Removal of Reactive Black 5 dye from Aqueous Media using Powdered Activated Carbon – Kinetics and Mechanisms

The textile industry is considered one of the major environmental polluters, primarily due to the quantity and composition of wastewater. It is therefore important to examine its diff erent treatment methods. For this purpose, the isothermal adsorption of Reactive Black 5 dye on powdered activated carbon at 25(±1) °C and 45(±1) °C was carried out to determine the eff ect of initial dye concentration, contact time and temperature on the adsorption process. In order to investigate the mechanism of adsorption of Reactive Black 5 dye on activated carbon, kinetic studies have also been carried out. Experimental data were analysed using a pseudo-fi rst-order and pseudo-second-order kinetic models, as well as an intraparticle diff usion model. Standard Gibbs free energy values of the adsorption process were also calculated, while the morphological analysis of activated carbon before and after adsorption was performed using a scanning electron microscope. The effi ciency of activated carbon as an adsorbent for Reactive Black 5 dye is evidenced by the fact that more than 60% of dye is adsorbed after 30 minutes regardless of initial concentration and temperature. The experimental data also showed that adsorption is kinetically controlled assuming a pseudo-second-order process, and that intraparticle diff usion is not the only process that infl uences the adsorption rate. Negative values of standard Gibbs free energy indicate that the adsorption reaction is spontaneous, while a higher negative value for temperature of 45 °C compared to 25 °C shows that a higher temperature is more energetically favourable for the adsorption of Reactive Black 5.


Introduction
At the end of the 20 th century, with all of the inventions and technical aids used every day, the relationship between man and nature was disrupted, and the Earth's ecosystem broken. Due to human knowledge and creativity over the last 250 years, a modern way of life, industrial development, advancing technology and agriculture, the use of large amounts of energy and natural resources, as well as the pursuit of a better and more comfortable life have led to the emergence of environmental hazards and their consequences (i.e. acid rain, desertifi cation, ozone depletion, climate change and other harmful phenomena). A great deal of attention worldwide has been given to environmental protection, the eff ect of harmful substances on human health and the entire ecosystem. Particular attention is given to industry, as it is no longer possible to continue expanding production and waste accumulation without infl icting permanent damage to nature. Th e textile industry is one of the major environmental polluters. It uses signifi cant quantities of water [1], which is eliminated as wastewater aft er being processed. Taking into account total industrial water pollution, textile fi nishing is considered one of the largest water consumption and pollution processes. Th e quantity and composition of wastewater depend on the type of basic raw material being processed (cotton, wool, fl ax, silk, synthetic fibres, blends, etc.) and textile fi nishing process used (laundering, bleaching, dyeing, printing, etc.), i.e. the type of chemicals used during some processes and the number of consecutive processes during production. In addition, the use of certain special fi nishing agents oft en results in specifi c and greater pollution of wastewater. Textile industry wastewater has a wide range of pH values and temperatures, it may be coloured and mainly contains various types of pollutants, most commonly dyes, surfactants, pesticides, oils, fats, solvents, heavy metals, inorganic salts, waste fi bres, etc. [2]. Special attention is given to dyes and pigments because they are highly visible materials so that even the minimum amount released into the environment may cause the appearance of colour in open waters. Besides being aesthetically defi cient, coloured water prevents the penetration of light into natural water, which negatively aff ects the entire natural water ecosystem, i.e. organisms depending on water quality. Th ere are more than 100,000 commercially available dyes with over 700,000 tonnes produced annually. Wastewater stream from the textile dyeing process contains unutilised dyes (about 8−20% of the total pollution load due to incomplete dye exhaustion) and auxiliary chemicals [3]. Besides, some dyes have been proven to have allergenic, toxic and/or carcinogenic properties that make them not only dangerous or potentially harmful to the environment, but also to human health in contact with human skin [4]. Th e continuous monitoring of dye content in wastewaters should therefore be an integral part of the technological process of textile materials and permitted concentrations regulated by law. Dyeing and printing should be carried out by achieving the maximum eff ect using a minimum amount of dye, not only for ecological reasons (reduced quantity of unused textile auxiliaries in wastewater), but also for economic reasons. Taking all of the above-mentioned facts into consideration, it is necessary to treat the wastewater by reducing the amount of harmful substances to legally limited concentrations. There are numerous methods for the treatment and disposal of textile industry wastewater: coagulation and fl occulation, chemical oxidation, biological treatment, membrane separation, reverse osmosis, etc. [3,5,6]. Although high quality effl uents can be obtained through water treatment with each of the previously mentioned processes, some of them have limitations/disadvantages, such as the use of excess chemicals, the formation of larger amounts of sludge that needs adequate care, and the incomplete removal of colour. Removal of dyes from wastewater is generally diffi cult. They are usually non-biodegradable, resistant to aerobic digestion, kinetično nadzorovana, pri čemer se predpostavlja proces psevdo drugega reda in da difuzija delcev ni edini proces, ki vpliva na hitrost adsorpcije. Negativne vrednosti standardne Gibbsove proste energije nakazujejo, da je reakcija adsorpcije spontana, višja negativna vrednost pri temperaturi 45 °C v primerjavi s 25 °C pa kaže, da je višja temperatura energijsko ugodnejša za adsorpcijo barvila Reactive Black 5. Ključne besede: adsorpcija, aktivno oglje, Reactive Black 5, kinetika, difuzija delcev and stable to light, heat and oxidising agents. One of the most commonly used and effective treatment methods for coloured water is adsorption, which is based on the accumulation of the substance from the solution in a solid phase. Adsorption is one of the best treatment methods due to its fl exibility, simplicity of design, and sensitivity to toxic pollutants. Furthermore, adsorption generally does not result in the formation of harmful substances [3, 7−9]. Activated carbon is the most popular adsorbent used for wastewater treatment due to its adsorption effi ciency and great capacity, although some other adsorbents are less expensive and easily available. It is effi cient in the removal of different types of dyes, including reactive dyes [10−15]. It is known that adsorption is a time-dependent process, and it is therefore necessary to know the adsorption rate for removal of dyes from wastewater. Th e most important factor in the design of adsorption systems is thus prediction of the rate of adsorption for a given system. Th is paper aims to present adsorption effi ciency of commercial activated carbon for the removal of Reactive Black 5 dye (RB5). Th e eff ect of contact time, initial dye concentration and temperature on adsorption were monitored. Th e pseudo-fi rst-order and pseudo-secondorder models were used to correlate the adsorption kinetic data, while intraparticle diff usion was used to evaluate the diff usion mechanism of the adsorption process. Activated carbon was also characterised before and aft er the adsorption of RB5 using a scanning electron microscope.

