Characterization of Activated Coconut Shell Charcoal as a Zinc Absorbent for Used Oil Lubricant

Used oil lubricant containing zinc is dangerous for human health and the environment and is categorized as B3 waste (hazardous and toxic materials). Zinc separation can be performed by the adsorption method using activated coconut shell charcoal. This study aims to determine the efficiency of zinc adsorption on lubricating oil with particle size and percentage of solvent variations. Adsorbent characterization is estimated by the Langmuir and Freundlich equilibrium model, while kinetic adsorption is estimated by pseudo-first and pseudo-second order kinetic models. Coconut shells are heated at a temperature of 300OC for one hour, and the results are soaked with HCl 25% for 18 hours. After the netralization process, charcoal is activated for three hours in a temperature of 500OC and saved in closed storage. Activated coconut shell charcoal and a number of volumes of H2SO4 as a solvent are stirred together using oil lubricant with stirring speed variation for every sample for two hours. Samples are taken every 30 minutes for the destruction process using HNO3 68% for two hours. The zinc concentrations before and after the absorption process are analyzed using Atomic Absorption Spectroscopy. The highest efficiency in the 0.5% v/v solvent and the particle size of -100 mesh variations ae 95.06% and 80.32%, respectively. The maximum adsorption power in the Freundlich isotherm is 5.35 g/g. Freundlich isotherm equilibrium and pseudo-second order are suitable methods to describe the characteristic evaluation of activated coconut shell charcoal and an adsorption kinetic model.


