Degradation of a leather-dye by the combination of depolymerised wood-chip biochar adsorption and solid-state fermentation with Trametes villosa SCS-10

Adsorption into biochar-derived materials and mycoremediation are promising technologies for removing dyes from solid and liquid matrices. This study presents a combined treatment with adsorption into wood-chip biochar and mycodegradation under solid-state fermentation by Trametes villosa for removing the leather-dye Acid Blue 161. In the first stage, untreated wood-chip biochar, NaOH–depolymerised biochar and KMnO4–depolymerised biochar were assessed for their dye removal efficiency by adsorption. KMnO4–depolymerised biochar exhibited the highest adsorption (85.1 ± 1.9%) after 24 h of contact. KMnO4–depolymerisation modified some physical and chemical properties on the untreated wood-chip biochar, increasing the surface area (50.4 m2 g–1), pore size (1.9 nm), and presence of surface functional groups. Response surface methodology coupled with a Box–Behnken design was used to optimise the AB161 adsorption into the KMnO4–depolymerised biochar. The optimised conditions, pH 3.0, dye concentration 100 mg L–1 and sorbent dosage 2 g L–1, led to a higher dye removal efficiency by adsorption (91.9 ± 1.0%). In a second stage, the wood-chip biochar supplemented with nutrients (1% malt extract and 0.5% peptone) was employed as a solid matrix for growing T. villosa and regenerating the dye-saturated material. After 15 days, T. villosa was able to grow (86.8 ± 0.8%), exhibit laccase activity (621.9 ± 62.3 U L–1), and biodegrade (91.4 ± 1.3%) the dye adsorbed into the KMnO4–depolymerised biochar. Finally, the mycoregenerated biochar was reutilised in a new cycle of adsorption reaching 79.5 ± 2.0% of dye removal efficiency by adsorption. This study revealed the potential of the combined treatment and is an initial assessment for developing commercial alternatives for treating leather industry wastewaters.


Introduction
The worldwide production of dyes has increased to satisfy the demand of different industries such as paper and pulp, textile, and leather. Wastewaters from these industries must be treated appropriately to avoid environmental issues. Besides textile, the leather industry is one of the main industries responsible for releasing dye-containing wastewaters. Tanneries contribute to about 8% of the worldwide release of dyes in the environment (Katheresan et al. 2018). Despite this, the treatment of dye-containing wastewaters from tanneries has been less studied than that other industries. Although both textile and leather industries employ similar dyestuffs, leather-dyes have different chemical structures from textile-dyes, since they are synthesised to dye different types of materials (cotton and leather fibres). Acid dyes are the most commonly used dyestuff in the leather industry. Among them, azo and metal complex dyes are preferred due to their broad spectrum of shades, colour quality and uniformity they confer to the final product. Furthermore, wastewaters produced in textile and leather processes may have different characteristics such as pH, total organic carbon, chemical oxygen demand, and chemicals concentration, among others (Pandi et al. 2019).
Wastewaters polluted with dyes are usually related to environmental problems including inhibition of aquatic photosynthesis, depletion of dissolved oxygen and a high concentration of metals such as chromium, cobalt, and copper. In some cases, dyeing effluents can also be associated with health problems, acute and chronic toxicity, carcinogenic and mutagenic effects (Katheresan et al. 2018;Vikrant et al. 2018).
The development of advanced, cost-effective and environment-friendly technologies for treating tannery wastewaters is still necessary to preserve the natural resources. Adsorption has been a widely used technology for removing dyes and other recalcitrant compounds from wastewaters. Activated carbon is the most commonly employed adsorbent because of its excellent adsorption capacities. However, activated carbon production is an expensive and high energy demanding process (Sumalinog et al. 2018;Dai et al. 2019). Various alternative materials have been efficiently utilised as adsorbents of synthetic dyes including leather waste (Gomes et al. 2016), cattle hair waste , algae and fungal biomasses (Da Fontoura et al. 2017;Puchana-Rosero et al. 2017) and biochar-derived materials (Chen et al. 2018;Zazycki et al. 2018).
Biochar has recently acquired great relevance because of its high adsorption performance and low cost (Jiang et al. 2018). Biochar is a carbonaceous by-product from biomass thermal conversion (gasification and pyrolysis), produced from a variety of low-cost and waste-derived feedstocks (Chen et al. 2018;Zazycki et al. 2018). Biochar properties include abundant surface functional groups, high surface area, porosity and stability (Braghiroli et al. 2018;Dai et al. 2019;Oliveira et al. 2017). Biochar chemical depolymerisation has been used for the production of chemicals and advanced nanomaterials such as carbon dots, humic substances and nanosilica (Placido and Capareda 2015;Placido et al. 2019a). Although the main reason for the chemical depolymerisation is the production of advanced nanomaterials, the chemical reaction creates different modifications in the depolymerised biochar structure that improves the adsorption capacity by modifying the surface area, porosity, pH, and presence of functional groups on its surface (Braghiroli et al. 2018).
One of the major negative aspects of adsorption is handling or removing the concentrated pollutants retained in the adsorbent. During adsorption, pollutants are accumulated on the adsorbent surface, resulting in a progressive reduction of the adsorption capacity until saturation (Salvador et al. 2015). Regeneration is a desorption or degradation process that aims to reuse or recycle the sorbent material. Dye-saturated biochar materials have been regenerated by solvent desorption (Lian et al. 2016;Vyavahare et al. 2018), chemical desorption (Bharti et al. 2019;Meili et al. 2019) and thermal desorption (Zhu et al. 2019). Most of these methods are focused on the desorption of the biochar and did not address the further degradation of the dye. Therefore, a complementary process needs to be performed after dye adsorption to avoid the transfer of the pollutant from one matrix to another and guarantee its degradation.
Mycoremediation has been proved for removing synthetic dyes from liquid and solid matrices. In a previous study, the Brazil native fungus Trametes villosa SCS-10 removed 96.84% of the leather-dye Acid Blue 161 in submerged-state fermentation conditions . Most of the studies on biodegradation of dyes are performed in submerged fermentation conditions. However, in recent years, the culture of white-rot fungi on solid matrixes appeared as an alternative treatment for textile dyes (Bankole et al. 2018;Jaramillo et al. 2017). Solid-state fermentation (SSF) is a technique in which microorganisms are cultured on a solid matrix in near-absence or absence of free water. The solid matrix acts as a support for the growth and metabolism of the microorganism (Thomas et al. 2013).
Combined treatments applying adsorption and mycodegradation have been efficient for removing textile dyes: the synthetic dye Red 40 previously adsorbed into corncob waste was biodegraded by a strain of Trametes versicolor under SSF conditions (Jaramillo et al. 2017). Similarly, Cibacron Brilliant Red 3B-A was biodegraded by a consortium of white-rot fungi in SSF (Bankole et al. 2018). To our knowledge, biochar and depolymerised biochar have not been used either as an adsorbent for leather dyes or as SSF matrix, and mycoremediation in SSF has not been employed as a regeneration method for saturated biochar. Therefore, coupling biochar and SSF mycoremediation is a novel method for treating dye-containing wastewaters from the leather industry.
This study aimed to evaluate the combination of depolymerised wood-chip biochar adsorption and solid-state fermentation for biodegradation of the leather-dye Acid Blue 161 and regeneration of the sorbent matrix.
The article includes the following specific aims: • Selection of a depolymerised biochar as an adsorbent matrix; • Assessment of wood-chip biochar and depolymerised biochar as suitable supports for solid-state fermentation with T. villosa SCS-10; • Optimisation of biochar dye removal using a Box-Behnken design (BBD) and response surface methodology (RSM); • Degradation of the adsorbed dye using solid-state fermentation (SSF) with T. villosa SCS-10; • Reutilisation of the mycoregenerated adsorbent.

