To conform or to transform? A comparative case analysis of the societal embedding of biogas systems

Abstract The diffusion of new socio-technical systems is essential for tackling contemporary sustainability challenges. Against the backdrop of literature on societal embedding this paper explores the diffusion of socio-technical systems as a process of co-constructing innovations and their societal environments. This paper uses a comparative research design to analyze the diffusion of biogas systems across four Brazilian states. In doing so, this paper makes two contributions. First, it contributes with nuances regarding the fit-and-conform and stretch-and-transform typology, showing that innovations exhibit not only hybrid patterns across societal environments but also across different sectors (e.g. agriculture, sanitation, and waste management). Furthermore, innovations exhibit hybrid conform and transform patterns across different administrative levels (e.g. municipal, state, and national). Second, it broadens the empirical base of societal embedding studies to the Global South and biogas technologies which represent a fragmented context and a complex innovation, respectively. Altogether, the paper contributes to further understanding of why multi-functional socio-technical systems, such as biogas systems, diffuse in certain contexts and not in others.


Introduction
Due to extensive agricultural activities and a sizable population, Brazil has one of the world's largest potentials for biogas production from organic residues.This potential remains largely unexploited [1,2].The introduction of biogas systems can lead to an overall reduction in environmental impact [3] and can simultaneously play several roles in sustainability transitions -for example, replacing fossil fuels, enabling the recirculation of nutrients, treating organic waste, and reducing greenhouse gas emissions.Although biogas production has accelerated in Brazil in the past few years, it has been limited to a handful of states.This variation in biogas diffusion between states has been previously attributed to technical aspects like the availability of substrates and sectoral contexts [4].Even though these aspects are important to enable the diffusion of biogas systems, they alone cannot explain the contrast in biogas systems' diffusion across Brazilian states.This is because the diffusion of socio-technical systems happens across multiple societal environments, such as regulatory, user, business, cultural, and trans-local [5,6].
Several frameworks investigate diffusion from different perspectives, conceptualizing endogenous and exogenous factors that can affect the diffusion process.Rogers (7) describes diffusion as a process of technology adoption among members of a social system.Diffusion studies tend to favor innovation itself and rarely address its consequences.The context is treated as static and homogenous, and adopters are put in the center of the analysis without much attention to other actors [8], thus ignoring that user populations, contexts, and the technology itself are dynamic and socially constructed entities [9,10].Although useful to understand the diffusion of single products (e.g.computers, fax machines, birth control pills), diffusion of innovation models struggle to incorporate the complexity of interactions between context and innovations that have the potential to transform entire socio-technical systems.Diffusion as a process of scaling up (in number, speed, and size) and scaling out (geographical expansion) comprises several important elements that influence the wider social adoption of technologies [11].However, scaling up and scaling out processes assume that what has worked in one context can be easily replicated in another without much attention to complex realities [12].
Diffusion can be understood as a process of societal embedding conceived as the co-construction of entire new configurations of socio-technical systems [5].Societal embedding is a reciprocal process that requires mutual shaping and adjustment of the technology and society [13].The societal embedding framework outlined by [5] suggests that alignments and struggles between technology and society happen across several societal environments through regulation, users' routines, market structures, cultural values, and interactions between spatially distributed actors and institutions.According to [6], the diffusion of socio-technical systems can develop either to fit and conform to unchanged societal environments or to stretch and transform societal environments to adopt features of the new socio-technical system.Furthermore, conform and transform patterns can appear in hybrid form in which interactions can follow a conform pattern in one societal environment but transform another [14].
Biogas systems are multifunctional systems spanning several societal domains and emerging at the interface of different sectors, such as energy, transportation, waste management, and agriculture [15].Different combinations of organic material for the anaerobic digestion (AD) -for example, manure, food waste, sewage sludge, stillage from ethanol production -and final use of the biogas produced -for instance, power, heat, vehicle fuel -give rise to different biogas system configurations.To implement such systems, a multitude of actors, technologies, infrastructure, and institutions need to come together [16,17].Hence, unique configurations of biogas systems can arise in diverse contexts with potentially different types of sociotechnical interactions.
The paper explores the ongoing diffusion of biogas systems across four Brazilian states as a process of societal embedding with the aim to understand how local contexts can shape the diffusion and configuration of these systems.By understanding the particularities of each state, it is possible to provide guidelines to system developers, technology providers, and policymakers on how to enable the diffusion of biogas systems locally.Specifically, this paper addresses the following research question: How does the alignment between technology and societal environments shape the diffusion of biogas systems?
The paper makes two contributions to the literature on the diffusion of socio-technical systems.First, the paper contributes with nuances to the literature on the societal embedding of socio-technical systems, particularly by highlighting different conform and transform patterns [6] in socio-technical interactions across different sectors and administrative levels.Second, by expanding the scope of empirical studies of diffusion to the Global South, the paper provides evidence of how socio-technical systems diffuse in contexts where informal institutions such as norms, and culture have a central role and shape formal institutions or prevail on the absence.This contribution is against the backdrop that sustainability transition studies are dominated by empirical cases from the Global North (c.f.18,19].
Following this introductory section, the theoretical framework elaborates on societal embedding literature.Next, the rationale for case selection and methods for data collection and analysis are presented followed by the description of the cases.The comparative analysis section explores the dynamics of societal embedding of biogas systems across four Brazilian states.Finally, the last sections discuss the main findings and conclusions.

