Introduction

The development of new electrochemical sensors that are sensitive, selective, portable, stable, and low-cost has gained notoriety mainly due to their high potential for application in environmental, pharmaceutical, food, forensic, and clinical areas [1, 2]. In this scenario, laboratory-made sensors stand out. These sensors are produced by different technologies, one of which is the production of composite electrodes from insulating and conducting materials that, after mixing, will present the desired properties [3,4,5,6].

Recently, the development of new composites has been a goal of increasing interest in the areas of electroanalytical and materials engineering. Different insulating materials are used in the composite fabrication, providing interesting mechanical characteristics such as resistance and viscosity [2, 7,8,9]. In that sense, the compound acrylonitrile butadiene styrene (ABS) stands out. It is a thermoplastic polymer that can be melted by heating and later reversibly solidified by cooling. This polymer, when used in the manufacture of a sensor, provides interesting physical characteristics such as resistance to impacts and tenacity. These properties qualify it as an excellent insulating material in the preparation of conductive composites [2, 7, 8].

Regarding conductive compounds, carbon-based materials, including graphite (Gpt), are widely employed [3, 10, 11]. The Gpt configuration is characterized by the presence of multiple flat layers, called graphene sheets. In each of these, the carbon atoms assume an sp2 hybridization with the existence of π bonds, which allows electronic mobility resulting in electrical conductivity [11, 12]. Additionally, other materials can be incorporated together with carbonaceous compounds seeking even better performances [4, 6]. In this context, the use of metallic oxides such as aluminum oxide can be highlighted.

Aluminum oxide (Al2O3), also called alumina, does not present electrical conductivity. Nevertheless, it can be used in the manufacture of composites thanks to its excellent dielectric properties, good thermal conductivity, high mechanical rigidity, and strong resistance to acids and bases [13,14,15]. Though alumina presence is not common in electrodes, its usefulness as a modifying agent in a polylactic acid (PLA) matrix has been demonstrated for minoxidil sensing [16]. Also, its use in conventional electrode modification has been reported in the literature for improved dopamine determination [17].

Dopamine (DOP), a neurotransmitter classified as a catecholamine, is produced by specific nerve cells and plays a crucial role in facilitating communication between neurons, feelings of euphoria, pleasure, motor coordination, and sexual motivation. Metabolic precursor of adrenaline and noradrenaline acts on specific receptors present in the central nervous system, mesenteric, renal, and coronary vessels. Atypical (high) levels of DOP can cause diseases such as Alzheimer’s and Parkinson’s disease [18]. Under physiological conditions in nervous and bodily fluids, the normal concentration of DOP is between 0.01 and 1 µmol L−1 [19]. Importantly, its use as a medication has been increasing considerably, and consequently, its appearance in water bodies, especially domestic or hospital effluents, is notorious [20, 21]. Residues of DOP in environmental waters can cause serious neurological and cardiac damage in animals [22]. As a result, it is highly required the development of sensitive methods for the determination of this analyte in several matrices.

Within these circumstances, numerous reviews have discussed the use of chemically modified electrodes, mainly with carbon-based materials such as graphene, carbon nanotubes, and carbon black for the electrochemical sensing of DOP [23,24,25,26]. Strategies with metallic nanomaterials, conductive polymers, and biological materials, as well as the combination of these with carbonaceous materials, have also been investigated [25, 27, 28]. Although an improvement in selectivity and sensitivity is noted, such approaches are time-consuming and require expensive materials, and the modification processes are complicated and non-reproducible, being impractical for routine analysis.

Thus, electrochemical methods which employ affordable and disposable sensors has opened up new ways for determining DOP [29]. Such methods bring various benefits, such as ease of use, cost-effectiveness, sensitivity, and capacity for real-time measurements. Therefore, this study concentrates on the effortless creation of a cost-effective alternative sensor employing ABS combined with Gpt and alumina. In contrast to a previously reported study in the literature [16], here we use ABS as the insulating component, which is economically more accessible, and offers greater mechanical resistance and durability compared with the PLA [30]. Also, the applicability of the proposed sensor was demonstrated in quantifying DOP levels in synthetic biological fluids (saliva and urine) and tap water by batch injection analysis coupled with amperometric detection (BIA-AD) using minimal sample preparation [31].

