Bioresource Technology 101 (2010) 6361–6367

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Comparison of atmospheric pressure effects on the anaerobic digestion

of municipal solid waste
Jianguo Jiang a,b,*, Xuejuan Du a, Siio Ng a, Chang Zhang a
a
Department of Environmental Science and Engineering, Tsinghua University, Beijing 100084, PR China
b
Key Laboratory for Solid Waste Management and Environment Safety, Ministry of Education of China, PR China

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

Article history: This study compared the performance of anaerobic digestion (AD) of municipal solid waste under 101 kPa
Received 17 December 2009 (R system) and 65.8 kPa (RPC system) – the atmospheric pressure on the Qinghai-Tibetan Plateau, China.
Received in revised form 4 March 2010 Gas production, gas composition, degradation of volatile solids, pH, alkalinity, volatile fatty acids, and
Accepted 15 March 2010
ammonia concentration were analyzed to examine how the two systems responded to change in the
Available online 1 April 2010
organic loading rate. The RPC system had a stronger buffering capacity and lower gas production rate
and could achieve a higher loading rate. The pH in RPC recovered more rapidly in the starting stage
Keywords:
and remained higher than that in R during the whole experiment, with an average difference of 0.2. Dur-
Anaerobic digestion
Atmospheric pressure
ing days 38–55, 56–70, and 71–125, the average methane production rates were 380.16, 318.67, and
High plateau 402.21 L kg1 VS in R and 367.40, 299.04, and 275.06 L kg1 VS in RPC, respectively.
Municipal solid waste Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction mesophilic temperature, and the cumulative methane production

was increased about 35% in optimal pH. The starting stage of AD
Anaerobic digestion (AD) has been increasingly applied to treat are significantly influenced by two most important factors – inoc-
solid waste worldwide, including in high-altitude regions. A prop- ula (Castillo et al., 2006) and total solid content (Li et al., 2009).
erly functioning AD system can achieve highly efficient biogas pro- Many other factors, including feedstock (Carneiro et al., 2008a,b),
duction to reduce fossil fuel consumption and greenhouse gas HRT (Salminen and Rintala, 2002; Rincón et al., 2008), stirring
emission and can transform organic waste into high-quality fertil- (Stroot et al., 2001), and ammonia inhibition (Sung and Liu,
izer (Tafdrup, 1995; Edelmann et al., 2005). 2003; Fricke et al., 2007) may also affect the performance of AD.
The performance of AD of the organic fraction of municipal solid Álvarez et al. (2010) have developed a linear programming optimi-
waste (OFMSW) generally depends on operational parameters such sation method for determining the most adequate ratios of differ-
as temperature, pH, and organic loading. Previous studies have ent co-substrates that provide an optimised biodegradation
concluded that the rates of biogas production with thermophilic potential or biokinetic methane potential.
digesters are higher than those with mesophilic digesters regard- Highlands cover one-fourth of China’s territory and occur
less of hydraulic retention time (HRT), although thermophilic mostly in the Qinghai-Tibetan Plateau, which has an average alti-
digesters are lower in energy conservation and exhibit less stability tude of about 4000 m. Atmospheric pressure decreases with an
(Gallert and Winter, 1997; Kim et al., 2006). As the temperature increasing altitude. According to Henry’s Law, at a given tempera-
decreases, the AD digester performs lower degradation rate of vol- ture, a decrease in total pressure will result in a reduction of the
atile solids (VS) and less biogas production (Chen et al., 2009). The partial pressure and solubility of CO2. This then changes the equi-
value of pH is the pivotal factor that affects the methane produc- librium of CO2 in the liquid phase and consequently affects the pH
tion efficiency. Studies have suggested that a pH range of 7–8 is and buffering capacity of a system (Rittmann and McCarty, 2002).
suitable for obtaining higher biogas production and degradation Hayes et al. (1990) reported obtaining higher methane content un-
of VS (Dinamarca et al., 2003; Zhang et al., 2005a). By Liu et al. der a higher pressure in a digester and less methane at a lower
(2008), it was observed that the optimal value of pH is 7.10 under pressure. Álvarez et al. (2006) set up reactors on a high plateau
(3000–4000 m altitude) in Bolivia to study biogas production from
manure; their results showed little difference between high and
* Correspondence to: Jianguo Jiang, Department of Environmental Science and
Engineering, Tsinghua University, Haidian District, Beijing 100084, PR China. Tel./
low altitudes. However, the experiment only considered gas pro-
fax: +86 10 62783548. duction, and it is difficult to know whether the other parameters
E-mail address: [email protected] (J. Jiang). of AD or equipment restrictions limited the outcomes of their

