Environmental Assessment of Hybrid Waste-to-Energy System in Ghana

Abstract

Waste management in most parts of Africa is characterized by the disposal of mixed waste in unengineered landfills. The aim of this study is to assess the environmental impact of mixed waste received at a waste-to-energy plant in Ghana relative to the current model of landfilling. A Life Cycle Assessment was conducted using OpenLCA software version 2.3.1 based on the ReCiPe Midpoint method. For landfilling, LandGEM software version 3.03 was used. The results indicate that waste-to-energy has the potential to provide carbon savings of 3.52 tCO₂eq/ton of waste treated compared to landfilling. Pyrolysis is observed to have high avoided burden across all impact categories, with the lowest Global Warming Potential of −2.3 kgCO₂eq. Anaerobic digestion shows a near neutral environmental impact with the highest value of 47.56 kg 1,4DCB for Terrestrial Ecotoxicity, while Refuse-Derived Fuel and segregation processes show low environmental burdens. The net avoided burden is highest for global warming and non-carcinogenic human toxicity potential. Overall, the hybrid waste-to-energy model is concluded to be an environmentally preferred waste management option compared to conventional landfilling methods, and we recommend that decision-makers facilitate investments into it. It is also recommended for the development of local inventories and databases to encourage more country-specific environmental impact studies and to reduce uncertainty.

1. Introduction

Developing countries continue to lag behind as the 2030 target of the Sustainable Development Goals (SDGs) draws nearer [1,2]. Waste management and energy are crosscutting sectors with high environmental footprints mainly linked to SDGs 3, 6, 11 and 13. With a record high of 37.4 Gt energy related CO₂ emissions in 2023, greenhouse gas (GHG) emissions from these sectors are a key concern to the global agenda. Renewable energy is, in particular, crucial for reducing emissions [3,4]. The recently ended United Nations Climate Conference 2024 (COP 29) was strongly dominated by discussions of climate finance and carbon markets with stronger commitments to supporting developing countries [5]. Ironically, a study in South Africa, Nigeria, Cameroun and Ghana identified a lack of strong policy implementation as a challenge to renewable energy uptake and particularly waste-to-energy (WtE) projects in Ghana [6,7].

Waste is a resource material for energy generation with the potential to substitute fossil fuels and consequently reduce the environmental burden of poor waste management [8]. CEWEP estimates that in 2021 alone, the conversion of waste to energy (WtE) supplied 35,000 GWh of electricity to 21 million Europeans, together with 87,000 GWh of heat to 17 miilion residents, which translates into 22–44 million tons of CO₂ savings [9]. The end products of WtE include electricity, hydrogen, biodiesel, RDF and heat, although electricity and heat are the most common. They can generally be categorized into either energy or fuels [10].

Comparative environmental studies support key decision-making on WtE technology selection. Decisions can be made between WtE technologies and conventional/existing treatment options such as landfilling. For instance, WtE shows methane emissions reductions of about 52% for biomethane production compared to landfilling [11]. A decision can also be made on the selection of the most environmentally friendly technology or process. For instance, electricity/heat cogeneration was found to be the most environmentally friendly scenario, releasing 1.95 kgCO₂eq/m³_biogas compared to the combined end-products of electricity/biomethane [12]. In another city-scale scenario analysis, it was concluded that a thermal WtE option (incineration) provided the best GHG savings of (652,626 tCO₂eq/yr), which was more than twice that of two other scenarios: recycling/landfill gas to energy and recycling/composting [13].

Following the Paris agreement, countries like Ghana have set up a National Greenhouse Gas Inventory. Data from the directory indicate that the three major contributing sectors are energy, agriculture, forestry and other land uses (AFOLU) and waste. Between 1990 and 2019, the total national GHG emissions increased by 853.1% and 265.1% in the energy and waste sectors, respectively, with the energy sector alone contributing 45.7% in 2019. Strategies to reduce emissions as the population and economy continue to expand are critical [14]. The energy sector has been reported to be financially unsustainable due to its high dependence on externally sourced fossil fuels for power generation [15]. On the other hand, the country aims to achieve 10% renewable energy penetration by 2030 through the Intended National Determined Contribution (INDC), which is unlikely to be achieved due to low incentives [16]. Meanwhile, solar, wind and biomass (including waste) have been identified as attractive alternatives to achieve this target, considering the local abundance of these resources [17,18,19]. It is estimated that WtE technologies such as anaerobic digestion (AD), incineration and landfill gas to energy (LFGTE) have the potential to supplement 59% of the country’s renewable energy target in 2030 [20]. Ironically, the waste management sector in Ghana is plagued by the poor implementation of sustainable waste treatment options, with an estimated 931 tons of waste sent to landfilling daily from the capital and adjoining towns alone [21].

