1 Introduction

Hydroxylamine is an oxygenated derivative of ammonia used in many organic syntheses, including pharmaceutical and industrial production processes [1]. This raw material is involved in many biological and pathological processes as an important product or mediator of metabolism, such as inflammation and bio-nitrification [2, 3]. However, hydroxylamine can act as a mutagen and an acute toxicant for plants, animals and humans, and an agent for inducing physiological changes [4]. Environmentally, oceanic hydroxylamine generates nitrogen oxide and is therefore a hazardous substance [5]. Achieving a sensitive and selective analytical approach for the determination of hydroxylamine is of significant industrial, environmental and public health importance.

The hydroxylamine determination has been carried out previously using diverse methods, some of which are spectrophotometry [6], high-performance liquid chromatography [7], gas chromatography [8] and chemiluminescence [9]. Despite many advantages, these techniques lack fast response and are expensive. Electrochemical detection is particularly simple and accurate and needs less expensive tools. Electrochemical analysis has advantages over conventional analytical methods such as reliability, ease, cost-effectiveness, and time-saving [10,11,12,13,14,15,16]. Electrochemical analysis has high sensitivity for the detection of analytes, but sometimes suffers from difficulty in detecting compounds. This bottleneck for electrochemical sensors can be bypassed through conductive modifiers, in particular nanomaterials possessing great conductivity [17,18,19,20,21,22,23,24].

Nanomaterials have a high surface area and sizes in the nanoscale, and therefore have admirable physical and chemical properties. They have recently been used in various applied fields such as energy-based research, imaging, catalysis, medicine, and environmental applications [25,26,27,28,29,30,31]. They have been a promising modifier used in the fabrication of electrochemical sensors [32,33,34,35,36,37]. Modifying the electrode with nanomaterials results in reducing the detection limit and increasing the sensitivity and selectivity in the detection of various analytes [38,39,40,41,42,43].

Hollow carbon spheres (HCSs) as newly introduced carbon nanomaterials can be used in the construction of electrochemical analytical systems, electro-catalysis such as oxygen reduction reaction (ORR), oxygen evolution reaction (OER) and hydrogen evolution reaction (HER), and also batteries and supercapacitors [44,45,46,47,48]. Some special properties of HCSs are low specific density, large specific surface area, admirable mechanical strength and reduced transport length for mass and charge transport. Therefore, they are used as modified nanomaterials in many strategies of electrochemical detection [49,50,51]. Heteroatoms like nitrogen [52] and phosphorus [53] can improve the electrochemical response of carbon materials. For example, the special advantage of nitrogen is to enhance the electrocatalytic behavior of carbon materials due to charge polarization and the difference in electronegativity and electron spin density between carbon and heteroatom [54].

Accordingly, we constructed N-HCSs based on the silica template method to develop a new sensitive and selective electrochemical hydroxylamine sensor based on the modification of GCE surface (N-HCSs/GCE sensor). As-fabricated sensor exhibited admirable activity with a broad linear range, and low limit of detection towards the hydroxylamine determination. Moreover, N-HCSs/GCE had successful applicability for sensing the hydroxylamine found in real water specimens in accordance with standard addition method.

2 Experimental

2.1 Equipments

A Metrohm Autolab PGSTAT 320 N Potentiostat/Galvanostat Analyzer (the Netherlands) with GPES (General Purpose Electrochemical System-version 4.9) software was applied for all electrochemical determinations at ambient temperature. A Metrohm 713 pH-meter with glass electrode (Switzerland) was recruited to determine and adjust the solution pH. Direct-Q® 8 UV deionized water (Millipore, Germany) was used to prepare all solutions freshly. A MIRA3 scanning electron microscope (Tescan, Czech Republic) equipped with an energy dispersive X-ray spectroscopy (EDS) detector was utilized for field emission scanning electron microscopy (FE-SEM) images and elemental analysis.

2.2 Solvents and chemicals

All solvents and chemicals applied in our protocol had analytical grade belonging to Merck and Sigma-Aldrich. Phosphate buffer solution (PBS) was prepared by phosphoric acid and adjusted by NaOH to the desired pH value.

2.3 Preparation process of N-HCSs

2.3.1 Preparation of SiO2 spheres

The Stöber process, with slight modification, was followed to construct SiO2 spheres [55]. Thus, tetraethyl orthosilicate (TEOS) (6 mL) was dissolved as dropwise in a solution containing ethanol (75 mL), deionized water (DIW, 10 mL) and aqueous solution of ammonia (3.15 mL), followed by stirring for 5 h at ambient temperature. A centrifugation was applied to extract the SiO2 spheres that were then rinsed with ethanol/DIW to achieve white precipitate was vacuum dried in an oven for 12 h at 65 °C.

