Preparation, characterization, and phenol adsorption of activated carbons from oxytetracycline bacterial residue

Abstract

The oxytetracycline bacterial residue–activated carbon (OBR-AC) prepared from oxytetracycline bacterial residue with K₂CO₃ under chemical activation was studied. The effects of activation temperature, activation time, and activation ratio on the specific surface area (SSA) and methylene blue adsorption (MBA) were studied. Characterization of the optimum OBR-AC was performed by using scanning electron microscopy (SEM), pore structure (PS,) and Fourier-transform infrared spectroscopy (FT-IR). The optimum parameters were as follows: 800°C activation temperature, 3 hr activation time, and 1:3 activation ratio. The SSA and MBA under optimum conditions were 1593.09 m²/g and 117.0 mg/g, respectively. Adsorption equilibrium and kinetics data were determined for the adsorption of phenol from the synthetically prepared phenol solution. The results showed that the Langmuir model gave the best fit for equilibrium isotherm, whereas the kinetics data were well fitted by the pseudo-second order model.

Implications

In the past, the bacterial residues have been used for feed additives in China. Unfortunately, doubts of its suitability as a feedstock have been raised because of the small amount of antibiotics, a large number of the fermentation by-products and metabolic products and by-products remaining in the bacterial residues. So Oxytetracycline Bacterial Residue (OBR) is one of hazardous wastes in China. In order to solve the problem of OBR, the preparation of OBR-AC is studied, and OBR-AC under optimum operation parameters is characterized by Scanning Eldctron Microscopy (SEM), Pore Structure (PS) and Fourier Transfer-Infra Red (FT-IR). Moreover, the phenol adsorption isotherms and kinetics models for OBR-AC under optimum operation parameters are studied.

Introduction

It is well known that the biopharmaceutical industry plays an indispensable role in the national economy. In 2009, Chinese production and export of antibiotics were ranked first in the world at 14.7 million and 2.5 million tons, respectively. The North China Pharmaceutical Company (NCPC), the chief biopharmaceutical firm in China, generates bacterial residue about 10,000 tons each year. In the past, the bacterial residues have been used for feed additives in China. Unfortunately, doubts about its suitability as a feedstock have been raised because of the small amount of antibiotics and large number of fermentation by-products and metabolic products and by-products remaining in the bacterial residues. Thus, oxytetracycline bacterial residue (OBR) is one of the hazardous wastes in China. Therefore, it should be managed strictly in accordance with the hazardous waste. It is urgent to find more effective treatments (in technical and economic terms) for OBR that match the most environmentally significant OBR, quantifying their reduction in existing treatment plants and then using promising new treatment methods that can be combined with existing ones to secure the necessary removal.

The world demand for activated carbons is expected to expand 5.2% per year through 2012 to 1.2 million tons (Freedonia Group, Inc., 2008). Activated carbon had been prepared from different raw materials and activation reagents. The BET surface area of bamboo-based activated carbon prepared from bamboo origin and modified by phosphoric acid activation was high (Cui et al., 2010). Coal-based activated carbons were prepared from Jiaozuo (Henan Province, China) anthracite with NaOH as activating agent, and the specific surface area was 2483 m²/g, total pore volume was 1.41 cm³/g, iodine adsorption value was 2530 mg/g, and methylene blue adsorption value was 418 mg/g (Xing et al., 2010). The raw materials used for activated carbon production, such as coal, coconut shell, and wood, are expensive. OBR is a kind of hazardous solid waste and contains protein, fat, cellulose, and a large amount of carbon. OBR-AC will recycle the carbon resources of OBR.

Chemical activation is an efficient method to obtain carbons with high specific surface area and micropore distribution. To get high specific surface area, KOH and NaOH are frequently used as activation reagents, but they are hazardous chemicals. K₂CO₃ is classified as pregnancy category C by the U.S. Food and Drug Administration. There has been no report for preparing oxytetracycline bacterial residue–activated carbon (OBR-AC) by chemical activation with K₂CO₃. McKee (1983) studied the gasification of graphite by alkali metal compounds and found that K₂CO₃ was reduced by carbon in an inert atmosphere as follows:

In wastewater, phenol is considered as one of the priority pollutants and a serious threat to human health and natural water quality (Bhatnagar et al., 2007). The methods to remove phenol from wastewater include photocatalytic degradation (Wu et al., 2011), chemical oxidation (Wang et al., 2009), membrane separation (Busca et al., 2008), microbial degradation and adsorption, and so on (El-Naas et al., 2010). Yet the adsorption process using activated carbons is the most favorable method due to its efficiency, high adsorption capacity, and best adsorption process. OBR-AC will reduce the absorption costs of phenol and phenolic wastewater to some degree.

