Pursuing environmental sustainability in the vehicle painting process in view of compliance with formaldehyde emission limits

UNIVERSIDADE DE SANTIAGO DE COMPOSTELA ESCOLA TECNICA SUPERIOR ENXEÑARIA MASTER ENVIRONMENTAL ENGINEERING Written by: Aída Isabel García Muñoz Under supervision of: Daniel Granadero Rey and Dr. Gumersindo Feijoó Costa 28 Julio 2017 CERTIFICADO Prof. Dr. Gumersindo Feijoo Costa, Catedrático de Ingeniería Química de la Universidad de Santiago de Compostela; y Daniel Granadero Rey, Project Leader responsable de emisiones y permisos en la empresa de estudio, certifican que: Aida Isabel Garcia Muñoz ha realizado bajo su dirección el presente Trabajo de Fin de Máster bajo compromiso de confidencialidad, con título: “Pursuing environmental sustainability in the vehicle painting process in view of compliance with formaldehyde emission limits”. Y para que así conste, firmamos el presente informe en Santiago de Compostela a 28 de Julio de 2017. Fdo: G. Feijoo Costa Fdo: D. Granadero Rey Fdo: A.I. Garcia Muñoz CONFIDENTIALITY NOTE This Master Thesis contains confidential information and data that belong to the company where the project was developed. Because of that some specific information from the company was delete for this public version. Any publication or duplication of this work is not allowed. Only the Professors of the University correcting the work and the evaluation committee may have access to this work. Este Proyecto TFM contiene información confidencial y datos que pertenecen a la empresa donde se desarrolló el proyecto. Debido a eso, la información específica de la empresa ha sido borrada para esta versión pública. No se permite ninguna publicación o duplicación de este trabajo. ACKNOWLEDGMENT Quisiera agradecer a la empresa que me ha acogido durante los últimos 9 meses para la realización de este proyecto. Muchas gracias por la oportunidad que me brinda el haber trabajado para ustedes. En especial quisiera agradecer a mi departamento y todos los ingenieros que me han ayudado en todos los aspectos posibles, tanto en el desarrollo del proyecto como en el día a día. A Daniel, mi supervisor parte de la empresa, por su paciencia infinita y todo el tiempo dedicado al desarrollo y corrección de este proyecto. Sin olvidar al tutor por parte de la universidad, Gumersindo, por la disponibilidad para las reuniones y la dirección de este trabajo. Mención también a Isabel por su ayuda con el SimaPro, sin la cual no podría haber obtenido los ninguno de los resultados aquí discutidos. A los viejos amigos por apoyarme en este nuevo camino, por los consejos y horas de Skype que no me han dejado sentirme sola en ningún momento. A los nuevos amigos por haber enriquecido esta experiencia. Gracias por haberme acogido con tanto cariño y haber compartido conmigo vuestra cultura y vuestro tiempo. A toda mi familia por el apoyo incondicional, por creer en mí en todo momento y por haberme enseñado a luchar y a vivir. No podría estar más orgullosa de la familia que tengo. Especial mención a mi familia gallega sin los cuales este Máster y esta oportunidad no habrían sido posibles. Gracias por acogerme en vuestra casa y tratarme como a una más. A todos ellos gracias por acompañarme y arroparme durante este camino y por continuar haciéndolo en las próximas aventuras que me depare la vida. ABSTRACT The environmental impacts associated with the vehicle painting process of a European plant were investigated based on a life cycle assessment (LCA) gate to gate study. Due the new hazardous classification of Formaldehyde, the studied plant carried out several measurements of formaldehyde emissions in order to detect and study those areas where the emissions were higher. As a result, some improvement actions were implemented and proposed to reduce these emissions. The three main processes within the painting operations were studied. It was found that the higher contribution to the total impact is due to the Electrocoating and Top coat processes, being the Electrocoating materials and natural gas consumption the main cause. Moreover, four different improvement Scenarios were studied for the company to reduce formaldehyde/VOC emissions and as a result, the impact on the human toxicity. LCA comparative study was performed for all of these Scenarios. The Scenario A represents the current situation of the plant with some improvements already implemented that implies an increase of the natural consumption, thus increasing the total impact. Scenarios B and C represent different abatement equipment possibilities to reduce air emissions on Top cat ovens. An environmental impact and economic analysis was developed. Scenario C was found the most environmental sustainable option because it includes a reduction on the natural gas consumption. However, both Scenarios achieve the formaldehyde emissions reduction and, as a consequence, the human toxicity impact could be reduced. Due to the fact that Scenario B is more economically favourable, the company decided this option for the air emissions reduction. The last Scenario studied was a future improvement over the Scenario B. It was seen that this improvement would not improve the situation. Natural gas consumption increases the total impact despite of the reduction due to the VOC abatement. In all cases natural gas consumption plays an important role at the whole plant and all the Scenarios studied increase their total impact due to that. Despite of this, as it was pursued, in all Scenarios there is a considerable reduction on the human toxicity impact. INDEX 1. INTRODUCTION .................................................................................................................... 11 1.1 Vehicle manufacturing process ............................................................................................ 12 1.2 Painting process .................................................................................................................... 13 1.2.1 Pre-treatment. ............................................................................................................... 14 1.2.2 Electrocoating (ELPO-coat). ........................................................................................ 15 1.2.3 Under body sealant.......................................................................................................15 1.2.4 Primer coat. .................................................................................................................. 16 1.2.5 Top coat (Base-coat + Clear-coat). .............................................................................. 16 1.3 Paints .................................................................................................................................... 17 1.4 Formaldehyde. ...................................................................................................................... 18 1.4.1 Formaldehyde emissions legislation. ............................................................................ 19 2. GOAL AND SCOPE OF THE STUDY .................................................................................. 