1 Introduction

Mercury pollution in aquatic systems remains one of the most severe and persistent environmental issues of modern times, threatening both ecological integrity and human health worldwide [1]. Owing to its toxicity, persistence, and mobility, mercury poses significant risks given its pronounced potential for bioaccumulation and biomagnification across aquatic food webs [2]. The Environmental Protection Agency (EPA) and the World Health Organization (WHO) have set extremely stringent limits for inorganic mercury in drinking water 2 µg/L and 1 µg/L, respectively, reflecting the critical need for effective mercury control strategies [3]. Major anthropogenic sources such as coal combustion, mining, waste incineration, and industrial discharges continue to introduce considerable mercury loads into natural water bodies [4]. Mercury exists in three principal chemical forms, elemental (Hg⁰), inorganic (Hg²⁺), and organic (primarily methylmercury), each exhibiting distinct environmental behaviors and toxicities [5, 6]. Elemental mercury can volatilize and be transported atmospherically, while inorganic mercury compounds often adsorb onto sediments or aquatic biomass. The most hazardous species, methylmercury, is generated through microbial methylation of inorganic mercury, readily entering the aquatic food web and causing neurological, renal, and developmental damage upon human exposure [7,8,9]. Mercury’s strong affinity for sulfur and nitrogen containing groups within biomolecules enables it to disrupt cellular processes, enzyme activity, and membrane integrity [10, 11]. Thus, the persistence, reactivity, and biomagnification potential of mercury highlight the urgent need for sustainable and efficient removal technologies.

Widely used physicochemical remediation methods, including chemical precipitation, ion exchange processes, membrane filtration systems, and activated carbon method are effective but limited by high operational costs, energy demands, and secondary waste generation [12]. As a result, alternative eco-friendly technologies have been explored, among which biological and biomaterial-based approaches have shown promising potential [13]. In particular, biosorption using microalgal biomass either living or non-living has emerged as an efficient and sustainable strategy for heavy metal removal [14, 15]. While living microalgae can remove metals through both metabolism-dependent and independent pathways, non-living biomass relies solely on passive surface adsorption processes, governed by functional groups such as carboxyl, hydroxyl, amine, and phosphate groups present on the cell wall matrix [16]. Furthermore, non-living microalgal biomass offers several advantages over living cells for practical applications. The absence of metabolic activity eliminates the need for nutrient supply, aeration, and light, reducing operational complexity and cost [17]. Nevertheless, non-living biomass is not susceptible to metal toxicity, making it suitable for higher concentration ranges and fluctuating environmental conditions [18]. The preparation of non-living biomass through heat or chemical inactivation also enhances its stability and facilitates storage and reuse. The surface chemistry and porosity of such thermally treated biomass often provide additional active sites, improving biosorption capacity [19].

Previous studies have demonstrated that non-living microalgae can effectively remove various heavy metals such as lead, cadmium, chromium, and mercury from aqueous systems through mechanisms involving ion exchange, electrostatic attraction, and complexation [20]. However, most research has focused on single-strain microalgae cultivated under controlled conditions, whereas limited attention has been paid to mixed indigenous microalgal communities (MIMA) that naturally exist in local aquatic environments. These communities are inherently adapted to regional conditions, often exhibiting diverse surface characteristics and functional groups that could enhance their sorption capacity and resilience [21]. Despite the growing interest in microalgal biosorbents, there remains a considerable research gap concerning the effectiveness of non-living MIMA in mercury removal at environmentally relevant concentrations. Understanding their adsorption behavior, kinetics, and surface interactions is crucial for developing practical, low-cost treatment technologies.

While numerous biosorption studies have focused on pure cultures and mg/L-level mercury concentrations, very limited work has examined non-living MIMA at realistic trace levels (10–100 µg/L). Furthermore, no prior studies have provided a combined mechanistic assessment integrating SEM, FTIR, XPS, kinetic modeling, and equilibrium analyses for such low mercury concentrations. By addressing these gaps, the present work offers novel insights into the applicability and mechanistic behavior of indigenous biomass for trace-level mercury remediation. Accordingly, this study investigates the mercury biosorption capabilities of MIMA collected from local Qatari coastal waters. The objectives are to: (1) Evaluate the mercury removal efficiency of MIMA across concentrations representative of environmental conditions (10–100 µg L⁻¹); (2) Examine the influence of biosorbent dosage and contact time on mercury removal; (3) Analyze adsorption kinetics and equilibrium behavior using various isotherm and kinetic models; (4) Characterize the structural and chemical modifications of the biomass before and after mercury exposure using SEM, EDX, FTIR, and XPS analyses. By addressing these objectives, this work provides new insights into the biosorption potential of indigenous microalgal biomass for mercury remediation. The findings contribute to understanding the surface interactions governing Hg(II) uptake and lay the groundwork for developing efficient, low-cost, and sustainable biosorbent materials for mercury-contaminated water treatment.

2 Materials & method

2.1 Materials

2.1.1 Chemicals and reagents

A certified mercury standard solution (10 µg mL⁻¹ in 3% v/v HNO₃, prepared from Hg(NO₃)₂; Inorganic Ventures, Christiansburg, VA, USA) was employed as the source to formulate aqueous test solutions with final Hg concentrations between 10 and 100 µg L⁻¹. Solution pH was adjusted as required by the addition of 1 M NaOH or 1 M HCl of analytical grade. Ultrapure deionized water (18.2 MΩ·cm at 25 °C) was used in all solution preparations. The BG-11 culture medium was formulated using analytical-grade components with ≥ 99% purity, all purchased from Sigma-Aldrich (St. Louis, MO, USA), namely NaNO₃, K₂HPO₄, MgSO₄·7 H₂O, CaCl₂·2 H₂O, C₆H₈O₇, C₆H₈O₇·xFe·yNH₃, Na₂EDTA·2 H₂O, Na₂CO₃, H₃BO₃, MnCl₂·4 H₂O, ZnSO₄·7 H₂O, CuSO₄·5 H₂O, Na₂MoO₄·2 H₂O, and Co(NO₃)₂·6 H₂O. To eliminate potential metal contamination, all glassware and experimental containers were immersed in 10% HNO₃ for 24 h after which deionized water was used to thoroughly rinse the materials prior to experimentation. Sterilization of the BG-11 culture was achieved by autoclaving at 121 °C for a duration of 20 min prior to experimental application.

