Cu(II) traceability in industrial samples: innovating detection with modified nanoparticles and magnetic electrodes

Abstract

This paper presents a novel approach for the sensitive detection of Cu(II) ions in acidic industrial samples, used in the manufacture of printed circuits. The study outlines the synthesis and functionalization of Fe₃O₄ magnetic nanoparticles, emphasizing the optimization of parameters affecting Cu(II) concentration measurements. The NPs are surface-modified with APTES and succinic acid and characterized through different methods including TEM imaging and FTIR analysis. A method employing the magnetic NPs for bulk preconcentration of Cu(II) ions, followed by collection using a simple and home-made magnetic glassy carbon electrode (MGCE), is detailed. The electrochemical analysis showcases the efficiency of the proposed method for rapid and sequential measurements of Cu(II) ions adequate for industrial matrices. Results demonstrate the potential of this approach for sensitive Cu(II) sensing, offering a cost-effective and efficient alternative to conventional analytical techniques. Notably, the successful quantification of Cu(II) concentrations in a real sample obtained from an acid industrial electroplating bath of CuSO₄ highlights the practical applicability of the developed methodology.

---

Cecilia D. Costa is pursuing her PhD at INQUIMAE-DQIAQF at the University of Buenos Aires, Argentina. Her thesis is developed with funding from UBA and CONICET doctoral fellowships. Her research focuses on nanomaterials and electrochemistry, with an emphasis on their practical applications in heavy metal removal and sensing.

Delphine Talbot has a Master's degree in Chemistry. She works at the Phenix laboratory as a design engineer. She specializes in the synthesis, functionalization and characterization of nanoparticles and their incorporation into various matrices for use as adsorbent materials in the environment.

Agnès Bée has obtained in 1987 a PhD in Chemical Sciences from the University of Reims in Champagne, France. Her PhD focused on modelisation of the interactions between soil organic matter and trace metals. Since 1988, she has been an assistant professor at Sorbonne University, France. Her research revolves around formulation of magnetic adsorbent materials for applications in water pollution remediation.

After completing his Master's degree in Chemistry at the University of Montpellier (France), Sébastien Abramson has acquired a Ph. D. in Enantioselective Heterogeneous Catalysis in 2000 at the ENSCM (National Superior School in Chemistry of Montpellier). He obtained a permanent position as an assistant professor in 2002 at Sorbonne University. His current research mainly focuses on the environmental applications of magnetic nanomaterials.

Virginia Diz Graduate and Doctor in Chemical Sciences graduated from the Faculty of Exact and Natural Sciences of the University of Buenos Aires. Her current lines of research are the design, synthesis and characterization of hard and soft nanoparticulate systems, destined for the transportation of active ingredients and environmental remediation.

G. González holds a Bachelor's and Doctorate degree in Chemical Sciences, as well as a specialization in Chemical Sciences and Environment from the University of Buenos Aires, where she currently serves as an associate professor. Additionally, she is an independent researcher at CONICET. Her research focuses on the development of interfaces and studies of ionic transport for analytical applications, as well as the removal and/or recovery of contaminants.

---

Environmental significance

The circular economy model ensures that materials remain in circulation for as long as possible and can be reintroduced into different stages of the production process. In this context, rapid and cost-effective analytical methods are required to evaluate process water in industrial matrices. This work aims to use a small amount of magnetic nanoparticles (MNPs) to quantify Cu(II) from a sample from the electroplating and printed circuit manufacturing industry with low pH and high ionic strength. These MNPs can be easily collected with a magnetic electrode, enabling rapid electrochemical quantification. This contributes to more efficient copper production cycles and reduces its disposal in effluents.

Introduction

The presence of heavy metals in wastewater is related to the growth of industrial and human activities. Poorly treated contaminated wastewater reaches the environment, threatening human health and the ecosystem. Considering that heavy metals are non-biodegradable, the presence of these metals in water even at small amounts could result in critical health issues to living organisms.¹

In particular, copper is the third most used metal in the world. It is extensively used in many industrial applications, such as metal finishing, electroplating, plastics, and etching. Thus, it is usually found at high concentrations in wastewater.^2,3 Importantly, copper is an essential micronutrient needed for human health in trace amounts,⁴ but at higher concentrations can cause problems in the liver, brain, kidneys, cornea, gastrointestinal system, lungs, immunological system, and haematological system.¹ The United States Environmental Protection Agency (USEPA) has fixed 1.3 ppm (20 μM) as the maximum contaminant level goal of copper ions in drinking water.⁵ In Argentina, regulation agencies fix a maximum of 1 ppm (16 μM) of Cu as the maximum permitted limit in drinking water. Additionally, the limit for treated sewers is 2 ppm (31 μM) for discharge into a sewer collector and 1 ppm (16 μM) for discharge into water bodies.⁶

Unfortunately, many countries, including Argentina, face challenges with several water sources contaminated with heavy metals, particularly copper, often surpassing permitted limits due to industrial effluent discharges without adequate treatment.^7–9 For this reason, industries must monitor their discharges, not only for environmental conservation but also to prevent economic losses and to adopt a circular economy strategy.

While wastewater can exhibit a wide range of concentrations depending on its source or pre-treatments, when considering reuse, it is essential to establish the concentrations for the analysed processes. The average concentration of Cu in electrorefining is 0.65 M,^10,11 while in printed circuit board manufacturing, it is approximately 0.4–0.5 M with an allowable reduction of no more than 25%, according to the manufacturer. The inclusion of various organic or inorganic species in the industrial samples as additives can fulfil specific roles, such as enhancing leveling or optimising the chemical, physical, and technological characteristics of the deposited metal. These characteristics may include corrosion resistance, brightness, reflectivity, hardness, mechanical strength, ductility, internal stress, wear resistance, and solderability.^12,13 Typically, the composition of commercial additives is extremely complex and unknown by users.¹⁴ Common organic additives used in electrodeposition include brighteners and surfactants. Brighteners are small aromatic and aliphatic compounds with functional groups such as aldehydes, ketones, carboxylic acids, and amines, which contribute to a reflective finish on the deposit.¹⁴ Surfactants serve as levelers and grain refiners in the plating bath, improving the overall surface quality.¹⁴ The introduction of these additives aims to broaden the operational range of the electrodeposition process, which may also enhance the solubility of secondary additives and function as wetting agents.¹⁵ The conventional analytical techniques for quantifying copper in water samples are flame atomic absorption spectrometry,¹⁶ spectrophotometry,¹⁷ inductively coupled plasma-mass spectrometry,¹⁸ and X-ray fluorescence.¹⁹ These methods are sophisticated and time consuming, and many of these detection techniques are expensive or need skilled operators. Therefore, the development of simple, rapid, and cost-effective methods is crucial. Electrochemical techniques offer an ideal solution, serving as reliable, simple, rapid, and sensitive methods for metal ion detection without requiring extensive expertise.²⁰ The design of electrode interfaces sensitive to target ions remains crucial in this quest. In the work of Zamani et al. (2005)²¹ and Zamani et al. (2007)²²), two Cu(II)-sensitive electrodes are presented, applicable to samples from electroplating baths. However, the original samples must be diluted to use their device, and EDTA needs to be added to the measuring solution.

