Ecofriendly Application of Calabrese Broccoli Stalk Waste as a Biosorbent for the Removal of Pb(II) Ions from Aqueous Media

Abstract

A new biosorbent obtained from Calabrese broccoli stalks has been prepared, characterised and used as an effective, low-cost and ecofriendly biomass to remove Pb(II) from aqueous solutions, without any complicated pretreatment. Structural and morphological characterisation were performed by TGA/DGT, FTIR and SEM/EDX; the main components are hemicellulose, starches, pectin, cellulose, lignin and phytochemicals, with important electron donor elements (such as S from glucosinolates of broccoli) involved in Pb(II) sorption. The biosorbent showed values of 0.52 and 0.65 g mL⁻¹ for bulk and apparent densities, 20.6% porosity, a specific surface area of 15.3 m² g⁻¹, pH_pzc 6.25, iodine capacity of 619 mg g⁻¹ and a cation exchange capacity of 30.7 cmol kg⁻¹. Very good sorption (88.3 ± 0.8%) occurred at pH 4.8 with a biomass dose of 10 g L⁻¹ after 8 h. The Freundlich and Dubinin–Radushkevich isotherms and the pseudo-second-order kinetic models explained with good fits the favourable Pb(II) sorption on the heterogeneous surface of broccoli biomass. The maximum adsorption capacity was 586.7 mg g⁻¹. The thermodynamic parameters evaluated showed the endothermic and spontaneous nature of the Pb(II) biosorption. The chemical mechanisms mainly involved complexation, ligand exchange and cation–π interaction, with possible precipitation.

1. Introduction

The handling, storage and/or disposal of the large amount of food and agro-industrial waste generated at different stages of the production process is a well-known global problem. In most cases, they are neither processed nor disposed of properly, a situation that contributes to the increase in environmental pollution. Paradoxically, these wastes have an intrinsic potential to be exploited in different processes, including the manufacture of new products (such as in animal feed production), the reuse as feedstock for biofuels (bioethanol, biodiesel, biogas, energy biomass) and, alternatively, their use as sorbents in bioremediation for the removal of contaminants (for example, in the removal of heavy metals, dyes or hydrocarbons) in order to recover the state of the environment that has been damaged [1,2,3,4,5,6,7]. The production of fertilizers or micronutrient dietary supplements for animals or plants is also an attractive recent use of biomass, which is used when the enrichment of biomass is carried out in nutritionally significant elements [8].

The presence of heavy metals in aqueous media such as aquatic environments, wastewater or industrial effluents is an important ecological problem due to their toxic nature and their ability to accumulate in the food chain. These elements affect the water quality and tend to accumulate in living organisms, resulting in various disorders and diseases in the ecosystems. Among them, lead is a prominent pollutant with drastic health consequences. It is a cumulative toxicant that affects multiple body systems, principally harmful to young children, and can cause anaemia, kidney malfunction, brain tissue damage, and even death in cases of severe poisoning [9]. Lead pollution in water not only occurs naturally but can also be due to anthropogenic activities, such as battery manufacturing, metal plating and finishing, glass manufacturing, anti-knocking synthesis, ceramic industries and others [10]. According to the WHO guidelines for drinking-water quality, the maximum tolerable limit is 0.005 mg L⁻¹ (24 nmol L⁻¹) for total lead amounts; moreover, the Environmental Protection Agency (EPA) allows the permissible limit of 0.05 mg L⁻¹ (0.24 µmol L⁻¹) for lead in wastewater. However, the usual content of lead in industrial wastewater ranges from 200 to 500 mg L⁻¹, exceeding water quality standards [11].

The use of ecofriendly biomaterials for sorption processes applied in the cleaning of waters contaminated by heavy metals, such as lead, has gained huge attention [12]. The ideal sorbent for wide application needs to be inexpensive, available in large quantities, non-toxic, with little or no processing, and with known kinetic parameters and good sorption characteristics [13]. In recent years, these studies have been focused on the use of waste materials, especially those obtained from food and agricultural waste as citrus peel, olive stone, groundnut shell, potato or cucumber peels, barley straw, cranberry kernel shell, tea waste, potato, strawberry or canola stems, among others [2,7,14,15,16,17,18,19,20]. This type of sorbents offers great potential for innovation in metal removal from aqueous media at a very low cost [21].

Agro-materials are usually composed of lignin and cellulose as main constituents and may also include other polar functional groups, such as alcohols, aldehydes, ketones, carboxylic, phenolic and ether groups, with an average composition of 40–50% cellulose, 20–30% hemicellulose, 20–25% lignin and 1–5% ash [22,23,24]. The removal of heavy metals in aqueous media will depend on the sorption capacity of these functional groups towards metal ions, among other different factors.

Broccoli (Brassica oleracea var. italica) is a valuable plant from the Brassicaceae (or Cruciferae) family (genus Brassica); the floret is the main edible part of this vegetable. The broccoli plant has several components, standing out the glucosinolates (mainly glucoraphanin and sulforaphane) [25], like other cruciferous vegetables; these sulphur-containing compounds give broccoli its characteristic odour and taste. The chemical composition of broccoli stalk also includes soluble and insoluble fibre (pectins, hemicellulose, cellulose, sugars, starch, lignin, etc.), minerals (K, Ca, Mg, P, Fe, etc.), vitamins (A, C (ascorbic acid), K and some B vitamins like B9 (folate)) and other phytochemicals (organic acids, organophosphorus and phenolic compounds such as flavonoids, phytates, etc.) and aminoacids (particularly, phenylalanine, tryptophan and cysteine) [26,27,28,29,30]. It has acquired a considerable relevance in the last few years as a health-promoting food, rich in antioxidants and anti-inflammatories. However, broccoli marketable florets (flower heads) represent only a minor part of the total above-ground plant biomass (<25% of the total yield), which generates a huge amount of agricultural waste [31,32]. So far, the use of broccoli waste is restricted to flour and fibre, standard extraction or characterisation of glucosinolates, and dairy cattle feed production [33,34,35,36]. The waste produced from broccoli by horticulture represents around 60–75% of the broccoli production [37,38]. Considering the worldwide production in 2017 (25,984,758 tonnes, reports combined with cauliflower) and if assuming that 60% of production is wasted, more than 15 million tonnes of waste were generated in 2017 [28]. Research studies that used broccoli stalk waste as sorbents of heavy metals are scarce in the literature. So far, there is only one study that focuses in the application of broccoli stalk waste (as raw material and as a precursor of carbon-based sorbents) in the removal of Cd(II), Ni(II), Cu(II) and Zn(II) [39].

Therefore, this study aims to add value to broccoli stalk waste by using it as a biosorbent for the removal of toxic metals, such as Pb(II) ions, from aqueous solutions. The application of this biomass in sorption processes can be an interesting and ecofriendly remediation strategy to clean polluted environments, to be applied to treat industrial wastewater or effluents, or for any other separation purposes.