2.1.2 Equipment

The surface characteristics and elemental distribution of the samples were characterized by scanning electron microscopy using a Nova NanoSEM 450 system (Field Electron and Ion, USA) equipped with an energy-dispersive X-ray detector. Chemical surface states were further analyzed by X-ray photoelectron spectroscopy employing an Escalab 250 Xi instrument (Thermo Fisher Scientific, USA). Total mercury content in aqueous phases and algal biomass was evaluated using DMA-80 evo (Milestone, Italy). Functional group analysis was performed by Fourier transform infrared spectroscopy with a Spectrum 400 spectrometer (PerkinElmer, USA). In addition, optical microscopy was performed using a BX53 microscope (Olympus, Japan).

2.2 Preparation and cultivation

2.2.1 Collection, isolation, and cultivation

MIMA were sourced from Al-Wakrah Beach (Qatar; 25.1816° N, 51.6164° E) during low tide by collecting surface seawater (0–30 cm) in sterile glass vessels. The collected seawater was promptly subjected to filtration using a 20 μm phytoplankton net to eliminate coarse particles and macro-organisms. The clarified suspension was then placed in sterile vessels supplemented with BG-11 culture and delivered to the lab within 2 h while maintaining a temperature of 25 ± 2 °C. Subsequent seperation and refinement of MIMA consortium were achieved through the preparation of a ten-fold dilution series, with subsequent inoculation of BG-11 agar. The isolated MIMA consortium predominantly consisted mainly of Chlorella sp., and Arthrospira sp. (commercially known as Spirulina). The composition of MIMA consortium was determined using morphological and microscopic examination following standard keys. The culture was consistently dominated by Chlorella sp. (60–70%) and Arthrospira sp. (30–40%), based on repeated microscopic counting across multiple sub-samples. Molecular sequencing was not performed because the objective of this work was to evaluate functional biosorption behavior rather than species-level phylogenetic resolution. To further support biomass consistency, protein and carbohydrate content analysis is identified as a valuable characterization step for future work, as these biomolecules are traditionally implicated in heavy metal biosorption by algal biomass. Similar microalgae-based biosorption studies have relied on morphological identification for mixed indigenous cultures where functional performance is the primary outcome [22, 23]. Nevertheless, the dominant genera were clearly distinguishable due to their characteristic cell shapes, sizes, and pigmentation.

The cultivation conditions were carefully regulated to maintain a constant temperature of 25 °C using a thermostatically controlled heater positioned beneath the photobioreactor. Uniform illumination of the photobioreactor was achieved using LED panels positioned around the glass surface, resulting in a photon flux density of 60 µmol photons m⁻² s⁻¹. Prior to inoculation, the culture medium pH was set to 7.5 by the addition of 1 M NaOH or 1 M HCl, and pH values were continuously tracked during the cultivation period. Aeration was supplied via airstones at a controlled flow rate to ensure gentle yet efficient mixing, promoting uniform cell and nutrient distribution with reduced shear effects. When required, CO₂ was injected through the aeration, serving as an additional carbon source for photosynthesis and contributing to pH stabilization via buffering effects associated with carbonate ions. Cells development was routinely inspected by assessing the observation of color intensity, microscopic evaluation of cell morphology and density, and monitoring of physicochemical conditions, including pH, temperature, and dissolved oxygen, which was performed throughout the experiments. The use of a transparent glass photobioreactor allowed real-time observation of the culture, which supported the early identification of operational issues such as contamination or cell clumping.

2.2.2 Preparation of non-living biomass

The preparation of non-living microalgal biomass followed a systematic multi-step process to ensure consistent quality and characteristics of the biosorbent material. Microalgal cells were initially harvested during the late exponential growth phase, which was determined through regular monitoring of culture density and typically occurred after 14–16 days of cultivation. The harvesting process began with the collection of microalgal suspension from the photobioreactor, followed by centrifugation at 7500 rpm for 7.5 min using 50 mL centrifuge tubes to separate the biomass from the culture medium. The concentrated biomass underwent a carefully controlled deactivation process designed to ensure complete cessation of biological activity while preserving the cellular structure important for biosorption. This process consisted of two critical steps: First, the harvested biomass was subjected to thermal treatment in an autoclave at 121 °C under 15 psi pressure for 10 min. This step ensured complete deactivation of all cellular metabolic processes and sterilization of the biomass. Following autoclaving, the biomass was immediately transferred to sterile glass petri dishes, ensuring an even distribution of material to facilitate uniform drying. The second step involved thermal drying in a laboratory oven at 105 °C for 24 h. These temperatures and durations were specifically selected to achieve complete moisture removal while minimizing potential thermal degradation of cellular components that could affect metal binding sites. The drying process was continued until a constant weight was achieved, indicating complete moisture removal. The dried biomass formed a solid cake that was then carefully collected from the petri dishes. To prepare the final biosorbent material, the dried biomass was mechanically processed to achieve uniform particle characteristics. The dried biomass cake was first gently ground using a clean mortar and pestle to break down larger aggregates. The ground biomass was sieved using a 100-mesh stainless steel sieve ≈ 150 μm to obtain a uniform particle size fraction. All biosorption experiments were performed using biomass with a particle size < 150 μm, which ensured consistent surface area exposure and minimized mass transfer limitations. The processed biomass appeared as a fine, homogeneous powder with a characteristic green-brown color.

The prepared biomass was immediately transferred to an airtight, clean glass container to prevent moisture absorption from ambient air. The container was stored at room temperature (25 ± 2 °C) in a dark, dry location to prevent any potential degradation from light or humidity exposure. This storage method ensured the stability and consistency of the biosorbent material throughout the experimental period. Prior to use in mercury removal experiments, the stored biomass was always checked for any signs of moisture absorption or aggregation to maintain experimental consistency.

