- Email: [email protected]
Removal of volatile fatty acids with immobilized Rhodococcus sp. B261
Bioresource Technology 96 (2005) 41–46
Removal of volatile fatty acids with immobilized Rhodococcus sp. B261 Soon-Il Yun
a,*
, Yoshiyuki Ohta
b
a
b
Food Science and Technology Major, Division of Biotechnology, Chonbuk National University, 664-14, Deokjin-Dong, Deokjin-Gu, Jeonju, Jeonbuk 651-756, South Korea Laboratory of Microbial Biochemistry, Faculty of Applied Biological Science, Hiroshima University, 1-4-4 Kagamiyama, Higashi-Hiroshima 739-8528, Japan Accepted 5 March 2004 Available online 18 May 2004
Abstract The removal of aqueous volatile fatty acids (VFA) in wastewater and spoiled waste-foods by immobilized Rhodococcus sp. B261 was investigated. The n-valeric acid (0.5%) was completely removed within 25 h under the following conditions; solution pH, 8.0; air flow rate, 0.2 l/min; superficial velocity, 0.96 h1 ; temperature, 37 C. Under the optimized conditions, the acetic (8525 ppm), propionic (7310 ppm) and n-butyric (4360 ppm) except n-valeric (2572 ppm) acids from the wastewater were completely removed by immobilized Rhodococcus sp. B261 in 24 h. The acetic (7810 ppm), propionic (8942 ppm) and butyric (5730 ppm) acids from the solution of spoiled waste-foods were effectively removed by immobilized Rhodococcus sp. B261 from 48 h within 60 h but n-valeric acid (3625 ppm) took 72 h. 2004 Elsevier Ltd. All rights reserved. Keywords: Malodor; Trickling biofilter; Volatile fatty acids; Wastewater
1. Introduction Volatile fatty acids (VFA) such as the n- and isoisomers of butyric and valeric acids are some of the main malodor compounds generating from livestock animal excreta (Hamano et al., 1972) and various kinds of wastewater caused by human life. The exhaust of these VFA, which were caused of malodor pollution, was restricted by the Ministry of Environmental Management in Japan (Kono, 1993). Microbial treatment of wastewater using biofilters and trickling biofilters has been applied in various industries due to their low operation costs and efficiency (Greiner and Timmons, 1998; Tracker and Ford, 1999; Kuai and Verstraete, 1999). Moreover, semi- and continuous systems of immobilized cells on various carriers have been used because of a prolongation of working life-span and protection of activity. The anaerobic elimination of VFA from wastewater has been done by the biofilter and biotrickling filters (Jimeno et al., 1990; Elefsiniotis and *
Corresponding author. E-mail address: [email protected] (S.-I. Yun).
0960-8524/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2004.03.006
Oldham, 1994; Ferchichi et al., 1994; Sai Ram et al., 1994; Gijzen and Kansiime, 1996; Wu et al., 1996). However, little is known about the aerobic elimination of VFA by immobilized microorganisms. We have reported the physiological properties of microorganisms that are capable of utilizing VFA as their sole carbon source (Yun and Ohta, 1997). One of the strains, Rhodococcus sp. B261 has the ability to utilize n-valeric acid, which is a stronger malodorous compound than the other VFA. In this study, the optimal conditions for the removal of the aqueous phase VFA by immobilized Rhodococcus sp. B261 were investigated using a trickling biofilter on a laboratory scale. Moreover, the removal of the aqueous phase VFA from the wastewater and the solution of spoiled waste-foods was performed.
2. Methods 2.1. Microorganism Rhodococcus sp. B261, used throughout the experiments, was isolated in our laboratory from the
42
S.-I. Yun, Y. Ohta / Bioresource Technology 96 (2005) 41–46
deodorized pig feces (Ohta and Ikeda, 1978). The cells on the slants cultured at 37 C and stored at 4 C till next subculture. 2.2. Medium The modified medium of Ohta and Sato (1985) was used for the cultivation and the immobilization of Rhodococcus sp. B261. The composition of the modified medium (g/l) was as follows: peptone, 5; KH2 PO4 , 1; MgSO4 Æ 7H2 O, 0.5; CaCl2 , 0.05; yeast extract, 0.5; CoCl2 Æ 6H2 O, 10 lM and 5 ml of VFA mixture (Yun and Ohta, 1997). The pH of the medium was adjusted to 8.0 (call this medium A for referring to later). When necessary, the VFA mixture was replaced with 0.5% (v/v) n-valeric acid. 2.3. Preparation of carrier The ceramic beads were composed of SiO2 (/3.5 mm) and supplied by Nippon Sharyo Co., Japan. They were prepared by washing with distilled water, drying at 100 C for 48 h and autoclaving at 121 C for 20 min. The physical properties of the ceramic beads and the basal operating conditions were shown in Table 1. The above basal conditions were replaced, when necessary. 2.4. Preparation of wastewater The 4 l of wastewater generated from the treatment of municipal waste was supplied from the Environmental and Hygienic Center of Higashi-Hiroshima, Japan, was filtered through cotton cloth to remove solid materials and pH was adjusted to 8.0. The VFA concentrations of the solution were as follows; acetic acid, 8500 ppm;
Table 1 The physical properties of porous ceramic beads and operating conditions for removal of aqueous VFA Ceramic beads
Bulk density Porosity
0.85 g cm3 62.5%
Column
Size (height · diameter) Packed material volume Void volume
50 cm · 5 cm 589 m3 393 m3
Reservoir
Volume Sample liquid volume Composition of circulating liquid
2l 500 ml n-Valeric acid only wastewater or solution of spoiled waste-food (pH 8.0)
Operating conditions
Superficial velocity (SV)
0.48 h1
Column temperature Moisture content
37 C 35%