Several studies have used magnetic nanoparticles (MNPs) such as magnetite (Fe₃O₄) as electrochemical sensors of heavy metals. Usually, a drop of MNP suspension is let dry on the electrode surface before each measurement.^23,24 This method has two main problems. First, it requires long periods of time to regenerate the electrode surface before each measurement and, second, there is a mass transport limitation of the analyte to the electrode surface. The solution to these problems can be found with the same MNPs. Many studies have presented the use of Fe₃O₄ for heavy metal adsorption due to the easiness of the adsorbent separation based on its magnetic properties.^25–32 If first the MNPs are distributed throughout a water sample, at a bulk scale, they rapidly adsorb and preconcentrate the analyte. Then, the analyte enriched MNPs can be easily recovered from the medium using a magnetic electrode where the voltammetric measurement is performed.^33,34

Yantasee et al. (2008)³⁵ presented a Pb(II) sensor using Fe₃O₄ NPs functionalized with dimercaptosuccinic acid on a magnetic electrode with a carbon paste surface. To mitigate severe physisorption and subsequent damage of the active surface observed at the carbon paste surface, they also introduced an electromagnet electrode with a glassy carbon active surface.³⁵ Thus, the creation of a regenerable interface is facilitated. After this publication, a few other studies showcased the utilization of magnetic carbon paste electrodes as sensors: Alizadeh et al. (2014) utilized MNPs coated with a shell of molecularly imprinted polymer (MIP) for 2,4,6-trinitrotoluene sensing,³⁶ Madrakian et al. (2015) employed synthesized Fe₃O₄@silica@MIP composites for mefenamic acid determination,³⁷ and Fayazi et al. (2022) used Au-modified Fe₃O₄ with L-cysteine for Cu(II) detection.³⁸ Other publications explored alternative active surfaces: Banerjee et al. (2010) employed Fe₃O₄@Au NPs with an organophosphorous ligand for uranium detection on a gold film surface,³⁹ Yang et al. (2013) developed an Ag(I) sensor using commercial Fe₃O₄@Au MNPs with a homemade glassy carbon magnetic electrode⁴⁰ and Hassan et al. (2018) utilized Fe₃O₄@silica@MIP nanocomposite particles for sensing methyl parathion on a graphite–epoxy composite electrode.⁴¹ The improvement in selectivity has been one of the objectives of these functionalizations. While this parameter is critical in environmental determinations, it often isn't as significant in the analysis of industrial samples. In such cases, complex samples with high ionic concentration, extreme pH, and the presence of organic additives typically pose different challenges. The determination of copper in samples from electroplating or printed circuit board manufacturing serves as a clear example of the type of problem addressed in this work.

In the present study, we synthesized Fe₃O₄ NPs with surface modification, initially employing 3-aminopropyltriethoxysilane (APTES), followed by a second step involving succinic acid. These MNPs were subsequently employed for the bulk preconcentration of Cu(II), which was then efficiently collected using a custom-made magnetic glassy carbon electrode (MGCE). Electrochemical measurements were performed on the MGCE, enabling rapid and sequential analyses of multiple samples. Notably, the preconcentration step took place in the bulk solution, away from the electrode, allowing for swift surface regeneration between measurements through a brief procedure. Following method optimization, the approach was successfully applied to the sensitive detection of Cu(II) in the industrial samples. The importance of the printed circuit manufacturing industry and the commercial value of copper highlight the significance of this work.

Experimental

Materials and methods

In this study, the following reagents were employed for the NP synthesis: iron(II) chloride tetrahydrate (FeCl₂·4H₂O, CAS: 13478-10-9), iron(III) chloride hexahydrate (FeCl₃·6H₂O, CAS: 10025-77-1, 3-(aminopropyl)triethoxysilane (APTES, CAS: 919-30-2), succinic anhydride (≥99%, CAS: 108-30-5) from Sigma Aldrich, ammonia solution (NH₄OH, CAS: 1336-21-6) from Merck and dimethyl sulfoxide (DMSO, CAS: 67-68-5) from Carlo Erba. For incubation and electrochemical measurements: copper sulfate pentahydrate (CuSO₄·5H₂O, CAS: 7758-99-8) from Mallinckrodt, potassium ferrocyanide (K₄[Fe(CN₆)]·3H₂O, CAS: 14459-95-1) from Carlo Erba, potassium ferricyanide (K₃[Fe(CN₆)], CAS: 13746-66-2) from Merck, potassium nitrate (KNO₃, CAS: 7757-79-1) from Carlo Erba, sodium chloride (NaCl, CAS: 7647-14-5) from Carlo Erba, potassium chloride (KCl, CAS: 7447-40-7) from Carlo Erba and sodium nitrate (NaNO₃, CAS: 7631-99-4) from Carlo Erba were used. For pH adjustment: 65% nitric acid (HNO₃, CAS: 7697-37-2) from Erbatron and potassium hydroxide (KOH, CAS: 1310-58-3) from AnalytiCals were employed. 37% hydrochloric acid (HCl, CAS: 7647-01-0) from Carlo Erba was used for cleaning all materials in contact with NPs. Additionally, ninhydrin (CAS: 485-47-2) from Sigma Aldrich was used for amine identification. All reagents were of analytical grade and utilized following the manufacturer's instructions and in compliance with relevant safety regulations. All solutions were prepared with double deionized water (resistivity <18 MΩ cm, Milli-Q Water System, Millipore, Bedford, MA, USA).

Synthesis

From synthesis reported by Carlos et al. (2012), Liu et al. (2013) and Wang et al. (2011), a successful new route to obtain magnetic nanoparticles covered with carboxylic groups is presented here.^42–44 Each synthesis step was performed under an argon (Ar) atmosphere, and the MNPs obtained at every stage were washed three times with ultrapure water and three times with ethanol. The magnetic properties of MNPs were used to separate them from the supernatant in each step. The black precipitate was in all cases dried at 40 °C under vacuum, since higher temperatures were avoided to prevent oxidation to maghemite.

