Organic Fraction of Municipal Solid Waste Recovery by Conversion

Subscriber access provided by REGIS UNIV

Article

Organic fraction of municipal solid waste recovery by conversion into added-value polyhydroxyalkanoates and biogas Francesco Valentino, Marco Gottardo, Federico Micolucci, Paolo Pavan, David Bolzonella, Simona Rossetti, and Mauro Majone ACS Sustainable Chem. Eng., Just Accepted Manuscript • DOI: 10.1021/ acssuschemeng.8b03454 • Publication Date (Web): 18 Oct 2018 Downloaded from http://pubs.acs.org on October 21, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sustainable Chemistry & Engineering

Organic fraction of municipal solid waste recovery by conversion into added-value polyhydroxyalkanoates and biogas

Francesco Valentinoa*, Marco Gottardob, Federico Micoluccic, Paolo Pavanb, David Bolzonellad, Simona Rossettie, Mauro Majonea

aDepartment bDepartment

of Chemistry, “La Sapienza” University of Rome, P.le Aldo Moro 5, 00185 Rome, Italy of Environmental Science, Informatics and Statistics, Via Torino 155, 30170 Venice

Mestre, Italy cDepartment

of Chemical Engineering, Lund University, PO Box 124, SE-221 00 Lund, Sweden

dDepartment

of Biotechnology, University of Verona, Strada Le Grazie 15, 37134 Verona, Italy

eWater

Research Institute, National Research Council (CNR), Via Salaria km 29, 300, 00015

Monterotondo (RM) Italy

Authors list *corresponding author: Francesco Valentino; [email protected] Marco Gottardo: [email protected] Federico Micolucci: [email protected] Paolo Pavan: [email protected] David Bolzonella: [email protected] Simona Rossetti: [email protected] Mauro Majone: [email protected]

ACS Paragon Plus Environment

1

ACS Sustainable Chemistry & Engineering 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Page 2 of 35

2

Page 3 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Abstract The integrated-multistage process proposed herein is a practical example of a biorefinery platform in which the organic fraction of municipal solid waste (OFMSW) is used as valued source for polyhydroxyalkanoates (PHA) and biogas production. Technical and economical feasibilities of this approach have been demonstrated at pilot-scale, providing a possible upgrade to traditional biowaste management practices, presently based on anaerobic digestion (AD). A pH-controlled OFMSW fermentation stage produced a liquid VFA-rich stream with high VFA/CODSOL ratio (0.90 COD/COD) that was easily used in the following aerobic stages for biomass and PHA production. The solid fraction was valorised into biogas through AD, obtaining energy and minimizing secondary flux waste generation. The reliable biomass enrichment was demonstrated by a stable feast-famine regime and supported by microbial community analysis. The selected consortium accumulated PHA up to 55% wt. Compared to the traditional AD process in an urban scenario of 900,000 AE, the integrated approach for OFMSW valorisation is preferable and it is characterized by an electrical energy production of 85.7 MWh/d and 1.976 t/d as PHA productivity. The proposed process was also evaluated as economically sustainable if the PHA is marketed from 0.90 €/kg as the minimum threshold to a higher market price.

Keywords Polyhydroxyalkanoates (PHA), Mixed Microbial Culture (MMC), Organic Fraction of Municipal Solid Waste (OFMSW), Volatile Fatty Acid (VFA), Anaerobic Digestion, Renewable Feedstock.


3


Page 4 of 35

Introduction European food production generates approximately 90 million tons of waste per year; this amount includes organic waste produced at the household level (40%) and bio-waste produced by the food service sector (15%) as well as at the retail level (5%).1 In urban areas, the total amount of organic waste also includes biodegradable garden and park waste. Current national legislations avoid landfilling organic wastes by treating them through thermal and/or biological processes (incineration, anaerobic digestion, composting) with a high disposal cost (75-125 €/ton).2 The recovery of added-value products from biowaste could be a strategy for both decreasing the cost of disposal and tackling the problems related to the increasing production of organic wastes; this can be achieved by using innovative technologies related to the circular economy, which is based on the concept of profit maximization from wastes without the generation of additional fluxes to be disposed of (closed loop technology).3,4 Even though biowastes vary greatly in composition, they are characterized by high moisture content and biodegradability. Organic waste quality is favoured by an efficient system of source separate collection, which makes it easier to obtain added-value products with biological processes not only via composting or anaerobic digestion5 but also with more recent biotechnologies that produce biopolymers and, in particular, the family of polyhydroxyalkanoates (PHA).6,7 This group of biodegradable thermoplastic polyesters are biologically produced by specific bacteria strains as intracellular carbon/energy sources. The current industrial PHA production processes are based on pure culture cultivation in sterile conditions.8 A sterile condition causes an increase in production cost (up to 8.0 €/kg) that consequently renders these polymers not cost-competitive with conventional oil based polymers.9 To decrease production costs, PHA can be produced from renewable organic resources using mixed microbial cultures (MMC) instead of pure cultures.10,11 PHA-accumulating organisms can


4

Page 5 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


be selected from the waste activated sludge from wastewater treatment, always available in the fullscale plants (WWTP), by applying transient conditions such as an aerobic dynamic feeding process (ADF). The selection of the MMC can be obtained through alternating feeding periods (feast and famine) with fermented feedstock rich in volatile fatty acids (VFA).12 The typical process applied for PHA production from MMC is the so-called three-step process.11 Many studies have been conducted in the last several years on MMC-PHA production using different types of waste, comprehensively listed in a recent review.11 The interest in biopolymer production from urban waste valorisation is relatively recent, even though it has been recognized as a key platform chemical raw material within the biorefinery framework.13,14 Few studies described PHA production from OFMSW sources, such as leachate,15 percolate16 and a mixture of primary sludge and OFMSW.17 These studies mainly concern laboratory scale tests. Pilot scale trials are unavoidably important to better understand the process feasibility and for integration in existing WWTPs to further their advancement. The pilot scale study proposed herein is an example of an integrated approach to treat and simultaneously valorise the OFMSW (main constituent of urban organic waste) through the production of PHA (via open MMC) and biogas (via anaerobic digestion), minimizing the need to dispose of secondary waste fluxes. The piloting facilities are located in the Treviso (northeast Italy) municipal full-scale Wastewater Treatment Plant (WWTP), where the OFMSW is collected and conventionally treated via anaerobic digestion.

Experimental Section (Materials and Methods) The process scheme of the units is presented in Figure 1. This integrated innovative approach includes a first unit (Stage I), consisting of a Continuous Stirred Tank Reactor (CSTR) for OFMSW fermentation. The fermented OFMSW, which are rich in VFA and are the precursors for PHA synthesis, are then conveyed to the following two aerobic stages: a Sequencing Batch Reactor (SBR)


5


Page 6 of 35

for biomass selection/enrichment under a feast-famine regime (Stage II); and a batch accumulation reactor for PHA production (Stage III). A final anaerobic digestion (CSTR) for residual and/or overflow conversion into biogas is also included. The operation of each unit was automated via a National InstrumentsTM cRIO device; centrifugation and feedstock filtration activities and VFA-rich stream feeding to Stage III were manual operations. The renewable feedstock was the OFMSW obtained from the door-to-door collection in the Treviso municipality. Next, the source-sorted OFMSW collection was transferred to the full-scale WWTP after its squeezing and homogenization, which was achieved by using a mobile continuous screw-press Elios BPM B600. This pretreatment was also used to remove inert materials, such as plastic and metals. Figure 1.