Materials
Reactive Black 5 dye (Drimarene Black R-3B, supplied by Clariant GmbH, C.I. 20505, chemical formula: C 26 H 21 N 5 Na 4 O 19 S 6 , M r = 991.82), was used for the adsorption experiment. Th e chemical structure of the dye prepared by the Accelrys Draw program is shown in Figure 1. One thousand milligrams per litre stock solution of dye was prepared by dissolving the required amount of dye in deionised water. Th e solutions of appropriate concentrations were prepared by diluting the stock solution with deionised water. Powdered activated carbon was purchased from the Croatian company Kemika. Adsorbent was dried in an oven at 105 °C for 24 hours and stored in a desiccator until it was used.

Batch mode adsorption studies
Adsorption studies were conducted by contacting 50 ml of dye solution of different initial concentrations (c 0 = 300, 500 and 700 mg/dm 3 ) with 0.2 g of activated carbon in glass bottles. The experiments were repeated three times under identical conditions to confi rm their repeatability. Experimental points presented in fi gures are the average values of three the repetitions. Suspensions were shaken at different contact times (15,30,45, 60, 90, 120, 180, 240 and 360 minutes and 16 hours until equilibrium was reached) with an impeller speed of 250 rpm at 25(±1) °C and 45(±1) °C (Heidolph Unimax 1010 with Incubator 1000). Experiments after 360 minutes for initial concentration of c 0 = 300 mg/dm 3 were not performed due to extremely low dye concentration after an adsorption time of 240 minutes. Suspensions were fi ltered after agitation through fi lter-paper blue ribbon. The residual liquid-phase dye concentration after adsorption was determined spectrophotometrically by monitoring the absorbance using a UV-Vis spectrophotometer (Lambda 20, Perkin Elmer) at a maximum absorbance wavelength (λ max = 598 nm). The calibration graph of absorbance versus concentration followed a linear Beer-Lambert relation. Th e amount of adsorbate adsorbed at any time t, q t (mg/g), and the amount of adsorbate adsorbed at equilibrium, q e (mg/g), were calculated using the following equation: where c 0 (mg/dm 3 ) represents the initial dye concentration, c t (mg/dm 3 ) represents the dye concentration in the liquid phase aft er appropriate time of adsorption and when equilibrium is reached (t = 16 hours), V represents the volume of the liquid phase (dm 3 ), and m represents the mass of the adsorbent (g). Percentage of adsorbed dye (% ads.) is calculated using the equation: where c s (mg dm −3 ) represents the concentration of the adsorbed dye in a solid phase (c s = c 0 -c t ).

