Fuel 96 (2012) 482–486

Contents lists available at SciVerse ScienceDirect

Fuel
journal homepage: [Link]/locate/fuel

Anaerobic treatment of spoiled milk from milk processing industry for energy
recovery – A laboratory to pilot scale study
Pandian Sivakumar a, Margandan Bhagiyalakshmi b, Kamalakannan Anbarasu c,⇑
a
Department of Chemical Engineering, Anna University, Chennai 600 025, Tamilnadu, India
b
Department of Chemistry, University College of Engineering Arni, Anna University, Chennai 600 025, Tamilnadu, India
c
Quality Control Department, Tamilnadu Co-operative Milk Producers Federation Limited, Aavin, Sholinganallur, Chennai 600 119, Tamilnadu, India

a r t i c l e i n f o a b s t r a c t

Article history: An anaerobic process for the treatment of spoiled milk from the milk processing industry was
Received 28 September 2011 investigated. It was carried out in a laboratory scale batch reactor and using the obtained data, it was
Received in revised form 19 January 2012 scaled up to a pilot Anaerobic Sequential Batch Reactor (ASBR). The processes include acclimatization
Accepted 20 January 2012
of seed consortia and start up of the reactor under room condition. The batch reactor was studied at dif-
Available online 5 February 2012
ferent pH, inoculum dosage, and Chemical Oxygen Demand (COD) loading. The scaled up pilot plant is of
100 L capacity and it was observed continuously for 30 days. At steady state, the efficiency of the pilot
Keywords:
plant in removing COD was found to be 92.80%, and it was able to produce biogas volume of 0.311 m3
Spoiled milk
Biogas
for 1 kg COD degradation. The results obtained showed that biogas contained a maximum of 69.18% of
COD methane.
Batch reactor Ó 2012 Elsevier Ltd. All rights reserved.
Anaerobic sequential batch reactor

1. Introduction subjected to pasteurization. In metro dairies, milk waste water

is treated normally by the aerobic method, which is a cost-inten-
Dairy industries release large quantities of wastewater, often sified process.
in the order of thousands of cubic meters per day [1]. The dairy In developed countries, there has been an increased interest in
wastewater is similar to most other agro-industries wastewater the development of technologies for harnessing renewable energy
characterized by higher biological oxygen demand and COD [2]. sources such as biomass either directly or through conversion
High concentrations of organic matter in dairy wastewater cause routes. The anaerobic digestion process has been applied for the
pollution problems to the surroundings [3]. Aerobic treatment treatment of wastes and bio solids for decades [9]. Recently,
(activated sludge process) is traditionally used for treatment of anaerobic digestion has received added attention since it produces
wastewater [4]. Aerobic methods for treatment and disposal of biogas and releases nutrients that can be recycled to be used in
dairy wastewater [5,6] have been less efficient, mainly due to agriculture as bio fertilizer [10,11]. The nutrients in the raw
bulking and excessive growth of biomass [7]. Anaerobic methods material remain in the mineralized digested sludge as accessible
for the treatment of dairy wastewater are attracting the compounds. Various reactor configurations like anaerobic
attention of researchers because of the presence of high organic contact reactor [12–14], upflow anaerobic sludge blanket reactor
content in the waste, low energy requirement of the process, [15–19], fluidized bed reactor, and anaerobic fixed film reactor
lesser sludge production, and generation of fuel in the form of [20,21] have been developed to treat wastewaters from different
methane [8]. industries.
In India, milk is collected in villages through a first-tier The main objectives of the study are to segregate spoiled milk at
collection system called society. From the society, milk is source and prevent them from mixing up with the effluent,
transferred to nearby chilling centers for the immediate chilling generate energy, produce clean effluent for recycle and irrigation,
to prevent microbial activity. From the chilling centers this milk reduce the odors associated with the waste products, concentrate
is transported to the processing dairies in metro cities through the nutrients in a solid sludge as biofertilizer, and reduce
insulated road vehicles. In the processing dairies, this milk is pathogens associated with the waste. Keeping the above
mentioned points, the treatment of spoiled milk using a lab scale
⇑ Corresponding author. Tel.: +91 9840574323. batch reactor followed by a pilot plant ASBR were investigated.
E-mail address: biofuelresearchers@[Link] (K. Anbarasu). The results obtained are presented and discussed.