In order to solve the problem of OBR, the preparation of OBR-AC is studied, and OBR-AC under optimum operation parameters is characterized by scanning electron microscopy (SEM), pore structure (PS), and Fourier-transform infrared spectroscopy (FT-IR). Moreover, the phenol adsorption isotherms and kinetics models for OBR-AC under optimum operation parameters are studied.

Experimental

Materials

OBR was obtained from the North China Pharmaceutical Company, Hebei, China, and was sieved (<0.6 mm) and dried at 105°C. The contents of C, H, O, N, and S of OBR were measured by a Vario EL CUBE elemental analyzer. The results were as follows: C 41.36%; H 4.996%; O 33.988%; N 8.850%; S 0.874%. K₂CO₃ (Tianjin Kemiou Chemical Reagent Co., Ltd., Tianjin, China) was used as the activation reagent.

Thermogravimetry–differential thermal analysis

Thermogravimetry–differential thermal analysis (TG-DTA) was performed with a DTG-60H instrument from Shimadzu. The sample was immersed in the solution of K₂CO₃ for 24 hr at 40°C, and then dried for 24 hr at 105°C. A 2-mg sample was heated from 20°C to 1000°C at a ramping rate of 10°C/min under a nitrogen gas atmosphere (Wimonrat et al., 2010).

Preparation of OBR-AC

Before use, OBR was dried at 105–110°C in an electric blast drying oven for 24 hr, and sieved to give fractions <0.9 mm in diameter. According to the different activation ratio, OBR was immersed in K₂CO₃ solution for 24 hr at 40°C. The impregnated samples were put in the ceramic crucibles with double cover (avoiding oxidation). In a muffle furnace (temperature range from 20°C to 1000°C, bought from Tianjin Taisite Instrument Co., Ltd., Tianjin, China) the samples were heated from room temperature to activation temperature and then kept at fixed activation temperature and activation time, and the heating rate was about 10°C/min. Then samples were cooled to room temperature. Then the samples were washed with HCl and distilled water to remove residual organic matters and K₂CO₃. The washing and filtration steps were repeated until the pH of the filtrating solution became neutral. The washed samples were dried for 2–3 hr at 100–105°C, and then OBR-AC samples were obtained. Specific surface area and methylene blue adsorption were selected as performance properties for preparation conditions in the paper. Specific surface area (SSA) was measured by the NOVA 2000e fully automatic surface area analyzer (Shimadzu, Japan). The concentration of methylene blue adsorption (MBA) was measured by using an ultraviolet (UV) 1102 spectrophotometer at 665 nm (GB/T12496.10-1999, China). The activation ratio was defined as follows:

1

where M_K2CO3 is the weight of K₂CO₃ (g) and M_OBR the weight of OBR (g).

Characterization methods

Surface morphology of OBR-AC (under optimum operation parameters) was investigated using S-48001 scanning electron microscopy. The samples were placed on the aluminum tub and coated with gold for electron reflection. The samples were vacuumed about 10 min before analysis.

The pore structure of OBR-AC (under optimum operation parameters) was characterized by N₂ adsorption using a NOVA 2000e analyzer. The samples were degassed at 300°C for 2 hr to remove any moisture or adsorption contaminants that may have been present on their surface. The software provided BET surface area (S_BET) of the carbons by applying the BET equation to the adsorption data. The micropore surface area (S_mic) and mesopore surface area (S_mes), the total pore volume (V_T), and micropore volume (V_mic) were evaluated by the t-plot method, and mesopore volume (V_mes) was estimated by the Barret–Joyner–Halenda method (Barret et al., 1951). The mean pore diameter (D_p) was calculated from Dp = 4V_T/S_BET (Gregg et al., 1967).