20 2.1 Goal of the study .................................................................................................................. 21 2.2. Scope of the study ............................................................................................................... 21 2.2.1 Process description. Case of study ................................................................................ 21 2.2.2 Functional unit. .............................................................................................................. 21 2.2.3 System Boundaries. ....................................................................................................... 21 3. DESCRIPTION OF IMPROVEMENT SCENARIOS ......................................................... 23 3.1 Chronology ........................................................................................................................... 23 3.2 Scenario A ............................................................................................................................ 24 3.3 Improvement Scenario B and C. ........................................................................................... 25 3.3.1 Improvement Scenario B. .............................................................................................. 26 3.3.2 Improvement Scenario C. .............................................................................................. 27 3.4 Improvement Scenario D. ..................................................................................................... 27 4. METHODOLOGY ................................................................................................................... 29 4.1 Impact assessment. ............................................................................................................... 30 4.1.1 Characterisation. ............................................................................................................ 31 4.1.2 Normalisation. ................................................................................................................ 31 5. LIFE CYCLE INVENTORY .................................................................................................. 32 5.1 System definition and Data quality ...................................................................................... 32 5.2 Types and sources of data. .................................................................................................... 33 5.3 Inventory ............................................................................................................................... 34 6. RESULTS .................................................................................................................................. 36 6.1 Painting process Base case analysis ..................................................................................... 36 6.1.1 Elpo process ................................................................................................................... 37 6.1.2 Primer process. ............................................................................................................... 39 6.1.3 Top coat process. ........................................................................................................... 40 6.2 Scenario A. Current situation ............................................................................................... 41 6.3 Improvement Scenarios at Top coat ovens. Scenarios B and C ........................................... 42 6.3.1 Economical analysis ....................................................................................................... 45 6.4 Scenario D. Future improvement Scenario. .......................................................................... 47 6.5 Painting process comparative results. ................................................................................... 48 7. CONCLUSIONS. ...................................................................................................................... 51 8. BIBLIOGRAPHY .................................................................................................................... 53 ANNEX .......................................................................................................................................... 57 LIST OF FIGURES Fig. 1. Vehicle manufacturing process……………………………………………………….......12 Fig. 2. Example of the main painting processes in a vehicle manufacturing painting plant….......14 Fig. 3. Multilayer coating of cars………………………………………………………………....17 Fig. 4. Process summary and system boundary………………………………………………......22 Fig. 5. Timeline of the different improvement Scenarios to comply with new emission limits....23 Fig. 6. Trend line of CO, total carbon and NOx according to the temperature of incineration inside the abatement installation. Adapted data from VDI.............................................................25 Fig. 7. RTO abatement scheme.......................................................................................................27 Fig. 8. TAR abatement scheme.......................................................................................................28 Fig. 9. System definition and data quality......................................................................................32 Fig. 10. Characterisation of the system “Painting process Base case”...........................................36 Fig. 11. Characterisation of the subsystem (SS1) “ELPO process Base case”...............................37 Fig. 12. Characterisation of the subsystem (SS2) “Primer process Base case”..............................39 Fig. 13. Characterisation of the subsystem (SS3) “Top Coat process Base case”..........................40 Fig. 14. Human toxicity Characterisation comparison between both Scenarios............................42 Fig. 15. Comparison of Scenarios B and C versus the Base case from the human toxicity point of view. Characterisation....................................................................................................................44 Fig. 16. Eco-efficiency analysis. Scenarios B and C Versus Scenario A taken into account the impacts directly related to formaldehyde emission........................................................................45 Fig. 17. Eco-efficiency analysis. Scenarios B and C Versus Scenario A from the human toxicity impact point of view.......................................................................................................................46 Fig. 18. Total normalisation results. Comparison between the Scenarios B and D.......................47 Fig. 19. Normalisation data of the system Painting process evaluating the impacts with a weight higher than 2%................................................................................................................................48 Fig. 20. Human toxicity normalisation data for all the Scenarios studied......................................49 LIST OF TABLES Table 1. Percentages of variation for each impact for the Scenarios B and C with respect to the current situation of the plant (Scenario A) impact at Top coat process………………………….43 1. INTRODUCTION 17 Under the painting booths in Primer, base coat and clear coat processes there are water scrubbers to collect the remaining paint. This wastewater mixed with waste paints is taken to the same water tank that contains coagulants in order to separate the water from the paints. This way, the water can be recirculated and used at the same system and the paint can be collected for its disposal. 