Anaerobic system for OFMSW fermentation and anaerobic digestion The pilot scale anaerobic process was characterized by a 200 L acidogenic fermentation CSTR with a hydraulic retention time (HRT) of 3.3 d, and an average organic load (OLR) of 20.5 ± 0.7 kgTVS/m3 d. The downstream solids separation was conducted by means of a coaxial filter bag (5.0 m porosity) equipped centrifuge. The liquid fraction was used in the aerobic PHA line. The discharged solid-rich fraction of the OFMSW fermentation, not suitable for PHA production, was used as feed for a 760 L CSTR anaerobic digester (AD), operated at 3.9 ± 0.4 kgTVS/m3d as OLR, and 12.7 d as HRT. The AD was inoculated with digestate from the Treviso full-scale plant. Fermentation and the AD were conducted at thermophilic conditions (55°C ± 0.1). The optimal pH value (above 5.0) in the fermentation stage was controlled by the digestate recirculation from the AD stage, since the digestate was rich in buffer agents. The recirculation ratio (RR) was defined as the volumetric fraction of digestate with respect to the total volume (digestate and pre-treated OFMSW) fed to the fermenter. The volume of recirculated digestate was automatically ACS Paragon Plus Environment

6

Page 7 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


dosed based on the measured pH, preventing unavoidable pH decreases due to VFA production and ensuring a stable fermentation activity.18

Aerobic stages for biomass selection/enrichment and PHA production Stage II consisted of a 140 L working volume. The SBR was fed with different feeding solutions: a) synthetic acetic acid solution, days 1st-49th (start-up); and b) pre-treated fermented OFMSW, days 50th129th. The synthetic solution consisted of acetic acid diluted with tap water and anaerobic digestate for the supply of nutrients. The SBR was inoculated with thickened activated sewage sludge from the Treviso WWTP. The HRT was set at 1.0 d, equal to the SRT, and the cycle length at 6 h (consisting of 3 min for dosing the feed, reaction time 354 min and biomass discharge 3 min).19 The SBR run was conducted for approximately 4.5 months, long enough to reach biomass adaptation (two weeks maximum with these operating parameters) and to characterize the process performance under a steady state. The reactor was aerated by means of linear membrane blowers (Bibus EL-S-250). The temperature (T) and pH were continuously measured but not controlled. The temperature changed seasonally between 14-18°C in March and 26-29°C in July. The pH fluctuated between 8.0-8.5 in the whole SBR cycle. The applied OLR was initially equal to 3.0 gCODSOL/L d and then maintained approximately 2.5 gCODSOL/L d by using fermented feedstock. The PHA storage potential of the selected biomass was exploited through fed-batch accumulation tests, performed with both synthetic (acetic acid), and pre-treated fermented OFMSW. The substrates were dosed in two separate spikes, setting the VFA/biomass ratio at 2.0 (COD basis) as a maximum value. The operative conditions of both aerobic reactors were chosen based on previous experimental laboratory scale proofs of concept.19,20 The dissolved oxygen and pH were monitored by Hamilton® industrial probes.


7


Page 8 of 35

Microbial community analysis in selection/enrichment SBR Fluorescence In Situ Hybridization Aerobic sludge samples (10 mL) were taken during the SBR operation at the end of the feast phase and fixed in formaldehyde as previously described.21 After fixation, the samples were kept at -20°C to be further analysed by Fluorescence In situ Hybridization (FISH). Oligonucleotide probes (Table S1; Supporting Information) targeting the Bacteria and Archaea domains and the main bacterial phyla were employed following the hybridization conditions reported elsewhere.22 The analysis was performed by epifluorescence microscopy (Olympus, BX51). Images were captured with an Olympus F-View CCD camera and handled with CellˆF software (Olympus, Germany). DNA extraction DNA extraction for the NGS analysis was performed on samples collected throughout the SBR operation. In detail, 10 ml sludge samples were collected and centrifuged at 15,000 g for 15 min. The pellet was processed for DNA extraction with a Power Soil DNA extraction kit (MoBio, Italy) following the manufacturer’s instructions. Purified DNA from each sample was eluted in 100 L sterile Milli-Q water, and 10 ng of extracted DNA was used for the following NGS analysis. Next Generation Sequencing (NGS) A 16S rRNA Amplicon Library Preparation (V1-3) was performed as detailed in Matturro et al. (2016).22 10 ng of extracted DNA was used as a template in the PCR reaction (25 μL) containing dNTPs (400 nM of each), MgSO4 (1.5 mM), Platinum® Taq DNA polymerase HF (2 mU), 1X Platinum® High Fidelity buffer (Thermo Fisher Scientific, USA) and barcoded library adaptors (400 nM)

containing

V1-3

primers

(27F:5’-AGAGTTTGATCCTGGCTCAG-3’;

534R:5’-

ATTACCGCGGCTGCTGG-3’). All PCR reactions were run in duplicate and pooled afterward. The amplicon libraries were purified using the Agencourt® AMpure XP bead protocol (Beckmann Coulter, ACS Paragon Plus Environment

8

Page 9 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


USA). The library concentration was measured with a Quant-iTTM HS DNA Assay (Thermo Fisher Scientific, USA) and the quality validated with a Tapestation 2200, using D1K ScreenTapes (Agilent, USA). The purified sequencing libraries were pooled in equimolar concentrations and diluted to 4 nM. The samples were paired end sequenced (2 × 301 bp) on a MiSeq (Illumina) using a MiSeq Reagent kit v3, 600 cycles (Illumina) following the standard guidelines for preparing and loading samples on the MiSeq. A 10% Phix control library was spiked in to overcome the low complexity issue often observed with amplicon samples. The data analysis was performed as detailed in Matturro et al. (2016).22

Analytical methods and calculation Suspended solids (total and volatile, TSS and VSS), ammonia and PHA analyses were performed as described in Valentino et al. (2014);19 total and volatile solids (TS and TVS), COD, VFA and phosphate were quantified as illustrated by Micolucci et al. (2014).23 The gas production in the anaerobic reactors was monitored by two flow meters (Ritter CompanyTM, drum-type wet-test volumetric gas meters). A comprehensive methods explanation is given in the Supporting Information. SBR and accumulation stages calculations were made as previously indicated in Valentino et al. (2014).19 Detailed explanations are also given in the Supporting Information. The mass and energy balance, as well as the energy yield, of the proposed multi-stage process has been compared to a conventional single stage anaerobic digestion process (CSSP). The aim of this comparison was to evaluate the different energy yields between the two scenarios and estimate the minimum economic value of the PHA produced to cover the economic income from the amount of biogas that would have been produced in CSSP. For the mass balance in the CSSP, an SGP of 0.75 Nm3/kgTVS has been considered.24