Experimental

Chemicals and solutions

All chemicals used were of analytical grade and were used without further purification. All solutions were produced with deionized water of resistivity at least 18.2 MΩ cm–1 from a Milli-Q water purification system Arium® Pro Sartorius (Göttingen, Germany). DOP (99.5% w/w), Al2O3, commercial Gpt powder (particles ≤ 20 µm), glucose (99.5 % w/w), caffeine (99 % w/w) were acquired from Sigma-Aldrich (St. Louis, EUA). Potassium chloride (KCl) (99.5% w/w) was purchased from Merck (Darmstadt, Germany), perchloric acid (HClO4) (72% v/v) from Hexis (São Paulo, Brazil). Acetone (98% v/v), sodium nitrite (97% w/w), sodium nitrate (99%, w/w), sodium chloride (99% w/w), ammonium chloride (99% w/w), calcium chloride dihydrate (78% w/w) from Labsynth (Diadema, Brazil). Ferricyanide ([Fe(CN)6]3−) and ferrocyanide ([Fe(CN)6]4−) (both 99.0% w/w), urea (99% w/w), citric acid (99% w/w), and sodium sulfate (99% w/w) were obtained from Vetec (Rio de Janeiro, Brazil). Pellets of ABS were acquired from GTMax 3D (São Paulo, Brazil).

A 0.1 mol L−1 HClO4 solution was used as supporting electrolyte for electrochemical measurements [32]. Stock and diluted solutions of DOP were prepared prior to the experiments employing background electrolyte in the presence of dissolved oxygen.

Fabrication of composite electrodes

The electrodes’ engineering was adapted from previous studies in the literature [3, 7, 16]. Summarily, commercial Gpt powder, ABS, and Al2O3 were added to beakers in the proportions provided in Table 1. These proportions were defined taking account the compromise between electroanalytical performance and physical characteristics necessary to the electrodes. After this, the mixtures were solubilized in acetone at room temperature (~ 25 °C). Subsequently, they were mixed manually, using glass sticks, for approximately 30 min, until their consistency was homogeneous. Next, the resultant pastes were introduced into commercial syringes of 1.0 mL (4.5 mm of internal diameter), which contained a piece of copper wire, used for performing electrical contact of the sensors. The syringes, filled with paste, were kept under pressure using a press for 7 days until acetone completely evaporated. After drying, the lower part of the syringe was cut so that the surface of the composite was exposed. Then, before being used, the electrode was polished with sandpaper of increasing number of grains (600 to 1500 grains of sand per cm2) and, finally, polished with a sheet of A4 paper until it was mirrored. Figure 1 shows a schematic representation of the steps involved in electrode material preparation.

Table 1 Proportions of conductive material (Gpt), insulator (ABS), and modifier (Al2O3) in the composite electrodes
Full size table
Fig. 1
figure 1

Schematic illustration of the composite electrode manufacturing process

Full size image

Instrumentation

Electrochemical experiments (cyclic voltammetry (CV), amperometry, and electrochemical impedance spectroscopy (EIS)) were conducted using a potentiostat/galvanostat Ivium CompactStat Technologies® B09118 (Eindhoven, Netherlands), operated by a desktop computer running IviumSoft 2.5 software for data acquisition and treatment. For all experiments, a pencil graphite, a miniaturized Ag|AgCl|KCl(sat) electrode [33], and a homemade composite electrode described in Sect. "Fabrication of composite electrodes" served as the counter, reference, and working electrodes, respectively.

In this study, for DOP detection, it was adopted a BIA-AD apparatus [31, 34]. For this, a lab-fabricated ABS-based BIA system was employed. The BIA apparatus consisted of an electrochemical cell with a capacity of 250 mL, a support and cover of the cell, and a micropipette positioner, all manufactured using 3D printing. This system stands out due to its reduced size and cost-effectiveness, making it highly suitable for on-site analysis [31]. For sample injections under hydrodynamic conditions, an electronic micropipette (Multipette R stream, Eppendorf, Hamburg, Germany) was employed. The micropipette’s tip was precisely positioned in a wall-jet configuration [31, 34] and allowed accurate injection volumes (from 10 to 1000 μL) under dispensing rates from 16.9 to 298.5 μL s−1. Fig. S1 presents the BIA system assembled with all the components.

