Professional Documents
Culture Documents
Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech
a r t i c l e i n f o a b s t r a c t
Article history: Assessing contemporary anaerobic biotechnologies requires proofs on reliable performance in terms of
Received 22 February 2011 renewable bioenergy recovery such as methane (CH4) production rate, CH4 yield while removing volatile
Received in revised form 16 April 2011 solid (VS) effectively. This study, therefore, aims to evaluate temperature-phased anaerobic sequencing
Accepted 27 April 2011
batch reactor (TPASBR) system that is a promising approach for the sustainable treatment of organic frac-
Available online 30 April 2011
tion of municipal solid wastes (OFMSW). TPASBR system is compared with a conventional system, mes-
ophilic two-stage anaerobic sequencing batch reactor system, which differs in operating temperature of
Keywords:
1st-stage. Results demonstrate that TPASBR system can obtain 44% VS removal from co-substrate of sew-
Temperature-phased anaerobic digestion
Co-digestion
age sludge and food waste while producing 1.2 m3CH4/m3system/d (0.2 m3CH4/kgVSadded) at organic load-
Anaerobic sequencing batch reactor ing rate of 6.1 gVS/L/d through the synergy of sequencing-batch operation, co-digestion, and
Sewage sludge temperature-phasing. Consequently, the rapid and balanced anaerobic metabolism at thermophilic stage
Food waste makes TPASBR system to afford high organic loading rate showing superior performance on OFMSW
stabilization.
Ó 2011 Elsevier Ltd. All rights reserved.
0960-8524/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2011.04.088
H.-W. Kim et al. / Bioresource Technology 102 (2011) 7272–7279 7273
production of renewable bioenergy (i.e., CH4) and a maximized sta- As a control system for comparison, a mesophilic two-stage
bilization efficiency of OFMSW because it integrates both advanta- anaerobic sequencing batch reactor (MTSASBR) system was oper-
ges of rapid thermophilic and stable mesophilic anaerobic ated simultaneously. We compare its co-digestion performance
digestion simultaneously (Kim et al., 2002; Lv et al., 2010). with the TPASBR system applying similar OLR. The system is sim-
Although higher energy requirement for the maintenance of ther- ilarly consisted of two mesophilic ASBRs and connected in series.
mophilic conditions was pointed out (Han et al., 1997), from the The first stage as well as the second stage is operated at the con-
kinetic point of view, this staged reactor system obtains a rapid stant temperature room (35 °C) using the same sequence with
substrate conversion in the 1st-stage while effluent polishing oc- the TPASBR system.
curs in 2nd-stages by establishing a plug-flow pattern thus this
system results in a very low effluent concentration of VFA and a 2.2. Operating procedures
high degree of biomass retention (van Lier et al., 2001).
If TPAD is combined with co-digestion and sequencing batch Table 1 shows the mode of sequencing batch operation for
operation, the systemic integration may yield extra dividends TPASBR and MTSASBR systems. In this operation, one cycle of each
simultaneously such as enhanced biodegradation of poorly digest- reactor consisted of four sequences: fill (1 h), react (17 h), settle
ible OFMSW, decreased VFA concentrations in effluent even at high (5 h), and draw (1 h). The 24 h cycle lasted continuously until the
OLR, increased system stability, and reduced fecal coliform levels. end of experiments and all the sequence were managed by electri-
Relevant literatures which have investigated this systemic integra- cal on/off switching timers managing controllers of peristaltic
tion are scarce. pumps (Cole-Parmer, USA).
