In order to aggregate value to the grape stem (wastes), this research aim was to increase the adsorption capacity of Cd2+ by chemical modifications on grape stems. The grape stems were milled and sieved, resulting in the biosorbent, which was used for the chemical modifications resulting in E. H2O2, E. H2SO4 and E. NaOH. These were characterized by such means as its pHPZC, Fourier transform-infrared (FTIR) spectroscopy, porosimetry, thermal stability and scanning electron microscopy. The ideal adsorption dose, the pH influence on adsorption, kinetics, equilibrium and thermodynamics studies were carried out. The FTIR spectroscopy suggests the occurrence of carboxyl, amine, and phenolic acting in Cd2+ sorption. The modification on grape biomass caused small increase in pore volume and specific surface area. The grape-based adsorbents have similar thermal stability, with irregular appearance and heterogeneity. 5.0 g kg−1 is the best adsorption dose. The modified adsorbents exhibited increase in Cd2+ removal of 66% for E. NaOH, 33% for E. H2O2 and 8.3% for E. H2SO4. The use of grape stem as adsorbent is an attractive alternative, because its wastes have great availability, low cost and great potential for metal adsorption processes.

One of the major issues related to environmental problems is the preservation of freshwater in quantity and quality for current and future human consumption (Schwantes et al. 2018).

Several techniques are used for remediation of contaminated water; however, these techniques do not always present economic feasibility. Among the conventional methods mostly used for the wastewater treatment are chemical precipitation, oxidation or reduction, filtration, electrochemical treatment, membrane separation processes and solid phase extraction. Some of these technologies are infeasible due to technical or economic unfeasibility, especially when referring to the removal of toxic metals, since these are usually present in large volumes of water and in relatively low concentrations (Schwantes et al. 2016).

Among the toxic metals, cadmium (Cd) deserves attention because of its high toxicity and deleterious effects on humans and the environment, even at low concentrations. One of the main problems associated with Cd is its final destination in the food chain, as it can reach the soil or the air, by the burn of municipal waste or fossil fuels, thus polluting the environment and causing damage to the ecosystem (Nacke et al. 2017).

In humans, the inhalation of this heavy metal may cause problems in respiratory tract and kidneys, in the case of oral intoxication, when a significant amount of Cd is ingested, may generate an immediate poisoning and damage to the liver and kidneys; already in contact poisoning, genetic alterations may occur (Kim et al. 2018).

There are many modern techniques being developed for cadmium removal purposes, such the use of nanoscale zero valent iron (nZVI) (Dong et al. 2017), or reverse osmosis and nanofiltration membranes (Kheriji et al. 2015).

However, one of the most promising techniques in the removal of toxic metals from water is the use of natural adsorbents. According to Gonçalves et al. (2016), regarding the conventional water treatment, the use of biomass as adsorbents exhibit great advantages, due to its ability to accumulate contaminants, to withstand several cycles of sorption and desorption, and also because they require little processing and are abundant in industry and agriculture, being considered materials of low cost.

Studies report the use of chemical modifications, whose purpose is to introduce functional groups into the structure of these adsorbents, or to increase their porosity and adsorption capacity (Dos Santos et al. 2010, 2011; Schwantes et al. 2016).

According to Hokkanen et al. (2016), methods of adsorbent modifications include physical modifications, chemical modifications and other methods. In order to make the biomass even more efficient, they suggest that chemical modifications should be carried out, for increasing adsorptive capacity.

In this scenario, the grape processing industry generates a large amount of residual biomass, that consists of bark, seeds, stems and, in smaller quantities, fruit pulp. Grape is produced either for consumption of grapes in natura or for the production of wine.

China leads the world production with 14,763,000 metric tons of grapes, while Brazil and Portugal produced, in 2016, respectively, 984,481 and 773,904 metric tons. The world grape production is around 77,438,929 metric tons (FAO 2016).

According to de Mello & da Silva (2014), the processing of grapes in winemaking and juice production generates approximately 20–25% of its weight in grapes solid wastes, especially grape stem. Thus, it can be estimated that the Brazilian, Portuguese and world annual production of grape solid wastes are respectively 246,120, 193,476 and 19,359,732 metric tons, respectively.

This research aimed to aggregate value to an agro-industrial waste, which occur worldwide, the grape stem, transforming it into adsorbents with high capacity of Cd2+ removal, and also to study the adsorption mechanisms of Cd2+ retention into the biomass.

Obtaining the adsorbent materials

The grape stems were obtained in the agroindustry of wine production located at the School of Agriculture (ISA), located in Lisbon, Portugal. The stems were dried at 60 °C for 48 h, crushed and sieved in order to standardize the particle size (material retained between 14 and 65 mesh), resulting in biosorbent of stem or E. in natura.

Chemical modifications were applied to E. in natura, aiming to increase favorable characteristics for the adsorption of Cd2+, such as surface contact area, porosity and the number of adsorption groups. For that purpose, the biosorbent of grape stems (E. in natura) was treated separately with H2O2, H2SO4 or NaOH 0.1 mol L−1 for 60 °C and 6 h (Schwantes et al. 2015, 2016). After that process, the residual treated biomass was washed with ultrapure water for the removal of any residue from the modifying solutions, and again dried at 60 °C for 48 h, resulting in the modified adsorbents called E. H2O2, E. H2SO4 and E. NaOH.

Characterization of the biosorbent and modified adsorbents

The point of zero charge (pHPZC) of adsorbents, which refers to the pH at which the resultant of the surface charges of adsorbent is zero, was determined according to adaptations on Mimura et al. (2010) method, with 4 h of shaking samples.

The identification of functional groups of the biosorbents was performed by Fourier transform-infrared spectrophotometry (FTIR), with Shimadzu Infrared Spectrophotometer FTIR-8300 Fourier Transform, with a resolution of 4 cm. The morphological characterization of the adsorbents was evaluated by scanning electron microscopy (SEM), using a FEI Quanta 200 microscope (The Netherlands) operating at 30 kV voltage.

