close

Вход

Забыли?

вход по аккаунту

?

j.biortech.2017.09.006

код для вставкиСкачать
Bioresource Technology 245 (2017) 590–597
Contents lists available at ScienceDirect
Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech
Minimizing mixing intensity to improve the performance of rice straw
anaerobic digestion via enhanced development of microbe-substrate
aggregates
MARK
⁎
Moonkyung Kim, Byung-Chul Kim, Yongju Choi, Kyoungphile Nam
Department of Civil and Environmental Engineering, Seoul National University, 1 Gwanak-ro, Gwanak-gu, Seoul 08826, Republic of Korea
G RA P H I C A L AB S T R A C T
A R T I C L E I N F O
A B S T R A C T
Keywords:
Anaerobic digestion
Mixing intensity
Hydrolysis
Microbial aggregates
Rice straw
The aim of this work was to study the effect of the differential development of microbe-substrate aggregates at
different mixing intensities on the performance of anaerobic digestion of rice straw. Batch and semi-continuous
reactors were operated for up to 50 and 300 days, respectively, under different mixing intensities. In both batch
and semi-continuous reactors, minimal mixing conditions exhibited maximum methane production and lignocellulose biodegradability, which both had strong correlations with the development of microbe-substrate
aggregates. The results implied that the aggregated microorganisms on the particulate substrate played a key
role in rice straw hydrolysis, determining the performance of anaerobic digestion. Increasing the mixing speed
from 50 to 150 rpm significantly reduced the methane production rate by disintegrating the microbe-substrate
aggregates in the semi-continuous reactor. A temporary stress of high-speed mixing fundamentally affected the
microbial communities, increasing the possibility of chronic reactor failure.
1. Introduction
greatest among domestic agricultural by-products (KEC, 2013). Approximately 37.6% of the annually generated rice straw in Korea is
recycled for animal feeding and other purposes, whereas the remaining
62.4% is treated as waste (KEC, 2013). Rice straw is a heterogeneous
polymeric material composed of cellulose, hemicellulose, lignin, and
other nonstructural carbohydrate components (Kaparaju et al., 2009; Li
Immense amounts of agricultural by-products such as rice straw,
rice husk, wheat straw, corn stover, and fruit branch are produced
worldwide every year (Kadam et al., 2000; Li et al., 2011). The annual
rice straw generation amounts to 6,034,000 tons in Korea, which is the
⁎
Corresponding author.
E-mail address: kpnam@snu.ac.kr (K. Nam).
http://dx.doi.org/10.1016/j.biortech.2017.09.006
Received 27 June 2017; Received in revised form 31 August 2017; Accepted 1 September 2017
Available online 05 September 2017
0960-8524/ © 2017 Elsevier Ltd. All rights reserved.
Bioresource Technology 245 (2017) 590–597
M. Kim et al.
digestion. Digested sludge collected from the Jungnang Wastewater
Treatment Plant located in Seoul, Korea was used as an inoculum. This
inoculum was passed through a sieve with 0.5 mm openings and stored
at 4 °C for a maximum duration of 48 h until use.
et al., 2011) that—if treated properly—can be converted into valuable
resources such as methane, bioplastic, bioethanol, and protein
(Abraham et al., 2016; Ahn et al., 2016; Huang et al., 2009; Lei et al.,
2010; Mussoline et al., 2013; Sommart et al., 2000). The conversion of
rice straw into useful resources can be an effective and sustainable alternative to treat the massive amount of the material that is currently
disposed.
Anaerobic digestion has been widely used for valorization of organic wastes and their conversion into renewable energy such as methane gas (Chandra et al., 2012; Ge et al., 2016). Anaerobic digestion is
known to have environmental and economic benefits such as energy
recovery from its organic components and significant reduction of
biomass (Lehtomäki and Björnsson, 2006; Nges et al., 2012). Organic
matter in wastewater sludge, food waste, and high-strength wastewater
is the typical substrate for anaerobic digestion, but use of lignocellulosic
materials such as rice straw has recently gained interest as a means of
producing renewable energy and to reducing greenhouse gas emissions
(Mussoline et al., 2013; Sawatdeenarunat et al., 2015; Yan et al., 2015).
Although challenges exist in the methanization of lignocellulosic materials because of their low conversion efficiency, the feasibility of this
methanization can be improved by optimized digester design and operation, substrate pretreatment, or co-digestion with other organic
wastes (Sawatdeenarunat et al., 2015).
Mixing intensity is one of the key operational variables that determine the performance of an anaerobic digestion process. Mechanical
mixing homogenizes the contents of an anaerobic reactor and enhances
the mass transfer of organic substrates to microbial biomass (Kaparaju
et al., 2008). In addition, agitation helps to release trapped gas bubbles
in the reactor, yielding uniformity of proper temperature and preventing sedimentation of denser particulate materials (Appels et al.,
2008). A recent study demonstrated that the shear force generated by
mechanical mixing is critical for formation and maturing of microbial
granules via aggregation of flocculent biomass (Zhou et al., 2014).
However, mixing has been shown to have not only positive but also
negative impacts on the performance of anaerobic digestion. It was
reported that overly intense mixing may lead to digester instability,
reduced biogas production, and increased vulnerability to shock loadings (Kaparaju et al., 2008; Lindmark et al., 2014; Stroot et al., 2001).