2. Experimental

2.1. Chemical Reagents and Equipment

Different chemical reagents of analytical grade were used according to the methodologies to be applied or the parameters to be analysed: (a) iodine adsorption capacity: iodine (0.01 mol L⁻¹ I₂ solution, Panreac, Barcelona, Spain), sodium thiosulfate pentahydrate (Na₂S₂O₃·5H₂O, Panreac, Barcelona, Spain), starch (1% (C₆H₁₀O₅)_n for HPLC, Panreac, Barcelona, Spain), potassium iodide (KI, Merck, Darmstadt, Germany) and potassium dichromate (K₂Cr₂O₇, Merck, Darmstadt, Germany); (b) acid and basic surface groups, the pH value at the point of zero charge (pH_pzc) and the cation exchange capacity of the biomass: sodium hydroxide (NaOH), potassium hydrogen phthalate (KC₈H₅O₄), sodium carbonate (Na₂CO₃), sodium hydrogen carbonate (NaHCO₃), hydrochloric acid (HCl, 37%), barium acetate (Ba(CH₃COO)₂), silver nitrate (AgNO₃) and sodium chloride (NaCl) (Panreac, Barcelona, Spain); (c) extraction of natural fats and oils: n-hexane (hexane mixture of isomers RS for HPLC Isocratic Grade, Carbo-Erba, Sabadell, Spain); (d) sorption experiments: solutions of Pb(NO₃)₂ (100%, Panreac, Barcelona, Spain) as Pb(II) precursor and acetate buffer (prepared conventionally using acetic acid (CH₃COOH, 96%, Merck, Darmstadt, Germany) and sodium hydroxide (NaOH, Panreac, Barcelona, Spain)); (e) standard solutions of the calibration curves for metal analysis: dilutions of an ICP standard solution of 1000 mg L⁻¹ Pb in 2–3% HNO₃ (Certipur, Merck, Darmstadt, Germany).

Pure (type II, <1 µS cm⁻¹) or ultrapure water (type I, 18.2 MΩ cm) was used as required for aqueous solutions and obtained by an Autwomatic system coupled with an Ultramatic Plus system (Wasserlab, Barbatáin, Spain).

A wide variety of instrumental equipment was used: (a) basic equipment: a J.P. Selecta oven (Selecta, Barcelona, Spain), a Crison Basic 20+ pH-meter with a 50–10 T combined glass-Ag/AgCl wire (Crison Instruments, Barcelona, Spain), an IKA HS 501 D open air laboratory shaker platform (Labortechnik, Wasserburg am Bodensee, Germany) and a D95 Dinko vacuum pump (Dinko Instruments, Barcelona, Spain); (b) advanced instrumentation: a TGA7 Thermogravimetric Analyzer (PerkinElmer, Waltham, MA, USA), a Shimadzu IRAffinity-1S Fourier transform infrared spectrophotometer (Shimadzu Corporation, Kyoto, Japan) with the PIKE MIRacle™ ATR sampling accessory (PIKE Technologies, Fitchburg, WI, USA), the SEM-FEI Nova NanoSEM 450 with the secondary electrons detector (TLD-SE) (Nova, Fort Worth, TX, USA) and the EDAX detector 100 mm² surface (AMETEK^®, Newark, DE, USA), the Microtrac NANOTRAC Wave DLS (dynamic light scattering) analyser, and an iCE 3000 series Atomic Absorption Spectrometer (Thermo Scientific, Waltham, MA, USA).

The data modelling for isotherm, kinetic and thermodynamic studies was processed on Excel 2016 program (Microsoft, Redmond, DC, USA).

2.2. Preparation of Broccoli Stalks as a Natural Sorbent

Calabrese broccoli was purchased at a local market. The broccoli stalk and big stems present in the vegetable were separated, washed several times with pure water and chopped into small chunks. After that, they were oven-dried at 45 °C until constant weight, crushed and ground in a mortar. Biomass particles of 125 < x < 249 µm and x < 125 µm sizes were obtained by using 60-mesh and 120-mesh nylon sieves (CISA, Barcelona, Spain), respectively. A portion of the biomass was defatted applying the Soxhlet method. For that, 6 g of dry broccoli biomass were extracted by heating to reflux with 130 mL of n-hexane for 8 h. The defatted and non-defatted biomass obtained were stored dry until further studies.

2.3. Methodologies for Biomass Characterisation

The following methodologies and techniques were applied for the characterisation of the biomass using the 125 < x < 249 µm particle size fraction. Thermogravimetric and derivative thermogravimetric analyses (TGA and DTG) were performed under nitrogen atmosphere at 10 °C min⁻¹ until 850 °C. The FTIR spectra in attenuated total reflection mode (ATR) were obtained with a 4 cm⁻¹ resolution in the region from 4000 to 650 cm⁻¹. A scanning electron microscope (SEM) (operated at an accelerating voltage of 5 kV) provided images of the biomass surface and a coupled energy-dispersive X-ray spectrometer (EDXS) allowed the microanalysis of the raw and Pb-loaded biomass after the sorption process. The micrographs were taken with the sample previously deposited onto a carbon grid and coated by sputtering with a 15 nm gold layer.

The procedures for the determinations of the iodine adsorption capacity [40], cation exchange capacity (CEC) [41], functional groups present in the biomass [42,43], biomass pH and conductivity [44], point of zero charge (pH_pzc) [45], bulk density (BD) [40] and apparent density (AD) [46] are described in Table S1 of the Supplementary Material. The percentage of porosity was calculated according to the following equation [47]:

$$porosity\left(\%\right)=\left[1-\frac{BD}{AD}\right]\times 100$$

(1)

The specific surface area (SSA) was evaluated by dynamic light scattering measurements. All the determinations were performed in duplicate.

2.4. Metal Removal Experiments

The capacity of the biomass for Pb(II) removal was determined by batch experiments. The broccoli biomass was added to 50 mL of an aqueous solution containing different¹ metal concentrations according to each study. The pH of the solution was adjusted to the selected value with 0.1 mol L⁻¹ acetate buffer for each experiment. Experiments were performed in duplicate using 100 mL polypropylene containers with a screw cap and the suspensions were shaken on an orbital shaker at 200 rpm under a controlled temperature. The suspensions were filtered using a vacuum pump, a glass filter holder and 4.7 cm (ϕ) glass microfiber filters (Whatman^®, Maidstone, UK). After the experiments, the concentrations of Pb(II) in the filtered suspensions were analysed by atomic absorption spectroscopy.

