Scholarly article on topic 'Investigation of BTEX compounds adsorption onto polystyrenic resin'

Investigation of BTEX compounds adsorption onto polystyrenic resin Academic research paper on "Chemical sciences"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{Adsorption / "Polystyrenic resin" / Isotherm / BTEX}

Abstract of research paper on Chemical sciences, author of scientific article — Thobeka P. Makhathini, Sudesh Rathilal

Abstract In this study, the adsorptive capacity of polystyrenic resin was evaluated for the removal of benzene, toluene, ethylbenzene and isomers of xylene (BTEX) from an aqueous solution. Batch studies were performed to evaluate the effects of various experimental parameters such as mixing strength, contact time, internal diffusion, adsorbates and initial concentration on the removal of the BTEX compounds. The equilibrium isotherms for the adsorption of the adsorbates on the PAD 910 polystyrenic resin were analyzed by the Langmuir and linearized Dubinin–Radushkevich models at pH of 5.86. The Langmuir model fitted the data adequately; is concluded that the latter is the most practical model in representing the adsorption of aromatic compounds. The Langmuir model indicated that resin has the highest adsorption capacity of 79.44 mg/g. At temperature of 25 °C, resin was found to adsorb 98% of benzene, 88% of toluene, 59% of ethylbenzene, 84% m-;p-xylene and 90% o-xylene at an initial concentration of 14.47 mg/l. The pseudo-second order rate model fitted better to the adsorption kinetics. The thermodynamic analysis resulted in a negative equilibrium enthalpy change suggesting an exothermic process.

Academic research paper on topic "Investigation of BTEX compounds adsorption onto polystyrenic resin"

Accepted Manuscript

Investigation of BTEX compounds adsorption onto polystyrenic resin Thobeka P. Makhathini, Sudesh Rathilal

PII: S1026-9185(16)30026-9

DOI: 10.1016/j.sajce.2017.03.001

Reference: SAJCE 23

To appear in: South African Journal of Chemical Engineering

Received Date: 30 May 2016 Revised Date: 22 November 2016 Accepted Date: 1 March 2017

Please cite this article as: Makhathini, T.P., Rathilal, S., Investigation of BTEX compounds adsorption onto polystyrenic resin, South African Journal of Chemical Engineering (2017), doi: 10.1016/ j.sajce.2017.03.001.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Investigation of BTEX compounds adsorption onto polystyrenic

1 2 Thobeka P. Makhathini and Sudesh Rathilal

1Mangosuthu University of Technology, Faculty of Engineering, Department of Chemical Engineering, P.O. Box 12363 Jacobs, Durban - Umlazi 4026, South Africa, Email: thobeka@mut.ac.za

2Durban University of Technology, Faculty of Engineering and the Built Environment, Department of Chemical Engineering, P.O. Box 1334, Durban 4000, South Africa, Email: rathilals@dut.ac.za

In this study, the adsorptive capacity of polystyrenic resin was evaluated for the removal of benzene, toluene, ethylbenzene and isomers of xylene (BTEX) from an aqueous solution. Batch studies were performed to evaluate the effects of various experimental parameters such as mixing strength, contact time, internal diffusion, adsorbates and initial concentration on the removal of the BTEX compounds. The equilibrium isotherms for the adsorption of the adsorbates on the PAD 910 polystyrenic resin were analyzed by the Langmuir and linearized Dubinin-Radushkevich models at pH of 5.86. The Langmuir model fitted the data adequately; is concluded that the latter is the most practical model in representing the adsorption of aromatic compounds. The Langmuir model indicated that resin has the highest adsorption capacity of 79.44 mg/g. At temperature of 250C, resin was found to adsorb 98% of benzene, 88% of toluene, 59% of ethylbenzene, 84% m-;p-xylene and 90% o-xylene at an initial concentration of 14.47 mg/l. The pseudo-second order rate model fitted better to the adsorption kinetics. The thermodynamic analysis resulted in a negative equilibrium enthalpy change suggesting an exothermic process.

Key words: Adsorption, polystyrenic resin, isotherm, BTEX

1. INTRODUCTION

The mass of organic wastes produced in our society and the disposal cost has persistently increased over the years and this creates a challenge to process, civil and environmental engineers to develop better cost effective and environmentally secured treatment and discarding methods. The dangers of these organic compounds make their reduction and elimination necessary. Organic elements are commodities and raw materials of industries like petrochemical, oil refinery, explosives, dye, pigment and pesticides industries [1]. The unacceptable discharge of organic compounds in water banks for a long period of time can cause worsening of water environments, while its consumption by human beings and animals causes liver function failure, kidney damage, central nervous coordination injury, diarrhoea and secretion of murky urine [2]. Currently, an approval of trade effluent for release into South African local municipality sewage clearance system does not specify the limits for the BTEX compounds. It is anticipated that, in the upcoming times, stringent regulations in South Africa will be compulsory on the concentrations of these compounds in wastewater before discharge. BTEX compounds have already been selected by the US Environmental Protection Agency as main concern chemicals that need to be condensed to low levels of concentration

before discharge [3]. Treatment of BTEX in wastewater becomes a fundamental part of wastewater treatment for oil and chemical industries.

