Академический Документы
Профессиональный Документы
Культура Документы
ABSTRACT
A lab-scale anaerobic hybrid (combining sludge blanket and filter) reactor was operated in
a continuous mode to study anaerobic biodegradation of distillery-spent wash. The study
demonstrated that at optimum hydraulic retention time (HRT), 5 days and organic loading
rate (OLR), 8.7 kg COD/m3.d, the COD removal efficiency of the reactor was 79%. The
anaerobic reduction of sulphate increases sulphide concentration, which inhibited the
metabolism of methanogens and reduced the performance of the reactors. The kinetics of
biomass growth i.e. yield coefficient (Y, 0.0532) and decay coefficient (Kd, 0.0041 d-1) was
obtained using Lawrence and McCarty model. However, this model failed in determining
the kinetics of substrate utilization. Bhatia et al. model having inbuilt provision of process
inhibition described the kinetics of substrate utilization, i.e. maximum rate of substrate
utilization (R, 1.945 d-1) and inhibition coefficient values (Ki, 0.032 L/mg). Modeling of the
reactor demonstrated that Parkin and Speece, and Bhatia et al. models, both, could be used
to predict the effluent substrate concentration. However, Parkin and Speece model predicts
effluent COD more precisely (within ±2%) than Bhatia et al. model (within ±5-20%) of the
experimental value. Karhadkar et al. model predicted biogas yield within ± 5% of the
experimental value.
(Key words – Anaerobic digestion, spent wash, anaerobic hybrid reactor, sulfide toxicity,
kinetics of biodegradation, mathematical modelling)
1
1 Introduction
Distillery spent wash is perceived as one of the serious pollution problems of the countries
producing alcohol from the fermentation and subsequent distillation of sugar cane
molasses. The distillery spent wash is characterized as one of the caramelized and
recalcitrant wastes containing extremely high COD, BOD, SS, inorganic solids, color and
low in pH (Biradar, 2005, Shin et al., 1992; Saha et al., 2005). Being rich in sugarcane
yield, all the ethanol, in India, is produced by the way of fermentation of molasses and its
subsequent distillation (Inamdar, 1991; Saha et al., 2005). At present, there are 285
distilleries in India that producing 2.7 billion liters of alcohol and generating 40 billion
Guiot and van den Berg, 1985, Ramjeawon et al., 1995, Fang
2
The process kinetics plays an important role in
3
2.1 Experimental set-up & design
The experimental set-up of the anaerobic hybrid reactor was designed as per the guidelines
given by Lettinga and Hulshoff (1991) and is shown in Fig. 1. The reactor was made of 6
mm thick transparent acrylic sheet with inner dimensions of 100 mm x 100 mm, height
1500 mm. The reactor was provided with hopper bottom of 150 mm length with top and
bottom widths of 25 mm and 100 mm, respectively. The feed inlet pipe of diameter 2.5 mm
was provided at the hopper bottom and the feeding system was designed to ensure the
uniform distribution of feed to the entire of the sludge bed area. The outlet pipe of diameter
2.5 mm was provided at the top of the reactor with the provision of sampling ports to
facilitate sampling. Gas Liquid Solid Separator (GLSS) device is made of square pyramid
provided at the top with inclined walls at 500 and provision of gas outlet pipe.
The actual spent wash collected from SSK Distilleries Ltd., Niphad, Nasik and
neutralized with the lime to a pH of 7.0±0.5. The neutralized wastewater was fed to the
reactor at various HRT and organic loading to study the effect of HRT on the performance
of the reactor as shown in the experimental design of the study in (Fig. 2).
