medium at lower potential [33, 36].

In this current study, to the best of our knowledge, oxidized GCE (OGCE) and oxidized MWCNT/GCE (OMWCNT/GCE) are prepared by electrochemical treatment method in basic medium that was reported for the first time. The sensor not only exhibited strong catalytic activity toward the oxidation of DA and DOX but also separated the originally overlapped signals of DA and DOX oxidation at the bare electrode into two well-defined peaks. The analytical performance of the proposed system was evaluated, and the optimal condition was used for the simultaneous determination of DA and DOX in complex biological matrix.

Materials and methods

Reagents

All chemicals were used as received without further purification. Dopamine (DA), doxorubicin (DOX), ascorbic acid (AA), uric acid (UA), tyrosine, tryptophan and sodium hydroxide were of supplied by Merck Company (Darmstadt, Germany) or Aldrich Company (USA). All other reagents were of analytical grade. In all the measurements, the supporting electrolyte used was Briton–Robinson buffer solution (BRBS) of pH 5.0. The stock solutions of 5.0 × 10−3 mol L−1 DA and DOX were prepared and working solutions were prepared through diluting the stock solution with buffer solution. Pure multiwall carbon nanotubes (diameter 10–20 nm, 30 μm length) were obtained from Neutrino Company (Iran).

Apparatus

The cyclic voltammetry (CV) and square wave voltammetry (SWV) experiments were performed using an Autolab PGSTAT302N potentiostat/galvanostat controlled with the Nova version 1.7 software. A conventional three-electrode system was used with an OMWCNT/GCE as working electrode, an Ag/AgCl (saturated KCl) as reference electrode, and a platinum wire auxiliary electrode. All the values of pHs were adjusted by a Metrohm Model 713 pH lab (Herisau, Switzerland). The structural morphology and elemental analysis of the electrode was studied using a VEGA3 TESCAN scanning electron microscope (SEM) and the energy dispersive X-ray spectroscopy (EDS) analysis. All experiments were carried out in an unstirred electrochemical cell at room temperature.

Electrode preparation

A bare GCE was polished successively with 0.3 μm Al2O3 water slurry using a polishing cloth and it was rinsed with doubly distilled water, sonicated subsequently in a 1:1 aqueous HNO3 solution, ethanol, and doubly distilled water (DDW) each for 10 min. Further, the GCE was oxidized by performing 40 cycles in 0.5 mol L−1 NaOH between 0.0 and 0.9 V at 30 mV s−1. The oxidized GCE was washed by deionized water and transferred to pH 7.0 BRBS for further studies. The fabrication of MWCNTs is described as follows: the MWCNTs’ suspension was prepared by dispersing 1.0 mg MWCNTs in 5.0 mL 3:1:1 mixture of DDW, ethanol and sodium dodecyl sulfate (SDS) under sonication for 30 min. A 10 μL aliquot of black suspension was dropped directly onto the clean GCE surface and dried at room temperature to form a MWCNT film at the GCE surface and prepare a MWCNT/GCE. Oxidized MWCNT/GCE (OMWCNT/GCE) was prepared by immersing MWCNT/GCE in 0.5 mol L−1 NaOH solution and applying cyclic voltammetry (CV) from 0.0 to +0.9 V with the scan rates 30 mV s−1 for 50 cycles. The OMWCNT/GCE was washed by DDW and transferred to pH 7.0 BRBS for further studies.

Results and discussion

Electrochemical oxidation of GCE and MWCNT/GCE

The GCE and MWCNT/GCE were oxidized by electrochemical method in basic medium (0.5 mol L−1NaOH) over the range of 0.0 and +0.9 V at 30 mV s−1. The current decreased gradually when the number of cycles was between 1 and 5 and then increased at +0.90 V for sequential cycles (from 5 to 40 cycles). For MWCNT/GCE the current decreased from 1 to 8 cycles and then increased when the number of cycles was between 8 and 50. These changes in currents indicate that the new surfaces are forming at GCE and MWCNT/GCE. These can be due to the formation of functional groups like carbonyl, carboxyl, and hydroxyl radical species on the electrode surface [32, 33].

Morphology and characterization of electrodes

The surface morphology of modified electrode was analyzed by scanning electron microscopy (SEM). Figure 1a depicts the SEM micrograph of OMWCNT/GCE.

