Microsoft Word - 164.docx CHEMICAL ENGINEERINGTRANSACTIONS VOL. 56, 2017 A publication of The Italian Association of Chemical Engineering Online at www.aidic.it/cet Guest Editors: JiříJaromírKlemeš, Peng Yen Liew, Wai Shin Ho, JengShiun Lim Copyright © 2017, AIDIC ServiziS.r.l., ISBN978-88-95608-47-1; ISSN 2283-9216 A Factorial Analysis Study on Removal of Mercury by Palm Oil Fuel Ash Adsorbent Imla Syafiqaha, Hafizuddin Wan Yussof*,a, Abdul Aziz Mohd Azoddeina, Shivananda Chandraseagara, Faizal Wan Ishakb aFaculty of Chemical Engineering & Natural Resources, Universiti Malaysia Pahang, 26300 Gambang, Pahang, Malaysia bFaculty of Industrial Science and Technology, Universiti Malaysia Pahang, 26300 Gambang Pahang, Malaysia hafizuddin@ump.edu.my Mercury is an extremely toxic pollutant that currently being emitted and distributed globally. The exposure of the mercury will mainly cause health effect and the exposure can be in term of dose, age of person exposed and duration of exposed. One of the adsorbent can remove mercury is an activated carbon from palm oil fuel ash (POFA). POFA was used as a low cost adsorbent. An adsorbent prepared from POFA) was successfully synthesised and used to remove mercury (Hg2+) from mercury model solution in a batch process. The effects of pH [A], contact time [B], initial concentration of the Hg2+ solution [C], adsorbent doses [D] and agitation speed [E] were screened by using response surface methodology (RSM). A half fraction two-level factorial analysis with five factors was selected for the experimental design to determine the best adsorption conditions for mercury removal. From this study, it was found the highest mercury removal was 98.03 % at adsorbent capacity 0.10 mg of Hg2+/g of the adsorbent with conditions of pH 2, contact time 4 h, initial Hg2+ concentration of 5 mg/L, adsorbent doses 0.25 g and agitation speed 200 rpm. The most important factor which gives the highest percentage contribution for mercury removal efficiency is contact time [B] with 22.76 % followed by initial mercury ion concentration [C] of 5.50 %. The model obtained in this present study is significant and had a maximum point which is likely to be the optimum point and possible for the optimisation process later. These results suggest that POFA has a potential used as an effective, low cost, and eco-friendly green adsorbent for the removal of mercury from waste water. 1. Introduction The presence of heavy metal ions in natural or industrial waste water and their hazardous impact has generated considerable concern in many years (Chowdhury et al., 2011). In Malaysia, waste water is contaminated by various substances and one of them is heavy metals which are hazardous to the environment. Mercury is one of the most toxic metals present in the environment. Mercury has attracted wide attention due to its high toxicity to human and other organisms (Zhang et al., 2016). An effective way should be developed for the selective monitoring and effective removal of mercury contamination in water or wastewater generated by both natural processes and human activities. According to Sharma et al. (2015) the removal of these heavy toxic metals by conventional methods generate a huge quantity of toxic chemical sludge and are not economical both in terms of operating and capital costs. Several researchers have developed a new alternative and low cost removal technique at laboratory scales which may have great potential for commercial applications. The commonly used methods for the removal of heavy metal ions from the wastewater include precipitation, coagulation or flocculation, ion exchange, reverse osmosis and electrochemical operation (Khoramzadeh et al., 2013). Among various mercury removal method currently in study, the adsorption process which offers as an effective and economic method for wastewater treatment (Sharma et al., 2015). In the adsorption process, a numerous number of adsorbents have been investigated to remove mercury. These include coconut pith char (Johari et al. 2016), silica (Samana et al., 2014), chitosan (Nasirimoghaddam and Sabbaghi, 2015) and palm shell (Ismaiel and Yusoff, 2013). This study aimed to find the best condition in the mercury removal efficiency. A half fraction two-level factorial DOI: 10.3303/CET1756251 Please cite this article as: Syafiqah I., Wan Yussof H., Azoddein A.A.M., Chandraseagar S., Wan Ishak F., 2017, A factorial analysis study on removal of mercury by palm oil fuel ash adsorbent, Chemical Engineering Transactions, 56, 1501-1506 DOI:10.3303/CET1756251 1501 analysis with five factors was studied for their effect to the mercury removal efficiency. This work is crucial for the optimisation stage later because each main effect had the tendency to affect one another. 