Simple adsorptive removal of crystal violet, a triarylmethane dye, from synthetic wastewater using Fe (III)-treated pine needle biochar

Untreated and Fe (III)-treated pine needle biochar (PNB) were evaluated at different pH for the removal of toxic crystal violet (CV) dye from synthetic wastewaters. Adsorption kinetics followed the pseudo-first-order kinetics involving intra-particle diffusion process. The adsorption rate constant increased with Fe treatment of PNB especially at pH 7.0. Adsorption data of CV conformed well to Freundlich adsorption isotherms and both adsorption capacity (ln K) and order of adsorption (1/n) of CV were nearly doubled with Fe (III) treatment of PNB at pH 7.0. Desorption of adsorbed CV from both untreated and Fe (III)-treated PNB could be accounted satisfactorily by third-degree polynomial equations. An increase in ionic strength and temperature enhanced dye adsorption onto untreated and Fe (III)-treated PNB. Adsorption of CV was an endothermic and spontaneous reaction with an increase in entropy of the system. FTIR spectra revealed that C = O of carboxylic acid aryls and C = O and C–O–C in lignin residues of PNB reacted with Fe (III) besides the formation of some iron oxyhydroxide minerals. The changes in FTIR confirmed the possible bonding of positively charged moiety of CV with the untreated and Fe-treated PNB. Scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS) revealed the porous surfaces of PNB with clear accumulation of Fe (III) after treatment and deposition of CV dye on surfaces and pores of PNB. Iron (III)-treated PNB at pH 7.0 can serve as an ecofriendly and cost-effective adsorbent for the efficient removal of CV dye from wastewaters.


