Anaerobic co-digestion of chicken litter with food and agro-industrial wastes/residues, an Australian case study: use of carbon-to-nitrogen ratio for substrate mixing and semi-solids versus wet anaerobic digestion

Zahan, Z 2019, Anaerobic co-digestion of chicken litter with food and agro-industrial wastes/residues, an Australian case study: use of carbon-to-nitrogen ratio for substrate mixing and semi-solids versus wet anaerobic digestion, Doctor of Philosophy (PhD), Engineering, RMIT University.


Document type: Thesis
Collection: Theses

Attached Files
Name Description MIMEType Size
Zahan.pdf Thesis application/pdf 3.48MB
Title Anaerobic co-digestion of chicken litter with food and agro-industrial wastes/residues, an Australian case study: use of carbon-to-nitrogen ratio for substrate mixing and semi-solids versus wet anaerobic digestion
Author(s) Zahan, Z
Year 2019
Abstract In Victoria, Australia, around 0.45 million tonnes of chicken litter (CL) are produced every year. CL has high potential for bioenergy production, mainly in the form of biogas (methane), however, it has not been widely practiced because it has high levels of protein and uric acid which are known to cause inhibition, during anaerobic digestion (AD), in the form of ammonia. The current CL management practice is to transport CL to a composting facility which has limited benefit to the broiler industries. This has triggered research to assess the feasibility of biogas production from broiler farms¿ CL co-digested with other substrates. Among the agro-industrial wastes usually available in the proximity of chicken farms are agro-industry processing wastes and agricultural residues such as yoghurt whey (YW), wheat straw (WS) and hay grass (HG). Municipal waste i.e. the food wastes (FW) in the household bins is also available to use. These wastes have high methane potential; but varying characteristics and composition, e.g. carbon-to-nitrogen (C/N) ratio, pH, alkalinity, ammonia, structural recalcitrance.  Therefore, these wastes need to be characterised, require understanding if they are suitable for co-digestion and what ratios they can be mixed together.

Also, most of the literature concerning co-digestion of different wastes focused on mixing one or two wastes; but little research has been done on the co-digestion of a wide variety of solid waste streams. Therefore, an efficient and economic energy recovery from those wastes through anaerobic co-digestion, an investigation into optimum process conditions is required.
Wet anaerobic digestion (W-AD) is a well-established technology operating at solids concentration of<10% total solids (TS), whereas, high solids AD (HS-AD), (10-15% TS) and dry AD (D-AD), <15% TS) are not a common practice. This is due to long retention times and types of infrastructure required, but the most important factors are the lack of knowledge concerning treatment operating conditions, loading and composition. However, as most of the agro-industrial wastes have high TS, the research has focused on both conventional wet and high solids AD.

The aims of this research are  to assess the potential of anaerobic co-digestion (ACoD) of CL and co-substrates and to determine the optimum AD process  conditions where the process is stabilised and show no inhibition, for different AD configurations, and to assess  the performance of HS-AD and W-AD fed with a range of co-substrates, under continuous feeding conditions for different organic loading rates (OLRs). The research focused on developing a method that can be used to determine the optimum mixing ratio of agro-industrial wastes of varying characteristics, optimum running conditions for continuous AD, pre-treatment of waste to improve bio-degradation and use of the optimum conditions to develop a high solid AD process.

The wastes have been characterised thoroughly before doing their biochemical-methane potential (BMP) under conventional W-AD solids concentrations (2-4% volatile solids (VS)). Four different kinetic models i.e. the first order kinetic model, modified Gompertz model, transfer function model and the cone model were used to fit the biogas yield of the single substrates and the modified Gompertz model (R2: 0.93-0.99) showed better fit among all the models which explains the variation in lag phase and methane production rate depending upon the substrate characteristics.

The second batch of BMP tests assessed the co-digestion of a range of different wastes mixed using C/N ratio as the control parameters. The batch experiments were designed according the response surface model (RSM). All the BMP tests were carried out in batch assays under mesophilic conditions in duplicates with 1:2 g/g VS of substrate to inoculum. Analysis of the BMP results using Matlab indicated that the maximum methane production could be achieved for a feedstock of 30-35% VS of CL and 65-70% VS of agro-industrial waste (i.e. YW, FW, HG and WS) that have a total C/N ratio of 26-27.5.