Fe₃O₄ MNPs. 100 mL of a 0.3 M FeCl₃ (8.16 g of FeCl₃·6H₂O) and 0.15 M FeCl₂ (3.00 g FeCl₂·4H₂O) (Fe³⁺ : Fe²⁺ = 2) solution in a round bottom flask was heated under magnetic stirring at 90 °C using a silicone bath. Then 10 mL of 25% ammonium was added rapidly. The mixture was magnetically stirred at 90 °C for 30 min and then cooled to room temperature before washing and drying.

Fe₃O₄@APTES MNPs (denoted as @APTES). 1 g of Fe₃O₄ was dispersed by ultrasonication (10 min) in 75 mL of a 1 : 1 ethanol : H₂O solution and heated to reflux at 70 °C. To this suspension, 4 mL of APTES was added under magnetic stirring. The mixture was magnetically stirred at 70 °C for 4 h and then cooled to room temperature, washed and dried.

Fe₃O₄@APTES@COOH MNPs (denoted as @COOH). 50 mg of Fe₃O₄@APTES was dispersed by ultrasonication (10 min) in 20 mL of DMSO. To this suspension, under magnetic stirring, was added a solution of 500 mg of succinic anhydride, previously dissolved in 5 mL of DMSO. The mixture was left to react at room temperature for 24 h, before washing and drying.

Characterization

FTIR spectroscopy. Fourier transform infrared (FTIR) spectra of the samples were recorded in the wavelength range of 4000–400 cm⁻¹ using an FTIR Nicolet 8700 instrument. KBr pellets of Fe₃O₄, Fe₃O₄@APTES and Fe₃O₄@APTES@COOH were prepared and the spectra were taken with 32 scans and a resolution of 4 cm⁻¹.

Zeta potential. The NP suspensions were prepared at a concentration of 60 mg L⁻¹.⁴⁵ The stabilization was done at the natural pH of NP suspension with 10 min of sonication. To adjust the pH, titration with solutions of KOH and HNO₃ was used and the pH was measured using a pH meter until stabilization. It is important to note that the pH adjustments were carried out sequentially, resulting in a slight dilution of the sample at extreme pH values of approximately 0.8. However, it was considered that this had a negligible effect on the zeta potential.

Measurements were conducted using a Zetasizer Nano ZS from Malvern PANalytical; zeta potential measurements were carried out using disposable folded capillary cells (DTS1070). Prior to each measurement, the cells were washed with water and rinsed with the NP suspension to ensure proper sample preparation. For each sample, three measurements were taken.

SEM and TEM microscopy. The morphology of the NPs was investigated using field emission scanning electron microscopy (SEM) on a SEM-FEG Hitachi SU-70 apparatus. Elemental microanalysis was conducted using energy-dispersive X-ray spectroscopy (EDX) with an Oxford X-Max 50 mm² detector. Prior to analysis, the samples were coated with a thin shell of gold by sputter deposition.

Transmission electron microscopy (TEM) measurements were carried out on a JEOL JEM 2011 transmission electron microscope.

Ninhydrin test. The ninhydrin test for primary and secondary amines was conducted using a 2% ninhydrin solution in ethanol. 10 mg of MNPs were mixed with 800 μL of ultrapure water and 1 mL of the ninhydrin solution. The mixture was placed in a boiling water bath for 15 minutes. Afterward, the supernatant was measured using UV-vis spectroscopy to detect the formation of Ruhemann's purple, the reaction product.

Electrochemical measurements

Electrode. The magnetic glassy carbon electrode (MGCE), as shown in Fig. 1a) and b), consists of a 3 mm diameter glassy carbon disc secured in place with epoxy resin inside a Teflon® hollow capsule. This configuration exposes both circular faces of the glassy carbon, with the internal face in contact with a NdFeB magnet (12 000 Gauss) covered with nickel to prevent oxidation. The magnet is held in position with a screw, which also serves as the connector for the electrode. This device makes the internal magnet very easy to change or remove. The external surface of the glassy carbon was first polished with sandpaper, hence, with an alumina abrasive having 0.3 μm grains and then with an alumina abrasive having 0.05 μm grains. This last step was repeated before each measurement and then the electrode was rinsed with double deionized water.

Fig. 1 a) Photo of a magnetic glassy carbon electrode and b) scheme of a vertical cut (not in scale); c) scheme of experimental design from NP incubation to electrochemical measurement.

Incubation and measurement. A suspension of MNPs at a concentration of 0.5 mg mL⁻¹ was achieved through 10 minutes of sonication. Subsequently, 125 μL of this suspension was incubated in a solution containing 175 μL of CuSO₄ (final volume of 2 mL), and after a 10 minute incubation period, the MNPs were efficiently collected using the MGCE. The electrode, with MNPs magnetically immobilized on its active surface, was rinsed in a 100 mM KNO₃ solution and was then placed in a three-electrode cell (stainless steel counter electrode and an Ag/AgCl reference electrode). Cyclic voltammetry was performed from 0.2 V to −1.0 V and from −1.0 V to 1.0 V at a scan rate of 50 mV s⁻¹, with a 100 mM KNO₃ solution serving as the electrolyte. Fig. 1b) depicts a schematic representation of the methodology.

The impact of pH during incubation was investigated using electrochemical impedance spectroscopy (EIS). Following the same methodology, the MNPs were incubated in a solution containing a 100 mM KNO₃ with pH adjusted using HNO₃. EIS measurements were conducted in 10 mM K₃[Fe(CN)₆]/K₄[Fe(CN)₆] (1 : 1 molar ratio) as a redox probe in 100 mM KCl solution as a supporting electrolyte. The experimental parameters were the following: a frequency range of 1 Hz to 50 kHz, an amplitude set at 10 mV, and a potential of 0.229 V vs. Ag/AgCl, corresponding to the formal potential of the redox couple.

Results and discussion

Characterization of the nanomaterials

The FTIR spectra of the Fe₃O₄, Fe₃O₄@APTES and Fe₃O₄@APTES@COOH samples (thereafter denoted as Fe₃O₄, @APTES and @COOH, respectively) are shown in Fig. 2. The vibrations around 550–650 cm⁻¹ observed in the three spectra are characteristic of the Fe–O bond, which can confirm the presence of magnetite NPs.⁴⁶ Functionalization with APTES can be confirmed by broad and overlapping bands corresponding to stretching vibrations of Si–O–Si bonds at 900–1200 cm⁻¹, while CH₂ stretching appears as bands at 2800–2900 cm⁻¹.⁴⁷ N–H stretching (3200–3500 cm⁻¹) and C–N stretching (1030–1230 cm⁻¹) respectively overlap with the broad absorption band of the silanol and OH groups (3000–3700 cm⁻¹), and with the Si–O–Si vibrations (900–1200 cm⁻¹).⁴⁸ The functionalization with carboxylic acid can be identified by the two bands at 1560 cm⁻¹ and 1400 cm⁻¹ corresponding to the asymmetric and symmetric stretching of the carboxylate ions.⁴⁹

Fig. 2 FTIR spectra of Fe₃O₄, @APTES and @COOH samples. To favour comparison, the spectra are normalized by intensity with respect to the FeO stretching band at 550–650 cm⁻¹ and corrected by the blank KBr spectrum.

