A framework for integrating safety and environmental impact in the conceptual design of chemical processes

Introduction

Multiple factors influence chemical process design and technology selection, including technical, economic, environmental, and safety considerations [1]. Traditionally, a techno-economic analysis has been used to select a base case design, while safety and environmental impact have been subsequently assessed [1, 2]. This inadvertently leaves out designs that may exhibit better environmental and safety performance than the selected base case at a very early stage of design, where abundant opportunities for incorporating these objectives are present. Furthermore, modifications to the process to address environmental and safety issues also require more effort in addition to becoming more expensive at later stages in the hierarchy of process design. This may impact the capital & operating costs and thus, the overall economics of the process. Thus, there is a growing awareness of the simultaneous consideration of these objectives during the conceptual process design phase of a project in order to select the most sustainable process route [2]. Although safety is an integral part of the overall sustainability of a chemical plant, historically safety has been addressed separately from sustainability [3]. There are a number of internationally recognized guidelines that may serve to guide sustainability efforts by the chemical industry. The AIChE Sustainability Index assesses the sustainability performance of facilities from seven perspectives [4], as shown in Fig. 1. The metrics include safety performance and environmental performance. Safety performance encompasses employee safety, process safety, and plant security while environment performance is based on resource use, emissions, and compliance management. Cefic proposes a safe and sustainable-by-design concept, according to which, a safe and sustainable design will require a set of harmonized criteria within an assessment framework [5]. Cefic has detailed a long list of design criteria in order to achieve this as can be seen in Fig. 2. The list includes a number of environmental impact criteria as well as occupational health elements within the health & safety and SHE profile criteria. Nevertheless, process safety which is related to the prevention of major accidents in process plants is not sufficiently emphasized in the guideline. In this context, a broader understanding of sustainability in the chemical process industries (CPIs) needs to be extended to include safety, as CPIs deal with hazardous substances and processes that can have a major impact on the well-being of workers, surrounding communities, and natural resources.

Fig. 1:

Elements of the AIChE sustainability index (information from [4]).

Fig. 2:

Elements of the Cefic safe and sustainable by-design concept (information from [5]).

In order to perform sustainability studies, it is important to know at which stage of the process design we intend to work. Process-based methods are usually classified into three categories: (i) process chemistry, (ii) conceptual process design, and (iii) process modeling and simulation [6]. The approaches differ in terms of information/time requirements and resulting accuracy [7]. In the chemistry selection stage, when no process data is available, the process chemistry-based method may be used for sustainability assessment. To increase the level of detail, it is necessary to move up from process chemistry to process design considerations. Table 1 presents the hierarchy of process-based methods for conducting sustainability assessment [6]. The conceptual design phase is particularly important as it establishes the initial idea and process objectives are translated to design parameters and working principles. For this reason, it is considered the most critical design phase in the overall design cycle [8]. The conceptual design phase poses valuable opportunities for incorporating safer and environmentally friendly technologies. Also, considering the fact that the conceptual design stage involves the best trade-off between data requirement and level of detail, performing the sustainability assessment at this stage can be considered ideal. However, conventional sustainability assessment tools, for example, the ISO 14040 standard, are not practically applicable for process design, as they require a large amount of data on the entire life cycle and the procedure can become overly complicated [9]. The conventional sustainability techniques are also limited to environmental impacts only and do not take into consideration safety, economic or other dimensions of sustainability. The key to an effective sustainability assessment method for selecting the most sustainable process route involves the parsimonious selection of adequate metrics which define the sustainability profile of the process and an integrated multi-criteria decision-making (MCDM) framework.

The present work investigates gaps in conceptual process design and existing sustainability assessment methods. An overview of existing environmental impact and safety assessment methodologies/tools is presented in this regard. Next, a possible workflow that incorporates safety and environmental impact in a holistic multi-criterion decision-making framework (MCDM) has been proposed to select the most sustainable process route. The use of this framework is illustrated through a simple case study involving assessing solvent alternatives for palm oil recovery to highlight the scope and significance of the proposed framework. The paper concludes with some final remarks about the proposed framework and how existing methods and tools may be incorporated into it.