Introduction
Water pollution poses a global threat to both living species and the environment. This necessitates a greater attention on waters, especially management of polluted water. Wastewaters from the paper, textile, cosmetics, food processing, and plastic industries contain a variety of organic and hazardous chemicals, such as dyes (Kyi et al., 2020). Dyes are one of the major causes of water pollution which not only diminish the clarity of water bodies but also affect the aquatic system with a direct or indirect impact on human health (Khamparia & Jaspal, 2016). Nearly more than 100,000 commercial dyes and 7 × 10 5 tons of dyestuff are produced annually (Yagub et al., Vol:. (1234567890) 2014). Approximately 12% of synthetic dyes are wasted during manufacture and processing, and about 20% of them end up in industrial effluent (Essawy et al., 2008). During the dyeing process, about 30% of the world's dye production is lost, and wastewater can contain up to 10-50 mg dye L −1 (Nethaji & Sivasamy, 2011). In the absence of natural degradation, dyes are abundant in the industrial wastewaters. Studies have shown that the synthetic origin and complex aromatic structures of dyes increase their stability to light, heat, and oxidizing agents. Dyes present in the effluent obstruct the passage of light into water and harm the ecosystem (Kulkarni et al., 2017). Organic dyes are teratogenic, carcinogenic, and mutagenic, and their unintentional and illegal discharge into the environment poses a serious risk to human health (Sun et al., 2015a, b). Dyes used in the industries have been classified into three categories: cationic dyes (all basic), anionic dyes (direct, acidic, and reactive), and non-ionic dyes (dispersive). Cationic dyes are more toxic as they can easily interact with negatively charged cell membrane surfaces and can enter into cells and accumulate in the cytoplasm. These dyes dissociate into positively charged ions in the aqueous systems (Rahmat et al., 2019). Crystal Violet (CV), a cationic dye, belongs to triarylmethane group which is one of the most commonly used synthetic dyes for purple coloration. It is used in the production of black and blue inks for ballpoint pens and printer ink jet manufacturing industries. It is also used for color paints, pharmaceuticals, leather, detergents, fertilizers, varnish, and waxes. CV, like most other dyes, is toxic and carcinogenic with an obstinate classification due to its non-biodegradability, persistence in different environments, and poor microbial metabolization. It destroys cells, binds to DNA, and can cause gene mutations and cancer (Khan et al., 2021). As a result, CV must be removed from wastewater prior to discharge for the survival of both humans and aquatic organisms. It is estimated that 10-20% of triarylmethane dyes from dyeing industries are released to the natural water surface without prior treatment (Schoonen & Schoonen, 2014).
Many technologies have been recently investigated for the removal of dyes in wastewater, including ion exchange, ozonation, coagulation, membrane filtration, catalytic degradation, precipitation, electrochemical treatment, and adsorption. However, the majority of these treatment methods are expensive and require a lot of energy. As a result, adsorbents that are low in cost, are environment friendly, and have a high capacity to adsorb have been deemed as appealing materials for remediation purpose. Many plant-based waste materials (Elsherif et al., 2021;Homagai et al., 2022;Imran et al., 2022;Loulidi et al., 2020) and chemically treated leaf powder (Al-shehri et al., 2021) have been tried for the removal CV dye from waters. To make the adsorption process more viable and cost-effective, a range of bio-based activated carbons (biochars) have been produced from agricultural wastes (Mishra et al., 2021). Biochar is a carbon-rich solid product formed during pyrolysis from thermo-chemical decomposition of biomass at a predetermined temperature of around 400-500 °C in the absence or limited supply of oxygen . It contains functional groups like carboxyl, hydroxyl, and phenolic on its surface and also possesses a high specific surface area (Jung et al., 2013) and surface reactivity to exhibit enhanced adsorption capacity. Plantbased biochars (Khan et al., 2021;Wathukarage et al., 2019), magnetic biochar (Foroutan et al., 2021;Sun et al., 2015a, b), and modifications of activated carbon (Wang et al., 2021) have been evaluated for the scavenging of CV from wastewaters. Biochars derived from seaweeds like Ulva reticulata (Rajagopalan et al., 2022), Ulva prolifera (Ravindiran et al., 2022a, b, c) and Caulerpa scalpelliformis (Ravindiran et al., 2022d) by low-temperature pyrolysis have been used for the efficient removal of some anionic textile dyes. In the Himalayan forests, the disposal of an enormous amounts of pine (Pinus roxburghii) needles is a challenge being faced by the local people as well as natural resource managers as it is one of the main causes of devastating forest fire due to their high resin content (Myers & Rodríguez-Trejo, 2009). The slow decomposition of pine needles also makes their disposal even tougher for manure preparation (Kainulainen & Holopainen, 2002). In the present investigation, we preferred pine needle biochar pyrolyzed at 450 °C (PNB) for sorption studies of CV as it had higher surface area and cation exchange capacity compared to the pine needle biochar prepared at 300 °C (Labanya et al., 2022). Both untreated PNB and Fe (III)-treated PNB were used as adsorbents for CV dye. Iron (III) treatment of PNB was preferred as Fe 2 O 3 particles have been reported to enter in biochar pores to produce more micropores and surface area and increase the abundance of C-OH and O-C = O functional groups (Wang et al., 2020). Moreover, Fe (III) treatment was also preferred in view of its relatively low cost and lower toxicity hazard to the environment compared to other transition metals (Egorova & Ananikov, 2017). This study reports the performance of Fe (III)-treated PNB which is simple to prepare and adopt at the user's end for the effective removal of dye contaminants. The adsorptiondesorption behavior of CV onto untreated and Fe (III)-treated PNB in relation to contact time, dye concentration, pH, ionic strength, and temperature has been investigated in detail.

Materials
Crystal violet dye was purchased from MolyChem Ltd., India. Silver nitrate (AgNO 3 ), ferric chloride (FeCl 3 ), calcium chloride (CaCl 2 ), methanol (CH 3 OH), and other chemicals used in this study were of analytical grade and the glassware were of Borosil (India) made. Pine needle biochar was procured from the Department of Soil Science, College of Agriculture, G. B. Pant University of Agriculture and Technology, Pantnagar, Uttarakhand, India.
Preparation of pine needle biochar and its treatment with iron Dry pine (Pinus roxburghii) needles were collected from the hills of Uttarakhand and washed with tap water (three to five times) to eliminate dust. Pyrolysis of biomass under limited oxygen at 450 °C was performed in controlled temperature muffle furnace for biochar production. Biochar obtained had 59.89% C, 3.75% H, 34.34% O, 1.54% N, 6.57 pH in 1:10 solidwater suspension, 154.2 m 2 g −1 specific surface area, and 0.88 meq cation exchange capacity. Prepared biochar was washed with double distilled water (10 times), 10% HCl, and hot water and again with double distilled water to remove soluble residues. Complete removal of soluble ions was confirmed by the non-appearance of cloudiness in test with 1% AgNO 3 solution. This PNB was then kept in an electric oven (70 °C) to remove all the traces of moisture. About 20 g of the dried PNB was treated with ferric chloride by mixing it with 200 mL of 1 M FeCl 3 solution for 24 h. After washing with distilled water, treated PNB was kept in an electric oven till it was completely dry. Untreated and Fe (III)-treated PNB samples were kept in desiccators to keep them moisture free. No other treatment was given before adsorption experiments.