The experiment used the NICOLET 6700 Fourier-transform infrared analyzer to investigate the presence of OBR-AC (under optimum operation parameters) surface functional groups. A disc containing fine carbon with 0.1% (weight percentage) potassium bromide was prepared before analysis. The disc was then placed inside the analysis chamber and exposed to infrared light at a wavelength ranging from 400 to 4000 cm⁻¹ (Mohd-Din et al., 2009).

Phenol adsorption methods

OBR-AC (under optimum operation parameters) was boiled in deionized water at 100°C for 1 hr and then was dried in the electric blast drying oven for 24 hr before use. The pH in the experiment was always within the range 6–7. A certain mass of OBR-AC (under optimum operation parameters) was put into the glass bottles, and then a series of different initial concentrations of phenol solutions ranging from 100 mg/L to 300 mg/L was kept in the sealed glass bottles, which not only avoided the loss of phenol by volatilization, but also prevented the entry of oxygen (Terzyk et al., 2007). For the e adsorption equilibrium study, the samples were kept on a shaker (HY-4, China) at the room temperature (25 °C) for 24 h to reach equilibrium. For the kinetics study, the samples were withdrawn at regular intervals (El-Naas et al., 2010). Then the suspension of OBR-AC (under optimum operation parameters) was filtered, and diluted if needed (Girods et al., 2009). The residual phenol concentration was measured by UV 1102 spectrophotometer at 510 nm. The amount of phenol adsorption on OBR-AC was then calculated by difference between the initial (known) and the final (measured) phenol concentrations. The experiment data was calculated according to eq 2 and eq 3.

2

3

where q_e is the amount of adsorption at equilibrium condition (mg/g), q_t the amount of adsorption at time t (mg/g), C_o the initial adsorption concentrations (mg/L), C_e the equilibrium adsorption concentrations (mg/L), V the volume of solution (L), M the mass of OBR-AC (g), and C_t the adsorption concentration at a certain time (mg/L).

The adsorption isotherm provides essential physiochemical data for assessing the applicability of the adsorption process as a complete unit operation (Aydin et al., 2006). Langmuir and Freundlich adsorption isotherm models are widely used to investigate the adsorption process (Lata et al., 2008). The Langmuir adsorption isotherm is developed on the assumption that the adsorption process will only take place at specific homogeneous sites within the adsorption surface with uniform distribution of energy level. The Freundlich adsorption isotherm is based on the assumption that the adsorption occurs on heterogeneous sites with nonuniform distribution of energy level. It describes reversible adsorption and is not restricted to the formation of a monolayer (Ng et al., 2002). The equations of Langmuir and Freundlich are represented as follows:

4

5

where q_e is the amount of adsorption at equilibrium (mg/g), C_e the equilibrium concentration of the adsorption (mg/L), K_L the Langmuir adsorption isotherm constant (L/g), a_L the Langmuir adsorption isotherm constant (L/mg), K_F the Freundlich adsorption constant (mg/g)(g/L)^1/n , and n the Freundlich adsorption constant (dimensionless).

In order to study the adsorption of phenol solution on the OBR-AC (under optimum operation parameters) and to interpret the experimental data obtained, two kinetic models, the pseudo-first-order (Tutem et al., 1998) and pseudo-second-order (Ho et al., 2000), are listed as follows:

6

7

where q_e is the amount of adsorption at equilibrium (mg/g), q_t the amount of adsorption at time t (mg/g), k₁ the rate constant of the pseudo-first-order equation, and k₂ the rate constant of the pseudo-second-order equation.

The equations of standard deviation, (%), are defined as:

8

where n is the the number of data points (dimensionless), q_exp the experimental adsorption capacity (mg/g), and q_cal the calculated adsorption capacity (mg/g).

Results and Discussion

TG-DTA analysis

In Figure 1, TG-DTA curves show the mass loss at 20°C to 1000°C, and this mass loss can be divided into three stages. The first and maximum rate of weight loss can be seen about 300°C, and is caused by a large amount of crude protein of OBR decomposition in this temperature segment. The second peak of weight loss occurs at 510–620°C, mainly due to high organic contents in OBR. In addition, the chemical bond of the components is relatively weak and a large amount of organic contents is decomposed when the temperature reaches a certain level. The third peaks appear at 700–900°C. The main reason is decomposition of K₂CO₃. Therefore, the following experiments discuss the activation temperature from 500°C to 900°C.

