Bioenergy production from the anaerobic digestion of agri-industrial residues

The worldwide wine and cheese-making industries generate an estimated 11 million metric tonnes of solid waste (grape marc, GM) and over 200 million m3 of watery waste (cheese whey, CW), respectively, every year. Despite traditional absorption of such wastes into composting and animal feeds, by-product disposal issues regularly occur with the production of agro-industrial commodities such as wine and cheese, prompting the dispersion of excess GM or CW on arable lands and in landfills. Consequently, high levels of environmental contamination may arise due to the leaching of pollutants into groundwater reservoirs leading to eutrophication, the depletion of available dissolved oxygen from the release of COD (chemical oxygen demand), and an outflow of volatile organic compounds linked to detrimental long-term environmental damage. To mitigate the increasing anthropogenic footprint on the environment, chiefly driven by a growing global population, a circular economy management strategy has gained momentum in recent times. This framework is often dubbed a “zero waste” approach implemented through deliberate reduction of earth’s natural resources, maximum conservation of commodity produced through extended reuse, and relentless recycling of discards by allied industries. Tapping into GM and CW waste streams as potential low-cost feedstock, the industrial bio-refinery approaches commonly in place proceed by physical, biological, chemical, or an assortment of extraction techniques for resource recoveries. Furthermore, thermal treatments such as combustion, gasification, and pyrolysis are often employed on solid wastes in energy generation schemes. Valorisation from a wide array of wastes, especially when moisture content is high like in CW, can only proceed at much lower calorific (thermal energy) requirements and sound economic competitiveness in anaerobic digestion (AD) technology for energy generation. Other advantages of AD over conventional thermal methods include, but not limited to, a carbon neutral footprint because of effective carbon sequestration; a production of a sanitised digestate, following digestion at thermophilic temperatures, which can later be used as fertiliser; and the dilution of biochemical toxicity thresholds to microorganisms when utilising wastes of complementary nutritional characteristics. In view of the above, the overarching objective of the research was stable energy production with concomitant residue reduction from the novel co-digestion of GM and CW in AD at high total solids. To minimise energy requirements, the research work proceeded in non-agitated digesters during incubation. A literature review of the state-of-the-art of GM valorisation is presented in Chapter 1, followed by a series of studies clearly delineated in Chapters 2 to 6, designed to provide optimal bioenergy production from GM and CW substrates at ultimately competitive energy costs. The aims of the research were to assess the potential for methane production at thermophilic temperature by co-digesting GM and CW in defined ratios (Chapter 2); to investigate the anaerobic treatment potential of the co-digestion of GM and CW in unstirred conditions at high working volume (Chapter 3); to achieve bioenergy production by the targeting of operational controls as a multi-level anaerobic digestion performance enhancer (Chapter 4); to determine the optimal substrate-to-inoculum (S/I) ratio for the treatment of GM and CW without any pretreatment (Chapter 5); and to evaluate the energy balance for the co-digestion at ambient conditions while eradicating the dependency on calorific energy inputs (Chapter 6). A proof-of-concept study (Chapter 2) was conducted to explore the feasibility of biogas production from co-digesting milled GM and full-strength CW. An L8 Taguchi orthogonal array was designed to determine the impact of feedstock ratio, initial pH, and incubation temperature on methane production. A sludge inoculation dose of 10 mL was added to 100 mL feedstock volume (10:1 S/I ratio) of GM and CW (w/w) incubated in ratios 1:3, 2:2, 3:1, and 4:0 GM/CW, respectively, at 35 and 55 ⁰C for AD over a period of 144 d. The optimal bioenergy profile was in 3:1 GM/CW digesters where cumulative methane production equalled 24.43 ±0.11 L CH4 kg-1 VS. Also, there was poor waste digestibility (18.63 % COD removal) and a lag time of 14.84 d in the start-up phase. A scale up into a 5 L W8 anaerobic digester (Armfield, UK) was utilised in Chapter 3 allowing greater understanding of the changes in bacterial diversity, and an advanced assessment of the possible economic returns. Therefore, the study at high treatment capacity (Chapter 3) processed 3:1 GM/CW (w/w) over 120 d at 45 ⁰C. There was 0.363 m3 CH4 kg-1 VS for cumulative methane production coinciding with 61.60 % COD removal efficiency whereas the lag time for the start-up phase was positively reduced to a duration of 7.9 d. Physicochemical factors such as increased salinity (85.71% change) indicative of an active