Effective Microporosity for Enhanced Adsorption Capacity of Cr (VI) from Dilute Aqueous Solution: Isotherm and Kinetics

Chemical Science International Journal 19(4): 1-12, 2017; Article no.CSIJ.34082 ISSN: 2456-706X (Past name: American Chemical Science Journal, Past ISSN: 2249-0205) Effective Microporosity for Enhanced Adsorption Capacity of Cr (VI) from Dilute Aqueous Solution: Isotherm and Kinetics Lloyd Mukosha1,2*, Maurice S. Onyango1, Aoyi Ochieng3 and John Siame2 1 Department of Chemical, Metallurgical and Materials Engineering, Tshwane University of Technology, Pretoria, Private Bag X680, South Africa. 2 Department of Chemical Engineering, Copperbelt University, Kitwe, P.O.Box 21692, Zambia. 3 Department of Chemical Engineering, Vaal University of Technology, Vanderbijlpark 1900, South Africa. Authors’ contributions This work was carried out in collaboration between all authors. Authors MSO and AO designed the study, organized experiment requisites and interpretation of results. Authors LM and JS conducted experiments, managed literature searches and analyzed samples and results. Author LM wrote the first draft of the manuscript. All authors read and approved the final manuscript. Article Information DOI: 10.9734/CSJI/2017/34082 Editor(s): (1) Nagatoshi Nishiwaki, Kochi University of Technology, Japan. (2) T. P. West, Department of Chemistry, Texas A&M University-Commerce, USA. Reviewers: (1) Farid I. El-Dossoki, Port-Said University, Egypt. (2) Jelena Kiurski, University Business Academy, Novi Sad, Serbia. (3) Saima Fazal, South China University of Technology, China. Complete Peer review History: http://www.sciencedomain.org/review-history/19612 th Original Research Article Received 12 May 2017 Accepted 13th June 2017 th Published 19 June 2017 ABSTRACT The adsorbent pore structure significant to enhanced adsorption capacity of Cr (VI) from dilute aqueous solution is evaluated. As reference, low-cost micro-mesoporous activated carbon (AC) of high basicity, mesoporosity centred about 2.4 nm, and effective microporosity centred about 0.9 nm was tested for removal of Cr (VI) from dilute aqueous solution in batch mode. At pH 2 the low-cost AC exhibited highly improved Langmuir Cr (VI) capacity of 115 mg/g which was competitive to high performance commercial AC. A Comparison with treated characterization results of literature adsorbents/ACs showed that moderate to high effective micropore volume of average pore-size _____________________________________________________________________________________________________ *Corresponding author: E-mail: loimwimba@yahoo.com; Mukosha et al.; CSIJ, 19(4): 1-12, 2017; Article no.CSIJ.34082 about 0.9 ± 0.1 nm is critical for increased adsorption capacity of Cr (VI) from dilute aqueous solutions. The mesostructure of the tested low-cost AC was associated with rapid kinetics that was fitted by the Pseudo-second kinetics model. While Biot numbers suggested slight significant contribution of intraparticle diffusion. It is hoped that this study may be a useful contribution to development of effective adsorbents for the efficient abatement of toxic Cr (VI) from wastewater and water. Keywords: Cr (VI) adsorption; dilute solution; adsorbent properties; micro-mesoporous; effective microporosity. 1. INTRODUCTION capacity for Cr (VI) than micro-mesoporous activated carbon (AC) of wide PSD centered about 3.2 nm. Contrary to above both reports, it is known that the adsorption process would follow a sequential occupancy of adsorption sites based on level of adsorption potential [13-15]. In this sense, for adsorption from dilute solutions, the adsorption potential would be highest in effective micropores with monolayer or primary volume filling capacity due to strong overlapping of adsorption potential from opposite walls, as opposed to surface coverage in wider pores. This means that enhancement of adsorption potential in effective micropores depends on the ratio of micropore size to adsorbate size. In related study, the high butane adsorption capacity from dilute solution was reported to occur in effective micropores with primary volume filling [16]. This should be contrasted with adsorption from concentrated solutions when capacity could be a function of total micropore volume. The necessity for removal of highly toxic Cr (VI) from wastewaters and the highly feasible use of the economical and efficient adsorption technology have been adequately discussed [1-3]. It is at low concentration of Cr (VI) in water bodies that is significant to environmental protection as reflected in stringent Cr (VI) contaminant limits of 0.05 mg/L in drinking water and 0.1 mg/L for wastewater discharge to surface waters [4]. For solution Cr (VI) concentration < 1000 mg/L, the predominant species are hydrogen chromate oxyanion (HCrO4-) in pH 1 – 6.5 and chromate oxyanion 2(CrO4 ) in pH 6.5 – 10 [2]. With reference to diameter of CrO42- of ca. ≈ 0.50 nm [5] and its spatial arrangement, it could be assumed that the diameter of HCrO4- (a product of protonation 2+ reaction: CrO4 + H ↔ HCrO4 ) would be similar at about 0.50 nm. Literature studies on the adsorption of Cr (VI) in solution pH 2 - 5 has showed that adsorbent specific surface functionality (acidic or basic) would enhance affinity for HCrO4 either by ion-exchange interaction [6] or electrostatic attraction [7,8]. It is evident from X-ray Photoelectron Spectroscopy analysis that HCrO4 /biosorbent