NC State
BioResources
Chen, H., Sha, L., Zhang, Y., Liu, Y., Kong, F., and Zhao, X. (2019). "Facile control of corncobs-based carbons with eutectic salt ZnCl2/NaCl templated for the adsorption of organic aldehyde," BioRes. 14(4), 8835-8848.

Abstract

Corncob, a renewable biomass waste, has been successfully explored as a low-cost crude carbon source to prepare controlled morphology, and result in higher value-added carbons through a ZnCl2/NaCl treatment and direct pyrolysis process. The synthesis route is simple, green, and results in carbons with different morphology and porous structures by varying the ratio of the eutectic salt ZnCl2/NaCl. Moreover, the adsorption capacities of different carbon materials for formaldehyde and butyraldehyde were tested further. It was demonstrated that NaCl and ZnCl2 could be utilized as a mesopore template, and as a micropore activation agent for carbon materials, respectively. Although the mesopores can provide a fast diffusion channel, and the micropores can enhance adsorption ability, the specific surface area originating from the mesopores was proportional to the amount of butyraldehyde adsorption, and the specific surface area of the micropores was beneficial to the amount of formaldehyde adsorption taking place. The adsorption isotherms follow Langmuir and pseudo first-order kinetic modeling. Additionally, the whole process belongs to physical adsorption.


Download PDF

Full Article

Facile Control of Corncobs-based Carbons with Eutectic Salt ZnCl2/NaCl Templated for the Adsorption of Organic Aldehyde

Honglei Chen, Lei Sha, Yujie Zhang, Yu Liu, Fangong Kong, and Xin Zhao *

Corncob, a renewable biomass waste, has been successfully explored as a low-cost crude carbon source to prepare controlled morphology, and result in higher value-added carbons through a ZnCl2/NaCl treatment and direct pyrolysis process. The synthesis route is simple, green, and results in carbons with different morphology and porous structures by varying the ratio of the eutectic salt ZnCl2/NaCl. Moreover, the adsorption capacities of different carbon materials for formaldehyde and butyraldehyde were tested further. It was demonstrated that NaCl and ZnCl2 could be utilized as a mesopore template, and as a micropore activation agent for carbon materials, respectively. Although the mesopores can provide a fast diffusion channel, and the micropores can enhance adsorption ability, the specific surface area originating from the mesopores was proportional to the amount of butyraldehyde adsorption, and the specific surface area of the micropores was beneficial to the amount of formaldehyde adsorption taking place. The adsorption isotherms follow Langmuir and pseudo first-order kinetic modeling. Additionally, the whole process belongs to physical adsorption.

Keywords: Corncobs; Eutectic salt; Controlled morphology; Adsorption ability

State Key Laboratory of Biobased Material and Green Papermaking, Qilu University of Technology, Shandong Academy of Sciences, Jinan, Shandong, 250353, P.R. China;

* Corresponding author: zhaoxin_zixi@126.com (Xin Zhao)

INTRODUCTION

Low molecular weight organic aldehydes in the atmosphere have been major pollutants for many years due to their strong, toxic effects on the eyes, skins, and respiratory tracks of human beings. Several strategies have been developed to remove organic aldehydes, such as oxidation, biotechnology, directly blow-off, and adsorption approach (Falco et al. 2018; Kante et al. 2019). The adsorption process removes low-concentration harmful substances that are difficult to separate by common methods (Han et al. 2019). It has the advantages of high purification efficiency, useful recyclability, obvious energy saving qualities, economy, and environmental protection (Kanta et al. 2019). Carbon materials, as promising adsorbents, remove organic aldehydes from the atmosphere due to their high surface area, hierarchical porous structure, and chemical inertness (Fuchs et al. 2016; Valentín-Reyes et al. 2019). In addition, biomass-derived carbon materials have been explored.

A number of biomass wastes, such as sawdust, Euonymus japonicas leaves, palm shell, and corncob have been used to prepare carbon materials (Zhu et al. 2015; Karnan et al. 2017; Yek et al. 2019; Zhao et al. 2019). As an agricultural waste, corncob is produced at high levels because corn is widely cultivated across the planet. The spongy cellular structure of corncob may be favorable to the formation of porosity, and its wide availability also serves as an abundant and inexpensive feedstock option for the preparation of carbon materials.