Theoretical framework
In the early stages of diffusion, technologies are nurtured, empowered, and protected from selection environments in technological niches [6].Selection environments are dynamic domains that restrict the development of niche technologies and influence the diffusion process but also are partially shaped by it [9,13].Hence, the new socio-technical system emerges through interactions between technology and society [13], where characteristics of this new socio-technical system are not known in advance but are co-constructed throughout the diffusion process [5].Because selection environments are multi-dimensional, socio-technical interactions and embeddedness are multidimensional.
Socio-technical interactions are contested processes full of choices that can lead to different trajectories of societal embedding [20], which in turn results in regional differences in socio-technical systems [21].External pressure from the selection environment may steer emerging socio-technical systems to fit and conform to pre-existing societal environments according to the terms of established institutions [6].For sustainable technologies, adapting to the norms and values of pre-existing institutional structures may jeopardize the broader sustainability aspects of the technology.On the other hand, emerging socio-technical systems can stretch and transform selection environments, thus challenging and re-configuring pre-existing institutional structures to adopt some features of the niche [6].Both processes of conforming and transforming rely not only on niche actors but also on selection environments and the broader society.
Building upon previous contributions [5,9,14,22,23], this paper distinguished between five key societal environments for the societal embedding of biogas systems.Firstly, the regulatory environment refers to policies that can influence production, consumption, market development, and the use of new technologies.The regulatory environment can be quite dynamic, involving policy changes and discontinuities.The role of regulatory agencies evolves with the diffusion of the technology, from supporting learning mechanisms and the articulation of demand in the early stages to applying redeployment mechanisms that allow the new technology to grow beyond niche markets [24].Because biogas systems cut across several sectors, the biogas policy landscape is complex and influences a variety of sectors in different ways [25].
Secondly, the user environment refers to the integration of the new technology into users' practices, development of preferences, and adaptation to new routines.User-technology linkages are important mechanisms for creating linkages between technology and society, but they can be hard to undo, creating barriers to the introduction of innovations [13].In this paper, the user environment refers to the adopters of biogas technologies, i.e. the users are the producers of biogas.
Thirdly, the business environment refers to the development of new industries and markets, business models, and supply chains.As biogas systems span several sectors, they need to be integrated into different markets and business practices, which may present different structures and challenges.Systems actors need to adjust to the variety of system arrangements and their requirements.As the system expands, bottlenecks and tensions may arise, requiring actors to constantly respond to these challenges [17].
Fourthly, the cultural environment refers to the acceptance of emerging socio-technical systems by the wider society.Articulation of discourses, narrative, storytelling and even negative discourses against existing technologies are activities performed by system actors to create support and cultural legitimation for the new socio-technical system [5,9].Narratives are empowering political devices designed to reshape social norms, enable institutional reforms [6], and shape markets [17].
Finally, analogous to [5], trans-local environments refer to external environments that influence the diffusion of socio-technical systems locally.The trans-local environment integrates interactions between spatially distributed actors and institutions that cut across various spatial levels from local to global [26].Trans-local relations help shape embeddedness processes locally, but they are also influenced by locally embedded processes [27].
Societal embedding literature provides the relevant societal environments for socio-technical interactions where the new socio-technical system can either conform to preexisting social environments or transform these to benefit the new socio-technical system.However, the societal embedding framework builds on cases from the Global North therefore focusing on successful cases of diffusion and particular institutional settings.

Methods
This paper presents comparative case studies of the diffusion of biogas in systems in four Brazilian states.The studies relied on qualitative data from semi-structured interviews, documents, and quantitative data from statistics on the potential, production, and use of biogas.

Case selection
Brazil has a large potential for biogas production across several sectors but especially from residues from the sugarcane industry and agribusiness [28].Substrate availability is unevenly distributed across the nation and concentrated in a few states [28] varying according to the number of inhabitants and local industries.In the past few years, biogas production has accelerated.Nevertheless, around three-quarters of the biogas produced in Brazil came from only a handful of states in 2020 [29].To analyze the diffusion of biogas systems, four states were selected as case studies based on the potential for biogas production (type and availability of substrate), contrasting diffusion patterns -in terms of scale of production, substrate used for biogas production, and use of the biogas produced -, realized potential, and diversity in terms of socio-economic contexts.Table 1 provides an overview of the selected states and motives for each selection.
As one of the largest ethanol producers in the world, the Brazilian sugarcane sector could produce up to 21 billion Nm 3 of biogas annually [28], and more than half of this potential is in São Paulo state.Regardless of this large potential in the sugarcane sector, biogas production in São Paulo comes mainly from landfill gas capture.In comparison with the other Brazilian states, Rio de Janeiro's potential for biogas production is insignificant (around 1% of Brazil's potential for biogas production); nonetheless, the state is the second-largest biogas producer in Brazil.Both São Paulo and Rio de Janeiro states produce biogas mostly from landfill gas capture to generate electricity despite their considerably different potentials for biogas production.In contrast, the third-largest biogas producer, Paran� a state, produces biogas mostly from food industry residues for heat generation.Paran� a's potential for biogas production also lies across the food value chain.Analogous to Paran� a and São Paulo, Goi� as has a strong agribusiness sector, besides being the second-largest ethanol producer in Brazil.However, unlike its peers, the state has seen little diffusion of biogas systems even so Goi� as has one of the largest potentials for biogas production among Brazilian states.