interaction could plausibly proceed through adsorption-coupled reduction and chemisorption or complexation of 3+ resultant Cr dependent on amount of acidic surface groups [8-10]. The contribution of knowledge of the relevant PSD for enhanced Cr (VI) capacity from dilute solutions would be significant for optimal design of future adsorbents. Interestingly, many studies have explored the potential application of abundantly available biomass wastes for development of low-cost adsorbents/ACs for abatement of Cr (VI) from wastewater [1,2,17]. The low-cost adsorbents/ACs were developed using the backward approach of first developing the adsorbent and then testing for quality in adsorption of Cr (VI). However, most of developed low-cost adsorbents/ACs have showed low Cr (VI) capacities from dilute aqueous solutions. In practical application, low Cr (VI) capacities on low-cost adsorbents would translate into usage of large quantities of adsorbent and design of large sized adsorption units. The handling and disposal of large quantities of spent adsorbent could also be an environmental problem. Therefore, there is still need to develop new low-cost adsorbents of high efficient removal of Cr (VI) from dilute aqueous solutions. At present, however, the outstanding fact about Cr (VI) adsorption is the lack of conclusive empirical evidence on the relevant pore-size distribution (PSD) for enhanced adsorption capacity from dilute industrial wastewaters [typically < 400 ppm] significant to environmental protection. Referenced to silica nanoparticles of mean adsorption pore size about 7.0 nm, the researchers [11] concluded that adsorbent broad PSD, not surface area or average pore size, is important for enhanced removal of Cr (VI) from wastewater. In another study [12], ordered mesoporous carbon (OMC) of PSD centered about 22 nm was reported to have very high 2 Mukosha et al.; CSIJ, 19(4): 1-12, 2017; Article no.CSIJ.34082 In this study, attempt is made to evaluate the effective microporosity for enhanced Cr (VI) capacity from dilute aqueous solution and, thereby, promote the forward approach of incorporating relevant adsorbent properties at development stage. Low-cost sawdust AC (denoted L-AC) developed at our previously determined optimum conditions [16] was submitted for removal of Cr (VI) from dilute aqueous media. The L-AC was characterized for relevant adsorptive properties and batch adsorption studies to evaluate effect of pH, equilibrium capacity and kinetics were conducted. Importantly, the pore structure the L-AC is discussed in relation to adsorption potentials for Cr (VI) from dilute aqueous media. The equilibrium capacity was compared with commercial AC. Thereafter, using our L-AC as reference, literature adsorbent/AC characterization results were treated and their relevance to results of comparative analysis was made to validate the effective micropore average size available for adsorption for enhanced Cr (VI) capacity from dilute solutions. and the char cooled to ambient temperature in N2 flow. To develop L-AC, the prepared char was heated in continuous preheated N2 flow (180°C, 570 mL/min) at 10 K/min to 800°C and superheated steam (180°C, 1.6 bar, and 780 mL/min) activated for 1.5 hrs, and cooled to ambient temperature in N2 flow and stored in a desiccator. The during activation or in-situ modification of surface groups occurred at the activation temperature in flowing N2-steam mixture. The outgassed L-AC sample was analyzed for textural properties using 77K N2-adsorption (TriStar II 2030, Micromeritics) and 273K CO2adsorption (ASAP 2020, Micromeritics), and surface morphology using scanning electron microscope (SEM, Cambridge Instrument 360). While the total acidity and basicity were determined by the standardized Boehm titration method and the pH point of zero charge (pHPZC) from proton binding isotherm obtained by the manual potentiometric titration method. Details of instruments and procedures were previously reported [16]. The bulk density was determined using the graduated cylinder method. The average pore size (LAVE, nm) of micropores available for adsorption was calculated from the geometrical relation [19]: 2. EXPERIMENTATION 2.1 Materials The low-cost AC (L-AC) was developed from Pine tree (Pinus patula) sawdust obtained locally (Singisi Sawmill, KwaZulu Natal, South Africa). The commercial AC (Norit RO 0.8), vendor specified as suitable for wastewater treatment was purchased from Sigma, USA and used for comparison. Potassium dichromate (K2Cr2O7, 99+ %) powder was purchased from Sigma, USA. Hydrochloric acid (HCl, 32%) Sodium Chloride standard solution (NaCl, 99.5%) and Sodium Hydroxide pellets (NaOH, ≥ 95%) were purchased from SMM, South Africa. Stock solutions were prepared using deionised water. All chemicals used were of analytical reagent grade. = 2 × 10 3 / (1) 2 where, VMIC (cm /g) and SMIC (m /g) are t-plot (analogue of α-plot [13]) micropore volume and surface area respectively. 2.3 Batch Equilibrium Equilibrium adsorption studies were conducted in sample bottles with 0.8 g/L-AC in solutions of known Cr (VI) concentrations shaken in a thermostated water shaker (Labcom, Maraisburg) at 200 rpm and 25°C for 24 hrs equilibration, paper filtration and the filtrates Cr (VI) concentrations determined by the EPA recommended 1, 5-Diphenylcarbazide method using UV-Vis spectrophotometer (Libra S12, UK) at 543 nm. The effect of pH was studied using triplicate samples. Solutions pH was adjusted using HCl or NaOH solutions, and using precalibrated pH meter (Orion 4 star, USA). Subsequent to determining the adsorption optimum pH all other adsorption tests were carried out at that optimum pH. The uptake qe (mg/g) and percentage removal φ (%) were calculated according to equations (2) and (3) respectively. 