Efficient Polyhydroxyalkanoates Production from a ... - ACS Publications

Sep 11, 2009 - discharge (AFD) process was investigated by the use of WAS alkaline fermentation liquid as the carbon source. It was observed that comp...
4 downloads 0 Views 377KB Size
Environ. Sci. Technol. 2009, 43, 7734–7741

Efficient Polyhydroxyalkanoates Production from a Waste-Activated Sludge Alkaline Fermentation Liquid by Activated Sludge Submitted to the Aerobic Feeding and Discharge Process YAMIN JIANG, YINGUANG CHEN,* AND XIONG ZHENG State Key Laboratory of Pollution Control and Resources Reuse, School of Environmental Science and Engineering, Tongji University, 1239 Siping Road, Shanghai 200092, China

Received May 15, 2009. Revised manuscript received August 21, 2009. Accepted August 21, 2009.

It was reported in our previous publication that the accumulation of short-chain fatty acids (SCFA) was significantly enhanced when waste-activated sludge (WAS) was anaerobically fermented at pH 10.0 (Yuan, et al., Environ. Sci. Technol. 2006, 40, 2025-2029). In this paper, the production of polyhydroxyalkanoate (PHA) by activated sludge with an aerobic feeding and discharge (AFD) process was investigated by the use of WAS alkaline fermentation liquid as the carbon source. It was observed that compared with other PHA synthesis processes reported in the literature, the AFD process showed the highest PHA production. The PHA content in sludge reached 72.9% when activated sludge was submitted to the AFD process. This was the highest PHA content obtained so far by activated sludge using wastes as the renewable carbon source. Although nitrogen and phosphorus were released into the WAS alkaline fermentation liquid, their presence did not affect PHA synthesis, which indicates that it is unnecessary to remove the released nitrogen and phosphorus, and the fermentation liquid can be used directly for PHA production. The accumulated PHA was mainly composed of 3-hydroxybutyrate (3HB) (73.5 mmol C%), 3-hydroxyvalerate (3HV) (24.3 mmol C%), and 3-hydroxy-2-methylvalerate (3H2MV) (2.2 mmol C%). Further investigation showed that SCFA rather than protein and carbohydrate in the alkaline fermentation liquid made the main contribution to PHA production. The PHA produced from WAS alkaline fermentation liquid had a molecular weight of 8.5 × 105 Da and a melting point of 101.4 °C. Analysis using the 16S rRNA gene clone library revealed that γ-Proteobacteria, R-Proteobacteria, and β-Proteobacteria were the dominant microorganisms in the PHA production system.

Introduction Polyhydroxyalkanoates (PHAs) have attracted much interest as an alternative to traditional plastics because they are biodegradable and biocompatible. However, most studies on PHA synthesis are conducted with pure cultures (in wild and recombinant strains) using petrochem-derived sub* Corresponding author phone: 86-21-65981263; fax: 86-2165986313; e-mail: [email protected]. 7734

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 20, 2009

strates (e.g., acetate or propionate) as carbon sources (1, 2), which require aseptic conditions and consume unrenewable resources. It has been reported that some microorganisms in activated sludge have the ability to accumulate PHA as an intracellular reserve of carbon and energy under unfavorable environmental conditions (3). Also, many kinds of wastes (in liquid or solid form) can be used as a renewable carbon sources. The combined use of activated sludge and waste organic carbon for PHA production does not need sterile conditions and can save petrochemicals. Thus, the use of activated sludge as the microorganisms and organic wastes as the carbon source for PHA production has become one of the focuses in the PHA biosynthesis field, and many efforts have been made toward the study of efficient PHA production techniques and use of renewable carbon sources (4-7). In the literature, anaerobic-aerobic- (An-Ae) and aerobicactivated sludge processes have been reported to synthesize PHA, but most studies used pure chemicals as the carbon sources. In the An-Ae process, PHAs are accumulated under anaerobic conditions with the uptake of short-chain fatty acids (SCFA) (8, 9). Nevertheless, as seen in the literature, PHA content is around 50% under anaerobic conditions. In recent years, PHA accumulation with aerobic-activated sludge process has been reported by several researchers. The feast and famine regimes (conditions with and without exogenous substrate available) give favorable conditions for microorganisms capable of storing organic matter such as acetic, propionic, and lactic acids as PHA (10, 11), which hereby can take up available substrate very fast and utilize it to gain dominant growth (12). With acetic acid as the carbon source, the aerobic PHA content in sludge can reach more than 70%. It is shown in the literature that most of the studies used petrochemicals to produce PHA, and the aerobic PHA synthesis was more efficient than the anaerobic one. The use of organic wastes as the renewable carbon sources for PHA synthesis has recently drawn much attention (5, 6, 13). Every year, large amounts of waste-activated sludge (WAS) are produced due to the widespread use of biological wastewater treatment processes. WAS is a troublesome waste, but it contains high levels of organic matter. In a previous investigation, we found that the SCFA production from WAS was significantly enhanced under alkaline conditions due to the increased hydrolysis of sludge protein and carbohydrate and decreased SCFA consumption by methanogens, and the alkaline fermentation liquid was primarily made up of SCFA, protein, and carbohydrate (14). If the WAS alkaline fermentation liquid could be used as the carbon source for PHA production, some petrochemicals (acetate, propionate, et al.) would be saved. Cai et al. reported that PHA could be produced by activated sludge microorganisms with volatile fatty acids generated by thermal alkaline fermentation of excess sludge (15). However, the maximal PHA content in activated sludge was only 56.5%. In this paper, an efficient PHA production process, aerobic feeding and discharge (AFD) was developed by using activated sludge as the biomass and WAS fermentation liquid as the renewable carbon source. In the AFD process, the activated sludge was cultured aerobically, and the WAS alkaline fermentation liquid was fed to the reactor in batches at different aerobic times. The mixture in the reactor was settled, and the supernatant was discharged before the next feeding. The effects of operational parameters such as feeding regimen, ammonia nitrogen (NH3-N) and soluble ortho-phosphate (SOP) concentrations, feeding times, and aeration time on PHA production were investigated. As the WAS alkaline fermentation liquid mainly consisted of 10.1021/es9014458 CCC: $40.75

 2009 American Chemical Society

Published on Web 09/11/2009

SCFA, protein, and carbohydrate, the roles of these carbon sources on PHA synthesis were studied. Also, some physical properties of the sludge-derived PHA were assayed. Finally, the microbial community involved in PHA synthesis by the AFD process was investigated by the method of the 16S rRNA gene clone library.

