Substitution potential of rare earth catalysts in ethanol steam reforming

This study suggests the possibility of substituting rare earths containing catalysts in ethanol steam reforming by meansofsustainabilityassessment.FourNi-catalystsareassessed;twooftheNi-catalystsarerareearthscontain-ing catalysts Ni/Al 2 O 3 -CeO 2 and Ni/La 2 O 3 while the other two are non-rare earths containing catalysts Ni/Al 2 O 3 and Ni/ZnO.The sustainability assessmenttoolused coversenvironmental, healthand safetyand economic indi-catorsinconjunction with a linear scaletransformation (Max) normalization technique and ananalyticalhierar-chy method to evaluate the sustainability performance of the catalysts. The sustainability assessment results obtained demonstrated that Ni/Al 2 O 3 is the best performing catalysts in terms of the overall sustainability of theethanolsteamreformingreaction.Therareearthscontainingcatalystsupports,i.e.ceriumandlanthanumox-ides can be successfully substituted with aluminum oxide catalyst support whilst still maintaining the overall sustainability performance of the reaction.


Introduction
Rare earth elements (REEs) are classified as critical raw materials [1] because they are highly demanded worldwide, are prone to supply risk, and besides it is very challenging to find suitable substitutes for them in certain applications (e.g.REEs permanent magnet in wind turbines), and their rate of recyclability is very low [2,3].The United States Department of Energy (DoE) identified the five most critical REEs as neodymium, yttrium, dysprosium, europium, and terbium with regards to their reserves, as well as their demand and supply levels [4].Consequently, cerium and lanthanum are usually used as catalytic materials and are more abundant than neodymium, praseodymium, and dysprosium used in permanent magnet applications.According to Roskill in 2016, catalysts have the largest global REEs consumption by tonnage, however, magnets have the largest application area by value (up to 24%) [3,5].
In recent years, rare earth oxides (REOs) have been investigated as catalysts support, and promoters in ethanol steam reforming (ESR) for the production of hydrogen, particularly lanthanum oxides (La 2 O 3 ) and cerium oxide (CeO 2 ).La 2 O 3 is known to increase nickel and support interaction as well as nickel dispersion.Additionally, La 2 O 3 are favored in ethanol steam reforming because of their high alkalinity, which favors dehydrogenation reaction.[6] CeO 2 acts as an oxygen supplier to help prevent catalyst deactivation caused by carbon deposition on the catalyst surface.[7][8][9][10][11] In ESR, the substitution of REEs containing catalysts by non-REEs containing catalysts is desired to reduce the criticality effect of REEs.The substitution of REEs by non-critical metals represents an important approach to combating their criticality.Some of this approach could be a total material substitution, using less critical REEs and the use of REEs in smaller amounts [12].Substitution of REEs can occur in various ways e.g. by direct substitution, gradefor-grade substitution, technology-for-element substitution, and system-level substitution [2,3].Nickel (Ni)-catalysts are often preferred for use in ESR mainly because they are cheaper than the noble metal catalysts' counterparts [13].Several studies have been conducted over the years using Ni-catalysts [6,[13][14][15][16][17], it has been found that when combined with supports like aluminum oxide (Al 2 O 3 ), CeO 2, and La 2 O 3 , they are known to exhibit high steam reforming activity in ESR.
ESR is important because it provides pollution-free sources for hydrogen production.It utilizes ethanol, which is beneficial because it is mostly renewable, is readily available, is biodegradable, has low toxicity, and can easily be transported [13,18].ESR may consist of other simultaneous reactions in addition to the reaction in Eq. (1) [18].
Ethanol steam reforming (ESR), is represented by the following overall stoichiometric reaction [19].
The formation of acetone (CH 3 COCH 3 ) as an intermediate by-product may lead to coke deposition on the catalyst surface, which in turn, can lead to the deactivation of a solid catalyst.[17,18].To minimize the coke formation and to reduce the number Sustainable Materials and Technologies 26 (2020) e00237 ⁎ Corresponding author.

