**1. Introduction**

The efficient use of energy is of utmost importance for process sustainability and emission reduction [1]. This is an area of increasing research and practical interest that has persisted to this day [2]. All major economic sectors are under investigation, including industry [3], transportation [4] and agriculture [5].

The use of all types of resources and the impacts of processes on the surroundings can be related to the use of the energy necessary to complete the tasks. The evaluation of industrial systems is most frequently performed by using mathematical models for the consistent estimation of their thermodynamic properties and related energy use. Commercial simulators provide this functionality [6].

When comparing alternative processes, the energy demand is not always a suitable metric because it may not take into account the quality of the energy used. In this context, exergy is the property that can be used as a combined indicator of energy quality and quantity [7]. This property of exergy allows the optimisation of the process design and operation based on realistic estimates of how much energy can be sourced, converted, supplied or reused. Extended Exergy Analysis also takes into account the

economic aspects of a resource; e.g., a wind turbine in a more windy place has a higher exergy efficiency because it produces more energy with similar investment and operation costs [8]. This concept can be applied to an isolated unit (e.g., a wind turbine), to an industrial process (e.g., concrete industry [9]) or a farming system (e.g., canola [10]).

Process systems can no longer be considered in isolation [11], dealing only with the maximisation of their standalone efficiency. While process efficiency is important for obtaining profit, its environmental impact spans beyond the boundaries of the current system. This conflict between the usefulness of the streams and resources of a process and their effect on natural storage is solved by the concept of circularity [12], in which the overall life cycle is considered.

The exergy concept has been shown to be key to addressing sustainability issues [13]. The use of renewable resources is beneficial, as this takes advantage of natural energy flows across the Earth, without depleting accumulated terrestrial energy reserves, such as fossil fuels [14]. Therefore, the share of renewable resources used in the economy should be increased, although the exergy obtained in some of the harvesting paths may be small. Another confirmation of the usefulness of exergy for sustainability modelling comes from the domain of water management and water treatment plants [15]. However, despite being proven to be potentially useful, the use of the exergy concept is frequently limited only to the estimation of the exergy efficiency of various process contexts [16], such as the exergy efficiency of a process or the share of exergy from the renewables provided to a process.

There are examples of exergy assessment in the literature. Changes in the concentration of a solvent give rise to a massive exergy loss, indicating the importance of solvent selection [17]. An exergy analysis to evaluate the performance of a continuous Directional Solvent Extraction (DSE) desalination process using octanoic acid was presented in [18]. Extractive solvent regeneration is a potential method to substitute stripping and reduce the exergy demands of CO2 capture systems [19].

To compare process alternatives using exergy, the selection of system boundaries and reference points (e.g., ambient conditions) must provide comparable output streams. The same issue is also typical for the implementations of the Life Cycle Assessment (LCA) framework [20], where the choice of the system boundary and selection of life cycle stages is crucial to obtaining credible results. This similarity is useful for the potential integration of exergy-based criteria within the framework.

There have been many attempts to define a universal reference state [21]. The restricted dead state is defined as the physical thermodynamic equilibrium with the reference state. However, a dead state which takes the chemical equilibrium into account is required for environmental assessments. A widely used variant is based on an Earth similarity condition [22], where a reference substance is chosen for every element [23]. Substance exergies are determined to start from those of the reference substances, considering balanced chemical reactions. Regretfully, some chemical exergies are negative, and the reference is not entirely consistent [24].

The initial reference state has been updated according to new and more accurate geochemical and geological information. Thanatia [25] is a thermodynamically dead planet in which all materials have reacted, dispersed and mixed; i.e., it represents a complete dispersed state of minerals and the complete combustion of fossil fuels. Thanatia is not a reference state but a baseline used to calculate concentration exergies, therefore providing the exergy replacement costs. To assess the exergy degradation of the natural capital, the reference environment has evolved to a Thermo-Ecological Cost (TEC) methodology which in combination with the concept of Exergy Replacement Costs (ERC) results in the TERC (Termal-Exergy Replacement Cost) methodology, which is used to assess the degradation of fossil and mineral capital [25].

The choice of the reference conditions can also have a significant effect on the evaluation of the chemical exergy of particular substances such as fuels [26]. This is even more important for the evaluation of the exergy efficiency of large-scale systems, such as the Turkish industrial sector. A study of the trends in this area [27] revealed an increase from 25% to 29% when the ambient reference temperature decreased from 298 to 273 K.

From the perspectives of ecological modelling and the life cycle, it is possible to use the concept of embodied exergy: the cumulative amount of exergy inputs necessary to deliver a product or a service [28]. The cited work has applied the concept to exergy costing and accounting for energy sector applications, linking exergy spending to monetary costs.

