Optimization of gold nanoparticles conjugated with four thiol-based ligands for the colorimetric detection of cadmium ions in water samples

Abstract

This study optimized colorimetric gold nanoparticle (AuNP) probes conjugated with four thiol-based ligands, namely, 1,3,4-thiadiazole-2,5-dithiol (TDDT), 2-amino-5-(methylthio)-1,3,4-thiadiazole (AMTTD), mercaptopropionic acid with l-alanine (MNA–Ala), and N-acetyl-l-cysteine (NAC), for cadmium (Cd²⁺) detection by tuning ligand loading, pH, salt, and incubation. UV-vis spectroscopy, FTIR spectroscopy, and HR-TEM data supported probe formation and response trends. Among the four ligands, MNA–Ala–AuNPs exhibited the most reliable performance under the optimized conditions of 10 mM MNA, 1 mM Ala, pH 10, and 100 mM NaCl with a 60-min pre-incubation, yielding a linear A₆₄₀/A₅₂₀vs. log[Cd²⁺] fit (R² = 0.9933). TDDT, AMTTD, and NAC showed only minor spectral/color changes. The enhanced Cd²⁺ sensitivity of MNA/Ala compared to TDDT, AMTTD, and NAC is rationalized by its flexible donor architecture, enabling effective Cd²⁺ chelation, unlike the rigid, sterically hindered, or partially masked thiol donors of the other ligands. MNA–Ala–AuNPs were used to identify the presence of Cd²⁺ in a household water sample collected from Kelapa Dua, Depok, and the analytical results indicated that the sample did not contain Cd²⁺.

Cd²⁺ bridges the carboxylate (MNA) and amino (Ala) groups on AuNP surfaces, cross-linking adjacent particles. This shortens interparticle distance, red-shifting and broadening the SPR band, giving a red-to-blue/purple colorimetric change.

Introduction

Cadmium (Cd²⁺) can exert toxic effects on the kidneys, skeletal system, and respiratory system, and it is classified as a human carcinogen.¹ The presence of Cd²⁺ in water, the environment, food, and beverages is therefore strictly regulated to prevent harmful health effects. Given its hazardous and carcinogenic effects, reliable and sensitive analytical methods for Cd²⁺ detection are particularly essential in water, which is widely used for drinking, agricultural, industrial, and domestic purposes.

Various methods have been developed for Cd²⁺ detection, including electrochemical techniques,² atomic absorption spectrophotometry (AAS),³ high-performance liquid chromatography (HPLC),⁴ and fluorescence sensors.⁵ Methods such as HPLC, AAS, and related techniques offer high sensitivity and selectivity due to advanced instrumentation for Cd²⁺ detection. However, these approaches are associated with several drawbacks, including high costs, the need for skilled operators, limited portability, and time-consuming sample preparation procedures. Consequently, there is a demand for simple and rapid analytical methods. In this context, colorimetric methods are attractive because they are simpler, faster, and more cost-effective and allow detection results to be observed with the naked eye.⁶

Colorimetric detection of heavy metals can be achieved using gold nanoparticles (AuNPs) owing to their unique optical properties and high sensitivity.^7,8 AuNPs exhibit a pronounced color change upon aggregation, a phenomenon known as surface plasmon resonance.⁹ This color change occurs when metal ions, including Cd²⁺, interact with the surface of AuNPs, inducing aggregation and causing a shift in the absorption wavelength that can be exploited for colorimetric detection.¹⁰ In addition, AuNPs are chemically stable, resistant to oxidation, and relatively non-toxic, making them suitable for analytical applications.⁹

Several studies have reported the determination of Cd²⁺ using colorimetric methods based on conjugated AuNPs. Previous work has involved the conjugation of AuNPs with DNA and peptides, such as glutathione,¹¹l-cysteine,¹² and 3-(1H-tetrazol-5-yl)aniline.¹³ The use of DNA and peptide conjugates offers high specificity but generally requires intensive labor, higher costs, and longer preparation times.¹⁴ In contrast, conjugation with other chemical compounds, such as dl-mercaptosuccinic acid and trithiocyanuric acid, requires less time and simpler sample preparation compared to methods employing more complex conjugates.¹⁵ AuNPs readily interact with carboxyl, thiol, and hydroxyl functional groups.⁹ Cd²⁺ can bind to hydroxyl groups,¹⁶ carboxyl groups,⁹ as well as amino and thiol groups.¹⁷ Therefore, conjugates containing these functional groups are essential, as such interactions play a crucial role in the development of sensitive colorimetric methods for Cd²⁺ determination.

