INTRODUCTION:

Caffeine (C₈H₁₀O₂) is a xanthine alkaloid, naturally present in more than sixty plant species. Caffeine containing by-products and wastewaters generated from coffee and tea processing plants comprise a major part of the agricultural and industrial wastes in coffee growing area¹. Caffeine has been detected in surface water, ground water and wastewaters at a high concentration (~10g caffeine/L) due to the discharge of caffeine containing wastewaters in the surrounding water bodies^2,3. Kolpin et al. ⁴ evaluated caffeine as fourth most frequently detected pollutant out of 95 different Organic Wastewater Contaminants (OWCs).

In recent years caffeine has been considered as an emerging contaminant^5,6. The emerging contaminants may pose a potential hazard to humans and aquatic animals⁷. Presence of caffeine in the agricultural soil, affects soil fertility as it inhibits seed germination and growth of seedlings viz. Trigonella foenum-graecum L.⁸ , Capsicum annuum L.⁹ , Helianthus annuus L.¹⁰. The ingestion of caffeine has severe effects on the physiological system. Caffeine’s lethal and sublethal effects on the freshwater species viz. Ceriodaphnia dubia (water flea) (LC₅₀ = 60 mg/L), Pimephales promelas (fathead minnow) (LC₅₀ = 100 mg/L), Chironomus dilutus (midge) (LC₅₀ = 1.230 g/L) and Xenopus laevis (African clawed frog) (LC₅₀ = 0.35 mg/mL) have also been reported ¹¹. The conventional methods of caffeine removal viz. using charcoal, carbon ¹²and membrane filtration ^13,14 , solid-liquid extraction ¹⁵and supercritical carbon dioxide extraction ¹⁶methods have major disadvantages. In case of solvent extraction, solvents such as trichloroethylene, methylene chloride or some other similar chlorinated compounds were used for caffeine extraction. This method has to meet stringent environmental restriction while discarding the solvent ¹⁷. The disadvantage of supercritical carbon dioxide extraction is the high cost equipment due to the usage of high pressure and also it needs large quantity of CO₂ which is very costly¹⁸. Moreover, the use of membrane or carbon filters in caffeine removal is expensive and the commercialization of the process becomes less viable¹⁷. Therefore, there is a need for alternate method to conventional processes involved in the removal of caffeine from industrial wastewater. Adsorption technique has attracted attention in this context and is being widely used for the removal of several pollutants. Limited number of studies is available on caffeine adsorption ^19,20. Biologically based adsorption process uses low cost biological materials viz. living or dead microorganisms²¹ . So far, there is no report available on the usage of dead yeast biomass as adsorbent for caffeine removal. Among microorganisms, yeast cells offer several advantages. Yeasts are inexpensive, readily available source of biomass. They not only grow rapidly like bacteria, but like filamentous fungi they also have the ability to resist unfavorable environments like low pH etc.²² . Many adsorbents do not exhibit their full potential in the raw (untreated) form and their uptake capacity has been found to improve significantly upon chemical pre-treatment. So, in the present study, the possible use of dead yeast biomass pretreated with NaOH and surfactants viz. Sodium dodecyl sulphate (SDS) and Tween 80 have been investigated to improve the caffeine adsorption potential of yeast biomass. Surfactant molecules can reduce surface and interfacial tensions in aqueous phase ²³. For industrial or technical operations, immobilized microbial systems provide additional advantages viz. regeneration and reuse of the biomass, minimal clogging in continuous column application and easier solid-liquid separation ²⁴. PVA-alginate based immobilization matrix can be considered best among other matrices in case of reusability, thermal stability and cost-effectiveness ²⁵. The present work investigated the caffeine adsorption using dead biomass of the yeast, Trichosporon sp. VITLN01 in native as well as in pretreated form. The influence of adsorption system variables including pH, temperature, adsorbent dosage, initial caffeine concentration on caffeine removal was studied. The data were fitted to adsorption isotherm models viz. Langmuir, Freundlich and Temkin and kinetic models viz. pseudo-first order, pseudo-second order and intraparticle diffusion model. The role played by the surface morphology and organic functional groups (present on yeast biomass) in adsorption process was examined by SEM, FT-IR and potentiometric titration. The experiment on caffeine removal from coffee processing industrial wastewater was carried out using PVA-alginate immobilized Tween 80 treated dead biomass of Trichosporon sp. in batch mode.

MATERIALS AND METHODS :

Preparation of Adsorbent:

The yeast species was isolated from caffeine contaminated soil under coffee cultivation area, collected from Coffee Board, Yercaud, India. The yeast cells were grown in Yeast Extract Peptone Dextrose (YEPD) medium containing (g l^-1): yeast extract, 10; peptone, 20; and dextrose, 20. The pH and temperature was maintained at 6.5 and 28 °C. Growth was allowed to proceed for five days on a rotary shaker operating at 120 rpm. After the yeast growth, the biomass and the culture medium were separated by centrifuging at 8000 rpm for 10 min and the resulting biomass was washed several times thoroughly with distilled water. The biomass obtained was subjected to drying at 60 °C until a constant weight of biomass was obtained. The dried native biomass of 1 g was suspended in each 100 ml of NaOH (0.1 M), Tween 80 (4 mM) and SDS (3 mM) and stirred at room temperature. After 24 h, biomass was separated by centrifugation, washed, dried and ground to a fine powder, which was sieved through a mesh (150 µm) sieve. The undersized fraction was used for caffeine adsorption studies.