2.2.3 Hg(II) solutions preparation

Initial Hg levels of 10, 40, 70, and 100 µg/L were employed to encompass concentrations relevant to environmental scenarios that commonly observed in polluted marine environments, especially in areas affected by industrial effluents or artisanal mining activities. In such regions, mercury concentrations frequently surpass the WHO and EPA regulation limits of 1–2 µg/L and may often fall within the range of tens to hundreds of micrograms per liter [24]. Thus, the chosen range effectively represents realistic contamination conditions for experimental exposure. Mercury test solutions were formed using a controlled dilution approach designed to ascertain precision and reliability. A commercially certified Hg standard served as the primary source solution. Target concentrations of 10, 40, 70, and 100 µg L⁻¹ were obtained by diluting appropriate aliquots of the stock solution with ultrapure deionized water. The required stock volumes for each concentration were determined based on standard dilution calculations using the dilution equation C₁V₁ = C₂V₂, where C₁ is the concentration of the stock mercury solution (µg/mL), V₁ is the required volume of stock solution (mL), C₂ is the desired final mercury concentration (µg/L), and V₂ is the final solution volume (mL). The solutions were prepared fresh before each experiment to minimize potential mercury losses caused by volatilization or surface adsorption [26]. Prior to initiating the experiments, the pH of the microalgal suspensions was set to 7.5 by the stepwise addition of 1 M NaOH or 1 M HCl, with measurements taken using a calibrated pH meter. All experimental conditions were conducted in triplicate, and reported results represent the corresponding mean values.

Quality control measures were implemented throughout the solution preparation process. All glassware used in solution preparation was acid-washed with 10% HNO₃ followed by extensive washing with deionized water before utilization. The prepared solutions were stored in amber glass bottles when not in immediate use to minimize potential photochemical reactions. Furthermore, the actual concentration of each prepared solution was verified before conducting the experiments to ensure the accuracy of the initial mercury concentrations. This comprehensive preparation protocol ensured consistency and reliability in the initial mercury concentrations across all experiments.

2.2.4 Non-living biomass experiments

Four test systems were established and labeled D-10, D-40, D-70, and D-100. In this notation, the prefix “D” refers to non-living microalgal biomass, while the numerical suffix indicates the corresponding initial mercury concentration (10–100 µg L⁻¹). In preparing the four different test systems labeled D-10 until D-100, specific amounts of prepared biomass powder were carefully weighed using an analytical balance: 0.0050 g for D-10, 0.0100 g for D-40, 0.0150 g for D-70, and 0.0050 g for D-100. The biosorbent dosages were selected to maintain an approximately proportional Hg/MIMA ratio across the intermediate concentrations (D-40: 0.010 g; D-70: 0.015 g). For D-10 and D-100, the same dosage (0.005 g) was intentionally applied to serve contrasting purposes: at D-10, this dosage provided sufficient binding capacity relative to the low mercury load, while at D-100, it was deliberately kept low to avoid premature saturation and to enable observation of maximum uptake capacity and concentration-driven adsorption behavior under binding-site-limited conditions. This design allowed mechanistic comparison between low-concentration/low-dose and high-concentration/low-dose scenarios. The objective of the study was not to optimize the interaction between variables, but rather to assess the biosorption behavior of MIMA under environmentally relevant mercury concentrations while maintaining controlled experimental conditions. Similar proportional dosage scaling is widely used in biosorption studies to avoid confounding saturation effects and to ensure comparability across concentration ranges. Calibrated 50 mL volumetric flasks were employed to prepare the mercury working solutions and diluted with deionized water prior to transfer into 250 mL Erlenmeyer flasks. The pre-weighed biomass was then added to each respective flask for the biosorption process. Experimental conditions were conducted at 25 °C and pH 7.5. All experiments were performed in triplicate for each mercury concentration to achieve statistically reliable outcomes, with values reported as mean ± SD. Data were statistically analyzed using [OriginPro, MS Excel], and one-way analysis of variance (ANOVA) was conducted using the built-in ANOVA tool in OriginPro, to assess differences in removal efficiencies among treatments. A significance level of p < 0.05 was adopted for all statistical comparisons. Additionally, to evaluate non-biological mercury losses, parallel control experiments were conducted using mercury-only solutions operated under the same experimental conditions and designated as C-10 through C-100. Control samples were processed identically to experimental samples to ensure reliable comparison and accurate determination of mercury removal efficiency for each concentration level. In this study, the pH 7.5 and temperature 25 °C were intentionally fixed in order to isolate the effect of mercury concentration and biomass dosage on biosorption performance. This approach is widely used in biosorption studies involving non-living microalgal biomass, where the objective is to characterize adsorption equilibrium, kinetics, and mechanistic behavior under controlled conditions. Evaluating multiple environmental parameters simultaneously may mask the mechanistic interpretation of uptake pathways. The pH of 7.5 was selected based on preliminary optimization from our similar previous study [27] and supported by literature showing that non-living microalgae exhibit optimal Hg(II) adsorption in the neutral to slightly alkaline range [16, 28]. Desorption and recyclability tests were not included in the current scope, but are identified as essential future investigations.

Although the total exposure duration was three days, kinetic sampling was conducted over the first 300 min, during which equilibrium was reached (200–250 min). The remaining duration served to ensure complete equilibrium attainment and to maintain consistent conditions for subsequent characterization analyses. No significant changes in mercury uptake were observed beyond 300 min, confirming that equilibrium had been established within the kinetic sampling window. Upon completion of the three-day exposure phase, the biomass was separated from the mercury-containing solutions by vacuum filtration using filter papers with 0.45 μm pore size. This filtration method was selected over centrifugation to ensure complete recovery of the biomass particles. The filtered solutions from each experimental condition were subsequently retrieved and preserved in suitable containers for subsequent mercury concentration determination. The filter papers containing the biomass from each solution (D-10 through D-100) were collected and air-dried for 12 h at room temperature before being prepared for subsequent detailed characterization that was carried out using scanning electron microscopy, X-ray photoelectron spectroscopy, energy-dispersive X-ray analysis, and Fourier transform infrared spectroscopy.

The quantitative evaluation of Hg(II) removal performance was based on two primary parameters. The removal efficiency (RE%) of mercury was expressed as a percentage and calculated according to the following equation:

$$\:RE\%=\left[\frac{{C}_{0}-{C}_{e}}{{C}_{0}}\right]\times\:100$$
(1)

where in this context, the starting mercury concentration is represented by C₀, while Cₑ refers to the concentration measured at equilibrium (µg L⁻¹). The equilibrium uptake capacity of mercury (qₑ) was determined using the equation below:

$$\:{q}_{e}=\left[\frac{{(C}_{0}-{C}_{e})\times\:\:V}{m}\right]$$
(2)

In this equation, qₑ indicates the mercury uptake per gram of biomass (mg g⁻¹), while V and m represent the solution volume (L) and the biomass quantity employed (g), respectively.