Materials and Methods Sludge Sources. The WAS used in alkaline fermentation was obtained from the secondary sedimentation tank of a municipal wastewater treatment plant in Shanghai, China. Seed sludge used for the enrichment of PHA production organisms was from a laboratory-scale anaerobic-aerobic sequence batch reactor (SBR) showing the profile of glycogenaccumulating organisms (16). WAS Alkaline Fermentation. WAS was prepared and fermented as described previously (14). The characteristics of WAS before fermentation are as follows: pH 6.7 ( 0.2, total suspended solids (TSS) 14786 ( 1677 mg L-1, volatile suspended solids (VSS) 9864 ( 1081 mg L-1, soluble chemical oxygen demand (SCOD) 396 ( 51 mg L-1, total chemical oxygen demand (TCOD) 14892 ( 1827 mg L-1, total carbohydrate 986 ( 108 mg COD L-1, total protein 8674 ( 953 mg COD L-1, and lipid and oil 182 ( 20 mg COD L-1. After fermentation, the mixture was centrifuged at 100 g for 10 min, and the supernatant, i.e., alkaline fermentation liquid was directly collected for PHA production unless otherwise stated. The characteristics of the alkaline fermentation liquid are as follows: SOP 137.6 ( 15.1 mg L-1, NH3-N 272.8 ( 26.2 mg L-1, SCOD 6569 ( 804 mg L-1, SCFA 2606 ( 263 mg COD L-1, soluble carbohydrate 187.7 ( 18.4 mg COD L-1, and soluble protein 1125.9 ( 100.3 mg COD L-1. Parent SBR Operation. Seed sludge was inoculated into an aerobic SBR with a working volume of 4 L. Each SBR cycle consisted of a 15 min feeding, 2 h aerobiosis, 1.5 h settling, and 15 min decanting of 1 L of supernatant, which was replaced with 1 L of fresh culture medium during the feeding time. The total cycle duration was 4 h. The sludge retention time (SRT) was maintained at 3 days by discharging the sludge before the end of aerobic period. Oxygen was supplied with an air pump, and the dissolved oxygen (DO) was around 6 mg L-1. The reactor temperature was maintained at 21 ( 1 °C. In each cycle, the reactor was mixed with the magnetic stirrer except in the settling and decanting phases. The WAS alkaline fermentation liquid diluted with tap water at a ratio of 1:4 was used as the feeding. The influent pH was adjusted to pH 7.0 ( 0.2 by adding 2 M HCl. After cultivation for 45 days, the sludge PHA content reached stability, and then the investigations of operational parameters affecting PHA synthesis were conducted. All of the following experiments were duplicated. Comparison among Different Feeding Regimens Affecting PHA Synthesis from Alkaline Fermentation Liquid. At the end of aerobiosis but before settling, 1.5 L of activated sludge mixture was taken from the parent SBR. The mixture was centrifuged at 100 g for 5 min to remove the supernatant, and the sludge was resuspended in tap water with a final volume of 0.75 L before being divided equally into three reactors (A1, A2, and A3) with working volume of 1.5 L each. A total of 0.75 L of alkaline fermentation liquid was added to A1, A2, and A3 with different pulse feeding regimens (corresponding to A1, A2, and A3, Figure 1). In A1 and A2, 0.75 L of fermentation liquid was fed to the reactors with three pulses (0.25 L in each pulse). In A1, the mixture was settled for 20 min after aeration, and 0.25 L of supernatant was discharged 10 min before 0.25 L of fermentation liquid was fed to the reactor at the time of 2.5 and 5 h. The operation involved in reactor A1 was called the AFD process. In A2, there were no settling and discharge phases before 0.25 L of fermentation liquid was fed at 2 and 4 h. In A3, all

FIGURE 1. Operations with the experiments of different feeding regimens affecting PHA synthesis. fermentation liquid was fed to the reactor at the beginning of the test. During the aerobiosis, the DO in three reactors was around 6 mg L-1. All reactors were maintained at 21 ( 1 °C and mechanically stirred in the aerobic time. Before feeding to the reactors, pH of the alkaline fermentation liquid was adjusted to pH 7.0 ( 0.2 by adding 2 M HCl, but the pH in the reactors was not controlled. Effect of Feeding Times (Loading of Fermentation Liquid) on PHA Production by the AFD Process. In this study, the fermentation liquid loading was controlled by the changes of feeding times. A total of 0.5 L of activated sludge mixture was taken from the parent SBR, centrifuged at 100 g for 5 min, resuspended in tap water with a final volume of 0.25 L, and put into a 1.5 L reactor. After 0.25 L of fermentation liquid was fed, the mixture was aerated for 2 h. In the next 0.5 h, the mixture was settled, and 0.25 L of supernatant was discharged. Then, the same operations were conducted for the second, third, and fourth pulse feeding, aeration (2 h), settlement, and discharge (0.5 h). After the fifth pulse feeding, the mixture was aerated for 10 h, and samples were taken per 2 h of aerobiosis for PHA analysis. All other operations were the same as those described above in the AFD process (Figure 1). Effect of NH3-N and SOP Concentrations on PHA Production by the AFD Process. Significant NH3-N and SOP were released during WAS fermentation. Before their influences on PHA production were evaluated, most of the released NH3-N and SOP were removed simultaneously in the form of magnesium ammonium phosphate hexahydrate (MgNH4PO4 · 6H2O). The MgCl2 · 6H2O and KH2PO4 were added to simultaneously recover NH3-N and SOP under the conditions of Mg/N ) 1.80, pH 10.41, and P/N ) 1.16, according to the method reported previously (17). After the first recovery, the precipitation (struvite) was removed from the fermentation liquid by centrifugation at 100 g for 10 min, and the supernatant was collected for a second NH3-N and SOP recovery to make sure that their concentrations in fermentation liquid were as low as possible. The added amounts of MgCl2 · 6H2O and KH2PO4 in the second recovery were according to the ratios of Mg/N ) 1.8 and P/N ) 1.13, respectively. After the second recovery, the supernatant with SOP 5.0 ( 0.6 mg L-1, NH3-N 72.8 ( 6.8 mg L-1, SCOD 5308 ( 648 mg L-1, SCFA 2552 ( 286 mg of COD L-1, soluble carbohydrate 162.8 ( 15.3 mg COD L-1, and soluble protein 995.2 ( 107.4 mg COD L-1, was used for PHA synthesis. A total of 3.5 L of activated sludge mixture taken from the parent SBR before the end of aerobic time was washed three times with tap water to remove the residual NH3-N and SOP and then resuspended in tap water with a final volume of 1.75 L before being divided equally into seven reactors (1.5 L each). A total of 0.75 L of fermentation liquid was added to each reactor. NH4Cl and KH2PO4 were supplied to control the VOL. 43, NO. 20, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7735