Data collection and analysis
In total, nineteen semi-structured interviews were carried out with twenty-two professionals from several sectors, for example, biogas, energy, transport, sanitation, gas, and agriculture (Appendix A).Among the interviewees, there were representatives from private and state-owned companies, governmental agencies, associations, research institutes, academia, and intergovernmental organizations.The participants were selected according to the organization's involvement in the diffusion of biogas systems and area of activity.Interviews with civil society representatives such as professional associations and research institutes were included in the study.These civil society representatives were selected based on their long-term involvement and experience in the biogas sector across the cases.Participants with an overall understanding of their area of activity were prioritized rather than users whose perspective could be limited to their experience with biogas.All participants had experience with biogas systems, most across two or more of the cases.The interviews were structured following the five environments of societal embedding and thematically analyzed accordingly.The interviews lasted, on average, 45 min and were conducted in two rounds.The first between April and July 2020 and the second to update the findings in September 2021.The indepth interviews helped to understand the reasoning behind actors' actions and interactions as well as how contextual factors influence these.Although interviews with people's representatives could enhance the political dimension of the study these were not included in the study because these representatives appear across several governmental organizations, and they may change every election cycle.Instead, the study relied on documents from governmental agencies and official gazettes to better understand the political positions of these representatives regarding biogas systems.
Documents from governmental agencies and research centers, policies, official gazettes of the states, and news articles, complemented the qualitative data and played a supporting role in strengthening the findings from the interviews.Furthermore, reports from local news outlets helped to establish a timeline, identify the local actors involved in the development of biogas systems in each state, and construct a narrative.
Quantitative data on the estimated potential for biogas production was retrieved from Abiog� as 1 .Their estimations present the maximum theoretical potential for biogas production according to the source of substrate per state.The data set on the production and use of biogas from 2003 to 2020 was retrieved from the Biogas Map presented by CIBiog� as 2 , a science and technology institution dedicated to developing the biogas value chain.
The data was analyzed using societal embedding to differentiate between five relevant societal environments for socio-technical interactions.By referring to actors' actions and statements, the analysis focuses on how the alignment of biogas systems and societal environments either aims to fit and conform or stretch and transform to pre-existing societal environments.Whereas conformative aims portray biogas systems as competitive and conventional within prevailing societal environments, transformative aims antagonize existing societal environments and argue for institutional change.

Case description
This section presents the case descriptions.For each case, an overview of the states and a summary of biogas production and use are first presented, followed by the main developments in important sectors for biogas production, and relevant state policies for the diffusion of biogas systems.Appendix B presents a summary of main developments in each sector per state.

São Paulo
São Paulo state has the largest economy among Brazilian states, representing approximately 30% of Brazil's gross domestic product (GDP) in 2018 [30].The state has a diversified industry, which includes knowledge-intensive sectors such as pharmaceutical and aerospace.Out of the 55 biogas plants in operation in 2020, 42 used the biogas produced to generate electricity (Figure 1).Small-scale biogas plants generate electricity for internal processes or are connected to the electricity grid through compensation schemes.In contrast, large-scale biogas plants sell electricity in national electricity auctions.Furthermore, the sugarcane sector also stands out in the state economy as the state is the largest ethanol producer in Brazil.São Paulo is also the largest biogas producer among Brazilian states and holds the largest potential for biogas production.
While São Paulo's potential lies mostly in the sugarcane sector, approximately 80% of the biogas produced in São Paulo comes from sanitation and waste management sectors, mostly from landfill gas capture (Figure 2).

Sanitation
In the 1980s, Sabesp 3 experimented with anaerobic digestion in sewage treatment stations implementing two test facilities, where one of them upgraded biogas to fuel the company's own vehicle fleet.These projects faced opposition as they were seen by some as a deviation from the Sabesp core business.Biogas and biomethane production were later discontinued following price drops in fossil fuel prices.In the early 2000s, the interest in anaerobic digestion in the sanitation sector was restored by research projects.The ENERG-BIOG Project installed and tested power generators in biogas plants in a sewage treatment station, and a biogas power plant was implemented in the sewage treatment facility at USP 4 as part of the Program of Rational Energy Use and Alternative Sources (PUREFA).These projects push the development of biogas systems in the sanitation sectors and influence the development of policies at state and federal levels.In 2011, the first largescale biogas plant in a sewage treatment station opened in São Paulo.In the following year, Sabesp resumed biomethane production in the Franca station.The sewage treatment station was later chosen as a demonstration plant to upgrade biogas for transportation.

Urban solid waste
In 2004, the São Paulo municipality approved the capture of landfill gas in the Bandeirantes landfill to reduce greenhouse gas emissions and generate carbon credits for the municipality.Initially, the biogas captured was flared, and only the carbon credits generated revenue; later, the biogas power plant was installed onsite.Similar projects were developed in large landfills around Sao Paulo city.These projects are coordinated by private companies through concession contracts, but the municipality owns half of the carbon credit generated.