2.2 Development and Characterization of Activated Carbon The carbonization and activation of carbon samples were performed in a stainless steel vertical fixed-bed tubular reactor. Equipments and details of preliminary optimization of carbonization/activation processes have been previously reported [18]. Briefly, the P. patula sawdust was carbonized in continuous N2 flow (570 mL/min) at 10 K/min to 800°C and 2 hrs, 3 Mukosha et al.; CSIJ, 19(4): 1-12, 2017; Article no.CSIJ.34082 =( )/ − = 100[( adsorption mass transfer Biot (Bi) number, equation 10 [23] with the batch film transfer coefficient (kfb, cm/s) determined by the Mathews & Weber (M&W) linear driving force (LDF) rate law (equation 12) [24] and the effective 2 intraparticle diffusion coefficient (Deff, cm /s) approximated by the Patterson diffusion model (equations 13 – 17) [23]. (2) ) / − (3) ] where, CO is the solute initial concentration (mg/L), Ce is the equilibrium solute concentration in solution (mg/L), VS is the solution volume (L), and m is mass of AC (g). = = =∑ ( )= = = ( ) (4) ( ) ( )= (5) ( ) ( ( ) ) (6) =( =4 +3 2.4 Batch Kinetics / (11) (12) ( ( − ( ))/( / )/ −3 ) 1+ ) 1+ − √ √ ) (13) (14) (15) (16) =0 (17) where, q1 and q2 are adsorbed quantities (mg/g), k1 and k2 are rate constants, h is initial sorption rate (min), qO adsorbed quantity (mg/g) in equilibrium with CO, dp is particle diameter (cm), 3 ρp is particle density (g/cm ) estimated by equation 11 [23], ρb is bulk density (g/cm3), ɛb is bed porosity of typically about 0.4 for AC [25], U(t) is the fractional attainment to equilibrium; w is the equilibrium partition ratio (0 < w < 1 indicates finite solution case i.e. concentration of adsorptive in solution continuously changes from initial to equilibrium, and w > 1 indicates infinite solution case i.e. concentration of adsorptive remains constant in solution), τ is dimensionless time constant and, the ϴ and β terms are the negative and positive roots respectively, of equation (17). Batch adsorption kinetics tests were conducted for the evaluation of the rate constants and diffusion parameters. A laboratory batch stirred tank with 600 mL solution of 50 mg/L Cr (VI) and stirring speed of 400 rpm was used by the method detailed in [21]. Syringe filters (0.2 µm) were used to withdraw 5 mL aliquot samples at predetermined time intervals. The variable studied was initial concentration. The uptake with time t (min), q(t) (mg/g) was calculated by: − ( ) (10) − − (9) ) =− ( )=( (8) ) = ⁄(1 − 1− where, Qmax is the monolayer adsorption capacity (mg/g), and KL is the Langmuir constant (L/mg). KF is Freundlich constant related to the adsorption capacity (mg/g) and Fr is the heterogeneous factor. ( )= (1 − ( )= The isotherm parameters were evaluated using the isotherm models of the Langmuir (equation 4) and Freundlich (equation 5) [20]. The agreement of model to data was validated by nonlinear regression of minimizing the Chi-squared (χ2) function per equation 6 [20], with solutions generated using Microsoft Excel 2010, Add-in solver. (7) where, C(t) is solute concentration in (mg/L) at time, t. The kinetics rate constants were evaluated using the adsorption surface reaction models of the pseudo first-order (equation 8) and the pseudo second-order (equation 9) [8,22], with conformity of model to data assessed through minimum Chisquared function, equation 6. While the diffusion controlling stage was evaluated from the 3. RESULTS AND DISCUSSION 3.1 Activated Carbon Characterization Fig. 1 shows the 77K N2-adsorption/desorption isotherm curve of L-AC. The exhibited type IV 4 Mukosha et al.; CSIJ, 19(4): 1-12, 2017;; Article no.CSIJ.34082 no. isotherm with H3-type type hysteresis loop in the range P/PO (0.4 – 0.8) was indicative of narrow distribution of relatively uniform or ordered mesopores. es. From SEM micrograph of Fig. 2, platelike particles with some slit-pores slit and dominant similar circular pores were observed. The high N2 adsorption start point in Fig. 1 suggested N2 molecule diffusional restrictions in narrow micropore (< 0.5 nm) at 77K and P/PO < 0.06 in our short experiment time of 8 hrs [15] and, therefore, narrow microporosity was evaluated from CO2-adsorption adsorption isotherm data at 273K and P/PO < 0.04 and 8 hrs [insert in Fig. 1]. measured high N2-BET BET surface area (SBET) in Table 1 indicated presence of wide micropore (0.5 – 2.0 nm). Using equation (1), the calculated average pore width (LAVE) of wide micropores available for adsorption was 0.9 nm. Fig. 3. The mesopore-size size distribution ((●) of L-AC [Insert: The narrow micropore micropore-size distribution (■) of L-AC] AC] AC showed high SBET of From Table 1, the L-AC 2 1079 m /g with wide micropore surface area (SMIC) at ca. 54%, and total pore volume (V ( T) of 0.71 cm3/g with moderate wide micropore 3 volume (VMIC) of 0.25 cm /g. The high proportion of mesoporosity (VT – VMIC) at ca. 65% was important for reduced diffusion limitation of HCrO4- oxyanion to interior wide micropores that would be sites of high adsorption potential [13-15]. 