Introduction

Cadmium (Cd) is a toxic heavy metal that has multiple toxic effects on the human kidney, liver, nervous and cardiovascular systems1. With the increase of industrial activities such as smelting operations, electroplating, pigments, fertilizers and mining, a large amount of industrial wastewater containing cadmium was discharged without treatment, seriously polluting drinking water and soil2. The consumption of cadmium-contaminated water and rice poses a great threat to human health3. Therefore, sensitive and rapid determination of cadmium in food is of great significance for environmental protection and food safety.

Conventional methods for cadmium detection include atomic absorption spectrometry (AAS), atomic fluorescence spectrometry (AFS) and inductively coupled plasma mass spectrometry (ICP-MS)4,5,6. Although these methods possess high sensitivity, accuracy and stability, they need complex instrument operation, long operation time and high operating cost, which are not suitable for rapid field detection7. Anodic stripping voltammetry is considered as a powerful tool for heavy metal detection8,9. In this method, Cd2+ is enriched to the electrode surface at a constant potential and reduced to Cd0. Then Cd0 is oxidized to Cd2+ under scanning voltage and stripped out into the electrolyte to generate a current. The current value is proportional to the content of metal ions, and each metal ion has its characteristic oxidation potential10,11.

Electrode modification is an important means to improve the analytical sensitivity of electrochemical sensors. Mercury film is the conventional material for electrode modification, however, mercury toxicity has restricted its use12. Alternative functional materials such as bismuth (Bi), antimony (Sb), and silver (Ag) raised attention13,14. Among them, bismuth modified electrode has been widely used in ASV-based electrochemical sensors because of its low toxicity, insensitivity to dissolved oxygen and wide operating potential window that is close to mercury electrode15,16,17. There are many cases where sensitive Cd determination and the simultaneous determination of Cd, Pb and Zn were achieved using bismuth modified electrode18,19,20. The modification of bismuth can be finished by in-situ deposition, ex-situ deposition or dropping ways21,22. In-situ deposition owns great advantage for that the co-deposition of Bi3+ and Cd2+ can enhance the enrichment of Cd2+ by forming Bi-Cd alloy, thus improving the detection sensitivity23. Besides, in-situ deposition is easy handling and time-saving, avoids the risk of degradation and uneven distribution of modified bismuth24.

Except for metal material modification, electrochemical activation can also enhance the detection sensitivity of electrode25. Pre-anodization is an effective method for electrode activation by accelerating the electron transfer between substance and electrode26,27. It has been used in the activation of screen-printed electrode to remove the organic binders during electrode preparation, thus enhancing the conductivity of electrode28. pre-anodization owns the advantages of less reagent and simple operation29 Moreover, to achieve point-of-need quantitation of heavy metal Cd2+, a portable and low-cost device capable of transmitting, collecting and displaying electrochemical signals is urgently required.

Herein, by combining pre-anodization technique and in-situ deposition method, an in-situ bismuth modified pre-anodized screen-printed carbon electrode (Bi/Pre-anodized SPCE) for Cd2+ determination was prepared for the first time, and the detection was conducted on a portable, self-made and low-cost potentiostat coupled with a stirring device (Fig. 1). The method is easy-handling and suit for point of care testing (POCT). The pre-anodization was carried out in PBS (pH = 9) by cyclic voltammetry (CV) method. Then Bi3+ and Cd2+ were co-deposited on the pre-anodized electrode and Cd2+ was stripped out for determination by square wave anodic stripping voltammetry (SWASV). The developed electrochemical sensor demonstrated good performance for Cd2+ determination with limit of detection as low as 0.15 μg/L and 3.55 μg/L using commercial and self-made potentiostat, respectively. The test could be completed within 3 min with good repeatability and specificity. The recovery rates in water and rice samples ranged from 91.7 to 107.1%.

Figure 1
figure 1

Schematic diagram of Cd2+ determination with in-situ bismuth modified pre-anodized SPCE and portable device. The stirring device contains an SPCE connector, a sample cell and a time- and speed-regulated stirring motor. The self-made PSoC Stat potentiostat connects to SPCE connector by dupont wire. The SPCE was pre-anodized in PBS (pH = 9) by cyclic voltammetry to active its surface. Then Bi3+ and Cd2+ were co-deposited on pre-anodized electrode, and Cd2+ was stripped out for determination. The stirring function is applied in the deposition process.

