A Case Study of Renewable Natural Gas Techno-Economics and Emissions at a Wastewater Treatment Plant

Abstract

Renewable natural gas derived from biogas presents a viable pathway for decarbonizing natural gas systems. Wastewater treatment plants equipped with anaerobic digesters often flare or utilize biogas for heat and electricity generation, missing the potential of renewable natural gas. This study investigates the techno-economic feasibility and emission impact of a renewable natural gas system at a small wastewater treatment plant in the Southwestern United States. Due to a lack of existing data on biogas composition and contaminants from plants originating within the United States, samples were tested seasonally for a year, and gas components are reported in this analysis. Using hourly biogas production data, the analysis incorporates costs and renewable fuel credits at 2022 market prices and the cost to remove the biogas contaminants. The results show a 15-year net present value of USD 16.3 million, with a payback period of three years, surpassing the economic performance of combined heat and power systems previously assessed for the same facility. Additionally, renewable natural gas systems achieve a 22% reduction in site emissions compared to combined heat and power systems. These findings highlight renewable natural gas as a profitable and environmentally superior alternative for biogas utilization in small-scale wastewater treatment plants, contingent on access to renewable fuel credits.

1. Introduction

Renewable natural gas (RNG), also known as biomethane generated from biogas offers an additional pathway to profit from biogas production while reducing carbon emissions. Biogas is naturally produced in the process of anaerobic digestion, in which organic matter is broken down by microbes without the presence of oxygen. Biogas produced by anerobic digestion primarily consists of methane (CH₄) and carbon dioxide (CO₂) along with other trace contaminants that must be removed before the gas can be used [1]. The gas can be cleaned in a process called “upgrading” in which the contaminant concentrations are reduced to the required levels, turning the raw biogas to pipeline quality natural gas [1]. Once this gas is compressed to pipeline operating pressure, the upgraded biogas can displace natural gas in multiple end uses. This enables processes that require natural gas to utilize a lower net emission and renewable fuel source.

Due to the environmentally harmful nature of biogas, it must be processed and should not be released to the environment. Anaerobic digestion is one process used to generate biogas from the sludge developed in the wastewater treatment process and thereby recover some of the available chemical energy. In addition to the chemical energy, anerobic digestion prevents contaminants from municipal wastewater—like nitrogen and phosphorus—from eutrophicating into other freshwater streams like lakes and rivers. Currently, most biogas produced from wastewater treatment plants (WWTPs) is flared or used on site instead of upgraded for reuse [2]. Flaring is the process of burning biogas after it is produced through anerobic digestion to turn environmentally harmful gases like methane into less harmful CO₂ [3]. This process burns the biogas as waste and does not typically allow for heat recovery after combustion. The flaring process is operated inexpensively and requires little maintenance and upkeep compared to alternative methods [4]. Two alternatives to flaring biogas are cleaning and upgrading the gas to be reused off site or for it to be used on site in combined heat and power (CHP) which can offset thermal energy demand required from the anerobic digestion process and electricity at the plant [5,6,7]. Both options reduce emissions of the WWTP and offer potential revenue streams [4,8]. While CHP and RNG may reduce emissions, they may not be the most economical choices for certain-sized plants, which is why many may opt to flare biogas [9]. Biogas from WWTPs is estimated to be able to meet 12% of the United States electricity demand [10]. Despite the potential, many factors prevent this large and renewable energy supply from being tapped, which will be further explored.

There are over 16,000 WWTPs in the United States, but only 1300 plants take advantage of anaerobic digestion, while many other plant types use alternatives instead [2,8,11]. These 1300 WWTPs represent a large potential source for easily usable RNG in the U.S., and only about 860 of those plants utilize any of their biogas for CHP or RNG, while the rest flare their biogas to the environment [11]. Smaller WWTPs (<400 m³hr⁻¹ biogas flowrate) that operate with less total flow of wastewater make up the majority of plants in this category and tend not to be profitable for RNG or CHP due to low biogas production and significantly longer paybacks due to the high initial capital costs [2,4,8,9]. Additionally, other factors affecting biogas produced from wastewater are its additional contaminants that must be removed compared to other waste streams due to the nature of the source of the waste [8,12,13,14,15]. These components consist of food additives such as siloxanes and other common household chemicals which are present in municipal wastewater but not observed in biogas produced from other waste streams. Other constraints involve the location of the plant or distance from the nearest pipeline, which increases capital costs, and other potential supply chain issues in obtaining the required plant materials themselves [8,16,17]. These factors can increase the capital costs or other recurring costs involved in transporting RNG to the nearest fueling station, making small-scale projects historically unprofitable for smaller WWTPs.

