Adsorption and Recovery Studies of Cadmium and Lead Ions Using Biowaste Adsorbents from Aqueous Solution

1. Introduction

Over the past few decades, water pollution caused by both natural processes and human activities has become a major concern [1]. Heavy metal contamination is one of the most significant forms of pollution which occurs when toxic metals and metalloids such as lead, mercury, cadmium, arsenic, chromium, and others are deposited into the environment [2]. While these metals naturally exist in trace amounts in the Earth's crust, human activities can release them at much higher concentrations, posing significant risks. Their ability to persist in the environment, bioaccumulate in living organisms, and cause toxic effects makes them a serious threat to ecosystems and human health [3].

From previous studies, biowaste sorbents, such as potato peels and pumpkin peels, showed great potential for adsorbing cadmium ions in aqueous solutions [4,5,6,7]. Adsorption of metal ions onto biosorbents provides a relatively low cost and efficient method to reduce these metal ions in aqueous medium. Biowaste sorbents offer numerous advantages for heavy metal ion adsorption, making them an attractive option for environmental remediation [8]. One of the most significant benefits is their cost-effectiveness, as biowaste materials such as fruit and vegetable peels are abundant and inexpensive compared to synthetic alternatives [9]. This makes them a viable option for large-scale water treatment applications [10]. The presences of functional groups on the surface of biowaste sorbents such as COOH, C=O, C-O, CONH, NH, and -OH, which can physically adsorb, ion-exchange, or chemically bind with metal ions, retaining them, particularly when these groups are above the iso-electric point [11,12]. Studies have provided evidence that chemical treatments can significantly enhance the adsorption capacity of biowaste sorbents by activating potential binding sites and minimizing structural interferences [13].

Luffa plants are well known for producing natural sponges. Luffa presents significant potential for various environmental applications due to its sustainable and biodegradable properties. The increasing interest in luffa for environmental technologies underscores its value as a cost-effective and sustainable solution for pollution mitigation and resource conservation [14,15]. A study showed that when the moisture percentage was minimized to 12.40% in luffa peels, the composition resulted in 38.94% carbohydrate, 14.26% protein, 6.10% fat, 20.60% fiber and 7.7% ash [16]. It was also shown that luffa peels contain phenolics flavonoids. oleanolic acid, carotenoid and chlorophyll [17]. While the inner parts of luffa have been the subject of various studies, limited research has been conducted on the adsorption capacities of luffa peels for different heavy metal ions. Only one study has shown that luffa peels are effective in adsorbing dyes and heavy metal ions, such as Pb²⁺ and Ni²⁺, from water on a small scale. The adsorption capacity is attributed to the presence of -OH and -COOH functional groups, which facilitate the binding of these contaminants [18].

Chamomile is a flowering plant from the Asteraceae family, best known for its calming properties when used in teas or essential oils. Chamomile flower has a long history of use in medicine and health products, known for its wide range of pharmacological activities [19]. Studies have identified a variety of organic acids, flavonoids, coumarins, and volatile oils in chamomile, which are abundant in functional groups such as -COOH, -OH, C=C, and C-O. These functional groups are believed to play a crucial role in the herb's ability to absorb heavy metals [19,20]. A study demonstrated that nano chamomile waste is an effective material for use as a solid phase extractant for the removal of heavy metals. The extraction of ions such as Pb²⁺, Cu²⁺, and Fe³⁺ achieved a removal efficiency exceeding 98% [21].

This current research focused on investigating the adsorption capacity of cadmium and lead ions in water onto luffa peels and compared them to chamomile flowers, aiming to find alternative biosorbents for heavy metal ion contaminated wastewater treatment. Efforts were also extended to evaluate the recovery of these metal ions from aqueous solutions with luffa peels.

2. Materials and Methods

2.1. Chemicals and Reagents

All solutions were made and diluted with Ultrapure (UP) deionized water obtained from a Direct-Q 3 UV-R purification system with a product resistivity of 18.2 M Ω·cm at 25 °C.