The percentage of Pb(II) removal by the biomass was determined as follows:

$$\%{Pb}_{removal}=\frac{C_0-C_t}{C_0}\times 100$$

(2)

where C₀ and C_t were the initial concentration and t-time concentration of Pb(II) ions (mg L⁻¹) in the aqueous solution, respectively.

The sorption capacity (q_t, mg g⁻¹), defined as the amount of Pb(II) ions (mg) sorbed per unit weight of sorbent (g), was calculated from the following equation:

$$q_t=\frac{{(C}_0-C_t)\times V}{m}$$

(3)

where m was the mass of the biosorbent (g) and V was the volume of the solution (L).

Finally, the equilibrium sorption capacity (q_e, mg g⁻¹) was obtained according to the following equation:

$$q_e=\frac{{(C}_0-C_e)\times V}{m}$$

(4)

where C_e is the concentration of Pb(II) (mg L⁻¹) in the aqueous solution at equilibrium.

2.5. Sorption Isotherms

The sorption isotherms were performed by adding 0.5 g of biomass (particle size of 0.125 mm < x < 0.249 mm) to 50 mL of aqueous metal solution at pH 4.8 and 23 °C for 8 h (equilibrium time) following the procedure described in Section 2.4. The initial concentrations of Pb(II) used were 0.25; 0.5; 1 and 1.5 mmol L⁻¹. The fit of the experimental data with the linear forms of the Langmuir, Freundlich, Dubinin–Radushkevich (D-R) and Temkin isotherm models was studied (equations in Supplementary Material Table S2) [48,49,50,51].

2.6. Sorption Kinetics

Pseudo-first-order (or Lagergen kinetic model), pseudo-second-order, Elovich, Ritchie’s second-order, first-order reversible and intraparticle diffusion models were applied to study the dynamic sorption process by broccoli biomass. The kinetic equations, their linear forms and the axes of the plots are summarised in Table S3 of the Supplementary Material [49,52,53,54,55,56]. The experiments were carried out for different time periods (n = 10) under the same conditions as the sorption isotherms study using 0.25, 0.5, 1 and 1.5 mmol L⁻¹ Pb(II) solutions.

2.7. Thermodynamic Studies

The thermodynamic parameters of the sorption process were studied through batch experiments carried out at different temperatures (18, 23 and 30 °C) under the same conditions as the study of sorption isotherms using 1 mmol L⁻¹ Pb(II) solutions for 8 h. The equations for the changes of standard Gibbs free energy, enthalpy and entropy, and also the activation energy and the sticking probability of biomass [57], are shown in Table S4 of the Supplementary Material.

3. Results and Discussion

3.1. Previous Studies to Select the Biomass Preparation

The aim of this study was to propose an ecofriendly use of broccoli stalks waste as a biosorbent for Pb(II) ions in aqueous solution. One aspect to take into account was to evaluate the suitability of a short preparation. To achieve this, the drying of biomass (particle size: x < 125 µm) was studied using two different treatments: (1) at low temperature, oven-dried at 45 °C until constant weight and (2) applying a freeze-drying process. The batch Pb(II) sorption experiments, as described in 2.4, were carried out during 24 h and a ratio biomass/aqueous solution of 0.5 g/50 mL (10 g L⁻¹) was applied. The results obtained were comparable, yielding removal rates of 83.2 ± 0.8% and 84.3 ± 1.3%, respectively. A comparison was drawn between the removal efficiencies of particles sizes of 125 < x < 249 µm and x < 125 µm; these were oven-dried in both cases. The percentage of Pb(II) sorption exhibited negligible variations with particle size. However, a slightly higher removal efficiency of 88.5 ± 0.5% was achieved with the larger particle size (0.125 mm < x < 0.249 mm). On the other hand, it was demonstrated that the defatting process did not contribute to the enhancement of metal sorption. As a result, the amount of Pb(II) sorbed by the broccoli biomass after fat and oil extraction was found to be 88.0 ± 0.8%. Therefore, further experiments were carried out employing the oven-dried broccoli biomass at 45 °C using a particle size of 125 < x < 249 µm. In this way, the pretreatment was very efficient, marked by a minimal consumption of resources, with only a few steps, being a simple process.

3.2. Sorbent Characterisation

The results for the characterisation of the sorbent are shown in Figure 1, Figure 2, Figure 3 and Figure 4 and

Table 1. Characterisation of broccoli stalk biomass.

Parameter	Average Value	SD
Iodine adsorption capacity (mg I2 g−1 biomass)	619.7	33.3
Cation exchange capacity, CEC (meq H+ exchange 100 g−1 biomass = cmol kg−1 biomass)	30.7	0.01
Functional groups (mmol g−1)
Basic	1.085	0.021
Acid	1.257	0.012
Carboxylic group	1.216	0.012
Lactonic group	0.006	0.013
Phenolic group	0.035	0.013
pH of biomass	6.88	0.02
pHpzc	6.25	0.02
Conductivity (µS cm−1)	1566	22
Bulk density, BD (g mL−1)	0.52	0.01
Apparent density, AD (g mL−1)	0.65	0.03
Porosity (%)	20.6	3.2
Specific surface area, SSA (m2 g−1)	15.3	2.7

. Figure 1 shows the TGA/DTG profiles obtained for broccoli biomass where different processes can be distinguished, contributing to a total weight loss of 89.9%: (a) the first loss of 11% due to moisture removal, meaning the presence of significant hydrophilic functional groups until 148 °C [58]; (b) the loss of 47.9% up to 400 °C and 16.4% up to 500 °C due to the removal of volatile materials (principally hemicellulose and cellulose, respectively); and (c) the pyrolysis of the more stable lignins in the range of 500–600 °C with a loss of 8.0%; this high thermal stability of lignin may be due to the formation of pseudolignin [59]. Also, the pyrolysis of char produced from the thermal degradation of pectins and hemicelluloses may be involved in this loss of weight [60]. The pyrolysis analysis was over at 850 °C with a low weight loss of 6.7%. The ash weight at 850 °C was 9.7% of the initial biomass.

