Ecology/environmental microbiology

Supplementation of ammonium bicarbonates for the treatment of fruit cordial

wastewater by anaerobic digestion process
Ghufran Redzwan
*
, Charles J. Banks
University of Southampton, School of Civil Engineering and the Environment, Higheld, Southampton, SO17 1BJ, UK
a r t i c l e i n f o
Article history:
Received 18 October 2007
Received in revised form
7 April 2009
Accepted 29 April 2009
Available online 18 May 2009
Keywords:
Anaerobic digestion
Ammonium bicarbonate
Toxicity test
Fruit cordial wastewater
Specic mathematical function
a b s t r a c t
Lack of nitrogenous substrate and buffering capacity have been identied as causing failure in previous
work on the treatment of fruit cordial wastewater using anaerobic continuous stirred tank reactors. In
this study, ammonium bicarbonate was proposed to be used as the substrate for nitrogenous and
buffering resources. In order to determine the toxicity effect of the ammonium salts on the anaerobic
system, a series of concentration from 0 to 40 mg L
1
was tested. Biogas production was used as the
indicator for NH
4

toxicity. The results showed no indication that methanogen was affected by the
additional ammonium salt within the dosing regime. Application of the specic mathematical function
(G G
m
k/t
) to describe the kinetic of biogas production, suggested that the optimal concentration of
ammonium bicarbonate that can be used is 10 mg L
1
. This study also shows that the dosage regime up to
40 mg L
1
can be used to supplement the lack of nitrogenous and buffering capacity for the anaerobic
digestion process of the fruit cordial wastewater using CSTR.
2009 Elsevier Ltd. All rights reserved.
1. Introduction
An experimental operation for the treatment of fruit cordial
wastewater using anaerobic continuous stirred tank reactor (CSTR)
has previously been carried out [1]. That study suggested that such
wastewater requires supplementary substrates to sustain its critical
operational parameters such as alkalinity, pH and biomass. Obser-
vation during the operation showed that those parameters had
gone below recommended levels. Subsequently, after increasing
the chemical oxygen demand (COD) mass loading to the level of
1.33 kg m
3
, the whole experimental operation had failed. The
failure was off-set by a sudden drop of pH and increasing concen-
tration of volatile fatty acid (VFA). It is well known that these two
factors are critical to the anaerobic digestion process, especially to
the sensitive methanogen group of bacteria. However, the amount
of COD loading at 1.33 kg m
3
was found to be too low to cause
failure in anaerobic CSTR [2,3]. Therefore, it was concluded that
inadequate buffering control and disruption of microbial pop-
ulation balance between non-methanogen and methanogen to
convert carbonaceous organic to CH
4
, were the main causes of
operational failure. Obstacles during operation should be rectied
since this process has been shown to be an efcient alternative to
treat such wastewater with similar characteristics and to produce
CH
4
as the bioenergy [2,4].
This study proposes the use of ammonium bicarbonate
(NH
4
HCO
3
), for its buffering requirement against acidity during the
treatment operation and also for the microbial population. NH
4

will
perform important roles in the anaerobic digester as the preferred
bacterial nutrient for nitrogen and buffering capacity in an anaer-
obic digester [5]. However, high NH
4
HCO
3
concentrations cause
free ammonia toxicity especially to the methanogen [5,6]. There-
fore the optimal dosage for NH
4
HCO
3
used as supplement in
anaerobic digestion process should be determined.
2. Materials and methods
2.1. Preparation of substrate
The fruit cordial wastewater in question is a purple soft drink
that contains the juice of grapes, raspberries and blackcurrants,
which has been analysed to be lacking in alkalinity and nitroge-
nous resources [1]. Preparation of fruit cordial wastewater was
carried out according to a previous study by diluting the stock
liquor [7]. The dilution of concentrated fruit cordial gave
a consistent concentration of wastewater up to 2000 mg L
1
of
Abbreviations: COD, chemical oxygen demand; CSTR, continuous stirred tank
reactor; G, biogas production; G
m
, maximum biogas production, k, specic math-
ematical coefcient; t, time; VFA, volatile fatty acid.