Full size image

Materials and methods

Reagents and materials

Sodium acetate, sodium bromide, acetic acid and hydrogen peroxide (w/w 30%) were purchased from Guangzhou Chemical Reagent Factory. K3[Fe (CN6)] and K4[Fe (CN)6] were purchased from Shanghai Maclin Biochemical Technology Co., LtD. All the chemicals used in the study were analytical agent grade. The 0.22 μm filter membrane was purchased from Beekman Biological Corporation. Stock solution of bismuth ion and cadmium ion (1000 μg/mL) were purchased from Beijing General Research Institute of Non-Ferrous Metals. Screen-printed carbon electrode (diameter 2.8 mm) was obtained from Wuhan Zhongke Zhikang Biotechnology Co., LtD. All solutions were prepared using ultrapure water produced by Milli-Q ultrapure water (Millipore Company, USA).

Instruments

Scanning electron microscope (SEM) images were acquired on a a field emission scanning electron microscope (Quattro S, Thermo Fisher Scientific, USA). Energy dispersive X-ray (EDX) spectrum was collected on an energy dispersive spectrometer (Ultim Max Oxford Instruments LtD., U.K.). Electrochemical impedance spectroscopy (EIS) was performed with a potentiostat (Autolab 302N, Metrohm Autolab B.V., Switzerland). Other electrochemical measurements were performed on a CHI1440 constant potential instrument (Shanghai Chenhua Instrument Co., LtD.) and a self-made PSoC Stat potentiostat (Fig. S1). A stirring device containing an SPCE connector, a sample cell and a time- and speed-regulated stirring motor was self-made for the experiment (Fig. S2).

Experimental methods

Pre-anodization of screen-printed carbon electrode

The screen-printed electrode was pre-anodized by cyclic voltammetry method. The SPCE was scanned for 5 cycles in 0.1 mol/L PBS phosphate buffer solution (pH = 9), with a scanning range of 0.5–1.7 V and a scanning rate of 0.1 V/s. Then the SPCE was rinsed thoroughly with ultra-pure water and dried at room temperature.

SWASV for Cd2+ determination

1 mL of 0.1 mol/L acetate buffer (pH 4.5) containing 150 μg/L Bi3+, 20 μmol/L NaBr and a certain concentration of Cd2+ were added into the sample cell. Then the pre-anodized SPCE were immersed into the solution. The detection parameters of SWSV were set as follows: deposition potential is − 1.4 V, deposition time is 180 s, potential increment is 4 mV, amplitude is 25 Hz, scanning range is − 1.4 to − 0.2 V. In the deposition process, the sample cell was rotated by the stirring motor with a stirring rate of 200 rpm. No stirring was provided in the stripping process. All tests were performed at room temperature (26 ± 0.5 °C).

Recovery study

Tap water and rice samples without Cd were selected for recovery studies. The tap water was filtered by 0.22 µm microporous filter membrane. Filtered water was adjusted to pH 4.5 by nitric acid, then mixed with an equal volume of 0.2 mol/L acetate buffer solution (pH = 4.5) to obtain the pre-treated tap water sample.

The pretreatment of rice samples was referred to the method reported by Wang30. The rice samples were crushed into powder and dried in an oven for 2 h. 1 g rice powder was dissolved in 10 mL nitric acid and boiled until nearly dry. Then 3 mL H2O2 was added and heated until dry. The residue was dissolved in 25 mL 0.1 mol /L acetic acid. After complete vortex oscillation, the solution was filtered by 0.22 μm membrane. The filtrate was adjusted to pH 4.5 by NaOH solution to obtain the pre-treated rice extract.

Standard solution of Cd2+were added into the pre-treated tap water or rice extract. Then the spiked samples were analyzed by both the established electrochemical method described in 2.3.2 and ICP-MS.