propionic acid, 7310 ppm; n-butyric acid, 4370 ppm; nvaleric acid, 2530 ppm. Waste-foods was obtained from restaurant at Hiroshima University, Japan. The Waste-foods was contained varies kinds of foods such as rice, noodles, vegetables, fishes, and meats, etc. Five kg of waste-foods was kept into 10 l bottle with 5 l of tap water and then the mixture was left at 30 C for 5 days to spoil. The spoiled waste-foods was filtered through cotton cloth to remove the solid compounds. The pH of the solution was also adjusted to 8.0. The VFA concentrations of the solution determined by GC were as follows; acetic acid 7810 ppm; propionic acid 8942 ppm; n-butyric acid 5730 ppm and n-valeric acid 3625 ppm. 2.5. Culture condition and immobilization Four ml of the precultured was transferred into the 4 l fresh medium A containing the VFA mixture and cultivated at 37 C for 24 h on a rotary shaker at 240 rpm. The cells were harvested by centrifugation at 17; 000g for 15 min. and washed twice with distilled water. The harvested cells were resuspended in 100 ml of fresh medium A and subjected for immobilization. Five hundred grams of ceramic beads were put into a vacuum bottle covered with a silicone rubber stopper which was evacuated. The cell suspension was injected into the bottle with a syringe and shaken. The cells adhered and absorbed onto the ceramic beads were subsequently transferred to a sterile beaker and covered with a thin metal foil and incubated at 37 C for 24 h. Initial cell count was approximately 107 –108 CFU/g-ceramic beads. The moisture content of the ceramic beads was adjusted to 35% (v/w) by adding the fresh medium A. 2.6. Viable immobilized cell count The initial and final number of cells absorbed on the ceramic beads were determined as follows; five grams of ceramic beads with cells added as above were ground in a sterile mortar with a pestle and suspended in a 50 ml portion of saline solution. Hundred ll of the serially diluted suspension (105 –107 CFU/g-ceramic beads) was spread on medium A agar plates. After incubating at 37 C for 48 h, the colonies were counted as viable immobilized cells. 2.7. Experimental apparatus A schematic diagram of a deodorizing trickling biofilter is shown in Fig. 1. The ceramic beads (500 g, dry weight) immobilized with cells were packed into a glass column measuring 5 · 50 cm. The solution in a 2 l bottle was circulated by a peristaltic pump into the deodorizing column. Air was supplied by an air pump and its flow
S.-I. Yun, Y. Ohta / Bioresource Technology 96 (2005) 41–46
Fig. 1. Schematic diagram of a biofilter system, (a) air in; (b) air pump; (c) flow meter; (d) ceramic beads with cells; (e) 0.5% n-valeric acid (500 ml); (f) peristaltic pump; (g) NaOH solution (100 ml); (h) air out; (i) water in; (j) syringe; (k) glass column (50 · 5 cm); (l) 2 l reservoir; (m) 250 ml flask.
rate was controlled with a flow meter (Kofloc Model RK-1250, Japan). The solution passed through the column with air where it came in contact with the coated ceramic beads after which it finally dropped through the bottom. The temperature of the column was kept at 37 C using a water circulation system. The air emitted through the bottom of the column was trapped in 100 ml of NaOH (pH 8.0) solution. Samples (10 ml) from the 2 l bottle and from the NaOH solution were withdrawn with a syringe and subjected to gas chromatography at intervals. 2.8. Analysis of volatile fatty acids The concentration of VFA solution was determined by gas chromatography (Shimadzu, model GC-14B, Japan) equipped with FID. The column and injector temperature was 140 C; nitrogen carrier gas flow rate, 30 ml/min; air flow rate, 30 ml/min. The results were represented by averages of duplicates.
3. Results and discussion
43
Fig. 2. The effect of the concentration on the removal of n-valeric acid. Operating conditions; air flow rate, 0.2 l/min; SV, 0.48 h1 ; column temperature, 37 C; initial immobilized cell numbers, 5.7 · 108 CFU/ g-ceramic beads; moisture content was adjusted to 35% by adding the basal medium; initial column and n-valeric acid solution pH, 8.0. Error bars represent the standard deviations of duplicate determinations. Symbols (concentration, v/v); ( ) 0.1%; (j) 0.5%; ( ) 1.0%; () 2.0%; (M) 4.0%.
removals were 58%, 25%, and 12%, respectively, for 54 h. Initial cell number was 3.74 ± 0.52 · 107 . After the run, the final cell number enumerated from 0.1%, 0.5%, 1.0%, 2.0%, and 4.0% of n-valeric acid were 1.59 ± 0.18 · 109 , 3.12 ± 0.92 · 109 , 5.01 ± 0.73 · 109 , 6.16 ± 0.47 · 109 , respectively. The actual amount (mg/1) of n-valeric acid removed were about 5800 (from 1% solution), 5000 (from 2% solution), and 4600 (from 4% solution) ppm in 42 h (Fig. 3). The removed amount of 4% n-valeric acid solution was low level of 430 mg l1 in 18 h. It was suggested that the cells were acclimatised in the high concentration of acid for 18 h. And the removal was dynamically increased with same level of 1% and 2% n-valeric acid solution until 42 h. The removal of 2%, and 4% of nvaleric acid were lower than that of 1% solution after 42 h. This shows that the higher concentration of n-valeric acid than 1% is cause of decreasing removal. The fact was guessed that the lower removal of n-valeric acid compare with other experiment such as air flow rate and pH was caused by the lower level of initial cell number.
3.1. Effect of initial concentration of n-valeric acid on the removal rate
3.2. Effect of air flow rate
As shown in Fig. 2, n-valeric acid was completely removed in 0.1% and 0.5% n-valeric acid in 54 h. When 1.0%, 2.0% and 4.0% n-valeric acid were used, the
When the air flow rate was 0.2 and 0.4 l/min, the 0.5% of solution was completely removed in 24 h. The highest removal at an air flow rate of 0.2 1/min was shown. The
44
S.-I. Yun, Y. Ohta / Bioresource Technology 96 (2005) 41–46
3.3. Influence of solution pH
Fig. 3. The accumulative removed n-valeric acid from different concentration. Symbols (ppm); ( ) 1000; (j) 5000; ( ) 10000; () 20000; (M) 40000.