15]. Reported theoretical calculations have indicated that the important parameter for enhanced adsorption potential in effective ropore size to micropores is the ratio of micropore adsorbate size, with upper ratio limit for primary volume filling in the range 1.5 – 2.0 dependent on actual shape (slit-like like or cylindrical) pores [13]. With reference to approximate size of HCrO4- (≈ ≈ 0.50 nm), narrow micropores (< 0.5 nm) would be closed due to size exclusion effect. Also, the adsorption capacity would be reduced in micropores of same size as HCrO4- due to kinetic limitations. Thus, considering the theoretical upper ratio limit of effective micropore size for primary filling g at 2.0, it could be assumed that for enhanced adsorption capacity of HCrO4 from dilute solutions the effective wide micropore PSD would be in the range 0.50 – 1.0 nm. Thus, the evaluated PSD of L-AC AC of moderate wide micropore volume of average pore width widt (LAVE = 0.9 nm) and high proportion of narrow mesoporosity was promising for enhanced capacity and fast kinetics of Cr (VI) oxyanion, subject to enhanced affinity of HCrO4- for the insitu created surface groups. Fig. 1. 77K N2- adsorption/desorption AC [Insert: 273K CO2isotherms (●) of L-AC adsorption isotherms (♦) ♦) of L-AC] L Fig. 2. SEM micrograph of L-AC. AC. Particle 1000 +850 µm; Mag.: 500 x 10µm size:-1000 The mesopore PSD was determined by the Pierce method [13] and results shown in Fig 3, where it was confirmed that the L-AC L had uniform mesopores in the narrow range 2.0 – 8.0 nm pore diameter centred at about 2.5 nm pore adsorption data, the diameter. From CO2-adsorption instrument software calculated narrow microporosity distribution (Horvath-Kawazoe (Horvath method) is shown as insert in Fig. 2. It was observed that the narrow microporosity was mostly 0.4 nm in diameter. However, the 5 Mukosha et al.; CSIJ, 19(4): 1-12, 2017;; Article no.CSIJ.34082 no. Table 1. Textural properties pr of L-AC 2 SBET (m /g) 1079 2 SMIC (m /g) 580 3 VT (cm /g) 0.71 3 VMIC (cm /g) 0.25 LAVE (nm) 0.9 Dave (nm) 2.4 Dave: HJB average pore diameter; ρb: Bulk density Fig. 4 shows that the L-AC AC was highly basic with pHPZC about 10, consistent with the Boehm titration calculations of high total basicity (940 µmol/g) against total acidity (71 µmol/g). The high activation temperature (800°C) C) basicity of L-AC AC could be due to presence of thermally stable surface oxygen basic c groups (pyrone and electrons system chromene) and delocalized ᴨ-electrons on basal graphene layers [26]. ρb (g/L) 162 The results in Fig. 5 showed that percentage removal was high at pH 2 (ca. 98 ± 1.6%) and minimal for pH > 2. A similar trend has been reported [27,28]. On the other hand, the equilibrium pH slightly increased for initial pH (2 – 8), and was similar to initial pH for pH (10 – 12). An only slight increase in equilibrium pH at initial pH 2 suggested that the solution sol direct reduction of Cr (VI) to Cr (III), catalyzed by surface acidic electron-donor donor groups and complexation of Cr (III) as suggested in [8,29] may not be the main mechanism of Cr (VI) adsorption. Otherwise, the equilibrium pH would have substantially increased due to consumption of large quantities of solution protons in the reduction process as per equation (18) [17]. The negligible direct reduction of Cr (VI) at pH ≥ 2 has been similarly reported [6,8,17]. For CO = 100 mg/L, the predominant species is HCrO4 in pH 1 – 6.5 [2] and L-AC AC surface is protonated as shown in Fig. 4. The observed slight increase in equilibrium pH for initial pH < 10 in Fig. 4 may be explained by surface protonation. Significant ion exchange of surface protons for HCrO4 at pH 2 as per equation (19) proposed in [6] may also not was be the main mechanism, because the L-AC L highly basic with low concentration of acidic groups that would be main sites of ion exchange to account for the measured high ca. 98 ± 1.6% removal for CO = 100 mg/L at pH 2. Fig. 4. Proton binding isotherm for L-AC. L 298K; < 45 μm 3.2 Effect of pH The pH influences the ionization of AC surface groups and Cr (VI) in water and, therefore, the evolution of uptake with pH could be used to elucidate the type of surface interactions e.g. complexation or ion exchange or electrostatic attraction or combinations. Fig. 5 shows the effect of pH (2 – 12) on percentage removal of Cr (VI) on L-AC. - + 3+ HCrO4 + 7H + 3e- ↔ Cr +4H2O (18) AC-OH2+ + HCrO4- ↔ AC-OH2CrO4- + H+ (19) + - + - AC-OH2 + HCrO4 ↔ AC-OH2 (HCrO4 ) (20) Therefore, the high adsorption at pH 2 was suggestive of primary physical non non-specific interactions of electrostatic attraction between HCrO4- and protonated surface groups e.g. equation (20). Since the level of acidic groups was low on L-AC, AC, it would be expected that only part of surface bound HCrO4- was reduced and complexed. In this study,, experiments not shown, the adsorption-coupled coupled reduction and complexation of part of surface bound HCrO4 could be supported by loss of Cr (VI) adsorption capacity at ca. 52% on reusability of column regenerated Cr (VI) saturated L--AC bed using 3M NaOH eluent. The observed low adsorption Fig. 5. Effect of pH on sorption of Cr (VI) on L-AC. CO: 50 mg/l; 298K; -425 425 + 106 μm: Percentage removal (♦); Equilibrium pH (X) 6 Mukosha et al.; CSIJ, 19(4): 1-12, 2017;; Article no.CSIJ.34082 no. that the Cr (VI) capacity on sawdust low low-cost AC of this study was on the higher side of sawdust low-cost cost adsorbents/ACs, and