Bioresource Technology 112 (2012) 67–74

Contents lists available at SciVerse ScienceDirect

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

Co-digestion of polylactide and kitchen garbage in hyperthermophilic

and thermophilic continuous anaerobic process
Feng Wang a,⇑, Taira Hidaka a, Hiroshi Tsuno a, Jun Tsubota b
a
Department of Environmental Engineering, Kyoto University, Kyoto-Daigaku-Katsura, Nishikyo-ku, Kyoto 615-8540, Japan
b
Energy Engineering Department, Osaka Gas Co., Ltd., 11-61, Torishima 5-Chome, Konohana-ku, Osaka 554-0051, Japan

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

Article history: Two series of two-phase anaerobic systems, consisting of a hyperthermophilic (80 °C) reactor and a
Received 1 December 2011 thermophilic (55 °C) reactor, fed with a mixture of kitchen garbage (KG) and polylactide (PLA), was
Received in revised form 3 February 2012 compared with a single-phase thermophilic reactor for the overall performance. The result indicated that
Accepted 13 February 2012
ammonia addition under hyperthermophilic condition promoted the transformation of PLA particles to
Available online 22 February 2012
lactic acid. The systems with hyperthermophilic treatment had advantages on PLA transformation and
methane conversion ratio to the control system. Under the organic loading rate (OLR) of 10.3 g COD/
Keywords:
(L day), the PLA transformation ratios of the two-phase systems were 82.0% and 85.2%, respectively,
Anaerobic co-digestion
Polylactide
higher than that of the control system (63.5%). The methane conversion ratios of the two-phase systems
Kitchen garbage were 82.9% and 80.8%, respectively, higher than 70.1% of the control system. The microbial community
Hyperthermophilic condition analysis indicated that hyperthermophilic treatment is easily installed to traditional thermophilic anaer-
Ammonia obic digestion plants without inoculation of special bacteria.
Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction of the AD process. Compared with the generally operated

mesophilic (37 °C) processes, thermophilic (55 °C) and hyper-
Kitchen garbage (KG) is an organic solid waste suitable for thermophilic (over 55 °C) anaerobic digestion processes have the
anaerobic digestion (AD) treatment (Kim et al., 2003). A large advantages of effective organic particles solubilization and higher
amount of plastic materials used as boxes and bags are typically biogas production (Nielsen and Petersen, 2000; Scherer et al.,
commingled with the collected KG, which makes it necessary to re- 2000). Two-phase systems employing hyperthermophilic
move these plastics before the AD process to avoid the adverse ef- (70–80 °C) and thermophilic (55 °C) AD process have been devel-
fect on the treatment performance due to the non-biodegradability oped for KG and sewage sludge treatment (Lee et al., 2008,
and mechanical troubles in mixers and pumps of the reactors. The 2009). The hyperthermophilic AD treatment for PLA is expected
extra separation cost weakens the economic benefit of the AD pro- to promote the hydrolysis and biodegradation performance, but
cess; an option is to replace the non-biodegradable plastics with generally applicable and effective anaerobic digestion methods
polylactide (PLA), one of the well-known biodegradable plastics for PLA have not been proposed yet.
(Armentano et al., 2010; Perepelkin, 2002), and treat PLA and KG Aqueous ammonia solutions have been widely used as reaction
simultaneously. An increase of biogas production by the PLA deg- media for the degradation of polymer materials, such as poly-
radation is anticipated. Consequently, an effective anaerobic diges- ethylene naphthalate, poly-ethylene terephthalate, polycarbonate
tion system for co-digestion of PLA and KG is important to be and poly-hexamethylene carbonate. These degradation processes
developed on the basis of biodegradation characteristics of PLA. were accomplished under hydrothermal or supercritical conditions
For the last decade, numerous studies have focused on the PLA of over 120 °C and over 10 MPa (Arai et al., 2010; Zenda and
degradation under various treatment conditions including hydro- Funazukuri, 2008). Supposing that PLA is able to be degraded via
lysis, hydrothermal and compost. Previous research results indi- enzymatic reactions under temperature and pressure acceptable
cated that temperature is one of the key factors affecting the PLA for AD process with existence of ammonia, the hydrolysis product
biodegradation since the rate of PLA degradation increased with of protein, PLA is expected to be transformed to methane easily.
temperature (Copinet et al., 2004; Dunne et al., 2000; Ghorpade Wang et al. (2011) performed batch experiments and demon-
et al., 2001). Higher temperature also promotes the performance strated that PLA was hydrolyzed under hyperthermophilic pre-
treatment with ammonia addition and converted to methane by
⇑ Corresponding author. Tel.: +81 75 383 3352; fax: +81 75 383 3351. methanogens under the following thermophilic anaerobic condi-
E-mail address: wang.feng.7s@kyoto-u.ac.jp (F. Wang). tions. However, few researches focusing on PLA biodegradation