Therefore, it is important to provide an adequate intensity of mixing for
a stable and efficient operation of an anaerobic digester. The optimal
mixing intensity for anaerobic digestion should vary for different types
of substrates and different modes of reactor operation (e.g., batch or
continuous-flow) (Kaparaju et al., 2008; Lindmark et al., 2014; Li et al.,
2015; Ong et al., 2002). The effect of mixing on anaerobic digestion has
been mostly studied using relatively easily degradable substrates such
as cattle manure and sewage sludge (e.g., Kaparaju et al., 2008; Ong
et al., 2002; Stroot et al., 2001), and it is currently not well understood
how mixing affects the anaerobic digestion of lignocellulosic biomass
such as rice straw.
The objective of this study was to improve the understanding of the
effect of mixing intensity on the performance of anaerobic digestion of
rice straw. Laboratory-scale batch and semi-continuous reactors were
used to investigate how the physical environment developed via different mixing conditions affects the interaction of microorganisms with
the particulate substrate and how it relates to the hydrolysis of lignocellulosic materials in rice straw and, consequently, to the performance of anaerobic digestion.
2.2. Batch reactors
Batch testing was conducted using a 600 mL serum bottle with
500 mL of inoculum and 5.76 g rice straw (5.0 g as volatile solids (VS)).
The contents of the batch reactors were purged with nitrogen (> 99%
purity) at the beginning of the test, and the bottle was sealed with a
rubber stopper and an aluminum cap to maintain anaerobic condition.
Seven different mixing conditions were applied: no mixing; once-aweek, twice-a-week, and once-a-day intermittent mixing; and 50, 150,
and 300 rpm continuous mixing. For the intermittent mixing conditions, end-over-end mixing was applied manually 10 times per mixing
event, otherwise the reactors remained undisturbed. Continuous mixing
was applied using a horizontal shaker. The batch reactors were prepared in triplicate for all mixing conditions and were operated for
50 days at a temperature of 35 °C in a thermostatic room for mesophilic
anaerobic digestion.
2.3. Semi-continuous reactors
Three semi-continuous reactors were operated at a working volume
of 2.5 L and headspace of 1.5 L, with connections to allow continuous
biogas ventilation and periodic substrate addition and effluent extraction. The substrate addition and effluent extraction were conducted on
a daily basis, and the reactors were operated at a solid retention time
(SRT) and hydraulic retention time (HRT) of 30 days throughout the
operation. Identical operational conditions were applied for the three
reactors during the initial period of 100 days for adaptation. A regime
of 50 rpm continuous mixing was applied using a pitched-blade turbinetype impeller and the organic loading rate (OLR) was gradually increased from 0.1 to 1.0g VS/(L d) during the adaptation period.
After the adaptation period, different mixing regimes were applied
for the three semi-continuous reactors. One reactor was operated with
50 rpm continuous mixing throughout the test. This reactor is referred
to as “Reactor 1” henceforth. Intermittent mixing was applied in the
second reactor. Once a day, the contents of the reactor were agitated 10
times via manual end-over-end mixing, otherwise the reactor was left
undisturbed. The daily effluent extraction was conducted immediately
after the manual mixing such that homogeneous mixed liquor could be
extracted. This reactor is referred to as “Reactor 2” hereafter. The third
reactor was operated at 50 rpm continuous mixing with occasional (i.e.,
two) high-speed (150 rpm) mixing events (henceforth referred to as
“Reactor 3”). The high-speed mixing was applied for 19 days for the
first event and 15 days for the second. Each of the three semi-continuous reactors was operated for a minimum of 300 days in a thermostatic room with a temperature of 35 °C. After the 100-day adaptation period, the OLR was maintained at 1.0g VS/(L·d) until the end of
operation for all reactors.
2.4. Sampling and analysis
2.4.1. Biogas sampling and analysis
The biogas produced was sampled using 1-L Supel™ Inert Gas
Sampling bags connected to each reactor, and its volume was recorded
on a daily basis. The biogas contents (i.e., methane, nitrogen, and
carbon dioxide) were analyzed using a gas chromatograph (GC; 6000
Series, ACME 6100, USA) using an 80/100 Porapak N column (Agilent
Technologies, USA; 305 × 2.1 mm) and a thermal conductivity detector (TCD). Helium (> 99% purity) was used as the carrier gas. The
oven was programmed as follows: 80 °C for 2.5 min, heated to 120 °C at
15 °C/min, and then held for 1.5 min.
2. Materials and methods
2.1. Preparation and analysis of substrate and inoculum
Rice straw was obtained from the Korea Rural Development
Administration (Seoul, Korea). The rice straw was cut, ground with a
mortar and pestle, passed through a sieve with 2 mm openings, and
dried at 60 °C for 24 h prior to use as a substrate for anaerobic
591
Bioresource Technology 245 (2017) 590–597
M. Kim et al.
from the NDF contents, as suggested in the literature (Krishnamoorthy
et al., 1982).
The lignin contents were analyzed following the Klason procedure
described by White (White, 2007). In 300 mg dry sample, 3 mL 72%
sulfuric acid solution was added, and then the sample was incubated for
1 h at 30 °C. After the incubation, 84 mL DI water was added, and the
sample was autoclaved at 121 °C for 1 h and then vacuum-filtered. The
absorbance of the filtrate was measured at 240 nm to determine the
Klason lignin contents. For ADF analysis, 100 mL acid detergent solution, prepared by dissolving 20 g/L cetyltrimethylammonium bromide
in 1 N sulfuric acid solution, was added to 2.0 g dry sample. After
adding 2 mL decalin, the sample was boiled for 1 h and the solids were
obtained using a glass filter with vacuum. After washing the solids
twice with hot water then with acetone until the wash solution was
clear, the solids were dried at 105 °C for 4 h. Finally, the weight of the
solids was measured and reported as the amount of ADF. For NDF
analysis, neutral detergent solution with the following composition was
used: 30.0 g/L sodium lauryl sulfate, 18.6 g/L EDTA disodium salt,
6.81 g/L sodium borate, 4.56 g sodium phosphate, and 10.0 mL/L
ethylene glycol monoethyl ether, with sodium carbonate or hydrochloric acid to adjust the pH to 6.9–7.1. In a beaker, 1.0 g dry sample,
100 mL of the neutral detergent solution, 2 mL decalin, and 0.5 g sodium sulfite was combined. Then, the sample was treated following the
boiling, solids separation, washing, and drying procedures used for the
ADF analysis. The final weight of the solids was determined as the
amount of NDF.