Several WtE studies have been carried out in Ghana, mostly incorporating techno-economic assessments. Existing and operational WtE plants include the Safi Sana and HPW plants, which incorporate AD and solar PV technologies [22]. It is clear that such projects require the strict assessment of impacts, including environmental impacts and the incorporation of risk mitigation measures, to avoid adverse impacts [23]. Published studies have considered the potential for establishing WtE plants under various settings in Ghana such as rural electrification [24], university sewage treatment [25], urban municipalities [26,27,28], incineration at the national scale [29] and landfill sites [30]. All the studies focused on one key product, electricity to grid, and concluded on technical feasibility, considering varied conditions.

The German government has recently supported the establishment of a pilot WtE plant in Ghana in Gyankobaa in the Ashanti Region. It has been projected to generate 810 MWh of electricity per annum with other associated fuels such as RDF, plastic pellets and compost. The technical model of the plant, which involves a hybrid of solar PV, AD and plastic pyrolysis, has been proven to have the potential to be replicable in other parts of the country [31]. A detailed techno-economic assessment of this case study plant has been previously conducted and published. The results indicate that a standalone WtE plant relying only on grid connected electricity is not feasible. A business case, which involves the strong integration of solar PV at a larger capacity than the current plant, is presented [32]. In an assessment of product models, it was concluded that the conversion of generated energy into fuels and materials such as green hydrogen, biomethane, RDF and compost will potentially make the plant more feasible [33]. These studies are intended to strategically direct the business model when the plant becomes fully operational. One important element of business models is the environmental impact. Also, at the national level, WtE is generally expected to generate negative environmental burdens compared to energy generation from fossil fuels or the landfilling of waste. A research gap is therefore identified for the environmental assessment of WtE using the hybrid WtE plant as a case study.

The aim of this study is therefore to investigate the potential avoided environmental burden from the pilot plant compared to the current landfilling alternative. Also, this study will conduct a comparative assessment of the environmental performance of the subsystems at the plant. It is expected that the results will enhance investment from both the private and public sectors to support decision-making in waste management projects.

The most standardized environmental assessment technique for WtE systems is the Life Cycle Assessment (LCA) [34]. LCA is a well-developed environmental management technique used to assess the environmental impacts of technologies, scenarios or systems. The ISO 14040 and 14044:2006 provides principles and detailed guidelines for conducting an LCA [35,36]. For WtE systems, the most commonly used LCA methods are the CML impact assessment series, IPCC, ReCiPe and Eco-indicator. Global Warming Potential (GWP) is the most analyzed impact/damage category. Generally, landfilling has the highest GHG emissions compared to WtE technologies [37]. The CML method is one of the earliest and most established LCA methods and focuses on individual impact categories. IPCC is primarily considered in assessing climate change impacts such as GWP. ReCiPe, on the other hand, covers a wide range of impact categories, which is relevant for such a study [38].

2. Materials and Methods

2.1. Study Scope and System Boundaries

The study focuses on the environmental assessment of multiple technologies at a pilot WtE plant located at Atwima Nwabiagya Municipality in Ghana. The Business-as-Usual (BAU) is the current waste management model where municipal waste is collected, transported and disposed at the Kyereso landfill, which is basically an unengineered landfill/dumpsite [39]. This study uses the LCA methodology, which follows four major stages, as recommended by ISO 14040 (1997): definition of scope, inventory development, impact assessment and interpretation of results [35]. Figure 1 shows the system boundary and the process flow within the plant beginning from the gate of the treatment facility/dumpsite to the end of the treatment. Although the plant also has the capacity to treat other forms of waste through recycling and composting, this study focuses on the environmental aspects of energy/fuel production [33].

2.2. Functional Unit

Based on the goal of the study, the functional unit (FU) considered is one ton of mixed MSW waste received at the treatment facility. Since the capacity of the WtE plant is 50 t/day, it is compared with the same quantity at the landfill site over an active lifespan of 20 years.