Calibration curve
A calibration curve was plotted between the absorbance and different concentrations of CV dye (0,10,20,30,40, and 50 mg L −1 ) solution (pH 7.0 and 9.2) at the predetermined λ max (584 nm) on a UV-Vis spectrophotometer (Genesys 10S). Chemical risk assessment was performed using material safety data to detect and control any dangers connected with the usage of CV dye.

Adsorption kinetics of CV dye
Batch equilibrium studies were conducted by taking two sets of 10 centrifuge tubes in duplicate: 0.1 g of untreated PNB or Fe (III)-treated PNB was taken along with 1 mL of stock solution of CV dye (50 mg CV dye L −1 ) and 1 mL of 0.2 M CaCl 2 in all the centrifuge tubes. The final volume was made to 20 mL by adding buffer solution of pH 7.0 or 9.2. All the tubes were kept in a shaker incubator at 25 °C for different time intervals, viz., 0, 0.25, 0.5, 1, 2, 4, 6, 12, 24, and 36 h. The centrifuge tubes were removed in duplicate after each time interval, and were centrifuged at 8000-9000 rpm for 15 min. The supernatant solutions were decanted and their absorbance values were recorded at 584 nm at pH 7.0 and 9.2 using UV-Vis spectrophotometer. The adsorbed amount of CV onto samples was calculated as the difference between initially added concentration of CV and the final concentration of CV left in the solution.
The kinetic data of CV dye adsorption onto untreated and Fe (III)-treated PNB at equilibrium pH 7.0 and 9.2 were fitted to different kinetic equations: (2) First − order model where Q 0 , Q t , and Q e indicated the quantity of adsorbed CV (mg kg −1 ) at 0, t, and equilibrium time, respectively. The constants represented by k 0 , k 01 , k 1 , k 02 , and k 2 indicated the rate constants for zero-, pseudofirst-, first-, pseudo-second-, and second-order model, respectively. The goodness of data fit to a model was examined by linear coefficient of determination (R 2 ) and standard error of estimate (SE est. ). In order to understand the involvement of mechanistic process(es) controlling the adsorption of CV onto untreated and Fe-treated PNB, the kinetic data were fitted to intra-particle (Ho et al., 2000) (Eq. 6) and film diffusion (van Lier, 1989) (Eq. 7) models.
where Qt indicated the adsorbed amount of CV onto PNBs at time "t" and at equilibrium, respectively. C t and C 0 indicated the concentration of nutrient in solution at time "t" and at t = 0, respectively, while k i.d. and k f.d. were intra-particle and film diffusion rate coefficients, respectively. The value of "c" in Eq. 6 indicated the initial adsorption or the thickness of the boundary line and a greater value of "c" showed higher magnitude of boundary line effect. In Eq. 7, the values of w (g), S.A. (m 2 g −1 ), and V (mL) indicated the weight of PNB sample, specific surface area, and volume of solution, respectively.

Effect of concentration on adsorption of CV
For studying the effect of concentration of CV dye on the adsorption onto untreated and Fe (III)-treated PNB, two sets of six centrifuge tubes in duplicate were taken. In each centrifuge tube, 0.1 g of untreated or Fe-treated PNB was added. To each set, 1 mL solution containing 0, 10, 20, 30, 40, and 50 mg CV L −1 and 1 mL of 0.2 M CaCl 2 were added. The final volume in each tube was made to 20 mL by adding buffer solution of pH 7.0 or 9.2. All the tubes were agitated in an incubator shaker (25 °C) for 24 h (equilibrium time). After equilibration, the contents were centrifuged at 8000-9000 rpm for 15 min, the supernatants were decanted, and their absorbance values were recorded at 584 nm using UV-Vis spectrophotometer. The amount of CV adsorbed by untreated or Fe (III)-treated PNB was computed as mentioned in the kinetic study.