The masses of lignin, cellulose, and hemicellulose remaining in the
batch reactors after incubation could be calculated as the total amount
of reactor solids remaining in the reactor, the fraction of digested rice
straw in the reactor solids, and the lignin, cellulose, and hemicellulose
mass composition in the digested rice straw were all measured.
Similarly, the mass extraction rates of lignin, cellulose, and hemicellulose from the semi-continuous reactors could be calculated.
2.4.2. Liquid and solid sampling and liquid fraction analysis
The reactor liquid and solid contents were collected after operations
of the batch reactors ceased and from the effluents extracted on a daily
basis for the semi-continuous reactors. The liquid fraction was analyzed
for volatile fatty acid (VFA) concentration and dissolved organic carbon
(DOC) content. The aqueous phase was separated via centrifugation at
4000g for 10 min and filtration using a 0.45 μm syringe filter for VFA
and DOC analyses. A high performance liquid chromatograph (HPLC;
Model YL9100; Young Lin Instrument, Korea) with Aminex® HPX-87H
(Bio-Rad Laboratories, USA; 300 × 7.8 mm) column was used for the
VFA analysis. The column temperature was maintained at 50 °C and
0.02 N H2SO4 was used as the mobile phase. The VFAs analyzed were
formic acid, acetic acid, propionic acid, isobutyric acid, butyric acid,
isovaleric acid, valeric acid, isocaproic acid, hexanoic acid, and nheptanoic acid. Each VFA was quantified to 0.1 mM. Total organic
carbon (TOC) content was analyzed using a TOC analyzer (TOC-VCPH,
Shimadzu, Japan).
2.4.3. Total extractable extracellular polymeric substance (EPS) content
and particle size distribution analyses
A portion of the reactor solids harvested by the centrifugation
procedure described above was used to analyze particle size distribution and total extractable extracellular polymeric substance (EPS)
content. Particle size analysis was conducted for both solids without
water washing and those with water washing. To obtain solids with
water washing, the reactor solids were washed with DI water with
gentle abrasion on a 0.1 mm-opening sieve to remove microorganisms
and organic residue. The digested rice straw was visibly distinguishable
from other types of reactor solids. By the washing, the color of the solid
matter turned from blackish to brownish color, which was close to the
color of the raw rice straw. No loss of digested rice straw from the
washing procedure was observed. Particle size distributions were analyzed using a Microtrac® S3500 particle size analyzer. A tri-laser technology was applied to allow accurate measurement in the sub-micron
size range. Particles were assumed to be spherically shaped for the interpretation of particle size distribution results. EPSs were extracted
from the solids via a procedure described in Liu and Fang (2002). First,
the harvested solids were resuspended with 15 mL 0.05% NaCl and
centrifuged at 4000g for 10 min, after which the supernatant was collected. The EPSs extracted using this procedure were regarded as either
readily extractable or loosely bound (Hung et al., 2005). Afterwards,
50 mL of 70 °C 0.05% NaCl solution was added to the sample residues.
After vortex mixing for 1 min and centrifugation at 4000g for 15 min,
the supernatant was collected. The EPSs extracted in this step were
regarded as tightly bound (Aguilera et al., 2008). The TOC contents of
both loosely and tightly bound EPSs were determined, and the sum of
the two TOC content values was defined as the total extractable EPSs.
2.6. Statistical analysis method
Data for triplicate measurements are expressed as mean ± standard
deviation. One-way ANOVA followed by the Duncan’s Studentized
range test was used to identify the differences of mean values of the
total methane production and the total extractable EPS contents in the
solids in the batch tests with different mixing regimes. The means were
determined to be statistically different if the p value was smaller than
0.05. All statistical analyses were performed using SPSS 15.0 software
(SPSS Inc., Chicago, USA).
3. Results and discussion
2.5. Lignin, cellulose, and hemicellulose content analyses
3.1. Effect of mixing intensity on total methane production in the batch
reactors
The solid fraction was also analyzed for lignin, cellulose, and
hemicellulose contents. In this case, a composite sample was prepared
because at least 10 g of dry weight was needed for the analysis. For the
semi-continuous reactors, the solids collected from effluents sampled
during 20 consecutive days was combined. For the batch reactors, the
solids obtained from three replicates were combined. The composite
samples of the reactor solids were washed with DI water to remove
microorganisms and organic residue as described above for particle size
analysis. The fraction of the digested rice straw in the reactor solids
recovered using this procedure was measured and recorded. The digested rice straw obtained was sent to the Korea Feed Ingredients
Association (Daejeon, Korea) for lignin, acid detergent fiber (ADF), and
neutral detergent fiber (NDF) content analyses. Undigested rice straw
was analyzed in parallel. The cellulose contents were determined by
subtracting the lignin contents from the ADF contents, and the hemicellulose contents were determined by subtracting the ADF contents
3.1.1. Performance of the batch reactors
The cumulative methane production measured over time for the
batch reactors is plotted in Fig. 1. For all mixing conditions, active
methane production occurred from the beginning of incubation to approximately 40 days, with little additional production observed thereafter. The total methane production within an overall incubation period
of 50 days was significantly greater for the intermittent and no mixing
conditions compared to the continuous mixing conditions (one-way
ANOVA, p < 0.05). Among the intermittent (i.e., once-a-week, twice-aweek, and once-a-day) and no mixing conditions, the differences in
total methane production were not statistically significant (one-way
ANOVA, p > 0.05). The total methane production for the 50, 150, and
300 rpm continuous mixing conditions was 13%, 25%, and 38%
smaller, respectively, than that for once-a-week mixing. It was evident
that the total methane production was reduced by increasing the mixing
speed for the continuous mixing conditions.