2-Amino-5-(methylthio)-1,3,4-thiadiazole (AMTTD) consists of a five-membered heterocyclic ring and contains three amino groups, one methyl group, and two sulphur-containing groups, where the methylthio can be converted into a thiol (–SH) group via sulfur demethylation. 1,3,4-Thiadiazole-2,5-dithiol (TDDT) is an organosulphur compound containing a thiadiazole moiety with two amino groups and two thiol (–SH) groups at the 2 and 5 positions. N-Acetyl-l-cysteine (NAC) consists of a cysteine backbone with an acetyl group attached to the nitrogen atom (N-acetyl) and a thiol (–SH) group on the side chain. Meanwhile, 3-mercaptopropionic acid (MNA) is an organosulphur compound that consists of a propionic acid chain with a thiol (–SH) group attached to the third carbon (beta position). The presence of a thiol group enables strong interactions with heavy metals, particularly Cd²⁺. In this study, AMTD, TDDT, MNA, and NAC are proposed as novel AuNP conjugates that have not previously been investigated for the colorimetric detection of Cd²⁺ in water. This research aims to evaluate the capability of these AuNP conjugates to detect Cd²⁺ using a colorimetric method in aqueous media, offering an approach that is cost-efficient, rapid, and suitable for direct or real-time analysis.

Materials and methods

Materials and reagents

For probe creation, gold nanoparticles (10 nm diameter, OD 1) dispersed in 0.02 M sodium citrate were obtained from Sigma-Aldrich, USA. 1,3,4-Thiadiazole-2,5-dithiol (TDDT), 3-mercaptopropionic acid (MNA), l-alanine (ALA), N-acetyl-l-cysteine (NAC), CdCl₂, and sodium chloride were purchased from Sigma-Aldrich, USA. 2-Amino-5-(methylthio)-1,3,4-thiadiazole (AMTTD) was obtained from BLD Pharm, China. Phosphate buffer (0.01 M) with pH ranging from 5 to 11 and distilled water were used in this study. The water sample was household water from Kelapa Dua, Depok.

Initial characterization of AuNPs

In the initial stage, the characterization of AuNPs was carried out using a Shimadzu UV-1900i UV-visible spectrophotometer, and ligand-functionalised AuNPs were prepared for Cd²⁺ sensing. UV-vis baselines were recorded by diluting 700 µL of AuNPs with distilled water to final volumes of 1850 µL (TDDT and AMTTD systems), 1900 µL (MNA–Ala system), or 2550 µL (NAC system).

TDDT- and AMTTD-conjugated AuNPs were obtained by mixing 700 µL of AuNPs with 1 mL of phosphate buffer (pH 7) and 50 µL of 10 µM ligand solution, followed by dilution to 1850 µL. MNA–Ala-conjugated AuNPs were prepared by combining 700 µL of AuNPs with 1 mL of acetate buffer (pH 10), 50 µL of 10 mM MNA, and 50 µL of 1 mM alanine, and diluting to 1900 µL. The NAC-conjugated AuNPs were prepared by mixing 700 µL of AuNPs with 1 mL of phosphate buffer (pH 7) and 50 µL of 10 mM NAC, followed by dilution to 2550 µL.

Preparation and optimization of ligand-functionalized AuNPs

Ligand-functionalized AuNPs were prepared by mixing 700 µL of AuNP solution with the corresponding ligand under buffered conditions. For TDDT- and AMTTD-functionalized AuNPs, 700 µL of AuNPs were combined with 1 mL of phosphate buffer (pH 7) and 50 µL of 10 µM TDDT or AMTTD solution, followed by dilution with distilled water to a final volume of 1850 µL. MNA–Ala-functionalized AuNPs were prepared by mixing 700 µL of AuNPs with 1 mL of acetate buffer (pH 10), 50 µL of 10 mM 3-mercaptopropionic acid, and 50 µL of 1 mM l-alanine, and diluting with distilled water to a final volume of 1900 µL.

To investigate the interaction sequence among Cd²⁺ ions, ligands, and AuNPs, preliminary orientation tests were conducted. For TDDT and AMTTD systems, 50 µL of 1 mM Cd²⁺ and 50 µL of 40 mM NaCl were mixed with TDDT or AMTTD. For MNA–Ala functionalized-AuNPs, the formulated AuNPs were mixed thoroughly with 50 µL of 1 mM Cd²⁺ and 50 µL of 20 mM NaCl. The color change was visually observed before recording the UV-vis spectrum.

For the NAC-functionalized AuNPs, 400 µL of 1 mM Cd²⁺ was mixed with 400 µL of 50 µM NAC and incubated for 30 minutes. Following incubation, each preformed Cd²⁺–ligand mixture was added to 700 µL of AuNPs and diluted with 1050 µL of distilled water before UV-vis spectral acquisition. In selected experiments, 400 µL of 1 mM Cd²⁺ solution was mixed with 400 µL of 50 µM NAC and incubated for 30 minutes, followed by the addition of 700 µL of AuNP solution, 1 mL of phosphate buffer (pH 7), and 50 µL of 10 mM NaCl before measurements.