Contents lists available at ScienceDirect
Sustainable Materials and Technologies j o u r n a l h o m e p a g e : w w w .e l s e v i e r .c o m / l o c a t e / s u s m a t of by-products generated in the process, the development of efficient and selective catalysts is required [17].The catalysts used in ESR are required to have the following properties: high activity and selectivity to hydrogen, the ability to resist deactivation, and stability throughout the use in reforming reaction [20].
Sustainable development often interchanged with sustainability has various definitions making it a somewhat confusing concept.However, a broader and more commonly accepted definition of sustainability was coined by the Brundtland Commission previously known as the World Commission on Environment and Development (WCED) where Sustainable development was defined as "The development that meets the needs of the present without compromising the ability of future generations to meet their own needs".[22] An alternative definition of sustainable development is finding ways to strike a balance to environmental, social, and economic needs [23].A widely accepted sustainability concept is that it aims to integrate the economic, social, and environmental aspects, these three aspects are known as the three pillars of sustainability [23,24]; this is also similar to the "triple bottom line" (TBL) concept.TBL was coined by Elkington in 1997.TBL was coined to further provide an understanding of the concept of sustainability as well as the importance of achieving sustainability.Its introduction into sustainability simplified the environmental (planet), economic (profit), and social (people) aspects of sustainability in such a way that it can be put into practice by decision-makers, policymakers, and organizations.It entails the interrelationship between the three aspects of sustainability by ensuring that they maintain their individuality, while simultaneously working together in harmony [25,26].In Elkington's sustainability approach, the "people" and "planet" can be viewed as being mutually beneficial to all parties involved, while that of "profit" can be viewed as a private interest.However, in 2002 the World Summit on Sustainable Development aimed to address this disparity by replacing "profit" with "prosperity" [27].
The increased awareness of sustainability in recent years has led to the need for efficiency of chemical processes, health, and safety consideration, as well as the utilization of renewable resources such as bioethanol derived from biomass [21].The chemical industry accounts for about 7% of the world's industrial greenhouse gas emissions.As a result, sustainability concerns are expected to increase in the chemical industry across processes, production systems, operations, etc. [28].Therefore, it is of utmost importance for sustainability assessment to be incorporated and to aid in the decision-making process in the chemical industry.Sustainability assessment can be described as a decisionmaking process that guides stakeholders and decision-makers into making decisions that uphold sustainability.Sustainability assessment takes into consideration the environmental implications as well as the social and economic impacts of projects, processes, services, etc. [29].There are several types of sustainability assessment methodologies currently available.One of the most prominent sustainability assessment methodologies is the Multi-Criteria Decision Analysis (MCDA).
MCDA was developed to evaluate the sustainability performance of processes and production systems and to provide clarity on how sustainability assessment could be conducted [26,30].
MCDA offers a formal approach which seeks to help decision-makers to make informed decisions based on multiple criteria.MCDA can be described as a set of tools, methods, and approaches that combine mathematical methods and technical information to choose the optimal solution amongst alternatives.The MCDA is often used to assess the sustainability of materials and processes and is very useful when comparing varying sustainability (environmental, social, economic) indicators.It provides transparency and helps to reduce subjectivity in decision-making.[26,31,32] One of the most widely used MCDA groups that specialize in analysing complex decision problems is the Analytic Hierarchy Process (AHP) developed by Saaty in the early 1970s [33].The AHP is a multi-attribute technique that enables decision making through the use of pairwise comparisons either individually or together as a group.The AHP can be carried out by either of the following two methods, (i) the aggregation of individual judgements and (ii) the aggregation of individual priorities.In the aggregation of individual judgements, the decision-making group makes a decision together as one unit.Alternatively, in the aggregation of individual priorities, the decision-makers make decisions as separate individuals [33][34][35].
An important criterion for sustainability assessment based on MCDA is the normalization of data to get comparable results.[36] Criterion in this context refers to the standard by which judgment is based on.[37] It has been suggested that for MCDA to successfully support sustainability assessment, it has to obtain input from other assessment tools [38].Combining MCDA with other sustainability assessment tools would facilitate the interpretation of sustainability assessment results.MCDA is useful in identifying the most preferred option amongst several alternatives, however, on its own, it is not very robust as a sustainability assessment tool.Therefore, there is a need to integrate it into other well-rounded sustainability assessment tools [31,38].This will provide an informative assessment that can facilitate decision making at the laboratory phase of processes.
While there is existing literature on the substitution of REEs in battery, lighting, magnetic, and catalyst applications [3,[39][40][41][42][43][44][45][46][47][48][49][50][51], information on detailed sustainability assessment of REEs containing catalysts is lacking.This study as an example focuses on the substitution potential of REEs containing catalysts, i.e. cerium and lanthanum in ESR.Sustainability assessment of four Ni-catalysts (REEs and non-REEs) used in ESR is performed and their respective sustainability performance is ascertained.The sustainability assessment tool covers environmental, health and safety, and economic sustainability aspects and is used in conjunction with the MCDA AHP method.

