Sorghum-Waste-Derived High-Surface Area KOH-Activated Porous Carbon for Highly Efficient Methylene Blue and Pb(II) Removal.

With the development of the environment and human society, the removal of metal ions and dyes in wastewater treatment remains an urgent problem to solve. In this work, two biomass carbon adsorbents were synthesized by a KOH activation and carbonization route using sorghum stem and root as carbon precursors. In comparison with the samples without KOH activation, the pore structure of the KOH-activated carbon has been dramatically improved. The findings show that the specific surface areas of the adsorbents by sorghum stem (S1) and sorghum root (R1) were 948.6 and 168.1 m2 g-1, respectively. Meanwhile, the abundant OH- and COO- groups on the surface of these adsorbents endow them with negative polarity, thereby exhibiting excellent adsorption performance for removing methylene blue (MB) and Pb(II) from wastewater. The adsorption amount and removal rate of S1 were 98.1 mg g-1 and 98.08%, respectively, for MB, whereas those of R1 were 197.6 mg g-1 and 98.82% for the Pb(II) ion, respectively. Our findings offer an invaluable insight into designing and synthesizing a highly efficient sustainable adsorbent to remove MB and Pb(II) based on biomass agricultural waste.

Introduction

With the rapid development of human society, a large amount of industrial wastewater is discharged and treated in the world every year, which releases a variety of pollutants.[1] Among these pollutants, dyes and heavy metals occupy a prominent position.[2−4] It is worth noting that dyes and heavy metal ions cause serious damage to human health because of their carcinogenic, mutagenic, and bioaccumulative effects.[5−7] Therefore, it is urgent to solve the problems of dye and heavy metal pollution in wastewater.[8,9]

In recent years, the physical adsorption method has attracted much attention because of its simple operation, high efficiency, and robust flexibility.[1,3,10] The various adsorbents reported in previous studies are activated carbon,[11] ion-exchange resins,[12] metal oxides,[13] zeolites,[14] graphene oxides,[15] and bentonite.[16] Among them, activated carbon materials account for a large proportion because of their eco-friendliness and easy handing.[17,18] What is more important, the precursors of carbon materials play a vital role in affecting their morphology, pore structure, and adsorption properties.[10] Therefore, seeking for suitable precursors is an inevitable trend in forming high-efficiency carbon-based adsorbents. The agricultural waste used as precursors seems a sustainable strategy to synthesize high-performance carbon-based adsorbents for removing the dyes and heavy metals because of the extensive source of raw materials, low cost, and no secondary pollution,[19] including coconut shells,[20] jute,[21] olive,[22] and so on. For example, Chen et al.[9] prepared a three-dimensional wood film that can effectively treat dye wastewater, and the degradation efficiency of methylene blue (MB) was as high as 99.8%. Meanwhile, Singh et al.[23] produced activated carbon with the adsorption degradation rate of 97.95% to remove heavy metal Pb(II) using tamarind wood as a precursor. In addition, Acharya et al.[24] used tamarind wood with zinc chloride activation to prepare activated carbon, and the maximum adsorption degradation rate of chromium(VI) reached 99%. Therefore, it is necessary to explore green agricultural waste with preparation in a simple way for obtaining high-efficiency adsorbents.

As the world’s fifth-largest cereal crop, sorghum biomass produces large amounts of agricultural wastes every year.[25] In addition to some of the animal feed, a large quantity of sorghum wastes has been used as cellulosic biofuels, which causes environmental pollution and violates current ecological protection and sustainable development concepts.[25,26] Therefore, the utilization of sorghum wastes has become an inevitable trend. With this in mind, for the first time, the sorghum stem and root are used as raw materials to synthesize carbon adsorbents by KOH activation, which is named S1 and R1. After activated by KOH, the prepared carbon adsorbents acquire vast specific surface areas, and the opulent OH– and COO– groups on the surface of the adsorbent endow the adsorbent with negative polarity, thereby exhibiting excellent adsorption performance for MB and heavy metal Pb(II).