Fig. 1
figure 1

a SEM image of OMWCNT/GCE, b EDS analysis of MWCNT/GCE, c EDS analysis of OMWCNT/GCE

Full size image

As shown in Fig. 1a, a network-like structure of MWCNTs without aggregation was observed on the electrode, which indicated that the MWCNTs were immobilized on the GCE surface. The EDS (Fig. 1b, c) of the obtained MWCNT/GCE and OMWCNT/GCE reveals the existence of C and O elements. At the OMWCNT/GCE, the percentage of O is higher than at the MWCNT/GCE. These results clearly confirm that the oxidation of MWCNT on the modified electrode surface in basic medium occurred and MWCNT was functionalized with carbonyl or hydroxyl groups.

Electrochemical behaviors of DA and DOX and their mixture at different electrodes

The CV responses of the bare GCE, OGCE, MWCNT/GCE, and OMWCNT/GCE toward DA (50 μmol L−1) and DOX (50 μmol L−1) are shown in Fig. 2. At the bare GCE, DA (Fig. 2a) and DOX (Fig. 2b) show quasi-reversible behavior with a small CV peak response. For DA, the oxidation and reduction peaks appear at 413 mV and 164 mV, respectively, and the peak potential separation (ΔEp) is about 249 mV. For DOX, the oxidation and reduction peaks appear at 591 and 444 mV, respectively, and ΔEp is about 147 mV. At the OGCE, the peak currents increased greatly, the peak potentials shifted negatively and show more reversible behavior for both DA and DOX. The MWCNT/GCE significantly enhanced the redox peak currents compared with the bare GCE and OGCE which can be related to electrocatalytic behavior and the high surface area of MWCNT. The voltammograms (d) in Fig. 2a, b show that the corresponding oxidation and reduction peak currents of DA and DOX increase further at the OMWCNT/GCE. In addition, the peak’s potential shifted to lower value which confirms the electrocatalytic behavior of OMWCNT.

Fig. 2
figure 2

Cyclic voltammograms of a 50 μmol L−1 DA and b 50 μmol L−1 DOX in BRBS (pH 5.0) at the (a) GCE (b) OGCE (c) MWCNT/GCE, (d) OMWCNT/GCE at 50 mV s−1

Full size image

The CVs of different modified electrodes in BRBS (pH 5.0) containing 50 μmol L−1 DA and 50 μmol L−1 DOX were investigated. At the bare GCE (Fig. 3, blue curve), the oxidation peaks of DA and DOX overlap and a broad oxidation peak was observed at 418 and 592 mV, respectively, which revealed that it is impossible to simultaneously determine these compounds. For OGCE (Fig. 3, red curve), the peak currents of DA and DOX was improved and the selectivity was better than that at the bare GCE. At the MWCNT/GCE (Fig. 3, green curve) redox peak currents of DA and DOX increased compared to the bare GCE and OGCE. The OMWCNT/GCE (Fig. 3, dark blue curve) exhibited good electrocatalytic activity for the oxidation of DA and DOX, since peak potentials shifted negatively and the peak current of the two compounds increased more at the OMWCNT/GCE compared to the MWCNT/GCE. From above, we can see OMWCNT/GCE shows the highest electrocatalytic activity toward DA and DOX among all the modified GCEs and it is suitable for simultaneous determination of DA and DOX (all CV peaks’ baseline currents were corrected, and the net current was used for comparison).

Fig. 3
figure 3

Cyclic voltammograms of the mixture containing 50 μmol L−1 DA and 50 μmol L−1 DOX in BRBS (pH 5.0) on (a) GCE, (b) OGCE, (c) MWCNT/GCE, (d) OMWCNT/GCE at 50 mV s−1

Full size image

Effects of the scan rate

The effect of potential scan rate on the oxidation responses of DA and DOX was investigated. Relationships between the redox peak currents and scan rates of DA in the range of 5–200 mV s−1 and DOX in the range of 5–150 mV s−1 are linear. The linear relationship between peak currents and scan rates suggests that the redox reactions of each two compounds at OMWCNT/GCE are adsorption-controlled processes at such scan rates.

Influence of solution pH

As the protons took part in the electrode reaction process of DA and DOX, pH of the working buffer is very important for the detection of these species. Thus, the effect of solution pH value on peak current and peak potential was investigated by recording the CVs of DA and DOX with concentrations of 40 and 30 μmol L−1, respectively, in a series of BRBS of varying pH in the range 2.0–9.0 (Fig. 4a). One can see from Fig. 4b that the peak currents of DA increase with the increasing pH value from 2.0 to 5.0 and reach a maximum at pH 5.0, and decrease when the pH value increases gradually. This may be because the surface charge of the electrode gets more negative and positive charge of DA decreases when increasing the pH (pKa1 of DA is 8.9 [2]). However, there is maximum interaction between DA and electrode surface at pH 5.0. So, maximum current was achieved. Similarly, we can also see from Fig. 4a that the peak current response of DOX slowly increases with the increasing pH value from 2.0 to 4.0 and remains almost unchanged until 5.0 then it sharply decreases when the pH value increases gradually. With increasing pH from 2.0 to 5.0, positive charge of DOX decreases (at 2 < pH < 8 DOX has one positive charge at the amino sugar group [37]). In contrast, with increasing pH, negative charge of electrode surface is increased. So, at pH 4.0 or 5.0 maximum interaction between DOX and electrode surface occurs. Considering sensitivity and physiological condition, BRBS pH 5.0 was selected for further experiments.

