CHEMICAL ENGINEERING TRANSACTIONS 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 Congo Red Removal by HNO3-Modified Resorcinol- Formaldehyde Carbon Gels Shu Hui Tanga,b, Muhammad Abbas Ahmad Zaini*,a,b aCentre of Lipids Engineering & Applied Research (CLEAR), Ibnu Sina ISIR, Universiti Teknologi Malaysia, 81310 UTM Johor Bahru, Malaysia. bFaculty of Chemical & Energy Engineering, Universiti Teknologi Malaysia, 81310 UTM Johor Bahru, Malaysia. abbas@cheme.utm.my Resorcinol-formaldehyde (RF) carbon gel has gained much attention in various commercial applications including electro-catalysis, ion exchange resin and hydrogen and electrical energy storage owing to its highly sensitive nanostructure. In this work, the resorcinol-formaldehyde carbon gels were synthesised via a sol-gel method from polymerisation of resorcinol and formaldehyde with sodium carbonate (Na2CO3) as a catalyst. Different resorcinol to catalyst (R/C) ratios (mol/mol) of 1,000 (RC3) and 2,000 (RC4) were used, and the effect of nitric acid (HNO3) modification was investigated. The oxidised RC3 and RC4 were denoted as RC7 and RC8,. It was found that a further increase in R/C ratio above 1,000 caused a decline in surface area, pore volume and pore size. RC7 gave the highest surface area of 711.49 m2/g with pore size of 7.68 nm. The HNO3 modification step increased the surface area of RC7, but slightly decreased the surface area of RC8 probably due to crystallisation of the gel to graphite. Adsorptive study was carried out using congo red (CR) as an adsorbate. The carbon gels with higher surface area, pore volume and pore size exhibit a better CR adsorption performance. The HNO3-modified resorcinol-formaldehyde carbon gels gave a better performance for the removal of congo red. The kinetics data of congo red adsorption could be explained by the pseudo- second-order kinetic model. The maximum adsorption of congo red on resorcinol-formaldehyde carbon gel is comparable to other adsorbents and activated carbons. 1. Introduction The surge in water pollution due to dye wastewater is a serious cause for concern in the current times. A rich diversity of dyes is widely used in various industries including textiles and apparel, paper, food and pharmaceutical to impart the desired hue. The most common textile industrial dyes are basic, acid and reactive dyes (Salleh et al., 2011). The effusion of dye wastewater into downstream can bring about adverse effect to living organisms in aquatic ecosystem (Tang and Zaini, 2015a). Thus, wastewater treatment prior to release is necessary to ensure that the water quality complies with the regulation standards. Adsorption using porous material is one of the most extensively employed dye removal technique owing to its simple, economical and efficient operation, and no sludge is generated. Different adsorbents have been studied over the years, like agricultural waste and sewage sludge (Chan et al., 2016). These adsorbents are mainly in granular or powder form, and their mesopores or micropores density can hardly be manipulated to match specific applications. Small particles are prone to rapid pressure drop (hydraulic resistance) during large scale continuous column adsorption. This can lead to deteriorating column performance and exorbitant operating cost. Carbon gel is a promising porous carbon substitute with a highly flexible shape, pore texture and nanostructure. It can be moulded into a particular shape to elicit the desired type of column packing, so that the surface contact between solute and sorbent can be maximised, and the hydraulic pressure inside the column can be averted. Synthesis of carbon gel comprises of three main stages - (i) preparation of sol-gel mixture (gelation, aging and solvent exchange), (ii) drying and (iii) carbonisation or activation (Al-Muhtaseb and Ritter, 2003). In the first stage, condensation-polymerisation (gelation) of resorcinol and formaldehyde takes place in the presence of an alkaline catalyst (Na2CO3) and a solvent (water or acetone). Aging step is DOI: 10.3303/CET1756140 Please cite this article as: Tang S.H., Zaini M.A.A., 2017, Congo red removal by hno3-modified resorcinol-formaldehyde carbon gels, Chemical Engineering Transactions, 56, 835-840 DOI:10.3303/CET1756140 835 carried out in order