2.3. Waste Composition and Segregation

Mixed MSW arriving at the plant or landfill is estimated to have the following characteristics according to a related study within the same municipality [39]: fermentables (37.8%), plastics (17.1%), paper (1.2%), wood (1.5%), cardboards (3.2%), metals (1.2%), glass (1.3%), textiles (4.4%), special wastes (4.2%), inerts/fine ores (16.8%) and miscellaneous materials including e-waste, sanitary materials and household repair debris (11.3%). The first stage of treatment is the four-stage manual and mechanical sorting system, similar to what is used in other Material Recovery Facilities (MRF) in Ghana, such as the Kumasi Compost and Recycling Plant (KCARP) [40,41]. After this, the various fractions are transferred to the respective processes, as shown in Figure 1.

2.4. Life Cycle Inventory Analysis (LCIA)

LCIA involves stock-taking of the inputs and outputs of materials, emissions and energy from each process within the boundaries of the product system [37]. Sources of data are primary data obtained from the plant, other related studies and calculated values based on literature. Elementary flows for emissions were estimated using emission factors from sources such as [42,43]. Pollutant emissions focus on greenhouse gases (CH₄, CO₂, N₂O) and acidic pollutants (SO₂, NO_x) [44]. The inputs and outputs of materials and energy for each process within the WtE plant were summarized in the following sections.

2.4.1. Business-As-Usual (BAU)

In BAU, mixed waste is disposed of at the dumpsite without any treatment. Landfill Gas Emissions Model (LandGEM) version 3.03 was used to estimate the emissions from 50 t/day MSW deposited at the Kyereso dumpsite [45]. The LandGEM software employs a first-order decomposition rate equation, as described in [43]. Localized values were calculated for the methane generation rate and the potential methane generation capacity to replace the default values (

Table 1. Key input variables for landfill gas modelling in LandGEM.

Input Variable	Value	Units	Details	Reference
Precipitation	1318.34	mm/yr	Regional annual average for Ashanti Region (1990–2021)	[46]
Methane correction factor	0.8	-	Default for deep, unmanaged landfills	[47]
Temperature	26.98	°C	Regional annual average for Ashanti Region (1990–2021)	[46]
Methane fraction in landfill gas	50	%	Default	[47]

).

Landfill gas is generally a mixture of two major components (CH₄ and CO₂) and several trace gases. LandGEM generates default outputs for Total LFG, CH₄, CO₂ and NMOC [45]. The GWP of landfill gas for 100-year horizon (GWP₁₀₀) in kgCO₂eq/yr is computed as the sum of GWP of GHGs. According to IPCC, GWP₁₀₀ for CO₂ (GWP₁₀₀) and non-fossil methane (GWP_CH4) are 1 and 27, respectively [48]. Not all methane generated in a landfill is directly emitted into the atmosphere. Losses occur through the surface cover, methane oxidation in cover soils and lateral migration or losses during storage, which is assumed to be 10% of the generated methane. Thus, mass uncontrolled methane emissions (UMP_CH4) from landfills without Gas Collection and Control Systems (GCCS) and subsequently (GWP₁₀₀) is estimated by [49]

$${UM}_{CH4}=M_{CH4}\times 0.9$$

(1)

$${GWP}_{100}={(UM}_{CH4}\times {GWP}_{CH4})+\left(M_{CO2}\times {GWP}_{CO2}\right)$$

(2)

2.4.2. Anaerobic Digestion (AD)

The inventory for this study follows a similar assessment in a related publication [50]. Electricity generated is computed based on [51,52]. The source of energy for the engines is the solar PV system, while the AD receives recycled heat energy from the CHP. Sources of emission from the AD system are mainly from digester and the CHP engine. Methane losses from the digester are estimated to be 5%, and CO₂ is ignored because it is of biogenic origin, which has insignificant environmental impact [44]. Emissions from the CHP engine are estimated using the emission factor for GHG, acidifying pollutants are computed [53] and the inventory is presented in

Table 2. Inventory for the AD process.

Flow Direction	Parameter	Units	Value
1. Bio-digestion
Inputs	Organic waste	tons	1.00
	Electricity (solar PV)	kWh	38.45
	Heat (CHP)	kWh	237.80
	Water (groundwater)	m3	0.37
Outputs	Methane	m3	62.14
	Digestate	tons	0.88
	Methane (emissions)	kg	1.34
2. Cogeneration
Input	Methane	m3	62.14
Outputs	Electricity	kWh	216.20
	Avoided heat	kWh	237.80
	Carbon monoxide	kg	5.16 × 10−9
	N2O	kg	0.0012
	NOx	kg	0.16
	SO2	kg	0.02

.