Ionic strength
Two sets of three centrifuge tubes were taken in duplicate and 0.1 g untreated or Fe-treated PNB was transferred to each centrifuge tube. One milliliter of 50 mg CV dye L −1 solution and 0.01, 0.1, and 1 mL of 2 M CaCl 2 were added to each set of centrifuge tubes in duplicate. The final volume was made up to 20 mL by adding buffer solution of pH 7.0 or 9.2. The contents were equilibrated in a shaker incubator at 25 °C for 24 h. The contents were centrifuged and their absorbance values were recorded at 584 nm to compute the adsorbed amount of CV onto untreated or Fe (III)treated PNB as mentioned in the kinetic study.

Temperature
Effect of temperature on adsorption of CV dye onto untreated or Fe (III)-treated PNB was studied by taking two sets of six centrifuge tubes in duplicate, and 0.1 g untreated or Fe-treated PNB was taken in each set. Two milliliters of 500 mg CV dye L −1 solution and 1 mL 0.0.2 M CaCl 2 were added to all the centrifuge tubes. The total volume in each tube was made to 20 mL by adding buffer solution of pH 7.0 or 9.2. The respective tubes were equilibrated in an incubator shaker for 24 h at 5 °C, 10 °C, 15 °C, 25 °C, 35 °C, and 45 °C. After equilibration, the contents were centrifuged at 8000-9000 rpm for 15 min, the supernatants were decanted, and their absorbance values were measured at 584 nm. The adsorbed amount of CV was computed as described in the kinetic study.
The observations on effect of temperature on adsorption of CV dye onto untreated or Fe-treated PNB were also used to calculate the thermodynamic parameters such as equilibrium constant (K c = (Q e )/C e )), standard Gibbs free energy change (∆G°), standard enthalpy change (∆H°), and standard entropy change (∆S°). The standard Gibbs free energy change (∆G°) for adsorption of CV onto PNBs was estimated as ∆G° = -RT ln K c . The Q e and C e indicated the adsorbed amount of CV Vol.: (0123456789) at equilibrium and concentration of CV in the equilibrium solution, respectively. The K c indicated equilibrium constant for adsorption, R gas constant, and T temperature (K). Standard enthalpy change (∆H°) and standard entropy change (∆S°) were computed as intercept and slope, respectively, from the graphical plots drawn between ∆G° and T (K).
Characterization by Fourier transform infra-red (FTIR) spectral analysis, scanning electron microscopy (SEM), and energy-dispersive X-ray spectroscopy (EDS) The FTIR spectral features of untreated and Fe (III)treated PNB before and after CV adsorption were examined using a FTIR spectrophotometer (model Perkin Elmer) at Indian Institute of Technology (IIT), Roorkee (India). Scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS) were also performed to get an idea about the changes in surface morphology of untreated and Fe (III)-treated PNB before and after the adsorption of CV using a SEM (model Zeiss Gemini) at IIT, Roorkee (India).

Simulation of CV desorption from PNBs
The desorption of CV adsorbed onto untreated and Fetreated PNB was simulated by equilibrating untreated or Fe (III)-treated PNB loaded with 4 mg CV g −1 in 200 mL of buffers of pH 7.0 or 9.2, and the process was sequentially repeated five times. At each step, the released concentration of CV was monitored by recording the absorbance of equilibrium solution at 584 nm.