Alkalinity, pH, COD, BOD5, TSS, TDS, SO4, Cl-, sulphide and VSS were analyzed in
conformity to Standard Methods for the Examination of Water and Wastewater (APHA,
collection period of 3 hrs, twice in a day and average value for the particular day has been
estimated. For measuring the biogas, the vent pipes of the reactors are connected with 5
liter aspirator bottle filled with water. As the biogas enters the bottle filled with water, it
displaces same volume of biogas through the stop cock which was collected and measured
4
to find out total amount of biogas produced during the given period. Methane content of the
biogas was measured by injecting 1 ml of biogas through gas chromatograph equipped with
Thermal Conductivity Detector (TCD) using Porapack - Q (80-100), SS column 8' X 1/8"
(Karim, 2001; Karhadakar, 1988). The analysis with Porapack column was carried out at
oven temperature 1100C and detector temperature 2000C. The carrier gas, hydrogen, was
Volatile fatty acid in the effluent was measured by injecting 2 µl of filtered and
acidified samples through gas chromatograph equipped with Flame Ionization Detector
(FID) using 10 % FFAP on 60/80 Chromosorb WHP/0.1 % H3PO4 SS column. The analysis
was carried out at an oven temperature of 1500C, injector temperature of 1800C and
detector temperature of 2500C. Hydrogen and zero air were used to fuel the flame while
The reactor was inoculated with 232.5 g of anaerobically digested sludge (on VSS basis)
collected from sewage treatment plant. Table 1 shows the characteristics of the
anaerobically digested sludge used in the study. The quantity of seed sludge was estimated
as per the guidelines mentioned by Hickey et al. (1991b). The reactor was started using 1%
spent wash (v/v) as substrate and later the dilution was gradually decreased to acclimatize
the biomass for higher concentration of spent wash. The acclimation periods in case of
hybrid and UASB reactors were 185 days and 200 days, respectively.
To study the effect of HRT, the reactor was operated in continuous mode at different
HRT’s ranging from 4 - 8 days. HRT of the reactors was gradually increased i.e. 4 days, 5
days, 6 days, 7 days and 8 days in order to find out the optimum hydraulic retention time at
5
which the COD removal and CH4 yield is maximum. The influent pH and alkalinity was
kept 7.00 ± 0.2 and 3950 ± 50 mg/l as CaCO3, respectively. The reactor was operated at
pseudo steady state condition for 20-25 days at every HRT and average value with standard
deviation for each parameter was worked out. The pseudo study state in the present study
was considered when the variation in effluent COD falls within ±5% of the average value.
According to Jenkins et al. (1991), the true bicarbonate alkalinity of the reactor is given as
under:
(1)
TBA = ( ALK 4.3 − 0.833VFA)
Where,
The linearised forms of the Lawrence and McCarty`s Model (1970) as given by Eqs. 2 and
3 were employed for the determining the kinetics of bacterial growth and substrate
utilization, respectively.
(2)
1 = γ m [(S0 − S ) θX ] − K d
θc
( Xθ S0 − S ) = [{(K s k ) *1 S } + 1 k ]
(3)
The linearised form of Bhatia et al Model (1985b) considering the process inhibition due to
Sθ ( S 0 − S ) = 1 R + ( K i R ) * I s (4)
Where,
6
θc = Sludge retention time, days
θ = HRT, days
The models Bhatia et al. model (1985b) and Parkin and Speece (Non-competitive) models
(1982) as employed for the prediction of effluent substrate concentration are given by Eqs.
4 and 5, respectively.
S = S0 θ {(1 θ ) + R (1 + K i * I s )}
(5)
S = S0 − kXθ (1 + Tx K i ) (6)
Where,
θ = HRT, days
7
Ki = Inhibition coefficient, L/mg
Karhadkar et al model (1990) as given by Eq. 6 was employed for prediction of biogas
yield from the reactor. Accordingly, the biogas yield can be expressed by Eq. 7 as follows:
The characteristics of the spent wash are given in Table 2. It can be seen from the table that
the spent wash is highly acidic in nature with a pH value ranging from 3.3 to 3.9. Apart
from the high COD (90,000-1,30,000 mg/L) and BOD (45,000-60,000 mg/L), spent wash
also contains high dissolve (70,000-78,000 mg/L) and suspended solids (9,000-10,000
mg/L). The SO4- and Cl- concentration in spent wash was found in the range of 6000-6500
8
mg/L and 5500–6000 mg/L, respectively. Analysis of the data demonstrated that the
characteristics of the spent wash are in agreement with the reported values by other
researchers (Preeti et al., 2006 and Karhadkar et al., 1990). A little variation in the
sources, storage conditions of molasses, and efficiency of the process adopted for the
The process performance data of HRT study i.e. COD removal; biogas generation and
sulphate reduction are shown in Figures 3, 4 and 5 respectively. It can be seen from Figures
3 and 4 that percent COD removal and methane yield per kg of COD removed, increased
with increase in HRT from 4 to 5 days and later decreased with increase in HRT beyond 5
days. The maximum COD removal, (79± 0.05%) and methane yield (0.344± 0.005%) was
found to be at HRT, 5 days and OLR, 8.70 kg COD/m3.day. Decrease in the performance of
reactor at higher HRT can be attributed to the reduction of sulphate into sulphide, which
might have inhibited the metabolism of methanogenic bacteria and reduced the COD
removal and methane yield. This can be seen from the sulphate reduction profile of the
reactor (Fig. 5). The sulphate reduction continuously increased with increase in HRT,
which subsequently increased the sulphide concentration inside the reactor and decreased
the COD reduction and methane yield at higher HRT. Many researchers (Bal, et al., 2001,
Parkin et al., 1983 and Lawrence et al., 1964) have reported sulphide toxicity to the
Methanogens particularly Acetate Utilizing and H2 Utilizing, who become inactive and
reduce the conversion of intermediates into final end products leading to accumulation of
VFA followed by decrease in methane yield. Bal et al. (2001) have reported that sulphide
concentration more than 100 mg/L is toxic for the acetate and H2 utilizing methanogens.