Characterization of Adsorbents:

The surface area and pore volume of the adsorbents were measured by BET (Brunauer Emmett-Teller nitrogen adsorption technique). Leach ability of the adsorbents was characterized by agitating 1 g of adsorbent with 50 mL of deionized water in an Erlenmeyer ﬂask on a rotary shaker for 24 h. The adsorbent was separated by settling and the Chemical Oxygen Demand (COD) of the supernatant was determined as per standard analytical method ²⁶.

pH of point of zero charge :

The pH of the point of zero charge (pHPZC) of the adsorbent was analyzed using 0.01M of KCl solutions (each of 50 mL). The pH values of the solution were adjusted between 2 and 12 with and without adsorbent (10 g/L). These solutions were periodically agitated and allowed to equilibriate for 48 h. The pH values of the supernatant liquids were determined. The ΔpH was calculated by measuring the difference between the pH of the solutions with and without the adsorbent. The pH_KCl is the pH of the solutions without the adsorbent. The point of zero charge is the pH at which pH_KCland ΔpH are zero. The point of zero charge was calculated by plotting ΔpH against pH_KCl.

Potentiometric Titration of the Adsorbents:

Potentiometric titrations were carried out following the method proposed by Boehm ²⁷. The titration was performed using the pH meter (Elico, India) and Erlenmeyer ﬂasks kept over the magnetic stirrer. NaOH (0.1M) and HCl (0.1M) were used to neutralize the total acidic and basic sites respectively. The yeast biomass (1 g) was suspended in 0.1N HCl (50 mL) and stirred for 2 h on a rotary shaker. The titrant (0.1 M NaOH) of 0.25 mL was added to the stirred suspension. After each addition, adsorbent slurry was allowed to equilibrate until the pH was stable. The total volume of NaOH used for neutralization was noted for total acidic sites calculations. Similarly, the total basic sites were estimated by suspending the adsorbent (1 g) in 0.1M NaOH solution (50 mL) and titrated similarly as described above with 0.1N HCl as titrant.

Thermal analysis:

Thermogravimetric analyses of native and pretreated adsorbents were carried out under high purity helium supplied at a purge gas flow rate of 0-1000 mL min^-1 (Diamond TG/DTA, Perkin Elmer, USA). All samples were subjected to 10 °C min^-1 heating rate and were characterized between 50 and 1000 °C.

Batch adsorption studies:

Experiments were conducted in 250 mL Erlenmeyer flasks containing 100 mL of synthetic caffeine solution. The flasks were agitated on a rotary shaker at 120 rpm for 12 h at room temperature (30 ±2 °C) to ensure equilibrium was reached. Adsorbent free blanks were used as control. For determining the effect of pH on caffeine adsorption by native and treated adsorbents, initial pH of each caffeine solution (10 g L^-1) was adjusted to desired value (4-8) using 0.1 M HCl or 0.1 M NaOH. The effect of temperature on the adsorption capacity was investigated in the temperature range of 20-40 °C. The adsorbent concentration was varied between 10 and 40 g L^-1 caffeine solution for determining the effect of adsorbent dosage on caffeine sorption capacity. The effect of initial caffeine concentration on the adsorption capacity was studied at optimum pH, temperature and adsorbent dosage. The initial concentration in the adsorption medium was varied between 2-14 g L^-1. To optimize the contact time, caffeine sorption at 1 h interval was determined by analyzing residual caffeine in the supernatant. The samples collected at different time intervals were centrifuged at 6000 rpm for 5 min and the liquid supernatant was analyzed for the residual caffeine concentration.

Analysis of Caffeine Concentration:

HPLC analysis was carried out on a Waters instrument equipped with a dual λUV–VIS detector and a C18 column. The mobile phase used was water: methanol (70:30) at a flow rate of 1 ml/min for 10 min. Synthetic caffeine at 15 g L^-1 was used as the standard. Retention time of caffeine was found to be 3.49 min at 30 °C. Detection of caffeine was done at 253 nm. The amount of caffeine adsorbed was calculated from the differences between initial caffeine concentration added to the solution and caffeine content of the supernatant.

Adsorption Isotherms and Kinetics :

From the experimental data obtained from the studies on effect of initial caffeine concentration on adsorption capacity at constant temperature, the applicability of Langmuir²⁸, Freundlich²⁹and Temkin³⁰ isotherm models were judged. Pseudo-first-order³¹, Pseudo-second-order³² reaction kinetic models and Intraparticle diffusion model ³³were applied to the data obtained from the experiments carried out at different contact time. Here the caffeine solution (12 g L^-1) containing optimized adsorbent dosage was maintained at optimum pH and temperature.

Thermodynamics of Adsorption:

The data obtained from the experiments carried out at different temperature were used to study the thermodynamics of caffeine adsorption. The values of Gibbs energy, enthalpy and entropy (ΔG, ΔH, ΔS), for the adsorption process were calculated using the standard equations ³⁴.

Fourier Transform Infrared (FTIR) Spectroscopic Analysis:

Infrared spectra of native and pretreated yeast biomass before and after caffeine adsorption were recorded on an Avatar 330 model (Thermo Nicolet Co., USA) FT-IR spectrometer. For IR studies, 5 mg of each adsorbent were encapsulated in 400 mg of KBr. Translucent discs were obtained by pressing the ground material with the aid of bench press. The spectrum was recorded in the range of 4000-400 cm^-1.