Safety vs. sustainability in chemical process design

Before moving forward, it would be useful to consider the safety/health and environmental concerns of some commonly used chemicals in order to appreciate the fact that a trade-off is often necessary to select the most sustainable process/product. Table 2 lists a variety of chemicals, which are known to be physically hazardous, or harmful to human health and the environment. The information in the table has been collected from the Safety Data Sheets (SDS) of the respective chemicals (GHS category). The table shows an interesting trend that environment-friendly chemicals in the list are either reactive/flammable or hazardous to human health. On the other hand, nontoxic and non-flammable chemicals are mostly bioaccumulative. Hydrogen peroxide, for instance, may cause fire or explosion and is a strong oxidizer, however, according to US EPA, it is a chemical of low environmental concern. The exceptions in the list are hexane and monoethanolamine, which possess all three types of hazards i.e. they are flammable, occupational hazards as well as harmful to the environment. Despite the hazards, it is the single most used solvent for the extraction of vegetable oil from oil cake due to its functional properties i.e. high selectivity, simple recovery, non-polar nature, and low latent heat of vaporization. Conversely, Monoethanolamine, both health and environmental hazard is being used to save the environment i.e. to capture CO₂ as an absorbent.

Overview of methods and tools for assessment of sustainability and safety

A large number of methods and tools for environmental impact and safety assessment are found in the literature. In this section, selected methodologies and tools for environmental impact and safety assessment are discussed.

The concept of sustainable development stemmed initially from environmental concerns. Environmental impact assessment has been recognized as an essential part of chemical process design for more than 20 years [10]. Various performance indicators have been proposed and used in different environmental impact assessment studies. Table 3 lists the most common environmental impact categories, together with their characterization and equivalence factors which have been used in earlier studies. Environmental impacts can be quantified along the complete life cycle of a process by life cycle assessment (LCA) [11]. The first step in an LCA involves defining the system boundary. This requires a decision to be made on which life cycle stages are to be included in the assessment. Though LCA studies should preferably consider the entire life cycle of a process (the so-called cradle-to-grave approach) [12], narrower system boundaries may be sufficient in many cases. These include cradle-to-gate, gate-to-gate, or gate-to-cradle approaches. A cradle-to-gate approach ignores all emissions beyond a certain defined boundary. Such a cradle-to-gate approach is appropriate when comparing processes for which the life cycle after the gate is identical. For example, if the desired output is the same chemical, it is not necessary to extend the LCA study beyond the product formation stage. Similarly, a gate-to-grave approach may be suitable for alternative waste treatment options. A gate-to-gate analysis can be used in comparative assessments in very special cases and is generally not recommended [12].

Several methodologies focus on the performance of a system by using technological indicators such as resource usage, reuse of materials, recoverability of waste materials, renewability, etc. Along with these comprehensive methods, several shortcut tool kits are available which may be used if the means for assessment are limited. These may give a first indication of the environmental impact of a product or process. In recent times, many LCA software has also become available, a number of which are open source. This software has various features that include a user-friendly interface for visualizing results, and options for alternatives comparison and sensitivity analysis. Table 4 lists some selected environmental impact assessment methods/tools of the types discussed.

Researchers have also proposed various safety evaluation methods over the years to check the inherent safety prospect of a design. These methods have used a wide variety of indicators to characterize the inherent safety of the process. Table 5 lists the most common safety indicators used in these methods. The methods found in literature can be classified into four categories: (i) consequence-based assessment, (ii) indexing methods, (iii) graphical assessment, and (iv) risk-based assessment. In the consequence-based approach, safety is evaluated based on the probable consequences of accidents. The earliest example of this approach includes Dow’s index and the Mond index. Even though these two methods have been widely used in industry for a long time, they are not usable in the early stage of process design, and the results are difficult to interpret [27]. Indexing methods are more suited for earlier design stages as they require less information. In these methods, a score is assigned for each safety parameter and an aggregate index is calculated for the process route. Gupta and Edwards [28] argue that adding parameters of different dimensions and comparing the aggregate value may not be scientifically sound. Further, making the terms dimensionless and scoring the parameters based on their hazard rating is time-consuming. They proposed graphical procedures to overcome this limitation. These methods visually depict the effect of different chemical and process safety parameters such as temperature, pressure, toxicity, etc., which help in decision making. Risk-based assessment methods evaluate the risk inherent to a process by combining probabilistic data and consequence determination. Table 6 lists selected safety assessment methods/tools of the types discussed. For a more comprehensive overview of various inherent safety assessment tools, readers may refer to reviews by Roy et al. [29] and Park et al. [30].