Material characterization
A comparison of FTIR spectra of untreated and Fe (III)-treated PNB revealed that with Fe (III) treatment, the absorption feature due to C = O groups of carboxylic acid aryls (1715 cm −1 ) became more accentuated (Fig. 1a, b). A shift of peak due to -OH group from 3425 (untreated PNB) to 3433 cm −1 (Fe (III)-treated PNB) with accentuated absorbance possibly indicated the presence of hydrolyzed Fe (III) besides the abundance of C-OH groups in the Fe-treated PNB. Wang et al. (2020) also noted that Fe (III) modification of biochar increased the H/C and O/C ratio of treated biochar due to abundance of C-OH and O-C = O groups, respectively. A peak due to C = C groups of the lignin residues in untreated PNB (1615 cm −1 ) shifted to a sharp and broader peak at 1621 cm −1 in Fe-treated PNB which reflected some possible restriction in the bond stretching modes. Furthermore, some contribution of C-O-C and C = O groups of alcohols and esters in untreated PNB (1176 cm −1 and 1043 cm −1 ) could also be anticipated for the retention of Fe (III) as the absorbance at these wave numbers was reduced in Fe (III)-treated PNB. A new peak at 880 cm −1 indicated possibly the formation of some oxyhydroxide minerals of Fe in Fe (III)-treated PNB.
Some changes were observed in the absorption peaks of CV dye loaded onto untreated PNB at pH 7.0 and pH 9.2. A shift in -OH peak from 3425 to 3435 cm −1 for CV loaded-untreated PNB at pH 7.0 and to 3430 cm −1 for CV loaded-untreated PNB at pH 9.2 suggested some significant hydrogen bonding of CV onto PNBs (Fig. 1d, f). The peak at 1587 cm −1 in untreated PNB shifted to higher wave number after CV loading and became more intense for pH 9.2 than pH 7.0 solution which possibly corresponded to the involvement of carboxyl group in the adsorption process. The peak at 1176 cm −1 in untreated PNB shifted to 1178 cm −1 and 1191 cm −1 after CV loading onto untreated PNB at pH 7.0 and pH 9.2, respectively, and plausibly suggested the aggregation of CV molecules with the functional groups of PNB. An increase in absorbance was also seen in the peak at 606 cm −1 at both the pH values which represented more C-N stretching. The absorption peaks of C-N vibrations in CV around 1364 cm −1 and 1160 cm −1 shifted to higher wave number and the increased absorption intensities indicated adsorption of CV onto untreated PNB at pH 7.0 and 9.0.
Some changes in the peaks were also noted in CV dye loaded onto Fe (III)-treated PNB at pH 7.0 and pH 9.2 (Fig. 1e, g). A shift to higher wave number and increase in the absorption intensity of peak of C-N stretch and also a shift and decrease in the absorption intensity of peak of C = C indicated the role of these moieties in the bonding of CV onto Fe (III)-treated PNB where N + might directly exchange with exchangeable cations present on Fe (III)-treated PNB or a possible bonding of CV dye due to π-π electron donor-acceptor interaction between π-electron-rich aromatic graphene in the carbonized region of PNB and π-electron deficient aromatic ring of positively charged CV molecules. The absorption peaks of C-N vibrations in CV around 1364 cm −1 and 1160 cm −1 shifted to higher wave number and the increased absorption intensities indicated adsorption of CV onto Fe-treated PNB at pH 7.0 and 9.0; the effects were more pronounced for Fe-treated PNB at pH 7.0.
Scanning electron microscopy (SEM) and electrondispersive X-ray spectroscopy (EDS) analyses provided an idea about morphology of the surface of adsorbents. The SEM/EDS images of untreated PNB, Fe (III)-treated PNB, CV dye loaded onto untreated PNB (pH 7.0 and 9.2), and CV dye loaded onto Fe (III)-treated PNB (pH 7.0 and 9.2) are given in Fig. 2a-f. The structure of untreated PNB showed heterogeneous but highly porous structures (Fig. 2a). On Fe (III)-treated PNB, a clear accumulation of Fe (III) as polymerized hydroxyl species besides welldeveloped pores was noticed (Fig. 2b). A clearly visible thick accumulation of CV dye onto Fe (III)-treated PNB at pH 7.0 (Fig. 2d) was also seen in contrast to PNB at pH 7.0, f CV dye loaded onto untreated PNB at pH 9.2, and g CV dye loaded onto Fe (III)-treated PNB at pH 9.2 CV dye load onto untreated PNB at pH 7.0 (Fig. 2c). This change could be attributed to higher adsorption of CV dye onto Fe (III)-treated PNB than untreated PNB. In Fig. 2e and f, relatively lower adsorption of dye was noted which could be related to some dissolution/degeneration of untreated PNB surfaces at pH 9.2 and also separation of some Fe-hydroxyl compounds from Fe (III)-treated PNB at pH 9.2. The EDS was also performed to identify the accumulation of elements under different PNB samples before and after adsorption of CV. The EDS spectra depicted the % of each element in different PNBs. Untreated PNB had high content of C, O, Mg, K, and Ca while Fe (III)-treated PNB contained high amount of Fe, Ag, Si, and some trace of Ca. A decrease of Fe content was noticed in CV dye loaded onto Fe (III)-treated PNB at pH 7.0 and 9.2.