Dye and chemicals
Acid Blue 161 (AB 161 ) was supplied by Lanxess (São Leopoldo, RS, Brazil). AB 161 is an acid/metal complex dye with an azo chromophore, used in the leather industry. Psychochemical characteristics of AB 161 are included in Additional file 1.
Media components used for microbial growth: malt extract agar, malt extract broth, and potato dextrose agar were purchased from Oxoid (UK). All chemicals were analytical grade.

Fungal strain
Trametes villosa SCS-10 is a Brazilian native strain, isolated in a previous study and selected for its high laccase activity and efficient biodegradation of leather-dyes from aqueous solution and real wastewaters . T. villosa SCS-10 was conserved on malt extract agar (MEA) at 4 °C. For inoculum preparation, the strain was reactivated by cultivation on MEA for 5 days at 30 °C.

Preparation of wood-chip biochar
Wood-chip biochar (WB), NaOH-depolymerised biochar (NWB) and KMnO 4 -depolymerised biochar (KWB), were assessed for their performance as dye adsorbents and as support matrices for T. villosa solidstate fermentation. WB was kindly provided by Common Visions LTD (Swansea, Wales, UK). WB was crushed using a mortar and sieved using a 1 mm mesh. KWB and NWB were prepared by chemical depolymerisation of the wood-chip biochar, according to Placido and Capareda (2015). The reaction mixture consisted of wood-chip biochar (10%) and KMnO 4 (10%) or NaOH (10%) solutions. The chemical depolymerisation reaction was performed at 120 °C for 1 h at 15 psi in an autoclave (Priorclave, London, UK). After the chemical depolymerisation, the liquid and solid phases in the biochar solutions were separated by centrifugation (5000 rpm for 20 min, Legend RT, Sorvall).
The solid phase (depolymerised biochar) was washed with distilled water (1:10 w/v), recovered via vacuum filtration (0.22 µm Millex syringe filters, Millipore, USA) and dried in a convection oven at 105 °C for 24 h. NWB and KWB were stored at room temperature (20-25 °C) for further experiments.

Dye removal performance
Dye removal studies were conducted with WB, NWB, KWB and commercial activated carbon (AC). 20 mg of WB, NWB, KWB or AC were transferred into 50-mL flasks containing 10 mL of 100 mg L -1 of AB 161 at pH 7.0. Flasks were incubated at 150 rpm and 25 ºC using an orbital shaker (Minitron Infors orbital shaker, Switzerland). The contact time was studied for 48 h, taking samples at 0, 3, 6, 12, 18, 24 and 48 h. Dye removal efficiency by adsorption (REA, %) was determined by measuring the solution's absorbance at the dye λ Max (578 nm) (UV-Vis spectrophotometer U3310, Hitachi, Japan) and employing Eq. 1: where REA represents the dye removal efficiency by adsorption (%), C 0 (mg L -1 ) is the concentration at t = 0 and C e is the dye concentration at different sampling times (mg L -1 ).
The amount of dye uptake at equilibrium (q e ) was determined using Eq. 2: where q e represents the amount of dye uptake per gramme of sorbent at equilibrium (mg g -1 ), V is the volume of the dye-containing solution (L), and M is the sorbent mass (g).

Characterisation of wood-chip biochar
Biochars were characterised by microscopic, morphologic and spectroscopic techniques. The biochar with the highest REA was examined before and after the dye removal assays, in order to elucidate the uptake mechanism. A surface area analyser (Nova 4200e model, Quantachrome Instruments, USA) was used to determine the specific surface area (S BET ) by the Brunauer, Emmett, and Teller (BET) multipoint method and the pore size distribution by the Barrett, Joyner, and Halenda (BJH) method. Phase structures were determined using X-ray diffraction (XRD; Bruker D2 Phaser X-ray diffractometer, Germany). Scanning electron microscopy (SEM; Hitachi S4800 SEM Microscope, Japan) was employed to analyse the surface morphology. The surface functional groups were determined by Fourier transform infrared spectroscopy (FT-IR; Frontier spectrometer, Perkin Elmer, USA). FT-IR analyses were carried out in the mid-IR region of 4000-400 cm -1 . The point-of-zero charges (pH pzc ) of the biochars was determined using a Zetasizer NANO-Z S90 (Malvern Instruments, UK). For the pH pzc analysis, the particles were suspended in ultrapure water and sonicated for 10 min. The pH was adjusted using 0.1 N HCl or 0.1 N NaOH solutions.