Results and Discussion

In Scheme , S1 and R1 are synthesized by a two-step process of KOH activation and carbonization. First, the sorghum stem and root were completely immersed in 1 mol L–1 KOH solution and then activated at 60 and 120 °C, respectively, for 12 h, which is beneficial to adsorb as much KOH as possible and promote the decomposition of organic matter in the stems and roots. After drying, the activated samples were placed in a tube furnace and carbonized at 800 °C for 2 h under a nitrogen atmosphere, thereby promoting the formation of porous carbon adsorbents with a rich specific surface area (see the Experimental Section for details).

Scheme 1

Schematic Illustration of the Fabrication Route of S1 and R1

Scanning electron microscopy (SEM) was used to observe the surface morphology of S1 (Figure a,b) and R1 (Figure d,e). Both S1 and R1 reveal a rich circular pore structure in the parallel direction, as shown in Figure a,d. More importantly, as demonstrated by the red dashed lines in Figure b,e, S1 and R1 have long straight pores of parallel shape on the vertical side, which makes contaminants enter the inner surface of carbon materials easily.[9] Besides, as shown by the yellow dotted line in Figure b, the inner surface of the vertical direction of S1 still contains a large number of mesopores, which is beneficial for the pollutant’s adsorption. In contrast, as shown in Figure S1, the surfaces of S0 and R0 that were not activated with KOH had no obvious pore structure, indicating that KOH is an activator, which can generate a large number of long and straight pore channels during the preparation process. The increase in the pore structure of the adsorbents can provide a large number of active centers for the adsorption of organic pollutants and greatly improve their adsorption performance.[23,24] Furthermore, it was observed by transmission electron microscopy (TEM) that S1 (Figure c) and R1 (Figure f) exhibited a wrinkled graphene-like structure of dispersed pores, which further confirmed their SEM results. On the contrary, the TEM images of S0 and R0 (Figure S2) show that the adsorbents have smooth surfaces, which is consistent with the corresponding SEM results.

Figure 1

SEM images of S1 (a,b) and R1 (d,e). TEM images of S1 (c) and R1 (f).

The N2 adsorption–desorption isotherms and pore size distributions of all samples were tested and analyzed based on Brunauer–Emmett–Teller (BET), as shown in Figures and S3. The specific surface areas of S1 and R1 were 948.6 and 168.1 m2 g–1, which were in sharp contrast with those of S0 and R0, only 0.4 and 0.5 m2 g–1, respectively. After activation, the specific surface area of the carbon adsorbents becomes more significant because of the increase in the pore structure of the adsorbents during the KOH-activated pore-etching process.[29] In addition, the pore distribution curves of Barrett–Joyner–Halenda are shown in Figure c,d. After KOH activation, the distribution of mesopore-dominated pore size appears in S1 and R1, which is more conducive to the adsorption of pollutants. The average pore size of S1 and R1 is 3.12 and 4.48 nm, respectively, and their total pore volume is 0.55 and 0.13 cm3 g–1, respectively, as shown in Table S1. Combined with SEM and TEM results, the KOH-activated method successfully formed a complex pore network structure for S1 and R1, which improves the adsorption rate of the adsorbents effectively.

Figure 2

N2 adsorption–desorption isotherms of (a) S1 and (b) R1. The corresponding pore size distributions of (c) S1 and (d) R1.

To further explore their graphitization structures, Raman spectroscopy was performed, as shown in Figure a,c. It was apparent that the D and G peaks of the samples were located at 1335 and 1580 cm–1, respectively. The intensity ratio (ID/IG) of D and G bands was an indicator of the defect density of carbon materials. The larger the value is, the lower the graphitization degree of the material is and the higher their structural defect is.[30] In Figure a, the values of ID/IG of S0 and S1 were 0.94 and 1.04, respectively, indicating an increase in defects of the adsorbent S1 after activation because of planar damage caused by sp2 carbon activation of the adsorbent. The phenomenon was similar to the value trend of ID/IG of R0 and R1, which was 0.98 and 1.03, as shown in Figure c, respectively. The X-ray diffraction (XRD) patterns (Figure b,d) showed the crystal structure of the activated and inactivated carbon layer. The diffraction peaks at about 24.0 and 43.0° corresponded to the typical graphite (002) and (101) planes, respectively.[31,32] As can be seen from the results of Figure b,d, the crystallinity of S0, S1, R0, and R1 is similar. After KOH activation, the (002) characteristic diffraction peaks of S1 and R1 become wider and weaker, indicating a low degree of graphitization.[33] This result is consistent with the conclusion of Raman measurement.