It is evident from the literature review summarized above that environmental sustainability has been dealt separately from safety and there is indeed scope for further study to integrate these two factors for design purposes.

Suggested framework for integrating chemical safety and environmental impact in conceptual design

In light of the discussion in the earlier sections, it can be said that there is a need for simultaneous consideration of multiple objectives during the conceptual process design phase in order to select the most sustainable process route. Also, it is important to integrate safety along with environmental performance in the sustainability assessment procedure. In this context, a holistic and integrated multi-criteria decision making (MCDM) framework is proposed which incorporates environmental, safety and economic considerations as a starting point towards selecting the most sustainable process route during conceptual design stage. The workflow for selecting the most sustainable process route is shown in Fig. 3.

Fig. 3:

Work-flow for selecting the most sustainable process route (adapted from Ortiz-Espinoza et al. [42]).

The key steps of the proposed work flow are described as follows:

STEP 1 Collect data: Collect initial data such as production capacity, feedstock composition, product specification and process chemistry in order to proceed with process simulations and further work.

STEP 2 Alternatives identification and simulation: Study the alternative designs of the process and simulate the process alternatives in a process simulator. The outputs of process simulators provide the process and inventory data such as mass/energy balances, operating conditions etc. needed to perform the subsequent process assessment.

STEP 3 Metric evaluation: Assess the inherent safety, environmental friendliness and economic performance of the selected designs. Inherent safety assessment should include chemical properties such as flammability, explosiveness, toxic limit, chemical interaction, heat of main/side reaction and process parameters including inventory, temperature and pressure. Environmental impact assessment should evaluate processes on the basis of parsimoniously selected global environmental parameters including global warming potential, ozone depleting potential etc. and ecological environmental indicators including acidification potential, eutrophication potential etc. Finally, economic performance assessment needs to include both capital and operating costs and can be performed using widely used methods such as NPV method, internal rate of return (IRR) method, benefit–cost ratio (BCR) method, etc.

STEP 4 Multi-Criteria Decision Making (MCDM): Choose and perform a suitable MCDM technique such as TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) to identify the best route among the alternative designs.

Case study

The significance of the issues discussed above is highlighted in this section through a case study involving assessing the inherent safety and environmental impact of solvent alternatives for palm oil recovery. A solvent is used to recover about 5–6 % of the residual palm oil that still exists in pressed palm fiber [43]. n-hexane is the preferred solvent for this purpose because of the economy it offers, however it has drawbacks of high flammability and toxicity. There has been a number of works exploring alternative solvents for residual palm oil recovery. Ahmad et al. [44] used Computer-aided Molecular Design (CAMD) to identify 3 potential alternatives to this solvent, which are all mixtures of n-hexane with another chemical. The authors first applied CAMD to generate alternative single solvents as substitutes for n-hexane which satisfy a number of target properties including boiling point and flash point. Once a number of suitable single solvent candidates were identified, the authors formulated a linear optimization problem with an objective function to minimize the n-hexane composition to design mixtures of n-hexane and the other identified solvents (ethanol, acetone and butanol) that satisfy the desired properties. The resulting mixtures are:

1. 65 % n-Hexane+ 35 % Ethanol

2. 30 % n-Hexane+ 70 % Acetone

3. 40 % n-Hexane+ 60 % Butanol

In the present case study, we shall explore the inherent safety and environmental impact characteristics of these options. First of all, we do a preliminary hazard assessment of the alternative chemicals identified. The health, environmental and physical hazards associated with these chemicals are presented in Table 7.