During the synthesis of carbon materials through direct carbonization, it is difficult to control the porous structures and dimensions. The ion thermal approach provides a facile, sustainable, and large-scale strategy for synthesizing carbon materials. As shown in pioneering experiments, carbon materials were prepared by carbonization ionic liquid mixed with eutectic inorganic salts at high temperature (800 °C to 1000 °C) under inert atmosphere (Fellinger et al. 2013). The compatibility of the organic and the inorganic molten salts facilitated the molecular mixing of the two phases throughout the decomposition and carbonization process of the organic materials. This leads to the nanoseparation of the inorganic salt and the final carbon. Zinc chloride (ZnCl2) is one of the activating agents most widely used in the preparation of different carbon materials with a high carbon yield and surface area (Liu and Antonietti 2014). Potassium chloride (KCl) and sodium chloride (NaCl) are porogens for carbon materials, which result in the different porous structures (Ma et al. 2014). The combination of an activating agent with a porogens leads to a “one-stone-two-birds” situation of controlling pore structure and surface area. However, it has seldom been applied to the carbon preparation that occurs from biomass carbonization.

The present research proposes a simple approach for the use of the agricultural waste, corncob, to produce novel carbon materials. The morphology and porous structures of corncob-based carbon was tailored by a modification of the ratio of a eutectic salt ZnCl2/NaCl solution and the carbonization temperature. The prepared corncob-based carbon, with its proposed method, was employed as an adsorbing material to remove selectively organic aldehydes, namely formaldehyde and butyraldehyde gas. Furthermore, the adsorption properties and influencing factors of carbon materials on organic aldehydes were systematically investigated. The adsorption thermodynamics and kinetics theory were further analyzed for its application on industrial wastewater and the air pollution of biomass-derived carbon.

EXPERIMENTAL

Preparation of Carbon Materials

The carbon samples were prepared as follows. The corncobs (Jinan farm, China) were triturated with the corncob powders passing through 60 mesh sieves. The material was boiled for 0.5 h. Finally, the material was dried. A series of eutectic salt solutions were developed by dissolving 70 g of ZnCl2/NaCl (Sigma, Augusta, America) with a different mass ratio of 1:0, 2:1, 1:1, 1:2, and 0:1 in 200 mL deionized water, respectively. The processed corncob powders (10 g dry weight) were each impregnated in a eutectic salt solution for 4 h, and then dried at 100 °C. Once the solution was dried, all the samples were further carbonized at 900 °C under a nitrogen atmosphere. The prepared samples were denoted as C-Zn, C-2, C-1, C-0.5, and C-Na, respectively. For comparison, the control was produced from corncob powders via a direct carbonization at 900 °C under nitrogen atmosphere (denoted as carbon).

Characterization

Scanning electron microscopy (SEM) was performed using a JSM-7401F microscope (JEOL, Hokkaido, Japan) at an acceleration voltage of 20 kV. The images of the transmission electron microscopy (TEM) were obtained on a JEOL 2011 (JEOL, Hokkaido, Japan) apparatus that was operating at 200 kV. The sample was first dispersed in ethanol and then subjected to ultrasound. Further, the upper layer of liquid was taken out, and then ethanol was added, followed by a continuation of dispersing with ultrasound. After several dispersed-ultrasound process cycles, the supernatant solution was placed dropwise on the test copper mesh and dried at room temperature. Powder X-ray diffraction (XRD) patterns of the samples were measured using a Brucker D4 (Bruker, Frankfurt, Germany) powder X-ray diffractometer with Cu Kα radiation at 40 kV and 40 mA. Nitrogen sorption isotherms were measured with a Micromeritics ASAP 2020 sorptometer (Maike, Augusta, America) using nitrogen as the adsorbate at 77 K. All samples were degassed at 300 °C for more than 10 h before the analysis. The specific surface area (SBET) was calculated using the NLDFT method based on the adsorption data in the relative pressure range of 0.05 to 0.2, and the total pore volume was determined at the highest relative pressure.

Adsorption Tests

The measurement of the adsorption capacity of organic aldehyde by Intelligent Gravimetric Analyzer (IGA, HIDEN company, UK) was carried out. The IGA contains a precise pressure, temperature control system, and an ultra-sensitive thermostatic balance with an accuracy of 0.1 μg. Firstly, the adsorbent sample was vacuumed to purify the substance at high temperature, and then the adsorption pressure point was set to slowly increase the pressure of the whole system through computer control. Meanwhile, the computer tracked and recorded the changes of adsorbent relative to time and weight, and the temperature control accuracy of the adsorbent was ±0.1 C. The computer detects the rate of adsorption close to equilibrium and analyzes the data to ensure the accuracy of the data (Chen et al. 2012).