Scanning Electron Microscopic (SEM) Analysis:

SEM analysis of native and pretreated yeast biomass before and after caffeine adsorption was performed using scanning electron microscopy, Hitachi (Model: S-3400N).

Use of Immobilized Adsorbents for Removal of Caffeine from Wastewater:

Coffee processing industrial wastewater was collected from Coffee Board, Yercaud, India. The initial caffeine concentration in the wastewater was measured using HPLC. PVA (10 g L^-1) and sodium alginate (10 g L^-1) were mixed thoroughly in distilled water at 80 °C to obtain a homogeneous suspension. After cooling PVA- alginate mixture to 40 °C, 20 g L^-1 of the adsorbent (native or pretreated) was added and mixed. The PVA- alginate mixture was extruded gently in 0.2 M CaCl₂to form PVA-alginate beads of diameter ~2 mm.The resulting beads were washed with saline and used for adsorption experiments. PVA-alginate beads without adsorbent were also prepared as control. The beads were added to 100 mL of coffee industrial wastewater in 250 mL of Erlenmeyer flasks, adjusted to pH 7.5 and maintained at temperature 30 °C. Liquid samples were collected at desired time intervals to estimate the caffeine adsorption capacity.

RESULTS AND DISCUSSION:

Characterization of native and pretreated adsorbent:

The external surface area and total pore volume of the adsorbents were measured from N₂ adsorption isotherms with a sorptiometer. These properties were listed in the Table 1. Among all the adsorbents, Tween 80 treated adsorbent was found to have the highest surface area and pore volume. The adsorbents exhibiting a large surface area are generally assumed to adsorb large amount of adsorb ate than the adsorbents with lower surface area. The micropore volume of adsorbent has major contribution on the adsorption capacity of adsorbate molecules to pass through the pores ²⁰. The leaching characteristics of the adsorbents were measured in terms of COD. Tween 80 treated adsorbent showed the least COD value and NaOH treated showed the highest values of about 10 mg L^-1 and 29 mg L^-1 respectively (Table 1).

Table 1. Physical properties of native and pretreated adsorbents.

S. No.	Adsorbents	BET surface area (m²g^-1)	Total pore volume (cm³ g^-1)	Leachability (mg L^-1)
1	Native	15.16	0.012	13
2	Tween 80 treated	29.14	0.023	10
3	SDS treated	23.21	0.017	14
4	NaOH treated	19.36	0.009	29

pH of point of zero charge (pH_PZC):

The pH_PZC values of native and treated adsorbents were shown in the Table 2. It was observed that pretreatment of the native adsorbent affected the pH_PZCvalue. When the pH of solution was higher than pH_PZC,the adsorbent surface exhibited net negative charge and could interact with positive charge containing species while at pH lower than pH_PZCthe surface was positively charged and could interact with negative species ³⁵

Table 2. The pH values of the point of zero charge (pH_PZC) attained by adsorbents:

S.No.	Adsorbents	pH_PZC
1	Native	5.0
2	Tween 80 treated	7.2
3	SDS treated	6.9
4	NaOH treated	6.3

Potentiometric titrations of the biomass:

The number of active (acidic and basic) sites and functional groups present on the adsorbents were estimated by potentiometric titrations as developed by Fourest and Volesky ³⁶. The identified functional groups and their site densities were summarized and presented in Table 3 and Table 4 respectively. Comparison of native and pretreated adsorbents shows the phenomenon of varying number of functional groups, their density and conformational changes in the adsorbent surface during pretreatments. And it may be inferred that acidic groups were phosphate (pK a 6.1–6.8) and carboxylic (pK a 1.7–4.7) and the basic groups were comparable to the values reported for sulfhydryl (pK a 8.0–10.0), amines (pK a 8.0–11.0) and hydroxyl (pK a 9.5–13) groups ^{37 – 39}.

Table 3. Functional groups determined by potentiometric titrations.

S. No	Adsorbents	pK_avalues	Functional groups
1.	Native	3.72	Carboxyl
		6.45	Phosphate
		8.63	Amine
		8.91	Amine
2.	Tween 80 treated	3.42	Carboxyl
		4.17	Carboxyl
		10.28	Amine
		12.23	Hydroxyl
		12.51	Hydroxyl
		12.72	Hydroxyl
3.	SDS treated	2.45	Carboxyl
		9.21	Sulfhydryl
		9.94	Amine
		11.00	Amine
4.	NaOH treated	3.29	Carboxyl
		6.34	Phosphate
		10.39	Amine
		12.49	hydroxyl

Table 4. Site densities of functional groups determined by potentiometric titrations.