* Corresponding author. University of Malaya, Faculty of Science, 50603 Kuala
Lumpur, Malaysia. Tel.: 60 379676797; fax: 60 379674178.
E-mail address: ghufran@um.edu.my (G. Redzwan).
Contents lists available at ScienceDirect
Anaerobe
j ournal homepage: www. el sevi er. com/ l ocat e/ anaerobe
1075-9964/$ see front matter 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.anaerobe.2009.04.008
Anaerobe 16 (2010) 3437
COD, which is in the range of medium strength wastewater [8].
Fibrous matter arising from the depulping process was separated
at the initial stage of production either by centrifugal process or
water-pressing, and managed under solid waste management.
This study was carried out to represent treatment for the unmixed
wastewater without the presence of brous components.
2.2. Toxicity test
The toxicity test of NH
4
HCO
3
(BDH Ltd.) on the anaerobic
digestion of soft drink wastewater was done by submerging a series
of air tight containers (500 mL) in a water bath at a temperature of
35

C. These containers were connected to biogas measurement

apparatus. Prior to testing by batch operation, each container was
lled with 300 mL of stabilised anaerobic sludge as the seeding
sludge and 100 mL of wastewater with COD concentration of
2000 mg L
1
. This was to simulate the non-critical COD mass
loading at 0.5 kg m
3
, which also avoided the seed experiencing
substrate shock loading. Duplicates of ve containers with an
incremental series of NH
4
HCO
3
concentrations from 0 to 40 mg L
1
were prepared by adding concentrated stock of ammonium salt.
The supplementary regime from 0 to 40 mg L
1
was chosen as the
trial study on the optimal dosage. All containers were gently shaken
every 10 min for mixing purpose and also to facilitate the release of
biogas.
The optimal concentration for ammonium salt was determined
according to the cumulative volumetric biogas production. It could
be assumed that rapid biogas production would take place within
the period of 3 h of batch operation for the same substrate [7].
Nevertheless, comparison of the maximum biogas production
could indicate the toxicity of NH
4
HCO
3
especially to the metha-
nogen in the system [6]. Composition of CH
4
:CO
2
in the previous
study is at the ratio of 20:80. However the biogas production
volume in this study was expected to be too low for composition
analysis by gas analyser (GA94A, Geotechnical Instrument (UK),
Ltd.). This study used the same methods of CO
2
removal by passing
the biogas through a high concentration alkaline solution while
measuring it by liquid displacement [1].
3. Results and discussion
The previous operational failure which was instigated by the
accumulation of VFA could have occurred with other limitations
such as micronutrients (Fe, Ni, Cu, Co, Mg) and phosphorus [8].
Nevertheless, hypothetically the lack of micronutrients could be
dismissed by the information on mineral content given on the label
of this particular softdrink cordial. Meanwhile the lack of phos-
phorus was also not being addressed because the inoculum was
taken from freshly digested sewage sludge, in which the presence
of phosphorus as the micronutrient should be adequate. Mean-
while, the presence of ammoniacal nitrogen as the macronutrient
in a stabilised digested sludge is also considered to be at a residual
concentration after the process of denitrication is complete [8].
The other salt that could be used to supplement the lack of
nitrogenous substrate is sodium nitrate. However, as for sodium
nitrate, the release of NO
3
1
would increase the oxidationreduction
potential (ORP) of the digester. The potential ORP of the digester
should be allowed to increase above 300 mV. At lower ORP
methane-forming bacteria could not produce methane, for example
at ORP of 300 mV [5]. While for the buffering salts, caution should
be exercised in the choice of chemical used for alkalinity adjust-
ment. Precipitation of CaCO
3
creates unwanted solids, and the large
quantity of single cation, for example Ca, presents the potential for
alkali metal toxicity.