0960-8524/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2012.02.064
68 F. Wang et al. / Bioresource Technology 112 (2012) 67–74

with ammonia under continuous anaerobic operation have been The operational conditions are summarized in Table 1. The
published to date and the optimum configuration of the system initial organic loading rate (OLR) of the three systems was
including hyperthermophilic treatment is poorly understood. 3 g COD/(L day). After the systems got stable for biogas production,
In this study, three types of anaerobic digestion systems were the OLRs were increased to 4.5 and 6.8 g COD/(L day) gradually.
operated continuously for 592 days with a co-substrate of PLA The solids of the discharged sludge from each thermophilic reactor
and KG to evaluate the PLA biodegradability. The overall treatment was returned after it was concentrated to a high density with
performances of these systems were compared in terms of PLA centrifuge under 3000 rpm for 10 min until the end of Run 3. In
degradation, methane conversion, and dewatering property. Run 4, the solid return operation was terminated while the OLR
was unchanged and the biogas production decreased obviously.
Consequently, the HRT of the systems was increased in Run 5. After
2. Methods the biogas production was recovered, the HRT was decreased in
Run 6 and the PLA 2 was used as substrate instead of PLA 1. In
2.1. Reactor configuration and operation Run 7, the overall OLR was increased to 10.3 COD/(L day). In Run
b8 to Run b9 of S1, and in Run c2 to Run c9 of S2, NH4Cl was added
One control reactor (M0) operated under a thermophilic (55 °C) into the reactors to evaluate the promotion effect of ammonia on
condition, and two series of two-phase systems were operated PLA degradation under hyperthermophilic condition. S1 was oper-
continuously fed with a mixture of PLA plastic and KG as shown ated without pH control while pH in S2 was maintained at 7.5 by
in Fig. 1. Both of the two-phase systems consisted of a thermophilic automatic titration of 10 N KOH in Run c2 to Run c9 to promote
reactor (M1, M2) and a hyperthermophilic (80 °C) reactor (S1, S2). In the PLA degradation as pH drops by degradation of PLA to lactic
the post-dissolution system of M1 + S1, the co-substrate was di- acid, and it was indicated that higher pH of aqueous solution was
luted by the liquor from the hyperthermophilic reactor S1 and then favorable for the PLA hydrolysis (Karst et al., 2008; Tsuji and Ikar-
fed into the reactor M1, and a given part of the mixed liquor from ashi, 2004).
M1 was fed into S1. In the pre-dissolution system of S2 + M2, the co-
substrate was diluted by the mixed liquor from the reactor M2 and 2.2. Characteristics of PLA and KG
then fed into the reactor S2. The liquor discharged from S2 was fed
into M2. The temperature was controlled by setting the reactors in Two kinds of plastic bags (PLA 1 and PLA 2) were used. PLA 1
oil bath. In each reactor, a steel stirrer was used for stirring at was composed entirely of polylactide, while PLA 2 had a polylac-
200 rpm. Withdrawing and feeding were conducted once a day. tide content of 70%. Both of the two plastics were 0.1 mm thick
The seed sludge was obtained from a continuously operated anaer- and were cut into small pieces (2  2 mm) using a shredder
obic digestion system (Lee et al., 2009). (M-450Cs, Fellowes, Japan) before conducting the experiments.

(1) Control system

Q Q
M0
C inf, M0 C enf, M0

(2) Post-dissolution system (M 1+S 1)

C inf, (M1+S1) C inf, M1 C enf, M1 C enf, (M1+S1)

M1
Q (1+r)•Q (1+r)•Q Q

C enf, S1 C inf, S1
S1
r•Q r•Q
Cenf, M1 = Cenf, (M1+S1) = Cinf, S1
Cinf, M1 = [Cinf, (M1+S1) • Q + Cenf, S1 • r • Q] / [(1+r) • Q]

(3) Pre-dissolution system (S 2+M 2)

C inf, (S2+M2) C inf, S2 C enf, S2 C inf,M2 C enf, M2 C enf, (S2+M2)

S2 M2
Q (1+r)•Q (1+r)•Q (1+r)•Q Q

r•Q

Cenf, S2 = Cinf, M2
Cenf, M2 = Cenf, (S2+M2)
Cinf, S2 = [Cinf, (S2+M2) • Q + Cenf, M2 • r • Q] / [(1+r) • Q]

Fig. 1. Configuration of three systems in the continuous operation. S1, S2: Hyperthermophilic reactors; M0, M1, M2: Thermophilic reactors; Q: Substrate flow rate;
r: recirculation ratio; Cinf, X and Cenf, X: PLA concentration in the influent and the effluent of reactor or system X.
 F. Wang et al. / Bioresource Technology 112 (2012) 67–74 69

Table 1
Operation condition.