0960-8524/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2010.03.068
6362 J. Jiang et al. / Bioresource Technology 101 (2010) 6361–6367

experiment. In fact, by changing the total pressure in a digester, 2. Methods

toxicity effects can be diminished or avoided. For example,
decreasing the pH by raising the partial pressure of CO2 can reduce 2.1. Apparatus
the non-ionized ammonia concentration and avoid ammonia inhi-
bition. Conversely, increasing the pH can lower the level of non- The experiments were conducted using two semi-continuous
ionized hydrogen sulfide (Vavilin et al., 1995). This suggests that AD systems: one with pressure control (RPC) and the other without
by appropriately adjusting the total pressure in a digester, several (R), as shown in Fig. 1a and b, respectively. Both systems had a
inhibitory effects on AD may be eliminated. Zhang et al. (2005b) working volume of 5 L.
tried absorbing CO2 in the gas phase to reduce the CO2 partial pres- Fig. 1a illustrates the RPC system. The digester was cylindrical,
sure and found it to be helpful in preventing acidification. with a diameter of 15 cm and a height–diameter ratio of 3. The in-
This study examined the influence of atmospheric pressure on let and outlet were placed at the top and bottom of the digester,
the performance of AD of OFMSW, using a self-designed AD exper- respectively. A bivalve design was applied to the inlet to minimize
imental system that simulates different pressures. The reactors the pressure change caused by incoming air. Other ports on the top
were operated semi-continuously under two pressures: 101 and of the digester included a gas exit, pressure probe jack, tempera-
65.8 kPa, with the latter representing the average atmospheric ture probe jack, pH probe jack, and spare jacks. A stirring device
pressure of the Qinghai-Tibetan Plateau. Gas production, gas com- was connected to the digester. Pressure, temperature, and stirring
position, degradation of volatile solids (VS), pH, alkalinity, volatile frequency in the digester could be adjusted using the multi-func-
fatty acids (VFAs), and ammonia concentration were analyzed and tion controller. Hot water passed through the water pipe inside
compared. the digester to maintain the temperature of the materials inside.

3
Connected to
4 the components
5

Gas
exit 7

8 9
1 6

10 Connect to
atmosphere
13
2

14
11 12

(a)

T
6
Inlet

Outlet
T
4 5 7
(b)
Fig. 1. Schematic diagram of the two anaerobic digestion systems: (a) RPC: (1) heater; (2) water pump; (3) inlet bottle; (4) valve; (5) pressure gauge; (6) multi-function
controller; (7) electronic switch; (8) stirrer; (9) hand switch; (10) hot water pipe; (11) outlet collector; (12) buffering bottle; (13) vacuum pump; (14) flow meter; (b) R: (1)
lagging cloth; (2) resistance wire; (3) stirrer; (4) valve; (5) temperature controller; (6) stirrer controller; (7) flow meter.
 J. Jiang et al. / Bioresource Technology 101 (2010) 6361–6367 6363