3.1. Biogas production
Fig. 1 shows the cumulative production of biogas during the
toxicity test of ammoniumsalt on the anaerobic digestion system. It
shows the typical rst order reaction rate on the conversion of
carbonaceous organic to biogas. It decelerates with the cessation of
raw materials. Total cumulative biogas production was observed to
increase when the ammonium salt was added. However, the
cumulative production of biogas was observed to decline when
more than20 mg L
1
of ammoniumsalt were addedintothe system.
The effect of C:N ratio was not investigated during this testing. The
Fig. 1. Cumulative biogas production during the toxicity test for serial concentration of NH
4
HCO
3
in anaerobic digestion system.
G. Redzwan, C.J. Banks / Anaerobe 16 (2010) 3437 35
ratio will not be consistent at all times for the batch operation due to
substrate consumption and the denitrication process [8].
3.2. Use of specic mathematical function
The usual statistical analyses, i.e. ANOVA and t-test, are not
suitable for a limited number of samples. The data description as
shown in Fig. 1 was not able to clearly indicate the effect of
ammonium salt on the production of biogas. Therefore, quantita-
tive analysis of the volumetric production of biogas was carried out
with the use of specic mathematical function as shown in Eq. (1).
This specic mathematic function was introduced and later adap-
ted for kinetic interpretation of methane gas production by
anaerobic digestion [7,9]. Similarly, the equation could be used to
determine the maximum volume of biogas production (G
m
)
without running the experiment to exhaustion. The linearised
equation is rearranged as Eq. (2). Table 1 shows the value of biogas
production (G) at time t, which was used to plot the data points for
the determination of coefcients as listed in Eq. (2). The plot of
linear data points for lnG against 1/t is shown in Fig. 2. The G
m
value
could be used to demonstrate the effectiveness of supplementing
ammonium salts to the system.
G G
k=t
m
(1)
lnG lnG
m
k=t (2)
where G cumulative biogas production, G
m
maximum biogas
production, k slope coefcient in linear equation, and t time.
All data points, which are plotted in Fig. 2, show that the G
m
could be determined by the specic mathematical function because
they have been plotted to be linear. Table 2 lists the estimated
maximum volumetric biogas production during the toxicity test. It
is determined by re-inversing the cut-off values at the y-axis in
Fig. 2 as written in Eq. (2). The estimation gives the highest biogas
production when 10 mg L
1
of ammonium salt was used in the
system.
Previous study shows that the range of methane (CH
4
) yield per
unit of COD removed for the fruit cordial wastewater was within
0.260.30 [1]. In this study, the amount of CH
4
yield was assumed
to remain on the basis that the same substrate was being digested.
The maximum CH
4
yield which could be expected from this study
could be equal to 60 mL, while the calculated maximum biogas
production based on the specic mathematical function (Eq. (2))
ranged between 52 and 59 mL, as listed in Table 2. The actual
amount of CH
4
in this study was not determined due to an inade-
quate volume of gas for analysis. Deductively, anaerobic digestion
including methanogenesis took place in this experiment without
difculty.
Table 1
Cumulative production of biogas during the toxicity test for serial concentration of
NH
4
HCO
3
in anaerobic digestion system.
Time (min) Average cumulative biogas production (mL)
Ammonium concentration (mg L
1
)
0 10 20 30 40
15 16 17 17 14 14
30 20 22 23 18 18
45 23 26 26 21 22
60 26 30 29 25 26
75 30 35 33 28 29
90 35 38 37 31 32
105 39 41 40 34 36
120 42 45 42 39 40
135 45 47 45 41 42
150 48 50 48 44 45
165 50 52 50 46 47
180 53 54 52 48 48
Table 2
Estimated maximum volumetric biogas production during the toxicity testing with
serial concentration of ammonium bicarbonate.
Ammonium concentration (mg L
1
)
0 10 20 30 40
lnG
m
4.04 4.08 4.01 3.95 3.98
G
m
(mL) 57 59 55 52 53
Fig. 2. Determination of coefcients for G
m
and k by linearising Eq. (2) for each of ammonium salt concentrations.