2. Experimental procedures 2.1 Material preparation The palm oil fuel ash (POFA) was collected from palm oil mill boiler at Kilang Sawit Lepar, Kuantan. It was washed with deionized water for several times to remove foreign particles and oven dried overnight at 110 ˚C to remove moisture. It was sieved through 100 µm and stored in airtight container. Mercury stock solution was prepared at 1,000 mg/L. This stock solution was then diluted to 1 mg/L and 5 mg/L using deionised water. All the adsorption experiment was carried out at room temperature. The required quantity of Hg2+ solution was dissolved in ultrapure water. Then, a small amount of HNO3 was added to preserve the solution. The pH of the solution was adjusted using 0.1 M of HCl and 0.1 M NaOH. The mixture of the sample is shaken by using incubator shaker at the constant room temperature. The suspension is filtered using a vacuum pump and the filtrates are analysed using the direct mercury analyser Milestone DMA-80 (ATS Scientific, USA). 2.2 Adsorbent characterisation Characterisation of POFA was done by EVO 50 SEM (Carl Zeiss, Germany) for raw POFA and after mercury sorption. Next, FTIR of raw POFA and after mercury sorption were recorded on a Spectrum 100 FTIR spectrometer (Perkin Elmer, USA). The thermal stability was determined using TGA with a TA Instruments model Q-500 (USA) on POFA. 2.3 Two-level factorial analysis experimental setup The experimental design for factorial analysis was performed using Design Expert 7.0.0 (Stat-Ease Inc., USA) software. The effects of five independent variables as shown in Table 1 were screened by using response surface methodology (RSM). The condition ranges chosen were based on the other researcher’s previous work (Rio and Delebarre, 2003) and from (Chowdhury et al., 2010). The factors were constructed in factorial designs of 25-1 to screen their effect on the response of mercury removal efficiency. All experiments consist of 16 run listed in Table 2. Table 1: Low and high values for each parameter Factor Units Low value (-1) High value (+1) A: pH 2 6 B: Contact time h 1 4 C: Initial concentration of the Hg2+ solution mg/L 1 5 D: Adsorbent doses g 0.10 0.25 E: Agitation speed rpm 100 200 Several factors were tested earlier and obtained from literature study. To identify the factors that affects the mercury adsorption. Five independent variables which are pH [A], adsorption time [B], initial concentration of the Hg2+ solution [C], agitation speed [D] and adsorbent doses [E] were selected for factorial designs of 25-1 in order to screen their effect on the response of mercury removal efficiency. Factorial design was applied to screen the mercury removal from mercury model solution. The ranges of the individual factors were chosen from the preliminary test. Statistical analysis was performed using Design Expert software (Stat-Ease, Inc.). Mercury removal efficiency was calculated using the following Eq(1). R= × 100 (1) where R is the percent removal of mercury, Co and Ce are the initial and residual concentration (mg/L) of mercury ion. The amount of mercury adsorbed by POFA was calculated from the difference between the initial Hg2+ concentration (Co) and the equilibrium concentration (Ce) was calculated and used to determine the adsorptive capacity (qe) as follows in Eq. (2). qe=(Co–Ce)V/M (2) where V is the total volume of mercury (II) solution (mL), M is the mass of adsorbent used (g), Co is the initial concentration of mercury ion solution (mg/L), and Ce is the residual Hg 2+ concentration(mg/L). 