Figure 3

Raman spectra (a,c) and XRD patterns (b,d) of S0, S1, R0, and R1.

Typically, the chemical composition and functional groups of the adsorbent were determined by Fourier transform infrared (FT-IR) spectroscopy and X-ray photoelectron spectroscopy (XPS). In Figure a,b, it can be observed that after KOH activation, the functional groups on the surface of the adsorbents become more abundant. All samples had significant absorption peaks around 3430 and 1455 cm–1, corresponding to tensile vibrations of O–H and C=C, respectively. In addition, the absorption peak of the adsorbent after KOH activation was significantly broader than that of the inactivated adsorbent, which was a result of an increasing amount of the functional groups. In addition, the peaks at 780 and 870 cm–1 were due to the bending vibration of the aromatic C–H, as shown in Figure a, and the absorption peak at 1100 cm–1 was attributed to the stretching vibration of C–O–C, as shown in Figure b. It is particularly noteworthy that the C–O–C absorption peak of R1 disappears with the destruction of the ether bond and shows a sharp C–O peak near 1010 cm–1, indicating that KOH activation destroys the fiber structure of R0 and forms a rich pore structure, which was proved by the increasing specific surface area of R1. Additionally, the adsorbents S1 and R1 have distinct absorption peaks around 1630 cm–1 because of C=O stretching vibration.[33,34] The surface functional groups of the adsorbent were further analyzed employing XPS, as shown in Figure c,d. Two distinct sharp peaks were observed in the XPS survey spectra of all samples, corresponding to C 1s and O 1s, respectively, indicating that the adsorbents are mainly composed of C and O elements. In addition, in Figure e,f, the high-resolution C 1s spectra of the adsorbent show three peaks at 284.9, 286.1, and 289.4 eV, which are attributed to the C=C, C–O, and COO– groups, respectively. At the same time, all adsorbents showed a peak at 532.0 eV in the high-resolution O 1s spectrum, which was attributed to the C–O group (Figure S4).[35,36]

Figure 4

FT-IR spectra of (a) S0 and S1 and (b) R0 and R1. XPS survey spectra of (c) S0 and S1 and (d) R0 and R1. High-resolution C 1s spectrum of (e) S0 and S1 and (f) R0 and R1.

For adsorbents, contact time is an essential factor affecting the adsorption efficiency of the adsorbent. The shorter the time required for the adsorbent to reach its maximum Qe is, the more apparent the advantage of the adsorbent is. The kinetic studies of Pb(II) or MB adsorption provide some important information about the mechanism of the adsorption process, as shown in Figure a. The adsorption rate of Pb(II) or MB is speedy at the beginning and then increases slowly until the plateau of adsorption equilibrium is reached because a large amount of adsorbent active sites in the initial stage causes a rapid increase in the adsorbent adsorption amount. However, with the further increase in the contact time, the aggregation of Pb(II) or MB molecules hinders the movement and flow of the adsorbent molecules so that the adsorption amount of the adsorbent gradually becomes stable.[37] The maximum Pb(II) adsorption capacity of R1 is 197.6 mg g–1, whereas the maximum MB adsorption capacities of S1 and R1 are 98.1 and 92.75 mg g–1, respectively, which are much higher than other previously reported biomass adsorbents, as listed in Tables S2 and S3. In addition, Figure b,c shows the UV–visible spectrum of the MB solution before and after being adsorbed by different adsorbents. After the treatment of S1 and R1, the characteristic absorption peak of MB (664 nm) completely disappeared. In contrast, after the MB solution was treated with S0 and R0, the characteristic absorption peak of MB was still obvious. This result confirms the key role of KOH activation for the successful preparation of highly efficient adsorbents S1 and R1. Meanwhile, in Figure d, after the adsorption of S0 and R0, the MB solution hardly changed. Conversely, the blue MB solution becomes almost colorless after being adsorbed by S1 and R1, which further visually proves the excellent adsorption performance of adsorbents S1 and R1.

Figure 5

(a) Effect of contact time on the adsorption of Pb(II) or MB by different adsorbents. The UV–vis spectra of the MB solution before (red) and after adsorption (blue) by (b) S1 and S0 and (c) R1 and R0. (d) Photographs of MB solution before and after being absorbed by S0 and S1, and R0 and R1.