Pressure control was the synergic result of the controller, pres- et al., 2005; Carneiro et al., 2008a,b). To avoid this influence in the
sure probe, electronic switch, and vacuum pump. A predetermined experiment, artificial organic waste was applied as described in
pressure could be set with the controller. When the pressure in the Table 1, based on an investigation of the composition of municipal
digester (measured by the pressure probe) rose above the designed solid waste in Tibet. Materials were crushed into small pieces
level, the electronic switch opened and repeatedly discharged the (<10 mm) before being blended and stored at 4 °C until use. Tables
excess gas until the pressure decreased. The pressure variation 1 and 2 show the compositions and properties of the artificial or-
was less than ±5 kPa. ganic waste.
Fig. 1b depicts the R system. The digester is the same as in the
RPC system except that the inlet and outlet are not connected to 2.3. Experimental design
pressure control components but to the atmosphere. Heating is
provided by the resistance wire and the lagging cloth working to- The experiments were conducted in Beijing, where the pressure
gether, under the control of a temperature controller. A gas exit is is about 101 kPa. The atmospheric pressures of R and RPC were set
located at the top and connected to a flow meter. at 101 and 65.8 kPa, respectively. Except for pressure, the two sys-
tems were operated under the same reacting conditions, including
2.2. Feedstock temperature (35 °C), stirring (10 min every 4 h at a speed of
120 rpm), and amounts of input and output.
AD performance can vary significantly for different kinds of The whole experiment lasted 146 d which was divided into
waste (Mandal and Mandal, 1997; Krzystek et al., 2001; Bouallagui three phases, and Table 3 lists the duration and operating param-
eters of each phase. Phase A (0–37 d) was the starting stage, where
Table 1 about 70% feedstock and 30% inoculation sludge, obtained from the
Composition of artificial organic waste.
digester of the Beijing Gaobeidian Wastewater Treatment Plant,
Constituent Percentage were mixed and placed in the reactors. The pH of the digester con-
Rice 34.65 tents was monitored daily and adjusted with 1 M NaHCO3. When
Vegetables 27.49 the pH rose above 7, the digester contents were regularly with-
Meat 13.55 drawn from each reactor and replaced with a fresh feed of the
Beans 5.90
Oil 0.83
same weight via the inlet, following the feeding method in
Salt 0.80 Table 3. Phase B started at day 38, when the pH stabilized above
Paper 1.88 7, and was divided into three stages according to the organic load-
Garden waste 5.96 ing rate (OLR) and feeding method. The OLR was low in the early
Other 8.94
stage of the experiment, to achieve a more stable state of perfor-
mance for comparison, and was then gradually raised to investi-
gate the influence of different OLRs on reactor performance. The
Table 2 OLR was raised to 6.82 kg m3 d1 in phase C on day 126 and main-
Properties of artificial organic waste. tained at this level until the end of the experiment to examine the
Item Value abilities of the systems to accommodate a high OLR.
Total solids (TS, % wet) 23.73
Volatile solids (% wet) 22.51 2.4. Analyses
C/N 24.83
Density (g mL1) 1.012
The cumulative gas production was measured by a gas flow me-
ter (LML-2, Changchun Automotive Filters Co., Ltd.) at the relevant
pressure and designed temperature. Methane and CO2 concentra-
Table 3 tions in the biogas were tested using a gas chromatograph – SHI-
Operating parameters of each phase of the experiment. MADZU GC2010. The pH of the digester contents was determined
Phase A B C with a pH-meter (Delta 320A, Mettler-Toledo International Inc.)
B1 B2 B3
at a specified time every day.
Total solids (TS) and VS were determined according to Standard
Duration (d) 0–37 38–55 56– 71–125 126–
Methods (Jiang, 2008). The TS content was determined after heat-
70 146
Feeding method 25 g/d when pH > 7 25 g/ 25 g/ 50 g/ 100 g/d ing (105 °C) for 24 h. VS was determined by igniting the residue
2d d 2d produced in the TS analysis in a muffle furnace (SX2-5-12, Tianjing
OLR (kg m3 d1) 0.78 1.76 1.76 6.82 Zhonghuan Experiment Electric Stove Co., Ltd.) at a temperature of
600 °C for 2 h.

Phase A Phase B Phase C

7.5

6.5
pH

R RPC

5.5

4.5
0 25 50 75 100 125 150
Time (d)

Fig. 2. pH profile of anaerobic digestion of artificial waste at 101 kPa (R) and 65.8 kPa (RPC).
6364 J. Jiang et al. / Bioresource Technology 101 (2010) 6361–6367

Ammonia was measured by an UV spectrophotometer (UV- where [A] represents the summation of the molar concentrations
2000, UNICO Instruments Co., Ltd.) with salicylic acid spectropho- of all weak acid salts present, except bicarbonate and carbonate.
tometry from the Standard Methods (MEP, 2002). The C/N ratio The units of alkalinity are mg/L as CaCO3, and [CO2 (g)] is the partial
was determined with an elemental analyzer (CE-440, Exeter Ana- pressure of CO2.
lytical Inc.). Alkalinity analyses were carried out using a pH-meter Eq. (1) indicates that the pH is controlled by the concentrations
(Delta 320A, Mettler-Toledo International Inc.) with potentiomet- of alkalinity and VFAs in the liquid phase and CO2 in the gas phase
ric titration method (MEP, 2002). The digester contents in the of the reactor, assuming that CO2 equilibrium exists between the
amount of 1 lL at a time were used for VFA determination using gas and liquid phases. According to Eq. (1), pH increases when
a gas chromatograph (GC3486, Beijing Beifen Huapu Analysis [CO2 (g)] decreases. Therefore, a low-pressure environment would
Equipment Technology Co., Ltd.) with a polar chromatographic col- enable AD to proceed at a higher pH, such that the AD system
umn (GDX-103, Beijing ZXYD Technology Co., Ltd.). would be better able to resist acidification and achieve a higher
OLR. Zhang et al. (2005b) succeeded in reducing acidification dur-
3. Results and discussion ing AD by absorbing CO2 to reduce the partial pressure of CO2.
Their result implies that a higher OLR can be applied to the AD of
3.1. pH OFMSW in a high-altitude area.

