Sage Journals: Discover world-class research

Abstract

This study proposes the use of activated charcoal made from Umbaúba wood as an adsorbent for the removal of naphthenic acid in an aviation kerosene model mixture. The activated charcoal was characterised as mesoporous with a carbon graphite profile and presented pH_pzc equal to 10.5. The best working conditions were obtained for activated charcoal levels of <0.09 mm and 300 r min⁻¹. The system reached the equilibrium after 360 min, without significant statistical difference for the pseudo-first- and pseudo-second-order kinetic models. The Weber–Morris and Boyd models corroborated the conclusion that adsorption is not controlled only by the intraparticle diffusion step. For the equilibrium study, the adsorptive capacity obtained was of 1.1 g g⁻¹, with the Brunauer–Emmett–Teller model better correlating with the experimental data. Given the results obtained, the activated charcoal demonstrated to have a remarkable potential for removing naphthenic acid in an aviation kerosene model mixture.

Keywords

Adsorption agro-industrial residues naphthenic acids Umbaúba wood

Introduction

The sources of crude oil in South America are the most acid in the world, consisting of heavy oils with a high relative density (<20° API), high viscosity and high levels of total acidity. This high acidity can lead to more severe corrosion due to the presence of certain compounds, such as naphthenic acids, which are also responsible for forming insoluble salts and deposits under heating (Farah, 2012; Gruber et al., 2012).

Naphthenic acids, when present in aviation kerosene (AVK), are the main source of corrosion in jet engines (Farah, 2012). For being chemically stable and non-volatile, these acids can act as emulsion surfactants. Furthermore, they are highly toxic for several organisms, with a harmful effect to the environment, mainly to aquatic environments (Clemente and Fedorak, 2005).

Several ways of dealing with this problem have already been suggested, such as liquid–liquid extraction (Shi et al., 2008), biodegradation (Quesnel et al., 2011) and catalytic esterification (Wang et al., 2014). However, these technologies are not entirely efficient or are not economically viable in the treatment of these compounds, due to their restricted use, associated with high costs, the destruction of the molecules of naphthenic acids and the treatment time.

Among the treatment processes suggested, adsorption has proved to be an efficient method for the removal of contaminants, as well as offering the possibility of regeneration, recovery and recycling of the adsorbent material (Eren et al., 2010).

The study of the removal of naphthenic acids, by adsorption, is relevant due to the non-destructive character of the process, thus enabling the purification and commercialisation of the acids removed. This allows the development of recycling methods of these removed acids, justifying the economic viability of the removal process (Gruber et al., 2012).

Among the adsorbents recommended, activated charcoal is one of the most used for the removal of industrial contaminants. The material is industrially produced by the oxidative pyrolysis of wood, mineral coal, bones and coconut shell, among others. Its adsorption capacity depends on the precursor material as well as on the activation and carbonisation methods used (Gonzalez and Cid, 2005).

The use of agro-industrial residues as adsorbents is considered to reduce the environmental impact in two different ways, with the polluting residual biomass accumulated being removed from its manufacturing plant or from the place it is deposited, and allowing the residual water to be treated with this biomass (Bhatnagar and Sillanpää, 2010).

In this context, researchers have investigated the development of new types of activated charcoal made from biomasses, such as nut shells and olive seeds (Martínez et al., 2006), mandioca shell (Sudaryanto et al., 2006), pine fruit (Royer et al., 2009), coconut shell (Cazetta et al., 2011), cotton fibres (Sun et al., 2012), bamboo (González et al., 2014), rice hulls (Muniandy et al., 2014), pine tree (Tonucci et al., 2015), potato residue (Zhang et al., 2015), papaya shell (Abbaszadeh et al., 2016) and guava seeds (Pezoti et al., 2016) and used in the removal of pollutants, such as heavy metals and organic compounds.

The use of Umbaúba wood as a precursor for the manufacturing of activated charcoal is advantageous due to the rapid growth of the tree, making it a renewable source of feedstock. The Umbaúba tree is characteristic of riparian forests, of large forest clearing, and is found from Mexico until Argentina, being dominant throughout all of the Brazilian territory (Hernández-Terrones et al., 2007). In the literature consulted, no reports were found of its use as a precursor in the preparation of activated carbon, nor of its application in the removal of naphthenic acids in AVK.

Taking this into account, this work was aimed at physically evaluating the use of activated carbon with CO₂ made from the Umbaúba wood, as an adsorbent for the removal of naphthenic acids present in the model mixture of AVK and also determining the kinetic and isothermal parameters of the adsorption process.

Materials and methods

Preparation and quantification of the AVK model mixture

The n-dodecanoic carboxylic acid was chosen as the model compound to represent naphthenic acids, due to its physical properties and similarity with the acid compounds frequently found in AVK, according to Nascimento et al. (2014).

