1 Introduction

The amount of sewage sludge is increasing due to population growth, urbanization, and industrial development, and more stringent standards for the discharge of wastewater to aquatic bodies. Sewage sludge is mainly composed of substances responsible for the toxic and pathogenic nature of non-treated wastewater and, therefore, its management is an issue of particular concern. Sewage sludge is a heterogeneous mixture of water, organic matter, microorganisms, and inorganic substances. The amount of the inorganic fraction is usually higher than 50 wt% [1]. Dry sewage sludge contains nitrogen and phosphorus in different forms. Sludge also contains high concentrations of inorganic salts including ions (CO32−, PO43−, SO42−, and NO3−), heavy metals (e.g., Zn, Pb, Ni, Cd, Cr, Cu, As, and Hg), and other elements (e.g., Si, Al, K, Na, Ca, and Mg) [2]. The composition of sewage sludge is highly variable, since it is affected by different parameters such as the specificity of the sludge source area, the processes carried out in the wastewater treatment plant, and the season [3].

Common routes of utilization and disposal, including land filling, application to farmland, and forestry and incineration, have serious limitations because of increasing costs and more stringent environmental standards to avoid secondary pollution [4]. Among the alternatives, the production of adsorbents is a promising way because, apart from being an eco-friendly alternative for the valorisation of sludge, it has the advantage of allowing its reuse in water treatment.

An increasing number of studies have been recently conducted to convert sewage sludge into adsorbents [5,6,7,8,9,10,11]. The activation procedure comprises several steps which initiate with drying of wet sewage sludge, with the aim of minimizing the requirement of energy for the subsequent thermal treatment [12]. The subsequent activation procedure can be either chemical or physical. In chemical activation, the precursor (either sewage sludge or pre-carbonized sludge) is mixed with a reagent, such as H3PO4 [13, 14], H2SO4 [15, 16], NaOH [17, 18], KOH [17, 19, 20], K2CO3 [14, 15], or ZnCl2 [21, 22] and then heated up to a high temperature under an inert atmosphere. Physical activation is usually carried out through a two-step process. The first step involves the pre-carbonization of the precursor at a moderate temperature (400–700 °C) to break down the cross-linkage between carbon atoms [23]. The second stage comprises the selective burn-off of atoms from the carbonaceous structure by using gasifying agents such as steam [24, 25], air or oxygen [26, 27], and CO2 [19, 28, 29] at high temperature. A post-activation washing of biochar with acids can generally remove the activation reagent residue and extraneous reaction products from the new free interstices formed, and reduce the inorganic content of the carbonaceous material [30]. In the current research, HF and HCl are the most used acids for sewage sludge–derived biochar [31]. The deashing of biochar by acid washing results in the opening of the pore channels; thus, the specific surface area and pore structure are greatly improved [32]. However, the removal of minerals also leads to a reduction of the biochar production yield of about 50% [33, 34].

An alternative to the physical or chemical activation is the use of combined multiple activation methods, an issue not sufficiently investigated in the production of sludge biochar [2]. In the preparation procedure used in this study, both chemical and physical activation processes are integrated in a single step. The preparation method takes advantage of the heating step of the chemical activation to introduce an oxidizing gas (CO2, in this case), instead of an inert atmosphere. Compared to the chemical activation, it has the potential to further improve the textural properties and the adsorption capacities of the biochars [35] without adding a new heating step.

The combined activation in one step has been reported to have a significant impact both on the porosity and on surface functional groups of materials prepared from carbonaceous precursors such as agricultural wastes [36, 37]. Nevertheless, it has been scarcely studied for sewage sludge. Wang et al. [38] studied the combined activation of sewage sludge with KOH and steam, and more recently, the activation of sewage sludge with HCl under CO2 atmosphere has been investigated [28, 33]. Nevertheless, to the best of the author’s knowledge, there are no reports concerning the combined activation of alkali-treated samples of sewage sludge under CO2 atmosphere.

The objective of the present paper is to investigate the preparation of porous biochars through a procedure that combines chemical activation with an alkali (NaOH or K2CO3) and physical activation with CO2 in a single step. The contribution of this work can be summarized as follows: (i) Utilization of a promising and sustainable combined activation method, not reported in the literature for the preparation of sludge-based biochar. The sustainability is based on the utilization of CO2 as oxidizing gas and on the absence of a pre-carbonization step, thus leading to an important reduction of the preparation cost, due to energy saving, higher yield of biochar, lower capital and operational costs, and less processing time [37]. (ii) Study of a wide activation temperature range (600–1000 °C), including higher temperatures than those commonly reported in the literature, to ensure an effective activation using CO2, based on its low reactivity, due to its large molecular size [37] and the endothermicity of its reaction with carbon [4]. (iii) Proposal of an activation mechanism that takes into account the effect of each activating agent. (iv) Investigation of the application of the biochars in the removal of aqueous pollutants using two model compounds, methylene blue and phenol.

2 Materials and methods

2.1 Raw material

Anaerobically digested and dewatered sewage sludge was collected from an urban wastewater plant. Raw sludge was dried at 105 °C for 48 h in a convection oven. According to the proximate analysis of the material, its high water content (73.3 wt%) is noteworthy. Dried sewage sludge has a similar percentage of ash and volatile matter (42.2 wt% and 49.2 wt%, respectively), and the content of fixed carbon is relatively low (8.6 wt%).

Table 1 shows the results of the elemental analysis and the concentration of heavy metals corresponding to the feedstock used in this study. As it generally occurs, carbon is the most abundant element. Carbon can be present in organic compounds (e.g., aliphatic and aromatic hydrocarbons) and inorganic compounds (such as carbonates) [39]. Hydrogen mainly includes aromatic hydrogen, fatty hydrogen, and hydrogen in functional groups [40]. Nitrogen is mainly present in proteins in biomass [41]. Oxygen can also take part of organic constituents (e.g., hydroxyl, carboxyl, and carbonyl) and inorganic constituents (e.g., bicarbonate, carbonate, and phosphate) [41].

Table 1 Results of elemental analysis and concentration of heavy metals of sewage sludge. Elemental analysis, wt%. Heavy metals, mg/kg
Full size table

2.2 Preparation of sludge carbon

Dried sewage sludge was ground with a mortar and sieved. Particles within the 0.5–1.0 mm size range were selected. The impregnation ratio was established at 60 mmol activation agent/gprecursor (dried sewage sludge), which is within the normal range used for the chemical activation of sewage sludge [14, 42, 43] and ensures sufficient interaction between the activation agent and the precursor. To carry out the impregnation, about 1 g of sewage sludge was added to 20 cm3 of a 3 M solution containing the activating agent (NaOH or K2CO3).