592
Bioresource Technology 245 (2017) 590–597
Total methane production (mL CH4/g TC)
300
200
No mixing
Once-a-week mixing
Twice-a-week mixing
Once-a-day mixing
50 rpm mixing
150 rpm mixing
300 rpm mixing
100
0
(a)
380
360
340
R2 = 0.8073
320
300
280
260
240
220
40
50
60
70
80
90
Lignocellulose biodegradability (%)
0
10
20
30
40
(b)
50
100
Lingnocellulose degradability (%)
Duration (day)
Fig. 1. Cumulative methane production measured in the batch reactors over time for
different mixing conditions. The error bars represent the standard deviation of triplicate
measurements.
Table 1
Amounts of residual lignin, cellulose, and hemicellulose in the reactor solids after the
batch reactor runs, compared with the amounts initially fed to each reactor.
Sample
Mixing
regime
Lignin (g)
Cellulose (g)
Hemicellulose (g)
Lignocellulose
biodegradability
(%)
Initial feed
–
0.436
3.043
2.805
–
Residuals
None
Once-aweek
Twice-aweek
Once-aday
50 rpm
150 rpm
300 rpm
0.306
0.289
0.816
0.683
0.073
0.094
81.0
83.0
0.290
0.886
0.104
79.6
0.392
1.579
0.194
65.5
0.379
0.406
0.415
1.512
1.986
1.899
0.467
1.101
0.973
62.5
44.4
47.7
400
380
90
R2=0.9354
360
80
340
320
70
300
60
280
R2=0.8069
260
50
240
40
220
160
180
200
220
240
Total methane production (mL CH 4 / g TC)
Cumulative methane production (mL CH 4/g TC)
M. Kim et al.
260
Total extractable EPS content (mg TOC/g TS)
Lignocellulose degradability (%)
Total methane production (mL CH4/g TC)
Fig. 2. Correlation of (a) the lignocellulose biodegradability with total methane production and (b) the total extractable EPS content with lignocellulose biodegradability and
total methane production for the batch tests under different mixing conditions.
organic compounds produced via hydrolysis of rice straw were nearly
complete for all mixing conditions.
In sum, the performance of the anaerobic digestion of rice straw in
batch reactors was the highest when intermittent mixing with minimal
mixing frequency was applied. Low residual VFA and DOC contents for
all mixing conditions indicate that hydrolysis was the bottleneck step
for the rice straw anaerobic digestion. The hydrolytic conversion efficiency of lignocellulosic contents of rice straw was remarkably enhanced by reducing the mixing intensity, reaching up to 83% for the
minimal mixing condition. The specific methane yield for the once-aweek reactor was 425 mL/g VS, which is substantially higher than the
reported range of 92–280 mL/g VS at mesophilic conditions (Mussoline
et al., 2013).
The amounts of residual lignin, cellulose, and hemicellulose for each
mixing condition are compared to the amounts in the initial feed in
Table 1. As expected, only a slight reduction in the amount of lignin
(5–34%) was observed by the incubation, indicating limited biodegradability of lignin. On the other hand, the amounts of cellulose and
hemicellulose were reduced by 35–78% and 61–97%, respectively, after
the incubation, showing partial to nearly complete degradation of those
components. By comparing the sum of residual lignin, cellulose, and
hemicellulose with the sum of the materials in the initial feed, the
lignocellulose biodegradability was determined for each reactor (data
shown in Table 1) and is plotted against the total methane production
in Fig. 2a. It can be clearly observed that total methane production and
lignocellulose biodegradability were positively correlated. The mixing
condition with the highest total methane production (i.e., once-a-week
mixing) showed the greatest lignocellulose biodegradability of 83.0%.
Continuous mixing conditions (i.e., 50, 150, and 300 rpm mixing),
which showed smaller total methane production than once-a-week
mixing did, exhibited much smaller lignocellulose biodegradability,
ranging from 44.4% to 62.5%.
The VFA concentrations in the liquids after the batch tests were
below the quantification limit of 0.1 mM for all VFAs analyzed and for
all mixing conditions. The DOC contents in the liquids were also much
lower than the amount of cellulose and hemicellulose carbon degraded
during the batch reactor operation. These results indicate that acidogenesis, acetogenesis, and methanogenesis of the simple, soluble
3.1.2. Development of microbe-substrate aggregates
Comparison among the particle distributions for the rice straw before digestion and the reactor solids after digestion showed that the
particles increased in size after incubation (Fig. 3a). The size distribution for reactor solids for all mixing conditions shifted to the right from
that of the initial feed. Particle size analysis was also conducted for the
reactor solids washed with DI water with gentle abrasion on a sieve to
remove microorganisms, as described in detail in Section 2.4.2. The
particle size for the washed solids was similar or slightly larger than
that of the rice straw before digestion, suggesting that the size of the
rice straw itself did not decrease significantly during digestion. Slightly
larger particle sizes observed for the gently washed solids from some of
593
Bioresource Technology 245 (2017) 590–597
Total extractable EPS content (mg TOC/g TS)
M. Kim et al.
(b)
a
a, b
a, b
b, c
250
c
200
e
d
150
0
k
k
ing
ay
ee
ee
ad
mix
aw
aw
ce
e
e
No
n
c
c
i
O
On
Tw
r
50
pm
0r
15
pm
0r
30
pm
Fig. 3. Characteristics of the reactor solids harvested after the batch reactor operations with different mixing regimes: (a) particle size distribution compared with that of the rice straw
before digestion and (b) total extractable EPS contents in the solids. The error bars in panel b represent the standard deviation of triplicate measurements. Pairs without a significant
difference (p < 0.05), as determined by ANOVA, are assigned the same letter in panel b.