Optimisation of AB 161 dye removal
A response surface methodology (RSM) coupled with a Box-Behnken design (BBD) was used to determine statistically the optimum conditions for dye removal.
The optimisation experiments were performed with the depolymerised biochar with the highest REA as described above. The independent variables, pH (X 1 ), initial dye concentration (mg L -1 ) (X 2 ) and sorbent dosage (g L -1 ) (X 3 ), were studied at three different levels (-1, 0 and 1) and the response variable was REA (%). The range and levels (coded and uncoded) of variables are listed in Table 1. Specific dye concentrations were prepared via dilutions of a stock solution containing 1000 mg L -1 of AB 161 in ultrapure water. The initial pH was adjusted with 0.1 N HCl or 0.1 N NaOH solutions. The experimental design had 17 experiments, including 12 factorial points and 5 replicates at the central point ( Table 2). The experimental matrix was designed using the Design Expert ® v.11.0 software (Stat-ease, USA). The predicted response, dye removal (Y), was calculated with a second-order polynomial regression model (Eq. 3), as follows: where Y is the predicted response value, b 0 is a constant term, b i is the linear effect term, b ii is the squared effect term, b ij is an interactive effect term, x i and x j are independent variables, and ε is the error of the model. An experiment under the optimal dye removal conditions was performed to confirm the model's prediction.

Desorption assays
The dye-containing depolymerised biochar (KWB-AB 161 ) obtained from the optimal condition experiments was used in the desorption assays. The KWB-AB 161 was washed with distilled water (1:10 w/v) and dried (105 °C for 24 h). Hydrochloric acid, glacial acetic acid, sodium chloride, sodium hydroxide, ethanol and methanol, and distilled water were assessed as eluents (Vyavahare et al. 2018;Daneshvar et al. 2017). Desorption experiments were conducted by mixing 20 mg of dried KWB-AB 161 in 10 mL of each eluent solution. The mixtures were stirred at 130 rpm, 20 °C for 120 min and then separated by centrifugation (3000 rpm for 5 min, Legend RT, Sorvall) (Daneshvar et al. 2017). Dye desorption was calculated by measuring the absorbance of the solution at the λ Max of the dye (578 nm) and using Eq. 4:  where DE is the dye desorption efficiency (%), V e is the volume of the eluent used (L), M s is the mass of the saturated sorbent (g), C e is final dye concentration (mg L -1 ) and q e is the amount of dye uptake per gramme of sorbent at equilibrium (mg g -1 ). The eluents assessed did not show any significant effect on the λ Max of AB 161 under the conditions studied.

Solid-state fermentation in wood-chip biochar
Trametes villosa SCS-10 solid-state fermentation in wood-chip biochar (WB) was assessed in a single factor experiment. Four different nutrient conditions combining malt extract, glucose and peptone were evaluated ( Table 3). The first condition (T 1 , high nutrient supply) utilised the optimal medium composition obtained in a previous study with T. villosa SCS-10 (Ortiz-Monsalve et al. 2017) and consisted of 1% (w/v) malt extract, 1% (w/v) glucose and 0.5% (w/v) peptone. Additional conditions were variations of T 1 concentrations: reduced nutrient supply (T 2 ), low nutrient supply (T 3 ) (Ortiz-Monsalve et al. 2019) and without additional nutrients (T 4 ). The SSF process was performed in 100 mm Petri dishes containing 2 g of WB mixed with 5 mL of sterile water and nutrients. The WB Petri dishes were inoculated with four agar plugs (diameter, 1 cm) from actively growing T. villosa SCS-10 and was incubated at 30 °C for 9 days. The experiment was performed in triplicates using destructive assays for each sampling day. Fungal growth (G, %) and enzyme activity (U L −1 ) were measured at days 0, 3, 6, 9, 12 and 15. Fungal growth was determined with the image processing software Image J ® v. 1.48 (NIH, USA) by analysing photographic records, according to Copete-Pertuz et al. (2019). The fungal growth was expressed in terms of Growth Area (A, %) following Eq. 5: where G is the fungal growth (%), A was the growth area obtained by the Image J software at different days sampled (cm 2 ) and Petridishtotalarea (56.7 cm 2 ).

Enzyme extraction and activity assays
The extracellular enzymes were extracted by transferring the content of each SSF plate into 50-mL centrifuge tubes with 20 mL of citrate-phosphate buffer (0.15 M, pH 5.0) and stirred on an orbital shaker at 4 ºC for 120 min (Jaramillo et al. 2017). After extraction, extractant and biochar were centrifuged, and the supernatant was used for ligninolytic activity measurements. The assays were carried out using 100 μL of enzyme extract in a total volume of 3 mL. Enzyme activities were expressed as U L -1 and defined as the amount of enzyme that catalysed 1 μmol of substrate per minute. Laccase (Lac) activity was determined by following the oxidation of 0.5 mM ABTS in 0.1 M sodium acetate buffer at pH 5.0. The change in absorbance was monitored at 405 nm (ε 405 = 3.6 × 10 4 M -1 cm -1 ) for 5 min (Bourbonnais et al. 1995). Manganese peroxidase (MnP) activity was measured following the catalysis of 0.1 mM DMP in 0.1 M sodium tartrate buffer at pH 5.0, containing 0.1 mM H 2 O 2 and 0.1 mM MnSO 4 . The increase in absorbance was determined at 469 nm (ε 469 = 2.75 × 10 4 M -1 cm -1 ) (Martinez et al. 1996). Lignin peroxidase (LiP) activity was determined by monitoring the oxidation of veratryl alcohol to veratraldehyde. The reaction consisted of 50 mM sodium tartarate buffer (pH 3.0), 2 mM veratryl alcohol and 0.4 mM H 2 O 2 . The change in absorbance was measured at 310 nm (ε 310 = 9.3 × 10 4 M -1 cm -1 ) for 3 min (Arora and Gill 2001). MnP and LiP assays were performed in the absence/presence of cofactors (H 2 O 2 and MnSO 4 ).