competitive with high performance commercial ACs. Due to low lowcost, the high Cr (VI) capacity on L L-AC would translate into high adsorption capacity per cost of adsorbent compared with expensive commercial ACs [40], and can be used on a once once-through basis without considerations for or the costly regeneration [32]. This result qualified the L L-AC to be very economical and efficient for abatement of Cr (VI) from wastewater, even at large scale. The spent AC would be safely disposed off through a strict toxic sludge containment protocol, otherwise use of the common spent AC incineration method would have a major problem 196°C of Cr (VI) thermal decomposition at about 196 to highly thermal stable and carcinogenic chromium (III) oxide [41]. of chromium oxyanions in pH (2 – 8) in Fig. 5 was attributed to very low level of surface protonation as seen in Fig. 4. While for pH > 10 the L-AC AC surface was deprotonation and repulsion of chromium oxyanions occurred. Further adsorption tests in this study were conducted at pH 2. 3.3 Adsorption Isotherm and Comparative Cr (VI) Capacities The equilibrium isotherm data was treated to evaluate the maximum adsorption capacity of LL AC for Cr (VI) from dilute aqueous solution. Generally, for economical and efficient operation of adsorption units, the AC must have high capacity for the solute so that relatively small quantities can be used to effect a given wastewater treatment. Fig. 6 compares the isotherm curves of Cr (VI) on L-AC and L commercial AC (Norit RO 0.8). Both ACs exhibited L-type type isotherms indicative of weak competition ion between Cr (VI) and water for active adsorption sites [30]. The isotherm parameters were calculated using the Langmuir (equation (14)) and Freundlich (equation (15)) models and results presented in Table 2. In majority of studies on adsorption of Cr (VI) on AC the pore structure was not fully characterized. Nevertheless, for dilute solutions, the results in Table 3 indicate that Cr (VI) capacity is not influenced by surface area. Analysis of surface functionality show that low and high Cr (VI) capacities have been obtained on either acidic or size distribution basic ACs, suggesting that pore-size (PSD) strongly influences HCrO4 capacity on AC. For assumed spherical HCrO4- oxyanions (size ≈ 0.5 nm); we calculated the monolayer equivalent surface area (SEQ) of adsor adsorbed HCrO4- on L-AC AC and compared it with the measured wide micropore surface area ((SMIC) of L-AC. It was found that SEQ was less than SMIC. With reference to the average wide micropore size available for adsorption (LAVE = 0.9 nm) of LL AC, and the assumed range ange 0.5 – 1.0 nm (section 3.1) of effective micropore PSD for HCrO4- primary volume filling, the calculated SEQ < SMIC showed that only effective micropores with monolayer filling were involved in the adsorption from dilute solution. To support this finding, ing, we treated literature AC characterization AC. For adequately results and compared with L-AC. characterized ACs in Table 3, It was found that always SEQ < SMIC, confirming that only an effective microporosity was involved in adsorption from dilute solution,, in agreement with results in related study [16]. The high Cr (VI) AC of this study could be capacity on the L-AC related to presence of large volume of effective micropores with primary filling capacity. It could be argued that the same reason may explain tthe reported high Cr (VI) capacity on mesoporous silica nanoparticles (SNP) having broad PSD [11]. While the minimal capacity for AC with Fig. 6. Adsorption isotherm of Cr (VI) on L-AC and Norit AC (RO 0.8). 298K; pH 2.0 ± 0.3; ●); Norit AC A (▲); -200+150 μm: L-AC (●); Langmuir (▬); ▬); Freundlich(Freundlich - -) According to Table 2, the isotherm data of both ACs were well fitted by the Langmuir model (lowest χ2) that underpins monolayer coverage, coverag suggesting uniform distribution of active L showed adsorption sites [20]. However, L-AC high monolayer capacity for Cr (VI) than Norit AC (RO 0.8). A further comparison of Cr (VI) capacities from dilute solutions (< 400 mg/L) in pH 1 - 3 with some literature ure sawdust low-cost low adsorbents/ACs and commercial ACs was summarized in Table 3. It was noticed in Table 3 7 Mukosha et al.; CSIJ, 19(4): 1-12, 2017; Article no.CSIJ.34082 LAVE = 0.6 nm (Table 3) could be attributed to almost absence of effective microporosity as confirmed by very large difference between SEQ and SMIC, suggesting that this AC had majority of micropores below 0.6 nm and size exclusion effects could have dominated. On the other hand, for moderate to high VMIC of ACs, the results in Table 3 seem to indicate gradual decrease of capacity with monotonic increase of LAVE above 0.9 nm. The decrease of capacity could be attributed to progressive loss of effective micropores with increase in LAVE > 0.9 nm. The exceptional very low capacity at LAVE = 1.0 nm (Table 3) could be explained in terms of very low VMIC. Interestingly, for ordered mesoporous carbon (OMC) it was observed that SEQ > SMIC: The very high Cr (VI) capacity on OMC of large sized pores may be explained in terms of bilayer or secondary volume filling considering the estimated average pore size available for adsorption (LAVE = 1.7 nm), and high CO upto 3000 mg/L used in the equilibrium adsorption experiments. The treated results in Table 3 confirm the suggestion that an effective micropore PSD is significant for enhanced sorption capacity of Cr (VI) from dilute aqueous solutions. 