In this study, therefore, long-term continuous flow experiments
combining a temperature-phased anaerobic sequencing batch
2.3. Inoculums
reactor (TPASBR) system with co-digestion strategy are performed
with various OLR to investigate the performance improvements on
Inoculation was carried out for both TPASBR and MTSASBR sys-
the CH4 recovery and stabilization efficiency of co-substrate made
tems using a seed sludge taken from a conventional egg-shaped
of sewage sludge and food waste. As a control experiment, a mes-
anaerobic digester in the D city WWTP. About 80% of digester
ophilic two-stage anaerobic sequencing batch reactor (MTSASBR)
working volume was filled with inoculums and sparged with N2
system are also operated to provide a quantitative comparison
before sealing. The pH, alkalinity, volatile suspended solids (VSS)
based on individual performance parameter. The main goals of this
and VSS/TSS of the inoculum were 6.5–7.0, 567–611 mg/L as
research are to verify excellences in unique combination of TPAD,
CaCO3, 14.8 g/L and 0.75, respectively.
sequencing-batch operation, co-digestion and to discuss the appli-
cability of TPASBR system for maximum stabilization of sewage
sludge and food waste with a high-rate production of a renewable 2.4. Feedstock preparation
bioenergy under the wide range of OLR.
Sewage sludge was sampled from a thickener at the same D city
WWTP, Korea. Average total solids (TS) concentration, volatile sol-
2. Methods ids (VS) concentrations, VS/TS, and moisture content during overall
operational period were about 25 g/L, 15 g/L, 0.61, and 98%,
2.1. Experimental set-up respectively.
Food waste was selectively collected from a dining hall of an
A TPASBR system was operated continuously as increasing OLR academic institute in D city, Korea. The food waste was addition-
(6.5 gVS/L/d) by using an optimal mixture of sewage sludge and ally crushed by an electrical blender to promote homogeneity of
food waste suggested by Kim et al. (2007). The system was con- substrate as well as disintegration of particulate organics. Average
sisted of two ASBRs, two gas collectors, a control panel with timers total solids concentration, volatile solids concentrations, VS/TS, and
and peristaltic pumps. Each reactor was 4.0 L in working volume moisture content of food waste were 229 g/L, 218 g/L, 0.95, and
and the first thermophilic stage ASBR was followed by the second 81%, respectively.
mesophilic stage ASBR in series. The thermophilic stage was con- A previous study, which tested similar source of co-substrates
trolled to 55 °C using a water bath circulator and a built-in water (Kim et al., 2007), revealed that optimal co-substrate condition is
jacket while the mesophilic stage was operated at 35 °C. The oper- attained when sewage sludge and food waste were mixed with a
ating sequence was maintained by controlling three peristaltic VS ratio of 60:40. Due to seasonal and operational variation of sub-
pumps with three timers on a control panel. The system was strates taken from WWTP, solid concentration of sewage sludge
placed in a constant temperature room (35 °C). and food waste were adjusted to target concentrations by a manual
Table 1
Operational sequence of TPASBR and MTSASBR system.
gravity settling or a dilution to meet the required sewage sludge tively, in a sequencing-batch mode with increasing OLR step-wisely
and food waste ratio and overall concentration as a designated at steady-state. Fig. 1 shows the dynamics of CH4 production rates
co-substrate. Prepared co-substrates filtered through a stainless (MPR) for each reactor of TPASBR and MTSASBR systems treating
steel sieve (sieve openings of 2.00 mm equivalent to US Mesh No. co-substrates of food waste and sewage sludge. In the TPASBR sys-
10) to screen deleterious particles and stored at 4 °C. tem (Fig. 1a), the MPR of thermophilic stage linearly increases to
2.3 l CH4/L/d as OLR increases up to 6.1 gVS/L/d. The CH4 content
2.5. Startup and operating procedures of the biogas was measured to be 57–68% at standard temperature
and pressure condition as dry gas basis. As can be seen, most biogas
To cultivate thermophilic methanogens as a dominant species was produced at thermophilic stage of TPASBR system, which indi-
in thermophilic stage of TPASBR, the seed sludge was directly accli- cates the thermophilic methanogens are conducting rapid and dom-
mated at 55 °C. During about 1 week, we checked the increase of inant methanogenesis. Effluent quality parameters for first stage
biogas production in the collector and CH4 content (>50%) as an thermophilic reactor such as pH, total alkalinity, and total VFA con-
acclimation period. Then the systems were operated at a HRT of centrations as soluble COD were stable as 7.0–7.8, 2600–3200 mg/L
30 days by feeding co-substrates. Starting-up OLR was 0.5 gVS/L/ as CaCO3, and 710–1280 mg COD/L, respectively.