The surface analysis, size and pore volume was performed using Quantachrome Nova 1200e equipment. To this end, 500 mg of material were heated to 200 °C under vacuum for approximately 4 h. Subsequently, the processes of adsorption and desorption of N2 were performed. The surface size and pore volume were calculated using the standard Brunauer, Emmett and Teller (BET) (Brunauer et al. 1938) and a pore size was obtained using the method of Barrett–Joyner–Halenda (BJH) (Barrett et al. 1951).

The thermal analysis was performed using a TGA 4000 Perkin Elmer thermogravimetric analyzer, where the materials were heated at a temperature ranging from 30 °C to 900 °C under heating rate 10 °C min−1 in nitrogen atmosphere.

The morphology of all adsorbents was evaluated by SEM using a JEO JSM 6360-LV microscope, equipped with dispersive energy microscopy.

Biosorption experiments

The adsorption studies were performed with solutions fortified with Cd2+, prepared from salts of cadmium nitrate (Cd(NO3)2 4H2O P.A. ≥99.0% Sigma-Aldrich) and ultrapure water.

Studies aiming to determine the ideal adsorbent dose and the pH influence on cadmium adsorption were carried out. For this, a rotational composite central design was used, with mass values studied from 250 to 1,250 mg (proportion of 5.0 to 25 g L−1 of Kg m−3). Experimentally, 50 mL of solution containing Cd2+ at 10 mg L−1 was added to Erlenmeyers with increasing masses of each adsorbent, which were then arranged in a thermostated Dubnoff system with constant stirring at 200 rpm and 25 °C for 1 h.

After the sorption process, the samples were centrifuged and aliquots were taken to determine the concentrations of remaining Cd2+ by EAA/Flame (Welz & Sperling 1999). The obtained results were evaluated by multivariate analysis using Statistica 5.0.

The adsorption kinetics of Cd2+ by grape adsorbents was evaluated by placing the adsorbent biomass in contact with solutions of Cd2+ in increasing amounts of time. For that, 200 mg of each adsorbent was separately placed in an Erlenmeyer flask with 50 mL of Cd2+ solution at 10 mg L−1, pH 5.00, 25°C and contact time from 10 to 180 min. After the centrifugation, the remained concentration of Cd2+ was determined by flame atomic absorption spectroscopy (FAAS) and the results linearized by the models of pseudo-first-order (Lagergren 1898), pseudo-second-order (Ho & McKay 1999), Elovich (Roginski 1948) and intraparticle diffusion (Weber & Morris 1963).

Adsorption isotherms were constructed using 200 mg of each adsorbent in Erlenmeyer flask containing 50 mL of Cd2+ solution at increasing concentrations (5 to 200 mg L−1) at pH 5.00 and 25 °C at 200 rpm in thermostated Dubnoff system for 1 h. After centrifugation, the remaining concentration of Cd2+ was determined by FAAS, and the results linearized by Langmuir (1916), Freundlich (1906), Dubinin & Radushkevich (1947) and Sips (1948).

Thermodynamics parameters were also evaluated, using 200 mg of each adsorbent in an Erlenmeyer flask with 50 mL of Cd2+ solution at 10 mg L−1, pH 5.00, 200 rpm with increasing temperatures of 15, 25, 35, 45 and 55 °C for 1 h. After the centrifugation, the remaining concentration of Cd2+ was determined by FAAS for the linearization and estimative of ΔG, ΔH and ΔS.

All equations and models used in this research are exhibited in Table 1 (supplementary material, available with the online version of this paper).

Table 1

SSA, volume and pore diameter for grape stem adsorbents

BiosorbentSSA (m2 g−1)Pore volume (cm3 g−1)Pore diameter (nm)
E. in natura 0.148 9.43 × 10−4 3.94 
E. H2O2 0.470 18.49 × 10−4 1.55 
E. H2SO4 0.143 23.79 × 10−4 64.73 
E. NaOH 0.238 12.93 × 10−4 1.93 
Pinus barks (H2O2)a 0.760 56.00 × 10−4 14.66 
Cassava barks (H2SO4)b 0.464 17.9 × 10−4 3.29 
Cd2+ ions hydration radiusc 0.275 nm 
BiosorbentSSA (m2 g−1)Pore volume (cm3 g−1)Pore diameter (nm)
E. in natura 0.148 9.43 × 10−4 3.94 
E. H2O2 0.470 18.49 × 10−4 1.55 
E. H2SO4 0.143 23.79 × 10−4 64.73 
E. NaOH 0.238 12.93 × 10−4 1.93 
Pinus barks (H2O2)a 0.760 56.00 × 10−4 14.66 
Cassava barks (H2SO4)b 0.464 17.9 × 10−4 3.29 
Cd2+ ions hydration radiusc 0.275 nm 

Characterization of adsorbents

The values obtained for the pHPZC (Figure 1) of adsorbents are 4.28 to E. in natura, 4.09 for E. H2O2, 2.41 for E. H2SO4 and 7.10 for E. NaOH. The net charge on the adsorbent surface is positive when the pH of solution is inferior to the pHPZC value, and negative, when the pH of solution is superior to pHPZC value (Silveira Neta et al. 2012).

Figure 1

pHPZC (above) and FTIR spectroscopy, for grape stem adsorbents (below).

Figure 1

pHPZC (above) and FTIR spectroscopy, for grape stem adsorbents (below).

The change in pHPZC occurred according to the acidification or alkalization from each solution, causing protonation, deprotonation or hydroxylation of chemical groups of grapes biomass. Thus, when the pHenvironment > pHPZC, the surface of the adsorbent is electronegative, favoring the adsorption of Cd2+; however, if the pHenvironment < pHPZC, the surface of the adsorbent is electropositive, in this state, the H+ ions compete with Cd2+, repelling or reducing the surface of adsorption (Schwantes et al. 2016).

It is possible to observe vibrational stretching at 3,395 to 3,422 cm−1, 2,919 to 2,922 cm−1, 1,735 to 1,738 cm−1, 1,614 to 1,622 cm−1, 1,513 to 1,522 cm−1, 1,441 to 1,443 cm−1, 1,390 to 1,400 cm−1, 1,272 cm−1, 1,060 cm−1 and 408 to 435 cm−1 (Figure 1).