The stock solution of the 8% (v/v) AVK model mixture was prepared from 6 g of n-dodecanoic carboxylic acid (Merck, 99% purity) dissolved in 100 ml of n-dodecanoic solvent (Merck, 99% purity). The working solutions (1–6%) were obtained by diluting the stock solution, pH 6 (natural solution pH).

The concentration analyses of n-dodecanoic acid were carried out by gas chromatography, using a Varian Chromatograph CP-3380 (USA), with a Carbowax capillary column (30 m × 0.32 mm × 0.25 µm) and a flame ionisation detector, under the following conditions: oven temperature at 220°C, temperature of the vaporiser at 250°C, temperature of the detector set to 260°C, H₂ used as the mobile phase, 2.0 ml min⁻¹ outflow pressure of 5 psi and injected volume of 1 µl. The samples were analysed before and after each experiment, with blank experiments being carried out following the same procedures.

Charcoal preparation and activation

The charcoal made from Umbaúba wood, used in the activation process, was provided by Elephant Chemical Industry Ltd.

The physical activation process took place in two separate steps. In the first step, the Umbaúba wood was submitted to a 10°C min⁻¹ heating ramp, starting at ambient temperature (25 ± 2°C) up until 600°C, in a N₂ inert atmosphere at an outflow of 100 ml min⁻¹, being kept at this temperature for 60 min. Subsequently, the N₂ outflow was changed to CO₂, at an outflow of 100 ml min⁻¹ for an additional 60 min. After the material had reached room temperature, it was classified in a Tyler sieves series into three different groups of particle sizes <0.09, 0.09–0.15 and 0.15–0.21 mm.

Activated charcoal characterisation

The activated charcoal was characterised through the following techniques:

Point of zero charge. The value of pH at the point of zero charge (pH_pzc) of the activated (C_c) charcoal was estimated using the pH measurement of water before and after the contact with the solids. A total of 0.1 g of charcoal for 25 ml of the solution with pH ranging from 2 to 11 was used in the trial, being continuously stirred for 24 h. The pH of the solutions was adjusted using a pH meter (Quimis, Q400AS) with HCl and NaOH (0.1 mol l⁻¹), respectively.

Textural characterisation. The specific surface area of the material was determined through the adsorption of N₂ at 350 ± 5°C in a BELSORP-MINI equipment, from Bel Japan Inc. Initially, 200 mg of the sample was pre-treated at 333 K under vacuum conditions (DEGASS) for 3 h.

X-ray diffraction. The charcoal was characterised in a BRUKER X-ray diffractometer (model D8 Advance), using a Cu-Kα radiation source, at a voltage of 30 kV and a current of 30 mA. The data were collected at a 2θ range from 5° to 80° with a spacing of 0.05° and a time of 2.0 s.

Fourier transform infrared spectroscopy. The absorption spectra were obtained from a BRUKER spectrometer (model VERTEX 70), using the attenuated total reflection technique. The data were collected in the infrared region between 4000 and 400 cm⁻¹, with a spectral resolution of 2 cm⁻¹.

Scanning electron microscopy with energy dispersive X-ray spectroscopy. The images were obtained using energy dispersive spectrophotometer connected to a SHIMADZU scanning electron microscope (Superscan SS-550 model) at 15 K. The charcoal was placed on a carbon tape and, subsequently, a thin layer of gold was added to further improve the sample handling.

Factorial design

The factorial design was used to evaluate the influences of the variables particle size (<0.09, 0.09–0.15 and 0.15–0.21 mm) and stirring speed (0, 150 and 300 r min⁻¹) on the adsorption process. The trials were carried out in a random order, and the factorial design used was a 2² type with a central point run performed in triplicate, in order to guarantee the required repeatability of the experimental data. The response used to determine the efficiency of the process was the adsorptive capacity, q (g g⁻¹).

The values of q were obtained from equation (1)

q_{e} = \frac{(C_{0} - C_{f}) V_{L}}{m_{A}}

(1)

with C_o and C_f being the initial and final concentrations of the solute in solution (g l⁻¹), V_L the volume of the solution (l) and m_A the mass of the adsorbent (g).

The 0.05 g of C_c that remained in contact with the 5 ml of the AVK model solution was used at 1% (v/v) during 360 min at a temperature of 28 ± 2°C, on a shaker (IKA, KS 130 control), according to Nascimento et al. (2014).

The calculations of the effects of the factors and interactions among them, as well as the respective standard errors, were made according to Barros Neto et al. (2007) using the Statistica 6.0 programme.

