Abstract

Activated carbons were prepared by carbonization of tomato paste processing industry waste at 500°C followed by chemical activation with KOH, K2CO3, and HCl in N2 atmosphere at low temperature (500°C). The effects of different activating agents and impregnation ratios (25, 50, and 100 wt.%) on the materials’ characteristics were examined. Precursor, carbonized tomato waste (CTW), and activated carbons were characterized by using ultimate and proximate analysis, thermogravimetric analysis (TG/DTG), Fourier transform-infrared (FT-IR) spectroscopy, X-ray fluorescence (XRF) spectroscopy, point of zero charge measurements (), particle size analyzer, scanning electron microscopy (SEM), energy dispersive X-ray (EDX) spectroscopy, nitrogen adsorption/desorption isotherms, and X-ray diffraction (XRD) analysis. Activation process improved pore formation and changed activated carbons’ surface characteristics. Activated carbon with the highest surface area (283 m3/g) was prepared by using 50 wt.% KOH as an activator. According to the experimental results, tomato paste waste could be used as an alternative precursor to produce low-cost activated carbon.

1. Introduction

Activated carbons (ACs) are materials generally composed of carbon which exhibit well-developed internal surface area and porosity [1]; they have been widely used in different areas such as in gas storage or separation, catalysts and catalyst support, pollutant or odor removal, supercapacitors, pharmaceuticals, and wastewater treatment [2, 3]. ACs are low-toxic, inexpensive in production, and easily utilized which can be produced from any natural or synthetic carbonaceous solid precursor such as coal, wood, sawdust, local waste products, and synthetic polymers [4, 5].

The structural properties of ACs are related to their raw material and the method of preparation. Organic materials are the mostly used precursors to produce ACs because of their high carbon and low ash content. Due to the advantages of environmental protection, agricultural or industrial wastes are considered to be critical precursor as they are inexpensive, renewable, safe, available at large quantities, and easily accessible sources [5]. The production of ACs can be performed by a two-step process: carbonization and activation. The first step involves carbonization or pyrolysis of the carbonaceous materials at elevated temperatures (500–900°C) in an inert atmosphere, in which several complex, rival, and consecutive reactions lead to the achievement of a fixed carbon mass due to release of volatile compounds. The second step is the activation of the material obtained from carbonization step by chemical or physical processes [1, 6]. The physical activation of the char is accomplished in the presence of activating agents such as air, CO2, oxidizing agents, or steam at 700–900°C [3]. Chemical activation is performed by mixing the precursor with an activating agent such as ZnCl2, H3PO4, FeCl3, H2SO4, K2CO3, NaOH, HCl, or KOH [1, 5, 6].Chemical activation has several advantages compared to the physical activation such as lower pyrolysis temperature, short activation time, low energy consumption, high yield, and high surface area [2, 7].

Activating agents of HCl, KOH, and K2CO3 were used for activated carbon preparation from moso bamboo [3], palm shell [8], maize stalks [9], pistachio shell [10], siris seed pod [11], potato peel [12], cotton stalk [13], deoiled canola meal [14], cherry stone [4], and Jatropha husk [15]. Turkey is the world’s third largest tomato producer; tomato production occurs in ten months of the year averaging more than 11 million tons annually [16]. The high volume of waste constitutes an environmental problem and its reutilization is useful; indeed the high volatile matter (77.35%) and lignocellulosic content (75.6%) of tomato paste waste promote its use as precursor for production of low-cost activated carbon. Using tomato waste as an activated carbon precursor was limited but considered a feasible approach to properly transform the resource into more valuable materials. On these bases, tomato waste is considered to be good lignocellulosic material and inexpensive source alternative to more expensive and polluting precursors such as coal. Therefore, the objective of this study is to investigate the structure of raw material and carbons which were obtained by carbonization of tomato paste waste and chemical activation of chars with different impregnation ratios (25, 50, and 100% wt.) of HCl, K2CO3, and KOH.

2. Materials and Methods

2.1. Raw Material

Tomato is one of the major crops planted in Turkey. The feedstock tomato paste waste (TW) was obtained from Food Factory in Bursa, Turkey; then it was air-dried in the laboratory, crushed, and sieved to obtain mean sizes. The average particle size of TW is about 0.655 mm.