1.3 Paints As it was explained before, the whole process needs three different kinds of paints for the three main painting steps of coat application on the BiW; Primer, Base coat and Clear coat. Those paints have essentially the same principal components and different thicknesses are required. The next figure shows the different painting layers in a vehicle and its required thickness. Fig. 3. Multilayer coating of cars [13]. The main components of the typical paints are [15-17]:  Pigments: To give colour, opacity and usually to improve the resistance against the corrosion. The clear coat paint is the only one without pigment. 1. INTRODUCTION 18  Additives: To improve aspect, biological properties, conservation, transports, etc. and to prevent defects in the coating like bubbles, poor levelling or flocculation. Different additives are used depending on the requirements of the paints.  Binders or resins: Provide adherence and resistance against mechanical and chemical strain. Amino resins derived from melamine are the most used resins in the automotive industry. These resins are made by a reaction between melamine, formaldehyde (CH2=O) and alcohols (ROH).  Fluidising medium. Depending on the chosen fluidising medium the paints could be waterborne or solvent borne. Additionally, there is another kind of paints based on a powder system but this is not commonly used in Europe. All of them contain different quantities of solvents in their composition. It is important to determine the operational conditions at the plant depending on the kind of paints used at each painting step. 1.4 Formaldehyde Formaldehyde is a simple organic volatile compound, which chemical formula is H-CHO. All organic life forms produce formaldehyde that is emitted to the environment. Moreover, it has a wide range of applications so it is also manufactured industrially. In the last years, almost 9% of the total formaldehyde production in Europe was melamine Formaldehyde resins. These resins are fast-curing and can resist high temperatures and chemical damages so that, they are usually used in the composition of surface coatings of vehicles [18-20]. Through any vehicle manufacturing painting processes, formaldehyde can be released in two different ways. Firstly, a small percentage of free formaldehyde is present in the paints and therefore it can be emitted to the atmosphere through the exhaust air coming directly from the paint booths. And secondly, during the paint curing process at high temperatures the melamineformaldehyde structure is broken, releasing formaldehyde which will be emitted to the atmosphere if no abatement equipment is installed or if the abatement system is not able to completely destroy the formaldehyde [14]. 1. INTRODUCTION 19 1.4.1 Formaldehyde emissions legislation. The Regulation (EC) No 1272/2008 of the European Parliament and of the Council on classification, labelling and packaging of substances and mixtures [22] (CLP Regulation) has the main purpose to ensure a high level of protection of human health and the environment by harmonising the criteria for classification of substances and mixtures, and the rules on labelling and packaging for hazardous substances and mixtures. To reflect the adaptions to technical and scientific progress, this regulation needs to be constantly revised and corrected, thus the classification of chemicals is regularly updated and new substances can be reclassified as hazardous. Besides, the Directive 2010/75/EU on industrial emissions [9] establishes the rules of integrated pollution prevention and control from industrial activities and is applicable to the industrial processes associated with vehicle assembly painting plants. This Directive sets emission limit values for carcinogens, mutagens, or substances toxic to reproduction. With one of the latest Adaptions to Technical Progress of the CLP Regulation in 2015 [8], formaldehyde has been reclassified to receive the Hazard Statements H350, may cause cancer, and H341, suspected of causing genetic defects. With this higher cancer-causing classification, the Industrial Emissions Directive is immediately applicable and compliance with the new applicable emission limit value for formaldehyde emissions must be guaranteed, forcing all European vehicle manufacturers to evaluate their legal situation and to implement technical and operational countermeasures to reduce or eliminate formaldehyde emissions. 2. GOAL AND SCOPE OF THE STUDY 20 2. GOAL AND SCOPE OF THE STUDY 2.1 Goal of the study Due to the new formaldehyde classification, as it was explained before in this work, the company performed formaldehyde measurements in all its panting plants across Europe. Some modifications were implemented by the paint manufacturers in the paints composition that helped to reduce the formaldehyde emissions from the painting booths. Additionally, technical actions were already taken at the painting installation to reduce those emissions from the Primer coat process. Also, it is being studying different abatement technologies have been evaluated for its installation at the Top coat process to eliminate formaldehyde emissions. This project is based on the LCA of those improvements in order to analyse their environmental impacts comparing the results with the situation of the plant on2015. The main goals of this project are:  Detect the hot spots from the main painting processes that cause the highest environmental impacts.  Propose possible future actions to reduce the impacts based on the hot spots identified.  Study the environmental impacts of the different corrective actions that are already implemented at the plant to reduce the formaldehyde emissions as a single Scenario in order to be able to compare the environmental situation before and after those actions.  Comparative study of the potential environmental impacts of the two possibilities that are being studied by the company to reduce the formaldehyde emissions at the Top coat ovens.  Economic comparative study between the two different alternatives for abatement technologies. Discussion between the results and the real decision of the company.  Study of another additional improvement Scenario. Future possible application.  Comparative analysis of the different Scenarios of study from the direct emissions and human toxicity point of view to evaluate if the target of the Industrial Emissions Directive is achieved through those actions. 