Kinetic adsorption study

Based on the conditions established by the factorial design, a kinetic study was carried out aiming at determining the system’s equilibrium time. The kinetic trials were carried out by adding the 0.05 g of C_c to the 1% (v/v) AVK model solution (5 ml), with the samples being filtered at time intervals from 5 to 540 min, at a temperature of 28 ± 2°C.

The kinetic pseudo-first-order (equation (2)) and pseudo-second-order (equation (3)) kinetic models, the Weber–Morris model (equation (4)) and the Boyd model (equation (5)) were adjusted to the experimental data obtained using a non-linear regression model with the Origin 8.0 computer programme. The models that best adjusted were compared using an F-test, according to Montgomery (2012)

\frac{d q_{t}}{d t} = k_{1} (q_{e} - q_{t})

(2)

\frac{d q_{t}}{d t} = k_{2} {(q_{e} - q_{t})}^{2}

(3)

with q_e and q_t being the amounts of solute adsorbed by unit weight of adsorbent (g g⁻¹), in the equilibrium and at a time t (min), respectively. k₁ is the pseudo-first-order adsorption rate constant (min⁻¹) and k₂ is the pseudo-second-order adsorption rate constant (g g⁻¹ min⁻¹) (Ho and Mckay, 1999; Lagergren, 1898)

q_{t} = k_{d i f} t^{\frac{1}{2}} + C

(4)

where k_dif is the intraparticle diffusion constant (g g⁻¹ min^−1/2) and C (g g⁻¹) a constant related with diffusion resistance (Weber and Morris, 1963)

B t = - 0.4977 - \ln (1 - \frac{q_{t}}{q_{e}})

(5)

with Bt a mathematical function of

\frac{q_{t}}{q_{e}}

(Boyd et al., 1947).

Adsorption equilibrium study

Based on the conditions established by the factorial design, an equilibrium study was developed aimed at determining the maximum adsorptive capacities for the C_c charcoal.

It was utilised, in this study, 5 ml AVK model solution with concentration ranging from 0.5 to 5%, being in contact with 0.05 g of the adsorbent under constant stirring at 300 r min⁻¹ at 28 ± 2° C. After 360 min, the samples were filtered and quantified.

The Langmuir (equation (6)), Freundlich (equation (7)), Fritz–Schlünder (equation (8)) and BET (equation (9)) adsorption models were adjusted to the experimental data by using the MATLAB version 7.8.0 programme, in order to evaluate the equilibrium data of the process

q_{e} = \frac{q_{m} \cdot K \cdot C_{e}}{1 + K \cdot C_{e}}

(6)

with C_e being the concentration of solute at equilibrium (g g⁻¹); q_m the Langmuir parameter that represents the maximum solvent capacity (g g⁻¹) and K the adsorption constant of equilibrium, which expresses the affinity between the adsorbent and the adsorbate (l g⁻¹) (Langmuir, 1916)

q_{e} = k {(C_{e})}^{\frac{1}{n}}

(7)

where k (g g⁻¹) (g l⁻¹)^−1/n and n, experimental constants; k indicating the adsorption capacity of the adsorbent and n indicating the effect of the concentration on the adsorption capacity and representing the adsorption intensity (Freundlich, 1906)

q_{e} = \frac{K_{F S} \cdot C_{e}^{b_{1}}}{1 + a \cdot C_{e}^{b_{2}}}

(8)

with K_FS (l g⁻¹)^−b₁ and a (g l⁻¹)^−b₂ Fritz–Schlünder constants and b₁ and b₂ are heterogeneity factors (Fritz and Schlünder, 1981)

\begin{array}{l} q_{e} = \frac{q_{S} b C_{e}}{(C_{S} - C_{e})} \\ \times \frac{1 + (\frac{n_{C} g}{2} - n_{C}) {[\frac{C_{e}}{C_{S}}]}^{n_{C} - 1} - (n_{C} g - n_{C} + 1) {[\frac{C_{e}}{C_{S}}]}^{n_{C}} + (\frac{n_{C} g}{2}) {[\frac{C_{e}}{C_{S}}]}^{n_{C} + 1}}{1 + (b - 1) [\frac{C_{e}}{C_{S}}] + (\frac{b g}{2} - b) {[\frac{C_{e}}{C_{S}}]}^{n_{C}} - (\frac{b g}{2}) {[\frac{C_{e}}{C_{S}}]}^{n_{C} + 1}} \end{array}

(9)

where q_S being the saturation concentration of the monolayer; C_s the saturation concentration on the monolayer (g l⁻¹); b the interaction parameter between the adsorbate and the adsorbent solid surface; n_C the number of layers predicted by the BET model; and g the ratio between the desorption and adsorption rate constants on each layer, being a function of the extra energy required to desorb the adsorbate (Q), given by equation (10) (Do, 1998)

g = \exp [\frac{Q}{R T}]

(10)

with R the universal gas constant and T the temperature (K).