to enhance the crosslinking density by boosting the condensation reaction of the hydroxymethyl groups, whereas solvent exchange involves the removal of old solvent which acts as a reaction medium during gelation. The hydrogel or lyogel is then dried via subcritical drying, supercritical drying or freeze drying, which subsequently produces xerogel, aerogel and cryogels. The organic gels are converted into resorcinol-formaldehyde (RF) carbon gels after pyrolysis or activation in the final phase. The density of mesopores and micropores of carbon gel can be tailored independently by adjusting the process condition to control the voids formed during aggregation of nanoparticles (Mukai, 2012). In this work, resorcinol-formaldehyde (RF) carbon gels were synthesised via sol-gel method and then modified by oxidation using nitric acid. An anionic azo dye, congo red (CR) was selected as a model dye pollutant to evaluate the feasibility of oxidised and unoxidised carbon gel as adsorbents for the CR removal. The RF carbon gels were characterised and the batch adsorption outcomes were analysed. 2. Materials and methods 2.1 Material Formaldehyde (HCHO, MW = 30.03 g/mol, 37 wt% in water), resorcinol (C6H6O2, MW = 110.11 g/mol, assay 99 %), sodium carbonate (Na2CO3, MW = 105.99 g/mol, assay 99.5 %), tert-butyl alcohol ((CH3)3COH, MW = 74.12 g/mol, assay 99 %), nitric acid (HNO3, MW = 63.01, 65 %) and congo red (C32H22N6Na2O6S2; MW = 696.7 g/mol) were used in the synthesis of RF carbon gels and adsorption studies. The chemicals and CR dye were obtained from Wako Pure Chemical Industries Ltd and R&M Chemicals, and are of analytical reagents grade. 2.2 Synthesis and characterisation of RF carbon gel The sol mixture was initially prepared using resorcinol, formaldehyde, water (solvent) and sodium carbonate (alkaline catalyst). The resorcinol/catalyst (R/C) ratios of 1,000 and 2,000 mol/mol were selected and denoted as RC3 and RC4. The resorcinol/water (R/W) and resorcinol/formaldehyde (R/F) ratios were fixed at 0.5 g/mlLand 0.5 mol/mol. Approximately 25 g of resorcinol, 0.241 g of Na2CO3 and 29.7 g of water were mixed in a disposable cup, followed by addition of 36.85 g of formaldehyde. The sol mixture was stirred and poured into a mould to allow gelation of RF sol solution at 35 °C for 2 d (Al-Muhtaseb and Ritter, 2003). Next, the aging step was carried out by heating the cured gel at 60 °C for 3 d. After that, the RF hydrogel was placed into a capped bottle filled with tert-butyl alcohol (TBA) and retained at 50 °C for 3 d to eliminate the surplus water by solvent exchange process. The old TBA was exchanged with a fresh one twice a day. The RF organic gel was then dried at 110 °C for 2 d and crushed to obtain uniform particle size of 1 mm. Lastly, the organic gel was carbonised at 600 °C for 3 h to produce RF carbon gel and was labelled as RC3 (R/C = 1,000 mol/mol) and RC4 (R/C = 2,000 mol/mol). The additional oxidation step was performed by soaking the carbon gel in HNO3 (65 - 66 %) at room temperature for 3 h, followed by heating at 60 °C for 5 h. The oxidised RC3 and RC4 were labelled as RC7 and RC8. The resultant materials were washed with distilled until a constant pH was elicited. The textural properties of the samples were examined using a surface area analyser (Micrometrics PulseChemiSorb 2705, USA) at liquid nitrogen temperature of 77 K. 2.3 Batch adsorption studies Congo red (CR) adsorption tests were executed in batch mode by weighing 0.03 g of carbon gel and adding it to 20 mL of CR solution with different initial concentrations (2 - 50 mg/L). The original solution pH was maintained and monitored using pH meter (HANNA HI8424) during the adsorption experiments. After 5 d (120 h), the dye solution was extracted and separated using a centrifuge. The clear supernatant was analysed for CR concentration on a visible spectrometer (Dynamica Halo VIS-10) at a maximum wavelength of 510 nm. The equilibrium adsorption capacity (qe) was calculated as Eq(1), qe = (Co − Ce) m V (1) where qe (mg/g) is the amount of adsorbate adsorbed per unit mass of adsorbent at equilibrium, Co (mg/L) is the initial concentration of dye solution, Ce (mg/L) is the equilibrium concentration of dye solution, m (g) is the mass of carbon gel and V (L) is the volume of dye solution. 