2. GOAL AND SCOPE OF THE STUDY 21 2.2. Scope of the study 2.2.1 Process description. Case of study This work is focused on the three most important processes within the coating operations because they were identified as the most relevant from the formaldehyde emissions viewpoint. The indicated processes are electrodeposition (ELPO), Primer and Top coat. The detailed scheme of the material and energetic balance of the three important steps of the painting process taken into consideration in this study cannot be included on this public version due to the confidentiality agreement. *Working conditions of the plant cannot be shown in this public version due to the confidentiality agreement* 2.2.2 Functional unit The functional unit provides a reference to which the inputs and outputs are related [23-24]. The functional unit chosen was 1hour of operation. Thus, all collected data and calculations are referred to one hour of operation in 2015. 2.2.3 System Boundaries The system boundaries determine which unit process is going to be included within the LCA. The selection of the system boundary has to be done in accordance with the goal of the study [23-24]. The following scheme shows a summary of all painting processes. Those processes inside the red square are the ones included in the studied system. 2. GOAL AND SCOPE OF THE STUDY 22 Fig. 4. Process summary and system boundary. *Assumptions made for the development of the project are strongly related with the working procedures of the plant, therefore cannot be shown in this public version due to the confidentiality agreement* Cationic paste Elpo binder Paint Base Coat Paint CC lacquer Solvent Purge solvent loop BC Paint PRETREATMEN PVCSealing Degreasin Phosphatin Passivation SEALING Cleaning Rinsing Oven Abateme Electrodeposition bath SS1. ELPO SS2. PRIMER Primer booth Flash off Oven Abateme SS3. TOP COAT Oven Flash off ClearCoat Intermediate Flash off Base Coat Abateme REPAIR Repair Inspection Painted car Natural Gas Energy Water HZ Solid wastes Paint waste (incineration/ separation) Water emissions Air emissions Boundaries 3. DESCRIPTION OF IMPROVEMENT SCENARIOS 23 3. DESCRIPTION OF IMPROVEMENT SCENARIOS In order to analyse the environmental impact of the vehicle painting process under different improvement modifications that were introduced or are currently being implemented at the studied painting plant, different Scenarios were studied. 3.1 Chronology The next figure shows the implementation timeline of the different Scenarios studied in this project for the plant of study. Fig. 5. Timeline of the different improvement Scenarios to comply with new emission limits. The starting Scenario was defined as “Basic case” and represents the painting plant as it was operated in 2015 when the first measurements were performed. In 2016, the first set of corrective actions was implemented and it was included in the first studied Scenario, “Scenario A”. This consists of a reduction of the amount of free formaldehyde in the paints carried out by paint manufacturers and a decrease of the emissions from the Primer coat process by introducing technical and operating improvements. After that, the company efforts were focused on reducing the emissions from the Top coat process. For that, the company is currently installing a regenerative thermal oxidiser for the exhaust air of the Top coat ovens projected to be finished in 2018. This was analysed as “Scenario B”. This comparison was made to determine 3. DESCRIPTION OF IMPROVEMENT SCENARIOS 24 which of both alternative Scenarios would have less environmental impact. Furthermore, a possible future improvement Scenario was analysed as “Scenario D”. It is important to bear in mind that only the environmental impacts caused by the painting process working under these new modifications were analysed and not the impact of the implementation itself. 3.2 Scenario A The first Scenario includes different corrective actions already implemented at the plant. Because of the new hazardous classification of formaldehyde, the paint manufacturers improved their production process to reduce the amounts of free formaldehyde in their paints. In this Scenario, Primer and Top coat paints have reduced amounts of free formaldehyde in its composition. The emissions from the flash off area were redirected to the oven to be treated commonly. It was increase the temperature of the incinerators at the Primer ovens in order to improve their abatement capacity.* Detailed explanation of this implementation is not shown in order to protect the confidentiality of the company* The next chart shows the trend line of CO, total carbon and NOx emissions in relation with the temperature of the incinerator. 3. DESCRIPTION OF IMPROVEMENT SCENARIOS 25 Fig.6. Trend line of CO, total carbon and NOx according to the temperature of incineration inside the abatement installation. Adapted data from VDI [14]. CO and total carbon follow the same trend, that is, the higher the temperature, the lower the amount emitted of these pollutants. However, NOx has the opposite behaviour, it increases with the temperature. It is necessary to achieve an equilibrium to have the lowest concentrations possible of all pollutants. The green square shows the most ideal situation, where the amounts of all parameters, CO, C total and NOx, remain in their lowest values, so the ideal working temperature has to be within this range. 3.3 Improvement Scenario B and C To decrease formaldehyde levels at the Top coat ovens, the option selected by the company was the addition of abatement systems at this oven. The abatement equipment commonly used at vehicle painting plants is a thermal oxidiser. Oxidisers can recycle the 95% of the heat excess and use 20 times less heat requirement for the burner than other kind of abatement. There are two different oxidisers used by the vehicle manufacturing companies: Regenerative (RTO) and Recuperative (TAR) Thermal Oxidisers [14, 25]. 3. DESCRIPTION OF IMPROVEMENT SCENARIOS 26 Although the company already decided to install RTO abatement system at the plant to reduce the emissions at Top coat ovens, this project reflects a comparison between these two alternatives from the environmental impact and economical point of view. Calculations were made by the company to estimate the number of systems for each oxidiser studied needed to ensure the required treatment of the emissions. Also, an evaluation of the energy demanded for each type was made by the company. 3.3.1 Improvement Scenario B A Regenerative Thermal Oxidiser (RTO) is a regenerative pot-combustion emission control system that uses a flame to oxidise the organic compounds at optimal temperatures around 800-850°C. The raw gas enters to the system and is heated by a ceramic that covers the device. This raw gas is burned and the final clean gas is cooled using a mixed-bed as a heat exchange. The heat is stored at the ceramic materials allowing to reuse it to heat the incoming raw gas, which reduces the amount of fuel needed. The destruction efficiency of this kind of abatement is between the 95-98%. This system is independent to the whole painting process, meaning, if the abatement malfunction suffers any malfunction, it is not necessary to stop the production [14]. The next figure explains the different parts of a common Regenerative Thermal Oxidiser. 