Scanning electron microscopy (SEM) images were acquired using Vega 3 LMU (TESCAN, Brno-Kohoutovice, Czech Republic) at 20 kV, employing a secondary electron detector to analyze surface morphology. Energy dispersion X-ray spectra (EDX) were derived from the SEM images using the appropriate INCA X-Act detector from Oxford Instruments (Abingdon, UK).

Raman spectroscopy was performed using Ar ion laser at 532 nm and 2% incident power, employing a LabRAM HR Evolution microscope from HORIBA (Kyoto, Japan). The materials were also characterized by X-ray diffraction from Shimadzu XRD 6000 (Kyoto, Japan) using Cu Kα as the radiation source.

Thermogravimetric analysis (TGA) was conducted using a Shimadzu® TG/DTA 60 instrument (Kyoto, Japan). Approximately, 10 mg of each sample was placed in alumina crucibles. The samples were subjected to a heating program ranging from 25 to 1000 °C at a heating rate of 10 °C min⁻1, under a synthetic air atmosphere with a flow rate of 50 mL min⁻1.

Sample preparation

A synthetic urine sample was produced according to Brooks and Keevil’s procedure [35]. This sample was composed of citric acid (2.1 mmol L−1), urea (166.5 mmol L−1), sodium bicarbonate (25.0 mmol L−1), calcium chloride (0.4 mmol L−1), sodium chloride (89.0 mmol L−1), sodium sulfate (9.9 mmol L−1), magnesium sulfate (4.1 mmol L−1), ammonium chloride (24.3 mmol L−1), monopotassium phosphate (7.0 mmol L−1), dipotassium phosphate (9.0 mmol L−1), and sodium nitrite (0.7 mmol L−1), which resulted in a pH of 6.4. Alternatively, the artificial saliva sample was prepared according to John Mary and colleagues’ protocol [36]. This sample consisted of KCl (5.4 mmol L−1), sodium chloride (6.8 mmol L−1), calcium chloride (8.2 mmol L−1), sodium dihydrogen phosphate (5.1 mmol L−1), and urea (16.7 mmol L−1) at pH 6.6. A sample of tap water was collected on the laboratory premises.

Before analysis, the synthetic saliva and urine samples were diluted 20 times in the appropriate background electrolyte. The tap water sample was diluted twice. Recovery studies using two concentration levels, 20 and 40 µmol L−1, were conducted to check the method’s accuracy.

Results and discussion

Composite material characterization

The chemical composition of the electrodes (E0 and E6) was confirmed by elemental analysis using EDX derived from the SEM images (Fig. 2A–C). In the case of E0, peaks indicating the presence of carbon from Gpt, as well as silicon and oxygen from the substrate (silicon-wafer), were observed. Conversely, for E6, peaks indicating carbon from Gpt and signals associated with aluminum and oxygen from the Al2O3 structure were detected. EDX mapping revealed that the presence of Al2O3 was dispersed intermittently along the electrode.

Fig. 2
figure 2

EDX spectra of E0 (A) and E6 (B) electrode surfaces; EDX mapping of E6 composite material from the SEM image region displayed in C1 with C2, C3, and C4 indicating the Al, C and O components, respectively (C); Raman (D) and FT-IR (E) spectra and XRD patterns (F) of E0 (black) and E6 (red) electrodes, and thermogravimetric analysis of the materials (G)