Sugarcane energy industry
São Paulo state is the largest sugarcane producer in Brazil; accordingly, the state has a large potential for biogas production in this sector.However, biogas production from sugarcane energy industry (residues of ethanol and sugar) has posed great technical and economic challenges.The topic has been thoroughly studied through projects that gathered private actors and actors from academia and government.In 2016, Ra� ızen 5 was the first ethanol producer to win an ANEEL 6 auction to sell electricity from biogas.Another project in the sugarcane sector aims to produce biogas from ethanol production residues for transportation.The Cocal 7 Narandiba ethanol plant will produce biomethane to supply a local gas grid that serves two municipalities in the Southwest region of São Paulo.G� as Brasiliano 8 is responsible for piped gas distribution in the western region of São Paulo state, where most sugarcane energy industries are located.The company has a strategic interest in introducing biomethane to their portfolio as it reduces dependence on fossil gas and diversifies the sources of gas.Both the sugarcane energy industry and the state gas distributor are investing in this project in addition to receiving financial support from transnational organizations for infrastructure development.

Policy
The State Energy Plan, developed by the CEPE 9 in 2010, acknowledges the potential São Paulo has for biogas production and suggested measures to realize this potential including policy intervention.Therefore, ARSESP 10 proposed biogas-specific state legislation to promote the development of biogas systems regionally.The proposed policy received contributions from several actors in the state, including industry associations, gas distributors, and gas users.In 2012, São Paulo was the first state in Brazil to pass a biogas-specific policy.The policy aims at increasing the share of renewable energy and reducing GHG  emissions besides including a minimum percentage of biomethane in the gas network.However, the mandatory minimum percentage was never implemented by the regulatory agency as it faced strong opposition from the industrial association.Shortly after, ABiog� as was founded in São Paulo to provide institutional support to actors across the biogas value chain.Although the association represents actors from around the country, most of them are in the South and Southeast regions.

Rio de Janeiro
Rio de Janeiro has the second-largest economy in Brazil, representing approximately 10% of Brazil's GDP in 2018 [30].The state's economy is diversified, yet most of its wealth comes from oil and gas exploration and tourism.Rio de Janeiro state is connected to all major gas pipelines in Brazil, and more than 50% of all gas-powered vehicles in the country are in this state.Rio de Janeiro state is the second-largest biogas producer and reached the highest realized potential among Brazilian states.Between 2017 and 2019, biogas production increased fivefold (Figure 3).In 2021, there were ten biogas plants in operation across the state, and nine of these generate electricity with the biogas produced.Biogas in Rio de Janeiro state comes mainly from landfill gas capture (Figure 4).

Sanitation
Attempting to reduce pollution in Guanabara Bay in Rio de Janeiro, the state joined efforts with the federal government and municipalities around the bay to improve waste and sewage treatment services.In the early 1990s, the Guanabara Bay Depollution Program (PDBG) was approved and shortly after received financial support from global organizations.The PDGB was structured around five areas of intervention: sanitation, solid waste management, rainwater management, complementary environmental projects, and digital mapping.The former received the largest share of investments for the expansion of the sewage network and the construction of new sewage treatment stations.Among the new sewage treatment station, Alegria operated by CEDAE 11 is the most successful case.Since 2005, the sewage treatment station has used anaerobic digestion technologies to produce biogas that generates electricity for the company.

Urban solid waste
Although the PDBG also included solid waste management, less than 3% of investments were direct to this area [31].The state, as the main financier of the program, was responsible for the construction of new waste treatment facilities, which included new landfills, recycling stations, and transfer stations, while the municipalities were responsible for operating these facilities.However, the municipalities lacked the resources and competencies to keep these facilities open.When the PDBG ended, the program had left behind a large state debt, unfinished projects, and only a small fraction of facilities built were operating.The lack of collaboration between and within administrative levels and frequent changes in leadership within governmental agencies harmed the execution of the program [31].
Gramacho landfill, once the largest landfill in Latin America, was the first landfill in Rio de Janeiro state to capture biogas.In 2008, the Rio de Janeiro municipality approved the project after five attempts to stop the public hearings.Until 2010, the biogas captured was flared and carbon credits sold internationally.Later, an upgrading facility was built on-site, and a 20-year concession contract was signed with Petrobras 12 .As part of the concession contract, a share of the revenue from biomethane sales goes to a social fund for former waste pickers.The fund aims to help these workers to enter the formal labor market.
In the 1980s, Rio de Janeiro municipality first experimented with biomethane for transportation.Comlurb 13  had 150 gas-powered vehicles besides fuel stations that served the public.This first project only lasted five years; after this period, natural gas supplied Comlurb's vehicle fleet.In 2018, Comlurb opened an anaerobic digestion plant to treat the organic fraction of urban solid waste.The Figure 3. Use of the biogas produced in Rio de Janeiro state (based on [29]).organic waste is separated in the transfer station by former waste pickers.The biogas is upgraded to biomethane and then used as a vehicle fuel and to generate electricity for the company.Comlurb's goal is to extend the life of landfills by reducing the amount of organic waste that goes to landfills, thus reducing environmental impact.In 2020, around 3% of the total volume collected by Comlurb went to AD plants.

Policy
Rio de Janeiro was the second state to implement a biogas-specific state policy in 2012 that aimed to promote production of biogas in the waste management sector.This policy determined that gas distributors had to acquire all biomethane produced in the state, up to 10% of the commercialized natural gas volume.However, at the time, there was no national standard defining biomethane.As a result, the state regulatory agency did not implement and enforce the policy.In 2019, the state government approved a second state policy on biogas with a wider scope.The policy aims to support and encourage the production of biogas and secondary products as an instrument for promoting regional socio-economic development and the circular economy.

