Introduction

Water is a prerequisite for all forms of life and most industrial activities. Thus, the need for clean water is increasing due to the increase of world population, changes in the global water cycle caused by the climate change, and the deterioration of water quality in several areas. One source contamination for water resources are heavy metals and persistent organic pollutants such as dyes, detergents, or petroleum products, to name just a few examples (Gautam et al. 2014; Raza et al. 2019; Quesada et al. 2019; Zamora-Ledezma et al. 2021; Ortúzar et al. 2022; Luukkonen et al. 2022).

Synthetic dyes are utilized extensively in different sectors such as the paint, printing, textiles, leather, plastics, cosmetics, food colouring, and pharmaceutical industries (Rezaei and Salem 2016; Pang et al. 2017). Currently, over one hundred thousand commercial dyes are available with a combined production of approximately 700,000 tonnes in each year (Kaya-Özkiper et al. 2022a). Dyes can include acidic, basic, azo, and diazo moieties or can be based on metal complexes (Hunger 2007; Chavan 2011). Approximately 10–15% of the dyes are discharged into water without any treatment, which has adverse impacts on the aquatic environments and, subsequently, on the health of human populations (Chan et al. 2015; Hussain et al. 2020). Dye concentrations even lower than 1 mg/L can be visually observable and unhealthy for human consumption (Moradihamedani 2022). Many of the commercial dyes are not biodegradable, and thus remain highly stable in the environment (Joseph et al. 2019). Moreover, some dyes (e.g., Disperse Blue 373, Disperse Orange 37, and Disperse Violet 93) can be carcinogenic or mutagenic (Alves de Lima et al. 2007; Subramanian et al. 2022). The presence of dyes in water interrupts the penetration of light (Pandey et al. 2007) and also causes increased biochemical and chemical oxygen demand (Yao et al. 2009).

Methylene blue (MB), a basic cationic dye, is one of the most widely consumed dyes in several industries (Parakala et al. 2019; Koyuncu and Kul 2020; Khan et al. 2022). MB is also used in microbiological staining and as pharmaceutical compound where the safe dose is less than 2 mg/kg of body weight (Ginimuge and Jyothi 2010; Wainwright et al. 1997). Chronic exposures to high MB concentrations (> 2 mg/kg) can cause low blood pressure, gas exchange deterioration in the lungs, abnormal heartbeat, coronary vasoconstriction, increased pulmonary vascular pressure, or pulmonary vascular resistance (Ginimuge and Jyothi 2010). The concentration of MB found in typical wastewaters generated by the textile industry is between 10 and 200 mg/L (Kadam et al. 2011). Water contamination by MB has been recognized as one reason for the shortage of clean water in the affected areas (Walker et al. 2019; Oladoye et al. 2022). MB produces a highly coloured solution even at a very low concentration in an aquatic environment (Khan et al. 2022). Because of the high molar absorption coefficient value of MB [~ 8.4 × 104 L/(mol × cm)], it can decrease the transmittance of sunlight through the water and, consequently, disturb the photosynthesis of aquatic organisms (Kosswattaarachchi and Cook 2018; Khan et al. 2022). MB-contaminated water, depending on the concentration and exposure duration, can have acute effects on human health, for example causing vomiting, diarrhoea, nausea, or difficulties in breathing (Gupta and Suhas 2009; Banerjee et al. 2015).

Various techniques including membrane separation, chemical or electrochemical oxidation, (electro-)coagulation and flocculation, adsorption, photodegradation, biological treatment, or combinations of these have been investigated to remove MB from aquatic environment (Hou et al. 2018; Cheng et al. 2020; Ahmed et al. 2022). Adsorption technology is considered a promising method to remove MB because of its simplicity in use, low operational costs, and potential high efficiency (Kuang et al. 2020; Eldeeb et al. 2022; Ye et al. 2023). However, further research is required to find more environmentally friendly, low cost, and efficient adsorbents.