Results and discussion

Study of electrode pre-anodization

The effect of solution type on pre-anodization was firstly evaluated. The screen-printed electrode was pre-anodized with 0.1 mol/L HNO3, HCl, H2SO4, NaOH, PBS (pH = 5), PBS (pH = 7) and PBS (pH = 9), respectively. Then the electrochemical characteristics of the electrodes were evaluated by cyclic voltammetry in 5 mmol/L [Fe (CN)6]3−/4− containing 0.1 mol/L KCl. The scanning range was − 0.6–1.0 V, the scanning rate was 0.1 V/s. As shown in Fig. 2a, compared with HNO3, HCl, H2SO4 and NaOH, the highest redox current and smallest peak potential difference were obtained when using 0.1 mol/L PBS (pH = 9), and the peak current increased by 50% in PBS compared with that in H2SO4, indicating an excellent electron transfer ability. Subsequently, PBS solution with different pH values were compared. As shown in Fig. 2b, PBS (pH = 9) demonstrated the best performance. In addition, compared with bare electrode, pre-anodization definitely improves the electron transfer ability. This is because that pre-anodization treatment can clean the impurities attached to the electrode surface, promote the electron transfer between substance and electrode28. In addition, it can significantly increase the carbon–oxygen functional groups on the working electrode surface, produce defect/edge like sites, and increase the capacitive current26. PBS buffer solution with pH = 9 provides a suitable alkaline environment for pre-anodization, which is conducive to the occurrence of this process29.

Figure 2
figure 2

Optimization of pre-anodization conditions. (a) CV curves of SPCE pre-anodized in different solutions. (b) CV curves of SPCE pre-anodized in PBS with different pH. (c) CV curves of SPCE treated with different scanning turns. (d) The relationship between the number of scanning cycles and redox current values.

Full size image

Next, the effect of scanning turns in pre-anodization was studied. SPCE was pre-anodized in 0.1 mol/L PBS solution (pH = 9) for different scanning cycles, then the SPCEs were characterized by CV in 5 mmol/L [Fe (CN)6]3−/4− containing 0.1 mol/L KCl. As shown in Fig. 2c, as the number of scanning cycles varies from 0 to 5, the redox current increases significantly and the peak potential difference decreases markedly. This is because that the impurities on the electrode surface were removed and the carbon–oxygen functional groups increased in the pre-anodization process, which improved the electron transfer ability of electrode. When the number of scanning turns exceeds 5, the peak potential difference is basically unchanged and the redox current decreases slightly (Fig. 2d), which is due to elimination of impurities and the saturation of carbon–oxygen functional groups on working electrode surface. Therefore, scanning turn was set as 5 in the pre-anodization of SPCE.

Morphological characterization and element analysis of SPCE

SEM and EDX were used to characterize the surface of SPCE during modification process. As shown in Fig. S3a, c, e, pre-anodization and bismuth modification have little effect on the morphology of electrode surface. Since pre-anodization treatment has a good activation effect on SPCE, it indicates that the main mechanism of activating SPCE by pre-anodization is not through factors that cause great changes in the morphology of electrode. The in-situ deposited bismuth cannot be observed directly through SEM images. It is attributed to the small size and low content of bismuth. EDX spectrums demonstrate that there were no elemental changes after pre-anodization, while Bi and Cd occurred after in-situ deposition of bismuth (Fig. S3b, d, f), which is consistent with the modification process. The pre-anodized and bismuth modified electrodes will be further characterized by electrochemical methods.

Electrochemical characterization of SPCE

Electrode performance evaluation with cyclic voltammetry

Pre-anodized SPCE was characterized by cyclic voltammetry in 5 mmol/L [Fe (CN)6]3−/4− and 0.1 mol/L KCl solution at different scanning rates. The scanning range is − 0.6–1.0 V, the scanning rate is 0.05–0.3 V/s. The redox peak current enhanced with the increasing scanning rate (Fig. 3a). Ipa and Ipc were linearly correlated with the square root of scan rate (Fig. 3b), indicating that the redox process is mainly based on diffusion-controlled reactions31.

Figure 3
figure 3

(a) Cyclic voltammetry curves of pre-anodized SPCE at different scanning rates. (b) The relationship between redox peak current value and the square root of scanning rate. Cyclic voltammetry curves (c) and Electrochemical impedance spectroscopy curves (d) of screen-printed electrode with different modifications.

Full size image

Then different electrodes were evaluated by cyclic voltammetry with scanning rate set as 0.1 V/s. As shown in Fig. 3c, compared with bare electrode, the redox current increased nearly three times and the peak potential difference decreased after pre-anodization, which is attributed to the improved electron transfer ability by pre-anodization treatment. On this basis, in situ deposition of metallic bismuth further improves the peak current. In addition, the presence of NaBr in bismuth modification enhanced the peak current. It is because that Br− can complex with Bi3+ and improve the deposition of Bi on carbon-based electrode32,33. The EIS result was consistent with the CV test for that the Rct decreased with the electrode modification (Fig. 3d). This suggests that pre-anodization and bismuth modification improve the charge transfer rate of screen-printed electrode34.

