Gold nanoparticles-based green nano-plasmonic sensors derived from Annona squamosa leaf extract for visual and UV–visible detection of cadmium contamination in water

1 Introduction

Cadmium (Cd) is a fully toxic heavy metal and from last century its became important for industry (Khan Z. et al., 2022). It has certain specific physical and chemical properties that have made it desirable, including chemical resistance, good electrical conduction, stability in electrochemistry, and resistance to high temperature (Adriano, 2001). It can be used in a number of technologies in many industries including rechargeable nickel–Cd batteries, metal coatings, electroplating, pigments, farming and pigments, stabilizers in plastics, photovoltaic devices, fertilizers, mining and alloys. It is used the most in the production of batteries, electroplating, and pigments. Rapid urbanization, industrialization, agricultural intensification, and poor management of industrial waste have caused Cd to be released into ecosystems (Adriano, 2001). The global concern of Cd contamination has escalated because of its non-degradable properties, its long environmental persistence, and its ability to biomagnification in organisms.

It can enter the ecosystems through soil and water as well as the food chain through crops, water, and livestock and further increases its multi-level toxicity (Khan Z. et al., 2022). It inhibits the photosynthesis efficiency of plants, slow down their growth, and also alters plant’s ability to uptake nutrients leading to chlorosis (Hu et al., 2025; Lee et al., 2021; Lin, 2011). Organisms such as fish and algae are vulnerable to Cd exposure and may experience several hazards such as reproductive disorders, metabolic disruptions, and even death (Lee et al., 2023; León-Vaz et al., 2020). Cd has high environmental persistence and can alter the biodiversity by modifying the interactions of different species and the ecosystems where they reside.

Its exposure can cause chronic health issues such as damage to the kidneys, bones, lungs, heart, and liver (Brzóska et al., 2003; Åkesson et al., 2005; Charkiewicz et al., 2023; Briffa et al., 2020). It can occupy the same spaces as calcium and zinc, which are necessary for the proper functioning of biological systems. This exposure can disrupt metabolism and enzymatic activities. It can also create oxidative stress, eventually leading to damage to deoxyribonucleic acid (DNA). The regulatory agencies have provided guidelines to restrict Cd exposure as much as possible. The World Health Organization and US Environmental Protection Agency have listed Cd in drinking water with the highest allowable limits of 3–5 μg/L, while regulating the presence of the compound in a variety of consumables, including vegetables, seafood, and cereals (US EPA, 2026; ATSDR, 2026; Program, 2024; WHO, 2006). There are numerous sources of Cd pollution that create a need for constant monitoring. Mining, industry, and waste incineration create large, sudden, concentrated releases, and the fertilizer’s use amplifies its presence in the environment over time.

Thus, there is a pressing need to develop analytical methods to detect Cd at low concentrations in environmental systems quickly, reliably, sensitively, and economically. Common methods for Cd analysis include atomic absorption spectrometry (AAS), inductively coupled plasma mass spectrometry (ICP-MS), inductively coupled plasma optical emission spectrometry (ICP-OES), anodic stripping voltammetry (ASV), X-ray fluorescence (XRF), and high-performance liquid chromatography (HPLC) (Liu et al., 2025; Li et al., 2018; Hu et al., 2002; Yi et al., 2021; Anoja et al., 2024). Although these methods have a great deal of sensitivity and selectivity, they have a lot of practical downsides (Lurie, 2005; Grabarczyk et al., 2023; Hyde et al., 2004; Wilschefski and Baxter, 2019). Traditional methods have high costs, require and sophisticated laboratory infrastructure, skilled operators, complex sample preparation, consume a lot of energy and time. Additionally, the transportation of environmental samples to centralized laboratories increases operational costs and decreases the practicality of field research. These constraints have created a demand to design and construct novel, simple, green, easy to use sensing platforms for the determination of heavy metals.

Nanotechnology-based sensing systems have become viable alternatives for environmental monitoring in the last several years (Mutalik et al., 2025; Singh et al., 2023). Recent reports have also demonstrated the use of other nanostructured optical materials for the detection and remediation of organic compounds and trace metal ions in aqueous medium (Singh et al., 2026; Kukkar et al., 2018; Singh et al., 2019). Among a number of the various types of nanomaterials, gold nanoparticles (GNPs) have received considerable interest due to their highly developed optical, electronic, catalytic and plasmonic properties (Kumalasari et al., 2024). GNPs show LSPR, which is due to the collective oscillation of conduction electrons when they are exposed to light. The light absorption and strong scattering in the visible region give rise to highly sensitive optical responses to environmental changes. Color changes and shifts in UV-Visible spectra of nanoparticles (NPs) due to changes in size, aggregation, surface chemistry, and interparticle distance create opportunities for analyte detection through visual observation and UV spectrophotometric analysis. Due to these characteristics, GNPs are extensively studied for detection of heavy metals, pesticides, antibiotics, biomolecules, pathogenic organisms, and environmental pollutants.