Fig. 2 shows the variation in pH of the two systems. Significant 3.2. Biogas production
t-test results of phases A, B1, B2, and B3 were all 0.000 (a = 5%),
indicating that the pH of the two systems differed significantly. Fig. 3 shows cumulative gas production, gas production rates,
In the early stage of phase A, acidification was noted in the two and pH in the three phases of the AD systems. Gas production rates
reactors, as the pH decreased to 4.7–4.9 on days 7–8. During this in B1 and B3 were calculated as the averages of every 2 d because
period of acidification, the pH was adjusted with 1 mol L1 NaH- of the feed mode adjustment. Gas production rates in phases A, B1,
CO3, whereupon the pH recovered to above 7. B2, and B3 were assessed by t-tests (a = 5%), and the Sig. results
During days 38–55, the pH in R and RPC fluctuated between were 0.017, 0.330, 0.202, and 0.000, respectively.
7.09–7.24 and 7.19–7.46, with average pH values of 7.13 and
7.31, respectively. It was found that the pH remained higher in 3.2.1. Phase A
RPC compared with the system under normal pressure and that Fig. 3a presents the cumulative gas production in phase A. The
this trend continued during the whole experiment. curves were similar at the beginning but started to separate at
During days 56–70, pH values in both systems decreased as a day 23. The cumulative gas production on day 37 was 26.36 L in
response to the OLR increase. In R, pH declined from 7.14 to 6.84, R and 32.61 L in RPC. As shown in Fig. 3b, the gas production rate
with a larger difference of 0.3 than that in RPC, where the pH de- curves of R and RPC were close in phase A, with no clear distinc-
creased from 7.4 to 7.26 for a difference of only 0.14. Feeding tions from each other (t-test, a = 5%, Sig. = 0.017). Obvious acidifi-
was stopped on day 70 to prevent the systems from acidifying, cation occurred in both systems in the early stage, restraining
and after 1 d, the pH in R and RPC increased dramatically to 7.25 the activity of methanogenic bacteria and thus leading to a low
and 7.4, respectively, because a majority of the VFAs were con- gas production rate. This phenomenon can be explained by the
sumed. Thereafter, the feed mode was changed to 50 g/2 d, and imbalance of hydrolysis/acidogenesis and methanogenesis. The
pH remained steady in both systems until the end of phase B. In optimal pH of acidogenic bacteria is 5.8, while methanogenic bac-
addition, inspection of the variation in pH at the time at which teria need a condition of pH above 6.2. The pH was adjusted using
OLR changed showed that the waves in the RPC curve were smaller NaHCO3, and as the pH was increased, bacteria groups developed
and recovered more quickly than those in the curves for the other and became acclimated, increasing the gas production rate.
reactor. The pH curves in Fig. 3b show a large increase starting on day 9,
The adjustment of the OLR in phase C led to significant changes when the NaHCO3 was added. The line representing RPC recovered
in the AD systems. At the beginning of phase C, the pH in R de- more rapidly, increasing above 7 2 d earlier than the line of R.
creased dramatically, from 7.29 to 6.07. These results indicate that Thereafter, the pH in RPC was higher than that in R (t-test,
the acidification of R was a consequence of the decrease in the gas a = 5%, Sig. = 0.000), with an average difference of 0.23 between
production rate. A different result was seen in RPC, where the pH the two curves.
variation was considerably smaller. The pH decreased, reaching a
value of 7.14 at day 131, and then rose to stabilize at about 7.38 3.2.2. Phase B
for 6 d, followed by another decrease until the end of the experi- Fig. 3c presents the cumulative gas production in phase B. As
ment. The pH in RPC at day 146 was 7.07, still in the optimal range shown in Fig. 3d, during days 38–55, the OLR was 0.78 kg m3 d1,
for the AD system (6.8–7.2). and the gas production rates of R and RPC were 2.10 and 2.12 L/d,
The pH is determined according to the following equation (Ritt- respectively. The difference between the two curves was not signif-
mann and McCarty, 2002): icant (t-test, a = 5%, Sig. = 0.330). During this period, pH rose from
7.09 to 7.24 in R and from 7.19 to 7.46 in RPC. Methanogenic bac-
½Alkalinity=50; 000 teria are more active than acidogenic bacteria at this range of pH,
pH ¼ pKa;1 þ log ; ð1Þ
½CO2 ðgÞ=½KH  and consumed a great deal of VFAs. Meanwhile, VS degradation
in the two systems did not differ significantly (t-test, a = 5%,
where Sig. = 0.028), suggesting similar activity of hydrolysis/acidogenesis
bacteria in both systems. The average methane production rates
½Hþ ½HCO3 
¼ Ka;1 ¼ 5  107 ð35  CÞ; (AMPR) during days 38–55 were 380.16 L kg1 VS, in R and
½H2 CO3 
367.40 L kg1 VS in RPC.
During days 56–70, the OLR was twice that in the previous per-
½Hþ  þ ½Alkalinity ¼ ½A  þ ½HCO3  þ 2½CO2 
3  þ ½OH ; iod, reaching 1.76 kg m3 d1. The gas production rates of R and
RPC in this stage stabilized at around 3.42 and 3.31 L/d, represent-
½CO2 ðgÞ 1 ing increases of 62.9% and 56.1%, respectively, due to the increase
¼ KH ¼ 38 atm L mol ð35  CÞ;
½H2 CO3  in substrate. However, the VFAs produced could not be consumed
 J. Jiang et al. / Bioresource Technology 101 (2010) 6361–6367 6365