The procedure was continued out as follows: First, 10 mg samples were washed and removed from the water. The obtained samples were placed in the sample chamber under a temperature of 180 °C and a high vacuum setting (<10-3 Pa) for 10 h, after which the temperature was cooled down to 25 °C. After the cooling step, formaldehyde and butyraldehyde gases of different pressures were added to the sample chamber and the adsorption equilibrium of the samples were tested at each pressure level. Finally, the adsorption isotherms of the organic aldehyde were obtained.

RESULTS AND DISCUSSION

All the samples showed different morphological features (Fig. 1). As can be observed in Fig. 1a, the corncob-derived carbon without a eutectic salt template displayed irregularity in its morphology from the decomposition of cellulose, hemicellulose, lignin and the structure of the corncob itself (Jiang et al. 2016). The C-Zn sample (Fig. 1b) appeared as a connected, reef-like structure. This kind of phenomenon may be due to the ZnCl2 effect on the cellulose of biomass, where swelling, peptizing, and dissolving of the pores formed occurs. Additionally, the ZnCl2 effect leads to a shrinkage of the macropores that takes place after the pores dissolve (Spagnoli et al. 2017). By adding the NaCl active agent, C-2 (Fig. 1c) shows a regularity of spheres that were about 2 μm on the reef. When the ZnCl2/NaCl reached to 1, C-1(Fig. 1d) displays clear spherical structures, as well as further aggregation between the spheres. However, as can be seen in Fig. 1c, it is interesting that the spheres gradually broke down. By contrast, C-0.5 (Fig. 1e) shows a cracked tube-like structure with a 4 μm width, as well as a large number of pores in the wall of the tube. However, the C-Na sample (Fig. 1f) shows macropores with larger holes at about 10 μm in diameter. This finding is attributed to the ZnCl2’s ability to catalyze the conversion of the biomass derivatives via stripping away the hydrogen and oxygen atoms in the carbon source. Meanwhile, it is clear that the ratio of the eutectic salt ZnCl2/NaCl has an obviously influence on the sample morphology. The ZnCl2/NaCl catalyzes the conversion of the corncobs to the C=C and C=O containing compounds and then further carbonizes the compound to form the carbon materials rich in oxygen-containing groups (Kim and Jeon 2015). These interconnected macropores might be derived from the combination of inherent macropores in the cellular structures of the corn husks with different eutectic salt ratios. However, these morphologies will serve as reservoirs to adsorb the metal ions, and these thicker pore walls can provide a large amount of interface for the ion storage and adsorption (Hattori et al. 2017).

Fig. 1. SEM images of the different carbon samples (a, Carbon; b, C-Zn; c, C-2; d, C-1; e, C-0.5; f, C-Na)

The TEM images of the as-prepared carbon materials are shown in Fig. 2. Each of the samples displayed an intrinsic and special microstructure, as well as abundant porosity. The porous structures had slight changes from disordered crack-like pores on the control to ordered point-like pores on the C-Zn, to disordered worm-like pores on the C-2, C-1, and C-0.5, and to finally ordered to ripple-like pores of the C-Na. The results can be ascribed to the presence of mineral salts that affected the charring and aromatization of the corncob’s molecular skeleton and led to the creation of diversiform porous structures (He et al. 2013). The carbon materials with the ordered pores were derived from corncob when a single salt (ZnCl2 or NaCl) was used as the functional agent. The disordered worm-like pores were generated with the utilization of the eutectic salts (ZnCl2 and NaCl), which may be ascribed to reciprocal inhibition from both of the salts. Nevertheless, these ordered or disordered porous structures provide a high surface area and adsorption ability for the prepared corncob-derived carbon materials.

Fig. 2. TEM images of the different samples (a, Control; b, C- Zn; c, C-2; d, C-1; e, C-0.5; f, C-Na)

The XRD analysis showed that all the samples had similar XRD patterns. Two diffraction peaks appeared at 2θ of 23° and 44° with large widths and low intensities, which are the typical characteristic peaks of the graphitic carbon (Gao et al. 2014).

Fig. 3. XRD images of the different carbon samples

There were no obvious diffraction peaks resulting from the porogen salts in the XRD images. These findings indicate that only a trace amount of porogen salts were present in the as-prepared carbon materials. This suggests that residue from most of the salts were removed via evaporation and a washing that followed.