Biodegradation and regeneration of the dye-containing depolymerised biochar by solid-state fermentation with T. villosa
To produce a significant amount of KWB-AB 161 , the adsorption process was scaled up 100 times, while maintaining the optimised operational conditions determined by the RSM (pH 3.0, dye concentration 100 mg L −1 and sorbent dosage 2 g L −1 ). The KWB-AB 161 was prepared in 2-L flasks filled with 1000 mL of dye-containing solution at 100 mg L −1 and 2 g L −1 of KWB. Flasks were stirred in a shaker at 150 rpm and 25 ºC for 24 h. REA and q e were calculated according to Eqs. 1 and 2. At the end of the process, KWB-AB 161 was separated by centrifugation (5000 rpm for 10 min), washed with distilled water (1:10 w/v) and dried in an oven (105 °C for 24 h). The solid-state fermentation (SSF) was performed in Petri dishes. 2 g of dried KWB-AB 161 were mixed with 5 mL of water and supplemented with the T 2 nutrient condition (1% malt extract, 0.5% glucose and 0.5% peptone). KWB-AB 161 was inoculated with 4 agar plugs (diameter, 1 cm) from actively growing T. villosa SCS-10 and incubated at 30 °C for 15 days. Lac, MnP and growth were measured as previously described and reported at days 0, 3, 6, 9, 12 and 15. In order to find the dye biodegradation achieved by T. villosa SCS-10 during the SSF, the plates' content was subjected to a desorption process with ethanol (1 M) following the protocol described before. A comparison between the desorption efficiency before and after the SSF was used to determine the dye removal efficiency by mycoremediation (REM,%), according to Eq. 6: where REM represents the dye removal efficiency by mycoremediation (%), m a is the mass of desorbed dye from KWB-AB 161 before the SSF and m d is the mass of desorbed dye from KWB-AB 161 after the SSF sampled at different days.

Reutilisation assays of the mycoregenerated biochar
The biochar recovered from the SSF was considered as mycoregenerated KWB (MKWB), and it was tested in a new adsorption cycle. MKWB was obtained after 15 days of SSF with T. villosa, harvested and dried in a convection oven at 50 °C for 24 h (Zeng et al. 2015). The weight difference between mycelia and MKWB facilitated their separation during drying, and the remaining mycelia were manually removed with a clamp. The dye removal assays were carried out at the optimised conditions described above (pH 3.0, initial dye concentration 100 mg L -1 and sorbent dosage 2 g L -1 ). 20 mg of MKWB were transferred into 50-mL flasks containing 10 mL of AB 161 solution (100 mg L -1 ) at pH 3. Flasks were incubated at 150 rpm and 25 ºC for 24 h. REA and q e were determined using Eqs. 1 and 2, respectively.

Dye removal performance
An initial screening of the dye removal performance of wood-chip biochar-derived materials (WB, NWB, and KWB) was carried out using the leather-dye Acid Blue 161 (AB 161 ). The results obtained were compared with commercial activated carbon (AC). The contact time (0 to 48 h) was analysed to determinate the equilibrium time for maximum removal. The dye removal efficiencies by adsorption (REAs) are shown in Fig. 1. It can be observed that all the biochars removed AB 161 to a different extent. After 48 h, the highest REA was displayed by AC (93.5 ± 0.9%) followed by KWB (85.7 ± 1.7%). The depolymerised biochar exhibited a higher REA than its precursor WB (46.1 ± 1.7%).
In the case of KWB and AC, the AB 161 removal increased sharply within the first 3-24 h and after 24 h remained relatively constant with time, reaching the equilibrium. Hence, 24 h was selected as KWB contact time. NWB and WB's REAs increased during the first 6 h of treatment; however, they did not change significantly after 12 h. WB and NWB's REAs were significantly lower than that of KWB (the ANOVA and Duncan's test are presented in Additional file 1). The highest amounts of dye uptake (q e ) were also obtained with AC (47.9 ± 1.5 mg g -1 ) and KWB (43.9 ± 0.4 mg g -1 ). The high REA and q e values displayed by commercial activated carbon were related to its well-known characteristics as adsorbent (great surface area, high ion exchange capacity and strong affinity to organic and inorganic compounds) (Nautiyal et al. 2016;Sewu et al. 2017;Sumalinog et al. 2018). Though, it should be noted that KWB's REA was 8% less than that obtained with AC. The high REA of KWB indicates its potential for removing AB 161 from wastewaters. Different from activated carbon, WB and KWB are byproducts from waste-to-energy processes and advanced material production where chemicals and carbonaceous nanomaterials are synthesised (Das et al. 2018;Oliveira et al. 2017;Placido et al. 2019a). The use of wastes makes biochar from these processes a superior economic and environmentally friendly option than activated carbons.

Characterisation of wood-chip biochar-derived sorbents
Some physical and chemical properties of WB and KWB were analysed to elucidate the dye removal phenomena. KWB was characterised before and after AB 161 removal assays because it displayed the highest REA and q e in the dye removal screening. Table 4 shows some physicochemical properties of WB and KWB.

XRD diffraction analysis
XRD evidenced that WB and KWB did not exhibit sharp peaks, indicating the non-crystalline nature of the materials and suggesting amorphous structures (Fig. 2a). The empty spaces in the disordered structures of both WB and KWB allowed the coupling of AB 161 molecules with their surface functional groups. A similar crystalline nature was described by XRD diffraction analyses of pecan nutshells biochar (Zazycki et al. 2018), microalgae biochar (Behl et al. 2019) and saccharide-derived biochar (Tran et al. 2017).