5. LIFE CYCLE INVENTORY 33 After that, all inventory data were transferred to the chosen software to analyse the environmental impacts affected for the different processes. It is important to bear in mind that some iterations with the software were needed in order to achieve the highest accuracy as possible. When the hot spots are detected, it is necessary to ensure that the source of the related data is reliable enough to obtain results as solid and certain as possible. 5.2 Types and sources of data Data selected for an LCA depends on the goal and scope of the study. For the development of this project, most of the data were collected from the production site associated with the unit within the system boundary. Besides, other data from alternative sources as books or papers, other similar plants, internal communications or additional information were needed in order to complete all the required information to elaborate the processes´ inventories. All data collected in this project include a combination of measured, calculated and estimated data. In this chapter, all data inventories are shown indicating the source of each figure and, when required, how it was calculated. Data were classified according to their level of reliability. 1. Data from the plant. As far as possible, real data from the studied plant were used. Different reports with relevant information were checked, as well as reports with real air emissions measurements made at the plant. The location of the studied plant is Europe and all the data compiled for the inventories of the base case Scenario were from 2015. 2. VOC and material balances from the plant. To determine the VOC emissions and the materials used at the plant, data were taken from the VOC emission balance report that the plant engineers had developed with real data from 2015 and 2016. 3. Safety Data Sheets from the materials (REACH) and other information from the material´s suppliers. This information came from external companies. 4. Calculations based on data given by the plant. Assumptions and calculations were made to have a good allocation of some data using information of the material and energy consumption. 5. LIFE CYCLE INVENTORY 34 5. Internal communications. To understand the whole system, several meetings were needed with different engineers and experts on the paints used at the plant, the processes and water and energy management. Despite of these data not being reflected in any report from the plant, the information from this source was considered reliable enough to be taken for the purpose of this project. 6. Data from a different plant of the same company with similar processes and characteristics. Some data were not available from the studied plant, therefore reports from other similar plants of the company were taken. 7. Bibliographic information. Books, papers, legislation and BREF reports were reviewed to understand the common vehicle painting processes and to verify and allocate the data to the studied processes. 8. Other calculations. It is referred to estimations made through different sources or following simple mathematical rules. Most of the data were compared with different sources to ensure its veracity, especially when those data were obtained as a percentage or an interval value. 5.3 Inventory As it was mentioned previously, the functional unit chosen was 1 hour of operation. According to that, all the data exposed below are referred to that unit. It was needed to calculate the production hours during a year in order to have all the data in the same units. This calculation was made through the assembly production calendar of 2015 given by the plant. The detailed inventories of inputs and outputs following explained define the initial situation of the plant as it was working in 2015. This Scenario was called in this project “Base case”. For the analysis of the different Scenarios, modifications in the natural gas consumption, paint composition and in the air emissions were needed. *This project was elaborated under a strict confidentiality agreement. Because of that, the inventories used for each process are not available in this public version*. 5. LIFE CYCLE INVENTORY 35 5.3.1 Inventory for Scenarios To analyse the impact of the different studied Scenarios, several modifications in the inventories were required. *The table that compiles those changes has been delated due to the confidentiality agreement*. The following changes were made on the different Scenarios:  Paints. Scenario A contemplates a change on the composition of all the paints used at the plant. The amount of free formaldehyde in the paints was reduced to a value below 0.1% so the suppliers do not need to label their products as carcinogenic as CLP Regulation reclaims. It was not possible to obtain the real amount of free formaldehyde that is currently present in the paints. Thus, the higher amount possible was assumed, i.e. 0.099%. This new paints´ composition remains in all the different Scenarios.  Natural Gas variation. The variation of the natural gas for the different Scenarios were calculated following some information from the engineers of the plant.  VOC and formaldehyde emissions were estimated or calculated by VOC mass balance simulating the different characteristics for each one.  NOx and CO. For the improvement Scenarios that are not implemented at the plant yet, air emissions data related to the abatement installations (NOx and CO) were not available. Data from other plants with similar characteristics were taken and adapted to the studied plant.  Economic analysis. Additionally, for Scenarios B and C, where two different abatement systems for the Top coat ovens were proposed, an economic analysis was made. For that, was taken into account the lifetime of both abatements, costs of maintenance and installation and natural gas consumption. The prevision of the cost due to the natural gas consumption through the lifetime of the abatement alternatives was performed using the IPC data from the last ten years [34] and the current natural gas cost [35]. 6. RESULTS 36 6. RESULTS This chapter includes the Life Cycle Impact assessment and interpretation. The assessment is the phase of the LCA that involves the evaluation of the magnitude and significance of the potential environmental impacts for the system. The interpretation is the evaluation of the results related to the goals and scope of the project to reach conclusions and recommendations [23-24]. Once all data were collected in the inventories described in chapter 5 and inserted in the SimaPro software, results were obtained and analysed. This chapter is structured following the goals described in chapter 2. 