(a) Cumulative gas production (L) 40 (d) 8 8

Gas production rate (L d-1)

7 7.5
30 R-biogas 6
7
RPC-biogas R-biogas
5 RPC-biogas 6.5

pH
4 R-pH
20 RPC-pH
6
3
5.5
10 2
1 5

0 0 4.5
0 10 20 30 40 38 56 66 82 102 122
Time (d) Time (d)

(b) 4 7.5
R-biogas (e) 500

Cumulative gas production (L)

3.5 RPC-biogas
Gas production rate (L d-1)

7 R-biogas
R-pH
3 450 RPC-biogas
RPC-pH
6.5
2.5
400
2 6

pH
1.5 350
5.5
1
5 300
0.5
0 4.5 250
0 10 20 30 125 135 145
Time (d) Time (d)

(c) (f) 20 8
Cumulative gas production (L)

380 18
Gas production rate (L d-1)

7.5
330 16
R-biogas 7
14
280 RPC-biogas
12 6.5
230

pH
10
180 8 6
R-biogas
6 RPC-biogas
130 R-pH
5.5
4 RPC-pH
80 5
2
30 0 4.5
40 60 80 100 120 126 131 136 141 146
Time (d) Time (d)

Fig. 3. Comparison of gas production rates and cumulative gas production of the two systems: (a) cumulative gas production in phase A; (b) gas production rates in phase A;
(c) cumulative gas production in phase B; (d) gas production rates in phase B; (e) cumulative gas production in phase C; (f) gas production rates in phase C.

immediately and pH decreased from 7.14 to 6.84 in R and from 7.4 of RPC was continuously higher than that of R. Together, this evi-
to 7.26 in RPC. The average methane production rates during days dence indicates that the difference in gas production between
56–70 were 318.67 L kg1 VS for R and 299.04 L kg1 VS for RPC, the two systems resulted from acidogenic bacteria being restrained
indicating that there was no obvious difference in the activity of by the high pH in RPC. Further proof of this suggestion was pro-
methanogenic bacteria in the systems. vided by the VS degradation results for this phase, which were
During days 71–125, the feed mode was changed from 25 g/d to 2–4% lower in RPC than in R.
50 g/2 d to study the influence on the performance of the AD sys-
tems, and the gas production rate was calculated as the average 3.2.3. Phase C
of every 2 d. As shown in Fig. 3d, the gas production rates in the Cumulative gas production in phase C is shown in Fig. 3e. To
two systems fell at the beginning of the increase in OLR, possibly investigate the response of the two systems to a high loading rate,
because the bacteria systems were unable to immediately acclima- the OLR was further increased to 6.82 kg m3 d1 during this
tize to the new conditions. The subsequent increase in R from days phase. The gas production rate of the R system rapidly increased
74 to 82, and in RPC from days 72 to 80, suggests that the bacteria to 9.74 L/d at day 127, about 235% of the prior level, and then grad-
community had gradually adapted to the new conditions. The aver- ually decreased to 7.43 L/d at day 132. The pH of this system de-
age gas production rate in R was around 4.31 L/d, a 26.0% increase creased continuously, from 7.29 to 6.37, inhibiting the activity of
compared to the previous period, whereas that in RPC was around methanogenic bacteria.
3.01 L/d, a decrease of 9.1%. The pH in R fluctuated between 7.08 Unlike the R system, the gas production rate of RPC increased
and 7.54, 0.2–0.3 lower than that in RPC, which fluctuated between steadily until day 143, reaching 10.99 L/d, approximately 373% of
7.22 and 7.56. The average methane production rate of R during the previous production, and the pH fluctuated between 7.14 and
days 71–125 increased to 402.21 L kg1 VS, whereas the AMPR of 7.4. After this, the RPC gas production rate gradually decreased to
RPC remained at about 275.06 L kg1 VS. 4.79 L/d, while pH decreased to 6.85. Clearly, in this phase, the sys-
In phase B, the gas production of the AD system under lower tem under lower pressure performed better at a high loading rate.
pressure was remarkably less than that under normal pressure. In this phase, average methane production rate in RPC was
The results of VFA tests showed no accumulation, and the pH curve 177.21 L kg1 VS.
6366 J. Jiang et al. / Bioresource Technology 101 (2010) 6361–6367