The 0.5% n-valeric acid solution was completely removed at pH 8.0 in 32 h (Fig. 5). When the initial pH is adjusted to 5.0, 6.0, 9.0 and 11, the removal was, respectively, 67%, 85%, 90% and 78% in 32 h. This indicates the necessity to control the initial solution pH to 8.0 and 9.0 for the effective removal of n-valeric acid. However, the control of the pH during the operation for the removal of VFA by Rhodococcus sp. B261 is not required because the column pH increases from initial pH of 7.0 to about 8.5–8.9 by the growth of the cells. This fact was suggested that the change in pH will be caused by formation of NHþ 4 by growing cell. The pH of 11 or 5, 6 are correctly not optimum pH for growth of Rhodococcus sp. B261. Despite the pH is not optimum for growth of the cell, the acids were up taken by the cells. But the pH of 5 was very lower removal than that of 11. It means that the initial pH of the column have to control up to 5 to obtain higher removal of acids in this system. 3.4. Effect of immobilized cell number The immobilized cell numbers were 7.50 ± 1.80 · 104 , 1.85 ± 0.33 · 106 , 5.05 ± 0.48 · 108 , and 1.40 ± 0.35 · 109 CFU/g-ceramic beads (Table 2). When the initial cell number of 1.40 ± 0.35 · 109 CFU/g-ceramic beads was used, 0.5% n-valeric acid was completely removed in 30 h under the following conditions: superficial velocity
Fig. 4. The effect of air flow rate on the removal of n-valeric acid. Operating conditions: inlet n-valeric acid concentration, 0.5%; SV, 0.48 h1 ; column temperature, 37 C; initial column and n-valeric acid solution pH, 8.0; moisture content was adjusted to 35% with the medium, initial immobilized cell number were presented in Table 2. Symbols (l/min); (}) 0.1; (j) 0.2; (M) 0.4; ( ) 0.8; (N) 1.2.
removal for an air flow of 0.1, 0.8 and 1.2 l/min were lower than that of 0.2 and 0.4 l/min (Fig. 4). It was suggested that the higher air flow rate than that of 0.4 l/ min caused low removal because of the short retention time of substrate in the column even though oxygen was enough. Moreover the low air flow rate also caused of low removal of n-valeric acid. It could be guessed that the oxygen was not enough to grow of the cells even though oxygen was not determined.
Fig. 5. The effect of solution pH on the removal of n-valeric acid. Operating conditions; inlet n-valeric acid concentration, 0.5%; SV, 0.48 h1 ; air flow rate, 0.2 l/min; column temperature, 27 C; initial moisture content, 35%. Error bars represent the standard deviations of duplicate determinations. Symbols: (M) pH 5.0; ( ) pH 6.0; (j) pH 8.0; () pH 9.0; ( ) pH 11.0.
S.-I. Yun, Y. Ohta / Bioresource Technology 96 (2005) 41–46
45
Table 2 Changes in cell numbers of Rhodococcus sp. B261 immobilized onto the ceramic beads in various conditions for removal of n-valeric acid solution Air flow rate (l/min)
0.2
0.4
0.6
0.8
Initial (·108 ) Final (·109 )
5.41 ± 0.67 4.72 ± 0.58
4.95 ± 1.04 5.24 ± 1.21
6.43 ± 0.95 4.98 ± 0.86
5.23 ± 0.81 3.95 ± 0.75
5.17 ± 0.71 2.04 ± 1.01
Initial pH Initial (·108 ) Final (·109 )
5.0 6.20 ± 1.25 1.50 ± 1.45
6.0 5.30 ± 1.61 6.70 ± 1.57
8.0 4.25 ± 0.48 3.00 ± 0.41
9.0 4.20 ± 0.81 2.70 ± 0.31
11.0 5.15 ± 0.64 1.60 ± 0.27
1.85 ± 0.33 · 106 7.16 ± 0.57 · 107
5.05 ± 0.48 · 108 9.02 ± 1.05 · 109
1.40 ± 0.35 · 109 9.83 ± 0.72 · 109
Immobilized cell number Initial 7.50 ± 1.80 · 104 Final 4.94 ± 0.95 · 104
1.2
All values (in CFU/g-ceramic beads) were the average ± standard deviation (STDEVP).
3.5. Removal of aqueous volatile fatty acids from wastewater and waste-foods From the above results, the operating conditions were optimized as follows: cell number, 5.6 ± 0.72 · 108 for wastewater, and 3.7 ± 0.13 · 108 cells/g-ceramic beads for waste-foods; air flow rate, 0.2 l/min; solution pH, 8.0; superficial velocity, 0.48 h1 ; temperature, 37 C. The acids from the wastewater were completely removed for 24 h by the immobilized Rhodococcus sp. B261 as shown in Fig. 7. Acetic acid (8525 ppm) and propionic acid (7310 ppm) with high concentration were effectively removed for 12 h. However, n-buryric (4360 ppm) and n-valeric acid (2572 ppm) were completely removed within 34 and 48 h, respectively. The pH 8.0 of the solution was increased to 9.67. After 24 h, the colour
Fig. 6. The effect of the initial immobilized cell number on the removal of n-valeric acid. Operating condition: inlet n-valeric acid concentration, 0.5%; air flow rate, 0.2 l/min; SV, 0.48 h1 ; column temperature, 37 C; initial column and n-valeric acid solution pH, 8.0; moisture content was adjusted to 35% by adding the medium. Error bars represent the standard deviations of duplicate determinations. Symbols (CFU/g-ceramic beads); () cell free; ( ) 1.40 ± 0.35 · 109 ; (j) 5.05 ± 0.48 · 108 ; ( ) 4.85 ± 0.33 · 106 ; (}) 7.50 ± 1.80 · 104 .
(SV), 0.48 h1 ; air flow rate, 0.2 l/min; solution pH, 8.0. The removal rates by 5.05 ± 0.48 · 108 , 1.85 ± 0.33 · 106 and 7.50 ± 1.80 · 104 CFU/g-ceramic beads, were only about 85%, 63% and 43%, respectively (Fig. 6). This indicates that the more cells immobilized, the more efficient n-valeric acid could be removed. The immobilized cell number after the reaction increased approximately by one log-unit as shown in Table 2. In addition, the higher removal rate could be explained with regard to the cell number which could proliferate better in the presence of the medium (data not shown) due to the growth of the cells in the immobilized form. The same result was observed in the degradation of 50% of 2,4-dichlorophenoxy-acetic acid solution by the 108 CPU/beads viable cell count (Kochar and Kahlon, 1995).