or biofilm on solid surfaces, EPS contents in the reactor solids can be
used as a quantitative measure of the development of the microbesubstrate aggregates. Fig. 3b shows the significant differences in the
total extractable EPS contents in the reactor solids for different mixing
conditions. Once-a-week mixing, which was the minimal mixing condition employed in this study excluding no mixing, exhibited a maximum total extractable EPS content of 261.1 ± 1.7 mg TOC/g total
solids (TS). This value was significantly different statistically from the
values for twice-a-week, 50 rpm, 150 rpm, and 300 rpm mixing (oneway ANOVA, p = 0.033, 0.024, 0.010, and 0.010, respectively) and
63% higher than the minimum value of 160.2 ± 15.1 mg TOC/g TS for
300 rpm mixing. The total extractable EPS contents for the 150 and
300 rpm mixing conditions were distinctively smaller than those for the
other conditions. The EPS analysis confirmed that minimal mixing was
favorable for the development of the microbe-substrate aggregates,
whereas continuous mixing with high mixing speed was detrimental
because of the shear force applied to the solids. Dangcong et al. (1999)
also reported that the strong shear force generated by intense mixing
could prevent the formation of larger microbial aggregates.
the batch reactors can be attributed to incomplete removal of microorganisms attached to the digested rice straw by gentle washing. The
particle size analysis suggested that the microorganisms attached to the
rice straw (cut into pieces < 2 mm in length), which possibly served as
nuclei to form microbe-substrate aggregates. Cirne et al. (2007) reported that a significant portion of reactor microorganisms were bound
either loosely or strongly to the solid substrate in a hydrolytic reactor
for digestion of sugar beet and grass/clover silage.
In general, the particle size of the reactor solids was larger for intermittent mixing conditions compared to continuous mixing conditions. For example, the size distribution curve peaked at 44 μm for
once-a-week mixing, whereas the peak for 300 rpm mixing was found at
22 μm (Fig. 3a). The results suggest that the relatively high shear stress
applied to reactor solids for the continuous mixing conditions inhibited
growth of the microbe-substrate aggregates. The only exception was the
150 rpm mixing condition, which showed a comparable size distribution curve with those for intermittently mixed conditions for unknown
reasons.
As EPSs are responsible for the development of microbial aggregates
594
Bioresource Technology 245 (2017) 590–597
M. Kim et al.
The total extractable EPS contents correlated well with the lignocellulose degradability and total methane production during the incubation (Fig. 2b). In other words, the low mixing intensity corresponded to enhanced development of the microbe-substrate aggregates,
high lignocellulose biodegradability, and high total methane production. These results suggested that the development of the microbesubstrate aggregates played a significant role in the performance of
anaerobic digestion of rice straw. It is likely that via the microorganisms’ binding to the solid substrate, the destruction of cellulose and
hemicellulose in rice straw was facilitated, supplying simple organic
molecules that were eventually converted to methane and carbon dioxide.
Previous studies suggested that the benefit of low intensity mixing
on methane production in anaerobic digesters could be attributed to the
enhanced syntrophy between acetogens and methanogens (Kaparaju
et al., 2008; Lindmark et al., 2014; Stroot et al., 2001). Development of
granular biosolids, which offer a favorable microenvironment and
juxtaposition for syntrophic microorganisms, was considered as the
main reason for the enhanced syntrophy in minimally mixed reactors
(Amani et al., 2010; de Bok et al., 2004; Gómez et al., 2006; Ong et al.,
2002). Using substrates that are relatively easier to be hydrolyzed than
lignocellulosic materials, these studies reported a notable increase in
VFA concentrations under intense mixing conditions. For example,
Kaparaju et al. (2008) observed that the VFA concentration in a laboratory-scale reactor with a mixing strategy of repeated 10 h continuous mixing to 2 h withholding was 20.5-fold greater than that in a
reactor with minimal mixing (10 min mixing at 12 h intervals), using
animal manure as the substrate. Stroot et al. (2001) reported up to
3000 mg/L of propionate buildup by operating an anaerobic digester
with continuous mixing using municipal solid waste and biosolids as codigestate. In contrast, as VFA buildup was not observed in any mixing
conditions in the current study, the benefit of low intensity mixing
could not be explained by the syntrophic relationship between acetogens and methanogens in granular biosolids. Recently, Lindmark et al.
(2014) also reported that the low methane production rate observed
under an intense mixing condition was not accompanied with VFA
buildup in an anaerobic digester using the organic fraction of municipal
solid waste as a substrate.
For anaerobic digestion of particulate lignocellulosic substrates,
such as rice straw, for which hydrolysis is rate limiting (Chandra et al.,
2012; Noike et al., 1985), the benefit of minimal mixing may be exhibited differently from the digestion of easily-hydrolyzable substrates.