Period 0–16 17–170 171–213 214–325 326–464 465–524 525–553 554–568 569–592
Control Run a1 Run a2 Run a3 Run a4 Run a5 Run a6 Run a7
HRT (d) 25 25 25 25 41 30 30
OLR (g COD/(L day)) 3 4.5 6.8 6.8 6.7 6.7 10.3
(M1 + S1) Run b1 Run b2 Run b3 Run b4 Run b5 Run b6 Run b7 Run b8 Run b9
Overall system HRT (day) 25 25 25 25 41 41 30 30 30
OLR (g COD/(L day)) 3 4.5 6.8 6.8 6.7 6.7 10.3 10.3 10.3
ra 1.5 1.5 1.5 1.5 1 1 0.5 0.5 0.5
M1 HRT (d)b 9.4 9.4 9.4 9.4 20 20 19.7 19.7 19.7
S1 HRT (day)b 1 1 1 1 1 1 1 1 1
NH4Cl addition (g N/L) No No No No No No No 2 3
(S2 + M2) Run c1 Run c2 Run c3 Run c4 Run c5 Run c6 Run c7 Run c8 Run c9
Overall system HRT (d) 25 25 25 25 41 41 30 30 30
OLR (g COD/(L day)) 3 4.5 6.8 6.8 6.7 6.7 10.3 10.3 10.3
ra 1.5 1.5 1.5 1.5 1 1 0.5 0.5 0.5
M2 HRT (day)b 9 9 9 9 19.5 19.5 19 19 19
S2 HRT(day)a 1 1 1 1 1 1 1 1 1
pH control No 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5
NH4Cl addition (g N/L) No 1.5 1.5 1.5 1.5 1.5 1.5 2 3
a
r is defined as the ratio of the recirculation flow rate from S1 or M2 to the influent flow rate of the fresh substrate.
b
HRT for single pass.

The COD and VS of the two types of plastics were 1.38 ± 0.07 (PLA n-butyrate (n-HBu), iso-valerate (i-HVa) and n-valerate (n-HVa))
1) and 1.40 ± 0.09 (PLA 2) g COD/g, and 0.99 ± 0.02 (PLA 1) and were measured using HPLC (CDD-10Avp, Shimadzu, Japan) with
0.98 ± 0.01 (PLA 2) g VS/g. The artificial KG consisted of 14 types the column of Shim-pack SCR-102H (Shimadzu, Japan), and soluble
of food items (Lee et al., 2008, 2009) on the basis of a survey con- phase of 5 mM p-toluenesulfonic acid monohydrate under the
ducted in Tokyo Metropolitan City, Japan (Tanigawa et al., 1997). column temperature of 43 °C. The produced gas was collected in
The average COD value of the raw KG was 230 g/L and in Runs 1 a gasbag and was analyzed for composition and volume using
to 4, it was diluted with tap water and then mixed with PLA based gas analyzers (GC-14B with thermal conductivity detector, and
on a COD ratio of 4:1. In other runs, the raw KG without dilution CGT-7000, Shimadzu, Japan). The column of GC-14B was SHINCAR-
was mixed with PLA based on the same COD ratio. The character- BON ST and carrier-gas was helium. In Run a3 to a7, b3 to b9 and c3
istics of the substrate in each run are summarized in Table 2. to c9, the dewatering characteristic of the sludge was evaluated
using the capillary suction time (CST) test (APHA and WEF, 1998).
2.3. Chemical analysis
2.4. Microbial analysis
Once every 2 or 3 days, sample was taken from the effluent of
each reactor for chemical analysis. Total solids (TS), volatile solids In Runs a2, b2 and c2, the systems got stable with PLA degrada-
(VS), suspended solids (SS), volatile suspended solids (VSS), total tion after 100 days operation, which indicated the acclimation of
COD (TCOD), soluble COD (SCOD), total ammonia (TAN) and pH the microbes to PLA and sample was taken from M0, M1 and M2
were measured according to the Standard Methods (APHA and reactor for microbial analysis. DNA was extracted using a dneasy
WEF, 1998). Free ammonia (FAN) was calculated from the TAN as tissue kit (Oiagen, Hilden, Germany). A primer set of UNIV519F
follows (Hansen et al., 1998): (50 -CAGCMGCCGCGGTAATWC-30 ; Lane, 1991) and UNIV1406R
8 91 (50 -ACGGGCGGTGTGTRC-30 ; Lane, 1991) was used to amplify
>
< >
= approximately 900 and 700 bp fragments, respectively. DNA
10pH
½FAN ¼ ½TAN  1 þ   ð1Þ extraction, PCR amplification, gel extraction and cloning operation
>
:  0:09018þ2729:92
>
;
TðKÞ were described by Lee et al. (2009). The sequencing of 16S rRNA
10
was conducted by TaKaRa Bio Dragon Genomics Center in Japan.
where T (K) is temperature of Kelvin. Obtained sequence data were compared with similar sequence in
Seven kinds of main organic acids during AD process (lactate National Center for Biotechnology Information data using the
(HLa), acetate (HAc), propionate (HPr), iso-butyrate (i-HBu), BLAST program (Altschul et al., 1990).

Table 2
Mixed substrate characteristics.

Control Run a1 Run a2 Run a3 Run a4 Run a5 Run a6 Run a7

M1 + S1 Run b1 Run b2 Run b3 Run b4 Run b5 Run b6 Run b7 Run b8 Run b9
S2 + M2 Run c1 Run c2 Run c3 Run c4 Run c5 Run c6 Run c7 Run c8 Run c9
TS (g/L) 60.6 90.9 136 216 216 248
VS (g/L) 58.8 78.4 132 210 210 239
COD of mixture (g/L) 75.1 113 169 276 276 310
COD of KG (g/L) 60.1 90.4 135 221 221 248
PLA type 1 1 1 1 2 2
PLA concentration (g/L) 10.9 16.3 24.5 40 27.6 31.0
70 F. Wang et al. / Bioresource Technology 112 (2012) 67–74