2.4.3. Pyrolysis

The pyrolysis system has the capacity to process 30–50 kg/h of mixed single-use plastics (SUPs). The process is described in a recent related publication [54]. It begins with the pretreatment stage involving shredding, washing and drying, which results in a 5% loss of weight due to the elimination of impurities and moisture. The feedstock is then introduced into the reactor. The main process outputs (wt%) are bio-oil (79.38%), gas (15.22%) and char (5.4%). The bio-oil and gas are combusted in separate CHP engines to generate electricity. Electricity from bio-oil is directed for grid supply, while electricity from the gas is internally recycled to provide the process energy for the reactor. The energy output (kWh/ton) of input waste from bio-oil estimation is based on [55]. Other outputs from the system are the wastewater from washing and emissions from the CHP engine. The main source of emissions is the CHP since the reactor is specifically designed to operate under airtight conditions. Emission factors from [55] are used. The inventory analysis is summarized in

Table 3. Inventory for plastic pyrolysis process.

Flow Direction	Parameter	Units	Value
1. Pretreatment
Inputs	Waste–plastic mixture	tons	1.00
	Water (well)	m3	0.15
	Electricity	kWh	1.00 × 10−5
Outputs	Waste–plastic mixture	tons	0.95
	Wastewater	m3	0.165
2. Pyrolysis
Inputs	Waste–plastic mixture	tons	0.95
	Electricity (solar PV)	kWh	0.009
Outputs	Bio-oil	tons	0.75
	Char	tons	0.05
	Syngas	m3	0.13
3. Cogeneration
Inputs	Bio-oil	tons	0.75
Outputs	Electricity	kWh	0.0033
	Heat	kWh	0.0093
	Carbon dioxide (non-fossil)	kg	2.15
	N2O	kg	0.0003
	Methane (non-fossil)	kg	0.0001
	NO2	kg	0.0039
	SO2	kg	0.0209

below.

2.4.4. RDF Production

The conversion of residuals to RDF involves three main stages: separation to remove complex waste and recyclables such as glass, metals or e-waste, cutting/shredding to homogenize waste into required particle sizes and finally compacting to reduce volume for easier transportation [56]. Energy demand for these processes has been computed using the power ratings of the equipment onsite and the 8 h of daily operating. Energy is sourced from internal solar PV system; thus, pollutant emissions are negligible. The inventory for the RDF process is specified below in

Table 4. Inventory for RDF production.

Flow Direction	Parameter	Units	Value
Inputs	Residual materials	tons	1.00
	Electricity (solar PV)	kWh	51.58
Outputs	RDF pellets	tons	0.95
	Heat (loss)	kWh	33.0
	Trace particles	kg	0.05

.

2.5. Life Cycle Impact Assessment (LCA)

The OpenLCA 2.3.1 software was used to model each process following the guidelines of the ISO 14040/14044:2006 [35,36]. According to ISO 14042 [57], LCA process follows three mandatory steps: selection of characterization methods, classification and characterization. Additional optional steps are also recommended: normalization, grouping, weighing and data quality analysis [57].

The following steps were therefore followed with the aid of the OpenLCA software. Firstly, the impact categories, indicators and characterization models were selected. ReCiPe2016 midpoint (H) impact assessment method was used, which identifies 17 midpoint characterization factors that are linked to three endpoint protection areas [58,59]. This study focuses on ten midpoint impact categories due to their relevance to the scope, as is done in similar studies and this is represented in

Table 5. Selected impact categories and their relevance to the study.

Impact Categories	Relevance to the Study	Link to Endpoint Impact	Unit
GWP	Risk to human health and ecosystem	Human health	kgCO2eq
PMFP	Increase in respiratory diseases		kg PM2.5 eq
HTP/C	Increased risk of carcinogenic diseases		kg 1,4-DCB
HTP/NC	Increase in other diseases		kg 1,4-DCB
WCP	Reduced access for consumptive uses		m3
TETP	Risk to terrestrial spp.	Ecosystems	kg 1,4-DCB
FETP	Risk to freshwater spp.		kg 1,4-DCB
AP	Risk to terrestrial spp.		kg SO2 eq
METP	Risk to marine life		kg 1,4-DCB
FFP	Increased scarcity and cost of extraction	Resource availability	kg oil eq

[60].

The second stage involves assignment of LCIA results (classification). In OpenLCA, the inventory results were input for each process within the system boundary. The Ecoinvent Database v3.5 provided secondary data in the calculation of impacts. The category indicator results were then calculated by the software. Normalization is recommended in LCIA to assess the magnitude of the category indicator results relative to a reference value. This was carried out based on the World 2010 normalization method [61].