Full size image

Figure 2D displays the Raman spectra of the E0 and E6 electrode surfaces. All samples exhibit three distinct peaks at 1345, 1594, and 2680 cm−1, corresponding to D, G, and 2D bands, respectively, all characteristic of the Gpt structure on the electrode. The G band corresponds to the optical E2g phonons at the Brillouin zone center, representing the bond stretching of sp2 carbon pairs within both rings and chains. The D band corresponds to the breathing mode of carbon atoms in aromatic rings, indicating the presence of defects, while the 2D band is associated with the presence of multilayer graphene within the graphite structure [3, 37, 38]. Moreover, the FTIR spectrum of the materials reveals characteristic peaks from ABS (Fig. 2E). The stretching vibrations of the aliphatic C-H bonds in ABS are observed in the range of 3000–2800 cm−1. The absorption at 1491 cm−1 corresponds to the stretching vibration of the aromatic ring from the styrene unit. The CH2 group deformation is identified at 1446 cm−1. The deformation of C-H bonds attached to alkenic carbons appears at 965 cm−1 and 705 cm−1 for 1,4-butadiene units and at 911 cm−1 for 1,2-butadiene units [39, 40].

The XRD peaks for the E0 and E6 electrodes are depicted in Fig. 2F. Miller indices highlighted in green correspond to the crystalline structure of Gpt, while those in blue indicate the phase attributed to the rhombohedral structure with space group R-3c (JCPDS no. 10–0173) of α-Al2O3 found in the E6 electrode [17, 41].

TGA analyses were also carried out to check whether the Gpt and Al2O3 have been incorporated into the ABS polymer matrix and to assess the thermal stability of the composite created. Thus, TGA curves were recorded for ABS (black line), Gpt (blue line), Al2O3 (red line), and Gpt/Al2O3/ABS (electrode E6, green line) as shown in Fig. 2G. ABS degrades in two steps. The first step begins at 350 °C, ends at 530 °C, and the second from 530 until 620 °C, as observed elsewhere [42, 43]. On the other hand, Gpt suffers significant mass loss between 640 and 900 °C [44], while Al2O3 does not degrade thermally until 1000 °C [45]. Given that the composite developed (Gpt/Al2O3/PLA) suffered 22.75% mass loss between 350 and 620 °C, 65% between 635 and 915 °C, and that 12.25% mass remained, since alumina does not degrade in the temperature range allowed by the equipment, it can be asserted that the process of synthesizing the composite was successful.

SEM analysis was conducted to examine the morphology of E0 and E6 electrodes at various magnifications. Figure 3A and B shows pronounced roughness, characteristic of the Gpt materials in E0. In contrast, Fig. 3C and D reveal the existence of distinct regions on the electrode surface attributable to the presence of Al2O3 nanoparticles in E6. This observation aligns with EDX mapping, which illustrates that Al2O3 nanoparticles are intermittently distributed across the electrode’s surface.

Fig. 3
figure 3

SEM images acquired of E0 (A and B) and E6 (C and D) electrode surface

Full size image

Electrochemical measurements

Preliminary studies via CV were conducted using DOP as the proof-of-concept molecule in a 0.1 mol L−1 HClO4 medium (supporting electrolyte), as shown in Fig. 4. These electrochemical measurements aimed to understand the influence of Al2O3 on the electrochemical response of composite electrodes. The first electrode tested was E0 (control electrode, black line), which exhibited discrete electrochemical processes for DOP. On the other hand, there was a significant improvement in anodic (Ipa) and cathodic (Ipc) peak currents, as well as peak-to-peak separation (ΔEp), as the amount of Al2O3 was increased. The better electrochemical behavior was achieved when the E6 electrode was used, a material set at 12.25% w/w Al2O3. Above this level, the composite material’s mechanical stability and electrochemical performance were significantly compromised. Comparing E6 with the one without the presence of aluminum oxide (E0 – control electrode), it is possible to observe that Ipa increased more than four times, together with a more than threefold decrease in ΔEp. A summary of Ip’s, Ep’s, and ΔEp’s obtained from Fig. 4 is provided in supplementary material (Table S1). The electrochemical oxidation of DOP on carbon-based surfaces involves two electrons and two protons, as well documented in the literature [23] (see inset in Fig. 4).