2.3.2 Preparation of SiO2@polydopamine (PDA) core–shell spheres

An auto-polymerization procedure was run to prepare the SiO2@PDA in accordance with previous report by Shahrokhian and et al., [56]. Thus, a certain amount of pre-prepared SiO2 particles (40 mg) was rinsed several times with Tris-buffer solution (50 mM, pH 8.5), followed by centrifugation for several times and then appending to Tris-buffer solution (20 mL, 50 mM, pH 8.5) carrying dopamine hydrochloride (40 mg) and high-rate stirring at ambient temperature for 24 h to achieve SiO2@PDA core–shell spheres. The appeared dark brown product was subjected to centrifugation and washing thoroughly with Tris-buffer to get rid of the brown color of solution to obtain the precipitate subsequently being dried in the oven for 12 h at 65 °C.

2.3.3 Preparation of SiO2@nitrogen-doped carbon spheres (N-CSs)

SiO2@PDA core–shell spheres were carbonized for 5 h at 800 °C with a heating rate of 5 °C/min in exposure to N2 atmosphere to extract SiO2@N-CSs with uniform structure.

2.3.4 Preparation of N-HCSs

In this phase, the silica cores were removed through etching SiO2@N-CSs with 2 M NaOH solution for 24 h at 60 °C to form N-HCSs following the centrifugation and repeatedly Tris-buffer washing. As-fabricated N-HCSs was dried for 12 h at 65 °C.

2.4 Preparation of the N-HCSs/GCE sensor

A drop-casting technique was followed to fabricate the N-HCSs/GCE. Thus, a certain amount of as-prepared N-HCSs (1 mg) was dispersed in DIW (1 mL) subsequently under 15 min ultra-sonication. Then, the well-dispersed suspension (3 µL) was coated on GCE surface as dropwise and dried at the laboratory temperature.

3 Results and discussion

3.1 Characterization of N-HCSs and relevant nanostructures by FE-SEM

The FE-SEM images were applied to explore the SiO2, SiO2@N-CSs and N-HCSs for morphology. The FE-SEM image captured for SiO2 is shown in Fig. 1a. The monodispersity was good for the SiO2 spherical particles, showing a uniform size of around 170 nm. Following the SiO2@PDA carbonization, the FE-SEM images of SiO2@N-CSs confirmed a homogenous and thin layer of N-doped carbon on the silica sphere surface, as shown in Fig. 1b. Following the etching process of NaOH for the elimination of the SiO2 cores, the original spherical morphology was conserved for the N-HCSs and the hollow structure was evident, see Fig. 1c.

Fig. 1
figure 1

FE-SEM images of a SiO2, b SiO2@N-CSs, and c N-HCSs

Figure 2a and b illustrates chemical composition of SiO2@N-CSs and N-HCSs using EDS spectra. The peaks can be seen for the elements of C, O, N and Si present in the structure of SiO2@N-CSs, according to the EDS spectrum (curve a). The silica core elimination was completely successful based on the reduction in Si peak, as observed in the EDS spectrum (curve b) captured for N-HCSs. The high intensity of N peak in the N-HCSs verified its successful doping in this carbonic shell.

Fig. 2
figure 2

EDS analysis of a SiO2@N-CSs; and b N-HCSs

3.2 Electrochemical response of hydroxylamine on diverse electrodes

The electrochemical response of hydroxylamine oxidation in the 0.1 M PBS adjusted to variable pH values (2.0 to 9.0) was explored to determine the influence of electrolyte solution pH. The results showed that the peak current of hydroxylamine oxidation depended on the pH value, so that it reached a maximum with increasing pH up to 7.0 and then decreased with further pH values. Hence, the pH value of 7.0 was considered to be the optimum for subsequent electrochemical determinations.

CV was performed to clarify the electrochemical behavior of hydroxylamine on unmodified (bare) and modified GCE surfaces. Figure 3 compares the bare GCE and N-HCSs/GCE for 100.0 µM hydroxylamine oxidation in 0.1 M PBS at the pH value of 7.0. The hydroxylamine oxidation displayed a tiny and wide peak (1.94 µA) at the potential of 1060 mV on the bare GCE surface. The N-HCSs-modified GCE exhibited a shift in the peak current towards the more negative potentials (960 mV) by raising the amount of current (8.8 µA). Such an improvement can be appeared because of appreciable catalytic impact of N-HCSs for the hydroxylamine oxidation.

Fig. 3
figure 3

Cyclic voltammograms captured for oxidation of hydroxylamine (100.0 µM) in PBS (0.1 M; pH 7.0) on a unmodified GCE and b N-HCSs/GCE with the scan rate of 50 mV/s

3.3 Effect of scan rate

The CVs were captured for the oxidation of hydroxylamine (90.0 µM) on the N-HCSs/GCE under variable scan rates (Fig. 4). There was an apparent gradual elevation in the oxidation peak by raising scan rate ranged from 10 to 300 mV/s. As seen in Fig. 4, the anodic peak current (Ipa) had a linear association with scan rate square root (υ1/2). The regression equation was obtained to be Ipa (μA) = 1.1683 ν1/2 (mV s−1)1/2 – 0.0804 (R2 = 0.9976), meaning controlled diffusion process of the hydroxylamine oxidation on the N-HCSs/GCE.