The solutions were introduced in 50 cm3 borosilicate amber glass vials and maintained under constant stirring at 150 rpm in a reciprocating shaker at room temperature (20 ± 2 °C) for 48 h, to ensure the access of the agent to the interior of the particles. Samples were then filtered, transferred to a convection oven, and dried at 80 °C for 24 h. A part of the precursor, used as reference, was not impregnated.

The thermal treatment (conventional slow pyrolysis) was conducted in a quartz tube furnace, under CO2 atmosphere (120 cm3/min of flowing gas, corresponding to 8 min of residence time). The impregnated sludge was heated from room temperature to 600–1000 °C, using a low heating rate of 15 °C/min. Samples were soaked at the final temperature for 30 min, and then cooled down in nitrogen atmosphere. The obtained biochars were washed with distilled water until a neutral pH was reached.

The samples of sludge carbon (SC) prepared by physical activation only were coded based on the temperature: SC-600, SC-700, SC-800, SC-900, and SC-1000. The samples subjected to a combined physical and chemical activation process were coded based on the activation reagent—NaOH (SCN) or K2CO3 (SCK)—and temperature. For instance, SCN-600 refers to sludge carbon impregnated with NaOH and activated in CO2 atmosphere at 600 °C.

2.3 Characterization

The precursor (dried sewage sludge) was subjected to elemental analysis using CHNS analyzer (Euro-Vector EA-3000). The total concentration of heavy metals (Cr, Ni, Cu, Zn, and Pd) in the precursor was determined by a high-performance inductively coupled plasma mass spectrometer (ICP-MS, 7500ce Agilent Technologies). Prior to heavy metal determination, samples were microwave digested (Speedwave Four, Berghof) using an acid mixture (HNO3:HF = 3:1).

The precursor and samples impregnated with NaOH and K2CO3 were subjected to thermogravimetric analysis (TG) in a thermobalance (T.A. Instruments SDT 2960), under inert (nitrogen) atmosphere. The alumina crucible of the thermobalance was loaded with 15–20 mg of sample, and subjected to heating from room temperature to 800 °C, using a heating rate of 10 °C/min. The textural properties of sludge biochar were determined by N2 adsorption/desorption at 77 K (ASAP 2010, Micromeritics). Prior to measurements, samples were outgassed under N2 flow at 200 °C for 15 h. Specific surface area was determined using the Brunauer–Emmett–Teller (BET) method. Surface area and pore volume in the mesopore and macropore range were obtained using the Barrett, Joyner, and Halenda (BJH) method, while values in the micropore range were calculated based on the t-plot method.

pH was determined following the method described by Tessmer et al. [44]. Ash content was measured by heating the samples under air atmosphere for 1 h at 815 °C (UNE 32004 standard). Fourier transform infrared (FTIR) measurements were carried out by means of a Thermo Nicolet 6700 equipment in the absorbance mode, using the KBr self-supported pellet technique. Spectra were collected in the 400–4000 cm−1 range with a resolution of 2 cm−1. X-ray powder diffraction patterns were collected using a Philips X’pert PRO automatic diffractometer operating at 40 kV and 40 mA, in theta-theta configuration, secondary monochromator with Cu-Kα radiation (λ = 1.5418 Å) and a PIXcel solid state detector (active length in 2θ 3.347°). Data were collected from 5 to 90° 2θ (step size = 0.026 and time per step = 1000 s) at RT. A fixed divergence and antiscattering slit giving a constant volume of sample illumination were used. X-ray fluorescence (XRF) analysis was conducted using a Pananalytical AXIOS X-ray fluorescence spectrometer. Borate fusion was used to prepare sample beads, at a dilution ratio of 20:1.

Raman analysis was carried out in a Renishaw InVia Raman spectrometer coupled to a Leica DMLM microscope, using a laser of 514 nm (ion-argon laser, Modu-Laser). The power density of the laser beam was reduced in order to avoid the photo-decomposition of the samples. In order to improve the signal-to-noise ratio, 40 s was used for each spectrum and 10 scans were accumulated at 10% of the maximum power of the 514 nm laser, in the 1000–2000 cm−1 spectral window.

The chemical composition and surface properties of the materials were analyzed by a scanning electron microscope (JEOL JSM-7000F) equipped with an energy-dispersive X-ray (EDX) detector. Measurements were taken at a live time of 120 s with a voltage of 20 kV.

2.4 Adsorption experiments

Methylene blue (MB) and phenol were used as target adsorbates because they possess different physicochemical properties and molecular size. MB has a basic nature, and can be used as an indicator of the adsorption capacity in the micro- and mesopore range. Phenol has an acidic nature, and, owing to its smaller size, can be used as an indicator of micropores.

Single solute adsorption isotherms were obtained using the conventional bottle-point technique. Sludge carbon (10–15 mg) was contacted in stoppered glass bottles with 10 cm3 of aqueous solutions with known concentration of MB or phenol. The flasks were shaken for 72 h by means of a rotary mixer placed in a thermostatic chamber at 20 °C ± 0.5 °C. The speed of the rotary mixer was set at 12 rpm. Preliminary tests revealed that a holding time of 72 h is enough to reach the equilibrium. After 72 h, samples were subjected to centrifugation, and the residual concentration of the solute in the supernatant was analyzed by UV/VIS spectrophotometry (Jasco V-630). The concentration of MB and phenol was obtained by collecting the UV absorbance at a wavelength of 662 and 270 nm, respectively. Blank runs (without sludge carbon) were carried out to take into account any effect of the experimental system.

The adsorption capacity of sludge carbon (qe, mg/g) was determined by mass balance:

$${\mathrm q}_{\mathrm e}=({\mathrm C}_0-{\mathrm C}_{\mathrm e})\;\mathrm V/\mathrm m$$
(1)

where C0 and Ce are the initial and equilibrium concentrations of the adsorbate (mg/L), respectively. V is the solution volume (L) and m is the adsorbent mass (g).

Randomly selected experiments were carried out in triplicate, and the mean values were reported.

The adsorption isotherms of both adsorbates were determined, by fitting the equilibrium data to the Redlich-Peterson equation (Eq. 2):

$${\text{q}}_{\text{e}}=\frac{{\text{K}}_{\text{R}}\cdot {\text{C}}_{\text{e}}}{{1}+\alpha \cdot {{\text{C}}_{\text{e}}}^{\beta }}$$
(2)

where KR is the Redlich-Peterson constant (L/g), α is a constant with units of (L/mg)β, and β is an exponent between 0 and 1.