Although exergy is very useful for assessing the loss of resource quality, its use has not been widespread in environmental impact evaluation. LCA is one of the well-established techniques with which exergy has been combined to conduct the exergy analysis of a complete product life cycle [29].

One method to quantify the environmental impact of a process based on exergy is the use of the environmental compatibility indicator, which takes into account the input exergy to the process and the exergy requirements for the abatement of process emissions and waste [30]. In an ideal case (no impact considered), the discussed process emits only heat.

Furthermore, the highest exergy efficiency does not correspond to minimum costs [31] or minimum environmental impact [32]. Exergy efficiency, in that sense, is a local evaluation criterion and is only appropriate to specific energy conversion or use schemes.

Circular economic flow is based on the separation of technology and the economy as the main condition [33]. This concept considers the inputs and outputs of operations during industrial production and focuses on cause-effect relationships. The author considers the circularity concept in terms of temporally repeating cycles of economic activity and presents the realisation that the economy cannot be considered separately from the environment.

Different industrial approaches to the improvement of the sustainability of human society and the environment have been attempted. The simple approaches to the substitution of materials and the end-of-pipe reduction of harmful emissions have been superseded by LCA-based methods for ecological design and economics [34]. The understanding of the interconnections, inputs and outputs for the entire supply leads to the goals of the circular economy [35]. In this context, close attention has to be paid to the full life cycle, including the facility construction and decommissioning, as has been shown in an analysis of the reuse of materials from wind turbines after their end of service [36].

The utilisation and reuse of different types of waste may be analysed by systematic approaches: e.g., P-Graph offers a solution for closed-loop processing and the analysis of its impact [37]. Process Integration also has great potential for analysing circular flows, especially in improving the sustainability of energy systems [38].

For the effective application of targeting and optimisation models in the design, operation and retrofitting of industrial processes for the circular economy, it is necessary to have flexible and scalable modelling concepts and tools. Conventional logic treats process streams as either inputs or outputs, where the outputs are either products or waste streams [39]. The waste streams were traditionally thought of as needing to be treated and disposed of. The circular economy paradigm for process design [40] requires non-product outlet streams to be treated as sources of potential resources as well.

Besides research, regulatory action has also been taken; for example, the EU action plan for the circular economy [41]. Some ideas related to circularity have been developed previous to the popularisation of the circularity concept; e.g., reuse, remanufacturing or recycling [42]. Sustainable Consumption and Production (SCP) tools have been identified as a booster of circularity [43]. The implementation of circularity has resulted in innovation opportunities [44]. This is the case with the redesign of pharmaceutical supply chains to prevent the waste of medical supplies [45].

A clear example of circularity is the mass flow in nature [46]: a mixture of dead biomass is decomposed by microorganisms and fungi to simple molecules that are captured by plant roots to generate complex molecules again using solar energy. This nutrient flow takes place in natural environments but not in agriculture, where the products are transported away to consumers without returning back to fields [46], breaking the natural cycle.

There is intensive research available in the literature about circularity in the industry, such as in metals processing [47], including copper [48] and steel [49]. Other fields have also been researched, such as construction [50] or forest wood harvesting and utilisation [51]. However, the global economy is not circular because large amounts of materials are used only once to provide energy or commercial value and are thus not available for recycling [52].

Examples of circularity in the chemical industry are related to plastics recycling as a consequence of the strategy of the European Commission [53]. The practices include plastic sorting [54], product design [55], or the design of chemical bonds suitable for biodegradation [56].

Many authors have defined circularity and its advantages and provided tools to quantify it. Examples include Corona et al. in 2019 [57], who focused on the circularity metrics, and Sassanelli et al. [58], who dealt with the assessment methods and the identification of the systematic taxonomy of the indicators used for circular economy evaluation by Saidani et al. [59].

The provided state-of-the-art review has shown that various tools and practices are available for process network optimisation, allowing the identification of the potential reuse paths for material components. However, accounting for the reuse of multiple resources within complex networks, containing multiple loops, creates a multi-dimensional optimisation problem if only approached directly. This observation reveals the need for an accounting framework and concepts that would measure the degree of sustainability and favorability of process networks adequately, taking into consideration the heterogeneous nature of the networks both in terms of their activities and the multitude of resources tracked.

The current work presents a system of analytical concepts, a framework and tools for evaluating the impacts of process systems based on thermodynamics. The framework is based on the concept of exergy as the unifying performance metric. It defines the tools of exergy assets and liabilities that enable the assessment of the sustainability of the considered systems. The trade-offs between the different feedstock and product flows and environmental impacts are modelled using the exergy assets and liabilities, leading to the calculation of the exergy footprint. The remaining content of the article presents the model and framework (Section 2), followed by illustrative case studies (Section 3) and a concluding discussion in Section 4.