Morphological analysis of adsorbent and dye-adsorbent samples
A fi eld emission scanning electron microscope (Mira, Tescan) was used for visualization of the adsorbent's morphology before and aft er adsorption. Th e accelerating voltage was 10.00 kV, while scanning was performed in situ on a sample powder. Samples were pre-coated with gold/palladium in a sputter coater. Optical micrographs were recorded using a Nikon Elipse E 400 microscope.

Eff ect of contact time, temperature and initial dye concentration on the adsorption process
Th e objective of this work was to assess the eff ectiveness of activated carbon for the treatment of   dye-rich textile wastewater with special focus on the reduction of colour polluters. Th e initial concentration of RB5 varied from 300 to 700 mg/dm 3 , which is a possible range of dye concentration in textile industry wastewater aft er the process of dying. Th e dependence of the amount of adsorbate adsorbed aft er appropriate time (q t ) versus time (t) is presented in Figure 2, while data are given in Table 1. It is evident from Figure 2 that a higher initial concentration of the dye increases the adsorption capacity and that the amount of adsorbed RB5 is higher at higher temperature for all concentrations. As equilibrium for the initial concentration c 0 = 300 mg/ dm 3 was reached very quickly, the amount of adsorbate adsorbed was monitored from 15 minutes to 4 hours, while for the other two initial concentrations were monitored from 15 minutes to 6 hours. Th e plots can be approximately divided into three regions. Th e fi rst region includes a very fast initial adsorption, probably governed by a rapid external diffusion process, which mainly includes adsorption of dye on the surface of activated carbon. Aft er this step follows the second region, with a milder and gradual increase of adsorbed dye, and the third region where a state of equilibrium was almost reached.
The sorption capacity at equilibrium at both temperatures increased from ca. 74 to ca. 173−174 mg/g with an increase of the initial dye concentration from 300 to 700 mg/dm 3 (Table 1). Initially, differences in adsorbed RB5 between two temperatures for the same initial dye concentration were more significant, while after a longer period of adsorption, quantities of adsorbed RB5 became very similar. Also, as expected, less time was needed to obtain high dye adsorption percentage values for lower concentra-tions of dye ( Figure 3). Thus, approximately 90% of RB5 was adsorbed at 25 °C for initial concentrations of c 0 = 300, 500 and 700 mg/dm 3 after 1, 3 and 6 hours, respectively (Table 1). At 45 °C, these times are signifi cantly shorter, and they were between 15 and 30 minutes for c 0 = 300 mg/dm 3 , 1 hour for c 0 = 500 mg/dm 3 and 2 hours for c 0 = 700 mg/dm 3 . The effi ciency of activated carbon for the adsorption of RB5 is also evidenced by the fact that more than 60% of dye is adsorbed after 30 minutes, regardless of initial concentration and temperature.

Kinetics of adsorption
Kinetic studies were performed in order to investigate the mechanism of adsorption and potential rate controlling steps. Kinetic study is important to an adsorption process because it depicts the uptake rate of the adsorbate and controls the residual time of the whole adsorption process for a given system. Th e experimental data were analysed using three kinetic models: pseudo-fi rst-order and pseudo-second-order kinetic models, and an intraparticle diffusion model. Th e pseudo-fi rst-order and pseudo-second-order models are used most frequently for determining of kinetic parameters.