Key experimental parameters affecting Cd²⁺ sensing were systematically optimized for each ligand–AuNP system. For TDDT-functionalized AuNPs, the ligand concentration was varied from 10 µM to 100 mM, followed by incubation for 30 s prior to the addition of NaCl and Cd²⁺; the mixtures were then allowed to stand for 10 min at room temperature before absorbance measurements. The conjugation reaction time was optimized using 1 mM AMTTD with incubation periods ranging from 0 to 180 min. The effect of pH was examined using phosphate buffers (pH 5–11) with 1 mM TDDT incubated for 25 min. The NaCl concentration was optimized over the range of 0–500 mM using 1 mM TDDT, and the incubation time after Cd²⁺ addition was varied from 0 to 60 min. For the AMTTD system, ligand concentration (10 µM–10 mM), conjugation time (0–180 min at 500 µM AMTTD), buffer pH (pH 5–11 using 10 µM AMTTD incubated for 45 min), NaCl concentration (0–500 mM at 500 µM AMTTD), and post-Cd²⁺ incubation time (0–60 min) were evaluated. For the MNA–Ala system, MNA (0–1 M) and l-alanine (1–500 mM) concentrations were optimized, followed by the evaluation of buffer pH (acetate buffer, pH 6–11), NaCl concentration (0–500 mM using 1 M MNA and 10 mM Ala), and incubation time after Cd²⁺ addition (0–60 min). For the NAC system, a preliminary orientation test was conducted by preincubating Cd²⁺ (1 mM) with NAC (50 µM) for 30 min prior to the addition of AuNPs; measurements were performed in either distilled water or phosphate buffer (pH 7) containing 10 mM NaCl.

Characterization of conjugated AuNPs

Characterization of the ligand-conjugated AuNPs was carried out under their respective optimized conditions. UV-visible absorption spectra were recorded to evaluate the optical response of AuNPs, while Fourier-transform infrared (FTIR) spectroscopy (Shimadzu IRXross, Japan) was employed to identify the functional groups involved in ligand coordination with Cd²⁺ ions and AuNP surfaces. The morphology, size distribution, and aggregation behavior of bare and ligand-conjugated AuNPs before and after Cd²⁺ exposure were examined using high-resolution transmission electron microscopy (TEM, Tecnai G2 from FEI Company, USA).

Sensitivity and selectivity tests of conjugated AuNPs

The selectivity and sensitivity of the AuNP-based probes were evaluated using MNA–Ala as ligands. For selectivity testing, various heavy metal ion solutions (Ca²⁺, Pb²⁺, Hg²⁺, Cd²⁺, Fe²⁺, and Zn²⁺; 100 µM) were added to MNA–Ala-conjugated AuNPs, and the mixtures were incubated for 60 min prior to UV-vis measurements (400–800 nm). For sensitivity tests, Cd²⁺ with concentrations ranging from 1 µM to 100 mM was preincubated with 400 µL of 1 mM MNA–Ala, followed by the addition of 700 µL of AuNPs, 1 mL of phosphate buffer (pH 6), and 50 µL of 10 mM NaCl prior to absorbance measurements (400–800 nm).

Application of conjugated AuNPs for Cd²⁺ detection in the water sample

50 µL of the household water sample was added to the MNA–Ala–AuNPs sensing solution prepared under optimized conditions (concentration, pH, and incubation time). Two experimental variations of the sample were tested: an unspiked sample and a sample spiked with Cd²⁺. The absorbance spectra were recorded in the wavelength range of 400–800 nm.

Results and discussion

Characterization of AuNPs pre- and post-treatment

The optical behavior of ligand-functionalized AuNPs toward Cd²⁺ was systematically evaluated through UV-vis spectroscopy by monitoring changes in the surface plasmon resonance (SPR) band. Because the SPR of AuNPs is highly sensitive to variations in local dielectric environment, surface charge distribution, and interparticle spacing, spectral shifts provide insights into the interaction mechanism between Cd²⁺ and the surface-bound ligands.^18,19 The results demonstrate that the sensing performance strongly depends on ligand structure and its ability to mediate interparticle interactions.

This study investigated gold nanoparticles (AuNPs) conjugated with TDDT, AMTTD, MNA–Ala, and NAC for the colorimetric detection of Cd²⁺ ions. The UV-vis spectra revealed that pristine AuNPs exhibited a characteristic surface plasmon resonance (SPR) peak at 523 nm with an absorbance of 0.172 and a wine-red color, consistent with previous reports.²⁰

In the initial tests, the AuNPs–TDDT conjugate showed a slight shift in the SPR peak from 523 nm to 525 nm upon addition of 10 µM TDDT, accompanied by a minor color change. However, the addition of 1 mM Cd²⁺ led to a decrease in absorbance without significant peak shifting, indicating a weak interaction between Cd²⁺ and the AuNPs–TDDT conjugate (Fig. 1a).

Fig. 1. UV-vis spectrophotometric results of (a) AuNPs–TDDT, (b) AuNPs–AMTTD, (c) AuNPs–MNA–Ala, (d) AuNPs–NAC before and after the addition of Cd²⁺.

