You are on page 1of 7

Waste Management xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Waste Management
journal homepage: www.elsevier.com/locate/wasman

Continuous dry fermentation of swine manure for biogas production


Chuang Chen a, Dan Zheng a, Gang–Jin Liu a,d, Liang–Wei Deng a,b,c,⇑, Yan Long a, Zhan–Hui Fan a
a
Biogas Institute of Ministry of Agriculture, Chengdu 610041, PR China
b
Laboratory of Development and Application of Rural Renewable Energy, Ministry of Agriculture, Chengdu 610041, PR China
c
Southwest Collaborative Innovation Center of Swine for Quality & Safety, Chengdu 611130, PR China
d
Bioprocess Control AB, Scheelevägen 22, 223 63 Lund, Sweden

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

Article history: A down plug-flow anaerobic reactor (DPAR) was designed for the feasibility study on continuous dry fer-
Received 7 May 2014 mentation of swine manure without any additional stirring. Using fresh swine manure as the feedstock
Accepted 23 December 2014 with TS concentration (w/w) of 20%, 25%, 30%, and 35%, stable volumetric biogas production rates of 2.40,
Available online xxxx
1.92, 0.911, and 0.644 L(L d)1 and biogas yields of 0.665, 0.532, 0.252, and 0.178 L g1VS were obtained
respectively, and the TS degradation rates were 46.5%, 45.4%, 53.2%, and 55.6%, respectively. With the
Keywords: increase of feedstock TS concentration, the concentration of ammonia nitrogen grew up to the maximum
Ammonia inhibition
value of 3500 mg L1. Biogas production was obviously inhibited when the concentration of ammonia
Biogas
Down plug-flow anaerobic reactor (DPAR)
nitrogen was above 3000 mg L1. The maximal volumetric biogas production rate of 2.34 L(L d)1 and
Dry fermentation biogas yield of 0.649 L g1VS were obtained with TS concentration of 25% at 25 °C without inhibition.
Swine manure Liquidity experiments showed that TS concentration of digestate could be less than 15.8%, and the flow
rate of digestate more than 0.98 m s1 when the feedstock TS concentration was less than 35%, which
indicated the digestate could be easily discharged from a DPAR. Therefore, it is feasible to conduct a con-
tinuous dry fermentation in a DPAR using fresh swine manure as the feedstock with TS concentration less
than 35%, whereas the feedstock TS concentration should not exceed 30% to achieve the maximal biogas
production rate and biogas yield.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction et al., 2011). However, since swine manure is high in nitrogen con-
tent, substantive ammonia nitrogen may accumulate and then
In recent years, intensive animal breeding farms produce a large inhibit the biogas fermentation process (El-Mashad et al., 2004;
quantity of manure in a restricted area, which posed serious pres- Hansen et al., 1999; Strik et al., 2006). Wet fermentation of swine
sure on environment. Therefore, finding effective methods of treat- manure became the focus of previous researches because of less
ment and application is quite urgent. At present, biogas ammonia inhibition, whereas dry fermentation gained less atten-
fermentation is a widely employed process for the treatment of tion. Nevertheless, biogas production from separated pig slurry sol-
swine manure (De Bere, 2000; Fierro et al., 2014; Lu et al., 2007). ids is more profitable than using untreated raw slurry because
Biogas fermentation processes are classified into wet fermenta- there is a higher CH4 potential per unit fresh matter (Hothan
tions (TS concentration less than 10%) (Deng et al., 2014), semi-dry et al., 2013). In the meantime, considered the specificity of the
fermentations (TS concentration ranging from 10% to 20%) dry fermentation feedstock, batch fermentation and two-phase fer-
(Bolzonella et al., 2003; Dong et al., 2010), and dry fermentations mentation processes were mostly adopted for the study on dry bio-
(TS concentration more than 20%) (Abouelenien et al., 2009; gas fermentation (Abouelenien et al., 2009; Borja et al., 2003;
Kusch et al., 2008). Dry fermentation has attracted increasingly Kusch et al., 2008; Massé et al., 2003; Yabu et al., 2011). A batch
extensive attentions in the studies of biogas fermentation with fermentation process is not suitable for industrial-scale production
advantages of water-saving, convenient operation, high yields, due to its complex operation (Ahn et al., 2010; Guendouz et al.,
high fermentation slurry concentration, and better energy- 2010; Lantz, 2012). The two-phase fermentation process can regu-
recovery (Fdez-Güelfo et al., 2010; Kafle and Kim, 2013; Yabu late the content of organic acid during methane-generating phase,
thus enhancing methane production rate and removal rate of vol-
atile fatty acids (VFA). However, in two-phase fermentation pro-
⇑ Corresponding author at: Biogas Institute of Ministry of Agriculture, Section
cess, the separation of acid-producing and methane-generating
4-13, Renmin Road South, Chengdu 610041, PR China. Tel./fax: +86 28 85236376.
E-mail address: dengliangwei@caas.cn (L.-W. Deng). phases is not easy to achieve, and it also requires high cost for