Paran� a
Paran� a has the fifth-largest economy among Brazilian states representing around 6% of Brazil's GDP in 2018 [30].Paran� a has a diversified and industrialized economy with an emphasis on the food and beverage sector.The agribusiness sector in Paran� a has one of the strongest participation rates across Brazilian states, reaching 33% of Paran� a's GDP in 2018.
In the last decade, biogas production has increased rapidly in Paran� a (Figure 5).In 2020, there were 143 biogas plants across the state producing more than 200 million Nm 3 of biogas annually.Manure from livestock activities and waste from food and beverage industries account for the large majority of substrate used for biogas production (Figure 6).Across the food value chain, biogas is used for heat generation but also to generate electricity.While heat is usually used to supply internal processes, biogas power plants are normally connected to the electricity grid through compensation schemes.

Sanitation
Sanepar 14 provides sewage services to 74% of Paran� a's inhabitants through contracts with the municipalities.The sewage collected by the company is treated across 225 sewage treatment plants, of which 215 use biogas technologies.Sanepar first experimented with biogas technologies in the 1980s; at the time, the biogas produced at the Londrina sewage treatment plant was upgraded and used to fuel the company's own vehicle fleet.However, a few years later, the project was discontinued because it was no longer economically feasible after the drop in fossil fuel prices.Biogas technologies were reintroduced by the company in the late 2000s when state actors developed a series of demonstration plants across different sectorssanitation, solid waste, agriculture -driven by environmental concerns.The expansion of biogas technologies in the sanitation sector in Paran� a was also supported by interactions with national governmental agencies, which promoted knowledge exchange within Brazil and with German actors.
The application of the biogas varies according to the volume produced in each sewage treatment station.In small to medium plants, biogas is used to supply heat for internal processes.In large plants, biogas is also used to generate electricity for internal processes.Whereas the digestate is mostly landfilled, the rising costs of landfilling encouraged the company to search for other solutions.Together with local farmers, Sanepar is redirecting part of the digestate to be used as a fertilizer in specific types of crops.

Agriculture and food industry
During routine quality tests, the Itaipu 15 identified eutrophication of the Paran� a Basin waters caused by the extensive agricultural activities in the surrounding areas, mostly poultry, cattle, and swine farming [32].The Ajuricaba micro-basin, located in the municipality of Marechal Cândido Rondon, presented the highest levels of water pollution in the region because of the large concentration of animals in a small area.Therefore, the region was considered ideal for the development of a bioenergy cooperative.The development and implementation of the Ajuricaba cooperative involved 33 small farmers from around the micro-basin, the municipality, and state companies under the coordination of Itaipu.Initially, the project generated extra income for biogas producers as the electricity was sold to the state electric utility company.However, after the distributed generation program was implemented nationally, the farmers only received compensation through their electricity bills, which decreased their interest in the project.Although the project had some setbacks, it served as a learning experience in the state.Later projects were able to overcome the income problem by selling the biogas produced to the local municipalities, which in turn generated electricity for public buildings.
The agribusiness sector in Paran� a is one of the strongest across Brazilian states, reaching 33% of Paran� a's GDP in 2018.Paran� a's agribusiness is formed by several specialized actors that are interconnected throughout the food value chain.Several food industries in Paran� a are managed by agricultural cooperatives that have a strong local presence and close connection to the negative impacts of the agribusiness sector.Therefore, biogas is seen as an opportunity for mitigating environmental impact while improving profitability.In contrast to farms, food industries use the biogas produced mainly to supply heat for internal processes.The expansion of biogas systems across Paran� a goes together  with the development of a local technical support structure headed by actors on the west side of the state.Paran� a's agribusiness sector presents a large variety of technological and productive arrangements.Thus, the availability of technical support close to substrate owners made the expansion of biogas systems possible in the state.

Policy
Paran� a's Biogas and Biomethane State Policy regulates the limits of actions of each actor across the biogas value chain and excludes biogas infrastructure from the domains of the state piped gas monopoly.In other words, the state policy allows biogas developers to build infrastructure independent from the state gas company.Furthermore, the state policy authorizes the state regulatory agency to define a minimum percentage of biomethane in the gas grid.However, industrial users fearing the increase in gas prices have blocked the implementation of this minimum percentage.

Goi� as
Goi� as contributed a little less than 3% of Brazil's GDP in 2018 [30].Analogous to Paran� a and São Paulo, Goi� as also has a strong agribusiness sector, besides being the secondlargest ethanol producer in Brazil.Like most of Brazil's gas infrastructure along the coast, Goi� as does not have access to the piped gas network.Goi� as has one of the largest potentials for biogas production among Brazilian states.However, the development of biogas systems has been slow -in terms of volume of production -in comparison to other states with similar potential for biogas production.Yet, it has steadily increased in the past decade (Figure 7).In 2020, there were 44 biogas plants.Out of these, 41 were farm-based biogas plants using manure and other residues from agricultural activities as a substrate for biogas production (Figure 8).Except for one industrial biogas plant, biogas was used to generate electricity.

Sanitation
In the South of Goi� as, Rio Quente is a popular holiday destination known for its thermal waters and adventure tourism.The municipality annually receives 300 times more tourists than the number of inhabitants in the city.Thus, a large hotel group in the region developed a private sewage treatment facility where the hotels' sewage is treated using anaerobic digestion, and the biogas produced generates electricity for the hotels.