TEM images presented in Fig. 3 show a spherical morphology for the nanoparticles. Notably, a thin functionalization layer is evident on @APTES, accompanied by a higher degree of agglomeration that can be induced by the drying process. While agglomeration persists for @COOH, the functionalization layer is less distinct, potentially attributed to the washing of the excess APTES layer during the 24 hour reaction step with succinic anhydride. This observation is further supported by FTIR spectra in Fig. 2, where, upon normalizing the spectra by the Fe–O band, around a 15% decrease in the areas of the superimposed bands at 900–1200 cm⁻¹ (Si–O–Si stretching) is noted between @APTES and @COOH.

Fig. 3 From left to right, TEM images of Fe₃O₄, @APTES and @COOH.

The particle sizes were determined by analysis of the TEM images using the ImageJ® software. The resulting distribution can be fitted with a log-normal function where σ = 0.2 is the standard deviation, and d_TEM = 11.5 nm is the mean size obtained from the log-normal distribution. This value is similar to the 10 nm crystallite size calculated using the Scherrer equation on the d₄₄₀ XRD peak (see Fig. S1 in the ESI†). Globally, the XRD pattern confirms that the magnetic core of the NPs is composed of Fe₃O₄.

The EDX technique is commonly used for semiquantitative elemental composition analysis. The results obtained at 5 keV are presented in Table 1, showing the atomic percentage of the elements N, O, Si, and Fe. As seen in the table, N, being a light element, is typically not easily detectable at low concentrations with this technique. It is interesting to analyse the Fe/O ratio, which is close to the theoretical value of 0.75 for magnetite (Fe₃O₄). As expected, Fe₃O₄ has a lower oxygen content compared to the functionalized NPs due to the oxygen content of organic functionalization. The presence of APTES functionalization is indicated by the presence of Si in @APTES and @COOH. The Si/Fe ratio is the same for both NPs; however, the percentage error is too large to draw strong conclusions from this result.

Table 1 Atomic percentage of each element of interest for the MNPs

	Fe3O4	@APTES	@COOH
N (%)	—	0.0	0.7 ± 0.8
O (%)	55.3 ± 0.9	56.5 ± 0.9	56 ± 1
Si (%)	—	1.3 ± 0.2	1.5 ± 0.4
Fe (%)	45 ± 1	42 ± 1	42 ± 1
Fe/O	0.81 ± 0.03	0.75 ± 0.03	0.74 ± 0.03
Si/Fe	—	0.03 ± 0.01	0.04 ± 0.01

---

The zeta potential is the relative potential at the double layer (slipping plane). This means that the zeta potential is relative to the surface charge, but not the surface charge of the particle itself.^50,51 The zeta potential values of the three MNPs dispersed in water, as a function of pH, are depicted in Fig. 4. A noticeable shift in the point of zero charge (PZC) is observed with surface functionalization. The bare Fe₃O₄ NPs exhibit a PZC at pH 7.6, in accordance with the reported values in the literature.^52,53 It's important to note that the PZC value may vary depending on the synthesis method, washing procedure, and sonication process. However, when the surface was coated with amine groups (@APTES), the PZC shifted to pH 10, which aligns well with the pK_a of amine groups on APTES.^54,55 In addition, the @COOH NPs, functionalized with carboxylic groups, displayed their PZC values at pH 3.9. This finding is in good agreement with the pK_a value of 4.87 for propionic acid.⁵⁶

Fig. 4 Zeta potential as a function of pH for Fe₃O₄, @APTES and @COOH. The corresponding PZC values are displayed on the figure.

To rule out the possibility of silica layer hydrolysis at high pH values during the zeta potential measurements, the FTIR spectra of @APTES and @COOH, before and after incubating in KOH solutions at pH 10 for 10 minutes and 2 hours, were analysed (data not shown). After 10 minutes at pH 10, no changes in the FTIR spectrum were observed, particularly in the Si–O band at 900–1200 cm⁻¹. After 2 hours, no changes were observed for @COOH, while there was a decrease of the silane bands by around 30% for @APTES. It can be concluded that for the duration of the zeta potential experiment, minimal or no damage was caused to the functionalization. Additionally, to confirm that the positive zeta potential values of @COOH weren't due to free and accessible amine groups present on the @COOH surface, a colorimetric ninhydrin test for primary amines was done on the @APTES and @COOH samples. The reaction is presented in Fig. S4.† The purple colour reaction product, indicative of the presence of the amine group, was observed by UV-vis (Fig. S5†) only for the @APTES NPs. This indicates that the remaining amine groups on the @COOH sample are not accessible on its surface, consistent with the result of the basic hydrolysis where no damage to the silane layer was observed on @COOH, probably due to the hindered access created by the succinic layer.

Optimization of the parameters affecting electrochemical Cu(II) measurements

Since extreme pH and high ionic strength often pose challenges in the analysis of samples with complex matrices, this study presents the effect and selection of optimal conditions for two parameters typically most relevant in electrochemical measurements for this type of sample: pH and the addition of a supporting electrolyte (NaCl).

The glassy carbon surface of the electrode requires activation.^57,58 Indeed, we observed that a polishing step with 0.05 μm alumina enhances electron transfer in ferrocyanide–ferricyanide solution (see Fig. S2 in the ESI†). Furthermore, it should be noted that the MGCE has a wide potential range of work, although this range is slightly shortened when Fe₃O₄ MNPs are magnetically immobilized on the MGCE surface leaving a work range between −1.0 and 1.0 V. To test the stability of @COOH at the acidic pH of the electroplating sample, three cyclic voltammetry scans were done in pH 2.5 KNO₃ observing no @COOH damage (see Fig. S3 in the ESI†). Since the nanoparticles are immobilized on the electrode surface solely by the magnetic field, they are easily removed by thoroughly rinsing the electrode with a wash bottle. Afterward, the surface is polished with alumina to restore it to its original state and rinsed again with water. No detectable concentrations of Cu were observed after this washing process.

