Optimization of batch study parameters for the adsorption of lead(II) ions onto spent tea grains

Contamination of surface and sub-surface water bodies due to rapid urbanization and industrialization is now a well-known phenomenon (Ahluwalia & Goyal 2005). Detection of toxic heavy metals in our natural environment has been a major problem for scientists worldwide (Singh & Singh 2012; Omar et al. 2022). These toxic heavy metals are affecting the health of our overall ecosystem and are a major health concern for human beings due to their non-biodegradability and bioaccumulation in different food chains (Yoshita et al. 2009). The awareness of today's toxicity driven limits is the major driver for our scientific community to find different methods to reduce toxic contaminates in the ecosystem. Discharge studies of surface and sub-surface water are also important for toxicity assessment (Omar 2015; Shekhar et al. 2021).

Lead, a highly toxic heavy metal, is one of the priority pollutants. Lead can be released into the environment by a vast number of industries, like battery manufacturing, electroplating, mining, pigments, paper, and pulp (Mondal 2010). WHO limits detection of a maximum of 0.005 mg/l in drinking water because at even very low concentrations it is causing various diseases and disorders. Therefore, the removal of lead from our ecosystem is currently one of the most important aims of our research community (Zuorro & Lavecchia 2010). Adsorption of toxic heavy metals onto low-cost locally available adsorbent is an area under development in the past decade (Bailey et al. 1999; Babel & Kurniawan 2003; Demirbas 2008). The various agricultural wastes are widely being investigated for their adsorption capabilities. These wastes are locally available in vast amounts, and they are low-cost raw materials. Many researchers investigated the capability of such materials in removing different heavy metals from aqueous solutions like sugarcane, bagasse, leaves, rice husk (Marshall & Champagne 1995), sawdust (Shukla & Pai 2005), peanut hulls (Periasamy & Namasivayam 1995), peat moss (Simón et al. 2022), coconut shell (Sekar et al. 2004), and crop milling waste (Saeed et al. 2005).

A low-cost locally available adsorbent is one which can be found abundantly in nature or is a by-product or a waste from the nearby community (Lavecchia et al. 2010). Developing low-cost and more effective adsorbents for lead removal from biomass waste is a novel and promising area (Guclu et al. 2021; Ang et al. 2022). Tea, due to its well-known health benefits, is a widely consumed beverage worldwide. Spent Tea Grains (STG) are an agricultural waste compromising cellulose, hemicelluloses, condensed tannins, structural proteins, and lignin (Sahu et al. 2018). These above-mentioned constituents make tea waste a great candidate to replace activated carbon used in industry as an adsorbent in future as it contains various functional groups which help in metal scavenging, mainly carboxylate, phenolic hydroxyl, and oxyl groups (Hussain et al. 2018). The objective of the present study is to evaluate the efficiency of tea waste as an adsorbent for lead removal from an aqueous solution. A batch study experiment has been done to understand the effect of various parameters such as particle size, pH, initial concentration of metal, adsorbent dose and contact time.

Lead(II) salt was used to prepare synthetic wastewater aqueous solutions. Double distilled water was used for preparation and dilution to the required concentration. The 1,000 mg/l stock solutions containing Pb (II) are obtained from Pb (NO₃)₂ salt of analytical grade. All glasswares used during the experimentation were properly washed and sterilised before and after the experimentation using chromic acid and then washed with distilled water.

The STG were obtained from different tea stalls on the campus of IIT BHU, Varanasi. The collected biomass waste is first dried in an oven at 80 °C for 12 h; however, instead of this collected biomass can be dried under sunlight for 2 days. Different constituents other than tea leaves/grains were then separated from it during preliminary screening. After this, the collected biomass waste was repeatedly boiled at above 100 °C and washed using distilled water for colour removal. After colour removal of the tea waste was completed, it was then oven dried by applying a temperature of 80 °C for 24 h. The sieving process was being applied to the prepared STG, particle range selected was in between 600 and 75 μm to be used for various batch experiments. The prepared adsorbent was stored in polyethylene bags at room temperature.