http://dx.doi.org/10.1016/j.wasman.2014.12.024
0956-053X/Ó 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Chen, C., et al. Continuous dry fermentation of swine manure for biogas production. Waste Management (2015), http://
dx.doi.org/10.1016/j.wasman.2014.12.024
2 C. Chen et al. / Waste Management xxx (2015) xxx–xxx

installation, resulting in its limited application. Moreover, stirring


is responsible for up to 54% of the power consumption of current
biogas plants (Kowalczyk et al., 2013). Consequently, the develop-
ment of a continuous dry fermentation process used in a swine
manure treatment plant is quite essential.
The selected type for digesters depends on operational factors,
including the nature of the waste to be treated, e.g. its solid
content. Typically, plug-flow digesters are suitable for ruminant
animal excreta with high solid concentrations (Hilkiah Igoni
et al., 2008). Most applied technologies for dry processes are
Dranco, Valorga, Linde and Kompogas, all working in the range
30–40% of total solids in the reactor feeding (Lissens et al., 2001).
In this study, a down plug-flow anaerobic reactor (DPAR) treat-
ing swine manure without any additional stirring was constructed
with the feedstock TS concentration of 20–35% to investigate
biogas yield, ammonia inhibition, digestate liquidity, and the
feasibility of continuous dry fermentation of swine manure for
Fig. 1. Diagrammatic sketch of the down plug-flow anaerobic reactor (DPAR).
biogas production.

2. Material and methods

each test, swine manure with a different TS concentration (20%,


2.1. Swine manure and inoculation sludge
25%, 30%, or 35%) was used as the feedstock, and the amount of
fresh manure was 100, 80.0, 66.7, and 57.2 g, respectively, accord-
Fresh swine manure was used as the feedstock in this study,
ingly the hydraulic retention time (HRT) was 56.2, 75.0, 96.3, and
taken from a pig farm in central Sichuan Province, China. The inoc-
121 d. In all instances, the amount of feeding equaled to that of dis-
ulation sludge was collected from an anaerobic digester for the
charging every day. The durations of the experiments were deter-
treatment of swine waste in our laboratory. Table 1 shows the
mined by the feeding rate and stability of biogas production, which
characteristics of the swine manure (SM) and inoculation sludge
were 67 d, 46 d, 37 d, and 42 d with feedstock TS concentrations of
(IS).
20%, 25%, 30% and 35%, respectively. The whole experiment was
conducted for 190 d, and feeding and discharging were carried
2.2. Experimental set-up out once a day. Moreover, biogas production and the pH value were
measured once a day, the concentrations of TS and ammonia nitro-
An anaerobic reactor was designed (defined as a down plug- gen in the discharged sludge were determined every two days, and
flow anaerobic reactor, DPAR) with a feeding inlet (internal diam- the composition of the collected biogas was analyzed once the
eter of 20 mm) on the top and a digestate outlet (internal diameter biogas container was full (13 L).
of 20 mm) at the bottom (Fig. 1). The reactor was a double-layer
cylinder made of plexiglass with a height of 375 mm and an inter-
nal diameter of 125 mm. The total volume of DPAR was 5.0 L, while 2.3. Analytical methods
its effective volume was 4.5 L. In the experiment, the swine
manure was fed into the DPAR through the inlet on the top and The biogas production was measured by a wet-type gas flow
the digestate was discharged through the outlet at the bottom. meter (LML-1, China). The composition of biogas was determined
Additionally, circulating hot water in water bath jacket was used by a biogas component analyzer (ADOS401, Germany). The pH
for thermal control. value was analyzed using an acidometer (pHS-3C+, China). The
In accord with the objective of the experiment, the TS concen- determinations of total solid (TS) and volatile solid (VS) were mea-
tration of the swine manure was adjusted to 20%, 25%, 30%, and sured as follows. Firstly, the sample was dried by a thermostat
35% by oven-drying, air-drying, and adding water to prepare the oven (CS101.2, China) at 105 ± 2 °C to stable weight, and then the
feedstock. TS concentration was calculated. Secondly, the sample was burned
During the startup of the experiment, the reactor was filled up in a muffle furnace (KSL-1100X, China) at 550 ± 20 °C, and then the
with 4.5 L inoculation sludge (TS concentration of 7.65%) and was VS concentration was also calculated (APHA, 2012). The concentra-
placed at the temperature of 25 ± 2 °C in water bath for 5 days for tion of ammonia nitrogen was determined by the UV–VIS spectro-
the purpose of activating sludge. At first, with temperature of 18– photometer (UV-2450, SHIMADZU, Japan). Digestate liquidity was
20 °C and organic loading rate (OLR) of 4.44 gTS(L d)1, fresh measured in the following steps: the digestate from the reactors
swine manure with TS concentration of 20% was loaded into the with various feedstock TS concentrations was collected, then
DPAR. After the indexes stabilized (67 d), the temperature was mixed and prepared. In the DPAR, given the stable temperature
maintained at 25 ± 2 °C with the same organic load, and the feed- at 25 ± 2 °C, a fixed liquid level and volume of digestate (500 mL),
stock with TS concentrations of 25%, 30%, and 35% were loaded the time of digestate discharging from the reactor was determined,
successively and the corresponding indexes were determined. In and the relevant flow and liquidity were calculated.