2.6. Interpretation of Environmental Impacts

The interpretation of results of LCIA was conducted according to ISO 14044:2006 [35]. This involves evaluation of results through checks for consistency, completeness and analysis of results. Results were then assessed through an iterative process where significant issues were searched, checked and analyzed. When significant issues were detected, corrections were made, and the simulation was re-run [57]. The characterization and normalization results were then imported from OpenLCA to be analyzed with Microsoft Excel. The characterization results were adjusted to percentage to assess the relative contribution of various impact categories. However, uncertainty results were exported from the OpenLCA software and analyzed using Python (Jupyter Notebook).

Interpretation of results includes sensitivity analysis and uncertainty analysis. This is done to check for the robustness, consistency and completeness of the data used in the analysis. A Monte Carlo simulation was conducted using 1000 iterations to perform the uncertainty analysis of the results. The data quality pedigree matrix was used to assign uncertainty values to the dataset used during the life cycle inventory stage [62].

3. Results and Discussion

3.1. Impact Assessment of Subsystems

This study focuses on an assessment of four subsystems at the WtE plant. Figure 2 shows the relative environmental impact of the various subsystems compared to the impact factors. Generally, pyrolysis shows a high negative impact across all impact categories, representing a high avoided environmental burden. The other subsystems show a positive impact, but one that is not so high. Segregation and RDF have positive values showing environmental burdens. AD, on the other hand, shows a generally neutral environmental impact. A neutral GWP was also recorded in a similar study, which assessed an AD system using different energy consumption scenarios [50]. In another study comparing AD with incineration, which is also a thermal conversion technology, AD showed near-neutral impacts, while incineration showed a high negative impact.

Figure 3, on the other hand, shows the percentage comparison of impact categories to the various subsystems. Pyrolysis for instance shows better environmental impact due to the production of new products such as char and digestate in the AD process compared to the negative impact the processes themselves generate [63]. Pyrolysis oil can also be considered a fuel similar to diesel [64]. Sources of environmental impact for the AD system include methane leakages from the biodigesters and the combustion of methane in the internal combustion engine. The highest environmental impact on soil and water in terms of toxicity may potentially originate from the digestate [65]. Overall, the magnitude of environmental burdens of the processes can be summarized as follows: RDF > segregation >AD > pyrolysis. RDF production and segregation, on the other hand, release process emissions, especially particulates [66].

3.2. Impact Assessment According to Impact Categories

Figure 4 shows the net contribution of all processes to the impact categories. The net emissions represent a balance of positive (burdens) and negative (offsets). This highlights the actual burden after accounting for reductions. The total net GWP for the WtE plant is estimated at −2.382 tCO₂eq/FU.

Figure 5 shows the relative contribution of impact categories based on normalized results. It corroborates the results of Figure 3, where the dominant subsystems are pyrolysis and AD.

3.2.1. GWP

The GWP is strikingly the highest impact category with the highest avoided burden compared to the other impact categories. GWP is an important metric for environmental emissions and climate change. Pyrolysis had the lowest negative GWP value of −2.3 tCO₂eq/FU, indicating high carbon savings (Figure 3). Net emissions are important in determining the burdens.

On the other hand, the GWP₁₀₀ is presented in Figure 6 for the same 50 t/day of waste disposed of at the landfill site. This is estimated to be 1.133 tCO₂eq/ton of waste-in-place by the end of the lifespan of the landfill. By the time of closure, the landfill would have emitted 4409.56 tons of uncontrolled methane into the atmosphere, although GHG emissions would continue for another 100 years. The total GWP over the lifespan is estimated to be 413,564.21 tCO₂eq since the landfill under consideration has no gas control infrastructure. The GWP of the landfill is based on uncontrolled methane emissions and non-biogenic CO₂.