0016-2361/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/[Link].2012.01.046
 P. Sivakumar et al. / Fuel 96 (2012) 482–486 483

2. Materials and methods

2.1. Materials

The spoiled waste milk for the present investigation was

collected from the milk processing industry near Chennai, India.
All the chemicals used during the experiments were of analytical
grade purchased from Merk (Mumbai, India). Double distilled
water from an all-glass apparatus was used whenever necessary.

2.2. Characteristics of the spoiled milk

The characteristics of the spoiled milk were determined as per

APHA standard methods [22]. The inlet and outlet COD were
determined by ISO 6060 method [23]. The biogas collected was
measured volumetrically in a scaled wet gas meter and its
composition was analyzed using a Varian Analytical Instruments
CP-3800 series gas chromatograph. The types of detectors
employed were the flame ionization detector and thermal
conductivity detector. Calcium, magnesium, sodium, copper,
manganese, nickel, and iron contents were analyzed by using
inductively coupled plasma–optical emission spectroscopy
(PerkinElmer Optima 7000 DV, USA).

2.3. Inoculum
Fig. 1a. The schematic representation of experimental laboratory scale biogas plant
The inoculum consists of an undefined mixed culture from an showing (A) batch reactor; (B) effluent outlet; (C) feeding neck; (D) CO2 scrubber
equivalent system that is actively digesting a similar feedstock (optional); (E) scaled wet gas meters; (F) gas sampling port; (G) displaced water.
[24]. Digested cow dung was used as an inoculum for seeding
the reactor. The cow dung was diluted with water (1:10) and
filtered through a 40 mesh screen to remove larger particles. The
filtered sludge was used as a seed culture for preparing inoculum
for the studies. The sludge was acclimatized with the spoiled milk
for 3 months under anaerobic condition to make sure that there is
no undigested matter present in the digester. This acclimatized
inoculum will help in the quick start up of the biogas digester.

2.4. Batch reactor

The anaerobic digestion of spoiled milk was performed in a

laboratory scale batch (Figs. 1a and 1b). The total designed volume
of the reactor was 1.75 L with a working volume of 1.25 L. The
reactor was made of glass having a provision for inlet and outlet.
The reactor was kept at room temperature at 32 ± 3 °C. The spoiled
milk (if it is in curdling form) is completely homogenized using a
blender before it is fed into the reactor so that there are no
settleable solids during the feeding. The feed was introduced into
the reactor via an inlet feeding port and the same quantity of outlet
was collected from the bottom of the reactor by means of gravity.
In order to resuspend the sediments and avoid scum layers, the
reactor was shaken once a day. The biogas produced was collected
in a scaled wet gas meter at room temperature. In addition, biogas
production by the inoculum without the feed was also recorded as
control. The pH was adjusted to the required level using 5%
aqueous lactic acid and sodium bicarbonate solutions.

2.5. The pilot plant

The pilot plant facility (Figs. 2a and 2b) comprised of a floating

drum ASBR having an operational volume equal to 100 L. The
material used for construction was fiber glass reinforced plastic.
The influent with a required pH was introduced continuously into
the reactor. The influent pH was regulated by aqueous solutions of
sodium bicarbonate, which were dosed through a 5 L conditioning
tank. The study was conducted at room temperature at 32 ± 3 °C. Fig. 1b. Pictorial view of experimental laboratory batch biogas plant.
484 P. Sivakumar et al. / Fuel 96 (2012) 482–486

Table 1
Characteristics of spoiled milk.