Fig. 4
figure 4

CVs obtained in the presence of 1.00 mmol L−1 DOP in 0.1 mol L−1 HClO4 (supporting electrolyte) for E0 (black line), E1 (green line), E2 (blue line), E3 (cyan line), E4 (magenta line), E5 (yellow line), and E6 (red line) electrodes. The corresponding blanks are shown in dashed lines. The inset illustrates the electrochemical oxidation mechanism of DOP. Experimental conditions: scan rate of 100 mV s−1; step potential of 5 mV

Full size image

Also, CVs in the presence of ferri-ferro cyanide, a well-known electrochemical redox probe, were recorded to check the performance of the electrodes for inorganic species. These results are provided in supplementary material (Fig. S2 and Table S2). From these results, it is possible to state that the same trend was reached when compared with the DOP, since the greater amount of Al2O3 in the electrode provided better electrochemical responses.

The electroactive surface areas for the E0 and E6 electrodes were estimated employing Randles–Sevcik’s protocol for a reversible and diffusion-controlled electrochemical process (Eq. 1) [3, 46], in which Ip represents the peak current, D the diffusion coefficient for K3Fe(CN)6 (6.39 × 10−6 cm2 s−1), n the number of electrons involved in the process (n = 1), A the electroactive area, C the concentration of electroactive species, and v the scan rate.

$$Ip=\;2.69\;\times\;10^5n^{3/2}AD^{1/2}Cv^{1/2}$$
(1)

Therefore, CVs in the presence of 1:1 mmol L−1 K3Fe(CN)6/K4Fe(CN)6 in 0.1 mol L−1 KCl (background electrolyte) were recorded at scan rates varying from 10 to 80 mV s−1. Fig. S3 displays these voltammetric responses. The acquired area values were 4.7 and 14.5 mm2 for E0 and E6 electrodes, respectively, i.e., a threefold increase using Al2O3-containing electrode. In addition, EIS measurements were carried out using both electrodes and the obtained Nyquist plots and a typical Randles equivalent circuit are shown in Fig. 5B. In both spectra, the formation of a semicircular portion related to the resistance to charge transfer (Rct) followed by a linear profile characteristic of diffusional behavior was observed. Considering the smaller diameter of the semicircle portion of the Nyquist plot for E6, it can be assessed that this electrode provided a facilitated electronic transfer for the ferri/ferro redox couple compared with the electrode without aluminum oxide (E0) [47, 48]. The achieved Rct values were 1990 Ω and 11,500 Ω for E6 and E0, respectively, meaning that E6 reduced Rct around 83%. These results agree with those shown in Fig. 5A, since the CVs obtained for the ferri/ferro cyanide showed better responses for E6 (red line), in terms of Ip’s, and ΔEp’s, if compared with the E0 (black line).

Fig. 5
figure 5

(A) CVs registered using E0 (black line) and E6 (red line) in the absence (dashed lines) and presence (solid lines) of 1:1 mmol L−1 ferri-ferro cyanide in 0.1 mol L−1 KCl (supporting electrolyte). (B) Nyquist plot from EIS spectra obtained for E0 (black circle) and E6 (red square). The inset shows the respective Randles equivalent circuit. CV experimental conditions: scan rate of 100 mV s−1 and step potential of 5 mV. EIS conditions: the applied potential of + 0.22 V, frequency range from 50 kHz to 0.01 Hz with a signal amplitude of 10 mV, and 10 data points per frequency decade

Full size image

The heterogeneous electron transfer (HET) rate constants (\({k}^{0}\)) were also calculated for E0 and E6 electrodes, employing the method of Nicholson (Eq. 2) [46, 49].

$$\Psi ={k}^{0}\frac{1}{\sqrt{\left(\frac{\pi Dn\nu F}{RT}\right)}}$$
(2)

In which \(\Psi\) is a kinetic parameter, D is the diffusion coefficient of K3Fe(CN)6 (6.39 × 10−6 cm2 s−1), n the number of electrons involved in the process (1), F corresponds to Faraday constant, R to gas constant (8.314 J/mol K), and T to temperature (298 K). The function \(\Psi \left(\Delta {E}_{p}\right)\) is given by Eq. 3, which was used to determine \(\Psi\) from experimentally recorded voltammograms [46], by inserting data already shown in Table S2. Then, the linear plot of \(\Psi\) versus \(\frac{1}{\sqrt{\left(\frac{\pi Dn\nu F}{RT}\right)}}\) at different scan rate values is used to calculate \({k}^{0}\).