The electrochemical behaviour of adsorbed Cu(II) was studied by cyclic voltammetry. After incubation of the Cu(II) solutions with a suspension of Fe₃O₄, @APTES, or @COOH MNPs, the nanoparticles were collected with the magnetic electrode resulting in their immobilization on the glassy carbon surface. The nanoparticles are rapidly immobilized on the electrode surface solely by the magnetic field generated by the electrode's internal magnet as shown in the Preliminary study on the behaviour of the MNPs on the magnetic electrode in the ESI† (Fig. S6).

The first measurements were made at pH 2.5 and are displayed in Fig. 5a). For the Fe₃O₄ and @APTES, a small reduction signal and no oxidation signal occur suggesting a low level of Cu adsorption and a probable Cu release from the surface after reduction. The results are comparable to the blank test, made in the absence of MNPs. Higher reduction currents for @COOH and the presence of oxidation currents show the enhanced performance of nanoparticles when they are functionalized with a carboxylic group, even at low pH values when the electrostatic interaction is not predominant.

Fig. 5 Cyclic voltammetry of: a) Fe₃O₄, @APTES and @COOH after NP incubation in pH 2.5 10 mM CuSO₄. b) @COOH after NP incubation in pH 2.5 and pH 4 10 mM CuSO₄. The blank measurement corresponds to a test without NPs in the incubation flask followed by the collection and rinse procedure.

Considering the acid–base properties of the functionalized samples, we studied a higher pH of incubation for @COOH. Given that the precipitation of Cu at 100 mM occurs at pH ≈ 5, we chose a pH of 4. At this pH, the @COOH sample is close to its isoelectric point (IEP), and more carboxylic groups on its surface are deprotonated, favouring the electrostatic interaction with Cu(II) and enhancing its adsorption. This behaviour is well-documented in the literature, with several studies reporting the adsorption of divalent cations on Fe₃O₄ functionalized with carboxylic acids.^19,51,52 The electrochemical response of @COOH at pH 2.5 and 4 is shown in Fig. 5b), where a smaller reduction signal and no oxidation signal are observed at pH 4. This improved electrochemical performance at pH 2.5, a pH level where Cu(II) adsorption is not optimal due to the high concentration of H⁺ ions, indicates that the pH affects the electrical connectivity of adsorbed Cu(II) ions to the surface. To explore this effect, electrochemical impedance spectroscopy (EIS) measurements were conducted. Hence, @COOH incubation took place in a pH 2.5 or pH 5.6 KNO₃ aqueous solution (without Cu(II)) for 10 minutes, followed by the standard collection of MNPs using the MGCE. Subsequently, EIS measurements using the ferrocyanide–ferricyanide couple as a redox probe were performed. The resulting Nyquist plot is presented in Fig. 6. As the imaginary component of the impedance of a resistor is null, the diameter of the semicircle represents the electron transfer resistance (R_ct), which is measured to be 140 ± 10 Ω and 280 ± 80 Ω for incubation at pH 2.5 and 5.6, respectively. The lower R_ct observed at low pH might account for the better electrochemical performance observed at this pH level.⁵⁹ Electrochemically, we will only observe the reduction of the ‘connected’ Cu, and we aim to demonstrate that despite the lower adsorption at pH 2.5, if the connected quantity is higher, we may observe a stronger signal.

Fig. 6 EIS measurements of @COOH incubated in pH 2.5 and pH 5.6 0.1 M KNO₃ solutions for 10 minutes. EIS measurements were conducted using 10 mM ferrocyanide–ferricyanide as a redox probe in 100 mM KCl aqueous solution (f = 1 Hz–50 kHz; A = 10 mV; E° = 0.229 V vs. ref).

Alongside the low pH, another complicating factor in these samples is the high concentration of salts, which can interfere with the interaction between @COOH and Cu(II). To study the effect of different salts during incubation, a 100 mM NaCl, KCl, NaNO₃ or KNO₃ solution was added to the 10 mM CuSO₄ incubation solution, in the presence of the @COOH MNPs. Then, the electrochemical signal of Cu ions was measured with the MGCE, after the usual collection and rinsing procedure. As presented in Fig. 7, comparing between cations, at fixed anions, Na⁺ shows a stronger signal compared to K⁺. This disparity could partially be explained by the Hofmeister series, where Na⁺ induces a greater decrease in water surface tension, promoting higher hydrophilicity.⁶⁰ Consequently, this can enhance the interaction of @COOH functional groups with the surrounding solvent favouring the interaction with the analyte. When comparing between anions at fixed cations, a stronger signal is observed for Cl⁻ than for NO₃⁻ but in this case the effect is contrary to what it is expected with the Hofmeister series for anions. This discrepancy was previously observed by Vaid et al., who proposed that Cu(II) adsorption on alginic acid due to the carboxylic acid moieties is dictated by the anion adsorption which induces the proton abstraction of the COOH.⁶¹

Fig. 7 Area of oxidation peak current for incubation of a 10 mM CuSO₄ solution with a 100 mM salt solution (NaCl, KCl, NaNO₃ and KNO₃) or without salt (w/o) at pH 2.5. Peak area is used instead of peak height to account for slightly different peak shapes.

The optimized values regarding the mass of nanoparticles incubated were selected based on the best signal-to-noise ratio from experiments using different volumes of NP suspension. The stability of the NPs and their functionalization after preparing the suspension was also confirmed by FTIR (data not shown).

Calibration curve and industrial sample

To evaluate the performance of the developed electrochemical sensor, a series of solutions containing concentrations of CuSO₄ in the range of 0.5–50 mM and 100 mM NaCl at pH 2.5 were analysed using the MGCE and the @COOH MNPs under the conditions described in the Incubation and measurement section. The calibration curve, depicting the variation of the oxidation peak height as a function of Cu(II) concentration, is given in Fig. 8. The equation which can be obtained by linear regression on the first part of the curve is Ip_a = 21.8 [Cu] − 10 (r² = 0.9995), in which Ip_a is the height of the anodic peak (μA) and [Cu] is the Cu(II) concentration in mM. Each point in this calibration curve is the mean value of triplicate independent experiments. The limit of detection (LOD) and limit of quantification (LOQ) were calculated from the standard deviation of the analytical results predicted by the calibration curve (So = 0.08 mM) as LOD = 3 × So, LOQ = 10 × So, giving LOD = 0.2 mM and LOQ = 0.7 mM. Usually, very low limits of detection and/or quantification are desirable in the development of analytical methods. In this study, within the context of the industrial application, the figure of merit must be suitable for determining the concentration of Cu within the working range of the solutions used in the manufacture of printed circuits, which is above the LOD and LOQ reported here.

Fig. 8 The calibration curve of CuSO₄ in 100 mM NaCl. Experimental conditions: incubation at pH 2.5 for 10 min with the @COOH MNPs, cyclic voltammetry in 100 mM KNO₃, scan rate 50 mV s⁻¹. Inset: Oxidation peak of cyclic voltammograms.