Fig. 4
figure 4

CVs captured for the oxidation of hydroxylamine (90.0 µM) on the N-HCSs/GCE under variable scan rates (a–l: 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, and 300 mV/s); Inset: the correlation of Ipa with ν1/2

A Tafel plot (Fig. 5) was achieved on the basis of data related to the rising domain of current–voltage curve at a low scan rate (10 mV/s) for hydroxylamine (90.0 µM) to explore on the rate-determining step. The linearity of E vs. log I plot clarifies the involvement of electrode process kinetics. The slope from this plot could present the count of transferred electrons during the rate-determining step. Based on Fig. 5, the Tafel slope was estimated to be 0.4008 V for the linear domain of the plot. The Tafel slope value reveals that the rate-limiting step is one-electron transfer process considering a transfer coefficient (α) of 0.85.

Fig. 5
figure 5

CV for hydroxylamine (90.0 µM) at the scan rate of 10 mV/s. Inset: The Tafel plot from the rising domain or the respective voltammogram

3.4 Chronoamperometric analysis

Chronoamperometry was recruited to explore the hydroxylamine catalytic oxidation on the N-HCSs/GCE surface. Chronoamperometric analysis was done for variable hydroxylamine contents on N-HCSs/GCE at the working electrode potential of 950 mV. The chronoamperograms captured for variable hydroxylamine contents on the N-HCSs/GCE is seen in Fig. 6. The Cottrell’s equation explains the current (I) for electrochemical reaction of an electroactive material with a D value (diffusion coefficient) under a mass transport limited condition. Figure 6 A shows a linear relationship of the I value with t−12 for the oxidation of variable hydroxylamine contents. The slopes form the obtained straight lines were plotted against variable hydroxylamine contents (Fig. 6B). The plotted slope and Cottrell equation estimated the D value to be 9.8 × 10−6 cm2/s for hydroxylamine.

Fig. 6
figure 6

Chronoamperometric behavior of N-HCSs/GCE in PBS (0.1 M; pH  7.0) at potential of 950 mV for variable hydroxylamine contents (a–d: 0.1, 0.5, 1.0, and 1.5 mM); Insets: A Plots of I vs. t−1/2 and B plots of the slopes from the straight lines vs. hydroxylamine level

3.5 DPV analysis of hydroxylamine

DPV analysis was done for variable hydroxylamine contents to explore linear dynamic range, LOD and sensitivity of the N-HCSs/GCE under optimized experimental circumstances (Fig. 7). As expected, the elevation in hydroxylamine level enhanced the peak current. Figure 7 shows a linear proportional of the oxidation peak currents to variable hydroxylamine contents (10.0–100.0 µM) with the linear regression equation of Ipa (µA) = 0.06787Chydroxylamine + 2.1453 (R2 = 0.9989), and the sensitivity of 0.0917 µA/µM. In the equation of LOD = 3σ/m, the σ stands for the standard deviation of the response for blank solution, and m for the slope from the standard graph. The LOD was estimated at 3.0 µM for hydroxylamine determination on N-HCSs/GCE. The performance of N-HCSs/GCE is compared with other electrochemical sensors for hydroxylamine detection, as summarized in Table 1.

Fig. 7
figure 7

DPVs captured for the oxidation of variable hydroxylamine contents on the N-HCSs/GCE (a–j: 10.0, 20.0, 30.0, 40.0, 50.0, 60.0, 70.0, 80.0, 90.0, and 100.0 µM); Inset: Calibration curve of voltammetric response (Ipa) against hydroxylamine level

Table 1 The comparison of the previous electrochemical sensors with the presented sensor (N-HCSs/GCE) for the determination of hydroxylamine

3.6 Analysis of real specimens

The practical applicability of N-HCSs/GCE was tested by sensing hydroxylamine in water specimens using DPV procedure and standard addition method, the results of which can be seen in Table 2. The recovery rate between 97.0 and 102.7%, and all RSD values were ≤ 3.4%. According to the experimental results, the N-HCSs/GCE sensor possessed a high potential for practical applicability.

Table 2 Voltammetric sensing of hydroxylamine in real specimens using N-HCSs/GCE. All concentrations are in µA (n = 5)

4 Conclusion

A sensitive electrochemical hydroxylamine sensor was fabricated on the basis of GCE modification with N-HCSs (N-HCSs/GCE sensor). The results from electrochemical characterization revealed the appreciable electrochemical activity of N-HCSs/GCE for hydroxylamine oxidation. Under the obtained optimal circumstances, as-developed sensor displayed the linear electrochemical current of the electrode in relation to the hydroxylamine concentration, with the dynamic range as wide as 10.0–100.0 µM and the LOD as narrow as 3.0 µM. Moreover, N-HCSs/GCE had successful applicability for sensing the hydroxylamine found in real water specimens, with recovery rates of 97.0 to 102.7%.