Due to small-scale plants having poor paybacks, many previous studies have focused on larger plants. This causes many datasets to skew towards already-existing large-scale biogas plants, which typically only represent the largest WWTPs. Based on data reported to the EPA and Argonne National Lab, the median RNG facility in the U.S. operates with a biogas flowrate of 1074 m³h⁻¹, which represents 283 facilities with reported flowrates [18,19]. The distribution of the flowrate of biogas to the upgrading system in m³h⁻¹ is shown in Figure 1. Just under 40% of plants fall below 500 m³h⁻¹ of flowrate, with the majority of these plants utilizing agriculture products (both crop and livestock waste) to produce biogas in the anerobic digestion process instead of municipal wastewater. A breakdown by type of feedstock is shown in Figure 1B. These plants do not have additional cleaning costs to remove impurities that are present in wastewater or landfills, which results in greater profitability at lower flowrates. Only 28 plants produce RNG from wastewater, and these plants have an average biogas flowrate of 1054 m³h⁻¹ of biogas [18]. The plant studied in this report has a flowrate below 450 m³h⁻¹, which firmly puts it well below the average plant. Cucchiella et al. showed how the lack of profitability is a major detractor for low-flow biogas systems from wastewater in the European market by simulating biogas plants below 1000 m³h^-1 and finding a negative net present value for smaller-sized plant configurations [20]. Sun et al. reviewed common upgrading technologies and show that lower flow systems have higher costs per unit flowrate of biogas [15]. Other studies have shown that lower flow systems have higher operating and electricity costs per unit of biogas upgraded compared to higher flow systems, but few studies have focused on smaller plants in the U.S. [4,8,12,14,15]. In addition, many previous studies focus on the European market, which has historically higher natural gas prices, making the systems significantly more valuable over time compared to U.S.-based systems [15,20,21].

There are four major cleaning systems that upgrade biogas to renewable natural gas, and the optimal system is dependent on the plants’ flowrate, gas composition, and other local factors. All methods may require additional steps to remove siloxane contamination depending on local requirements [22,23]. Pressurized water scrubbing is the most commonly used technique to clean biogas [8,24,25,26,27,28]. It offers a cost-effective way to remove carbon dioxide and hydrogen sulfide by absorbing the contaminant gases into high-pressure water [24,25,26]. Pressure swing adsorption, which is the second most utilized cleaning method, compresses the gas and removes contaminants by utilizing molecular sieves or other similar adsorbents to purify the methane stream [8,28,29,30,31,32,33]. Solvent scrubbing utilizes a chemical solvent to absorb contaminant gases, which can later be regenerated [32,33,34]. The final main method used to clean biogas is utilizing membranes to selectively allow methane to pass through while blocking contaminants [35,36,37]. There is no clear “best” technology, as many have different tradeoffs (ability to use waste heat, smaller plant footprint, etc.) which leads to most plants utilizing the best technology for their specific constraints [28]. Figure 2A,B show the distribution of cleaning technologies for biogas-producing landfills and manure-based plants [19]. These plots show that the lower-flowrate manure-based systems tend to use membrane or water scrubbing systems, which are typically used at facilities of these sizes [8]. For the purposes of this paper, all technologies and their costs are utilized, and a function is fitted to the data to determine system cost per biogas flowrate.

More work is needed to assess the techno-economics of RNG installations in the U.S. because less is known about the biogas composition in the U.S. and other economic factors need to be considered [38]. Accounting for all biogas technologies including CHP, Europe generated 70% of the world’s biogas, while North America only generated about 16% in 2017 [39]. Analyzing the biogas composition is essential to accurately calculate upgrading costs and requirements. Despite this, few studies have analyzed biogas composition generated from anerobic digestors in U.S.-based WWTPs. Admasu et al. analyzed the biogas composition produced from sludge originating from the beverage industry using gas chromatography, but this composition may not be applicable to WWTPs [40,41,42]. In addition, anaerobic digesters have optimal temperature ranges and other factors that affect the design and biogas composition, which can cause variation between locations [22,42,43]. Few results have been published showing the biogas composition for WWTPs located in the U.S., especially with common contaminants like siloxane [41,44]. Kazimierz compiled a dataset of biogas plants throughout Europe and their relevant siloxane components [45]. Their data showed large variations in siloxane composition between plants that must be accounted for in upgrading costs. These variations can have significant effects on biogas cleaning capital and long-term operating costs.

Additional economic constraints exist in the U.S. that should be considered in RNG assessments, including the Renewable Fuel Standard (RFS) program, which was created in 2005 [44,46]. The RFS creates renewable fuel identification numbers (i.e., RINs) to be traded on the market, which are created by renewable fuel producers and assigned a D-code depending on the source of the feedstock. WWTPs produce D-5 advanced biofuel RINs at the time of this writing [46]. The RFS program can be a significant source of income for any WWTP [8,47]. Previous studies have shown that the incorporation of RIN credits and other low-carbon fuel credit systems significantly increase the profitability of biogas on a volume basis [44,48,49]. Yang et al. showed that the inclusion of RINs and California low-carbon fuel standard credits provided the optimal pricing structure when including carbon capture compared to CHP and RNG without carbon capture [49]. While their analysis considered the U.S. market, they considered biogas production from agriculture crops which takes advantage of higher-value RIN credits and have different biogas production rates compared to WWTP digesters [11,46,49].