The aim of this electrochemical sensor was to determine the concentration of Cu(II) in industrial samples and to evaluate its applicability. Therefore, a Cu bath from an etching process was evaluated. The sample was provided by a local printed circuit board manufacturing SME located in Buenos Aires, Argentina. The solution contains 100 g L⁻¹ (400 mM) heptahydrate copper sulfate, 138 mL L⁻¹ sulfuric acid (78%), 167 μL L⁻¹ hydrochloric acid (37%) and 6.7 mL L⁻¹ additive Cuprostar® (commercial additive based on 2-chlorobenzaldehyde and 2-butyne-1,4-diol by GC-MS). This gives a solution of 400 mM in Cu(II) and pH < 1 with the commercial organic additive. In order to achieve a concentration within the linear range of the calibration curve, a 1/50 dilution of this solution in 100 mM NaCl was done, the pH was adjusted with KOH to 2.5, and the Cu concentration was analyzed with the MGCE and @COOH MNPs as described in the Incubation and measurement section. After considering the dilution carried out and performing the error propagation, this gave a concentration for the industrial sample derived from the manufacture of printed circuits of 396 ± 14 mM, where the error was determined as So × t_(99%,n−2) (t_(99%,n−2) = 3.7465). Considering the reuse of the solutions, since a concentration of 400 mM with a maximum reduction of 25% is recommended for electroplating processes, diluting by a factor of 50 would be appropriate to evaluate both a solution of 400 mM and a solution of 300 mM using the proposed system. For its implementation in industrial samples of different origins, it would only be necessary to perform a calibration curve in that matrix to rule out potential interferences and the necessary dilution, thus validating its use.

Conclusions

This study presents a comprehensive measurement platform—including the synthesis of the nanomaterial, electrode design, and measurement conditions—aimed at the reuse of copper solutions in one of the most relevant industries today: printed circuit manufacturing, a critical step in electronic development, thus contributing to the implementation of circular economy principles. Electrochemical sensors enable these measurements in cost-effective equipment and in a less time-consuming manner compared to other instrumental techniques, such as atomic absorption. To achieve this, magnetic Fe₃O₄ nanoparticles functionalized with carboxylic moieties (@COOH) were synthesized. The successful functionalization at each synthesis step was validated through FTIR, zeta-potential, UV-vis, and TEM analyses. Exploiting the magnetic properties, a magnetic glassy carbon electrode (MGCE) was designed to, first, recover the magnetic nanoparticles post-incubation in the samples and then, to measure the Cu(II) concentration by cyclic voltammetry. The impact of pH on the sensor performance revealed higher anodic currents at pH 2.5, not due to increased adsorption capacity but owing to reduced electron transfer resistance. Studying the impact of various ions revealed increased currents, with the order of influence as NaCl > KCl ≈ without salt ≈ NaNO₃ > KNO₃. Based on this finding, a calibration curve for Cu(II) in the range of 0.5–50 mM was established using 100 mM NaCl solution at pH 2.5. This methodology accurately determined the concentration of a sample obtained from a Cu bath in the etching industry.

Data availability

The data analysis scripts of this article are available in the interactive notebook https://drive.google.com/file/d/1RhKnpMKO2GsE6T-fJhlaTIf5q391c7vF/view?usp=sharing.

Conflicts of interest

There are no conflicts of interest to declare.

Acknowledgements

This work was partially supported by Universidad de Buenos Aires UBA (20020170100341BA), CONICET (11220150100291CO), OPCW (L/ICA/ICB/210497/17), CNRS (Centre National de la Recherche Scientifique) and Sorbonne Université. G. G. is a research staff member of CONICET and V. D. is a research staff member of UBA. C. C. acknowledges UBA and CONICET for his doctoral fellowship. The authors gratefully acknowledge David Montero for the SEM and TEM analyses and ADIMRA (Asociación de Industriales Metalúrgicos de la República Argentina) for providing the industrial copper process water samples.

