Fabrication of magnetic activated carbons from corn cobs using the pickle liquor from the surface treatment of iron and steel

WANG Fang, DANG Yan-qiu, TIAN Xun, Steven Harrington, MA Lei, MA Yan-qing,

(1. Key Laboratory of Materials Chemistry in Binzhou City, Department of Chemical Engineering and Safty, Binzhou University, Binzhou256600, China; 2. School of Precision Instrument & Opto-Electronics Engineering, Tianjin University, Tianjin300072, China; 3. Tianjin International Center for Nanoparticles and Nanosystems, Tianjin University, Tianjin300072, China)

Abstract: Corn cobs were impregnated with the pickle liquor from the surface treatment of iron and steel at ambient temperature for 24 h, dried at 393 K for 12 h and carbonized at 573, 773 and 973 K for 1 h to obtain magnetic activated carbons (MACs). The adsorption of methyl orange (MO) onto the MACs was investigated under different conditions. Results show that the surface area of the MACs increases with carbonization temperature from 423.5 to 784.76 m2 g-1. The MACs consist of hematite (Fe2O3) and magnetite (Fe3O4). The MAC obtained at 973 K has the highest MO monolayer adsorption capacity of 555.56 mg g-1 at 298 K in a water sample containing MO dye. The adsorption is endothermic and obeys pseudo-first-order kinetics. The MACs can be easily separated from the dye water by an external magnet.

Key words: Corn cob; Carbon materials; Hydrochloric acid pickling water; Methyl orange

1 Introduction

Most organic dyes from textile, leather, paper, and plastics industries have complex aromatic molecular structures, which are highly toxic to some organism s and hence disturb the ecosystem. Various techniques such as coagulation, adsorption, chemical oxidation and froth floatation have been used for the removal of organic compounds from wastewaters[1]. Among these chemical and physical method, the adsorption technique has been found to be superior for the treatment of these dye-bearing wastewaters[2]. Biomass activated carbon is one of the most commonly used adsorbents for removing hazardous compounds from industrial waste gases or wastewaters owing to its high specific surface area and high porosity as well as renewable, cheap, and environmental friendly. However, conventional biomass activated carbon is difficult to be separated and recovered except by means of high speed centrifugation or filtration. Magnetic activated carbon (MAC) adsorbents can easily be separated from solutions using a magnetic separator even if the solution contains a significant concentration of other solids[3,4]. However, most previous studies are focused on the preparation of MAC via a two-step method, which has several disadvantages, such as complexity, high cost and the loss of adsorption capacity during recycle[5-7]. To the best of our knowledge, only a few attentions have been paid to the one-step method[8].

In steel industries, the pickling process of iron and steel objects to remove oxides formed by atmospheric corrosion and hot processing generates a considerable quantity of hydrochloric acid pickling water containing the dissolved metal salts of iron, chromium and nickel, as well as residual free acid[9,10]. If one could combine the advantages of cheap agricultural residues and magnetic activators, such as FeCl3and HCl in the hydrochloric acid pickling water, to fabricate MAC with high surface area, appropriate pore size, and magnetic separability, a promising novel adsorbent may be accessible.

The work presented here successfully prepared the MAC with higher adsorption capacity and excellent separation properties from corn cobs using hydrochloric acid pickling water as an activator. Then, the obtained activated carbon was utilized for methyl orange (MO) removal under different experimental conditions in order to evaluate the equilibrium isotherms, kinetics and thermodynamics.

2 Experimental

2.1 Preparation of the activated carbons

The corn cob used for preparation of the activated carbons was obtained locally. The raw material was first washed with distilled water, dried, cut and sieved to the desired size. To prepare the samples, thirty grams of corn cob sample was impregnated with 300 mL hydrochloric acid pickling water for 24 h, and then dried in an oven at 393 K for 12 h. Subsequently, the impregnated sample was heated at a ramping rate of 10 K min-1under a nitrogen flow and carbonized at different activation temperatures (573, 773 and 973 K) for 1 h. Afterwards, the activated carbons in the furnace were cooled to room temperature with a continuous nitrogen flush. The cooled solids were washed by the deionized water for several times until the pH value of the filtrate became neutral and filtered . Finally, the samples were dried for testing and were denoted as MAC (573), MAC (773) and MAC (973).The yields of the activated carbons were calculated based on the following equation:

(1)

Wherewfis the weight of final activated carbon products (g) andw0is the weight of dried corn cob (g).

2.2 Characterization of the activated carbons

The functional groups on the surface of MACs were identified by using FT-IR. The MACs were diluted with KBr, compressed into a wafer, and the FT-IR spectra were recorded by an AVATAR 360 (Thermo Nicolet Co., USA) FT-IR spectrophotometer. The X-ray diffraction patterns (XRD) were obtained via a Philips Xpert MPD instrument using Cu Kradiation in a scanning angle range of 10-90°at a scanning rate of 0.5°min-1at 40 mA and 50 kV. The specific surface areas and mesoporous structures of the adsorbents were measured by the Brunauer-Emmett-Teller (BET) method on a Micromeritics ASAP-2020 with N2as the absorbent at the 77 K. Scanning electron microscopy (SEM) was carried out using a JSM-6490LV to study the activated carbon surface textures and the development of porosity.