An alternative for the removal of BTEX from a solution is by adsorption, of which the recovery of such compounds for likely recycling to the manufacturing process is permitted. The adsorption process is one of the prominent treatment options for the reduction of BTEX pollutants from aqueous media since it is possible to recover the adsorbent and adsorbate [4]. The adsorption of BTEX compounds can be achieved by macro reticular resin [5, 6] and activated carbon [7, 8]. Other adsorbents may include carbon nano tube, diatomite [4, 9] and organo-clay. All of the studies mentioned above investigate removal at higher concentration. None have been recorded on the low range concentration (less than 1 ppm) adsorption of BTEX using a polystyrenic resin. Hence, it was decided to embark on this investigation.

The main objective of this study was to evaluate the ability of untreated polystyrenic PAD 910 resin in the batch process to remove BTEX compounds from aqueous solution. In this work, the influence of mixing strength, contact time, internal diffusion, temperature, adsorbates and initial concentration on the removal of the BTEX compounds were investigated. The Langmuir and linearized Dubinin-Radushkevich models were used to examine fitness with the equilibrium data. The kinetic models and intra-particle diffusion models were also applied to evaluate the nature of the sorption process.

1.1 Isotherm models

The adsorption isotherms are defined by many mathematical formations often based on empirical fittings while simplifying the physical properties and processes involved in the adsorption. The major models that have mostly been used in fitting equilibrium adsorption isotherm of hydrocarbons in the literature are highlighted in this section.

Langmuir isotherm model

This is the simple and most commonly used model. It is a semi-emperical isotherm originating from a proposed kinetic mechanism [10]. This is normally based upon four assumptions that are summed up in a homogeneous sorbent: (a) Localized adsorption; molecules are adsorbed at a set number of localized sites; (b) There is no interaction in between adsorbed molecules; (c) Monolayer adsorption: one adsorbate molecule can be held per site and (d) Equivalence of energetic sites.

The model is derived from rational considerations, hence is defined as:

Where: qe is the mass of adsorbate adsorbed per unit mass of adsorbent (mg adsorbate/g activated carbon); KL is the Langmuir constant; Ce is the equilibrium concentration of adsorbate in solution after adsorption (mg/L) and qmax is the maximum amount of adsorption equivalent to complete monolayer coverage on the surface (mg/g).

The Dubmm-Radushkevish isotherm model (D-R model)

The D-R model is often utilized to give insight into the adsorption mechanism by finding the apparent free energy of the adsorption process. The model is given by Dubinin [11]:

i l.ii

qe = qD exp(-/?£2)

Where:

In the above equations, qe and qD are the equilibrium adsorption capacity and the maximum adsorption capacity under optimized conditions in the D-R isotherm model in mg/g

respectively; Ce is the equilibrium concentration in mg/L; /? is a constant related to energy in

molVkJ" and e is the Polanyi potential. The linearized form of Eq. (2) is used in obtaining the parameter /? from which the mean free energy in kl/mol is obtained by [12].

2. MATERIALS AND METHODS

2.1 Adsorbate material

The aqueous solution used in this research was synthetically prepared with a composition which is similar to that discharged by the local petrochemical companies to the mixing point before it is pumped to sea. The samples were collected randomly for six months so as to determine the reference average concentrations for a long time period. All samples were taken at different times of the day, which included steady and unsteady state operating conditions. Dilution of required quantities of benzene, toluene, ethyl-benzene and o-, m-, p-xylene were prepared to make a stock solution in the required volume of distilled water to find solutions of adsorbate for different initial concentrations (C0) within the range of 0.35 -14.47 ppm. Ten samples of different initial concentration were prepared within the given range.

Fresh solutions were prepared on a daily basis with the initial concentration of the adsorbate calculated before the start of each experiment. The pH value for the aqueous solution containing the mixture of benzene, toluene, ethyl-benzene and o-, m-, p-xylene (0.35 to 14.47 ppm) was measured and found to be between 5.86 and 5.99 before the adsorption tests were conducted. All pH measurements were done with a pH/conductivity meter (Orion Star A215), where an average of two results was recorded. Every flask was closed with a cap and overturned several times to blend the materials carefully and then allowed to stabilize for 15 minutes in the pyrex conical flask until experiments were run.

2.2 Adsorbent

PuroSorb PAD 910 polystyrenic adsorbent is a synthetic polymer that is highly cross linked and also has a highly porous structure as part of the polymer matrix. This adsorbent in many cases may replace carbon which is a common generic adsorbent with an advantage of being regenerated. The graded resin utilized in this investigation was sourced from Purolite Co. Ltd. in China. The pH of the PAD 910 polystyrenic was measured after 2.5 g of resin was mixed with 10 ml of distilled water and shaken at 25°C for 6 hrs. The procedure was replicated from Battelle [13]. The mixture was filtered and the pH was measured as 6.85. The resin was stored in a cool dry place (approximately 22°C) in the Laboratory until experiments were carried out.