Considering the importance of green chemistry and sustainable development, biologically-synthesized GNPs, using plant extracts or microorganisms, are being studied (Khan F. et al., 2022). This synthesis is considered to be more advantageous since it uses non-toxic chemicals and is more cost effective, environmentally safer, and biocompatible. Many phytochemical constituents of plant extracts such as flavonoids, alkaloids, terpenoids, tannins, phenolics, proteins, and polysaccharides are capable of reducing gold ions to form stable NPs (Singaravelu et al., 2025). Annona squamosa (AS) is one of the plants explored for green synthesis of GNPs due to its high phytochemical content and medicinal properties (Malik et al., 2022).

Annona squamosa, or custard apple is a tropical plant that contains leaves rich in bioactive compounds such as acetogenins, flavonoids, alkaloids, phenolics, saponins, and antioxidants (Kumar et al., 2021). These bioactive compounds can assist in the facile synthesis and stabilization of GNPs. The leaf extract of Annona squamosa has been used to synthesize GNPs which have been used in the areas of biomedicine, catalysis, antioxidants, and environmental sciences (Moussa et al., 2024). The biosynthesized GNPs have shown catalytic activity, biocompatibility, significant antimicrobial activities towards pathogenic microorganisms, and a highly effective antioxidant activity. Phytochemicals improve the natural capping, NP’s stability, and their interaction with target analytes in aqueous mediums.

Using GNPs from Annona squamosa as a Cd²⁺ sensor is advantageous as the presence of phytochemicals enables the interaction with metal ions and may control the aggregation of NPs and the plasmon resonance (Singh et al., 2018; Singh et al., 2021). Cd²⁺ induced aggregation of these NPs may exhibit a visible colorimetric and UV-Vis spectral change. Thus, this work aims to use the Annona squamosa leaf extract (ASLE)-derived GNPs as a nano-plasmonic tool for the colorimetric and UV-Vis detection of Cd²⁺. This work is primarily concerned with the formulation and characterization of ASLE-GNPs and the evaluation of their optical response, selectivity, and analytical activities for Cd²⁺ in the lab setting.

2 Methodology and tools

2.1 Materials

Fresh Annona squamosa leaves were obtained in September 2025 from the S.A.S. Nagar region of Punjab state, India. Chloroauric acid (HAuCl₄), hydrochloric acid (HCl), nitric acid (HNO₃), and sodium hydroxide (NaOH) at analytical grade, as well as metal ion analytes, were purchased from Merck. Ultra-pure Milli-Q (MQ) water with a resistivity of 18 MΩ·cm was used for the dispersion and reaction media. The laboratory glassware was cleaned with a 4% ES-7X aqueous cleaning solution, rinsed with MQ water, and dried to minimize contamination. The pH of the reaction mixtures was adjusted with NaOH or the plant extract’s naturally occurring phytochemical constituents. Experiments on synthesis and sensing were done using an N₂-filled thermometer from GH Zeal Ltd, London, England, to measure temperature and using a digital analytical pH meter from Mettler Toledo, Columbus, USA, to measure pH.

2.2 Methods

2.2.1 Preparation of Annona squamosa leaf extract (ASLE)

The ASLE was produced using leaves of the Annona squamosa plant that is edible and usually considered agricultural waste. The leaves extracts were using a modified extraction method based on previously published methods (Malik et al., 2022). The leaves were thoroughly washed and cleaned using MQ-water several times to remove dust and surface contaminants. The clean leaves were then crushed and mixed with MQ-water. The extraction was performed at grinded leaves/MQ-water volume ratio equal to 1:3 and heated over 100 °C until the original volume reduced to its half. Crude extract was filtered using Whatman filter paper to remove plant debris and suspended particles. The sample was purified by centrifugation at 13,500 rpm for 1 h at room temperature, and supernatant was filtered using a 0.22 µm syringe membrane filter. The ASLE was kept at 4 °C for future experiments.