Several characteristics of FTIR spectra were observed related to the main compounds of broccoli stalk: hemicellulose, starches, pectin, cellulose, lignin and some phytochemicals. The absorption intensities generally increased for Pb-loaded biomass (decreasing transmittance) (Figure 2). Aldehyde and ketone C–O stretching, alkyl C–H stretching, and alcohol O–H stretching from carbohydrates were associated with peaks at 1022 cm⁻¹ and 2910 cm⁻¹, and with a broad band around 3320 cm⁻¹, respectively [25]. The alteration of this broad band suggests that Pb(II) ions compete with the hydrogen bonding between OH groups associated with cellulose and proteins of the biomass [61]. This behaviour has been considered as a ligand exchange process and it occurs when hydroxyl groups exist on the adsorbent as covalent bonds and not ions [62]. The peak around 1080 cm⁻¹, corresponding to the symmetrical stretching vibrations of the C–O–C ether bond and related to the glycosidic linkage in polysaccharides, was affected by Pb(II) sorption, as well as the small peak at 900 cm⁻¹. Similarly, the peak around 1730 cm⁻¹ was attributed to C=O stretching present in some carbohydrates, such as acetyl groups in hemicellulose, in flavonoids or phenolic acids. The peaks between 2130 and 2058 cm⁻¹ were correlated with asymmetric stretching of N=C=S of glucosinolates [63,64]. The C=S stretching vibration was also identified by the peak at 1022 cm⁻¹ that shifted after Pb(II) sorption [65]. The peaks between 1200 cm⁻¹ and 1400 cm⁻¹ related to the symmetric and asymmetric stretch of S=O were also found to be characteristic of this type of glucosides and the interaction of this group with the metal was shown with higher absorption intensities [25]; also, they could be related to the P=O stretch of organophosphorus compounds such as phytates [66]. FTIR analysis also showed the absorption bands of C–H bending vibrations at 1352 cm⁻¹ of alkyl, aliphatic and aromatic groups from cellulose, hemicellulose and lignin [29]. In addition, the peak at 1600 cm⁻¹ could be attributed to the aromatic C=C bonds, such as those in glucosinolates, and showed a small shift after lead sorption to a lower wavenumber [25]. The peak at 2335 cm⁻¹ was attributed to N–H stretching vibrations or to C=O stretching vibrations where oxygen or nitrogen could be involved in Pb(II) sorption, increasing the intensities of the IR spectrum after metal exposure [67].

Figure 3 shows SEM micrographs of broccoli stalk biomass before and after Pb(II) ions sorption and Figure 4 includes the EDX analysis. The SEM analysis revealed the presence of particles with irregular sizes and non-rounded shapes, featuring fibrous structures with internal voids. After the metal sorption by the biomass, a slightly degraded and fluffy surface was observed. This degradation can be attributed to the erosion caused by solution during agitation, likely associated with the hydrolysis of natural fibers, the exchange of Na ions from the buffer solution, and the interactions with the Pb(II) ions. The weight percentages of the chemical composition of the raw biomass were as follows: 45.06% of C, 40.84% of O, 6.06% of K, 3.46% of N, 2.18% of Cl, 1.13% of Na, 0.38% of S, 0.33% of Ca, 0.25% of P, 0.2% of Mg and 0.11% of Cu (Figure 4a), according to the nature of broccoli stalks described in the introduction (e.g., sulfur-, nitrogen- and phosphorus-containing organic compounds (glucosinolates, proteins, phytic acid), minerals (Ca, K, Mg, P), etc.) [68]. In the same sense, the O/C ratio of 0.91 was closed to the unit, showing good hydrophilic properties of the biomass due to carboxylic, alcohols, ethers and aldehydes groups with a high degree of chemical functionality. This value can be provided from broccoli stalk components such as cellulose, hemicellulose, starch and pectin with an O/C ratio of 0.83, sugars (fructose, sucrose, mannose…) with O/C ratio of 1, or different carboxylic acids (lactic, gallic, malic, citric, succinic…) with an O/C of ≈1. The lower content of lignin in broccoli biomass is consistent with the lower O/C ratio of 0.35 described in the literature for this organic polymer [69,70]. These results suggest the adequate use of raw broccoli stalk biomass to produce chemical sorbents for ions in aqueous solutions rather than to produce bioenergy. The N/S ratio of 9.1 obtained was similar to that found in crops of Brassica species and associated with the glucosinolates concentration when fertilization treatments were used [71]. The comparison of the elemental analysis before and after the sorption of Pb(II) ions showed the presence of the loaded metal in the biomass (see spectra in Figure 4) and the effect of the presence of Na(I) ions from the acetate buffer. K(I) ions were exchanged by Na(I) ions, increasing the Na(I) weight % from 1.13 to 5.83 and decreasing the K(I) weight % from 6.06 to 1.33. A slight exchange of Mg(II) and Ca(II) ions by Pb(II) could also take place, resulting in the decrease in the weight % from 0.2 to 0.01 and from 0.33 to 0.25, respectively.

The iodine capacity (I, mg g⁻¹) can provide significant information about the surface of powder materials. It is defined as the mass of iodine in mg that is consumed by grams of a chemical substance and is often used to determine the amount of unsaturation in the form of double bonds; consequently, the higher the iodine capacity, the higher the number of C=C bonds present. In the present work, the iodine capacity of the biomass was found to be 619.7 ± 33.3 mg g⁻¹ (Table 1). This result is higher than that of other biosorbents used for Pb(II) removal, such as coconut-shell-activated carbon (334.2 mg g⁻¹) [72], pine cone powder (23.7 mg g⁻¹) [40] or rice husk (68 mg g⁻¹) [73], showing the great potential of the Calabrese broccoli stalk biomass as a sorbent.

An intermediate CEC value of 30.7 ± 0.01 cmol kg⁻¹ was found for broccoli biomass, being suitable for the sorption of divalent ions by exchange. This value is comparable to the CEC range of 11.7–34.8 cmol kg⁻¹ described for agricultural biochar [74]. Higher values can be found for other biomass such as green compost (67.8 cmol kg⁻¹) and peat (84.6 cmol kg⁻¹) [75] with significant exchange properties for Pb(II). Lower values can also be found for biochar from mushroom compost (5.76 cmol kg⁻¹) [76] or for woody biomass (1.3–3.0 cmol kg⁻¹) with lower mineral contents [74], indicating the minor role of ion exchange during the metal adsorption.

Regarding the functional groups on the biomass surface, the results showed that the acid groups (ag: 1.257 ± 0.012 mmol g⁻¹) were slightly higher than the basic ones (bg: 1.085 ± 0.021 mmol g⁻¹) (Table 1). Carboxylic groups (1.216 ± 0.012 mmol g⁻¹) were predominant compared to phenolic (0.035 ± 0.013 mmol g⁻¹) and lactonic groups (0.006 ± 0.013 mmol g⁻¹). Other biomasses recently used for Pb removal showed higher values for acid groups such as blue-green algae (ag: 0.52/bg: 0.02 mmol g⁻¹) [77], corncob-activated carbon (ag: 1.22/bg: 0.57 mmol g⁻¹) [78], olive stone (ag: 0.96/bg: 0.41 mmol g⁻¹), pine nut shell biochar (ag: 0.206–0.266/bg: 0.020–0.029 mmol g⁻¹) [79] or nanche stone (ag: 0.1037/bg: 0.046 mmol g⁻¹) [80]. The carboxylic groups are Pearson hard basic sites on the biomass surface, while Pb(II) is a borderline softer acid ion [81]. The biomass–metal ionic electrostatic interaction with these groups might not be the main one, and the Pb(II) ions could also show covalent complexation, cation–π interactions [74,82,83] or other biosorption mechanisms. The large configuration of electron cloud ([Xe]6s²4f¹⁴5d¹⁰) of Pb(II), with a more diffuse electron distribution, increases the probability and the intensity of interactions between the ions and the sorbent, facilitating the sorption processes and explaining this type of interactions [83]. In this sense, it has been reported that Pb(II) was readily bound by cation–π interactions, attributing the adsorption to electron donor ligands of aromatic functional groups such as the content found within the broccoli biomass [84,85].