3. Results and discussion 3.1 Physical characteristics of RF carbon gels The textural properties of RF carbon gels are summarised in Table 1. It is discovered that a higher R/C ratio gives a better yield owing to the greater quantity of resorcinol (reactant) compared to Na2CO3 (catalyst). 836 During carbonisation, almost all hydrogen groups and residual oxides are eliminated, resulting in a highly dense pure carbon nanostructure. Table 1: Physical properties of RF carbon gels. The mass reduction of RF carbon gels is around 50 % after pyrolysis at 600 °C and these findings (Table 1) are consistent with the study of Al-Muhtaseb and Ritter (2003). The diminished mass in the gel is closely attributed to the burnout of the organic matters, which gives rise to the formation of new voids or micropores and mesopores, thus increasing the surface area of carbon gel. In this work, the average pore diameter of the synthesised RF carbon gels is in the range of 3.56 to 7.68 nm, suggesting a higher concentration of mesopores (2 - 50 nm) or mesoporosity (59.2 - 75.1 %). Table 1 shows that there is no notable change in the yield and textural properties of carbon gel after oxidation. The surface area, pore volume and mesoporosity and average pore diameter of RC7 increased slightly, while that of RC8 decreased slightly except for mesoporosity and average pore diameter which remained almost the same. Al-Muhtaseb and Ritter (2003) also reported similar outcomes. 3.2 Adsorption isotherm Adsorption isotherm is indispensable in understanding the interaction between the adsorbed dye on carbon gel and the remaining dye in the bulk solution. In the present work, the equilibrium data are fitted into two isotherm models - Langmuir and Freundlich to determine the adsorption behaviour for efficient design of adsorption process. Adsorption equilibrium is attained when the dye adsorption rate is equivalent to the desorption rate (Tang and Zaini, 2015b). The Langmuir equation is given as Eq(2), qe = qmbCe (1 + bCe) (2) where b (L/mg) is the Langmuir isotherm constant and qm (mg/g) is the maximum monolayer coverage. Table 2 revealed that the Langmuir isotherm is a better fit for the equilibrium data and this indicates that the CR adsorption is a monolayer adsorption onto homogeneous surface of RF carbon gel. The maximum adsorption capacity (qm) was found to be in the range of 2.25 to 10.58 mg/g. The oxidised carbon gels have a higher qm than the unoxidised carbon gels owing to greater mesoporosity and average pore diameter. The separation factor (RL) is a dimensionless constant of Langmuir isotherm and is expressed as RL = 1/(1 + bCo). The affinity between adsorbate and adsorbent can be determined through RL: unfavourable (RL > 1), linear (RL = 1), favourable (0 < RL < 1) and irreversible (RL = 0) (Chan et al., 2016). As shown in Figure 1(b), the RL values are in the range of 0 to 1 and decreased with increasing Co, indicating a favourable adsorption process. The Freundlich isotherm is used to describe multilayer adsorption over a heterogeneous surface of adsorbents. The Freundlich equation is given as, qe = KF(Ce) 1 n (3) where KF ((mg/g)(L/mg) 1/n) and n are the Freundlich isotherm constants. The values of n and KF are listed in Table 2. The lower R2 values showed that the Freundlich isotherm is a poor fit. The adsorption of CR molecules onto carbon gel is unlikely a multilayer. Higher KF values demonstrate greater adsorption capacity or affinity towards CR anions. The constant n is the heterogeneous factor: linear adsorption (n = 1), cooperative adsorption (n < 1) and ideal Langmuir isotherm (n > 1) (Prahas et al., 2008). All the 1/n values are below 1, denoting an ideal Langmuir adsorption, and this further confirms the good applicability of Langmuir isotherm. Figure 1 illustrates the CR equilibrium adsorption onto carbon gels. The equilibrium removal increases with increasing CR concentration until a maximum adsorption capacity is achieved (horizontal plane). When there Oxidised Unoxidised RC3 RC4 RC7 RC8 R/C ratio (mol/mol) 1,000 2,000 1,000 2,000 Yield (%) 42.6 53.5 42.6 53.5 BET surface area (m2/g) 639 631 711 586 Total pore volume (cm3/g) 0.632 0.536 0.817 0.500 Micropore volume (cm3/g) 0.191 0.219 0.203 0.203 Mesopore volume (cm3/g) 0.441 0.317 