2.2.2 Green synthesis of GNPs using ASLE

ASLE was used to reduce and stabilize GNPs by means of green synthesis. The reducing potential of leaf extract for chloroauric acid (HAuCl₄) was optimized by varying different physicochemical parameters to study the effect of concentration of extract and precursor. The optimal synthesis conditions of the ASLE-GNPs were sought by varying the volumes of 1 mM HAuCl₄ while keeping a constant amount of ASLE. The optimal precursor-to-extract ratio (P:E) was 6.2:1 (v/v), which was determined to be 3,100 μL of 1 mM HAuCl₄ and 500 μL of ASLE. The successful reduction of the Au³⁺ ions to Au⁰ NPs was confirmed by the gradual formation of a ruby-red color that is characteristic feature of the GNP’s SPR phenomenon. The most stable and efficient ASLE-GNPs were achieved by optimizing the solution pH = 8.2, heating the reaction to 100 °C, and using a P:E of 6.2:1 (v/v). Under these conditions, stable, colloidal, and substantially dispersed ASLE-GNPs were formed in the aqueous medium that possess reliable optical characteristics. The synthesized NPs were subsequently purified by repeated washing with deionized water (DI-water) followed by centrifugation at 13,500 rpm to remove excess phytochemicals and unreacted species. The purified NPs were then reconstituted in deionized water for further studies.

2.2.3 Sensing of metal ions

The sensing performance of the ASLE-GNPs for different metal ions was analyzed using UV–visible spectroscopy to study the LSPR behavior. The effects of various metal ions due to the addition of different metal chlorides of Al³⁺, Ba²⁺, Co2⁺, Fe²⁺, Hg²⁺, In²⁺, K⁺, Li⁺, Mg²⁺, Mn²⁺, Na⁺, Ni²⁺, Pb²⁺, Pt²⁺, Cd²⁺, Sn²⁺, and Zn²⁺ on the optical response of the colloidal ASLE-GNPs were investigated. For initial selectivity studies, one part per million (ppm) aqueous metal ion solutions were incubated within the aqueous suspensions of GNPs for 5 min and then subjected to UV-visible analysis. Among the studied analytes, Cd²⁺ showed the strongest binding with ASLE-GNPs, demonstrated by significant bathochromic shifts of the LSPR peak and noticeable color change. Additional studies confirmed that the sensors were able to detect Cd²⁺ with ASLE-GNPs across a wide range of 0.025–4 ppm. Interference studies were also performed to study the selectivity of the sensing system in the presence of other metal ions. Even in the presence of other competing ions, the ASLE-GNP system demonstrated an optical response that was solely directed towards Cd²⁺.

2.2.4 Basic quality assurance studies (QAS)

Quality assurance of the analytical sensing performance of ASLE-GNPs was assessed determine the influence of signal-to-noise ratio (S/N) on UV-Visible spectra (UVS) and to establish the LOD and the LOQ using formulae of 3.3σ/m and 10σ/m, respectively. The ‘σ’ calculated from the standard deviation of UVS-readings of blank samples, whereas the slope ‘m’ calculated from the line equation ‘y = mx + c’ of linear fitting curve produced from the UVS-reading of ASLE-GNPs treated with different concentrations of Cd²⁺ ions followed by incubation for 10 min. This work was carried out to establish the analytical performance of the ASLE-GNPs in sensing Cd²⁺ in the stated laboratory conditions.

2.3 Characterization techniques

Optical studies of the synthesized ASLE-GNPs both before and after interaction with metal ions were performed using the UV-vis Shimadzu UV-2600+ with quartz cuvettes with a 10 mm optical path length. Particle size and zeta potential were assessed with a Malvern Nano-ZS to study the NP’s uniformity and the suspension’s stability. The Perkin-Elmer Spectrum-2 FTIR instrument in attenuated total reflection (ATR) was used to analyze the functional groups and phytochemical interactions linked to ASLE and ASLE-GNPs. High-resolution transmission electron microscopy (HR-TEM) of ASLE-GNPs was carried out with a 200K V accelerating voltage. For TEM imaging, the nanoparticle (NP) suspensions were deposited and dried on carbon-coated copper grids. The HR-TEM was used to analyze the elemental compositions of the NPs and conduct simultaneous energy-dispersive X-ray spectroscopy (EDX) of the NPs.

3 Results and discussion

The synthesis of ASLE-GNPs was systematically optimized for two physicochemical parameters including pH, and gold precursor-to-extract (P/E) ratio.