$$\Psi =\frac{-0.6288+0.0021n{\Delta \text{E}}_{\text{p}}}{1-0.017n\Delta {\text{E}}_{\text{p}}}$$
(3)

\({k}^{0}\) values were 0.000279 cm s−1 and 0.00179 cm s−1, for E0 and E6, respectively. So, as E6 presented a heterogeneous electron transfer rate constant 6 times higher than E0, it can be assessed that the presence of Al2O3 positively impacted the kinetics, corroborating EIS results. These improvements in the electrochemical properties of the Al2O3-loaded electrode are in line with previous studies [14, 17]. As a result, the E6 electrode was employed in further experiments.

DOP’s electrochemical behavior

Previous studies reported in the literature have shown that HClO4 is a useful supporting electrolyte for the electrochemical oxidation of DOP [32]. Thus, the influence of its concentration on the electrochemical response of DOP was studied at 0.001, 0.05, 0.1, and 0.5 mol L−1 (Fig. S4). As can be seen, a more reversible electrochemical behavior was achieved using 0.1 mol L−1 (green line), and for this reason, the HClO4 was kept at this concentration level in the next measurements.

The mass transport regime of DOP on the E6 electrode was examined varying scan rates from 10 to 100 mV s−1 (Fig. S5A). Considering the oxidation process at + 0.575 V vs. Ag|AgCl|KCl(sat), a linear adjustment between log I and log v (Fig. S5B, r2 = 0.997) was obtained with a slope value of 0.70, suggesting both diffusion- and adsorption-controlled mass transport phenomenon [50]. Subsequently, the electrode’s reproducibility was also investigated by CV using three (n = 3) different E6 electrodes in the presence of the ferri/ferro redox probe and DOP (Fig. S6A–B). Relative standard deviation (RSD) values of 5.3 and 5.7% were achieved for the anodic peak currents of ferri/ferro and DOP, respectively. These findings suggest that the construction of our device is reproducible. Furthermore, successive measurements (n = 50) using a single electrode (Fig. S6C) resulted in an RSD of 1.7%, indicating the high repetitiveness of the electrochemical response of the proposed device. Additionally, the electrode demonstrated stable performance over a 5-day period (n = 5), with an RSD of 2.8%, when stored under ambient temperature and humidity conditions (Fig. S6D).

Electrochemical determination of DOP by BIA-AD

The BIA-AD system has been selected for the development of a method aimed at detecting DOP. This analytical system offers several advantages over stationary voltammetric measurements, such as (i) high analytical frequency, (ii) minimal sample and reagent consumption, (iii) increased sensitivity, and (iv) reduced surface fouling effects [31]. Initially, its instrumental parameters (working potential, dispensing rate, and injection volume) were investigated individually through authentic replicas (n = 3). Hydrodynamic voltammetry in a potential range between 0.4 V and 1.2 V was accomplished to attest to the best potential for DOP detection (Fig. S7A). As can be seen, appropriate current values and lower deviations were achieved when + 0.7 V was employed. Consequently, this potential was used in further experiments. Other conditions for the BIA-AD technique, such as dispensing rate (16.9 to 298.5 μL s−1) and injection volume (25 to 125 μL) were also evaluated (Fig. S7B and C, respectively). Aiming for a compromise between precision, analytical signal, and lower sample consumption, the selected conditions were 227.3 μL s−1 and 75 μL for dispensing rate and injection volume, respectively. The summary of these results is provided in Table S3. Under BIA-AD selected conditions, the analytical features of the proposed method were estimated. Calibration curves in increasing and decreasing direction of DOP levels in the range of 10 to 1000 μmol L−1 were constructed (Fig. 6A), and the respective linear regressions are shown in Fig. 6B. Slope values of 0.308 ± 0.002 and 0.295 ± 0.006 µA L µmol−1 were obtained for ascending and descending concentration orders, respectively. Using the paired Student’s t-test at 95% confidence, there was no significant evidence between the slope values since tcalc (2.55) was lower than tcrit (4.303). Thus, it can be stated that there were no significant surface fouling effects even in a wide range of DOP concentrations. Additionally, a comparison between the analytical curves obtained for E0 and E6 electrodes was performed (Fig. S8). While the E6 provided wide linear behavior between 10 and 1000 µmol L−1 (r2 > 0.99), E0 generated a smaller linear range (10 to 200 µmol L−1 (r2 > 0.99). Furthermore, E6 furnished greater sensitivity to the calibration curve, when compared with E0.