The test was also conducted on the AuNPs–AMTTD conjugate. Upon conjugation with 10 µM AMTTD, the absorbance of the AuNPs increased, and the SPR peak shifted slightly to 525 nm, suggesting successful ligand attachment and minor modifications to the local dielectric environment. When 1 mM Cd²⁺ was added, a small red shift to 537 nm was observed, but the color change was minimal, indicating that the interaction between Cd²⁺ and the AuNPs–AMTTD system was weak, with limited aggregation occurring under these conditions. The results are shown in Fig. 1b.

In the AuNPs–MNA–Ala system, the SPR peak shifted to 516 nm with a decrease in absorbance, suggesting an interaction between MNA, alanine, and the AuNP surface (Fig. 1c). However, the addition of 1 mM Cd²⁺ led to a significant red shift in the SPR peak to 640 nm. This large shift indicates aggregation, likely due to coordination bridging between Cd²⁺ ions and the functional groups of MNA and alanine, which causes interparticle plasmon coupling and a visible color change. These results highlight the potential of this system for colorimetric Cd²⁺ detection.

For the NAC-conjugated AuNPs, the SPR peak was observed at 517 nm in the unmodified AuNPs (Fig. 1d). After conjugation with 50 µM NAC, the SPR peak shifted to 521 nm, indicating the interaction between NAC and the AuNP surface. Upon the addition of 10 mM Cd²⁺, the SPR peak shifted slightly to 519 nm, and the absorbance decreased, suggesting partial destabilization or modification of surface charge, likely due to the formation of Cd–NAC complexes prior to their interaction with the AuNP surface.²¹ This effect contrasts with the aggregation observed in the AuNPs–MNA–Ala system and suggests a different mode of interaction between Cd²⁺ and NAC.

Gold nanoparticles exhibit high sensitivity because their size and composition can be precisely tuned for colorimetric detection. The size and shape of gold nanoparticles significantly influence their optical properties; specifically, smaller nanoparticles generate surface plasmon resonance peaks at shorter wavelengths. This tunability enhances analytical accuracy and improves the capability to identify and discriminate among various target analytes, making gold nanoparticles highly suitable for sensitive and selective sensing applications.²²

To clarify the donor groups available for Cd²⁺ coordination and for AuNP surface anchoring, the structures of the four ligands (TDDT, AMTTD, MNA, and NAC) together with l-alanine (the co-ligand used with MNA) are shown in Fig. 2. TDDT presents two thiol groups on a rigid, planar 1,3,4-thiadiazole ring; AMTTD presents one primary amine and a ring-bound methylthio group that requires in situ demethylation to expose a reactive thiol; NAC presents a single thiol together with a carboxylate and an amide carbonyl on a flexible aliphatic backbone; and MNA presents a single thiol and a terminal carboxylate on a short, flexible propanoate chain, co-administered with l-alanine, whose free amino and carboxylate groups can further participate in Cd²⁺ coordination once alanine is co-adsorbed at the AuNP surface. This comparison is used to rationalize why the flexible, multidentate MNA/Ala pair outperformed the two rigid heterocyclic thiadiazoles and the sterically encumbered NAC.

Fig. 2. Molecular structures of the four thiol-based ligands (TDDT, AMTTD, MNA, and NAC) and the MNA co-ligand, l-alanine.

The marked difference in Cd²⁺ response among the four ligands can be rationalized from their structures (Fig. 2) and is consistent with the UV-vis results above. (i) Denticity and flexibility: MNA/Ala offers two independently rotatable donor sets, a thiol for Au anchoring plus a carboxylate, and a second, co-adsorbed amino acid bearing both amine and carboxylate groups. In short, conformationally flexible aliphatic chains allow the Cd²⁺ ion to be chelated and to bridge across nanoparticles. TDDT and AMTTD, by contrast, present their donor atoms fixed within a rigid, planar thiadiazole ring, which constrains the geometry available for Cd²⁺ bridging and likely explains the negligible aggregation and unshifted 640 nm shoulder observed for both systems. (ii) Donor availability: in AMTTD, the second sulphur is masked as a methylthio (–SCH₃) group that requires demethylation to expose a reactive thiol. (iii) Steric accessibility: NAC contains a thiol and a carboxylate, similar to MNA, but its N-acetyl amide group is directly adjacent to the thiol-bearing carbon, which is expected to sterically crowd the Au–S anchor point and reduce the exposure of the carboxylate for Cd²⁺ coordination, consistent with the smaller SPR shift and only slight aggregation observed for AuNPs–NAC/Cd²⁺ relative to AuNPs–MNA–Ala/Cd²⁺. Taken together, these structural features, flexibility, donor-atom availability, and steric accessibility, provide a consistent chemical rationale for why the MNA/Ala pair outperformed the other three thiol ligands under the conditions tested.