2.3 Adsorption Kinetic Study

Dilutions of the stock solution were prepared in distilled water (simulated waste water) in a series of 250 ml bottles. From 100 mg to 400 mg of PAD 910 resin were dosed into the series of sample bottles containing 50 ml of the simulated wastewater and the bottles were placed in an incubator at 25°C (±1°). The samples were placed in an orbital shaker (Stuart - SSL1) running at 150 rpm, having determined this was the mixing strength at which external diffusion was minimized negligibly. The simulated wastewater was prepared from stock

BTEX solutions and then stored in the refrigerator at 4 C to prevent any possible evaporation since the BTEX compounds are highly volatile.

2.4 Batch studies

Batch studies were performed to obtain the equilibrium data for the adsórbate and the adsorbent (PAD 910 polystyrenic resin). The adsorbates were the mixture of 99.9 % benzene, 99.7% toluene, 99.9% ethylbenzene and 99.8% of each isomer of xylene from Merck Chemical Ltd, SA. The batch technique was used because of its straight forwardness to obtain equilibrium data. The initial solute concentration was recorded before adding the sample to the polystyrenic resin. Adsorption experiments for short time kinetics and isotherms were repeated twice and the duplicate adsorption data is reported, while the repeatability of the different instruments used in analyzing the four BTEX compounds are reported in terms of % relative standard deviation (RSD). Figure 1 protocol that was followed to carry out the adsorption kinetic experiments.

Figure 1: Adsorption Kinetic Experimental Protocol

For every adsorption data point, a 50 ml aliquot of the adsórbate mixture had an initial concentration of benzene, toluene, ethylbenzene, m-;p-xylene, o-xylene of 3.70 mg/1, 6.66 mg/1, 0.34 mg/1, 2.57 mg/1 and 1.20 mg/1 respectively. The adsorbent-to-simulated waste water ratio on a mass/volume basis was varied from 1 to 5 [14]; when it was introduced into a series of 100 ml Pyrex conical flasks at a temperature of 25°C. The conical flasks were positioned and shaken in a shaker where equilibrium was attained since no further adsorption occurred after 6 hrs of residence time and at a steady stirring speed of 150 revolutions per minute. The samples were enclosed using aluminium foil during the experiment and the supernatant solution filtered after different time intervals of agitation through Whatman microfiber filters. The concentration of the filtrate left was then determined using GC/MS and recorded, as shown in figure 1. The adsorbent capacity for BTEX adsorption was calculated using Eq. (4) and the removal percentage used Eq. (5):

{C0-Ce)v

100 (c0-ce) Co

Where C0 is the initial sorbate concentration (mg/1), Ce the equilibrium sorbate concentration (mg/1), M is the mass of the adsorbent (g) and V i s the volume of solution (1).

3. RESULTS AND DISCUSSION

3.1 Adsorbent physical properties

Particle size distribution analysis was manually done. Initially, a resin sample of 400 g was analyzed. This sample was passed through different sieves and an amount of fine and course particles were measured in the US Standard mesh size number 30-45. The nominal particle size was found to be 0.45 mm. Density of the adsorbent is particularly important for adsorption. The Brunauer-Emmett-Teller (BET) surface area of PAD 910 polystyeric resin was measured and found to be 480 m2/g, which was close to the manufacture's specification of 510 mVg. The porous structural properties of the samples were analyzed by obtaining the nitrogen adsorption and desorption isotherms at 77 K using the Autosorb 1 specific surface area analyzer. The average pore diameter was given as 18 A for PAD 910 polystyeric resin by the manufacturer. These results showed that this adsorbent is reasonably good for adsorption.

3.2 Adsorbent characterization

Surface morphological images obtained by scanning electron microscopy (SEM) for some of the samples used in this study are shown in Figure 2.

3300 2300 1300 Wavelength (cm1)

Figure 2: (a) SEM Image and (b) FTIR spectra of PAD 910 polystyrenic resin

The image in Figure 2 (a), gives evidence to the porous nature of the resin. However there was essentially little porosity seen in the commercial PAD 910 polystyrenic resin. The porous structure of the sample of PAD 910 resin appears to be honeycomb (or spongy) like, and it is

in agreement with the pores size distribution [15-17]. The FTIR spectra in Figure 2 (b) displayed a peak between 2800 ~ 3200 cm-1. The peaks ascribed to the stretching vibration of groups such as -OH associated with the aromatic and carboxylic functional groups [14]. The band around 1150 cm-1 is attributed to the stretching vibration of C-O associated with the lactonic group [18]. The band between 1600 ~ 1800 cm-1 is ascribed to the oscillation of C=O associated with carbonyl and carboxylic groups [15-17].

