- Email: [email protected]
Potential alternative for water and energy savings in the automotive industry: case study for an Austrian automotive supplier
Journal of Cleaner Production 34 (2012) 146e152
Contents lists available at SciVerse ScienceDirect
Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro
Potential alternative for water and energy savings in the automotive industry: case study for an Austrian automotive supplier Peter Enderle*, Otto Nowak, Julia Kvas Resources e Institute for Water, Energy and Sustainability, Joanneum Research, Elisabethstrasse 16/I, 8010 Graz, Austria
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 July 2011 Received in revised form 10 November 2011 Accepted 12 November 2011 Available online 20 November 2011
This paper show the alternative possible optimization measures to increase the water and energy efficiency in the automotive industry. Within a potential study the technological system optimization of an automotive supplier was performed by combining the fields of process water reuse and heat recovery. Possible optimization measures to retrofit existing processes with the focus on high pressure die casting and component cleaning were worked out. Furthermore, limiting factors for a successful and extensive implementation in the case of retrofitting existing processes and systems were evaluated. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: Aluminium die casting Component cleaning System optimization Process water reuse Heat recovery
1. Introduction The automotive industry is one of Austria’s most important industrial sectors comprising the complete value chain from engineering services, metal working, metal processing, electronics manufacturing up to the production of die cast components (in particular lightweight components made out of aluminium and magnesium alloys). In contrast to other branches of industry, Original Equipment Manufacturers (OEMs) have outsourced large parts of their production and R&D activities to the supplier industry over the last decades. Thus, automotive suppliers are an integral part of automobile production, whereby their portfolio has changed over the years from the production of single components to complete systems and devices (e.g. driveline systems, engines, etc.), also including in-house R&D activities. Today, a modern automobile comprises up to 10,000 components, whereby nearly 80% of them are produced by external partners. Furthermore, beside automotive suppliers many other branches of industry are directly or indirectly linked to the automotive sector as an integral part of upstream and downstream production steps (e.g. chemical industry, textile industry, recycling industry, etc.). Therefore, economic and environmental challenges within the automobile sector have * Corresponding author. Tel.: þ43 316 876 2420; fax: þ43 316 8769 2420. E-mail addresses: [email protected] (P. Enderle), otto.nowak@ joanneum.at (O. Nowak), [email protected] (J. Kvas). 0959-6526/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.jclepro.2011.11.013
immediate impacts to the development of other industrial sectors (Orsato and Wells, 2007). A number of studies have been published in the recent years mainly focused on special issues regarding ecoinnovations, new technology developments and vehicle recycling (Christensen, 2011; Passarini et al., 2012; Alves et al., 2010; Zapata and Nieuwenhuis, 2010; Cui and Roven, 2010; Granata et al., 2009; van den Hoed, 2007; Duval and MacLean, 2007). These studies underline the dynamic change of market needs with an immediate impact on production and process needs within the automotive sector but also on other branches of industry. Although the automotive sector as a whole can not specified as water and energy intensive branch of industry and considerable improvements to reduce the environmental impact of car production have been achieved within the last decades, there are still certain production areas and production steps with a high potential to increase the resource efficiency, also including good housekeeping practices. Good housekeeping is focused on measures with low investment requirements, e.g. in order to minimize material losses, to reduce the consumption of process chemicals, to prevent unnecessary waste generation (Telukdarie et al., 2006) but also to raise employees’ environmental awareness so that employees adopt a positive attitude to pollution prevention (Kjaerheim, 2005). Regarding the automotive industry, the production areas of die casting, mechanical processing, paint finishing and hardening can be seen as those areas with a high potential of process water reuse and heat recovery measures. The more often applied die casting
P. Enderle et al. / Journal of Cleaner Production 34 (2012) 146e152
process within the automotive sector is High Pressure Die Casting (HPDC), due to the possibility to reach high output rates and to produce complex lightweight components with low wall thicknesses (Tharumarajah, 2008). In general, the demand for lightweight components is growing worldwide, as aluminium and magnesium alloys offer excellent material properties and can contribute measures to reduce the fuel consumption of automobiles (Cui and Roven, 2010; Neto et al., 2009). Especially for larger companies, there are possibilities to integrate waste heat streams from the die casting line into the hot water supply system in a worthwhile way and thus increasing the energy efficiency to a more economical and profitable performance. Water is primarily used for cooling purposes, e.g. to cool the casting forms and for die casting cooling baths in the case of heat treatment. This is primarily done via internal closed cooling cycles, which allows a substantial reduction in the fresh water demand and wastewater formation. Water is also used for cleaning purposes and as carrier medium for water soluble release agents. Generally, these release agents are highly responsible for the casting quality, the surface finish, the ease of cavity fill and the ease of casting ejection. These release agents can also speed up the casting rate, minimize maintenance requirements and reduce the accumulation of material on the die face (EIPPCB, 2005). The reuse of excess release agents without adequate treatment has to be seen critical, as negative side effects to the casting surface as well as to the die surface can not be ruled out due to the presence of leaked liquids from the hydraulic system of the HPDCs. In the case of HPDC mainly hydraulic fluids on water glycolic base (HFC fluids) are used for the reason of fire protection (Keudel, 2011). Therefore, the direct reuse of these excess release agents is not applied in most of the cases and the wastewater stream is treated using end-of-pipe measures. However, if the HPDCs are equipped with two separate collecting tanks, the possibility to reuse excess release agents could be increased, as the risk of contaminating the excess release agent with leaked liquids and therefore with glycolic substances could be minimized (Keudel, 2011). Furthermore, many manufacturing processes require a cleaning step within the process chain for degreasing and cleaning metal surfaces in order to gain an optimal product quality in the following process steps. In many cases, these cleaning processes need a high amount of water, chemicals and energy and in the majority of the