Methods
The method used in this study was divided into six stages (Fig. 1): (i) systems boundary definition, (ii) sustainability indicators definition, (iii) data collection, (iv) ESR process design and simulation, (v) sustainability assessment, and (vi) decision analysis.In the context of this study, the system boundary is first defined.The system boundary shows raw materials inputs, the output products for the catalyst preparation phase, and the ESR phase, as well as the energy flows for the ESR phase (Fig. 2).ESR is performed in laboratory scale.

Definition of sustainability indicators
The triple bottom line approach to sustainability aims to integrate the social and environmental impacts associated with a product, process, or service to its economic performance in a way that can be measured [52].In other words, the three aspects of sustainability should be present in any sustainability assessment.
Over the years, several approaches have been developed to evaluate sustainability performance.Approaches that integrate the three aspects of sustainability have been widely used [53].Sustainability assessment usually depends on a variety of indicators.The definition of sustainability indicators is a very crucial step in determining the outcome of the sustainability assessment.Sustainability indicators can be used to transform sustainability-related issues into quantitative measures of environmental, social, and economic achievement [54].Some important requirement that a sustainability assessment should have are as follows (i) the expected outcome should foster sustainable development, (ii) sustainability indicators are to be founded on key elements and clear guiding vision (iii) should possess ecological and societal fairness (iv) should take into consideration the triple bottom line concept i.e. people, planet and profit (prosperity), and (iv) should set goals aimed at implementing reasoning of logic models.[55] It is worth noting that sustainability indicators may not match precisely with the 17 sustainable development goals.The reason for this is because sustainable development goals are formulated for policymakers focusing mainly on macro, socio, and development economics and as such do not focus on process-relevant assessments [52].
In this study, the indicators used were based on the selected indicators from Green Engineering Principles [56].A similar approach was used by Saavalainen et al. (2017) to perform a sustainability assessment of an early stage chemical process design using Green Chemistry principles [57].Fig. 4 gives an overview of the sustainability indicators used in this study.
The 12 principles of Green Engineering were first developed by Anastas and Zimmerman in 2003 to achieve sustainability through the design of products, processes, or systems that are benign to human health and the environment.It aims at achieving sustainability through science and technology; and is especially useful in integrated systems.The 12 principles of Green Engineering provide a sustainability framework that serves as a guide for engineers and scientists when designing products, processes, or systems.It is based on twelve principles that cover the environmental, social, and economic sustainability aspects.The Green Engineering principles go beyond ensuring that only engineering quality and safety specifications are met to also covering the three aspects of sustainability (environmental, social, economic).These are considered key factors during the early design phase of processes, products, or systems.[56] Table 2 presents the 12 principles of Green Engineering, as well as the selected indicators used for the sustainability assessment in this study.From this table, all the Green Engineering principles except for Principle 5, 6, 8, 9, and 10 were used to propose the selected indicators used in this study.Principles 5 and 10 are not applicable because all the evaluated catalysts had similar raw materials, amounts used in the reaction, only the catalyst used varied.Principle 6 applies to commercial scaled processes.Principle 9 is not applicable here because the main product of the ESR is hydrogen.
Green Engineering principles were chosen as a guideline for use in the development of sustainability indicators used in this study.The sustainability indicators were selected to suit the laboratory stage nature of the process.Other relevant considerations for the selection of indicators were based on catalyst activity, their specificity, and challenges in the reaction (deactivation caused by carbon formation) as well as their performance in the reaction.Table 4 shows the assessment methods and the selected indicators used in this study.
The selected indicators consist of quantitative and semi-quantitative indicators.Quantitative assessment methods are based on numerical data expressed as scores and have a standard unit of measurement.On the other hand, qualitative assessment methods have usually yes or no answers that cannot be directly measured but could be described and narrated effectively as semi-quantitative numerical data.[58] Table 5 lists the sustainability indicators used, their description, and the assessment method each represents.