Álvarez et al. (2006) found that the pressure effect (at pressures et al. (2008) an appropriate buffering capacity and high stability
of 65.8 and 101 kPa) was not significant in any case. In the present of the experimental system for HRT in the range of 17–108 d and
study, the same result was obtained under a low loading rate. a destabilization and deterioration of the process and an inhibition
However, the situation was quite different at a high loading rate, of anaerobic system occurred at an HRT of 15 d.
where the performance of the two systems under different pres- As an intermediate of AD from organic matter degradation, VFA
sures varied significantly. concentration is an important index of the metabolism of anaero-
bic microbes. High VFAs lead to acid inhibition of the digestion sys-
3.3. Degradation of VS tem. The concentration of VFAs in an AD reactor can reflect the
bioactivity of methanogenic bacteria and the degree of system
Fig. 4 shows the variation in the VS degradation rate and the deterioration. In phase B, the VFA concentrations in both systems
VFAs during the experiment. On days 38–55, the average degrada- were less than 500 mg/L and did not accumulate. In phase C, the
tion rates of VS in R and RPC were 93.3% and 94.3%, respectively, VFA concentrations increased rapidly, resulting in a pH decrease
which were both at high levels due to the low OLR and long HRT. in both systems.
On days 56–125, the OLR of both systems remained unchanged,
but the VS degradation rates in R and RPC gradually decreased 3.4. Gas composition
from 94.02% and 91.85% (at 57 d) to 85.06% and 84.80% (at
122 d), respectively. The degradation rate of VS in R was always The average methane contents of R and RPC biogas during days
higher than that in RPC, especially during days 71–125, possibly 56–125 were 59.25% and 60.01%, respectively. The results of t-tests
because high pH inhibited the activity of acid-producing bacteria revealed no significant differences in the methane contents of R
in RPC. In phase C, the OLR was increased to 6.82 kg m3 d1, short- and RPC biogas (Sig. = 0.021). Theoretically, a decrease in pressure
ening the HRT to 30 d, and the degradation rate of VS in RPC further reduces the solubility of gas, leading to its discharge from the sol-
decreased to 78%. vent. As the solubility of CO2 is 40 times that of CH4 (pH 7.0 and
As expected, when the OLR was increased, the degradation rate 35 °C), more CO2 will be discharged when the pressure decreases.
of VS decreased, which may be explained by the shorter HRT of or- Hayes et al. (1990) found that the pH and gas pressure of the sys-
ganic matter in the reactors. Similar results were reported by Bu- tem were the main factors that influenced gas composition, and
joczek et al. (2000). Resch et al. (2006) suggested that HRT > 30 d the methane content of two systems with a pressure difference
would be beneficial for obtaining a high level of degradation and of 70 kPa deviated by 5%. In the present study, there was no obvi-
gas conversion of organic waste, which contains large amounts of ous difference in the methane content of R (101 kPa) and RPC
long-chain fatty acids and proteins. It was observed by Rincón (65.8 kPa).

Phase A Phase B Phase C

100 5000

95 4500
4000
90
3500
R-VS
85
VFA (mg/L)

RPC-VS 3000
VS (%)

80 R-VFA 2500
RPC-VFA 2000
75
1500
70
1000
65 500
60 0
9 17 32 42 52 62 72 82 92 102 112 122 132 142
Time (d)

Fig. 4. Variation in VS degradation and VFAs at 101 kPa (R) and 65.8 kPa (RPC).

Phase A Phase B Phase C

3000
R N-NH3 RPC N-NH3 8000
2500 R alkalinity RPC alkalinity 7000
Alkalinity (mg·L -1)

6000
N-NH3 (mg·L-1)

2000
5000
1500 4000

1000 3000

2000
500
1000
0 0
9 17 32 42 52 62 72 82 92 102 112 122 132 142
Time (d)

Fig. 5. Variation in alkalinity and ammonia at 101 kPa (R) and 65.8 kPa (RPC).
 J. Jiang et al. / Bioresource Technology 101 (2010) 6361–6367 6367