In the second phase of the experimental work, semi-continuous anaerobic co-digestion were performed at 4-6% TS based on the predictions and conditions from the batch assesses to observe process performances. The AD reactors were operated at organic loading rates (OLRs) of 2.0-3.0 g TS/L. d and hydraulic retention time (HRT) of 20 days. The optimum feedstock (substrates mixture) was CL: FW:WS of 60:20:20, where 73%, 167% and 117% increase in total biogas production at OLR of 2.0, 2.5, 3.0 g TS/L. d, respectively, compared to that from CL, was achieved. Principal component analysis (PCA) was applied with the characteristics parameter and 68.1% of data variability was explained with second principal component. During semi-continuous work, a new concept, that C/N ratio and lignocellulosic structure degradation might be related during AD Digestate, is applied and found that carbohydrate degradation i.e. cellulose and water soluble contents plays a major role in explaining the variation in performance and produced biogas for different feedstocks of balanced C/N ratio.

As most of the agro-industrial wastes are lignocellulosic in nature which resists biodegradability, therefore, the next phase of experimental work focused on selective fractionation of lignocellulosic biomass using sequential alkaline (AKP) and dilute acid (DAP) pre-treatments to increase biodegradability. NaOH pre-treatment was applied both on CL (as it had included bedding material in it) and WS under independent factors of NaOH concentration (1-5% w/v), reaction time (30-90 min) and temperature (60-120°C), and experiment design using RSM. The optimum conditions were analyses with Minitab and the optimum conditions are NaOH concentration 1% (W/V) for 30 min at 120ºC and NaOH concentration 5% (W/V) for 90 min at 120ºC for AKP of WS and CL, respectively. Sequential DAP with H2SO4 (1%-3% w/v) was applied with the best conditions found for the AKP using RSM 2 with same conditions as AKP i.e. same temperatures and times.  With sequential DAP+AKP, higher removal of lignin and hemicellulose and an increment of 25% in biogas was obtained compared to single pre-treatment.

Finally, High solids AD were performed in sequential batch assays to give it enough time to degrade and to observe the long-time process performances. sequential ACoD at 15% TS was performed and reported with the optimum conditions from the batch assays (W-AD) i.e. CL: FW: WS mixed at a ratio of 35:32.5:32.5 to have C/N ratio of 26.5 over 215 days in six cycles. The reactors that were fed with untreated substrates produced 321.6±13.4 mLN biogas/g VSadded, which increased by 88%, when CL and WS were sequentially pre-treated using AKP and DAP. The VS removal in the reactors that received pre-treated substrates showed improved biogas production compared to those that received untreated feedstock. A VS removal of 55% were observed with AKP+DAP pre-treated substrates fed reactors, compared with only 36% VS removal from untreated substrate fed reactors. A reduction in ammonia and cellulose with an increase in water soluble contents was also observed in the reactors that received these pre-treated substrates. Additionally, it was also noted that biogas production using sequential SS-AD at 15% was almost 38% less than W-AD, however this was negated with the pre-treatment of substrates, indicating that co-digestion at high TS of 15% is achievable.
Degree Doctor of Philosophy (PhD)
Institution RMIT University
School, Department or Centre Engineering
Subjects Environmental Engineering not elsewhere classified
Environmental Engineering Modelling
Environmental Technologies
Keyword(s) Anaerobic digestion (AD)
Wet anaerobic digestion
Semi-solid anaerobic digestion
Chicken litter
Responce surface methodology (RSM)
Sequential Alkaline and acid pre-treatment
Wheat straw
Food wastes
Yoghurt whey
Hay grass
Biogas
Bio-chemical methane potential
Agro-industrial wastes
Carbon-to-Nitrogen ratio (C/N ratio)
Kinetic Model
Surface optimisation
Batch anaerobic digestion
Semi-continuous anaerobic digestion
Anaerobic co-digestion (ACoD)
Organic loading rate
Lignocellulosic fractionation
Anaerobic digestion reactor performance
Principal component analysis (PCA)
Sequential anaerobic co-digestion
High solids loading
Digestate fractionation
Alkaline pre-treatment
Semi-continuous anaerobic co-digestion
Versions
Version Filter Type
Access Statistics: 20 Abstract Views, 22 File Downloads  -  Detailed Statistics
Created: Fri, 15 Nov 2019, 10:04:31 EST by Adam Rivett
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us