Figure 1. TG-DTA of OBR (activation ratio 1:1; impregnation 40°C for 24 hr).

Preparation of OBR-AC: Effects of processing parameters

Effects of activation temperature

In Figure 2, each sample was heated from room temperature to activation temperature (500°C, 600°C, 700°C, 800°C, 900°C) and kept at fixed activation temperature (500°C, 600°C, 700°C, 800°C, 900°C) for 2 hr. It can be clearly observed that SSA increases from 350.73 m²/g to 758.14 m²/g with increasing activation temperature from 500°C to 800°C. Similarly, increasing activation temperature from 500ºC to 800°C shows an enhancement of MBA from 3.0 mg/g to 93.5 mg/g. Over 800°C, both SSA and MBA decrease.

Figure 2. Effects of activation temperature (activation ratio 1:1; activation time 2 hr).

The numbers of micropore and mesopore increase with the significantly increase of activation temperature. Also, SSA and MBA of the OBR-AC increase. The progressive activation temperature increases the C-K₂CO₃ reaction rate, resulting in increasing carbon “burn-off.” Concurrently, K₂CO₃ decomposes into K₂O and at the high-temperature K₂O is reverted as the atoms of the metal potassium, and the metal potassium enters into the carbon tissue, enhancing the amount of pores and increasing the performance of OBR-AC (Yang, 2000). At low temperatures, activation is not completed due to slow reaction rate between the reagent and OBR. As temperature increases, the porosity and specific surface area increase due to the release of tars from the cross-linked framework generated by the treatment of K₂CO₃. Over 800°C, SSA and MBA decrease, which may be due to the collapse of micropores formed during previous stages, as shown in Figure 2. Therefore, the activation temperature of 800 °C is chosen as the optimum condition for further experiments.

Effects of activation time

In Figure 3, increasing activation time shows an enhancement of SSA and MBA from 654.33 m²/g to 874.13 m²/g and from 74.5 mg/g to 125.5 mg/g, respectively. The phenomenon indicates that some volatile compounds in internal part of OBR-AC can be evaporated with longer activation time. The porosity and specific surface area increase with time (≤3 hr) due to the release of tars from cross-linked framework generated by the treatment of K₂CO₃. After 3 hr, the channels are broken and pore structures collapse. Therefore, the activation time of 3 hr is chosen as the optimum condition for further experiments.

Figure 3. Effects of activation time (activation ratio 1:1; activation temperature 800°C).

Effects of activation ratio

Effects of activation ratio on SSA and MBA of OBR-AC are shown in Figure 4. It can be observed that SSA increases from 588.71 m²/g to 1593.09 m²/g with increasing activation ratio from 2:1 to 1:3. Similarly, increasing activation ratio from 2:1 to 1:2 indicates an enhancement of MBA from 62.5 mg/g to 151.5 mg/g, and then decreases. Over 1:3, it shows that both SSA and the adsorptive capacity of OBR-AC decrease. For a fixed mass of OBR and K₂CO₃, the yield of OBR-AC increases with the increasing of the ratio of M_K2CO3/M_OBR. Under the condition of sufficient activation, the development of porosity is associated with the reduction of K₂CO₃ under an inert condition to form K and CO. The potassium would diffuse into the internal structure of carbon, widen the existing pores, and create new porosities, according to the reactions (McKee, 1983). However, when the activation ratio is over 1:3, both SSA and MBA decrease, with too low an activation ratio to activate fully the raw material.

Figure 4. Effects of activation ratio (activation temperature 800°C; activation time 3 hr).

On the basis of these experiments and considering the factors of economy and technology, the optimum conditions as follows: 800°C activation temperature, 3 hr activation time, and 1:3 activation ratio.