In relation to the functional groups, the pH of the biomass was found to be 6.88 ± 0.02, the pH_pzc value was 6.25 ± 0.02 (data in Table S5 of Supplementary Material), and the conductivity of the biomass was 1566 ± 22 µS cm⁻¹ (Table 1). These pH values have a great importance in metal sorption because they determine the ionization of the chemically active sites in the sorbent and the charge of the sorbent surface during sorption processes, affecting the sorption mechanism. Its quantification is of great importance when the main process is physisorption and provides information on the affinity of the biomass for cationic or anionic species depending on the pH of the liquid phase [86]. Thus, at solution pH higher than pH_pzc = 6.25, the biomass surface will be negatively charged, facilitating the interaction with positive species, while at pH lower than this value, the behaviour of the solid surface will be the opposite. Lower or similar values can be found in the literature for other biomasses used for Pb biosorption: blue-green algae (pH_pzc 1.3) [77], corncob-activated carbon (pH_pzc 3.8) [78], nanche stone (pH_pzc 6) [80], mixture of coffee grounds and orange barks residues (pH_pzc 5.2) [87], olive stone (pH_pzc 6.61) [88] or untreated and alkaline-treated apricot shell (pH_pzc 4.9 and 5.7, respectively) [89].

Other physico-chemical parameters such as bulk density (BD) and apparent density (AD) were measured, with values of 0.52 ± 0.01 g mL⁻¹ and 0.65 ± 0.03 g mL⁻¹, respectively (Table 1), with a porosity of 20.6 ± 3.2% and a specific surface area (SSA) of 15.3 ± 2.7 m² g⁻¹. These values were comparable with those of other biomasses found in the literature: mixture of coffee grounds and orange barks residues (BD = 0.43 g mL⁻¹) [87], untreated and alkaline-treated apricot shell (BD = 1.15 g mL⁻¹ and 1.10 g mL⁻¹; AD = 1.50 g mL⁻¹ and 1.58 g mL⁻¹; SSA = 15.4 m² g⁻¹ and 20.0 m² g⁻¹, respectively) [89], green compost (BD = 0.38 g mL⁻¹) and peat (BD = 0.25 g mL⁻¹) [75] or olive leave biomass (BD = 0.25 g mL⁻¹) [90]. As defined by American Water Work association, bulk density of biomass higher than 0.25 g mL⁻¹ is adequate for metal removal [91]; thus, the BD for broccoli was adequate for this purpose. Additionally, the comparison of the SSA of some biochars derived from agricultural biomass such as: corncob (SSA = 53.71 m² g⁻¹), rice husk (SSA = 51.39 m² g⁻¹) or wheat straw (130.14 m² g⁻¹) [92,93], showed higher values for the surface area, although these biochars require greater pretreatments. The physical adsorption process is mainly affected by porosity and specific surface area which provide more adsorption sites. Therefore, it seems unlikely that the mechanism by which broccoli removes Pb(II) is mainly based on physical adsorption if it has a lower number of voids [62].

3.3. Metal Removal Experiments

3.3.1. Effect of Ratio Mass of Biosorbent/Volume of Pb(II) Solution

In order to reduce the amount of biomass used in the sorption experiments, studies on the ratio of mass of biosorbent/volume of aqueous metal solution were carried out in the range of 5–10 g L⁻¹. The yield obtained with the lowest ratio was 67.9 ± 0.8%, while the ratio of 10 g L⁻¹ yielded 88.5 ± 0.5% after 24 h of exposure (Figure S1 of Supplementary Material). Therefore, a ratio of 10 g L⁻¹ was used for further experiments, i.e., 0.5 g of biomass were used in 50 mL of aqueous solution loaded with Pb(II).

3.3.2. Effect of pH

The effect of the solution pH on the effectiveness of Pb(II) ions removal from aqueous solutions using broccoli biomass as sorbent was studied at pH values of 3, 4.8 and 7. Higher pH values were not studied due to the rapid precipitation of the lead-loaded solution. Lower pH values could lead to the breaking of some biomass molecules. The results for 24 h were 85.8 ± 2.6, 88.5 ± 0.5, and 40.8 ± 1.2, respectively (Figure S1 of Supplementary Material). The pH dependence observed suggested that the removal of Pb(II) ions was favoured by a slightly acid pH value of 4.8.

The influence of pH on the sorption of metal ions is complex and depends mainly on the acidic properties of the cation and the acid-base character of the biomass surface. Concerning the cation, the pH speciation of Pb(II) indicates Pb(OH)₂ as the dominant species at pH > 5 and Pb(OH)⁺ at pH < 5. Thus, precipitation processes occur at pH values higher than 5, and the insoluble Pb(OH)₂ particles may be deposited from the bulk solution; these particles may also block the biomass pores and be absorbed [65,75]. The values obtained in the broccoli experiments were in agreement with the Pb(II) speciation, suggesting that it is more appropriate not to perform the experiments at cation precipitation values.

Related to the broccoli biomass, the value of pH_pzc determines when the net charge is 0. Below this value, it is positive, and the H⁺ ions competes for the active sites with Pb(II) ions, thereby protonating the oxygen-containing organic groups (carboxylate, phenolic hydroxyl groups, glucosinolates, phytates, etc.) leading to repulsion which increases as the pH decreases [94]. The pH_pzc value of 6.25 for broccoli biomass was higher than the pH of 4.8, at which the net charge of the biomass was positive. Nevertheless, the sorption was satisfactory. This phenomenon has been previously described in the literature and can be explained by considering other sorption mechanisms. The adsorption mechanisms of Pb(II) include: ion exchange (with negative surface of biomass), ligand exchange (with group providing hydrogen bonding such –OH group (phenolic hydroxyl)), precipitation (depending on the metal speciation), complexation (with electron donor elements in the complexing groups), cation–π interactions (with electron donor functional groups with delocalized π-electrons), physical sorption on surface sites (contact adsorption) and intraparticle diffusion [62,85,93]. Therefore, based on the pH of the aqueous solution, both ion exchange and physical electrostatic adsorption were mechanisms that were less favoured using broccoli biomass at pH 4.8, as already mentioned in Section 3.2. Microprecipitation on the biomass surface could also occur given the proximity to the pH value of 5.