The pH of the aqueous solutions used during different experiments was measured by using a digital pH meter. As pH is one of the most important parameters, its calibration was done with requisite precision. To identify different functional groups, present in STG, Fourier Transform Infrared Spectroscopy was investigated. Functional groups indicate adsorbent capability towards heavy metal scavenging, it also gives us an idea about the adsorption mechanism (Hammud et al. 2016). A spectra range of 600–4,000 cm⁻¹ was selected to determine different functional groups present in the STW samples. SEM-EDS analysis was done to understand the morphological characteristics of the prepared adsorbent. SEM-EDS analysis was performed at an acceleration voltage of 5–15 kV and an EDAX detector using a microscope equipped with an energy dispersive analytical system. Elemental analysers were used for determining the elemental composition of the prepared adsorbent STG. Proximate analysis for determining fixed carbon, ash, moisture, and volatile matter was performed according to the ASTM D 2866-94, IS code 1350 (Part 1), and ASTM D 2867-95 methods. For investigating the structural properties of the adsorbent, X-ray diffraction (XRD) analysis was performed on the prepared adsorbent. XRD analysis is used to indicate crystal structural characteristics which give us an idea about its metal scavenging capabilities. Thermal degradation characteristics were investigated by thermogravimetric analysis (TGA). The TGA is very useful for determining the type of biomass composition, its moisture content and volatile matter composition. TGA is one of the effective techniques that are being used for pyrolysis optimization.

The point-of-zero charge (pHzc) of the prepared adsorbent was investigated by the solid addition method. A series of 50 ml aqueous solutions of 0.1 M KNO₃ was transferred into a conical flask of capacity 100 ml. The pH for its determination was adjusted from 2.0 to 12.0 by using dilute HNO₃ and dilute KOH. Then, 1.5 g of prepared adsorbent from STG was added to each conical flask. The conical flasks were then placed in a rotary shaker for 24 h at 150 rpm. After 24 h, the pH values of the supernatant are noted down. By analysing these values, we get the value of pHzc for the prepared adsorbent STG.

Response surface methodology (RSM) is a powerful mathematical and statistical tool to analyze a wide variety of factors over a wide range to obtain responses, to identify the region of optimal response or to determine the near-optimal response by developing a designed experiment (Kavand et al. 2020). Central composite design under RSM was used for the optimization study. The advantage of using CCD using multilevel factorials in comparison to full factorial design is it will need fewer experiments to provide us with reliable information for statistical analyses. The ideal values of different process variables were calculated using the CCD model and then further validated by experiments (Sujatha & Sivarethinamohan 2021). The Design-Expert 13.0.5.0 (Stat-Ease) software was used for the regression and graphical analysis of the obtained data.

Firstly, some preliminary experiments were performed on lead(II) adsorption using STG as an adsorbent and the range and central points of all independent factors were determined. Different process variables and their ranges used in the model's computation were 5 and 50 mg/l for initial lead(II) ion concentration. The lower and upper limits of pH were 2 and 7, respectively. Similarly, the adsorbent dose was 1 and 10 g/l, respectively, the contact time was 30 and 180 min, respectively, stirring speed taken was 25 and 200 rpm, respectively. When this model is employed we will get 32 combinations of different test variables, namely A, B, C, D, E, AB, AC, AD, AE, BC, BD, BE, CD, CE, DE, A², B², C², D² and E².

Analysis of variance (ANOVA) tool is used for above developed CCD model with 95% confidence limits (α = 0.05). Consideration of the model is only done after comparing the variance due to regression to that of the total variance (Gholamiyan et al. 2020). If they are different, then we will only consider the model developed to be validated. F-value and p-value will provide us with the significance of each factor. To provide the precise significance of the relationship between input variables, responses, and influential factors the ANOVA tool is utilized.

To acquire kinetics and equilibrium data, different adsorption experiments were performed in a batch manner using an aqueous solution of Pb(II). For the determination of optimum values of different parameters which will affect adsorption efficiency, batch studies were performed. The pH values of the aqueous solutions were changed using weak acid, i.e. 1.0 M HNO₃ or weak base, i.e. 0.1 M NaOH solution whenever required during various experiments. Various factors considered during experiments were investigated including pH of the aqueous solutions (2.0–7.0), contact time (15–180 min), initial metal ion concentration from 5 to 50 mg/l, stirring speed (30–180 rpm), and adsorbent dose (1–5 g/l). The efficiency of adsorption is always dependent on the surface area available for adsorption. Therefore, the effect of particle size was investigated in the range of 75–150 μm, 150–300 μm, 300–600 μm, 600–900 μm, 900 μm–1.18 mm at room temperature, and maintaining other parameters as optimum during the experimentation.

