USE OF AGROINDUSTRIAL MATERIALS AS ACTIVATED CARBON PRECURSORS FOR CAFFEINE REMOVAL: GLOBAL OVERVIEW

Objective: Prepare a global overview of agro-industrial residues used for activated carbon (CA) production, seeking caffeine removal of water bodies. Theoretical reference: Aiming the adequate understand of involved processes, as well as each stage when adsorption occurred. The theoretical concepts might help to analyse how important is pollution of emerging contaminants in water bodies, as a current environmental problem. Method: A data survey was developed in general perspective, using information from literature published in online repositories, such as Scopus and Science Direct, trough investigations that seek the use of agro-industrial residues for production of activated carbons and, also, their caffeine removal parameters analysis. Results and conclusion: The analysis results of the different methodologies applied in studies for activated carbon production from agro-industrial residues, showed that good caffeine adsorption process and the adequate characteristics of each adsorbent material, depend on processes and conditions adopted by each researcher. In this way, the different studies here analysed showed the importance of using industry waste, helping to reduce waste volumes in landfills and, at the same time, the search for new technologies that could help to remove emerging contaminants, such as caffeine. Originality/value: This study provides different basic and methodological concepts related to carbonization processes from agro-industrial waste, verified and studied by different authors for caffeine elimination; and also the importance of tracking new economical and environmentally friendly precursor materials.


INTRODUCTION
Erroneous production and consumption standards of energy, water and material resources in society, is one of the main causes of environmental problems (Ghazinoory, 2005, Grejo & Lunkes, 2022. However, hydric resources contamination has become one of the most relevant environmental issues and has recently caused international concern, mainly due to various contaminants entering in aquatic systems, as result of global human population growth, increased industrialization, unplanned urbanization, agricultural activities, as well as the excessive use of chemical products (Vakili et al., 2014).
In recent times, different substances were discovered in water bodies known as contaminants of emerging concern (CECs) (Ouda et al., 2021;Khan et al., 2022). The list of these substances is particularly wide, including drugs such as antibiotics, anti-inflammatories, antidiabetics, analgesics, antidepressants, natural and synthetic hormones, illicit drugs and endocrine disruptors; cosmetics and personal care products; industrial matherials such surfactants, additives; microplastics and pathogens. The number of chemical substances changes constantly, both in terms of original and their degradation materials (Crini et al., 2022).
In the last years, relevant efforts have been applied to develop adaptable and costeffective strategies to address different environmental concerns, using sustainable and multifunctional materials aiming to eliminate toxic contaminants from different outcomes, which is a cause of global apprehension (Chawla et al., 2022). The adsorption technique is recognized for being flexible and easy to use. Therefore, becomes necessary the use of simple, renewable and low-cost energy sources for carbonaceous adsorbents production (Kadim & Abd, 2022). At the same time, adsorption processes have many advantages compared to traditional techniques, as they require less energy to work, are characterized by their versatile designs and can operate in different scenarios for water purification (Koul et al., 2022).
Activated carbon is a carbon material with specific surface properties and highly developed porosities, which belongs to porous carbon materials. Different literature revisions revealed great interest in using activated carbon materials (Ahmad et al., 2010), due to the extensive base of raw materials, diverse production methods, high chemical and thermal resistance, as well as good adsorption and catalytic agents (Serafin et al., 2022). In this context, the theoretical contribution of this work, defining both the part of ECs and caffeine, and the different agro-industrial residues used for the production of activated carbon, as well as a practical one revealing a wide diversity of studies developed for caffeine adsorption, this work can be and report important scientific references in this subject, justifying this investigation.
This investigation has the purpose study agro-industrial residues for activated carbon production in order to eliminate ECs, such as caffeine.

THEORETICAL REFERENCE
Theoretical concepts are important tools to analyze and present theories that exist about adsorption, being divided into different relevant concepts focusing the analysis of different studies, adding scientific knowledge and greater clarity in this subject. In this way, it is pertinent to study previous concepts, helping to improve the understanding of CECs.