9


Page 10 of 35

The price of electrical energy and the disposal cost of the digestate (at a 25% TS level) were assumed to be 130 €/MWh (no incentive) and 100 €/ton, respectively. The cost of tap water was considered 5.55 €/m3. All the parameters were adapted to a scaled-up version of both processes (reference parameters and boundary conditions are given in Table S2; Supporting Information). The payback time of the investment for the proposed process (two-stage anaerobic process coupled with PHA production) was calculated as an upgrade of the already existing full-scale CSSP. The payback period was defined as the time of recovery of the invested amount that was used to build up the fermentation reactor and PHA-related reactors, in a hypothetical platform of 900.000 Personnel Equivalent (PE). The capital cost for the fermenter was assumed to be 750 €/m3 reactor.25 For both aerobic PHA production stages, a cost of 5.54 €/PE was assumed, by analogy with the traditional aerobic wastewater treatment.26 For the PE quantification, the influent COD mass flow into the PHA production line was considered.

Results and Discussion Characteristics of the pre-treated OFMSW The pre-treated OFMSW was analysed twice a week after its collection inside the full-scale WWTP. For the general chemical-physical characteristics, the pre-treated OFMSW showed an average dry matter content of 280 ± 44 gTS/kg and 250 ± 33 gTVS/kg. Indeed, the volatile organic matter accounted for 90% ± 2% of the total solids, meaning that the fraction of putrescible and fermentable material was high enough to support the fermentation process. This important feature of the feedstock was favoured by the source-sorted collection inside the municipality. The COD values were typically greater than 900 gCOD/kgTS (906 ± 58). The content of nitrogen and phosphorus was 27 ± 3 gN/kgTS and 4.0 ± 0.2 gP/kgTS on average, respectively. For the nutrients, this substrate showed a COD:N:P ratio of 100:2.9:0.4 (on average), not particularly rich in nutrients but potentially usable for the


10

Page 11 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


following aerobic stages (Table S3; Supporting Information). To sustain the set OLR in the fermenter, the pre-treated OFMSW was diluted almost 4 times with tap water and recirculated digestate.

Fermentation stage The objective of the fermentation step was to produce a VFA-rich stream from the selected bacteria to be efficiently used for PHA production. If related to PHA-MMC technology, this step must be performed by considering two crucial aspects: a) a high VFA/CODSOL ratio compared to that of unfermented stream, and b) the VFA distribution must remain as stable as possible during the entire operation. While a high VFA/CODSOL ratio increases the biomass selection efficiency in the SBR, a stable VFA spectrum implies a predictable and reproducible PHA monomers production and, as a consequence, a stability in physical and mechanical PHA properties.7,11 Under a pH-controlled fermentation strategy, the RR was in the range of 0.43 – 0.59 v/v. Consequently, the pH fermentation value was maintained between 5.2 - 5.6 (Figure S1; Supporting Information). Through a stable pH maintained by the automated control strategy, the fermentation activity was consequently stable and optimal for VFA production. As matter of fact, the alkalinity of the fermented feedstock (894 ± 104 mg CaCO3/L, pH 5.0 ± 0.2) also excluded the necessity of NaOH addition into the medium, at the same time avoiding a pH decrease (inhibiting for the fermentative microorganisms). Nearly two weeks of operation (corresponding to approximately 4.2 HRTs) were necessary to achieve process stability in terms of COD solubilization (0.20 ± 0.03 CODSOL/CODin). VFA yield (0.31 ± 0.02 CODVFA/CODin) and VFA concentration (16.0 ± 0.7 gCOD/L). In this stable operation period, the average VFA/CODSOL ratio was equal to 0.91 ± 0.02, almost six times higher than the initial value of the unfermented stream (0.16 ± 0.09). As consequence, the quantified fermented liquid stream was equal to 3.86 kgCOD/d, of which 1.05 kgCOD/d was in the soluble fraction and 2.81 kg COD/d was in the particulate fraction.


11


Page 12 of 35

In terms of VFA distribution, the fermented stream was primarily composed of butyric (38.0 ± 0.3% COD basis), acetic (21.5 ± 0.2%), propionic (12.7 ± 0.3%), valeric (11.6 ± 0.5%) and caproic (10.0 ± 0.1%) acids (Figure 2). Lower levels of isobutyric (3.80 ± 0.06%), isovaleric (1.61 ± 0.04%) and isocaproic (0.72 ± 0.04%) acids were also detected. Indeed, the net prevalence of acids with even numbers of carbon atoms means a higher 3-hydroxybutyrate (HB) monomer synthesis compared to the 3-hydroxyvalerate (HV) synthesis, which is generally favoured by the presence of VFA with an odd number of C atoms (propionic, valeric and isovaleric acid). Figure 2.

Anaerobic digestion stage The fermented product was subjected to a solid/liquid separation process by a centrifuge. The filtered mass flow was sent to a PHA production unit, and the solid-rich mass flow fraction (“cake”) was sent to the anaerobic digestion stage. As for the general chemical characteristics, the cake had an average solids content of 200 ± 35 gTS/kg and 175 ± 29 gTVS/kg (Table S3; Supporting Information). The COD:N:P ratio was equal to 100:2.7:0.56, which meant there were enough nutrients to sustain the anaerobic process; the nitrogen content was not extreme enough to inhibit the methanogens as a consequence of the adopted pHcontrolled fermentation strategy. The ammonia digestate concentration was 900 ± 67 mg N-NH4+ /L, much lower than the inhibition value for the methanogenic activity.23 The stability of the process was also proved by the VFA/partial alkalinity ratio, which was less than 0.3 mg acetic acid/mg CaCO3 for the overall period of experimentation (approximately 120 days).24 The average specific biogas production (SGP) was 0.71 Nm3/kgVS fed with a composition of 65% v/v and 35% v/v of methane and carbon dioxide, respectively. Considering the overall process (fermentation and AD stages), the percentage of methane in the biogas was 53%, while the remainder


12

Page 13 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


was carbon dioxide and other compounds in a minor percentage. This result indicates that biogas production from the solid-rich overflow was a part of a multi-step integrated approach, which helps to increase the profit of the overall process from the discharged substrate fraction that is not suitable for PHA production.