Alkali-activated materials (AAMs), and geopolymers (a subclass of AAMs), are mainly utilized and studied for the construction material applications at the moment, but they have also several high-value uses such as materials for water treatment (Luukkonen et al. 2022). Alkali activation is a process in which aluminosilicate precursors (e.g., calcined clays, metallurgical slags, or fly ashes) and an alkali activator solution (e.g., alkali metal hydroxide, silicate, or carbonate) react at low temperature (< 100 °C) to form aluminosilicate products. Blast furnace slag (BFS), a by-product from iron manufacturing, is one of the typical precursors used for the preparation of high-calcium-content AAMs (Aziz et al. 2020). The availability of BFS is expected to decrease in the near future as the steel industry transitions into the reduction of iron ore by H2 instead of coke and using electric arc furnaces to further process the formed sponge iron (Patisson and Mirgaux 2020). Nevertheless, BFS can be replaced by other high-Ca content aluminosilicate precursors in the future. Therefore, BFS-based AAMs can be considered as useful model systems of those containing calcium/sodium–aluminium–silicate–hydrate gels (or C/N–A–S–H in the cement chemist notation). BFS-based alkali-activated materials have drawn attention for their environmental applications, for example as adsorbents in wastewater treatment (Luukkonen et al. 2016; Runtti et al. 2016; Medina et al. 2020; Sundhararasu et al. 2021; Humberto Tommasini Vieira Ramos et al. 2022; Abdelbasir and Khalek 2022). Gheibi et al. (2022) and Eftekhari et al. (2021) have earlier reported excellent adsorption performance of aerated autoclaved concrete (where the predominant phase is calcium–silicate–hydrate, C–S–H) and its composite with chitosan for cationic dyes malachite green, methyl violet, and MB (maximum adsorption capacities between 256–833 mg/g for powders). Their works are highly encouraging related to the present study since the BFS-based AAM consists essentially aluminum-substituted C–S–H (i.e., C–A–S–H).

Alkali-activated highly porous materials are a new emerging topic related to the use of AAMs as adsorbents since they can be regenerated more readily and require no separation steps unlike powder-form adsorbents. The direct foaming method is commonly used to prepare alkali-activated porous materials (Bai and Colombo 2018): there, a blowing agent, such as hydrogen peroxide (H2O2), peracetic acid (CH3COOOH), or metallic aluminium (Al0), is used to introduce pores to the AAMs, and a surfactant is used to prevent the coalescence of bubbles (Bai and Colombo 2018; Bhuyan et al. 2023).

Alkali-activated BFS (AABFS) has already been proven as a promising adsorbent for water and wastewater treatment (Luukkonen et al. 2016; Runtti et al. 2016; Latorrata et al. 2021; El Alouani et al. 2021). The development of solid foam by alkali-activation of BFS with the direct-foaming method and its use as an adsorptive filter in wastewater treatment was reported earlier (Bhuyan et al. 2022). From this work, it was observed that surfactants greatly influence the properties of AABFS foam, for example, porosity, but the effect of different types of surfactants used in the AABFS foam preparation on the adsorption performance is not studied in detail. The use of different surfactants is expected to impact the porosity and pore structure of the foams through the change of viscosity of the fresh-state paste resulting from the reaction of the surfactants with soluble compounds, which can cause differences in the adsorption performance.

Lignin is a component of wood that is obtained as a by-product from pulp production and is mainly used as a fuel for power and heat generation (Sharma et al. 2021). The amount of lignin generated from the pulp industry is more than 70 million tonnes annually (Saad and Hawari 2013). It is, after cellulose, the second most abundant natural biopolymer, comprising 18–35 wt% of wood (Shorey et al. 2021). Globally, the availability of lignin in the biosphere is approximately 300 billion tonnes (Bajwa et al. 2019; Shorey et al. 2021). Because of the presence of numerous functional groups, for example, ether, carbonyl, aliphatic, and aromatic hydroxyl groups, lignin can have various chemical interactions with different pollutants (such as MB) via electrostatic interaction, H-bonding, and π–π interactions (Suteu et al. 2010). The application of lignin as an adsorbent in removing different pollutants, for example, dyes, potentially toxic elements, or pesticides from water has been reported (Mariana et al. 2021; Wang et al. 2022). The addition of lignin to AAMs to form a composite has been also reported (Ye et al. 2018), but the application of such biomass-AAM composites for wastewater treatment has been investigated only to a very limited extent, and thus represents a research gap. It is hypothesized that the introduction of additional functional groups to the surface via the development of composite materials would increase the adsorption performance of AAMs via enhanced and more diverse molecular interactions between the AAM composite surface and adsorbates (Açışlı et al. 2022).

The main target of the present study is to introduce lignin to the AABFS structure to enhance the adsorption capacity of MB in a dynamic adsorption set-up. The first part of the study addresses the AABFS foam without lignin by (a) selecting the optimum surfactant type (cationic or non-ionic) based on the adsorption performance of MB in a dynamic adsorption set-up; (b) optimizing the factors affecting adsorption: pH, flow rate, and initial concentration of MB; and (c) characterization (by infrared spectroscopy and zeta potential measurements) of the AABFS foams before and after MB adsorption and after thermal regeneration. In the second part of the study, the optimum lignin amount in AABFS foam is selected based on compressive strength, porosity, and specific surface area characterization. The optimized lignin-containing AABFS is used in dynamic adsorption experiments to observe breakthrough curves in comparison to AABFS foam without lignin. The breakthrough curves were fitted to kinetic models, namely, Adams–Bohart, Thomas, and Yoon–Nelson models. This is the first study in which a lignin-AAM composite material is applied for an adsorption application providing insights about the significant improved adsorption performance. This study was carried in Oulu, Finland from October 2021 to November 2022.

