Co-digestion of water hyacinth, municipal solid waste and cow dung: A methane optimised biogas–liquid petroleum gas hybrid system

Highlights

• Co-digestion, modelling and optimisation incorporated substrate seasonal variations.

• CH₄ is maximised whilst CO₂, NH₃, and H₂S are minimised.

• Methane-optimised biogas is hybridised with liquid petroleum gas.

• Optimal results showed an increase of 174.58% in annual biogas output.

• 6.97% and 18.24% annual cost savings in winter and summer respectively.

Abstract

Fossil fuels are still the major source of energy in developing countries, howbeit expensive and environmentally unsustainable. Co-digestion substrate proportions and the respective biogas potentials for a huge number of biomaterials for anaerobic digestion are yet to be ascertained let alone optimised. This paper presents a novel methane-optimised biogas–liquid petroleum gas hybrid system concept. Herein this research, biogas is produced from the anaerobic co-digestion of water hyacinth, municipal solid waste and cow dung. A model that incorporated seasonal variations of biomass feedstocks was developed; an optimisation problem was formulated and solved using the Optimisation Interface tool (OptiTool) in combination with the Solving Constraint Integer Programs (SCIP) toolbox in Matrix Laboratory (MATLAB). The biogas production reactions are optimised in such a way that the methane component of the biogas is maximised, and the other components minimised by the integration of a model which necessitates the feed in of optimal substrate masses as per the ratios ascertained for the substrates considered thereby yielding a high quality combustible biogas product. The methane-optimised biogas is channelled towards some community gas demand and liquid petroleum gas comes in to fill the discrepancy between the methane-optimised biogas and the gas demand. Consideration of seasonality changes in the availability of substrates in the modelling and optimisation led to an increase of 174.58% in annual biogas output. A 6.97% annual lowest cost savings was realised in winter and 18.24% annual highest cost savings was realised in summer from the methane-optimised biogas–liquid petroleum gas hybrid system.

Introduction

The heating, cooling and transport sectors, which account for $80\text{\%}$ of global total final energy consumption, are lagging behind in view of meeting Sustainable Development Goal 7 (SDG 7) - (affordable and clean energy) and thus require accelerated action towards the renewable energy transformation [1]. One lucrative avenue towards solving the issue is venturing into biofuels such as biogas which is a form of bioenergy. Bioenergy can be regarded as the most substantial renewable energy source due to its cost-effective advantages and its great potential as an alternative to fossil fuels [2]. It is a renewable energy that is derived from biomass material which is any biological organic matter obtained from plants or animals. Bioenergy is obtained from a broad variety of resources and produced in many diverse routes [3].

Biomass energy sources include but are not limited to terrestrial plants, aquatic plants, timber processing residues, municipal solid wastes, animal dung, sewage sludge, agricultural crop residues and forestry residues. These different types of biomass have to be linked to the various energy flows and conversions in order to meet both renewable energy needs and solve waste management challenges [4]. Bioenergy is one of the most versatile among other renewable energies since it can be made available in solid, liquid and/or gaseous forms [5]. Biogas is one such bioenergy source in the form of a gaseous biofuel. In contrast with other biofuels, biogas production is flexible to different substrates on condition that they are biodegradable. Biogas is produced by the process of anaerobic digestion of biodegradable organic matter. Anaerobic digestion is the breakdown of biomass materials with the aid of bacteria in the absence of oxygen producing a mixture of gases [6]. Production of biogas through the anaerobic digestion process is an environmental friendly process utilising the increasing amounts of organic wastes produced [7]. This technology reduces greenhouse gas emissions and as such a sustainable form of energy, biogas, a biofuel is obtained [8].

Rozy et al. [9], experimentally investigated the effect of varying physicochemical parameters on biogas production from water hyacinth (WH) in combination with cow dung and obtained enhanced yield parameters. They however, emphasised the need to enhance and optimise methane generation from WH and other such substrates. In anaerobic co-digestion combinations, it is of paramount importance to know the mass ratio of each substrate to be fed in the blend mixture so as to achieve the highest possible proportion of methane in the output biogas. In as much as WH is a nuisance to waterways and sources, municipal solid waste (MSW) and cow dung (CD) are as well pausing some detrimental effects to the environment. Anaerobic co-digestion of these bio-materials among others leads to increased biogas yields when compared to mono-digestion of the same due to enhanced bio-degradability, bio-accessibility, and bio-availability among other synergisms in the process reactions [10].

Biogas production can be enhanced by utilisation of high-methane potential substrates, enzymes and microbial addition, optimisation of process conditions and parameters, co-digestion of various substrates, pre-treatment of the feed material and separating the digestion process into phases (multi-stage digestion) [11]. Dependable anaerobic co-digestion modelling is essential to clearly forecast the consequence of blending substrates in a reactor and do away with possible undesirable outcomes from blending combinations established on arbitrary and/or heuristic conclusions. To optimise is to determine the maximum or minimum values of a specified function that is subject to certain constraints.¹ Hagos et al. [12], highlighted that process optimisation and improvement of biogas production still needs more investigations to be done and that the use of modelling and simulation ways can lead to realisation of substantial enhancement of biogas yields. Diverse optimisation approaches are established in literature in a bid to obtain the best reaction conditions, best reaction parameters and best substrate ratios for different feed stocks so as to enhance and optimise the biogas production process. Sreekrishnan et al. [13], also notes that use of additives, recycling of slurry and slurry filtrate, variation of operational parameters like temperature, hydraulic retention time (HRT) and particle size of the substrate and use of fixed film/biofilters are some of the techniques for enhancing biogas production.