Agriculture and food industry
In 2003, the first biogas plant in Goi� as was installed in the Rio Verde municipality, in the South of the state.Of the 41 biogas plants operating in 2021 in Goi� as, 22 are in the Rio Verde municipality; hence, the region has the highest concentration of biogas plants in the state.Besides the location, these biogas plants have other aspects in common: all are small-scale farm-based facilities that use the biogas produced to generate electricity.Although these biogas plants are in the same sector, these plants were developed in isolation by (mostly) private actors and without a common denominator.
In 2013, Brazil signed a cooperation agreement with Germany that aimed to contribute to expanding efficient energy use of biogas, reducing greenhouse gas emissions, and strengthening relations between public, private, and academic actors in both countries.Between 2013 and 2017, four projects were selected involving universities from Goi� as, São Paulo, Cear� a, and Rio de Janeiro, and German research groups.Within the framework of this project, a demonstration biogas plant in a food industry in Goiânia (Goi� as capital city) was designed and implemented.The project included a comprehensive feasibility study, explored several technological arrangements, and helped develop the biogas value chain in the state.
The first biogas plants in the dairy industry in Brazil started operations in 2015.The plant is part of climate change mitigation efforts by a national dairy industry.The wastewater from milk and dairy production is treated using anaerobic digestion, and the biogas produced is used to generate heat to supply the dairy industry's internal demand.

Policy
In 2020, the Goi� as state government implemented the State Biogas and Biomethane Policy.This policy targets the development of biogas value chains in Goi� as as a means to reduce environmental impact in agriculture, promote the correct treatment of organic waste, and create incentives for the use of biogas.Goi� as biogas policy differentiates between agricultural residues and other substrates, giving preferential treatment to the former.The main policy instruments include public procurement of biogas and biomethane, authorizing special loans and grants, and promoting the creation of cooperatives and associations across the biogas value chain.

Comparative analysis
The comparative analysis shows that the societal embedding of biogas systems was influenced by several societal environments, including regulatory, user, business, cultural and trans-local environments, which affected the configuration of biogas systems.

Regulatory environment
All four states have implemented biogas-specific policies, but they vary in terms of the targeted sector, goals, and overall impact on the biogas value chain.Besides, these policies were implemented in different years against the background of different policies at the national level, which affected the overall impact of state policies.Coherence issues appear both across administrative areas -for example, energy, water, gas, and waste -and between policies at the national level and state level, which shaped the regulatory precondition for the development of biogas systems across the Brazilian states [33].This spatial diversity in the regulatory environment is also seen in other countries with federalist structures, such as the USA [5].
Biogas actors in São Paulo and Rio de Janeiro targeted higher administrate levels, while in Paran� a, biogas actors promoted changes to the regulatory environment at the state level.In 2012, Rio de Janeiro implemented the state's first biogas-specific policy, which focuses on biogas production from urban solid waste and in the use of landfill gas.Rio de Janeiro's policy conforms to national policies that aimed to end dumpsites and promote landfills, which in turn allowed the state to capture national resources to invest in the construction of landfills and landfill gas capture.However, the lack of standardization of biogas and biomethane at the national level at the time undermined the state's attempt to promote the use of landfill gas, especially as a substitute for natural gas.Only five years later, after several interactions between national and state actors, mainly from São Paulo and Rio de Janeiro, the national regulatory agency accepted the injection of biomethane in the gas grid and defined quality standards for biomethane from landfills.
Because of the growing development of biogas systems in Paran� a, in 2018, the state government implemented a series of policies that defined the area of action, rights, and duties of actors across the biogas value chain and excluded biogas from the domains of the state monopoly on piped gas.Biogas production and use had grown and gained momentum in the state; thus, biogas systems gained the support of state actors and were able to transform the regulatory environment in Paran� a and eventually in other states as well.Other Brazilian states, among them Goi� as and Rio de Janeiro, followed the footsteps of Paran� a and implemented similar state policies.

User environment
The user environment plays an important role in the embedding of biogas systems since biogas technologies require constant monitoring and management [34].Across the cases, there were five main groups of users: livestock farmers, food and beverage industries, sugarcane energy industry, sanitation, and waste management sectors.Biogas systems fit or transform the routines and practices of these user groups in different ways.
For actors across the food value chain, both farmers and food industries, the adoption of biogas technologies transform their practices and routines.However, food industries have a larger capacity to capture financial and human resources.Because biogas technologies are configurational, the development and implementation of each biogas plant is unique.Therefore, farm-based biogas plants need easy access to specialized technical support.In Paran� a, the local presence of state actors that offered technical support and coordinated the development of biogas systems helped to reduce uncertainties regarding the adoption of biogas technologies.Furthermore, the integration of biogas systems into cooperatives conformed with the business culture of Paran� a's agribusiness.Geographical and cultural proximity are important factors for knowledge creation, learning, and innovation [35,36].Hence, the integration of biogas systems into the structure of food value chain in Paran� a can help explain the faster uptake of biogas systems in Paran� a in comparison to São Paulo and Goi� as, even though these states have similar potentials in the agribusiness sector.
In contrast to farmers and the food industry, the sugarcane energy industry operates in the energy sector as a biofuel producer.Actors in this sector can integrate their capacities and redeploy resources to biogas production and biogas markets.Although it represents a technological change, the adoption of biogas technologies by sugarcane energy industries fits into their routines and processes.