2.3 Kinetic studies

To elucidate the mechanisms and rate-controlling steps associated with mercury uptake, the experimental kinetic data were evaluated using three different kinetic models. Among these, the pseudo-first-order model (Lagergren model) was employed and is described by the following equation:

$$\:\text{log}\left({q}_{e}-{q}_{t}\right)=log{q}_{e}-\left(\frac{{k}_{1}}{2.303}\right)t$$
(3)

Here, the equilibrium adsorption capacity is represented by qₑ (mg g⁻¹), while qₜ indicates the mercury adsorbed at time t (mg g⁻¹); k₁ is the pseudo-first-order kinetic constant with units of min⁻¹. Additionally, the pseudo-second-order kinetic model was applied according to the equation given below.

$$\:\frac{t}{{q}_{t}}=\frac{1}{{k}_{2}{q}_{e}^{2}}+\frac{t}{{q}_{e}}$$
(4)

The parameter k₂ denotes the pseudo-second-order rate constant (g mg⁻¹ min⁻¹), and this kinetic model is commonly associated with improved agreement between predicted and experimental results in biosorption studies. Both linear and nonlinear regression methods were applied to the pseudo-first-order and pseudo-second-order models. Nonlinear regression was considered the primary fitting approach, while linearized forms were included for comparison, as recommended in biosorption kinetic literature.

In addition to the pseudo first and second order models, the intraparticle diffusion model (Weber–Morris model) was applied to further examine whether diffusion within the biomass matrix contributed to the overall rate-controlling mechanism. The model is represented by:

$${q}_{t}=\text{k}_\text{i}\text{d}{t}^{\frac{1}{2}}+\text{C}$$
(5)

where kid is the intraparticle diffusion constant and C is the boundary layer thickness. Multi-stage diffusion behavior is typically indicated when the plot of qₜ versus t1/2 deviates from linearity or consists of multiple linear regions.

2.4 Isotherm studies

To assess mercury distribution at equilibrium, adsorption isotherm experiments were carried out focusing on the interaction between the aqueous phase and microalgal biomass. The resulting experimental data were fitted to multiple isotherm models to characterize Hg(II) removal behavior and estimate the highest uptake capacity of the biomass.

The Langmuir adsorption model was employed to describe monolayer mercury uptake occurring on a uniform surface composed of a limited number of equivalent adsorption sites, as expressed by the following equation:

$$\:\frac{{C}_{e}}{{q}_{e}}=\frac{1}{\left({K}_{L}{q}_{m}\right)}+\frac{{C}_{e}}{{q}_{m}}$$
(6)

Here, the equilibrium adsorption capacity is given by qₑ (mg g⁻¹) and the equilibrium mercury concentration in solution by Cₑ (mg L⁻¹). The maximum biosorption capacity and Langmuir constant are represented by qₘ (mg g⁻¹) and KL (L mg⁻¹), respectively. The favorability of the Langmuir isotherm is commonly assessed using the dimensionless separation factor (RL), calculated as shown below:

$$\:{R}_{L}=\frac{1}{(1+{K}_{L}{C}_{o})}$$
(7)

In this expression, C₀ denotes the starting mercury concentration (mg L⁻¹).

The Freundlich model was also employed to represent adsorption processes occurring on heterogeneous surfaces under non-ideal and reversible conditions with varying adsorption energies and site affinities, as expressed by the following equation:

$$\:log{q}_{e}=log{K}_{F}+log{C}_{e}^{\frac{1}{n}}$$
(8)

Here, the Freundlich coefficient KF [(mg/g)(L/mg)1/n] describes the relative sorption capacity of the biosorbent, while the heterogeneity factor \(\:\frac{1}{n}\) is related to the strength and favorability of the adsorption process.

Adsorbate–adsorbate interactions during adsorption were evaluated using the Temkin isotherm model which assumes a linear decline in adsorption heat with increasing surface coverage. Temkin isotherm was utilized and expressed by the following equation:

$$\:{q}_{e}=\left(\frac{{R}_{T}}{{b}_{T}}\right)\text{l}\text{n}\left({K}_{T}{C}_{e}\right)$$
(9)

Here, the Temkin constants KT (L g⁻¹) and bT (J mol⁻¹) describe the equilibrium binding strength and heat of sorption, respectively. R refers to the universal gas constant (8.314 J mol⁻¹ K⁻¹), and T indicates system temperature expressed in Kelvin.

The Dubinin–Radushkevich (D–R) isotherm model, commonly employed to describe adsorption processes characterized by a Gaussian energy distribution over a heterogeneous surface, was applied in this study. The model is mathematically represented as follows:

$$\:{q}_{e}={q}_{m}\text{e}\text{x}\text{p}(-K{}^{2})$$
(10)
$$ε\:=RT\:\text{l}\text{n}(1+\frac{1}{{C}_{e}})$$
(11)
$$\:E=\frac{1}{\sqrt{2K}}$$
(12)

Here, the D–R isotherm parameters include K, which is related to adsorption energy (mol² kJ⁻²), and qₘ, representing the theoretical maximum uptake (mg g⁻¹). The Polanyi potential is denoted by ε, with R (8.314 J mol⁻¹ K⁻¹) and T (K) indicating the gas constant and system temperature, respectively. The sorption process is further characterized by the mean free energy, E (kJ mol⁻¹).

All isotherm parameters were obtained using nonlinear regression. Linearized isotherm plots were used only for visualization, while the parameter values reported in Table 1 correspond to the nonlinear fitting results, in accordance with recommendations from adsorption modeling literature. These analyses offered valuable understanding of the mechanisms governing biosorption, surface characteristics, and binding favorability and strength of the biomass toward mercury ions under the investigated operating parameters. Furthermore, the isotherm evaluation elucidated the type of the interactions occurring at the interface between mercury species and the microalgal cell surface.