3.3.3. Effect of Sorption Time and Initial Concentration of Pb(II) on the Removal

The removal of Pb(II) at different concentrations of 0.25, 0.5, 1, and 1.5 mmol L⁻¹ was studied over time. The experiments were conducted under the conditions mentioned in Section 2.4, at pH 4.8 and 23 °C, using 0.5 g of biomass (particle sizes: 0.125 mm < x < 0.249 mm) in 50 mL of aqueous solution. The percentage of Pb(II) removal was calculated with Equation (2) and the experimental data obtained are shown in Figure 5. It can be observed that Pb(II) removal increased rapidly with time during the initial intervals, followed by a slower increase until equilibrium was reached. This effect decreased at higher concentrations, requiring more time to reach equilibrium. The removal efficiency does not seem to increase with the increasing concentration gradient from the solution to the sorbent, as occurs mainly in physical processes, which rely more on diffusion. Equilibrium was reached after 6–8 h for all initial concentrations, making it unnecessary to extend the removal times. The percentage of removal at equilibrium was similar in the range of 0.25–1 mmol L⁻¹ Pb(II), with an average of 88.1 ± 1% after 6 h and 88.3 ± 0.8% after 8 h, with lower yields for 1.5 mmol L⁻¹ Pb(II) of 83.2 ± 0.1% after 6 h and 83.4 ± 0.3% after 8 h. When the Pb(II) concentration was too high, this efficiency was slightly lower, probably due to the saturation of the binding sites on the biomass. The precision of the method was evaluated using a concentration of 1 mmol L⁻¹ of Pb(II) by six replicate measurements at a confidence level of 95%, and was found to be 0.23%.

3.4. Mechanisms of the Pb(II) Sorption

3.4.1. Sorption Isotherm

Adsorption isotherm models are used as a tool to elucidate the interactions between sorbents and target substances. An isotherm describes the relationship between the amount of substance sorbed per unit mass of sorbent and the remaining concentration in the aqueous phase at equilibrium, under controlled temperature conditions. Experimental data on this relationship and their fitting to different predictive models can be used to define the nature of the interactions in the sorption process. The mechanisms of adsorption, sorption favourability, and sorbate–sorbent affinity can be explained by the isotherm models [95]. Four models were applied by using the equations described in the Supplementary Material (Table S2). The constants and determination coefficients of the models, calculated from the linear plot, are shown in

Table 2. Results of the fits of parameters and determination coefficients for the different isotherm models.

Isotherm Models	Isotherm Constants and Determination Coefficients a
	Parameter	Values at 23 °C
Langmuir	KL (L mg−1//L mmol−1)	0.013//2.78
	qmax (mg g−1//mmol g−1)	64.9//0.31
	RL	0.19–0.62
	R2	0.873
	MSE	0.023
Freundlich	KF (mg g−1//mmol g−1)	1.11//0.005
	1/n	0.82
	R2	0.969
	MSE	0.005
Dubinin–Radushkevich	B (mol2 J−2)	7.28 × 10−9
	qmax (mg g−1//mmol g−1)	587//2.83
	E (kJ mol−1)	8.29
	R2	0.981
	MSE	0.018
Temkin	B	9.93
	KT (L mg−1//L mmol−1)	0.24//50.4
	bT (J mol−1)	248
	R2	0.980
	MSE	2.69

^a The model equations and parameters are defined in Table S2.

at the equilibrium time of 8 h.

The Langmuir isotherm is widely used and applies to the monolayer sorption process on a surface with a finite number of identical sites and equivalent sorption energies. Once the sites are occupied by the sorbate, no further sorption can take place. Additionally, it assumes that there are no interactions between the adsorbed species [51,96]. K_L is the Langmuir isotherm constant and q_max is the maximum adsorption capacity. The model predicts the suitability of the sorption process as a function of the values of the dimensionless constant separation factor R_L, which depends on K_L (Table S2). The sorption is favoured within the range of 0 (more favourable) to 1 (less favourable), while sorption processes with out-of-range values are not suitable. In this work, the fit of the Langmuir isotherm provided the lowest R² value (R₂ = 0.873) among the four models applied for the Pb(II) removal experiments. The values of R_L varied significantly depending on the initial concentration of Pb(II): 0.62 (for C₀ = 0.25 mmol L⁻¹); 0.39 (for C₀ = 0.5 mmol L⁻¹); 0.33 (for C₀ = 1 mmol L⁻¹); and 0.19 (for C₀ = 1.5 mmol L⁻¹), indicating that the Pb(II) sorption with broccoli biomass was suitable at all concentrations and more favourable at higher concentrations. The calculated q_max was significant, with a value of 64.9 mg g⁻¹ compared to other similar Langmuir values for stalk or stem biomass with Pb (II) (corn stalks (biochar): 40.98 mg g⁻¹ [97]; grape stalks: 37.7 mg g⁻¹ [98]; rooibos shoot: 19.8 mg g⁻¹ [99]; and tobacco stems: 5.54 mg g⁻¹ [100]).

The Freundlich isotherm model assumes multilayer adsorption on the heterogeneous sorbent surface characterised by non-uniform sorption energies. The Freundlich isotherm constant K_F is an indicator of adsorption capacity related to the surface heterogeneity and activity. Additionally, the parameter 1/n indicates the strength of adsorption: the smaller the 1/n value, the greater the heterogeneity. If the value of 1/n is less than 1, it indicates normal adsorption. Values of n in the range of 1 < n < 10 indicate good adsorption [21]. From the plot of experimental data using the Freundlich model, the value of K_F was 1.11 mg g⁻¹ and 1/n was 0.823 with a good coefficient of determination (R² = 0.969) (Table S2), indicating a favourable Pb(II) sorption on the heterogeneous surface of the broccoli biomass. These results were similar to those determined for some stalk or stem biomass mentioned above, such as rooibos shoot: K_F = 1.47 mg g⁻¹ and 1/n = 0.34 [99], and tobacco stems: K_F = 0.95 mg g⁻¹ and 1/n = 0.723 [100]. The values for corn stalks (biochar): K_F = 16.44 mg g⁻¹ and 1/n = 7.25 [97] did not result in favourable sorption, and the Freundlich model was not studied for grape stalks [98].

The Langmuir and Freundlich models can be used to determine whether the adsorption process is monolayer or multilayer, but they do not really describe the nature of the sorption. The Dubinin–Radushkevich model (D-R) can be used for this purpose as it allows the determination of the type of adsorption by calculating the average free energy of adsorption (E) per mole of sorbate. Energy values in the range of 8–16 kJ mol⁻¹ are indicative of chemical sorption, whereas values below 8 kJ mol⁻¹ suggest physical sorption [95]. The study of the experimental data with broccoli biomass using the D-R model yielded a very good fit (R² = 0.981) (Table S2), similar to that obtained with the Freundlich model. The value of q_max was significantly high (q_max = 586.7 mg g⁻¹) and the nature of the sorption was defined as chemisorption based on the calculated E value (E = 8.29 kJ mol⁻¹). No data were found for other stalk or stem biomasses mentioned above using the D-R model.