Parameter Unit Concentration

pH – 6.6 ± 0.5
1
Total solids mg L 150,715 ± 1705
1
Total volatile solids mg L 101,960 ± 1804
1
Suspended solids mg L 1730 ± 340
1
Volatile suspended solids mg L 1830 ± 255
1
Total COD mg L 203,020 ± 60,250
1
TKN mg L 12,072 ± 1011
1
Total phosphorus mg L 158 ± 18
1
Lipid mg L 30,662 ± 415
1
Chloride mg L 633 ± 93
1
Calcium mg L 1200 ± 120
1
Magnesium mg L 97 ± 12
1
Sodium mg L 1003 ± 200
1
Potassium mg L 1523 ± 160
1
Copper mg L 5±1
1
Manganese mg L 7±2
1
Nickel mg L 4±1
1
Iron mg L 17 ± 3
1
Fig. 2a. The schematic representation of experimental ASBR biogas pilot plant Zinc mg L 6±1
showing (A) feed inlet; (B) effluent outlet; (C) gas outlet; (D) floating drum; (E)
cleaning eye.

COD and the nutrient level of the spoiled milk were very high, indi-
cating the complexity in treating the milk waste.

3.2. Batch test

3.2.1. Effect of pH on cumulative biogas production

The effect of pH was monitored to obtain maximum biogas
production in a batch reactor. The initial COD loading was
150,000 ± 100 mg L 1 with 10% inoculums, and a Hydraulic
Retention Time (HRT) of 48 h was maintained for all pH studies.
The pH was varied from 5 to 8. Lactic acid and sodium bicarbonate
solutions were used to adjust the pH to a required level. As shown
in Fig. 3, a lower volume of biogas production was observed at
lower pH. It is expected that the drop in pH inhibits the growth
of hydrogen-utilizing methanogenic bacteria, causing a rise in the
hydrogen partial pressure, which causes further accumulation of
volatile fatty acids. Therefore, methane generation is reduced
considerably when pH drops [25,26]. Maximum bio gas production
was recorded when the pH ranged between 6.5 and 7.5, with an
optimum at 7 pH. It was also observed that there was a decrease
in biogas volume with increase in pH after 7.5 pH. This can be
attributed to the sensitivity of methanogens to high pH [26].

3.2.2. Effect of inoculum potential on cumulative biogas production

Instability during both the start up and operation of the
anaerobic degradation process can be problematic due to the low

Fig. 2b. Pictorial view of experimental ASBR biogas pilot plant.

60

The daily biogas production was measured. Excess effluent was

Biogas volume (L)

removed regularly to maintain a constant reactor volume. The 40

biogas was vented out after sampling and measuring the volume
at room temperature whenever required.

20
3. Results and discussion

3.1. Characteristics of the spoiled milk

0
4.5 5.5 6.5 7.5 8.5
The physicochemical characteristics of the spoiled milk were
pH
given in Table 1. A random analysis of the sample was done for a
period of 3 months with a regular interval. The values were Fig. 3. Effect of pH on cumulative biogas production for 48 h HRT, COD
documented to evaluate standard deviation. It was observed that 150,000 ± 100 mg L 1, and 10% of inoculum concentration.
 P. Sivakumar et al. / Fuel 96 (2012) 482–486 485

specific growth rate of the methanogenic microorganism involved. Table 2

[27]. Anaerobic microorganisms acclimatized to spoiled milk can Volume of gas produced per gram of COD destruction at various initial COD
concentration.
be good inoculum for a fast start up of the operation of the reactor.
The inoculum dosage was varied from 5% to 25% in volume, as Initial COD concentration Volume of gas produced per gram of COD
shown in Fig. 4. A little difference in the optimum inoculum (mg L 1) destruction (L g 1)

percentage was observed for cumulative biogas production. A 50,000 ± 100 0.344
10% concentration of inoculum was relatively more efficient at a 100,000 ± 100 0.341
150,000 ± 100 0.335
given concentration treatment, yielding a larger volume of biogas, 200,000 ± 100 0.332
than that of a 5% concentration. There was relatively no change in 250,000 ± 100 0.315
biogas volume noticed at other higher inoculum concentrations of
15% and 20%, but a higher rate of biogas production was observed.
Therefore, 10% of inoculum concentration was taken as optimum
4.5
value for corresponding experiments.
4 50000±100 mg/L
3.5 100000±100 mg/L