9
Parkin et al. (1983) have reported that sulphide toxicity to the bacterial population could be
A relationship between volatile fatty acids, true bicarbonate alkalinity and total
alkalinity was worked out and is shown in Fig. 6. True bicarbonate alkalinity of the effluent
was estimated using the Eq. 1 (Jenkins et al., 1991) and compared with the experimental
value observed during the HRT study. It can be seen from the Figure 6 that the total
alkalinity of the effluent increased with increase in HRT. However, the bicarbonate
alkalinity of the effluent remained almost constant (3910 ± 74.16 mg/l) through out the
HRT study. The increase in total alkalinity with increase in HRT can be attributed to
increase in VFA concentration which contributed to VFA alkalinity in the effluent. Further,
it can also be seen that the observed bicarbonate alkalinity are almost same with the TBA as
estimated by the Jenkins et al. equation with standard deviation of ±18.08 mg/l (Fig. 6).
This indicates that Jenkins et al. formula can be used to find out true bicarbonate alkalinity
The VFA/alkalinity ratio varies from 0.02 to 0.14 at all HRT’s. Moreover, the
effluent alkalinity is higher than the influent alkalinity. This indicates that enough buffering
capacity was present in the reactor due to which decrease in pH was not observed even after
increase in the concentration of VFA in the reactor. The pH of the effluent at all HRT was
found in the range of 7.8 to 8. This indicates that the efficiency of the reactors decreased
due to sulphide inhibition rather than VFA inhibition. The increase in VFA concentration in
the reactor merely represents the incomplete conversion of VFA into final end product
10
A relationship between specific gas production and organic loading rate was
established (fig. 7), which yielded a good correlation (R2 = 0.994) between organic loading
rate and specific gas production. This indicates biogas production is strongly correlated
with the organic loading rate and is in accordance with the fundamentals of anaerobic
treatment.
In order to determine the kinetics of biological growth, the parameter 1/θc versus (So-S)/Xθ
of Eq. 2 were calculated from the process performance data of the reactor and plotted as
shown in Figure 8. The plot of 1/θc versus (So-S)/Xθ yielded a straight line (R2 = 0.985)
whose slope gives biomass yield coefficient (Y = 0.0532) and intercept gives bacterial
decay coefficient (Kd = 0.0041d-1). These values compare well with the reported values of
Y and Kd by many researchers (Karhadkar et al., 1990, Prakash, 1998 and Karim, 2001)
The relationship among the parameters 1/S versus Xθ/(So-S) were plotted (Fig. 9) based on
model. For a meaningful model, the values of k and Ks should be positive, which do not
seem to be possible in the present case. The reason for non-compliance of this model for
• The model does not take into consideration the process inhibition. However, in
present case, COD reduction and methane formation is strongly related to sulphide
11
• Secondly, the model assumes growth-associated methanogenesis, which does not
was observed in the present case despite substrate conversion to methane. The fact
is well supported from the mass balance of COD reduction and gas production
(Table 3) which shows that out of total substrate consumed, most of the substrate
present system. Similar observations were made by Karhadkar et al. (1990) and
The relationship for the substrate utilization as given by Eq. 4, is not bound with the
limitations of Lawrence and McCarty`s model as discussed earlier (Bhatia et al., 1985b). In
supports the observations made in this study i.e. high rate of substrate conversion to
methane despite of low biomass yield. Because of the above-mentioned reasons, Bhatia et
al. (1985b) model offering the following advantages over Lawrence and McCarty`s model
• It explains the high rate of substrate conversion to methane despite of low biomass
• Substrate utilization rate has inbuilt provision for incorporating process inhibition
due to toxicity.