2.1.1. Metal Ion Solutions

Cadmium sulfate octahydrate with a purity of 99% was purchased from JT Baker Chemicals. A stock solution of 1000 ppm of cadmium ion solution was prepared by adding 2.282 g of cadmium sulfate octahydrate in a 1L volumetric flask with UP deionized water. Using the stock solution, serial dilutions were made to make 500, 250, 100, 50, 20 and 10 ppm of cadmium ion solutions.

Lead (II) nitrate crystals with a purity of 99% were purchased from JT Baker Chemicals. A stock solution containing 1000 ppm of lead ions was prepared by adding 1.598 g of lead (II) nitrate to a 1L volumetric flask with UP deionized water. Serial dilutions were followed to make 500, 250, 100, 50, 20 and 10 ppm solutions of lead ions.

2.1.2. Buffer Solutions

ACS grade glacial acetic acid and solid sodium acetate were purchased from JT Baker Chemicals. To prepare a 0.1 M acetic acid/acetate buffer solution, 800 mL of UP deionized water was added to a 1-liter volumetric flask. Then, 340 microliters of glacial acetic acid and 7.7 grams of sodium acetate were added to the volumetric flask and mixed thoroughly. The pH was measured with an Orion pH electrode connected to Thermo Scientific Orion STAR A112 pH-meter and adjusted accordingly using either acetic acid or sodium acetate to reach a pH of 5.6. UP deionized water was finally added for the volumetric flask to reach 1 liter and solution was mixed thoroughly.

Trizma base (Tris(hydroxymethyl)aminomethane) was purchased from Millipore Sigma. The four Trizma buffers were prepared separately in a similar manner to the acetic acid/acetate buffer by adding 800 mL of UP deionized water to 1-liter volumetric flasks. Then, 12 grams of Trizma base were added to each volumetric flask. The pH was measured again and adjusted accordingly with nitric acid to reach a final pH of 6.7, 7.0, 7.5, and 8.0. UP deionized water was then added to each volumetric flask to reach 1 liter. These buffers were used only for cadmium solutions, as the lead solutions precipitated at any pH above 7. For lead solutions, the acetic acid/acetate buffer was used for all measurements.

2.1.3. Nitric Acid for ICP-OES

Ultra-pure nitric acid (67%–70%) was purchased from JT Baker ULTREX^® II ultrapure reagents. A 4% nitric acid stock solution was prepared by adding 800 mL of UP deionized water to a 1-liter beaker. Then, 29 mL of 67%–70% nitric acid was added slowly to the beaker with constant stirring. The solution was transferred to a 1-liter volumetric flask, and UP deionized water was then added to the volumetric flask to reach 1 liter. This 4% stock solution was mixed with every sample before conducting the ICP-OES.

2.2. Instrumentation

2.2.1. ICP-OES

Cadmium and lead ion concentrations after biosorbent adsorption were all measured using Thermo Scientific™ iCAP™ 7200 ICP-OES, inductively coupled plasma atomic emission spectroscopy. The wavelengths used for cadmium and lead measurements were 214.438 nm and 220.353 nm both in radial mode. Calibration curves were obtained every time before analysis with 1, 5, 10, 25, 50, 75, 100, 125 ppm of both Cd²⁺ and Pb²⁺ solution to maximize the accuracy. All samples were mixed with 4% nitric acid ²⁺1:1 ratio one day before performing the experimentation. Linear correlation curves were obtained every day in use with correlation coefficient R² equal to 1 for both ions in the range of 1 to 125 ppm.

2.2.2. FTIR

FTIR analysis was performed using a Perkin Elmer Spectrum 100 FTIR spectrometer (LR 64912C) with a diamond-attenuated total reflectance (ATR) accessory. Measurements were taken with 4 scans at 1 cm⁻¹ resolutions and were recorded within the 4000–650 cm⁻¹ region. A background measurement of air was taken before every first measurement of the day. In between each measurement, the sample compartment was cleaned using an acetone-dipped Q-tip and left to dry for at least 30 seconds.