Table 1
The characteristics of swine manure (SM) and inoculation sludge (IS).

Item pH TS (%) VS/TS (%) NH+4–N (mg L1) TKN/TS (%) TC/TS (%) TP/TS (%) Ca/TS (%) Fe/TS (%) C/N
SM 7.32 23.6 81.3 400 2.28 32.9 2.12 1.96 0.21 14.5
IS 7.55 7.65 57.1 395.3 — — — — — —

NH+4–N: Ammonia Nitrogen; TKN: Total Kjeldahl Nitrogen; TC: Total Carbon; TP: Total Phosphorus.

Please cite this article in press as: Chen, C., et al. Continuous dry fermentation of swine manure for biogas production. Waste Management (2015), http://
dx.doi.org/10.1016/j.wasman.2014.12.024
C. Chen et al. / Waste Management xxx (2015) xxx–xxx 3

3. Results and discussion

3.1. Biogas production rate and components of biogas

As mentioned above, the experiment was implemented with


sufficient inoculation sludge and semi-continuous feeding. TS con-
centration of the inoculation sludge was 7.65% with the VS: TS
ratio of 0.571. Without temperature control (ranging from 18 °C
to 28 °C), acidification arose in the DPAR (pH < 6.3) just 10 days
after the start of the experiment. Biogas production has practically
stopped, because the feedstock of dry fermentation was high in
concentration with the inhomogeneity of the substrate compared
with wet fermentation, so that accumulation of intermediate prod-
ucts appeared during the start-up process, resulting in acid poison-
ing (Forster-Carneiro et al., 2008; Martin et al., 2003). At this point,
feeding was stopped, and the substrate in the DPAR was stirred and
mixed before the further experiment. The pH value in the reactor
began to recover and biogas production restarted after 23 days.
As Fig. 2(a) shows, there were two peaks of biogas production dur-
ing the experiment; one was at 43 d (feedstock TS concentration of
20%) and the other at 107 d (feedstock TS concentration of 25%).
Whereafter, the biogas production rate fell to approximately
5 L d1 at 50 d with a steep drop in the temperature 18–20 °C,
and biogas production at this low temperature continued for about
10 days. After 70 days, because of the seasonal change, the temper-
ature started falling. To carry on the experiment, thermal control
(25 ± 2 °C) was launched with the feedstock changing to swine
manure with a TS concentration of 25%. Then the biogas produc-
tion began to rise, which was shown in Fig. 2. On day 107, the bio-
gas production reached its second peak with the biogas production
rate of 10.2 L d1, holding steady at 8.65 L d1 on day 113. When
the feedstock TS concentration was added to 30% and 35%, the bio-
gas production rate continued to drop, stabilizing at 4.10 L d1 and
2.90 L d1 on day 150 and 180, respectively. An explanation could
be that, as the feedstock TS concentration increased during the
process of the experiment, ammonia nitrogen accumulated in the
fermentation liquor, then when its concentration exceeded
3000 mg L1, there would be ammonia inhibition to the fermenta-
tion (both data and analysis were demonstrated in Section 3.3).
Another explanation could be that, the inhomogeneity of mass
transfer, which was aroused by the increasing TS concentration
in the reactor system during the fermentation process, prohibited
the contact between the anaerobic microorganisms and the
substrate, thereby affecting the fermentation rate.
To a biogas plant, the volumetric biogas production rate and
biogas yield are the two significant indicators for evaluating mate-
rial utilization rate and digester efficiency. As shown in Fig. 2(a)
and (b), when the feedstock TS concentration was 20% with the
temperature at 25–28 °C on day 43, the volumetric biogas produc-
tion rate reached its maximal value of 2.40 L(L d)1. The value was Fig. 2. Biogas production rate and volumetric biogas production rate (a), biogas
close to the result of 2.55 L(L d)1, which was obtained in the yield (b), CH4 content and CH4 yield (c) under different feedstock TS concentrations.

experiment of treating swine manure with TS concentration of


12.1% in a modified up-flow solid reactor at a 38 °C (Wang et al., Generally, CH4 content (v/v) (more than 60%) of the biogas gen-
2009). After day 43, the temperature fell to 18–20 °C, the stable erated is a key indicator of good performance for anaerobic diges-
volumetric biogas production rate was 1.11 L(L d)1, and the bio- tion (Bujoczek et al., 2000). As shown in Fig. 2(c), on day 40 the CH4
gas yields were 0.250 L g1TS and 0.307 L g1VS, respectively. The content reached 65% when the first peak in biogas production was
part of the experiment conducted with feedstock TS concentration obtained, and the CH4 yields were 0.344 L g1TS and 0.424 L g1VS
of 20% showed that dry fermentation of swine manure was greatly respectively, with organic loading rate of 4.44 gTS(L d)1 and feed-
influenced by temperature. As the temperature dropped by 5 °C, stock TS concentration of 20%. After day 45, because of the temper-
the volumetric biogas production rate would fall by 53.7%, with ature declining to 18–28 °C, the CH4 content dropped to 55% and
the feedstock TS concentration of 20%. At 25 ± 2 °C, when the feed- then rose to around 60% after thermal control was launched from
stock TS concentrations were 25%, 30%, and 35%, the stable volu- day 50. Fig. 2(c) also showed that with the feedstock TS concentra-
metric biogas production rates could reach 1.92, 0.911, and tion of 25% the CH4 content increased gradually as the biogas pro-
0.644 L(L d)1, and the biogas yields were 0.432, 0.205, and duction rate increased, and achieved 65% on day 107, when the
0.145 L g1TS, and 0.532, 0.252, and 0.178 L g1VS, respectively. second peak in biogas production was obtained, and the CH4 yields