0.614 0.297 Mesoporosity (%) 69.8 59.2 75.1 59.3 Average pore diameter (nm) 6.89 3.56 7.68 3.57 837 is a contrast between the amount of adsorbate in bulk solution and surface of adsorbent, a concentration gradient is created and it acts as a driving force in adsorption process. The saturation point is achieved at the horizontal level, whereby almost all the active sites of RF carbon gels are occupied by dye molecules. Table 2: Langmuir and Freundlich isotherm constants of CR adsorption onto carbon gels. Carbon gels Langmuir Freundlich qm (mg/g) b (L/mg) R 2 N KF ((mg/g)(L/mg) 1/n) R2 RC3 2.25 25.6 0.995 10.0 1.78 0.751 RC4 2.51 6.34 0.975 8.09 2.41 0.177 RC7 10.6 0.72 0.914 1.37 2.48 0.863 RC8 4.36 1.00 0.651 1.53 2.30 0.523 Figure 1: (a) Equilibrium adsorption of CR predicted by Langmuir model (b) Relationship between RL and Co (adsorbent dosage = 0.03 g, solution volume = 20 mL, CR initial concentration = 2 to 50 mg/L, contact time = 120 h) Figure 2: Relationships between maximum capacity with (a) BET surface area; (b) mesoporosity The textural properties of carbon gel play a significant role in the dye removal performance. Figure 2 shows the relationships between the maximum adsorption capacity, and BET surface area and mesoporosity. The adsorption performance increased significantly with an increase in mesoporosity and BET surface area. Higher surface area and mesoporosity provide more number of active sites available for adsorption of CR anions. The carbon gels with greater mesoporosity or mesopore (2 - 50 nm) volume are more favourable to 0 2 4 6 8 10 12 0 10 20 30 E qu ili br iu m a ds or pt io n, q e ( m g/ g) Equilibrium concentration, Ce (mg/L) RC3 RC4 RC7 RC8 0 0.1 0.2 0.3 0.4 0.5 0 20 40 S ep ar at io n fa ct or , R L ( m g/ g) Initial concentration, Co (mg/L) 0 2 4 6 8 10 12 0 200 400 600 800 M ax im um a ds or pt io n ca pa ci ty , q e ( m g/ g) BET surface area (m2/g) 0 2 4 6 8 10 12 0 20 40 60 80 M ax im um a ds or pt io n ca pa ci ty , q e ( m g/ g) Mesoporosity (%) (a) (b) (a) (b) 838 accommodate CR molecules with a single dimension of 2.62 nm (width), 0.74 nm (depth) and 0.43 nm (thickness) (Cotoruelo et al., 2010). An obstruction in CR adsorption process may occur for microporous adsorbents because the width of CR molecule is greater than 2 nm. 3.3 Adsorption kinetics To further comprehend the sorption mechanism of CR onto carbon gels, the kinetics data were simulated with the pseudo-second-order kinetic model, which can be expressed as, qe = k2qe 2t 1 + k2qet (4) where t (h) is the period of adsorption, qt (mg/g) is the amount of dye adsorbed at time t and k2 (g/mg.h) is the rate constant for pseudo-second-order sorption. The calculated qe and k2 values are tabulated in Table 3. Higher k2 values corresponds to greater affinity of carbon gels towards CR at certain concentration. RC7 has the highest k2 value at Co of 10 mg/L. Table 3 also shows that the experimental qe are close to the theoretical qe (high R 2), indicating that chemisorption may govern the adsorption process whereby valence forces through exchange or sharing of electron between adsorbent and adsorbate predominates (Prahas et al., 2008). The effect of contact time on the uptake of CR by carbon gels was studied for 120 hat concentrations of 5 and 10 mg/L. From Figure 3, it can be observed that the kinetics behaviour of RC7 and RC8 consists of three phases - a rapid initial adsorption over 10 h, followed by a significantly slower adsorption (40 h) and finally a gradual equilibrium contact time. Table 3: Parameters of pseudo-second-order kinetic model for CR adsorption. Carbon gels qe (mg/g) Pseudo-second order Cal. qe (mg/g) k2 (g/mg.h) R 2 RC3 (5 mg/L) 3.26 2.94 0.0153 0.837 RC4 (5 mg/L) 2.99 2.98 0.0354 0.975 RC7 (5 mg/L) 3.22 3.19 0.180 0.999 RC8 (5 mg/L) 3.37 3.01 0.0670 0.966 RC3 (10 mg/L) 4.04 3.58 0.0102 0.801 RC4 (10 mg/L) 4.57 3.93 0.0142 0.690 RC7 (10 mg/L) 4.98 4.73 0.379 0.993 RC8 (10 mg/L) 4.73 4.73 0.104 0.999 Figure 3: Effect of contact time on CR adsorption (m = 0.03 g, t = 120 h, Co = 5 and 10 mg/L) 3.4 Comparison of CR removal with various adsorbents The maximum monolayer adsorption (qm) of CR in this study was compared with other adsorbents as summarised in Table 4. It can be seen that the oxidised RF carbon gel (RC7) portrayed a