The strong and broadband at 3,395 to 3,422 cm−1, may result from a stretching asymmetric hydroxyl OH groups present in water and cellulose, and the symmetrical stretching of N-H bonds associated with primary (aliphatic and aromatic) and secondary amines (Barbosa 2007).

The peaks among 2,919 to 2,922 cm−1 can be referred to the stretching vibration of C-H bonds, and may be symmetrical or asymmetrical, due to the presence of alkanes, acyl, aliphatic acids and lipids (Abidi et al. 2014).

The peaks between 1,735 to 1,738 cm−1, tend to be axial stretching of C = O bonds of aldehydes and esters groups, originating from polysaccharide, lipid and hemicellulose (Barbosa 2007).

The vibrational waves between 1,614 to 1,622 cm−1, may refer to C-C phenyl ring stretch, but can also be associated with nucleic acids peaks due to carbonyl and carboxyl groups (Schulz & Baranska 2007). The peaks among 1,513 to 1,522 cm−1, may be related to the N-H group, referring to secondary amide, probably of protein degradation.

The stretching vibrational around 1,441 to 1,443 cm−1, tend to angular deformation of C-H2 or C-H3, related to lipids and fatty acids, and aromatic compounds, respectively (Barbosa 2007).

The wave in 1,390 cm−1 can be related to carbon particles, as the peaks present in E. in natura and E. H2SO4, the 1,399 cm−1, arise mainly from the vibrational modes of methyl and methylene groups of proteins, lipids and amino groups being a symmetric bending mode of CH3. The adsorbent E. NaOH exhibits a peak at 1,440 cm−1, derived from symmetric stretching vibration of the COO group linked to fatty acids and amino acids, and symmetrical stretching of CH3 protein.

The stretching vibration at 1,060 cm−1 may be related to the group CO deoxyribose from nucleic acids or cellulose and polysaccharide degradation bonds such as CO, CC and OCH (Barbosa 2007).

Importantly, new groups, not previously seen in adsorbent E. in natura can be observed, such as vibrational stretch in 1,271 cm−1, found only in E. H2SO4, which refers to a mass balance in the C-H bond (Schulz & Baranska 2007). A peak at 1,514 cm−1 is possibly related to the presence of a carotenoid structure that may be a cellular pigment.

The chemical modifications with H2O2 and NaOH caused a small increase in the specific surface area (SSA) of grape stem (Table 1). In addition, it is observed that the pore volume increased significantly when compared to the biosorbent (E. in natura), with increases of 1.9× for E. H2O2, 2.5× for E. H2SO4 and 1.8 for E NaOH.

The adsorbent E. H2SO4 exhibits the higher pore diameter (64.73 nm) which, according to IUPAC (1972) constitutes a macroporous material, while E. in natura is generally mesoporous (∼2–50 nm) and E. H2O2 predominantly microporous (0–2 nm). These results are similar to other biosorbent materials (1–2) after simple chemical modification as Pinus barks treated with H2O2 and H2SO4.

The thermogravimetric analysis (TG and DTG) was performed to verify the thermic stability of grape stems in natura and after the treatments with H2O2, H2SO4 and NaOH. The TG and DTG curves can be seen in Figure 2.

Figure 2

TG and DTG curves for the grape stem adsorbents.

Figure 2

TG and DTG curves for the grape stem adsorbents.

The adsorbent E. in natura (Figure 2) exhibited three events, the first at 76 °C with mass loss of 6%, possibly related to the loss of water and volatiles. In the second event, with 211 °C and with mass loss of 12%, with the decomposition of hemicellulose. The third event is exhibited at 325 °C, with a mass loss of 62%, possibly related to cellulose breakage (Melzer et al. 2013).

For adsorbent E. H2O2, it is exhibited the first mass loss at 64 °C, with a reduction of 9% of mass, possibly due to loss of water. The second event occurred at 352 °C, due to the degradation of cellulose, lignin, hemicelluloses, minerals and condensed tannins, reaching 60% of its mass.

For adsorbent E. H2SO4, the first event occurred at 75 °C, with mass loss of 8%, possibly due to volatiles and water content. Another event is exhibited at 321 °C, with decomposition of 63% of mass, due to the breakdown of carbohydrates, such as hemicellulose and cellulose (Rambo et al. 2015).

The adsorbent E. NaOH exhibited three distinct events, the first at 76 °C, possibly related to the evaporation of water and volatile compounds, resulting in mass loss of 8%. The second mass loss (60%) occurred at 298 °C, with probable degradation of compounds such as cellulose and hemicellulosis. The third event occurred at 880 °C, with 25% of mass loss, probably due to the incineration of the remaining material, resulting in ash formation (Melzer et al. 2013).

The SEM at Figure 3 illustrates, for E. in natura, an irregular and heterogeneous surface. The SEM for E. H2O2, which resulted from a modification with a powerful oxidizing agent, reveals an area with many cavities, and heterogeneous, spongy and tubular aspects.

Figure 3

SEM for adsorbents E. in natura at 600 (a) and 2,400× (b), E. H2O2 at 600 (c) and 2,400× (d), E. H2SO4 at 600 (e) and 2,400× (f), and E. NaOH at 600 (g) and 2,400× (h).

Figure 3

SEM for adsorbents E. in natura at 600 (a) and 2,400× (b), E. H2O2 at 600 (c) and 2,400× (d), E. H2SO4 at 600 (e) and 2,400× (f), and E. NaOH at 600 (g) and 2,400× (h).

The micrograph for E. H2SO4, which resulted from a strong acid treatment, known also as a powerful dehydrating agent, exhibits a heterogeneous surface, though with numerous cavities of spongy appearance.

For E. NaOH, which was treated with a strong base, known for its corrosion and solubilization of numerous organic compounds, the SEM exhibits a heterogeneous surface, however irregular, also with numerous cavities.

As noted, all the obtained adsorbents exhibit heterogeneous morphologic characteristics that resemble the biomass origin (E. in natura); however, each applied chemical treatment generated adsorbents with particular characteristics, according to the employed modifying agent.