In the aforementioned Equation (2), Q_e is the amount of adsorbed lead(II) ions by the adsorbent STG in mg/g. The mass of adsorbent (i.e. STG) is represented by m in mg. The volume of the solution (ml) is indicated by V in Equation (3); Y (%) defines the removal efficiency of lead(II) ions by the adsorbent. C_o and C_e are, respectively, the initial and final concentrations of lead(II) in mg/l.

Proximate analysis of STG will provide us with information about moisture, volatile matter, fixed carbon, and ash content, respectively. From the analysis of STG, we get the values fixed carbon content at 19.6%, there was a low ash content of 5.4%.

From the ultimate analysis of the STG, we found a reasonable amount of carbon content of approximately 47.541%. Detailed estimation of various parameters is mentioned in Table 2.

The initial weight loss in the first stage (20–200 °C) is around 8.29%. The second region of the STG transpires at around 250–415 °C. At a temperature of around 200 °C, a rapid devolatilization of the components starts and lasts until 410–415 °C, indicating the degradation of the fibre content of STG of hemicellulose, cellulose and attributed to the decomposition of holocellulose and partial decomposition of lignin. The contribution of weight loss during the second stage is about 51.48 wt.% and the maximum decomposition temperature was found to be in the region of 415 °C. In the second stage, the high weight loss may be due to the degradation of its main components by decarbonization, dehydration, and decarboxylation. The third stage of the STG is at 415–1,100 °C. The maximum decomposition temperature in this region is 1,100 °C. The weight loss in the third stage was due to the degradation of the lignin content of the STG. When the temperature reached 1,100 °C, the total weight loss of the STG is 93.49 wt.%.

Therefore, we can conclude from TGA that during our preparation of adsorbent at the initial stage when we applied a temperature of around 80 °C for 24 h weight loss occurred only due to loss of moisture, or loss of weak bonds, and there will be no chemical and structural modifications. Basically, during the preparation of STG, we are avoiding the degradation of lignin and cellulose functional groups responsible for the adsorption of lead(II) onto STG.

The FTIR spectroscopy technique was used to measure the various IR peaks of different functional groups of the STG. FTIR analysis was performed in the range of 500–4,000 cm⁻¹ wave number. From Figure 2, we can see the complex IR spectra showing numerous different types of functional groups on the surface of STG. The difference between the IR peaks of different functional groups onto STG before and after lead(II) ions shows possible functional group involvement during the process of adsorption. Specific IR peaks due to unique energy bands before and after adsorption onto STG are provided as a useful tool to identify the involvement of certain functional groups during the adsorption process onto the surface of the STG.

At 3,351.2 cm⁻¹ there is a presence of broad absorption bands, which signifies –OH stretching and confirms the presence of functional groups related to cellulose and lignin. The peaks appearing at about 3,016.6–2,859.5 cm⁻¹ could be attributed to the aliphatic C–H groups, stretching in CH, CH₂, and CH₃ groups. The band observed at 1,641.96 cm⁻¹ could be assigned to the asymmetric stretching vibrations of C = O stretching of the amide group and the peak appearing at around 1,475.51 cm⁻¹ could be due to the aromatic compound group owing to the presence of N–H bending types of bonds of amide (II).

The various elemental compositions of STG before and after the adsorption of lead(II) ions were analysed by EDX analysis as shown in Figure 5(c) and 5(d). The major constituents of the raw STG were found to be only carbon and oxygen. The EDX analysis of STG after lead(II) ions adsorption indicates the existence of lead(II), confirming its attachment to the biosorbent.

The quadratic model was developed using the CCD design matrix. The matrix was established to investigate the influence of five independent factors (pH, contact time, initial metal ion concentration, stirring speed, and adsorbent dosage of STG). Five levels for different variables were defined (i.e. low (encoded −1), high (encoded +1) and rotatable (encoded ±α)). The adsorption uptake capacity and removal efficiency, taken as the main response, were predicted using a polynomial regression equation in which the main, interaction and quadratic effects of the variables were modelled.

The CCD model gives us an experimental run of 32 runs considering all independent variables which will be replicated three times and an average of the results will be taken for the modelling purpose. All the experimental runs are shown in Table 3 with its responses of removal efficiency and adsorption uptake capacity. It is observed during the analysis of the CCD model that the majority of independent process variables affect the responses. The quadratic equation is developed through the CCD design matrix and its main contents area is shown in Table 4. High F-value, correlation coefficient (R) tending to 1, low p-value and most importantly the results of lack of fill test indicates the high statistical significance of the developed model with high accuracy.