The batch study results strongly suggest that the microbe-substrate
aggregates developed in the anaerobic reactors accelerated the rice
straw hydrolysis. As the hydrolytic conversion efficiency is a critical
factor in determining the feasibility of biogas production from lignocellulosic biomass (Mussoline et al., 2013; Sawatdeenarunat et al.,
2015), optimizing the mixing intensity to favor the development of the
microbe-substrate aggregates is expected to significantly improve the
feasibility of the process.
3.2. Behavior of the semi-continuous reactors under different mixing
regimes
The biogas composition of the three semi-continuous reactors from
the beginning to 300 days of operation is presented in Fig. 4. At the end
of the adaptation period (i.e., Day 100), all reactors were stable for
anaerobic digestion with methane and carbon dioxide contents of
48.1 ± 2.7% and 45.5 ± 3.1%, respectively, in the biogas. Nitrogen gas
composition was relatively higher in Reactor 1 at Day 100, but the
value gradually decreased thereafter and remained at 7.8 ± 2.0% after
150 days. The OLR was maintained to be constant after Day 100, and a
mixing regime of 50 rpm continuous mixing was applied for the whole
period of reactor operation for Reactor 1.
For Reactor 2, the mixing condition was changed from 50 rpm
continuous mixing to once-a-day intermittent mixing at Day 100. The
Fig. 4. Changes in the biogas composition for the semi-continuous reactors: (a) Reactor 1,
(b) Reactor 2, and (c) Reactor 3.
methane content in biogas was slightly greater for Reactor 2 than it was
for Reactor 1. The methane and carbon dioxide contents during Days
100–300 were 53.2 ± 1.7% and 47.2 ± 3.3% for Reactor 2, respectively, and 50.8 ± 2.1% and 49.4 ± 1.2% for Reactor 1. As shown in
Table 2, lignocellulose biodegradability was also slightly higher for
595
Bioresource Technology 245 (2017) 590–597
M. Kim et al.
Table 2
Comparison of feeding and extraction rates of lignin, cellulose, and hemicellulose for the
semi-continuous reactors with different mixing regimes.
Conditions
Feeding/extraction rate (g/d)
Feedinga
Extractionb
Reactor 1
Reactor 2
Reactor 3
Lignocellulose
biodegradability (%)
Lignin
Cellulose
Hemicellulose
0.138
1.277
0.886
–
0.140
0.136
0.137
0.564
0.542
1.014
0.187
0.106
0.335
61.3
65.9
35.4
a
Same feeding rate was applied for all reactors.
Determined using a composite sample of Day 217–237 effluent solids for all reactors.
The second event of 150 rpm mixing for Reactor 3 was Day 215–230.
b
Reactor 2 (65.9%) than for Reactor 1 (61.3%). These results showed
that, as was observed in batch reactors, intermittent mixing condition
performed better than continuous mixing condition in semi-continuous
reactor operation.
The negative effect of overly intense mixing was observed in
Reactor 3. For this reactor, the mixing speed was changed unintentionally from 50 rpm to 150 rpm at around Day 123 because of an
operational error. The error was identified at Day 142, and the mixing
speed was immediately recovered back to 50 rpm. Fig. 4c shows that
the higher mixing intensity substantially affected the reactor performance. During the 150 rpm mixing event (i.e., Days 123–142), the
methane and carbon dioxide contents were reduced from 43% to 10%
and 39% to 3%, respectively. By adjusting the mixing speed back to
50 rpm, the reactor performance was rapidly recovered to obtain
42 ± 5% methane and 50 ± 5% carbon dioxide contents at Days
141–210. To study the effect of the intense mixing in a more controlled
manner, the second event of 150 rpm mixing was applied at Days
215–230. It is clear that the methane and carbon dioxide contents responded to the change in the mixing speed, dropping to minimum values of 7.3% and 15%, respectively, during the second 150 rpm mixing
event.
The daily-extracted solids of Reactor 3 during and shortly after the
second high-speed mixing event were combined to analyze lignocellulosic material contents. The data in Table 2 show that lignocellulose biodegradability was substantially smaller for the 150 rpm
mixing period of this reactor compared to that of the other reactors.
Considering the SRT of 30 days for the semi-continuous reactors, the
actual lignocellulose biodegradability should have been even smaller
than the measured value of 35.4%. As was observed in the batch reactors, these results indicate that the high-speed mixing mainly affected
the hydrolysis of rice straw, resulting in reduced biogas production.
Neither VFA accumulation nor significant reduction in pH in the reactors was observed in any of the reactors for any period of operation.
Therefore, the poor performance during the 150 rpm mixing period is
not likely to be caused by the inhibition of methanogens.
After the mixing intensity for Reactor 3 was recovered to 50 rpm at
Day 234, the methane and carbon dioxide contents gradually recovered
to their levels prior to the second 150 rpm mixing event and remained
stable until Day 272. However, the contents of methane and carbon
dioxide decreased sharply after Day 272 and were measured as 3.2%
and 27.3%, respectively, at Day 300 (see Fig. 4c). Further operating
Reactor 3 up to Day 350 did not result in the recovery of the methane
content in the biogas. Two explanations are possible for this reactor
malfunction after two 150 rpm mixing events. It is possible that the
semi-continuous reactor had capacity to recover from the stress of a
single high-speed mixing event, but could not absorb the stress of
multiple events. Another possibility is that the reactor performance
could only be temporarily recovered after a stress, with the reactor
destined to fail after long-term operation. The reactor performance
recovered after the first high-speed mixing event and remained stable
Fig. 5. Characteristics of reactor solids harvested from the semi-continuous reactors with
different mixing regimes: (a) the total extractable EPS contents in the solids during Days
213–253 and (b) particle size distribution at Day 230. The second event of 150 rpm
mixing for Reactor 3 was Day 215–230.
for approximately 70 days, but it cannot be disregarded that reactor
failure would have eventually occurred without the second high-speed
mixing event. For both possible modes of failure, it is certain that the
stress from the high-speed mixing events had a fundamental impact on
reactor performance, possibly by affecting the microbial community in
the microbe-substrate aggregates involved in hydrolysis of lignocellulosic materials. Deublein and Steinhauser (2011) demonstrated
that even a slight change or damage to the microbial community could
severely affect the performance of an anaerobic digester. High-intensity
mixing events during the operation of the semi-continuous reactors in
the current study resulted in increased susceptibility to reactor failure,
which was characterized by a reduced rate of hydrolysis without any
accumulation of VFAs or reactor acidification.