1502 3. Results and discussions 3.1 Material preparation The Fourier Transform Infrared Spectroscopy (FTIR) technique is an important tool to identify the characteristic functional groups, which are instrumental in adsorption of aromatic compounds (Daffalla et al., 2010). The change in intensity or shift in peak position gives information about interaction adsorbate and adsorbent. As shown in Figure 1, the adsorption peak at 3,356 cm-1 indicates the existence of phenolic and aliphatic OH group and the broad band was attributed to the presence in POFA after sorption with mercury similar to Kuncoro and Fahmi, (2013). After contacting with Hg, there is a change in intensity of spectral, then transmittance intensity is lower. The small band at 1,630 cm-1 indicates the presence of alkenes and primary amines group. The peaks at 1,020 cm-1 for the POFA after sorption with mercury assigned for Si-O-Si asymmetric stretching vibration of SO4 -2 group. POFA sample at this band has highest SiO2 content (Kipcak, et al., 2012). After adsorption of Hg2+, the interaction between the POFA surface and Hg2+ was evidenced by a shift in the O-H and N-H bands from 3,356.40 cm-1 and 1,081.65 cm-1 to 3,312.56 cm-1 and 1,019.75 cm-1. These results provide strong evidence for the adsorption of Hg2+ onto the POFA surface. According to Esmaeili et al. (2012), the presence of carboxylic acid and sulphate groups can act as binding sites for metals. Figure 1: FT-IR spectra of the raw POFA and after adsorption with mercury In this thermal gravimetric analysis (TGA) test, the mass loss was measured while the specimens were gradually exposed to increasing temperatures. The thermal stability of the POFA form was studied by thermal TGA in the conditions conducted at 24.58 °C which is almost at room temperature up to 895.53 °C for the process to complete. The results for POFA are displayed in Figure 2 below. It can be observed that there were three main areas of weight loss. The first was showed sharp decrease weight before 250 °C as recorded in TGA thermogram is attributed to the loss of evaporable water in the POFA as observed in Bakri et al. (2012). The weight loss which marked the major decomposition of the POFA took place between 200 °C and 650 °C, which saw a 2.53 % weight lost, caused by the decomposition of CaCO3 and the burning of residual coal of the ash (Santos, et al., 2014). The third weight loss, little change occurred in the percentage of weight remaining beyond 650 °C. Above 900 °C, the ash had thermogravimetric stability andthe average total percentage of weight remaining was 95.07 %. Similar results reported by Santos et al. (2014) where the small weight loss is associated with high ash content found in POFA. Scanning electron microscope (SEM) presented in Figure 3 indicates the surface morphology of POFA. The image was captured at 1,000× magnification. From the SEM image, it can be seen that the POFA mainly consists of alumina silicate spherical particles with smooth outer surfaces, also known as cenospheres (Nyale et al., 2013). It has been seen clearly that the adsorbents are highly porous in nature which is a desirable characteristic of any adsorbent. 3.2 Factors affecting mercury removal efficiency The highest percentage of mercury removal was obtained at 98.03 % with the adsorption conditions at 100 rpm, pH 2, 4 h, 5 mg/L of initial concentration of the Hg2+ solution, and 0.25 g of adsorbent dosage as shown in Table 2. The relative size of effects was visually demonstrated by Pareto chart in Figure 4, where the bar length is proportional to the absolute value of estimated effect. For main effects, an effect is said to be positive when an increase to its high level will cause an increase in the response while the negative effect is when an increase in its high level will result in a decrease in the response. For interactions, the positive effect is when 1503 both factors were a chance to the same level (low or high), the response will increase. The negative effect is when both factors were change to the opposite level (one at its low and the other at its high), the response will increase (Martendal et al., 2007). Effects of t-value limit (black line) are considered statistically