The model F-value for percentage removal of 8.70 implies the model is significant. The model F-value for uptake capacity of 3.98 implies the model is significant. There is no indication of noise affecting F-values.

Low p-values for removal efficiency less than 0.05 indicate model terms are significant. Low p-values for adsorption uptake capacity less than 0.05 indicate model terms are significant. The Lack-of-Fit F-value for % removal of 605.44 implies the Lack-of-Fit is significant. There is only a 0.01% chance that a Lack-of-Fit F-value this large could occur due to noise. The Lack-of-Fit F-value for uptake capacity of 48,484.97 implies the Lack-of-Fit is significant. There is only a 0.01% chance that a Lack-of-Fit F-value this large could occur due to noise.

In the present study, the influence of pH on the adsorption capacity and % removal efficiency were studied. The pH in the range of 2.0–7.0 was adjusted using dilute acid/base. The initial pH value of the prepared aqueous solution containing lead(II) ions was found to be around 4.2 ± 0.4. For detailed estimation of optimum pH, three different adsorbent doses were considered for the experiment (low dose of 1 g/l, average dose 2.5 g/l, and high dose 5 g/l), and four different initial metal ion concentrations were considered (5, 10, and 15 mg/l). Other parameters remain constant, i.e. contact time of 90 mins, shaker speed of 150 rpm.

When we analysed the Figure 11(a) and 11(b) with respect to both adsorption capacity and removal efficiency we get our optimum value of pH, i.e. pH = 5. At pH 5, the uptake capacity was found to be maximum with very good % removal efficiency.

Further increasing pH from 5 to 7 increases our removal percentage but decreases adsorption uptake. When the effects of pH value on the removal percentage and adsorption uptake of the metals were examined simultaneously, it was found that the optimum value for Pb was pH 4.96 ± 0.12. At lower pH values the H+ ions compete with metal cations for the electrostatic surface charges in the system decreasing the percentage of sorption. These outcomes regarding the pH amount are also parallel to various studies that used acidic pH. The experimental results showed that the maximum removal percentage of Pb is found to be 96.67% at the optimum pH, C_o = 5 ppm, contact time = 90 min, adsorbent dose = 0.5 g/100 ml, stirring speed = 150 rpm.

Here, C_e refers to the balance amount of lead ions (mg/l). Q_e refers to the adsorption capacity of adsorbent STG at equilibrium in mg/g. Q_m refers to the the maximum adsorption capacity of adsorbent STG in mg/g, K_L refers to the Langmuir constant (L/mg). K_F and 1/n refer to the Freundlich constants. Kinetics models used will give a clear indication about the changes in the adsorption mechanism within a stipulated time-period and provide significant information required for the experimental setup.

The linear Langmuir isotherm plots STG is shown in Figure 17(a). R² greater than 0.99 is a good indication of the suitability of lead(II) ions adsorption onto STG. Using Langmuir, the isotherm maximum adsorption capacity is found to be 24.272 mg/g. R_L is of utmost importance which indicates the condition of adsorption onto adsorbent. R_L greater than 1 is unfavourable and signifies non-optimum adsorption, linear adsorption for R_L = 1, and favourable optimum adsorption occurs when R_L values lie in between 0,1. Now in our present study, we found R_L = 0.1444 which indicates optimum lead(II) adsorption onto STG. All the calculated data are presented in Table 5.

When we study the Freundlich isotherm in its linear form it also shows a good correlation coefficient R² = 0.9665 which suggests STG possesses high affinity and high adsorption capacity. As 1/n < 1 it suggests optimum adsorption condition, it also indicates multilayer adsorption onto heterogeneous surfaces.

Temkin isotherm model investigation shows a very good correlation coefficient of R² > 0.99, values of equation parameters b_T = 247.486 and A_T = 7.4644. The value of the correlation coefficient for the D–R model is low (R² = 0.835) as compared to the other models, suggesting that the data is not fitted into the D–R isotherm models. Moreover, the value of maximum adsorption capacity calculated using the D–R isotherm model is not close to the value calculated using the Langmuir isotherm model.

In the aforementioned equation Q_e measured in mg/g, is the adsorption capacity of STG at equilibrium and Q_t measured in mg/g is the adsorption capacity of STG at a certain time t; k₁ and k₂ are the pseudo-first-order rate constant and pseudo-second-order rate constant, respectively. The slope and intercept from the linear plot of ln (Q_e − Q_t) versus t as defined by the pseudo-first-order model will provide the values of k₁ and Q_e while in the case of the pseudo-second-order model, the linear plot of t/Q_t versus t will define the values of Q_e and k₂.