average concentrations of NH3-(2.60, 6.17, 9.74, 12.39, 2.60, 2.60, and 2.60 mM N) and SOP (0.08, 0.08, 0.08, 0.08, 1.03, 2.22, and 3.10 mM P) in the seven reactors. All seven reactors were then operated for PHA synthesis by the AFD process, which was described in Figure 1. Effect of Aerobic Time Configuration on PHA Production by the AFD Process. A total of 2 L of activated sludge mixture was taken from the parent SBR, centrifuged, resuspended in 1 L of tap water, and then divided equally into four reactors with working volumes of 1.5 L each. A total of 0.25 L of alkaline fermentation liquid was added into each reactor, and the mixture in four reactors was aerated for 1, 2, 3, and 1 h, respectively. After settling for 20 min, 0.25 L of supernatant was discharged within 10 min. A total of 0.25 L of fermentation liquid was then fed to each reactor, and the mixture was aerated for 1, 2, 3, and 2 h, respectively, before settlement and discharge of the 0.25 supernatant. Then, each reactor was fed with 0.25 L of fermentation liquid, aerated for 10 h, and sampled every 2 h for PHA assay. The aerobic time in the four reactors was marked as 1 h-1 h, 2 h-2 h, 3 h-3 h, and 1 h-2 h, respectively. Effect of Main Organic Carbon Composition of Alkaline Fermentation Liquid on PHA Synthesis. The main compositions of the fermentation were acetic acid, propionic acid, isovaleric acid, carbohydrate, and protein. To explore their effects on PHA production, we conducted batch experiments with synthetic wastewaters of acetic acid, propionic acid, isovaleric acid, bovine serum albumin (BSA, a model compound of protein used in this study), and glucose (a model compound of carbohydrate). Eight reactors were operated, and the carbon sources in these reactors were as follows: 30 mM C acetic acid (reactor 1), 30 mM C propionic acid (reactor 2), 30 mM C isovaleric acid (reactor 3), 30 mM C glucose (reactor 4), 30 mM C BSA (reactor 5), 30 mM C acetic acid plus 10 mM C glucose (reactor 6), 30 mM C acetic acid plus 10 mM C BSA (reactor 7), 30 mM C acetic acid plus 10 mM C glucose, and 10 mM C BSA (reactor 8). A total of 4 L of activated sludge from the parent SBR was centrifuged and resuspended in 2 L of tap water before being divided equally into eight reactors (1.5 L each). After the above organic carbons were added into each reactor, the volume of the mixture in each reactor was adjusted to 0.5 L with tap water. Then, the mixture was aerated for 10 h, and samples were taken per 2 h. The maximal PHA content was then reported. The DO in all reactors was controlled at 6 mg L-1, and the pH of sludge mixture before aeration in all reactors was adjusted to pH 7.0 ( 0.2 by the addition of 2 M NaOH or 2 M HCl. Analytical Methods. The measurements of SOP, NH3-N, SCFA, soluble carbohydrate and protein, sludge glycogen, lipid, SCOD, TSS, VSS, 3-hydroxybutyrate (3HB), 3-hydroxyvalerate (3HV), and 3-hydroxy-2-methylvalerate (3H2MV) were the same as described in our previous publications (14, 18). The total SCFA content was calculated as the sum of measured acetic, propionic, n-butyric, iso-butyric, nvaleric, and isovaleric acids. PHA was calculated as the sum of measured 3HB, 3HV, and 3H2MV. To analyze the molecular weight and thermal property of PHA produced from different carbon sources,we first synthesized PHA under optimal conditions with the carbon source of WAS alkaline fermentation liquid, acetic acid, propionic acid, isovaleric acid, BSA, and glucose, respectively. Then, PHA were extracted from the lyophilized cells with chloroform at 30 °C for 6 h. After filtration, the chloroform extract was concentrated, and the dissolved PHA were precipitated in chilled methanol (19). To calculate the weightaverage molecular weight (Mw), number-average molecular weight (Mn), and polydispersity index (Mw/Mn), we subjected the polymer to a Waters 150C gel permeation chromatography (GPC) analysis system (Waters, U.S.A.) using the 7736