According to Nacke et al. (2016, 2017) irregular appearance, heterogeneity, many cavities, are generally characteristics that generate high SSA, which are usually characteristic of adsorbents with high adsorption capacity of ions in solution.

Rubio et al. (2015) in evaluating crambe biosorbents obtained SEM with an irregular and heterogeneous structure, which, according to the authors, was experimentally demonstrated to be a primordial characteristic for high adsorption rates of Pb2+ ions in the aqueous medium.

Experiments with Cd2+ adsorption

Significant differences were found at 1% for adsorbent E. in natura, E. H2O2, E. H2SO4 and E. NaOH regarding the source of variation ‘adsorbent mass’ (for linear and quadratic models), indicating that the amount of adsorbent mass influences the process of adsorption of Cd2+ (Table 2).

Table 2

Summary of the analysis of variance (ANOVA) for tests involving adsorbent mass and pH of Cd2+ solution

Sources of variationDegrees of freedomAverage squares
E. in naturaE. H2O2E. H2SO4E. NaOH
Mass (L) 0.658324a 1.059795a 0.563293a 1.886175a 
Mass (Q) 0.171143a 0.201129a 0.140963a 0.409658a 
pH (L) 0.002291ns 0.010476ns 0.078108b 0.000035ns 
pH (Q) 0.00338ns 0.002326ns 0.030001ns 0.001817ns 
Mass × pH 0.002113ns 0.002002ns 0.01655ns 0.00001ns 
Residue 0.00498 0.002645 0.006695 0.010799 
Total 11     
Sources of variationDegrees of freedomAverage squares
E. in naturaE. H2O2E. H2SO4E. NaOH
Mass (L) 0.658324a 1.059795a 0.563293a 1.886175a 
Mass (Q) 0.171143a 0.201129a 0.140963a 0.409658a 
pH (L) 0.002291ns 0.010476ns 0.078108b 0.000035ns 
pH (Q) 0.00338ns 0.002326ns 0.030001ns 0.001817ns 
Mass × pH 0.002113ns 0.002002ns 0.01655ns 0.00001ns 
Residue 0.00498 0.002645 0.006695 0.010799 
Total 11     

ns, not significant.

aSignificant at 1%.

bSignificant at 5%.

Significant difference at 5% was found for E. H2SO4, for the ‘pH of Cd2+solution’ (for linear and quadratic models), indicating that for this specific adsorbent, the pH range has influence in the adsorption process of Cd2+. For other adsorbents, no significant differences for pH ranges were found, suggesting that in the evaluated range (3.00 to 7.00), the pH of the solution does not cause influence on Cd2+ adsorption.

This result is extremely favorable, because it demonstrates grape stem adsorbents can remedy Cd2+ from aquatic environments with high removal capacity within the pH range of 3.00 to 7.00, without requiring any adjustment to the pH of the liquid medium (Figure 4).

Figure 4

Surface of response for adsorbed amount at equilibrium (Qeq) versus pH of Cd2+ solution versus mass of grape stem adsorbents. x = adsorbent mass, from 250 to 1,250 mg (5.0 to 25 g L−1 or Kg m3), y = pH of the solution containing Cd2+ at 10 mg L−1, z = quantity of adsorbed Cd2+ at equilibrium (mg g−1). E. in natura z = 1.20239 − 0.00254378x + 0.0000013077x2 + 0.161047y − 0.0114731y2 − 0.000045822xy R2: 98.1%. E. H2O2 z = 1.58617 − 0.00293286x + 0.000001417x2 + 0.154181y − 0.00951766y2 − 0.000044605xy R2: 98.8%. E. H2SO4 z = −0.0634065 − 0.00188951x + 0.0000011868x2 + 0.507761y − 0.0341814y2 0.00012824xy R2: 94.5%. E. NaOH z = 2.61556 − 0.00442359x + 0.0000020232x2 + 0.083276y − 0.00841089y2 + 0.000003090xy R2: 97.9%.

Figure 4

Surface of response for adsorbed amount at equilibrium (Qeq) versus pH of Cd2+ solution versus mass of grape stem adsorbents. x = adsorbent mass, from 250 to 1,250 mg (5.0 to 25 g L−1 or Kg m3), y = pH of the solution containing Cd2+ at 10 mg L−1, z = quantity of adsorbed Cd2+ at equilibrium (mg g−1). E. in natura z = 1.20239 − 0.00254378x + 0.0000013077x2 + 0.161047y − 0.0114731y2 − 0.000045822xy R2: 98.1%. E. H2O2 z = 1.58617 − 0.00293286x + 0.000001417x2 + 0.154181y − 0.00951766y2 − 0.000044605xy R2: 98.8%. E. H2SO4 z = −0.0634065 − 0.00188951x + 0.0000011868x2 + 0.507761y − 0.0341814y2 0.00012824xy R2: 94.5%. E. NaOH z = 2.61556 − 0.00442359x + 0.0000020232x2 + 0.083276y − 0.00841089y2 + 0.000003090xy R2: 97.9%.

As can be seen at Figure 4 and in Table 1, the pH ranges do not cause significant variation in Cd2+ adsorption, except for adsorbent E. H2SO4. In general, the greater Cd2+ removal rates (measured by the adsorbed amount Qeq or Qads), were obtained for adsorbents masses close to 200 mg (ratio of 5.0 g L−1 or Kg m3).

In this way, it can be estimated that at least 50% of the grape stem produced in Brazil in 2016 were used in the manufacture of biosorbents, about 123,000 tons of adsorbents would be produced, with the possibility of removing an enormous amount of cadmium from contaminated water. As already mentioned, the availability of plant waste is one of the factors that makes its use so attractive.

According to Meneghel et al. (2013) and Rubio et al. (2013), the relation between adsorbent mass and volume of contaminated water is critical. In certain cases, a decrease in the adsorbed amount occurs due to the formation of agglomerates, which will reduce the total surface area and therefore the number of available active sites.

Grape stem materials exhibited the following Cd2+ adsorption capacity: E. in natura 1.20, E. H2O2 1.60, E. H2SO4 1.20 and E. NaOH 2.00 mg g−1, i.e. an increase of the adsorption capacity of 33% for E. H2O2 and 66% for E. NaOH, when compared to the biosorbent E. in natura.