References

1. N. A. A. Qasem, R. H. Mohammed and D. U. Lawal, Removal of heavy metal ions from wastewater: a comprehensive and critical review, npj Clean Water, 2021, 4, 36.
2. S. A. Al-Saydeh, H. El-Naas and S. J. Zaidi, Copper removal from industrial wastewater: A comprehensive review, J. Ind. Eng. Chem., 2017, 56, 35–44.
3. Z. Shabbir, A. Sardar, A. Shabbir, G. Abbas, S. Shamshad, S. Khalid, Natasha, G. Murtaza, C. Dumat and M. Shahid, Copper uptake, essentiality, toxicity, detoxification and risk assessment in soil-plant environment, Chemosphere, 2020, 259, 127436.
4. R. Manne, M. M. R. M. Kumaradoss, R. S. R. Iska, A. Devarajan and N. Mekala, Water quality and risk assessment of copper content in drinking water stored in copper container, Appl. Water Sci., 2022, 12, 27.
5. Ministerio de Salud - Argentina, https://e-legis-ar.msal.gov.ar/htdocs/legisalud/migration/html/5551.html.
6. RESIDUOS PELIGROSOS - Decreto 831/93 - Reglamentación de la Ley N° 24.051. - Bs. As., 23/4/93, https://www.argentina.gob.ar/normativa/recurso/12830/texact/htm.
7. E. Sci, P. Res, L. Regaldo, M. F. Gutierrez, U. Reno, V. Fernández, S. Gervasio, M. R. Repetti and A. M. Gagneten, Environmental Science and Pollution Research Water and sediment quality assessment in the Colastiné-Corralito stream system (Santa Fe, Argentina): impact of industry and agriculture on aquatic ecosystems, Environ. Sci. Pollut. Res., 2018, 25, 6951–6968.
8. R. J. M. Serafini, S. Arreghini, H. E. Troiani and A. R. F. de Iorio, Copper, zinc, and chromium accumulation in aquatic macrophytes from a highly polluted river of Argentina, Environ. Sci. Pollut. Res., 2023, 30, 31242–31255.
9. M. Baria, R. U. Escaray and J. E. Bustingorry, Copper, zinc and chromium in water, sediments and biota in the pampean Chascomus Lake (Argentina), Nat. Neotrop., 1999, 30, 67–69.
10. G. Cifuentes, J. Simpson, F. Lobos, L. Briones and A. Morales, Copper Electrowinning Based On Reactive Electrodialysis, J. Chil. Chem. Soc., 2009, 54(4), 334–338.
11. K. S. Barros, V. S. Vielmo, B. G. Moreno, G. Riveros, G. Cifuentes and A. M. Bernardes, Chemical Composition Data of the Main Stages of Copper Production from Sulfide Minerals in Chile: A Review to Assist Circular Economy Studies, Minerals, 2022, 12, 250.
12. J. Dini, Electrodeposition: the materials science of coatings and substrates, 1993, vol. 28.
13. N. Kanani, Electroplating: Basic Principles, Processes and Practice, Elsevier, 2006.
14. C. D. Costa, V. E. Diz and G. A. González, Electrochemical Study of the Effect of Zinc Baths with Additives on Steel Surfaces Coating by Electroplating, Port. Electrochim. Acta, 2024, 42, 255–272.
15. J.-C. Hsieh, C.-C. Hu and T.-C. Lee, The Synergistic Effects of Additives on Improving the Electroplating of Zinc under High Current Densities, J. Electrochem. Soc., 2008, 155, D675.
16. G. Bagherian, A. Chamjangali, H. S. Evari and M. Ashrafi, Determination of copper(II) by flame atomic absorption spectrometry after its perconcentration by a highly selective and environmentally friendly dispersive liquid–liquid microextraction technique, J. Anal. Sci. Technol., 2019, 10, 3.
17. B. Topuz, Ş. M. Adanur and A. Yalçuk, A new method for simultaneous determination of trace amounts of Cu(II) and Ni(II) ions by preconcentration and spectrophotometric analysis, Turk. J. Chem., 2017, 41, 619–629.
18. M. Radaelli, E. Scalabrin, G. Toscano and G. Capodaglio, High Performance Size Exclusion Chromatography-Inductively Coupled Plasma-Mass Spectrometry to Study the Copper and Cadmium Complexation with Humic Acids, Molecules, 2019, 24, 3201.
19. M. I. Maretti Silveira Bueno and L. C. Do Amaral, X-ray fluorescence determination of adsorbed copper on activated charcoal after glycerin complexation, Quim. Nova, 1998, 21, 434–436.
20. R. Tahaei, H. Shayani-Jam and M. R. Yaftian, Voltammetric determination of trace copper(II), cadmium(II), and lead(II) using a Schiff base modified glassy carbon working electrode, Monatsh. Chem., 2021, 152, 51–59.
21. H. A. Zamani, G. Rajabzadeh, M. R. Ganjali and S. M. Khatami, Highly Selective and Sensitive Copper(II) Membrane Sensors Based on 6-Methyl-4-(1-phenylmethylidene)amino-3-thioxo-1,2,4-triazin-5-one as a New Neutral Ionophore, Electroanalysis, 2005, 17, 2260–2265.
22. H. A. Zamani, G. Rajabzadeh, A. Firouz and M. R. Ganjali, Determination of copper(II) in wastewater by electroplating samples using a PVC-membrane copper(II)-selective electrode, J. Anal. Chem., 2007, 62, 1080–1087.
23. H. Yan and S. Hu, Electrochemical Sensing of Heavy Metal Ions based on Monodisperse Single-crystal Fe3O4 Microspheres, J. Wuhan Univ. Technol., Mater. Sci. Ed., 2018, 33, 1422–1427.
24. M. Baghayeri, A. Amiri, B. Maleki, Z. Alizadeh and O. Reiser, A simple approach for simultaneous detection of cadmium(II) and lead(II) based on glutathione coated magnetic nanoparticles as a highly selective electrochemical probe, Sens. Actuators, B, 2018, 273, 1442–1450.
25. M. Ozmen, K. Can, G. Arslan, A. Tor, Y. Cengeloglu and M. Ersoz, Adsorption of Cu(II) from aqueous solution by using modified Fe3O4 magnetic nanoparticles, Desalination, 2010, 254, 162–169.
26. W. Yantasee, C. L. Warner, T. Sangvanich, R. S. Addleman, T. G. Carter, R. J. Wiacek, G. E. Fryxell, C. Timchalk and M. G. Warner, Removal of heavy metals from aqueous systems with thiol functionalized superparamagnetic nanoparticles, Environ. Sci. Technol., 2007, 41, 5114–5119.
27. A. Bashir, A. H. Pandith, L. A. Malik, A. Qureashi, F. A. Ganaie and G. N. Dar, Magnetically recyclable L-cysteine capped Fe3O4 nanoadsorbent: A promising pH guided removal of Pb(II), Zn(II) and HCrO4- contaminants, J. Environ. Chem. Eng., 2021, 9(5), 105880.
28. N. Kobylinska, L. Kostenko, S. Khainakov and S. Garcia-Granda, Advanced core-shell EDTA-functionalized magnetite nanoparticles for rapid and efficient magnetic solid phase extraction of heavy metals from water samples prior to the multi-element determination by ICP-OES, Microchim. Acta, 2020, 187, 289.
29. S. Villa, P. Riani, F. Soggia, E. Magi and F. Canepa, Thiol-functionalized magnetic nanoparticles for static and dynamic removal of Pb(II) ions from waters, J. Nanopart. Res., 2019, 21, 44.
30. P. Miao, Y. Tang and L. Wang, DNA modified Fe3O4@Au magnetic nanoparticles as selective probes for Simultaneous detection of heavy metal ions, ACS Appl. Mater. Interfaces, 2017, 9, 3940–3947.
31. Y. Bagbi, A. Sarswat, D. Mohan, A. Pandey and P. R. Solanki, Lead and Chromium Adsorption from Water using L-Cysteine Functionalized Magnetite (Fe3O4) Nanoparticles, Sci. Rep., 2017, 7, 1–15.