3.4 Batch Kinetics and Intraparticle Mass Transfer Besides high adsorption capacity for Cr (VI), the Cr (VI) uptake rate on L-AC is integral to optimal design of adsorption units for it would affect the throughput. Fig. 7 shows the effect of initial concentration and contact time on sorption of Cr (VI) on L-AC. The L-AC exhibited fast adsorption kinetics that could be associated with the mesostructure that offered reduced diffusion resistance for HCrO4 oxyanions to interior wide micropores the sites of high adsorption energy. Table 2. Isotherm parameters of Cr (VI) on L-AC and Norit AC (RO 0.8) at 298K Adsorbent QMAX (mg/g) 115 94 L-AC Norit AC Langmuir constants KL (L/mol) 0.83 0.55 χ2 9.14 3.04 Freundlich constants KF (mg/g) Fr χ2 52.0 0.18 10.66 41.0 0.19 4.00 Table 3. Comparison of Cr (VI) capacities from dilute aqueous solutions on low-cost adsorbents and commercial activated carbon at room temperature and pH (1 - 3) Adsorbent SAC SAC SD CAC SAC J. AC CAC P. AC L. AC LS.AC L. AC P. AC B. AC CAC SAC CAC MCS WB.AC SNP OMC [A] [A] [A] [A] [B] [A] [A] [B] [B] [B] [B] [B] [A] [B] SBET (m2/g) 0.86 1674 650 1305 874 751 1459 1512 1363 86 718 1402 1079 1150 4.4 760 553 VMIC (cm3/g) 0.55 0.35 0.33 0.82 0.61 0.36 0.09 0.32 0.46 0.25 0.36 SMIC (m2/g) 1178 952 701 180 585 1370 580 415 SEQ (m2/g) LAVE (nm) 339 142 376 54 360 24 465 1.4 1.3 1.0 1.0 1.1 0.6 0.9 764 1.7 QMAX (mg/g) 42.50 44.05 12.78 147 22.35 141 132 106 84 35 93 16 89 5.9 115 94 200 59.23 111 189 Ref. [17] [28] [31] [32] [33] [34] [34] [34] [35] [1] [35] [36] [37] [38] This study This study [27] [39] [11] [12] SAC: Sawdust activated carbon; SD: sawdust; CAC: commercial activated carbon; J: Jatropha; P: Peanut shell; L: Lignin; LS: Longan seed; B: Bagasse;MCS: Modified corn stalk;WB: Waste bambo; SNP: Silica nanoparticles; OMC: Ordered mesoporous carbon; [A]: Acidic; [B]: Basic; SEQ = monolayer equivalent surface area = (Am Qmax NA)/MW; Am: molecular area; NA: Avogadro constant; MW: molecular weight; LAVE = 2000vmic/smic 8 Mukosha et al.; CSIJ, 19(4): 1-12, 2017;; Article no.CSIJ.34082 no. The applicable rate constant was determined first order (equation 4.5) from fitting the pseudo-first second order (equation 4.7) to and pseudo-second kinetics data, and results presented ed in Table 4. The kinetics rate constant is an essential parameter for optimal design of batch adsorption units by use of the contact time model [42]. The results in Table 4 indicated that the pseudopseudo second order model adequately described the kinetics data (lowest χ2), ), as confirmed in Fig. 7. The Pseudo-second second model is formulated on chemisorption kinetics [43] and, therefore, its agreement with kinetic data suggested involvement of chemical adsorption, further support of occurrence of some coupled reduction of Cr (VI). The analysis of adsorption-reduction order kinetic parameters revealed that second-order the initial adsorption rate, h,, increased with an increase in CO, consistent with results in Fig. 7. On the other hand, the adsorption rate constant, k2, decreased with an increase in CO. Similar variation of k2 with CO was reported for sorption of Cr (VI) on sawdust [17]. The increase of h with an increase in CO was attributed to high concentration gradients at high initial concentration causing fast diffusion of HCrO4oxyanion to the AC surface. As would be expected of equilibrium uptake, the q2 values increased with an increase in CO. This was attributed to large number of HCrO4 oxyanion available for adsorption and hence long equilibrium times as CO increased. film round the solid [23,25]. The knowledge of diffusion limiting stage would allow for proper ess conditions to enhance Cr (VI) design of process uptake. To determine the effective intraparticle diffusion coefficient (Deff), the Patterson model (equations 13 - 17) was fitted to kinetics data and the model performance shown in Fig. 8. While Fig. 9 shows the fitting off the M&W (LDF) rate law (equations 4.12 – 4.14) to kinetic data at CO = 50 mg/L Cr (VI) in the initial adsorption time < 5 min for determination of the [batch] liquid-film liquid mass transfer coefficient (kfb). Fig. 8. Fractional attainment to equilibrium for sorption of Cr (VI) on L-AC: AC: 10 mg/L ((□); 30 mg/L (∆); ∆); 50 mg/L (○); Patterson diffusion model (▬) Fig. 9. Application of linear driving force rate law for sorption of Cr (VI) on L-AC. L t < 5.0 mins Fig. 7. Effect of initial concentration on batch 296K; kinetics for sorption of Cr (VI) on L-AC. L ▲); 30 mg/L (♦); ( 50 -600+300 µm: 10 mg/L (▲); mg/L (●); 2nd order kinetics (▬) ( It can be seen that the Patterson model fit to kinetics data was excellent, with an average error between experimental and model values of ca. ≤ 8% for all concentrations studied. Equally, the M&W (LDF) fit to kinetic data was excellent with 2 the correlation coefficient of R > 0.98 for all concentrations studied. The estimated values of Deff and kfb and Bi are summarized in Table 5. The Patterson model is applicable to intraparticle diffusion control and finite solution case, with The batch diffusion controlling stage was evaluated through calculation of [batch] Biot number (Bi), equation 10. The Biot number is a ratio of mass transfer resistance in the solid phase to mass transfer resistance in the liquidliquid 9 Mukosha et al.; CSIJ, 19(4): 1-12, 2017; Article no.CSIJ.34082 Table 