2.2. Preparation of Activated Carbon

Heating rate and contact time are two important parameters in the activated carbon production. There are various applications of these parameters in the literature: (i) heating rate applied as 2, 5, 10, 15, and 30°C/min and (ii) activation time utilized as 15, 30, 45, 60, 120, and 240 min [1726]. In the light of the above information, the raw material was heated in a muffle furnace at a rate of 15°C/min from room temperature to 500°C and maintained at this temperature for 1 h under static atmosphere to obtain carbonized material (CTW). The activation process was performed under impregnation ratios of 25, 50, and 100% (wt) of activating agent (HCl, K2CO3, and KOH): carbonized material, where carbonized material was mixed with activating agents, stirred for 2 h, and kept in a dark room at room temperature for 24 hours. Then, impregnated samples were heated in a tubular furnace at the heating rate of 15°C/min from room temperature to 500°C and maintained at this temperature for 15 min under N2 flow for 100 cm3/min. After cooling, the resulting ACs were washed with distilled water until pH became neutral for removal of waste from the activating agent and by-products produced during the process. The washed samples were dried at ambient temperature. Activated carbons prepared by chemical activation with KOH, K2CO3, and HCl were coded as KAC, KCAC, and HAC, respectively. Impregnation ratios were stated after the codified names, in such a way that KAC25 implied the 25% KOH impregnated activated carbon.

2.3. Characterization

The effects of the chemical activation of tomato paste waste were investigated by analyzing the improvements in the structural, physical, and chemical properties.

Structure and preliminary analysis were performed to complete the proximate analysis of TW; elemental analyses of TW, carbonized TW, and ACs were done by elemental analyzer (Leco CNH628 S628) to find carbon, hydrogen, nitrogen, and oxygen contents of materials by using helium, dry air, and oxygen gases. The complete combustion of all organic samples was carried out by operating elemental analyzer’s furnace at 950°C.

The mineral composition of tomato paste waste was determined by using X-ray fluorescence (XRF) spectrometer (Empyrean PANalytical Axios Max Minerals).

The thermal behavior of TW was measured with TG analyzer (Setaram Labsys Evo). About 10 mg of sample material was heated from room temperature to 1000°C at a ramping rate of 10°C/min under argon gas atmosphere with the low rate of 20 mL/min, until no further weight loss was detected.

FT-IR spectroscopy was used to identify the chemical groups present in the raw material and ACs. The infrared transmittance measurements of TW and ACs were carried out on Perkin Elmer Spectrum 100 FT-IR spectroscopy in the 4000–400 cm−1 wavelength range. The FT-IR spectrums were obtained using the ATR technique (with a diamond protected attenuated total reflectance crystal unit) with a resolution of 4 cm−1 after 100 scans.

The point of zero charge () is defined as the pH of the mixtures at which surface charge density on the material is zero. The pHPZC value was determined by the solid addition method [27, 28]. The solutions (20 mL) with pH varying from 1 to 8 were transferred to a series of 100 mL conical flasks. The values of each solution were adjusted by adding either 0.1 N HCl or NaOH and were measured. The carbonized TW was added to each flask; the suspensions were then manually shaken and the zeta potential of carbonized tomato waste was determined by using Malvern Zeta Sizer Nano Series Nano-ZS. Zeta potential (mV) of particles was plotted against pH and the point of intersection of the resulting curve with abscissa, at which pH is 0, gave .

The surface physical morphologies of TW, carbonized TW, and ACs were identified by using a scanning electron microscope (Zeiss Supra VP 40) with an accelerating voltage of 5 kV equipped with an energy dispersive X-ray (EDX) spectrometer. The samples were sputter-coated with platinum (Quorum Q 150 R ES DC Sputter).

X-ray diffraction (XRD) analyses of the tomato paste waste, the carbonized TW, and the ash of activated carbon with the highest surface area were achieved in a PANalytical X’Pert Pro Materials Research Diffractometer with ( = 0.15405 nm) radiation as the X-ray source. Scans were recorded with a scanning rate of 0.01°/s, typically in the angle range between 5° and 80°.

The particle size analysis of carbonized TW was performed by using Malvern Mastersizer Hydro 2000S.

N2 adsorption/desorption isotherms at 77 K were measured by using Micromeritics Asap 2020 analyzer in order to determine the surface areas and pore volumes of ACs. The surface area was determined by Brunauer-Emmett-Teller (BET) equation and total pore volume () was defined as the maximum amount of nitrogen adsorbed at relative pressure of . The pore size distribution was obtained by applying the Barrett-Joyner-Halenda (BJH) analysis method.