To further understand the adsorption mechanism of the adsorbent, we studied the adsorption kinetics of S1 adsorption on MB and adsorption of R1 on MB and Pb(II). For the present investigation, the adsorption process of the adsorbent was fitted using a pseudo-first-order, pseudo-second-order, Elovich, and intraparticle diffusion model, as expressed in the following equations[38,39]where Q (mg g–1) is the MB adsorption amount at time t (min), Qe (mg g–1) is the equilibrium adsorption amount, K1 and K2 are pseudo-first-order and pseudo-second-order rate constants, and α (mg g–1 min) and β (g mg–1) correspond to the adsorption rate and desorption rate, respectively. Also, k (mg g–1 min0.5) and C (mg g–1) represent the diffusion rate within the particle and the constant associated with the thickness of the boundary layer, respectively.[39,40]

On the basis of the classic adsorption model, the dynamic behaviors of R1 and S1 along with the time were calculated. The adsorption phenomena were analyzed by the pseudo-first-order, quasi-second-order, Elovich, and intraparticle diffusion model, as shown in Figure a–d, respectively, and all the fitting parameters are summarized in Table S1. For the adsorbent S1, the corresponding R2 value (>0.99) of the pseudo-second-order kinetic model was larger than that of the pseudo-first-order kinetic model (<0.92), and the fitting line of the pseudo-second-order dynamic model of Qe is more consistent with the experimental results, indicating that the pseudo-second-order kinetic model is more suitable for describing S1.[38] Meanwhile, the results of adsorption of MB and Pb(II) by adsorbent R1 show that the R2 of the pseudo-first-order and pseudo-second-order kinetic models is higher than 0.96, which signifies that the adsorption of MB and Pb(II) on R1 is consistent with the pseudo-first-order and pseudo-second-order dynamics model. The Elovich model (Figure c) can be used to determine chemisorption. It can be concluded that S1 and R1 are chemisorbed by the corresponding R2 values, as shown in Table S4 (higher than 0.96), and the fitted line and experimental data. At the same time, the large α value and small β value indicate the feasibility of adsorbing MB and Pb(II) and the irreversibility of desorption, respectively.[41] In Figure d, the fit of the experimental data and the intraparticle diffusion kinetics model of the relevant parameters can prove that the main adsorbent systems for MB and Pb(II) include external surface adsorption (a sharper linear region cross section) and intraparticle diffusion.[42]

Figure 6

Plots of (a) pseudo-first-order kinetics, (b) pseudo-second-order kinetics, (c) Elovich kinetics, and (d) intraparticle diffusion model for the adsorption of MB and Pb(II) on R1 and MB on S1.

To further explore the adsorption mechanism of the adsorbent, a zeta potentiometer was used to measure the electrostatic attraction of the adsorbent and MB dispersed in solution. As shown in Figure a, the zeta potentials of S1 and R1 were −24.5 and −23.0 mV, respectively, which was due to the increase in the number of OH– groups and COO– groups on the surface of the adsorbent after KOH activation (Figure ). In contrast, the zeta potential of MB is 11.3 mV, indicating a positive polarity. Therefore, the electrostatic attraction between the adsorbent and MB is beneficial to the effective adsorption of MB. At the same time, the OH–/COO– and the pollutant Pb(II) on the surface of the adsorbent can form a surface complex, which can improve the adsorption of Pb(II) to a certain level. In addition, it should be noted that the surface areas of S1 and R1 has been increased sharply compared to S0 and R0 (Figure ), and it is well-known that the larger surface areas are good for providing more adsorption sites and can adsorb more pollutants on the surface of the adsorbent. Therefore, the combined effect of electrostatic attraction and physical adsorption promotes excellent adsorption performance of S1 and R1 (Figure S5). Moreover, the recyclability of the adsorbents S1 and R1 was investigated by performing an adsorption–desorption cycle of the contaminants. After five adsorption–desorption cycles, the adsorption efficiency of the pollutants remained stable, and the adsorption rate was still as high as 96.60%, as shown in Figure b, indicating the stability of the adsorption activity of the prepared adsorbents S1 and R1. Therefore, excellent recyclability makes S1 and R1 potentially suitable for efficient, low-cost, and extensive water pollution treatment.