Furthermore, the effective area of the electrode was calculated by Randles-Ševčik equation 31. The equation is shown in (1).

$$ I_{{\text{P}}} = {2}.{69} \times {1}0^{{5}} {\text{n}}^{{{3}/{2}}} AD^{{{1}/{2}}} v^{{{1}/{2}}} C $$
(1)

Ip is the redox peak current (A), n is the number of transferred electrons (n = 1), A is the active surface area of working electrode (cm2), D is the diffusion coefficient of electroactive substance (cm2/s), C is the concentration of electroactive substance (mol/cm3) and v is the scan rate (V/s). For 5 mmol/L of [Fe (CN)6]3−/4−, n = 1, and D = 6.30 × 10−6cm2/s. After calculation, the effective area of pre-anodized SPCE (0.057 cm2) is about 2.59 times that of the bare electrode (0.022 cm2), and further increased after the in-situ deposition of bismuth, reaching 0.064 cm2 (without NaBr) and 0.078 cm2 (with NaBr), respectively. The results demonstrated that the active surface area of the screen-printed electrode is enlarged by pre-anodization and bismuth modification, which is beneficial to improve the sensitivity of the electrochemical sensor.

Electrochemical response of different electrodes towards Cd2+

Bare SPCE, in situ bismuth modified electrode (Bi-SPCE), Bi/Pre-anodized SPCE without NaBr (Bi/Pre-anodized SPCE (-NaBr)), and Bi/Pre-anodized SPCE containing NaBr (Bi/Pre-anodized SPCE(+ NaBr)) were used to detect Cd2+ by SWASV. As shown in Fig. 4, when Cd2+ was detected by bare electrode, an irregular stripping peak of Cd2+ appeared and the signal was weak. After modification of bismuth, the stripping peak shape of Cd2+ is good and the peak height increased. In addition, the stripping peak of Bi3+ was also observed at − 0.42 V, indicating that bismuth was successfully modified onto electrode through in-situ deposition. This result demonstrates that the co-deposition of Bi3+ and Cd2+ promotes the enrichment of Cd2+35. When Cd2+ was detected using a Bi/Pre-anodized SPCE without NaBr, the stripping peak current of Cd2+ increased significantly. The enhancement in Cd2+ peak current benefits from the superior electron transfer ability of pre-anodization treatment. When a Bi/Pre-anodized SPCE containing NaBr was used, the highest stripping peak signal and the best peak shape of Cd2+ was obtained, indicating good sensitization effect of NaBr, for that the complexation of Bi3+ and Br- can improve the deposition of Bi on carbon-based electrode32,33. Taken together, benefited from co-deposition of Bi, pre-anodization activation and sensitization of NaBr, Bi/Pre-anodized SPCE containing NaBr demonstrated the best performance for Cd2+ determination. As for the stripping peak of Bi, the shifts in peak potential were due to the changes in electrode characteristics by different treatments. In addition, the stripping peak height is lower than that of Cd, indicating the stripping condition is more favored for Cd. Moreover, the stripping peak height of Bi was not consistent in the electrodes with different modifications. This may be due to the different balance in the deposition and stripping amount of Bi in different electrodes.

Figure 4
figure 4

Square wave stripping voltammetry curves of Cd2+ with different electrodes.

Full size image

Optimization of experimental parameters

In order to obtain the best detection effect of Bi/Pre-anodized SPCE for Cd2+ determination, several experimental parameters including Bi3+ concentration, NaBr concentration, electrolyte type, pH value of electrolyte, deposition potential and deposition time, stirring rate were studied.

Concentration of Bismuth and NaBr

Firstly, the effect of Bi3+ content on the stripping peak current of Cd2+ was studied. As shown in Fig. 5a, the peak current increases with gradual Bi3+ concentration ranging from 0 to 150 μg/L, and reached a plateau at 150 μg/L. Therefore, Bi3+ content of 150 μg/L was selected for subsequent experiments. Then, the influence of NaBr content on stripping peak current of Cd2+ was explored. As shown in Fig. 5b, the stripping peak current of Cd2+ rose with the increased NaBr concentration ranging from 0 to 20 μmol/L, and kept relatively stable as the NaBr concentration rose continuously. Therefore, the optimal concentration of NaBr was 20 μmol/L.