Materials and methods

This section is divided into sub-sections describing the used raw materials and chemicals, preparation of alkali-activated foams and introduction of lignin, material characterization methods, and adsorption experiments conducted with MB.

Materials and chemicals

The BFS used in this study (tradename KJ400) was obtained from Finnsementti (Finland), and its composition (in wt%) is 38.5% CaO, 32.3% SiO2, 9.5% Al2O3, 10.2% MgO, 4.0% SO3, and 1.2% Fe2O3. The alkali activator was prepared by stirring a Na2SiO3 solution (Merck, extra pure, molar SiO2/Na2O ≈ 3.5, 65 wt% water) and NaOH pellets (98.7%, VWR) to obtain a SiO2/Na2O molar ratio of approximately 1.2 and the amount of water approximately 56 wt%. H2O2 (30%, VWR, w/v) was used as a blowing agent, and four different surfactants—Triton X-405 (70% solution, Sigma-Aldrich), Triton X-114 (100% solution, VWR Acros), Triton X-100 (100% solution, VWR), and cetyltrimethylammonium bromide (CTAB; ≥ 99 wt% solid, VWR)—were employed in the preparation of foams. Among these surfactants, Triton surfactants are non-ionic and CTAB is a cationic surfactant. The non-ionic surfactants have a difference in their ethylene oxide chain length and cloud point; the chain length of Triton X-114 is the shortest while Triton X-100 is the longest, and Triton X-405 has the highest cloud point among all three. Methylene blue (MB, TCI, ˃ 70%) was used to prepare synthetic wastewater, and lignin (alkaline, VWR TCI) was used to functionalize the AABFS. MB solutions of three different pH were prepared by using a buffer solution of pH 4 (0.1 M acetic acid buffer), 7 (0.05 M phosphate buffer), and 9 (0.05 M bicarbonate buffer).

Preparation of AABFS with different surfactants

The preparation of AABFS foams were conducted by the direct foaming method (Bhuyan et al. 2022). BFS, alkali activator solution and water were mixed for 4 min with a high-shear mixer (IKA EUROSTAR 20) using a speed of 3000 rpm. H2O2 (0.06 wt% of paste) and four distinct surfactants—CTAB, Triton X-100, Triton X-405, and Triton X-114 (0.001 mmol/g paste)—were added, and mixing (speed 3000 rpm) was continued for additional 2 min. Then, the paste was cast in column moulds of diameter 4.3 cm and length 10.2 cm and curing was done at 60 °C inside the oven for 4 h by keeping it in an airtight bag. In the next step of curing, the airtight bag containing the samples was stored at room temperature. Before their use in the adsorption experiments, the neutralization of the pore solution of the foams was done by using 0.1 M acetic acid.

Preparation of AABFS-lignin (AABFS-Lig) composite foams

To prepare the AABFS-Lig composite foam, different percentages (0–1 wt% of the precursor) of alkaline lignin (VWR TCI) were mixed with BFS for 1 min, then the alkali activator solution and water were added in, and mixing was continued for 4 min to get a homogeneous paste. Next, H2O2 and Triton X-114 were poured in the paste and mixed (speed 3000 rpm) for additional 2 min. Triton X-114 was used with composite foams based on the comparative adsorption experiments between the foams prepared with different surfactants. The paste containing lignin was cast, cured, stored, and pre-treated similarly as described above.

Characterization of the AABFS and AABFS-Lig foams

Mechanical strength

Foam samples (50 mm × 50 mm × 40 mm) were characterized for compressive strength by employing a universal testing machine, ZwickRoell Z10 (10 kN), with 1 mm/min loading rate. The measurement was carried out for 3 replicates at 7 d sample age. Equation (1) was employed to determine the compressive strength (σ, [MPa]) of the foam samples:

$${\upsigma } = \frac{{\text{F}}}{{\text{A}}}$$
(1)

here A indicates the surface area (mm2) under compression, and F is the force (N).

Fourier transform infrared spectroscopy (FTIR)

The alteration in the chemical environment of the AABFS after neutralization with acetic acid, and MB adsorption were studied by FTIR (Bruker VERTEX 80v). Powdered samples (particle size < 500 µm) were used for the analysis, and the wavenumber range during the collection of spectra was 500–4500 cm−1 and resolution was 2 cm−1. In the similar way, the change in the chemical environment of AABFS-Lig before and after adsorption of MB was studied.

BJH and BET

Nanometer-scale pore volume and the surface area of the foams were studied by applying the Barrett–Joyner–Halenda (BJH) method and the Brunauer–Emmet–Teller (BET) isotherm for the adsorption and desorption of liquid N2 at − 196 °C. A Micromeritics ASAP 2020 instrument was employed for this analysis. Before measurement, the foam samples were crushed into 4–8 mm pieces, freeze-dried for 72 h, then degassing of the samples was conducted at 70 °C for 24 h.