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 20, 2009

method described by Lee et al. (7). Tetrahydrofuran was used as the eluent at a flow rate of 1.0 mL min-1, and a sample concentration of 1.0 mg mL-1 was applied. Polystyrene standards with low polydispersity were used to plot the standard curve. The thermal characteristics of PHA were measured with a TA Q100 differential scanning calorimetry (DSC) coupled with a liquid nitrogen cooling system (TA, U.S.A.) according to the method described by Bhubalan et al. (20). Melting point (Tm), glass transition temperature (Tg), and enthalpy of fusion (∆Hm) were recorded on a DSC instrument equipped with a cooling accessory. The weight of sample was about 20 mg, and the sample was heated from -80 to 200 °C at a heating rate of 20 °C min-1 under 70 mL min-1 of nitrogen purge. Before microbial community analysis, the activated sludge mixture was centrifuged at 10000 rpm for 5 min, and the total genomic DNA was extracted using the method described by Purkhold et al. (21) with minor modification. The centrifuged precipitate (0.5 g, wet weight) was washed three times with STET buffer (8% sucrose, 5% Triton X-100, 50 mM EDTA, 50 mM Tris, pH 8.0) and finally resuspended in 360 µL STET buffer. After addition of 40 µL of lysozyme (50 mg mL-1), the mixture was incubated for 10 min at 37 °C. A total of 20 µL of 10% sodium dodecyl sulfate (SDS) and 2 µL of proteinase K (20 mg mL-1) were then added, and the mixture was incubated at 37 °C for 60 min. A total of 50 µL of sodium chloride (5 M NaCl) and 50 µL of 10% cetyltrimethylammonium bromide (CTAB) were added and incubated at 65 °C for 10 min. After that, 0.5 mL of Tris-saturated phenol, 0.5 mL phenol-chloroform-isoamyl alcohol (25:24:1), and 0.5 mL chloroform-isoamyl alcohol (24:1) were used, respectively, to extract the nucleic acids, which were then precipitated by incubation with 0.1 volume of sodium acetate (3 M NaAc, pH 5.2) and 2 volumes of ethanol for 1 h at room temperature and subsequently centrifuged at 13000 rpm for 10 min. The pellets were washed with 500 µL of 70% ethanol, dried at room temperature, and finally resuspended in 50 µL of elution buffer (10 mM Tris, pH 8.5). The extracted DNA was checked by 1% agarose electrophoresis using ethidium bromide as the staining dye, and the amount and purity of DNA were determined spectrophotometrically at 260 and 280 nm. The nearly complete 16S rRNA gene fragments of bacteria were amplified using the universal bacterial primers 27F (5′AGAGTTTGATCCTGGCTCAG-3′) and 1492R (5′-GGTTACCTTGTTACGACTT-3′) (22). PCR amplification was carried out in a total volume of 25 µL containing 10 ng of template DNA, 1 × Ex Taq reaction buffer, 2U Ex Taq polymerase, 3.0 mM MgCl2, 0.2 mM dNTPs, and 0.5uM primers (TaKaRa, Japan) by using the StepOne Real-Time PCR System (Applied Biosystems, Foster City, CA). The amplification program consisted of an initial denaturation step of 94 °C for 5 min, 30 cycles of denaturation at 94 °C for 30 s, annealing at 55 °C for 30 s, and extension at 72 °C for 60 s, followed by a 5 min final extension at 72 °C. The amplified DNA was purified using the TaKaRa Agarose Gel DNA Purification Kit (TaKaRa, Japan), ligated into the pMD19-T vector (TaKaRa, Japan), and transformed into Escherichia coli DH5R cells (TaKaRa, Japan) with ampicillin selection and blue/white screening according to the manufacturer’s instructions. The white clones were randomly selected, and cloned inserts were identified using the vector primers M13-47 and RV-M. The positive clones were sequenced using the ABI PRISM 3730 automated DNA sequencer (Applied Biosystems) and then grouped into operational taxonomy units (OTUs) according to the 97% similarity threshold (23). The closest matching sequences in the GenBank database were searched using the BLAST program (24). Multiple alignments were generated using the ClustalX 2.0 (25), and then the phylogenetic tree was constructed with MEGA 4.0 (26) using the Jukes-Cantor model for the neighbor-joining algorithm (24).

FIGURE 2. Changes of PHA (a) and SCFA (b) with different feeding patterns. In A1, the fermentation liquid was fed in three pulses with the discharge of supernatant before the next pulse (AFD). In A2, the fermentation liquid was fed in three pulses, but no supernatant was discharged. In A3, all fermentation liquid was fed in one pulse. Error bars represent standard deviations of duplicate tests. Accession Numbers. The nucleotide sequences reported in this paper have been deposited in the GenBank, EMBL, and DDBL nucleotide database under the accession numbers FJ623276-FJ623387.

Results and Discussion Comparison among Different Feeding Regimens Affecting PHA Synthesis. As shown in Figure 2, three feeding patterns had different effects on PHA accumulation. With an increase in reaction time, the PHA content increased in all three reactors, and the maximal PHA contents in A1, A2, and A3 were obtained at 6 h, 6 h, and 4 h, respectively, which were 50.8% in A1, 41.7% in A2, and 38.6% in A3. It is obvious that splitting alkaline fermentation liquid into three pulses and discharging the supernatant before the next pulse led to higher PHA content. PHA synthesis and degradation have been reported to have a simultaneous existence during PHA production (27, 28). Under the conditions favorable for PHA synthesis, intracellular PHA content shows net accumulation. Under nonfavorable conditions, PHA net degradation is observed. It is also known that the higher organic carbon concentration is favorable for PHA accumulation (10). Obviously, feeding in one pulse resulted in a relatively higher initial SCFA concentration, which was the main reason for a higher specific PHA synthesis rate in A3. However, after 4 h, the SCFA concentration in A3 became relatively lower, which resulted in the net degradation and decrease in PHA content. Although in A1 and A2 the alkaline fermentation liquid was added in three pulses, the supernatant in A1 was discharged, while it was not discharged in A2. Thus, A1 had a relatively higher SCFA concentration than that of A2 most of the time, and a greater PHA content in A1 was obtained. Effect of Feeding Times (Loading of Fermentation Liquid) on PHA Production by the AFD Process. Figure 3 presents the variations of sludge PHA and glycogen with feeding times. The initial sludge PHA content was 1.58%. After the first feeding, the PHA content reached 21.4%, which was observed to increase to 63.3% with an increase in feeding times to four. Nevertheless, when the feeding times were increased to five, the PHA content was 64.7%, which showed very little increase compared to the feeding times of four. It seems that the suitable alkaline fermentation liquid feeding times was four. It is well-known that PHA is a more reduced compound than SCFA. The synthesis of PHA from SCFA requires reduction equivalent (NADH), and glycogen degradation usually provides the reduction equivalent under anaerobic conditions (29). In this study, it was observed that under aerobic conditions accompanying PHA accumulation and SCFA consumption (Figure S1 of the Supporting Information),