While numerous studies have evaluated the potential of renewable natural gas systems, significant gaps remain in understanding their application at small-scale WWTPs in the United States. Most existing analyses focus on large-scale facilities with high biogas flowrates, often overlooking smaller WWTPs, which constitute a substantial portion of potential RNG producers. Additionally, many studies emphasize the European market, where higher natural gas prices and distinct regulatory frameworks drive profitability, limiting the applicability of findings to U.S.-based systems. There is also a lack of comprehensive data on biogas composition from U.S. WWTPs, particularly regarding siloxane contamination, which can significantly affect cleaning costs and system design. Furthermore, while techno-economic comparisons often explore CHP systems, few studies have directly contrasted the economic and environmental impacts of RNG and CHP at smaller facilities. Addressing these gaps is critical for guiding the deployment of RNG technologies in underutilized markets and ensuring accurate assessments of their feasibility and benefits.

This paper will address the potential benefits of installing an RNG system at a small WWTP located in the Southwestern U.S. that already has an anaerobic digester. The average biogas flowrate of the plant is less than 450 m³h⁻¹, which places it below the bottom 40% of all biogas-to-RNG producers in the U.S. and is smaller than most of the facilities assessed in the literature [4,8,18,19]. The results will be compared to other technologies like CHP, which has been shown to be profitable for this facility within a 20-year lifetime in a separate report [49,50]. The techno-economic model will include profit from renewable fuel credits. In addition, this report will compare the economics and emissions of CHP to RNG over its lifecycle, utilizing data from actual plants located in the United States. The biogas composition from a WWTP anaerobic digester located in the Southwestern U.S. is analyzed and reported for the first time and used to gauge potential cleaning costs for the system. The main components of the biogas are analyzed and larger hydrocarbons, including siloxanes, are quantified using gas chromatography.

2. Materials and Methods

2.1. Gas Chromatography

Analysis of the biogas composition was conducted with a gas chromatograph (Shimadzu gc 2014) with a polysiloxane-fused silica column, a setup that has not been reported and required the development of a procedure. The column is 30 m in length with an inner diameter of 0.32 mm and a film thickness of 0.25 µm. The total flowrate of gas was 11.5 mL/min with a purge flow of 3 mL/min and a split ratio of 4. After the biogas sample was inserted, the column temperature was held at 60 °C for 4 min and increased to 250 °C at a rate of 8 °C per minute. The temperature was held at 250 °C for a total of 6 min. The column was calibrated for siloxanes ranging from D3 to D5 and L2 through L5 listed in

Table 1. List of common siloxanes found in wastewater-derived biogas, including chemical names, abbreviations, and molecular formulas.

Organic Silicon (Siloxanes)	Abbreviation	Molecular Formula
Decamethylcyclopentasiloxane	D5	C10H30O5Si5
Dodecamethylpentasiloxane	L5	C12H36O4Si5
Octamethylcyclotetrasiloxane	D4	C8H24O4Si4
Decamethyltetrasiloxane	L4	C10H30O3Si4
Hexamethylcyclotrisiloxane	D3	C6H18O3Si3
Octamethyltrisiloxane	L3	C8H24O2Si3
Hexamethyldisiloxane	L2	C6H18OSi2

and was able to detect siloxanes at levels as low as 0.3 mg/m³ of biogas. Gas was obtained directly from the outlet of the anaerobic digester and stored in Tedlar plastic sample bags, which have been previously shown to be suitable for biogas storage [51]. Gas was obtained and stored at ambient temperature. The standard gases and small hydrocarbons (i.e., <C₆) were analyzed in a separate Shimadzu 2014 gas chromatograph having a flame ionization detector (FID) and two thermal conductivity detectors. The biogas composition was measured in August, October, and December 2021 and February 2022.

2.2. Techno-Economic Model

A techno-economic model was developed to determine the cost savings of upgrading biogas to RNG at the WWTP facility located in the Southwestern U.S. The costs utilized in the techno-economic model assumes that an anerobic digestor has already been installed and that only the upgrading technology needs to be installed on site. The biogas components used directly in this analysis measured at the facility are shown in

Table 2. Techno-economic model inputs and values based on the specific wastewater treatment plant located in the American Southwest.

Parameter	Value
CH4 concentration	62.24 vol%
CO2 concentration	37.36 vol%
Siloxanes	$12.9 mg Si/m^3$
H2S	126 ppmv
Max biogas flowrate (BGmax)	641 m3h−1
Discount rate (r)	5%
Capital cost (CC)	USD 6186${\left({\displaystyle \frac{m^3}{hr}}\right)}^{-1}$
Pipeline cost (PC)	USD 372,824/km
Interconnect cost (IC)	USD 400,000
O&M cost (O&Mcost)	$USD 0.106{\left({\displaystyle \frac{m^3}{hr}}\right)}^{-1}$
Natural gas cost (NGC)	USD 0.24 m−3

, with a full list shown in

Table 3. Individual siloxane compositions and concentrations present in biogas measured over a year from an anerobic digestor located in the Southwestern United States. Compositions of major biogas components and contaminants measured over the same period.