Emerging Contaminants
According to Gil et al. (2012), CECs refer to compounds of different origin and chemical nature, whose presence in the environment is considered insignificant in terms of distribution and concentration; therefore, the largest groups of contaminants do not include CECs, and they are not considered for final effluent evaluation, in different sources. However, CECs are being widely detected, their persistence and toxicity can affect human health, and other living beings, by metabolism alterations, as well as causing negative effects on environment, especially for aquatic ecosystem (Varsha, Kumar & Rathi et al. al., 2022) The biggest concern is the quantity of substances released into the environment, since these compounds do not have emission standards. Some countries adopted concentration limits as parameters for substances in water bodies and, therefore, its management due their relevance from environmental contamination point of view. The "Policy 2008/105/EC" of the European Parliament and "The Council and the Canadian Environmental Protection Act", show the concentration limit values for these contaminants, however there are no defined limit values for caffeine (Richardson & Kimura, 2016 ;Luo et al., 2014).
It has been established that these compounds enter the environment through some sources, such as domestic and industrial wastewater, waste from treatment plants, hospital effluents and agricultural activities that contain a large amount of specific organic components and CECs, which are transported, in different concentrations, to surface waters. These contaminants can be easily transported and are highly persistent in air, water, soil, sediments and ecological receptors, even at low concentrations (Gavrilescu et al., 2015). Figure 1 shows the different CECs input sources Figure 1. Input ways of some contaminants of emerging concern (CECs) from sources to receptors. Source: Adapted from Gavrilescu et al. (2015).

Caffeine
According to World Health Organization (WHO), the definition of caffeine is an alkaloid, classified as a psychotropic drug which alters the central nervous system. Its molecular composition, basically, contains oxygen, hydrogen, carbon and nitrogen as shown in Table 1. Caffeine is widely used in the industrial sector, mainly in the pharmaceutical sector. It is part of a variety of drinks and numerous food products consumed by human population. Different studies reported caffeine impacts on water bodies. Golovko et al. (2021) for example, verified the occurrence and removal of 164 CECs in wastewater treatment plants and their ecotoxicological impact on water systems with caffeine concentration of 64,000 ng/L, and high concentrations after pre-treatment in wastewater samples. In another study carried out by Sui et al. (2010), if it was identified that the most abundant compound in consumer products managed at different treatment stations is caffeine, with concentrations of 3.4 to 6.6 mg/L, the authors justified that its presence is due to greater consumption beverages such as coffee, tea, energy drinks, etc. Yang et al. (2021) showed caffeine risk assessment, suggesting that this CEC poses relatively high ecological risks for most species, such as algae and flowers in aquatic ecosystem, especially around the water treatment plant gates.

Activated Carbons
Activated carbons (AC) are versatile materials widely studied due their characteristics, such as high surface areas, high adsorption potential, specific surface properties, thermal stability, mechanical and electrical properties, low acid/base reactivity and controllable pore structure, rapid adsorption kinetics and relative ease of regeneration (Üner & Bayrak, 2018). Another important attribute is their origin, they are generally obtained from residues, from sources forestry and agricultural co-products.
In the AC production process, activation is the main step, since it will provide its characteristics that would help in adsorption process (Danish et al., 2011). Therefore, there are two activation processes: • Physical activation: That consists of two steps: (i) carbonization and (ii) gasification control of carbonaceous precursor using gas stream (O2, CO2, water vapor, etc.) at high temperatures and, then, comes the chemical activation, applied directly to the raw material and at relatively high carbonization temperatures, above 800 °C (Ozdemir et al., 2014).

•
Chemical activation: The precursor is impregnated with a chemical agent, with the resulting solid activated at temperatures lower than those used in physical activation (Njoku & Hameed, 2011). The raw material is transformed into carbon, acquiring properties with great activity and adsorption power. The carbonization/activation is carried out with treatment temperatures ranging between 400 and 900 °C. After carbonization process, the charcoal obtained is cooled, washed and then dried for use (Motato & Arrieta, 2016). The chemical agents used are phosphoric acid (H3PO4), sodium hydroxide (NaOH), zinc chloride (ZnCl2), hydrochloric acid (HCl), sulfuric acid (H2SO4), among others (De Costa, Furmanski & Dominguini, 2015).