The best fitting parameters were calculated by non-linear regression. The goodness of fit between the experimental and predicted values was determined using the average percentage error (APE):

$$\mathrm{APE}=\frac{\sum\limits_{\mathrm i=1}^{\mathrm n}\left|\left({\mathrm q}_{\mathrm e,\exp}-{\mathrm q}_{\mathrm e,\mathrm{pred}}\right)/{\mathrm q}_{\mathrm e.\exp}\right|}{\mathrm n}\cdot100$$
(3)

where qe,exp is the experimental uptake capacity, qe,pred is the calculated uptake capacity, and n is the number of experiments.

3 Results and discussion

3.1 Characterization of raw sludge and investigation of the activation mechanism

The nature of raw sludge was analyzed elsewhere [33]. The results reveal that the precursor is an anaerobically digested and well-stabilized sewage sludge, with a low degree of polymerization and aromatization.

The weight loss of raw sewage sludge during the pyrolysis under nitrogen atmosphere can be divided into three main sections (Fig. 1), in good agreement with the profiles reported in the literature [45]. Nevertheless, the TG and DTG curves of municipal sewage sludge may have differences in shape, due to its complex composition [46]. The first stage (below 175 °C) could be attributed to moisture loss. The most important weight loss takes place at intermediate temperatures, between 175 and 550 °C. The processes involved in this stage include the release of constitution water and the decomposition and volatilization of organic matter [47, 48]. The DTG profile (Fig. 1b), similar to those reported by others [49, 50], shows two overlapping peaks with maxima around 290 and 330 °C, and a shoulder close to 450 °C. According to Xiaohua and Jiancheng [46], the peaks near 300 °C are related to the decomposition of aliphatic compounds, whereas the shoulder is attributed to the decomposition of carbohydrate and proteins. The final mass loss, at temperatures higher than 550 °C, is commonly attributed to the decomposition of minerals like calcium carbonate [45].

Fig. 1
figure 1

Thermogravimetric results of raw sludge and sludge impregnated with NaOH and K2CO3 under inert (N2) atmosphere. a TG curves; b DTG curves

Full size image

The samples impregnated with NaOH and K2CO3 exhibit different TG and DTG profiles (Fig. 1). The first stage of mass loss (below 175 °C), attributed to the evaporation of adsorbed water, is more pronounced, thus evidencing the hygroscopic nature of both reagents. The weight loss above 600 °C is higher for the sample impregnated with K2CO3, as expected, due partly to the thermal decomposition of the incorporated carbonates.

The weight loss at the intermediate temperature range (175–550 °C) shows important differences between raw sludge and impregnated samples. The peaks appear more clearly defined in the case of the impregnated samples. The peak at around 290 °C is shifted to lower temperature, thus evidencing the interaction between the precursor and the reagent. Furthermore, the impregnated samples lead to a lower weight loss at intermediate temperatures, which could be explained by the high amount of salts incorporated during the impregnation. According to the results of XRF analysis (Table 2), SCK samples possess a remarkable amount of K2O (about 5 wt% vs. 0.3–0.8 wt% of the other samples), whereas SCN samples exhibit a remarkable Na2O amount (11–12 wt% vs. 0.1–0.5 wt% of the other samples).

Table 2 Results of XRF analysis for selected samples of sludge biochar. Main elements, expressed as weight percentage of the oxide. Other elements, expressed as weight percentage of the element
Full size table

In our previous works, a preliminary study on the physical activation mechanism of raw sewage sludge with CO2 was performed [28, 33]. Moreover, based on an extensive research concerning the chemical activation mechanism of a carbonaceous precursor using different reagents, including NaOH and K2CO3 [51], the following reactions and processes can be proposed for the activation with CO2 of non-impregnated sludge: (i) the decomposition/volatilization of organic matter; (ii) the decomposition/volatilization of inorganic constituents, such as calcium carbonate; (iii) the desorption of water; (iv) the gasification of carbon through two reactions: the reverse Boudouard reaction (Eq. (4) and the reaction of carbon with steam (Eq. (5); (v) the water–gas shift reaction (Eq. (6); and (vi) the reaction of carbon and OH–, to produce compounds such as cyanides, hydrocarbons, and carbonates, as discussed below. All these reactions are gasification reactions, except for Eq. (6), and, thus, are expected to generate porosity in the solid matrix.

$$\mathrm C\;(\mathrm s)+{\mathrm{CO}}_2\;(\mathrm g)\rightarrow2\;\mathrm{CO}\;(\mathrm g)$$
(4)
$$\mathrm C\;(\mathrm s)+{\mathrm H}_2\mathrm O\;(\mathrm g)\rightarrow{\mathrm H}_2\;(\mathrm g)+\mathrm{CO}\;(\mathrm g)$$
(5)
$$\mathrm{CO}\;(\mathrm g)+{\mathrm H}_2\mathrm O(\mathrm g)\leftrightarrow{\mathrm H}_2(\mathrm g)+{\mathrm{CO}}_2\;(\mathrm g)$$
(6)

Two pathways have been proposed for the reaction of carbon constituent with OH− ions. The first one results in the production of compounds such as cyanides and hydrocarbons. CN− ions were detected during the activation of different carbonaceous precursors, including sewage sludge [28, 33, 51, 52]. Robau-Sánchez et al. [52] proposed several mechanisms for the activation with KOH of Quercus agrifolia char that leads to the formation of cyanides, such as Eq. (7):

$$6\;\mathrm{KOH}\;(\mathrm l)+5.5\;\mathrm C\;(\mathrm s)+{\mathrm N}_2\;(\mathrm g)\rightarrow{\mathrm K}_2\;{(\mathrm{CN})}_2\;(\mathrm g)+2\;{\mathrm K}_2\;{\mathrm{CO}}_3\;(\mathrm s,\mathrm l)+1.5\;{\mathrm{CH}}_4\;(\mathrm g)$$
(7)

That mechanism that implies the reaction of carbon, nitrogen, and OH− ions was proposed in our previous papers for non-impregnated carbonaceous precursors—bone char [51] and sewage sludge [28, 33]—and it was concluded that structural nitrogen (constituent of the precursor) takes part in the activation process. Moreover, FTIR analyses reveal the presence of OH−, as discussed below.