3. Results and discussion tion ratio in S1 and S2 was comparable to that of Lee et al. (2008),
who employed a two-phase system consisting of a hyperthermo-
3.1. Hydrolysis in the hyperthermophilic reactors philic (80 °C) reactor and a following thermophilic (55 °C) methane
fermentation reactor to treat the KG with same composition as that
Fig. 2 shows the PCOD dissolution ratio and the PLA transforma- used in this study. HRT of that hyperthermophilic reactor was
tion ratio in the two hyperthermophilic reactors, S1 and S2. The sol- longer than 4 days (Lee et al., 2008), while HRT of S1 and S2 was
ubilization ratio of particulate COD (PCOD) were calculated as 1 day in this study.
follows: Both of the PCOD dissolution ratio and PLA transformation ratio
in the hyperthermophilic reactors were linearly related with the
PCOD solubilization ratio ð%Þ ¼ ½ðPinf  Penf Þ=Pinf   100 ð2Þ ammonia concentration as shown in Runs b7 to b9 and Runs c7
where, Pinf and Penf are the influent and the effluent concentra- to c9. These results accorded with the previous research conducted
tions of PCOD (g/L), respectively. The definition of ‘‘soluble’’ in this with batch experiments (Wang et al., 2011). The pretreatment with
study was any material passing through filter with pore size of ammonia solution can promote the dissolution of biomass waste
1 lm (ADVENTEC). such as lignin and hemicellulose before the subsequent fermenta-
The PLA transformation ratio for each single reactor or system tion (Kurakake et al., 2001; Lee et al., 2010). Aqueous ammonia was
was calculated as follows: used as reaction media for polymer depolymerization under
hydrothermal conditions (Arai et al., 2010; Zenda and Funazukuri,
PLA transformation ratio ð%Þ ¼ ½ðC inf  C enf Þ=C inf   100 ð3Þ 2008). In this study, ammonia was effective for PLA transformation
under the biologically acceptable temperature and pressure. This is
where, Cinf (g/L) and Cenf (g/L) are the PLA concentration in the
a mild reaction condition compared with supercritical or hydro-
influent and the effluent for one single reactor or system as shown
thermal treatment for other polymer materials and the reaction
in the Fig. 1, respectively. Cenf (g/L) was calculated as follows:
rate was higher than that of compost treatment.
C enf ¼ C total lactic  C lactic ð4Þ
where Ctotal lactic (g/L) is the total lactic acid concentration in the
3.2. Effect of ammonia on methane production
effluent after chemical dissolution, and Clactic is the lactic acid con-
centration in the soluble phase. To measure Ctotal lactic, the sample
Fig. 3 shows the ammonia concentration, methane production
(2 mL) and 5 N NaOH (5 mL) were mixed to dissolve all proportions
rate and organic acid concentrations under the OLR of 10.3
of PLA into lactic acid, and the lactic acid concentration in the fil-
g COD/(L day) in the three methane fermentation reactors. The
tered solution was measured by HPLC. This ratio was determined
average TAN in Run a7 of the M0 reactor was 1210 mg N/L and
by our preliminary examination by comparing 0.1–10 N NaOH,
the average methane production rate was 2.7 L CH4/(L day). NH4Cl
and the linier correlation between added PLA and measured lactic
was added into S1 in Runs b8 and b9, and the ammonia concentra-
acid with R2 = 0.996 was obtained.
tion in M1 was correspondingly increased. The average methane
As shown in Fig. 2 (a), the PCOD dissolution ratio in the two
production rate was 2.9, 3.1 and 3.3 L CH4/(L day) in Runs b7, b8
reactors varied in the range of 10.8–15.7% (S1, Runs b1 to b6)
and b9, respectively. The increase in methane production was
and 12.6–16.8% (S2, Runs c1 to c6), respectively. The PCOD dissolu-
attributed to the improvement of PCOD dissolution and PLA trans-
formation caused by ammonia addition in S1 since anaerobic diges-
(a) 30
tion of solid wastes is rate-limited by the hydrolysis step; and
promotion of solubilization can improve the overall system perfor-
PCOD dissolution ratio (%)

S1: y = 0.0071x + 4.2677

25 2 mance (Mata-Alvarez et al., 2000). The ammonia concentration in
R = 0.9997
20 M2 increased from 824 mg N/L in Run c7 to 871 mg N/L in Run
c8. The methane production rate of M2 increased slightly from
15 3.0 CH4/(L day) in Run c7 to 3.1 L CH4/(L day) in Run c8. However,
10 S2: y = 0.0071x + 2.887 the ammonia concentration in Run c9 increased to 2010 mg N/L,
R2 = 0.9997 and the methane production rate decreased to 2.8 L CH4/(L day).
5
This was due to an inhibiting effect of ammonia on methane
0 fermentation (Kadam and Boone, 1996). The average FAN concen-
0 1000 2000 3000 4000 tration in M1 was 148 (Run b7), 209 (Run b8) and 229 (Run b9)
Ammonia concentration (mg N/L) mg N/L, respectively. The FAN in M2 was 270 (Run c7), 274 (Run
(b) 60 c8) and a much higher value of 422 (Run c9) mg N/L. FAN is the
S1: y = 0.0058x + 30.949 active component which causes inhibition (Hansen et al., 1998;
PLA transformation ratio (%)