Although only tested in laboratory scale, several authors have already reported the use of modified biosorbents in bed columns, aiming the removal of metals such as Cd from industrial effluents (Ahmad & Haydar 2016), various heavy metals from wastewater (Abdolali et al. 2017), Ni2+ and Cu2+ (Barquilha et al. 2017), and many others.

In 20 min of contact-time, the adsorption system enters in chemical equilibrium, not exhibiting great variations on the Cd2+ adsorption rate after this time interval (Figure 5 left), i.e. the adsorption of Cd2+ by grape stem materials is a fast process.

Figure 5

(left) Kinetics of Cd2+ adsorption by grape stem materials. Experimental conditions: Initial concentration of Cd2+ of 10 mg L−1, 200 rpm, 25 °C, pH of 5.00, mass of adsorbent/volume of solution relation of 5.0 g L−1. (right) Cd2+ removal percentage at increasing concentrations. Experimental conditions: 200 rpm, 25 °C, pH of 5.00, mass of adsorbent/volume of solution relation of 5.0 g L−1, 60 min of contact time.

Figure 5

(left) Kinetics of Cd2+ adsorption by grape stem materials. Experimental conditions: Initial concentration of Cd2+ of 10 mg L−1, 200 rpm, 25 °C, pH of 5.00, mass of adsorbent/volume of solution relation of 5.0 g L−1. (right) Cd2+ removal percentage at increasing concentrations. Experimental conditions: 200 rpm, 25 °C, pH of 5.00, mass of adsorbent/volume of solution relation of 5.0 g L−1, 60 min of contact time.

Elovich, pseudo-first-order and intraparticle diffusion models don't fit well to the experimental data (Table 3), however, good adjustments (R2 values) are exhibited for pseudo-second-order model, suggesting the occurrence of Cd2+ chemisorption by E. in natura, E. H2O2, E. H2SO4 and E. NaOH (Ho & McKay 1999). In addition, the calculated adsorption capacity predicted by pseudo-second (Qeq(cal.)) approach the experimental values (Qeq(exp.)).

Table 3

Parameters of pseudo-first-order, pseudo-second-order, Elovich and intraparticle diffusion for the removal of Cd2+ by E. in natura, E. H2O2, E. H2SO4 and E. NaOH

Models/AdsorbentsE. in naturaE. H2O2E. H2SO4E. NaOH
Pseudo-first-order 
K1 (min−1−0.0031 −0.0037 −0.0094 −0.0149 
Qeq(cal.) (mg g−10.5961 0.2209 0.2250 0.2845 
R2 0.914 0.847 0.927 0.853 
Pseudo-second-order 
K2 (g mg−1 min−10.1740 0.1877 −0.3335 0.1584 
Qeq(cal.) (mg g−11.4511 1.9170 1.3777 2.5176 
 R2 0.997 0.997 0.998 1.000 
Elovich 
a (mg g−1 h−11.0891 1.5396 1.1460 1.8762 
b (g mg−10.0596 0.0606 0.0696 0.1202 
R2 0.939 0.642 0.930 0.892 
Qeq(exp.) (mg g−11.3823 1.8133 1.4302 2.3511 
AdsorbentsIntraparticle diffusionLine ALine B
E. in natura Kid (g mg−1 min−1/20.0308   
Ci (mg g−11.0350   
R2 0.905   
E. H2O2 Kid (g mg−1 min−1/20.1069 0.0118  
Ci (mg g−11.2290 1.6776  
R2 0.937 0.735  
E. H2SO4 Kid (g mg−1 min−1/20.0223   
Ci (mg g−11.2505   
R2 0.9594   
E. NaOH Kid (g mg−1 min−1/20.0248   
Ci (mg g−12.1707   
R2 0.971   
Models/AdsorbentsE. in naturaE. H2O2E. H2SO4E. NaOH
Pseudo-first-order 
K1 (min−1−0.0031 −0.0037 −0.0094 −0.0149 
Qeq(cal.) (mg g−10.5961 0.2209 0.2250 0.2845 
R2 0.914 0.847 0.927 0.853 
Pseudo-second-order 
K2 (g mg−1 min−10.1740 0.1877 −0.3335 0.1584 
Qeq(cal.) (mg g−11.4511 1.9170 1.3777 2.5176 
 R2 0.997 0.997 0.998 1.000 
Elovich 
a (mg g−1 h−11.0891 1.5396 1.1460 1.8762 
b (g mg−10.0596 0.0606 0.0696 0.1202 
R2 0.939 0.642 0.930 0.892 
Qeq(exp.) (mg g−11.3823 1.8133 1.4302 2.3511 
AdsorbentsIntraparticle diffusionLine ALine B
E. in natura Kid (g mg−1 min−1/20.0308   
Ci (mg g−11.0350   
R2 0.905   
E. H2O2 Kid (g mg−1 min−1/20.1069 0.0118  
Ci (mg g−11.2290 1.6776  
R2 0.937 0.735  
E. H2SO4 Kid (g mg−1 min−1/20.0223   
Ci (mg g−11.2505   
R2 0.9594   
E. NaOH Kid (g mg−1 min−1/20.0248   
Ci (mg g−12.1707   
R2 0.971   

Many other studies aiming the use of biosorbents for metal removal found excellent adjustments for pseudo-second-order: Schwantes et al. (2016) using cassava barks for adsorption of Cd2+, Pb2+ and Cr3+; Schwantes et al. (2015) with crambe pie for Cd2+, Pb2+ and Cr3+ removal; Schwantes et al. (2018) with Pinus barks aiming Cd2+, Pb2+ and Cr3+ removal; Ngabura et al. (2018) using durian peels for Zn2+ adsorption; Nacke et al. (2016, 2017) using Jatropha curcas L. biosorbents for Cu2+ and Zn2+ removal.

These results demonstrates that numerous lignocellulosic-based adsorbents, when in contact with heavy metals in liquid phase, undergo chemisorption. In all these cited biosorbents is recurrent the existence of carboxylic, hydroxyl, methoxy, and phenolic groups that are potentially active in metals binding (Hokkanen Bhatnagar & Sillanpaa 2016).