Adsorption of CV dye onto untreated and Fe (III)-treated PNB
The adsorption isotherms of CV onto untreated and Fe (III)-treated PNB at equilibrium pH 7.0 and 9.2 are shown in Fig. 3. The adsorption of CV onto untreated and Fe (III)treated PNB regularly increased with the increasing concentration of CV in equilibrium solution indicating higher preference of adsorbent for the adsorbate. It is interesting to note that the adsorption isotherms of CV both onto untreated and Fe (III)-treated PNB at equilibrium pH of 7.0 showed a straight line relationship indicating a constant partitioning of the adsorbate onto adsorbent, while at equilibrium pH 9.2, both the adsorbents showed a unique distribution of points conforming to an exponential relationship. This is indicative of a case of cooperative adsorption where the already adsorbed molecules promote further adsorption of the adsorbate and this might be anticipated for a cationic dye like CV with Cl − .
The adsorption data of CV onto untreated and Fe (III)-treated PNB at equilibrium pH 7.0 and untreated PNB at equilibrium pH 9.2 fitted well to Freundlich adsorption isotherm with R 2 = 0.877 (significant at p ≤ 0.05) to 0.946 (significant at p ≤ 0.01) as shown in Table 1. However, in the case of Fe (III)-treated PNB at the equilibrium pH 9.2, the value of coefficient of determination (R 2 ) was 0.746, statistically not significant at p ≤ 0.05. Other investigators (Wathukarage et al., 2019) also reported the suitability of Freundlich adsorption isotherms to account CV adsorption data onto biochars.
In general, the values of ln K (adsorption capacity) and 1/n (intensity of adsorption) were almost doubled with Fe (III) treatment of biochar at equilibrium pH 7.0. On the other hand, an increase in ln K and 1/n with Fe (III) treatment of biochar at equilibrium pH 9.2 was not that spectacular and the values were much smaller in magnitude compared to those at the equilibrium pH 7.0. The loss in sorption capacity of both untreated and Fe (III)-treated PNB at pH 9.2 could be due to a change in the architecture of biochar at higher pH because of the dissolution of some organic components from PNB and possible precipitation of entrapped metal oxides to clog the pores. Iron (III) treatment of PNB at equilibrium pH 9.2 might have also caused hydrolysis and precipitation of noncomplexed Fe (III) ions originally retained on the exchange sites of PNB.

Effect of ionic strength (I) on adsorption of CV onto PNB
The data on the effect of ionic strength (I) on the adsorption of crystal violet onto untreated and Fe (III)-treated PNB at equilibrium pH 7.0 and 9.2 are presented in Table 2. Iron (III) treatment of PNB significantly increased the adsorption of CV onto biochar both at pH 7.0 and 9.2; the effect was more accentuated at pH 7.0. An increase in the ionic strength from 0.003 to 0.03 significantly increased the adsorption of CV onto PNBs possibly due to compression of double diffused layer bringing CV closer to adsorbent surface; however, the increase recorded for Fe (III)-treated PNB at pH 9.2 was statistically not significant probably due to the precipitation of constituent metal oxides and non-complexed Fe (III) present on PNB. Furthermore, an increase in ionic strength from 0.03 to 0.3 caused a significant decrease in CV adsorption onto all PNBs which could be attributed to "salting in" effect where the increase in ionic strength could increase the solubility of a solute.