Proposed Cd²⁺-sensing mechanism

Fig. 3 summarizes the proposed mechanism for the best-performing MNA–Ala–AuNPs system. At pH 10, the terminal carboxylate of MNA and the amino/carboxylate groups of co-adsorbed alanine are deprotonated and project away from the gold surface, while the thiol of MNA anchors the ligand shell to the AuNPs through an Au–S bond. Cd²⁺, a borderline-to-soft Lewis acid, is bridged simultaneously by the exposed carboxylate/amino donors on adjacent nanoparticles (Cd²⁺–O and Cd²⁺–N coordination), analogous to the Cd²⁺–thiolate/carboxylate cross-linking mechanisms reported for other carboxyl- and thiol-bearing AuNP probes.^15,25,26 This interparticle bridging pulls neighbouring AuNPs closer than the plasmon-coupling threshold (typically <2.5 particle diameters), red-shifting and broadening the SPR band from ∼520 nm to ∼640 nm and producing the visible red-to-blue/purple transition used for the colorimetric readout. The same bridging chemistry was far less effective for TDDT and AMTTD, whose thiol/amine donors are held rigidly in the plane of the thiadiazole ring and are less able to reach across the interparticle gap (Fig. 2), and for NAC, whose bulky N-acetyl group is proposed to sterically hinder close approach of adjacent particles.

Fig. 3. Proposed mechanism of Cd²⁺-induced aggregation of MNA–Ala–AuNPs: (red) dispersed, ligand-capped AuNPs; (purple) Cd²⁺ bridges surface carboxylate/amino donors, cross-linking adjacent particles; and (blue) interparticle plasmon coupling produces the red-to-blue/purple colour change (λ_max 520 → 640 nm) used for colorimetric readout.

Method optimization

Optimization was performed during the preparation of ligand-functionalized AuNPs to identify the parameters that yielded AuNPs with optimal sensitivity for the detection of Cd²⁺. For the AuNPs–TDDT system, optimization revealed that 10 µM TDDT led to minimal AuNP aggregation (A₆₄₀/A₅₂₀ ratio = 0.073), with a slight shift in the SPR peak from 523 to 525 nm. The aggregation began to increase at 50 µM TDDT (A₆₄₀/A₅₂₀ = 0.444), reaching a maximum at 1 mM TDDT (A₆₄₀/A₅₂₀ = 0.468), though no secondary peak at 640 nm was observed, indicating weak Cd²⁺ interaction. At higher concentrations (10 and 100 mM), excessive stabilization by TDDT reduced aggregation. The optimal pH was 7, where the system showed the best color response, although no secondary peak at 640 nm appeared. A 25-minutes reaction time produced stable AuNPs, while longer times caused aggregation. NaCl concentration of 40 mM was optimal, producing the most distinct purple hue. The highest aggregation (A₆₄₀/A₅₂₀ = 0.468) was achieved after 40 minutes of incubation with 1 mM Cd²⁺.

In the AuNPs–AMTTD system, 500 µM AMTTD produced the strongest colorimetric response with a red shift to 537 nm and an A₆₄₀/A₅₂₀ ratio of 0.356. Lower concentrations (<100 µM) showed weak responses, while higher concentrations (>1 mM) inhibited aggregation due to electrostatic repulsion. The optimal reaction time was 45 minutes, ensuring uniform surface modification. The maximum A₆₄₀/A₅₂₀ ratio was achieved at pH 7, indicating that neutral conditions favored Cd²⁺ coordination. The amine group in AMTTD is pH-sensitive, undergoing protonation at low pH and deprotonation at neutral to basic conditions, which affects its binding affinity toward Cd²⁺ and its aggregation efficiency.^23–25 The NaCl concentration also influenced aggregation, with 40 mM NaCl providing the best balance of stability and responsiveness.²⁶ The highest sensitivity was observed after 25 minutes of incubation with Cd²⁺, with an A₆₄₀/A₅₂₀ ratio of 0.203, indicating optimal interaction, which suggests over aggregation or restabilization of the colloid.^27,28

The optimal concentrations for MNA and l-Ala were found to be 1 M and 10 mM, respectively, as these conditions provided effective AuNP surface modification. The SPR peak shifted from 520 nm to 523 nm upon conjugation, with increased absorbance at 640 nm, indicating aggregation. The system showed a color change upon the addition of Cd²⁺ at pH 6 to 10. The highest sensitivity was observed at pH 10, where Cd²⁺ aggregation was most pronounced due to the reduction in free Cd²⁺ concentration from hydroxide formation. NaCl concentrations of 100 mM and a 60-minutes incubation time produced the highest A₆₄₀/A₅₂₀ ratio (2.038), indicating optimal aggregation and Cd²⁺ detection.

NAC-modified AuNPs exhibited maximum aggregation at 1 mM NAC, indicated by a red shift to 523 nm in the SPR spectrum. The optimal reaction time was 50 minutes, with an A₆₄₀/A₅₂₀ ratio of 0.876, reflecting aggregation. The optimal pH was 6, where the NAC–AuNP system showed the highest sensitivity toward Cd²⁺, as minimal aggregation occurred under extreme pH conditions. The addition of 10 mM NaCl produced the highest A₆₄₀/A₅₂₀ ratio (1.185), suggesting aggregation and a colorimetric response.