6.1 Painting process base case analysis Firstly, the global system was analysed to understand which process contributes with the higher environmental impact to the whole system. Next chart shows the characterisation obtained from the data of the whole process in 2015, when the concerns about formaldehyde emission levels started and the first measurements took place. The components taken into account for this chart were the three subsystems from ELPO, Primer and Top coat processes and all inputs and outputs that could not be allocated to the different processes. Fig. 10. Characterisation of the system “Painting process base case”. 6. RESULTS 37 Top coat is the largest process with the highest material consumption and emissions. Therefore, as it was expected, it can be seen in the chart that Top coat process has the highest contribution to the environmental impact for most of the categories of impact. However, it has to be taken into account that the ELPO process plays an important role in all the impact categories being the highest contribution in the categories of climatic exchange (CC), terrestrial acidification (TA), particular matter formation (PMF), water depletion and fossil depletion (FD) essentially because of the materials used in this process. In the following subchapters, the results from the analysis of each process are shown in detail to determine the components that constitute the potential cause for the impacts. Some actions were also proposed to be studied and implemented in the future to decrease those impacts following the indications from the STS BREF. 6.1.1 ELPO process The first subsystem within the boundaries of the system is the ELPO process. The figure below represents the characterisation of this process obtained through the software using the inventory data explained in chapter 5. Fig.11. Characterisation of the subsystem (SS1) “ELPO process base case”. 6. RESULTS 38 The ELPO cationic paste has the most important contribution to the total environmental impact from this process. ELPO cationic paste and binder were not found in the Ecoinvent database on the software. Therefore, those materials were also introduced as a subsystem inside the ELPO process. For that, the composition of the materials was analysed. Due to the high impact related to ELPO cationic paste in most of the impact categories, its characterisation was carried out. In this characterisation, it was observed that the use of epoxy resins on its composition is the main cause of the impact essentially because of the CO2 emissions during its production. According to this result, a future recommendation is the research and development of different resins that can be used for the production of ELPO materials with a lower environmental impact by the paint producers, or a modification in their production process leading to a reduction in the CO2 emissions. It is also important to highlight the high contribution on freshwater and marine ecotoxicity of the wind energy. Data for wind electricity was taken from the Ecoinvent using the “Electricity, high voltage {ES}| electricity production, wind, >3MW turbine, onshore | Alloc Def, U” parameter after being checked that this is the most suitable to the case of study. Analysing the data and the characterisation of this type of energy, it was concluded that this ecotoxicity character is a consequence of the use of copper on the construction of the wind turbines. Copper is a heavy metal that affects essentially the water ecosystems. We can find those unusual values from the wind energy in each of the processes. The supplier of the energy for the plant is an external company and it is not related exclusively with vehicle manufacturers. For this reason, an improvement to the plant in order to reduce these impacts cannot be suggested. With regard to the direct emissions from this process, the results indicate that they are not relevant in any of the impact categories. For this reason, the following improvement Scenarios studied at this project are not affecting this process. 6. RESULTS 39 6.1.2 Primer process After ELPO process, the following step studied at the painting plant was the Primer process. The results of the characterisation of this process are shown in the next chart. Fig. 12. Characterisation of the subsystem (SS2) “Primer process base case”. Most of the impact categories are highly affected by the natural gas consumption. This impact is associated with its extraction and use. Currently, there is no feasible alternative to the natural gas used as a fuel to feed the burners for the abatement equipment, so that the impact related to that cannot be reduced. The percentage of the renewable energy from each process was taken from the Spanish electrical mix, so that percentage is the same on all processes studied on this project. As it was observed for the ELPO process, wind electricity represents a high contribution to the impact in certain categories. In this process, the direct emissions have the main contribution to the marine eutrophication and photochemical oxidant formation impacts due to the NOx emissions. Although the direct emissions are not the principal cause of the toxicity categories, it also plays an important role on them. The first Scenario studied in the section 6.2 contemplates an improvement on this process in order to reduce the air emissions and, as a consequence, the impact related to human toxicity. 6. RESULTS 40 6.1.3 Top coat process Finally, the last painting step was analysed. Top coat includes Base coat and clear coat processes, being one of the most complex within the system. As in the previous processes, the results of the characterisation of this step are shown in the graph below. Fig.13. Characterisation of the subsystem (SS3) “Top coat process base case”. To analyse this process, the elaboration of three subsystems for the paints was needed. These subsystems were “Base coat paint”, “white paint” and “clear coat lacquer”. However, to make the chart more visual, the impact of these three paints was analysed together as “Total paints”. As it was observed on the characterisation of the paints on ELPO process, the three different paints used at Top coat have high levels of epoxy resin on their composition which is the main cause of the impact related to the paints. Marine eutrophication, terrestrial acidification, photochemical oxidant formation and particular matter formation impact categories are the most affected by the direct emissions. Almost the whole photochemical oxidant formation impact is due to direct emissions, mainly because of the NOx emissions at the exit of the incinerators. As in Primer process, direct emissions are not the main cause of the human toxicity, nevertheless as a consequence of the new classification of formaldehyde, the Scenarios B, C and D studied 6. RESULTS 41 different technical improvements to reduce the formaldehyde emissions and therefore the human toxicity. 6.2 Scenario A. Current situation As explained in previous chapters, Scenario A considers the current situation of the paining plant after the implementation of the following modifications:  Less formaldehyde in all the paints (Primer, Base coat and clear coat). In 2015 the amount of formaldehyde at the paints was around 0.3% as average and nowadays it contains less than 0.1%.  