FIGURE 3. Variations of sludge PHA and glycogen with feeding times. Analyses of PHA and glycogen were conducted after the individual feeding and then aeration for 2 h. Error bars represent standard deviations of duplicate tests. sludge glycogen degradation was observed (Figure 3). As shown in Figure 3, with an increase in PHA content from 1.6% to 63.3%, 7.71 mM C glycogen degradation occurred. However, when the feeding times increased from four to five, the PHA content and glycogen concentration changed very little (from 63.3% to 64.7% and 10.95 to 10.80 mM C, respectively). It seems that under aerobic conditions, glycogen degradation provided the reduction equivalent for PHA synthesis from SCFA. Effect of Aeration Time Configuration on PHA Synthesis by the AFD Process. Different aeration time configurations showed different effects on SCFA consumption and PHA accumulation. As shown in Figure 4a,b, after the first pulse feeding, the highest PHA content increment occurred when aeration time configurations were 1-1 (25.5%) and 1-2 (25.8%), and the corresponding residual SCFA concentrations were in the range of 5-7 mM C. For the aeration time configurations of 2-2 and 3-3, during the first aeration not only the residual SCFA concentrations but also the observed PHA contents were lower than those of 1-1 and 1-2, which indicated that the extension of aeration time caused partial degradation of produced PHA. The same observations were made in the second pulse feeding. After the third pulse feeding, the highest final PHA content (65.7%) was observed with the aeration time configuration of 1-2, i.e., the aeration time after the first and second pulse feeding was 1 h and 2 h, respectively, and the aeration time after the third pulse of feeding was 3 h. Corresponding to this maximal PHA content, the residual SCFA concentrations after aeration were in the range of 5-7 mM C. VOL. 43, NO. 20, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7737

FIGURE 4. Effect of aeration time configurations on PHA content increase (a) and residual SCFA concentration (b). The notation 1-1 refers to the aeration time after the first and second pulse feeding was 1 h and 1 h, respectively, and the rest can be deduced by analogy. PHA and SCFA were assayed at the end of each aeration time with the first and second pulse feedings, whereas in the third pulse feeding, the PHA and SCFA were assayed every 2 h, and the data with maximum PHA content and corresponding SCFA concentration were reported. Error bars represent standard deviations of duplicate tests.

FIGURE 5. Effects of influent NH3-N (a) and SOP (b) concentrations on PHA accumulation. Error bars represent standard deviations of duplicate tests. It has been reported that intracellular PHA can be used as the carbon and energy sources of microorganisms under famine conditions (30), and PHA degradation will be observed once the external carbon source concentration becomes low such as 0.6, 0.8, and 1.9 mM C in this study. According to the data in Figure 4a,b it seems that a longer aeration time resulted in PHA degradation, but a shorter aeration could not provide enough time for SCFA consumption and PHA synthesis. Effects of NH3-N and SOP Concentrations on PHA Synthesis by the AFD Process. As illustrated in Figure 5a, the maximal PHA contents were 56.2%, 53.8%, 55.7%, and 54.8%, respectively, at NH3-N initial concentrations of 2.60, 6.17, 9.74, and 12.39 mM N, which indicated that the increase of NH3-N concentration had little influence on PHA synthesis. The same observation was made with the effect of SOP on PHA (Figure 5b, variations of NH3-N and SOP are shown in Figure S2 of the Supporting Information). The average initial concentrations of NH3-N and SOP in the PHA synthesis reactors were 9.74 mM N and 2.22 mM P, respectively, which suggested that it is unnecessary to remove the released NH3-N and SOP or to supplement either of them during PHA production with the WAS alkaline fermentation liquid as the renewable carbon source. PHA Production under Optimal Conditions. According to above study, the optimal conditions for PHA production can be summarized as follows. A total of 1 L of activated sludge mixture taken from the parent SBR was centrifuged for 5 min, resuspended in tap water with a final volume of 7738

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 20, 2009

0.5 L, and divided equally into two parallel reactors for duplicate tests (the working volume of each reactor was 1.5 L). A total of 1 L of alkaline fermentation liquid without NH3-N and SOP recovery was then fed to each reactor with four pulses (0.25 L in each pulse) by the AFD process. The aeration time in each reactor after the first, second, and third pulse feeding was 1 h, 2 h, and 3 h, respectively. After the fourth pulse feeding, the mixture was aerated for 10 h, and samples were taken per 2 h. The maximal PHA content, 72.9 ( 5.5%, was achieved at 10 h (Figure S3 of the Supporting Information). As far as we know, it was the highest PHA content with activated sludge as the microorganisms and wastes as the carbon source. As shown in Table 1, the 3HV fraction in PHA produced in this study was much higher than that reported in the literature. Effect of Main Organic Carbons of Alkaline Fermentation Liquid on the Accumulation and Composition of PHA. It was observed that the type of main organic carbons in WAS alkaline fermentation liquid influenced PHA content and composition (Table 2). Among the three SCFA, acetic acid resulted in the highest PHA content and 3HB fraction, while isovaleric acid showed the lowest PHA content. When glucose or BSA was used as the carbon source, there was no PHA accumulation observed in this study. The combined use of acetic acid plus glucose and/or BSA did not increase the accumulation of PHA significantly and did not show quite a difference in PHA composition compared with only acetic acid. All of these indicated that the activated sludge microorganisms enriched in this work could not use glucose or