He-gas pycnometry

A helium gas pycnometer of model AccuPyc II 1340 (Micromeritics) was used to investigate the open and total porosity of the prepared foam samples. Dried foam samples (at 60 °C) were used to calculate the geometric density (ρg [g/cm3]) by determining sample dimensions and their mass by using a calliper and analytical balance, respectively. The helium gas pycnometer was used to determine the apparent density (ρa [g/cm3]) and true density (ρt [g/cm3]) from crushed foam particle sizes of < 1 cm and < 500 µm, respectively. Equations (2, 3) were employed to calculate the open and total porosities, respectively.

$${\text{Open }}\;{\text{porosity }}\left[ {{\% }} \right] = { }\frac{{{\uprho }_{{\text{a}}} - {\uprho }_{{\text{g}}} }}{{{\uprho }_{{\text{a}}} }} \times 100{ }$$
(2)
$${\text{Total }}\;{\text{porosity }}\left[ {{\% }} \right] = { }\frac{{{\uprho }_{{\text{t}}} - {\uprho }_{{\text{g}}} }}{{{\uprho }_{{\text{t}}}}} \times 100{ }$$
(3)

Zeta potential analysis

A Zetasizer pro (Malvern) was employed to determine the zeta potentials of the AABFS foams at pH of 4–9. The powders (0.1 wt%, particle size ≤ 100 µm) were dispersed in 0.1 M HCl, and the dispersion was used for the zeta potential measurements. NaOH and HCl were used to adjust pH.

Adsorption experiments

MB adsorption by AABFS foams using different surfactants

To compare the MB adsorption performance by AABFS foams prepared with different surfactants, the foams were cast in column moulds of diameter 4.3 cm and length 10.2 cm. The geometric density of the foams was 0.85–1.13 g/cm3 depending on the selection of the surfactant (see Table S1, Supporting information, SI). It should be noted that the weights of the foams were detected after curing without drying, and thus they also contain some contribution from water (~ 2 wt%) (Bhuyan et al. 2022). These values were used to calculate the adsorbed amount of MB per mass of the adsorbent (referred from this on as an adsorption amount). The foams were neutralized by using 4 L of 0.1 M acetic acid and 4 L of deionized water by pumping them through the foam with a 1.0 L/h flow rate. A 5 mg/L MB solution was prepared in a 0.05 M phosphate buffer. A peristaltic pump (MINIPLUS 3) was used to distribute 6 L of MB solution through the foam with a flow rate of 1.0 L/h (i.e., empty bed contact time ≈ 9 min). Using 15 min intervals, the filtrate was collected and analysed immediately by using a Shimadzu UV-1800 ultraviolet–visible light (UV–Vis) spectrophotometer at a wavelength of 664 nm. Calibration was conducted for the MB solutions in a concentration range of 1 to 5 mg/L, as shown in Fig. S1 (SI). Based on these experiments, an AABFS prepared with Triton X-114 surfactant was chosen for the optimization of flow rate (0.5 or 1.0 L/h), pH (4, 7, or 9) and influent concentration of MB (5, 10, 15, or 20 mg/L). The cumulative adsorption amount per mass of the foam, qt (mg/g) was determined by employing Eq. 4. However, it should be noted that the cumulative adsorption amount does not reflect the maximum adsorption capacity of the foam as complete saturation was not reached.

$${\text{q}}_{{\text{t}}} = \frac{{{\text{Q}}\mathop \smallint \nolimits_{{0}}^{{\text{t}}} \left( {{\text{C}}_{{0}} - {\text{C}}_{{\text{t}}} } \right){\text{dt}}}}{{\text{m}}}$$
(4)

In Eq.(4), C0 (mg/L) indicates the concentration of the influent, Ct (mg/L) indicates the concentration of the effluent at time t (min), and m (g) indicates the mass of the foam after curing including the contribution of water (~ 2 wt%).

MB adsorption by AABFS-Lig

AABFS-Lig composite foams with different percentages of lignin (0.0–0.8 wt%) were prepared and tested similarly as described above. The geometric densities of these foams were 1.05–0.63 g/cm3 (i.e., higher lignin content decreased the density), as shown in Table S2 (SI). Based on the highest cumulative adsorption amount, the optimum AABFS-Lig composite foam (i.e., with 0.8 wt% of lignin) was chosen to investigate the effect of different initial concentrations of MB. Throughout all experiments, the flow rate of 1.0 L/h, and a pH of 7 were maintained.