Increase of the temperature at the Primer incinerators.  Redirection of the flash off emissions to the Primer oven. The change on the paints´ composition made by the paint producers slightly decreases all impact categories at Top coat process, being the highest decrease in urban land occupation decreasing the impact by 0.04%. The main reason of this change on the composition was the new formaldehyde classification, therefore human toxicity was separately analysed. Human toxicity at Top coat decreases its impact by only 0.025%. This Scenario is fundamentally focused on the reduction of the formaldehyde emissions at Primer process with the aim at reducing the impact at the toxicity categories. Increasing the temperature at the incinerators involves an important increase of the natural gas consumption. As a consequence of the higher natural gas consumption, all impact categories dramatically increase its impact. However, as it can be observed in the next chart, human toxicity decreases its impact considerably which was the target of the implemented actions included in Scenario A. 6. RESULTS 42 Fig.14. Human toxicity Characterisation comparison between both Scenarios. The graph shows a comparison between the results of the characterisation for the Human toxicity category between both Scenarios. Human toxicity impact was reduced by 7% with the implemented actions included in Scenario A. Besides, terrestrial ecotoxicity impact is decreased by 4% with these countermeasures. 6.3 Improvement Scenarios at Top coat ovens. Scenarios B and C In this section, a comparison between the two different abatement technologies that were studied by the company to be implemented for the Top coat process was performed. As it was explained before, the company is currently installing an RTO at Top coat process to treat the air emissions from the ovens. This implementation is projected to be finished in 2018. In this work, this improvement was studied under Scenario B. Nevertheless, another alternative of abatement was also analysed. Scenario C reflects the installation of several TAR instead of an RTO at the Top coat ovens. One of the main advantages of this Scenario is that the total natural gas consumption is reduced because the heating excess can be used to heat the ovens, and therefore the burners for the ovens are no longer necessary. Next Table collects the reduction or increase on the different environmental impact categories percentage related to the Scenario A for the Top coat process. 6. RESULTS 49 increase of the total impact on Scenarios B (2%) and a decrease due to Scenario C (-1%). Finally Scenario D involves an increase by 1% compared with Scenario B. As it was explained before these increases or decreases in all Scenarios are strongly linked to the natural gas consumption. The impact is being reduced in some categories as human toxicity or photochemical oxidant formation due to the abatement of air emissions but other impacts as terrestrial acidification or natural land transformation are increased because of the natural gas consumption. As a result these variations reach an equilibrium being the impact practically the same as the initial situation. Furthermore, focusing on the target of study, reduction on formaldehyde emissions, impact categories related with formaldehyde were taken into account for the next criterion. The same tendency was observed as it is shown in the figure 19. Formaldehyde is strongly related to human toxicity, so that the results for this impact category were also analysed. Fig.20. Human toxicity normalisation data for all the Scenarios studied. One of the main goals of the study of the different Scenarios was to achieve a reduction on the Human toxicity impact through the formaldehyde air emissions reduction. In this last chart, is could be seen that the target is achieved in all Scenarios comparing them with the base case Scenario. 6. RESULTS 50 At the chat it can be seen that the current situation of the plant (Scenario A) accomplishes a reduction on the normalised index of human toxicity by 1%. Scenarios B and C improve the situation with regard to the Scenario A by 1.54% and 5.26% respectively, being the latter the most environmental sustainable Scenario. Scenario D does not involve a great change with respect to Scenario B. In fact, the human toxicity normalised index is slightly increased by 0.07%. As it was observed on all the graphs, direct emissions are not the only component affecting human toxicity. Other inputs as the natural gas consumption and paints are related to this impact as well. Therefore, even though formaldehyde emissions can be almost eliminated due to these actions, the human toxicity impact cannot be completely reduced. 7. CONCLUSIONS 51 7. CONCLUSIONS Through the Life Cycle Assessment, the following conclusions were obtained based on the analysis made in this work.  Analysing the hot spots from the main painting processes, Electrocoating (ELPO) and Top coat processes cause the highest environmental impacts due to the use of the materials, natural gas and wind electricity.  A possible future action to reduce the impacts based on the hot spots identified would be the research and development of different resins that can be used for the production of ELPO materials with a lower environmental impact by the paint producers, or a modification in their production process leading to a reduction in the CO2 emissions.  Scenario A is focused on the formaldehyde emission reduction in the Primer process and therefore a reduction on the human toxicity impact. Comparing the results from this Scenario (current situation of the plant) and the situation in 2015 (Base case Scenario), it can be determined that there is a decrease on the human toxicity impact by 7% in this process so the target of this improvement was achieved. Besides, terrestrial ecotoxicity is reduced. However, due to the increase of the natural gas consumption there is an increase on the rest of the impact categories.  Scenarios B and C studied the installation of new abatement equipment for the Top coat ovens, RTO (Regenerative Thermal Oxidiser) or TAR (Recuperative Thermal Oxidiser) with energy recovery, to reduce VOC and formaldehyde emissions. Comparing both Scenarios with the current situation of the plant (Scenario A) at the Top coat process, it can be concluded that: o Natural gas plays an important role on all the impacts for Top coat process, what leads to an impact increase in Scenario B (4%), where the addition of new burners is needed. On the other hand, in Scenario C the natural gas consumption decreases and therefore the total impact is being reduced (2%). o Focusing the study on Human toxicity, a reduction is obtained with both Scenarios. Scenario B reduces human toxicity by 3% while Scenario C reduces it by 11%. 7. CONCLUSIONS 52 Although Scenario C has a higher reduction and therefore it is more environmental sustainable, both Scenarios achieve the target of the new formaldehyde classification. o From an economic point of view, Scenario B achieves the target of the company (formaldehyde and human toxicity reduction) and it is economically more favourable which explains the decision made by the plant of building an RTO.  