Organic Silicon (Siloxanes)	Conc. Unit	August	October	December	February
Decamethylcyclopentasiloxane (D5)	mg/m3	6.77	7.15	5.7	5.62
Dodecamethylpentasiloxane (L5)	mg/m3	2.67	0.37	0.069	<0.3
Octamethylcyclotetrasiloxane (D4)	mg/m3	2.01	2.64	0.34	5.37
Decamethyltetrasiloxane (L4)	mg/m3	0.64	0.52	2.81	2.24
Hexamethylcyclotrisiloxane (D3)	mg/m3	0.2	0.28	0.13	<0.3
Octamethyltrisiloxane (L3)	mg/m3	<0.3	<0.3	1.02	2.13
Hexamethyldisiloxane (L2)	mg/m3	0.48	0.4	0.3	0.37
Component	Conc. Unit	August	October	December	February
Methane	%	61.07	63.2	63.39	61.31
Carbon Dioxide	%	38.43	36.41	36.28	38.32
Nitrogen	%	0.31	0.29	0.25	0.3
Oxygen	%	0.11	0.1	0.08	0.077
Hydrogen Sulfide	ppmv	228.8	215	24	36.5
Total Siloxane	mg/m3	12.77	11.36	10.369	15.73

[49]. These components were used to estimate RNG upgrading capital and operations (O&M) costs based on the flowrate of biogas during the most recent year of the data provided. All equipment metrics including capital costs (CCs) and O&M costs per unit of biogas were determined from an EPA report which is based on a linear fit to real data and is shown in Table 2. This model does not utilize a specific technology type and instead is averaged across all technology types from all plants across the U.S. to create a more generalized model to estimate costs. To further develop the model, the discount rate of investment (r) is used. Costs were adjusted from the EPA’s published model to the value of the United States dollar in 2022 using the U.S. Bureau of Labor Statistics inflation calculator [8,52].

RNG upgrading requirements depend on the end use and transportation pathway. This model assumes pipeline injection as the proposed RNG pathway and the end use as vehicle fuel to obtain RIN credits. Based on the literature, current pipeline injection requires biogas to be upgraded to >97% CH₄ and <3% CO₂ and compression to a pressure of 10 MPa, but requirements can vary based on the agreement with the service provider [1,53,54]. In addition to the removal of CO₂ and water vapor, siloxanes must be removed to acceptable levels. Other viable RNG pathways include local usage as liquid natural gas (LNG) for use in transportation vehicles or by local municipalities [55]. When creating LNG, RNG is further purified to >99% CH₄ in the liquification process, generally cryogenically [12]. Converting biogas to LNG may have more stringent cleaning requirements for siloxanes and other contaminants, which can cause damage to combustion engines and will have a higher O&M cost [1,4,8]. If the gas is used locally in boilers, some compression and cleaning costs can be saved, but the gas may not qualify for the RFS and other types of renewable fuel credits [56].

Capital costs were estimated using the upper bound of the EPA cost recommendations published in their RNG handbook [8]. The EPA model is based on actual capital and operating costs for RNG plants across the U.S. but is not specific to installations at WWTPs. All values assume that an anaerobic digester is already installed on site and that gas flow is measured leaving the digester. The EPA last updated their RNG handbook in 2021 at the time of writing, and the model is assumed to be up to date [8]. The capital cost of the biogas cleaning and compression system was estimated using the highest flowrate value of biogas from the anerobic digester to ensure the system is not undersized. Decreasing the max flowrate of the cleaning system can result in capital expenditure savings. Careful analysis should be performed to ensure an appropriate value is chosen based on predicted future plant demand. The calculation of the total capital cost, $TCC$, is shown using Equation (1).

$$TCC=CC\times B{G}_{max}+PC+IC$$

(1)

In Equation (1), $CC$ is the capital cost from Table 2 and $B{G}_{max}$ is the max flowrate of biogas at the facility. PC is the pipeline cost for less than 1.6 km of the estimated pipeline needed to be installed, and IC is the cost for the interconnection to the NG grid [8]. Next, the O&M costs were calculated for the facility. This was calculated on an hourly basis to match the historic biogas flowrate for the plant and then summed over the period of a year. This cost is all-encompassing, including the annual fee of pipeline injection of USD 0.002370 per MJ, which can vary by utility [8]. Equation (2) shows the calculation of operating cost ($OC$) for a single day’s flowrate.

$$O{C}_{RNG,hourly}=O\&M_{cost}\times B{G}_{hourly}$$

(2)