50 R2 = 0.9832 Sterling et al., 2001). Except in Run c9 (reactor M2), the lactic acid
concentration in the three reactors was lower than 1 g COD/L, indi-
40
cating that the lactic acid generated from KG and PLA hydrolysis
S2: y = 0.0055x + 38.693
30 was transformed to methane gas. In all runs, the total organic acid
R2 = 0.9669 concentrations in M1 were lower than those in the control reactor
20 and M2. An accumulation of acetic acid was observed in M2 and
10 S1 S1 (Run b7-b9)
this was the premier indicator of an ammonia inhibition as ace-
S2 S2 (Run c7-c9) tate-utilizing methanogens are most easily inhibited (Angelidaki
0 and Ahring, 1993). The deterioration of M2 performance corre-
0 1000 2000 3000 4000 sponded with the fact that ammonia was a significant inhibitor
Ammonia concentration (mg N/L) of methane production (Sung and Santha, 2003). Addition of
ammonia in the hyperthermophilic reactors can promote the PLA
Fig. 2. PCOD dissolution and PLA transformation in the hyperthermophilic reactors,
(a) PCOD dissolution ratio; (b) PLA transformation ratio. Runs b7 to b9 and c7 to c9
hydrolysis rate, however the ammonia concentration should be
were operated under the same condition except for the addition amount of controlled with caution to avoid the adverse effect on methane
ammonia in the hyperthermophilic reactors, respectively. production in the methane fermentation reactors.
 F. Wang et al. / Bioresource Technology 112 (2012) 67–74 71

(a) Run a7
25.0

Organic acid concentrations (g COD/L)

3000 3.5
M0
2500 3.0
20.0

Methane production rate

Ammonia concentration
2.5
2000

(L at STP/(L•d))
15.0
2.0

(mg N/L)
1500
1.5 10.0
1000
1.0
5.0
500 0.5

0 0.0 0.0
526 532 538 544 550 556 562 568 574 580 586 592
Time (d)
(b) Run b7 Run b8 Run b9
25.0

Organic acid concentrations (g COD/L)

3000 3.5
M1
2500 3.0
Ammonia concentration

20.0

Methane production rate

2.5
2000

(L at STP/(L•d))
(mg N/L)

15.0
2.0
1500
1.5 10.0
1000
1.0
5.0
500 0.5

0 0.0 0.0
526 532 538 544 550 556 562 568 574 580 586 592
Time (d)
Run c7 Run c8 Run c9
(c) 25.0

Organic acid concentrations (g COD/L)

3000 3.5
M2
2500 3.0
20.0
Methane production rate
Ammonia concentration

2.5
2000 (L at STP/(L•d))
15.0
(mg N/L)

2.0
1500
1.5 10.0
1000
1.0
5.0
500 0.5

0 0.0 0.0
526 532 538 544 550 556 562 568 574 580 586 592
Time (d)

Enf. TAN Enf. FAN Methane production rate

Lactic acid Acetic acid Total organic acids

Fig. 3. Methane production rates, ammonia concentration and organic acid concentrations in methane fermentation reactors, (a) Control system; (b) Post-dissolution system;
(c) Pre-dissolution system.

3.3. Dewatering property of the sludge in thermophilic reactors

Dewatering property of sludge is a factor affecting the economic 30

benefit of practical AD processes. Fig. 4 shows the results of CST M0: y = 1.845x - 151.91
25 R2 = 0.9557
measurement tests in the three methane fermentation reactors. M1: y = 0.1118x + 10.283
CST/TS (s/(g/L))

There was a linear relationship between VS/TS and CST/TS. The 20 R2 = 0.7187
CST/TS ratio of M1 and M2 was lower than that of the control
15
reactor, which indicated that the dewatering property was M2: y = 0.1773x + 4.26
improved after hyperthermophilic treatment was introduced. The 10 M0
M R2 = 0.7061
0
deterioration in dewatering properties is associated with the accu- M1
5 M 1
mulation of proteins and polysaccharides in the colloidal size frac-
M2
M2
tion (Bivins and Novak, 2001). Consequently, reduction of the 0
colloidal materials may improve the dewatering property. There 50 60 70 80 90 100
are two possible reasons for the improvement of dewatering prop- VS/TS (%)
erty with hyperthermophilic treatment. One is the increase of the
overall system performance such as organic proportion removal; Fig. 4. Relationship between VS/TS and CST/TS.
72 F. Wang et al. / Bioresource Technology 112 (2012) 67–74