3.3 Effect of mixing strength

The mixing strength is a major factor in the adsorption process. This is due to the fact that external mass transfer of aromatic hydrocarbon molecules from the bulk liquid to the surface of the adsorbent is dependent on the dynamics of the system whereby the process occur due to the boundary layer effect at the solid-liquid interface [19]. The boundary layer distorts surrounding non-viscous flow, which will then limit the adsorption process [20].

100 90

♦ ♦ ♦

5 80 X.

X £ J A 4 ♦ Benzene

£ 70 v

• Toluene ■ Ethylbenzene Xo-Xylene

40 Am-;p-Xylene

0 50 100 150 200

Mixing Strength, rpm

Figure 3: Effect of mixing strength on adsorption of a mixture of BTEX for 6 hrs (0.25 g of resin, 25 ml simulated waste water with 3.69 ppm Benzene, 6.66ppm Toluene, 0.34 ppm Ethylbenzene, 2.56 ppm Ethylbenzene, 2.56 ppm m-;p-Xylene and 1.18ppm o-Xylene)

Figure 3 shows the removal percentage against the varying mixing strengths. It demonstrates that an increase in the mixing strength increases the % removal of the BTEX compounds. This is attributed to the increased turbulence, which then reduces the boundary layer effect.

Given these results, subsequent experiments were conducted at a mixing strength of 120 rpm, 150 rpm and 180 rpm, in order to comfortably assume negligible external diffusion limitations. Figure 4 shows data obtained to verify the validity of this assumption in a non competitive kinetic experiment over the time regime to be studied. The trend shows that at the mixing strength of 150 and 180 rpm, the same result of Ct/Co is obtained hence it can be confirmed that adsorption is not dependant on the mixing strength beyond 120 rpm. Hence subsequent experiments were carried out at a fixed mixing strength of 150 rpm.

1.2 1 0.8 0.6 0.4 0.2 0

♦ 120 rpm 150 rpm ▲ 180 rpm

20 25 30 Time, min

Figure 4: Effect of mixing strength on adsorption of Toluene onto 0.25 g resin. Initial concentration ~ 6.6 ppm)

3.4 Effect of internal diffusion

With the initial concentration of BTEX aqueous solution of 14.47 ppm, a long term kinetic study was conducted with PAD 910 polystyrenic resin. Figure 5 shows the evolution of the BTEX concentration with time. The equilibrium concentration of 2 ppm was achieved for the adsorbent after 3 hrs. The global kinetic parameters were obtained and are given in Table 1.

Figure 5: Long Term BTEX adsorption on 0.25 g of PAD 910 polystyrenic resin in simulated waste water

Short term adsorption is based on the initial phase of adsorption which is 45 minutes and the long term adsorption is the final phase of the adsorption which occurred at 6 hours.

Table 1: Global rate parameters for short term BTEX adsorption onto polystyrenic resin

Model Benzene Toluene Ethylbenzene m-;p-Xylene o-Xylene

Pseudo 1st Order

ki (min1) 0.10 0.22 0.11 0.16 0.06

qe l (mg/g) 0.13 0.25 0.19 0.19 0.14

r2 0.966 0.992 0.961 0.956 0.987

Pseudo 2nd Order

k2 (g.mg^.min1) 0.98 0.5 0.13 0.25 0.41

qe2 (mg/g) 1.14 2.5 0.08 0.22 0.19

r2 0.993 0.995 0.983 0.990 0.994

Elovich

b (g.mg"1) 3.86 10.25 1.22 62.44 84.33

a (mg.g^.min1) 3.45 X 1014 2.09 X 1032 3.23 X104 00 00

r2 0.442 0.624 0.852 0.421 0.392

Obviously, the Elovich model does not fit the experimental data adequately as shown by the low correlation coefficients hence it was not considered in subsequent analyses. The choice between the pseudo-first order and pseudo-second order was however relatively more subtle given the very high correlation coefficients. The pseudo-second order model as shown in Figure 6 (b) has been identified as the best performing model for this study based on the following reasons.

A plot of the predicted residual concentration at different sampling times versus the experimental residual concentrations for all adsorption kinetics experiments is presented models in Figure 5. It is observed that the pseudo-second order model represents the experimental values better for the whole concentration range.

Figure 6: Validation of Kinetic Model Choice a) pseudo-first order; b) pseudo-second order

The Pseudo-first order model also seemed to represent the experimental data fairly well. Another validity criterion commonly used in the literature for pseudo-first order is the plot of In(qe — qt) versus t, which yields a linear relationship in which the intercept should give a value close to In qe [21, 22]. Performing this validity test did not exactly yield values of qe

that were close to those obtained for the pseudo-first order model in all cases. In addition, one way ANOVA was used to validate the effect of pseudo first and second order models on the experimental data. There was a significant effect of p = 0.027 < 0.05 level for all conditions. In addition, a post hoc comparison was done which indicated that the mean score for BTEX concentration had a significant difference for Figure 6 (a) while there is no significant difference in Figure 6 (b).