3.1 Effect of reaction pH

The SPR analysis of ASLE-GNPs showed absorbance peaks at 535, 548, 541, and 521 nm at pH 3.39, 4.93, 7.04, and 8.20, respectively (see Figure 1A). Comparison of the absorption peaks of pH 7.04 and 8.20 with those at pH 3.39 and 4.93 showed sharper peaks, indicative of the formation of larger particles and partial particle aggregation (Singh et al., 2018). The suspensions turned bluish-black after 24 h and formed aggregates, resulting from inadequate stabilization due to the protonation of the phytochemical functional groups. Unlike the ASLE-GNPs at the other pH values, ASLE-GNPs synthesized at pH 8.20 remained ruby-red and stable. This observation indicated that under alkaline conditions the capping and electrostatic repulsion forces were stronger and the particle size was smaller, thus improving the colloidal stability of ASLE-GNPs at pH 8.20.

3.2 Effect of gold precursor-to-extract (P/E) ratio

The ratio between chloroauric acid and ASLE strongly influenced NP nucleation, growth, and stability (Singh et al., 2021). Controlled addition of small volumes of 1 mM HAuCl₄ to a fixed 0.5 mL volume of ASLE extract (dissolved in 25 mL of DI-water) resulted in a sequential color change of the solution from transparent to faded pink, pink, and finally ruby red, confirming the GNP’s formation (Figure 1B). The extent of Au³⁺ reduction and NP formation was directly related to the HAuCl₄ concentration (1 mM) and its volume (3,100 μL) due to adequate phytochemical capping and increased NP growth. Optimized P/E ratios of 6.2:1 (v/v) provided improved stabilization and balanced the growth and nucleation of the ASLE-GNPs.

3.3 Synthesis of ASLE-GNPs

Once a stable matrix was produced, a GNP design with controlled dimensions and improved detection capability was achieved by optimizing synthesis conditions at pH = 8.2 and P/E = 6.2/1, and reaction temperature = 100 °C (Dinh et al., 2025). In these conditions, the reaction media underwent a distinct change from light yellow to ruby red following the addition of HAuCl₄, confirming the GNP’s formation by their SPR features. ASLE compounds reduced the Au³⁺ ions to GNPs and subsequently prevented the NP suspensions from agglomeration in water. The GNPs synthesized showed a uniform size and a high degree of stability in colloidal form for at least 1 month. The synthesized GNPs were characterized using appropriate analytical techniques in the following sections.

3.4 Ultraviolet–visible spectroscopic analysis

The optical properties and SPR of the GNPs were examined by using UV–visible spectroscopy and presented in Figure 2A. The UV–visible absorption spectrum of the synthesized GNPs showed a strong absorption peak in the range of 500–550 nm, thus confirming the formation of GNPs in the aqueous medium (Singh et al., 2018). The maximum SPR absorption peak (λ_max) was recorded at approximately 521 nm, indicating that the GNPs were in a colloidal stable form. Moreover, the strong absorption peak indicated that a sufficient concentration of NPs was formed in the aqueous medium. Along with the SPR peak associated with GNPs, the ASLE also showed UV absorption peaks due to the presence of bioactive phytochemicals (Kumar et al., 2021). The absorption peaks observed at 260 and 316 nm were attributed to flavonoids such as flavanones and flavones, which were present in the leaf extract. These phytochemicals likely assisted the reduction of Au³⁺ ions and the stabilization of the synthesized NPs by forming a protective layer on the surface. Thus, the UV–visible absorption spectra confirmed the presence of ASLE-GNPs.

3.5 Fourier transform infrared (FTIR) analysis

FTIR Spectroscopy was used to investigate the surface functional groups of ASLE and ASLE-GNPs in (Figure 2B). A distinct band in the FTIR spectrum of ASLE was observed at 3,746 cm^-1. It can be ascribed to the stretching vibration of free or weakly hydrogen-bonded O–H groups, indicating the presence of the phytochemicals in ASLE which contain hydroxyl groups (Das et al., 2016). The absorption bands at 2,926 and 2,859 cm^-1 are attributed to the asymmetric and symmetric stretching vibrations of the aliphatic C–H bond, which indicates the presence of aliphatic hydrocarbon chains in the phytochemicals derived from the leaves. The strong band at 1745 cm^-1 was assigned to the stretching vibration of carbonyl (C=O) and was considered to be of aldehydes, ketones, esters, or carboxylic acids. The 1,553, 1,464, and 1,386 cm^-1 bands are reported to be of polyphenolic constituents, which are consistent with the presence of phytochemicals in ASLE. Further, a strong absorption at 1,033 cm^-1 was assigned to the C–O stretching vibrations of phenolic compounds.