S. No	Adsorbents	Functional groups	Site density (mmol g^-1)
1.	Native	Carboxyl	0.52 ± 0.06
		Phosphate	0.41 ± 0.02
		Amine	0.50 ± 0.03
2.	Tween 80 treated	Carboxyl	1.25 ± 0.01
		Amine	0.30 ± 0.03
		Hydroxyl	0.85 ± 0.01
3.	SDS treated	Carboxyl	0.52 ± 0.07
		Sulfhydryl	0.16 ± 0.03
		Amine	0.38 ± 0.04
4.	NaOH treated	Carboxyl	0.77 ± 0.05
		Phosphate	0.16 ± 0.05
		Amine	0.23 ± 0.02
		hydroxyl	0.52 ± 0.06

Thermal Analysis:

Thermo gravimetric and differential thermal (TG/DTA) analysis of the native and pretreated adsorbents were carried out for better understanding of the influence of temperature on the properties of adsorbents. The initial weight loss (7-9 %) in the small temperature range of 100 to 200 °C could be due to the loss of adsorbed water molecules. The drastic weight loss due to decomposition was observed in the temperature range of 200 to 700 °C. The maximum weight loss (76.155 %) was observed in case of NaOH treated adsorbents followed by SDS treated adsorbent (70.355 %), native adsorbent (68.21 %) and Tween 80 treated adsorbent (65.352 %). This might be due to the degradation of the organic fractions of the studied adsorbents.

Batch adsorption studies:

Effect of pH:

pH of the solution is one of the most important factors that influences the chemical state of functional groups and affect the adsorption process^40,41. The effect of pH on caffeine adsorption was studied by varying the pH from 4 to 8. For each pH value, temperature (30 °C), adsorbent dosage (10g L^-1), initial caffeine concentration (2 g L^-1) were kept constant.. In all the cases, caffeine uptake increased with increasing pH and reached maximum at pH 7.5. These results suggest that caffeine adsorption is controlled by ionic attraction between caffeine and functional groups at the adsorbent surface.

Effect of Temperature:

The adsorption of pollutants increases with increase in temperature along with faster rate of diffusion of adsorb ate molecules from the solution to the adsorbent by increasing the pore size in the adsorbents ^42,43 . The present data of caffeine adsorption onto the native and pretreated adsorbents at the range of temperatures between 20 and 40 °C and at constant pH (pH 7.5) was noted. The results showed that the adsorption capacity increased with increasing the temperature up to 30 °C and above this temperature, the adsorption capacity of caffeine was decreased. This decrease at high temperature may be due to the denaturation of active binding sites in the adsorbents⁴⁴.

Effect of Adsorbent Dosage:

The number of sites available for the caffeine adsorption depends on the concentration of adsorbents. The effect of adsorbent concentration (10-40 g L^-1) on caffeine uptake was noted. Caffeine uptake was found to increase with increase in adsorbent concentration from 10 to 20 g L^-1. Beyond this dosage the uptake was reduced. From these results it may be concluded that at lower adsorbent dosage (i.e., below 20 g L^-1), caffeine molecules were competing for adsorption at limited amount of adsorption sites. However, as the adsorbent concen tration was increased, the adsorption sites were easily available resulting in higher uptake of caffeine. The significant decrease in caffeine uptake at adsorbent dosage greater than 20 g L^-1 may be attributed to the presence of surplus binding sites on the adsorbent surface than the available caffeine molecules in the solution at constant concentration of 10 g L^-1.

Effect of Initial Caffeine Concentration:

The initial caffeine concentration provides an important driving force to overcome mass transfer resistance of caffeine between solid and aqueous phases⁴⁵. The caffeine uptake capacity of native and pretreated adsorbents as a function of initial caffeine concentration was recorded. The initial caffeine concentration in the solution was varied between 2-14 g L^-1. The uptake of caffeine increased linearly with increase in initial caffeine concentration and then reached a plateau at 12 g L^-1 demonstrating the saturation of binding sites.

Effect of Contact Time:

The contact time for an adsorption process has quite practical application significance^46,47. Typical adsorption process shows a speedy initial uptake, followed by a slower process. At the initial 180 min, caffeine molecules in the solutions were rapidly adsorbed by the four adsorbents. Afterwards the uptake decreased or attained equilibrium. As The adsorption equilibrium time of native adsorbent was 10 h, while the equilibrium time of Tween 80 treated, SDS treated and NaOH treated adsorbents were attained in about 3 h, 6 h and 9 h respectively. Among all adsorbents, Tween 80 treated adsorbents showed the maximum caffeine uptake (0.5 g g^-1) and shorter equilibrium time.

Adsorption Isotherms:

The experimental data were analyzed using Langmuir, Freundlich and Temkin isotherm models. Langmuir constants q_max and b were determined from the slope and intercept of the linear plot of 1/q_eq against 1/C_eq. The parameter q_max is the maximum amount of caffeine adsorbed per unit weight of adsorbent to form a complete monolayer on the adsorbent surface and b is the constant. The R² values shown in Table 5 suggested that the Langmuir isotherm provides a good fit to the isotherm data. The values of Freundlich constants were calculated from the intercept and slope of the plot log q_eqvs C_eq. K_f (g/g (Lg) ^1/n and 1/n are indicators of adsorption capacity and adsorption intensity respectively. Values of n>1 represents favorable adsorption condition^48,49. The value of Freundlich exponent n is in the range n>1, indicating a favorable adsorption condition. From the linear plot of q_eqversus ln C_eq, Temkin isotherm constants A and B were determined. A is the equilibrium binding constant and the constant B is related to the heat of adsorption. The R² values of Freundlich and Temkin isotherm models were lower than Langmuir isotherm (Table 5). It can be seen from the isotherm model parameters and correlation coefficient (R²) values that Langmuir isotherm model showed better fit than Freundlich and Temkin isotherm models. Thus in this study, the results indicated that caffeine adsorption by native and pretreated adsorbents was apparently with monolayer coverage on adsorbent surface that is homogenous.