3.5. Other parameters Bouallagui, H., Touhami, Y., Cheikh, R.B., Hamdi, M., 2005. Bioreactor performance in
anaerobic digestion of fruit and vegetable wastes. Process Biochemistry 40,
989–995.
Two additional parameters, alkalinity and ammonia concentra- Bujoczek, G., Oleszkiewicz, J., Sparling, R., Cenkowski, S., 2000. High solid anaerobic
tion, were monitored in the experiment (Fig. 5). digestion of chicken manure. Journal of Agricultural Engineering Research 76,
Alkalinities of the two systems, which was affected by bicarbon- 51–60.
Carneiro, T.F., Pérez, M., Romero, L.I., 2008a. Thermophilic anaerobic digestion of
ate and ammonia produced from the degradation of organic waste, source-sorted organic fraction of municipal solid waste. Bioresource
tended to rise during the experiment. The highest alkalinities in R Technology 99, 6763–6770.
and RPC were 6976, and 6498 mg/L, respectively, and were higher Carneiro, T.F., Pérez, M., Romero, L.I., 2008b. Anaerobic digestion of municipal solid
wastes: dry thermophilic performance. Bioresource Technology 99, 8180–8184.
in R than in RPC. Rittmann and McCarty (2002) pointed out that as Castillo, E.F., Cristancho, D.E., Arellano, V., 2006. Study of the operational conditions
alkalinity increases, the pH in the reactor becomes higher. How- for anaerobic digestion of urban solid wastes. Waste Management 26, 546–556.
ever, pH is not sensitive to increases in alkalinity after the pH Chen, Z.Y., Cai, C.D., Shi, D.W., 2009. Effect of different temperature on anaerobic
digestion of livestock manures. Guizhou Agricultural Sciences 37, 148–151.
and alkalinity reach about 7.4 and 5000 mg/L (as CaCO3), respec- Dinamarca, S., Aroca, G., Chamy, R., Guerrero, L., 2003. The influence of pH in the
tively. Therefore, the variation in pH in the later stages of the hydrolytic stage of anaerobic digestion of the organic fraction of urban solid
experiment did not have a strong association with alkalinity. waste. Water Science and Technology 48, 249–254.
Edelmann, W., Baier, U., Engeli, H., 2005. Environmental aspects of the anaerobic
An increasing trend was also found in ammonia concentration. digestion of the organic fraction of municipal solid wastes and of solid
The concentration curves of the two reactors were almost the same agricultural wastes. Water Science and Technology 52, 203–208.
on days 0–70, with both tending to increase slowly. The ammonia Fricke, K., Santen, H., Wallmann, R., Huttner, A., Dichtl, N., 2007. Operating problems
in anaerobic digestion plants resulting from nitrogen in MSW. Waste
concentrations of R and RPC rose to 616.73 and 601.86 mg/L at day
Management 27, 30–43.
67, respectively. Although the C/N of the artificial organic waste Gallert, C., Winter, J., 1997. Mesophilic and thermophilic anaerobic digestion of
was about 25, i.e., within the appropriate range for AD, a certain source-sorted organic wastes: effect of ammonia on glucose degradation and
proportion of C was locked in the lignin of vegetables, paper, and methane production. Applied Microbiology and Biotechnology 48, 405–410.
Hayes, T.D., Isaacson, H.R., Pfeffer, J.T., Liu, Y.M., 1990. In situ methane enrichment
grass, making decomposition difficult and thus resulted in ammo- in anaerobic digestion. Biotechnology and Bioengineering 35, 73–86.
nia accumulation in both systems. With the change of feedstock, Jiang, J.G., 2008. The physical and chemical characteristics of solid wastes. In: Solid
the ammonia concentrations of R and RPC mushroomed after day Waste Disposal and Resource, first ed. Chemical Industry Press, Beijing, pp. 26–
37.
71, with averages on days 71–125 of 1214.63 mg/L for R and Kapp, H., 1992. The effect of higher pressures on the mesophilic and the
1101.55 mg/L for RPC. In phase C, the ammonia concentration of thermophilic anaerobic sludge digestion. In: Proceedings of International
R rose to 1544.59 mg/L, and that of RPC was maintained at Symposium on Anaerobic Digestion of Solid Waste, Venice, 14–17 April 1992.
pp. 85–94.
1175.78 mg/L. At the higher loading rate, ammonia accumulation Kim, J.K., Oh, B.R., Chun, Y.M., Kim, S.W., 2006. Effects of temperature and hydraulic
was not apparent in the RPC system but was in the R system. Kapp retention time on anaerobic digestion of food waste. Journal of Bioscience and
(1992) reported that ammonia accumulation at high temperature Bioengineering 102, 328–332.