Please cite this article in press as: Chen, C., et al. Continuous dry fermentation of swine manure for biogas production. Waste Management (2015), http://
dx.doi.org/10.1016/j.wasman.2014.12.024
4 C. Chen et al. / Waste Management xxx (2015) xxx–xxx

were 0.296 L g1TS and 0.364 L g1VS respectively. The probable


cause is that, thermal control raised the temperature from 18–
28 °C to 25 ± 2 °C, thus improving the activity of methanogenic
bacteria. When the feedstock TS concentration was 30%, the CH4
content fell to 60% with the decrease of biogas production rate.
However, the CH4 content obviously increased to 75% with the
decrease of biogas production rate, while the feedstock TS concen-
tration was 35%. This may have been caused by a decrease of CO2
content in the biogas (mainly composed of CH4 and CO2) generated
during anaerobic digestion as a drop in CO2 content would cer-
tainly raise the CH4 content in biogas. The production of CO2 was
caused by the hydrolysis of organics and the methanation of VFA.
It was reported that a decrease in CO2 content could have been
the result of acidification inhibition during substrate hydrolysis
in a reactor system (Kowalczyk et al., 2013; Strik et al., 2006). Fur-
thermore, Calli et al. (2005) pointed out that the influence of
ammonia nitrogen accumulation in restraining acetate methano-
genesis by bacteria was stronger than the influence of H2/CO2 on Fig. 3. Dynamic changes in pH of digestate with different feedstock TS
CH4 production, and therefore the accumulation of high ammonia concentrations.
concentration would partly reduce the CO2 content in the biogas.
In addition, Kotsyurbenko et al. (1993) suggested that, at low tem-
perature (less than 25 °C), most of the CO2 may be converted into bacteria was restrained and the pH was slightly downward. When
acetate by the growth of homoacetogens, reducing the CO2 content the feedstock TS concentrations were 25% and 30%, the pH value
in the biogas. Table 2 showed the variations in biogas production slowly rose from 7.25 to 7.5 due to the increase of methanogenic
under different feedstock TS concentrations. bacteria and ammonia nitrogen accumulation, and then decreased
Compared to other dry fermentation studies, most of which are slightly on days 90 and 131. The concentration of ammonia nitro-
batch processes, the DPAR operated continuously in this study had gen in digestate demonstrated that, when the feedstock TS concen-
better performances. Chae et al. (2008) demonstrated CH4 yields of trations were 25% and 30%, the variations in pH value were
0.312, 0.383 and 0.319 L g1VS at temperatures of 25, 30 and 35 °C, obviously affected by the concentration of ammonia nitrogen
respectively, for anaerobic digestion of swine manure with TS con- (Fig. 4). As the feedstock TS concentration increased from 30% to
centration of 20% in batch experiments, but the results were lower 35%, the pH value increased from 7.5 to 8.1, while the concentra-
than that of 0.424 L g1VS obtained with the same TS concentra- tion of ammonia nitrogen increased to around 3000 mg L1. An
tion at 25 °C in the present study. Additionally, in a batch-operated explanation is that, with the increase of feedstock TS concentration
reactor, Ahn et al. (2010) illustrated a CH4 yield of 0.337 L g1VS for in the DPAR, the concentration of the fermentation liquor increased
dry anaerobic digestion of swine manure at 55 °C with TS concen- (35% in the feedstock and 14% in the digestate), and the substrate
tration of 15%, which was a little lower than that of 0.364 L g1VS of the upper area could hardly make contact with the microbial
obtained by the DPAR with TS concentration of 20% only at 25– sludge of the bottom area. Therefore, the interactions between
28 °C. the substrate and the microorganisms were severely hindered,
affecting the mass transfer of the hydrolysis and acidification reac-
tions. A large number of microorganisms inhabited at the bottom
3.2. Variations in pH of the digestate of the reactor, making it the major methane-producing area, where
VFA could be rapidly consumed and the pH raised due to the acid-
The pH value in the course of anaerobic digestion is the synactic alkali imbalance.
result of the acid-alkali balance, the liquid-solid CO2 balance, and
the solubility equilibrium. The pH value decreased with the
increasing of organic acids, whereas increased with the increasing 3.3. Variations in concentrations of ammonia nitrogen and free
of ammonia, which was the product of the decomposition of ammonia in the digestate
nitrogenous organics (Chae et al., 2008). Fig. 3 shows that, when
the feedstock TS concentration was 20%, the pH of digestate The optimum C/N ratio is 20–30 for the microorganisms of bio-
decreased from an initial value of 7.55 to 5.80, and then increased gas fermentation. The C/N ratio of the swine manure in this study
to about 7.25 on day 40. This may be attributed to the complete was 14.5. Nitrogen surplus in the process of biogas fermentation
inoculation and semi-continuous feeding, providing abundant was likely to cause the accumulation of ammonia nitrogen.
microorganisms which could rapidly decompose organics and pro- As can be seen in Fig. 4, the concentration of ammonia nitrogen
duce VFA at the start-up stage. It also revealed that, on day 45 at increased from an initial value of 395 mg L1g to a maximal value
the low temperature of 18–20 °C, the activity of methanogenic of 3500 mg L1. Given the same organic loading rate of