The spectra comparison before and after the NPs synthesis reflected the changes in the peak positions and intensities which can be attributed to the surface phytochemicals contributions during the NP’s formation (Gangapuram et al., 2018). The peak positions indicating the functional groups of the phytochemicals shifted from 1,464 to 1,407 cm^-1, 1,386 cm^-1 to 1,370 cm^-1, and 1,033 cm^-1 to 1,095/1,023 cm^-1. The changes in the spectra reflected the possible formation of a local chemical environment after the NP’s formation. Furthermore, the intermediate phytochemicals, which are suspected to be primarily carbonyls, phenols, and alcohols, reduced the gold ions (Au³⁺) and were most likely stabilized on the gold (Au) NPs by the capping agents. Thus, this study confirmed the establishment and stabilization of the nano-plasmonic colloidal GNP system using ASLE.

3.6 Morphological and elemental analysis

To assess nano-plasmonic sensing applications, morphological and elemental characterizations were conducted on ASLE-GNPs to investigate their surface architecture, structural uniformity, elemental composition, and physicochemical stability (Ngo et al., 2022). TEM characterization revealed that nanoscale crystals having uniform quasi-spherical morphology and size distributions were formed successfully within range of 3–11 nm (Figure 3A). Further, the size analysis of 20 NPs (selected from TEM graph) also showed a mean particle size of 6.85 ± 1.84 nm, showing a good agreement with the previously reported particle size distribution of the synthesized NPs showing a distribution in the nanoscale (Figure 3B). (Singh et al., 2021) The HRTEM graph in Figure 3C revealed the lattice fringes of interplanar spacing in range of 0.246–0.250 nm corresponding to (1, 1,1) crystal planes of face centred cubic (fcc) GNPs. The synthesized GNPs demonstrated dispersed structural features, indicating phytochemical constituents providing efficient capping and stabilization to GNP crystals.

Based on elemental analysis, energy peaks depicted at 2.12, 9.71, and 11.4 KeV confirms the presence of Au as the primary metal (see Figure 4A). While the energy peak at 0.28 KeV reveals the availability of carbon either from the capping of phytochemicalsf absorbed on the GNPs or because of copper-coated carbon. The copper peaks at 8.04 and 8.90 KeV were the result of a copper-coated carbon support which is used to prepare samples for imaging.

Additionally, the zeta potential of ASLE-GNPs was measured at −27.66 ± 0.21 mV (Figure 4B). These nanoparticles were stabilized in water due to a negative surface charge, and electrostatic repulsion was generated from ionizable groups of the phytochemical constituents of ASLE-GNPs. Furthermore, the high negative zeta potential exhibited the role of phytochemicals in the stabilization of ASLE-GNPs (Egorova et al., 2020).

Besides it, the ASLE-GNPs possessed a Z-average hydrodynamic diameter of 33.22 nm, with a polydispersity index (PDI) of 0.190 (Figure 4C). The relatively low PDI defined a good colloidal uniformity and a narrow hydrodynamic size distribution. The DLS measurement, which defined a significant difference between the GNP hydrodynamic diameter and the GNP diameter from TEM, was attributed to the surface-bound phytochemical constituents and solvation layer of the GNPs. The phytochemical constituents of ASLE may include polyphenolics, flavonoids, and other biomolecules, which may act as stabilizing and capping agents in the process of GNP synthesis (Egorova et al., 2020).

4 Solution-based spectroscopic and colorimetric detection of trace metal IONS

GNPs are responsive to changes at their surface and exhibit a strong relationship between plasmonic absorption properties and their interaction with external analytes (Zhao et al., 2026). Therefore, UV–visible spectroscopy was employed to investigate the LSPR behavior of the synthesized ASLE-GNPs toward different trace metal ions in aqueous media. Such interactions can induce aggregation, or surface coordination effects, leading to shifts in the LSPR absorption band accompanied by visible colour changes in the NP suspension. Owing to these unique optical characteristics, the synthesized ASLE-GNPs were explored as nano-plasmonic probes for the spectroscopic and visual detection of heavy metal ions.

The UV–visible spectrum of ASLE-GNPs displayed a characteristic LSPR absorption maximum at approximately 521 nm. When GNPs interacted with different metal ions for 15 min (1 ppm each), each of the metal ions caused a different optical signature (see Figure 5A). Among the ions tested, the Cd²⁺ ion interacted most strongly with ASLE-GNPs, resulting in a significant bathochromic shift of the SPR band from 521 to 554 nm. Simultaneously, the colour of the GNP-suspension gradually transformed from ruby red to bluish coloration, indicating aggregation and formation of unstable NP–metal ion assemblies (see Figure 5B). These aggregated nanostructures became increasingly visible with prolonged incubation time. However, the other studied metal ions caused little to no significant spectral variation in the LSPR band of ASLE-GNPs, demonstrating that ASLE-GNPs exhibited a selective optical response to Cd²⁺ ions under the investigated experimental conditions.