For pseudo-first-order, K1 assumes negative values, indicating that the concentration of solutes in solution decreases with increasing time (Table 3). However, the observed data fit to pseudo-second-order model. The pseudo-second-order kinetics describes well the processes of chemical adsorption, involving donation or exchange of electrons between the adsorbate and the adsorbent, as covalent and ion exchange forces (Ho & McKay 1999). In this type of adsorption, the molecules are not attracted by all the points of the surface of the solid, but specifically to the active sites, forming a single layer initially, and other layers may be formed by physisorption.

A mechanistic study was carried out to evaluate the diffusion of the adsorbate using the Morris–Weber model for the studied system and, despite the low porosity of the adsorbent material (Table 1), there is an occurrence of intraparticle diffusion. Cd2+ ions hydration radius of 0.275 nm (Tagliaferro et al. 2011), i.e. lower than the average pore diameter at the surface of the adsorbents, allows the diffusion of Cd2+ ions into the adsorbent pores. In spite of this fact, the low obtained adjustments (R2) suggest that intraparticle diffusion is not the governing mechanism of the adsorption process in this case.

According to equilibrium tests, E. NaOH was the adsorbent with the higher values of Cd2+ removal (Figure 5 right), with an average of 70% of metal removal in the interval of 5 to 200 mg L−1 of Cd2+ in solution.

Grape stem adsorbents presented good fit (R2) to Langmuir model (Table 4), which suggests adsorption of Cd2+ in monolayers. It should also be noted that the adsorption process predicted by Langmuir model was favorable since RL values ranged from 0 to 1, for E. in natura, E. H2SO4 and E. NaOH (Langmuir 1916).

Table 4

Linear parameters of Langmuir, Freundlich, D-R and Sips for adsorption of Cd2+ by E. in natura, E. H2O2, E. H2SO4 and E NaOH

ParametersE. in naturaE. H2O2E. H2SO4E. NaOH
Langmuir Qm (mg g−1) 1.785 4.591 2.881 14.923 
KL (L mg−1) 1.592 −0.036 0.441 0.000 
RL 0.003 −0.159 0.011 0.988 
R2 0.993 0.996 0.999 0.993 
Freundlich Kf (mg g−1) 0.061 3.128 1.886 7.797 
N 1.358 9.573 12.367 5.074 
R2 0.959 0.894 0.910 0.992 
D-R Qd (mol L−1) 9.017E-05 4.762E-05 4.885E-05 2.495E-04 
E (KJ mol−1) 8.853 36.140 32.308 17.961 
R2 0.984 0.580 0.604 0.997 
Sips n 0.607 0.818 0.720 1.795 
Ks (L mg−1) 0.070 0.222 0.106 1.578 
R2 0.995 0.885 0.999 0.999 
ParametersE. in naturaE. H2O2E. H2SO4E. NaOH
Langmuir Qm (mg g−1) 1.785 4.591 2.881 14.923 
KL (L mg−1) 1.592 −0.036 0.441 0.000 
RL 0.003 −0.159 0.011 0.988 
R2 0.993 0.996 0.999 0.993 
Freundlich Kf (mg g−1) 0.061 3.128 1.886 7.797 
N 1.358 9.573 12.367 5.074 
R2 0.959 0.894 0.910 0.992 
D-R Qd (mol L−1) 9.017E-05 4.762E-05 4.885E-05 2.495E-04 
E (KJ mol−1) 8.853 36.140 32.308 17.961 
R2 0.984 0.580 0.604 0.997 
Sips n 0.607 0.818 0.720 1.795 
Ks (L mg−1) 0.070 0.222 0.106 1.578 
R2 0.995 0.885 0.999 0.999 

In this context, the adsorbent E. NaOH, which had the best value of Qm, 14.9 mg g−1, i.e. with an increase of 8.4× in the precursor biosorbent. Good fit was also obtained by Freundlich for E. NaOH, suggesting the simultaneous occurrence of mono and multilayer Cd2+. In this case, n is higher than 1, indicating, for E. NaOH, active sites of high reactivity (Schwantes et al. 2018).

The Qm obtained for E. NaOH (14.9 mg g−1) is higher than obtained by Lo et al. (2012) using bamboo active coal (0.67 mg g−1 for Pb2+) and obtained by Schwantes et al. (2018) using Pinus biosorbent (10.83 mg g−1 for Cd2+).

E. in natura and E. NaOH exhibit good fit for D-R, with E (sorption energy) values greater than 8 KJ mol−1, indicating Cd2+ chemisorption (Dubinin & Radushkevich 1947).

The values of ΔH indicate endothermic reactive systems (ΔH > 0) or exothermic (ΔH < 0) (Wan Ngah & Fatinathan 2010). The results of Table 5 indicate that the adsorption of Cd2+ by E. in natura and E. H2O2 is endothermic, whereas the adsorption by E. H2SO4 and E. NaOH is exothermic.

Table 5

Thermodynamic parameters for adsorption of Cd2+ by E. in natura, E. H2O2, E. H2SO4 and E. NaOH

AdsorbentsTemp. °C Qeq (mg g−1)ΔG (KJ mol−1)ΔH (J mol−1)ΔS (J mol−1 K−1)R2
E. in natura 15 0.00 45.44 15.58 − 103.71 0.93 
25 0.11 46.48 
35 0.17 47.52 
45 0.14 48.56 
55 0.10 49.59 
E. H2O2 15 1.51 9.29 5.67 12.57 0.98 
25 1.57 9.41 
35 1.61 9.54 
45 1.56 9.67 
55 1.67 9.79 
E. H2SO4 15 0.92 −30.05 − 29.71 1.20 0.87 
25 0.93 −30.05 
35 0.99 −30.05 
45 1.03 −30.05 
55 1.02 −30.05 
E. NaOH 15 2.40 −33.68 − 21.71 41.55 0.91 
25 2.39 −33.68 
35 2.37 −33.68 
45 2.46 −33.68 
55 2.38 −33.68 
AdsorbentsTemp. °C Qeq (mg g−1)ΔG (KJ mol−1)ΔH (J mol−1)ΔS (J mol−1 K−1)R2
E. in natura 15 0.00 45.44 15.58 − 103.71 0.93 
25 0.11 46.48 
35 0.17 47.52 
45 0.14 48.56 
55 0.10 49.59 
E. H2O2 15 1.51 9.29 5.67 12.57 0.98 
25 1.57 9.41 
35 1.61 9.54 
45 1.56 9.67 
55 1.67 9.79 
E. H2SO4 15 0.92 −30.05 − 29.71 1.20 0.87 
25 0.93 −30.05 
35 0.99 −30.05 
45 1.03 −30.05 
55 1.02 −30.05 
E. NaOH 15 2.40 −33.68 − 21.71 41.55 0.91 
25 2.39 −33.68 
35 2.37 −33.68 
45 2.46 −33.68 
55 2.38 −33.68 