2.3. Biosorbents Treaments

Luffa and chamomile flowers were ordinary kinds obtained from New York City stores. For luffa, the peels were collected, rinsed with tap water, then with distilled water several times and air dried for three days at ambient conditions. The dried luffa peels and chamomile flowers were hand-milled in a mortar and pestle and sieved through an 18-mesh sieve. To find the best adsorption conditions, biosorbents were treated with acid or base. Their treatment was done by adding 6.0 g of the biosorbent to 200 mL of 0.4 M HNO₃ or NaOH. The mixture was left for 24 hours, and then filtered with filter paper and washed with UP deionized water until the filtrate was clear before they were collected. The biosorbents were left to air dry in the laboratory for 3 to 5 days before use.

2.4. Batch Experiments

2.4.1. Equilibrium Adsorption

The equilibrium study was performed under isothermal conditions at 20 ± 2 °C for 24 hours using a batch equilibration procedure. Each biosorbent (0.100 g) was treated with 10 mL of Cd²⁺ and Pb²⁺ solutions at 10, 20, 50, 100, 250, and 500 ppm, mixed with 10 mL of acetic buffer at pH 5.6 for both ions and Tris buffer at pH 8.0 only for Cd²⁺, in 20 mL scintillation vials with stirring for 24 hours. The solution was then filtered with filter paper followed by a Millipore syringe filter. Two milliliters of the filtered solution was mixed with 2 mL of 4% HNO₃ for ICP-OES analysis. The biosorbents were collected for FTIR analysis. Three experimental sets were conducted to obtain average values.

The adsorption efficiency (A%) was calculated using the initial concentration introduced into the solution (C₀) and the equilibrium concentration (C_eq), as shown in Equation (1) below. The mass ratio between the amount of adsorbed species and the amount of adsorbent (Q_e) can be calculated as shown in Equation (2) below.

$\mathrm{\%}A={\displaystyle \frac{C_0-C_{eq}}{C_0}}$ (1)

$Q_e={\displaystyle \frac{\left(C_0-C_{eq}\right)V}{M}}$ (2)

The Freundlich isotherm model is an empirical model to analyze the relationship between the concentration (C_e) of an adsorbate in solution and Q_e onto the heterogenic surface at equilibrium. At low-solute concentrations, the amount of adsorbate adsorbed increased with an increase in solution concentration. The non-linear form of the Freundlich isotherm equation is shown in Equation (3) and the linear form in Equation (4) below [22,23].

$\left(Q_e\right)=K_f\times C_{eq}^{1/n}$ (3)

$log\left(Q_e\right)=log{K}_f+{\displaystyle \frac{1}{n}}log{C}_{eq}$ (4)

By plotting Log(Q_e) vs. LogC_eq, a straight line is expected. Correlation coefficient R² will be used to evaluate the significant linearity. The K_f constant is given by the intercept, and the n value is calculated from the slope of the straight line. Large K_f and 1/n value favor the adsorption with this model. The definitions of the constant and all the symbols used are listed at the end.

The Langmuir isotherm model is also used as an empirical model for equilibrium adsorption studies. It enables the determination of the maximum monolayer adsorption capacity (Q^o_max) for an adsorbent. The nonlinear and linear forms of the Langmuir isotherm are represented by Equations (5) and (6) below [24].

$Q_e={\displaystyle \frac{Q_{max}^o{K}_L{C}_{eq}}{1+K_L{C}_{eq}}}$ (5)

${\displaystyle \frac{C_{eq}}{Q_e}}={\displaystyle \frac{1}{Q_{max}^o{K}_L}}+{\displaystyle \frac{1}{Q_{max}^o}}{C}_{eq}$ (6)

Here Ceq/Q_e will be plotted vs the C_eq to evaluate the linear fit, Q^o_max can be obtained from the slope of the curve, K_L then can be obtained from the intercept.

2.4.2. Adsorption Kinetics

Kinetic studies were performed in batch mode with 0.100 g of biosorbents and 20 mL of 250 ppm initial concentration of both metal ions mixed with buffer under constant stirring for 1, 2, 3, 5, 10, 15, 20, 30, and 60 minutes. The supernatants were immediately collected and filtered using a Millipore syringe filter. Three experimental sets were conducted, and the same manner was used for sample preparation for ICP-OES analysis as in the equilibrium studies. The adsorption kinetics can be fitted to the Pseudo-second order model, as shown in Equation (7) below [25].