and the efficient removal of larger proportion of colloidal materi- total organic acid and ammonia concentrations in M2 were higher
als. The other is that more filamentous bacteria can be destructed than those in M1 and M0 (Fig. 5(c, d)). The reactor configuration of
in a higher treatment temperature (Marneri et al., 2009). the M1 + S1 system had the advantage over the S2 + M2 system be-
cause the liquor from S1 with high ammonia concentration was di-
3.4. Performance comparison of different digestion systems luted by the fresh substrate. This operation can avoid deterioration
of methanogens activity caused by the instant loading of substrate
The performance comparisons of the three systems are summa- with a high ammonia concentration as that in M2. In the control
rized in Fig 5. The methane conversion ratio was calculated as reactor, there was no obvious organic accumulation; consequently
follows: the lower treatment performance of the control reactor was possi-
bly due to lower hydrolysis efficiency under the thermophilic
Methane conversion ratio ð%Þ ¼ ðCOD of methane=COD of substrateÞ  100
condition.
ð5Þ Fig. 6 shows the COD balance of the three systems under the
In each system, the PLA transformation ratio (Fig. 5(a)) grad- same OLR (Runs a7, b9 and c8). The COD balance of the three sys-
ually increased, which indicated the acclimation of the microbes tems was closed to 100% and the difference between the fresh sub-
to PLA. The two-phase systems had obvious advantage of the strate COD and the total effluent COD due to measurement error
overall PLA transformation ratio compared with the control was less than 5%. The methane production in hyperthermophilic
reactor (Fig. 5(a, b)). This accorded with the previous researches reactors was less than 1% of the substrate COD and this demon-
results that higher temperature was favorable for PLA biodegra- strated the successful separation of the two phases. The PLA trans-
dation (Copinet et al., 2004; Itavaara et al., 2002; Wang et al., formation ratio and the methane conversion ratio of the control
2011). system were lower than those of the two-phase systems.
The methane conversion ratio in the three thermophilic reactors In a series of batch experiments conducted in the previous re-
showed a decline in Runs a4, b4 and c4. This was possibly caused search (Wang et al., 2011), the same plastics were pretreated under
by the reduction of SRT because in these runs the effluent was dis- the hyperthermophilic condition with ammonia addition and the
charged without centrifuge and the solid was not returned to the hydrolysis product was converted to methane gas in the following
reactors. The SS concentrations were 30.3, 20.1 and 18.4 g/L in thermophilic reactor with the ratios of 81.8% (PLA 1) and 77.0%
Runs a3, b3 and c3, respectively, and the values decreased to (PLA 2). The present results demonstrated that the continuous
15.8, 13.7 and 17.6 g/L in Runs a4, b4 and c4, respectively. After anaerobic digestion processes were applicable for PLA degradation
the HRT of the three reactors were increased (OLR was kept con- with same range of transformation ratio and methane conversion
stant), the SS concentrations recovered to 25.5, 21.0 and 23.4 g/L ratio.
in Runs a5, b5 and c5, respectively. Climenhaga and Banks
(2008), Nges and Liu (2010) showed that SRT is one of the key fac- 3.5. Microbial analysis
tors to keep the high treatment performance. In this study, most
methanogens were inactivated in the hyperthermophilic reactors; The microbial diversity of the three methane fermentation reac-
hence the SRT should be controlled with more caution in the meth- tors is summarized in Table 3. In each reactor, 96 clones were ana-
anogenic reactor than in traditional two-phase mesophilic or ther- lyzed. All clones having a sequence similarity of more than 97%
mophilic treatment. The M1 + S1 and S2 + M2 systems had also with each other were grouped into an operational taxonomic unit
higher COD removal ratio than the control system. However, the (OTU). The detected OTUs with an occupation ratio over 3% in each

(a) 100 (b) 95

Methane conversion ratio (%)
PLA transformation ratio (%)

85
80
75
60
65
Control Control
40 M1+S1 55 M1+S1
S2+M2 S2+M2
20 45
1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
Run Run
(c) 30 (d) 2.5
Control Control
Total organic acids (g COD/L)

25 M1+S1 2.0 M1+S1

Ammonia conc. (g N/L)

S2+M2 S2+M2
20
1.5
15
1.0
10

5 0.5

0 0.0
1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
Run Run

Fig. 5. Performance comparison of three digestion systems, (a) PLA transformation ratio; (b) Methane conversion ratio; (c) Total organic acid concentrations; (d) Ammonia
concentration.
 F. Wang et al. / Bioresource Technology 112 (2012) 67–74 73

(a) Control system

COD3: 70.1 COD1: PLA excluded
COD2: PLA
COD3: Methane
COD1:
86.7 M0
COD1: 20.3
COD2: COD2: 5.0
13.3

(b) Post-dissolution system

COD1: COD3:
COD1:
94.3 82.7 COD1: 20.4 COD1: 13.6
86.7
M1 COD2: 2.4
COD2: COD2: COD2: 3.6
13.3 13.9
COD1: 7.6 COD3: 0.2 COD1: 6.8

COD2: 0.6 S1 COD2: 1.2

(c) Pre-dissolution system

COD3:
COD1: 101.2
COD3: 0.5 80.2
COD1: COD1:
COD1: COD1: 24.6 16.4
86.7 94.9 M2
S2
COD2: COD2: COD2: COD2:
13.3 14.5 3.5 2.3
COD2: 7.8

COD1: 8.2
COD2: 1.2

Fig. 6. COD balance (Runs a7, b9, c8; OLR = 10.3 g COD/(L day)), (a) Control system; (b) Post-dissolution system; (c) Pre-dissolution system.