Characterization of conjugated AuNPs

FTIR spectroscopy: Fourier-transform infrared (FTIR) spectroscopy was performed to analyze the functional groups on the surface of AuNPs before and after conjugation with different ligands and subsequent exposure to Cd²⁺ ions. This analysis helps in understanding the interaction between the functional groups and the metal ions, providing insights into the mechanism of AuNP aggregation during Cd²⁺ detection.²⁹

In the AuNPs–TDDT system, the –SH absorption band was absent, suggesting that the thiol groups of TDDT likely reacted with the AuNP surface. However, since the pure TDDT spectrum also lacked this band, this suggests that the binding was not fully established. The absence of significant shifts in the C N (1500–1600 cm⁻¹), N N (1400–1500 cm⁻¹), and C–S (600–700 cm⁻¹) bands across all samples indicates no strong interaction between AuNPs–TDDT and Cd²⁺ ions. These findings suggest that TDDT does not form a stable or strong complex with cadmium ions, implying that Cd²⁺ detection occurs primarily through electrostatic or colloidal interactions, rather than covalent Au–S bonding. The minimal color change observed in the AuNPs–TDDT–Cd²⁺ system further supports incomplete conjugation, leaving some free TDDT in solution.

For AMTTD, absorption bands at 3307 and 3275 cm⁻¹ correspond to N–H stretching vibrations. These bands did not show significant shifts after conjugation with AuNPs or after the addition of Cd²⁺, indicating that coordination between the amino group (–NH₂) and Cd²⁺ was not optimally formed. Slight shifts observed around 1635 cm⁻¹ (C N or N–H bending) suggest weak interactions, but not strong enough to confirm stable complexation. Additionally, characteristic CH₃ stretching peaks appeared at 1466 and 1373 cm⁻¹, and the expected C–S stretching band (660–630 cm⁻¹) was not clearly detected, likely due to overlapping with nearby signals. This suggests that the –SCH₃ group does not form strong Au–S bonds, as –SCH₃ exhibits weaker coordination with gold than thiol (–SH) groups due to steric hindrance and lack of reactive hydrogen.^30,31 Overall, the spectra of AuNPs–AMTTD and AuNPs–AMTTD/Cd²⁺ were nearly identical to those of AMTTD alone, indicating minimal chemical interaction or stable coordination between AuNPs, AMTTD, and Cd²⁺.

The FTIR spectra of NAC, AuNPs–NAC, and AuNPs–NAC–Cd²⁺ revealed only minor intensity changes and negligible peak shifts (e.g., around 3309 cm⁻¹ and 1635 cm⁻¹), suggesting that chemical modifications of NAC upon exposure to AuNPs and Cd²⁺ were minimal. No significant disappearance or pronounced shift of thiol-related bands was observed, and the weak features around 430–450 cm⁻¹ were ambiguous, possibly arising from overlapping signals or instrument noise rather than a clear M–S stretching.

The FTIR analysis of the MNA–Ala–AuNPs–Cd²⁺ system demonstrated key interactions during conjugation and Cd²⁺ binding. The pure MNA spectrum showed a strong band near 1700 cm⁻¹, corresponding to C O stretching of the carboxyl group. Upon conjugation with AuNPs, this band decreased in intensity and shifted slightly, indicating deprotonation and attachment of MNA to the nanoparticle surface. The band at 3300–3400 cm⁻¹, corresponding to N–H and O–H stretching from l-alanine, remained visible, suggesting that these groups remained accessible on the surface after conjugation. Significant changes were observed in the C O stretching bands at 1580–1650 cm⁻¹ and 1380–1420 cm⁻¹ after the addition of Cd²⁺, confirming the coordination between Cd²⁺ and the carboxylate groups. A new or intensified band around 500–600 cm⁻¹ indicated Au–S bond formation, confirming that MNA binds to AuNPs via thiol groups. Overall, these FTIR results demonstrate that MNA binds to AuNPs through Au–S bonds, alanine integrates through amino and carboxyl groups, and Cd²⁺ interacts with carboxylates, promoting nanoparticle aggregation, which is essential for the colorimetric sensing mechanism (Fig. 4).

Fig. 4. Overlay FTIR spectra showing the functional groups of MNA, AuNPs–MNA, and AuNPs–MNA/Cd²⁺.