BET surfaces analysis
The BET surface areas (S BET ) were 3.40 m 2 g -1 (WB) and 50.4 m 2 g -1 (KWB). KWB evidenced a significant augment in the S BET (approximately 15-fold). The average pore size of KWB (1.9 nm) was also higher than that of WB (0.6 nm). WB and KWB were classified into the microporous region (average pore size < 2.0 nm) (Thommes et al. 2015). According to the International Union of Pure and Applied Chemistry (IUPAC), the KWB N 2 adsorption/desorption isotherm was classified as Type 4 with an H 4 hysteresis loop (Fig. 2b), a typical characteristic of micro-mesoporous carbon materials (Thommes et al. 2015). KMnO 4 is a strong oxidising agent that might deform some nanoporous and microporous structures on the biochar surface, creating new MnO structures which led to the formation of more and larger micro-mesopores (Song et al. 2014). The KMnO 4 reaction increased KWB's surface area, offering more and larger active sites for interacting with AB 161 molecules. The depolymerisation also enhanced KWB's pore size, facilitating the pore diffusion of dye molecules. These results suggested that physisorption (physical adsorption) via pore diffusion was involved in the dye removal mechanism by KWB. Previously, Jiang et al. (2018) augmented the specific surface area and total pore volume of swine manure biochar with a KMnO 4 reaction of the raw material. The chemical modification on the biochar surface improved the adsorption capacity of metals (Cu 2+ and Zn 2+ ) and antibiotics. Similarly, Song et al. (2014) augmented the surface area and porosity of corn straw biochar by KMnO 4 reaction, improving the Cu (II) adsorption capacity of corn straw biochar.

FT-IR analysis
FT-IR spectroscopy was utilised to determine the changes in the surface functional groups during the treatment. FT-IR spectra of WB and KWB, before and after AB 161 removal, are shown in Fig. 2c. All the wood-chip biochar-derived sorbents displayed similar FT-IR profiles, related to common polymers from woody biomass. The bands between 3500-3300 cm -1 indicated the -OH stretching of phenol and alcohol groups associated with cellulosic-derived materials (Chaukura et al. 2017;Zazycki et al. 2018). The non-depolymerised WB had several functional groups on the surface, with signals at 3350 cm -1 (-OH stretching), 1700 cm -1 (COOH, C=O stretching), and 750 cm -1 (aromatic C-H) (Boakye et al. 2019). The peaks at 1590 and 1390 cm −1 were related to C=C-C stretch and C=C skeletal vibrations (Sewu et al. 2017;Bernardino et al. 2017). The peak at 1200 cm -1 was related to C-O-H symmetric bending vibrations (Xu et al. 2016). The FT-IR spectrum of the KWB showed signals at 1580, 1390 and 750 cm −1 related to carbon linkages and the disappearance of peaks at 1200 and 1700 cm −1 , also related to carbon vibration, indicating that the depolymerisation process generated chemical modifications to the biochar structure, where carbonaceous compounds were released into the liquid phase producing a reduction of carbonaceous linkages signals (Placido et al. 2019a, b). Additionally, it should be noted that bands at 3350, 1350 and 1250 cm −1 , ascribed to O-containing groups, exhibited enhanced intensity. The KMnO 4 depolymerisation augmented the number of Mn-O bonds, modifying O-containing functional groups, especially the C-O bond (1350 cm −1 ) and OH stretching (3350 cm −1 ) (Jiang et al. 2018; Song et al. 2014). New intense peaks in the FT-IR profile of KWB were detected between 400-500 cm -1 and were associated with the presence of K-OH (Placido et al. 2019c). Carboxyl groups act as electron acceptors leading to the formation of π-π-electron-donor-acceptor interaction with aromatic molecules on the dye structure. Hydroxyl groups in biochar surface act as π-electron donor sites (Dai et al. 2019).
After dye removal, the KWB-AB 161 FT-IR spectra exhibited the change and disappearance of many peaks demonstrating the interactions between KWB surface functional groups and the AB 161 molecules. The phenolic functional group's peak shifted (from 3350 to 3333 cm −1 ) and reduced its intensity, exposing its participation in the removal phenomena. Similarly, the disappearance of the signals at 1350 and 1250 cm −1 confirmed the involvement of carboxyl functional groups on the dye removal process. These data were in accordance with previous studies with biochar (Xu et al. 2016;Sewu et al. 2017;Ahmed et al. 2019). The appearance of new signals at 950 and 595 cm -1 can be related to S=O and C-N stretching vibrations, evidencing the presence of AB 161 on the biochar as those chemicals bonds belonging to the AB 161 structure (Additional file 1). The FT-IR results offered more evidence about the dye removal mechanism. The KMnO 4 depolymerisation provided a stronger presence of functional groups in the surface, mainly active oxygen-containing groups such as hydroxyl (-OH) and carboxylic (-COOH). FT-IR results revealed the possible participation of physisorption mechanisms via electrostatic attraction/repulsion interactions between the functional groups on the KWB surface and the AB 161 molecules, as additional mechanisms to pore diffusion.

Zeta potential analysis
Zeta potential measurements determined the pH value corresponding to the point-of-zero charges (pH pzc ). KWB zeta potential decreased from about 10.0 mV at pH 3.0 to 1.0 mV at pH 4.5. The pH pzc of KWB was obtained at pH 4.6. Then, the Zeta potential decreased gradually to about − 30 mV at pH > 7.0 (Fig. 2d). The pH pzc of WB was 5.1, similar to other lignocellulosic-derived biochar as switchgrass biochar (pH pzc = 5.1) (Park et al. 2019), sugarcane bagasse biochar (pH pzc = 5.3) (Vyavahare et al. 2018) and wood-chip biochar (pH pzc = 6.0) (Wathukarage et al. 2017). According to data, the pH pzc of WB reduced with the KMnO4 depolymerisation. For pH values lower than 4.6, KWB had a positive surface charge. When the pH value of the solution was higher than pH pzc , the KWB was negatively charged. Biochar-derived materials have numerous functional groups, whose ionisation potential strongly depends on their pH. Thus, the biochar surface charge is also pH-dependent. The charge on the biochar surface can affect the electrostatic attraction between biochar and organic dyes. For pH values < pH pzc , attraction occurs between the positively charged biochar surface and negatively charged compounds. In contrast, when pH > pH pzc , the negatively charged biochar surface attracts cationic compounds (Jiang et al. 2018). The initial dye removal assays were carried out at neutral pH (pH = 7.0 > pH pzc ). At pH 7.0, the KWB surface was negatively charged. AB 161 is an anionic dye with sodium sulfonate groups, negatively charged in aqueous solutions (Aksu et al. 2008;da Fontoura et al. 2017). These initial results suggested that electrostatic attraction was a weaker mechanism than pore diffusion. However, these phenomena were further studied in the optimisation and desorption experiments.