Introduction
Lubricating oil is important in mechanical activities. It serves to prevent and reduce wear because of the friction or mechanical contact. Motor oil is the biggest application of oil lubricant (Awad and Mohammed, 2014). Motor oil can experience deterioration caused by heat, oxidation, and fuel contamination. This has an impact on the other parts of the engine. The degree of degradation depends on the severity of the condition of the engine and the length of time of its use. Therefore, it is necessary to change lubricant oil periodically (Dai et al., 2016).
The increasing use of oil lubricant is simultaneous with the increasing total of motor vehicles. Used oil lubricant consists of 20% zinc and 4-5% lead. The content of zinc in used lubricating oil can cause pollution. The sources of pollutant zinc are generated from industries, such as general industry and mining, plating, fertilizers, paper products, and fibers (Parmar and Thakur, 2013). Zinc is an anorganic material that sometimes causes problems to the environment. Pollution comes from the contamination of mining or industrial waste. Zinc waste can cause death to plants, invertebrate animals, and fish when the concentration is higher than 1 ppm. In order to prevent zinc pollution before releasing it into the environment, it must be treated using lubricant oil.
Various regenerative technologies are used to generate reusable lubricant by the following five methods, namely acid/clay; distillation; solvent de-asphalting; Thin Film Evaporation (TFE) with hydrofinishing, TFE with clay finishing, and TFE with solvent finishing; TDA with clay finishing and Thermal De-Asphalt (TDA), with hydrofinishing process (Jafari and Hasanfour 2015). The first step of acid/clay process is the addition of sulfuric acid into the used lubricant oil. The product is namely dehydrated waste oil. Acid sludge is separated by deposition. The colloid, organic acid, foreign and wave substrances are removed by porcelain and aluminum silicate as the types of clay. The end of the process is filtering process for getting reusable oil (Kajdas, 2000). The second method is distillation process. The process is relatively same with acid/clay process. The processes step is dehydration and removal of light, vacuum distillation, and clay finishing (Kajdas, 2000).
The third method is solvent de-aspalting process. This process uses dehydration and removal of light, vacuum distillation, solvent extraction, centrifugation, stripping, vacuum distillation, and clay finishing. Many researchers using solvent for treatment used lubricant oil such as propane (purity 95%) (Rincón et al., 2003), MEK methylethylketone (Alhamed and Al-Zahrani, 1999) and chloroform (Lee et al., 2007). On the other hand, Kim et al. (2003) studied the use of bentonite, SBS polymer, sodium hydroxide, and the lime as solvent. Another researcher using n-hexane for purification used lubricating oil (Hasanpour et al., 2013).
The fourth method is TFE. Steps of the TFE process is dehidration removal of oil, vacuum, hydrofinishing, and filtering. The aim of hydrofinishing is to remove nitrogen, hydrogen, and sulphure compounds.
The fifth method is popular with the name of solvent extraction hydrofinishing. This process uses combination between solvent extraction and hydro-finishing. Solvent is used for eliminating foreign substance and hydrofinishing for fortifying oil quality. Two processes are combined for eliminating sulfur, nitrogen, and oxygen (Hsu et al., 2009). The fifth method is famous with TDA with hydrofinishing process and TDA with hydrofinishing. In the process, liquid petroleum gas (LPG) condensate and stabilized condensate (SP) are used as extraction solvents. The extracting process is continued with the clay treatment using clay material (Hamad et al., 2005). Table 1 describe differences between regenerative technologies for the used lubricant oil.
In addition, adsorption technique is also commonly applied to remove zinc metal in the used lubricating oil (Labied et al., 2018). Adsorption is a method of waste management that utilizes the ability of certain solids to bind certain substances from the solutions and bring them to the surface (Li et al., 2017). The parameters studied to optimize the use of nonconventional adsorption in waste management are adsorbate and adsorbent characteristics, adsorbent concentration, contact time, pH of solution, adsorbate concentration, particle size, and thermodynamic parameters (Bazrafshan et al., 2016). In previous studies, the absorption of zinc with clay waste as an adsorbent resulted in pH, adsorbent concentration, and contact time influencing greatly the adsorption rate.
Other researchers have studied zinc adsorption using precipitated calcium carbonate. The results of these studies show that the Langmuir isotherm is the suitable model for zinc adsorption, with a correlation coefficient of 0.541. This correlation coefficient is less than the maximum because the correlation coefficient value is relatively small. Adsorption without activation of an adsorbent gives less-thanoptimal results, with a k value and adsorption power lower than those in adsorption by activation.
Researches about adsorption process of the used lubricating oil has been explored by some researchers by combining the perfluoroethylene oxide and perfluoromethylene oxide (Ikenaga et al., 2007), silica (SiO2), alumina (Al2O3), silica-alumina (SiO2-Al2O3) supported iron oxide (Fe2O3) catalysts (Bhaskara et al., 2004), graphite (Yao et al., 2016), bentonite (Oduola and Okwonna, 2016). In this research, coconut shells are used as absorbent. The advantages of coconut shells as an adsorbent are that they have excellent natural structure and low ash content (Song et al., 2014). Coconut shells will be burned before turning into charcoal; then, they will be activated with HCl for 18 hours to make the reagent experience an expansion of the surface area and able to bind the lead maximally.
Hlaing et al. (2011)  This research studies activated carbon of coconut shell charcoal as an adsorbent for adsorption liquid for waste consiting of Zn 2+ . The effect of particle size and percentage of solvent on the adsorption capacity of Zn have been evaluated. Based on these studies, Freundlich and Langmuir isotherm models were used to fit the equilibrium data. Finally, the adsorption kinetic model using pseudo-first-order and pseudo-second-order were investigated.
A plot of log Cµ versus log Ce is conducted with regression linear equation to obtain 1/n as slope and log k as intercept. The magnitude of 1/n < 1 indicates the favorability of the adsorption process.
where, k: adsorption capacity n: adsorption intensity

Kinetics Modelling
The rate constants were interpreted by using pseudo-first-order and pseudo-second-order models.

Pseudo-First-Order Kinetics Model
The reaction rate constant using a pseudofirst-order kinetic model was proposed by (Yuh-Shan, 2004) which is revealed by Equation 8.
If a plot of ln (Ce -Cµ) is applied to t, we will obtain k1 and Ce values.