reactor are listed in this table. In the control reactor and M1, two OTU closely related to Clostridium sp. PR67 was also detected. In
detected OTUs with the highest occupation were closely related M2, an OTU closely related to Geobacillus sp. MJU 148-2 dominated
to Thermotogaceae bacterium FR850164 and Geobacillus sp. MJU (12.5% of the total colons). The other OTUs with occupation ratios
148-2. They were reported in thermophilic anaerobic digesters over 3% were related to T. bacterium FR850164 (4.2% of the total co-
treating excess sludge and cheese whey. In the two reactors, an lons), Clostridium sp. 5-8 (3.1% of the total colons), and Geobacillus
sp. NBM49 (3.1% of the total colons). The dominant microbes were
bacteria which are reported to be surviving in thermophilic anaer-
Table 3 obic conditions. Comparing the microbial distribution in the three
Results of clones sequence in three methane fermentation reactors.
reactors, there was no obvious difference. This implies that the
Reactor OTU Comment Similarity Cloning Occupation microbial communities were not affected by the reactor configura-
(%) No. ratioa (%) tion of the three systems markedly. HRT in the hyperthermophilic
M0 (Run a-2) FR850164 Thermotogaceae 99.4 13/96 13.5 reactors for single pass was set at 1 day, and this did not possibly
bacterium affect the microbial diversity. Wang et al. (2011) showed that
EU093964 Geobacillus sp. 95.9 7/96 7.3
PLA degradation was partially contributed by biological and/or
MJU 148-2
AB174828 Clostridium sp. 95.9 4/96 4.2 enzyme activity in the hyperthermophilic treatment by the batch
RR67 experiments. This hyperthermophilic treatment does not require
M1 (Run b-2) FR850164 Thermotogaceae 99.2 11/96 11.5 special bacteria adapted to this extreme temperature condition,
bacterium
but bacteria in the thermophilic condition can be utilized, so the
EU093964 Geobacillus sp. 96.0 10/96 10.4
MJU 148-2
hyperthermophilic solubilization can be easily installed to
AB174828 Clostridium sp. 95.9 3/96 3.1 traditional thermophilic anaerobic digestion plants.
RR67
M2 (Run c-2) EU093964 Geobacillus sp. 96.0 12/96 12.5
MJU 148-2
4. Conclusion
FR850164 Thermotogaceae 99.6 4/96 4.2
bacterium
FJ808605 Clostridium sp. 94.0 3/96 3.1 The results of this study showed that PLA was degraded to
5-8 methane in the continuous anaerobic digestion process. The
HQ703944 Geobacillus sp. 96.0 3/96 3.1
hyperthermophilic treatment promoted the PLA transformation
NBM49
and methane conversion of the two-phase systems compared with
a
Detected clones to total clones. the control reactor. The highest PCOD dissolution ratio and PLA
74 F. Wang et al. / Bioresource Technology 112 (2012) 67–74