Adsorption Processes
Adsorption is the mass transfer operation, as ability of certain solids to concentrate, on their surface, certain substances existing in liquid or gaseous fluids allowing their components separation. Since the adsorbed components are concentrated on external surface, the greater this external surface per solid mass unit, the adsorption will be more favorable (Ruthven, 1984).
The adsorption process is considered the best alternative for water and effluents treatment, due to its convenience and ease of operation, gaining importance as separation and purification process (Bhatnagar, Sillanpää & Witek-krowiakc, 2015). This method has been the interest of researchers since the beginning of the century, presenting technological and biological importance, as well as practical applications in industry and environmental protection, making it a useful tool in many manufacture sectors (Cooney, 1999;Mckay, 1996;Dabrowski, 2001;Crini, 2005;Do Nascimento et al., 2020). Regarding the environmental point of view, adsorption is one of the most efficient processes for drinking water and wastewater treatment, and is being used in industries aiming to reduce the quantity of toxic compounds in their effluents (Santana et al., 2020;Oliveira et al., 2020;Moreira et al., 2016;Do Nascimento et al., 2020).
The factors that influence adsorption process are different such as surface area, adsorbent and the adsorbate properties, the temperature system, the solvent nature and the adsorbent medium pH .

METHOD
For the present systematic literature review, an advanced search was developed in Scopus and Science Direct databases, using the bibliometrix tool with the RStudio software, to analyze the different studies using keywords such as: (Adsorption OR Adsorption Process OR Adsorption Technologies) AND (Emerging Contaminants OR Emerging Pollutants) AND (Agroindustrial waste OR Industrial Waste) AND (Activated carbon OR Biochar OR Organic Compounds) AND (Caffeine). A review of book chapters was also applied to address different specific concepts for this investigation.
After the collection of scientific papers, and for data selection, an inclusion criterion was the search for articles focused on CECs removal, specifically caffeine, considering this important, since it is most found by its high consumption, and represents ecotoxicological risks for aquatic organisms. In addition to this, studies that only used AC produced from agroindustrial residues were selected. On the other hand, those studies with similar pattern in AC parameters analysis were selected, as well as the type of activating agent used, analyzing the influential variables and the results obtained in each investigation. These data were organized in the form of incidence of each country in graphic format and it was also organized in the form of tables for better understanding.
A filter of the different studies published in the last 10 years was executed, because in this interval there is an increase in the number of publications referring to the subject of study with data extracted from the Scopus database (Figure 2), reflecting the current interest of the authors in finding new adsorbents with good properties for CECs removal, such as caffeine.   7 Therefore, the methodological exclusion criteria for data collection in this research were articles that do not represent high publication rate, in addition to AC with no agroindustrial or lignocellulosic origins, as well as other treatment different than adsorption process, repetitions of adsorbent materials and the non-inclusion of caffeine.

RESULTS AND DISCUSSIONS
The bibliographic focus in this research identified 131 articles with the search criteria. Those that did not deal with the topic of study were excluded, since most associated other types of activated carbon production unrelated to the topic of interest or also associated the use of technologies other than adsorption. With the chosen articles, it was decided to work on taking as representation by country the articles that work with similar analysis methods and by-product reuse purposes, or articles that worked with new adsorbent materials.
This is how different studies related to the use of agro-industrial waste for adsorbents production have been carried out, with focus in eliminating wastewater caffeine. Figure 3 shows caffeine adsorption studies that have influenced in recent years, and Table 1 shows the different authors with the AC production conditions and analysis techniques.