HR-TEM: High-resolution TEM (HR-TEM) was performed at 200 kV to observe AuNPs before and after conjugation. In Fig. 5a, the AuNPs appear nearly spherical with uniform size. After the addition of TDDT and Cd²⁺, the particles still show similar spherical shapes and size distribution (Fig. 5b). Since no noticeable morphological or dispersion changes are observed, the results suggest that the AuNPs–TDDT/Cd²⁺ system did not form or conjugate effectively. The HR-TEM images show that AuNPs maintained their spherical morphology after AMTTD and Cd²⁺ conjugation, with no significant changes in particle shape or size. This indicates the absence of strong chemical interactions or new bond formation, suggesting only weak surface interactions between AuNPs, AMTTD, and Cd²⁺ (Fig. 5c). After conjugation with NAC and Cd²⁺ (Fig. 5d), slight aggregation was observed, suggesting partial interaction or surface modification of AuNPs. However, the overall morphology and crystalline structure remained largely unchanged, indicating that the interaction between NAC/Cd²⁺ and AuNPs was relatively weak and did not significantly alter the nanoparticle core. After conjugation with MNA–Ala–Cd²⁺ (Fig. 5e), aggregation is observed, indicating surface interaction or modification due to ligand and Cd²⁺ binding. HR-TEM results reveal that the crystalline structure of AuNPs remains well-preserved after conjugation, with clearly visible lattice fringes. The aggregation observed suggests coordination between the carboxylate groups of MNA–Ala and Cd²⁺ ions, which forms the basis of the color change in the colorimetric sensing system.

Fig. 5. HR-TEM images of (a) AuNPs, (b) AuNPs–TDDT/Cd²⁺, (c) AuNPs–AMTTD/Cd²⁺, (d) AuNPs–NAC/Cd²⁺, and (e) AuNPs–MNA–Ala/Cd²⁺.

Sensitivity and selectivity tests of conjugated AuNPs

The colorimetric readout used throughout this study (the visible red-to-blue/purple transition and the A₆₄₀/A₅₂₀ ratio) is a surface plasmon resonance property of aggregated AuNPs; none of the four free ligands in solution possesses an SPR band, so a free ligand alone cannot produce a directly comparable colorimetric signal upon Cd²⁺ binding. The AuNP scaffold converts Cd²⁺–ligand binding into an optical readout in the first place.

MNA–Ala–AuNPs showed the most optimal results based on the UV-vis spectrum, which showed a wavelength shift to 640 nm, as well as changes in the FTIR spectrum, indicating the formation of MNA–Ala–AuNPs/Cd²⁺ interactions and the presence of aggregation observed in HR-TEM analysis. Based on these findings, this system was selected for further selectivity and sensitivity testing.

MNA–Ala–AuNPs exhibited a measurable response toward Cd²⁺ relative to the other five ions tested. CdCl₂ produced a mean A₆₄₀/A₅₂₀ ratio of 1.89 (n = 3). Accordingly, the MNA–Ala-modified system demonstrates selectivity for Cd²⁺ over Ca²⁺, Pb²⁺, Hg²⁺, Fe²⁺, and Zn²⁺. The AuNPs selectivity result is shown in Fig. 6.

Fig. 6. Selectivity of the method using MNA–Ala–AuNPs for Cd²⁺ detection compared with different metal ions (Ca²⁺, Pb²⁺, Hg²⁺, Fe²⁺, and Zn²⁺) at 100 µM.

The selection of metals for the selectivity test was based on several factors. Ca²⁺ and Fe²⁺ were included because they are the most abundant divalent cations in Indonesian ground and household water and therefore represent the most probable everyday interferents for a field-deployable sensor; Zn²⁺ was included as a chemically similar first-row transition/post-transition divalent ion frequently co-present with Cd²⁺ in plumbing and industrial run-off; and Pb²⁺ and Hg²⁺ were included because, like Cd²⁺, they are borderline-to-soft Lewis acids with high thiophilicity. Mechanistically, Cd²⁺ is classified as a borderline Lewis acid under the HSAB framework, giving it appreciable affinity for both the soft thiolate sulfur and the hard-to-borderline carboxylate/amine oxygen and nitrogen donors presented by MNA and alanine; this dual affinity is what allows Cd²⁺ to bridge two coordination modes simultaneously (Fig. 3) and is the basis for its aggregation-inducing effect. Ca²⁺ and Fe²⁺, being harder Lewis acids, bind the carboxylate/amine donors less effectively and were expected to aggregate the probe less; Pb²⁺ and Hg²⁺, being comparably soft to Cd²⁺, were expected and, per the modest differences reported above, were found to compete more closely for the same thiolate/carboxylate donor set.

Quantitatively, MNA–Ala–AuNPs displayed a log-linear response described by y = 0.0338x + 1.6722 (R² = 0.9933) (Fig. 7). The sensing mechanism is consistent with HSAB-driven Cd–thiolate coordination, which promotes interparticle cross-linking and localized surface plasmon resonance (LSPR) red-shifting.

Fig. 7. MNA–Ala–AuNPs Sensitivity at various Cd²⁺ concentrations (0.001, 0.01, 0.1, 1, and 10 mM).