Kinetic models

For scaling-up the column adsorption process, the prediction of the breakthrough curve is significant. To study the kinetics of a continuous fixed adsorption column, Adams–Bohart, Thomas, and Yoon–Nelson models (Eqs. 5–7 respectively) were employed in the prediction of breakthrough curve behaviour (Bohart and Adams 1920; Thomas 1944; Yoon and Nelson 1984):

$$ {\text{ln}}\frac{{{\text{C}}_{{\text{t}}} }}{{{\text{C}}_{{0}} }} = {\text{k}}_{{{\text{AB}}}} {\text{C}}_{{0}} {\text{t}} - {\text{k}}_{{{\text{AB}}}} {\text{N}}_{{0}} \frac{{\text{Z}}}{{\text{F}}} $$
(5)

In Eq. (5) (i.e., the Adams–Bohart model), \({k}_{AB}\) [L/(mg × min)] indicates the rate constant, F (cm/min) indicates the linear-flow velocity, \({N}_{0}\)(mg/L) indicates the adsorption capacity of the adsorbent per unit volume of the bed, Z (cm) indicates the bed height, and \({C}_{0}\) and \({C}_{t}\) indicate concentrations of inlet MB and effluent MB, respectively (mg/L).

$$ \ln \left( {\frac{{{\text{C}}_{0} }}{{{\text{C}}_{t} }} - 1} \right) = \frac{{{\text{k}}_{{{\text{Th}}}} {\text{q}}_{0} {\text{w}}}}{{\text{Q}}} - {\text{k}}_{{{\text{Th}}}} {\text{C}}_{0} {\text{t}} $$
(6)

In Eq. (6) (i.e., the Thomas model), \({k}_{Th}\) [L/(mg × min)] indicates the rate constant, \({q}_{0}\) (mg/g) indicates the adsorption capacity of the bed at the breakthrough point, and w indicates the mass of the bed (g).

$$ {\text{ln}}\frac{{{\text{C}}_{{\text{t}}} }}{{{\text{C}}_{{0}} - {\text{C}}_{{\text{t}}} }} = {\text{k}}_{{{\text{YN}}}} {\text{t}} - {\tau k}_{{{\text{YN}}}} $$
(7)

In Eq. (7) (i.e., Yoon–Nelson model), \({k}_{YN}\) (1/min) indicates the rate constant, \(\tau\) (min) indicates the time required for the breakthrough of 50% adsorbate, and t (min) indicates the breakthrough time.

To determine the Adams–Bohart, Thomas, and Yoon–Nelson models parameters, a nonlinear regression was performed by employing the Microsoft Excel Solver function (GRG nonlinear).

Results and discussion

This section is divided into two sub-sections describing the AABFS foam optimization, characterization, and application for MB adsorption without lignin and introduction of lignin to the foam and comparing the properties to the foam without lignin.

Optimization and characterization of AABFS foam without lignin

Selection of the optimal surfactant when preparing AABFS foams

Three non-ionic surfactants (Triton X-100, X-114, and X-405) and one cationic surfactant (CTAB) were used to prepare AABFS foams. The presence of different functional groups in non-ionic and cationic surfactants can modify the surface charge and bubble stability in the fresh-state geopolymer paste and, consequently, could have an effect on viscosity (Perumal et al. 2021). From earlier studies, it is known that CTAB produces higher porosity and specific surface area in comparison to non-ionic surfactants (Bhuyan et al. 2022). Due to this higher specific surface area and porosity, it is expected that AABFS foam with CTAB would show better performance in MB adsorption compared to other non-ionic surfactants. However, MB is positively charged at neutral or basic pH, similarly as CTAB, and, thus, there would be a repulsion between them that could have a negative influence on MB removal (Kuang et al. 2020).

The results obtained for MB removal by using AABFS foam prepared with different surfactants are presented in Fig. 1. The duration of the experiment for MB was 6 h with a flow rate of 1.0 L/h, initial MB concentration of 5 mg/L, and pH of 7. The foam prepared with Triton X-114 had the highest MB adsorption amount, possibly due to its highest specific area. MB removal (as %) by AABFS foam prepared with different surfactants are presented in Fig. S2 (SI), where AABFS foam prepared with Triton X-114 still had ~ 84% removal efficiency after 6 h. Thus, the cumulative adsorption amount reported in Fig. 1 does not represent the maximum adsorption capacity.