better maximum monolayer adsorption of CR (10.58 mg/g) than majority of the previously developed adsorbents. This shows that the oxidised RF carbon gel is a potential adsorbent for the removal of congo red and other anionic dyes. 0 1 2 3 4 5 6 0 20 40 60 80 100 120 140 A ds or pt io n at ti m e t, q t (m g/ g) Time, t (h) RC7 (5 mg/L) RC8 (5 mg/L) RC7 (10 mg/L) RC8 (10 mg/L) 839 Table 4: CR removal using various adsorbents Adsorbent Surface area (m2/g) pH qm (mg/g) Reference Activated carbon from apricot stone 88.1 - 32.9 Abbas and Trari (2015) Pineapple plant stem 2.39 - 12.0 Chan et al. (2016) RF carbon gel 711 5.40 10.6 This study Acid-activated red mud - - 7.08 Salleh et al. (2011) Benzyltrimethylammonium chloride (BTMA) bentonite - - 6.58 Fosso-Kankeu et al. (2016) Tris(hydroxymethyl)aminomethane (THMA) bentonite - - 2.82 Fosso-Kankeu et al. (2016) Kaolin 10.6 - 5.94 Zenasni et al. (2014) Eichhorniacrassipes root - - 1.58 Wanyonyi et al. (2014) Gold nanoparticle-coated AC - 6.50 0.50 Pal and Deb (2014) 4. Conclusion In the present work, RF carbon gels have been successfully synthesised via a sol-gel method, and oxidised with HNO3. The oxidised and unoxidised carbon gels were characterised for textural properties, and applied for the removal of CR from aqueous solution. The effects of initial concentration and contact time on CR adsorption were discussed. Langmuir isotherm provided a good fit for the equilibrium data, indicating a monolayer adsorption. RC7 possesses the highest qm of 10.58 mg/g, in which the oxidised carbon gels perform better than the unoxidised ones. The kinetics data fitted well to the pseudo-second-order kinetic model, suggesting that chemisorption may predominate. The oxidised RF carbon gel is a promising adsorbent for the CR adsorption process. Acknowledgments This work was supported in part by UTM-Research University Grant (Tier 1) No. 14H19. Reference Abbas M., Trari M., 2015, Kinetic, equilibrium and thermodynamic study on the removal of congo red from aqueous solutions by adsorption onto apricot stone, Process Saf. Environ. 98, 424-436. Al-Muhtaseb S.A., Ritter J.A., 2003, Preparation and properties of resorcinol –formaldehyde organic and carbon gels, Adv. Mater. 15 (2), 101-114. Chan S.L., Tan Y.P., Abdullah A.H., Ong S.T., 2016, Equilibrium, kinetic and thermodynamic studies of a new potential biosorbent for the removal of basic blue 3 and congo red dyes: pineapple (Ananascomosus) plant stem, J. Taiwan Inst. Chem. E. 61, 306-315. Cotoruelo L.M., Marqués M.D., Díaz F.J., Rodríguez-Mirasol J., Rodríguez J.J., Cordero T., 2010, Equilibrium and kinetic study of congo red adsorption onto lignin-based activated carbons, Transport Porous Med. 83 (3), 573-590. Fosso-Kankeu E., Waanders F., Fourie C.L., 2016, Adsorption of Congo Red by surfactant-impregnated bentonite clay, Desalin. Water Treat 57 (57), 27663-27671. Mukai S.R., 2012, Controlling the morphology of carbon gels, Boletin del Grupo Espanol del Carbon 26, 8-11. Pal J., Deb M.K., 2014, Efficient adsorption of congo red dye from aqueous solution using green synthesized coinage nanoparticles coated activated carbon beads, Appl. Nanosci. 4 (8), 967-978. Prahas D., Kartika Y., Indraswati N., Ismadji S., 2008, Activated carbon from jackfruit peel waste by H3PO4 chemical activation: pore structure and surface chemistry characterization, Chem. Eng. J. 140, 32-42. Salleh M.A.M., Mahmoud D.K., Karim W.A.W.A., Idris A., 2011, Cationic and anionic dye adsorption by agricultural solid wastes: a comprehensive review, Desalination 280 (1), 1-13. Tang S.H., Zaini M.A.A., 2015, Isotherm studies of methylene blue adsorption onto potassium salts-modified textile sludge, Jurnal Teknologi 74 (7), 57-63. Tang S.H., Zaini M.A.A., 2015, Chapter 2 - Dyes – Classification and Effective Removal Techniques, Ed. Taylor J.C., Advances in Chemistry Research, 30, Nova Science Publishers, New York, USA. Wanyonyi W.C., Onyari J.M., Shiundu P.M., 2014, Adsorption of congo red dye from aqueous solutions using roots of Eichhorniacrassipes: kinetic and equilibrium studies, Energy Procedia 50, 862-869. Zenasni M.A., Meroufel B., Merlin A., George B., 2014, Adsorption of congored from aqueous solution using CTAB-kaolin from becharalgeria. J. Surf. Eng. Mater. Adv. Technol. 4, 332-341. 840