Biomass selection/enrichment in SBR (Stage II) The start-up of the aerobic selection stage was initiated by using acetate synthetic feeding solution at 3.0 gCOD/L d as OLR. Since the system was not equipped with a temperature (T) control, the initial SBR temperature was slightly higher than 10°C (end winter), achieving almost 30°C in the last part of the experimentation (midsummer). This difference strongly affected the culture adaptation to the imposed feast-famine conditions. More than 40 STRs were needed to achieve a feast/cycle length ratio below 0.2 h/h necessary for a stable feast-famine regime with a satisfying biomass selective pressure.11,12 Most of the studies describing the MMC selection with the same process configuration were performed under a T-control at 25°C; these examples reported 10-15 SRTs maximum19,27,28 as window times to achieve a stable storage response, also in agreement with a change in the microbial community and stabilization of one major PHA-storing phylotype.19 The length of the feast phase stabilized after approximately 50 SRTs, as the temperature started to increase from approximately 20°C (Figure 3A). The temperature values (daily average values) progressively increased until the end of the operation, generating an increasing selective pressure over the biomass as demonstrated by the decreasing feast/cycle length ratios, often below 0.1 h/h in the last 50 days of operation. Before a reasonable feast-famine pressure was reached, strong fluctuations were observed for both the feast/cycle length ratio and the PHA concentration (Figure 3B), with values abundantly higher than average trends and positively correlated with each other. Indeed, the feast phase increased (i.e.,


13


Page 14 of 35

CODSOL uptake rate decreased) as PHA concentration in the medium increased. When the feast phase started to be consistently short, the PHA concentration profiles displayed net and constant differences between the end of the feast and the end of the cycle, as an indication of a stable storage response. This evidence was already identified in previous lab-scale SBR experiments, and usually discussed based on PHA concentration28,29 or storage yield trends:30 in the acclimation process, the biomass fits faster to storing PHA than consuming it, creating an increase in PHA cell content. The biomass storage capacity is consequently saturated, but recovered when the PHA concentration decreases, and stabilized if process conditions remains unchanged. Similar biomass behaviour was observed in this pilot scale approach, but with longer adaptation periods. A combined effect of increasing temperature and feedstock shift (day 49) probably increased the capacity of the culture to progressively adapt itself to the dynamic feeding regime. Figure 3. The feedstock shift caused only an immediate increase of the feast/cycle length ratio up to 0.4 h/h; however, the functional feast-famine regime was rapidly re-established since the biomass was largely acclimated to the process conditions. Table 1 summarizes the main parameters that were monitored and quantified in the SBR, in both periods when the acetic and fermented feedstock were used as the substrate. The quantification of the biomass storage properties, in terms of yields (YP/Sfeast, YP/VFAfeast) and rates (qPfeast), confirmed the efficiency of the selective pressure over the biomass when the SBR was fed with fermented OFMSW. These parameters were comparable with previous investigations applying the same three-step-based technology.31,32 Table 1.

Fed-batch PHA accumulation reactor (Stage III)


14

Page 15 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Preliminary accumulations were conducted to exploit the storage potential of the selected consortium by using acetate synthetic feeding without nutrient addition. Subsequent tests were carried out with fermented OFMSW, assessing the biomass PHA production capacity and the process productivity. All the fed-batch tests were performed after the 50th days of SBR operation, when the imposed selective pressure was stable and high enough to ensure a satisfying PHA accumulation performance. Acetate accumulations led to a 0.37-0.42 gPHA/gVSS PHA content, consistent with results previously reported by using synthetic acetate (0.12-0.76 gPHA/gVSS)33,34 or a VFA mixture solution (0.14-0.51 gPHA/gVSS).19 The accumulation capacity of the biomass was better expressed with fermented OFMSW, with a PHA content in the range 0.39-0.52 gPHA/gVSS. However, PHA saturation levels in the biomass were not achieved, even though the storage response was the prevailing mechanism of substrate removal alongside biomass growth. Recent investigations of the use of fermented OFMSW or similar sources reported a wide range of different MMC-PHA accumulation capacities. Amulya and co-workers achieved a maximum level of 0.24 gPHA/gVSS with fermented oil-free filtered food waste.35 The use of percolate was more comparable with the present investigation; the PHA biomass content was in the range of 0.40-0.48 gPHA/gVSS.16 Korkakaki and co-workers (2016)15 achieved even better performances by using pretreated leachate (close to 0.80 gPHA/gVSS), even though the biomass selection step was performed with a solution largely made up of synthetic VFA (75%-90% volume based). Although with some variability due to the fermentation performance and/or the maximum VFA content achieved, fed-batch accumulations indicated that fermented OFMSW triggered higher accumulation rates and productivities than did acetate with no nitrogen and phosphorus addition (Table 2). Table 2. Such differences in the accumulation response to different levels of nutrients have been widely investigated, and even the presence of nutrients at certain levels were associated with increased


15


Page 16 of 35

polymer productivities due to concurrent PHA storage and active biomass growth.11,36 Indeed, PHA storage was contingent upon nutrient level as well as growth response. Higher biomass growth response and yields (YX/Sbatch = 0.19-0.28 CODXa/CODSOL) were obtained with fermented OFMSW than with strongly growth-limiting nutrient levels of acetate solution (YX/Sbatch = 0.12-0.15 CODXa/CODSOL). The PHA storage response was greater in acetate accumulations and higher in terms of yield (YP/Sbatch = 0.61-0.64 CODPHA/CODSOL); however, the lack of nutrient availability strongly limited the PHA production to a large extent, since the PHA storage was poorly supported by the new storing active biomass growth. For this reason, PHA productivities were doubled or even tripled when using fermented OFMSW (0.28-0.49 vs 0.16-0.18 gPHA/L h with OFMSW and acetate, respectively) as a result of a not negligible growth response and faster kinetics for both substrate uptake and storage specific rates (Table 2). In these cases, a sustained PHA content (apparently not at saturation level, Figure S2; Supporting Information) alongside a growth of PHA-storing biomass increased PHA productivities.

PHA-accumulating microbial community in SBR The microbial composition and structure of the communities selected in the SBR were estimated by using in situ detection methods for cell-based quantification and high-throughput sequencing. Bacteria were the primary microbial component, accounting for 50-80% of the total biomass (Figure S3; Supporting Information). A FISH analysis revealed a marginal presence of archaeal cells, approximately 4% with acetate feeding (days 28, 50) and 10% of the total population with fermented OFMSW (days 62, 82, 91 and 108). The analysis with bacterial phylum specific probes revealed the dominance of Betaproteobacteria during the acetate feeding: 70% of the total bacteria (day 28, Figure S4; Supporting Information), which almost completely consisted of members of the Azoarcus/Thauera group (Figure S5; Supporting


16

Page 17 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Information). From the beginning of the SBR operation, a six-fold increase in the relative abundance of Azoarcus/Thauera was observed revealing members of this group as the primary PHA-accumulating microorganisms in the SBR with acetate as the sole carbon source, in accordance to previous findings.28,37,38,39,40 Temporal fluctuations of the microbial population were observed during the operation with fermented OFMSW

(Figure

S6;

Supporting

Information).