The yearly income from selling the natural gas was calculated on an hourly basis similarly to the operating cost. The natural gas price was determined from the EIA historical data from 2021 [57]. This is multiplied by the amount of methane in the biogas stream to find the hourly profit of natural gas. A total leak rate (LR) of 1% was assumed for the additional piping and upgrading process, which is based on EPA estimates [2,4,8]. Equation (3) shows the calculation for yearly natural gas profit ($NGP$) from natural gas sales.

$$NG{P}_{hourly}=B{G}_{hourly}\times NGC\times (1-LR)$$

(3)

In addition, carbon credits were included in the yearly profit. Upgraded biogas counts as a renewable fuel for the Renewable Fuel Standard (RFS) program [46]. In order to qualify for the RFS program, the renewable fuel producer must demonstrate that the RNG produced on site is used as transportation fuel [46]. There are different ways to demonstrate this, and an agreement may need to be formed with the local natural gas utility [8,46]. For this analysis, an average RIN value of USD 1.63 per RIN credit, or about USD 0.79 per meter cubed of natural gas produced, was assumed. This is the average value determined from 2022 RIN prices obtained directly from the RFS market [58]. A sensitivity study is performed on the values in a later section. Equation (4) below shows the calculation for profit generated by RIN credits.

$$RF{P}_{hourly}=B{G}_{hourly}\times RIN$$

(4)

In order to estimate the value of the project, the net present value (NPV) was determined. The discounted rate of return, r, was estimated based on previous studies for municipal WWTPs funded through bond measures [8]. Equation (5) shows the calculation for NPV.

$$NPV={\sum }_0^L{\displaystyle \frac{\sum _{Year}NG{P}_{hourly}+\sum _{Year}RF{P}_{hourly}-\sum _{Year}O{C}_{RNG,hourly}}{{\left(1+r\right)}^t}}-TCC$$

(5)

A negative $NPV$ indicates that a project will not pay off in the number of years, t, from the investment date to the end of life (L), estimated to be 15 years for RNG [8]. When $NPV$ is positive, it is considered that the discounted cashflow is greater than the total capital cost, TCC.

2.3. Model Inputs and Limits

Historic RIN values for D-5 RINs are shown in Figure 3. The market is subject to volatility like most markets and depends on the production of obligated parties who must purchase RIN credits to offset their emissions. The minimum and maximum possible prices for D-5 RINs are capped at USD 0.05 and USD 3.00 by the EPA [46]. The 2022 average RIN price was used in the model as the base case. The limits of the sensitivity analysis compare the base case to a 20% variation in RIN cost along with the minimum and maximum possible values RIN trades are capped at.

Citygate natural gas prices in the Southwestern U.S. used in the model were obtained from the EIA to find the profit per meter cubed of natural gas. Citygate natural gas prices represent the price of natural gas after transport from the refinery to the area it will be used in [57]. Another type of pricing examined in the analysis is Henry Hub pricing [59]. Henry Hub represents the commodity price for natural gas and does not include transportation costs. Due to its similarity with the citygate pricing, citygate will be used for the analysis as it includes region-specific variation caused by local supply and demand along with transportation to the municipality. The actual prices for natural gas can vary depending specifically on the location the gas is injected at, the particular pipeline used for injection, and any agreements formed with the utility that services the pipeline [54]. Figure 3 shows historical natural gas prices for the specific region in the Southwestern U.S. starting in 2016. Due to the high variability in natural gas prices in the last few years, the model used the 2022 average natural gas price for its base estimate. Other potential prices were analyzed and included in the sensitivity study, such as the average price since 2016 (6-year price), the previous 3-year average, and a 10% increase in price compared to the base cost.

2.4. Emission Analysis

RNG offers pathways to reduce greenhouse gas (GHG) emissions by capturing biogas from the anaerobic digester and displacing carbon emissions from its proposed end use. Methane is the main component in biogas and a potent GHG that is more than 25 times more effective at trapping heat in the atmosphere than carbon dioxide [4,60]. Other GHGs and volatile organic compounds (VOCs) are created through the process of anaerobic digestion, which negatively impact air quality and haze in the area [4,14,15]. Most facilities flare biogas, which burns the biogas to turn the methane into $C{O}_2$, which is less harmful for the environment. The NO_x emissions should not be modeled by chemical equilibrium due to overestimation as a result of the slow kinetics and should instead be modeled with kinetics based on experimental data [61].

Williams et al. have compiled experimental results from actual biogas flares, CHP, and RNG systems, which can be used to analyze emissions [4]. Their values were scaled and used to determine emission release for the local plant, which does not include the end use of renewable natural gas if used as vehicle fuel [2,4,62]. The experimental results are incorporated into the hourly results to compare emission release across different technologies including CHP, RNG, and biogas flaring.