Business environment
Paran� a's agribusiness encompasses a multitude of specialized actors that are interconnected across the food value chain.Paran� a's food industry presents a strong relationship with actors across the value chain, acting as an important integrator agent with the knowledge and resources to implement long-term biogas projects.Furthermore, these industries have a strong local presence and direct connection with the negative environmental impacts of food production.
Brazil has a large potential for biogas production from sugarcane industries' residues, yet because of technical and economic feasibility challenges, the adoption of biogas in this sector only happened recently.These industries are already biofuel producers, which means that they have the knowledge and qualified staff to implement large-scale biogas projects.Besides, the sugarcane sector is well-established in Brazil and holds political power both at the state level in São Paulo and the national level.Yet, the injection of biomethane to local gas grids requires new partnerships with local piped gas distributors.In Brazil, gas markets are regulated at state level and dominated by large state companies that hold the concession for local piped gas distribution.The injection of biomethane to local gas grids requires the transformation of gas distributors business models to allow decentralization of production.Although in the west of São Paulo, local actors were pushing for the transformation of gas markets, most grid operators across the four cases only support biomethane injection when conforming to established gas markets.
The Brazilian electricity market is regulated at the national level, and its structure provides privileges to large-scale producers.Before the implementation of compensation scheme, small-scale producers were able to sell the electricity generated from biogas to the electricity grid.However, the compensation scheme limited income possibilities for small-scale producers since these producers could only be compensated through discounts on the electricity bill.Yet many small biogas producers have the potential to generate more electricity than they need.In Paran� a, the agro-energy cooperatives were able to explore business models that were more attractive to small producers and create partnerships with state and local governmental actors.

Cultural environment
The development of biogas systems in Paran� a started in the agricultural sector as a solution to environmental problems caused by the extensive use of fertilizers and many animals in a small area.However, the expansion of biogas systems in the agribusiness sector challenges the identity of some actors as food producers.Farmers want to preserve their legal and cultural identity to maintain access to the services and financial support they already have access to as food producers.Although farmers do not lose financial support when producing biogas from food production residues, investments in biogas systems compete with investments on their core business, i.e. farmers have limited access to financial support to both acquire essential agricultural inputs and to improve waste treatment and energy efficiency.In contrast, the production of biogas in the sugarcane sector in São Paulo reinforces actors' identity as biofuels producers.Similarly, the agricultural sector in Goi� as is geared towards large-scale industrial farms and energy crops that are already coupled with the energy sector.
The level of engagement of state-level actors and the institutional stability of governmental agencies varied considerably between cases and appeared as a key factor for the diffusion of biogas systems.In Paran� a and Rio de Janeiro, the state government coordinated and financed biogas projects but with a focus on different biogas systems.However, the institutional instability of Rio de Janeiro state's politics negatively impacted the development of biogas projects, whereas, in Paran� a, biogas projects were able to create knowledge cycles within the state and ultimately grew into a regional development agenda.Furthermore, while Rio de Janeiro's state actors promoted end-of-pipe technologies in the waste management sector, other proactive biogas systems were seen as unfitted for the state context.
The implementation of landfill gas capture technologies in São Paulo and Rio de Janeiro has improved waste management practices and added another layer of complexity.Yet, for citizens in both states, the adoption of biogas systems has not changed daily practices since the initiative was not coupled with more proactive procedures such as the separation of organic waste and recycling.

Trans-local environment
The constellation of actors involved in regional activities and the interconnections between the various kinds of actors also presented considerable variation thus influencing the diffusion of biogas systems.In Goi� as, biogas projects are largely disconnected and developed by single actors, with a few exceptions, while biogas projects in São Paulo state had the support of well-established industrial sectors, state and national governmental agencies, and transnational actors.Both in São Paulo and Rio de Janeiro, the development of biogas systems set the agenda for advocacy groups at the national level as biogas associations and other advocacy groups are dominated by large firms in these states.Therefore, it is no surprise that the development of biogas systems in these states conforms to national advocacy groups' views on what are the desirable paths for the development of biogas systems.
Transnational companies in the food sector were early adopters of biogas technologies in Brazil.Partnerships between the Brazilian and German governments also influenced the adoption of biogas in the sanitation and waste management sector.Research groups and companies from Germany and the Netherlands were involved in early development in these sectors; nonetheless, these projects are carried out with local patterns.
In summary, biogas actors acted and reacted differently to the variety of challenges posed for the diffusion of biogas systems in each state, thus aiming to conform to or transform these environments.In the regulatory environment, biogas actors in different sectors and states responded differently to coherence issues between national and state-level policies.In the user environment, actors faced different challenges according to the user's capabilities and resources.In the business environment, electricity and gas markets influenced and directed biogas producers to different applications and system configurations.In the cultural environment, the level of engagement of actors, institutional stability, and cultural and legal identity of users' groups appeared as key factors in the diffusion of biogas systems.In the trans-local environment, there was a large variation in relation to the actors involved in the diffusion of biogas systems and their roles.