2.3 Adsorption studies

Adsorption isotherms were performed in a set of Erlenmeyer flasks (500 mL) where 200 mL of MO solutions with the initial concentrations of 100-500 mg L-1were placed in each flask. Equal masses of 0.2 g of the obtained MACs were added to each flask and kept in an isothermal shaker at 298 K for 6 h to reach equilibrium. The original pH values of the solutions were around 6.5. Similar procedures were followed for another two sets of Erlenmeyer flask containing the same initial dye concentrations and the same amount of activated carbons, but these were kept under 308 K and 318 K. All samples were filtered prior to analysis in order to minimize interference of the carbon fines with the analysis. Each experiment was duplicated under identical conditions. The concentrations of MO in the supernatant solutions after adsorption were determined using a T6 UV-vis spectrophotometer (Beijing, China) at 465 nm. The amount of adsorption at equilibrium,qe(mg g-1) was defined by the following formula,

(2)

WhereC0andCe(mg L-1) are the liquid-phase concentrations of MO at the initial time and at equilibrium, respectively.Vis the volume of the solution (l) andWis the mass of the dried MAC used (g).

Two famous isotherm equations, namely the Langmuir and Freundlich equations, were applied to fit the experimental isotherm data of MO adsorption on MACs. The linear forms of Langmuir isotherm and Freundlich equation are given as:

(3)

(4)

For the Langmuir isotherm, a separation factor,RL, is the essential characteristics of this isotherm, which can be obtained from the following equation:

(5)

WhereCeis the equilibrium concentration of the adsorbate (mg L-1),qeis the amount of adsorbate adsorbed per unit mass of adsorbent (mg g-1), andQ0andbare Langmuir constants related to adsorption capacity and rate of adsorption, respectively.KFandnare Freundlich constants withngiving an indication of how favorable the adsorption process is.KF((mg g-1) (L mg-1)) is the adsorption capacity of the adsorbent, which can be defined as the adsorption or distribution coefficient and represents the quantity of dye adsorbed onto activated carbon for a unit equilibrium concentration, withC0being the highest initial solute concentration.

Three models, primarily pseudo-first-order, pseudo-second-order and intraparticle diffusion models, were used to analyze the kinetic data. These models can be expressed as:

ln(qe-qt)=lnqe-k1t

(6)

(7)

qt=kpt1/2+C

(8)

Whereqeandqt(mg g-1) are the uptake of MO at equilibrium and at timet(min), respectively,k1(min-1) is the adsorption rate constant,k2(g mg-1min-1) is the rate constant of second-order equation,k3(mg g-1min-1/2) is the intraparticle diffusion rate constant, andC(mg g-1) is a constant that gives information as to the thickness of the boundary layer.

Thermodynamic behavior of MO adsorption onto MACs was evaluated by the thermodynamic parameters including the change in free energy (ΔG°), enthalpy (ΔH°), and entropy (ΔS°). These parameters were calculated using the following equations:

(9)

(10)

ΔG°=-RTlnKd

(11)

WhereKdis the distribution coefficient,qe(mg L-1) is the amount adsorbed on a solid at equilibrium andCe(mg L-1) is the equilibrium concentration.R(8.314 J mol-1K-1) is the universal gas constant, andT(K) is the absolute solution temperature.

3 Results and discussion

The yield, surface area and pore size of the obtained MACs are summarized in Table 1. The yields of MACs decrease from 73.7% to 56.6% with increasing the carbonization temperature from 573 to 973 K. The yields are higher than those reported by a previous study using ferric chloride[11]as an activating agent. This may be due to the cooperative effect of ferric chloride and hydrochloric acid in hydrochloric acid pickling water. In addition, as the activation temperature increases, the specific surface areas increase from 423.54 m2g-1at 573 K to 784.76 m2g-1at 973 K. This is expected, as increasing temperature will lead to releasing more volatiles, thereby resulting in the decreased yield and increased specific surface area. The pore size is another important characteristic and the results of Table 1 show that MAC-973-1 has a combination of micro-, meso- and macropores, with an average pore size of 4 nm.

Table 1 Yield, surface area, and pore size of MACs at different activation temperatures.

Fig.1 FTIR spectra of activated carbons before adsorption: (a) MAC (573), (b) MAC(773) and (c) MAC (973).

X-ray diffraction patterns of the MACs prepared at different activation temperatures are given in Fig.2. The X-ray diffraction patterns for MACs display a number of sharp peaks which are compatible

with the presence ofα-FeO(OH) (peaks at 2θ= 20.7° and 33.7°), Fe2O3(hematite) (peaks at 2θ= 32.5°, 37.5°, 38.8° and 42.3°) and of Fe3O4(magnetite)(peaks at 2θ= 30.7°, 35.5° and 43.2°)[9,10]. According to previous reports, hematite and magnetite are magnetic with magnetizations of 100 and 60 J·T-1·kg-1[15]. Therefore, the obtained MACs are magnetic, and can be separated easily from dye wastewater.