12
3.6 Determination of kinetic coefficients of Bhatia et al. model
A relationship was established among the parameters Sθ/(So-S) versus (Is) based on the
process performance data of reactor at various HRT’s (Fig. 10). The plot of these two
variables established a good correlation (R2 = 0.9743) and gives maximum rate of substrate
utilization (R) and inhibition coefficient (Ki) to be 1.945 mg/mg/d and 0.032 l/mg,
respectively. The kinetics of substrate utilization as obtained in the present case is in good
agreement with the values reported by other researchers (Karhadkar, 1988, and Prakash,
1998).
Using Bhatia et al. model (Eq. 5) and Parkin and Speece model (Eq. 6) the effluent
substrate concentration (S) was predicted at various HRTs and compared with the
experimental values as obtained during HRT study. Figures 11 and 12 show the plots of
observed and predicted effluent substrate concentrations using Bhatia et al model, and
Parkin and Speece model, respectively. The comparison of observed and predicted values
reveals a good agreement between the two. Bhatia et al. model predicts effluent substrate
within ±5-20% of the experimental value. However, the model suggested by Parkin and
values).
13
As per the Karhadakar et al. model (Eq. 7)), the biogas production at various HRT’s were
predicted and compared with the experimental values (Fig. 13). The value of conversion
factor (G, 0.52 m3/kg COD) was determined experimentally. The comparison of observed
and predicted values reveals a good agreement between the two. The model predicted
4. Conclusions
distillery-spent wash. The optimum COD removal efficiency of the hybrid reactor
found to be 79% corresponding to optimum HRT and organic loading rate of 5 days
2. Kinetics of bacterial growth (Y, 0.0532 and Kd, 0.0041 d-1) can be determined using
3. Bhatia et al. model having the inbuilt provision of process inhibition can be used for
determining the kinetics of substrate utilization (R, 1.945 d-1 and Ki 0.032 L/mg).
4. Parkin and Speece model can be more precisely used for modeling the performance
of anaerobic hybrid reactor for the anaerobic treatment of spent wash treatment than
Bhatia et al. model. The model predicts effluent substrate within ± 2 % of the
experimental value.
14
5. Kerhadakar et al. model can be used to predict the biogas yield of the anaerobic
hybrid reactor for the treatment of spent wash. The model predicts gas yield within
15
REFERENCES
_________________________________________________________________________
APHA, AWWA and WPCF (1989). Standard Methods for the Examination of Water and
Wastewater. American Public Health Association, American Water Works
Association and Water Pollution Control Federation, Washington. D.C.
Bal, A.S. and Dhagat, N.N., 2001. Upflow anaerobic sludge blanket reactor-A review.
Indian J. Environ Health. 43 (2), 1-83.
Bardiya, M. C., Hashia, R., and Chandna, S., 1995. Performance of hybrid reactor for
anaerobic digestion of distillery effluent. J. Indian Association of Environmental
Management. 22, 237 - 239.
Bardiya M.P., 1988. Diphasic anaerobic process for distillery spentwash, A Report
submitted to Department of Non conventional Energy sources, Govt. of India.
Bhatia, D., Vieth, W.R. and Vekatasubramanian, K., 1985b. Steady state and transit
behaviour in microbial methanification: II. Mathematical modeling and verification.
Biotech. Bioengrg. 27 (8), 1192-1198.
Biradar, A., 2005. Post-Anaerobic Treatment of Distillery Spent wash. Ph.D. Thesis Indian
Institute of Technology, Bombay (India).
Goyal, S. K., Seth, R., and Handa, B. K., 1996. Diphasic fixed film biomethanation of
distillery spent wash. Biores. Technol., 56, 239 – 244.
Guiot, S. R., and van den Berg, L., 1985. Performance of an anaerobic reactor combining a
sludge blanket and a filter treating sugar waste. Biotechnol. Bioeng. (xxvii), 800-
806.
Gupta M.K., 1989. Thermophilic process for treatment of distillery effluent – A case study.