By ignoring extremely low and high concentration values, the Temkin isotherm considers the fact that the heat of sorption for all molecules on the sorbent surface decreases linearly with increasing coverage due to interactions between the sorbent and the sorbate. It also suggests that sorption is defined by a uniform distribution of binding energies up to a maximum binding energy [51,95,101]. K_T is the equilibrium binding constant, corresponding to the maximum binding energy, and it indicates the strength of the interaction. b_T is the Temkin isotherm constant and is related to the heat of sorption. Although the Temkin equation is more suitable for predicting gas phase equilibria [102], a determination coefficient was obtained with the experimental data of Pb(II) sorption using broccoli biomass (R² = 0.980) (Table S2). This value was as good as that obtained for the Freundlich and D-R models. However, the value of the MSE (mean square error) was significantly higher than the values for other isotherms (MSE = 2.69 compared with MSE values of 0.005 and 0.018 for the Freundlich and D-R fits, respectively). This discrepancy can be attributed to the complexity of the liquid–solid interphase when different mechanisms coexist, which is not assumed by the Temkin model. For this reason, the Freundlich and D-R models provide better explanations for Pb(II) sorption with broccoli biomass.

It can be concluded that the Freundlich and D-R models can better explain the favourable Pb(II) sorption on the heterogeneous surface of the broccoli biomass, with different possible interactions and binding energy, preferably by chemical mechanisms and with a very high q_max of 586.7 mg g⁻¹. Based on the percentage of Pb(II) removal (Figure 3), the adsorption process did not encounter any resistance with a substantial adsorption force, demonstrating rapid kinetics at beginning of the process and reaching a high average removal of 88.3 ± 0.8% after 8 h for the Pb(II) concentrations range of 0.25–1 mmol L⁻¹. Increasing the number of adsorbed subsequent layers resulted in only a slightly decrease in the adsorption force, particularly notable for the highest Pb(II) concentrations of 1.5 mg L⁻¹, with an efficiency of 83.4 ± 0.3% after 8 h.

3.4.2. Sorption Kinetics

Kinetic models allow for the determination of the sorbate removal rate from solution by the biosorbent, as well as the elucidation of the mechanisms of the process. The kinetic behaviour of Pb(II) removal on broccoli biomass was investigated by six common models. The experimental data obtained for several concentrations of Pb(II) (0.25, 0.5, 1 and 1.5 mmol L⁻¹) were fitted to each model’s linear equation form. The parameters of these models were then calculated from the slopes and intercepts of the plots, as described in Table S3.

As can be seen in

Table 3. Results of the fits of the kinetic models, parameters and determination coefficients for different Pb(II) concentrations.

Kinetic Models	Kinetic Parameters and Determination Coefficients b
	Parameter a	0.25 mmol L−1	0.5 mmol L−1	1 mmol L−1	1.5 mmol L−1
Pseudo-first-order	k1 (min−1)	3.78 × 10−4	7.73 × 10−3	0.01	0.03
	qe (mg g−1//mmol g−1)	0.02//1.14 × 10−4	0.30//1.43 × 10−3	1.11//5.35 × 10−3	12.2//0.06
	qe exp (mg g−1//mmol g−1)	3.95//0.02	10.3//0.05	13.4//0.07	27.9//0.14
	R2	0.008	0.989	0.942	0.968
	MSE	0.05	1.43 × 10−3	0.03	0.05
Pseudo-second-order	k2 (g mg−1 min−1//mmol−1 min−1)	0.85//1.76 × 102	0.09	0.04//7.96	4.22 × 10−3//0.87
	qe (mg g−1//mmol g−1)	3.95//0.02	10.4//0.05	13.5//0.07	27.0//0.13
	qe exp (mg g−1//mmol g−1)	3.95//0.02	10.3//0.05	13.4//0.07	27.9//0.14
	h (mg g−1min−1//mmol g−1 min−1)	13.2//0.06	10.1//0.05	6.95//0.03	3.08//0.02
	R2	0.999	0.999	0.999	0.999
	MSE	0.03	1.86 × 10−3	2.11 × 10−3	0.03
Elovich	a (mg g−1 min−1)	non fitted	2.86 × 1048	1.26 × 1012	0.57
	b (g mg−1)		11.6	2.57	0.32
	R2		0.985	0.876	0.768
	MSE		2.10 × 10−4	0.05	5.55
Ritchie’s second order	k2_R (min−1)	1.00	0.88	1.19	3.55
	R2	0.056	0.965	0.755	0.751
	MSE	23.3 × 103	1.85 × 102	1.05 × 104	5.66 × 104
First order reversible	k1 (min−1)	0.02	0.03	0.02	0.03
	k−1 (min−1)	3.34 × 10−3	3.20 × 10−3	2.73 × 10−3	6.71 × 10−3
	R2	R2 < 0	R2 < 0	R2 < 0	0.947
	MSE	13.3	6.83	3.78	0.15
Intraparticle diffusion	Kd (mg g−1 min−1/2)	9.0 × 10−5	0.02	0.07	0.50
	C (mg g−1)	3.94	10.0	12.1	17.7
	R2	0.0007	0.9417	0.712	0.582
	MSE	6.74 × 10−4	8.19 × 10−4	0.11	10.0

^a The model equations and parameters are defined in Table S3. ^b R² < 0: the model is not able to adequately explain the adsorption behaviour observed in the experimental data.