Fig. 7. Removal of various volatile fatty acids from wastewater. Error bars represent the standard deviations of duplicate determinations. Symbols: ( ) acetic acid; ( ) propionic acid; () n-butyric acid; (j) n-valeric acid; (}) cell free.
46
S.-I. Yun, Y. Ohta / Bioresource Technology 96 (2005) 41–46
of the beads inside column was changed to yellow, due to the growth of Rhodococcus sp. B261. It is suggest that Rhodococcus sp. B261 was dominated compared with the extended cells in the wastewater, therefore all acids were effectively utilized by them. Advancing the operation, the solution became limpid because of the precipitation of solid materials. The pH was increased. It was probably caused by NHþ 4 generated from the cell growth. After the operation, the final living cell numbers in the column were increased from 5.6 ± 0.72 · 108 to 7.3 ± 1.04 · 109 CFU/g-ceramic beads. Moreover, the any other cell was not detected in the medium containing only VFA as carbon source. This indicates the effective removal of the acids was caused by Rhodococcus sp. B261. In the case of the solution from the waste-food, the acids were difficult to remove by Rhodococcus sp. B261 during the initial period of 48 h (Fig. 8). This may be due to the solid compounds which prevent the reaction between acids and cells. However, the optical density of the solution with solid compounds was declined from 7.3 to 1.5 (data not shown). The removal of acetic acid (7810 ppm), propionic (8942 ppm), n-butyric (5730 ppm) and n-valeric (3625 ppm) acids from the solution was initiated at 48 h and then acids except n-valeric acid were dynamically removed in 64 h by the cells. The n-butyric (5730 ppm) and n-valeric (3625 ppm) acid were gradually decreased and removed within 72 h. The viable cell number of Rhodococcus sp. B261 were increased from 3.7 ± 0.13 · 108 to 2.9 ± 0.47 · 109 CFU/g-ceramic beads. The acids in NaOH solution (pH 8.0) from the bottle (g) was not detectable until 36 h (for wastewater) and
Fig. 8. Removal of various volatile fatty acids from the solution of spoiled waste-food. Error bars represent the standard deviations of duplicate determinations. Symbols: ( ) acetic acid; ( ) propionic acid; () n-butyric acid; (j) n-valeric acid; (}) cell free.
60 h (for spoiled waste-foods solution) by gas chromatography. It shows that acids were not volatilized from the bottom of the column by the system. Even though some acids were volatilized, it was also effectively removed by the immobilized cells. This is a report on the removal of n-valeric acid by a trickling biofilter system. The removal of wastewater and waste-foods was longer than that of using pure acids. It was caused by many kinds of unknown materials in the wastewater and waste-foods. Also, the high concentration and different kinds of acids cause of the longer removal time for n-valeric acid. However, the data presented here show that trickling biofilter system under the optimum conditions can be applied to the removal of high concentration of VFA from wastewater and waste-foods under the aerobic condition.
References Elefsiniotis, P., Oldham, W.K., 1994. Anaerobic acidogenesis of primary sludge: the role of solids retention time. Biotechnol. Bieng. 44, 7–13. Ferchichi, M., Ghrabi, A., Grasmick, A., 1994. Urban wastewater treatment by trickling filter and rotating biological reactor. Water Res. 28, 437–443. Gijzen, H.J., Kansiime, F., 1996. Comparison of start-up of an upflow anaerobic sludge blanket reactor and a polyurethane carrier reactor. Water Sci. Technol. 34, 5–6. Greiner, A.D., Timmons, M.B., 1998. Evaluation of the nitrification rates of microbead and trickling filters in an intensive. Aquacult. Eng. 18, 189–200. Hamano, T., Oka, Y., Takada, O., Asano, T., 1972. Test of malodor composition in the feces of domestic animals. Bull. Hyogoprefect. Exp. Stan. Anim. Husbandry 9, 140–145. Jimeno, A., Bermudez, J.J., Canovas-Doaz, M., Manjon, A., Iborra, J.L., 1990. Degradation of isobalerate and 2-methylburate in an anaerobic trickling filter. Biol. Wastes 34, 241–250. Kochar, G.S., Kahlon, R.S., 1995. Degradation of 2,4-dichlorophenoxy-acetic acid (2,4-D) by immobilized cells of Pseudomonas putida. J. Gen. Appl. Microbiol. 41, 367–370. Kono, K.Y., 1993. Malodor Preventive Low: Editorial Supervision for Special Pollution Section of Air Preservation Department of the Ministry of Environment in Japan. Gyousei. Inc. pp. 13–35. Kuai, L., Verstraete, W., 1999. Autotrophic degasification with elemental sulphur in small-scale wastewater treatment facilities. Environ. Technol. 20, 201–209. Ohta, Y., Ikeda, M., 1978. Deodorization of pig feces by Actinomycetes. Appl. Environ. Microbiol. 36, 487–491. Ohta, Y., Sato, H., 1985. An artificial medium for deodorant microorganisms. Agric. Biol. Chem. 49, 1195–1196. Sai Ram, M., Singh, L., Suryanarayana, M.V.S., Alam, S.I., 1994. Effect of sulfate and nitrate on anaerobic oxidation of volatile fatty acids in rabbit waste at 20 C. J. Gen. Appl. Microbiol. 41, 181– 189. Tracker, B.K., Ford, C.G., 1999. In situ bioremediation technique for sites underlain by silt and clay. J. Environ. Eng. 125, 1169–1172. Wu, W.-M., Jain, M.K., Zeikus, J.G., 1996. Formation of fatty aciddegrading, anaerobic granules by defined species. Appl. Environ. Microbiol. 62, 2037–2044. Yun, S.-I., Ohta, Y., 1997. Some physiological properties of microorganisms capable of deodorizing farm animal feces. Bioresour. Technol. 60, 21–26.