d for both TPASBR and MTSASBR. We continuously monitored When OLR was further increased to 6.5 gVS/L/d, however, a
the systems more than 1 month until the CH4 contents of biogas drastic pH drop (pH < 4.7) was observed and MPR decreased at
and CH4 production rate (MPR) was stabilized. We then increased the same time. This reveals that maximum available OLR for
the OLR to next step by either reducing HRT (from 30 to 7 d) or TPASBR system is 6.1 gVS/L/d. The accumulation of VFAs must have
increasing substrate concentration (from 23 to 46 g VS/L) when induced instant pH drop as fast hydrolysis/acidogenesis of readily
the system reached a global steady state. biodegradable organics overwhelms methanogenesis, and corre-
sponding low pH inhibits the activity of thermophilic methanogens
eventually.
2.6. Specific methanogenic activity and biochemical methane potential
In the case of MTSASBR (Fig. 1b), the 1st-stage mesophilic reac-
tor shows inferior performance to that of TPASBR. When the OLR is
The specific methanogenic activity (SMA) tests on each stage’s
over 3.5 gVS/L/d, MPR starts to decrease by a pH drop. The maxi-
biomass of both systems were performed in serum vials (125 mL)
mum MPR of 1st-stage was limited to 0.9 l CH4/L/d at OLR of
following a method of Shin et al. (2001). Biomass samples were ta-
3.5 gVS/L/d showing stable effluent quality. pH, total alkalinity,
ken at each OLR for the SMA test. The SMA of the biomass was
and total VFA concentrations as soluble COD were 7.3–7.7, 2000–
measured for a substrate of 2000 mg/L as chemical oxygen demand
3600 mg/L as CaCO3, and 298–419 mg COD/L, respectively. Slow
(COD); in this study, acetate, propionate, and butyrate were used
hydrolysis/acidogenesis might have been rate-limiting though
individually as precursor substrates of CH4.
effluent VFA concentration of MTSASBR is comparably lower than
Biochemical methane potentials (BMP) was also determined for
that of TPASBR.
fresh co-substrates prepared following a method suggested by Kim
In one hand, at OLR or 4 gVS/L/d, the methanogenesis of 2nd-
et al. (2007) modifying procedures of Owen et al. (1979) and Owens
stage reactor seems to be activated as a complementary effect of
and Chynoweth (1993) for sewage sludge and food waste. Variabil-
deteriorating quality in the 1st-stage effluent, which is an influent
ity may have originated from small differences in acclimation to the
of 2nd-stage. Therefore, the MTSASBR system was turned into con-
new conditions, inoculum size, carryover of nutrients with the
ventional two-phase ASBR system, which is consisted of an acido-
inoculums thus substrate controls were tested to exclude extra
genesis reactor (1st-stage reactor) followed by a methanogenic
BMP in inoculum. Average BMPs of TPASBR system for co-sub-
reactor (2nd-stage reactor). This transition, however, did not last
strates were 0.32 (thermophilic biomass) and 0.23 l CH4/gVS (mes-
longer because it was apparent that the OLR eventually caused a
ophilic biomass), respectively, while those of MTSASBR system
pH drop for 2nd-stage as well unless 1st-stage significantly re-
were 0.24 (1st-stage) and 0.18 l CH4/gVS (2nd-stage), respectively.
duced organics. In order to maintain steady-state, therefore, we
should relieve pH inhibition from 2nd-stage afterwards. Because
2.7. Analyses the overload in 1st-stage not only produced high amount of VFAs,
but also directly transported incompletely acidified organic solids,
The contents of CH4 and CO2 in the biogas were analyzed by a gas the addition of alkalinity was inevitably necessary to ensure stable
chromatograph (GC, Gow Mac series 580) equipped with a thermal methanogenesis keeping pH neutral. Thus we monitored pH of
conductivity detector (TCD) and a 2 m 2 mm stainless-steel col- 2nd-stage and controlled it to be over 7 by using a pH controller.
umn packed with Porapak Q (80/100 mesh). During the experi- In such a case, MTSASBR system could be operated even at high
ments, 1.5 mL of liquid sample was collected at a proper timing OLR from 3.5 to 5.8 gVS/L/d though MPR is maintained only at
with a syringe. The sample was immediately filtered through 0.49–0.66 l CH4/L/d after the transition.