Asia
In China, Oni et al. (2022) used tamarind shell as material precursor for AC production, which was activated with phosphoric acid (H3PO4), to remove caffeine from wastewater. The method used for AC preparation consisted in carbonization-activation process. The characterization of the AC in this study was carried out by analyzing the moisture, ash, volatile matter, carbon, nitrogen and hydrogen contents, based on ASMT 537393 and ASMT 4239 analisys; also, metal detection was analyzed using adsorption spectrophotometer. Surface area and micropores volume analyzes were developed using the methodology proposed by Galhetas et al. (2014), according the methods applied by Dubinin & Radushkevich (1947), Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH), zero charge point and functional groups, SEM analysis and the Fourier Transform Infrared (FTIR) spectrophotometer. For the caffeine adsorption experiments, the initial conditions were: initial caffeine concentration of 5-40 g/L, pH 0-14, and residence time of 0-90 minutes. The highest adsorption capacities, 78 (model) and 72.60 mg/g (model), were obtained at pH of 6.0, contact time equal to 40 minuts and caffeine concentration equal to 10 g/L, obeying Langmuir isothermy with a determination coefficient (R 2 ) of 0.9808. The thermodynamic analysis indicated that this adsorption was an endothermic process, driven by entropy.
On the other hand in Malaysia, Danish et al. (2014), used acacia wood for AC preparation trough the experimental design method of response surface, aiming to measure adsorbent parameters. Briefly, the acacia wood samples were charred and impregnated with H3PO4 at differente concentrations, ranging from 6.45 to 48.5%, in an environment with temperature from 364 to 1036 °C and activation time from 19 to 146 minutes. In the characterization, surface area determination, pores volume and diameter, and morphological characteristics were registered, with techniques such as N2 adsorption-desorption isotherms, as well as Brunauer-Emmett-Teller (BET), Barrett-Joyner.-Halenda method (BJH) and scanning electron microscopy (SEM); finding a surface area of 957 m 2 /g, and pore volume of 0.526 cm 3 /g. This is how AC morphology was characterized by large and well-developed pores in the form of a honeycomb; this might be due to the fact of the activating agent effectiveness creating surfaces with well-developed pores. For optimization, the process was repeated with H3PO4 at a concentration of 40%, activation temperature of 900 °C and activation time of 45, which resulted in a yield of 20.3%. Yield percentage followed the quadratic model and surface area followed the linear model.