Effect of temperature on adsorption of CV onto PNB
The data on the effect of temperature on CV adsorption and associated changes in the equilibrium constant (K c ) and Gibbs free energy (ΔG 0 ), enthalpy (ΔH 0 ), and entropy (ΔS 0 ) onto untreated and Fe (III)-treated PNB at equilibrium pH 7.0 and 9.2 are presented in Table 3. The adsorption of CV regularly increased with the increase in temperature; however, the magnitude of relative increase in the adsorption of CV with the increase in temperature was statistically not significant beyond 35 °C for untreated PNB at pH 7.0 and beyond 25 °C at pH 9.2. Similarly, the magnitude of relative increase in the adsorption of CV with the increase in temperature from 5 to 15 °C and from 25 to 45 °C for Fe (III)-treated PNB at pH 7.0 was statistically not significant, whereas at pH 9.2, the relative increase in adsorption of CV onto Fe (III)-treated PNB with the increase in temperature from 5 to 15 °C was statistically not significant. In general, an increase in adsorption of CV onto PNBs with the increase in temperature signified that the interaction of CV with PNB was a sole physical process. The adsorption reaction of CV onto untreated and Fe (III)-treated PNB at equilibrium pH 7.0 and 9.2 was accompanied by large decrease in Gibbs free energy (∆F) indicating highly spontaneous nature of the adsorption reaction. The values of enthalpy change (∆H) in a high range indicated the endothermic nature of the adsorption reaction and also confirmed that higher temperatures could not only provide higher kinetic energy to CV molecules to bond onto untreated or Fe (III)-treated PNB but might also allow their possible entry into the porous structure of PNB. In the case of Fe (III)-treated PNB at pH 7.0, the magnitude of ∆H (39.81 kJ mole −1 K −1 ) was nearly equal to 40.0 kJ −1 K −1 which indicated the possible involvement of ligand exchange in CV adsorption (Sun et al., 2015a, b). Doan et al. (2021) also observed that adsorption of cationic and anionic dyes by nanomagnetite supported on biochar derived from Eichhornia crassipes and Phragmites australis stems was a spontaneous and endothermic reaction.
An increase in the entropy of the system (∆S) with the adsorption of CV onto untreated and Fe-treated PNB could be linked to the release of exchangeable H + /M +n along with water molecules from the hydration sphere of cations present on the PNB surfaces. Kinetics of CV dye adsorption onto untreated and Fe-treated PNB With increasing time intervals, the adsorbed amount of CV dye on PNB increased and then gradually slowed down until the equilibrium was reached (Fig. 4). The quasi-steady state equilibrium was attained at 24 h for both untreated and Fe (III)-treated PNB and only a minimal increase in adsorption was observed after 24 h. Higher adsorption of CV was observed at pH 7.0 compared to pH 9.2. This could be ascribed to the release of some humic acid-like C content from biochars at alkaline pH (Labanya et al., 2022). Ahmad et al. (2014) in his review on "Biochar as a sorbent for contaminant management in soil and water" also mentioned that there was an appreciable increase (96.7%) in the adsorption of MG dye by rambutan seed activated carbon (RSAC) up to pH 8.0, but further increase in pH decreased the amount of dye adsorbed. The values of the coefficient of determination (R 2 ) and standard error of estimation (SE est ) obtained with the fitting of adsorption data to different kinetic models are presented in Table 4. It is clearly evident from the values that the adsorption kinetics of CV onto untreated and Fe (III)-treated PNB could be best accounted by the pseudo-first-order kinetics as this model resulted in the coefficient of determination (R 2 values) in the range of 0.959 (Fe (III)-treated PNB at pH 9.2) to 0.995 (untreated PNB at pH 7.0), all significant at p ≤ 0.01 and the lowest value of SE est in the range of 1.7 (Fe (III)-treated PNB at pH 7.0) to 13.0 (untreated PNB at pH 9.2). Though fitting of the Table 3 Effect of temperature on CV dye adsorption and associated changes in equilibrium constant (K c ), Gibbs free energy (ΔG 0 ), enthalpy (ΔH 0 ), and entropy (ΔS 0 ) onto untreated and Fe-treated PNB Under a given equilibrium pH, the numerical values with dissimilar letters in the subscript indicate significant difference at p ≤ 0.05. U-PNB, untreated PNB, FeT-PNB, iron (III)-treated PNB kinetic data to pseudo-second-order kinetics gave the highest value of the coefficient of determination (R 2 values; all significant at p ≤ 0.01), yet the values of SE est were much higher in the range of 35.0 (untreated PNB at pH 9.2) to 68.2 (untreated PNB at pH 7.0) as compared to pseudo-first-order kinetics. Based on only the criterion of R 2 values, most of the earlier investigations (Wathukarage et al., 2019;Foroutan et al., 2021;Wang et al., 2021) have reported the suitability of pseudo-second-order kinetics for adsorption of CV onto different bio-adsorbents. However, we noted the superiority of pseudo-first-order model over the pseudo-second-order model to account the adsorption kinetics of CV adsorption onto different PNB preparations. The values of intercept (Q e ) and the rate constant (K 01 ) of pseudo-first-order kinetics and the model constants of intra-particle diffusion and film diffusion mechanistic models for adsorption kinetics of CV onto PNB preparations are presented in Table 5.
In general, the values of intercept (Q e ) of pseudofirst-order kinetics were lower for Fe (III)-treated PNB preparation compared to untreated PNB, both at equilibrium pH 7.0 and pH 9.2. The rate constant (K 01 ) was higher for Fe (III)-treated PNB compared to untreated PNB at equilibrium pH 7.0, while conversely, the K 01 value for Fe (III)-treated PNB was lower than the value for untreated PNB at equilibrium pH 9.2. This indicated that Fe (III) treatment of PNB at pH 9.2 slowed the adsorption of CV possibly due to the tendency of non-complexed Fe (III) to undergo hydrolysis and precipitation. As regards the mechanism of CV adsorption onto untreated and Fe (III)-treated PNB, the values of R 2 and SE est obtained with intra-particle and film diffusion model constants clearly showed that the kinetics of CV adsorption onto PNBs was best accounted by intra-particle diffusion model. Sun et al. (2015a, b) also reported that the intra-particle diffusion was the limiting step in the adsorption of CV onto Fe 3 O 4 -coated biochar. The higher values of the intercept (c) showed that intra-particle diffusion mechanism was not the sole mechanism controlling the adsorption of CV onto untreated and Fe-treated PNB at both equilibrium pH values of 7.0 and 9.2. The positive values of "c" indicated that some initial interaction of CV with the surfaces of untreated and Fe-treated PNB could be anticipated resulting in higher boundary line effect; however, the subsequent adsorption of CV by pore diffusion was likely to occur. In general, the adsorption process of CV might involve the diffusion of CV molecules from the outer surface to the bulk of the PNBs through internal diffusion within the pores of PNBs. On saturation of the adsorption sites onto PNBs, the adsorption rate of CV in the later stage of the adsorption could be slowed due to the attainment of equilibrium.