4. Batch kinetics rate parameters for sorption of Cr (VI) on L-AC Variable 10 30 50 qe(expt) (mg/g) Pseudo-first order q1 k1 (mg/g) (/ min) CO (mg/l) 11.66 11.34 0.35 36.54 35.48 0.19 60.76 58.84 0.13 χ2 Pseudo-second order χ2 q2 k2 h (mg/g) (g /mg min) (mg/ g min) (particle size:-600+300μm; T: 296K) 12.02 0.0417 6 0.1 38.20 0.0065 9 0.8 61.37 0.0025 10 1.2 0.3 1.0 1.4 Table 5. Estimated diffusion parameters for sorption of Cr (VI) on L-AC Parameter 10 30 50 w CO (mg/l) 0.977 0.971 0.961 3 8 2 kfb ( x 10 cm/s) Deff ( x 10 cm /s) (Particle size: -600+300μm; T: 296K) 7.34 6.16 7.45 3.09 5.30 2.24 finite solution case verified by calculated equilibrium partition ratios from equation (17) of 0 < w < 1 for all initial concentrations studied. The 2 magnitudes of estimated Deff (10-8 cm /s) and kfb (10-3 cm/s) are typical of metal ions diffusion into AC porous structure reported in literature [23,24] and, therefore, further validating the applicability of the Patterson model and M&W (LDF) rate law to kinetic data. The kfb values were similar at CO (10 and 30 mg/l), but decreased at CO = 50 mg/L. Choy et al. [24] have reported a similar trend for sorption of Zn (II) on bone char. The Deff values decreased with an increase in CO, indicative of increasing intraparticle resistance presumably due to HCrO4- anions crowding of the adsorbent surface and, therefore, anions competition for adsorbent pores. Consequently, the Bi increased with increase in CO. Intraparticle diffusion sorely controls the adsorption process for Bi > 30, while film diffusion controls the adsorption for Bi < 0.5 [23]. Therefore, the calculated (Bi) values indicated that intraparticle diffusion resistance was relatively significant for CO ≥ 20 mg/L. Bi 0.96 5.15 8.18 high proportion of mesoporosity is desirable for reduced diffusion limitations to interior effective micropores. The results of this study show that the forward approach of tailoring relevant adsorptive properties at adsorbent preparation stage is important than the backward approach of developing the adsorbent first and then testing for quality. The determining of adsorbent/AC micropore volume and an average effective micropore size from N2 adsorption isotherm data would be simple routine quality control procedure to ensure quality of developed adsorbent/AC. ACKNOWLEDGEMENT The authors gratefully acknowledge the financial support of this project by the National Research Fund of the Tshwane University of Technology, Pretoria, South Africa. COMPETING INTERESTS Authors have interests exist. 4. CONCLUSION declared that no competing REFERENCES Optimally developed low-cost adsorbent/ activated carbon (AC) of tailored microporosity distribution would provide a sustainable alternative for economical and efficient removal of Cr (VI) from dilute wastewaters. The results of this study showed that adsorbent effective microporosity strongly influences the sorption capacity of Cr (VI) from dilute solutions significant to environmental protection. Consequently, the optimum design of low-cost adsorbent/AC should factor moderate to high micropore volume and effective microporosity of average pore size about 0.9 ± 0.1 nm. Also, a 1. 2. 10 Jinbei Y, Meiqiong Y, Wentao C. Adsorption of hexavalent chromium from aqueous solution by activated carbon prepared from longan seed: Kinetics, equilibrium and thermodynamics. J. Ind. Eng. Chem. 2015;21:414–422. Mohan D, Pittman CU. Review of activated carbons and low cost adsorbents for remediation of tri- and hexavalent chromium from water. J. Hazard. Mater. 2006;B137:762–811. Mukosha et al.; CSIJ, 19(4): 1-12, 2017; Article no.CSIJ.34082 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Ghosh PK. Hexavalent chromium [Cr (VI)] removal by acid modified waste activated carbons. J. Hazard. Mater. 2009;171:116– 122. EPA (Environmental Protection Agency), Environmental pollution control alternatives, EPA/625/5-90/025, EPA/625/4-89/023, Cincinnati, US; 1990. Jenkins HDB, Thakur KP. Reappraisal of thermochemical radii for complex ions. J. Chem. Educ. 1979;56(9):576. Liu SX, Chen X, Chen XY, Liu ZF, Wang HL. Activated carbon with excellent chromium (VI) adsorption performance prepared by acid–base surface modification. J. Hazard. Mater. 2007;141: 315–319. Owlad M, Aroua MK, wan Daud WMA. Hexavalent chromium adsorption on impregnated palm shell activated carbon with polyethyleneimine. Bioresour. Technol. 2010;101:5098–5103. Yingxin Z, Wenfang Q, Guanyi C, Min J, Zhenya Z. Behavior of Cr (VI) removal from wastewater by adsorption onto HCL activated Akadama clay. J. Taiwan Inst. Chem. Eng. 2015;50:190–197. Hu J, Chen G, Lo IM. Removal and recovery of Cr (VI) from wastewater by maghemite nanoparticles. Water Res. 2005;39:4528–4536. Zhung-gia N, Jun-wei L, Hanita D, Si-ling N, Mohammed JKB. Reassessment of adsorption–reduction mechanism of hexavalent chromium in attaining practicable mechanistic kinetic model. Process. Saf. Environ. Prot. 2016;102:98– 105. Idris SA, Alotaibi KM, Peshkur TA, Anderson P, Morris M, Gibson LT. Adsorption kinetic study: Effect of adsorbent pore size distribution on the rate of Cr (VI) uptake. Microporous Mesoporous Mater. 2013;165:99–105. Chen T, Wang T, Wang DJW, Xue HR, Zhao JQ, Ding XC, Wu SC, He JP. Synthesis of ordered large-pore mesoporous carbon for Cr (VI) adsorption. Mater. Res. Bull. 2011;46:1424–1430. Gregg SJ, Sing KSW. Adsorption, surface area and porosity. 