Members

of

Proteobacteria

and

Cytophaga/Flexibacter/Bacteroidetes represented the main components at day 91, whereas a marked shift towards the dominance of Betaproteobacteria was found at day 108 along with the increase in temperature most likely driving the observed changes of the PHA accumulating SBR biomass. Conversely, from the start-up phase with synthetic feeding, the Thauera/Azoarcus group represented only a portion of the total Betaproteobacteria (Figure S4; Supporting Information) and gradually decreased until reaching the lowest value (~12% of the total Betaproteobacteria) at day 108. Highthroughput sequencing showed the occurrence of known PHA-accumulating microorganisms including Acidovorax spp. and Hydrogenophaga spp., the latter representing 52-79% of the OTUs affiliated with Betaproteobacteria (Figure S6-A; Supporting Information). Additionally, genera Amaricoccus spp., Meganema spp., Rhizobium spp. and Rhodobacter spp. were found (Figure S6-B; Supporting Information), as well as a variety of other Alphaproteobacteria occurring at very low relative abundances. Some of the taxa found to dominate with fermented OFMSW, such as Acidovorax and Hydrogenophaga, were previously found prevailing in MMC fed with a synthetic soluble fraction of municipal wastewater41 under ADF conditions or with fermented waste activated sludge under aerobic/anaerobic operating conditions.42 Both genera, commonly found in activated sludge, are aerobic even though some species are capable of the heterotrophic denitrification of nitrate. In addition


17


Page 18 of 35

to chemoorganotrophic metabolism, some strains of Hydrogenophaga are chemolithoautotrophic, using the oxidation of H2 as an energy source and CO2 as a carbon source. To date, very little is known regarding the key-microbes catalysing the PHA storage from OFMSW, and most of the indications were obtained by adopting low-resolution monitoring tools such as DGGE without providing any quantitative data as, instead, has been performed in this study. In particular, Brachymonas denitrificans, Corynebacterium, Xanthobacter and Azorhizobium were found with raw or pre-treated leachate obtained from OFMSW with VFA primarily composed by acetate, propionate and butyrate.15 The influence of aerobic and anoxic conditions on PHA production was evaluated in SBRs treating un-fermented and fermented food waste.43 A DGGE analysis was performed only on the biomass selected in the SBR treating unfermented food waste under anoxic conditions and revealed the occurrence of genera belonging to gammaproteobacteria (e.g., Pseudomonas, Aeromonas and Acinetobacter) followed by members of Betaproteobacteria, Epsilonproteobacteria, Bacteroidia and Firmicutes.

Mass and energy balance assessment of the integrated platform The data analysis of each separate pilot unit was transferred to a single industrial scheme, ideally identified in an urban scenario of 900,000 PE with a specific OFMSW production of 0.3 kg/PE d.24 Considering a recovery of 75% TS from the pre-treatment screw-press stage and a dry matter content of 28%, the amount of OFMSW to be treated is 60,143 kgTS/d which corresponds to 53,865 kgTVS/d. The mass balance discussed in this paragraph is illustrated in detail in Figure 4A. The pre-treated OFMSW stream conveyed to the industrial plant for PHA and biogas production needs to be abundantly diluted with tap water to reduce the TS level to 7% w/w and to maintain the applied OLR in the fermenter at approximately 20 kgVS/m3d. The gaseous effluent flow rate out of the


18

Page 19 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


acidogenic fermenter is 8,080 Nm3/d, corresponding to 10,479 kgTVS/d, almost 20% of the TVS influent amount (53,865 kgTVS/d). The solid/liquid separation unit allows the recovery of more than 70% of the volumetric OFMSW liquid flow rate: 574,839 kg/d of fermented stream with a TS content of approximately 0.5% w/w or below and a VFA/CODSOL ratio of 0.90 can be used for both aerobic PHA production steps. On the other hand, the more concentrated TS stream (cake) can be further valorised into biogas. Regarding the PHA production line, the liquid fermented OFMSW needs to be split into two secondary fluxes, properly quantified based on the OLR applied to both the SBR and accumulation reactors. The SBR step accounts for 49% of the influent carbon source (281,671 kg/d) and 100% of the dilution water consumption, as largely proven in this PHA process configuration.11 the dilution water is included because the SBR was modelled at 3.0 kgCODSOL/m3 d as the OLR, lower than the 40 kgCODSOL/m3 d which is the current OLR applied in the PHA accumulation reactor. Accordingly, the volumes of the selection and accumulation reactors are 1,690 m3 and 136 m3, respectively. The storage yield for both aerobic stages is 0.35 CODPHA/CODSOL, based on the consumed kgCODSOL/d for the biomass production (in the SBR) and the PHA synthesis (accumulation). The modelled multi-stage process has a production potential of 1,976 kgPHA/d; this means an overall polymer productivity of 1.08 kgPHA/m3 d. The cake previously discharged out of the PHA line represents the anaerobic digestion feed. Based on 0.69 Nm3/kgTVS as the digester SGP, the anaerobic digestion process produces approximately 28,410 Nm3/d of biogas, primarily composed of CH4 (65% v/v or 18,467 Nm3CH4/d) and CO2 (35% v/v, 9,944 Nm3CO2/d). The energy balance comparison between the proposed process and the CSSP was conducted using the same full-scale scenario of mass balance (900,000 PE basin). All the thermal and electrical energy


19


Page 20 of 35

items are summarized in Table 3 (reference parameters and boundary conditions in Table S2; Supporting Information). Table 3. Considering the thermal yield of the Combined Heat & Power unit (CHP) of 0.5,24 the thermal energy produced is approximately 464,828 MJ/d and 419,854 MJ/d from the CSSP and the proposed process, respectively. In both scenarios, the thermal balance is closed positively because the estimated thermal energy requests are approximately 40% (185,774 MJ/d) and 82% (342,314 MJ/d) of the thermal energy produced for the CSSP single stage and the proposed multi-stage processes, respectively. Regarding the income from the electrical energy produced, the estimated production is approximately 103.4 MWh/d for the CSSP and 93.4 MWh/d for the proposed process. By considering the overestimated electrical energy consumption for the aeration in the two aerobic steps (7.7 MWh/d), the net production is 85.7 MWh/d (Table 3). Considering the electrical energy value, the disposal cost of the digestate and the tap water cost, a gap of approximately 637,354 €/y exists between the CSSP and the proposed process. This gap can be easily covered since 1.976 t/d of produced PHA has to be marketed at the cost-efficient value of 0.90 €/kg, as a minimum threshold. A reasonable market price can overcome the economic income from that fraction of biogas, which is not produced in the CSSP, as a demonstration of the economical sustainability of the proposed multi-stage process. The capital cost for the implementation of the fermentation reactor (2,693 m3) is estimated as 2,020,000 €. Regarding the capital cost assessed for both the PHA production steps, the estimated PE is 88,621, corresponding to an incoming mass flow of 10,635 kgCOD/d. Therefore, the capital cost for the PHA production platform is assumed to be 490,000 €. Figure 4B shows the cumulative income versus the capital cost for the realization of the fermentation reactor and both PHA production reactors. The upgrade payback time of the whole platform is estimated to be rather short (less than two years) if the PHA is sold at the quantified minimum value of 0.90 €/kg. This promising outcome can be even more