2.5. CHP Analysis

A previous report was generated for the facility to determine CHP sizing and payback [49]. The analysis and results were similar to the results published by Elio et al. in a separate case study [50]. The CHP system utilized a reciprocating internal combustion engine sized at 884 kW. The system utilized an electric load following its design and incorporated facility demand and biogas production together.

3. Results

3.1. Gas Chromatography

The gas chromatography results of the biogas composition are shown in Table 3, and a full report is available in Supplementary Materials. A decrease in overall siloxane concentration was observed during the winter months compared to the summer months. This could be due to ambient temperatures as different types of siloxane have different volatilities, which impact their concentration in the anaerobic digester. Methane remained mostly constant throughout all months, varying between 61 vol% and 63 vol%. Winter months produced far less hydrogen sulfide gas and siloxanes compared to other months of the year. These results indicate that a plant in the Southwestern U.S. could have significantly higher operating costs in the summer months for biogas cleaning compared with the winter months. In addition, installed technology must be able to handle peak siloxane and hydrogen sulfide flows produced by the anaerobic digester and should be adequately sized.

Individual siloxane compositions are shown in Table 3. D5 siloxanes correspond to the largest siloxane detected by concentration for each month tested. This pattern is similar to previous results for biogas generated from wastewater in anaerobic digesters located in Europe [63].

3.2. Techno-Economic Analysis Results

The results of the analysis are shown below in

Table 4. Model results showing capital and operating costs along with a comparison showing NPV at 15 years with and without RIN credits. Breakeven price shows the minimum combined RIN and NG price required for profit in 15 years.

Model	Model without RINs	Model with RINs
Capital cost	USD 4,963,351
Year 1 operating cost	USD 407,039
Breakeven price per unit NG	USD 0.373 m−3
Yearly profit	USD 578,359	USD 2,456,793
NPV at year 15	USD (3,185,106)	USD 16,312,396
PBP (year)	29	3

, including an analysis with RIN credits and an analysis without RIN credits. The model shows that upgrading the biogas to RNG and pipeline injection is profitable if RIN values are considered in the profits. If RINs are not included, the model does not pay off within its lifetime of 15 years [8]. This demonstrates the importance incentives have for the overall profitability of the model.

A sensitivity analysis was performed to demonstrate how the RIN price effects the NPV and is shown in Figure 4A by varying the RIN cost by 20%. In addition, Figure 4A shows the NPV at the capped high and low prices. It is important to note that the RIN market has high volatility as shown in Figure 3A, but only 1% of historic values from the last 6 years fall below the breakeven cost at the current 2022 citygate natural gas cost in Arizona. To operate profitably, the lowest RIN value the facility should operate at is USD 0.267 per RIN. Ignoring any profit from natural gas, to be solely profitable on RIN sales alone, the facility would need to sell its RINs for a minimum value of USD 0.77 per RIN.

A similar analysis was performed using natural gas prices and is shown in Figure 4B. The figure shows that the price of natural gas does not impact the NPV as significantly as RIN values did in previous graphs. The base case is a 40% increase compared to the previous 2016–2022 average price but only impacts the overall NPV at its 15-year lifespan by 15%. The future of natural gas prices is uncertain, but this indicates a fair degree of resistance to fluctuations in natural gas cost in the model. This is in part due to the high value of the RIN credit making up for the lack of profit from the natural gas on its own.

Table 5. The 15-year NPV for a renewable natural gas plant with varying RIN prices and natural gas pricing. The percentage difference from the base 2022 average RIN value and the 2022 base biogas pricing are shown.

Method	$\mathbf{RIN} \mathbf{Price} \left(\frac{\mathit{U}\mathit{S}\mathit{D}}{\mathit{R}\mathit{I}\mathit{N}}\right)$	NPV at 15 Years $\left(\mathit{U}\mathit{S}\mathit{D}\right)$	NPV Percent Change from Base
Lowest possible	0.05	USD (2,587,023)	(115)%
Breakeven price	0.267	8657	(100)%
20% below base	1.304	12,412,896	(24)%
Base 2022 average	1.63	16,312,396	-
20% above base	1.956	20,211,896	24%
Max possible	3	32,699,867	100%
Method	Natural Gas Price $\left({\displaystyle \frac{\mathit{U}\mathit{S}\mathit{D}}{{\mathit{m}}^3}}\right)$	NPV at 15 years (USD)	NPV Percent Change from Base
Base + 10%	0.267	16,899,877	3.6%
2022 average (base)	0.243	16,307,459	-
2019–2022 average price	0.186	14,900,467	(8.6%)
2016–2022 average price	0.146	13,913,104	(15%)

summarizes results from Figure 4.

3.3. Emission Reduction

Figure 5 shows the annual amount of each type of emissions released for different pathways. These values were calculated based on experiential data obtained from Williams et al. on an hourly analysis to determine the annual emissions [4]. For all cases, $C{O}_2$ emissions include gases produced during the anerobic digestion process, along with those generated through flaring, RNG, or CHP (reciprocating engine). Emissions were reduced in the CHP case to account for the production of electricity [64]. To be consistent with other studies, only the local site emissions are accounted for in all three cases. Figure 5A illustrates the experimental data, which show RNG as consistently the lowest emitter, with CHP and flaring both alternating as the highest emitters.