3 Results & discussion

3.1 Characterization

3.1.1 SEM & EDX analysis

SEM micrographs of the raw biomass presented in Fig. 1b revealed a more compact and aggregated structure, characterized by irregular, flattened cellular formations with distinctive surface roughness. The thermal treatment process resulted in a collapsed network structure with numerous micropores and cavities distributed throughout the biomass matrix. This morphological reorganization suggests significant structural modifications during the deactivation process, potentially exposing additional binding sites that contribute to the observed higher uptake capacities. Following mercury exposure at 100 µg/L, examination illustrated in Fig. 1a revealed substantial surface modifications. The treated biomass displayed enhanced surface irregularity with notable changes in the microporous structure. The presence of smaller, more uniform cavities suggests possible mercury-induced surface restructuring. The mercury-loaded nonliving biomass exhibited uniform surface alterations [28]. The structural characteristics of the biomass provide mechanistic insights into its distinct kinetic behavior. The compressed and reorganized cellular structure, combined with the extensive network of micropores, explains the long equilibrium time (200–250 min). This architecture suggests that mercury removal involves not only surface binding but also intraparticle diffusion through the microporous network, consistent with the multi-stage uptake pattern observed in kinetic studies. Furthermore, the enhanced surface area resulting from thermal treatment-induced structural modifications correlates with the higher maximum uptake capacity (0.65 mg/g) achieved by biomass at 100 µg/L. Although EDX did not reveal detectable Hg(II), this result can be explained by the low mercury concentration employed (100 µg L⁻¹), which is below the typical EDX detection limit (~ 0.1 wt%). However, mercury could still be detected by XPS due to its significantly higher surface sensitivity and shallow sampling depth (~ 10 nm).

Fig. 1
Fig. 1
Full size image

SEM images of non-living MIMA biomass: (a) after Hg biosorption at 100 µg L⁻¹ and (b) prior to mercury exposure. Pore size ranging from 0.2 to 1.5 μm in diameter

3.1.2 FTIR analysis

To investigate functional moieties contributing to mercury binding and to infer potential biosorption mechanisms, FTIR analysis was conducted on the MIMA biomass. Spectra were collected over the range 4000 –400 cm⁻¹, comparing raw biomass with samples exposed to 100 µg/L mercury solutions. Comparison of FTIR spectra obtained from MIMA prior to and following mercury biosorption at 100 µg L⁻¹ indicated notable modifications across multiple functional groups, providing insights into the mercury binding mechanisms. In Fig. 2, the raw biomass spectrum exhibited several characteristic bands indicative of typical microalgal cellular components. A broad, intense band centered around 3400 cm⁻¹ corresponds to O-H stretching vibrations from hydroxyl groups and N-H stretching of proteins. After mercury biosorption, this band showed decreased intensity and a slight shift, suggesting these groups’ involvement in mercury binding through hydrogen bonding or complexation [29]. The region around 2900 cm⁻¹ displayed sharp peaks arising from asymmetric and symmetric vibrational modes of C–H bonds in aliphatic moieties. The spectral features in the 1800 –1500 cm⁻¹ region, particularly the bands around 1650 cm⁻¹ and 1540 cm⁻¹, are characteristic of amide I (C = O stretching) and amide II (N-H bending and C-N stretching) vibrations of proteins, respectively. Notable changes in these peaks’ intensity after Hg(II) exposure indicate protein participation in the biosorption mechanism. The region between 1500 –500 cm⁻¹ showed multiple bands related to various functional groups [30]. In addition to the major O–H and amide band shifts, a noticeable change was observed in the region around 1300 cm⁻¹, corresponding to C–O stretching in polysaccharides and P = O vibrations in phosphate-containing groups. These functional groups are commonly implicated in metal binding by non-living algal biomass and their shift after Hg(II) exposure suggests their participation in adsorption. While small shifts were also observed in the amide I and amide II bands, these should be interpreted cautiously as protein structures in non-living biomass are largely denatured. The pronounced changes in the O-containing functional groups (O–H, C–O, P = O) are more consistent with the Hg–O interactions confirmed by XPS.

Fig. 2
Fig. 2
Full size image

FTIR spectra of nonliving MIMA before and after Hg biosorption

3.1.3 XPS analysis

Surface chemical characteristics of non-viable MIMA biomass before and after mercury uptake (100 µg L⁻¹) were examined using X-ray photoelectron spectroscopy. The survey spectra (Fig. 3) were dominated by C 1s and O 1s signals, while Hg 4f features became detectable only after exposure, indicating effective mercury binding at the cell surface. Nevertheless, deconvolution of the C 1s high-resolution spectra for raw biomass presented in Fig. 4a revealed three clearly distinguishable components: C-C/C-H (284.0 eV), C-O/C-N (285.6 eV), and C = O (287.5 eV). After contact with mercury, notable peak shifts were observed as seen in Fig. 4b, with the C-O/C-N shifting to 285.8 eV and the C = O peak to 287.8 eV. The relative intensity changes and peak shifts suggest significant interaction between mercury ions and carbon-containing functional groups, particularly those involving oxygen-bearing moieties. The O 1s spectra illustrated in Fig. 4c and d showed detailed insights into the mercury binding mechanism. The raw biomass exhibited three characteristic peaks: C = O (531.0 eV), C-O (532.7 eV), and COOH (532.8 eV). After mercury biosorption, significant spectral changes were observed, including the newly observed presence of a prominent metal-oxide peak at 529.8 eV and a shift in the C = O peak to 531.45 eV. These changes indicate strong mercury-oxygen interactions and the formation of metal-oxide bonds at the biomass surface. High-resolution examination of the Hg 4f spectrum for the mercury-treated sample (Fig. 4e) displayed a distinct doublet, with binding energies centered at 101.76 eV for Hg 4f₇/₂ and 105.80 eV for Hg 4f₅/₂. These energy positions are characteristic of Hg(II), confirming that mercury remains in its divalent oxidation state following biosorption. Moreover, the clear peak definition and consistent spin–orbit splitting indicate the presence of specific surface interactions rather than nonspecific physical deposition.