Results and discussion

Characterisation of the carbonaceous materials

The C_c charcoal exhibited a pH_pzc equal to 10.5. Activated charcoals physically at high temperatures are generally characterised by their basic surface, according to Matos et al. (2011). A similar result was found by Largitte et al. (2016) who obtained a pH_pzc of 9.6 and 10.3 for the activated charcoal made of guava seeds and almond shells, activated with CO₂, respectively.

According to Ould-Idriss et al. (2011) below the pH_pzc the material exhibits positive charges on its surface, favouring the adsorption of anions. Consequently, above the pH_pzc, the material has negative charges on its surface, thus favouring the adsorption of cations. As the pH_pzc > pH of the model mixture of AVK (4.1), there is a predominance of particles positively charged on the surface of the charcoal, favouring the adsorption of naphthenic acid.

The results obtained in the textural characterisation by adsorption/desorption of N₂ for the C_c were surface area of 243.2 m² g⁻¹, pore volume of 0.13 cm³ g⁻¹ and a pore diameter of 22 Å. The charcoal studied exhibited a pore diameter within the range of 20–500 Å, being classified as mesoporous, according to the International Union of Pure and Applied Chemistry (Thommes et al., 2015). The surface area found was within the range of values reported for activated charcoal, according to Habila et al. (2015) and Rodrigues et al. (2011).

In the X-ray diffraction for the C_c, peaks overlapping the two halos (approximately 24° and 43°) can be noticed in the diffractogram; such peaks demonstrate the presence of crystalline particles in the materials, which consist of residues produced in the carbonisation process. The halos in such positions are characteristic of the diffraction profile of carbon graphite and are attributed to the reflections at both planes (002) and (101). The latter represents the overlapping of reflections at the 100 and 101 planes. Similar results were obtained by Huang et al. (2011) for activated charcoal of stalks lotus.

The adsorption spectra in the infrared region, with Fourier transform, of the material studied are exhibited in Figure 1.

Figure 1.

Infrared spectrum for the C_c.

For the C_c, the 3725–2950 cm⁻¹ band was obtained indicating the presence of functional groups associated with the phenolic, alcohol and carboxylic acids O–H. The band at 1640–1510 cm⁻¹ can be attributed to the oxygen functional groups, such as the highly conjugated C = O bonds, the stretching of the C–O bonds, in carboxylic, aldehyde, ketone, ester and carboxylate salt groups (Kumar and Jena, 2015). The band ranging between 1510 and 1335 cm⁻¹ can refer to the C–C stretching vibration in aromatic rings. Peaks at 877 and 750 cm⁻¹ are indicative of the presence of C–H in aromatic ring (Huang et al., 2015).

The C_c presented a wrinkled, heterogeneous and irregular surface, as well as a wide range of particle sizes. Similar results were found by Habila et al. (2015), who prepared activated charcoal from agro-industrial solid residues, obtaining a wrinkled surface with similar pores and cavities.

Factorial design

Based on the experimental data, and using the Statistica 6.0 computer programme, the main effects and interactions, with their respective standard deviations, were calculated, and the Pareto chart and response surfaces of the model in study were drawn.

The effects that were considered as being statistically significant at a confidence level of 95% on the levels studied are those placed after the line (p = 0.05) of the Pareto chart in Figure 2(a). The results of the effects of the factors and the interactions between them demonstrated that the effect Stirring Speed and the interaction between the effects Particle Size and Stirring Speed were statistically significant. The effect of the interaction between both effects can be better visualised through the surface presented in Figure 2(b), in which a higher adsorption capacity can be observed for the level < 0.09 mm and 300 r min⁻¹.

Figure 2.

(a) Pareto chart for the effects calculated and (b) response surface for the adsorption capacity (pure error: 0.00005).

Kinetic study of adsorption

The adsorption kinetics of n-dodecanoic carboxylic acid for the solid phase of the C_c as an adsorbent and the adjustment to the non-linear pseudo-first- and pseudo-second-order models are shown in Figure 3.

Figure 3.

Adsorption kinetics in solid phase using a finite bath system and the adjustment to the pseudo-first- and pseudo-second-order models.

By observing Figure 3, the adsorption capacity, q, increased with the increase of contact time with the C_c, and it can be observed that the removal mostly took place at around 180 min, reaching the equilibrium after 360 min, with the experimental adsorption capacity (q_exp) being equal to 0.57 g g⁻¹.

This trend is caused due to the fact that, at the start, there is a great number of empty sites available for adsorption, which, with time, decreases; other than that, the repulsion forces of the adsorbed molecules tend to inhibit the adsorption at other sites.