Adsorption Kinetics:

The value of pseudo-first-order rate constants (k₁) and calculated sorption capacity were determined from the equations of linear plots of log (q_eq-q_t) against time. Pseudo-second-order rate constants (k₂)and calculated sorption capacity (q_eq, _cal) were determined from the equations of linear plots of t/q_t against time. Table 6 shows the comparison of the pseudo-first-order, pseudo-second-order adsorption rate constants, calculated and experimental q_eqvalues obtained for different adsorbents. In case of Tween 80 treated adsorbent, the data showed well fit to pseudo-first order kinetics, since the variations between the calculated (q_eq,cal) and experimental (q_eq,exp) sorption capacity were minimal and correlation coefficient (R²) value was also found to be 0.999. Therefore, caffeine sorption may be explained as the passive uptake through physical adsorption or the adsorbent surface ion exchange. The data for other adsorbents showed better fit to pseudo-second order kinetics, thus in agreement with chemisorptions being the rate limiting step in these cases.

Intraparticle Diffusion:

The above studied kinetic models cannot identify the diffusion mechanism. The kinetics results can be analyzed by intraparticle diffusion model in order to elucidate the diffusion mechanism ^50,51. The linear plots revealed the occurrence of intraparticle diffusion in the sorption process of caffeine by the native and pretreated adsorbents. The rate constants for the intraparticle diffusion k_idwere listed in Table 7. The results showed that the rate constant was higher for Tween 80 treated adsorbents compared to other adsorbents.

Table 7. Intraparticle diffusion model constants and correlation coefficients for adsorption of caffeine on native and pretreated adsorbents.

	K_id (g g^-1min^-1)	C	r²
Native	0.01	0.007	0.988
Tween 80 treated	0.048	0.188	0.977
SDS treated	0.020	0.047	0.960
NaOH treated	0.013	0.002	0.957

Thermodynamics of Sorption:

In order to describe thermodynamic behavior of caffeine adsorption on native and pretreated adsorbents, thermodynamic parameters viz. Enthalpy change (ΔH), Gibbs energy (ΔG) and entropy change (ΔS) were estimated using the standard equations as shown in Table 8. The negative values of ΔG indicated that caffeine removal by all adsorbents were thermodynamically feasible and spontaneous in nature. The values of ΔH and ΔS in the caffeine adsorption process were determined from slope and intercept of the plot of log q_e/C_e vs. 1/T. The negative value of ΔH indicated the exothermic nature of caffeine adsorption on native and pretreated adsorbents. The positive value of ΔS showed the increased randomness at the solid/liquid interface during adsorption process and suggested good affinity of the caffeine towards the adsorbents (native and pretreated). Thus the thermodynamic studies clearly demonstrated that adsorption of caffeine on native and pretreated adsorbents were favorable, spontaneous and exothermic in nature.

FTIR Analysis:

FTIR spectral analysis was carried out to elucidate the mechanism of caffeine adsorption on native and pretreated adsorbents. The FTIR spectra of adsorbents before and after caffeine adsorption are shown in Fig.1- 2. After caffeine adsorption by native adsorbent, significant shifts from FTIR spectra were noted from 3421 cm^-1 to 3429 cm^-1 (NH stretch; amine group), 1653 cm^-1 to 1651 cm^-1 (C=O; carbonyl group), 1238 cm^-1 to 1234 cm^-1 (P-O-C stretch; phosphate group) (Fig. 1A.(a,b). This result signified that amine, carbonyl and phosphate groups were the major sites on native adsorbent for caffeine adsorption. The FTIR spectrum of Tween 80 treated adsorbent after adsorption of caffeine showed shift of several peaks (Fig. 1Bb) compared to before adsorption spectra (Fig. 2a). The most significant shifts were observed from 3423 cm^-1 to 3441 cm^-1 (NH stretch), 2926 cm^-1 to 2927 cm^-1, 2854 cm^-1 to 2856 (CH stretch; alkyl group), 1647 cm^-1 to 1651 cm^-1 (C=O group), 1082 cm^-1 to 1076 cm^-1 (C-OH stretch; alcohol group), indicated the involvement of amine, alkyl, carbonyl and hydroxyl group in caffeine adsorption by Tween 80 treated adsorbents. The shift in the hydroxyl group which was not observed in the native adsorbents might be the reason behind enhanced caffeine uptake by Tween 80 treated adsorbents. Fig. 2A (a, b) displayed the FTIR spectra of SDS treated adsorbents before and after caffeine adsorption. The shifts in the absorption peak were observed from 3448 cm^-1 to 3446 cm^-1 (NH stretch), from 1658 cm^-1to 1641 cm^-1 (C=O stretch), 1541 cm^-1 to 1546 cm^-1 (NH deformation) and 1236 cm^-1 to 1238 cm^-1 (S=O stretch ; sulfate group) specified the involvement of amine, carbonyl and sulfate group. FTIR spectra of NaOH treated adsorbent showed peaks at 3439 cm^-1(NH stretch) and 1647 cm^-1 (C=O stretch) had shifted respectively to 3456 cm^-1 and 1639 cm^-1 (Fig. 2B (a, b). This represented the involvement of amine and carbonyl groups in caffeine adsorption by NaOH treated adsorbents. To summarize the FTIR study, negatively charged groups (carbonyl, phosporyl, sulfate and hydroxyl groups) and few positively charged groups (alkyl and amine groups) were expected to significantly favor caffeine adsorption on native and pretreated adsorbents.