3. Results and Discussion

3.1. Ultimate, Proximate, and XRF Analysis

Ultimate and proximate analysis results of TW were given in Table 1. Based on its nature, cellulose, lignin, and hemicelluloses should be the main components of TW. Elemental compositions of carbonized TW and activated carbons were shown in Table 2. The content of oxygen was calculated as the difference between 100 and the total wt% of CHN constituents. The low carbon (49.69%) and high volatile (77.35%) content of tomato paste waste indicated that the precursor was not suitable enough for activated carbon production. Therefore, at first, tomato paste waste was carbonized to increase the content of carbon up to 67.44%. Hydrogen and oxygen elements in the TW are lost more easily than carbon element in the heat treatment, thus, producing a carbonaceous material after activation [29]. All carbon samples had higher carbon contents and lower hydrogen and oxygen contents when compared to TW. As a result, chemical activation had accelerated the hydrogen and oxygen removal resulting in increased carbon as expected.

The major chemical analysis of the tomato paste waste obtained using X-ray fluorescence spectroscopy was shown in Table 3. XRF analysis indicated that K was the most abundant element (39.23%) in the inorganic fraction of TW. As expected, other major elements in the TW were oxygen, calcium, and phosphorus.

3.2. Thermal Behavior of Raw Material

The thermal degradation (TG) and derivative thermal degradation (DTG) curves of TW were given in Figure 1. According to the literature, pyrolysis of different lignocellulosic materials occurred as the two weight-loss stages. The first one is correlated with the hemicellulose decomposition and the second with cellulose decomposition while degradation of lignin appeared over a wide temperature interval (150–480°C or higher). Lignocellulosic materials identify the decomposition temperatures of these components [30, 31]. TG and DTG curves can be divided into three stages of weight loss. The first stage that was obtained at 35–200°C indicated the loss of adsorbed water and light volatile materials such as glacial acetic, alcohols, and the physically adsorbed species [9]. In the second stage, sharp weight loss of the raw material was at 200–450°C in which carbonization process began and mainly hemicellulose and cellulose fractions decomposed [32]. Finally, consolidation of the char structure at 400–800°C resulted in a small weight loss. The DTG curve showed two endothermic peaks around 110 and 400°C which belonged to the water removal and hydroxyl deformation, respectively [2]. Accordingly, the carbonization temperature of 500°C was selected for the carbonized TW production.

3.3. Surface Chemistry Characterization: pHPZC and Infrared

The zero point charge () is defined as the pH at which the surface of material has net electrical neutrality [7]. A zeta potential analysis result of CTW was shown in Figure 2. The zeta potential of particles was between approximately 0 mV and −30 mV. The surface charge of carbonized TW between 1 and 3 was found as zero; lower value near 2 indicated that carbonized TW surface had acid functional groups on it.

The 4000–400 cm−1 infrared spectral region of raw material was shown in Figure 3. It can be said that the chemical structure of TW, being a lignocellulosic material, is made up of different atomic groupings and a large number of functional groups. O-H stretching vibration of hydroxyl functional groups including hydrogen bonding was detected at bandwidths of 3300–3200 cm−1; this peak shows a presence of alcohol, phenol, or carboxylic acids [33]. Other major peaks detected at 2923 and 2852 cm−1 were attributed to asymmetric and symmetric C-H stretching of aliphatic methyl and methylene. The carbonyl (-C=O) stretching vibration occurred at 1744, 1709, and 1641 cm−1, and peaks observed at 1530 and 1457 cm−1 were assigned to the aromatic C=C ring stretch. Bands at 1241 cm−1 were assigned to CH=CH stretching and also bands between 1100 and 1000 cm−1 were detected to C-O stretching vibrations of lignin [16]. The band at 600 cm−1 was related to out-of-plane angular deformations of benzene derivatives [1].

Functional groups determine the surface properties of the carbons and their quality, so they are significant characteristics of the activated carbons. The infrared spectrums of the activated carbons as a function of the impregnation ratio were given in Figure 4. The FT-IR spectrums of activated carbons were slightly different from TW, which is a result of the chemical and thermal treatment. The TW spectrum had more vibrational bands than ACs spectrums; this may be due to the chemical bonds breakage in the raw material during the carbonization process followed by chemical activation. In addition, the intensity of some absorption bands was decreased mainly, which signified a change in the tomato paste waste’s functionality [1, 34]. The band at 3100 cm−1 was attributed to C-H stretching vibration. The intensity of peak decreased after activation, which indicated that hydrogen was removed during the achievement process. The intensities of the -OH stretching vibration band at about 3300–3400 cm−1 and C=O vibrations at 1750 cm−1 decreased for all activated carbons after carbonization. The band at 1470–1430 cm−1 was ascribed to C-H bending vibrations in CH3 groups for all activated carbons. The band at 1592–1530 cm−1 was the unsaturated stretching of C-C bonds ascribed to aromatic C=C vibration for activated carbons.