The second reaction pathway between OH− ions and carbon was proposed by Lillo-Ródenas et al. [53, 54] during the chemical activation of anthracites with hydroxides:

$$6\;\mathrm{NaOH}\;(\mathrm s,\mathrm l)+\mathrm C\;(\mathrm s)\rightarrow2\;\mathrm{Na}\;(\mathrm s)+3\;{\mathrm H}_2\;(\mathrm g)+2\;{\mathrm{Na}}_2{\mathrm{CO}}_3\;(\mathrm s,\mathrm l)$$
(8)

Regarding the samples impregnated before the treatment with CO2, the occurrence of additional processes or the intensification of the aforementioned processes could be expected. In the activation mechanism proposed, the effect of the alkaline reagent can be classified as follows: (i) thermal decomposition of the reagent, (ii) interaction with the oxidizing gas CO2, (iii) interaction with constituents of the precursor, and (iv) catalyst of carbon gasification reactions.

With regard to the activation with NaOH, according to previous studies [55], the incorporated reagent can react with CO2, giving way to the production of Na2CO3 (Eq. (9), which is then decomposed at high temperature, along with the remaining NaOH (Eqs. (10–11)). These gasification reactions involve the generation of porosity. As mentioned before, remarkable amounts of Na were detected in SCN samples (Table 2).

$$2\;\mathrm{NaOH}\;(\mathrm s,\mathrm l)+{\mathrm{CO}}_2\;(\mathrm g)\rightarrow{\mathrm{Na}}_2{\mathrm{CO}}_3\;(\mathrm s,\mathrm l)+{\mathrm H}_2\mathrm O\;(\mathrm g)$$
(9)
$${\mathrm{Na}}_2{\mathrm{CO}}_3\;(\mathrm s,\mathrm l)\rightarrow{\mathrm{Na}}_2\mathrm O\;(\mathrm s)+{\mathrm{CO}}_2\;(\mathrm g)$$
(10)
$$2\;\mathrm{NaOH}\;(\mathrm s,\mathrm l)\rightarrow{\mathrm{Na}}_2\mathrm O\;(\mathrm s)+{\mathrm H}_2\mathrm O\;(\mathrm g)$$
(11)

Furthermore, according to the mechanism proposed by Zou et al. [55], NaOH and Na2O can react with SiO2 inorganic constituent of the precursor (Table 2), as shown in Eqs. (12)–(13). This reaction set does not imply gasification reactions and, consequently, does not contribute to the generation of porosity.

$$2\;\mathrm{NaOH}\;(\mathrm s)+{\mathrm{SiO}}_2\;(\mathrm s)\rightarrow{\mathrm{Na}}_2{\mathrm{SiO}}_3\;(\mathrm s)+{\mathrm H}_2\mathrm O\;(\mathrm l)$$
(12)
$${\mathrm{Na}}_2\mathrm O\;(\mathrm s)+{\mathrm{SiO}}_2\;(\mathrm s)\rightarrow{\mathrm{Na}}_2{\mathrm{SiO}}_3\;(\mathrm s)$$
(13)

The impregnation with K2CO3 would contribute to the development of porosity through a similar mechanism. Apart from the thermal decomposition of K2CO3 (Eq. (15), the transformation of carbonates into OH− ions in the presence of water (Eq. (14) has been reported [51, 56]. KOH would then be decomposed at high temperature (Eq. (16). The results of XRF (Table 2) evidence the presence of important amounts of K in SCK samples.

$${\mathrm K}_2{\mathrm{CO}}_3\;(\mathrm s,\mathrm l)+{\mathrm H}_2\mathrm O\;(\mathrm g)\rightarrow2\;\mathrm{KOH}\;(\mathrm s,\mathrm l)+{\mathrm{CO}}_2\;(\mathrm g)$$
(14)
$${\mathrm K}_2{\mathrm{CO}}_3\;(\mathrm s,\mathrm l)\rightarrow{\mathrm K}_2\mathrm O\;(\mathrm s)+{\mathrm{CO}}_2\;(\mathrm g)$$
(15)
$$2\;\mathrm{KOH}\;(\mathrm s,\mathrm l)\rightarrow{\mathrm K}_2\mathrm O\;(\mathrm s)+{\mathrm H}_2\mathrm O\;(\mathrm g)$$
(16)

Furthermore, the treatment with NaOH incorporates OH− ions, thus resulting in a greater extent of the reactions between carbon and OH− ions (Eqs. (7) and (8)). These reactions involve the gasification of sludge constituents and, in addition, would lead to the production of carbonates, which would then be decomposed at high temperature, thus enhancing the generation of pores. The formation of OH− ions through Eq. (14) upon the impregnation with K2CO3 would have the same effect (Fig. 2).

Fig. 2
figure 2

Pore size distribution of samples of sludge carbon. a Samples prepared by physical activation with CO2 only; b samples prepared by impregnation with K2CO3 followed by activation with CO2; c samples prepared by impregnation with NaOH followed by activation with CO2

Full size image

The impregnation would also enhance the gasification of carbon. Indeed, the catalytic role of K and Na metals is well known for both the reverse Boudouard reaction (Eq. (4), and the gasification of carbon with steam (Eq. (5), being K the most active [57, 58].

As discussed below (Sect. 3.3.4., Fig. 3), the alkali activation promotes the generation of magnetite (Fe3O4), absent in samples of biochar prepared by physical activation only. The high iron content of the prepared materials (Table 2) reflects the addition of iron-containing coagulants such as PFC (polyferric chloride) or PFS (polyferric sulfate), during the wastewater treatment [59].

Fig. 3
figure 3

XRD patterns of samples of sludge carbon. a Samples prepared by physical activation with CO2 only; b samples prepared by impregnation with K2CO3 followed by activation with CO2; c samples prepared by impregnation with NaOH followed by activation with CO2

Full size image

Two possible mechanisms are proposed for the generation of magnetite. The first one (Eq. (17) implies the direct formation of Fe3O4 from the reaction of Fe3+ with hydroxides. The second mechanism (Eqs. (18)–(19) involves the generation of Fe(OH)3 and its subsequent transformation into magnetite, following a scheme similar to that proposed by Tang et al. [59] for the formation of Fe2O3:

$$3\;\mathrm{Fe}^{3+}+8\;\mathrm{OH}^-+\mathrm e^-\rightarrow{\mathrm{Fe}}_3{\mathrm O}_4+4\;{\mathrm H}_2\mathrm O$$
(17)
$$\mathrm{Fe}^{3+}+3\;\mathrm{OH}^-\rightarrow\mathrm{Fe}{(\mathrm{OH})}_3$$
(18)
$$3\;\mathrm{Fe}{(\mathrm{OH})}_3+{\mathrm e}^-\rightarrow{\mathrm{Fe}}_3{\mathrm O}_4+4\;{\mathrm H}_2\mathrm O+\mathrm{OH}^-$$
(19)

The effect of the addition of the alkali reagent is clear in both reaction mechanisms. The treatment with NaOH directly incorporates OH− ions, whereas the impregnation with K2CO3 would have the same effect, as a result of the generation of OH− ions through Eq. (14).