According to Wan Ngah & Hanafiah (2008), when ΔG assumes negative values this is indicative of the spontaneous nature of the reaction, whereas positive values for ΔS indicate an increase in the disorder and randomness of the solid interface/solution during the sorting process.

Thus, due to the values assumed by ΔG, the adsorption of Cd2+ by E. in natura and E. H2O2 constitute non-spontaneous sorption processes, whereas E. H2SO4 and E. NaOH constitute spontaneous adsorptive processes.

The grape stem biosorbent, after simple chemical modification with H2O2, H2SO4 and NaOH, exhibit different values for pHPZC, possible functional groups such as carboxyl, amine, phenolic and other groups, modifications on adsorbents texture, but with similar thermal stabilities, and these results suggest that treatments carried out with H2O2, H2SO4 and NaOH caused certain modifications on the biomass grape stems.

The modified adsorbents of grape stem exhibited superior adsorption rates to its precursor biosorbent, especially E. H2O2, with 33% of increase in adsorptive capacity, and E. NaOH, with adsorption elevation by 66%.

The use of such solid residues (grape stems) as raw material for the production of modified adsorbents appears as an excellent alternative for the disposal of these residues, allowing added value to a currently discarded residue.

To Capes and CNPq for the funding of this research. To The School of Agriculture (ISA – Lisboa), for providing the required grape stem.