${\displaystyle \frac{t}{Q_t}}={\displaystyle \frac{1}{K_2{Q^2}_e}}+{\displaystyle \frac{t}{Q_e}}$ (7)

with

$Q_t={\displaystyle \frac{\left(C_0-C_t\right)V}{M}}$ (8)

Here C_t is the concentration at time t of the adsorption, by plotting t/Q_t, a linear line is expected. The equilibrium Q_e can be calculated with the slope of the curve, the K₂ can be calculated with the intercept in equation (7).

2.5. Fixed Bed Column Adsorption and Recovery

Waste treatment—dynamic adsorption of cadmium and lead from aqueous solution was carried out as follows. Six grams of base-treated luffa peels were packed in two 300 mm x 30 mm flash chromatography columns with UP deionized water for each metal ion study. The column for cadmium adsorption and recovery was preconditioned with Tris buffer at pH 8.0, while the column for lead adsorption and recovery was preconditioned with acetic acid buffer at pH 5.6. Adsorption was performed by feeding solutions of 250 ppm of cadmium or lead mixed with buffer continuously at a draining speed of 4 mL/min. The draining solution was collected in 80 mL batches initially, or 20 mL batches after 320 mL, and monitored with ICP-OES. The process was stopped when the biosorbents reached 80% saturation. The dynamic breakthrough curve was evaluated with the Thomas model shown in Equation (9) and the linear form in Equation (10) below [22,26].

${\displaystyle \frac{C_t}{C_0}}={\displaystyle \frac{1}{1+{exp}^{(\frac{K_T{Q}_{0}M}{Q}-K_T{C}_{0}t)}}}$ (9)

$\ln \left({\displaystyle \frac{C_0}{C_t}}-1\right)={\displaystyle \frac{K_T{Q}_{0}M}{Q}}-K_T{C}_{0}t$(10)

The breakthrough capacity (Q_b) was obtained using Equation (11), shown below.

$Q_b={\displaystyle \frac{Q}{1000}}(C_0-{\displaystyle \frac{C_b}{2}}){\displaystyle \frac{t_b}{M}}$ (11)

The adsorbed cadmium and lead ions were then recovered by feeding 0.4 M HNO₃ at a draining speed of 4 mL/min until no heavy metal ions were present in the drained solution, monitored by ICP-OES. The biosorbents were then washed with UP deionized water until the draining solution reached a neutral pH. The adsorption-desorption processes were repeated three times to evaluate the reuse ability of the luffa peels. The recovery efficiency (R%) was obtained using Equation (12), shown below.

$R\mathrm{\%}=\left(1-{\displaystyle \frac{M_T-M_R}{M_T}}\right)\mathrm{\%}$ (12)

3. Results and Discussion

3.1. Biosorbent Characterization

The bulk density of untreated (natural form) luffa peels was measured to be 0.42 g/cm³. After chemical treatments, a slight increase in bulk density was observed, with the base-treated luffa peels (luffa-base) exhibiting the highest density. In contrast, the untreated chamomile flowers exhibited a lower bulk density of 0.29 g/cm³. Unlike luffa peels, both acid and base treatments reduced the bulk density of chamomile flowers, as shown in Table 1.

Figure 1 illustrates the FTIR spectra of untreated, acid-treated, and base-treated luffa peels. All sorbent types exhibit characteristic peaks corresponding to O-H stretching at 3294 cm⁻¹, C-H stretching at 2919.8 cm⁻¹, and C=O stretching at 1731.3 cm⁻¹, indicative of the presence of carboxylic acid groups. The base treatment resulted in a significant reduction in the C=O stretching intensity at 1731.3 cm⁻¹, and two additional peaks were observed at 1601.5 cm⁻¹ and 1412.3 cm⁻¹, corresponding to C-O stretching. These peaks indicate the full deprotonation of carboxylic acids and the formation of carboxylates (COO⁻). In contrast, the acid treatment led to full protonation of the carboxylates, resulting in an increase in C=O stretching intensity.