The future application of Scenario D to reduce VOC emissions at Base coat booths involves a high natural gas consumption. As it was previously seen, this entails an increase in the total impact, specifically the impact would be increased by 2% having into account the Top coat process.  In the final comparison analysis performed for all different Scenarios, it has been observed that from the total impact and Human toxicity point of view the most environmental sustainable Scenario is the installation of incinerators (TAR) at Top coat process, i.e. Scenario C. Finally, with this work and the results obtained it was proved that Life cycle Assessment is a powerful tool to analyse different improvement alternatives that can be installed in a vehicle painting plant from the environmental sustainability viewpoint. 8. BIBLIOGRAPHY 53 8. BIBLIOGRAPHY [1] Geldermann J. and Rentz O. (2005) Decision Support through Mass and Energy Flow Management in the Vehicle- Refinishing Sector. Journal of Industrial Ecology 8 (4). 173-187. [2] STS BREF (2007). Reference Document on Best Available Techniques on Surface Treatment using Organic Solvents. European Commission. 6:163-186. [3] Rivera J.L. and Reyes-Carrillo T. (2016). A life cycle assessment framework for the evaluation of automobile paint shops. Journal of Cleaner Production 115. 75-87 [4] EEA (2008). Air pollution: Air pollution harms human health and the environment. European environmental Agency. Available at https://www.eea.europa.eu/themes/air/intro (Last access on 02/2017) [5] Heather L.M. and Lester B.L. (1998). A life-Cycle Model on automobile. Environmental Science & technology/news 322-330. [6] Hernandez M., Messagie1 M. Hegazy O., Marengo L. Winter O. and Van Mierlo J. (2017). Environmental impact of traction electric motors for electric vehicles applications. Int Journal Life Cycle Assess 22. 54–65. [7] Papasavva S., Kia S., Claya J. and Gunthert R. (2000). Characterization of automotive paints: an environmental impact analysis. Progress in Organic Coatings 43. 193–206. [8] Regulation (EC) No 2015/491 of the European Parliament and of the Council of 23 March de 2015 amending Regulation (EU) No 605/2014 amending, for the purposes of introducing hazard and precautionary statements in the Croatian language and its adaptation to technical and scientific progress, Regulation (EC) No 1272/2008 of the European Parliament and of the Council on classification, labelling and packaging of substances and mixtures. [9] Directive 2010/75/EU of the European Parliament and of the Council of 24 November 2010 on industrial emissions (integrated pollution prevention and control). Official Journal of the European Union. 8. BIBLIOGRAPHY 54 [10]* Confidential website* [11] Omar M.A. (2011). The automotive body manufacturing systems and process. WILEY. UK. 2-5:15- 233. [12] RESOLUCIÓN de 7 de enero de 2014, del Instituto Aragonés de Gestión Ambiental, por la que se actualiza la autorización ambiental integrada de la planta de fabricación de vehículos a motor, ubicada en el polígono Entrerríos, en los términos municipales de Figueruelas y Pedrola (Zaragoza), y promovida por General Motors España, S.L.U. (Número Expte. INAGA 500301/02/2012/2990). Boletín Oficial de Aragón. [13] Streitberg H.J. and Dössel K.F. (2008). Automotive paints and coats. 2° ed. VILEY-VCH. Verlag GmbH and Co. KGaA. Germany. 3-6: 61-210. [14] *Confidential report* [15] Stoye D. (1993) Paints, coatings and solvents. VCH. Germany. 1:3-7; 2: 54-84; 3:101-124; 11:243- 246 [16] Stoyer D. and Freitag W. (1996) Resins for coatings. HANSER. Germany. 6:103-118. [17] Wicks Z.W, Jones F.N., Pappas S.P., Wicks D.a. (2007). Organic coatings: Science and technology. 3° Ed. WILEY-INTERSCIENCE. New Jersey.8: 159-166; 10-11: 200-229. [18] Formacare. Formaldehyde information. Available at http://www.formacare.org/about-formaldehyde/ (Last access on 05/2017). [19] Bolt H.M., Johanson G., Nielsen G. D., Papameletiou D. and Klein C. L. (2016). SCOEL/REC/125 Formaldehyde. Recommendation from the Scientific Committee on Occupational Exposure Limits. European commission. [20] *Confidential book* [21] RAC (2012). Annex 1: Background document to the Opinion proposing harmonised classification and labelling at EU level of Formaldehyde. Committee for Risk Assessment. 8. BIBLIOGRAPHY 55 [22] Regulation (EC) No 1272/2008 of the European Parliament and of the Council of 16 December 2008 on classification, labelling and packaging of substances and mixtures, amending and repealing Directives 67/548/EEC and 1999/45/EC, and amending Regulation (EC) No 1907/2006. [23] UNE-EN ISO 14040: 2006. Environmental management. Life cycle assessment. Principles and framework. Ginebra: International Organisation Standardisation. [24] UNE-EN ISO 14044:2006. Environmental management. Life cycle assessment. Requirements and guidelines. Ginebra: International Organisation Standardisation. [25] Herbert F Lund (1971) Industrial Pollution Control Handbook. McGraw-Hill. USA. 7:1-10. [26] CMM. Regenerative and Recuperative Thermal oxidiser information. Available at http://www.thecmmgroup.com/ (Last access 05/2017). [27] Geldermann J. and Rentz O. (2005) Multi-criteria Analysis for Technique Assessment Case Study from Industrial Coating Journal of Industrial Ecology 9 (3). 127-142. [28] Goedkoop M., Heijungs R., Huijbregts M., De Schryver A., Struijs J. and van Zelm R. (2013) ReCiPe 2008 A life cycle impact assessment method which comprises harmonised category indicators at the midpoint and the endpoint level. Ministerie van VROM Rijnstraat. [29] ILCD (2010). Analysis of existing Environmental Impact Assessment methodologies for use in Life Cycle Assessment. Background document. European Commission. Joint Research Centre. Institute for Environment and Sustainability. [30] ILCD (2010). Framework and requirements for Life Cycle Impact Assessment models and indicators. European Commission. Joint Research Centre. Institute for Environment and Sustainability. [31] ENDESA (2015). Informe anual 2015. Documento resumen. [32] *Confidential website* 8. BIBLIOGRAPHY 56 [33] UNE-EN ISO 14045:2012 Environmental management. Eco-efficiency assessment of product systems. Principles, requirements and guidelines. Ginebra: International Organisation Standardisation. [34] INE. IPC data from the last ten years. Available at http://www.ine.es/jaxiT3/Tabla.htm?t=22344 (Last access 06/2017) [35] OMIP. Natural gas price 2018. Available at http://www.omip.pt/ (Last access 06/2017) ANNEX 57 ANNEX The original version contains 5 annex with detailed information from the plant of study. Those annex were: Annex 1: Flow diagram from the whole process Annex 2: Natural gas allocation for the base case. Annex 3: Natural gas variation for the different scenarios calculation. Annex 4: VOC balances. Annex 5: Economic analysis. *The content of these annexes is confidential, so they cannot be exposed on the public version* ANNEX 58