Rate (L Day -1)

3.2.3. Effect of COD loading on cumulative biogas production 3 150000±100 mg/L
The effects of initial COD concentration and digestion time on 2.5 200000±100 mg/L
cumulative biogas production are shown in Fig. 5. It was observed 2 250000±100 mg/L
that the volume of biogas produced was found to increase linearly 1.5
in the beginning, then nonlinearly at a slower rate, and finally 1
attained saturation, called complete digestion. The digestion time 0.5
was influenced remarkably by the variation in the initial COD 0
concentration. This is because of the food to microorganism ratio. 0 12 24 36 48 60 72 84
The initial microorganism population can only consume a certain Time (Hour)
amount of food in the form of COD. Therefore, the more the COD,
Fig. 6. Variations of instantaneous reaction rates with different COD concentrations.
the longer it takes to enhance the food to microorganism ratio. It
was observed that the biogas produced per gram of COD
destruction varied with varying the COD concentration and it Biogas production rates are given in Fig. 6. The maximum rates
decreased with an increase in COD concentration mass. The occurred at different reaction times that ranged from 18 to 24 h for
amount of biogas produced decreases from 0.344 to 0.315 L of 50,000–250,000 ± 100 mg L 1 COD, respectively. It was observed
biogas per gram of COD destruction for an increase in COD from that lower COD concentrations exhibit maximum biogas
50,000 to 250,000 mg L 1 as shown in Table 2. This instability is production rates at less time than higher concentrations. But a
usually witnessed by feed overload [28,29]. Whereas the higher COD concentration resulted in a higher peak rate.
cumulative biogas volume increases from 17.2 to 78.75 L with an
increase in initial COD.
3.3. Pilot plant studies

60 In pilot scale, anaerobic digestion of spoiled milk substrate was

carried out in an ASBR reactor. After determining the kinetic
Biogas Volume (L)

parameters, using experimental data at different optimized

50
operational conditions in batch system (e.g., inoculum concentra-
tion, influent pH, HRT), the reactor was fed according to the
40
optimum results. But a COD of 200,000 ± 100 mg L 1 was selected
based on the real time average COD of spoiled milk. A constant
volumetric loading of 5 L day 1 spoiled milk was maintained. The
30 reactor was fed daily for 30 days.
0 5 10 15 20 25 30 Fig. 7 shows the relationship between the daily biogas
Inoculum (vol %) production and the retention time, at a feeding rate of 1 kg COD
day 1. The average daily biogas production during 30 days of
Fig. 4. Effect of inoculum concentration on cumulative biogas production for
150,000 ± 100 mg L 1 COD at 7 pH in 48 h HRT.
retention time period was observed to be 0.3117 m3 of biogas
per day. It was observed that biogas production rate became stable

100 3.5
50000±100 mg/L
100000±100 mg/L 3
Volume of gas (m3)
Biogas volume (L)

80 200000±100 mg/L
250000±100 mg/L 2.5
300000±100 mg/L
60 2
Biogas
40 1.5 Methane
Carbondioxide
1
20
0.5
0 0
0 12 24 36 48 60 72 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Time (Hours) Hydraulic Retention Time (Days)

Fig. 5. Effect of COD loading on cumulative biogas production at 7 pH in 10% Fig. 7. Variation of methane and carbon dioxide content in produced biogas from
inoculum concentration. spoiled milk substrate at 7 pH for 200,000 ± 100 mg L 1 COD.
486 P. Sivakumar et al. / Fuel 96 (2012) 482–486