0.45 lm cellulose nitrate membrane filters (Whatman, USA) and Fig. 1c compares system MPRs obtained from both systems
then stored at 4 °C for further analysis. For the determination of indi- according to applied OLR on the basis of total volume. The TPASBR
vidual VFA concentrations, a high performance liquid chromato- system show approximately linear increases as OLR increases but
graph (HPLC; Spectra Physics P2000) was used with an Aminex the MTSASBR system shows relatively consistent trends at most
HPX-87H (300 mm 7.8 mm) column and a UV (210 nm) detector. of the OLR tested. Overall system MPR of MTSASBR ranges between
Other analytical methods such as total solids (TS), VS, total sus- 0.2 and 0.5 l CH4/L/d while that of TPASBR significantly increases
pended solid (TSS), VSS, and COD were determined according to from 0.2 to 1.2 l CH4/L/d. The difference in system MPRs enlarges
Standard Methods (American Public Health Association et al., 2005). as OLR increases. At OLR of 5.5 gVS/L/d, maximum available OLR
for MTSASBR, TPASBR shows about 3-fold higher system MPR
3. Results and discussion (1.1 l CH4/L/d) than MTSASBR (0.4 l CH4/L/d).
To verify the long-term performance of co-digestion, TPASBR and Fig. 2 compares the removal efficiencies of volatile solids at
MTSASBR systems were operated during 540 and 460 days, respec- each OLR during long-term continuous operation of TPASBR and
H.-W. Kim et al. / Bioresource Technology 102 (2011) 7272–7279 7275
Fig. 1. Variations of MPRs for: (a) TPASBR system, (b) MTSASBR system and (c) overall comparison of system MPRs according to OLR.
Fig. 3. Comparisons on: (a) overall CH4 yields, CH4 conversion efficiencies: (b) TPASBR system and (c) MTSASBR system.
Fig. 5. Specific methanogenic activities of biomasses from: (a) thermophilic of TPASBR system, (b) mesophilic of TPASBR system, (c) 1st-stage of MTSASBR system, and (d)
2nd-stage of MTSASBR system.
Fig. 6 illustrates variation of ammonium and organic nitrogen Table 2 compares overall performance data of TPASBR and
contents in effluents from each stage of TPASBR and MTSASBR MTSASBR systems together with similar parameter values in other
7278 H.-W. Kim et al. / Bioresource Technology 102 (2011) 7272–7279
Table 2
Comparison of overall system performance data obtained in MTSASBR and TPASBR system and those in other literatures.
Song, Y.C., Kwon, S.J., Woo, J.H., 2004. Mesophilic and thermophilic temperature co- van Lier, J.B., van der Zee, F.P., Tan, N.C.G., Rebac, S., Kleerebezem, R., 2001. Advances
phase anaerobic digestion compared with single-stage mesophilic and in high-rate anaerobic treatment: staging of reactor systems. Water Science and
thermophilic digestion of sewage sludge. Water Research 38 (7), 1653–1662. Technology 44 (8), 15–25.
Sung, S.W., Dague, R.R., 1995. Laboratory studies on the anaerobic sequencing batch vanLier, J.B., 1996. Limitations of thermophilic anaerobic wastewater treatment and
reactor. Water Environment Research 67 (3), 294–301. the consequences for process design. Antonie Van Leeuwenhoek International
Sung, S.W., Santha, H., 2003. Performance of temperature-phased anaerobic Journal of General and Molecular Microbiology 69 (1), 1–14.
digestion (TPAD) system treating dairy cattle wastes. Water Research 37 (7), Veeken, A., Hamelers, B., 1999. Effect of temperature on hydrolysis rates of selected
1628–1636. biowaste components. Bioresource Technology 69 (3), 249–254.