The nitrogen adsorption isotherms and the corresponding pore size distribution curves of the different carbon samples are shown in Fig. 4. The adsorption capacities of all the samples below the relative pressure (P/P0) of 0.1 exhibited curves indicating a typical microporous characteristic. Meanwhile, the hysteresis loop at the relative pressure (P/P0) of 0.4 to 0.95 is associated with the capillary condensation that occurs in the mesopores (Riachy et al. 2017). Carbon exhibited a type-I sorption isotherm of microporous characteristic, but the adsorption capacities are obviously low, which exhibits a porosity that is relatively poor for carbon. The C-Zn showed a type-I sorption isotherm. The higher adsorption capacity below the relative pressure (P/P0) of 0.1 for this sample is due to the increase in micropores (Tian et al. 2017). This finding indicates that the presence of ZnCl2 affected mainly the generation of micropores (Erdem et al. 2016). The enhanced adsorption capacity for the C-Zn is due to an increase in the contact area between the raw material and the activating agent. The ZnCl2 worked as a dehydration regent during carbonization, endowing the carbon materials with a porous structure and producing a high surface area by removing the hydrogen and oxygen (H2O) from the feedstock, and leaving behind much more of the elemental carbon (Zhao et al. 2017). The adsorption property was obviously improved by using ZnCl2 and NaCl simultaneously as processing agents, which indicates that the synergistic effect of ZnCl2 and NaCl could enhanced the adsorption ability of as-prepared carbon materials. When the ratio of ZnCl2 is dominant, a quantity adsorption of the C-2 was improved upon in comparison to the C-1 below the relative pressure (P/P0) of 0.1. This observation is probably due to the contraction of pore structure caused by the thermal breakdown of the corncob skeleton under the high ratio of the activating agent (ZnCl2) to the carbon precursor. When the ratio of NaCl2 was dominant, a hysteresis loop in the C-0.5 was enhanced in comparison to the C-1 at the relative pressure (P/P0) of 0.4 to 0.95. Further, when the NaCl agent was a single agent, the quantity adsorption saw a significant decrease. These results explain that the as-prepared C-Zn mainly showed microporous structure, and the C-Na exhibited mesopores structure. Moreover, eutectic salts were more favorable to the formation of pores than single salt, which could be attributed to the increase in the activating agent enhancing the thermal decomposition rate of the feedstock, and therefore, increasing the surface area of the as-prepared carbon.

The pore size distribution plots of the as-prepared samples are shown in Fig. 4b. The carbon exhibits a uniform pore size distribution, and its pore size mainly distributes in the range of 1.8 nm to 2.5 nm, with a peak at 2.0 nm. The as-prepared C-Zn displays an analogous pore size distribution and peak position to the carbon sample, but a higher peak occurred at 2.0 nm for more of the micropores. An intense and sharp peak of the C-2 appeared at 2.0 nm, indicating a uniform pore size distribution. However, with the increase of NaCl in the eutectic salt system, a larger number of mesopores were generated, while the number of microspores decreased for the C-1 and C-0.5, leading to a wider pore size distribution. Additionally, the C-Na exhibited a similar pore size distribution plot to the carbon except for the peak migration and the width of the pore size distribution. Obviously, the C-Na had a pore size distribution from 2.0 nm to 4.0 nm and the peak appeared at 2.8 nm, which indicates that larger mesopores were generated in the sample of C-Na than the carbon. The results show that the type and amount of porogens are an important influencing factor on the textural properties of the resultant carbon materials (Zhao et al.2017). Therefore, the pore size of the as-prepared carbon materials can be controlled by adjusting the ratio of the ZnCl2/NaCl in the eutectic salt system. The use of ZnClindependently promoted the formation of micropores and had little effect on the mesopores, i.e., in the as-prepared C-Zn. The addition of NaCl within a certain extent further facilitated the generation of micropores, i.e. in the as-prepared sample, C-2. Nevertheless, this trend was reversed with the increase of NaCl in the eutectic salt system, which resulted in reduced micropores and an increase in the mesopores of the as-prepared samples C-1 and C-0.5. The phase separation of the newly formed carbon materials and the ZnCl2/NaCl ratio may lead to the formation of pores in the carbon materials, which are dominated by mesopores and micropores. These findings indicate that NaCl played a role as the template for mesopores, and ZnCl2 works as an activation agent for micropores.