The conventional method of optimisation of anaerobic digestion comprise of laboratory batch experiments varying reaction conditions and parameters as well as co-digestion of varied feed stocks to evaluate the digestibility and biogas potential of different substrates. Co-digestion of varied substrates has shown that an improved biogas production potential can be realised as compared to mono-digestion of single substrates [14]. Artificial Neural Networks (ANNs) and Genetic Algorithms (GA) are some of the modern tools that are used to solve complex problems which cannot be unravelled by conventional solutions [15]. Linear programming approaches [16], response surface methodologies [17], as well as simplex-centroid mixture design and central composite design [18], are some of the optimisation approaches which have been applied in anaerobic digestion.

There has been a considerable increase in demand for energy in developing countries like Zimbabwe while the supply and/or generation capacity is lagging behind [19]. As a result consumers are shifting to alternative renewable energy options and also to other available fossil derived and imported fuels such as Liquid Petroleum Gas (LPG). The availability of non-renewable forms of energy such as LPG, derived from fossil fuels will continue to decrease while at the same time their costs will continue to increase [20]. The interchangeability of fuels has to be compared in terms of the Wobbe Index (WI) when considering shifting from one fuel to the other. The Wobbe Index (WI) is an indicator of the interchangeability of fuel gasses. It is the key pointer to the replacement of one fuel with another and is very useful in comparing the burning efficiency of fuel gases [21].

This research deduced from previous works/studies that solar PV–Biogas hybrid systems have been developed and optimised to ease the energy demand mainly being fostered by inadequate conventional energy supplies. Nawaz et al. [22], carried out a feasibility study on a solar photovoltaic–biogas hybrid system and also did an optimisation of the same. Kwok et al. [23], investigated the hybridisation of solar, wind and biogas in a bid to optimise energy generation from these renewable energy sources. In some instances the solar PV–Biogas systems have been tied with the grid to minimise energy costs and at the same time ensuring consistent supply of energy at all times [24]. It was also however, realised that not much has been reported and/or researched with respect to integrating optimised biogas systems and LPG for heating, lighting and power purposes. According to the authors’ literature survey, no research was found to report on optimised biogas–LPG hybrid systems. In order to meet the growing energy demands and to do away with waste disposal problems, the production of biogas and the respective optimisation of its bio-methane major constituent is of uttermost importance in addition to hybridisation of the energy supply alternatives such as LPG [25].

In the authors’ previous paper [26], a model for determining biogas production potential from water hyacinth (WH), municipal solid waste (MSW), and cow dung (CD) was presented and an optimisation of the co-digestion mixing ratios of these substrates was carried out in a bid to obtain the highest possible amount of biogas from the co-digestion mixture. The same substrates are used in this present work. However, the model developed and being reported in its own novel way in this current paper differs with the previous one in the following ways:

• Seasonal variations of the substrates are taken into consideration in the modelling and optimisation.

• Methane is maximised whilst carbon dioxide, ammonia, and hydrogen sulphide are minimised to obtain more methane in the biogas mixture and thus improving the quality of the biogas produced.

• The enhanced biogas produced is hybridised with Liquid Petroleum Gas (LPG) to supply some gas demand

It is hereby being emphasised that according to the authors’ knowledge and research investigations, no previous studies are reported to have looked at the effect of substrate/feedstock seasonal variations on co-digestion and at the same time incorporate the same in modelling and optimisations. As such, this current research is unique and innovative in that regard and the findings are one of their own kind, contributing immensely to the anaerobic digestion research niche. The purpose and contribution of this current work is the development of a model which facilitates the attainment of high quality biogas constituted of a high proportion of methane while at the same time taking into consideration the seasonality changes of the substrates. Consequently, the resultant co-digestion substrate blending ratios vary for each month and so does the biogas yield unlike in the previous work where a single average blend ratio and an annual average biogas output was obtained. The high quality optimised biogas produced is channelled towards the gas requirements of a community in a hybrid system with Liquid Petroleum Gas (LPG) where LPG meets the rest of the demand not met by the biogas. This work contributes to the reduction of reliance on imported energy and adds great value by supplying a high quality bio-methane gas thereby substituting a great proportion of LPG consequently reducing import costs as well as minimising environmental pollution. The model and the method used herein this study are general enough for use in many countries, and can be applied with many other varied biomass resources.

Section 2 of this paper gives the modelling and optimisation materials and methods, Section 3 gives a case study, Section 4 gives the results & discussion and Section 5 concludes the paper. The algorithm is given as the appendix.