It was evident that the development of microbe-substrate aggregates
was affected by the mixing intensity in the semi-continuous reactors.
Total extractable EPS contents in the reactor solids were substantially
decreased by the high-speed (i.e., 150 rpm) mixing event, as shown in
Fig. 5a. For both Reactor 1 and 2, the average total extractable EPS
contents at Days 213–253 were 260 ± 3 mg TOC/g TS. The total extractable EPS content for Reactor 3 during the 50 rpm mixing period
was also comparable to those of the other reactors, amounting to
596
Bioresource Technology 245 (2017) 590–597
M. Kim et al.
Chandra, R., Takeuchi, H., Hasegawa, T., 2012. Methane production from lignocellulosic
agricultural crop wastes: a review in context to second generation of biofuel production. Renew. Sust. Energy Rev. 16 (3), 1462–1476.
Cirne, D., Lehtomäki, A., Björnsson, L., Blackall, L., 2007. Hydrolysis and microbial
community analyses in two-stage anaerobic digestion of energy crops. J. Appl.
Microbiol. 103 (3), 516–527.
Dangcong, P., Bernet, N., Delgenes, J.P., Moletta, R., 1999. Aerobic granular sludge – a
case report. Water Res. 33 (3), 890–893.
de Bok, F.A., Plugge, C.M., Stams, A.J., 2004. Interspecies electron transfer in methanogenic propionate degrading consortia. Water Res. 38 (6), 1368–1375.
Deublein, D., Steinhauser, A., 2011. Biogas from Waste and Renewable Resources: An
Introduction, second ed. John Wiley & Sons, New Jersey.
Ge, X., Xu, F., Li, Y., 2016. Solid-state anaerobic digestion of lignocellulosic biomass:
recent progress and perspectives. Biores. Technol. 205, 239–249.
Gómez, X., Cuetos, M.J., Cara, J., Morán, A., García, A.I., 2006. Anaerobic co-digestion of
primary sludge and the fruit and vegetable fraction of the municipal solid wastes.
Renew. Energy 31 (12), 2017–2024.
Huang, C.F., Lin, T.H., Guo, G.L., Hwang, W.S., 2009. Enhanced ethanol production by
fermentation of rice straw hydrolysate without detoxification using a newly adapted
strain of Pichia stipitis. Biores. Technol. 100 (17), 3914–3920.
Hung, C.C., Santschi, P.H., Gillow, J.B., 2005. Isolation and characterization of extracellular polysaccharides produced by Pseudomonas fluorescens Biovar II. Carbohyd.
Polym. 61 (2), 141–147.
Kadam, K.L., Forrest, L.H., Jacobson, W.A., 2000. Rice straw as a lignocellulosic resource:
collection, processing, transportation, and environmental aspects. Biomass. Bioenerg.
18 (5), 369–389.
Kaparaju, P., Buendia, I., Ellegaard, L., Angelidakia, I., 2008. Effects of mixing on methane production during thermophilic anaerobic digestion of manure: lab-scale and
pilot-scale studies. Biores. Technol. 99 (11), 4919–4928.
Kaparaju, P., Serrano, M., Thomsen, A.B., Kongjan, P., Angelidaki, I., 2009. Bioethanol,
biohydrogen and biogas production from wheat straw in a biorefinery concept.
Bioresour. Technol. 100 (9), 2562–2568.
Korea Environment Corporation (KEC), 2013. Estimation of Agricultural Waste
Generation and Usage (in Korean), KECO2013-PE11-30. KEC, Korea.
Krishnamoorthy, U., Muscato, T., Sniffen, C., Van Soest, P., 1982. Nitrogen fractions in
selected feedstuffs. J. Dairy Sci. 65 (2), 217–225.
Lehtomäki, A., Björnsson, L., 2006. Two-stage anaerobic digestion of energy crops: methane production, nitrogen mineralisation and heavy metal mobilisation. Environ.
Technol. 27 (2), 209–218.
Lei, Z., Chen, J., Zhang, Z., Sugiura, N., 2010. Methane production from rice straw with
acclimated anaerobic sludge: effect of phosphate supplementation. Biores. Technol.
101 (12), 4343–4348.
Li, L.L., Tong, Z.H., Fang, C.Y., Chu, J., Yu, H.Q., 2015. Response of anaerobic granular
sludge to single-wall carbon nanotube exposure. Water Res. 70, 1–8.
Li, Y., Park, S.Y., Zhu, J., 2011. Solid-state anaerobic digestion for methane production
from organic waste. Renew. Sust. Energ. Rev. 15 (1), 821–826.
Lindmark, J., Thorin, E., Fdhila, R.B., Dahlquist, E., 2014. Effects of mixing on the result
of anaerobic digestion: review. Renew. Sust. Energ. Rev. 40, 1030–1047.
Liu, H., Fang, H.H., 2002. Extraction of extracellular polymeric substances (EPS) of
sludges. J. Biotechnol. 95 (3), 249–256.