Fig. 4. Nitrogen sorption isotherms (a) and pore size distribution plots (b) of the as-prepared carbon samples

Textural parameters of different carbon samples are listed in Table. 1. Carbon showcased a low specific surface area (130 m2/g) and pore volume (0.07 cm3/g), and the pore structure is dominated by micropores (Smicro/SBET = 80%). The higher specific surface areas (360 m2/g, 289 m2/g) and pore volumes (0.12 cm3/g, 0.10 cm3/g) of the C-Zn and C-Na are shown, but the Smicro/SBET increased to 87%, and decreased to 68% in each, respectively. The above results suggest that both of ZnCl2 and NaCl have important roles in controlling the surface area and porous percentage of resultant carbons. That is, the NaCl plays a crucial role in the formation of mesopores, and similarly the effects of ZnClis to promote the generation of micropores that result in a larger surface area. Furthermore, the amount of ZnCl2 also affected the SBET of the as-prepared samples when the eutectic salt ZnCl2/NaCl was used in the production of the carbon materials. The SBET increased from 582 m2/g in the C-0.5 to 762 m2/g in the C-1 sample, and then further reached 960 m2/g in the C-2 sample as the amount of ZnCl2 in eutectic salt system increased. This finding is attributed to the role of ZnClin increasing the number of micropores. Moreover, the average pore size decreased from 3.1 nm in the C-0.5 to 2.0 nm in the C-2, which provided further evidence of the function of ZnCl2 in the shrinkage of the pore size. As can be seen, the nitrogen adsorption capacities of as-prepared carbon materials derived from salt solution are clearly higher than those derived from precursors. And the synergistic effect of eutectic salt ZnCl2/NaCl could control the structure of the pores, pore size, and surface area. In particular, the high surface area of the material exposes the more active centers to the reactants, and an appropriate pore volume can improve the adsorption and diffusion of the reactant molecules (Liu and Antonietti 2014).

Table 1. Textural Parameters of Different Carbon Samples

The adsorption isotherms of formaldehyde gas (Fig. 5a) and butyraldehyde gas (Fig. 5b) distributed by the different carbon samples are shown in Fig. 5. The carbons exhibited an excellent adsorption performance for the two tested organic aldehydes.

Fig. 5. Adsorption isotherms of formaldehyde gas (a) and butyraldehyde gas (b) by the different carbon samples

The adsorption ability is in the order of C-2>C-1>C-Zn>C-0.5>C-Na>carbon for the formaldehyde gas. The formaldehyde gas adsorption capacity is interrelated with the microporous content of the carbons. This can be attributed to an increase in the microporous specific surface area, which led to more adsorption sites that allowed the formaldehyde gas to be absorbed onto the pore structure quickly and efficiently. The adsorption ability was in the order of C-1>C-0.5>C-Na>C-2>C-Zn>carbon for the butyraldehyde gas. This order is attributed to how the adsorption amount of the butyraldehyde is dependent on the specific surface of the mesopores, as well as how the mesopores can provide a fast diffusion channel and short diffusion distance. During the carbonization of the raw material, ZnCl2 has a skeleton role for the new carbon, leading to placement of the new carbon deposition on it. New carbon has a primary key to adsorption and can thus make the carbon and organic aldehyde stick together (Van et al. 2017). The results exhibit that the activity enhancement that occurs with the additions of ZnCl2 may be related to the increase in the surface area. The synergistic effect that occurs from the mixtures leads to a higher specific surface area, further achieving the higher adsorption ability for the organic aldehyde. The high surface area of the carbon materials exposes the more active centers to the reactants, and an appropriate pore volume will improve the diffusion of the reactant molecules, which may serve as the main reasons. The formaldehyde gas adsorption amount of the C-2 and the butyraldehyde gas adsorption amount of the C-1 reach to 460 mg/g and 480 mg/g, respectively.

Figure 6 exhibits the adsorption isotherm for the formaldehyde and butyraldehyde gas of the superior carbon (C-2 and C-1) through the Langmuir and Freundlich model. Superior carbon materials showed the better fitting curve of the Langmuir model, and the coefficients of determination R2 for the C-2 (0.997) and C-1 (0.982) were closer to 1 (Table 2), indicating a better composite of the Langmuir model for the adsorption process. The value of 1/n is between 0.1 to 0.5 for the formaldehyde gas of the C-2 sample through the Freundlich model and is between 0.5 to 2 for the butyraldehyde gas of the C-1 sample through Freundlich model. The results indicated it easily occurred for the adsorption process, which is a physical adsorption.

Fig. 6. Formaldehyde gas for the C-2 (a) and the butyraldehyde gas for the C-1 (b) by the Langmuir and Freundlich model isotherm

Table. 2 Summary of the Langmuir Adsorption Parameters of the Formaldehyde and Butyraldehyde on the C-2 and C-1 Samples