Data collection
The data source used in this study was taken from the laboratory experiments performed by Sánchez-Sánchez and Navarro (2007) and Yang et al. (2006); and are presented in Table 1  shows the weight percentage values of Ni loading of the catalyst and the waste generated during the catalyst preparation stage.

Ethanol steam reforming process design and simulation
For calculating mass-energy balance, a process simulation model was developed using an Aspen Plus process engineering software (Fig. 3).The process flowsheet was similar for all the cases because the same reaction conditions were applicable for all the catalysts.The process flowsheet consists of ethanol and water feed streams, a mixer, a reactor, a separating unit, a product stream, and a side product stream.The feed water to ethanol molar ratio used for all the catalysts was 8:1 (total flow 1 kmol/h), the operating temperature was 500 °C and the pressure was 1 bar.The outlet stream (S2) from the reactor was fed into the separating column (SEP) at a pressure of 10 bar.The separating column split the outlet stream into two streams consisting of the main product stream, i.e. the H 2 stream, and a side stream (S3) consisting of CO, CO 2 , CH 4 , C 2 H 4, and H 2 O.

Calculation
The sustainability assessment methodology utilizes information from process configuration as well as materials and energy flows obtained from simulation with Aspen Plus.It integrates process simulation and sustainability assessment.

Quantitative and semi-quantitative assessment method
The sustainability assessment method used in this study is grouped into quantitative and semi-quantitative as shown in Table 5.Additionally, the sustainability assessment results for all the catalysts were interpreted based on their sustainability performance using a minmax normalization method similar to [59].The min-max normalization method is one of the most commonly used normalization methods when performing a sustainability assessment and has been used in several studies [53,[59][60][61][62].It was chosen to avoid overcompensating the sustainability assessment results and it is more suited when the sustainability indicators are a mix of quantitative and qualitative methods.Normalization enables the comparison of different indicators that have varying measurement units as well as a range of values [63].
In the min-max normalization method, if a higher score is more beneficial, for instance in the case of the indicators hydrogen selectivity, ethanol conversion, Eq. ( 2) can be used.The normalized value score N i j is given as: X ij is the indicator being considered, X min is the minimum performance rating amongst alternatives for the indicator being considered and X max is the maximum performance rating amongst alternatives for the indicator being considered.
If a lower score is more desirable, for instance in the case of the indicators catalyst cost, energy used, waste gas emission" etc., Eq. ( 3) can be used.The normalized value score N i j is given as: The evaluation of semi-quantitative indicators was performed using the Likert scale.The Likert scale is useful for assigning quantitative values to indicators in an assessment to make them comparable.It is useful for evaluating qualitative and semi-quantitative indicators [64,65].For this study, the Likert scale having the degree of severity of an indicator ranging from "bad to worst" is allocated with scores from "4-0".The score of "4" is assigned to the case with the least impact while the score of "0" is assigned to the case with the most impact.

Economic indicators evaluation
The economic indicators consist of four quantitative and two semiquantitative indicators.The indicators ethanol conversion and hydrogen selectivity were evaluated with Eq. ( 2), while catalyst cost and energy used indicators were evaluated with Eq. ( 3).Subsequent criticality and stability of catalyst indicators were evaluated using Likert scale.The catalyst cost is equal to the sum of the cost of the active metal and the support of the catalyst.

Health and safety indicators evaluation
The health and safety indicators consist of semi-quantitative indicators.In assessing the health and safety impacts during the catalysts preparation and ESR, three things were taken into consideration.Firstly, the substances and chemicals used during the preparation of a catalyst were identified, secondly, the substances and chemicals were reviewed for a potential hazard in safety data sheets, and thirdly the substances and chemicals were classified and the degree of hazard was taken into consideration.Safety data sheets were used to obtain information on the potential hazard possessed by the chemicals used during the catalyst preparation and the evaluation was carried out using the Likert scale.