Fig. 1
figure 1

Comparison of the cumulative adsorption amount of MB after 6 h adsorption experiment by AABFS foams prepared with different surfactants. The other parameters of the experiment were: flow rate of 1.0 L/h, initial MB concentration of 5 mg/L, and pH of 7

Full size image

Optimizing the adsorption conditions: initial MB concentration, flow rate, and pH

The effect of changing the influent concentration of MB from 5 to 20 mg/L when using the Triton X-114-based AABFS foam is presented in Fig. 2. A representative concentration of MB in textile industry wastewater is typically between 10 and 200 mg/L (Kadam et al. 2011). With the increase of the initial concentration of MB, the cumulative adsorption amount of AABFS (again, the cumulative adsorption amount was calculated after 6 h of continuous adsorption experiment) increased. The likely reason for the increased adsorption when increasing the influent concentration of MB is that the mass transfer driving force (concentration gradient) is enhanced (Nasuha et al. 2010). After a 6 h experiment, the MB removal (as %) in the influent concentration range of 5–15 mg/L dropped by approximately 14–15%, whereas for the influent concentration of 20 mg/L, it dropped by approximately 20% from the initial value of 100% (Fig. S3, SI). This decrease is likely due to coverage of adsorbent active sites with MB molecules (Ebadollahzadeh and Zabihi 2020).

Fig. 2
figure 2

The effect of initial MB concentration on cumulative adsorption amount (calculated after 6 h of experiment) of AABFS foam prepared with Triton X-114 surfactant. The other parameters of the experiment were: flow rate of 1.0 L/h and pH of 7

Full size image

The effect of flow rates (i.e., 0.5 and 1.0 L/h corresponding to ~ 18 and ~ 9 min empty bed contact times, respectively) of MB solution (5 mg/L) on the adsorption efficiency and cumulative adsorption amount of AABFS over 6 h at pH 7 are shown in Fig. S4 (SI). Enhanced MB adsorption was observed with a lower flowrate or longer empty bed contact time, which is a typical behaviour for dynamic adsorption systems (Sittipol et al. 2021). Longer contact time allows more mass transfer of MB to occur into the pores of the foam materials (Cruz-Olivares et al. 2013).

MB adsorption on AABFS prepared with Triton X-114 at pH 4, 7, or 9 was investigated by distributing 6 L of MB solution of 5 mg/L through the foam with a flow rate of 1.0 L/h, as shown in Fig. 3. The duration of the experiment was 6 h. At pH 9, a slightly higher extent of MB adsorption was observed compared to pH 7, which could be because of a more negative surface charge at pH 9 (as shown in Fig. S5, SI). The electrostatic attractions or repulsions between the surface of the foam and MB species in the solution are dependent on pH. At pH 7 and 9, methylene blue exists 100% in a cationic form, but at pH 4 it is approximately 50% in non-ionic undissociated form (Salazar-Rabago et al. 2017). At pH 7 and 9, the cationic MB species are adsorbed by the negatively charged surface of the foam. Surprisingly, however, the highest MB adsorption amount by the foam was found at pH 4, although the surface of the foam is positively charged. This indicates that the foam can possibly adsorb undissociated MB molecules at pH 4, which could be because of another mechanism (e.g. hydrogen bonding) rather than electrostatic interaction (Eldeeb et al. 2022). A similar finding, that is, the highest MB adsorption at pH 3 compared to pH of 4–8, was reported for MB adsorption by the geopolymer prepared from rice husk ash, metakaolin, palm oil fuel ash, and ground granulated blast furnace slag (Maleki et al. 2020). The MB removal by AABFS for pH 4, 7, and 9 were approximately 84–100%, as shown in Fig. S6 (SI).

Fig. 3
figure 3

The effect of solution pH on the MB adsorption capacity (cumulative adsorption amount was calculated at 6 h) by the AABFS foam prepared with Triton X-114 surfactant. The other parameters of the experiment were: initial concentration of 5.0 mg/L and flow rate of 1 L/h

Full size image

Characterization of the AABFS foams before and after MB adsorption

The FTIR spectra of AABFS foam (prepared with the Triton X-114) before and after neutralization with acetic acid, and after MB adsorption are shown in Fig. 4. In every case, the bands at ~ 3475 cm−1, ~ 2350 cm−1, and ~ 1650 cm−1 are the stretching and bending vibration of hydroxyl groups on the foam surface caused by adsorbed water molecules (Puertas et al. 2011; Razeghi et al. 2022). The band at ~ 1425 cm−1 isfrom the carbonation of the AABFS foam surface (Cao et al. 2020). The band at ~ 1125 cm−1 (i.e. Si–O–Si/Al stretching (Bernal et al. 2015)) shifted towards a higher wavenumber after acetic acid-washing, or adsorption of MB, which indicates changes in the chemical structure. After MB adsorption, two weak bands at ~ 2500 cm−1 and ~ 1750 cm−1 appear, which are from the MB structures, namely Chet–N(CH3)2 vibrations and Chet = N+(CH3)2 stretching vibrations (where Chet is a carbon atom in the heterocycle of MB) (Ovchinnikov et al. 2016).