Europe
In Spain, Torrelas et al. (2015) used peach seeds to produce AC with H3PO4, resulting in a product with high mesoporosity, efficient for removal of several CECs, including caffeine. The precursor material was charred and impregnated with H3PO4and then modified by a liquid oxidation treatment with HNO3 to increase the amount of functional groups. The textural characterization of this CA was analized by N2 adsorption-desorption, using the Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) methodology, micropore analysis by Dubinin & Radushkevich (1947), electron microscopy of scanning (SEM), characterization of functional groups by Fourier transform infrared spectroscopy (FTIR), programmed thermal decomposition (TPD) to determine the zero charge point and isoelectric points, using the mass titration method proposed by Noh & Schwarz (1989).
The adsorption experiments were executed using the UV-Vis spectrophotometer with different contaminants and ultrapure water. The AC presented large surface area and materials with well-developed pores; TPD and FTIR profiles showed carboxylic, phenolic, carbonyl, and quinone functions on carbonaceous surfaces; the adsorption isotherms were type S, observed at 30 °C. It was observed that for caffeine concentrations higher than 20 mg/L, sigmoid trend was observed due to a synergistic effect, favoring caffeine molecules adsorption in solution, by adsorbate-adsorbate interactions. Caffeine adsorption on the three adsorbent samples was performed at pH of 4.8, reaching equilibrium up to 20 mg/L.
In Portugal, Mestre et al. (2014) evaluated the use of pre-treated industrial cork from renewable biomass, the carbon was chemically activated with potassium carbonate (K2CO3) and potassium hydroxide (KOH) and it was also physically activated by steam. The adsorbent characterization was developed by N2 and CO2 adsorption-desorption, pHPCZ according the method proposed by Noh & Schwarz (1989), infrared spectroscopy (FTIR), microvolume analysis of the pores proposed by Reinoso et al. (1987). ash content, scanning electron microscopy (SEM) and morphology using the field emission scanning electron microscope (FE-SEM).
Liquid phase adsorption analysis was performed with various contaminants, including caffeine. The chemically activated samples presented type I isotherms, and those subjected to physical activations, presented type I, with some type IV isotherms. Apparent surfaces ≥ 900 m 2 /g were reached. Physically, AC was able to adsorb all pharmaceutical compounds, with 40 and 90% removal efficiency for caffeine. The structure of the chemically AC presented some microporous characteristics, while the micro/mesoporous peculiarities were verified at physically AC. The Langumir adsorption capacity was equal to 174.4 mg/g for ibuprofen, and 149.5 mg/g for caffeine.
In France, Francoeur et al. (2021) used Sargassum (sp) for AC production, to remove caffeine in aqueous medium. Sargassum was charred and activated with H3PO4; then it was pyrolyzed under N2 atmosphere. The AC resulted here was characterized by N2 adsorptiondesorption isotherms, FTIR spectroscopy, X-ray photoelectrons (XPS), Boehm (1994) titration, and the point of zero charge (pHPCZ) method. The best caffeine adsorption was observed at pH 6.0, with residence time of 90 minuts and adsorbate concentration of 5000 mg/L. Langmuir's nonlinear model presented a better fit to the experimental data; adsorption isotherms equilibrium was evaluated at 25 °C, for maximum adsorption capacity of 221.61 mg/g.
In United Kingdom, Wurzer & Mazek (2021) used softwood and wheat straw for AC, in order to obtain synergistic studies with iron (Fe) catalytic activity, using ocher on solid and gaseous products from biomass pyrolysis. The samples were charred and activated by pyrolysis, driven by CO2 in solid phase and, in the phase through equipment with infrared sensor (CO, CO2, CH4 and H2), and physical activation with CO2. Biochar characterization was performed by energy dispersive spectroscopy (EDX); structures analysis was performed by (XPS); textural properties, by N2 adsorption-desorption; solids tempered density functional theory (QSDFT) was applied to calculate the specific surface area and pore size distribution, using a technique proposed by Thommes et al. (2015), the adsorption experiments were developed with different contaminants, and UV-Vis was used to analyze equilibrium concentrations.
The isothermal data fit better in Langmuir model, with high R 2 value, indicating monolayer caffeine adsorption. The maximum adsorption capacity of softwood was higher than observed in wheat straw. The softwood exhibited a qmax adsorption capacity of 135 mg/g; increased to 173 mg/g in softwood mixed with 5% ocher suspension, and 227 mg/g in softwood mixed with 10% ocher suspension (+68%). Wheat straw with 42 mg/g and 123 mg/g in wheat straw mixed with 5% ocher suspension, similar to wheat straw mixed with 10% ocher suspension (+193%). In solid pyrolysis, small changes were found from bio-oil aiming to produce non-condensable gas, with increased ocher addition. Fe doping resulted in relevant changes on pyrolysis yield distribution, with higher gas yields (+50%) and gas energy content (+40%), reducing energy costs for production. Upon physical activation completion, the ocher transformation into magnetite/maghemite occurred, resulting in activated magnetic biochars, leading to a 4-fold increase for micropollutants adsorption capacity.