It was also confirmed that the adsorption process was extremely fast within the first 45 mins in both cases, which was followed by a slower transport stage. This suggested a shift of the BTEX molecule transport regime from macropore/mesopore to the micropore channel of the resin. A shorter time regime experiment method was then designed to capture the process better and also for analysis of the diffusion process. An aqueous system of initial BTEX concentration was reduced by 50% to 7 ppm and then was adsorbed using PAD 910 polystyrenic resin and the concentration-time dependence plots are shown in Figure 7.

100 Time, min

♦ Experimental — — — pseudo-2nd-order model

Figure 7: Short Term BTEX adsorption on 0.25g of PAD 910 polystyrenic resin

When assuming that Ce tends to zero, the linear Eq. (5) was obtained from which the rate parameter in terms of the residual concentration, k'2 could be obtained.

The rate parameters are reported in Table 2.

L = k't + L

Table 2: Global parameters for short term BTEX adsorption

Model Benzene Toluene Ethylbenzene m-;p-Xylene o-Xylene

Pseudo 2nd Order (RESIN) k2 (L.mg^.min"1) 0.03 0.52 0.17 0.16 0.22

r2 0.923 0.992 0.962 0.976 0.994

3.5 Effect of initial concentration

The effect of initial BTEX concentration on the adsorption process was investigated using three different batches of 25 ml BTEX simulated wastewater each treated with 0.25 g of PAD 910 polystyrenic resin. The quantity-time dependence plot is shown in figure 8.

^ 30 >

----«4

/A •

• ^ «r r - *

- V--*

7ppm Experimental 14ppm Experimental 21ppm Experimental

- 7ppm Pseudo-2nd order -14ppm Pseudo-2nd order

- 21ppm Pseudo-2nd order

120 Time, min

Figure 8: Plot of BTEX sorbed capacity versus time showing initial concentration effect

Rate parameters obtained from the experimental data are highlighted in table 3, while figure 8 shows the typical sorption data curves for the effect of initial BTEX concentration on the quantity of BTEX adsorbed per time.

Table 3: Rate parameters for initial BTEX concentration effect on the adsorption process

Pseudo-2nd order Model 7 ppm 14 ppm 21 ppm

k2 (L.mg^.min1) 0.39 0.044 0.021

h0 (mg.g^.min1) 1.8 5.6 14.3

r2 0.99 0.94 0.92

Table 3 shows that the initial concentration influences the rate parameters, which demonstrates that the adsorption process is controlled by the diffusive mass transport of BTEX compounds onto the adsorption remaining "active sites". The dependence is inconsistent with the diffusive mass transport theory with a negative dependence of the rate constant with the concentration trend. However, the unit of the second order rate constant demonstrate that it has an inverse relation to the BTEX concentration in the aqueous phase. It may be concluded that the rate constant is positively influenced by the concentration of BTEX in the aqueous phase. This positive dependence of the rate on the BTEX concentration is attributed to the greater concentration driving force for the transport process. A simple inspection of figure 8 shows that the initial concentration influenced the contact time required to reach equilibrium. Exaggerated results for residual concentration and rate of adsorption at 7 ppm are observed, this result could be attributed to the fact that less quantity is adsorbed at

lower concentration which is logical because there is less amount of BTEX available in the aqueous system.

In terms of the initial concentration and the initial rate of adsorption (h0) as time tends to zero, the rate law of the adsorption process could be written as

A non linear regression of h0 against the initial concentrations was used to determine the order (n) of the process as 1.23. This may be the reason behind both the pseudo-lst order and the pseudo-2nd order models predicted data giving a relatively high coefficient of regression with respect to the experimental data in all cases.

4. STUDY OF THE ADSORPTION EQUILIBRIUM

4.1 Effect of internal diffusion

The adsorption isotherm of BTEX at 25°C onto PAD 910 polystyrenic resin was studied and the experimental data with the Langmuir model fitted plot is shown in figure 8.

4.90 5.10 5.30 5.50 5.70 5.90 6.10

Equilibrium Concentration, ppm

♦ Experimental RESIN

Langmuir RESIN

Figure 9: BTEX adsorption isotherm on PAD 910 polystyrenic resin at 25 C (0.25 g adsorbent in 25 ml simulated wastewater of initial concentration of 15 ppm)

The isotherm parameters from the Langmuir and D-R models are highlighted in table 4. From the maximum adsorption capacity parameter qm obtained from the Langmuir model, it is evident that there is a direct correlation between this parameter and the specific surface area (SSA) of the adsorbent. This is due to the adsorption capacity being reached as the number of 'active sites' available for the monolayer adsorption (as assumed by Langmuir) is saturated. The amount of the adsorption 'active sites' is expectedly proportional to the specific surface area. However, on normalizing the maximum adsorption capacity by the SSA, the influence of the pore size distribution is demonstrated.