Ahmad et al. [44] performed inherent safety assessment for each of the options with respect to flammability, toxicity, reactivity, and explosiveness. The values obtained from Ahmed et al.’s study is given in Table 8. Furthermore, we have used the Green Degree method proposed by Zhang et al. [13] to calculate the environmental impact of the solvents.

The equation for calculating the green degree of a substance is [13]:

where, φ i , j N = φ i , j φ i , j max , φ i , j max = max ( φ i , j ) , ∑ j 9 α i , j = 1 and i = 1, 2, 3, …; j = 1, 2, 3, …, 9.

Here, G D i s u is environmental impact potential of substance i for impact category j with respect to 9 impact categories: global warming potential, ozone depleting potential, photochemical ozone creation potential, acidification potential, eutrophication potential, eco-toxicity potential to water, eco-toxicity potential to air, human carcinogenic toxicity potential to water, and human non-carcinogenic toxicity potential to water. φ i , j N is the relative impact potential obtained by normalizing φ_i,j by φ i , j max which is the maximum value for category j among all reported substances. α_i,j is the weighting factor of substance i for impact category j [13].

Green degree of a mixture is calculated by [13]:

GD^mix has units of green degrees per kilogram of mixture (gd/kg of mixture). G D k s u is the green degree of component k, and x_k is the mass fraction of component k in the mixture.

Using eqs. 1 and 2, the green degree values for n-hexane and the alternatives have been calculated and given in Table 8.

It may be seen that option C, 30 % n-Hexane+ 70 % Acetone ranks the best in terms of both inherent safety and environmental impact. Option A, n-hexane, though being the worst in terms of inherent safety, ranks second in terms of environmental impact, while option D, 40 % n-Hexane+ 60 % n-Butanol ranks second in terms of inherent safety, but has the worst environmental impact.

This example might seem relatively straightforward as option C ranks first in terms of both the objectives considered. However, in more complicated situations, several complexities may arise:

1. One option might rank best in terms of one objective, while another might rank best in terms of another objective.

2. There might be more objectives to consider.

3. There might be only a marginal difference between the different options for a particular objective.

The case study emphasizes and reiterates the need for a systematic framework for the consideration of multiple objectives during the conceptual process design phase in order to select the most sustainable process route. In more complex situations, there may be numerous different routes that need to be compared. Hence, a framework as described in the previous section that incorporates the existing knowledge, while making the process more efficient and robust is necessary for a complete alternative assessment procedure.

A number of researchers have attempted to formulate a multi-objective optimisation problem incorporating safety and sustainability to select the most sustainable process route [2, 45, 46]. Guillen-Cuevas et al. [2] for example, demonstrate the conflicting objectives that may arise in a design problem using a case study on methanol synthesis. Higher pressures improve economic performance and reduce carbon footprint, but increase the explosiveness potential, thus affecting safety. The authors have applied a return on investment (ROI) based metric extended to include safety and sustainability-relevant performance criteria to balance these conflicting objectives to determine an optimal pressure that considers all these objectives.

Conclusions and the way forward

In this paper, the need for integration of safety and environmental impact at the conceptual design stage has been discussed and a possible workflow that incorporates these objectives in a holistic multi-criteria decision-making framework has been presented. This shall enable parallel consideration of multiple objectives in the decision-making process as opposed to the traditional method in which safety and environmental impact are assessed after the selection of a base-case design considering economics only. We have focused on the conceptual process design stage when the availability of data is limited but it is crucial to perform a sustainability assessment at this stage considering the increased efforts and costs associated with assessments at later stages. Additionally, a case study to highlight the existing gaps and an overview of selected existing methodologies and tools for environmental impact and safety assessment are discussed which best fit the concept of the proposed integrated sustainability assessment framework. The different environmental performance and safety assessment techniques are different with respect to (i) purpose, (ii) indicators chosen, (iii) system boundary chosen, (iv) information requirement and level of detail (v) complexity. Depending on the available information and purposes, these existing methods/tools may be incorporated into the proposed framework in order to study the sustainability performance of the process from multiple perspectives and determine its overall sustainability profile.