North America
In the United States of America, Kante et al. (2012) used coffee bean residues to prepare AC. The residues were charred and activated with ZnCl2. Subsequently, hydrogen sulfide evaluation (H2S) was carried out to determine adsorbents penetration capacity, as well as an AC analysis without activation. In the characterization, the AC surface area was measured using Brunauer-Emmett-Teller (BET) and Barret-Joyner-Halenda (BJH) methods; micro and mesopores volume and total pore volume were also obtained. Isotherms were calculated, followed by spectroscopy, Fourier transform infrared spectroscopy (FTIR), the morphology was obtained using field emission scanning electron microscope (FE-SEM), X-ray spectroscopy (XPS), and elemental analysis. Residual inorganic material participates in salt retention was analized, a srelevant activating agent promoting the formation of a large pore volume, around 30 A°. Although the activation process could be altered, the presence of nitrogen in the precursor (caffeine) was an important asset of this specific organic residue; in addition, the nitrogen presence in functional groups, which played catalytic role for CECs oxidation.
In Canada Sanford et al. (2012) sought to evaluate chitosan and three derivatives efficacy as natural precursors. The method was based on crude chitosan pretreatment with H2O2 and another chitosan for carbonization-activation process. The structures were synthesized using organic surfactants; in sequence, chitosan was prepared under different conditions. Caffeine adsorption tests were performed for 72 hours, under neutral and acidic conditions. Sample characterization analyzes were performed using techniques such as reverse phase solid phase extraction (SPE) cartridges, Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) methods, FTIR spectroscopy, mesopores (TEM) and mass spectrometry chromatography. AC was found to have high affinity for caffeine (98% removal).
Testing under different conditions, as well as increasing the dose, had no effect on performance. Chemical modifications in chitosan included calcined mesoporous and noncalcined materials, which increased caffeine adsorption to 29% and 40%, respectively. Chitosan pretreated with hydrogen peroxide had the best performance of chitosan-based adsorbents, and achieved 46% caffeine extraction. AC showed good fit to Langmuir model, and was the most effective adsorbent for caffeine removal, extraction almost 100% in 72 hours.

South America
In Brazil, there are different investigations related to agro-industrial residues use for AC aiming CECs elimination, such as caffeine. Although it does not harm human health, significantly, in low concentrations, caffeine presence is indicative of anthropic action for aquatic matrix, related to other pollutants presence (Quadra et al., (2020);Martinez et al., (2016 ), Li et al., (2020).Thus, different authors investigated the possibility of using industrial residues for AC production. DeAlmeida et al. (2021) used açaí seeds as precursor material for AC production, caffeine removal. The AC was obtained by pyrolysis process at temperatures and heating rates, at 400 °C and 10 °C/min and 600 °C and 50 °C/min, respectively. K2CO3 was used to impregnate the precursor. AC characterization was executed trough elemental analysis, thermogravimetric analysis, Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) methods, energy dispersive spectroscopy (EDX), X-ray diffraction (XRD), transformation of Fourier (FTIR), N2 adsorption-desorption chemical hardness and global electrophilicity based on Density Functional Theory (DFT). From the increase in pyrolysis, the solid product growth at 400 °C and 10 °C/min, the chemical activation increased pores size, with final diameter pore, in average of 2.6 A°, with a surface area of 13% that corresponds to 1,150.3 m 2 /g and an increase of 24% and 28% of micro and mesopores volume. AC showed adsorption capacity of 176.8 mg/g, with a 20% to 60% caffeine removal and a good fit to pseudo-second order model. Beltrame et al. (2018) studied pineapple leaves as material precursor for AC production, with the aim of eliminate caffeine content in the aqueous medium. The material was charred and activated with H3PO4, by slow pyrolysis with low N2 flow. AC was characterized by N2 adsorption-desorption isotherms, proximal analysis, Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) methodS, scanning electron microscopy (SEM), thermogravimetric analysis, Fourier transform spectroscopy (FTIR), X-ray diffraction (XRD), Raman spectroscopy, Boehm (1994) titration, and the point of zero charge (pHPCZ) methodS; then, adsorption and thermodynamic studies were performed. In the centesimal analysis, high levels of volatile materials and low ash contents were identified in the precursor material, with desirable characteristics for AC preparation. The charred material presented a surface area of 1,031 m 2 /g, attributed to the activator and slow pyrolysis process; mesopore volume of 1.27 cm 3 /g and pores with average diameter of 5.87 A°. In functional groups analysis, the acids predominated on material surface. The pseudo second order kinetics and the Langmuir isotherm showed good fit to the data. Caffeine adsorption was 155.50 mg/g, and thermodynamic studies revealed that adsorption process occured spontaneously, were exothermic and preferentially by physisorption. Portinho et al. (2017) used grape stalk as material precursor for CA synthesis, to caffeine removal. The stalks were handled and charred, therefore, samples were prepared without activation (GS) and with activation by H3PO4 (MGS), and through synthesis process (GSAC) based on Deiana et al. (2009) methodology. The adsorbents characterization consisted in th analisys of specific surface area, using the Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) methods, determination of microporous volume by the Dubinin & Radushkevich (1947) method, N2 adsorption-desorption and the zero charge point method (pHPCZ) by the methodology proposed by Schreier & Regalbuto (2004) were applied. Then, the adsorption studies were executed varying parameters like pH, residence time and adsorbent concentration to find ideal conditions.
The percentage of caffeine removal by AC went from higher to lower adsorbent amount, not justifying the variation in its concentration. Varying the pH, the best adsorption occurred in acidic conditions, with maximum adsorption capacities of 2,300 mg/g (MGS) and 0.938 mg/g (GS) at pH of 2.0; although, for GSAC the optimum pH was 4.0, leading to an adsorption of 19.575 mg/g. The residence time and adsorbent concentration were, in this order, for 30 minuts and 15 g/L (MGS), 40 minuts and 25 g/L (GS) and 30 minuts and 1 g/L (GSAC) with correspondent adsorption capacities of 1.484, 0.932 and 19.575 mg/g for MGS, GS and GSAC, respectively, resulting in caffeine removal of 85.1%, 75.0% and 96.4%. The equilibrium was evaluated through the adsorption isotherms construction; MGS, GS and GSAC, showing good fit to the Sips model with adsorption capacities of 89.2, 129.6 and 916.7 mg/g, respectively. AC surface areas were 4.2, 6.23 and 1099.86 m 2 /g and pore volumes were 0.002, 0.003 and 0.568 cm 3 /g for MGS, GS and GSAC, respectively.
The use of agroindustrial precursors for AC production, in different studies, leads to adsorbents with promising characteristics for CECs removal in different aquatic matrices, offering commercial benefits due to its low cost, as a technique considered eco-friendly with easy access; it should be noted that, with the use of agro-industrial by-products, or discardes biological metherials, it is possible to reduce the pressure of waste discarded in landfills. This is how the use of different adsorbents, in previously studies, led to 100% removal capacities, using chitosan as AC precursor, for caffeine concentration of 250 mg/L in 72 hours, data that was fitted to Langmuir model. However, other previously adsorbents showed good results for caffeine adsorption, representing economical and sustainable materials for AC production, as well as providing good structural and porous characteristics, making them potential materials for CECs removal. Therefore, the use of these industrial by-products in the conversion for effective precursors, should be appreciated, increasing their added market value as new raw materials for wastewater treatment.