Figure 7

(a) Zeta potential of S1, R1, and MB. (b) Recyclability of adsorbents S1 and R1 for the adsorption of MB or Pb(II).

Conclusions

In summary, biomass carbon adsorbents were prepared through the KOH activation and carbonization route, using the inexpensive and abundant sorghum waste in nature as carbon precursors. The obtained adsorbents S1 and R1 have larger specific surface areas, which are 948.6 and 168.1 m2 g–1, respectively. Because of the activation of KOH, the abundance of OH– and COO– groups on the surface of the adsorbent imparts negative polarity to the adsorbent, thereby exhibiting high-efficiency adsorption performance for MB and Pb(II) in wastewater. The adsorption capacity and removal rate of MB and Pb(II) in wastewater by S1 and R1 are 98.1 mg g–1 and 98.08% and 197.6 mg g–1 and 98.82%, respectively. Our finding provides green, low-cost, and feasible strategies for converting agricultural and forestry waste into high-value biomass carbon adsorbents, promoting the sustainable development of the economy and human society.

Experimental Section

Materials

Sorghum was from the rural areas of Shanxi province. All chemicals were provided by Sinopharm Chemical Co., Ltd (Shanghai, China) and used directly without further purification.

Synthesis of the Adsorbent

Typically, about 5 g of the sorghum stem and root were, respectively, immersed in 500 mL of a 1 mol L–1 KOH solution and then activated at 60 and 120 °C for 12 h, respectively. The activated samples were taken out and dried in an oven. The dried samples were placed in a tube furnace and carbonized at 800 °C for 2 h under a nitrogen atmosphere. The carbonized samples were washed to neutrality with dilute hydrochloric acid solution, deionized water, and absolute ethanol and then dried. Finally, the dried samples were ground into powders with a mortar and passed through 100 and 300 mesh sieves to obtain 100–300 mesh adsorbent particles, which were named S1 and R1, respectively. For comparison, we directly carbonized the corresponding sorghum stem and root at 800 °C for 2 h in a nitrogen atmosphere to obtain the corresponding control products, which were named S0 and R0, respectively.

Adsorption of Dyes and Heavy Metal Ions

About 0.02 g of adsorbents S1 and R1 was added to 100 mL of MB solution (20 mg L–1), respectively. Similarly, 0.02 g of R1 was added to 100 mL of Pb(II) solution (40 mg L–1) and then continuously mixed by magnetic stirring to achieve better mass transfer and high interface contact area. After adsorption at different time intervals, the residual concentrations of Pb(II) and MB in each sample were determined by atomic absorption spectrometry and UV–visible spectrometry, respectively. The formula for the removal rate of Pb(II) or MB is[27]where C0 and C (mg L–1) represent the initial and t-time solution concentrations, respectively.

The calculation of the amount of dye and heavy metal ion adsorption is performed using the following formula[28]where C0 and Ce (mg g–1) respectively indicate the concentration at the starting time of the solution and the concentration at the equilibrium (mg L–1); V and W represent the volume of the solution and the weight of the adsorbent, respectively.

Characterization

The surface morphology of all samples was examined using a scanning electron microscope (VP-EVO, MA-10, Carl-Zeiss, UK) operated at 10 kV and a transmission electron microscope (JEOL JEM-2100F) operated at 200 kV. X-ray powder diffraction (XRD, Mercury CCD, Rigaku) was performed using Cu Kα radiation to characterize the crystal structure of the adsorbents by a scan rate of 5°/min in the 2θ range of 10–90°. The identification of surface composition of all samples was accounted for by FT-IR spectroscopy using a Bio-Rad FTIR spectrometer FTS165. XPS was carried out using an AMICUS electron spectrometer on SHIMADZU using 300 W Al Kα radiations to analyze the chemical composition of the materials. Raman spectra were obtained with a Thermo Fisher Scientific in plus laser DXRxi Raman imaging with 633 nm. The BET surface area was determined by the nitrogen sorption of samples acquired on a Micromeritics Ga 30093-2901 U.S.A. instrument and the specific surface area was calculated by the BET method. UV–vis spectra were recorded using a Hitachi 3100 spectrophotometer. An atomic absorption spectrometer (PinAAcle 500) was used to record the atomic absorption spectrum.