Comparing these models using the F-test, it was found that F_cal (1) <Ftab (2.42); thus, both models did not exhibit any significant statistical difference at a confidence level of 95%.

The parameters of the kinetic models calculated are presented in Table 1.

Table 1.

Kinetic parameters calculated for the pseudo-first- and pseudo-second-order models.

Pseudo-first-order				Pseudo-second-order
k₁ (min⁻¹)	q_e (g g⁻¹)	r	Residual sum of squares	k₂ (g g⁻¹ min⁻¹)	q_e (g g⁻¹)	r	Residual sum of squares
0.009 ± 0.001	0.58 ± 0.03	0.94	0.94	0.013 ± 0.004	0.71 ± 0.05	0.94	0.94

Although by using the F-test, according to Montgomery (2012), no significant difference is found between the models evaluated, the q_exp was similar to the q_e for the pseudo-first-order model, but q_e obtained through the pseudo-second-order model is not in agreement with the q_exp. Therefore, it may be suggested that the adsorption of naphthenic acid on activated carbon from Umbaúba wood has a behaviour of first-order reaction according to Ho and Mckay (1999).

The adjustment of the experimental data to the Weber–Morris intraparticle diffusion model is presented in Figure 4.

Figure 4.

Weber–Morris intraparticle diffusion model for the adsorption of naphthenic acid by C_c.

According to this model, when the graph of q_t in relation to t^1/2 results in a straight line passing through the origin, the rate of the sorption process can be considered as being limited by the intraparticle diffusion process. However, as the results obtained were multilinear (Figure 4), the adsorption process was considered as not being controlled only by the intraparticle diffusion step, with two or more steps controlling the process.

The data are represented by three linear phases, with the initial phase (k_dif1) representing the effect of the boundary layer, with external mass transfer, in which the n-dodecanoic carboxylic acid was rapidly adsorbed by the C_c. After 60 min, the adsorption rate was reduced, resulting in the second phase, which extended until 300 min. According to Kumar and Porkodi (2007), this phase refers to the diffusion of molecules to the most internal adsorption sites of the adsorbent (k_dif2).

The equilibrium was reached after 360 min, leading to the decline of intraparticle diffusion, due to less availability of adsorption sites (k_dif3). Comparing the three diffusion constants of the Weber–Morris model (k_dif), it is concluded that k_dif1>k_dif2>k_dif3, as observed in Table 2. The molecules diffusion in the interior of the adsorbent is the determining step of the adsorption process, being confirmed by the decrease of k_dif3 value.

Table 2.

Intraparticle diffusion constants of the Weber–Morris model.

Step	Parameters
Step	k_dif	C	r
1	0.037	∼0.0	0.93
2	0.032	0.063	0.99
3	0.004	0.4799	0.94

According to Bhattacharyya and Sharma (2005), this is another indication that the intraparticle diffusion mechanism is not the only limiting step of the adsorption process, and that other interaction mechanisms (adsorption on the external surface and diffusion into the interior) might be acting simultaneously to it.

The Boyd model assumes, however, that the diffusion is the limiting step of the adsorption process, due to the boundary layer that involves the adsorbent particle. Aiming at ratifying the results obtained by the Weber–Morris model, the experimental data were adjusted to the Boyd kinetic expression (Figure 5).

Figure 5.

Boyd kinetic model for the adsorption of naphthenic acid by C_c.

It was therefore noticed that the experimental data did not follow a linear trend, not passing through the origin. This demonstrates that the adsorption process is controlled both by the effects of intraparticle diffusion and external diffusion, corroborating with the results obtained by the Weber and Morris model, presented previously.

Equilibrium study of adsorption

The adsorption isotherm of the C_c and the adjustment to the experimental data to the BET equation for types IV and V isotherms are presented in Figure 6.

Figure 6.

Equilibrium adsorption curve of naphthenic acid in the AVK model using C_c.

By analysing Figure 6, it can be observed that the C_c exhibited a maximum value of q_exp equal to 1.1 g g⁻¹. A similar result was obtained by Silva (2007), who obtained q equal to 1.48 g g⁻¹ for the adsorption of naphthenic acid using an MgMCM-41 molecular sieve.

The experimental data applied to the Langmuir, Freundlich and Fritz–Schlünder models are presented in Table 3.

Table 3.

Parameters of the Langmuir, Freundlich and Fritz–Schlünder models.