The FTIR analysis of chamomile flowers also indicated the presence of carboxylic acids and carboxylates after base treatment as illustrated in Figure 2. Both luffa peels and chamomile flowers possess similar active adsorption sites for heavy metal ions.

In an acidic environment, during the adsorption of heavy metal ions by carboxylic acids, the metal ions must compete with protons (H⁺) for binding sites. This competition causes ion exchange and deprotonation of carboxylic acids, resulting in the formation of carboxylate-metal complexes. In this case, Pb²⁺ exchanged protons, resulting a reduction of COOH and forming COO-Pb complexes. This is evidenced by the decrease in C=O stretching intensity at 1729.8 cm⁻¹ and appearance of deprotonated C-O stretching peaks at 1585.1 cm⁻¹ and 1414.9 cm⁻¹ at pH 2.8 as illustrated in Figure 3.

For base-treated sorbents, the deprotonation of most COOH groups is achieved through the treatment process. During the adsorption of heavy metal ions, the negative charge of the COO⁻ groups attract the metal ions, resulting in significantly less proton exchange. As shown in Figure 4, following the adsorption of Cd²⁺, the base-treated chamomile showed a decrease in C=O stretching intensity at 1727.7 cm⁻¹, attributed to ion exchange. Additionally, C-O stretching peaks at 1602.1 cm⁻¹ and 1410.6 cm⁻¹ are observed due to the adsorption of Cd²⁺.

The FTIR spectra of luffa peels indicated a similar adsorption mechanism to that of chamomile flowers, which can be attributed to the similarities in their surface functional groups. In an acidic environment, heavy metal ions primarily undergo ion exchange with protons, as shown in process (1) of Figure 5. In a basic environment, the adsorption mechanism, as illustrated in process (2), proceeds without the need for ion exchange with protons.

3.2. Batch Adsorption

3.2.1. Equilibrium Adsorption

The adsorption of Pb²⁺ was carried out using an acetic acid (AA) buffer at pH 5.6 to prevent precipitation. Figure 6(a) presents Q_e for both acid- and base-treated luffa peels at various initial Pb²⁺ concentrations in the range of 5-250 ppm. The base-treated luffa peels exhibited a consistently higher adsorption capacity as the concentration of Pb²⁺ increased compared to the acid-treated luffa peels. This could be attributed to the higher presence of deprotonated COO⁻ groups on the surface of the base-treated sorbents, which require less free energy for Pb²⁺ binding compared to the proton exchange required by COOH groups, resulting in a higher binding affinity. Figure 6 (a) plotted with the Q_e vs C₀ shows a L-shaped curve, the Freundlich isotherm model linear plots shown in Figure 6(b), for both types of luffa peels shows good R² values, indicating that the Freundlich model effectively describes the Pb²⁺ adsorption process, which indicates chemisorption plays a critical role. The slope 1/n values for Pb²⁺ both less than 0.5 indicating a favorable adsorption.

The adsorption of Cd²⁺ was conducted using luffa peels under the same conditions as Pb²⁺ adsorption for comparison. Figure 7 illustrates the isotherm plots, Q_e vs. C₀ 7(a), and the Freundlich linear plots for Cd²⁺ (7b) are very similar to those for Pb²⁺. Lower Q_e values were observed for all concentrations of Cd²⁺ compared to Pb²⁺ on both sorbents. In addition, K_f calculated from the intercept and the slope 1/n values for Cd²⁺ were higher than 0.5, all confirming that Cd²⁺ had a lower affinity for the binding sites on the luffa peels than Pb²⁺. The R² values from both linear plots suggest that the Freundlich model also effectively describes the Cd²⁺ adsorption process.

Due to the higher solubility of Cd²⁺ at pH values below 10, its adsorption is not confined to acidic conditions. A pH study was conducted on a 250 ppm Cd²⁺ solution to investigate the maximum adsorption capacity of the biosorbents. An AA buffer at pH 5.6 and Tris buffers at pH levels of 6.7, 7.0, 7.5, and 8.0 were used to identify the optimal pH conditions. As illustrated in Figure 8, the Q_e values of both base-treated chamomile and luffa peels increased significantly from pH 5.6 to 6.7 and slowly increased up to pH 8.0 with no precipitation observed. Consequently, pH 8.0 was selected for further analysis of Cd²⁺ adsorption.