Environmental indicators evaluation
The environmental indicators consist of two quantitative and two semi-quantitative indicators.The waste gas emissions indicator was evaluated using Eq. ( 3), and the catalyst regeneration indicator was evaluated using Eq. ( 2).Additionally, the waste generated during the Intermediate values between judgements 1/3, 1/5, 1/7, 1/9 (reciprocals) If an activity x has a numerical value allocated to it when in comparison with activity y, then activity y has the reciprocal value of activity x.
preparation of the catalyst and renewability of the catalysts indicator was evaluated using the Likert scale.
To evaluate the waste gas emitted during the reaction, the emitted gasses C 2 H 4 , CH 4, and CO were converted to their CO 2 equivalent (CO 2-e ) values [66] as follows: where x is the mass of gas in tonnes, and GWP is the global warming potential of the gas.In Table 6, the Global Warming Potential of the waste gases is presented.

Analytic hierarchy process
In this study, a commonly used MCDA approach known as Analytic Hierarchy Process (AHP) based on [67], is used to specify weights for each indicator in the sustainability assessment.Weighting is used to prioritize the sustainability indicators based on their perceived order of importance.AHP method can be applied to both qualitative and quantitative data.It takes into consideration the consistency of decision-makers [67].In AHP, weighting is used to prioritize the sustainability indicators based on their perceived order of importance.
In AHP, selected indicators are arranged in a hierarchy structure starting from the assessment goal (sustainability assessment of ESR catalysts), down to the criteria (indicators), and ends with the available alternatives (e.g. the catalysts to be assessed).Weights are calculated in AHP using pairwise comparison judgements whereby the indicators are measured against each other.Table 7 presents the AHP pairwise comparison scale.
The values obtained from the pairwise comparison scale used for the assignment of weights were decided by a 10-member team of decisionmaking experts in the Environmental and Chemical Engineering Unit, Faculty of Technology at the University of Oulu, Finland.The decision-making experts had years of experience in catalysis, reaction engineering, process design, chemistry, and sustainability The AHP method used by the decision-makers in this study is the Aggregation of Individual Priorities (AIP).Equal weighting was assigned to each decision-maker corresponding to a one-person-one-vote procedure.A questionnaire was sent out to all the decision-making experts.Table 12 presents a list of the decision-makers' codes and their fields of expertise.The AHP approach was executed using the following four-step procedure:  ) at the fourth level (Fig. 5).
Step 2: Development of a normalized pairwise comparison between the second level indicators and calculation of the indicator weights (Table 8).The normalized pairwise matrix is obtained by dividing the indicator value of each column by the sum of column indicators.
The normalized pairwise comparison (N m ) is equal to: where IV C is the indicator value of each column, and I C is the sum of column indicators.The indicator weight is calculated by dividing the sum of each row's indicator values by the total number of indicators.
The indicator weight (W I ) is equal to: where ∑ IV R is the sum of each indicator value of each row, and I T is the total number of indicators.
Step 3: Development of a normalized pairwise comparison between the third level indicators and calculation of the weights.This will form the local weights of the indicators (Table 9).The local indicator weight is calculated by dividing the sum of each row indicator values by the total number of indicators.The local indicator weight is calculated by dividing the sum of each row indicator values by the total number of indicators.
The local indicator weight (W L ) is equal to: where ∑ IV R is the sum of each indicator value of each row, and I T is the total number of indicators.
Step 4: Calculation of the global weights of the indicators.This will be the weights of each indicator from Step 2 multiplied by the local weight from Step 3 (Table 10).
The global weight of an indicator (W G ) is equal to: where W I is the indicator weight obtained from Step 2, and W L is the local weight obtained from Step 3.
In calculating the weighted score results (W s ), the normalized score of each indicator obtained in Eq. ( 2) and Eq. ( 3) was multiplied by its weight and summed: where v is the sustainability assessment score in consideration, and w is the weights assigned to the indicator in consideration.