Scanning Electron Microscopy (SEM):

Surface texture and morphology of the adsorbents before and after caffeine adsorption were examined using SEM analysis. The native adsorbent surface appeared smooth and clear with oval shaped cell (Fig. 3a). No noticeable changes were seen in the native adsorbent after caffeine adsorption (Fig.3b). Due to pretreatments, structural distortion and roughness of adsorbent surface was observed in case of all pretreated adsorbents before caffeine adsorption (Fig. 3 c, e, g). In case of Tween 80 treated adsorbent, formation of large amount of agglomerative mass of caffeine binding to the adsorbent surface was noted after caffeine adsorption (Fig. 3d).The surface of SDS treated adsorbent was changed after caffeine adsorption and caffeine was adsorbed as cylinder or fibers-like structure (Fig. 3f). The NaOH treated adsorbent showed non-uniformly distributed caffeine bodies on irregular adsorbent surface after caffeine adsorption (Fig. 3h).

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

Fig. 3. SEM images of adsorbents before and after caffeine adsorption. (a) native adsorbent, (b) native adsorbent + caffeine, (c) Tween 80 treated adsorbent, (d) Tween 80 treated adsorbent + caffeine, (e) SDS treated adsorbent (f) SDS treated adsorbent + caffeine, (g) NaOH treated adsorbent and (h) NaOH treated adsorbent + caffeine

Caffeine Removal from Industrial Wastewater using Immobilized Adsorbents:

The coffee processing industrial wastewater collected from Coffee Board, Yercaud, India, was found to contain 11.4 g L^-1 caffeine. Wastewater pH was adjusted to 7.5 and experiments were conducted in batch mode at 30 °C. Removal of caffeine from industrial wastewater was investigated using pretreated adsorbents immobilized in PVA-alginate matrix. Caffeine uptake was found to be maximum (0.570 g g^-1) in case of immobilized Tween 80 treated adsorbents followed by SDS treated adsorbent (0.378 g g¹), NaOH treated adsorbent (0.347 g g^-1) and native adsorbent (0.291gg^-1).

CONCLUSION:

PVA-alginate immobilized Tween 80 treated dead biomass of Trichosporon sp. VITLN01 has been successfully used as adsorbent for the removal of caffeine from aqueous medium. The batch experiments showed that all the parameters viz. pH, temperature, initial adsorbent dosage, initial caffeine concentration and contact time had effect on caffeine removal. Adsorption equilibrium data fitted well to the Langmuir model suggesting homogeneous monolayer adsorption. Kinetics of adsorption by Tween 80 treated adsorbent was better explained by pseudo-first order model which suggested physical adsorption. The thermodynamic calculation showed the feasibility, exothermic and spontaneous nature of caffeine adsorption. The FTIR spectral analysis showed the involvement of carbonyl, amine, hydroxyl, phosphate and sulfate groups. SEM analysis revealed surface morphological changes on pretreated adsorbents before and after caffeine adsorption.

ACKNOWLEDGEMENT:

Authors acknowledge the Sophisticated Analytical Instrumentation Facility (SAIF), IIT Bombay, Mumbai, India for providing facility for TG/DTA analysis and Madras University, Chennai, India for providing SEM facilities. We also wish to thank VIT University, Tamilnadu, India for providing financial support and laboratory facilities.

CONFLICT OF INTERESTS :

There is no conflict of interests among the authors.

REFERENCES:

1. Adams MR, Dougan J. Biological management of coffee processing. Tropical Science. 123 ; 1981 : 178–196.

2. Buerge IJ et al. Caffeine, an anthropogenic marker for wastewater contamination of surface waters. Environmental Science and Technology. 37 ; 2003: 691-700.

3. Glassmeyer ST et al. Transport of Chemical and Microbial Compounds from Known Wastewater Discharges: Potential for Use as Indicators of Human Fecal Contamination. Environmental Science and Technology. 39 ; 2005: 5157-5169.

4. Kolpin DW et al. Pharmaceuticals, hormones, and other organic wastewater contaminants in United States streams, 1999–2000 : A national reconnaissance. Environmental Science and Technology. 36 ; 2002: 1202–1211.

5. Bruton T et al. Fate of Caffeine in the Environment and Ecotoxicological Considerations. In: Contaminants of Emerging Concern in the Environment: Ecological and Human Health Considerations. Edited by Halden RU, American Chemical Society, Vol.1048 ; 2010 pp. 257-273.

6. Sodre FF, M.A.F. Locatelli MAF, Jardim WF. Occurrence of emerging contaminants in Brazilian drinking waters: A sewage-to-tap issue. Water Air Soil Pollution. 206 ; 2010: 57-67.

7. Bolong N, Ismail AF, M.R. Salim, Matsuura T. A review of the effects of emerging contaminants in wastewater and options for their removal. Desalination. 239 ; 2009 : 229-246.

8. Anis M, Wani AA. Caffeine induced morph cytological variability in Fenugreek, Trigonella foenum-graecum L. Cytologia. 62 ;1997 : 343-349.

9. Kumar G, Tripathi A. Mutagenic response of caffeine in Capsicum annum L., Journal of Indian Botanical Society. 83 ; 2004 : 136-140.