2nd ed. London: Academic Press; 1982. Dabrowski A. Adsorption-from theory to practice. Adv. Colloid Interface Sci. 2001;93:135-224. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 11 Marsh H, Rodriguez-Reinoso F. Activated carbon. 1st ed. UK: Elsevier Science & Technology Books; 2006. Centeno TA, Marban G, Fuertes AB. Importance of micropore size distribution on adsorption at low adsorbate concentrations. Carbon. 2003;41:843–846. Gupta S, Babu BV. Removal of toxic metal Cr (VI) from aqueous solutions using sawdust as adsorbent: Equilibrium, kinetics, and regeneration studies. Chem. Eng. J. 2009;150:352–365. Mukosha L, Onyango MS, Ochieng A, Kasaini H. Development of better quality low-cost activated carbon from South African pine tree (Pinus patula) sawdust: Characterization and comparative phenol adsorption. Int. J. Chem. Nuclear, Metal. Mater. Eng. 2013;7(7):228-238. Ohkubo T, Iiyama T, Nishikawa K, Suzuki T, Kaneko K. Pore-width-dependent ordering of C2H5OH molecules confined in graphitic slit nanospaces. J. Phy. Chem. 1999;B103:1859-1863. Foo KY, Hameed BH. Insight into the modeling of adsorption isotherm systems. Chem. Eng. J. 2010;156:2-10. Mukosha L, Onyango MS, Ochieng A, Taile YL. Sorption characteristics of mixed molecules of glutaraldehyde from water on mesoporous acid-amine modified low-cost activated carbon: Mechanism, isotherm and kinetics. J. Chem. 2014;1-13. Qiu H, Lu LV, Pan BC, Zhang QJ, Zhang WM, Zhang QX. Critical review in adsorption kinetic models. J. Zhejiang Univ. Sci. A. 2009;10(5):716-724. Inglezakis VJ, Poulopoulos SG. Adsorption, ion exchange and catalysis: Design of operations and environmental applications. UK: Elsevier; 2006. Choy KKH, Ko DC, Cheung WC, Porter JF, McKay G. Film and intraparticle mass transfer during the adsorption of metal ions onto bone char. J. Colloid Interface Sci. 2004;271:284-295. Cooney DO. Adsorption design for wastewater. USA: CRC Press; 1999. Villacanas F, Pereira MFR, Orfao JJM, Figueiredo JL. Adsorption of simple aromatic compounds on activated carbons. J. Colloid Interface Sci. 2006;293:128– 136. Suhong C, Qinyan Y, Baoyu G, Qian I, Xing X. Removal of Cr (VI) from aqueous Mukosha et al.; CSIJ, 19(4): 1-12, 2017; Article no.CSIJ.34082 solution using modified corn stalks: H3PO4 activation of lignin from kraft black Characteristic, equilibrium, kinetic and liquors. Water Res. 2004;38:3043–3050. thermodynamic study. Chem. Eng. J. 36. AL-Othman ZA, Ali R, Naushad MU. 2011;168:909–917. Hexavalent chromium removal from 28. Karthikeyan T, Rajgopal S, Miranda LR. aqueous medium by activated carbon Chromium (VI) adsorption from aqueous prepared from peanut shell: Adsorption solution by Hevea brasilinesis sawdust kinetics, equilibrium and thermodynamic activated carbon. J. Hazard. Mater. studies. Chem. Eng. J. 2012;184:238–247. 2005;B124:192–199. 37. Demiral H, Demiral I, Tumsek F, 29. Hu J, Chen C, Zhu X, Wang X. Removal of Karabacakoglu B. Adsorption of chromium chromium from aqueous solution by using (VI) from aqueous solution by activated oxidized multiwalled carbon nanotubes. J. carbon derived from olive bagasse and Hazard. Mater. 2009;162:1542–1550. applicability of different adsorption models. 30. Dabrowski A, Podkoscielny P, Hubicki ZM. Chem. Eng. J. 2008;144:188–196. Adsorption of phenolic compounds by 38. Gottipati R, Mishra S. Simultaneous activated carbon – a critical review. removal of trivalent and hexavalent Chemosphere. 2005;58:1049–70. chromium by activated carbon: Effect of 31. Kapur M, Mondal MK. Mass transfer and solution pH and pore size distribution of related phenomena for Cr (VI) adsorption adsorbent. Environ Prog Sustain Energy. from aqueous solutions onto Mangifera 2013;32(4):1030-1035. indica sawdust. Chem. Eng. J. 2013;218: 39. Tamirat D, Khalid S, Shimeles AK. 138–146. Adsorption of hexavalent chromium from 32. Aksu Z, Gonen F, Demircan Z. Biosorption aqueous solution using chemically of chromium (VI) ions by Mowital®B30H activated carbon prepared from locally resin immobilized activated sludge in a available waste of bamboo (Oxytenanthera packed bed: Comparison with granular abyssinica). Environ. Chem. 2014;1-9. activated carbon. Process Biochem. 40. Argun EM, Dursun S, Ozdemir C, Karatas 2002;38:175–186. M, Heavy metal adsorption by modified 33. Politi D, Sidiras D. Wastewater treatment oak sawdust: Thermodynamics and for dyes and heavy metals using modified kinetics. J. Hazard. Mater. 2007;141:77– pine sawdust as adsorbent. Procedia Eng. 85. 2012;42:1969–1982. 41. Guertin J, Jacobs JA, Avakian CP. 34. Gueye M, Richardson Y, Kafack TF, Blin J. Chromium (VI) handbook. UK: CRC Press; High efficiency activated carbons from 2004. African biomass residues for the removal 42. Ho SY, McKay G. A two-stage batch of chromium (VI) from wastewater. J. sorption optimized design for dye removal Environ. Chem. Eng. 2014;2:273–281. to minimize contact time. Trans. Icheme. 35. Gonzalez-Serrano E, Cordero T, Part B. 1998;76:313–318. Rodriguez-Mirasol J, Cotoruelo L, 43. Ho YS, McKay G. Pseudo-second order Rodriguez JJ. Removal of water pollutants model for sorption processes. Process with activated carbons prepared from Biochem. 1999;34:451-465. _________________________________________________________________________________ © 2017 Mukosha et al.; This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Peer-review history: The peer review history for this paper can be accessed here: http://sciencedomain.org/review-history/19612 12