147
cases a classification of the necessary surface cleanliness is not given. In fact, the operating requirements to the cleaning process are defined by the successful realization of the following process steps (e.g. assembling, hardening, etc.). These cleaning steps are not the core business of manufacturing companies and due to their classification as marginal process steps, the specific knowledge on cleaning processes can often be seen as limited within individual companies. Thereby, in many cases a high potential to enhance the operating performance and to reduce the specific costs on these processes is given. Furthermore, these cleaning processes become more and more important as independent process steps within the value added chain, as increasing quality standards require a defined cleanliness of component surfaces. In order to reduce the input of process utilities (e.g. coolants and lubricants) into the cleaning process, an integrated approach also requires an optimization of upstream manufacturing steps and the best possible adjustment when selecting treatment technologies and chemical systems. For example, dry-cutting technologies became more and more common within the production of automotive suppliers in the past few years (Wyman and Mueller, 2002). The difference to wetcutting technologies is the possibility to reduce coolants and lubricants significantly. This has positive effects to the economical and ecological performance of the cutting process as well as to downstream processes like component cleaning steps, as the introduction of impurities (e.g. coolants and lubricants) can be reduced. However, also the parameter “new part materials” has to be considered when talking about process and technology improvements. The use of new materials (e.g. high-temperature alloys) can limit the implementation of dry-cutting technologies again, as their low thermal conductivity leads to extremely high temperatures in the cutting zone (Pusavec et al., 2010), which makes the removing of heat by the use of specific heat carrier media necessary. 2. Materials and methods 2.1. System analysis and data acquisition Starting in December 2008 a comprehensive potential study was carried out by the author to improve the existing performance of an
Fig. 1. Production site and system boundaries including the relevant processes considered within the project.
148
P. Enderle et al. / Journal of Cleaner Production 34 (2012) 146e152
automotive supplier with the focus on the technological system optimization of the production line “die casting”. The investigated company produces driveline and chassis control systems. The machinery comprises more than 250 facilities including CNC machining centres, hardening and tempering furnaces, HPDC devices and component cleaning facilities. More than 48,000 single components (basis 2009) are produced per day in a 3-shift operation. Due to organizational and logistical reasons, the production area of approx. 40,000 m2 is divided into three production lines: mechanical processing, hardening and die casting (Fig. 1). Due to the complexity and variety of productions steps within the company, a detailed system analysis was carried out at the beginning of the potential study to define all relevant energy and water related processes. To enhance the possibility for an efficient heat recovery, the system boundaries of the production line die casting were expanded to the whole company and in doing so to the production lines hardening and mechanical processing (Fig. 1). The expression “system” was chosen as synonym for “production line”, as this term much better represents the interaction between single processes and there exchange of material, energy, water and information flows. Within the “expanded system die casting” relevant energy and water streams were defined and interpreted on the basis of quantitative and qualitative criteria. Technical documents and internal data (e.g. operating times, product output, resource input, process specifications, specific costs, etc.) were evaluated and missing data were collected through measurements and analysis. As mentioned by Deul (2002), a stepwise approach allows the review of the project at each stage, whereby the first step should be a plant-wide audit to provide relevant process information and detailed operation data. Thus, depending on the production capacity, the company site and the different production lines/processes a stepwise approach was used orientating on the companies goals: - Data acquisition of all energy and water related data within the company (demand and availability) - Calculation of the overall energy and water balance of the company - Identifying relevant processes and reviewing of possible measures for enhancing the overall performance (technological
improvements, best available technologies, reduction of heat losses and fresh water demand, etc.) - Thermodynamic calculations and design of a heat exchanger network (heat integration) - Laboratory and pilot tests and system optimization of the internal water management (process water reuse) - Economic and ecological analysis
2.2. Process optimization through heat recovery and process water reuse With the focus on heat recovery and process water reuse measures, the following areas were investigated in detail: - Production line die casting with the focus on melting/holding furnaces and external HPDC cooling (water soluble release agent) - Production line mechanical processing with the focus on component cleaning processes - Production line hardening with the focus on component cleaning processes and hardening furnace - Wastewater treatment and internal water management - Central heating with the focus on hot water supply Fig. 2 should give an overview to the considered in- and output flows of the investigated processes within the potential study. The output stream considered within the potential study include the wastewater streams WW I, WW II, WW III and WW IV and waste heat form melting/holding furnaces and central heating. In the case of component cleaning facilities all 15 devices were evaluated in detail taking into account the following criteria: -
Performance of the existing bath maintenance system Bath lifetime, specific chemical and fresh water demand Specific wastewater flow Process throughput, operating time and purity requirement.