Process simulation and sustainability assessment results
Table 11 illustrates the energy used, conversion, and selectivity obtained for the ethanol steam reforming catalysts.The energy balance results from the simulation are defined as energy used in kW and the overall specific energy used of the steam reforming reaction was expressed per kmol of H 2 .In Fig. 6, the sustainability assessment results for all the catalysts are presented.The sum of scores for economic, health and safety, and environmental indicators are presented for all the four catalysts in consideration.The overall scores for each catalyst cutting across all the indicators are also presented.
From Fig. 6, it can be seen that the difference between the catalysts were (a) 23% between the Ni/Al      (48.25).

Sensitivity analysis
Sensitivity analysis was performed to measure the robustness of the sustainability assessment results.The final priorities of the alternatives are highly dependent on the weights attached to each indicator.Changes in the relative weights could in turn result in changes in the final rankings of the catalysts.Considering that the weights are often based on highly subjective judegments, there is a need to test the stability of the ranking for the varying criteria weights.The sensitivity analysis method used in this study is similar to that of [68], in addition, equal weights variation and variation of expert answers (swapping lowest and highest weight was also employed).Fig. 7 shows the sensitivity analysis results.Fig. 7 (a) shows the sensitivity analysis results when all the indicators are assigned equal weights; Fig. 7 (b) shows the sensitivity analysis results when the selectivity and the waste gas emissions indicators weights are decreased by 5% and the rest of the indicators weights are increased proportionally to keep the sum of the weights to 100%; Fig. 7(c) shows the sensitivity analysis results when the selectivity and the waste gas indicators weights are increased by 5% and the rest of the indicators weights are decreased proportionally to keep the sum of the weights to 100%; Fig. 7 (d) shows the sensitivity analysis results when the selectivity and waste gas emission indicators weights are decreased by 10% and the rest of the indicators weights are increased proportionally to keep the sum of the weights to 100%; Fig. 7 (e) shows the sensitivity analysis results when the selectivity and waste gas emission indicator weights are increased by 10% and the rest of the indicators weights are decreased proportionally to keep the sum of the weights to 100%.Fig. 7(f) shows the sensitivity analysis result when the weights are varied by swapping lowest with highest.The results of the sensitivity analysis, it can be seen that for Fig. 7 (a) to (e), two things were observed; firstly, the catalyst with the highest to the lowest sustainability score are the Ni/Al 2 O 3 , Ni/La 2 O 3 , Ni/ZnO and Ni/ Al 2 O 3 .Secondly, the ranking order remained the same as in Fig. 6 for Fig. 7 (b), (c), (d), (e).What this means is that small changes to the indicator's input does not affect the overall ranking performance of the catalysts.However, in Fig. 7(f), the ranking was different than in Fig. 7(a) to (e).Here it was observed that the catalysts with highest to the lowest sustainability score are Ni/Al 2 O 3 , Ni/ZnO, Ni/La 2 O 3 , and Ni/ Al 2 O 3 .What this means is that if there is a wide variation of weight assignment by the decision making expert, the result of the assessment may differ.

Uncertainty of the results
It is important to identify uncertainties when performing a sustainability assessment.Uncertainty about the sustainability assessment results presented in this study arises because the experimental data used for the assessment comes from the laboratory scale studies of catalyst development for ESR.ESR process is yet to be a commercially established process; subsequently, some of the experimental results may not be completely reliable which might affect the accuracy of the sustainability assessment results.Uncertainty may also arise due to changes in reaction conditions (e.g.pressure, temperature), as well as changes in the price of the catalysts.Additionally, uncertainty (large) may also stem from a narrow research data utilized based on a few experiments that may not be completely reliable and precise.Finally, uncertainty may also arise because of the method used for assigning weights although the analytical hierarchy process gives flexibility and offers the decision-maker a tool to accurately accord judgements based on the pairwise comparison.However, it has a subjective nature as such that two different decision-makers can get different results given the same process parameters.
4.4.Substitution potential for reducing the rare earths containing catalysts' use in ethanol steam reforming Owing to the criticality challenge posed by REEs there is a need to finding plausible substitutes when possible as well as adopting sustainable usage (from secondary sources) methods to REEs containing catalysts.Substitution of the REEs containing catalysts represents one means to combating their criticality and sustainable usage of the REEs containing catalysts represents another means of combating their criticality effect.
There have been considerable research efforts in finding sustainable usage for REEs, mainly because they are often discarded as wastes in some mining operations [5][6][7].Therefore, finding sustainable use for cerium and lanthanum would cushion their criticality effects.Although the major application areas of REEs catalysts are fluid catalytic cracking and automobile exhaust catalyst [3].ESR is a marginal application and does not play a major role in REEs catalytic application area.That said, the REEs containing catalysts used in ESR plays a very minor role in the criticality effects of REEs use.However, the sustainable use of REEs containing catalysts used for ESR is desired.This can be achieved by first obtaining them from secondary sources and subsequently utilizing them.