3.2 Sludge carbon yield and ash content

The results of ash content and partial and overall yields are listed in Table 3. The impregnation yield (Yimpr) was calculated as the ratio between the weight of the sample after and before the impregnation with K2CO3 and NaOH. It is noteworthy that the impregnation with K2CO3 results in an increase in mass, due to the partial retention of the salt. On the contrary, the treatment with NaOH leads to a decrease in mass, suggesting an interaction between the reagent and the structure.

Table 3 Overall carbon yield and partial yields of different preparation steps (impregnation, activation with CO2), ash content, and pH
Full size table

The yield of the activation step with CO2 (YCO2, on a dry basis) was obtained from the mass of the sample prior and after treatment with CO2. The overall yield of sludge carbon (YSC) was calculated as the product of the individual yields. As expected, YCO2 decreases with temperature, whereas the ash content is increased. These results could be explained by the decomposition/volatilization or gasification of sludge constituents and/or incorporated species at high temperature, thus leading to the concentration of the mineral constituents. It is noteworthy that YCO2 is higher for the samples impregnated with K2CO3 and NaOH, compared to non-impregnated samples. Moreover, the highest overall yield corresponds to the samples activated with K2CO3. As commented before, this result could be attributed to the high amount of salts incorporated during the impregnation (Table 2).

3.3 Characterization of sludge biochars

3.3.1 Textural properties

Table 4 summarizes the textural properties of the materials, and Fig. 2 shows the results of pore size distribution (PSD) of sludge biochar, measured by the BJH method. The materials possess a hierarchical structure, with a multimodal pore-size distribution (peaks at around 20, 80–100, and 300–500 Ǻ).

Table 4 Textural properties of the prepared samples of sludge carbon. S m2/g; V cm3/g, Dp, Å
Full size table

Concerning the samples prepared by activation with CO2 only (SC), the temperature that provides the highest value of specific surface area (SBET of 113 m2/g) and pore volume (0.156 cm3/g) is 800 °C. The maximum value of micro- and mesoporosity (40.7 and 44.0 m2/g, respectively) is also obtained at this temperature. For SC-800, the main contribution to specific surface area comes from pores up to 80 Å, for which the effect of temperature is especially strong. Lower temperatures are not effective to develop porosity, in good agreement with other authors [23, 37], who state that high temperatures are required for the activation with CO2 to be effective. Temperatures higher than 900 °C have a detrimental effect. The destructive effect of the activation temperature of 1000 °C is noteworthy in the whole pore range (Fig. 2), due to pore deformation, cracking, or blockage [60].

Regarding the impregnation with K2CO3, lower temperatures are required to achieve suitable textural properties. The highest value of SBET (154 m2/g) is obtained at 700 °C. This value is slightly higher than that obtained at 600 °C (129 m2/g). SCK-700 exhibits the highest mesoporosity (51.3 m2/g and 0.1103 cm3/g) of the samples activated with K2CO3, which is mainly due to pores up to 100 Å (Fig. 2) whereas SCK-600 possesses the highest microporosity (74.4 m2/g and 0.0325 cm3/g), which represents around fivefold increase as compared to non-chemically treated sample (SC-600). An increasing temperature has the effect of decreasing microporosity, the effect being noteworthy above 800 °C. The highest total pore volume corresponds to SCK-700 and SCK-800 (0.286 and 0.291 cm3/g, respectively). In the case of the sample activated at 800 °C, this high value of porosity is mainly due to its high degree of macroporosity. The different pore size distribution of SCK-800 and SCK-900 should be highlighted. The sample activated at 800 °C exhibits a dramatic decrease in mesopores up to 100 Å (compared to samples activated at lower temperature), and an important increase in pores higher than 500 Å (large mesopores and macropores), likely due to pore cracking. Contrary to expected, further increase in temperature (from 800 to 900 °C) gives way to a rise in mesoporosity especially in the 40–100 Å range. This phenomenon should be attributed to the occurrence of reactions or processes at high temperature involving the activating agent (such as the aforementioned thermal decomposition of K2CO3) that results in the development of porosity.

Compared to samples subjected to physical activation only, at temperatures up to 700 °C, the use of K2CO3 leads to an outstanding increase in SBET (129 vs. 58.4 m2/g at 600 °C; 154 vs. 50.7 m2/g at 700 °C) and microporosity (74.4 vs. 15.0 m2/g, and 62.1 vs. 11.5 m2/g, respectively). On the contrary, at higher temperatures the treatment with K2CO3 has a detrimental effect on the development of porosity, the highest difference being obtained at 800 °C (SBET = 41.2 and 113 m2/g and Smicro = 13.9 and 40.7 m2/g, for SCK-800 and SC-800, respectively).

The impregnation with NaOH prior to the activation with CO2 results also in a trimodal PSD, though with a higher development of pores larger than 100 Å, compared to SC samples. Among the temperatures studied, the most suitable is 600 °C, since it provides the best textural properties: SBET (93.3 m2/g), microporosity (35.8 m2/g), mesoporosity (41.0 m2/g), and Vtotal (0.196 cm3/g). These values are better than those obtained at 600 °C using physical activation only (59.8% of increase in SBET and 139% in Smicro), but do not improve those obtained using K2CO3. Note that similar textural properties are attained without impregnation by a more severe thermal treatment, at around 800 °C. It is noteworthy that the same phenomenon observed for SCK samples takes place, evidenced by the different PSD of SCN-800 and SCN-900. Whereas the activation at 800 °C (SCN-800) causes a drastic decrease in mesopores up to 100 Å, with most of its specific area in large mesopores and macropores (possibly due to pore cracking, the average pore diameter being 192 Å);, further increase in temperature (900 °C) better preserves mesoporosity, up to 300 Å. This phenomenon could be related to the effect of the activating agent, noteworthy at high temperatures. As proposed for SCK samples, the reactions which generate carbonates, along with the subsequent thermal decomposition, could explain the generation of new porosity.

Excessive activation temperatures have an important detrimental effect on porosity (micro- and mesoporosity, mainly up to 80 Å), and the optimum value depends on the chemical treatment. For instance, the optimum values of specific surface area and pore volume are obtained at 800 °C for SC, 700 °C for SCK, and 600 °C for SCN samples.