solubilisation of organics by microbes, and a stable electrical conductivity (8.24% change) were also noted. Microbial species identified included Firmicutes, Bacteroides and Methanosarcina. The use of digestate as downstream inoculum in a follow-on study (Chapter 4) involved an increase of the inoculation dose by lowering the S/I ratio to 10:3 in the mono-digestion of milled GM in a W8 anaerobic digester (Armfield, UK), with a lowering of the temperature to 35 ⁰C to further diversify the bio-catalytic bacterial population over 42 d of treatment. The acclimated inoculum additionally reduced the lag of the start-up phase to 7 d. Mono-digestion of GM was favoured over co-digestion temporarily to fine-tune the biogas production conditions for GM as the main co-substrate when co-digesting. The cumulative methane production equalled 0.145 m3 CH4 kg-1 VS whilst 43.50 % COD removal efficiency was achieved for the mono-digestion. The next aim was to completely remove energetic requirements during pretreatment upstream of an AD operation to improve the overall energy balance of the bioenergy production system (Chapter 5). Therefore, small-scale digesters of unmilled GM and undiluted CW in 3:1 co-digestion ratio (w/w) were inoculated with a recirculated 120-day digestate (viz. inoculum) at S/I ratios of 0:10, 5:5, 7:3, and 9:1 over 58 d at 45 ⁰C. It was observed that lack of milling GM was linked to the presence of large particle sizes that likely resulted in slow hydrolysis of the recalcitrant lignocellulosic material and inhibited acidogenesis. The optimal 7:3 S/I ratio generated a cumulative 6.45 L CH4 kg-1 VS whereas the COD removal efficiency equalled 12.73%. The start-up to stable biogas production extended to 9.4 d when optimal S/I ratio; however, uneconomically longer lag times and lower biogas production prevailed at weaker inoculation doses (viz. 9:1 S/I ratio). The final aim of the research was to look at the viability of further improving the bioenergy profile of the GM and CW co-digestion by eliminating the thermal energy input when artificially heating digesters (Chapter 6). The assumption was that the digestion efficiency is reached wen digester temperature is closer to ambient temperature. This study processed unmilled GM and undiluted CW in 3:1 co-digestion ratio, respectively, at ambient temperature of 25 ±0.8 ⁰C over a period of 58 d. The waste inoculation dose was bumped up to 10:3 S/I ratio (slightly above the optimal S/I ratio) to compensate for the slower reaction kinetics in lower temperatures and non-thermoregulated conditions. Noteworthy, a recirculated 120-day digestate from an upstream digester run was used as downstream inoculum. When normalised to 250 mL of working volume, the highest cumulative methane production equalled 5.36 ±0.1 L CH4 kg-1 VS. Also, the highest total energy produced equalled +189 KJ kg-1 VS whilst the highest COD removal efficiency was 43.72%. Energy savings because of lack of pretreatment and “unplugged” digester operation (ambient temperature) resulted in stable biogas production and an overall positive energy balance. This study demonstrated that GM and CW can be co-substrates in a self-sustaining bio-based generator. Overall, this research has uncovered the suitability of an innovative co-digestion of grape marc and cheese whey. The use of unamended winery and dairy feedstock involved less processing infrastructure and freshwater use, reduced operating costs, and enhanced bioenergy production and bioremediation profiles.  There were also significantly greater methane yields when co-digesting feedstock, in optimal 3:1 GM/CW (w/w) ratio, than when substrates were incubated individually.  The maintenance of an active microbial culture through the recirculation of digestate from a previously successful digester run held a number of benefits namely the provision of ample inoculum for subsequent digester runs at the prescribed high inoculation dosage (7:3 and 10:3 S/I ratios); additionally, an inoculum of concentrated anaerobic microbial consortia of redundant metabolic capabilities likely allowed for a reduced lag time in the start-up for stable biogas production, thus translating to a profitable digester operation. At the same time, the energy balance of the GM/CW bioenergy system turned positive and independent from energy inputs for thermoregulation when pretreatment of feedstock was abandoned altogether, and digester power inlets were switched off during operation. In contrast, the literature is awash with complex, at times necessary, waste pretreatment strategies to increase organics solubilisation. However, energetic requirements during pretreatment and subsequently during digester operation may reverse possible energetic gains, thus rendering a bioenergy production enterprise uneconomical. Therefore, future large-scale experiments on the maximisation of the positive energy balance so far obtained would play an important role in the delivery of enhanced energetic valorisation from the grape marc and cheese whey bioresource to the joint sector in which it is produced.