Benchmarked against five previously reported thiol/chelator-functionalised AuNP colorimetric sensors for Cd²⁺, the MNA–Ala–AuNPs probe developed here exhibits an Limit of Detection (LOD) of 1.25 µM, which is higher (less sensitive) than all comparable systems, including the tetrazolylaniline–AuNPs (0.27 µM), Na₂EDTA–AuNPs (0.28 µM), dimercaptopurine–AuNPs (32.7 nM), TMT–AuNPs (3.5 nM), and glutathione–AuNPs (4.6 nM) probes; this reduced sensitivity is consistent with the more labile carboxylate/amine bridging mode of MNA/Ala, which lacks the multiple pre-organised soft-donor (thiol/N-heterocycle) sites that confer stronger, near-irreversible Cd²⁺ chelation in other ligand systems. Nevertheless, this trade-off is offset by the markedly broader, log-linear working range achieved (1 µM–100 mM, R² = 0.9933), spanning five orders of magnitude and substantially exceeding the narrow or unstated ranges reported elsewhere (e.g., TMT–AuNPs: 10–300 nM; dimercaptopurine–AuNPs: 0.75–3.0 µM), which makes the MNA–Ala probe advantageous for applications where Cd²⁺ concentrations may vary widely across samples without requiring dilution or repeated recalibration. Successful validation in household water (spiked and unspiked) further situates this work alongside other field-relevant matrices (tap water, lake water, and paper-based formats) rather than the food/biological-matrix studies, and its simple mix-and-read colorimetric readout, requiring no acid digestion or complex ligand synthesis, supports its practical value as a low-cost, field-deployable screening tool suited to semi-quantitative detection over a wide concentration range, complementing rather than replacing the more sensitive but narrower-range alternatives summarised in Table 1.

Table 1. Benchmarking of MNA–Ala–AuNPs against reported AuNP-based colorimetric Cd²⁺ sensors.

AuNPs conjugate	Sensing basis	Linear range	LOD	Real-sample validation	Ref.
MNA–Ala–AuNPs	Cd2+-bridged carboxylate/amino cross-linking (Fig. 3)	1 µM–100 mM (log-linear, R2 = 0.9933)	1.25 µM	Household water (spiked/unspiked)	This work
Glutathione–AuNPs	Cd2+-GSH complex destabilises salt-induced aggregation	Not stated (single-point LOD study)	4.6 nM	Rice (digested)	11
3-(1H-tetrazol-5-yl)aniline–AuNPs	Cd2+ chelation → aggregation	Not stated	0.27 µM	Rice, soil, tap water	13
Trithiocyanuric acid (TMT)–AuNPs	Cd2+–TMT coordination → aggregation	10 nM–300 nM	3.5 nM	Lake water	15
2,6-Dimercaptopurine–AuNPs	Cd2+ chelation → aggregation	0.75–3.0 µM	32.7 nM	Food, biological, and environmental samples (88–112% recovery)	32
Na2EDTA–AuNPs (cotton-pad format)	Cd2+-EDTA complexation destabilises AuNPs	Not stated	0.28 µM	Paper-based field format	33

Application of conjugated AuNPs for Cd²⁺ detection in the water sample

Cd²⁺ was successfully identified in a household water sample using the colorimetric MNA–Ala–AuNPs method, which demonstrated a greater absorbance ratio after Cd²⁺ addition compared to the sample without Cd²⁺. This shows how well the technique works for detecting cadmium in water samples.

The capacity of MNA–Ala–AuNPs to detect Cd²⁺ in the water sample from Kelapa Dua, Depok, was tested to illustrate its practical use. A 0.45 mm membrane was used to filter the water sample, and UV-vis spectrophotometry was used to assess the method's performance in matrices after known amounts of Cd²⁺ were added to the samples.

The findings demonstrated that the absorbance ratios (A_640/520) of the sample spiked with 10 mM Cd²⁺ were higher than those of the samples without additional Cd²⁺. The values of A_640/520 in the unspiked sample and spiked sample were 0.836 and 1.699, respectively, indicating that the system can distinguish the presence of Cd²⁺ within the matrix and that no Cd²⁺ is present in the household water sample. With a spectral response, the nanoparticles successfully identified Cd²⁺, demonstrating their potential as an environmental monitoring tool.

Conclusions

Across four thiol-based ligands, namely, TDDT, AMTTD, MNA–Ala, and NAC, MNA–Ala is the top performer under optimized conditions (10 mM MNA, 1 mM Ala, pH 10, and 100 mM NaCl, with 60-min pre-incubation) with a linear calibration (R² = 0.9933). This system is able to differentiate the identified Cd²⁺ within the matrix. From the testing results on household water samples from Kelapa Dua, Depok, Cd²⁺ was not detected.

Author contributions

Conceptualization: F. A. S.; methodology: F. A. S., S. F., and D. A. P.; software: S. F. and D. A. P.; validation: S. F., D. A. P., and F. A. S.; formal analysis: S. F. and D. A. P.; investigation: S. F. and D. A. P.; resources: F. A. S., Z. K., V. D., and R. K. C.; data curation: F. A. S., S. F. and D. A. P.; writing – original draft preparation: F. A. S., S. F., and D. A. P.; writing – review and editing: F. A. S., S. F., D. A. P., R. I., C. A. M., Z. K., V. A. D., and R. K. C.; visualization: S. F. and D. A. P.; supervision: F. A. S., R. I., and C. A. M.; project administration: F. A. S. and R. I.; and funding acquisition: F. A. S.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.