Isotherms were obtained for Cd²⁺ adsorption at pH 8 with Tris buffer. The F Freundlich linear plots data together with values for Pd²⁺ and Cd²⁺ ions at pH 5.6 were summarized in Table 2. It can be seen from Table 2 that chamomile flowers exhibited a higher Qe and adsorption capacity (K_f) than luffa peels at both pH values for Cd²⁺ adsorption, suggesting a greater number of active binding sites on its surface. Low Freundlich adsorption constants (K_f) were observed for Cd²⁺ adsorption at pH 5.6; however, when equilibrium adsorption was re-evaluated at pH 8.0, significantly higher adsorption capacities were observed for both chamomile and luffa peel adsorbents. These results indicated that both base-treated biosorbents had an average equilibrium adsorption Q_e exceeding 31 mg/g for Pb²⁺ ions at pH 5.6 and 36 mg/g for Cd²⁺ ions at pH 8.0 when C₀ was 250 ppm. The high R² values with the Freundlich model indicate that Cd²⁺adsorption may involve multilayer adsorption.

The equilibrium studies showed higher adsorption quantity (Q_e​) for both base-treated biosorbents across all pH ranges, suggesting that the base-treated biosorbents have more active surface binding sites due to the complete deprotonation of COOH groups at pH 8.0. At this pH, H⁺ ion exchange no longer occurs, effectively eliminating the H⁺ matrix effect observed with the acid-treated biosorbents. Density differences may also be a factor that acid and base treatment results in slightly different surface property composition-porosity, especially the chamomile treated sorbents.

Figure 9(a) displays the Langmuir isotherm model for Pb²⁺ adsorption using both base- and acid-treated luffa peels plotted by Q_e vs C_eq, shows an L-type (concave down) isotherm for both types of peels. The R² values from the linear plots, as shown in Figure 9(b), confirm that the Langmuir model effectively describes the Pb²⁺ adsorption process.

Although the adsorption of Pb²⁺ fits the Langmuir model well, the adsorption of Cd²⁺ did not follow the L-type Langmuir model well (R²<0.85). Table 3 summarizes the monolayer maximum adsorption capacity (Q^o_max), Langmuir adsorption affinity (K_L), and R² values for the linear plots. Higher K_L values with the base treated sorbents indicate a higher binding affinity. The maximum adsorption capacities are ranked in the following order: base-treated chamomile flowers > acid-treated chamomile flowers > base-treated luffa peels > acid-treated luffa peels, consistent with the results obtained from the Freundlich model. Based on the R² values, the Langmuir model provides a better description of Pb²⁺ adsorption. Which indicates that adsorption occurred on the adsorption sites, i.e. functional groups, as a monolayer adsorption.

3.2.2. Adsorption Kinetics

The kinetics study of Pb²⁺ adsorption was conducted using a 250 ppm Pb²⁺ initial solution over a time range of 1–60 minutes at pH 5.6, under the same conditions as the equilibrium studies. The adsorption process occurred rapidly, with approximately 32% of Pb²⁺ in the solution adsorbed onto both base-treated and acid-treated luffa peels within 1 minute of contact. This instantaneous adsorption process suggests that luffa peels have a strong affinity for Pb²⁺ ions in solution. The adsorptions reached equilibrium within approximately 60 minutes. At this time, the base-treated luffa peels exhibited a Pb²⁺ adsorption efficiency of 68%, while the acid-treated luffa peels showed an efficiency of 48%. Both values closely align with those obtained from the equilibrium studies, indicating that adsorption equilibrium was achieved at 60 minutes. Figure 10(a) shows the linear plot based on the pseudo-second order model for Pb²⁺ adsorption onto luffa peel biosorbent. The slope from the linear plot of the base-treated luffa peels is smaller, resulting in a higher Q_e due to the inverse relationship between Q_e and the slope of the linear plot shown in equation (7). Additionally, the strong linearity of both plots indicates that the pseudo-second order model effectively describes the bio-adsorption kinetics of the processes.