Pseudo-fi rst-order kinetic model
Lagergren [16] proposed a rate equation for the sorption of solute from a liquid solution based on solid capacity. Th e kinetic model of this the most widely used rate equation is expressed using the following equation: where k 1 represents the rate constant of the pseudofi rst-order (min −1 ). Integrating this equation for the boundary conditions t = 0 to t = t and q t = 0 to q t = q t results in: In(q e -q t ) = In q e -k 1 . t (4).
Th e kinetic constant k 1 can be determined by plotting ln(q e −q t ) against time (t), and if the fi rst-order equation is applicable, the plot should give a linear relationship, and facilitates the calculation of the rate constant of pseudo-fi rst-order (k 1 ) from the slope and amount of adsorbate adsorbed at equilibrium (q e,calc ) from the intercept. Th e values of the constants of the pseudo-fi rst-order model for adsorption of RB5 onto activated carbon are given in Table 2.
Th e values of the correlation coeffi cient (R 2 ) obtained from the linear plot (Eq. 4) are relatively high (from 94.6% to 99.8%), the exception being the initial concentration of c 0 = 300 mg/dm 3 at 25 °C (71.1%). However, for all concentrations and at both temperatures, there is considerable disagreement between the experimental and calculated values of the amount of adsorbed RB5 at equilibrium (q e,exp and q e,calc ). Th is suggests that this sorption system is not a fi rst-order reaction and that a pseudo-second-order model might provide a better correlation of the data.

Pseudo-second-order kinetic model
Ho and McKay [17,18] developed a second-order equation based on adsorption capacity. Th is kinetic model is illustrated by the following equation: where k 2 represents the rate constant of the pseudosecond-order (g mg −1 min −1 ). Integrating the equation (5) for the same boundary conditions used for the fi rst-order results in the equation presented below in the linear form:  If the pseudo-second-order equation is applicable, the plot of t/q t against time t should give a linear relationship and facilitates the calculation of the amount of adsorbate adsorbed at equilibrium (q e,calc ) from the slope, and then the rate constant of the pseudo-second-order (k 2 ) from the intercept.
Th e values are given in Table 2, while linear plots are presented in Figure 4. Based on this model, calculated q e values (q e,calc ) and experimental equilibrium values (q e,exp ) demonstrate a much better correlation (Table 2). Moreover, the values of correlation coeffi cients are very close to 1 (higher than 99.4%) for all initial dye concentrations and at both temperatures. Figure 5 shows the good correlation of the data with the pseudo-second-order equation. Th e experimental points are shown together with the theoretically generated lines, and they fi t nicely for all concentrations and at both temperatures, although slightly better for the temperature of 45 °C. Th e adsorption of RB5 dye on commercial activated carbon is thus kinetically controlled assuming a pseudo-secondorder rather than a pseudo-fi rst-order process. Th e pseudo-second-order model assumes chemical sorption or chemisorption as the rate-limiting process. Maximum k 2 values were obtained for the smallest initial dye concentration (c 0 = 300 mg/ dm 3 ), while values for temperature of 45 °C are approximately 2−4 times higher than those for 25 °C. Also, as time approaches zero, according to the pseudo-second-order model, the initial adsorption rate h (mg g −1 min −1 ) can be calculated using the following equation [18,19]: .
Th e obtained h values are also presented in Table 2.
As for k 2 , values of the initial adsorption rate increase with a decrease of initial dye concentration, and increase with an increase in temperature.

Intraparticle diff usion model
We used an intraparticle diff usion model to evaluate the diff usion mechanism for adsorption of RB5 on activated carbon. Most adsorption reactions are carried out using a multistep mechanism and involve several steps: (i) external mass transfer of the adsorbate from the solution to the adsorbent surface; (ii) adsorption at a site on the adsorbent surface; and6 (iii) intraparticle diff usion of the adsorbate in the pores of the adsorbent and adsorption on the site.
Step (ii) is oft en assumed to be very fast, and thus cannot be treated as a rate limiting step. Generally, the rate of adsorption is limited by external mass transfer for a system with poor mixing, low adsorbate concentration, its high affi nity to the adsorbent and the small adsorbents particles. Th e adsorption of large molecules, for which longer contact time is needed to reach equilibrium, is always considered to be diff usion controlled by external fi lm resistance and/or internal diffusion mass transport or intraparticle diff usion [20]. Th eoretical treatments of intraparticle diff usion yield complex mathematical relationships that diff er in form as functions of the geometry of the adsorbent particle. Th e intraparticle diff usion model is based on the following equation [19,20]: where k i represents the intraparticle diff usion rate constant (mg g −1 min −0.5 ). If the intraparticle diffusion is a rate limiting step of adsorption, i.e. intraparticle diffusion controls the rate of adsorption, then plot q t versus t 0.5 should be linear and pass through the origin. If the plot of q t versus t 0.5 exhibit multi-linearity, this indicates that two or more rate controlling steps occur in the adsorption processes [20−22]. Figure 6 shows the root time plots for the adsorption of RB5 onto activated carbon at temperatures of 25 and 45 °C. It is evident from the above fi gure that the plots are not linear, i.e. that they exhibit multi-linearity with several sections. It can thus be concluded that intraparticle diff usion is not the only process that infl uences the adsorption rate and that multiple steps took place during the adsorption process. Few stages can be distinguished during the dye adsorption. Th e adsorption rate is initially higher and corresponds to instantaneous adsorption, probably due to an electrostatic attraction between the dye and the external surface of the adsorbent. Th e amount of adsorbed substance on the adsorbent and diff usion decreases over time, and represents a gradual adsorption stage where diff usion rates decreased by increasing the contact time. Th is process usually includes the intraparticle diff usion of the molecules through the pores of the adsorbent. Th at stage is followed by the equilibrium stage when dye molecules occupy all active sites of the adsorbent [19]. Diff usion in the pores of the adsorbent is usually determined by the fact that there are several diff erent pore sizes in the adsorbent, and that controlling regions correspond to the dye diff usion to the activated carbon pores of diff erent dimensions.
Since the dye molecules diff use into the inner structure of the adsorbents, the pores for diff usion become smaller and thus the free path of the molecules in the pore decreases.