Further experiments including sensitivity and interference tests demonstrated that the presence of other interfering ions had little to no impact on the response of the developed nano-plasmonic system for sensing Cd²⁺ ions. For sensitivity test, ASLE-GNPs showed strong response to Cd²⁺ ions in the 0.025–1 ppm concentration range for an incubation period of 30 min. As the concentration of Cd²⁺ ions increased, the LSPR peak shifted to a more red-shifted region, and the color of the NP suspension changed to purple-blue (see Figure 6A). The extent of the shift in the spectrum was directly proportional to the concentration of Cd²⁺ ions, and strong absorbance variation was observed (Singh et al., 2021). At low concentrations of Cd²⁺, absorbance intensity was less affected due to NP aggregation and cluster formation. However, at higher concentrations of Cd²⁺ ions (2–4 ppm), the Cd ions were not reached toward the GNP’s surface much efficiently and get saturated as no active sites available for sufficient interactions.

The UVS reading for Cd²⁺–treated ASLE-GNPs and ASLE-GNPs are provided in Table 1. The analytical performance of the designed ASLE-GNP nano-plasmonic sensor was examined via a calibration curve that plotted the decrease in the absorbance signal at the characteristic LSPR wavelength against exposure to incrementally increasing Cd²⁺ concentrations (see Figure 6B). Each concentration of Cd²⁺ was measured three times. In the same experimental conditions, the control measurement, which was the ASLE-GNP suspension in the absence of Cd²⁺, was measured seven times. The LOD and LOQ were evaluated using the standard equations 3.3σ/m and 10σ/m, respectively, considering the standard deviation of the control measurements (σ) and the slope of the calibration curve (m). For the concentration range of 0.025–1.0 ppm of Cd²⁺, the calibration curve yielded an LSPR absorbance relationship with a linearity of (Adj. R ²) = 0.950. The calculated LOD and LOQ were 47.9 ppb and 145.2 ppb, respectively. These results made the ASLE-GNP nano-plasmonic sensor reliable for the spectrophotometric analysis of laboratory-prepared aqueous samples of Cd²⁺. Even with the limits of detection higher than the Cd²⁺ drinking-water standards, the sensor is applicable for the estimation of Cd²⁺ levels in laboratory conditions. Before practical application in real field scenarios, more studies are necessary to confirm the efficacy of the sensor in the analysis of real samples, and cost and feasibility studies are needed to design a portable version of the sensor.

TABLE 1. QAS of ASLE-GNPs plasmonic sensors treated with different concentrations of Cd²⁺ ions.

R1, UVS, reading cycle-1; R2, UVS, reading cycle-2; R3, UVS, reading cycle-3.

The colorimetric sensing ability of ASLE-GNPs was further assessed through interference studies in a mixed-metal system containing Cd²⁺ ions along with several competing ions of different elemental blocks (see Figures 7A–C). The samples of mixed-metal system were prepared according to table provided in Figure 7D. In the presence of Cd²⁺ ions, large bathochromic shifts and color changes were observed whereas in the absence of Cd²⁺ ions, there were no noticeable alterations in the spectra. These results indicate that the ASLE-GNPs exhibited a selective response toward Cd²⁺ in the presence of the investigated competing ions under the experimental conditions employed.

4.1 Spectroscopic sensing mechanism

Functional groups on the NP’s surface can help to stabilize the ASLE-GNPs, as well as assist in the recognition of the analyte through the formation of metal–ligand interactions (Yaraki et al., 2022). Phytochemicals including hydroxyl and carboxyl moieties has high affinity towards Cd²⁺ ions for metal–ligand bond formation. Thus, after GNPs–Cd²⁺ interaction the negative charge on the ASLE-GNP’s surface get neutralized to some extent by the Cd²⁺ ions coordination to functional groups. The ASLE-GNPs in solution will then repel less, leading to an increase in the NPs’ aggregation. Subsequently, the decrease in the NP’s distance promotes the coupling of adjacent NPs, and the shift in the LSPR band of NPs to lower energy and it is manifested by the colour change of the colloidal solution from wine red to blue.