, the pseudo-second-order model proved to be the most effective in describing the sorption of Pb(II), with very good R² values for all concentrations of Pb(II). In addition, the fitted equilibrium adsorption capacities (q_e (0.25 mg g⁻¹) = 3.95 mg g⁻¹; (q_e (0.5 mg g⁻¹) = 10.4 mg g⁻¹; (q_e (1 mg g⁻¹) = 13.5 mg g⁻¹; (q_e (1.5 mg g⁻¹) = 27.0 mg g⁻¹) were very close to the experimental data (q_e exp (0.25 mg g⁻¹) = 3.95 mg g⁻¹; (q_e exp (0.5 mg g⁻¹) = 10.3 mg g⁻¹; (q_e exp (1 mg g⁻¹) = 13.4 mg g⁻¹; (q_e exp (1.5 mg g⁻¹) = 27.9 mg g⁻¹). The suitability of the pseudo-second-order kinetic was further demonstrated by the plot of t/q_t versus t, showing a linear relationship (Figure 6). The equilibrium sorption capacity (q_e) of the broccoli biomass increased when increasing the initial Pb(II) concentration from 0.25 to 1.5 mmol L⁻¹ as the slope (1/q_e) decreased with the concentration (Figure 6). This increase indicated favourable interactions between the Pb(II) ions and the sites on the biomass. These results suggest that the adsorption processes of Pb²⁺ onto broccoli biomass may involve chemical interactions with electron sharing or exchange [103]. Previous studies (Table 1) have indicated that the CEC of broccoli biomass was intermediate, implying that ion exchange plays a less significant role. Therefore, interactions such as ligand exchange (with –OH group of the polysaccharide matrix), complexation with key donor elements (e.g., O, S, N, P) found in main compounds of broccoli stalks (e.g., cellulose, hemicellulose, glucosinolates, phytates, amino acids), and cation–π interactions (with aromatic functional groups) appear to be the main mechanisms controlling the Pb(II) sorption [104,105]. In the literature, the frequency of the different binding groups on biomass surface involved in the complexation of metal ions has been reported: carboxylates: 40.8%, aromatic ring: 15.5%, hydroxyl: 15.5%, amine: 12.7%, phosphate: 4.2%, carbonyl: 4.2%, thiol: 2.8%, amide: 2.8%, and sulfonate 1.4%. Pb ranks among the top heavy metals of concern for water pollution when complexed by ligands with S and –COOH groups, such as some of the broccoli biomass, due to the Pb(II) configuration and Pearson properties [106].

Other models provided good fits but not for all Pb(II) concentrations, which could be related to other concurrent sorption mechanisms that varied in effectiveness depending on the concentration. In this sense, the pseudo-first-order model exhibited good R₂ values except for the lowest Pb(II) concentration of 0.25 mmol L⁻¹. In addition, the predicted values of q_e differed significantly from the experimental values. This could also indicate that physical sorption was less efficient than chemical interactions.

3.4.3. Sorption Thermodynamic Studies

Thermodynamic studies are used to investigate the effect of temperature on the sorption capacity and to understand whether the sorption is spontaneous or not. The standard Gibbs free energy change (ΔG°), the standard enthalpy change (ΔH°), the standard entropy change (ΔS°), the activation energy (E_a) and the sticking probability (S*) are thermodynamic parameters that can be obtained by the equations detailed in Table S4 [57,107]. The values of enthalpy and entropy variation were calculated using the graphical representation of ln K_c vs. 1/T, being (−ΔH°/R) the slope of the curve and (ΔS°/R) the ordinate at the origin. From the modified Arrhenius and its subsequent linearization, the parameters E_a and S* could be determined. Thus, the representation of ln (1 − θ) vs. 1/T allowed the calculation of E_a/R as the slope of the curve and ln S* as the ordinate at the origin.

Biosorption experiments were carried out as described in Section 2.7 to evaluate the thermodynamic properties, and the results are presented in

Table 4. Thermodynamic parameters ^a for the Pb(II) sorption onto broccoli Calabresse biosorbent (50 mL of 1 mmol L⁻¹ Pb(II) solution at pH 4.8; biosorbent mass = 0.5 g).

Temperature (°C/K)	ΔG° (kJ mol−1)	ΔH° (kJ mol−1)	ΔS° (J mol−1 K−1)	S*	Ea (kJ mol−1)
18/291	−4.17	8.98	45.20	0.006	7.66
23/296	−4.40
30/303	−4.72

^a The model equations and parameters are defined in Table S4.

. The values of enthalpy (ΔH°) and entropy (ΔS°) were determined to be 8.98 kJ mol⁻¹ and 45.20 J mol⁻¹ K⁻¹, respectively. The positive value of ΔH° was consistent with the endothermic nature of the sorption process and slightly exceeded the indicative energy value for physical sorption (<8 kJ mol⁻¹). The positive value of ΔS° suggested an increase in randomness at the solid–liquid interface, while a low value of this parameter indicated that not remarkable change in entropy occurred. The values of the standard Gibbs energy change were calculated and found to be −4.17, −4.40 and −4.72 kJ mol⁻¹ at 18 °C, 23 °C and 30 °C, respectively. The negative values of ΔG° increased with increasing temperature, indicating the spontaneity of the sorption process and that the process was favoured by the increase in temperature.

The value of the activation energy (E_a) was 7.66 kJ mol⁻¹, which was in agreement with the positive value of ΔH° for the endothermic Pb(II) sorption. Understanding activation energy provides insight into the nature of sorption. Physisorption entails weak forces, with activation energy values below 4.2 kJ mol⁻¹. Conversely, chemisorption involves stronger and specific forces compared to physical adsorption, with activation energy exceeding this value [108]. The activation energy (E_a) value determined in this study indicates that the sorption of Pb onto the broccoli biosorbent is predominantly a chemisorption process. The sticking probability (S*) was 0.006, a very low value. The interpretation of this parameter can be described as follows: (a) S* > 1: sorbate unsticking to sorbent—no sorption; (b) S* = 1: linear sticking relationship between sorbate and sorbent—possible mixture of physical and chemical sorption; (c) S* = 0: indefinite sticking of sorbate to sorbent—predominance of chemical sorption; (d) 0 < S* < 1: favourable sticking of sorbate to sorbent—predominance of physical sorption [109]. The obtained value of S* suggests that the sorption of Pb(II) was governed by a mechanism on the borderline between chemisorption and physisorption. This value, being very close to zero, highlights the significance of chemical sorption, and also corroborates the findings from the kinetic study.

4. Conclusions

A low-cost and ecofriendly biosorbent obtained from Calabrese broccoli stalks showed significant potential for the removal of Pb(II) from aqueous solutions, exhibiting efficient adsorption performance without requiring complicated pretreatment processes. The electronic configuration of Pb(II) ions, along with the metal speciation at a pH of 4.8 in aqueous solution, and the nature of broccoli biomass, synergistically supported the sorption process. The Freundlich and Dubinin–Radushkevich models explained with good fits the favourable Pb(II) sorption on the heterogeneous surface of the broccoli biomass, with different possible interactions and binding energies, preferably by chemical mechanisms. The pseudo-second-order kinetic model was the best fit, further confirming the chemical nature of the sorption for the Pb(II) removal using broccoli biomass. These chemical mechanisms mainly involved complexation, ligand exchange, and cation–π interaction. Ion exchange and physical sorption occurred to a lesser extent, as did physical sorption. The pH of the solution could also induce microprecipitation on the biomass surface. The thermodynamic parameters evaluated showed the endothermic and spontaneous nature of the Pb(II) biosorption, and the sticking probability confirmed the borderline mechanism between chemisorption and physisorption. Hence, this novel biomass has proven to be a promising sorbent for the removal of Pb(II) ions from aqueous solutions.