TABLE 1. Comparison of PHA Composition and Content Achieved in This Study and Reported in Literature feeding process

cultures

PHA composition (mmoL C%)

substrate

diffusion flux of organic Ralstonia eutropha food waste fermentation carbon through membranes liquid aerobic dynamic feeding activated sludge acetic acid aerobic dynamic feeding activated sludge thermal alkaline fermentation liquid of excess sludge aerobic feeding and activated sludge alkaline fermentation liquid discharge (AFD) of waste activated sludge

PHA content (% of VSS) reference

3HB:3HV (96.5:3.5)

72.6

31

3HB 3HB:3HV (88.1:11.9)

78.5 56.5

11 15

3HB:3HV:3H2MV (73.5:24.3:2.2)

72.9

this study

TABLE 2. PHA Production and Composition with Different Organic Carbon Sourcesa PHA composition (mmoL C %)

a

organic carbon source

PHA content (% of VSS)

3HB

3HV

3H2MV

acetic acid propionic acid isovaleric acid glucose BSA acetic acid + glucose acetic acid + BSA acetic acid + glucose + BSA

29.7 ( 2.0 20.3 ( 1.3 5.9 ( 0.4 NDb ND 30.6 ( 2.2 29.1 ( 1.9 30.8 ( 2.1

97.8 ( 0.4 7.2 ( 1.0 92.4 ( 1.1 ND ND 96.9 ( 0.6 97.3 ( 0.5 96.3 ( 0.7

2.2 ( 0.4 83.0 ( 2.4 5.9 ( 0.8 ND ND 3.1 ( 0.6 2.7 ( 0.5 3.7 ( 0.7

0 9.8 ( 1.4 1.7 ( 0.3 ND ND 0 0 0

Data reported are the averages and their standard deviations of duplicate tests.

b

ND: Not detectable.

TABLE 3. Composition and Physical Properties of PHA Produced from Different Carbon Sourcesa composition (mmoL C%) carbon source

3HB

acetic acid 98.4 ( 0.3 isovaleric acid 90.8 ( 1.4 propionic acid 7.0 ( 0.8 alkaline fermentation liquid 73.5 ( 3.3 Biopol productsh 100-71.6

thermal properties

molecular weight

3HV

3H2MV

Tg (°C)b

Tm (°C)c

∆Hm (J g-1)d

Mw (× 105)e

PDIf

1.6 ( 0.3 7.4 ( 1.1 82.4 ( 1.2 24.3 ( 3.0 0-28.4

0 1.8 ( 0.3 10.6 ( 2.0 2.2 ( 0.3 0

NAg NA NA 2.68 ( 0.18 -8 to -9

157.9 ( 11.2 156.2 ( 8.8 96.1 ( 5.8 101.4 ( 6.2 102-175

83.8 ( 5.4 70.9 ( 5.5 58.9 ( 5.5 48.1 ( 3.0 NA

4.95 ( 0.31 6.13 ( 0.33 6.68 ( 0.33 8.50 ( 0.42 3.6-9.3

1.94 2.75 2.61 2.70 2.5-4.6

a Data reported are the averages and their standard deviations of duplicate tests. b Glass transition temperature (°C). Melting point (°C). d Enthalpy of fusion (J g-1). e Weight-average molecular weight (Da). f Polydispersity index: Ratio of weight-average molecular weight to number-average molecular weight. g NA: Data not available. h Reference 33.

c

BSA to accumulate PHA. In fact, it was observed in this study that there was no significant consumption of carbohydrate and protein during PHA production from WAS alkaline fermentation liquid under the optimal conditions (Figure S4 of the Supporting Information), which implied that when fermentation liquid was supplied as the carbon source, most of the PHA was from the fermentative SCFA. Molecular Weight and Thermal Properties of PHA Derived from WAS Alkaline Fermentation Liquid. It is known that the homopolymer PHB is highly crystalline, which limits its application (32), and the incorporation of 3HV or 3H2MV into PHB greatly influences its thermal and mechanical properties. Table 3 lists the partial physical properties of PHA produced from fermentation liquid under the optimal conditions. For comparison, the properties of Biopol products (33) and PHA synthesized from acetic acid, propionic acid, and isovaleric acid, respectively, are also included in Table 3. As shown in Table 3, with an increase in the 3HV percentage from 1.6% to 82.4% and 3H2MV from 0 to 10.6%, the Tm and ∆Hm of PHA decreased. PHA synthesized from alkaline fermentation liquid with 24.3% 3HB and 2.2% 3H2MV had a Tm of 101.4 °C, which was much lower than that of PHB homopolymer (180 °C) (34) and similar with Biopol products. Also, compared with the PHB homopolymer (34), the alkaline fermentation liquid-derived PHA had a lower Tg (2.68 against 4 °C), which was almost the same with Biopol products. However, Table

3 showed that with an increase in 3HV and 3H2MV percentages, the Mw increased. Although PHA obtained from propionic acid had the highest proportions of 3HV and 3H2MV, the alkaline fermentation liquid-derived PHA exhibited the highest Mw (8.5 × 105 Da), which was near the upper limit of Biopol products. This indicated that the PHA produced from alkaline fermentation liquid had a higher molecular weight. As also shown in Table 3, the alkaline fermentation liquid-derived PHA had a similar PDI value to that of Biopol products. Microbial Community Analysis of Parent SBR. With alkaline fermentation liquid as the carbon source of PHA synthesis, the microbial community structure of the parent aerobic SBR was assayed by the 16S rRNA gene clone library. Results showed that a high level of diversity among microorganisms was present in the current PHA production system (Figure S5 of the Supporting Information). The 112 clones were retrieved from the parent SBR and grouped into 53 OTUs. Among them, OTUs related to γ-Proteobacteria (42.0%), R-Proteobacteria (16.1%), and β-Proteobacteria (15.2%) were the most dominant, followed by those related to Bacteroidetes (5.4%), Chloroflexi (4.5%), candidate division SR1 (4.5%), Verrucomicrobiae (3.6%), Planctomycete (3.6%), ε-Proteobacteria (2.7%), and Sphingobateria (1.8%). On the basis of the analysis of the phylogenetic relationships of OTUs (Figure S5 of the Supporting VOL. 43, NO. 20, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7739