Conclusion
Rare earth elements are critical raw materials, they are highly demanded worldwide and are prone to supply risk.However, it is difficult to find suitable substitutes for them in many applications and besides their rate of recyclability is very low.To combat the criticality of rare earth elements, finding suitable substitutes is of vital importance.Additionally, the sustainable utilization of REEs obtained from secondary sources (e.g. via recycling) is also desired.In this study, the substitution potential of rare earths catalysts in ethanol steam reforming, ESR, was ascertained.Sustainability assessment of four different catalysts used in ESR was presented.Four-nickel containing catalysts were assessed, i.e.Ni/Al 2 O 3 -CeO 2 and Ni/La 2 O 3 , and two non-rare earth containing catalysts i.e.Ni/Al 2 O 3 and Ni/ZnO.The methodology used was combining process simulation, AHP, and sustainability assessment.The sustainability assessment tool utilized Green Engineering principles, as well as information from process parameters, materials, and energy flows obtained from simulations with Aspen Plus.The assessment was based on selected indicators and that covered the economic, health and safety, and environmental sustainability aspects.A min-max normalization technique (quantitative indicators) and, in some cases the Likert scale (semi-quantitative indicators) were used during the evaluation of the indicators.The analytical hierarchy process was conducted by a ten decision-making experts, and was used to specify weights for each indicator in the sustainability assessment based on their perceived order of importance.From the sustainability assessment result obtained, it can be seen the difference between the catalysts were catalyst had the best environmental performance.This was mainly attributed to the fact that because it had the least amount of generated waste gas emission .Sensitivity analysis conducted also demonstrated that there was no significant difference in the overall sustainability scores of the catalysts.In conclusion, the sustainable utilization of REEs obtained from secondary sources is to be encouraged.
. The catalysts selected for this study were Ni/Al 2 O 3 , Ni/ZnO, Ni/Al 2 O 3 -CeO 2 , and Ni/La 2 O 3 .The method used to prepare the Al 2 O 3 -CeO 2 support was incipient wetness impregnation with aqueous solutions of the corresponding metal nitrates.Additionally, the Al 2 O 3 support was obtained commercially.The ZnO support material was prepared by the decomposition of ZnCO 3 while the La 2 O 3 support was obtained commercially.[14,15] The Ni/ Al 2 O 3 , Ni/Al 2 O 3 -CeO 2 , Ni/ZnO and Ni/La 2 O 3 catalysts were prepared by impregnation with an aqueous solution of nickel nitrate.Table3

Fig. 2 .
Fig. 2. System boundary showing material inputs, and outputs of the catalyst preparation phase, and inputs, outputs and energy flow of the ethanol steam reforming phase.