FINAL CONSIDERATIONS
The present work allowed to identify different agro-industrial residues used for activated carbon production for caffeine removal in water bodies. This is how different authors worldwide are developing new technologies for CECs treatment present in hydric resources, looking for economical and sustainable alternatives. Therefore, future studies should focus on new industrial residues use found in large quantities, and that have not yet been used for activated carbon production, thus analyzing the interaction between the contaminants and adsorbent, trough different methods and optimal conditions for contaminants removal, contributing to new economic and efficient technologies implementation for removal contaminants that affect water quality.
In this investigation, it was observed that carbons under different activation processes, through chemical agents and some through physical activation, presented excellent characteristics, such as high porosity, different functional groups and good stability, which provided spontaneous caffeine adsorption and other emerging contaminants by activated carbon.
In this way, was also observed certain missing data related to activated carbon production articles, such as ideal conditions if the product would be produced at large scale, which is an important factor since it would allow us to know the adequate environmental for massive activated carbons production. This is how the present research is, also, limited to the general published data. On the other hand, it can be recommended that, in future, studies for new adsorbent materials production, research must be developed at laboratory level, but also on a real scale, that is, their application in treatment stations that receive different types of emerging pollutants, among which caffeine is found and, thus, conclude if activated carbon production worked correctly under conditions carried out at laboratory level. Finally, it is also necessary for future work in the social field seeking to create awareness in the community, and highlight the effects of emerging pollutants on natural environment, creating public policies for pollutants control and monitoring, looking for minimize their releases to water bodies and other vulnerable matrices.