Information), the majority of γ-Proteobacteria clones were represented by OTU30 (17/112), OTU28 (1/112), and OTU29 (1/112), which were closely related to Thiothrix sp. It was reported that under pure culture conditions Thiothrix sp. displayed fast acetate uptake and storage in the form of PHB (ranging from 45% to 64% of the observed production) (35), which might be one reason for the highest percentage of γ-Proteobacteria (42.0%) in the current PHA synthesis system. The rest of OTUs belonging to γ-Proteobacteria were distributed in a single cluster containing two subgroups. These OTUs (including OUT31, 32, 33, 34, 40, 45, 50, and 53) were also numerically abundant (24.1%) in the aerobic SBR but not closely related to any known genus of γ-Proteobacteria. The R-Proteobacteria and β-Proteobacteria were also the major members in the parent SBR. In the case of R-Proteobacteria, the dominant clones were close to Rhodobacter sp. (4.5%) and Meganema sp. (6.3%). Rhodobacter sp. and Meganema sp. have been reported to be able to synthesize PHA under aerobic conditions (36, 37). For β-Proteobacteria, the dominant OUT (OTU38) belonged to Hydrogenophaga sp., which was also shown to produce PHA copolyesters aerobically with many kinds of carbon sources (38). The minority of the aerobic SBR community contained those related to Bacteroidetes, Chloroflexi, candidate division SR1, Verrucomicrobiae, Planctomycete, ε-Proteobacteria, and Sphingobateria. As described above, the microorganisms related to γ-, R-, and β-Proteobacteria were dominant at 73.3% in the current PHA synthesis reactor, suggesting that these microorganisms might be more adaptive to the WAS alkaline fermentation liquid culture environment than those minor ones and, thus, played an important role in PHA synthesis.

Acknowledgments This work was financially supported by the Fok Ying Tung Education Foundation (101080), Foundation of State Key Laboratory of PCRR, and Program for NCET in University (06-0373).

Supporting Information Available Figures S1-S5. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Choi, J.; Lee, S. Y. Economic considerations in the production of poly(3-hydroxybutirate-co-hydroxyvale-rate) by bacterial fermentation. Appl. Microbiol. Biotechnol. 2000, 53, 646–649. (2) Steinbuchel, A.; Bernd, F. Bacterial and other biological systems for polyester production. Trends Biotechnol. 1998, 16, 419–427. (3) Salehizadeh, H.; Van Loosdrecht, M. C. M. Production of polyhydroxyalkanoates by mixed culture: Recent trends and biotechnological importance. Biotechnol. Adv. 2004, 22, 261– 279. (4) Dionisi, D.; Majone, M.; Papa, V.; Beccari, M. Biodegradable polymers from organic acids by using activated sludge enriched by aerobic periodic feeding. Biotechnol. Bioeng. 2004, 85, 569– 579. (5) Dionisi, D.; Carucci, G.; Petrangeli, M.; Papini, P.; Riccardi, C.; Majone, M.; Carrasco, F. Olive oil mill effluents as a feedstock for production of biodegradable polymers. Water Res. 2005, 39, 2076–2084. (6) Bengtsson, S.; Werker, A.; Christensson, M.; Welander, T. Production of polyhydroxyalkanoates by activated sludge treating a paper mill wastewater. Bioresour. Technol. 2007, 99, 509– 516. (7) Lee, W.-H.; Loo, C.-Y.; Nomura, C.; Sudesh, K. Biosynthesis of polyhydroxyalkanoate copolymers from mixtures of plant oils and 3-hydroxyvalerate precursors. Bioresour. Technol. 2008, 99, 6844–6851. (8) Kasemsap, C.; Wantawin, C. Batch production of polyhydroxyalkanoate by low-polyphosphate-content activated sludge at varying pH. Bioresour. Technol. 2007, 98, 1020–1027. 7740