Removal of volatile fatty acids with immobilized Rhodococcus sp. B261 Soon-Il Yun
a,*
, Yoshiyuki Ohta
b
a
b
Food Science and Technology Major, Division of Biotechnology, Chonbuk National University, 664-14, Deokjin-Dong, Deokjin-Gu, Jeonju, Jeonbuk 651-756, South Korea Laboratory of Microbial Biochemistry, Faculty of Applied Biological Science, Hiroshima University, 1-4-4 Kagamiyama, Higashi-Hiroshima 739-8528, Japan Accepted 5 March 2004 Available online 18 May 2004
Abstract The removal of aqueous volatile fatty acids (VFA) in wastewater and spoiled waste-foods by immobilized Rhodococcus sp. B261 was investigated. The n-valeric acid (0.5%) was completely removed within 25 h under the following conditions; solution pH, 8.0; air flow rate, 0.2 l/min; superficial velocity, 0.96 h1 ; temperature, 37 C. Under the optimized conditions, the acetic (8525 ppm), propionic (7310 ppm) and n-butyric (4360 ppm) except n-valeric (2572 ppm) acids from the wastewater were completely removed by immobilized Rhodococcus sp. B261 in 24 h. The acetic (7810 ppm), propionic (8942 ppm) and butyric (5730 ppm) acids from the solution of spoiled waste-foods were effectively removed by immobilized Rhodococcus sp. B261 from 48 h within 60 h but n-valeric acid (3625 ppm) took 72 h. 2004 Elsevier Ltd. All rights reserved. Keywords: Malodor; Trickling biofilter; Volatile fatty acids; Wastewater
1. Introduction Volatile fatty acids (VFA) such as the n- and isoisomers of butyric and valeric acids are some of the main malodor compounds generating from livestock animal excreta (Hamano et al., 1972) and various kinds of wastewater caused by human life. The exhaust of these VFA, which were caused of malodor pollution, was restricted by the Ministry of Environmental Management in Japan (Kono, 1993). Microbial treatment of wastewater using biofilters and trickling biofilters has been applied in various industries due to their low operation costs and efficiency (Greiner and Timmons, 1998; Tracker and Ford, 1999; Kuai and Verstraete, 1999). Moreover, semi- and continuous systems of immobilized cells on various carriers have been used because of a prolongation of working life-span and protection of activity. The anaerobic elimination of VFA from wastewater has been done by the biofilter and biotrickling filters (Jimeno et al., 1990; Elefsiniotis and *
Corresponding author. E-mail address: [email protected] (S.-I. Yun).
0960-8524/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2004.03.006
Oldham, 1994; Ferchichi et al., 1994; Sai Ram et al., 1994; Gijzen and Kansiime, 1996; Wu et al., 1996). However, little is known about the aerobic elimination of VFA by immobilized microorganisms. We have reported the physiological properties of microorganisms that are capable of utilizing VFA as their sole carbon source (Yun and Ohta, 1997). One of the strains, Rhodococcus sp. B261 has the ability to utilize n-valeric acid, which is a stronger malodorous compound than the other VFA. In this study, the optimal conditions for the removal of the aqueous phase VFA by immobilized Rhodococcus sp. B261 were investigated using a trickling biofilter on a laboratory scale. Moreover, the removal of the aqueous phase VFA from the wastewater and the solution of spoiled waste-foods was performed.
2. Methods 2.1. Microorganism Rhodococcus sp. B261, used throughout the experiments, was isolated in our laboratory from the
42
S.-I. Yun, Y. Ohta / Bioresource Technology 96 (2005) 41–46
deodorized pig feces (Ohta and Ikeda, 1978). The cells on the slants cultured at 37 C and stored at 4 C till next subculture. 2.2. Medium The modified medium of Ohta and Sato (1985) was used for the cultivation and the immobilization of Rhodococcus sp. B261. The composition of the modified medium (g/l) was as follows: peptone, 5; KH2 PO4 , 1; MgSO4 Æ 7H2 O, 0.5; CaCl2 , 0.05; yeast extract, 0.5; CoCl2 Æ 6H2 O, 10 lM and 5 ml of VFA mixture (Yun and Ohta, 1997). The pH of the medium was adjusted to 8.0 (call this medium A for referring to later). When necessary, the VFA mixture was replaced with 0.5% (v/v) n-valeric acid. 2.3. Preparation of carrier The ceramic beads were composed of SiO2 (/3.5 mm) and supplied by Nippon Sharyo Co., Japan. They were prepared by washing with distilled water, drying at 100 C for 48 h and autoclaving at 121 C for 20 min. The physical properties of the ceramic beads and the basal operating conditions were shown in Table 1. The above basal conditions were replaced, when necessary. 2.4. Preparation of wastewater The 4 l of wastewater generated from the treatment of municipal waste was supplied from the Environmental and Hygienic Center of Higashi-Hiroshima, Japan, was filtered through cotton cloth to remove solid materials and pH was adjusted to 8.0. The VFA concentrations of the solution were as follows; acetic acid, 8500 ppm;
Table 1 The physical properties of porous ceramic beads and operating conditions for removal of aqueous VFA Ceramic beads
Bulk density Porosity
0.85 g cm3 62.5%
Column
Size (height · diameter) Packed material volume Void volume
50 cm · 5 cm 589 m3 393 m3
Reservoir
Volume Sample liquid volume Composition of circulating liquid
2l 500 ml n-Valeric acid only wastewater or solution of spoiled waste-food (pH 8.0)
Operating conditions
Superficial velocity (SV)
0.48 h1
Column temperature Moisture content
37 C 35%