G. Redzwan, C.J. Banks / Anaerobe 16 (2010) 3437 36
Sustaining methanogenesis in the anaerobic digestion process is
the vital operational procedure for the removal of carbonaceous
materials in wastewater. Proliferation of a specic substrate in the
wastewater is equally important. It has been suggested that such
wastewater with high organic content should not be mixed with
other wastewater that is low in organic content to be treated with
the same process even if it is generated in the same vicinity [10].
Therefore, individual wastewater treatment processes for specic
strength and content of wastewater should be in place. Separating
the wastewater treatment processes in terms of their characteris-
tics would be more cost effective, if the treated wastewater is to be
reused [11]. In the case of wash waters from the production oper-
ation of fruit cordial processing, i.e. pre washing of the fruit,
cleaning of process vessels, pipelines and associated equipments
[4], application of anaerobic digestion as an individual system for
the treatment of its wastewater should be established, prior to the
grouping of wastewaters according to their strength and compo-
sition. The establishment would then support the use of tertiary
treatment for the prospect of treated wastewater being recycled
and reused.
4. Conclusion
Using biogas production as the indicator, this study has shown
that ammonium bicarbonate could be used as the supplement for
anaerobic digestion of fruit cordial wastewater. Application of the
specic mathematical function in this study suggests that 10 mg
L
1
of ammonium salt is the optimal concentration for the
substrate to be used when compared to the series of increments
up to 40 mg L
1
. This recommendation can only be applied to the
current study using CSTR design. Although there are other
designs of reactor that have better operational performance,
CSTR should be known for the simplicity of its operational
mechanism.
Acknowledgement
This study was supported in part by a grant from the University
of Malaya Kuala Lumpur for the training of academic staff (SLAB
8989).
Appendix A. Supplemental material
Supplementary data associated with this article can be found, in
the online version, at doi: 10.1016/j.anaerobe.2009.04.008
References
[1] Redzwan G, Banks CJ. An evaluation of soft drink wastewater treatment by
anaerobic digestion process. Malays J Sci 2007;26(1):3542.
[2] Borja R, Banks CJ. Kinetics of anaerobic digestion of soft drink wastewater in
immobilised cell bioreactors. J Chem Tech Biotechnol 1994;60:32734.
[3] Redzwan G, Banks CJ. Application of anaerobic process for the treatment of
organic fraction in municipal solid waste. Malays J Sci 2004;23:10313
(selected proceeding for LUCED-I&UA International Conference on Environ-
mental Management &Technology: A Clean Environment Towards Sustainable
Development, Renaissance Palm Garden, Putrajaya, Malaysia).
[4] Borja R, Banks CJ. Kinetics of anaerobic uidized bed system used for the
purication of fruit processing wastewater. Chem Eng J 1994;54:B2532.
[5] Gerardi MH. The microbiology of anaerobic digesters. Somerset, NJ, USA:
Wiley-Interscience; 2003. p. 103-108.
[6] Sawayama S, Tada C, Tsukahara K, Yagishita T. Effect of ammonium addition on
methanogenic community in a uidized bed anaerobic digestion. J Biosci
Bioeng 2004;97(1):6570.
[7] Redzwan G, Banks CJ. The use of a specic function to estimate maximum
methane production in a batch fed anaerobic reactor. J Chem Technol
Biotechnol 2004;79:11748.
[8] MetcalfandEddyInc. Wastewater engineering, treatment, disposal and reuse.
3rd ed. Singapore: McGraw-Hill, Inc.; 1991. p.359440.
[9] Daniel C, Wood FS. Fitting equations to data computer analysis and multifactor
data. 2nd ed. New York: John Wiley and Sons; 1979. p. 549.
[10] Chmiel H, Kaschek M, Blocher C, Noronha M, Mavrov V. Concepts for the
treatment of spent process water in the food and beverage industries.
Desalination 2002;152:30714.
[11] Miyakia H, Adachib S, Suds K, Yasunari K. Water recycling by oating media
ltration and nanoltration at a soft drink factory. Desalination
2000;13(1):4753.
G. Redzwan, C.J. Banks / Anaerobe 16 (2010) 3437 37