Desorption of adsorbed CV dye from PNB
The sequential desorption data of CV from untreated and Fe (III)-treated PNB preparations did not fit well to Freundlich desorption isotherms (R 2 = 0.560-0.695; all non-significant at p ≤ 0.05, data not presented here) but followed the polynomial patterns (Fig. 5).
In the above-mentioned polynomial equations, the values of intercept and regression coefficients clearly revealed that Fe (III) treatment of PNB enhanced the adsorption capacity of PNB and substantially decreased desorption of CV. The cumulative percent desorption values of adsorbed CV were 24.5, 22.1, 27.4, and 25.4% for untreated PNB at pH 7.0, Fe (III)-treated PNB at pH 7.0, untreated PNB at pH 9.2, and Fe (III)-treated PNB at pH 9.2, respectively. This indicated that Fe (III)treated PNB could largely hold CV dye irreversibly reducing its potential to pollute the aquatic system.
The mechanism of interaction between a cationic dye like CV with untreated PNB might, therefore, involve π-π stacking interaction, H bonding, ion exchange, electrostatic interaction, and pore diffusion (Doan et al., 2021), while with Fe (III)-treated PNB, an additional cation-π interaction could also be operational besides the accentuated effect of all already stated mechanisms in view of surface modifications brought about by Fe (III) treatment (Wang et al., 2020).

Conclusion
Thus, the adsorption of crystal violet by untreated and ferric (Fe) chloride-treated pine needle biochars at pH 7.0 and 9.2 conformed to pseudo-first-order kinetics and could be better accounted by intraparticle diffusion process. Adsorption of CV onto untreated and Fe-treated pine needle biochar conformed to Freundlich model, and Fe treatment of pine needle biochar at equilibrium pH of 7.0 almost doubled the adsorption capacity and the intensity of adsorption of CV onto pine needle biochar. Higher temperature and ionic strength of 0.3 encouraged the adsorption of CV onto pine needle biochars. The desorption of adsorbed crystal violet from pine needle biochars can be simulated using third-degree polynomial equations and the extent of CV desorption remains around 22.1% for Fe-treated pine needle biochar. Iron-treated pine needle biochar can be used for the removal of crystal violet dye from waste waters. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.