Abdolali
A.
,
Ngo
H. H.
,
Guo
W.
,
Zhou
J. L.
,
Zhang
J.
,
Liang
S.
,
Chang
S. W.
,
Nguyen
D. D.
&
Liu
Y.
2017
Application of a breakthrough biosorbent for removing heavy metals from synthetic and real wastewaters in a lab-scale continuous fixed-bed column
.
Bioresource Technology
229
,
78
87
.
Barbosa
L. C. D. A.
2007
Espectroscopia no infravermelho na caracterização de compostos orgânicos
.
UFV
,
Viçosa
,
189
.
Barquilha
C. E. R.
,
Cossich
E. S.
,
Tavares
C. R. G.
&
Silva
E. A.
2017
Biosorption of nickel(II) and copper(II) ions in batch and fixed-bed columns by free and immobilized marine algae Sargassum sp
.
Journal of Cleaner Production
150
,
58
64
.
Barrett
E. P.
,
Joyner
L. G.
&
Halenda
P. P.
1951
The determination of pore volume and area distributions in porous substances. I. Computation from nitrogen isotherms
.
Journal of the American Chemical Society
73
(
1
),
373
380
.
Brunauer
S.
,
Emmett
P. H.
&
Teller
E.
1938
Adsorption of gases in multimolecular layers
.
Journal of the American Chemical Society
60
(
2
),
309
319
.
de Mello
L. M. R.
&
da Silva
G. A.
2014
Availability and Characteristics of Waste From the Grape Processing Industry From Rio Grande do Sul
. .
Dong
H.
,
Zeng
Y.
,
Xie
Y.
,
He
Q.
,
Zhao
F.
,
Wang
Y.
&
Zeng
G.
2017
Single and combined removal of Cr(VI) and Cd(II) by nanoscale zero-valent iron in the absence and presence of EDDS
.
Water Science & Technology
76
(
5
),
1261
1271
.
Dos Santos
V. C. G.
,
Tarley
C. R. T.
,
Caetano
J.
&
Dragunski
D. C.
2010
Assessment of chemically modified sugarcane bagasse for lead adsorption from aqueous medium
.
Water Science and Technology
62
(
2
),
457
465
.
Dos Santos
V. C.
,
De Souza
J. V.
,
Tarley
C. R.
,
Caetano
J.
&
Dragunski
D. C.
2011
Copper ions adsorption from aqueous medium using the biosorbent sugarcane bagasse in natura and chemically modified
.
Water, Air, & Soil Pollution
216
(
1–4
),
351
359
.
Dubinin
M. M.
&
Radushkevich
L. V.
1947
The equation of the characteristic curve of the activated charcoal
.
Proceedings of the National Academy of Sciences
.
USSR Physical Chemistry Section
55
,
331
337
.
FAO
2016
Food and Agriculture Organization of the United Nations. Available: http://www.fao.org/faostat/en/#data/QC
.
Freundlich
H. M. F.
1906
Over the adsorption in solution
.
J Phys Chem
57
,
385
471
.
Gonçalves
A. C.
Jr
,
Coelho
G. F.
,
Schwantes
D.
,
Rech
A. L.
,
Campagnolo
M. A.
&
Miola
A.
Jr
2016
Biosorption of Cu(II) and Zn(II) with acai endocarp Euterpe oleracea M. in contaminated aqueous solution. Acta Scientiarum
.
Technology
38
(
3
),
361
371
.
Ho
Y. S.
&
McKay
G.
1999
Pseudo-second-order model for sorption process
.
Process Biochem
34
(5)
,
451
465
.
IUPAC
1972
Manual of Symbols and Terminology for Physicochemical Quantities and Units
.
Butterworths
,
London, UK
.
Kim
J.
,
Garcia-Esquinas
E.
,
Navas-Acien
A.
&
Choi
Y.-H.
2018
Blood and urine cadmium concentrations and walking speed in middle-aged and older U.S. adults
.
Environmental Pollution
232
,
97
104
.
Lagergren
S.
1898
Zur theorie der sogenannten adsorption geloster stoffe
.
Kungliga Svenska Vetenskapsakademiens Handlingar
24
(
4
),
1
39
.
Lo
S. F.
,
Wang
S. Y.
,
Tsai
M. J.
&
Lin
L. D.
2012
Adsorption capacity and removal efficiency of heavy metal ions by Moso and Ma bamboo activated carbons
.
Chemical Engineering Research and Design
90
(
9
),
1397
1406
.
Melzer
M.
,
Blin
J.
,
Bensakhria
A.
,
Valette
J.
&
Broust
F.
2013
Pyrolysis of extractive rich agroindustrial residues
.
Journal of Analytical and Applied Pyrolysis
104
,
448
460
.
Meneghel
A. P.
,
Gonçalves
A. C.
Jr
,
Strey
L.
,
Rubio
F.
,
Schwantes
D.
&
Casarin
J.
2013
Biosorption and removal of chromium from water by using moringa seed cake (Moringa oleifera Lam.)
.
Quim Nova
36
(
8
),
1104
1110
.
Mimura
A. M. S.
,
Vieira
T. D. A.
,
Martelli
P. B.
&
Gorgulho
H. D. F.
2010
Utilization of rice husk to remove Cu2+, Al3+, Ni2+ and Zn2+ from wastewater
.
Quim Nova
33
(
6
),
1279
1284
.
Nacke
H.
,
Gonçalves
A. C.
Jr
,
Campagnolo
M. A.
,
Coelho
G. F.
,
Schwantes
D.
,
Santos
M. G.
,
Briesch
D. L.
&
Zimmermann
J.
2016
Adsorption of Cu (II) and Zn (II) from water by Jatropha curcas L. as Biosorbent
.
Open Chemistry
14
(
1
),
103
117
.
Nacke
H.
,
Gonçalves
A. C.
,
Coelho
G. F.
,
Schwantes
D.
,
Campagnolo
M. A.
,
Leismann
E. A. V.
,
Conradi
E.
Jr
&
Miola
A. J.
2017
Removal of Cd(II) from water using the waste of jatropha fruit (Jatropha curcas L.)
.
Applied Water Science
7
(6)
,
3207
3222
.
Ngabura
M.
,
Hussain
S. A.
,
Ghani
W. A.
,
Jami
M. S.
&
Tan
Y. P.
2018
Utilization of renewable durian peels for biosorption of zinc from wastewater
.
Journal of Environmental Chemical Engineering
6
(
2
),
2528
2539
.
Rambo
M. K. D.
,
Rambo
M. C. D.
,
Almeida
K. J. C. R.
&
Alexandre
G. P.
2015
Study of thermo-gravimetric analysis of different lignocellulosic biomass using principal component analysis
.
Ciência E Natura
37
(
3
),
862
.
Roginski
S. Z.
1948
Adsorption and Catalysis on Inhomogeneous Surfaces
.
USSR Ac. of Sci. Publ. Moscow
,
Leningrad, Russia
.
Rubio
F.
,
Gonçalves
A. C.
Jr
,
Meneghel
A. P.
,
Tarley
C. R. T.
,
Schwantes
D.
&
Coelho
G. F.
2013
Removal of cadmium from water using by-product Crambe abyssinica Hochst seeds as biosorbent material
.
Water Science and Technology
68
(
1
),
227
233
.
Rubio
F.
,
Gonçalves
A. C.
Jr
,
Dragunski
D. C.
,
Tarley
C. R. T.
,
Meneghel
A. P.
&
Schwantes
D.
2015
A Crambe abyssinica seed by-product as biosorbent for lead (II) removal from water
.
Desalination and Water Treatment
53
(
1
),
139
148
.
Schwantes
D.
,
Gonçalves
A. C.
Jr
,
Coelho
G. F.
,
Campagnolo
M. A.
,
Santos
M. G.
,
Miola
A. J.
&
Leismann
E. A. V.
2015
Crambe pie modified for removal of cadmium, lead and chromium from aqueous solution
.
International Journal of Current Research
7
,
21658
21669
. .
Schwantes
D.
,
Gonçalves
A. C.
Jr
,
Coelho
G. F.
,
Campagnolo
M. A.
,
Dragunski
D. C.
,
Tarley
C. R. T.
,
Miola
A. J.
&
Leismann
E. A. V.
2016
Chemical modifications of cassava peel as adsorbent material for metals ions from wastewater
.
Journal of Chemistry
2016
,
1
16
.
Schwantes
D.
,
Gonçalves
A. C.
Jr
,
Campagnolo
M. A.
,
Tarley
C. R. T.
,
Dragunski
D. C.
,
De Varennes
A.
,
Silva
A. K. S.
&
Conradi
E.
2018
Chemical modifications on pinus bark for adsorption of toxic metals
.
Journal of Environmental Chemical Engineering
6
(
1
),
1271
1278
.
Silveira Neta
J. D. J. D.
,
Silva
C. J. D.
,
Moreira
G. C.
,
Reis
C.
&
Reis
E. L.
2012
Removal of the reactive blue 21 and direct red 80 dyes using seed residue of Mabea fistulifera Mart. as biosorbent
.
Revista Ambiente & Água
7
(
1
),
104
119
.
Sips
R.
1948
Combined form of Langmuir and Freundlich equations
.
The Journal of Chemical Physics
16
,
490
495
.
Tagliaferro
G. V.
,
Pereira
P. H. F.
,
Rodrigues
L. A.
&
Da Silva
M. L. C. P.
2011
Cadmium, lead and silver adsorption in hydrous niobium oxide prepared by homogeneous solution method
.
Quim Nova
34
(
1
),
101
105
.
Weber
W. J.
&
Morris
J. C.
1963
Kinetics of adsorption carbon from solutions
.
Journal Sanitary Engineering Division Proceedings. American Society of Civil Engineers
89
,
31
60
.
Welz
B.
&
Sperling
M.
1999
The Techniques of Atomic Absorption Spectrometry
, 3rd edn.
Atomic Absorption Spectrometry
,
Wiley-VCH, Weinheim, Germany
, pp.
335
475
.

Supplementary data