Fig. 4
figure 4

FTIR spectra of the AABFS foam prepared with Triton X-114 surfactant before and after 0.1 M acetic acid washing and after MB adsorption (conducted with acid-washed AABFS). Chet = carbon atom in the heterocycle structure of MB

Full size image

The zeta potential of the AABFS foam prepared with a Triton X-114 surfactant was measured at pH of 4, 7, and 9, and the results are presented in Fig. S5 (SI). The point of zero charge for the foam was observed at pH 4.8 (i.e., the overall surface charge is zero), and thus the surface has a positive charge when pH is < 4.8 and a negative charge when pH is > 4.8. The negative surface charge of AABFS at pH > 4.8 is approximately similar to the other alkali-activated materials (Gunasekara et al. 2015; Hertel et al. 2019). Thus, it is expected that at pH greater than 4.8, the foam can adsorb cationic MB dye via electrostatic interaction. As the pH increases from 7 to 9, the surface is yet more negatively charged, resulting in improved adsorption as reflected in the results of the pH optimization experiments (Fig. 3). The zeta potential of AABFS with the other three surfactants (Triton X-100, Triton X-405, and CTAB) was − 16.3 mV, − 17.5 mV, and − 17.5 mV, respectively, at pH 7.

Introduction of lignin into the AABFS foams

Compressive strength of AABFS-Lig composites

The first selection criteria for the addition amount of lignin (0‒1 wt% of the precursor) was the compressive strength of the foam (from this on, Triton X-114 was used to prepare all foam samples) as shown in Fig. 5. Compressive strength was used as a rough indication of the mechanical durability of the foams, that is, to remain intact in flowing water. There were no clear changes in compressive strength up to the addition of 0.8 wt% of lignin compared to AABFS without lignin; however, a reduction in compressive strength for the addition of 1.0 wt% of lignin was clear. The compressive strength of the foam samples depends on their porosity and the shape and size of pores (Ye et al. 2018). The reduction in compressive strength for the addition of lignin higher than 0.8 wt% could be because of a poor integration of lignin into the AABFS structure at high addition amounts.

Fig. 5
figure 5

Comparison of compressive strength of AABFS foam for the incorporation of different amount of lignin

Full size image

Porosity and specific surface area of AABFS-Lig composites

The apparent densities and porosities of AABFS foams with different addition amounts of lignin are shown in Table S3 (SI). The apparent density of AABFS foam without lignin was 2.33 g/cm3, and after the addition of lignin up to 1 wt%, no clear changes were observed. However, the addition of lignin decreased the specific surface area of foams from 86.48 m2/g to 72.13 m2/g when the lignin amount was increased from 0 to 0.8 wt%, respectively. The cumulative nano-scale pore volume (Fig. S7, SI) was, in all cases, mainly within 2–30 nm-sized pores (i.e., mesopores). On the other hand, a slight increase in total porosity (as measured with the helium gas pycnometer) was observed after the incorporation of different amount of lignin into the foams. The AABFS foam with 0.8 wt% lignin showed ~ 74% total porosity whereas without lignin it was ~ 69%. In all cases, the open porosity and the total porosity were almost equal, which indicated that the pores in the foams were connected (Bhuyan et al. 2022).

MB adsorption with AABFS-Lig composites

AABFS-Lig composites with different amounts of lignin (0–0.8 wt%) were employed for MB removal from MB solution with a concentration of 5 mg/L and pH of 7 (Fig. 6). The composite with 1.0 wt% of lignin was not employed for the MB removal as the foam had lower mechanical strength. Figure 6 shows the effect of the amount of lignin in AABFS on MB removal performance. As the lignin amount was increased, the MB adsorption amount increased up to ~ 93% with the 0.8 wt% lignin addition in comparison to the AABFS without lignin. This is likely due to introduction of various functional groups to the surface, for example, ether, carbonyl, and aromatic hydroxyl groups, which can form electrostatic and π–π interactions as well as hydrogen bonds with MB (Bai et al. 2020; Liu et al. 2021). These increased interactions likely contributed to a higher adsorption amount. A similar observation was reported in the literature for a sodium alginate/lignin composite (Chen et al. 2022). The MB removal (as %) by AABFS foams with different amounts of lignin (0‒0.8 wt%) is presented in Fig. S8 (SI): the MB removal remained at > 85% until 2 h for the AABFS foam without lignin, whereas for the AABFS foam with 0.8 wt% of lignin it was 100%.

Fig. 6
figure 6

The effect of lignin amount in AABFS foam on the MB cumulative adsorption amount. Cumulative adsorption amount was calculated in a 2 h experiment. Two litres of MB solution (5 mg/L, pH of 7) were passed through the foam with a 1 L/h flow rate

Full size image

The adsorption of MB was confirmed also with FTIR analysis of AABFS-Lig composite before and after MB adsorption (Fig. S9). A weak band at around ~ 2507 cm−1 was observed after MB adsorption by AABFS-Lig, which could be attributed as Chet–N(CH3)2 vibrations from the MB structure (Ovchinnikov et al. 2016). Two weak bands at ~ 1405 cm−1 and ~ 1480 cm−1 would be because of C–C stretching vibration and CH2 scissoring in lignin respectively (Piqueras et al. 2020).