Table 2
Variations in biogas production under different feedstock TS concentrations*.

Feedstock TS Temperature Biogas production rate CH4 content Volumetric biogas production rate Biogas yield Biogas yield
(%) (°C) (L d-1) (%) L(L d)1 (L g1TS) (L g1VS)
20 25–28 10.8 65.0 2.40 0.540 0.665
20 18–20 5.00 55.0 1.11 0.250 0.307
25 25 ± 2 8.65 65.0 1.92 0.432 0.532
30 25 ± 2 4.10 60.0 0.911 0.205 0.252
35 25 ± 2 2.90 75.0 0.644 0.145 0.178
*
The data in the table are average experimental results after the feedstock TS became stable.

Please cite this article in press as: Chen, C., et al. Continuous dry fermentation of swine manure for biogas production. Waste Management (2015), http://
dx.doi.org/10.1016/j.wasman.2014.12.024
C. Chen et al. / Waste Management xxx (2015) xxx–xxx 5

tration increased from 2000 mg L1 to 3500 mg L1 and then fell to
3000 mg L1 on day 131, the concentration of free ammonia was
up from 30 mg L1 to 100 mg L1 and then down to 55 mg L1.
When the feedstock TS concentration increased to 35%, the total
ammonia nitrogen concentration changed slightly, increasing from
3000 mg L1 to about 3250 mg L1. However, pH value went up
from 7.32 to 8.0, resulting in a rapid increase in the concentration
of free ammonia from 55 mg L1 to 200 mg L1. Accumulation of
free ammonia was mainly affected by three factors such as the
temperature, the total ammonia nitrogen concentration, and the
pH value. With the increase of the temperature, the total ammonia
nitrogen concentration and pH value, the accumulation of free
ammonia would be prone to occur (Bujoczek et al., 2000; Hansen
et al., 1998). By calculating, under the same conditions of total
ammonia nitrogen concentration and temperature, a variation in
pH of 0.1 would result in a variation in free ammonia concentration
Fig. 4. Variations in concentrations of free ammonia and ammonium nitrogen of of 25%. During the experiment, the concentration of free ammonia
digestate under different feedstock TS concentrations.
that caused inhibition was 55 mg L1 (feedstock TS concentration
of 25%). For the inhibition concentration, researchers obtained
4.44 g(L d)1, the concentration of ammonia nitrogen in the dige- inconsistent conclusions. Weiland (1993) and Koster (1986) found
state increased with the advance of experiment and the increase of that the free ammonia concentration of 50 mg L1 or 80 mg L1
TS concentration in feedstock, as a result of moisture content would have an impact on the biogas fermentation process. Accord-
reduction and ammonia nitrogen accumulation in the DPAR. When ing to McCarty and McKinney (1961), ammonia inhibition would
the feedstock TS concentration was 20%, the highest accumulated not occur until the concentration of free ammonia reached
ammonia nitrogen concentration of 2000 mg L1 could be 150 mg L1. However, Ripley et al. (1985) found that the reaction
achieved. could be carried out well even though the concentration of free
When the feedstock TS concentration was up to 25%, the con- ammonia was more than 345 mg L1. Over a long period of sludge
centration of ammonia nitrogen increased to 3000 mg L1. At this acclimation, Hansen et al. (1998) broadened the inhibition concen-
time, the biogas production rate began to decline because of tration to 1100 mg L1. As shown in Fig. 4, the concentration of free
ammonia inhibition. However, CH4 content in biogas stayed at a ammonia leading to inhibition was lower than those reported by
good level (more than 60%). These can be explained by the follow- other researchers, which could be attributed to the following rea-
ing facts. Although the accumulated ammonia nitrogen exceeded sons. On the one hand, the acclimation time was not so long that
the range of half inhibitory concentration for microorganisms of the microorganism is prone to be inhibited at a low concentration
biogas fermentation (more than 1700 mg L1), it still remained of free ammonia. On the other hand, variations in inoculum, sub-
below the toxicity range (more than 4000 mg L1), which brought strate, experimental conditions, and environment may cause the
about a state of ‘‘inhibited biogas production’’ in the reactor difference of the inhibition concentration.
(Angelidaki et al., 1993; Nakamura et al., 2004). When the feed-
stock TS concentration increased to 30% and 35%, the concentration 3.4. Variations in TS concentrations in digestate and its liquidity
of ammonia nitrogen accumulated to a maximum of 3500 mg L1
on day 131 and decreased to 2900 mg L1 on day 148. 3.4.1. Variations in TS concentrations
Ammonium ion (NH+4) and free ammonia (FA, NH3) are the two As demonstrated in Fig. 5, with the feedstock TS concentration
main forms of inorganic ammonia nitrogen in aqueous solution. of 20%, the digestate TS concentration increased from 7.65% to
Free ammonia has been suggested to be the governing cause of 13.6%. Because there were no measures of temperature control,
inhibition since it is freely membrane-permeable by passive experiment was conducted at ambient room temperature of 18–
expansion. In the experiment, the proportion of free ammonia 28 °C, wide fluctuations occurred on days 32 and 65. Since then,
(NH3) in the total ammonia nitrogen could be calculated by the fol- the temperature was kept at 25 ± 2 °C by circulating hot water in
lowing formula (Hansen et al., 1998):