Parameters
Langmuir
q_m (g g⁻¹)	1.1 ± 0.2
K (l g⁻¹)	0.2 ± 0.1
R	0.760
Residual sum of squares	0.106
Freundlich
k (g g⁻¹) (g l⁻¹)^−1/n	0.26 ± 0.07
n	2.5 ± 0.6
r	0.848
Residual sum of squares	0.067
Fritz–Schlünder
K_FS (l g⁻¹)^−b₁	0.1 ± 0.3
a (g l⁻¹)^−b₂	0.3 ± 1
b ₁	5 ± 5
b ₂	5 ± 5
r	0.772
Residual sum of squares	0.061

It is therefore concluded that the Langmuir, Freundlich and Fritz–Schlünder models did not adjust to the experimental data, with r < 0.9; with only the BET model having adjusted to the equilibrium data in the adsorption of n-dodecanoic acid in AVK using C_c. Since the obtained isotherm (Figure 6) was type IV, the adsorption of n-dodecanoic acid using CC occurred in multilayers. According to Do (1998), the BET isotherm model assumes the same hypotheses presented by the Langmuir mechanism. However, while the Langmuir model only describes the adsorption in a single layer, the BET one can describe the occurrence of a multilayer adsorption. A comparison between the experimental data and the data obtained from the adjustment curve for the adsorption isotherm of C_c is shown in Figure 7.

Figure 7.

Comparison between the experimental and calculated equilibrium data of the adsorption of naphthenic acid in AVK using C_c.

The value of the correlation coefficient (r) of 0.9598 demonstrates an agreement between the experimental and calculated data.

The parameters of this model were obtained from the adjustment of the equilibrium curve of the adsorbent material and are presented in Table 4.

Table 4.

Parameters of the BET model for the types IV and V isotherms for the adsorption of n-dodecanoic acid by C_c.

Reference	Adsorbent	C_o (g l⁻¹)	q_s (g g⁻¹)	b	C_S (g l⁻¹)	n_c	g
Silva (2007)	MgMCM-41	52.5	0.22	2217.10	0.123	7	286,690.00
Silva (2007)	MCM-41	37.5	0.098	225.34	0.026	7	97.62
Nascimento et al. (2017)	Sr-MCM-41	52.5	0.57	403.30	213.00	4	2513.00
This work	C_c	37.5	0.68	220.13	361.21	4	3041.40

BET: Brunauer–Emmett–Teller.

The comparison between the parameters obtained in this work for the BET model with the ones found by using other adsorbents found in the literature for the adsorption of n-dodecanoic acid in an AVK model mixture is presented in Table 4.

According to Table 4, the material studied exhibited a higher C_s in equilibrium with q_s, with that being an indication of higher adsorption on the first layer if compared to the remaining adsorbents, despite the lower number of layers.

Regarding the affinity between the adsorbate and the adsorbent surface (b), the material studied had a similar value to the MCM-41 molecular sieve – a remarkable result, given it is an activated charcoal. The activated charcoal exhibited more than three adsorption layers (n_c) due to the heterogeneity of the process.

The extra energy required to dissolve an adsorbate (Q) for the activated charcoals MgMCM-41, MCM-41 and Sr-MCM-41 was calculated from the values of g presented in Table 4, being equal to 20.1, 31.5, 11.5 and 19.6 kJ mol⁻¹, respectively. The higher the value of Q, the higher was the affinity between the adsorbent and adsorbate. The C_c had a higher energy for a higher desorption, if compared to both MCM-41 and Sr-MCM-41 indicating that the adsorbent studied had an affinity for naphthenic acid.

Conclusion

The activated charcoal produced from the Umbaúba wood is a mesoporous adsorbent, with a heterogeneous surface, and presented a wide range of positive charges on its surface, which favours the adsorption of naphthenic acid.

The best condition for the adsorption of naphthenic acid in the C_c was obtained for a particle size < 0.09 mm and for 300 r min⁻¹, at the levels studied. The equilibrium time of the adsorption process was of 180 min, and there was no significant difference between the pseudo-first- and pseudo-second-order models. The diffusion was considered as not being the limiting step of the adsorption process studied, according to the Weber–Morris and Boyd models. An adsorption capacity of 1.1 g g⁻¹ was obtained for the adsorption equilibrium, with only the BET being adjusted to the experimental data.

The adsorption study of naphthenic acids in an AVK model mixture using activated charcoal made from Umbaúba was considered as being significantly efficient. It is noteworthy that even though the charcoal studied was prepared from biomass, it presented an adsorption capacity similar to the ones found in the literature.

Footnotes

Acknowledgements

The authors are also thankful to Elephant Chemical Industry Ltda. for donating the charcoal. The authors are also thankful to CETENE, LACOM-UFPB and Analytical Center/DQF- UFPE for analysing the results of characterisation.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to thank PETROBRAS for providing the financial resources needed for this work.

Appendix

References

Abbaszadeh

Alwi

SRW

Webb

et al . (2016) Treatment of lead-contaminated water using activated carbono adsorbent from locally available papaya peel biowaste. Journal of Cleaner Production 118: 210–222.