The kinetics studies of Cd²⁺ were conducted at pH 8.0 under the same conditions as the Pb²⁺ kinetics studies. Rapid adsorption of Cd²⁺ was observed, with approximately 37% adsorption efficiency for both types of luffa peels within 1 minute contacting time. Similar to Pb²⁺, the Cd²⁺ adsorption reached equilibrium at 60 minutes, achieving 75% adsorption efficiency for the base-treated luffa peels and 61% for the acid-treated luffa peels. Higher Q_e of the base-treated luffa peels was found, as shown in Figure 10(b).

As summarized in Table 4, the adsorbed amount calculated from the pseudo-second-order kinetic plots (Q_e^cal) closely match the equilibrium adsorption values (Q_e^m) measured experimentally, further confirming that the adsorption process follows a pseudo-second-order kinetic model and reaches equilibrium within 60 minutes. The higher fitting of the adsorption data to both ions to this model implies that the rate limiting step involves interactions of the ions with surface functional groups. This agrees with the IR spectra changes of the COOH and OH bonds after the adsorption, that both ions bind with these sites.

3.3. Fixed Bed Column Adsorption

Although base-treated chamomile flowers exhibited a higher adsorption affinity for both Pb²⁺ and Cd²⁺ in batch adsorption studies, its physical properties and chemical stability limit its suitability as a column sorbent. For instance, the low density of chamomile flowers leads to large void spaces within the column in a self-made column without applied pressure, and its low stability causes blockages, resulting in inconsistent flow and operational issues. In contrast, base-treated luffa peels were easier to pack into the columns and produced consistent free solution flow rate. Therefore, only base-treated luffa peels were focused on the dynamic adsorption and recovery studies for the two metal ions.

The fixed-bed column adsorption of Cd²⁺ was performed using a 250 ppm Cd²⁺ solution at pH 8.0, while Pb²⁺ adsorption was performed at pH 5.6. Continuous flow was maintained at a rate of 4 mL/min until the C_t/C₀ ratio reached 0.8, indicating 80% saturation of the luffa peels. The initial fixed-bed height of approximately 15 cm reduced to ~10 cm after the first regeneration cycle and further to ~8 cm after the final adsorption-regeneration cycle, consistent across both columns. This reduction in bed height may have contributed to the observed decrease in dynamic adsorption capacity, as shown by the shifts in breakthrough curves in Figure 11.

As shown in Figure 11(a), the breakthrough points occurred when the C_t/C₀ ratio reached 0.1. The breakthrough curves for Cd²⁺ indicate a decrease in breakthrough time with each adsorption-regeneration cycle: 125 minutes for the first cycle, 92 minutes for the second cycle, and 78 minutes for the third cycle. The decrease in breakthrough time indicates a reduction in breakthrough capacity (Q_b), and the overall shift in the breakthrough curves reflects a continuous decline in dynamic capacity (Q₀). This suggests that the luffa peels lost active binding sites after each cycle, likely due to incomplete release of Cd²⁺ ions as well as the reduction in bed height which may result in a denser pack of the sorbents, that reduced the accessible sites for adsorption.

Similar to Cd²⁺ adsorption, Figure 11(b) illustrates the shift in breakthrough curves for Pb²⁺ after each adsorption-regeneration cycle. The breakthrough time for the first cycle was 145 minutes, indicating a higher adsorption affinity of luffa peels for Pb²⁺, consistent with the batch study results. The breakthrough times decreased to 90 minutes in the second cycle and 75 minutes in the third cycle, with a corresponding decrease in breakthrough capacity (Q_b) also observed. Both fixed-bed column studies stated that luffa peels maintained a high adsorption capacity even after three cycles.