propionic acid, 7310 ppm; n-butyric acid, 4370 ppm; nvaleric acid, 2530 ppm. Waste-foods was obtained from restaurant at Hiroshima University, Japan. The Waste-foods was contained varies kinds of foods such as rice, noodles, vegetables, fishes, and meats, etc. Five kg of waste-foods was kept into 10 l bottle with 5 l of tap water and then the mixture was left at 30 C for 5 days to spoil. The spoiled waste-foods was filtered through cotton cloth to remove the solid compounds. The pH of the solution was also adjusted to 8.0. The VFA concentrations of the solution determined by GC were as follows; acetic acid 7810 ppm; propionic acid 8942 ppm; n-butyric acid 5730 ppm and n-valeric acid 3625 ppm. 2.5. Culture condition and immobilization Four ml of the precultured was transferred into the 4 l fresh medium A containing the VFA mixture and cultivated at 37 C for 24 h on a rotary shaker at 240 rpm. The cells were harvested by centrifugation at 17; 000g for 15 min. and washed twice with distilled water. The harvested cells were resuspended in 100 ml of fresh medium A and subjected for immobilization. Five hundred grams of ceramic beads were put into a vacuum bottle covered with a silicone rubber stopper which was evacuated. The cell suspension was injected into the bottle with a syringe and shaken. The cells adhered and absorbed onto the ceramic beads were subsequently transferred to a sterile beaker and covered with a thin metal foil and incubated at 37 C for 24 h. Initial cell count was approximately 107 –108 CFU/g-ceramic beads. The moisture content of the ceramic beads was adjusted to 35% (v/w) by adding the fresh medium A. 2.6. Viable immobilized cell count The initial and final number of cells absorbed on the ceramic beads were determined as follows; five grams of ceramic beads with cells added as above were ground in a sterile mortar with a pestle and suspended in a 50 ml portion of saline solution. Hundred ll of the serially diluted suspension (105 –107 CFU/g-ceramic beads) was spread on medium A agar plates. After incubating at 37 C for 48 h, the colonies were counted as viable immobilized cells. 2.7. Experimental apparatus A schematic diagram of a deodorizing trickling biofilter is shown in Fig. 1. The ceramic beads (500 g, dry weight) immobilized with cells were packed into a glass column measuring 5 · 50 cm. The solution in a 2 l bottle was circulated by a peristaltic pump into the deodorizing column. Air was supplied by an air pump and its flow
S.-I. Yun, Y. Ohta / Bioresource Technology 96 (2005) 41–46
Fig. 1. Schematic diagram of a biofilter system, (a) air in; (b) air pump; (c) flow meter; (d) ceramic beads with cells; (e) 0.5% n-valeric acid (500 ml); (f) peristaltic pump; (g) NaOH solution (100 ml); (h) air out; (i) water in; (j) syringe; (k) glass column (50 · 5 cm); (l) 2 l reservoir; (m) 250 ml flask.
rate was controlled with a flow meter (Kofloc Model RK-1250, Japan). The solution passed through the column with air where it came in contact with the coated ceramic beads after which it finally dropped through the bottom. The temperature of the column was kept at 37 C using a water circulation system. The air emitted through the bottom of the column was trapped in 100 ml of NaOH (pH 8.0) solution. Samples (10 ml) from the 2 l bottle and from the NaOH solution were withdrawn with a syringe and subjected to gas chromatography at intervals. 2.8. Analysis of volatile fatty acids The concentration of VFA solution was determined by gas chromatography (Shimadzu, model GC-14B, Japan) equipped with FID. The column and injector temperature was 140 C; nitrogen carrier gas flow rate, 30 ml/min; air flow rate, 30 ml/min. The results were represented by averages of duplicates.
3. Results and discussion
43
Fig. 2. The effect of the concentration on the removal of n-valeric acid. Operating conditions; air flow rate, 0.2 l/min; SV, 0.48 h1 ; column temperature, 37 C; initial immobilized cell numbers, 5.7 · 108 CFU/ g-ceramic beads; moisture content was adjusted to 35% by adding the basal medium; initial column and n-valeric acid solution pH, 8.0. Error bars represent the standard deviations of duplicate determinations. Symbols (concentration, v/v); ( ) 0.1%; (j) 0.5%; ( ) 1.0%; () 2.0%; (M) 4.0%.
removals were 58%, 25%, and 12%, respectively, for 54 h. Initial cell number was 3.74 ± 0.52 · 107 . After the run, the final cell number enumerated from 0.1%, 0.5%, 1.0%, 2.0%, and 4.0% of n-valeric acid were 1.59 ± 0.18 · 109 , 3.12 ± 0.92 · 109 , 5.01 ± 0.73 · 109 , 6.16 ± 0.47 · 109 , respectively. The actual amount (mg/1) of n-valeric acid removed were about 5800 (from 1% solution), 5000 (from 2% solution), and 4600 (from 4% solution) ppm in 42 h (Fig. 3). The removed amount of 4% n-valeric acid solution was low level of 430 mg l1 in 18 h. It was suggested that the cells were acclimatised in the high concentration of acid for 18 h. And the removal was dynamically increased with same level of 1% and 2% n-valeric acid solution until 42 h. The removal of 2%, and 4% of nvaleric acid were lower than that of 1% solution after 42 h. This shows that the higher concentration of n-valeric acid than 1% is cause of decreasing removal. The fact was guessed that the lower removal of n-valeric acid compare with other experiment such as air flow rate and pH was caused by the lower level of initial cell number.
3.1. Effect of initial concentration of n-valeric acid on the removal rate
3.2. Effect of air flow rate
As shown in Fig. 2, n-valeric acid was completely removed in 0.1% and 0.5% n-valeric acid in 54 h. When 1.0%, 2.0% and 4.0% n-valeric acid were used, the
When the air flow rate was 0.2 and 0.4 l/min, the 0.5% of solution was completely removed in 24 h. The highest removal at an air flow rate of 0.2 1/min was shown. The
44
S.-I. Yun, Y. Ohta / Bioresource Technology 96 (2005) 41–46
3.3. Influence of solution pH
Fig. 3. The accumulative removed n-valeric acid from different concentration. Symbols (ppm); ( ) 1000; (j) 5000; ( ) 10000; () 20000; (M) 40000.