The WtE plant reduces emissions by 3.52 tCO₂eq/FU, representing the avoided emissions. The WtE plant is expected to treat 365,000 tons of MSW over its 20-year lifespan. With the calculated net carbon reductions, this could result in carbon savings of an estimated 1.28 million tCO₂eq. This may be associated directly or indirectly with the financial feasibility of the WtE plant if such reductions are adequately verified [67]. The net emissions show that the WtE plant reduces emissions by 310%. This may, however, be subject to differences if long-term monitoring is conducted over the entire lifespan of the WtE plant compared to the long-term considerations made for the landfill. In a comparative study in Nigeria, a hybrid AD/incineration system and an LFGTE had the potential to release 75–93% and 400% less, respectively, compared to conventional landfilling [44]. A similar result was obtained in a study in the US where landfilling released 75.24% higher GHG compared to gasification systems [68]. In another study in Australia, landfilling recorded the highest environmental burden in 9 of 10 midpoint impact factors assessed, indicating that landfilling is the least environmentally friendly option [60]. Since the source of GHG at landfill sites is the biodegradable components and non-biogenic CO₂, the biochemical process can be compared to the AD system at the WtE plant, which has a GWP of 0.0032 tCO₂eq/FU. Although recent studies in Ghana have estimated that LFGTE is a feasible option, AD is even more efficient, with a lower GWP compared to LFGTE [28]. This is attributed to higher losses due to landfill cover of 10% and gas capture efficiency of 75% compared to 5% losses from the biodigesters [43,69].

The carbon intensity of WtE technologies is linked to the source of energy [60]. Sources of emissions from the WtE plant include engines and leakages of methane from AD. Energy for the RDF and segregation equipment is sourced from internally generated electricity, solar PV systems and biogas, which has a low emission profile. Also, heat energy losses from the RDF and segregation systems add up to environmental burdens. Assuming electricity for RDF pelletizing is sourced from fossil sources (diesel), avoided emissions can be estimated using emission factors [70]. With an energy demand of 452 kWh/day at full capacity, the GWP of a diesel-fired RDF pelletizer machine is estimated at 13.84 kgCO₂eq/ton, which, in this study, can be considered an avoided emission. RDF produced using cleaner energy reduces the overall environmental footprint of the energy produced from it. In addition, combustibles and other non-biodegradable components of waste are more efficiently treated with thermal energy conversion technologies such as gasification and pyrolysis rather than storage in land space at a landfill.

Studies have shown that the pretreatment and dewatering of waste significantly reduce emissions [13]. The low GWP recorded for pyrolysis may be a combination of the source of energy (solar PV and syngas), incorporating pretreatment and potential carbon sequestration from char. Char produced from pyrolysis possess has the capacity to sequester carbon, leading to the negative release of carbon [63,71]. Additionally, bio-oil and syngas replace fossil fuels for the production of electricity, thus further accounting for the avoided release of stored/fossil carbon, which could potentially end up in the atmosphere.

3.2.2. Toxicity Potentials

Toxicity represents the environmental fate and accumulation of a chemical in the food chain and its associated effects [58]. The toxicity potentials analyzed in this study are Human Toxicity Potentials (HTP) and Ecotoxicity Potentials (EP) in terms of Terrestrial (TETP), Freshwater (FETP) and Marine Ecotoxicity (METP). HTP is presented in two parts by the ReCiPe method: carcinogenic (HTP/C) and non-carcinogenic (HTP/NC). Both are linked to endpoint damage to human health due to an increase in lifetime disease incidence as a result of the intake of substances such as contaminated water or food from contaminated soil [59]. Generally, AD showed a high environmental burden due to the release of digestate. Digestate is planned to be upgraded into a soil amendment product. In the case that high quantities of heavy metals and contaminants are present, digestate may potentially have an alarming disease-causing impact on humans [72].

HTP is a very important potential risk that should be further investigated due to the mixed nature of the waste. High concentrations of heavy metals such as zinc, copper, lead, chromium and nickel have been detected in waste from the same region [73]. Heavy metals and chlorines in RDF must also be avoided due to their potential risk when gasified at industrial plants [41]. HTP/NC recorded the second-highest avoided impact for pyrolysis due to the production of char. Char has the potential to adsorb such pollutants and has been effectively used in soil/wastewater remediation. This can, however, be substantial only if the char is utilized for this purpose and the sorption capacity is improved through techniques such as grinding [74].

The ecotoxicological effect measures the Potentially Affected Fraction (PAF) or the change in Potentially Disappeared Fractions (PDF) of ecological species due to exposure to unacceptable pollutant concentrations [59,75]. TETP, FETP and METP measure the impact on land-dependent, freshwater or marine organisms, respectively. EPs are notably high for AD, and the lowest avoided burdens for pyrolysis are in the order of TETP > METP > FETP. Digestate and methane losses from the AD are a major source of ET to ecological systems. Methane emissions from AD are a potent GHG, which can potentially have a greater negative impact on PAF and PDF. Marine ecosystems have been shown to be highly sensitive to changes in global climate changes from GHG. Also, the bioaccumulation of chemicals such as heavy metals in species and the food chain as a result of digestate is a source of toxicity to land, freshwater and marine environments [76]. Since this is a potential environmental burden, it is necessary to carefully monitor and control the use of digestate. In the case of pyrolysis, avoided emissions due to the substitution of internally generated energy and the avoided use of grid-powered electricity from fossil sources account for high negative impacts.