For a more accurate comparison of emission reductions between technologies and emission types, emissions can be normalized in terms of kg of CO₂ equivalent released to the environment. To normalize the values, the emission’s mass is multiplied by their global warming impact. The GWP for CH₄ is 27 and that for N₂O is 273 when compared to CO₂ [60]. The results from this comparison are shown in Figure 5. From Figure 5B, we can see that RNG has the lowest overall emission impact in all cases except methane when compared by kilograms of CO₂ equivalent. This is due to the assumed methane leak rate and unrecoverable methane during the upgrading and compression process. Despite these emissions, RNG still offers the lowest overall emissions compared to CHP and flaring when compared to the total local emissions released. Most emissions for RNG are moved downstream to other processes after compression and pipeline injection, which are out of the scope of this study. Despite this, RNG downstream emissions are considered lower than those of the comparison case when RNG is utilized as vehicle fuel [2,55].

3.4. Comparison to CHP

A previous report was created for the same facility by this lab that focused on hourly CHP analysis [49]. The report sized a CHP system at 884 kW of electrical power, and its values can be found in [49]. Elio et al. found its value at the end of its 20-year operating lifetime as USD 1,283,149 [49].

Table 6. Comparison between RNG and a previous report for the same facility’s CHP costs and NPV over their estimated lifecycle.

Method	RNG with RIN Credits (15-Year Lifespan)	RNG Without RIN Credits	CHP (20-Year Lifespan) [50]
Capital cost (USD)	4,963,351	4,963,351	5,229,856
Annual net revenue (USD) 1	2,049,754	171,320	310,143
NPV at end of lifetime (USD)	16,312,396	(3,185,106)	1,283,149
PBP (years)	3	29	17

¹ Annual revenue for CHP accounts for the discount rate over the course of its lifetime.

shows a comparison between CHP and RNG models at the end of their lifetimes. Comparing the models, RNG has a higher capital cost while having a significantly higher NPV at the end of its 15-year lifespan. While the numbers seem to be in favor of RNG compared to CHP, the analysis for RNG relies on renewable fuel credits to be profitable. If the RFS market is modified or removed, RNG may have a negative NPV and may not be a worthwhile investment.

3.5. Model Validation

The model used in this study was validated against a different UC Davis and EPA case study published in 2016 [4]. Their analysis was conducted on a yearly basis, and the results are shown in

Table 7. Model comparison and validation between the current 2021 EPA study and an older EPA 2016 study with a similar average biogas flowrate from the digester. Costs are adjusted to 2021-equivalent USD values utilizing the consumer price index.

Model	EPA 2021 [8]	EPA 2016 [4]	Percent Difference
Yearly average flow (m3h−1)	437	451	−3.1%
Methane content	62%	60%	3.3%
Capital cost (USD)	4,963,351	6,184,330	−20%
Yearly cost (USD)	407,039	690,180	−41%

below along with a comparison to the model used in this report, both with the same project lifetime. The 2016 model is significantly older than the more recent model used in this paper, which uses updated values from 2021. The older model was created from experimentally fitted data that outdates the current model, but both models are based on real data and have been updated to 2021-equivalent USD values utilizing the consumer price index [52].

Table 7 demonstrates that the EPA 2021 model utilized in this study and the older 2016 model’s inputs for both natural gas flowrate and methane concentration are similar. Both models have similar capital costs, but the 2021 model’s are slightly below the 2016’s predicted capital cost. The operating cost in the newer model is significantly lower than that in the previous report. This could be due to multiple reasons, including less expensive upgrading technology over time, resulting in lower costs, and differences in bounds of fitted data for the flowrate. The flowrate for this plant is lower than most flowrates studied in previous reports, which can lead to some inaccuracies in the fitted equation. These results indicate that the current 2021 EPA model, which primarily focuses on RNG derived from landfill gas, is appropriate to be utilized for RNG derived from wastewater. Due to the high NPV of the model at year 15, the difference does not severely impact this study’s finding.

4. Discussion

This study highlights the viability of renewable natural gas (RNG) systems for small-scale WWTPs in the American Southwest with access to renewable fuel credits. The system has a positive NPV of USD 16,312,396 and a rapid payback period of three years. This demonstrates that RNG systems are a profitable investment for smaller-scale WWTPs, encouraging the expansion of the technology and the utilization of previously flared biogas. If RINs or other renewable fuel credits are not included, RNG is not a good investment shown by the negative NPV. This underscores the critical role of incentives in making RNG systems economically attractive as a potential future fuel source, especially in smaller-scale applications where economies of scale are limited. Parker et al. found similar results for biogas systems in California utilizing both local credits unique to California, in which RIN credits were critical to system profitability [65]. Their systems utilized other waste streams which could be applicable to more valuable RIN credits [65]. This can cause potential future instability in the market, as any government incentive could have changing regulations that could impact RIN credit price. The short payback period (3 years) does help reduce the long-term risk of instability in the RIN price.