Table 5 summarizes the parameters derived from the Thomas model, with R² values indicating a strong fit of the model to the fixed-bed column adsorption data. Across all cycles, luffa peels showed a higher dynamic adsorption capacity for Pb²⁺, as seen in the Q_b and Q₀ values, consistent with the batch adsorption results. For Pb²⁺, the Q₀ in the first cycle was 32.9 mg/g, decreasing by 40% to 20.0 mg/g in the second cycle and by another 9% to 18.2 mg/g in the third cycle. Similarly, the Q₀ for Cd²⁺ started at 25.8 mg/g in the first cycle, dropped by 33% to 17.2 mg/g in the second cycle, and further by 8% to 15.9 mg/g in the third cycle. The slower rate of capacity loss across cycles suggests that luffa peels retain potential for adsorption even after three adsorption-regeneration cycles. The continuous increase in the Thomas adsorption rate constant (K_T) with each cycle may be due to the luffa peels becoming more homogeneous after prolonged soaking in the column.

Recovery

The luffa peels were regenerated by introducing a 0.4 M HNO₃ solution to both columns at a flow rate of 2-4 mL/min until the effluent concentration of Pb²⁺ or Cd²⁺ (C_e) dropped below 0.01 ppm. As shown in Figure 12(a) and Figure 12(b), rapid releases of Pb²⁺ and Cd²⁺ were observed, with the Pb²⁺ column exhibiting a faster rapid release rate, likely due to the higher quantity of adsorbed Pb²⁺ ions.

The recovery curves remained consistent across all cycles. By calculating the total mass of Pb²⁺ and Cd²⁺ adsorbed from the breakthrough curves and comparing it with the total mass recovered from the recovery curves, the recovery efficiency for both Pb²⁺ and Cd²⁺ was determined, as summarized in Table 6. The recovery efficiency remained above 87% for Pb²⁺ and 90% for Cd²⁺, indicating that luffa peels serve as a reliable adsorbent with strong reusability. These recovery studies also highlight the potential of luffa peels as an effective solid-phase extraction (SPE) sorbent for Pb²⁺ and Cd²⁺. To achieve recovery, higher nitric acid concentration might be needed, which could alter the nature of the sorbent.

The FTIR spectra (Figure 13) indicates that the luffa peels retained surface carboxylic acid (COOH) groups after three adsorption-desorption cycles. An increase in all peak intensities is observed, which may be attributed to the homogenization of the sorbent following prolonged soaking in solution. Additionally, the increased intensity of the C=O stretching band at 1729.8 cm⁻¹ and the absence of C–O stretches are likely due to the sorbent being at pH 4.5 after acid washing, resulting in protonation of the surface functional groups. The presence of COOH groups observed in the FTIR spectra further suggests that the luffa peels retain adsorption capacity even after three usage cycles. This retention of functional groups highlights the durability and reusability of luffa peels as an adsorbent in repeated adsorption-desorption processes.

3.4. Comparison of Other Studies

Table 7 and Table 8 summarize a comparison of the adsorption efficiency of luffa peels, chamomile flowers, and other adsorbents reported in the literature, under batch and column methods, respectively. According to these tables, both luffa peels and chamomile exhibited relatively high affinities for lead and cadmium in batch experiments, while luffa peels showed high dynamic adsorption affinity for these metals in column experiments. These results suggest that luffa peels and chamomile flowers are promising adsorbents for the removal of lead and cadmium from aqueous solutions.

4. Conclusions

The use of Luffa peels and chamomile flowers have been evaluated for the removal and recovery of lead and cadmium ions from aqueous solutions. FTIR spectra indicates that both adsorbents are rich in surface carboxylic acid (-COOH) groups, which contribute to adsorption through ion exchange or electrostatic interactions, depending on pH. The base-treated adsorbents have higher adsorption capacities for both ions likely due to higher deprotonation of the carboxylate and phenolate groups. The isotherm for both metal ions fit well with the Freundlich model, an indicator for chemisorption, although the Langmuir model fit better for the Pb(II) adsorption. Adsorption kinetics obeyed a pseudo-second-order model, confirming chemical interaction as the rate limiting step in the adsorption process. The successful fixed-bed dynamic column adsorption with base-treated luffa peels for both metal ions suggest that luffa peels have high potential to be used as low-cost reuseable biosorbents for the adsorption and recovery of toxic heavy metals.