3.2.3. Water Consumption Potential

Water is an important resource for the AD and pyrolysis process that is sourced from onsite groundwater (borehole). Water consumption/depletion is assessed as a negative impact when the abstracted water evaporates or is used in a production process far from the point of abstraction, which reduces human access to the resource at the point of abstraction [61]. The pretreatment stage of pyrolysis consumes water, which is a source of environmental burden, while the RDF is insignificant. However, in this LCA model, both pyrolytic water and liquid digestate can be reused/recycled which reduces the demand for freshwater. However, the burden of AD is potentially higher if digestate is transported to another site and cannot be reused.

3.2.4. Particulate Matter Formation Potential (PMFP)

PMFP measures the endpoint impact of fine particles/aerosols with a diameter of less than 2.5 µm (PM_2.5) on human health. These particles are a complex mixture of organic and inorganic substances, which has a potential effect on the cardio-pulmonary system and mortality after long-term exposure [58,59]. This is a very important parameter because the WtE plant under study requires human operations for at least 40 h a week. Although all the PM values recorded indicated a low impact (−0.21 to 0.03 kg PM_2.5 eq), the emission factors used are only limited to the engine. Data on site-specific operational emissions are not available. Air quality measurements are highly recommended for future studies to determine the exact potential PM emissions at various waste treatment plants across Ghana and the corresponding impact on humans and the ecosystem. In a previous study to assess the concentrations of PM in the breathing zones of informal e-waste workers in Ghana, the PM 2.5, 2.5–10 concentrations were 2–3 times above a control group and also above the WHO air quality standards. The study recommended the assessment of the impact of long-term exposure [77]. In the long term, PM may also impact the ecosystem of the area, which is predominantly agricultural, thus requiring further investigation.

3.2.5. Terrestrial Acidification Potential

Acidification Potential (AP) measures the potential impact of acidifying substances such as NO_x, NH₃ and SO₂ on soils and their dependent species [58]. The values of AP were generally low, with a net avoided impact of 0.31 kg SO₂ eq, which is linked to the low emissions of these pollutants directly from the systems. The use of electricity from the solar PV system contributes to avoided emissions.

3.2.6. Fossil Resource Scarcity

Fossil resources are subject to depletion apart from their high climate footprint once combusted. The increased substitution of fossil energy for renewable energy is important. The energy mix in Ghana is composed of 68% generation from thermal combustion of fossils [78].

3.3. Uncertainty Analysis

Each box plot shows the distribution of uncertainty values for each impact category (Figure 7). Uncertainty in the LCA results highlights the variability and spread of data within each impact category. High uncertainty is recorded for HTP/NC, HTP/C and FETP, which indicates a high level of uncertainty in the input data, suggesting that the impact assessments for this category are highly variable and potentially influenced by many factors. FETP, METP and GWP, on the other hand, show moderate uncertainty with several outliers, indicating some level of variability. FFP, WCP, PMFP and TETP show relatively low levels of uncertainty, with smaller IQRs and fewer outliers. This suggests that the data for these impact categories are more consistent and reliable. Categories with low uncertainty provide more confidence in the results and can be used more reliably in environmental impact assessments and sustainability planning. In summary, the uncertainty plot provides valuable insights into the reliability and variability of LCA results across different impact categories.

3.4. Other Potential Impacts

Other potential impacts include socio-economic and public health improvement in rural communities. According to the 2021 census data, 36.8% of rural households have no waste collection, while 57.3% use open dumps with potential health impacts. With the plant located within a municipality with 64.7% rural households, the proximity of a WtE plant will likely reduce the cost of waste collection services and encourage more residents to pay for WtE services. Also, the plant currently provides employment and income security to many residents, while compost may support agriculture at a lower cost since this is the most prevalent economic activity within the town [39]. Additionally, a buffer zone has been created around the plant site within a 1 km distance from the closest households for protection against fire outbreaks, odor and noise.