The characterization results after sensing provide strong evidence for the proposed aggregation mechanism. After exposure to Cd²⁺, the hydrodynamic diameter and PDI of the ASLE-GNPs were increased to 118.7 nm and 0.288, respectively, suggesting that the particles formed an aggregate in solution (see Figure 8A). The zeta potential also demonstrated a change from −27.66 ± 0.21 mV for the ASLE-GNPs to −14.63 ± 0.70 mV after the Cd²⁺ treatment (Figure 8B). The decrease of the zeta potential indicates that the NP’s surface became less negatively charged. This change suggests that the Cd²⁺ ions interacted with the surface phytochemical moieties of the ASLE-GNPs and as a result decreased the electrostatic repulsion. In concordance with the DLS and Zeta results, the TEM image of Cd²⁺-treated ASLE-GNPs showed that the NPs were packed closely together rather being dispersed (see Figure 8C). Overall, the DLS, Zeta, and TEM results show that after the treatment with Cd²⁺ the surface charge of the GNPs changed and interparticle aggregation was observed. This also explains the observed changes in the LSPR position and the colorimetric change associated with the sensing.

Other metal ions, such as Pb²⁺ and Hg²⁺, can interact with the oxygen and other functional groups of NPs coated with phytochemicals (Rasool et al., 2025; Hoshino et al., 2009; Zhao et al., 2018). However, they may interact with the surface-bound phytochemicals less favourably under the experimental setup used in this study. The same can be said for Fe²⁺, Co²⁺, Mn²⁺, Ni²⁺, and Zn²⁺ (Singh et al., 2021). Although they may interact with the surface functional groups to some extent, they did not produce a comparable change to the LSPR response. The aforementioned examples may be attributed to the differences in metal–ligand coordination behavior, metal hydration behavior, ionic dimensions, and interaction with the surface functional groups. Consequently, the selective behavior observed cannot only be attributed to the ionic radius, but must also consider metal–ligand interaction and the physicochemical behavior of the metal ions.

The selective spectroscopic response of ASLE-GNPs to Cd²⁺ observed in this study might be due to the interaction of Cd²⁺ ions with surface-bound phytochemicals, which results in the neutralization of the NPs surface charge and the enhanced interparticle aggregation of the NPs (Figure 9). The ionic dimensions and coordination characteristics of Cd²⁺ ions may be more favourable for this interaction. These effects combine to facilitate the aggregation of the NPs and promote plasmon coupling. This leads to the typical observable LSPR color changes and the bathochromic shift, along with changes in the UV–Vis spectra, when Cd²⁺ is detected.

4.2 Comparative analysis of ASLE-GNP-based Cd²⁺ sensing with previous studies

The analytical performance of the ASLE-GNP nano-plasmonic sensor was benchmarked against published GNP-based Cd²⁺ detection systems (Table 2). When needed, the LOD conversion to μg/L was applied as a basis for LOD comparison, along with the comparison of linear range, response time, sample matrix, selectivity, and surface functionalization. ASLE-GNPs demonstrated an LOD of 47.9 μg/L and an extended quantitative working range (25–1,000 μg/L) with a 10 min response time. Although some functionalized GNP-based sensors reported lower LODs, they often used additional recognition strategies and surface functionalization (Kim et al., 2001; Marzo et al., 2013; López_Marzo et al., 2013). For example, paper-based immunosensors reported LODs in the range of 0.20–0.40 μg/L but relied on cadmium-specific antibodies, Cd–EDTA–BSA conjugation, and recognition components (Marzo et al., 2013; López_Marzo et al., 2013). In contrast, the ASLE-GNP sensor was able to detect Cd²⁺ without any additional chemical functionalization of the sensor.

TABLE 2. Comparative analysis of ASLE-GNP-based Cd²⁺ sensing with previously reported GNP sensors.

AHMT, 4-amino-3-hydrazino-5-mercapto-1, 2,4-triazole; MC, momordica charantia; EDTA, ethylenediaminetetraacetic acid; BSA, bovine serum albumin.

The comparison also showed the differences in the sensing matrices and the way selectivity was assessed. Many of the reported systems were assessed in one of drinking water, tap water, rice digest, waste water, or fruit-based matrices (Singh et al., 2021; Marzo et al., 2013; López_Marzo et al., 2013; Guo et al., 2014), while this work was assessed in DI-water based media. Given that sample composition and matrix effects can greatly influence the optical response and NP’s aggregation, a more cautious and less direct comparison of the practical analytical performance is warranted. Regardless, ASLE-GNPs showed a selective response to Cd²⁺ from a total of 17 metal ions, which generated the most pronounced optical and LSPR response of all the metal ions tested, with no recognition strategies or other post synthetic molecules (Kim et al., 2001; Marzo et al., 2013; Guo et al., 2014).