Barros Neto

Scarminio

Bruns

(2007) Como Fazer Experimentos: Pesquisa e Desenvolvimento Na Ciência e Na Indústria. Campinas: Unicamp.

Bhatnagar

Sillanpää

(2010) Utilization of agro-industrial and municipal waste materials as potential adsorbents for water treatment – A review. Chemical Engineering Journal 157: 277–296.

Bhattacharyya

Sharma

(2005) Kinetics and thermodynamics of methylene blue adsorption on neem (Azadirachta indica) leaf powder. Dyes and Pigments 65: 51–59.

Boyd

Adamson

Meyers

(1947) The exchange adsorption of ions from aqueous solution by organic zeolites II. Kinetics. Journal of the American Chemical Society 69: 2836–2848.

Cazetta

Vargas

AMM

Nogami

et al . (2011) NaOH-activated carbon of high surface area produced from coconut shell: Kinetics and equilibrium studies from the methylene blue adsorption. Chemical Engineering Journal 174: 117–125.

Clemente

Fedorak

(2005) A review of the occurrence, analyses, toxicity, and biodegradation of naphthenic acids. Chemosphere 60: 585–600.

(1998) Adsorption Analysis: Equilibria and Kinetics. London: Imperial College Press.

Eren

Cubuk

Ciftci

et al . (2010) Adsorption of basic dye from aqueous solutions by modified sepiolite: Equilibrium, kinetics and thermodynamics study. Desalination 252: 88–96.

10.

Farah

(2012) Petróleo e Seus Derivados. Rio de Janeiro: LTC.

11.

Freundlich

HMF

(1906) Over the adsorption in solution. Journal of Physical Chemistry 57: 385–471.

12.

Fritz

Schlünder

(1981) Competitive adsorption of two dissolved organics onto activated carbon-I: Adsorption equilibria. Chemical Engineering Science 36: 721–730.

13.

Gonzalez

MFE

Cid

AAP

(2005) Adsorbentes de origen natural para la eliminación de colorantes textiles. In: Reinoso FR (ed.), Descontaminación Ambiental Mediante Adsorbentes, Espanha: CYTED, capítulo 5, pp. 99–122.

14.

González

Hernández-Quiroz

García-González

(2014) The use of experimental design and response surface methodologies for the synthesis of chemically activated carbons produced from bamboo. Fuel Processing Technology 127: 133–139.

15.

Gruber

LDA

Damasceno

Caramão

et al . (2012) Ácidos naftênicos no petróleo. Química Nova 35: 1423–1433.

16.

Habila

AlOthman

Al-Tamrah

et al . (2015) Activated carbon from waste as an efficient adsorbent for malathion for detection and removal purposes. Journal of Industrial and Engineering Chemistry 32: 336–344.

17.

Hernández-Terrones

Morais

SAL

Londe

et al . (2007) Ação alelopática de extratos de embaúba (Cecropia pachystachya) no crescimento de capim-colonião (Panicum maximum). Planta Daninha 25: 763–769.

18.

Mckay

(1999) Pseudo-second order model for sorption processes. Process Biochemistry 34: 451–465.

19.

Huang

Sun

Wang

et al . (2011) Comparative study on characterization of activated carbons prepared by microwave and conventional heating methods and application in removal of oxytetracycline (OTC). Chemical Engineering Journal 171: 1446–1453.

20.

Huang

Zhao

(2015) Thermal and structure analysis on reaction mechanisms during the preparation of activated carbon fibers by KOH activation from liquefied wood-based fibers. Industrial Crops and Products 69: 447–455.

21.

Kumar

Jena

(2015) High surface area microporous activated carbons prepared from Fox nut (Euryale ferox) shell by zinc chloride activation. Applied Surface Science 356: 753–761.

22.

Kumar

Porkodi

(2007) Mass transfer, kinetics and equilibrium studies for the biosorption of methylene blue using Paspalum notatum. Journal of Hazardous Materials 146: 214–226.

23.

Lagergren

(1898) Zur theorie der sogenannten adsorption gelöster stoffe. Kungliga Svenska Vetenskapsakademiens. Zeitschrift Für Chemie Und Industrie Der Kolloide 24: 1–39.

24.

Langmuir

(1916) The constitution and fundamental properties of solids and liquids. Journal of the American Chemical Society 38: 2221–2295.

25.

Largitte

Brudey

Tant

et al . (2016) Comparison of the adsorption of lead by activated carbons from three lignocellulosic precursors. Microporous and Mesoporous Materials 219: 265–275.

26.

Martínez

Torres

Guzmán

et al . (2006) Preparation and characteristics of activated carbon from olive stones and walnut shells. Industrial Crops and Products 23: 23–28.

27.