It is apparent from all the kinetics models as represented by Figure 18(a)–(d) that the adsorption capacity of STG increased rapidly for lead(II) in the first few minutes and then steadily increased until the adsorption equilibrium was achieved (i.e. 90 min). The parameters obtained from all the studied kinetic models for the removal of lead(II) onto STG was tabulated in Table 6. A correlation coefficient value >0.99 was found when we evaluated pseudo-second-order kinetics. After evaluation of all the parameters from all four kinetics models studied, it can be concluded that the experimental data is best represented by the pseudo-second-order model compared to the other kinetic model for lead(II) adsorption onto STG.

The lead(II) adsorption capacity of STG was compared with other biosorbents found in the literature. The values are given in Table 7. From the table, STG was found as an efficient adsorbent with good metal scavenging capabilities.

For maintaining the biosorption process to be cost effective as an alternative to other wastewater treatment schemes, the regeneration or recyclability of biosorbent is very important. In this experiment, 1 g/l of the STG was used for lead(II) adsorption. The initial metal ion concentration of lead(II) ion was taken as 50 mg/l at optimum pH and stirring speed. The duration of sorption taken was for 6 h. The adsorbent was regenerated using 0.1 M HNO₃, and then washed with double distilled water. After drying it is used in subsequent cycles. In the first cycle, the removal efficiency and adsorption capacity come at 87.37% and 19.66 mg/g,, respectively, which decreased to 78.43% and 17.648 mg/g, respectively. After the third cycle it again decreased slightly 71.63% and 16.118 mg/g. The sorption of the lead(II) ions from STG declined in the next cycles also. It was observed that regenerated STG using 0.1 M HNO₃ can be used for up to three cycles without losing much of its metal scavenging capabilities. This desorption study result might be due to a decrease in available active sites after 2–3 cycles of regeneration because they were inactivated in the first cycle. Thus, diffusion of the metal ions into the inner structure of the prepared adsorbent; therefore, the extractant has a weaker ability to recover such occupied sites.

Discharged heavy metals with wastewater cause many environmental and health effects. The solution would be preventing the discharge of heavy metal directly into the water bodies. Adsorption is one of the important processes for the removal of heavy metals from wastewater. STGs are a cheap and available material discarded as waste material from tea stalls to hotels in the environment without any treatment and can be converted into adsorbent for lead(II) removal from an aqueous solution.

RSM analysis indicates that STG is a potential adsorbent with a desirability function of 0.619. Desirability analysis provided the inference about the optimum response regarding the adsorption capacity of STG as 8.9087 mg/g when process variables were 38.75 mg/l (initial concentration), 5.20655 (pH), stirring speed of 119.32, and 3.25 g/l (STG dose) for a contact time of 135.05 min at a desirability value of 0.619.

STG characterization shows a high surface area for the adsorption. The STG was determined as a potential effective adsorbent in the removal of lead(II) from aqueous solutions. The optimum adsorption at which lead(II) ions uptake is maximum was at around pH 5. The rate of adsorption was found to be quite rapid during the initial contact.. Equilibrium is achieved within 90 min for adsorbing a significant amount of lead(II) ions onto STG. Four isotherm models were used for the analysis of experimental data (i.e. Langmuir, Freundlich, Temkin, D–R isotherm models). The adsorption process is showing monolayer formation and can be easily explained by the Langmuir isotherm having an adsorption capacity of 24.272 mg/g. By kinetics studies evaluation we can conclude that the pseudo-second-order kinetics model for lead(II) ions adsorption onto STG was the most suitable kinetic model.

The highest removal efficiency was found using kinetics models. It is found to be 94.33% when under optimum conditions for C_o = 5 ppm we use an adsorbent dose of 0.5 g/100 ml. The adsorption mechanism and its uptake capacity were dependent on various parameters such as the pH of the aqueous solution, STG dose, initial metal ion concentration, contact time, stirring speed and particle size. The results of this study showed that STG could be suitably used as an alternative and effective adsorbent material for the removal of lead(II) ions from aqueous solutions. Based on the present study, STG can be considered as a low cost, locally and freely abundantly available, eco-friendly, and efficient bio-adsorbent for removal of lead(II) from aqueous solutions. After detailed experimental investigation we can say that RSM modelling gives us a good indication of potential adsorbents, but developing an adsorbent which can be scaled-up requires detailed experimental investigation.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.