3.4. Morphology Analysis

Figure 5 showed the SEM photographs of TW and carbonized TW. As shown in Figure 5(a), TW had a relatively smooth surface without large defects. Carbonized TW surface presented more damage due to release of volatile compounds during the carbonization procedure (Figure 5(b)). According to EDX analysis, TW contained 61.40% C, 38.05% O, 0.41% Al, and 0.13% Ca, while carbonized TW included 77.92% C, 20.58% O, and 1.50% K; this result conformed to FT-IR analysis.

SEM images of activated carbons with different impregnation ratios were given in Figure 6. The chemically activated carbons exhibited larger cavities and rougher surfaces compared to TW. After activation process, activated carbons had discontinuous and irregular surfaces, caused by the decomposition of the sample matrix by the activating agents followed by the dehydrating action of them during heat treatment, which led to the porosity development [1, 35]. The large pores produced by the chemical treatment promoted the activation in the internal surface of the carbon particles [36]. Therefore, KOH, K2CO3, and HCl were proved to be effective activating agents for the production of high surface area activated carbons. Significant quantities of potassium were observed on the surface of activated carbons prepared with KOH and K2CO3, while chlorine was detected on HCl treated activated carbons by EDX analysis (see Table 4). The chemical composition of raw material and activating agent assisting in depolymerization, dehydration, and redistribution of constituent biopolymers caused significant changes in the pyrolytic decomposition of the lignocellulosic material [37].

3.5. Crystalline Structure

X-ray diffraction (XRD) technique is a substantial tool to analyze crystalline nature of materials. XRD profiles of raw material, carbonized TW, and the ash of activated carbon with the highest surface area (KAC50) were given in Figure 7. XRD spectrums of TW had wide and high peaks which were converted to narrower peaks after activation process. Diffraction peaks around 25° and 44° conformed with (0 0 2) and (1 0 0) diffraction of disordered stacking of graphite-like microcrystalites, respectively [31, 3840]. XRD results revealed that TW had mainly monoclinic KO2 and cubic K structures. Other components were cubic Ca and CaO and monoclinic P4O8. Carbonized TW had principally cubic MgO and Mg structures. Orthorhombic P and P2O5, cubic K and Ca, and monoclinic KO2 structures were also detected in the structure. The main components of the ash of KAC50 were found as cubic MgO, Ca, and K. Additionally, cubic CaO and K2O, hexagonal Mg, and orthorhombic P and P2O5 were detected in KAC50.

3.6. Particle Size Analysis and Textural Properties

According to particle size analysis, main particle size distribution of CTW was found as 86.42 nm. Activated carbon’s surface area extremely influences the reactivity and combustion behavior, so it is critical like other chemical and physical properties [29]. Results for the specific surface areas and pore properties of the prepared activated carbons were listed in Table 5. The activated carbons had different textural parameters related to activating agents and impregnation ratios. HCl treatment did not support a substantial development of porous structure, as denoted by the specific surface areas between 43 and 77 m2/g. The chemical activation with KOH confirmed that it was more efficient in the development of activated carbons’ porous structures. The increase in surface area was based on the decomposition of activating agent which assisted the formation of water and increased the formation of various gases such as CO, CO2, and H2. These compounds were released from the precursor forming the porous structure of the activated carbon [13, 14, 32, 40, 41]. Based on related research, chemical activation utilizing alkalis such as NaOH and KOH in excess attained in microporous ACs [41]. The higher surface areas are probably due to the opening of the restricted pores. The surface area increased with the impregnation ratio of activation process; an exception was found for sample 100% wt. KOH.

The pore size distribution is expressed as heterogeneity degree of porous materials [1]. The nature of the precursor and the production procedure strongly affect the porous structure of the produced activated carbon [3]. According to the IUPAC classification of pore dimensions, the pores could be classified into micropore (size < 2 nm), mesopore (2–50 nm), and macropore (>50 nm). A considerable amount of pores was distributed in the region of 9.4–15.2 nm, 6.6–13.1 nm, and 17.6–22.4 nm for KOH, K2CO3, and HCl impregnated samples, respectively, further indicating the presence of mesopores in the activated carbons. According to pore size distributions, KOH and K2CO3 impregnated activated carbons exhibit pores of diameter lower than 10 nm while HCl impregnated activated carbons presented a distribution shifted towards larger pore sizes (>20 nm).