Fig. 3
Fig. 3
Full size image

Comparison of X-ray photoelectron spectra of non-living MIMA prior to and following mercury uptake

Fig. 4
Fig. 4
Full size image

High-resolution XPS spectra of non-living MIMA: C 1s and O 1s regions before mercury biosorption (a, c) and after biosorption (b, d), along with the Hg 4f spectra (e)

3.2 Mercury removal performance

3.2.1 Influence of biosorbent dosage and initial Hg(II) concentration on non-living microalgae

The mercury removal capability of MIMA biomass was evaluated across concentrations ranging from 10 to 100 µg/L, employing a strategic variation in biosorbent dosages. As presented in Fig. 5a, initial mercury concentrations of 10, 40, 70, and 100 µg/L were reduced to 3.68 ± 0.21, 8.14 ± 0.37, 14.16 ± 0.62, and 34.32 ± 1.45 µg/L, respectively, demonstrating substantial removal efficiencies between 61.5% and 79% as shown in Fig. 5b. A systematic analysis of the biosorption process reveals interesting patterns when examining both removal efficiency and uptake capacity. At the lowest concentration (D-10, 10 µg/L) with 0.005 g biosorbent, the biomass achieved 61.5 ± 2.7% removal with an equilibrium uptake capacity (qe) of 0.06 ± 0.004 mg/g. When examining the highest concentration (D-100, 100 µg/L) using the same biosorbent dosage (0.005 g), the removal efficiency was comparable at 64.5 ± 3.0%, but notably, the uptake capacity increased dramatically to 0.65 ± 0.030 mg/g. This ten-fold increase in qe, despite similar removal percentages, demonstrates that the biosorbent’s capacity for mercury uptake is significantly enhanced at higher concentrations, likely due to the stronger concentration gradient driving the mass transfer process [31]. Furthermore, the intermediate concentrations showed progressive improvement in both removal efficiency and uptake capacity with increased biosorbent dosages. At 40 µg/L (D-40) with 0.010 g biosorbent, the biomass achieved 78 ± 1.8% rmoval with qe of 0.16 ± 0.009 mg/g, while at 70 µg/L (D-70) with 0.015 g biosorbent, a slightly higher removal efficiency of 79 ± 2.1% was achieved with a slightly higher qe of 0.18 ± 0.010 mg/g. This pattern demonstrates that the increased biosorbent dosage such as in (D-10, D-40, D-70) maintained the high removal efficiency while still allowing for improved uptake capacity per unit mass, indicating effective utilization of the additional binding sites even at higher biosorbent dosages.

Fig. 5
Fig. 5
Full size image

Temporal profiles of Hg(II) biosorption by non-living MIMA showing (a) residual mercury concentration (µg L⁻¹), (b) removal efficiency (%), and (c) adsorption capacity (mg g⁻¹) as functions of contact time

3.2.2 Influence of contact time on Hg(II) using non-living microalgae

The time-dependent removal of Hg(II) by MIMA showed a progressive progression toward equilibrium at all examined concentration levels between 10 and 100 µg L⁻¹. As illustrated in Fig. 5b and c, the corresponding trends in removal efficiency and uptake capacity (q), respectively, showed distinct patterns characterized by multiple stages of uptake over an extended period. The uptake capacity (q) versus time profiles revealed strong concentration-dependent behavior, most notably with D-100 showing significantly higher uptake capacity than other concentrations. The D-100 profile demonstrated a distinctive multi-stage uptake pattern: an initial rapid increase in the first 20 min, followed by a more gradual increase until about 150 min, and finally a slower approach to equilibrium, ultimately reaching approximately 0.65 mg/g after 300 min. The lower concentrations (D-70, D-40, and D-10) showed similar multi-stage patterns but with lower final uptake capacities, reaching approximately 0.18 ± 0.01, 0.16 ± 0.009, and 0.06 ± 0.004 mg/g, respectively.

The removal efficiency versus time curves revealed interesting and somewhat unexpected patterns. Despite D-100 showing the highest uptake capacity, its removal efficiency, approximately 65 ± 3% was lower than D-70 and D-40, which both achieved around 79% removal efficiency. D-10 showed similar removal efficiency to D-100, reaching about 61.5 ± 2.7%. This pattern suggests that while higher concentrations drive greater absolute uptake (q), the relative efficiency of removal may be optimized at intermediate concentrations. A notable feature of the biomass kinetics is the extended time required to reach equilibrium, typically 200–250 min. This slow kinetic behavior suggests a mechanism of mercury binding, likely involving gradual diffusion into the biomass matrix and interaction with internal binding sites [32]. The removal profiles also show distinct stages: (1) An initial rapid uptake stage in the first 20–30 min (2) A transition stage with moderate uptake rate (30–150 min) (3) A final slow approach to equilibrium (150–300 min). This multi-stage behavior is particularly evident in the D-100 profile, suggesting that different binding mechanisms or sites may become active as the process progresses. The gradual increase in uptake capacity over time might indicate a combination of surface adsorption followed by slower intraparticle diffusion processes [33]. However, the contrasting behaviors at different concentrations, particularly the similar removal efficiencies of D-10 and D-100 despite vastly different uptake capacities, suggest complex interactions between concentration, available binding sites, and mass transfer processes. This complexity likely arises from the physical structure of the biomass and the distribution of binding sites throughout the biomass matrix [34]. Based on the analyses and results, a proposed mechanism of Hg(II) removal by non-living MIMA is illustrated in Fig. 6.

Fig. 6
Fig. 6
Full size image

A proposed mechanism of Hg(II) removal by non-living MIMA

These characteristics have important implications for process design and optimization. The extended time required to reach equilibrium suggests that biomass would require longer contact times to achieve maximum removal efficiency [35]. However, the higher ultimate uptake capacities achieved, particularly at higher concentrations, indicate potential advantages for treating more concentrated mercury solutions where extended contact times are acceptable. Also, these findings have significant implications for process optimization and practical applications. The ability to achieve higher uptake capacity with lower biosorbent dosage at higher concentrations suggests potential economic advantages in treating more concentrated mercury solutions. However, the trade-off between removal efficiency and uptake capacity must be carefully considered when designing treatment systems. The data indicate that while higher biosorbent dosages can achieve better removal percentages, they may result in less efficient utilization of the biosorbent material itself.