Step 1 :
Development of a hierarchical structure with a goal (ESR) at the top level, main indicators (economic, social and environmental) at the second level, indicators (catalyst cost, hydrogen selectivity, energy used, etc.) at the third level, and the catalyst alternatives (Ni/Al 2 O 3, Ni/ ZnO, Ni/La 2 O 3 , and Ni/Al 2 O 3 -CeO 2
2 O 3 and Ni/ZnO catalysts; (b) 38% between the Ni/Al 2 O 3 and Ni/Al 2 O 3 -CeO 2 catalysts; (c) 30% between the Ni/Al 2 O 3 and Ni/La 2 O 3 catalysts.For the economic assessment, the catalyst with the highest score is the Ni/ZnO, with a score of 23.50.On the other hand, the catalyst with the lowest score with respect to the economic indicators is the Ni/La 2 O 3 catalyst having a score of 10.74.The most expensive catalysts are the REEs-containing catalyst Ni/ Al 2 O 3 -CeO 2 and the Ni/Al 2 O 3 catalyst.On the other hand, the least expensive catalysts are the REEs-containing catalyst Ni/La 2 O 3 and the Ni/ ZnO catalyst.In reviewing the H 2 selectivity, two catalysts give the highest selectivities to hydrogen in the reaction, namely the Ni/ZnO catalyst and the REEs-containing catalyst Ni/La 2 O respectively.In terms of ethanol conversion, Ni/Al 2 O 3 catalyst had a complete ethanol conversion to hydrogen, while Ni/Al 2 O 3 -CeO 2 catalyst ethanol conversion was 97%.The third best catalyst in ethanol conversion is Ni/ZnO.The catalyst with the lowest hydrogen selectivity is Ni/Al 2 O 3 -CeO 2 .Finally, in terms of the energy used, the catalyst that utilized the least amount of energy during the reaction is the REEs-containing Ni/Al 2 O 3 -CeO 2 followed by Ni/Al 2 O 3 .On the other hand, the catalysts that consumed

Fig. 5 .
Fig. 5. Analytic Hierarchy Process decision hierarchical structure for the sustainability assessment of ethanol steam reforming catalysts.

Fig. 7 .
Fig. 7. Sensitivity analysis results under the variation of indicators weight.(a) equal weights; (b) 5% decrease in selectivity and the waste gas emissions indicators weights and the rest of the indicators weights are increased proportionally to keep the sum of the weights to 100%; (c) 5% increase in selectivity and the waste gas emissions indicators weights and the rest of the indicators weights are increased proportionally to keep the sum of the weights to 100% (d) 10% decrease in selectivity and the waste gas emissions indicators weights and the rest of the indicators weights are increased proportionally to keep the sum of the weights to 100% (e) 10% increase in selectivity and the waste gas emissions indicators weights and the rest of the indicators weights are increased proportionally to keep the sum of the weights to 100%the rest of the indicators weights are decreased proportionally to keep the sum of the weights to 100%.(f) weights variation of decision making expert answers.

Table 1
Data sources used in this study.
Design of products, processes, and systems must include integration and interconnectivity with available energy and materials flowsNot applicablePrinciple 11: Design for commercial "afterlife"Products, processes, and systems should be designed for performance in commercial "afterlife"Catalyst regenerationPrinciple 12: Renewable rather than depleting Materials and energy inputs should be renewable rather than depleting Renewability

Table 5
Description of sustainability assessment indicators and the assessment methods used.

Table 6
Global warming potential for time horizon 100 years (GWP 100 ).

Table 8
AHP pairwise comparison matrix with intensity and weights for second level indicators.

Table 9
Normalized pairwise comparison matrix and local weights for third level indicators.

Table 12
AHP decision makers and their field of expertise.
L.Omodara, E.M. Turpeinen, S. Pitkäaho et al.Sustainable Materials and Technologies 26 (2020) e00237 the most amount of energy during the reaction are Ni/La 2 O 3 and Ni/ZnO respectively.In terms of health and safety considerations, the three catalysts with the highest score are Ni/Al 2 O 3 and the REEs-containing catalyst Ni/ La 2 O 3 and Ni/Al 2 O 3 -CeO 2 .All three catalysts had equal scores of 25.41.On the other hand, the catalyst with the lowest health and safety indicators score is the Ni/ZnO with a score of 22.77.In terms of the total environmental score, the catalyst with the highest score is the Ni/Al 2 O 3 catalyst with a score of 24.00, followed by Ni/La 2 O 3 with the second-highest score of 15.39.Ni/Al 2 O 3 is ranked as the best performing catalyst based on the environmental indicator because it had the least amount of waste gas emissions generated .The catalyst with the lowest score is the Ni/ZnO with a score of 8.00.The overall indicator score for all the catalysts was determined to find out the most sustainable catalyst cutting across the economic, health, and safety as well as environmental indicators.Overall, the highest to the lowest overall sustainability score for all the assessed catalysts are Ni/Al 2 O 3 (66.88),Ni/ZnO (54.27),Ni/La 2 O 3 (51.54),and Ni/Al 2 O 3 -CeO 2