3.7. Utilization of KAC50 Carbonaceous Material in Cobalt(II) Removal

In order to characterize the zeta potential-pH behavior of the KAC50 carbonaceous material, pH range was adjusted between 2 and 12. Figure 8 showed that the measured zeta potential was between a maximum of +5.99 mV and a minimum of −22.7 mV. The maximum measured zeta potential value was at a pH of 2 which corresponded to a zeta potential of +5.99 mV. The observed decrease in zeta potential with an increase in pH is dependent on the respective surface acidity constants. At higher pH values, the reduction in the rate of zeta potential change with pH could be due to the saturation of surface groups [42].

Batch adsorption experiments were conducted in a set of conical flasks containing 50 mL of solution to investigate the effects of pH (2–8). The effect of pH solution on cobalt(II) adsorption was examined by using 0.1 g KAC50 and 100 mg/L of cobalt(II) ion concentration for 1 hr at 293 K. The suspensions were then filtered and metal ion concentrations in the supernatant solutions were measured by Atomic Absorption Spectrophotometer (GBC933AA). The sorbed metal concentrations were calculated by the difference between initial and final Co(II) concentrations in aqueous solution. The metal ion removal efficiency is identified as follows [16]:where and  (mg/L) are the initial and final concentrations of the Co(II) solution, respectively.

Metal ions removal from aqueous solutions by sorption is highly influenced by the pH of the solution. Due to the results given in Figure 9, % removal efficiency increased acutely in the pH range of 6–8, while it was low under acidic conditions. The higher % removal efficiency was obtained as 89.36% at pH = 8. When pH of the solution increased from 2 to 8, the cobalt(II) adsorption capacity reached a range from 12.31 to 44.68 mg/g. H+ ions and metal ions compete severely for active sorption sites at lower pH values which leads to limited sorption of metal ions [16].

4. Conclusions

Tomato paste waste was selected as a suitable waste due to its abundance in Turkey to produce activated carbon. In order to obtain low volatilization and a high char yield, low heating rate was used. KOH, K2CO3, and HCl have been used as activators for preparation of activated carbon from tomato paste waste. Effects of chemical activation agents and impregnation ratios (25, 50, and 100% wt.) on structure and surface chemistry of activated carbons were investigated. Raw material, carbonized TW, and activated carbons were characterized by several techniques and methodologies to investigate the effects of carbonization and activation procedures.

Tomato paste waste was mainly constituted by carbon and oxygen. Due to the experimental results, characteristics of activated carbons are affected by chemical agents and impregnation ratios. Chemical activation also improved the carbon content and decreased the oxygen and hydrogen contents of the carbonized tomato waste and activated carbons as expected. According to FT-IR results, surface functional groups of TW were removed after preparation procedure. FT-IR spectrum of TW and activated carbons showed that the intensity of peaks decreased as a result of removing hydrogen during the applied process. SEM images showed that the morphology of the tomato paste waste changed after carbonization and activation processes. The external surfaces of the activated carbons were full of cavities. Activating agents caused pore, void, and cavity formation on precursor; also the activating process removed activating agents and improved surface areas. Activated carbon with the highest surface area (283 m2/g), which was prepared by 50% KOH activation, should be chosen as optimal. KOH was found to be the optimum activating agent for activated carbon production from tomato paste waste.

Metal bearing effluents can cause severe environmental pollution; thus removal of metal ions by adsorption is an important circumstance. The maximum cobalt(II) removal (%) was obtained as 89.36% at pH = 8. Since tomato waste was abundant and low-cost material, it could be successfully applied for the production of activated carbon and removal of cobalt(II) ions from aqueous solutions.

The results indicated that activated carbons with favorable physicochemical properties could be produced using different chemical activating agents and suggested that tomato waste was a good precursor material. Due to the obtained results, various chemical activators (H3PO4, ZnCl2, H2SO4, etc.) could be utilized to improve structural and surface characteristics of tomato paste based activated carbons. Furthermore, prepared activated carbons with optimum properties could be good candidates as aqueous phase adsorbents for inorganic and organic pollutant uptake.

Competing Interests

The authors declare that they have no competing interests.