Table 4: BTEX adsorption isotherm parameters for PAD 910 Resin

Langmuir Linear D-R

Kl qm SSA ß X 108 qD E

Adsorbent (L/mg) (mg/g) 2 (m /mg) R2 (mol2/J2) (mg/g) (kJ/mol) R2

RESIN 0.79 78.9 0.66 0.95 1.77 63.23 5.21 0.98

4.2 Effect of adsorbate

Figure 10 shows the adsorption isotherm data obtained from a comparative competitive study of the equilibrium adsorption of benzene, toluene, ethylbenzene and isomers of xylene with the adsorption conditions, while the parameters are tabulated in table 5.

■E 6

3.4 3.9 4.4 4.9

Equilibrium Concentration, ppm

5.00 5.50 6.00 6.50 7.00 Equilibrium Concentration, ppm

♦ Benzene — — — Benzene Langmuir

♦ Toluene — — — Toluene Langmuir

0.10 1.10 2.10 Equilibrium Concentration ,ppm

♦ Ethylbenzene

— — Ethylbenzene Langmuir

I 50.00

a 30.00

Equilibrium Concentration, ppm

♦ m-;p-Xylene

— — m-;p-Xylene Langmuir ▲ o-Xylene

— — — o-Xylene Langmuir

*E3 m,

■a <

Figure 10: Benzene, Toluene, Ethylbenzene and Xylene Adsorption isotherm on PAD 910 polystyrenic resin at 250C (0.25 g in 25 ml simulated wastewater of initial concentration of 7 ppm)

Table 5: Benzene, Toluene, Ethylbenzene, m-;p-Xylene and o-Xylene adsorption isotherm parameters

Langmuir Linear D-R

Kl qm qm ß X 108 qD E

Adsorbate (L/mg) (mg/g) (mmol/g) R2 (mol2/J2) (mg/g) (kJ/mol) R2

Benzene 17.23 9.89 0.078 0.94 0.93 10.56 7.13 0.96

Toluene 6.45 13.67 0.45 0.98 2.78 9.88 4.45 0.93

Ethylbenzene 3.67 12.9 0.11 0.99 5.78 14.45 3.98 0.94

m-;p-Xylene 21.78 11.45 0.67 0.91 3.12 11.56 4.99 0.99

o-Xylene 19.45 10.45 0.56 0.97 2.23 8.55 5.87 0.92

As demonstrated in Table 5, the equilibrium capacity of the PAD 910 polystyrenic resin used was greater for toluene. Compared to the other compounds, toluene has a smallest size which makes it easier for its molecules to diffuse into more micropores for the adsorption process to take place. This further gives proof to the argument that there is a form of molecular sieving in the adsorption process with some portion of the micropores being inaccesssible to some aromatic molecules. In a competitive study Valderrama, et al. [23] also observed a similar trend of the form of molecular sieving towards larger aromatic hydrocarbons.

4.3 Effect of temperature

The effect of adsorption of BTEX at different temperatures was investigated using PAD 910 polystyrenic resin and the experimental data is provided in Figure 11 while the parameters obtained from the isotherm models are given in Table 6. Evidently there is an increase in the adsorption equilibrium capacity as the temperature increased. The increasing sorption capacity is further indicative of the chemisorption process. For any chemical process, the positive influence of temperature has been well documented, as explained by the Arrhenius equations. Furthermore, according to Rodriguez and Linares [24], water molecules attached to carbon surface oxygen groups through hydrogen bonding and may cause steric hinderance to the adsorption of aromatic compounds by the adsorbents. This steric hinderance could be reduced at an elevated temperature and hence an observed increased adsorption equilibrium capacity.

Equilibrium Concentration, ppm

♦ Exp. 35 C ------Langmuir 35 C •

Exp. 25 C Langmuir 25 C ■

Exp. 45 C Langmuir 45 C

Figure 11: BTEX Adsorption Isotherm on PAD 910 polystyrenic resin at different

temperatures (0.25 g adsorbent in 25 ml simulated wastewater with initial concentration of 15 ppm)

However, the adsorption coefficient decreased as temperature increased. This may be indicative of decreased adsorption intensity associated with an exothermic process. Site [19] in his review paper, concluded that due to the dependence of both sorption coefficients and solubility with temperature, the measured effect of temperature on sorption isotherm is a result of combined sorption and solubility contributions. Unfortunately, there was no data available for the solubility of some BTEX compounds at the elevated temperatures of 350C and 450C, which could have been used to normalize the equilibrium capacities.