Among green-synthesized systems listed in Table 2, MC-GNPs have an LOD of 34.10 μg/L and a working range of 200–800 μg/L, and AMPuE-GNPs have an LOD of 8.16 μg/L and a wider working range of 8.25–1,632 μg/L (Singh et al., 2018; Singh et al., 2021). Thus, ASLE-GNPs do not have the lowest LOD in the green-synthesized GNP systems. However, the current platform integrates the synthesis from plant extracts, 10-min optical response selective to Cd²⁺, and no additional functionalization to the surface after synthesis. These features streamline both the synthesis and the sensing processes. The phytochemicals of ASLE serve both as reducing agents and as stabilizing/capping agents, while the synthesized nanoparticles’ surface chemistry offers a binding site for Cd²⁺. The overall comparison shows that ASLE-GNPs provide an easy, green-synthesized optical sensing method for initial detection of Cd²⁺ in water, although more testing will be needed to validate their use in diverse real-world environments to assess their true applicability.

5 Conclusion

This study illustrates the green synthesis of GNPs using ASLE and the study of their use as a nano-plasmonic platform for the detection of Cd²⁺. The synthesized ASLE-GNPs exhibited a quasi-spherical morphology with an estimated size of 3–11 nm, and an average size of 6.85 ± 1.84 nm based on the analysis of 20 nanoparticles. DLS analysis indicated a Z-average of 33.22 nm, with a PDI of 0.190, and a zeta potential of −27.66 ± 0.21 mV resulting in a negatively charged surface, due to the phytochemicals. These characteristics suggest that ASLE-derived phytochemicals contributed to the stabilization and colloidal dispersion of the GNPs. The ASLE-GNP exhibited a quantifiable and selective optical response towards Cd²⁺ during the laboratory experiment. The optimum concentration range for quantification was 0.025–1.0 ppm with an Adj. R ² of 0.950. The LOD and LOQ were found to be 47.9 ppb and 145.2 ppb, respectively. Even though the LOD did not meet the most demanding drinking-water guidelines, the device can be used for the first approximation of high concentration in Cd²⁺ within the aqueous samples prepared in the laboratory. Measurement of the sensing response was reached in less than 10 min and does not require any surface modification with post-synthesis ligands, antibodies, aptamers, or other chemicals.

Post-sensing characterization lends more support to proposed mechanisms for Cd²⁺ sensing. Following exposure to Cd²⁺, the average hydrodynamic diameter increased from 33.22 nm to 118.7 nm with a PDI of 0.288. Further, the zeta potential changed from −27.66 ± 0.21 to −14.63 ± 0.70 mV. A proposed mechanism involves the Cd²⁺ interaction with phytochemicals bound to the GNP’s surfaces, a partial reduction of the electrostatic repulsion, and significant NP’s aggregation. Plasmonic coupling of the NPs leads to the observed optical response and a colorimetric shift. When compared to other GNP-based Cd²⁺ sensors, the current system does not achieve the lowest detection limit. The current system does incorporate green NP synthesis, a 10-min response time, selective detection of Cd²⁺ under the conditions tested, and no post-synthesis surface functionalization. The systems must be compared with the sensors cited to evaluate detection conditions, sensing matrices, response calibration and range, and specificity and selectivity measures.

6 Future perspectives

The research focused on simple laboratory aqueous solutions, and thus, the performance of ASLE-GNPs is not known for the complex environmental matrices. Future work will describe real sample testing on spiked samples of tap water, groundwater, river water, and industrial waste. Recovery studies and comparisons with AAS, ICP-OES, and ICP-MS will be included for assessment of instrumentation performance and analyte matrix effect. Additional selectivity studies are needed to assess the impact of common aqueous constituents (e.g., Ca²⁺, Mg²⁺, Na⁺, K⁺, Cl⁻, SO₄ ²⁻, NO₃ ⁻, and CO₃ ²⁻, as well as competing metal ions) at environmentally relevant concentrations. While these studies are ongoing and will assess sensor robustness within intricate water matrices, more research is needed to address the development of a portable detection format and assess the cost of operation. Until then, it is necessary to consider this research as proof-of-concept at the laboratory scale for green-synthesized ASLE-GNPs, which will serve as a basic optical tool for preliminary detection of Cd²⁺ and will incorporate the environmental testing needed to support a conclusion regarding the practicality of the tool.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Daniela Šojić Merkulov, University of Novi Sad, Serbia

Reviewed by: Osama A. Fouad, Head of Nanostructured Materials and Nanotechnology Department, Egypt

Deepak Verma, United Arab Emirates University, United Arab Emirates