Figure 5
figure 5

Optimization of experimental parameters for Cd2+ determination. (a) Bi3+ content. (b) NaBr content. (c) Supporting electrolyte. (d) ABS with different pH. (e) Deposition potential. (f) Deposition time.

Full size image

Electrolyte types and pH values

The influence of supporting electrolyte type and pH value on stripping peak signal of Cd2+ was studied. Supporting electrolyte type was tested among 0.1 mol/L acetate buffer solution (ABS, pH = 4.5), phosphate buffer solution (PBS, pH = 7), HCl, NaOH and KCl. As shown in Fig. 5c, the highest stripping peak current was obtained when using acetate buffer solution. Furthermore, 0.1 mol/L acetate buffer solution with different pH values were investigated (Fig. 5d). The optimal Cd2+ stripping peak current was obtained when the pH of the acetate buffer solution is 4.5. Too low pH will easily lead to hydrogen evolution on the surface of working electrode and reduce the stripping response; Cd2+ is prone to hydrolysis under high pH, resulting in a decrease in the stripping peak current35. Therefore, the electrolyte was selected as 0.1 mol/L acetate buffer solution with a pH of 4.5.

Deposition potential and deposition time

The effects of deposition potential on the stripping peak current of Cd2+ were investigated. As shown in Fig. 5e, when the deposition potential moved from − 1.4 to − 1.8 V, hydrogen evolution was easy to occur under a low potential, and the hydrogen bubbles formed on the electrode surface hindered the stripping of Cd2+. Meanwhile, when the potential moved from − 1.4 to − 1.0 V, the enrichment of Cd2+ was impaired for the deposition potential was close to the stripping potential of Cd2+, resulting in a decrease in stripping peak current. Therefore, the deposition potential was set as − 1.4 V.

Then, deposition time was investigated. The stripping peak current of Cd2+ rose with the increase of deposition time, and presented a linear relationship after 120 s (Fig. 5f), which is due to the adsorption balance of Cd2+ between electrode surface and solution. Although increasing the deposition time can enhance stripping peak current and reduce the limit of detection, a longer time will prolong the detection period. Considering detection sensitivity as well as detection efficiency, the deposition time was set as 180 s.

Stirring rate

Concentration polarization can be reduced by stirring during the deposition process, which is beneficial to the enrichment of Cd2+. The effect of stirring rate was investigated and adjusted by our self-made stirring device. As shown in Fig. S4, the stripping peak current increased significantly with the increase of stirring rate and reached a plateau at 200 rpm. Therefore, 200 rpm was used in the deposition process.

Analytical performance of Bi/Pre-anodized SPCE for Cd2+ determination with portable potentiostat and stirring device

Under optimal experimental conditions, the Bi/Pre-anodized SPCE (+ NaBr) was used to detect different concentrations of Cd2+ through SWSV. The experiment was conducted on commercial potentiostat and portable PSoC Stat potentiostat, respectively. The PSoC Stat potentiostat was self-made according to a reported work and its published open-source program36. The main part of the device is PSoC 5LP single chip microcomputer (9.9 USD), and the detection resolution is improved by installing monolithic capacitor. In addition, a portable and low-cost stirring device was self-made for the electrochemical detection, which was fabricated by a time-delay relay (1.6 USD), a DC motor speed controller (0.7 USD), a motor (1.2USD), a sample cell (0.2 USD) and an SPCE connector (0.5 USD). The potentiostat was connected to the SPCE connector by dupont wire.

The stripping curves using self-made PSoC Stat potentiostat were shown in Fig. 6a. With the increase of Cd2+ concentration, the stripping peak potential shifted towards positive potential direction, and the stripping peak current enhanced. The peak current showed a good linear relationship with Cd2+ concentration in the range of 5–100 μg/L (Fig. 6b). The linear regression equation was Ip = 0.29C + 0.35, and the lowest detection limit of Cd2+ was 3.55 μg/L (S/N = 3). The stripping curves using commercial potentiostat were shown in Fig. S5. The limit of detection reached 0.15 μg/L, and the linear range was 1–100 μg/L. Compared with other electrochemical methods, our proposed method demonstrates competitive sensitivity and linear range (Table 1). More importantly, our method holds great advantage in the point-of-care testing. Although the limit of detection using self-made PSoC Stat potentiostat is higher than that using commercial potentiostat, it was lower than the maximum permissible level of Cd in drinking water (5 μg/L)42, which is sufficient for the on-site determination of Cd2+. The program needs be optimized and the single chip can be replaced by new ones with higher properties to improve the detection sensitivity of our self-made potentiostat, thus achieving the same level of commercial potentiostat. In addition, the Bi/Pre-anodized SPCE can simultaneously determine Cd2+ and Pb2+ (Fig. S6), demonstrating the ability of multiple detection.