20

Page 21 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


profitable with a higher PHA market price even though the market scenario of the PHA produced from the MMC still needs to be established in relation to the required purity and properties of the material obtained after downstream processing. Figure 4

Supporting Information Tables showing the oligonucleotide probes used for FISH analysis and employed in targeting Bacteria and Archaea domains and the main bacterial phyla; reference parameters and boundary conditions for energy balance as the data used for energy balance and the comparison of produced energy between the conventional AD process and the multi-step proposed platform; main features of the pre-treated unfermented OFMSW, fermented OFMSW and residual solid-rich fraction (“cake”) of pre-treated OFMSW. Figures showing the recirculation ratio (RR) and pH trends in the fermentation reaction; PHA-VFA and active biomass and PHA biomass content profiles in two accumulations performed with synthetic acetate solution and fermented OFMSW; relative abundance of total bacteria and archaea in SBR during acetate and fermented OFMSW feeding; the composition of total bacteria estimated by FISH analysis with group specific oligonucleotide probes; relative abundance of Betaproteobacteria and Thauera/Azoarcus estimated by FISH analysis; affiliation of the primary OTUs belonging to Betaproteobacteria and Alphaproteobacteria estimated by Illumina sequencing in samples taken in SBR during the fermented OFMSW feeding. Descriptions of methods for the VFA analysis, PHA analysis, and Carbon Dioxide and Methane analysis. Calculations of main performances and performances related to both aerobic PHA production steps.


21


Page 22 of 35

Acknowledgements The authors gratefully acknowledge the Italian National Institute for Insurance against Accidents at Work (Istituto Nazionale per l’Assicurazione contro gli Infortuni sul Lavoro, INAIL) for financial support of this research, in the frame of national Call BRIC 2013-2015 (ID15). The hospitality of Alto Trevigiano Servizi srl is also gratefully acknowledged.


22

Page 23 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


REFERENCES (1) European Commission Website 2010; Preparatory study on food waste across EU-27 for the European Commission, ec.europa.eu/environment/eussd/pdf/ bio_foodwaste_report.pdf. (2) Pfaltzgraff, L. A.; Cooper, E. C.; Budarin, V.; Clark, J. H. Food waste biomass: a resource for highvalue chemicals. Green Chem. 2013, 15, 307-314, DOI 10.1039/c2gc36978h. (3) Lieder, M.; Rashid, A. Towards circular economy implementation: a comprehensive review in context of manufacturing industry. J. Cleaner Production. 2016, 115, 36-51, DOI 10.1016/j.jclepro.2015.12.042. (4) Fava, F.; Totaro, G.; Diels, L.; Reis, M. A. M.; Duarte, J.; Beserra Carioca, O.; Poggi-Varaldo, H. M.; Sommer Ferreira, B. Biowaste biorefinery in Europe: opportunities and research & development needs. New Biotechnol. 2015, 32, 100-108, DOI 10.1016/j.nbt.2013.11.003. (5) Girotto, F.; Alibardi, L.; Cossu, R. Food waste generation and industrial uses: A review. Waste Management. 2015, 45, 32-41, DOI 10.1016/j.wasman.2015.06.008. (6) Tamis, J.; Luzkova, K.; Jiang, Y.; van Loosdrecht, M. C. M.; Kleerebezem, R. Enrichment of Plasticicumulans acidivorans at pilot-scale for PHA production on industrial wastewater. J. Biotechnol. 2014, 192, 161-169, DOI 10.1016/j.jbiotec.2014.10.022. (7) Morgan-Sagastume, F.; Hjort, M.; Cirne, D.; Gérardin, F.; Lacroix, S.; Gaval, G.; Karabegovic, L.; Alexandersson, T.; Johansson, P.; Karlsson, A.; Bengtsson, S.; Arcos-Hernández, M. V.; Magnusson, P.; Werker, A. Integrated production of polyhydroxyalkanoates (PHAs) with municipal wastewater and sludge treatment at pilot scale. Biores. Technol. 2015, 181; 78-89, DOI 10.1016/j.biortech.2015.01.046. (8) Chen, G. Q. Plastics from bacteria: Natural Functions and Applications. Microbiology Monographs; Vol.14, 351-352, 2010, DOI 10.1007/978-3-642-03287-5. (9) Gholami, A.; Mohkam, M.; Rasoul-Amini, S.; Ghasemi, Y. Industrial production of polyhydroxyalkanoates by bacteria: opportunities and challenges. Minerva Biotechnol. 2016, 28, 59-74. (10) Reis, M. A. M.; Albuquerque, M. G. E.; Villano, M.; Majone, M. Mixed culture processes for polyhydroxyalkanoate production from agroindustrial surplus wastes as feedstocks. In: Moo-Young, M. (Ed.), Comprehensive Biotechnology. Elsevier B.V., 669–683, 2011, DOI 10.1016/B978-0-08088504-9.00464-5. (11) Valentino, F.; Morgan-Sagastume, F.; Campanari, S.; Villano, M.; Werker, A.; Majone, M. Carbon recovery from wastewater through bioconversion into biodegradable polymers. New Biotechnol. 2017, 37, 9-23, DOI 10.1016/j.nbt.2016.05.007. (12) 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, DOI 10.1002/bit.20683.