Characterization analysis of OBR-AC under optimum conditions

Scanning electron microscopy

The SEM of OBR-AC is shown in Figure 5 under optimum conditions. The surface and internal aspects of OBR-AC are porous, mostly micropores and mesopores. The ashes are seen as bright zones unevenly distributed on the carbon grains, which may be not washed clearly. Such ashes clearly correspond to the already-mentioned filled pores, resulting in a little effect on OBR-AC. The structure of the pores is homogeneous, and the distribution of pore size is relatively centralized, while at the same time the macropores are distributed and inserted among the micropores and mesopores, so they greatly increase surface area and adsorption capability of OBR-AC.

Figure 5. SEM of OBR-AC (activation temperature 800°C; activation time 3 hr; activation ratio 1:3).

Pore structure

The analysis of PS under the optimum conditions is listed in Table 1 and Figure 6. In Table 1, it can be seen that the micropore specific surface area and micropore volume of OBR-AC are higher than those of the mesopores. The massive micropores can improve SSA and adsorption capability. These parameters have been described very well in a recent paper (Passe-Coutrin et al., 2008). N₂ adsorption/desorption isotherms were shown in Figure 6. At lower pressure, it is found that the isotherm is more of type I character as defined by International Union of Pure and Applied Chemistry (IUPAC) classification, which suggests that the OBR-AC is characteristic of a micropore size distribution. At higher relative pressure, the isotherms display a hysteresis loop, which demonstrates the existence of mesopores in the OBR-AC and that the size distribution of pores became wider.

Table 1. Pore structure (entry sequence: activation ratio–activation temperature–activation time)

	Specific surface area (m^2/g)			Pore volume (cm^3/g)
Entry	SBET	Smic	Smes	VT	Vmic	Vmes	Dp (nm)
1:1-500-2	350.73	264.32	56.85	0.25	0.2	0.03	2.85
1:1-600-2	461.71	345.89	98.43	0.34	0.28	0.03	2.95
1:1-700-2	655.95	512.56	87.64	0.43	0.37	0.04	2.62
1:1-800-2	758.14	589.32	106.73	0.62	0.45	0.15	3.27
1:1-900-2	623.81	327.46	216.49	0.63	0.42	0.2	4.04
1:1-800-1	654.33	513.42	120.19	0.41	0.32	0.05	2.51
1:1-800-1.5	667.29	542.89	98.26	0.53	0.4	0.07	3.18
1:1-800-2.5	761.42	576.43	149.33	0.66	0.47	0.18	3.47
1:1-800-3	874.13	702.97	114.53	0.69	0.48	0.2	3.16
1:1-800-3.5	513.68	432.18	65.39	0.65	0.39	0.25	5.06
2:1-800-3	588.71	465.18	100.76	0.58	0.35	0.15	3.94
1:2-800-3	1084.10	875.33	168.73	0.71	0.5	0.21	2.62
1:3-800-3	1593.09	1365.85	197.36	0.87	0.54	0.27	2.18
1:4-800-3	1071.45	819.73	152.27	0.69	0.4	0.25	2.58

Figure 6. Nitrogen adsorption–desorption isotherms of OBR-AC (activation temperature 800°C; activation time 3 hr; activation ratio 1:3).

Fourier-transform infrared

The surface chemistry FT-IR spectrum of OBR-AC under the optimum conditions is presented in Figure 7. The spectrum exhibits –OH stretching vibration at wavelength 3401 cm⁻¹ (Wimonrat et al., 2010). The presence of alkynes and aromatics can be observed through C≡C and C=C stretching vibrations, respectively, at wavelengths 2315 cm⁻¹ and 1570 cm⁻¹ (Mohd-Din et al., 2009). C–OH shows at wavelength 1060 cm⁻¹ (Wimonrat et al., 2010). The phenol hydroxyl and π electron of phenol rings form the p–π conjugate and the hydroxyl group generates electron delocalization and tends to the side of phenol rings, so the hydroxyl group in the phenol rings cannot manacle the hydrogen atom in the hydroxyl group, so it is very easy to ionize to produce a hydrogen ion. C≡C and C=C on OBR-AC contain π electrons, and the hydrogen ion is an electrophilic reagent, so they generate adsorption, improving adsorption capability.

Figure 7. FT-IR of OBR-AC (activation temperature 800°C; activation time 3 hr; activation ratio 1:3).