3.3 Isotherm models

Figure 7 illustrates the four adsorption isotherm models applied, while the corresponding models parameters are summarized in Table 1. Figure 7e presents the experimental mercury adsorption isotherm (qₑ vs. Cₑ), constructed using the four environmentally relevant mercury concentrations evaluated in this study. It is acknowledged that four data points impose inherent limitations on the statistical confidence of isotherm fitting, and the apparent linearity of the experimental isotherm (Fig. 7e) should be interpreted with caution. Nevertheless, the selected concentration range was deliberately constrained to environmentally relevant trace levels, and the purpose of isotherm modeling here is mechanistic rather than predictive, to characterize surface heterogeneity and adsorption energy distribution rather than to determine high-concentration saturation behavior. The Langmuir isotherm showed remarkably poor correlation (R² = 0.0206) for the biomass, the lowest among all models tested. The calculated parameters include a maximum adsorption capacity (qmax) of 16.3 mg/g and a negative Langmuir constant (KL = -0.508). The extremely low R² value and negative KL indicate that the biosorption process fundamentally deviates from the Langmuir model’s assumptions of monolayer adsorption on homogeneous surfaces [36]. This suggests that the binding mechanism for biomass entails the presence of adsorption interactions more complex than monolayer coverage. In addition, the Freundlich model exhibited a high level of consistency with the observed data (R² = 0.964), suggesting that mercury biosorption on biomass occurs through heterogeneous surface interactions. The Freundlich constant (KF) value of 16.9 indicates moderate adsorption capacity. Notably, the value of n is almost exactly 1.00, suggesting nearly linear adsorption behavior. This implies that the energy distribution for adsorption sites is relatively uniform, despite the surface heterogeneity, and that the adsorption affinity remains relatively constant as more mercury ions are bound [37]. Nevertheless, the Temkin isotherm exhibited good level of consistency (R² = 0.824), which may indicate moderate adsorbate-adsorbate interactions during the biosorption process. The model parameters include a bT value of 0.252 and a KT value of 259. The relatively low bT value suggests weaker heat of adsorption effects, while the high KT value indicates strong binding affinity at equilibrium [38]. Furthermore, the Dubinin-Radushkevich (D-R) isotherm showed very good correlation (R² = 0.946) and provided crucial insights into the biosorption mechanism. The mean free energy of sorption (E) was calculated as 5.00 kJ/mol. This E value, falling within the range of 1–8 kJ/mol, confirms that physical adsorption is also the predominant mechanism [21]. The maximum theoretical saturation capacity (qmax) was calculated as 1.95 mg/g, with a β value of 2.00 ⋅ 10− 8 mol²/kJ². The theoretical qmax derived from the D–R model exceeds the experimentally observed maximum uptake (0.65 mg/g), which is a commonly reported phenomenon in D–R modeling. This discrepancy arises because the D–R model estimates a theoretical saturation capacity assuming complete surface coverage across a Gaussian energy distribution, whereas the experimental qmax reflects uptake under the limited concentration range studied (10–100 µg/L), where full surface saturation was not reached. Therefore, the D–R qmax should be interpreted as a theoretical upper bound rather than a directly comparable experimental value.The isotherm analysis reveals several important characteristics underlying the biosorptive uptake process. The excellent result of the Freundlich model combined with the D-R isotherm suggests a heterogeneous adsorption process dominated by physical interactions. Moreover, the near-unity n value in the Freundlich model is particularly interesting, as it suggests more uniform binding site energetics. This could be attributed to the thermal treatment process during biomass preparation, which might have homogenized the surface binding sites to some extent, while still maintaining surface heterogeneity [39]. It is important to note that mechanistic interpretation was primarily based on the isotherm models that exhibited strong agreement with the experimental data (Freundlich and D–R), as indicated by their high correlation coefficients. The Langmuir and Temkin models showed lower goodness of fit and were therefore considered only for comparative purposes rather than detailed mechanistic inference.

Fig. 7
Fig. 7
Full size image

Isotherm modeling of Hg(II) biosorption onto non-living MIMA: (a) Langmuir model; (b) Freundlich model; (c) Temkin model; (d) Dubinin–Radushkevich (D–R) model; and (e) Experimental mercury adsorption isotherm

Table 1 Isotherm model parameters describing Hg(II) uptake by non-living MIMA biomass
Full size table

3.4 Kinetic models

The adsorption kinetics governing mercury uptake by MIMA demonstrated a notable pattern in terms of uptake and equilibrium achievement time. The process was analyzed using both linearized and non-linear representations of the pseudo-first-order and pseudo-second-order kinetic models in addition to the intraparticle diffusion kinetic model across the concentration range of 10–100 µg/L as shown in Fig. 8, revealing distinct patterns in the biosorption mechanism. The pseudo-first-order kinetic model showed good correlation across all concentrations, with R² values improving as concentration increased: 0.928 for D-10, 0.950 for D-40, 0.986 for D-70, and 0.984 for D-100. The rate constants (K₁) exhibited a gradual increase with concentration, from 0.00759 min⁻¹ for D-10 to 0.0126 min⁻¹ for D-100, indicating faster kinetics at higher concentrations. The calculated equilibrium uptake capacities (qe) exhibited an increasing trend with rising concentration, ranging from 0.0271 mg/g for D-10 to 0.588 mg/g for D-100, reflecting the strong concentration dependence of the biosorption process. On the other hand, superior model–data agreement was obtained using the pseudo-second-order kinetic model for all concentrations examined, with consistently higher R² values (0.932 to 0.992) and lower RMSD values compared to the pseudo-first-order model. A notable trend was observed in the rate constants (K₂), which decreased significantly with increasing concentration, from 1.054 g/mg·min for D-10 to 0.0236 g/mg·min for D-100. This inverse relationship between K₂ and concentration suggests that while overall uptake capacity increased, a decrease was observed in the mercury uptake rate relative to the available concentration at higher concentrations, possibly due to saturation effects [40, 41]. Parameter values of pseudo-first-order and pseudo-second-order models reported in Table 2 correspond to the nonlinear regression results. The application of non-linear modeling also facilitated improved visual interpretation of biosorption dynamics. The biomass showed a more gradual approach to equilibrium, typically requiring 200–250 min. The uptake profiles exhibited three distinct stages: an initial rapid uptake within the first 40 min, followed by a slower uptake stage until about 150 min, and finally a gradual approach to equilibrium. This multi-stage behavior was particularly evident in the D-100 experiment, where the uptake capacity reached 0.721 mg/g under pseudo-second-order conditions.