23


Page 24 of 35

(13) Kleerebezem R, van Loosdrecht MCM. Mixed culture biotechnology for bioenergy production. Current Opinion in Biotechnol. 2007, 18, 207-212, DOI 10.1016/j.copbio.2007.05.001. (14) Koller, M.; Gasser, I.; Schmid, F.; Berg, G. Linking ecology with economy: insights into polyhydroxyalkanoate-producing microorganisms. Eng. Life Sci. 2011, 11, 222-237, DOI 10.1002/elsc.201000190. (15) Korkakaki, E.; Mulders, M.; Veeken, A.; Rozendal, R.; Van Loosdrecht, M. C. M.; Kleerebezem, R. PHA production from the organic fraction of municipal solid waste (OFMSW): Overcoming the inhibitory matrix. Wat. Res. 2016, 96, 74-83, DOI 10.1016/j.watres.2016.03.033. (16) Colombo, B.; Favini, F.; Scaglia, B.; Sciarria, T. P.; D’Imporzano, G.; Pognani, M.; Alekseeva, A.; Eisele, G.; Cosentino, C.; Adani, F. Enhanced polyhydroxyalkanoate (PHA) production from the organic fraction of municipal solid waste by using mixed microbial culture. Biotechnol. Biofuels. 2017, 10, 201, DOI 10.1186/s13068-017-0888-8. (17) Basset, N.; Katsou, E.; Frison, N.; Malamis, S., Dosta, J.; Fatone, F. Integrating the selection of PHA storing biomass and nitrogen removal via nitrite in the main wastewater treatment line. Biores. Technol. 2016, 200, 820-829, DOI 10.1016/j.biortech.2015.10.063. (18) Gottardo, M.; Micolucci, F.; Bolzonella, D.; Uellendahl, H.; Pavan, P. Pilot scale fermentation coupled with anaerobic digestion of food waste - Effect of dynamic digestate recirculation. Renewable Energy. 2017, 114, 455-463, DOI 10.1016/j.renene.2017.07.047. (19) Valentino, F.; Beccari, M.; Fraraccio, S.; Zanaroli, G.; Majone. M. Feed frequency in a Sequencing Batch Reactor strongly affects the production of polyhydroxyalkanoates (PHAs) from volatile fatty acids. New Biotechnol. 2014, 31, 264-275, DOI 10.1016/j.nbt.2013.10.006. (20) Villano, M.; Valentino, F.; Barbetta, A.; Martino, L.; Scandola, M.; Majone, M. Polyhydroxyalkanoates production with mixed microbial cultures: from culture selection to polymer recovery in a high-rate continuous process. New Biotechnol. 2014, 31, 289-296, DOI 10.1016/j.nbt.2013.08.001. (21) Queirós, D.; Fonseca, A.; Rossetti, S.; Serafim, L. S.; Lemos, P. C. Highly complex substrates lead to dynamic bacterial community for polyhydroxyalkanoates production. J. Ind. Microbiol. Biotechnol. 2017, 44, 1215-1224, DOI 10.1007/s10295-017-1951-y. (22) Matturro, B.; Ubaldi, C.; Rossetti, S. Microbiome Dynamics of a Polychlorobiphenyl (PCB) Historically Contaminated Marine Sediment under Conditions Promoting Reductive Dechlorination. Front. Microbiol. 2016, 7, 1502, DOI 10.3389/fmicb.2016.01502. (23) Micolucci, F.; Gottardo, M.; Bolzonella, D.; Pavan, P. Automatic process control for stable biohythane production in two-phase thermophilic anaerobic digestion of food waste. Int. J. Hydrogen Energy. 2014, 39, 17563-17572, DOI 10.1016/j.ijhydene.2014.08.136.


24

Page 25 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(24) Micolucci, F.; Gottardo, M.; Pavan, P.; Cavinato, C.; Bolzonella, D. Pilot scale comparison of single and double-stage thermophilic anaerobic digestion of food waste. J. Cleaner Production. 2018, 171, 1376-1385, DOI 10.1016/j.jclepro.2017.10.080. (25) Leite, W.R.M.; Gottardo, M.; Pavan, P.; Belli Filho, P.; Bolzonella, D. Performance and energy aspects of single and two phase thermophilic anaerobic digestion of waste activated sludge. Ren Energy 2016, 86, 1324-1331, DOI 10.1016/j.renene.2015.09.069. (26) Vagliasindi, F. Strategies and Policies for a more sustainable use for water resource; use of marginal water resources under drought conditions – technical and financial issue. WUEMED Workshop, Rome, Italy, 2005. (27) Villano, M.; Lampis, S.; Valentino, F.; Vallini, G.; Beccari, M.; Majone, M. Effect of hydraulic and organic loads in a sequencing batch reactor on microbial ecology of activated sludge and storage of polyhydroxyalkanoates. Chem. Eng. Transaction. 2010, 20, 187-192, DOI 10.3303/cet1020032. (28) Villano, M.; Beccari, M.; Dionisi, D.; Lampis, S.; Miccheli, A.; Vallini, G.; Majone, M. Effect of pH on the production of bacterial polyhydroxyalkanoates by mixed cultures enriched under periodic feeding. Process Biochem, 2010, 45, 714-723, DOI 10.1016/j.procbio.2010.01.008. (29) Valentino, F.; Brusca, A. A.; Beccari, M.; Nuzzo, A.; Zanaroli, G.; Majone, M. Start up of biological sequencing batch reactor (SBR) and short-term biomass acclimation for polyhydroxyalkanoates production. J. Chem. Technol. Biotechnol. 2013, 88, 261-270, 10.1002/jctb.3824. (30) Albuquerque, M. G. E.; Torres, C. A. V.; Reis, M. A. M. Polyhydroxyalkanoate (PHA) production by a mixed microbial culture using sugar molasses: effect of the influent substrate concentration on culture selection. Wat. Res. 2010, 44, 3419-3433, DOI 10.1016/j.watres.2010.03.021. (31) Colombo, B.; Sciarria, T. P.; Reis, M. A. M.; Scaglia, B.; Adani, F. Polyhydroxyalkanoates (PHAs) production from fermented cheese whey by using a mixed microbial culture. Biores. Technol. 2016, 218, 692-699, DOI 10.1016/j.biortech.2016.07.024. (32) Oehmen, A.; Pinto, F. V.; Silva, V.; Albuquerque, M. G. E.; Reis, M. A. M. The impact of pH control on the volumetric productivity of mixed culture PHA production from fermented molasses. Eng. Life Sci. 2014, 14, 143-152, DOI 10.1002/elsc.201200220. (33) Wen, Q.; Chen, Z.; Tian, T.; Chen, W. Effects of phosphorus and nitrogen limitation on PHA production in activated sludge. J. Environ. Sci. 2010, 22, 1602-1607, DOI 10.1016/S10010742(09)60295-3. (34) 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, 145-160, DOI 10.1002/bit.20085. (35) Amulya, K.; Jukuri, S.; Venkata Mohan, S. Sustainable multistage process for enhanced productivity of bioplastics from waste remediation through aerobic dynamic feeding strategy: process integration for up-scaling. Biores. Technol. 2015, 188, 231-239, DOI 10.1016/j.biortech.2015.01.070. ACS Paragon Plus Environment

25


Page 26 of 35

(36) Valentino, F.; Karabegovic, L.; Majone, M.; Morgan-Sagastume, F.; Werker, A. Polyhydroxyalkanoate (PHA) storage within a mixed-culture biomass with simultaneous growth as a function of accumulation substrates nitrogen and phosphorus levels. Wat. Res. 2015, 77, 49-63, DOI 10.1016/j.watres.2015.03.016. (37) Serafim, L. S.; Lemos, P. C.; Rossetti, S.; Levantesi, C.; Tandoi, V.; Reis, M. A. M. Microbial community analysis with a high PHA storage capacity. Wat. Sci. Technol. 2006, 54, 183-188, DOI 10.2166/wst.2006.386. (38) Lemos, P. C.; Levantesi, C.; Serafim, L. S.; Rossetti, S.; Reis, M. A. M.; Tandoi, V. Microbial characterisation of polyhydroxyalkanoates storing populations selected under different operating conditions using a cell-sorting RT-PCR approach. Appl. Microbiol. Biotechnol. 2008, 78, 351-360, DOI 10.1007/s00253-007-1301-5. (39) Ciggin, A. S.; Orhon, D.; Rossetti, S.; Majone, M. Short-term and long term effects on carbon storage of pulse feeding on acclimated or unacclimated activated sludge. Wat. Res. 2011, 45, 31193128, DOI 10.1016/j.watres.2011.03.026. (40) Carvalho, G.; Oehmen, A.; Albuquerque, M. G. E.; Reis, M. A. M. The relationship between mixed microbial culture composition and PHA production performance from fermented molasses. New Biotechnol. 2014, 31, 257-263, DOI 10.1016/j.nbt.2013.08.010. (41) Valentino, F.; Morgan-Sagastume, F.; Fraraccio, S.; Corsi, G.; Zanaroli, G.; Werker, A.; Majone, M. Sludge minimization in municipal wastewater treatment by polyhydroxyalkanoate (PHA) production. Environ. Sci. Poll. Res. 2015, 22, 7281-7294, DOI 10.1007/s11356-014-3268-y. (42) Jiang, Y.; Chen, Y.; Zheng, X. Efficient polyhydroxyalkanoates production from a waste-activated sludge alkaline fermentation liquid by activated sludge submitted to the aerobic feeding and discharge process. Environ. Sci. Technol. 2009, 43, 7734-7741, DOI 10.1021/es9014458. (43) Venkateswar Reddy, M.; Venkata Mohan, S. Influence of aerobic and anoxic microenvironments on polyhydroxyalkanoates (PHA) production from food waste and acidogenic effluents using aerobic consortia. Biores. Technol. 2012, 103, 313-321, DOI 10.1016/j.biortech.2011.09.040.