Table 6: BTEX adsorption isotherm parameters at different temperatures

Langmuir Linear D-R

Kl qm ß X 108 qD E

Temp. 0C (L/mg) (mg/g) R2 (mol2/J2) (mg/g) (kJ/mol) R2

25 0.66 198 0.97 4.27 43.96 3.43 0.97

35 0.59 211 0.98 3.29 62.49 3.77 0.96

45 0.53 218 0.98 2.93 55.21 4.09 0.9

4.4 Thermodynamic analysis of BTEX adsorption

The mean free energy of adsorption, E, in kJ/mol, was obtained in the form of the D-R model and reported for all cases. According to Balan, et al. [25] values of the mean free energy that are smaller than 8 kJ/mol are characteristic of a physisorption mechanism while values

between 8-16 kl/mol indicate an ion exchange mechanism. In all cases, values of mean free adsorption energy were less than 8 kl/mol, suggesting the process involved physisorption. This seemed contrary to the chemisorption mechanism suggested by the kinetic study and the observed temperature effect, hence a thermodynamic study to investigate the adsorption enthalphy was explored. The thermodynamic property associated with the change in Gibbs gree energy (AG0) was derived for all the systems studied according to the equation:

AG° = -R xTx InKt (7)

Where, R, is the ideal gas constant, 8.134 I/mol.K; T is the absolute temperature in K\ Kl is the equilibrium coefficient of the aromatic adsorption on adsorbents, which is equivalent to the equilibrium adsorption coefficient constant in the Langmuir model where it is applicable (converted to L/mol). The values obtained ranged from -20.3 kl/mol to - 36.6 kl/mol, indicating the likely spontaneity of the adsorption process as expected. Similary, the (AG0) is also equivalent to:

AG0 = A H° -Tx A S° (8)

By which the other thermodynamic properties, entropy in I/mol, AS0 and enthalpy, in kl/mol, AH° could be calculated from combining both equations to yield:

ln A£!_AH!

L R R*T V '

Table 6, highlights the thermodynamics properties obained by linear regression of Kl values obained from the temperature study.

Table 7: Thermodynamics properties of the adsorption of BTEX onto PAD 910 polystyrenic resin

Temperature, KL -AG0 -AH0 AS0 R2

K L/mol kJ/mol kJ/mol J/mol

298 6.98 x 104 20.3 4.67 69.34 0.88

308 6.54 x 104 26.12 4.17 62.19 0.84

318 5.77 x 104 36.6 - - -

The positive value (although small) of entropy changes suggests an increase in randomness at the water/carbon interface during the BTEX molecules adsorption on the adsorbents. This is likely due to desorption on smaller molecules into the aqueous system as the adsorption process proceeded. The negetaive value of the enthalpy change indicates the exothermic naure of the adsorption process. This is consistent with conculsions made by Site [19], that adsorption processes are generally believed to be exothermic, although the influence of temperature on the process may however vary depending on the variations that occur in the solubility of the sorbate with increasing temperature.

The relatively small equilibrium enthalpy of adsorption is within the range for hydrogen bonded processes (about -7 kl/mol) as reported by Site [19]. In addition, as discussed earlier, hydrogen bonding is an intermediate between physisorption and chemisorption, with an

average enthalpy of betwen 4 kl/mol and 40 kl/mol. This explains why the process has often been described as an interplay of both physisorption (E < 7 kl/mol) and chemisorption process (i.e pseudo-second order fitting and positive correlation with temperature).

5. CONCLUSIONS

It was shown that the PAD polystyrenic resin can be successfully utilized as an adsorbent for the removal of benzene, toluene, ethylbenzene and isomers of xylene from wastewater. The equilibrium adsorption data is better fitted by the Langmuir adsorption model rather than the D-R adsorption model for these adsorbates. Due to the lower adsorption capacity of the PAD 910 polystyrenic resin, the adsorbent requirements for this resin will be high. This will lead to larger adsorbent bed volumes required for treatment when the PAD 910 polystyrenic resin adsorbent is used. For larger volumes of adsorbent to be accommodated to treat the influent with the PAD 910 polystyrenic resin, it means larger equipment is required. The mean free energy parameter, E from D-R model was however from 3.45 to 4.12 kl/mol, indicative of a rather weak interaction between BTEX hydrocarbons and the carbon surface. The thermodynamic analysis resulted in a negative equilibrium enthalpy change, AH° suggesting an exothermic process. The positive entropy change AS° shows a random increase which is likely due to desorption of smaller molecules as BTEX molecules were adsorbed onto the carbon surface. In addition, the positive temperature effect is consistent with the chemisorptions process. A method for maximum regeneration of adsorbents is an important aspect of this work and should be included in further studies.

References

[1] R. A. Edgehill and G. Q. Lu Max, "Adsoprtion characteristics of carbonized bark fro phenol and pentachlorophenol," Journal of Chemical Technology and Biotechnoloy, vol. Vol. 71, pp. 27 - 34, 1998.

[2] M. Sarkar and P. K. Acharya, "Use of fly ash for the removal of phenol and its analogues from contaminated water.," Waste Management, vol. Vol. 26, pp. 559 -570, 2006.

[3] S. H. Lin and C. Y. Huang, "Adsorption of BTEX from aqueous solution by macro reticular resins," Hazardous Materials, vol. 70, pp. 21-30, 1999.