SEM analysis
SEM micrographs agreed with the BET/BEJ and XRD results. WB had a homogenous structure to a certain extent, with visible pores on its surface (Fig. 3a). In contrast, KWB displayed a more heterogeneous structure with a rougher surface, higher superficial area, and multiple protuberances, cavities and pores. SEM micrographs confirmed the structural changes by the depolymerisation reaction. The thermochemical activation carried out on KWB, led to the formation of grooves, pores and cavities with different sizes on the material, which is adequate for dye removal. These cavities and porous were larger in KWB than WB, allowing a higher penetration of AB 161 molecules. Although the KWB surface was rough after the dye removal assays, it displayed a smooth packing of pores and cavities related to dye presence (Chaukura et al. 2017;Zazycki et al. 2018).
The characterisation analyses evidenced that the KMnO 4 reaction provided a greater surface area, larger pore size, and a higher presence of functional groups in the wood-chip biochar surface. These proprieties are related to high adsorption capacity. Therefore, the depolymerisation with KMnO 4 improved the adsorptive capacity of the biochar.
Dye removal by KWB was an adsorption process, where the AB 161 molecules were transferred from the aqueous solution to the biochar surface attracted by noncovalent forces and bonds with KWB functional groups. Physisorption via pore diffusion, π-π electron-donor acceptor interactions and OH-bonding were postulated as the mechanisms involved in AB 161 removal. Since the aim of this study was to evaluate the treatment of AB 161 by the combination of wood-chip biochar adsorption and mycodegradation in solid-state fermentation with T. villosa, kinetic and thermodynamic studies were not addressed.

Optimisation of AB 161 dye sorption into KWB
A Box-Behnken design (BBD) was applied to establish the combined effect of the three selected independent variables, pH (X 1 ), initial concentration of dye (X 2 ) and sorbent dosage (X 3 ), on the dye removal efficiency by adsorption (REA) and to determine the optimal conditions for the dye adsorption. As seen in Table 2, the predicted and experimental values of REA were very similar. The dye removal efficiency quadratic prediction model is described in Eq. 7: where Y represents the dye removal efficiency by adsorption (REA, %), X 1 is the pH, X 2 is the initial dye concentration (mg L −1 ) and X 3 is the sorbent dosage (g L −1 ). The quadratic prediction model was statistically significant to represent the response, and the lack of fit was not significant (Lack of fit = 0.0602 > 0.05). The predicted R 2 of 0.984 was in reasonable agreement with the adjusted R 2 of 0.990, confirming the prediction accuracy of the developed model. The model's ANOVAs are included in Additional file 1. Figure 4a displays the model accuracy by comparing the predicted and experimental dye removal for each run. Figure 4b-d illustrates the response surface plots of the (7) Y = 80.7 − 4.03X 1 − 19.45X 2 + 2.2X 3 + 3.68X 1 X 2 + 1.78X 1 X 3 + 0.675X 2 X 3 − 3.24X 2 1 − 12.39X 2 2 − 3.74X 2 3 , mutual effect of two variables on REA, while the third variable was kept constant at level 0. The double effect of the initial dye concentration and pH is shown in the three-dimensional response surface (Fig. 4b). Both of the variables had a significant effect on REA. The initial dye concentration played an important role in dye removal as REA decreased with the increase in dye concentration from 100 to 300 mg L -1 . These results suggested the saturation of the active sites and reduction of the available surface area of the adsorbent, which agrees with the KWB SEM images. Similarly, the pH also had a strong influence on REA as dye molecules have different ionisation potentials with the pH; thus, the dye net charge is also pH-dependent . KWB had a pH pzc of 4.6, and at pHs below this value, KWB had a positive surface charge. Then, at lower pH, electrostatic attraction took place between the positive surface charge of KWB and the negative charge of AB 161 . In contrast, when the pH increased, the repulsion between OH − ions and the negatively charged groups in the dye arose, affecting the REA. These results were consistent with previous reports about using acidic conditions (pH 2.0-4.0) as optimal for the adsorption of AB 161 into dried fungal and microalgae biomasses (da Fontoura et al. 2017;Puchana-Rosero et al. 2017). These conditions are significant for treating real effluents because leather dyeing wastewaters are generally acid, ranging between 3.0 and 4.0 . Figure 4c depicts the mutual effect of the initial dye concentration and adsorbent dosage. REA increased at higher adsorbent dose, since the number of active sites/pores available on KWB for binding the AB 161 molecules augmented. These results were in accordance with previous studies, where greater adsorption of AB 161 was obtained with a higher dosage of adsorbents (da Fontoura et al. 2017;Mella et al. 2017).
The optimised parameters for maximising the REA were pH 3.0, initial dye concentration 100 mg L −1 and sorbent dosage 2 g L −1 . Under these conditions, the predicted REA was 93%. In order to validate the RSM based on the BBD model, quintuplicate experiments were performed in the selected optimal conditions. In the optimal conditions, KWB led to a higher REA (91.9 ± 1.0%) and q e (46.7 ± 0.4 mg g −1 ). These results confirmed the model's validity (p-value < 0.0001) by the similarity between the predicted (93%) and experimental (92%) REA. The optimised operational conditions obtained REA and q e values similar to those obtained with activated carbon. A comparison between KWB and other adsorbents is also included in Table 4. KWB exhibited higher surface area and higher dye adsorption capacity than other biocharderived materials (e.g. algae biochar or pulp and paper sludge biochar) toward different dyes (e.g. Congo Red, Methyl Orange and Methylene Blue) (Table 4).