Fig. 4
figure 4

a Cyclic voltammograms of the mixture containing 40 μmol L−1 DA and 30 μmol L−1 DOX on OMWCNT/GCE in different pH values from 2.0 to 9.0. b The plots of anodic peak current, and c anodic peak potential (E pa) of these compounds versus pH values

Full size image

On the other hand, the effect of pH on the E pa and E pc of DA and DOX was investigated by CV in the solution containing DA (40 μmol L−1) and DOX (30 μmol L−1). The electrochemical responses of DA and DOX on the OMWCNT/GCE show a strong dependence on the pH in the range of 2.0–9.0 (Fig. 4c). As shown in Fig. 4c, the E pa of DA and DOX shifted linearly toward negative values as pH increased, with regression equations of E pa = 0.594−0.0548 pH (R 2 = 0.9946), for DA and E pa = 0.839−0.0586 pH (R 2 = 0.9911), for DOX, respectively. The slopes were 54.8 mV pH−1 for DA and 58.6 mV pH−1 for DOX, respectively, which were close to the theoretical value of 59 mV pH−1 at 25 °C [38, 39]. The results indicate that the equal number of protons and electrons are involved in the electrochemical reaction [6, 40–42] (Scheme 1).

Scheme 1
scheme 1

Oxidation mechanism for DOX (a) and DA (b)

Full size image

Effects of accumulation time and potential

The effect of the accumulation potential on the anodic peak current of DA and DOX was examined in the potential range −0.40 to +0.20 V vs. Ag/AgCl under the above optimum conditions. The largest peak current was obtained at an accumulation potential of −0.30 V vs. Ag/AgCl and then decreased with increasing potential. Therefore, −0.30 V vs. Ag/AgCl was selected as the accumulation potential in the procedure. The dependence of the maximum peak current on the accumulation time was also examined. Under the other optimum conditions, there was a linear relationship between the peak current and accumulation time in the range 0–120 s, above which it became constant. In these experiments, 120 s was selected as the accumulation time.

Selective determinations of DA and DOX

SWV technique was employed for the selective detections of DA and DOX in their mixture due to its higher sensitivity and better resolution than the CV method [43]. In a mixture of DA and DOX, the concentration of one specie changes while the concentrations of the other one remains constant. The results are shown in Fig. 5a and b. As shown in Fig. 5a, the current response of DA increases with the increasing DA concentration in the range of 0.03–60 μmol L−1, while the currents of DOX are nearly unchanged. Similarly, in fixed concentrations of DA, the peak current of DOX was positively proportional to its concentration (Fig. 5b). It can be observed that the peak current of DOX linearly increased with increasing the DOX concentration in the range of 0.04–90 μmol L−1 at DA fixed concentrations. This indicates that the OMWCNT/GCE can be used for the selective determinations of DA or DOX in the presence of the other species.

Fig. 5
figure 5

SWVs at OMWCNT/GCE in BRBS of pH 5.0 a containing 5 μmol L−1 DOX and different concentrations of DA from 0.03 to 60 μmol L−1 (1 to 12); b containing 2 μmol L−1 DA and different concentrations of DOX from 0.04 μmol L−1 to 90 μmol L−1 (1 to 16)

Full size image

Simultaneous SWV determination of DA and DOX

The next attempt was taken to simultaneously detect DA and DOX using the OMWCNT/GCE. The SWV results showed that concurrent determination of DA and DOX with well-distinguished two anodic peaks can be possible at OMWCNT/GCE. Figure 6a shows the SWVs results at OMWCNT/GCE when the concentrations of DA and DOX were simultaneously changed. The separation between the two peak potentials is sufficient enough for simultaneous determination of the two species. The peak currents of DA were proportional to the concentration in two concentration ranges: 3.0 × 10−2–1.0 μmol L−1 with a calibration equation of I p (μA) = 5.36C (μmol L−1) + 0.73 (R 2 = 0.995), and 1.0–55.0 μmol L−1 with a calibration equation of I p (μA) = 0.41C (μmol L−1) + 6.87 (R 2 = 0.994) (Fig. 6b).