The AABFS-Lig composite foam with 0.8 wt% of lignin was selected to study the effect of influent MB concentration. With an increase in the concentration of MB, the cumulative adsorption amount (calculated at 6 h) of the foam increased almost linearly shown in Fig. 7. Again, it should be noted that since the mass of the foam includes approximately 2–3 wt% of water, the actual adsorption amount was higher than the reported values. The MB removal (as %) is shown in Fig. S10 (SI): after a 6 h experiment, the removal-% for the initial MB concentration of 100 mg/L decreased to 82%, while for the initial concentration of 10 mg/L it was 92%.

Fig. 7
figure 7

The effect of initial MB concentration on the cumulative adsorption amount on AABFS with 0.8 wt% lignin, which was calculated after 6 h (flow rate of 1 L/h and pH of 7)

Full size image

To observe the complete saturation and the subsequent maximum adsorption capacity of the optimized foam (i.e. AABFS with Triton X-114 and 0.8 wt% of lignin), an adsorption experiment of 73 h was conducted with a 100 mg/L MB solution using a flow rate of 1 L/h and pH of 7. The experimental data was fitted to the Adams–Bohart, Thomas, and Yoon–Nelson models and the obtained parameters are listed in Table 1. The Adams–Bohart model assumes the correlation between the bed depth and the breakthrough time. The Thomas model assumes that the data follows the Langmuir isotherm and kinetics model of the second order. The Yoon–Nelson model assumes that the adsorption is proportional to the reduction in the adsorption amount (Mekonnen et al. 2021). The R2 values (both 0.97) for the Thomas and Yoon–Nelson models are higher than for the Adams–Bohart model (R2 = 0.85), which indicates a better agreement between the experimental and calculated results for the Thomas and Yoon–Nelson models, and thus these are discussed below. The q0 value for the Thomas model is 39.5 mg/g, which indicates the maximum adsorption capacity (in fact, the maximum adsorption capacity would be yet higher as the foam mass includes some weight of water) of the foam until complete saturation. This maximum adsorption capacity is higher than those reported for other alkali-activated foam materials, as shown in Table 2. From the Yoon–Nelson model, the parameter τ was estimated to be ~ 2262 min (corresponding to 37.7 L of treated water with 100 mg/L MB concentration and a foam mass of 96.7 g mass), which indicates the time needed for a 50% breakthrough. Figure 8 presents non-linear plots of the Adams–Bohart, Thomas, and Yoon–Nelson models fitted to the experimental data.

Table 1 Parameters of Adams–Bohart, Thomas, and Yoon–Nelson models for 100 mg/L MB influent, 1 L/h flow rate, and pH of 7
Full size table
Table 2 A comparison of MB adsorption capacities for different alkali-activated foam materials in the literature to the current study
Full size table
Fig. 8
figure 8

Fitting of a the Adams–Bohart, b Thomas, and c Yoon–Nelson models to the experimental breakthrough curves (○ = Ct/C0 experimental, — = Ct/C0 calculated, ∙∙‧‧‧‧‧ = 95% upper and lower confidence limits)

Full size image

Conclusion

In this study, the effect of different surfactants in the alkali-activation and direct foaming of blast furnace slag were assessed when the obtained foams were used in MB removal in dynamic adsorption experiments. The surfactant type (cationic, non-ionic, and the chain length of the surfactant) affects the porosity and pore structures in the obtained foams, but no evidence was found about the integration of the surfactant molecules to the structures of the foams after they were hardened and flushed with acetic acid. The highest adsorption performance was observed with a non-ionic surfactant, Triton X-114, which was selected for the optimization of operational parameters of dynamic adsorption (i.e., initial concentration of MB, flow rate, and pH).

Furthermore, lignin was introduced into the alkali-activated foam with an aim to enhance the adsorption performance. The optimum lignin addition amount was 0.8 wt% (in the range of 0–1 wt%). The incorporation of lignin in the foam significantly improved the adsorption performance as a ~ 93% increase in the adsorption amount of MB was observed in comparison to the foam without lignin. The highest adsorption capacity for the lignin containing AABFS was 39.5 mg/g, which is among the highest reported for alkali-activated monolithic and/or foam structures.

Limitations of the present study include that not all parameters affecting adsorption were assessed for lignin-containing AABFS in details. Furthermore, regeneration of the composite foam containing AABFS and lignin after adsorption of MB was not investigated in this study. However, this study provides a preliminary proof-of-concept for the enhancement of MB adsorption by a composite foam containing AABFS and lignin.