Matos

Nahas

Rojas

et al . (2011) Synthesis and characterization of activated carbon from sawdust of Algarroba wood. 1. Physical activation and pyrolysis. Journal of Hazardous Materials 196: 360–369.

28.

Montgomery

(2012) Introdução Ao Controle Estatístico Da Qualidade. Rio de Janeiro: Livros Técnicos e Científicos Editora.

29.

Muniandy

Adam

Mohamed

et al . (2014) The synthesis and characterization of high purity mixed microporous/mesoporous activated carbon from rice husk using chemical activation with NaOH and KOH. Microporous and Mesoporous Materials 197: 316–323.

30.

Nascimento

Duarte

MMMB

Sales

DCS

et al . (2017) Kinetic and equilibrium adsorption study for removal of naphthenic acids presents in model mixture of aviation kerosene. Chemical Engineering Communications 204: 105–110.

31.

Nascimento

Duarte

MMMB

Schuler

ARP

et al . (2014) Synthesis, characterization, and application of the mesoporous molecular sieve SR-MCM-41 in the removal of naphthenic acids from a model mixture of aviation kerosene by adsorption. Brazilian Journal of Petroleum and Gas 8: 1–13.

32.

Ould-Idriss

Stitou

Cuerda-Correa

et al . (2011) Preparation of activated carbons from olive-tree wood revisited. II. Physical activation with air. Fuel Processing Technology 92: 266–270.

33.

Pezoti

Cazetta

Bedin

et al . (2016) NaOH-activated carbon of high surface area produced from guava seeds as a high-efficiency adsorbent for amoxicillin removal: Kinetic, isotherm and thermodynamic studies. Chemical Engineering Journal 288: 778–788.

34.

Quesnel

Bhaskar

Gieg

et al . (2011) Naphthenic acid biodegradation by the unicellular alga Dunaliella tertiolecta. Chemosphere 84: 504–511.

35.

Rodrigues

Silva

Alvarez-Mendes

et al . (2011) Phenol removal from aqueous solution by activated carbon produced from avocado kernel seeds. Chemical Engineering Journal 174: 49–57.

36.

Royer

Cardoso

Lima

et al . (2009) Applications of Brazilian pine-fruit shell in natural and carbonized forms as adsorbents to removal of methylene blue from aqueous solutions – Kinetic and equilibrium study. Journal of Hazardous Materials 164: 1213–1222.

37.

Shi

Shen

Wang

(2008) Removal of naphthenic acids from Beijing crude oil by forming ionic liquids. Energy & Fuels 22: 4177–4181.

38.

Silva

(2007) Remoção de ácidos naftênicos por adsorção utilizando peneiras moleculares do tipo MCM-41 . Dissertation, Universidade Federal de Pernambuco, BR.

39.

Sudaryanto

Hartono

Irawaty

et al . (2006) High surface area activated carbon prepared from cassava peel by chemical activation. Bioresource Technology 97: 734–739.

40.

Sun

Yue

Gao

et al . (2012) Preparation of activated carbon derived from cotton linter fibers by fused NaOH activation and its application for oxytetracycline (OTC) adsorption. Journal of Colloid and Interface Science 368: 521–527.

41.

Thommes

Kaneko

Neimark

et al . (2015) Physisorption of gases, with special reference to the evaluation of surface area and pore size distribution (IUPAC Technical Report). Pure and Applied Chemistry 87: 1051–1069.

42.

Tonucci

Gurgel

LVA

Aquino

(2015) Activated carbons from agricultural byproducts (pine tree and coconut shell), coal, and carbon nanotubes as adsorbents for removal of sulfamethoxazole from spiked aqueous solutions: Kinetic and thermodynamic studies. Industrial Crops and Products 74: 111–121.

43.

Wang

Sun

et al . (2014) Removal of naphthenic acids from crude oils by fixed-bed catalytic esterification. Fuel 116: 723–728.

44.

Weber

Morris

(1963) Kinetics of adsorption on carbon from solution. Journal of the Sanitary Engineering Division 89: 31–60.

45.

Zhang

Luo

Liu

et al . (2015) A low cost and highly efficient adsorbent (activated carbon) prepared from waste potato residue. Journal of the Taiwan Institute of Chemical Engineers 49: 206–211.

Removal of naphthenic acids using activated charcoal: Kinetic and equilibrium studies

Abstract

Keywords

Introduction

Materials and methods

Preparation and quantification of the AVK model mixture

Charcoal preparation and activation

Activated charcoal characterisation

Factorial design

Kinetic adsorption study

Adsorption equilibrium study

Results and discussion

Characterisation of the carbonaceous materials

Factorial design

Kinetic study of adsorption

Equilibrium study of adsorption

Conclusion

Footnotes

Acknowledgements

Declaration of Conflicting Interests

Funding

Appendix

References