The 0.5% n-valeric acid solution was completely removed at pH 8.0 in 32 h (Fig. 5). When the initial pH is adjusted to 5.0, 6.0, 9.0 and 11, the removal was, respectively, 67%, 85%, 90% and 78% in 32 h. This indicates the necessity to control the initial solution pH to 8.0 and 9.0 for the effective removal of n-valeric acid. However, the control of the pH during the operation for the removal of VFA by Rhodococcus sp. B261 is not required because the column pH increases from initial pH of 7.0 to about 8.5–8.9 by the growth of the cells. This fact was suggested that the change in pH will be caused by formation of NHþ 4 by growing cell. The pH of 11 or 5, 6 are correctly not optimum pH for growth of Rhodococcus sp. B261. Despite the pH is not optimum for growth of the cell, the acids were up taken by the cells. But the pH of 5 was very lower removal than that of 11. It means that the initial pH of the column have to control up to 5 to obtain higher removal of acids in this system. 3.4. Effect of immobilized cell number The immobilized cell numbers were 7.50 ± 1.80 · 104 , 1.85 ± 0.33 · 106 , 5.05 ± 0.48 · 108 , and 1.40 ± 0.35 · 109 CFU/g-ceramic beads (Table 2). When the initial cell number of 1.40 ± 0.35 · 109 CFU/g-ceramic beads was used, 0.5% n-valeric acid was completely removed in 30 h under the following conditions: superficial velocity
Fig. 4. The effect of air flow rate on the removal of n-valeric acid. Operating conditions: inlet n-valeric acid concentration, 0.5%; SV, 0.48 h1 ; column temperature, 37 C; initial column and n-valeric acid solution pH, 8.0; moisture content was adjusted to 35% with the medium, initial immobilized cell number were presented in Table 2. Symbols (l/min); (}) 0.1; (j) 0.2; (M) 0.4; ( ) 0.8; (N) 1.2.
removal for an air flow of 0.1, 0.8 and 1.2 l/min were lower than that of 0.2 and 0.4 l/min (Fig. 4). It was suggested that the higher air flow rate than that of 0.4 l/ min caused low removal because of the short retention time of substrate in the column even though oxygen was enough. Moreover the low air flow rate also caused of low removal of n-valeric acid. It could be guessed that the oxygen was not enough to grow of the cells even though oxygen was not determined.
Fig. 5. The effect of solution pH on the removal of n-valeric acid. Operating conditions; inlet n-valeric acid concentration, 0.5%; SV, 0.48 h1 ; air flow rate, 0.2 l/min; column temperature, 27 C; initial moisture content, 35%. Error bars represent the standard deviations of duplicate determinations. Symbols: (M) pH 5.0; ( ) pH 6.0; (j) pH 8.0; () pH 9.0; ( ) pH 11.0.
S.-I. Yun, Y. Ohta / Bioresource Technology 96 (2005) 41–46
45
Table 2 Changes in cell numbers of Rhodococcus sp. B261 immobilized onto the ceramic beads in various conditions for removal of n-valeric acid solution Air flow rate (l/min)
0.2
0.4
0.6
0.8
Initial (·108 ) Final (·109 )
5.41 ± 0.67 4.72 ± 0.58
4.95 ± 1.04 5.24 ± 1.21
6.43 ± 0.95 4.98 ± 0.86
5.23 ± 0.81 3.95 ± 0.75
5.17 ± 0.71 2.04 ± 1.01
Initial pH Initial (·108 ) Final (·109 )
5.0 6.20 ± 1.25 1.50 ± 1.45
6.0 5.30 ± 1.61 6.70 ± 1.57
8.0 4.25 ± 0.48 3.00 ± 0.41
9.0 4.20 ± 0.81 2.70 ± 0.31
11.0 5.15 ± 0.64 1.60 ± 0.27
1.85 ± 0.33 · 106 7.16 ± 0.57 · 107
5.05 ± 0.48 · 108 9.02 ± 1.05 · 109
1.40 ± 0.35 · 109 9.83 ± 0.72 · 109
Immobilized cell number Initial 7.50 ± 1.80 · 104 Final 4.94 ± 0.95 · 104
1.2
All values (in CFU/g-ceramic beads) were the average ± standard deviation (STDEVP).
3.5. Removal of aqueous volatile fatty acids from wastewater and waste-foods From the above results, the operating conditions were optimized as follows: cell number, 5.6 ± 0.72 · 108 for wastewater, and 3.7 ± 0.13 · 108 cells/g-ceramic beads for waste-foods; air flow rate, 0.2 l/min; solution pH, 8.0; superficial velocity, 0.48 h1 ; temperature, 37 C. The acids from the wastewater were completely removed for 24 h by the immobilized Rhodococcus sp. B261 as shown in Fig. 7. Acetic acid (8525 ppm) and propionic acid (7310 ppm) with high concentration were effectively removed for 12 h. However, n-buryric (4360 ppm) and n-valeric acid (2572 ppm) were completely removed within 34 and 48 h, respectively. The pH 8.0 of the solution was increased to 9.67. After 24 h, the colour
Fig. 6. The effect of the initial immobilized cell number on the removal of n-valeric acid. Operating condition: inlet n-valeric acid concentration, 0.5%; air flow rate, 0.2 l/min; SV, 0.48 h1 ; column temperature, 37 C; initial column and n-valeric acid solution pH, 8.0; moisture content was adjusted to 35% by adding the medium. Error bars represent the standard deviations of duplicate determinations. Symbols (CFU/g-ceramic beads); () cell free; ( ) 1.40 ± 0.35 · 109 ; (j) 5.05 ± 0.48 · 108 ; ( ) 4.85 ± 0.33 · 106 ; (}) 7.50 ± 1.80 · 104 .
(SV), 0.48 h1 ; air flow rate, 0.2 l/min; solution pH, 8.0. The removal rates by 5.05 ± 0.48 · 108 , 1.85 ± 0.33 · 106 and 7.50 ± 1.80 · 104 CFU/g-ceramic beads, were only about 85%, 63% and 43%, respectively (Fig. 6). This indicates that the more cells immobilized, the more efficient n-valeric acid could be removed. The immobilized cell number after the reaction increased approximately by one log-unit as shown in Table 2. In addition, the higher removal rate could be explained with regard to the cell number which could proliferate better in the presence of the medium (data not shown) due to the growth of the cells in the immobilized form. The same result was observed in the degradation of 50% of 2,4-dichlorophenoxy-acetic acid solution by the 108 CPU/beads viable cell count (Kochar and Kahlon, 1995).