immediately after the second day of the digestion process. This [10] Odegaard H, Paulsrud B, Karlsson I. Waste water sludge as a resource, sludge
disposal strategies and corresponding treatment technologies aimed at
period is very short because milk is easily biodegradable when
sustainable handling of waste water sludge. Water Sci Technol
compared to other substrates like municipal garbage [30], oil seed 2002;46:295–303.
cake [31], poultry manure [32], kitchen waste, and night soil [33]. [11] Francese AP, Aboagye-Mathiesen G, Olesen T, Cordoba PR, Sineriz F. Feeding
Anaerobic digestion is a complex process of degrading organic approaches for bio gas production from animal wastes and industrial effluents.
World J Biotechnol 2000;16:147–50.
compounds through a variety of intermediates into methane and [12] Silva AJ, Varesche MB, Foresti E, Zaiat M. Sulphate removed from industrial
carbon dioxide by the action of a consortium of microorganisms waste water using a packed bed anaerobic reactor. Process Biochem
[34,35]. The maximum and minimum values of methane and 2002;37:927–35.
[13] Venkata Mohan S, Sarma PN. Pharmaceutical waste water and treatment
carbon dioxide were found to vary from 69.18% to 60.61% and technologies. Pharma Bioworld 2002;1:93–100.
27.18% to 26.0%, respectively, starting from first day to 30th day 14 Pfeffer JT. Anaerobic digestion processes. In: Stafford, Wheatley, Huges, editors.
of retention time. The average methane and carbon dioxide content Anaerobic digestion. London: Applied Science Publishers; 1979.
[15] Rebac S, Lier JB, Janssen MGT, Dekker F, Swinkers KM, Lettinga G. High-rate
over 30 days were found as 64.88% and 26.01%, respectively. anaerobic treatment of malting wastewater in a plant scale EGSB system under
Cumulative biogas, methane, and carbon dioxide production Psychorophilic conditions. J Chem Tech Biotechnol 1997;68:135–46.
over a 30-day retention period were 9.664, 6.27, and 2.514 m3 [16] Yu H, Gu G. Biomethanation from brewery waste water using an anaerobic
upflow blanket filter. J Cleaner Prod 1996;4:219–24.
respectively, from a total of 150 L of spoiled milk, therefore 1 L [17] Mukti Deb R, Nandy T, Kaul SN. Anaerobic upflow bioreactor for
milk can produce an average of 64.42 L biogas. The average outlet pharmaceutical waste treatment. Indian J Environ Prot 1993;13:41–6.
COD of the stabilized plant was found to be 14,390 ± 60 mg L 1. [18] Berrueta J, Lastrillon L. Anaerobic treatment of leachates in UASB reactors. J
Chem Tech Biotechnol 1992;54:33–7.
The COD removal efficiency of the pilot plant was 92.80%.
[19] Lettinga G, Grin PC, Roersma RE, Hobma SW. High rate anaerobic waste water
treatment using the UASB reactor under a wide range of temperature
4. Conclusion conditions. Biotechnol Genetic Eng Rev 1984;2:253.
[20] Mark CM, Vanioosdrecht Sef JH. Biofilm bioreactor for waste water treatment.
Trends Biotechnol 1993:117.
Spoiled milk was treated under anaerobic condition. [21] Henze M, Herremoes P. Anaerobic treatment of waste water in fixed film
Accordingly, an investigation was made on a laboratory scale to reactors- a literature review. Water Sci Technol 1983;15:1–101.
[22] APHA. Standard methods for examination of water and wastewater. 19th ed.
assess the feasibility and was then scaled up to a pilot plant. The Washington (DC, USA): American Public Health Association Inc.; 1995.
batch study suggested an optimum 7 pH with low COD [23] ISO 6060:1989 method, water quality – determination of the chemical oxygen
concentration and 10% inoculum. The rate of production of biogas demand.
[24] Griffin ME, McMahon KD, Mackie RI, Raskin L. Methanogenic population
was high during 18–24 h. Results showed that spoiled milk was
dynamics during start-up of anaerobic digesters treating municipal solid