Fig. 6
figure 6

a SWVs profiles at OMWCNT/GCE in BRBS (pH 5.0) containing different concentration of DA from 3.0 × 10−2 to 55.0 μmol L−1 and DOX from 4.0 × 10−2 to 90.0 μmol L−1; b and c show the plots of the peak current vs. the concentration of DA and DOX, respectively

Full size image

In addition, the obtained detection limit was 8.5 × 10−3 μmol L−1 (S/N = 3). For DOX, there were two linear dynamic ranges: 4.0 × 10−2–2.0 μmol L−1 with a calibration equation of I p (μA) = 3.11C (μmol L−1) + 0.71 (R 2 = 0.996), and 2.0–90.0 μmol L−1 with a calibration equation of I p (μA) = 0.12C (μmol L−1) + 7.32 (R 2 = 0.990) (Fig. 6c). The detection limit of 9.4 × 10−3 μmol L−1 (S/N = 3) was obtained. Therefore, it seems that the present-used electrode may be a valuable means for making voltammetric sensor to detect DA and DOX in clinical laboratories.

Interferences, stability, and reproducibility

In order to investigate the selectivity of the OMWCNT/GCE, several compounds from common co-existing substances were investigated by detecting the response of the modified electrode to DA (3.0 μmol L−1) and DOX (5.0 μmol L−1). The results indicate that there is nearly no interference (less than ±5 % relative error) from the following compounds: 1000-fold NaCl and KCl, 800-fold CaCl2, MgSO4, and ZnCl2, and 100-fold glucose and citric acid, 30-fold l-cysteine, tyrosine, and tryptophan, 20-fold acetaminophen, ascorbic acid and uric acid had almost no influence on the current responses of DA and DOX. The interference of levodopa and carbidopa was also investigated. The results shown that less than 5-fold concentrations of these substances had no interference with determination of DA. The stability of the electrode was also tested. The peak current only decreased less than 5 % after the electrode was stored at 4 °C for 20 days. In addition, in order to evaluate the reproducibility of the modified electrode, a series of five OMWCNT/GCEs were prepared for the detection of 0.1 μmol L−1 DA, and 0.3 μmol L−1 DOX. The RSD (n = 5) of peak currents for DA and DOX were calculated as 3.1 and 2.6 %, respectively, which suggests that the reproducibility of the proposed electrode was good. These results indicate that the modified electrode has good anti-interference ability, reproducibility, and repeatability.

Real samples analyses

The applicability of OMWCNT/GCE to the determination of DA and DOX in human serum and human urine was examined. All samples were collected from a healthy male volunteer with informed consent and all experiments were performed in compliance with the relevant laws and institutional guidelines. For serum and urine samples preparation, different amounts of DA and DOX were spiked to 1 mL of the samples and then 0.8 mL of acetonitrile was added to remove serum protein. The mixture was centrifuged for 15 min at 3500 rpm to remove the serum protein residues. Then, the supernatant was taken carefully and transferred into a 25-mL flask and diluted up to the volume with BRBR (pH 5.0). Finally, the electrochemical signal was determined using OMWCNT/GCE at optimum conditions. Concentrations were measured by applying the calibration plot using the standard addition method. The results are shown in Table 1. The recovery values indicate a good accuracy of the proposed method. From the above experimental results, it is very clear that this method has great potential for the determination of trace amounts of these compounds in biological systems.

Table 1 Determination results of DA and DOX in serum and urine samples (n = 5)
Full size table

The comparison of the results for the determination of DA and DOX by different modified electrodes and different analytical parameters in the literature [2, 6, 10, 19, 44–48] is given in Table 2. The comparative data suggested superiority of the present sensor over some earlier reported methods, especially for the detection limit and sensitivity. This can be attributed to the OMWCNT composite on the GCE surface with large surface area, excellent conductivity, and good electrocatalytic effect.

Table 2 Comparison of the analytical performance of the different modified electrodes for determination of DA and DOX
Full size table

Conclusion

In this paper, a novel electrochemical sensor was successfully fabricated. The fabricated OMWCNT/GCE associated to the excellent electrocatalytic activity of the oxidized MWCNT showed high selectivity and sensitivity for individual and simultaneous determination of DA and DOX with low detection limits and wide concentration ranges. The OMWCNT/GCE not only exhibited high electrocatalytic activities toward the electrooxidations of DA and DOX, but also resolved the overlapped anodic peaks, lowered the oxidation over potentials, and enhanced the oxidation currents of DA and DOX. In addition to the good properties of high selectivity, sensitivity, accuracy, and reproducibility, this could provide a simple and suitable method for the quantitative determination of nanomole level of DA and DOX for clinical laboratories.