3.3.2 SEM

SEM analysis (Figure S1 given in Online Resource 1) reveals that a temperature of 800 °C causes the most prominent modification of the surface morphology of non-impregnated samples of sludge biochar (SC), in line with the abovementioned trend in BET and pore volume data. Moreover, as will be discussed in Sect. 3.4, these features definitely enable an easy access for methylene blue and phenol molecules and also contribute to the adsorption on surface. Figure S2 (Online Resource 1) shows the SEM images of selected samples of biochars prepared through chemical impregnation prior to the activation with CO2 at 800 °C. The treatment with NaOH smooths the surface and causes a significant loss of BET area and pore volume (mainly micropore volume). The formation of tunnels and surface carvings upon the treatment with K2CO3 can be clearly observed. Such a significant alteration of the macropore structure is translated into an increased Smacro and Vmacro for sample SCK-800 (Table 4), whereas specific surface area and total pore volume decrease.

3.3.3 EDX and XRF

Local EDX analyses (Table 5) reveal the high degree of heterogeneity of both the precursor (dried sewage sludge) and the prepared samples of biochar. Consequently, apart from the matrix, the composition of individual particles visible in surface has been determined.

Table 5 Results of EDX analysis for selected samples of sludge biochar. Chemical composition of the prepared materials (wt%)
Full size table

The results show that in most cases the matrix is rich in oxygen, calcium, iron, silicon, phosphorus, and aluminum, in good agreement with the results of XRF (Table 2). As an exception, the sample activated at the highest temperature (SC-900) possesses a matrix rich in O, Fe, and Cr, and the other elements appear in hydroxyapatite-like individual particles.

The high amount of sulfur detected in many samples is noteworthy, either in the matrix or in individual particles, such as sulfates of Fe–Ca. These data reveal that S appears preferentially in the surface of biochars, given the low amount of sulfur detected by XRF (Table 2). Contrarily, the amount of Na measured is very low (with the exception of SC-900, in which Na was detected in the aforementioned individual particles), even in the sample impregnated with NaOH (SCN-800). The high amount of Na incorporated, detected by means of XRF (Table 2), evidences that this element is concentrated in the bulk of the material. Regarding the sample impregnated with K2CO3, important amounts of K are detected both in the matrix and in individual particles.

Apart from the aforementioned sulfates, other individual particles detected in biochars were identified as oxides or carbonates of Ca, phosphates of Fe–Ca, aluminosilicate-like aggregates, iron oxides, hydroxyapatite-like aggregates, and quartz (SiO2).

3.3.4 XRD

XRD analyses were performed to investigate the crystalline structure. Strong and sharp reflection XRD peaks (Fig. 3) characteristic of SiO2 are observed for all non-impregnated biochars. Also, calcite is observed for non-impregnated sludge carbons activated up to 800 °C (SC-600 to SC-800), whereas at the highest temperature (900 °C) it is absent. Moreover, at temperatures above 800 °C, sharp XRD peaks pertaining to more complex crystalline structures such as apatite (Ca5(PO4)3(OH)) (PDF 9–432), CaMgSi2O6 (PDF 75–1092) and whitlockite (Ca18,19(Mg1.17Fe0.83H1.62(PO4)14)) (PDF 70–1786) emerge as a result of the burning off of the volatile fraction.

In the case of the impregnated samples, the crystalline structure is significantly modified and new crystalline phases (i.e., magnetite, apatite) appear in sludge biochar even at the lowest temperature (600 °C). It is interesting to note that all the sludge carbons prepared by combined physical and chemical activation show magnetic properties, whereas the non-chemically treated samples do not contain magnetic particles (i.e., Fe3O4). It is interesting to note that the activation protocol used in this study is effective to generate biochar with magnetic properties through the transformation of the Fe introduced in the water treatment process into magnetite, without adding an iron source. The presence of magnetite is an advantage for the reuse of sewage sludge–based biochars in wastewater treatment, since it favors the removal of anionic surfactants [61] and heavy metals such as lead, copper, zinc, and manganese [62].

Also, in the impregnated samples, contrary to the non-impregnated SC samples, silicates are decomposed at high temperatures (i.e., no characteristic peaks of SiO2 are detected in samples SCN-800, SCN-900, and SCK-900) and new and more complex crystalline phases are formed (Fig. 3b, c). The treatment with K2CO3 results in the formation of KNO3⋅0.5H2O (PDF 35–927) or KAlSiO4 (PDF 33–988), as both phases have characteristic peaks in that range. Upon treatment with NaOH, new diffraction peaks emerge which can be ascribed to burkeite (Na6CO3(SO4)2) and hanksite (Na22K(CO3)2(SO4)9Cl) phases. The appearance of new peaks upon the impregnation with either K2CO3 or NaOH supports the occurrence of reactions between the incorporated Na(II) and K(II) with sludge constituents.

3.3.5 Raman

Raman spectroscopy was performed to gain insight into the carbon structure. Figure 4 shows the Raman spectra of the precursor (raw sewage sludge) and selected samples of biochar. Regarding the precursor, a prominent photoluminescence background is observed, due to the incoherent vibrations of the high content of hydrogen [63]. The photoluminescence background is reduced when samples are subjected to pyrolysis, reflecting the reduction of the hydrogen content with an increasing degree of carbonization.

Fig. 4
figure 4

Raman spectra of selected samples of sludge biochar

Full size image

Except for SCN-800, every sample exhibits two peaks near 1360 and 1590 cm−1, attributed to D and G bands, respectively. Generally, for highly ordered carbonaceous materials such as graphite or graphene, the D band indicates the presence of amorphous or disordered graphite, whereas the G band corresponds to graphitic crystallites [64]. Nevertheless, for highly disordered carbonaceous materials such as sewage sludge biochar, Li et al. [65] proposed that the D band should be more attributed to the breathing mode of sp2-bonded carbon atoms in hexagonal aromatic rings, whereas the G band should be ascribed to bond stretching of all pairs of sp2-bonded carbon atoms in both chains and rings. The D band is related to the structure disorder in the case of graphite, but for amorphous carbon, the development of the D band reveals structure ordering [66].

The ratio between the D and G bands can be used to determine the effect of the activation method on the carbonaceous structure of biochars. The deconvolution was necessary because of the large overlap between the D and G bands. Table S1 (Electronic Supplementary Material) provides the ID/IG and AD/AG ratios (corresponding to the intensity and area of the peaks, respectively). Both parameters follow the same trend. It is observed that the activation temperature has no visible effect on carbon structure, since the biochars prepared at 600–900 °C possess a similar ID/IG (0.91–0.94), slightly lower than that of the precursor (0.99). The chemical treatment, on the contrary, leads to an increase in the ID/IG ratio. This increase is moderate for K2CO3 (1.07 vs. 0.94), whereas for the sample activated with NaOH, only the D band is observed. Therefore, the chemical treatment results in the development of hexagonal aromatic rings and in the increase in the structural ordering of the carbon fraction.