Thermodynamics
Standard Gibbs free energy (ΔG 0 , kJ/mol) values of the adsorption process can be calculated using the equation: (a) (b) Figure 6: Root time plots for the adsorption of RB5 for three initial dye concentrations (c 0 : 1 300 mg/dm 3 , 2 500 mg/dm 3 , 3 700 mg/dm 3  where R represents the universal gas constant and T represents temperature. K c represents the equilibrium constant calculated from the concentration of the dye adsorbed on the solid at equilibrium (c s / mg/dm 3 ) and the concentration of the dye when equilibrium is reached in the liquid phase (c e / mg/dm 3 ) [23]:
The negative values of ΔG 0 indicate that the adsorption reaction is spontaneous. Thus, ΔG 0 values given in Table 3 refl ect the feasibility of the process, and that the adsorption of RB5 onto activated carbon was a spontaneous process in nature, in which no energy input from outside of the system was required. The higher negative value refl ects a more energetically favourable adsorption. It can thus be concluded that adsorption at 45 °C is energetically more favourable than at 25 °C.

Morphological analysis of the adsorbent and dye-adsorbent samples
In order to perform a morphological analysis of activated carbon before and aft er adsorption, photographs were taken using scanning electron microscopy (SEM). It is evident from recorded photographs of activated carbon that the particles diff er morphologically and have diff erent pore sizes, as assumed by the results of the intraparticle diff usion model (Figure 7). Figure 8 shows the appearance of activated carbon after the adsorption of RB 5 (adsorption time of 4 hours at a temperature 45 °C) for initial dye concentrations of c 0 = 300 mg/dm 3 (Figure 8a) and c 0 = 700 mg/dm 3 (Figure 8b). It is evident from the images that dye is adsorbed on the surface of activated carbon. For the initial concentration c 0 = 300 mg/dm 3 , some parts of the surface are almost completely straight, while the other parts are uneven. Th e activated carbon surface for the initial dye concentration of c 0 = 700 mg/dm 3 is fl atter on most parts, an indication that activated carbon is almost completely covered with dye molecules. From this it can be assumed that, for a higher initial concentration, a certain amount of dye is further adsorbed on activated carbon.

Conclusion
According to the results achieved, the adsorption of Reactive Black 5 onto activated carbon was very fast. This study also confi rmed that activated carbon is a very effi cient adsorbent and that adsorption at higher temperature is greater. For all initial concentrations, approximately 90% of RB5 was adsorbed at 25 °C after 1 to 6 hours. On the other hand, these times are signifi cantly shorter at a temperature of 45 °C, with the maximum value of 2 hours for the initial concentration of c 0 = 700 mg/dm 3 . Furthermore, the adsorption of RB5 dye on commercial activated carbon was kinetically controlled assuming a pseudo-second-order rather than a pseudo-fi rst-order process. This study also revealed that intraparticle diffusion was not the only process that infl uenced the adsorption rate and that multiple steps took place during the adsorption process. Finally, the negative values of ΔG 0 indicate that the adsorption reaction was spontaneous in nature and that adsorption at 45 °C was energetically more favourable than at 25 °C. The adsorption of Reactive Black 5 dye on activated carbon was also confi rmed by scanning electron microscopy.