Materials
Used oil lubricant was used as raw material which was collected from the motor vehicle repair shops around Surakarta, Central Java, Indonesia. Coconut shells were collected from the traditional market around Surakarta, Central Java, Indonesia. HNO3 (Merck Millipore), HCl (Merck Millipore), distilled water, H2SO4 (Merck Millipore), standard solution Zn (Merck Millipore), and Whattman 42 were chemicals used in this research.

Preparation of Absorbent
Coconut shells were cleaned and broken into smaller size and dried in an oven at 100 o C for 1 hour. After drying process, the coconut shells were put into furnace with a temperature of 300ᴼC for one hour for the charcoaling process. In these temperature, coconut shell charcoal does not turn to ash because coconut shell has a low ash content. Besides, it is not burned at its maximum temperature, which is 1500ᴼC (Tinga et al., 2016). Adsorbents at room temperature were pounded and filtered using 149 µm sieve. One hundred gram adsorbent was soaked into HCl 25% 250 mL for 18 hours (Bath et al., 2012). The result of the marinades was filtered and washed with distilled water to reach the neutral pH. The marinades were put into a furnace at a temperature of 500ᴼC for 3 hours for the activation process. Activated adsorbents were stored in the desiccator for 0.5 hours.

Adsorption Experiment
Used oil lubricant as much of 400 mL, 10 g adsorbents, and various percentage of H2SO4, namely 0% v/v, 0.5% v/v, 1% v/v, and 2% v/v were put into glass flask with volume of 600 ml, mixed using magnetic stirrer in the speed of 10 for 2 hours. Every 30 minutes (0, 30, 60, 90, and 120 minutes), the filtrate was taken as much of 50 mL. Then, it was transferred to 100 mL Erlenmeyer flask which had been labeled. The same procedures were repeated in the particle size variations, such as -20 + 40 mesh, -40 + 60 mesh, -60 + 100 mesh, and -100 mesh.

Sample Analyzing
Used oil lubricant had experienced adsorption (filtrate) in the various percentages of solvent and particle size of destruction process before being analyzed using Atomic Absorption Spectroscopy (AAS). Filtrate in the Erlenmeyer flask was added with 5 mL HNO3 68%; then, it was heated on electrical stove with constant temperature for 2 hours. In the first 1 hour, 5 mL HNO3 68% was added again. After heating, the result of destruction was cooled to room temperature. The filtrate was filtered using Whatman 42 paper. The digestion product was diluted with HNO3 in a 100 mL volumetric flask and shaken until homogeneous. The filtrate was filtered and analyzed using AAS Shimadzu AA-6650.

Coconut Shell Charcoal
The coconut shell charcoal has a particle size of 149 µm before being used as adsorbent and analyzed using Indonesian Industrial Standard Number 0258-79. The adsorbent test results are shown in Table 2. Based on Table 2, coconut shell adsorbent in the particle size of 149 µm is proper to be used for adsorbing the used oil lubricant. Figure 1 describes the relationship between time versus percentage efficiency in the percent solvent and particle size variations. Equilibrium adsorption time is the time when adsorption capacity to Zn is not changing or in constant condition. Based on Fiqure 1, the time for adsorption of the used lubricant oil using activated coconut shell charcoal is 120 minutes. It is compared to 60 minutes using bentonite (Oduola and Okwonna, 2016).