0 0 11 1
pH
½NH3  B B 10
¼ @1 þ @  C C
A A ð1Þ
½TNH3   0:09018þ2729:92
TðKÞ
10

where [NH3] is the concentration of free ammonia, [TNH3] is the


concentration of total ammonia nitrogen, pH is the pH value deter-
mined in the reactor, and T(K) is the temperature,
T(K) = (273.15 + 25)K.
The concentration of free ammonia calculated by the formula
was shown in Fig. 4. With the temperature of 18–20 °C and the
feedstock TS concentration of 20%, the highest free ammonia con-
centration accumulated was 30 mg L1, and no inhibition by free
ammonia occurred. When the temperature was 25 ± 2 °C and the
feedstock TS concentration were 25% and 30%, pH value ranged
from 7.25 to 7.50 in the reactor was maintained. Concentrations
of free ammonia and total ammonia nitrogen changed in accor- Fig. 5. Variations in TS concentrations of digestate and TS removal rates under
dance with the same trend. As the total ammonia nitrogen concen- different feedstock TS concentrations.

Please cite this article in press as: Chen, C., et al. Continuous dry fermentation of swine manure for biogas production. Waste Management (2015), http://
dx.doi.org/10.1016/j.wasman.2014.12.024
6 C. Chen et al. / Waste Management xxx (2015) xxx–xxx

When the TS concentration (w/w) of fresh swine manure was 20%,


25%, 30%, and 35%, the DPAR achieved stable volumetric biogas
production rates of 2.40, 1.92, 0.911, and 0.644 L(L d)1 and biogas
yields of 0.665, 0.532, 0.252, and 0.178 L g1VS, and the TS degra-
dation rates of 46.5%, 45.4%, 53.2%, and 55.6%, respectively.
The concentration of ammonia nitrogen increased in line with
the increase of the TS concentration of feedstock and reached
3500 mg L1 at its maximum. When the concentration of ammonia
nitrogen exceeded 3000 mg L1, biogas production was inhibited
obviously in the reactor. The liquidity experiment showed that
the digestate could be easily discharged from a DPAR when the
feed TS concentration was less than 35%. Considering the biogas
production rate and the utilization rate of feedstock, feeding TS
concentration should not exceed 30%.
Meanwhile, compared to previous dry mesophilic fermentation
studies, the DPAR in this study possessed several advantages for
further engineering application.

(1) Energy saving


Fig. 6. Relationship between digestate TS concentration and flow rate.