easily degradable in comparison to that of other agro-based raw waste and biosolids. Biotechnol Bioeng 1998;57(3):342–55.
materials. The scaled-up pilot plant shows an average COD removal [25] Bjornsson I, Murto M, Jantsch TC, Mattiasson B. Evaluation of new methods for
the monitoring of alkalinity, dissolved hydrogen and the microbial community
efficiency of 92.8% with 0.3117 m3 of biogas per kg of COD
in anaerobic digestion. Water Res 2001;35:2833–40.
destruction. It was found that treatment of spoiled milk by ASBR [26] Jenkins SR, Morgan JM, Zhang X. Measuring the usable carbonate alkalinity of
was a most cost-effective approach with profitable biogas operating anaerobic digesters. Res J Water Pollut Control Fed 1991;63:28–34.
recovery. [27] Bjornsson I, Mattiasson B, Henrysson T. Effects of support material on the
pattern of volatile fatty acid accumulation at overload in anaerobic digestion
of semi-solid water. Appl Microbiol Biotechnol 1997;47:640–4.
References [28] Lyberatos G, Skiadas IV. Modeling of anaerobic digestion – a review. Global
Nest Int J 1999;2:63–76.
[1] Wheatley AD, Johnson KA, Winstanley CI. The reliability of anaerobic digestion [29] Powell GE, Archer DB. On line titration method for monitoring buffer capacity
for the treatment of food processing effluent. In: Proceedings of 5th and total volatile fatty acid levels in anaerobic digesters. Biotechnol Bio Eng
symposium on anaerobic digestion, Bolonga; 1991. p. 135-46. 1989;33:570–7.
[2] Orhon D, Gorgon E, Grimly G, Rattan N. Biological treatability of dairy waste [30] Rao MS, Singh SP, Singh AK, Sodha MS. Bioenergy conversion studies of the
waters. Water Res 1993;27:625–33. organic fraction of MSW: assessment of ultimate bioenergy production
[3] Perle M, Kimchie S, Shelef G. Some biochemical aspects of anaerobic potential of municipal garbage. Appl Energy 2000;66:75–87.
degradation of dairy waste water. Water Res 1995;29:1549–54. [31] Chandra R, Vijay VK, Subbarao PMV, Khura TK. Production of methane from
[4] Kevin DT, Jack LM. Pharmaceutical waste water treatment comparisons of anaerobic digestion of jatropha and pongamia oil cakes; 2011. doi:10.1016/
process scale-up from bench to full-scale a case study. J Environ Progress [Link].2010.10.049.
1992;11:104–12. [32] Gelegenis J, Georgakakis D, Angelidaki I, Christopoulou N, Goumenaki M.
[5] Bull MA, Sterritt RM, Lester JN. The influence of COD hydraulic, temperature, Optimization of biogas production from olive-oil mill wastewater by
and pH shocks in stability of an unheated fluidized bed reactor. J Chem Technol codigesting with diluted poultry-manure. Appl Energy 2007;84:646–63.
Biotechnol 1983;33:221–30. [33] Koji Fuchigami, Ken-ichiro Mizuno, Torataro Minegishi, Hiroshi Atsuura,
[6] Jones HR. Pollution control in meat, poultry and sea food processing. New Satoru Ishimoto. Methane fermentation of night soil sludge and kitchen
Jersey: Noyes Data Corporation; 1974. waste mixture NKK tech rev 2001;85:44–50.
[7] Timmerans P, Haute A, Van JL. Influence of the type of organism on the biomass [34] Noykova N, Muller TG, Gyllenberg M, Timmer J. Quantitative analysis of
hold up in the fluidized bed reactor. Appl Microbiol Biotechnol 1993;19:36–43. anaerobic waste water treatment process, identifiability and parameter
[8] Ghaly AE, Pyke JB. Amelioration of methane yield in cheese whey fermentation estimation. Biotechnol Bio Eng 2002;78:89–103.
by controlling pH of the methanogenic stage. Appl Chem Biotechnol [35] Gujer W, Zehnder AJB. Conversion processes in anaerobic digestion. Water Sci
1991;27:217–37. Technol 1983;15:127–67.
[9] Verstraete W, Beer D, Pena M, Lettinga G, Lens P. Anaerobic bio processing of
organic wastes. World J Microbiol Biotechnol 1996;12:221–38.