[4] M. Aivalioti, P. Papoulias, A. Kousaiti, and E. Gidarakos, "Adsorption of BTEX, MTBE, and TAME on natural and modified diamite," Hazardous Materials, vol. 207208, pp. 117-127, 2012.

[5] P. Cornel and H. Sontheimer, Chemical Engineering Science, vol. 41, pp. 1791-1800, 1986.

[6] E. H. Crook, R. P. McDonnel, and I. I. McNulty, Industrial Engineering, vol. 14, pp. 113-118, 1978.

[7] E. G. Furuya, H. T. Chang, Y. Miura, and K. E. Noll, Separation and Purification Technology, vol. 11, pp. 68-78, 1997.

[8] K. E. Noll, V. Gounaris, and W. S. Hou, "Adsorption Technologies for Air and Water pollution control," 1994.

[9] M. Aivalioti, I. Vamvasakis, and E. Gidarakos, "BTEX and MTBE adsorption on to raw and thermally modified diatomite," Hazardous Materials, vol. 178, pp. 136-143, 2010.

[10] I. Langmuir, "The Adsorption of Gases on Plane Surface of Glass, Mica and Platinum," Journal of Am. Chem. Soc. , vol. No. 40, pp. 1361 - 1368, 1916.

[11] M. M. Dubinin, "The Potential Theory of Adsorption of Gases and Vapors for Adsorbents with Energetically Non-Uniform Surfaces," Chem. Rev, vol. Vol. 60, pp. 235 - 241, 1960.

[12] S. M. Hasany and M. H. Chaudhary, "Sorption potential of hare river sand for the removal of antimony from acidic aqueous solution," Appl. Rad. Isot., vol. No. 47, pp. 467 - 471, 1996.

[13] Battelle, "The development of a Fluidized Bed Technique for the Regeneration of Powdered Activated Carbon.," Washington: US Department of the Interior, 1992.

[14] J.-M. Chern and Y.-W. Chien, "Adsorption of nitrophenol onto activated carbon: isotherms and breakthrough curves," Water Research, vol. 36, pp. 647-655, 2// 2002.

[15] J. Byrd, "Characteristics Infrared Absorption Frequencies," 1998.

[16] Y. Gou and D. A. Rockstraw, "Physiochemical Properties of Carbons Prepared from Pecan Shell by Phosphoric Acid Activation.," Bioresource Technology, vol. Vol. 98, pp. 1513 - 1521, 2007.

[17] B. Jiang, Y. Zhang, K. Zhou, and S. Chen, "Effects of Chemical Modification of Petroleum Cokes on the Properties of the Resulting Activated Carbon.," Fuel, vol. Vol. 87, pp. 1844 - 1848, 2008.

[18] S. Laleh, T. Ali, K. Hossein, N. B. Gholamreza, A. A. Ali, N. Shapoor, et al., "Adsorption of BTEX on Surfactant Modofied Granulated Natural ZEolite Nanoparticles: Parameters Optimizing by Applying Taguchi Experimental Design Method.," Clean - Soil, Air, Water Journal, vol. Vol 39 (10), pp. 939 - 948, 2011.

[19] A. D. Site, "Factors affecting sorption of organic compounds in natural sorbent/water systems and sorption coefficients for selected pollutants: A review," Journal of Physical Chemistry. Ref. Data, vol. Vol. 30, No. 1, pp. 187 - 438, 2001.

[20] R. M. Allen-King, P. Grathwohl, and W. P. Ball, "New modeling paradigms for the sorption of hydrophobic organic chemicals to heterogeneous carbonaceous matter in soils, sediments and rocks," Advanced Water Resources, vol. Vol. 25, pp. 985 - 1016, 2002.

[21] Y. S. Ho and G. McKay, "Sorption of Dye from Aqueous Solution by Peat.," Chem. Eng. Journal, vol. Vol. 70(2), pp. 115 - 124, 1998.

[22] S. Hall, R. Tang, J. Baeyens, and R. Dewil, "Removing Polycyclic Aromatic Hydrocarbons from Water by Adsportion on Silica Gel.," Polycyclic Aromatic Compounds, vol. Vol. 29, pp. 160-183, 2009.

[23] C. Valderrama, J. L. Cortina, A. Farran, X. Gamisans, and C. Lao, "Kinetics of Sorption of Polyaromatics Hydrocarbons onto Granular Activated Carbon and Macronet Hyper-Cross-Linked Polymers (MN200)," Jornal of Collod Interface Science, vol. Vol. 310 (1), pp. 35 - 36, 2007.

[24] J. F. Rodriguez and S. A. Linares, "Chemistry and physics of carbon: a series of advances," MarcelDekker, Inc., 1989.

[25] C. Balan, D. Bilba, and M. Macoveanu, "Studies on Chromium (III) removal from aqueous solution by sorption on sphagnum moss peat.," J. Serb. Chem. Soc., vol. Vol. 74 (8-9), pp. 953 - 964, 2009.

Dear Editor,

I am happy with this article presented in black and white colour.

Regards,

Thobeka