Figure 6
figure 6

Analytical performance of in-situ bismuth modified pre-anodized SPCE for Cd2+ determination with self-made PSoC Stat potentiostat and stirring device. (a) The square wave anodic stripping curves of Cd2+. (b) The relationship between stripping peak current values and Cd2+ concentrations. Error bars represent the standard deviations calculated from three separate experiments. (c) Reproducibility of developed sensor for Cd2+ detection at the concentration of 50 μg/L. Ten measurements were conducted repeatedly. (d) Specificity evaluation of the electrochemical sensor. The interfering ion was added to the Cd2+ solution separately, and the corresponding peak current values of Cd2+ were compared with that of Cd2+ solution absent of any different ions (marked as Absence).

Full size image
Table 1 Comparison of different electrochemical sensors for Cd2+ determination.
Full size table

The repeatability of the proposed electrode for Cd2+ determination was evaluated by 10 consecutive tests of 50 μg/L Cd2+ standard solution (Fig. 6c). The relative standard deviation (RSD) of the measured peak current values was 4.4%, which indicates a good repeatability. Moreover, considering the interference from other potential metal ions in the determination of Cd2+ in actual samples, the anti-interference ability of Bi/Pre-anodized SPCE was studied. Ca2+, Mg2+, Fe3+ and K+ with a concentration 10 times higher than that of Cd2+, as well as a similar concentration of Pb2+, Hg2+ and Cu2+ was added to the supporting electrolyte containing 50 μg/L Cd2+ as interfering ion, respectively. As shown in Fig. 6d, Ca2+, Mg2+, Fe3+, K+ and Pb2+ had little interference on the determination of Cd2+, while Hg2+ increased the peak current of Cd2+, which is due to the co-deposition effect of Hg2+ for Cd2+ determination12. Cu2+ significantly reduced the stripping peak current of Cd2+. This inhibition may be due to the competition between Bi3+ and Cu2+ for the active site on the working electrode during deposition as well as the formation of intermetallic compounds among copper and cadmium46. The interference from Cu2+ can be eliminated by the addition of 40 µmol/L potassium ferrocyanide into the detection solution (Fig. 6d).

Recovery studies

To assess the feasibility of the developed heavy metal electrochemical sensor in practical applications, a spike-recovery method was used to detect Cd2+ in tap water and rice extract. The pre-treatment of tap water and rice samples were described in 2.4. The spiked amount of Cd2+ were set to four different levels. Each spiked sample was measured three times in parallel. Simultaneously, the spiked samples were determined by ICP-MS.

As shown in Table 2, when using our proposed electrochemical sensor, the recoveries in water and rice samples ranged from 91.7 to 107.1%, and the RSD were between 2.09 and 5.26%. In addition, the detection results were consistent with ICP-MS with deviation below 10%. These results demonstrate that the heavy metal electrochemical sensor owns great accuracy and reliability, and can be applied to the determination of Cd2+ in tap water and rice samples.

Table 2 Results of Cd2+ determination in tap water and rice samples using PSoC Stat potentiostat.
Full size table

Conclusion

A metal-bismuth modified pre-anodized screen-printed electrode was prepared based on the pre-anodization and in-situ deposition technique. Various electrochemical characteristics proved the successful preparation of the modified electrode. Benefiting from pre-anodization, bismuth co-deposition and bromine sensitization, the electron transfer ability was improved, the enrichment of Cd2+ was facilitated, and the sensitivity of the electrode for Cd2+ determination was significantly enhanced. Coupled with the self-made, portable and low-cost potentiostat and stirring device, the electrochemical sensor has a wide linear range of 5–100 μg/L and a low detection limit of 3.55 μg/L, possesses good repeatability and specificity. In addition, it can be applied to the determination of Cd2+ in drinking water and rice. Our work provides a promising electrode fabrication method and a point of need device for electrochemical determination of heavy metals.