Isotherm kinetics
A standard solution was prepared at various zinc concentrations, namely 0, ½, 1, 1 ½, and 2 ppm. Based on the plot, the relationship between concentrations and the adsorbance results in calibration curve. The calibration curve equation is y = 0.127x + 0.0101 with value of R 2 is 0.9949.
The equation becomes the target for calculating the sample absorbance that had been read using AAS. The relationship between the metal ion concentration after adsorption and the adsorption solute concentration consisting of zinc metal for Langmuir isotherm is presented in Figure 2 in the variation (a) solvent % and (b) particle size.  Table 3 and Table  4. Tables of 3 and 4 summarize all the constants and correlation coefficient of two isotherms used. In the Langmuir equation, Cµmak shows adsorption capacity level in the metal zinc. Based on Table 3, the more solvent percent is added, the more absorption capacity will increase. Based on Table 3 the adsorption capacity will increase with the smaller grain size. On the other hand, in the Freundlich isotherm, the relationship between metal ion concentration after adsorption and adsorption solute concentration on the Zn metal-containing solution for the Freundlich isotherm can be seen in Figure 3 in the variation solvent concentration (a) and particle size (b). Figure 3 (a) and Figure 3 (b), the k value, n in the variation percent of solvent and particle size can be seen in Table 4. Table 4 shows the Freundlich parameter. The plot of log Cm versus log Ce gives a straight line with slope 1/n. Parameter to illustrate adsorption capacity is k and n values. If the equilibrium adsorption Freundlich constant (k) is bigger, then adsorption capacity will be better. Table 4, the highest adsorption capacity in the equilibrium concentration 0.5% v/v is 5.35 g/g while in the smaller particle size, the adsorption capacity will decrease. 1/n is symbolized absorption capacity, the greater the adsorption capacity (1/n), the smaller the affinity of coconut shell charcoal to absorb the used oil.   This shows that the process of adsorption of the used oil using coconut shell charcoal is more suitable when approached with the Freundlich equilibrium model. Tables of 4 and 5 describe n constant between 2.71 to 2.98. However, the constant is in the range of 1.0-10.0, the constant shows that adsorption process better runs in the varioues particle sizes and solvent percent.

Adsorption Kinetics
The reaction kinetics is predicted using pseudo-first and second-order kinetics. The relationships between the changed concentration and the time are shown in Figures of 4 and 5. Figure 4 (a) shows that first-order kinetic data in the range of 0-120 minutes. The total metal Zn concentration increases to time. Figure 4 (b) describes that adsorption processs is effective in the interval 0 to 30 minutes while in the interval 60 to 120 minutes, desorption process is more dominant than adsorption because metal Zn concentration increases.   Based on Figure 5 (a), at the various solvent percentages using the second-order kinetic model in the range of 0-120 minutes, metal Zn concentration increases every time. The phenomena show that desorption is more dominant than adsorption process. On the other hand, Figure 5 (b) at the various particle sizes has the same phenomena as the solvent percent desorption, which is more dominant than adsorbtion process.
Pseudo-second-order kinetic reaction is more suitable for illustrating the adsorption process of zink because the value of correlation coefficients are in the range of 0.67-0.99 compared to pseudo-first-order kinetic, namely in the range of 3x10 -2 -0.47 (Table 6). While reaction rate constant in the range of 0.23-5.69 is bigger than that of the used oil adsorption using bentonite, expanded graphite, and microplastic.
Researches about used oil adsorption have been conducted by some researchers using adsorbent bentonite, expanded graphite, and microplastic. The adsorption constant in the pseudo-first and second-order can be seen in Table 7.

Conclusions
Activated coconut shell charcoal can be used as an adsorbent for oil lubricant consisting of metal ion using adsorption process. By using coconut shell active charcoal as a used oil adsorbent, it can increase the value of activated charcoal which has only a waste material without economic value. On the other hand, activated coconut shell charcoal can reduce the environmental pollution caused by the used lubricant oil. Isotherm Freundlich equilibrium model is more suitable to describe reaction equilibrium than isotherm Langmuir which is the maximum adsorption power of 5.35 g/g. The reaction constant rate in the first-order and secondorder reaction are 0.0207 min -1 and 0.24156 g/mg·min -1 , respectively. Li, T., Chen, C., Jin, Q., Zhao, J., Tang, X., Zhu, Y.