10. Khusheed T, Ansari MYK, Shahab D. Studies on the effect of caffeine on growth and yield parameters in Helianthus annuus L. variety modern. Biology and Medicine. 1(2) ; 2009 : 56-60.

11. Moore MT, Greenway SL, Farris JL. Assessing caffeine as an emerging environmental concern using conventional approaches. Archives Environmental Contamination Toxicology. 54 ; 2008: 31-35.

12. Feldman JR, Katz SN, Caffeine. In: Encyclopedia of chemical processing design. Marcel Dekker Inc, 1977, pp. 424–440.

13. Snydera SA et al. Role of membranes and activated carbon in the removal of endocrine disruptors and pharmaceuticals. Desalination. 202 ; 2007 : 156–181.

14. Yoona Y et al. Removal of endocrine disrupting compounds and pharmaceuticals by nanofiltration and ultrafiltration membranes. Desalination. 202 ; 2007 : 16–23.

15. Senol A, Aydin A. Solid-liquid extraction of caffeine from tea waste using battery type extractor: Process optimization. Journal of Food Engineering. 75 ; 2006 : 565–573.

16. Kim WJ, Kim JD, Kim J, Oh SG, Lee YW. Selective caffeine removal from green tea using supercritical carbon dioxide extraction. Journal of Food Engineering. 89 ; 2008: 303-309.

17. Gokulakrishnan S, Chandraraj K, Gummadi SN. Microbial and enzymatic methods for removal of caffeine. Enzyme and Microbial Technology. 37 ; 2005 : 225-232.

18. Katz SN, Decaffeination. ln: Coffee Technology ,Edited by Clarke RJ, Macrae R, Elsevier Applied Science, London, 1987 ; pp. 59-71.

19. Lin AY, Lin C , Tung H, Sridhara Chary N. Potential for biodegradation and sorption of acetaminophen, caffeine, propranolol and acebutolol in lab-scale aqueous environments. Journal of Hazardous Materials. 183 ; 2010 : 242–250.

20. Sotelo JL, Rodríguez A, Álvarez S, García G. Removal of caffeine and diclofenac on activated carbon in ﬁxed bed column. Chemical Engineering Research and Design. 90 ; 2012: 967–974.

21. Šafaříková M, Ptáčková L, Kibriková I, Šafařík I. Biosorption of water-soluble dyes on magnetically modified Saccharomyces cerevisiae subsp. uvarum cells. Chemosphere. 59 ; 2005: 831-835.

22. Yu Z, Wen X. Screening and identification of yeasts for decolorizing synthetic dyes in industrial wastewater. International Biodeterioration and Biodegradation. 56 ; 2005 : 109-114.

23. Rao RAK, Khan MA, Jeon BH. Utilization of carbon derived mustard oil cake for the removal of bivalent metal ions: Effect of anionic surfactant on the removal and recovery, Journal of Hazardous Materials. 173 ; 2010 : 273- 282.

24. Jianlong W, Liping H, Hanchang S, Yi Q. Biodegradation quinoline by gel immobilized burkholderia sp. Chemosphere. 44 ; 2001 : 1041-1046.

25. Nunes APM, Vila-Real H, Fernandes PCB, Ribeiro MHL. Immobilization of naringinase in PVA–Alginate matrix using an innovative technique. Applied Biochemistry and Biotechnology.160 ; 2010 : 2129–2147.

26. APHA, Standard methods for Water and Wastewater Examination, 20^th ed. American Public Health Association, Washington DC, USA, 2001.

27. Boehm HP, Some aspects of the surface chemistry of carbon blacks and other carbons. Carbon. 32 ; 1994: 759–769.

28. Langmuir I. The constitution and fundamental properties of solids and liquids. Journal of the American Chemical Society. 38 ; 1916 : 2221-2295.

29. Freundlich HMF. Over the adsorption in solution. Journal of Physical Chemistry. 57 ; 1906 : 385-470.

30. Temkin M, Pyzhev V. Kinetics of ammonia synthesis on promoted iron catalysis. Acta Physiochimica. URSS 12 ; 1940 : 385-470.

31. Lagergren S. About the theory of so-called adsorption of soluble substances. K. Sven. Vetenskapsakad. Handl. 24 ; 1898 : 1-39.

32. Ho YS, Mc Kay G. A comparison of chemisorptions kinetic models applies to pollutant removal on various sorbents. Process Safety and Environmental Protection B. 76 ; 1998: 332-340.

33. Weber WJ, Morris JC. Kinetics of adsorption on carbon from solution. Journal of Sanitary Engineering Division, American Society Chemistry Engineering. 89 ; 1963 : 31-59.

34. Khan SA, Rehman R, Khan MA. Adsorption of chromium (III), chromium (VI) and silver (I) on bentonite. Waste Management. 15 (4) ; 1995: 271-282.

35. Nurchi VM, Crisponi G, Villaescusa I. Chemical equilibria in wastewaters during toxic metal ion removal by agricultural biomass. Coordination Chemistry Reviews. 254 ; 2010 : 2181-2192.

36. Fourest E, Volesky B. Contribution of sulfonate groups and alginate to heavy metal biosorption by the dry biomass of Sargassum uitans. Environmental Science and. Technology. 30 ; 1996 ; 277-282.