3.3.6 FTIR

Figure 5 shows the FTIR spectra of all samples of biochar. Regarding the samples prepared by physical activation with CO2 (Fig. 5a), the small peaks detected in the 600–750 cm−1 range, which do not show a clear trend with temperature, are attributed to the vibration of complex components of sludge carbon [55]. All materials exhibit a similar and predominant peak close to 1060 cm−1 (peak I), related to the silicon content. This band is indicative of structures such as Si–O–Si, Si–O–X (X = Al, Fe, Ca, Mg, and Na) and Si–O–C [25, 55]. The small peak observed at 870 cm−1 could also be attributed to Si–O–Si structure. The peak with its maximum near 1400 cm−1 (peak II) is difficult to assign. It could be ascribed to either organic sulfates, CaCO3, or long-chain aliphatic structures (indicative of C–H bending band) [18]. This peak undergoes a dramatic decrease with temperature. Whereas a prominent and clearly defined peak is observed for SC-600, for SC-900 it almost disappears. This behavior can be ascribed to the decomposition of organic compounds and CaCO3 at high temperature, in good agreement with the XRD results discussed above. The samples prepared between 600 and 800 °C also show a shoulder near 1600 cm−1 (peak III), which could be associated to carbonyl groups (C = O) [25, 67]. The intensity of this peak also decreases with temperature and above 900 °C it is almost completely absent. The broad band with its maximum close to 3400 cm−1 (peak IV) should be related to –OH and –NH surface functional groups [18, 19, 68], and could also be partially attributed to adsorbed water. This peak is strong for SC-600, and undergoes a drastic decrease above 700 °C.

Fig. 5
figure 5

FTIR spectra of samples of sludge carbon. a Samples prepared by physical activation with CO2 only; b samples prepared by impregnation with K2CO3 followed by activation with CO2; c samples prepared by impregnation with NaOH followed by activation with CO2

Full size image

Figure 5 b displays the FTIR spectra of the samples impregnated with K2CO3 followed by the activation with CO2. In these spectra, the silicon content of the samples is still evident (band at 1060 cm−1, peak I, which becomes broader with temperature). Regarding the peaks in the 1400–1600 cm−1 range, the treatment with K2CO3 results in important changes. The peak at 1400 cm−1 (peak II), attributed to calcite and long-chain aliphatic structures in SC samples, appears diminished for temperatures up to 800 °C, compared to SC samples, and a decreasing trend is observed in line with the characteristic XRD reflection peaks of calcite in the SCK series. Taken into account that the impregnation step would result in an increase in the amount of carbonates, this result could be attributed to the enhancement of reactions that lead to the removal of aliphatic structures. For temperatures of 900–1000 °C, on the contrary, the size of this band remains almost constant, and two overlapping peaks (peaks II and V) can be distinguished (at lower activation temperature, a shoulder can be distinguished), suggesting the presence of new functionalities. Several authors [68,69,70] attributed a band near 1450 cm−1 to the presence of the C = C bond of aromatic rings polarized by oxygen atoms bound near one of the C atoms, suggesting the presence of basic oxygen-containing functionalities, such as chromene structures, diketones of quinone groups, and pyrone-like groups. The peak near 1490 cm−1 (peak V) might also signal the presence of nitrogen-containing groups, as reported in sludge-derived carbons prepared at high temperature (above 800 °C) [24, 71]. This phenomenon, that is, the occurrence of reactions and/or processes at high temperature level for samples impregnated with K2CO3, is in good agreement with the results of textural properties. As mentioned before, when the activation temperature is increased from 800 to 900 °C, there is an important increase in mesoporosity.

Regarding the band at 1600 cm−1 (peak III), related to C = O functionalities, it still appears as a shoulder. This peak shows a decreasing trend with temperature, as occurred for SC samples, but still can be distinguished at high temperatures (above 900 °C). The impregnation with K2CO3 has also the effect of increasing –OH and –NH functional groups (broad peak with its maximum close to 3400 cm−1, peak IV). This peak is maximum for the lowest temperature studied (600 °C), as occurred for SC samples. However, the decrease with temperature is less pronounced for SCK samples. In fact, SCK-900 and SCK-1000 still exhibit an appreciable peak, higher than that of SCK-700 and SCK-800, thus suggesting the occurrence of mechanisms that involve the generation of those functionalities, such as Eq. (14).

The FTIR spectra of the samples prepared by impregnation with NaOH followed by activation with CO2 are shown in Fig. 5c. As occurred with SC and SCK samples, the high intensity of peak I is related to the silicon content of the materials. Note that strong XRD diffraction peaks in SCN-800 and SCN-900, ascribed to Na2O·Al2O3·SiO2, can be observed. The most visible effect of the impregnation with NaOH is in the 1300–1500 cm−1 range. As occurred with SCK samples at high temperatures, two overlapping bands can be distinguished (peaks II and V), suggesting the occurrence of reactions that involve the generation of oxygen- and nitrogen-containing functionalities. These peaks are more pronounced than those obtained after the impregnation with K2CO3. Furthermore, they do not exhibit important variations with temperature. The peak of C = O (near 1600 cm−1) appears as a shoulder (peak III), as in SC and SCK samples. This peak exhibits its maximum at the lowest activation temperature studied (600 °C), and is lower than that of SC-600 sample. The amount of C = O groups decreases with temperature. Concerning the effect of the impregnation with NaOH on the amount of –OH and –NH surface functional groups (broad peak close to 3400 cm−1, peak IV), it is qualitatively similar but less pronounced than that of the impregnation with K2CO3. That is, the peak is maximum for the activation temperature of 600 °C, and the decrease with temperature is lower than for SC samples, the peak still being appreciable at 900 and 1000 °C.

3.3.7 pH

Table 3 shows the pH values of the prepared materials. The samples prepared by physical activation with CO2 are slightly basic in nature, in good agreement with the results reported in the literature [4], with values of pH ranging from 7.4 to 7.7. The impregnation with either K2CO3 or NaOH results in an increase of the pH, with values in the 9.4–10.4 range for SCK samples, and 9.9–10.2 for SCN samples. In all cases, the pH increases with the activation temperature, which can be attributed to the decomposition and/or desorption of acidic functional groups (C = O and –OH and –NH) [27, 72], as confirmed by FTIR analyses (Fig. 5). Moreover, the increase in the basic character of sludge biochar as a result of the impregnation with K2CO3 and NaOH is in line with the generation of basic oxygen-containing functionalities observed in the FTIR spectra.