26

Page 27 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure Captions Figure 1. Schematic overview of the pilot scale multi-step process for PHA and biogas production from the OFMSW. Figure 2. VFA evolution in the fermenter CSTR (Stage I). Figure 3. Feast phase/cycle length ratio and temperature monitored in the SBR (A); PHA concentration at the end of the feast and at the end of cycle (B) (Stage II). A vertical axis separates the two periods in which the synthetic acetate solution (days 1-49) and the fermented OFMSW (days 50-129) were used as feed. Figure 4. Mass balance of the proposed multi-stage process currently developed at pilot-scale (A); Trend of cumulative income versus the capital cost for the realization of the fermentation reactor and the aerobic PHA production reactors (B).

Table Captions Table 1. Main parameters monitored in the SBR and in the two representative periods with different feed solutions. Table 2. Summary of the PHA accumulation fed-batch tests performed with synthetic acetate solution and fermented OFMSW. Table 3. Assessment of the energy balance of the proposed integrated platform compared with that of the traditional anaerobic digestion process.


27


Page 28 of 35

Figure 1.

Figure 2.


28

Page 29 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3.


29


Page 30 of 35

Figure 4.


30

Page 31 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

aCoaxial


filter-bag equipped centrifuged (solid/liquid separation unit)

bCake


31


Page 32 of 35

Table 1. Synthetic acetate solution

Pre-treated fermented

(days 1st-49th)

OFMSW (days 50st-129th)

mg/L

1227 ± 144

1405 ± 132

PHA (end of feast)

mg/L

119 ± 55

200 ± 19

PHA (end of cycle)

mg/L

55 ± 23

69 ± 8

HB:HV content (end of feast)

Mass (g) %

100:0

88 ± 1:12 ± 1

Feast phase/cycle length ratio

h/h

0.42 ± 0.07

0.13 ± 0.02

320 ± 32

578 ± 77

46 ± 22

225 ± 33

Parameter

Unit

VSS (end of feast)

Substrate uptake rate

mgCODSOL/gCODXa/h

(-qSfeast) Storage rate

mgCODPHA/gCODXa/h

(qPfeast) Storage yield Observed yield

(YP/Sfeast)

CODPHA/CODSOL

(YP/VFAfeast)

CODPHA/CODVFA

(YOBSSBR)

CODVSS/CODSOL

0.40 ± 0.03

0.59 ± 0.05

°C

10.5 – 22.5

20 – 29.5

7.2-8.2

7.8-8.4

Temperature range pH Applied OLR

gCODSOL/L d gCODVFA/L d

0.16 ± 0.08

3.0


0.36 ± 0.04 0.41 ± 0.05

2.7 ± 0.2 2.3 ± 0.2

32

Page 33 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47


Table 2. Substrate

Accumulation

Unit Acetate Number Parameter 1 COD:N:P ratio Mass basis 100:0:0 Temperature °C 22.5 pH 7.8-8.5 Feast phase length h/h 0.13 (SBR cycle)* Accumulation time h 6 PHA content g/gVSS 0.42 Storage yield YP/Sbatch CODPHA/CODSOL 0.64 YP/VFAbatch CODPHA/CODVFA Growth yield YX/Sbatch CODXa/CODSOL 0.07 YX/VFAbatch CODXa/CODVFA Storage rate 264 CODPHA/CODXa/h batch (qP ) Substrate uptake rate 404 CODSOL/CODXa/h (-qSbatch) PHA productivity gPHA/L h 0.18 HB:HV content Mass% 100:0 * corresponding to the SBR cycle in which the biomass was sample

OFMSW 2 100:0:0 22-24 8.0-8.8

1 100:2.6:0.6 24-26 8.0-8.8

2 100:3.0:0.7 23.5-26 7.4-8.7

3 100:2.6:0.6 24-26.5 7.5-8.9

4 100:2.8:0.5 25-27 7.4-9.0

5 100:2.5:0.5 25-27 7.5-9.0

0.16

0.08

0.11

0.13

0.09

0.11

6 0.37

232

4.3 0.52 0.45 0.57 0.18 0.20 436

5 0.45 0.39 0.49 0.21 0.24 388

6 0.47 0.46 0.55 0.21 0.25 255

5.3 0.39 0.40 0.43 0.24 0.28 294

6 0.44 0.47 0.50 0.17 0.19 367

389

746

702

520

501

662

0.16 100:0

0.49 92:8

0.40 88:12

0.28 87:13

0.31 93:7

0.36 92:8

0.61 0.09


33


Page 34 of 35

Table 3. Process

Energy and process units Heating Water

173,989 MJ/d

Thermal Dissipation

10,491 MJ/d

Total

184,481 MJ/d

Thermal

464,828 MJ/d

Electrical

103.4 MWh/d

Fermentation

Heating Water

135,358 MJ/d

Reactor

Thermal Dispersion

3,282 MJ/d

Heating Water

58,926 MJ/d

Heating Water

135,358 MJ/d

Thermal Dispersion

8,000 MJ/d

Total

340,924 MJ/d

Thermal

438,804 MJ/d

Electrical

97.6 MWh/d

Electrical

7.7 MWh/d

Thermal Energy CSST

Digester

Required

Energy produced from CHP

Thermal Energy Integrated

Aerobic Reactors (PHA line)

Required

Digester

multi-stage process

Unit

Whole platform Energy Produced from CHP Energy consumed for aerobic reactors (PHA line)


34

Page 35 of 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Abstract Art

Synopsis Recovering bio-based products from the conversion of a bio-waste organic fraction solves open issues about bio-waste treatment and disposal.


35

Organic Fraction of Municipal Solid Waste Recovery by Conversion

Recommend Documents