Krzystek, L., Ledakowicz, S., Kahle, H.J., Kaczorek, K., 2001. Degradation of household
can be avoided by increasing pressure in an AD system. In the pres- biowaste in reactors. Journal of Biotechnology 92, 103–112.
ent study, ammonia accumulation in AD at a medium temperature Li, D., Sun, Y.M., Yuan, Z.H., 2009. Effect of solid content on start-up of mesophilic
occurred more readily at atmospheric pressure than at lower middle solid anaerobic digestion for water-sorted organic fraction of municipal
solid waste. The Chinese Journal of Process Engineering 9, 987–992.
pressure.
Liu, C.F., Yuan, X.Z., Zeng, G.M., Li, W.W., Li, J., 2008. Prediction of methane yield at
optimum pH for anaerobic digestion of organic fraction of municipal solid
4. Conclusions waste. Bioresource Technology 99, 882–888.
Mandal, T., Mandal, N.K., 1997. Comparative study of biogas production from
different waste materials. Energy Conversion and Management 38, 679–683.
The experimental results demonstrated the feasibility of apply- Ministry of Environmental Protection of the People’s Republic of China, 2002.
ing AD to treat OFMSW in high-altitude areas. Gas production was Standard Methods for Water and Wasterwater Monitoring and Analysis, forth
ed. China Environmental Science Press, Beijing.
comparatively lower in the system under 65.8 kPa than that under Resch, C., Grasmug, M., Smeets, W., Braun, R., Kirchmayr, R., 2006. Optimised
101 kPa. At a low loading rate and long HRT, pressure had no obvi- anaerobic treatment of house-sorted biodegradable waste and slaughterhouse
ous effect on the gas production of the AD system. However, at a waste in a high loaded half technical scale digester. Water Science and
Technology 53, 213–221.
high loading rate, the lower pressure enhanced the pH buffering
Rincón, B., Borja, R., González, J.M., Portillo, M.C., Jiménez, C.S., 2008. Influence of
capacity and had a positive effect on the stability of the AD system. organic loading rate and hydraulic retention time on the performance, stability
Therefore, an AD system constructed in the low atmospheric pres- and microbial communities of one-stage anaerobic digestion of two-phase olive
mill solid residue. Biochemical Engineering Journal 40, 253–261.
sure environment of the Qinghai-Tibetan Plateau can sustain a
Rittmann, B.E., McCarty, P.L., 2002. Environmental Biotechnology: Principles and
higher loading rate. To clarify the mechanism of pressure effects Applications. McGraw-Hill, New York, USA.
on AD, further studies of two-phase AD systems under different Salminen, E.A., Rintala, J.A., 2002. Semi-continuous anaerobic digestion of solid
pressures are recommended. poultry slaughterhouse waste: effect of hydraulic retention time and loading.
Water Resources 36, 3175–3182.
Stroot, P.G., McMahon, K.D., Mackie, R.I., Raskin, L., 2001. Anaerobic codigestion of
Acknowledgements municipal solid waste and biosolids under various mixing conditions—I.
Digester performance. Water Resources 35, 1804–1816.
Sung, S., Liu, T., 2003. Ammonia inhibition on thermophilic anaerobic digestion.
We thank the Ministry of Science and Technology, China, for Chemosphere 53, 43–52.
funding this study (2006DFB93580), and the Tibet Energy Re- Tafdrup, S., 1995. Viable energy production and waste recycling from anaerobic
sources Research and Demonstration Center for support during digestion of manure and other biomass materials. Biomass and Bioenergy 9,
303–314.
the course of the study. Vavilin, V.A., Vasiliev, V.B., Rytov, S.V., 1995. Modelling of gas pressure effects on
anaerobic digestion. Bioresource Technology 52, 25–32.
References Zhang, B., Zhang, L.L., Zhang, S.C., Shi, H.Z., Cai, W.M., 2005a. The influence of pH on
hydrolysis and acidogenesis of kitchen wastes in two-phase anaerobic
digestion. Environmental Technology 26, 329–339.
Álvarez, R., Villca, S., Liden, G., 2006. Biogas production from llama and cow manure
Zhang, B., He, Z.G., Zhang, L.L., Xu, J.B., Shi, H.Z., Cai, W.M., 2005b. Anaerobic
at high altitude. Biomass and Bioenergy 30, 66–75.
digestion of kitchen wastes in a single-phased anaerobic sequencing batch
Álvarez, J.A., Otero, L., Lema, J.M., 2010. A methodology for optimising feed
reactor (ASBR) with gas-phased absorb of CO2. Journal of Environmental
composition for anaerobic co-digestion of agro-industrial wastes. Bioresource
Sciences 17, 249–255.
Technology 101, 1153–1158.