The DPAR was operated successfully at ambient temperature


without any heating or stirring, guaranteeing low operational
water bath jacket. With the increase of TS concentration in the
costs.
feedstock, digestate TS concentration was up from 13.6% (65 d)
to 15.9% at the end of the experiment. And digestate TS concentra-
(2) High biogas yield
tions were 10.7%, 13.6%, 14.0%, and 15.8%, removal rates of TS were
46.5%, 45.4%, 53.2%, and 55.6% for feedstock with concentration of
On the premise of digestate discharging and without ammonia
20%, 25%, 30%, and 35%, respectively.
inhibition, the DPAR would achieve higher biogas yield given
TS degradation rates went up with the increases of the feed-
rather high TS concentration.
stock TS concentration, but the biogas production rate did not rise.
Main reasons could be attributed to: firstly, the proportion of non-
(3) Continuous operation
degraded TS from the original sludge was high when feeding TS
concentration was 20%, which led to a rise in total digestate TS con-
Unlike batch processes, the DPAR could run continuously and
centration and less removal rates of TS. Secondly, in the beginning
effectively for biogas production.
stage of the experiment, there was an obvious relative motion
between the high concentration feedstock and the low concentra-
tion inoculation sludge. No plug flow occurred in the reactor, and a Acknowledgements
large amount of non-degraded feedstock was present in the dige-
state, which lowered the feedstock degradation rate. The authors would be very grateful for the financial support
from China Agriculture Research System (CARS-36-10B).
3.4.2. The liquidity of digestate
In continuous dry fermentation process of swine manure, the TS References
concentration should not be too high due to poor liquidity of the
Abouelenien, F., Nakashimada, Y., Nishio, N., 2009. Dry mesophilic fermentation of
substrate. Whether the digestate would be easily discharged or chicken manure for production of methane by repeated batch culture. J. Biosci.
not is a main problem for continuous dry fermentation. In this Bioeng. 107, 293–295.
study, the flow rate and flow velocity were used to reflect the dis- Ahn, H.K., Smith, M.C., Kondrad, S.L., White, J.W., 2010. Evaluation of biogas
production potential by dry anaerobic digestion of switchgrass-animal manure
charge capability of digestate in the reactor. In Fig. 6, the flow rate mixtures. Appl. Biochem. Biotechnol. 160, 965–975.
and flow velocity of digestate decreased linearly with an increase Angelidaki, I., Ellegaard, L., Ahring, B.K., 1993. A mathematical model for dynamic
in TS concentration (coefficient of determination R2 = 0.983). With simulation of anaerobic digestion of complex substrates: focusing on ammonia
inhibition. Biotechnol. Bioeng. 42, 159–166.
the temperature of 25 ± 2 °C and the digestate volume of 500 mL,
APHA, 2012. Standard Methods for the Examination of Water and Wastewater,
the flow rate and flow velocity were 2.5 mL s-1 and 0.007 m s1 Twenty second ed. American Public Health Association/American Water Works
respectively, while the TS concentration reached 21.1%. In this Association/Water Environment Federation, Washington, D.C., USA.
Bolzonella, D., Innocenti, L., Pavan, P., Traverso, P., Cecchi, F., 2003. Semi-dry
case, the reactor cannot smoothly discharge digestate. In the
thermophilic anaerobic digestion of the organic fraction of municipal solid
experiment of continuous dry fermentation, when the digestate waste: focusing on the start-up phase. Bioresour. Technol. 86, 123–129.
concentrations were 10.7%, 13.6%, 14.0%, and 15.8%, the flow rates Borja, R., Rincón, B., Raposo, F., Alba, J., Martı́n, A., 2003. Kinetics of mesophilic
were 808, 569, 542, and 310 mL s1, respectively, correspondingly, anaerobic digestion of the two-phase olive mill solid waste. Biochem. Eng. J. 15,
139–145.
the flow velocities were 2.57, 1.81, 1.72, and 0.980 m s1, respec- Bujoczek, G., Oleszkiewicz, J., Sparling, R., Cenkowski, S., 2000. High Solid Anaerobic
tively, clearly higher than those when the digestate TS concentra- Digestion of Chicken Manure. J. Agric. Eng. Res. 76, 51–60.
tion was 21.07%. The liquidity experiment showed there was no Calli, B., Mertoglu, B., Inanc, B., Yenigun, O., 2005. Effects of high free ammonia
concentrations on the performances of anaerobic bioreactors. Process Biochem.
difficulty for the discharge of digestate in the DPAR using swine 40, 1285–1292.
manure as feedstock with TS concentration less than 35%. Chae, K., Jang, A., Yim, S., Kim, I.S., 2008. The effects of digestion temperature and
temperature shock on the biogas yields from the mesophilic anaerobic
digestion of swine manure. Bioresour. Technol. 99, 1–6.
4. Conclusions De Bere, L., 2000. Anaerobic digestion of solid waste: state-of-the-art. Water Sci.
Technol. 41, 283–290.
Deng, L., Li, Y., Chen, Z., Liu, G., Yang, H., 2014. Separation of swine slurry into
A down plug-flow anaerobic reactor (DPAR) using fresh swine different concentration fractions and its influence on biogas fermentation. Appl.
manure as the feedstock was successfully constructed in this study. Energy 114, 504–511.

Please cite this article in press as: Chen, C., et al. Continuous dry fermentation of swine manure for biogas production. Waste Management (2015), http://
dx.doi.org/10.1016/j.wasman.2014.12.024
C. Chen et al. / Waste Management xxx (2015) xxx–xxx 7