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 20, 2009

(9) Chua, A. S. M.; Takabatake, H.; Satoh, H.; Mino, T. Production of polyhydroxyalkanoates (PHA) by activated sludge treating municipal wastewater: Effect of pH, sludge retention time (SRT), and acetate concentration in influent. Water Res. 2003, 37, 3602– 3611. (10) Dionisi, D.; Majone, M.; Vallini, G.; Di Gregorio, S.; Beccari, M. Effect of the applied organic load rate on biodegradable polymer production by mixed microbial cultures in a sequencing batch reactor. Biotechnol. Bioeng. 2006, 93, 76–88. (11) Serafim, L. S.; Lemos, P. C.; Oliveira, R.; Reis, M. A. M. Optimization of polyhydroxybutyrate production by mixed cultures submitted to aerobic dynamic feeding conditions. Biotechnol. Bioeng. 2004, 87, 145–190. (12) Dircks, K.; Henze, M.; van Loosdrecht, M. C. M.; Mosbaek, H.; Aspegren, H. Storage and degradation of poly-β-hydroxybutyrate in activated sludge under aerobic conditions. Water Res. 2001, 35, 2277–2285. (13) Khardenavis, A. A.; Kumar, M. S.; Mudliar, S. N.; Chakrabarti, T. Biotechnological conversion of agro-industrial wastewaters into biodegradable plastic, poly β-hydroxybutyrate. Bioresour. Technol. 2007, 98, 3579–3584. (14) Yuan, H.; Chen, Y.; Zhang, H.; Jiang, S.; Zhou, Q.; Gu, G. Improved bioproduction of short-chain fatty acids from waste sludge under alkaline conditions. Environ. Sci. Technol. 2006, 40, 2025–2029. (15) Cai, M.; Chua, H.; Zhao, Q.; Sin, N. S.; Ren, J. Optimal production of polyhydroxyalkanoates (PHA) in activated sludge fed by volatile fatty acids (VFAs) generated from alkaline excess sludge fermentation. Bioresour. Technol. 2009, 100, 1399–1405. (16) Zhang, C.; Chen, Y.; Randall, A. A.; Gu, G. Anaerobic metabolic models for phosphorus- and glycogen- accumulating organisms with mixed acetic and propionic acids as carbon sources. Water Res. 2008, 42, 3745–3756. (17) Tong, J.; Chen, Y. Recovery of nitrogen and phosphorus from alkaline fermentation liquid of waste activated sludge and application of the fermentation liquid to promote biological municipal wastewater treatment. Water Res. 2009, 23, 2969– 2976. (18) Tong, J.; Chen, Y. Enhanced biological phosphorus removal driven by short-chain fatty acids produced from waste activated sludge alkaline fermentation. Environ. Sci. Technol. 2007, 41, 7126–7130. (19) Doi, Y.; Kunioka, M.; Nakamura, Y.; Soga, K. Nuclear magnetic resonance studies on poly(β-hydroxybutyrate) and a copolyester of β-hydroxybutyrate and β-hydroxyvalerate isolated from Alcaligenes eutrophus H16. Macromolecules 1986, 19, 2860–2864. (20) Bhubalan, K.; Lee, W.-H.; Loo, C.-Y.; Yamamoto, T.; Tsuge, T.; Doi, Y.; Sudesh, K. Controlled biosynthesis and characterization of poly (3-hydroxybutyrate-co-3-hydroxyvalerateco-3-hydroxyhexanoate) from mixtures of palm kernel oil and 3HV-precursors. Polym. Degrad. Stab. 2008, 93, 17–23. (21) Purkhold, U.; Pommerening-Roser, A.; Juretschko, S.; Schmid, M.; Koops, H.; Wagner, M. Phylogeny of all recognized species of ammonia oxidizers based on comparative 16S rRNA and amoA sequence analysis: Implications for molecular diversity surveys. Appl. Environ. Microbiol. 2000, 66, 5368–5382. (22) Lane, D. 16S/23S rRNA Sequencing. In Nucleic Acid Techniques in Bacterial Systematics; Stackebrandt, E., Goodfellow, M., Eds.; John Wiley & Sons: Chichester, U.K., 1991; pp 115-147. (23) Stackebrandt, E.; Goebel, B. Taxonomic note: A place for DNADNA reassociation and 16S rRNA sequence analysis in the present species definition in bacteriology. Int. J. Syst. Bacteriol. 1994, 44, 846–849. (24) Saitou, N.; Nei, M. The neighbor-joining method: A new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 1987, 4, 406–425. (25) Thompson, J.; Gibson, T.; Plewniak, F.; Jeanmougin, F.; Higgins, D. The CLUSTAL_X windows interface: Flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 1997, 25, 4876–4882. (26) Tamura, K.; Dudley, J.; Nei, M.; Kumar, S. MEGA4: Molecular evolutionary genetics analysis (MEGA) software, version 4.0. Mol. Biol. Evol. 2007, 24, 1596–1599. (27) Doi, Y.; Segawa, A.; Kawaguchi, Y.; Kunioka, M. Cyclic nature of poly(3-hydroxyalkanoate) metabolism in Alcaligenes eutrophus. FEMS Microbiol. Lett. 1990, 67, 165–169. (28) Taidi, B.; Mansfield, D. A.; Anderson, A. J. Turnover of poly(3hydroxybu-tyrate) (PHB) and its influence on the molecular mass of the polymer accumulated by Alcaligenes eutrophus during batch culture. FEMS Microbiol. Lett. 1995, 129, 201–206.

(29) Hood, C. R.; Randall, A. A. A biochemical hypothesis explaining the response of enhanced biological phosphorus removal biomass to organic substrates. Water Res. 2001, 35, 27582766. (30) Dionisi, D.; Renzi, V.; Majone, M.; Beccari, M.; Ramadori, R. Storage of substrate mixtures by activated sludges under dynamic conditions in anoxic or aerobic environments. Water Res. 2004, 38, 2196–2206. (31) Du, G.; Yu, J. Green technology for conversion of food scraps to biodegradable thermoplastic polyhydroxyalkanoates. Environ. Sci. Technol. 2002, 36, 5511–5516. (32) Lee, S. Y. Bacterial polyhydroxyalkanoates. Biotechnol. Bioeng. 1996, 49, 1–14. (33) Dai, Y.; Yuan, Z.; Jack, K.; Keller, J. Production of targeted poly(3hydroxyalkanoates) copolymers by glycogen accumulating organisms using acetate as sole carbon source. J. Biotechnol. 2007, 129, 489–497. (34) Doi, Y. Microbial Polyesters. VCH Publishers, New York, 1990.

(35) Majone, M.; Beccari, M.; Dionisi, D.; Levantesi, C.; Ramadori, R.; Tandoi, V. Effect of periodic feeding on substrate uptake and storage rates by a pure culture of Thiothrix (CT3 strain). Water Res. 2007, 41, 177–187. (36) Kranz, R. G.; Gabbert, K. K.; Locke, T. A.; Madigan, M. T. Polyhydroxyalkanoate production in Rhodobacter capsulatus: Genes, mutants, expression, and physiology. Appl. Environ. Microbiol. 1997, 63, 3003–3009. (37) Kragelund, C.; Nielsen, J. L.; Thomsen, T. R.; Nielsen, P. H. Ecophysiology of the filamentous Alphaproteobacterium Meganema perideroedes in activated sludge. FEMS. Microbiol. Ecol. 2005, 54, 111–122. (38) Choi, M. H.; Yoon, S. C.; Lenz, R. W. Production of poly(3hydroxybutyric acid-co-4-hydroxybutyric acid) and poly(4hydroxybutyric acid) without subsequent degradation by Hydrogenophaga pseudoflava. Appl. Environ. Microbiol. 1999, 65, 1570–1577.

ES9014458

VOL. 43, NO. 20, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7741