Fig. 8
Fig. 8
Full size image

Kinetic modeling of Hg(II) biosorption by non-living MIMA: (a) linearized pseudo-first-order model; (b) linearized pseudo-second-order model; (c) non-linear pseudo-first-order model; (d) non-linear pseudo-second-order model; and (e) intraparticle diffusion model

Table 2 Estimated parameters of kinetic models describing Hg(II) biosorption by non-living MIMA
Full size table

The slow kinetics might be attributed to the absence of active cellular processes and the reliance on passive diffusion and surface binding mechanisms, which are influenced by the surface area and porosity of the biomass [42, 43]. The gradual increase in uptake capacity over time suggests a complex interplay between surface adsorption and intraparticle diffusion processes, which are influenced by the pore structure and volume of the biomass [44]. In addition, the concentration dependence of both kinetic models’ parameters provides valuable insights into the biosorption mechanism. The systematic increase in first-order rate constants (K₁) with concentration indicates that higher mercury concentrations enhance the initial binding rate. Conversely, the decrease in second-order rate constants (K₂) suggests that surface saturation effects become increasingly important at higher concentrations, potentially limiting the rate of subsequent mercury uptake. To further examine whether diffusion within the biomass matrix contributed to the rate-controlling mechanism, the intraparticle diffusion model (Weber–Morris) was applied to all concentrations as shown in Fig. 8e. The plots of qₜ versus t1/2 did not pass through the origin, and all exhibited non-zero intercepts, indicating that intraparticle diffusion was not the sole rate-limiting step. The highest linearity was observed at 100 µg/L with R² = 0.95, suggesting a stronger contribution of intraparticle diffusion under higher driving-force conditions. The linearity progressively decreased at lower concentrations (R² = 0.92 for 70 µg/L, 0.82 for 40 µg/L, and 0.69 for 10 µg/L), consistent with a weaker diffusional driving force at lower concentration gradients. Moreover, the lower-concentration curves displayed multi-linear regions, confirming a multi-stage kinetic process consisting of: (1) external boundary layer diffusion, (2) surface adsorption, and (3) gradual intraparticle diffusion into the porous biomass matrix. These findings corroborate the slow attainment of equilibrium (200–250 min) and support the pseudo-second-order model results, indicating that while diffusion is involved, surface-controlled adsorption and boundary-layer effects dominate the early uptake stages. The best fit of the pseudo-second-order model should not be interpreted as definitive evidence of chemisorption, as this model frequently provides superior fits to biosorption data regardless of the underlying mechanism, including processes dominated by diffusion or surface heterogeneity rather than chemical reaction. The rate-limiting step is more accurately described as surface-controlled adsorption involving multiple concurrent processes. This interpretation is consistent with the D–R isotherm results (E = 5.00 kJ/mol), which confirm the predominance of physical interactions, and with the intraparticle diffusion analysis indicating significant boundary layer and diffusional contributions to the overall uptake rate.

4 Comparison, limitations, and future work

To contextualize the adsorption capacity obtained in this study, which achieved a highest of 0.65 mg/g, Table 3 compares the mercury biosorption performance of MIMA with other microalgal, biological, and lignocellulosic biosorbents reported in the literature. Several studies have reported mercury biosorption using algal biomass at mg/L concentrations, including Chlorella vulgaris [28], Spirogyra [45], and Chlamydomonas reinhardtii [46], and Spirulina sp [16]., all studies investigate Hg(II) removal at mg/L-level concentrations, which typically yield higher adsorption capacities. The lower qe observed in the present work is expected, given the mercury concentrations consistent with environmentally relevant conditions (10–100 µg L⁻¹) examined. Even though this research advances understanding of mercury removal by MIMA, certain limitations warrant consideration. First, the effects of coexisting cations (e.g., Ca²⁺, Mg²⁺, Pb²⁺), which are commonly found in natural aquatic systems and may influence adsorption behavior, were not explored. Assessing the impact of competing ions represents a crucial avenue for future research to more accurately replicate natural environmental conditions [23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47]. Second, the study did not examine the reusability of MIMA following mercury desorption. Investigating regeneration and multiple-use cycles is essential to determine the feasibility of large-scale applications. Moreover, future investigations should also assess the role of major operational parameters such as temperature, pH, and illumination, as these factors are anticipated to affect the removal performance. Additionally, integrating statistical and computational modeling approaches such as response surface techniques as well as machine learning–driven methods optimization could offer predictive insights into system behavior under varying conditions and support the design of scalable, high-efficiency treatment systems.

Table 3 Compares the mercury biosorption performance of MIMA with other microalgal, biological, and lignocellulosic biosorbents reported in the literature
Full size table

5 Conclusion

The present work provides insight into the Hg(II) biosorption behavior of MIMA at concentrations representative of environmental conditions, providing valuable insights into their potential as cost-effective biosorbents. While the maximum uptake capacity (0.65 mg/g) may appear modest relative to literature values, it should be noted that most reported biosorbents achieving higher capacities were evaluated at mg/L-level mercury concentrations, one to three orders of magnitude higher than those studied here. At environmentally relevant trace concentrations (10–100 µg/L), removal efficiency is the more meaningful performance indicator, and MIMA achieved 61.5–79% removal under these conditions. MIMA achieved substantial mercury removal efficiencies ranging from 61.5% to 79% across the tested range (10–100 µg/L), with a maximum uptake capacity of 0.65 mg/g. The process exhibited a multi-stage kinetic pattern and reached equilibrium after approximately 200–250 min, indicating the predominance of diffusion-controlled and surface-limited mechanisms rather than active transport. Isotherm analysis demonstrated an excellent fit to the Freundlich and Dubinin–Radushkevich models (R² = 0.964 and 0.946, respectively), confirming heterogeneous surface adsorption dominated by physical interactions with a mean sorption energy (E) of 5.0 kJ/mol. These findings align with the morphological and spectroscopic results, where SEM revealed thermally induced microporosity, FTIR confirmed the participation of hydroxyl, carbonyl, and amide groups in metal binding, and XPS spectra verified the presence of Hg(II) species forming stable mercury–oxygen bonds on the biomass surface. The use of non-living biomass offers significant advantages for scalability, as it eliminates the need for light, nutrients, and strict cultivation control, allowing processing in simple, low-cost reactors. The indigenous origin of the biomass also makes large-scale production feasible in coastal regions, and its thermal stability enables long-term storage and transport. While reusability and regeneration require further investigation, the present results indicate that MIMA has promising potential for deployment in decentralized or field-scale mercury remediation systems.