Fig. 7. Removal of various volatile fatty acids from wastewater. Error bars represent the standard deviations of duplicate determinations. Symbols: ( ) acetic acid; ( ) propionic acid; () n-butyric acid; (j) n-valeric acid; (}) cell free.
46
S.-I. Yun, Y. Ohta / Bioresource Technology 96 (2005) 41–46
of the beads inside column was changed to yellow, due to the growth of Rhodococcus sp. B261. It is suggest that Rhodococcus sp. B261 was dominated compared with the extended cells in the wastewater, therefore all acids were effectively utilized by them. Advancing the operation, the solution became limpid because of the precipitation of solid materials. The pH was increased. It was probably caused by NHþ 4 generated from the cell growth. After the operation, the final living cell numbers in the column were increased from 5.6 ± 0.72 · 108 to 7.3 ± 1.04 · 109 CFU/g-ceramic beads. Moreover, the any other cell was not detected in the medium containing only VFA as carbon source. This indicates the effective removal of the acids was caused by Rhodococcus sp. B261. In the case of the solution from the waste-food, the acids were difficult to remove by Rhodococcus sp. B261 during the initial period of 48 h (Fig. 8). This may be due to the solid compounds which prevent the reaction between acids and cells. However, the optical density of the solution with solid compounds was declined from 7.3 to 1.5 (data not shown). The removal of acetic acid (7810 ppm), propionic (8942 ppm), n-butyric (5730 ppm) and n-valeric (3625 ppm) acids from the solution was initiated at 48 h and then acids except n-valeric acid were dynamically removed in 64 h by the cells. The n-butyric (5730 ppm) and n-valeric (3625 ppm) acid were gradually decreased and removed within 72 h. The viable cell number of Rhodococcus sp. B261 were increased from 3.7 ± 0.13 · 108 to 2.9 ± 0.47 · 109 CFU/g-ceramic beads. The acids in NaOH solution (pH 8.0) from the bottle (g) was not detectable until 36 h (for wastewater) and
Fig. 8. Removal of various volatile fatty acids from the solution of spoiled waste-food. Error bars represent the standard deviations of duplicate determinations. Symbols: ( ) acetic acid; ( ) propionic acid; () n-butyric acid; (j) n-valeric acid; (}) cell free.
60 h (for spoiled waste-foods solution) by gas chromatography. It shows that acids were not volatilized from the bottom of the column by the system. Even though some acids were volatilized, it was also effectively removed by the immobilized cells. This is a report on the removal of n-valeric acid by a trickling biofilter system. The removal of wastewater and waste-foods was longer than that of using pure acids. It was caused by many kinds of unknown materials in the wastewater and waste-foods. Also, the high concentration and different kinds of acids cause of the longer removal time for n-valeric acid. However, the data presented here show that trickling biofilter system under the optimum conditions can be applied to the removal of high concentration of VFA from wastewater and waste-foods under the aerobic condition.
References Elefsiniotis, P., Oldham, W.K., 1994. Anaerobic acidogenesis of primary sludge: the role of solids retention time. Biotechnol. Bieng. 44, 7–13. Ferchichi, M., Ghrabi, A., Grasmick, A., 1994. Urban wastewater treatment by trickling filter and rotating biological reactor. Water Res. 28, 437–443. Gijzen, H.J., Kansiime, F., 1996. Comparison of start-up of an upflow anaerobic sludge blanket reactor and a polyurethane carrier reactor. Water Sci. Technol. 34, 5–6. Greiner, A.D., Timmons, M.B., 1998. Evaluation of the nitrification rates of microbead and trickling filters in an intensive. Aquacult. Eng. 18, 189–200. Hamano, T., Oka, Y., Takada, O., Asano, T., 1972. Test of malodor composition in the feces of domestic animals. Bull. Hyogoprefect. Exp. Stan. Anim. Husbandry 9, 140–145. Jimeno, A., Bermudez, J.J., Canovas-Doaz, M., Manjon, A., Iborra, J.L., 1990. Degradation of isobalerate and 2-methylburate in an anaerobic trickling filter. Biol. Wastes 34, 241–250. Kochar, G.S., Kahlon, R.S., 1995. Degradation of 2,4-dichlorophenoxy-acetic acid (2,4-D) by immobilized cells of Pseudomonas putida. J. Gen. Appl. Microbiol. 41, 367–370. Kono, K.Y., 1993. Malodor Preventive Low: Editorial Supervision for Special Pollution Section of Air Preservation Department of the Ministry of Environment in Japan. Gyousei. Inc. pp. 13–35. Kuai, L., Verstraete, W., 1999. Autotrophic degasification with elemental sulphur in small-scale wastewater treatment facilities. Environ. Technol. 20, 201–209. Ohta, Y., Ikeda, M., 1978. Deodorization of pig feces by Actinomycetes. Appl. Environ. Microbiol. 36, 487–491. Ohta, Y., Sato, H., 1985. An artificial medium for deodorant microorganisms. Agric. Biol. Chem. 49, 1195–1196. Sai Ram, M., Singh, L., Suryanarayana, M.V.S., Alam, S.I., 1994. Effect of sulfate and nitrate on anaerobic oxidation of volatile fatty acids in rabbit waste at 20 C. J. Gen. Appl. Microbiol. 41, 181– 189. Tracker, B.K., Ford, C.G., 1999. In situ bioremediation technique for sites underlain by silt and clay. J. Environ. Eng. 125, 1169–1172. Wu, W.-M., Jain, M.K., Zeikus, J.G., 1996. Formation of fatty aciddegrading, anaerobic granules by defined species. Appl. Environ. Microbiol. 62, 2037–2044. Yun, S.-I., Ohta, Y., 1997. Some physiological properties of microorganisms capable of deodorizing farm animal feces. Bioresour. Technol. 60, 21–26.