Mussoline, W., Esposito, G., Giordano, A., Lens, P., 2013. The anaerobic digestion of rice
straw: a review. Criti. Rev. Env. Sci. Tec. 43, 895–915.
Nges, I.A., Mbatia, B., Bjornsson, L., 2012. Improved utilization of fish waste by anaerobic
digestion following omega-3 fatty acids extraction. J. Environ. Manage. 110,
159–165.
Noike, T., Endo, G., Chang, J.E., Yaguchi, J.I., Matsumoto, J.I., 1985. Characteristics of
carbohydrate degradation and the rate-limiting step in anaerobic digestion.
Biotechnol. Bioeng. 27 (10), 1482–1489.
Ong, H.K., Greenfield, P.F., Pullammanappallil, P.C., 2002. Effect of mixing on biomethanation of cattle-manure slurry. Environ. Technol. 23 (10), 1081–1090.
Sawatdeenarunat, C., Surendra, K., Takara, D., Oechsner, H., Khanal, S.K., 2015.
Anaerobic digestion of lignocellulosic biomass: challenges and opportunities. Biores.
Technol. 178, 178–186.
Sommart, K., Parker, D., Rowlinson, P., Wanapat, M., 2000. Fermentation characteristics
and microbial protein synthesis in an in vitro system using cassava, rice straw and
dried ruzi grass as substrates. Asian. Austral. J. Anim. 13 (8), 1084–1093.
Stroot, P.G., McMahon, K.D., Mackie, R.I., Raskin, L., 2001. Anaerobic codigestion of
municipal solid waste and biosolids under various mixing conditions-I. Digester
performance. Water Res. 35 (7), 1804–1816 (5).
White, R.H., 2007. Effect of lignin content and extractives on the higher heating value of
wood. Wood. Fiber. Sci. 19 (4), 446–452.
Yan, Z., Song, Z., Li, D., Yuan, Y., Liu, X., Zheng, T., 2015. The effects of initial substrate
concentration, C/N ratio, and temperature on solid-state anaerobic digestion from
composting rice straw. Biores. Technol. 177, 266–273.
Zhou, D., Niu, S., Xiong, Y., Yang, Y., Dong, S., 2014. Microbial selection pressure is not a
prerequisite for granulation: dynamic granulation and microbial community study in
a complete mixing bioreactor. Biores. Technol. 161, 102–108.
259 mg TOC/g TS at Day 213. As the mixing speed increased to
150 rpm, the total extractable EPS content decreased, reaching approximately 160 mg TOC/g TS at Day 226. After changing the mixing
speed back to 50 rpm at Day 234, the total extractable EPS content
recovered gradually to 249 mg TOC/g TS at Day 243.
The particle size distribution data also indicate that the reactor
solids were significantly affected by the high-speed mixing event. The
particle size distribution was similar for Reactor 1 and Reactor 2 while
that for Reactor 3 during the high-speed mixing event was distinguishably smaller, indicating the disintegration of microbe-substrate
aggregates (Fig. 5b).
The total extractable EPS and particle size distribution results in the
semi-continuous reactors support the implication from the batch test
results that the development of microbe-substrate aggregates played a
significant role in the hydrolysis of rice straw. Overly intense mixing
lowered the performance of the semi-continuous anaerobic digester by
inhibiting the development of the aggregates or by disintegrating them.
The stress imposed to the reactor solids by the overly intense mixing
may cause chronic failure of the semi-continuous reactor for anaerobic
digestion of rice straw.
4. Conclusions
Mixing intensity had a profound effect on the performance of
anaerobic digestion of rice straw. Minimal mixing conditions favored
the development of the microbe-substrate aggregates, leading to a remarkable enhancement in the hydrolysis efficiency of cellulose and
hemicellulose in rice straw, which, in turn, results in a substantial enhancement in the methane production rate. An eventual failure of the
semi-continuous reactor after two high-intensity mixing events suggests
that a consistent maintenance of an adequate mixing regime is needed
for a stable operation of an aerobic digestion process for methanization
of lignocellulosic biomass.
Acknowledgements
This study received substantial support from the Geo-Advanced
Innovative Action (GAIA) Project of the Korea Environmental
Industry & Technology Institute (KEITI). This study was supported by
the Korea Ministry of Environment as a “Waste to Energy Human
Resource Development Project”.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.biortech.2017.09.006.
References
Abraham, A., Mathew, A.K., Sindhu, R., Pandey, A., Binod, P., 2016. Potential of rice
straw for bio-refining: an overview. Biores. Technol. 215, 23–36.
Aguilera, A., Souza-Egipsy, V., San Martín-Úriz, P., Amils, R., 2008. Extraction of extracellular polymeric substances from extreme acidic microbial biofilms. Appl.
Microbiol. Biot. 78 (6), 1079–1088.
Ahn, J., Jho, E.H., Kim, M., Nam, K., 2016. Increased 3HV concentration in the bacterial
production of 3-hydroxybutyrate (3HB) and 3-hydroxyvalerate (3HV) copolymer
with acid-digested rice straw waste. J. Polym. Environ. 24 (2), 98–103.
Amani, T., Nosrati, M., Sreekrishnan, T., 2010. Anaerobic digestion from the viewpoint of
microbiological, chemical, and operational aspects – a review. Environ. Rev. 18,
255–278.
Appels, L., Baeyens, J., Degrève, J., Dewil, R., 2008. Principles and potential of the
anaerobic digestion of waste-activated sludge. Prog. Energ. Combust. 34 (6),
755–781.
597
Документ
Категория
Без категории
Просмотров
4
Размер файла
1 011 Кб
Теги
2017, 006, biortech
1/--страниц
Пожаловаться на содержимое документа