significant at 95 % confidence level while effects below t-value limit are not likely to be significant. Effect above Bonferroni’s corrected t-value limit (red line) is almost certainly significant (Anderson et al., 2009). A quick analysis was performed on the selected effects using Pareto chart to statistically check for significance of the selected effects at 95 % confidence level. All the selected effects (A, B, C, D, E, AC, AE, BE, CD, CE, and DE) shown to be significant at both t-value limit and Bonferroni’s corrected t-value limit. The fitted model for the factorial analysis in coded form was shown in Eq(3). Figure 2: TGA results on POFA Figure 3: SEM image of POFA Figure 4: Pareto chart Mercury removal efficiency = 71.62 - 0.97*A + 9.69*B + 4.76*C – 3.83*D – 1.73*E + 8.67*A*C + 7.85*A*E + 2.97*B*E + 7.31*C*D – 7.67*C*E – 3.53*D*E (3) Based on the coded factors, the coefficients of A to E are small compared to the interception. This gives an indicator of the existent of the design plateu as the other coefficient for the factors are lower than the interception at 71.62. This plateau showed that the design had a maximum point which is likely to be the optimum point. 3.3 ANOVA The analysis of variance (ANOVA) test results are as shown in Table 3. As can be seen, the model obtained was significant with p-value 0.0047. Coefficient of determination (R2), adjusted R2 and predicted R2 values were used to evaluate the fitness of the model. Adjusted R2 is a modification of R2, which adjusts for the number of explanatory terms in a model relative to the number of data points (Anderson and Whitcomb., 2005). The R2 value of 0.9834 (R2adj = 0.9378) indicates that the predicted values obtained from the model is a good fit of the experimental data (Dong and Sartaj, 2016). 1504 Table 2: Experimental design of factor screening process using 25-1 factorial design with their response Std. Order Factors Response pH Contact time (h) Initial concentration of the Hg2+ solution (mg/L) Adsorbent doses (g) Agitation speed (rpm) Mercury removal efficiency (%) 1 2 1 1 0.10 200 74.67 2 6 1 1 0.10 100 42.67 3 2 4 1 0.10 100 97.33 4 6 4 1 0.10 200 97.33 5 2 1 5 0.10 100 73.67 6 6 1 5 0.10 200 71.67 7 2 4 5 0.10 200 65.33 8 6 4 5 0.10 100 80.93 9 2 1 1 0.25 100 59.67 10 6 1 1 0.25 200 44.87 11 2 4 1 0.25 200 74.33 12 6 4 1 0.25 100 44.00 13 2 1 5 0.25 200 37.73 14 6 1 5 0.25 100 90.53 15 2 4 5 0.25 100 98.03 16 6 4 5 0.25 200 93.20 Table 3: Analysis of variance (ANOVA) Source Sum of Squares Degree of freedom Mean Square F-Value p-value Model 6,486.72 11 589.70 21.55 0.0047 significant A-pH 15.07 1 15.07 0.55 0.4991 B-Contact time 1,500.98 1 1,500.98 54.86 0.0018 C-Initial Hg ion concentration 362.81 1 362.81 13.26 0.0219 D-Adsorbent doses 234.63 1 234.63 8.58 0.0429 E-Agitation speed 47.85 1 47.85 1.75 0.2565 AC 1,202.88 1 1,202.88 43.97 0.0027 AE 985.18 1 985.18 36.01 0.0039 BE 141.19 1 141.19 5.16 0.0856 CD 855.42 1 855.42 31.27 0.0050 CE 941.42 1 941.42 34.41 0.0042 DE 199.30 1 199.30 7.29 0.0541 Residual 109.43 4 27.36 Cor Total 6,596.15 15 C.V. = 7.30 %; R2 = 0.9834; Adjusted R2 = 0.9378; Adeq. Precision = 13.450. In percentage of contribution, the most important factor which gives the highest percentage contribution for mercury removal efficiency is contact time [B] with 22.76 % followed by initial mercury ion concentration [C] of 5.50 %. pH [A] shows the lowest contribution which is 0.23 %. Other variables that less contributed to the effect of the mercury removal efficiency were agitation speed [E] and adsorbent doses [D] by having a value of 0.73 % and 3.56 %. 