J. Process and Plant Engg. VII (3), 101-103.
Hayase, F.; Kim, S.M. and Kato, H., (1984). Decolorization and degradation products of
the melanoidins by Hydrogen peroxide. Agric. Biol. Chem. 48 (11), 2711–2717.
16
Hickey R.F. Wu, W.M., Veiga, M.C. and Jones, R., 1991b. Start-up, operation, monitoring
and control of high rate anaerobic treatment system. Wat. Sci. Tech., 24 (8), 207-
255.
Inamdar, S.N., 1991. The distillery industry growth prospects for the 90s. Chem. Engg.
World. XXVI (9), 43-46.
Jenkins SR, Morgan JM, Zhang X, 1991. Measuring the useable carbonate alkalinity of
operating anaerobic digesters. J Water Pollut Control Fed. 55: 448-453.
Jimenez, A.M, Borja, R. and Martin, A., 2003. Aerobic–anaerobic biodegradation of beet
molasses alcoholic fermentation wastewater. Proc. Biochem. 38, 1275–1284.
Karahadkar, P.P., 1988. Methane recovery from distillery spent wash. Ph.D. thesis
submitted Indian Institute of Technology, Bombay, India.
Karhadkar, P.P., Handa, B.K. and Khanna P., 1990. Pilot-Scale Distillery Spent wash
Biomethanation. J. Envir. Engrg. 116 (6), 1029-1045.
Kaul S.N. and Badrinath S.D., 1987. Energy recovery from Distillery Spentwash Using
Whole Cell Immobilized Reactor System, A Project Report, NEERI, Nagpur
(India).
Kimata, T., Kainoi, T., Tada, M., Tomkar, K., Shirabe, K., and Shirrizu, K., 1993.
Anaerobic treatment of thermal sludge conditioning liquor with granular sludge.
Water Environ. Res., 65, 6 – 14.
Lawrence, A.W. and McCarty, P.L., 1970. Unified basis for biological treatment design and
operation. ASCE, J. Sanitary Engg. Div. 96, SA3, 757.
Lawrence, A.W., McCarty, P. L. and Guerin, F., 1964. The effects of sulfide on anaerobic
treatment. Proc. 19th Purdue Univ. Ind. Waste Conf. 343.
Lettinga G. and Hulshoff Pol., 1991. UASB-process design for various types of wastewater.
Wat. Sci. Tech., 24 (8), 87-107.
Nandy, T., S. Shastry and S.N. Kaul., 2002. Wastewater management in cane molasses
distillery involving bioresource recovery. J. Environ. Manage. 65, 25–38.
17
Ozatijrk, I., Eroglu, V., Ubay, G., and Demir, I., 1993. Hybrid upflow anaerobic sludge
blanket reactor (HUASBR) treatment of dairy effluents. Wat. Sci. Tech., 28 (2), 77 -
85.
Parkin. G.F., Speece, R.E., Yang, C.H.J. and Kocher, W.M., 1983. Response of methane
formation systems to Industrial Toxicants. J Water Pollut Control Fed. 55, 44-48.
Pena, M.; Coca, M.; Gonzalez, G.; Rioja, R. and Garcia, M.T., 2003. Chemical oxidation of
wastewater from molasses fermentation with ozone. Chemosphere 51, 893–900.
Raghukumar, C.; Mohandass, C.; Kamat, S.; Shailaja, M.S., 2004. Simultaneous
detoxification and decolorization of molasses spent wash by the immobilized white-
rot fungus isolated from a marine habitat. Enzyme Microb. Technol. 35, 197-202.
Ramjeawon, T., Boguant, J. and Horan, N. J., 1995. UASB system for treating sugar cane
wastewater in Mauritius–A pilot scale study. J. Indian Association Environmental
Management. 22, 42 – 49.
Saha, N.K.; Balakrishnan, M. and Batra, V.S., 2005. Improving industrial water use: case
study for an Indian distillery. Resour. Conserv. Recycl. 43, 163–174.
Sheehan, G.J. and Green field, P.F., 1980. Utilisation, Treatment and Disposal of Distillery
Wastewater. Water Res. 14, 257-265.
Shin, H.-S.; Bae, B.-U.; Lee, J.-J.; Paik; B.-C., 1992. Anaerobic digestion of distillery
wastewater in a two-phase UASB reactor system. Wat. Sci. Tech. 25 (7), 361-371.
18
Nomenclature
19
Table Legend
Figure Legend
20