The promotion of WtE in such settings has climate mitigation impacts at both national and international levels. Ghana’s NDCs aim to reduce GHGs by 64 MtCO₂eq and create one million decent and green jobs by 2030. The expansion of WtE systems within a circular economy concept will support the attainment of such national targets with global impacts [79]. As of 2022, Ghana’s national energy mix was made up of 1.9%, 70% and 28.1% renewable energy, fossil fuels and hydro, respectively. The sustained substitution of WtE with fossil fuels is important to reaching the set targets and protecting global resources.

3.5. Limitations and Recommendations

One major limitation of this comparative study is the lack of an ongoing monitoring system with a recent, detailed localized database on waste generation and treatment pathways across cities [38]. The Ecoinvent Database, for instance, does not contain some locally specific data for Ghana, especially for the plant site in Atwima Nwabiagya. Where locally specific data were not available, data from the rest of the world were used during the modeling in OpenLCA. This provides a limitation to the results, which may have a substantial deviation from the site-specific environmental impacts.

It is strongly proposed that a country-specific database be developed for Ghana by the government. Since Ghana is a signatory to the Paris Agreement, such a database becomes a technical resource that will enhance the accuracy of climate monitoring and auditing. The development of a database can be facilitated through the following:

• Dedicating funding to this;
• Dedicating human resources for national coordination, such as incorporating it into the Ghana Carbon Market Office;
• Future research focused on carrying out detailed and up-to-date studies on aspects such as national waste composition similar to that carried out in 2014 [80]. Other important areas of concern are emission factors and operation-related emissions from waste treatment activities;
• Facilitating training programs for waste management companies, research institutions, agencies and decision-makers to enhance monitoring and compliance.

The uncertainty in LCA results is crucial for decision-making and policy development. High uncertainty in certain impact categories, such as HTP, indicates that more research and data collection are needed to improve the reliability of these assessments. Another major source of concern is the potential impact of mixed waste on the HTP as well as ETP through bioaccumulation in terrestrial, freshwater or marine species. The waste segregation model for the socio-cultural setting in Ghana needs to be studied and implemented to avoid such impacts.

Life cycle costing is highly recommended in future studies to incorporate the cost of environmental impact in decision-making. Carbon trading can potentially stimulate more emission reduction developmental projects in local communities [81]. It is important for WtE plants to obtain international certifications such as ISO 14001, which may return positive financial benefits for the case study plant, although it requires adherence to strict requirements. Environmental assessment studies such as this to support carbon financing mechanisms that require international certification [82].

4. Conclusions

This study conducted an environmental assessment of various processes within a pilot waste-to-energy plant in Ghana. The Life Cycle Assessment was conducted using OpenLCA software version 2.3.1. Landfill emissions were considered the baseline for estimating potential emission reduction. The results indicate that the GWP has the highest potential environmental impact followed by the toxicity potentials, mainly high Human Toxicity Potentials. Pyrolysis showed high avoided impacts, while AD showed near-zero impacts. The major attributing factors are the potential sequestering of the produced char end-product and the input of renewable energy. Overall, pyrolysis technology is the most environmentally friendly. This is relevant because a comparative assessment of the GWP of landfilling showed that the WtE plant has the potential to reduce emissions by 3.52 tCO₂eq/FU, resulting in a total of 1.28 million tCO₂eq over its 20-year lifespan based on the expected capacity. This can directly or indirectly potentially improve the business case of the plant. Uncertainty assessment also points to the need for a national database with relevant localized data to improve the reliability of future studies. Also, although AD showed a near-neutral environmental impact, the release of digestate may have high human and ecotoxicity impacts when substances such as heavy metals are present. This can be a potential risk in Ghana, where the source segregation of waste is largely non-existent. As such, the implementation of AD technology must be associated with strict regulations and monitoring systems on the management of digestate. Based on these findings, it is recommended that pyrolysis of end-of-life and non-recyclable plastics be incorporated strongly into the waste-to-energy models. Decision makers have a role to play by introducing innovative approaches to encourage source segregation. It is also recommended to use strict guidelines and monitoring systems to ensure that emission control is implemented by waste-to-energy plants according to environmental compliance. With the siting of the WtE plant in a rural community, the municipal authorities are encouraged to take advantage of this to stimulate community participation in environmental protection by providing incentives such as lower waste collection costs. In the long term, this WtE plant may contribute relevant data for improving decision-making based on environmental assessments in Ghana. It may also provide a basis for carbon trading and enhance the financing profile of the WtE model studied. Future studies may be extended to include other processes such as composting and recycling.