A notable finding of this study is the substantial reduction in site emissions achieved with RNG systems compared to combined heat and power and flaring technologies. The 22% reduction in emissions relative to CHP demonstrates RNG’s potential to contribute to local air quality improvements and broader greenhouse gas reduction goals. RNG’s capability to help reduce emissions is echoed in a similar study that focused on the total emissions reduced across the entire U.S. [66]. They found that RNG can play a critical role in reducing emissions caused by WWTPs [66]. Other waste streams, such as manure and food waste, could be mixed with wastewater and result in greater emission reduction. These waste streams generally produce biogas with a higher concentration of natural gas, and notably lack contaminants like siloxanes, making it less expensive to upgrade. These waste streams can be integrated into existing anerobic digestors commonly used at wastewater treatment plants and can be digested alongside standard municipal waste. This advantage is further amplified by the displacement of conventional natural gas when RNG is injected into pipelines or used as transportation fuel. However, it is important to note that the analysis did not account for downstream emissions associated with RNG usage, which may vary depending on the end-use application. Future studies should incorporate a lifecycle emission analysis to provide a more comprehensive assessment of RNG’s environmental impact.

Despite its benefits, implementing RNG systems at small WWTPs presents several challenges. The high initial capital costs, particularly for biogas cleaning and upgrading infrastructure, remain a barrier to adoption. Additionally, the variability in biogas composition, as observed in this study, highlights the need for flexible and robust upgrading technologies capable of handling seasonal fluctuations in contaminants such as siloxanes and hydrogen sulfide. The inclusion of gas chromatography analysis in this study provided valuable insights into the biogas composition, but more extensive datasets from diverse WWTPs across the United States are needed to develop generalized cost and performance models.

This study provides a direct comparison between RNG and CHP systems both techno-economically and on an emission basis. While RNG offers higher economic returns and greater emission reductions, its observed reliance on market incentives like RIN credits introduces an element of potential financial risk. RNG is not profitable without renewable fuel credits, indicating that it is less economical for use than conventional natural gas. In contrast, CHP systems, though less profitable overall, may offer a more stable investment in regions where renewable fuel credits are unavailable or uncertain. CHP also had a positive NPV over its lifetime, indicating the technology is more economical than utilizing conventional natural gas. Additionally, CHP systems are the more mature technology and allow the biogas to be used locally by the plant for both heat and power. These findings suggest that the optimal biogas utilization strategy for WWTPs will depend on site-specific factors, including biogas flowrates, local regulations, and market conditions. Biogas composition for a plant in the American Southwest has been reported in this paper for the first time, which can be generalized to other plants to help determine capital and upgrading costs.

5. Conclusions

The analysis shows that RNG is profitable for small-scale WWTPs if RIN credits are used. The NPV is positive within 3 years, indicating a very fast payback compared to CHP. If RIN credits are not applicable, the NPV of RNG is negative and will not pay off for a system of this size within a 15-year lifetime. For this reason, it should be ensured that any potential RNG facility should be able to take advantage of any potential credits. In order to have positive cashflow in year 15, the target income combining both RIN and natural gas pricing should be around a minimum of 37 cents per meter cubed of natural gas for this facility. Capital costs can be significantly reduced if a smaller plant size is chosen, but estimations should be made with future biogas flow expectations in mind. The NPV was compared to the CHP system size in the previous report. At 15 years, RNG has a total NPV of USD 16,312,396 compared to the NPV of CHP at USD 1,283,149. This indicates that RNG is a better overall investment if RIN credits can be applied. If RIN credits are not applicable, CHP has a higher overall lifetime NPV and is a safer investment.

The results of the sensitivity studies show that RIN value has the largest impact on the NPV in this model. As part of the analysis, only 1% of daily RIN prices in the previous 6 years fell below the minimum required value to operate the plant profitably, indicating a high level of safety in the investment. Other factors varied include the discounted rate of return. A higher value can significantly reduce the NPV of the system at 15 years, but the RNG system will still have a relatively quick 3-year payback. Natural gas prices played a much smaller role in the overall payback of the system. This is due to the high value of RIN credits in comparison, which reduces the effect the natural gas price has on the overall system.

RNG also offers the most localized emission reductions across all technologies, including CHP, analyzed in this study for smaller-scaled WWTPs. Most sites flare biogas as it is a safe and cheap way to dispose of it. Flaring and CHP both increase on-site emissions and create harmful pollutants that are released into the air. Due to the nature of the cleaning process, RNG creates very little toxic pollutants compared to its competing technologies. In total, RNG offers an environmentally friendly and economically viable alternative for biogas for a small-scale WWTP.