37. Loukidou MX, Zouboulis AI, Karapantsios TD, Matis KA. Equilibrium and kinetic modeling of chromium(VI) biosorption by Aeromonas caviae. Colloids and Surfaces. A 242. ; 2004 : 93-104.

38. Volesky B. Biosorption and me. Water Research. 41 ; 2007 : 4017-4029.

39. Ramrakhiani L, Majumder R, Khowala S. Removal of hexavalent chromium by heat inactivated fungal biomass of Termitomyces clypeatus: Surface characterization and mechanism of biosorption. Chemical Engineering Journal. 171 ; 2011 : 1060-1068.

40. Padmesh TVN, Vijayaraghavan K, Sekaran G, Velan M. Batch and column studies on biosorption of acid dyes on fresh water macro alga Azolla filiculoide. Journal of Hazardous Materials B. 125 ; 2005 : 121–129.

41. Yang F, Liu H, Qu J, Chen JP. Preparation and characterization of chitosan encapsulated sargassum sp. biosorbent for nickel ions sorption. Bioresource Technology. 102 ; 2011: 2821-2828.

42. Bernardin FE, Experimental design and testing of adsorption and adsorbates. In: Adsorption Technology: A Step-by-Step Approach to Process Evaluation and Application , Edited by Slejko FL, Chemical Industries, Marcel Dekker Inc., New York, vol. 19, 1985 : pp. 37–90.

43. Wahab MA, Jellali S, Jedidi N. Effect of temperature and pH on the biiosorption of ammonuium onto Posidonia oceanica ﬁbers: Equilibrium, and kinetic modeling studies, Bioresource Technology. 101 ; 2010 : 8606–8615.

44. Özer A, Özer D. Comparitive study of the biosorption of Pb (II), Ni (II) and Cr(VI) ions onto S. cerevisiae: Determination of biosorption heats. Journal of Hazardous Materials. 100 ; 2003 : 219-229.

45. Chou W, Wang C, Huang K, Chang Y, Shu C. Investigation of indium ions removal from aqueous solutions using spent coffee grounds. International Journal of Physical Sciences. 7(16) ; 2012 : 2445-2454.

46. Marques PA, Pinheiro HM, Teixeira JA, Rosa MF. Removal efficiency of cu²⁺, cd²⁺, Pb²⁺ by waste brewery biomass: pH and cation association effect. Desalination. 124(1) ; 1999 : 137-144.

47. Goksungur Y, Uren S, Guvenc U. Biosorption of copper ions by caustic treated waste baker’s yeast biomass. Turkish Journal of Biology. 27 ; 2003 : 23-29.

48. Treybal RE, Mass transfer operations, 2^nd ed., McGraw Hill, New York, 1968.

49. Ho YS, McKay G. Sorption of dye from aqueous solution by peat. Chemical Engineering Journal. 70 ; 1978: 115-124.

50. Weber WJ, Morris JC. Kinetics of adsorption on carbon from solution. American Society of Civil Engineering. 89 ; 1963: 31-60.

51. Wu J, Yu HQ. Biosorption of 2,4-dichlorophenol from aqueous solution by Phanerochaete chrysosporium biomass: Isotherms, kinetics and thermodynamics. Journal of Hazardous Materials B. 137 ; 2006 : 498-508.

Received on 01.08.2016 Modified on 10.08.2016

Research J. Pharm. and Tech 2016; 9(12):2113-2123.

DOI: 10.5958/0974-360X.2016.00430.3

S. No	Isotherm model	Parameters	Adsorbents
			Native	Tween 80 treated	SDS treated	NaOH treated
1	Langmuir	q_m(g/g)	0.379	0.766	0.471	0.448
		B	0.245	0.613	0.482	0.407
		r²	0.922	0.964	0.999	0.996
2	Freundlich	K_f(g/g) (l/g)^1/n	10.6	3.5	6.44	9.5
		N	1.745	2.217	1.561	1.527
		r²	0.866	0.884	0.837	0.876
3	Temkin	A (l/g)	0.562	1.105	0.587	0.726
		B	0.065	0.106	0.103	0.076
		r²	0.848	0.778	0.872	0.885

Adsorbent	q_eq,exp (g/g)	Pseudo-first order model			Pseudo-second order model
Adsorbent	q_eq,exp (g/g)	K₁	q_eq,cal(g g^-1)	R²	K₂ (g g^-1 min^-1)	q_eq,cal (g/g)	R²
Native	0.268	0.002	3.597	0.916	13.922	0.268	0.967
Tween 80 treated	0.500	0.007	0.51	0.999	0.012	0.401	0.971
SDS treated	0.345	0.006	0.373	0.953	9.21	0.349	0.971
NaOH treated	0.312	0.002	0.303	0.935	10.272	0.313	0.971

Adsorbents	Temperature n(K)	ΔG (kJ mol^-1)	ΔH (kJ mol^-1)	ΔS (kJ mol^-1 K^-1)	R²
Native adsorbent	283	-2.96	-2.76	0.0007	0.999
	293	-2.97
	303	-2.97
Tween 80 treated adsorbent	283	-9.4	-7.23	0.00769	1.000
	293	-9.48
	303	-9.56
SDS treated adsorbent	283	-7.55	-6.57	0.00347	0.990
	293	-7.58
	303	-7.62
NaOH treated adsorbent	283	-4.87	-4.39	0.0017	0.993
	293	-4.89
	303	-4.91