Fig. 6
figure 6

(A) Electrochemical responses for consecutive injections (n = 3) of DOP standard solution (a–o: 10 to 1000 µmol L−1) in increasing and decreasing concentrations using the BIA-AD system and E6 electrode; (B) respective calibration plots. BIA-AD conditions in Table S3.

Full size image

Repeatability was studied using 15 measurements alternating quintuplicate injections at two DOP concentration levels (20 µmol L−1 and 50 µmol L−1) upon the E6 electrode, as shown in Fig. S9. The RSD values resulted in acceptable values, lower than 5% for both concentration levels, confirming the method’s precision [51]. The limit of detection (LOD) was estimated according to IUPAC recommendations (S/N = 3) [52]. Table 2 summarizes the figures of merit for the proposed methodology.

Table 2 Analytical figures of merit obtained for the proposed method for DOP detection using the E6 electrode and BIA-AD
Full size table

Interference studies were also conducted, in the presence of compounds such as glucose, nitrate, urea, caffeine, and nitrite, expected to coexist with the analyte, in synthetic biological fluids (saliva and urine) or tap waters. Variations in the electrochemical response of 20 µmol L−1 DOP in the absence and presence of these concomitants were tested, using the ratios 1:1 and 1:5 between DOP and the interfering agents (Fig. S10). It can be noticed that these substances did not significantly influence the electrochemical response of the DOP (variations lower than 15%) [51], which demonstrates the method’s selectivity for applications in biological fluids and tap water.

The method’s accuracy was evaluated in terms of recovery tests, spiking the sample solutions with 20 and 40 µmol L−1 DOP. Figure 7 and Table 3 show the results for the quantification of DOP and the corresponding recovery values in synthetic biological fluids and tap water by the BIA-AD system. The recoveries between 90.5 and 107.3 attested the accuracy of the proposed method [51].

Fig. 7
figure 7

BIA-AD recordings for injections of increasing concentrations of DOP (a) 10, (b) 20, (c) 30, (d) 40, (e) 50, and (f) 60 µmol L−1, application of the method to non-spiked samples of synthetic saliva (S1), tap water (S2), and synthetic urine (S3) and spiked samples at two concentration levels: 20 µmol L−1 (F1) and 40 µmol L.−1 (F2). BIA-AD conditions in Table S3.

Full size image
Table 3 Recoveries obtained for DOP in synthetic biological fluids and tap water and samples by the BIA-AD method
Full size table

Table 4 presents a literature overview of different electrochemical methods for DOP determination considering technique, sensor, LOD, and linear range. As noted, the proposed method presents comparable results to others already reported in the literature in terms of LOD and is even better in terms of linear range. Additionally, it is important to highlight that the method here developed is fast (182 analyses per hour) and portable, offering an economical alternative for field analysis. Importantly, the proposed composite material is easy to prepare and can be an interesting and promising option for additively manufacturing accessible and enhanced electrodes.

Table 4 Literature overview for different electrochemical methods towards DOP determination
Full size table

Conclusion

This work proposed the manufacturing of an electrochemical sensor based on ABS, Gpt, and Al2O3 for DOP monitoring. Structural, morphological, and elemental characterizations combined with electrochemical measurements indicated the successful incorporation of Al2O3 into the composite material, as well as its positive impact on the DOP response. The proposed sensor was coupled with BIA-AD, a high-throughput tool, to determine DOP in biological fluids (saliva and urine) and tap water samples. Great analytical features were acquired, and such results were comparable to others already present in the literature. Furthermore, the sensor exhibited selectivity towards biomarker compounds and other species found in water. Hence, it is possible to conclude that our analytical strategy could be useful for DOP sensing in biological samples and water, given that it is simple, fast, affordable, accurate, precise, and portable.