4. Conclusion The best condition for the mercury removal efficiency using POFA was obtained at agitation speed 200 rpm, pH 2, 4 h, initial Hg2+ concentration of 5 mg/L, and adsorbent doses of 0.25 g with 98.03 % mercury removal. The most important factor which gave the highest percentage contribution for mercury removal efficiency is contact time [B] with 22.76 % followed by initial mercury ion concentration [C] of 5.50 %. The model obtained in this present study is significant and had a maximum point which is likely to be the optimum point and 1505 possible for the optimisation process later. POFA was performed excellently well as adsorbent in mercury removal at high contact time (4 h) and initial concentration of the Hg2+ solution (5 mg/L). It can be concluded that POFA can potentially be a suitable adsorbent for mercury removal from waste water as it can be obtained at low-cost. Acknowledgments The authors gratefully acknowledge financial support from the UMP grant RDU140352 and Faculty of Chemical and Natural Resources Engineering, University Malaysia Pahang (UMP) that made this work possible. Reference Anderson M.J., Whitcomb P.J., 2005, RSM Simplified: Optimizing Processes Using Response Surface Methods for Design of Experiments, Productivity Press, New York, USA. Anderson M.J., Whitcomb P.J., Kraber S.L., Adams W., 2009, Stat-Ease Handbook for Experimenters, Stat- Ease, Inc, Minneapolis, USA. Bakri A.M., Kamarudin H., Bnhussain M., Nizar I.K., Rafiza A.R., Zarina A.Y., 2012, The processing, characterization, and properties of fly ash based geopolymer concrete, Rev. Advance Material Science 30, 90-97. Chowdhury Z., Zain S., Rashid A.A., 2011, Equilibrium isotherm modeling, kinetics and thermodynamics study for removal of lead from waste water, E- J. Chem. 8, 333-339. Daffalla S.B., Mukhtar H., Shaharun A.M., 2010, Characterization of adsorbent developed from rice husk: effect of surface functional group on phenol adsorption, J. Appl. Sci. 10, 1060-1067. Dong S., Sartaj M., 2016, Statistical analysis and optimization of ammonia removal from landfill leachate by sequantial microwave/aeration process using factorial design and response surface methodology, J. Environ. Chem. Eng. 4, 100-108. Esmaeili A., Saremnia B., Kalantari M., 2012, Removal of mercury (ii) from aqueous solutions by biosorption on the biomass of Sargassum Glaucescens and Gracilaria Corticata, Arabian J. Chem. 8, 506-511. Ismaiel A.A., Yusoff M.K., 2013, Palm shell activated carbon impregnated with task specific ionic-liquids as a novel adsorbent for the removal of mercury from contaminated water, Chem. Eng. J. 225, 306-314. Johari K., Saman N., Tien S.S., Chin C.S., Kong H., Mat H., 2016, Removal of elemental mercury by coconut pith char adsorbents, Procedia Eng. 148, 1357-1362. Khoramzadeh E., Nasernejad B., Halladj R., 2013, Mercury biosorption from aqueous solutions by sugarcane bagasse, J. Taiwan Inst. Chem. Eng. 44, 266-269. Kipcak A., Acarali N.B., Kolemen S., Tugrul N., Derun E.M., Piskin A.S., 2012, Study on characterization of tuncbilek fly ash, international science index, Mater. and Metallurgical Eng. 6, 615-618. Kuncoro E.P., Fahmi, A.M., 2013, Removal of Hg and Pb in aqueous solution using coal fly ash, Procedia Earth Planet 6, 377-382. Martendal E., Budziak D., Carasek E., 2007, Application of fractional factorial experimental and Box-Behnken designs, J. Chromatogr. 1148, 131-136. Nasirimoghaddam S., Sabbaghi S.Z., 2015, Chitosan coated magnetic nanoparticles as nano-adsorbent for efficient removal of mercury contents from industrial aqueous and oily samples, J. Ind. Eng. Chem. 27, 79- 87. Nyale S.M., Babajide O.O., Birch G D., Böke N., Petrik L.F., 2013, Synthesis and characterization of coal fly ash-based foamed geopolymer, Procedia Environ. Sci. 18, 722-730. Rio S., Delebarre A., 2003, Removal of mercury in aqueous solution by fluidized bed plant fly ash, Fuel 82, 153-159. Samana N., Joharia K., Mat H., 2014, Synthesis and characterization of sulfur-functionalized silica materials towards developing adsorbents for mercury removal from aqueous solutions, Microporous Mesoporous Mater. 194, 38-45. Santos R.P., Martins J., Gadelha C., Cavada B., Albertini A.V., Arruda, F., Vasconcelos M., Teixeira E., Alves F., Filho J.L., Freire V., 2014, Coal fly ash ceramics: preparation, characterization, and use in the hydrolysis of sucrose, Sci. World J. 2014, 154651. Sharma A., Sharma A., Arya R.K., 2015, Removal of mercury (ii) from aqueous solution: a review of recent work, Sep. Sci. Technol. 50 (9), 1310–1320. Zhang A., Xing W., Zhang Z., Meng F., Liu Z., Xiang J., Sun L., 2016, Promotional effect of SO2 on CeO2-TiO2 material for elemental mercury removal at low temperature, Atmospheric Pollution Research 7, 895-902. 1506