3.4 Adsorption of methylene blue and phenol

The adsorption isotherms of methylene blue and phenol at 20 °C for the prepared materials are shown in Figures S3 and S4 (given in Online Resource 1).

Table 6 summarizes the adsorption capacity of each material, as well as the parameters of best fit for the Redlich-Peterson model.

Table 6 Best fit Redlich-Peterson isotherm parameters for MB and phenol adsorption and experimental uptake of both adsorbates
Full size table

Among the samples prepared by physical activation only, the highest adsorption capacity of both MB and phenol (30.2 and 13.5 mg/g, respectively) corresponds to SC-800. This sample has the highest values of SBET, Vmicro, Vmeso, and Vtotal. Regarding the materials prepared by impregnation with K2CO3 followed by activation with CO2, SCK-700 exhibits the highest removal ability of MB (56.1 mg/g) and phenol (25.3 mg/g). These values represent an increase of 87%, compared to the best value of SC samples, and in the case of MB, it is the highest value of all samples. The high adsorption capacity of SCK-700 could be explained by its suitable textural properties. This sample has the highest values of SBET and Vmeso and 98% of the highest Vmicro, along with a suitable surface chemistry—second highest band of carbonyl groups among SCK samples, which has been reported to favor the adsorption of MB [33]. Concerning the uptake of MB, the good value obtained with SCK-600 (52.8 mg/g) is noteworthy, only slightly lower than the maximum value. The latter has the advantage that a lower activation temperature is required. The good adsorption ability of SCK-600 could be explained by its good textural properties (SBET and Vmeso are lower than those of SCK-700, and Vmicro is maximum) and suitable surface chemistry (the amount of C = O functionalities is the highest of SCK series). Activation temperatures above 700 °C have an important detrimental effect on the porosity, and thus, moderate temperatures are advisable for the activation with K2CO3.

Regarding the removal of phenol, the small difference between the adsorption capacity of both SCK-600 and SCK-800 (17.5 vs. 21.2 mg/g) is remarkable, given the great differences in the textural properties of both samples (SBET = 129 m2/g and Smicro = 74.4 m2/g for SCK-600, and SBET = 41.2 m2/g and Smicro = 13.9 m2/g for SCK-800). These results evidence that other factors, apart from textural properties, are implied in the adsorption process of phenol. It has been reported that phenol could be adsorbed not only by physisorption, but also by surface polymerization [73], favored by the presence of metals on the surface of the material. The much higher amount of ash of SCK-800, compared to SCK-600 (82.1 vs. 64.7 wt%), could then favor its phenol removal ability.

Among the materials prepared by impregnation with NaOH prior to activation with CO2, the sample activated at 600 °C shows the highest MB and phenol removal ability (35.0 and 34.0 mg/g, respectively), the uptake capacity of phenol of SCN-600 being the highest of all the materials. This sample has the highest value of SBET, Vmicro, Vmeso, Vtotal, and C = O functional groups among SCN samples. Compared to only physically activated samples, the maximum removal ability is increased by 15.9% (MB) and 152% (phenol), with the advantage that a mild activation is required (600 °C for SCN samples vs. 800 °C for SC).

Regarding the adsorption of MB of SCN samples, temperatures higher than 600 °C result in a decrease, which could be attributed to the decrease in SBET, Vmeso, and Vtotal (Table 4) and carbonyl groups (Fig. 5). Concerning the adsorption of phenol, the good uptake capacity of all SCN materials is noteworthy, even for the materials prepared at high temperature. This result could be attributed to the aforementioned generation of new functionalities containing oxygen and/or nitrogen as a consequence of the chemical activation with NaOH (Fig. 5c, peak V), in good agreement with the results reported in the literature [74] for the adsorption of phenolic compounds. The size of the peak is similar for all activation temperatures and thus, it would result in a good adsorption capacity of phenol within the whole temperature range studied. The aforementioned hypothesis that phenol could also be adsorbed by surface polymerization, favored by the presence of metals on the surface of the material (their amount increasing with temperature, Table 3), also contributes to explain the high uptake capacity of SCN samples activated at high temperatures. The removal ability of phenol decreases with temperature, which is in good agreement with the decreasing trend with temperature of the microporosity as shown in Table 4.

4 Conclusions

The proposed activation mechanism of non-impregnated sludge includes (i) the decomposition/volatilization of organic and inorganic constituents; (ii) the desorption of water; (iii) the gasification of carbon; and (iv) the reaction of carbon and OH− to produce compounds such as cyanides, hydrocarbons, and carbonates. The impregnation with K2CO3 or NaOH would lead to (i) the thermal decomposition of carbonates and hydroxides, directly incorporated or formed from the reagent; (ii) the intensification of the set of reactions between carbon and OH−, due to the incorporated or generated OH− ions; (iii) the enhancement of carbon gasification, owing to the catalytic role of K and Na; and (iv) the generation of magnetite (Fe3O4) from the reaction of Fe3+ with hydroxides.

From the standpoint of textural properties, the impregnation results in a decrease in the optimum activation temperature: 800 °C for untreated (SC) samples, and 600–700 °C for the samples impregnated with K2CO3 (SCK) and NaOH (SCN). The highest value of SBET of all samples corresponds to SCK-700 (increase of 36% compared to SC-800), whereas SCK-600 possesses a slightly lower value, along with the highest microporosity (83% of increase vs. SC-800). The impregnation increases the amount of surface –OH and –NH functional groups, also detected in non-impregnated biochars, the effect being more pronounced for K2CO3. It also results in the generation of new functionalities (in the whole temperature range for NaOH and above 800 °C for K2CO3), attributed to oxygen and/or nitrogen containing groups.

SCK-700 has the highest MB removal ability, whereas SCK-600 exhibits an only slightly lower adsorption capacity. These samples combine suitable textural properties and surface chemistry (high amount of carbonyl groups). SCN-600 possesses the highest phenol adsorption capacity. The optimum uptake values represent an increase of 87% and 152% for MB and phenol, respectively, compared to the highest value of non-impregnated samples (SC-800), with the advantage of the lower temperature required. The good phenol uptake capacity of all SCN samples could be attributed to (i) the generation of the new functionalities containing nitrogen and/or oxygen and (ii) the adsorption of phenol by surface polymerization.

The chemical treatment develops magnetic properties in biochar, an advantage for its reuse in wastewater treatment, since it favors the removal of anionic surfactants and heavy metals such as lead, copper, zinc, and manganese.