Phenol Adsorption

Effects of temperature

Figure 8 shows that the amount of the phenol adsorption decreases with the temperature rising, but the adsorption capacity does not decrease significantly. Phenol is easy to remove from the exterior of activated carbon to the solution or the block active sites on activated carbon with the temperature increasing (Kilic et al., 2011).

Figure 8. Effects of temperature on phenol adsorption onto OBR-AC (pH 6–7).

Figure 9. Langmuir adsorption isotherm of phenol (pH 6–7; temperature 25°C).

Figure 10. Freundlich adsorption isotherm of phenol (pH 6–7; temperature 25°C).

Adsorption isotherm

Figures 9 and 10 and Table 2 exhibit Langmuir and Freundlich isotherms of phenol adsorption under the optimum conditions preparation OBR-AC. The R² values are evidence that the phenol adsorption is well fitted to the Langmuir adsorption isotherm. Because the Langmuir model takes place at the homogeneous surface, while the Freundlich adsorption model is suitable for the heterogeneous surface, these have researched by SEM for this paper. In addition, this phenomenon can be further explained by the surface chemistry of OBR-AC. Active sites with higher energy level tend to form heterolayer phenol coverage with robust support from strong chemical bonding, while active sites with lower energy level will induce monolayer coverage due to electrostatic forces (Mohd-Din et al., 2009).

Table 2. Langmuir and Freundlich coefficients for phenol adsorption

	Langmuir isotherm parameters				Freundlich isotherm parameters
Adsorbent	qe (mg/g)	KL (L/g)	a l (L/mg)	R ^2	KF (mg/g) (L/mg)^1/n	n	R ^2
OBR-AC	199.532	3.872	0.021	0.9802	19.143	4.07	0.8959

Adsorption kinetics models

Figure 11, Figure 12, and Table 3 show the applicability of the pseudo-first-order and pseudo-second-order equations. Adsorption kinetics describe the controlling mechanism of adsorption processes, which in turn govern the mass transfer and equilibrium time (Mestre et al., 2011). Kinetic data follows a pseudo-second-order model, which has higher correlation coefficients and lower standard deviation, . The pseudo-second-order model could describe the behavior over the whole process of adsorption. This may indicate that the adsorption process takes place via surface exchange reactions. With the processing of the reaction, most of the adsorption sites have been occupied, which limits activated sites for the phenol molecules to attach to.

Figure 11. (a) Pseudo-first-order kinetics for adsorption of phenol by OBR-AC (pH 6–7; temperature 25°C). (b) Standard deviation of pseudo-first-order kinetics for adsorption of phenol by OBR-AC (pH 6–7; temperature 25°C).

Table 3. Fitted kinetics parameters for the adsorption of phenol

		Pseudo-first-order				Pseudo-second-order
Concentration(mg/L)	qexp (mg/g)	qcal (mg/g)	k 1	R ^2	(%)	qcal (mg/g)	k 2	R ^2	(%)
100	169	153	0.0132	0.956	9.26	175	0.00108	0.9872	0.19
200	186	165	0.0207	0.918	5.82	192	0.00297	0.9889	0.15
300	201	179	0.0094	0.927	2.83	209	0.00543	0.9905	0.11

Figure 12. (a) Pseudo-second-order kinetics for adsorption of phenol by OBR-AC (pH 6–7; temperature 25°C). (b) Standard deviation of pseudo- second -order kinetics for adsorption of phenol by OBR-AC (pH 6–7; temperature 25°C).

Conclusion

The OBR-AC could be prepared by chemical activation with K₂CO₃ as the activation reagent. The optimum conditions for this were obtained as follows: 800°C activation temperature, 3 hr activation time, and 1:3 activation ratio. The Langmuir adsorption isotherm was well fitted for the equilibrium isotherm, which indicated a monolayer adsorption of phenol on the OBR-AC homogeneous surface. The pseudo-second-order model was the best model to describe the whole adsorption process of phenol. Conversion of OBR to activation carbon will help to solve the problem of pollution from OBR and part of the problem of phenol pollution.

Acknowledgments

The authors acknowledge the financial support provided by the Shijiazhuang Municipal Bureau of Science and Technology (project no. 09924393A-1). This research is supported by Natural Science Foundation of China (no. 21106033) and the Science and Technology Department of Hebei Province (no. B2012208037).