Dong, L., Yuan, Z., Sun, Y., 2010. Semi-dry mesophilic anaerobic digestion of water Kusch, S., Oechsner, H., Jungbluth, T., 2008. Biogas production with horse dung in
sorted organic fraction of municipal solid waste (WS-OFMSW). Bioresour. solid-phase digestion systems. Bioresour. Technol. 99, 1280–1292.
Technol. 101, 2722–2728. Lantz, M., 2012. The economic performance of combined heat and power from
El-Mashad, H.M., Zeeman, G., van Loon, W.K.P., Bot, G.P.A., Lettinga, G., 2004. Effect biogas produced from manure in Sweden – a comparison of different CHP
of temperature and temperature fluctuation on thermophilic anaerobic technologies. Appl. Energy 98, 502–511.
digestion of cattle manure. Bioresour. Technol. 95, 191–201. Lissens, G., Vandevivere, P., De Baere, L., Biey, E.M., Verstraete, W., 2001. Solid waste
Fdez-Güelfo, L., Alvarez-Gallego, C., Sales Márquez, D., Romero García, L., 2010. digestors: process performance and practice for municipal solid waste
Start-up of thermophilic–dry anaerobic digestion of OFMSW using adapted digestion. Wat. Sci. Technol. 44, 91–102.
modified SEBAC inoculum. Bioresour. Technol. 101, 9031–9039. Lu, S.G., Imai, T., Ukita, M., Sekine, M., 2007. Start-up performances of dry anaerobic
Fierro, J., Gómez, X., Murphy, J.D., 2014. What is the resource of second generation mesophilic and thermophilic digestions of organic solid wastes. J. Environ. Sci.
gaseous transport biofuels based on pig slurries in Spain? Appl. Energy 114, 19, 416–420.
783–789. Martin, D.J., Potts, L.G.A., Heslop, V.A., 2003. Reaction mechanisms in solid-state
Forster-Carneiro, T., Pérez, M., Romero, L.I., 2008. Influence of total solid and anaerobic digestion: ii. the significance of seeding. Process Saf. Environ. Prot. 81,
inoculum contents on performance of anaerobic reactors treating food waste. 180–188.
Bioresour. Technol. 99, 6994–7002. Massé, D.I., Masse, L., Croteau, F., 2003. The effect of temperature fluctuations on
Guendouz, J., Buffière, P., Cacho, J., Carrère, M., Delgenes, J.P., 2010. Dry anaerobic psychrophilic anaerobic sequencing batch reactors treating swine manure.
digestion in batch mode: design and operation of a laboratory-scale, completely Bioresour. Technol. 89, 57–62.
mixed reactor. Waste Manage. 30, 1768–1771. McCarty, P.L., McKinney, R.E., 1961. Salt toxicity in anaerobic digestion. J. Water
Hansen, K.H., Angelidaki, I., Ahring, B.K., 1998. Anaerobia digestion of swine Pollut. Control Fed. 33, 399–415.
manure: inhibition by ammonia. Water Res. 32, 5–12. Nakamura, S., Kishi, I., Sugimura, S., Hamajima, M., Mimura, R., Tsuji, Y., 2004. Effect
Hansen, K.H., Angelidaki, I., Ahring, B.K., 1999. Improving thermophilic anaerobic of ammonia inhibition and loading rate on thermophilic methane fermentation
digestion of swine manure. Water Res. 33, 1805–1810. of garbage. Resour. Process 51, 141–147.
Hilkiah Igoni, A., Ayotamuno, M.J., Eze, C.L., Ogaji, S.O.T., Probert, S.D., 2008. Designs Ripley, L.E., Kmet, N.M., Boyle, W.C., Converse, J.C., 1985. The effects of ammonia
of anaerobic digesters for producing biogas from municipal solid-waste. Appl. nitrogen on the anaerobic digestion of poultry manure. In: Proceedings, 39th
Energy 85, 430–438. Purdue Industrial Waste Conference, Purdue University, Butterworth
Hothan, A., Brauckmann, H.-J., Broll, G., 2013. Influence of storage on methane Publishers, Boston, MA, USA, pp. 73–80.
yields of separated pig slurry solids. Biomass Bioenerg. 52, 166–172. Strik, D.P.B.T.B., Domnanovich, A.M., Holubar, P., 2006. A pH-based control of
Kafle, G.K., Kim, S.H., 2013. Anaerobic treatment of apple waste with swine manure ammonia in biogas during anaerobic digestion of artificial pig manure and
for biogas production: batch and continuous operation. Appl. Energy 103, 61– maize silage. Process Biochem. 41, 1235–1238.
72. Wang, G., Zhao, M., Lang, X., Bao, Z., Zhou, R., 2009. Study on an improved USR for
Koster, I.W., 1986. Characteristics of the pH-influenced adaptation of methanogenic pig manure treatment. Chin. J. Environ. Eng. 3, 919–921 (in Chinese).
sludge to ammonium toxicity. J. Chem. Technol. Biotechnol. 36, 445–455. Weiland, P., 1993. One-and two-step anaerobic digestion of solid agroindustrial
Kotsyurbenko, O.R., Nozhevnikova, A.N., Zavarzin, G.A., 1993. Methanogenic residues. Water Sci. Technol. 27, 145–151.
degradation of organic matter by anaerobic bacteria at low temperature. Yabu, H., Sakai, C., Fujiwara, T., Nishio, N., Nakashimada, Y., 2011. Thermophilic
Chemosphere 27, 1745–1761. two-stage dry anaerobic digestion of model garbage with ammonia stripping. J.
Kowalczyk, A., Harnisch, E., Schwede, S., Gerber, M., Span, R., 2013. Different mixing Biosci. Bioeng. 111, 312–319.
modes for biogas plants using energy crops. Appl. Energy 112, 465–472.

Please cite this article in press as: Chen, C., et al. Continuous dry fermentation of swine manure for biogas production. Waste Management (2015), http://
dx.doi.org/10.1016/j.wasman.2014.12.024