Desorption assays
In the current study, desorption was used to remove the AB 161 dye retained on the KWB-AB 161 biochar and to analyse its biodegradation with T. villosa SCS-10. The AB 161 desorption from KWB-AB 161 biochar was evaluated with different eluents (Table 5). Ethanol exhibited the maximum desorption (50 ± 5%), followed by methanol (39 ± 4%). Most solvents are polar compounds that penetrate the porous adsorbent structures, reaching active sites and physically displacing the adsorbate molecules (Salvador et al. 2015). Ethanol was the most suitable solvent as it penetrated the mesopores of KWB, affecting the polarity of functional groups such as hydroxyl and Fig. 4 a Plot of the experimental and predicted dye removal efficiency by adsorption (REA); response surface 3D plots showing the effect of b pH and initial dye concentration (at sorbent dosage 2 g L −1 ), c pH and sorbent dosage (at initial dye concentration 200 mg L −1 ) and d sorbent dosage and initial dye concentration (at pH 5.0) on the dye removal efficiency carboxylic, and displacing the AB 161 molecules . Distilled water desorbed only 2%, indicating the strong interaction between the AB 161 molecules and the surface groups on KWB. NaCl and NaOH solutions exhibited lower desorption efficiency (21±3 and 19 ± 1%, respectively). Acid compounds such as HCl and CH 3 COOH desorbed less than 5%.
These results were consistent with the previous pH analysis and indicated that electrostatic attractions were also present in the adsorption process. All the further desorption assays were performed using ethanol (1 M).

T. villosa solid-state fermentation in wood-chip biochar
Although there are many alternatives for using biochar, to our knowledge biochar have not been used as a solidstate fermentation (SSF) matrix. In the current study, WB with four different nutrient conditions (Table 3) was used as a solid matrix for SSF with T. villosa SCS-10. The higher concentration of nutrients, treatment T 1, led to the highest growth (90.13 ± 0.40%). However, the reduced nutrient addition (T 2 ) also permitted substantial growth (86.76 ± 0.76%). The treatments with low nutrients concentration (T 3 ) and without nutrient supplementation (T 4 ) displayed a growth area ranging between 20 and 50%. The growth was significantly higher in the nutrient added conditions (T 1 , T 2 and T 3 ). However, T. villosa slightly grew in the non-nutrient condition indicating that WB provided some nutrients for fungal growth. Similar to growth percentage, high Lac activities (> 600 U L −1 ) were achieved on T 1 and T 2 treatments (Fig. 5a). T 2 displayed the highest Lac activity (758.85 ± 51.70 U L −1 ). In fact, this Lac activity was higher than that expressed in liquid state fermentations with the same strain (approx. 500 U L −1 ) (Ortiz- Monsalve et al. , 2019. Figure 5b illustrates the profile of fungal growth and Lac activity. The first sign of mycelium production was detected on the 3rd day of culture. The highest Lac values were reached between the 8th and 10th day of cultivation. Manganese or lignin peroxidases activities were not detected in the tested conditions. In SSF cultures, white-rot fungi grow on similar conditions as their natural habitats allowing the production of enzymes and metabolites secreted at those specific conditions (Daâssi et al. 2016).
The solid matrix could be a source of nutrients, or it could be an inert support for fungal growth (Thomas et al. 2013). The nutrient release capacity of biochar depends on feedstock nature. Wood-chip-derived biochars prepared from plant-based feedstocks are potential sources of nutrient such as C, N, P, K, S Mg, Na and Cu (Singh et al. 2010;El-Naggar et al. 2019). The Lac activity and growth of T. villosa were strongly dependent on the C/N ratio and carbon and nitrogen sources (malt extract, glucose or peptone). Similar to SSF culture, T. villosa SCS-10 liquid cultures evidenced a decline in the Lac activity with higher glucose concentration (Ortiz-Monsalve et al. 2019).
To our knowledge, studies using biochar as supports for SSF were not available in the current literature; therefore, a Lac activity comparison was made with studies  employing natural-derived compounds as supports for growing Trametes strains (Table 6). Lac activity expressed by Trametes villosa SSF in WB (approx. 700 U L −1 ) was higher than that of Trametes sp. in SSF of cassava residue (150 U L −1 ) (Li et al. 2014) and Trametes hirsuta in SSF of pine wood-chip and orange peel (480 U L −1 ) (Böhmer et al. 2011) and lower than that of Trametes pubescens in SSF of sunflower seed shells (5603 U L −1 ) (Rodríguez-Couto et al. 2009) and Trametes versicolor in SSF of corncobs (911 U mL −1 ) (Asgher et al. 2017). However, in these studies, laccase inducers, such as copper sulphate or phenols, were applied, or concentrated laccase extracts were produced. The wood-chip biochar supplemented with nutrients was considered a suitable support for the growth of T. villosa and Lac activity production. These results open the door to utilise biochar from thermal conversion as SSF matrices. T 2 was selected as the culture nutrient condition for the subsequent analyses because T 2 provided the best metabolic conditions for maximum Lac activity production.