transformation ratio were 25.0% and 54.9%, respectively. Under the Kadam, P.C., Boone, D.R., 1996. Influence of pH ammonia accumulation and toxicity
in halophilic, methylotrophic methanogens. Appl. Environ. Microbiol. 62 (12),
OLR of 10.3 g COD/(L day), the highest methane conversion ratios
4486–4492.
in the two-phase systems were 82.9% and 80.8%, respectively, Karst, D., Hain, M., Yang, Y.Q., 2008. Mechanical properties of polylactide after
higher than that of 70.1% in the control reactor. The systems repeated cleanings. J. Appl. Polym. Sci. 108 (4), 2150–2155.
including the hyperthermophilic reactors also had the advantage Kim, H.W., Han, S.K., Shin, H.S., 2003. The optimisation of food waste addition as a
co-substrate in anaerobic digestion of sewage sludge. Waste Manage. Res. 21
on dewatering property. The microbial analysis indicated that (6), 515–526.
hyperthermophilic treatment does not require special bacteria Kurakake, M., Kisaka, W., Ouchi, K., Komaki, T., 2001. Pretreatment with ammonia
adapted to this extreme temperature condition, and can be easily water for enzymatic hydrolysis of corn husk, bagasse, and switchgrass. Appl.
Biochem. Biotechnol. 90 (3), 251–259.
installed to traditional thermophilic anaerobic digestion plants. Lane, D.J., 1991. Nucleic Acid Techniques in Bacterial Sytstematics. John Wiley &
Sons, New York.
Lee, M., Hidaka, T., Tsuno, H., 2008. Effect of temperature on performance and
Acknowledgements microbial diversity in hyperthermophilic digester system fed with kitchen
garbage. Bioresour. Technol. 99 (15), 6852–6860.
The authors would like to express their deepest gratitude to Dr. Lee, M., Hidaka, T., Hagiwara, W., Tsuno, H., 2009. Comparative performance and
microbial diversity of hyperthermophilic and thermophilic co-digestion of
Masaki TAKAOKA and Dr. Kazuyuki OSHITA for their kind support kitchen garbage and excess sludge. Bioresour. Technol. 100 (2), 578–585.
in CST analysis. The authors would also like to express the respect Lee, J.M., Jameel, H., Venditti, R.A., 2010. A comparison of the autohydrolysis and
and appreciation to Mr. Yasunari KUSUDA and Mr. Takumi OISHI ammonia fiber explosion (AFEX) pretreatments on the subsequent enzymatic
hydrolysis of coastal Bermuda grass. Bioresour. Technol. 101 (14), 5449–5458.
for their great contribution. Marneri, M., Mamais, D., Koutsiouki, E., 2009. Microthrix parvicella and Gordona
amarae in mesophilic and thermophilic anaerobic digestion systems. Environ.
Technol. 30 (5), 437–444.
References Mata-Alvarez, J., Mace, S., Llabres, P., 2000. Anaerobic digestion of organic solid
wastes. an overview of research achievements and perspectives. Bioresour.
Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local Technol. 74 (1), 3–16.
alignment search tool. J. Mol. Biol. 215, 403–410. Nielsen, B., Petersen, G., 2000. Thermophilic anaerobic digestion and pasteurisation.
Angelidaki, I., Ahring, B.K., 1993. Thermophilic anaerobic-digestion of livestock Practical experience from Danish wastewater treatment plants. Water Sci.
waste - the effect of ammonia. Appl. Microbiol. Biotechnol. 38 (4), 560–564. Technol. 42 (9), 65–72.
Arai, R., Zenda, K., Hatakeyama, K., Yui, K., Funazukuri, T., 2010. Reaction kinetics of Nges, I.A., Liu, J., 2010. Effects of solid retention time on anaerobic digestion of
hydrothermal depolymerization of poly(ethylene naphthalate), poly(ethylene dewatered-sewage sludge in mesophilic and thermophilic conditions.
terephthalate), and polycarbonate with aqueous ammonia solution. Chem. Eng. Renewable Energy 35 (10), 2200–2206.
Sci. 65 (1), 36–41. Perepelkin, K.E., 2002. Polylactide fibres: fabrication, properties, use, prospects. a
Armentano, I., Dottori, M., Fortunati, E., Mattioli, S., Kenny, J.M., 2010. Biodegradable review. Fibre Chem. 34 (2), 85–100.
polymer matrix nanocomposites for tissue engineering: a review. Polym. Scherer, P.A., Vollmer, G.R., Fakhouri, T., Martensen, S., 2000. Development of a
Degrad. Stab. 95 (11), 2126–2146. methanogenic process to degrade exhaustively the organic fraction of
APHA, AWWA, WEF, 1998. Standard Methods for the Examination for Water and municipal ‘‘grey waste’’ under thermophilic and hyperthermophilic
Wastewater, 20th ed. conditions. Water Sci. Technol. 41 (3), 83–91.
Bivins, J.L., Novak, J.T., 2001. Changes in dewatering properties between the Sterling, M.C., Lacey, R.E., Engler, C.R., Ricke, S.C., 2001. Effects of ammonia nitrogen
thermophilic and mesophilic stages in temperature-phased anaerobic digestion on H-2 and CH4 production during anaerobic digestion of dairy cattle manure.
systems. Water Environ. Res. 73 (4), 444–449. Bioresour. Technol. 77 (1), 9–18.
Climenhaga, M.A., Banks, C.J., 2008. Anaerobic digestion of catering wastes: effect of Sung, S.W., Santha, H., 2003. Performance of temperature-phased anaerobic
micronutrients and retention time. Water Sci. Technol. 57 (5), 687–692. digestion (TPAD) system treating dairy cattle wastes. Water Res. 37 (7),
Copinet, A., Bertrand, C., Govindin, S., Coma, V., Couturier, Y., 2004. Effects of 1628–1636.
ultraviolet light (315 nm), temperature and relative humidity on the Tanigawa, N., Takemoto, T., Ohki, H., Kawasaki, T., 1997. Detailed component of
degradation of polylactic acid plastic films. Chemosphere 55 (5), 763–773. garbage. J. Jpn. Waste Manage. Assoc. 50, 116–119.
Dunne, M., Corrigan, O.I., Ramtoola, Z., 2000. Influence of particle size and Tsuji, H., Ikarashi, K., 2004. In vitro hydrolysis of poly(L-lactide) crystalline residues
dissolution conditions on the degradation properties of polylactide-co- as extended-chain crystallites – III. Effects of pH and enzyme. Polym. Degrad.
glycolide particles. Biomaterials 21 (16), 1659–1668. Stab. Stabil. 85 (1), 647–656.
Ghorpade, V.M., Gennadios, A., Hanna, M.A., 2001. Laboratory composting of Wang, F., Tsuno, H., Hidaka, T., Tsubota, J., 2011. Promotion of polylactide
extruded poly(lactic acid) sheets. Bioresour. Technol. 76 (1), 57–61. degradation by ammonia under hyperthermophilic anaerobic conditions.
Hansen, K.H., Angelidaki, I., Ahring, B.K., 1998. Anaerobic digestion of swine Bioresour. Technol. 102 (21), 9933–9941.
manure: inhibition by ammonia. Water Res. 32 (1), 5–12. Zenda, K., Funazukuri, T., 2008. Depolymerization of poly(ethylene terephthalate) in
Itavaara, M., Karjomaa, S., Selin, J.F., 2002. Biodegradation of polylactide in aerobic dilute aqueous ammonia solution under hydrothermal conditions. J. Chem.
and anaerobic thermophilic conditions. Chemosphere 46 (6), 879–885. Technol. Biotechnol. 83 (10), 1381–1386.