Discussion
The case analysis shows how patterns of alignment may differ across multiple sectors, as different user groups respond to challenges posed by pre-existing societal environments in diverse ways.The case analysis also demonstrates how patterns may change across different states because of variation on local assets such as infrastructure, formal and informal institutions, and industrial base.Moreover, it shows how patterns of alignment may vary between administrative levels because relevant biogas policies vary across states and national levels and thus influence the actions of biogas actors differently.In the analysis, this was particularly prominent in the regulatory environments.Because national and state levels of policy are clearly separated, and policy coherence issues can appear (cf.33] between these administrative levels, conformative or transformative aims towards the regulatory environment are more pronounced compared to other societal environments.In summary, the case analysis shows how alignment patterns between biogas systems and societal environments can vary across space (different states), scale (administrative level), and sectors (administrative areas).
The analysis draws attention to interactions that arise between societal environments.For example, the regulatory environment influenced actors' actions to conform with or transform other environments such as business environment and cultural environment.As national policies on waste management supported landfill gas capture and use, and electricity markets favor large-scale power plants, electricity generation has been the preferable choice for biogas producers in the waste management sectors.Furthermore, landfill gas capture avoids any disruption to waste management practices thus conforming to the cultural environment because sorting of waste is not a common practice in Brazil, and conforming to the user environment as waste management companies don't have to change their existing practices.These hybrid alignments (transform or conform patterns) between societal environments and biogas can explain why certain configuration of biogas systems diffused faster than others and in certain contexts.Specifically, when biogas systems configurations seek to conform to several pre-existing societal environments their diffusion is facilitated.On the other hand, when biogas systems configuration requires the transformation of several societal environments, their diffusion can be constrained.
The alignments also highlight important sustainability implications since strategies to fit and conform to unchanged societal environments may reduce the environmental benefits of biogas systems.Biogas could have a greater positive environmental impact in the transportation sectors in Brazil because this sector relies heavily on fossil fuels.However, biogas producers prefer to conform to favorable electricity markets than trying to transform the conditions for biogas use in the transportation sector.Essentially, these actors produce biogas and use the biogas to generate electricity.From an actors' perspective, this decision makes sense since electricity markets are easier to enter and offer better economic rewards.However, from environmental perspective, biogas could replace fossil fuels in the transportation sector.In São Paulo, the state government recognizes that biomethane presents positive externalities that are enjoyed by society, but it does not support the subsidies for either production or use, affirming that biomethane must be competitive in terms of natural gas markets.Similarly, the Rio de Janeiro municipality stated that the anaerobic digestion of organic household waste is not a suitable technology for developing countries because of high investment and high operational costs.At the same time, Rio de Janeiro municipality supports the implementation of landfill-gas capture technologies and the use of biogas and biomethane.This indicates that in both states, São Paulo and Rio de Janeiro, states governments believe that biogas systems should conform to pre-existing societal environments, thus fitting to price expectations and existing regulations.

Conclusion
Expanding the empirical scope of analyses to the Global South, this paper has contributed to the literature on the diffusion of socio-technical systems.Investigating how the alignment between new socio-technical systems and societal environments has shaped the diffusion of biogas across four Brazilian states, the comparative analysis has shown how different regional contexts have strong influences on which system configurations diffuse.
The societal embedding literature distinguishes five key societal environments for socio-technical interactions [5] which new socio-technical systems can either fit and conform or stretch and transform [6].Although useful to analyze the emergence and diffusion of biogas systems, these previous contributions provide partial understanding to explain limited diffusion in certain places, such as the Global South.The societal embedding framework builds on successful diffusion cases in the Global North.However, Global South contexts differ in terms of institutional stability.Societal environments in the Global South appear to be more fluid, and informal institutions appear to be more important than in cases in the Global North [19].
The Brazilian context thus presents some new analytical issues which have been given limited attention in previous societal-embedding research.Specifically, the instabilities in the regulatory environment may differ from typical cases studied in the Global North.Biogas systems are complex systems and thus need to be embedded in a stable institutional environment to allow learning, knowledge transfer, regulation, and funding.Indeed, previous studies [37] call for a destabilization of existing institutional configurations to open up windows of opportunities to scale new socio-technical systems.However, in contexts where institutions are unstable, actors from the new socio-technical system may struggle to scale up since there are no clear institutions to destabilize.Essentially, this discourages users from investing in new technology.Institutional stability is often lacking in the studied context of Brazil where institutions can be more fluid, fragmented and unstable [33,38].Moreover, it suggests that the fit-and-conform and stretch-andtransform patterns [6] are useful to highlight differences across societal environments at different scales, but it is necessary to recognize the instability of institutions in certain contexts.In the case of fluid and fragmented institutions, the fit and transform dichotomy becomes unclear and hard to operationalize because there are no clear institutions to conform to or transform.For innovations to diffuse, a certain degree of institutional stability is needed (cf.39] for niche actors to be able to scale up new socio-technical systems.In the absence of such stability, diffusion of such systems can be constrained by the lack of directionality, in-efficient use of public resources and fragmentation of niche actor efforts.

Figure 2 .
Figure 2. Source of substrate for biogas production (bars) and number of facilities (dotted lines) per sector in São Paulo state (based on [29]).

Figure 4 .
Figure 4. Source of substrate for biogas production(bars) and number of facilities (dotted lines) per sector in Rio de Janeiro state (based on [29]).

Figure 6 .
Figure 6.Source of substrate for biogas production(bars) and number of facilities (dotted lines) per sector in Paran� a (based on [29]).

Figure 8 .
Figure 8. Source of substrate for biogas (bars) and number of facilities (dotted lines) per sector in Goi� as (based on [29]).

Table 1 .
Nominal monthly household per capita, potential for biogas production per sector, estimated realized potential, and motive for selection.