Biodegradation of adsorbed AB 161 in depolymerised biochar by solid-state fermentation with T. villosa
In order to produce a high amount of dye-containing biochar, the adsorption process of AB 161 into KWB was scale-up 100 times while maintaining the optimal operational conditions. As WB was confirmed as a suitable support for T. villosa growth, the dye-containing depolymerised biochar (KWB-AB 161 ) was used as a matrix for the solid-state fermentation with T. villosa SCS-10 using the conditions previously selected. Desorption using ethanol (1 M) was assessed before and after the SSF and data were compared to determinate the dye biodegradation efficiency of the fungal strain.
KWB conserved the adsorption capacity in scale-up conditions achieving 90.1 ± 1.0% of REA. Desorption (57.8 ± 3.5%) also remained similar to the one obtained in the desorption screening (50.2 ± 4.76%). The SSF supported a fungal growth of 67.8 ± 5.6%. The growth of T. villosa on KWB was significantly reduced when compared to that achieved on WB (87.0 ± 2.0%). This negative effect was probably related to the toxic effect of AB 161 on biomass production  or the modifications on the surface of KWB. Although the fungal growth was significantly reduced, the Lac activity peak of 751.5 ± 55.9 U L −1 at the 9 th day of culture was slightly higher than that of the SSF on WB (621.8 ± 62.3 U L −1 ). A plausible explanation could be the Lac activity induction by the presence of phenolic groups in the KWB-AB 161 surface. Chemical compounds with phenolic nature such as dyes are laccase inducers because of their similarity with lignin's molecular structure (Mann et al. 2015;Rivera-Hoyos et al. 2013).
In the current study, the initial dye concentration before adsorption was 103.4 mg L −1 . After KWB adsorption, approximately 10.2 mg of dye remained in the effluent, and 90.1 mg were adsorbed into the depolymerised biochar. The desorption with ethanol recovered 54.6 ± 2.0 mg of AB 161 . This amount (54.6 mg) was selected as the initial dye mass to be removed in the solid-state fermentation.
The desorption after the SSF led to the recovery of 4.9 ± 0.4 mg of dye, indicating that T. villosa SCS-10 removed 91.36 ± 1.32% AB 161 from KWB. As shown in the dye removal profile (Fig. 6), decolourisation was strongly associated with Lac activity. In the first 3-6 days of solid culture, T. villosa removed just 10-20% of AB 161 . However, after the 6 th day, the dye removal efficiency by mycoremediation (REM) increased progressively until the 9 th day when it reached 85%. In this period, Lac activity increased from 100-300 U L −1 to 751.4 U L −1 , which was the highest peak of Lac activity. Different mechanisms may be involved in dye removal by fungal treatment: breakdown of the dye structure by enzymes (biodegradation) or binding of the dye molecules to the mycelia surface (biosorption) (Kaushik and Malik 2009). Trametes villosa maintained its original colour after the solid-state fermentation (Fig. 6b), suggesting that the main mechanism of dye removal was biodegradation (Kalpana et al. 2012;Saroj et al. 2014). In addition, during the SSF cultivation, the plate's content (KWB and mycelium) were not separated before desorption, and the biomass was also suspended in ethanol. According to this, AB 161 molecules that would be biosorbed in biomass were also removed in the solvent desorption. These results suggested enzymatic biodegradation as the main mechanism of dye removal in the T. villosa SSF.
UV-vis spectroscopy was used to complement the identification of the mechanisms involved in dye removal. UV-vis spectrum analyses (400-800 nm) of the adsorption and solid-state fermentation processes are shown in Fig. 7. During the adsorption on KWB, the characteristic peak of AB 161 at 578 nm decreased proportionally with the rest of the spectra until reaching its lowest value after 24 h (Fig. 7a). In contrast, the complete disappearance of the characteristic peak was observed during the solid-state fermentation with T. villosa. Additionally, changes in the peak patterns between the untreated and fungal treated KWB-AB 161 were observed in the UV-vis spectrum analysis (Fig. 7b). This is a typical behaviour of enzymatic biodegradation of dyes, where the reduction or disappearance of characteristics peaks are related to the breakdown of chromophore groups (Chen and Ting 2015;Kalpana et al. 2012). Therefore, laccase biodegradation was postulated as the main mechanism involved in AB 161 removal during T. villosa SSF. The REM of T. villosa SSF on KWB (over 90%) was similar to other studies with Trametes species (90-100%) using agroindustrial matrices as support for dye biodegradation. These studies reported dye biodegradation via in vitro and in vivo processes (Asgher et al. 2017;Li et al. 2014;Jaramillo et al. 2017). A comparison with these reports is presented in Table 6. Although in vitro biodegradation has been more efficient for treating synthetic dyes, this process may require the extraction of crude extracts, the addition of enzymatic mediators or/and inducers and may be restricted to optimal operating conditions such as specific pH or temperature ranges. The in vivo degradation with T. villosa was efficient without the addition of inducers or mediators, or the purification of crude extracts.

Reutilisation of the mycoregenerated KWB
The mycoregenerated KWB (MKWB, KWB after adsorption and SSF assays) was used for a new batch of adsorption, aiming a cyclic treatment. The adsorption assays utilised the optimised parameters for dye adsorption into KWB (pH 3.0, initial dye concentration 100 mg L −1 and sorbent dosage 2 g L −1 ). After 24 h of adsorption, MKWB removed 79.5 ± 2.0% of AB 161 from the aqueous solution. The decline in the REA was attributed to the saturation of surface functional groups (such as -COO and -OH). However, the q e of MKWB (41.4 ± 1.8 mg g −1 ) was similar to that obtained with KWB during the first batch of adsorption (46.6 ± 0.4 mg g −1 ). In previous studies, Zeng et al. (2015) also described the regeneration of Methylene Blue dye-saturated rice residues by SSF with Phanerochaete chrysosporium. The SSF system led to 85% of colour removal, related with MnP activity. The mycoregenerated sorbents were used in an additional cycle of adsorption and degradation, reaching a colour removal between 65 and 90%. There are few studies on the regeneration of biochar after adsorption of pollutants and the further handling or degradation of the compound. This article is the first report about mycoregeneration of dye-saturated biochar. SSF with Trametes villosa on KWB-AB 161 was an appropriate method for the regeneration and reuse of the saturated adsorbent. The mycodegradation of the adsorbed dye was also a simple and eco-friendly alternative for the final disposal of adsorbed AB 161 .