32. Y. S. Minaberry, C. Costa, V. Diz and M. Tudino, An ion imprinted magnetic organosilica nanocomposite for the selective determination of traces of Cd( <scp>ii</scp> ) in a minicolumn flow-through preconcentration system coupled with graphite furnace atomic absorption spectroscopy, Anal. Methods, 2022, 14, 2920–2928.
33. M. Ahmadi, A. Ghoorchian, K. Dashtian, M. Kamalabadi, T. Madrakian and A. Afkhami, Application of magnetic nanomaterials in electroanalytical methods: A review, Talanta, 2021, 225, 121974.
34. L. Gloag, M. Mehdipour, D. Chen, R. D. Tilley and J. J. Gooding, Advances in the Application of Magnetic Nanoparticles for Sensing, Adv. Mater., 2019, 31, 1–26.
35. W. Yantasee, K. Hongsirikarn, C. L. Warner, D. Choi, T. Sangvanich, M. B. Toloczko, M. G. Warner, G. E. Fryxell, R. S. Addleman and C. Timchalk, Direct detection of Pb in urine and Cd, Pb, Cu, and Ag in natural waters using electrochemical sensors immobilized with DMSA functionalized magnetic nanoparticles, Analyst, 2008, 133, 348–355.
36. T. Alizadeh, Preparation of magnetic TNT-imprinted polymer nanoparticles and their accumulation onto magnetic carbon paste electrode for TNT determination, Biosens. Bioelectron., 2014, 61, 532–540.
37. T. Madrakian, E. Haghshenas, M. Ahmadi and A. Afkhami, Construction a magneto carbon paste electrode using synthesized molecularly imprinted magnetic nanospheres for selective and sensitive determination of mefenamic acid in some real samples, Biosens. Bioelectron., 2015, 68, 712–718.
38. M. Fayazi, M. Ghanei-Motlagh and C. Karami, Application of magnetic nanoparticles modified with L-cysteine for pre-concentration and voltammetric detection of copper(II), Microchem. J., 2022, 181, 107652.
39. R. Banerjee, Y. Katsenovich, L. Lagos, M. Senn, M. Naja, V. Balsamo, K. H. Pannell and C. Z. Li, Functional magnetic nanoshells integrated nanosensor for trace analysis of environmental uranium contamination, Electrochim. Acta, 2010, 55, 7897–7902.
40. H. Yang, X. Liu, R. Fei and Y. Hu, Sensitive and selective detection of Ag+ in aqueous solutions using Fe3O4@Au nanoparticles as smart electrochemical nanosensors, Talanta, 2013, 116, 548–553.
41. A. H. A. Hassan, S. L. Moura, F. H. M. Ali, W. A. Moselhy, M. del P. Taboada Sotomayor and M. I. Pividori, Electrochemical sensing of methyl parathion on magnetic molecularly imprinted polymer, Biosens. Bioelectron., 2018, 118, 181–187.
42. L. Carlos, M. Cipollone, D. B. Soria, M. Sergio Moreno, P. R. Ogilby, F. S. García Einschlag and D. O. Mártire, The effect of humic acid binding to magnetite nanoparticles on the photogeneration of reactive oxygen species, Sep. Purif. Technol., 2012, 91, 23–29.
43. Y. Liu, Y. Li, X. M. Li and T. He, Kinetics of (3-aminopropyl)triethoxylsilane (aptes) silanization of superparamagnetic iron oxide nanoparticles, Langmuir, 2013, 29, 15275–15282.
44. S. Wang, S. Wen, M. Shen, R. Guo, X. Cao, J. Wang and X. Shi, Aminopropyltriethoxysilane-mediated surface functionalization of hydroxyapatite nanoparticles: synthesis, characterization, and in vitro toxicity assay, Int. J. Nanomed., 2011, 6, 3449–3459.
45. S. Housni, S. Abramson, J. M. Guigner, P. Levitz and L. Michot, Flocculation and magnetically-assisted sedimentation of size-sorted beidellite platelets mixed with maghemite nanoparticles, Nano Res., 2020, 13, 3001–3011.
46. L. Bondarenko, E. Illés, E. Tombácz, G. Dzhardimalieva, N. Golubeva, O. Tushavina, Y. Adachi and K. Kydralieva, Fabrication, microstructure and colloidal stability of humic acids loaded fe3o4/aptes nanosorbents for environmental applications, Nanomaterials, 2021, 11(6), 1418.
47. A. Durdureanu-Angheluta, A. Dascalu, A. Fifere, A. Coroaba, L. Pricop, H. Chiriac, V. Tura, M. Pinteala and B. C. Simionescu, Progress in the synthesis and characterization of magnetite nanoparticles with amino groups on the surface, J. Magn. Magn. Mater., 2012, 324, 1679–1689.
48. N. Majoul, S. Aouida and B. Bessaïs, Progress of porous silicon APTES-functionalization by FTIR investigations, Appl. Surf. Sci., 2015, 331, 388–391.
49. S. E. Noriega and A. Subramanian, Consequences of Neutralization on the Proliferation and Cytoskeletal Organization of Chondrocytes on Chitosan-Based Matrices, Int. J. Carbohydr. Chem., 2011, 2011, 1–13.
50. V. Selvamani, Stability Studies on Nanomaterials Used in Drugs, Elsevier Inc., 2018.
51. A. Barhoum, M. L. García-Betancourt, H. Rahier and G. Van Assche, Physicochemical characterization of nanomaterials: Polymorph, composition, wettability, and thermal stability, Elsevier Inc., 2018.
52. C. Yang, G. Wang, Z. Lu, J. Sun, J. Zhuang and W. Yang, Effect of ultrasonic treatment on dispersibility of Fe3O 4 nanoparticles and synthesis of multi-core Fe3O 4/SiO2 core/shell nanoparticles, J. Mater. Chem., 2005, 15, 4252–4257.
53. M. Kosmulski, The pH dependent surface charging and points of zero charge. IX. Update, Adv. Colloid Interface Sci., 2021, 296, 102519.
54. M. H. Mashhadizadeh and M. Amoli-Diva, Drug-carrying amino silane coated magnetic nanoparticles as potential vehicles for delivery of antibiotics, J. Nanomed. Nanotechnol., 2012, 3, 139.
55. T. Suteewong, H. Sai, M. Bradbury, L. A. Estroff, S. M. Gruner and U. Wiesner, Synthesis and formation mechanism of aminated mesoporous silica nanoparticles, Chem. Mater., 2012, 24, 3895–3905.
56. R. N. Goldberg, N. Kishore and R. M. Lennen, Thermodynamic Quantities for the Ionization Reactions of Buffers, J. Phys. Chem. Ref. Data, 2002, 31(2), 231–370.
57. I. F. Hu, D. H. Karweik and T. Kuwana, Activation and deactivation of glassy carbon electrodes, J. Electroanal. Chem. Interfacial Electrochem., 1985, 189, 59–72.
58. Q. L. Zhao, Z. L. Zhang, L. Bao and D. W. Pang, Surface structure-related electrochemical behaviors of glassy carbon electrodes, Electrochem. Commun., 2008, 10, 181–185.
59. M. Lu and R. G. Compton, Voltammetric pH sensing using carbon electrodes: Glassy carbon behaves similarly to EPPG, Analyst, 2014, 139, 4599–4605.
60. K. P. Gregory, G. R. Elliott, H. Robertson, A. Kumar, E. J. Wanless, G. B. Webber, V. S. J. Craig, G. G. Andersson and A. J. Page, Understanding specific ion effects and the Hofmeister series, Phys. Chem. Chem. Phys., 2022, 24, 12682–12718.
61. U. Vaid, S. Mittal and J. Nagendra Babu, Influence of anion induced proton abstraction on Cu(II) adsorption by alginic acid, React. Funct. Polym., 2015, 97, 48–55.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4en00459k

---