Fig.2 X-ray powder diffraction patterns for (a) MAC (573), (b) MAC (773) and (c) MAC(973).

To observe the morphology of the prepared samples, the -SEM images of samples are shown in Fig. 3. Micrographs of Fig. 2(a, b) show a sheet structure for the MACs prepared at 573 K and 773 K. There is a honeycomb shape in which small canals are clearly found on the surface of the MAC (973). The well-developed pores have led to the large surface area and porous structure of the activated carbon, which are in accordance with BET results. These results show that FeCl3and HCl in hydrochloric acid pickling water are effective in creating well-developed pores on the surface of the MACs, hence leading to the activated carbons with large surface area and well-developed porous structure.

Fig.3 SEM images for activated carbons carbonized at different temperatures: (a) MAC (573), (b) MAC (773), (c) MAC (973).

The experimental equilibrium data for MO adsorption on the MAC (973), calculated from Eq.(3) and Eq.(4), are fitted with Langmuir and Freundlich isothermals. All the constants of the two isotherm models along with the linear correlation coefficientR2are summarized in Table 2. From Fig. 4 and Table 2, it can be found that the Langmuir isotherms give the better correlation coefficients (R2values) of 0.9810, 0.9623 and 0.9749 at 298, 308 and 318 K, respectively. This result reveals that during the adsorption process the uptake of MO occurs on a homogenous surface by monolayer adsorption without any interactions between adsorbed MO. In addition, the values ofRLat temperatures of 298, 308 and 318 K are 0.036, 0.031 and 0.028, respectively. Therefore, the adsorption of MO on the obtained MAC (973) under the conditions used in this study is favorable[16].

Fig.4 Langmuir adsorption isotherms of MO onto MAC(973) at 298, 308 and 318 K.

From Fig. 5 and Table 2, it also can be seen that the slopes of 1/nare between 0 and 1, indicating the high adsorption intensity[17]. The maximum mon olayer adsorption capacity of MO at 298, 308 and 318 K are 555.56, 588.24 and 595.24 mg g-1, respectively. The results are higher than those reported previously, this result indicates that the MAC (973) prepared via the hydrochloric acid pickling water activation has great potential as an adsorbent for MO removal[18,19].

Fig.5 Freundlich adsorption isotherms of MO onto MAC(973) at 298, 308 and 318 K.

Temperature(K)Langmuir isothermQ0 (mg g-1)b(L mg-1)RLR2Freundlich isothermKF(mg g-1)(L mg-1)1/nR2298555.560.01720.0360.981021.310.640.9318308588.240.01610.0310.962320.330.620.9331318595.240.01710.0280.974921.150.640.9460

To evaluate the kinetics mechanism of MO adsorption on the MAC (973), kinetics data are interpreted by the pseudo-first-order (Eq.6), pseudo-second-order (Eq.7) and intraparticle diffusion (Eq.8)

models. The estimated constants of the three kinetic equations along withR2values at different initial MO concentration are listed in Table 3.

Table 3 Comparison of the pseudo-first-order kinetic, pseudo-second-order kinetic and intraparticle diffusion models for different initial MO concentrations at 298 K.

HighR2values of 0.980 0, 0.973 2 and 0.973 5 are obtained for the linear plot of ln(qt-qe) versust(Fig.6) at the initial MO concentrations of 100, 200 and 300 mg/L, respectively for the pseudo-first order equation. It can be found that the pseudo-first order kinetic model better represents the adsorption kinetics and the experimental and calculated adsorption capacity values are in close agreement. This suggests that the overall rate of the adsorption process is controlled by chemisorption which involved valency forces through sharing or exchange of electrons between the MAC (973) and MO[20].In order to determine the thermodynamic parameters, the sorption studies were carried out at different temperatures (298, 308 and 318 K). The values of ΔG°, ΔH° and ΔS° were calculated from Eq.9-Eq.11 and listed in Table 4. It can be found that the negative free energy changes (ΔG°) at all studied temperatures suggest that the adsorption of MO onto MAC (973) adsorbent is feasible and spontaneous thermodynamically. The positive value of ΔH° reveals that the adsorption is an endothermic process, indicating that the adsorption capacity increases with increasing temperature. It also can be found from Table 2 that the maximum monolayer adsorption capacity of MO increases from 555.56 to 595.24 mg g-1with increasing temperature from 298 to 318 K. This further confirms the endothermic nature of the adsorption process. Finally, the positive values of ΔS°show the affinity of the obtained MAC (973) for MO and an increased randomness at the solid-solution interface during the adsorption process.

Table 4 Thermodynamic parameters for adsorption of MO onto prepared MAC(973).

Fig.6 Pseudo-first-order kinetics for adsorption of MO onto MAC(973) at 298 K.

4 Conclusions

A novel and high-performance magnetic activated carbon was synthesized successfully from corn cob using hydrochloric acid pickling water as an activating agent for the first time in this paper. The prepared MAC (973) has high potential for the removal of MO and can be easily attracted from the aqueous solution by an external magnet.

For a better understanding of the properties of the activated carbons, it is necessary to expand this work to include optimization of the parameters in activated carbon preparation and the adsorption of other polluting molecules.