Based on the results of the evaluation step focused on the mentioned criteria, relevant optimization measures for each
Fig. 2. Considered in- and output flows within the investigated company.
P. Enderle et al. / Journal of Cleaner Production 34 (2012) 146e152
cleaning machine were worked out and the performance of one cleaning machine with the highest throughput was improved exemplary. As upstream processes (e.g. cutting-technologies) can have an immediate effect to the component cleaning process, the evaluation also included an analysis of used coolants and lubricants for upstream processes to verify there impact to the cleaning systems. 2.3. Laboratory and pilot tests With the aim to reduce the yearly excess release agent flow WW III (Fig. 2) by generating a filtrate which can be reused as solvent for the release agent, ultra filtration (UF) tests were made and critical/ limiting factors for a successful implementation were defined. These limiting factors were defined, as the water quality plays a major role regarding negative side effects to the product quality and the used die casting tools. By using an ultra membrane test facility (tube module, PES membrane; cut-off 100 kDa; surface area 0.82 m2) the principal application of this technology could be demonstrated. For the experimental tests, a cross-flow velocity of 15 m3/h with an average permeate flux of 70 L/h was chosen. The cross-flow velocity and the pressure difference p1 p2 was controlled over control valves. The ultra membrane test facility was controlled via a central control unit and measurement data were recorded online. The sampling of the filtrate (permeate) was taken from a sampling valve, the wastewater sample (feed solution ¼ excess release agent) was taken from an installed preliminary tank. The parameter COD (Chemical Oxygen Demand), TOC (Total Organic Carbon) and TIC (Total Inorganic Carbon) were analyzed to verify the separation efficiency. To reach a further reduction of the remaining COD load after ultra filtration, promising laboratory tests by freeze concentration were made. These tests were made with a laboratory test facility designed and build by Joanneum Research. This test facility is based on the principle of falling film crystallization. i.e. on the crystallization of the solvent (e.g. water) on a cold surface by a direct contact of the fluid to be treated. Generally, the technology of freeze concentration is a thermal process to separate particles and dissolved substances from aqueous systems based on a solideliquid phase separation. Test with different amounts of water and different flow cycles have been carried out. Impurity rations (%), efficiency (%), ice production rate und energy consumption have been measured. Organic carbon (analyzed as TOC) was used to verify the separation efficiency. To increase the economical and ecological performance of component cleaning processes, additional investigations were made within the study, especially following the aims: - improved removing of impurities (oil, grease, metallic dust, etc.) from the cleaning baths (pre-degreasing, degreasing, rinsing) - reduction of the organic load and selective recovery of cleaning chemicals (surfactants) - increased cleaning bath lifetime and reduced demand of cleaning chemicals Besides commonly used oil separators the practical application of membrane filtration technologies to separate impurities from the cleaning baths by recycling surfactants at the same time was evaluated. Therefore, in the course of laboratory tests with ultra filtration membranes the effective removal of oil and grease at the simultaneous recovery of surfactants was tested. The cross-flow ultra filtration tests were undertaken under different flow velocities (1.06e1.86 m/s) and the retention behaviour for oil/grease and surfactants was determined depending on the pressure difference.
149
As feed solution the degreasing bath of one component cleaning facility within the investigated company (modular cleaning system) was used, whereby different UF membranes were tested (regenerated cellulose membrane: cut-off 100 kDa; polyethersulfone membrane: cut-off 150 kDa; polysulfone membrane: cut-off 100 kDa). The regenerated cellulose membrane showed the best filtration and retention characteristics, whereby an average COD reduction of 75% and a selective recovery of surfactants (nonionic) of on average 30% could be reached. Based on these investigations, pilot tests were made with the pre-degreasing bath of one selected component cleaning facility (modular cleaning system) within the investigated company. Using the above mentioned ultra membrane test facility (tube module, PES membrane; cut-off 100 kDa; surface area 0.82 m2) a cross-flow velocity of 15 m3/h was chosen for the experimental tests, whereby the retentate was recycled to the preliminary tank and the filtrate (permeate) was collected in a downstream collection tank. The filtrate showed an average COD reduction of 80% and a remaining concentration of surfactants (nonionic) of 1100 mg/L. This would be sufficient to guarantee a good cleaning performance when reused for the pre-degreasing bath in the investigated case, although a trend towards a lower recovery rate of surfactants than to builder components could be observed. This correlates with literature and can be described by the partial fixing of surfactants at the removed oil (Brunn, 2001; Adams and Jelinek, 1999). Furthermore, the bath maintenance system (oil separator) of the component cleaning facility with the highest product throughput was adapted also including the continuous addition of the necessary cleaning chemicals. 3. Results and discussion 3.1. Energy-related considerations of the company site The specific gas demand of the main consumers within the expanded system die casting formed the basis for the further investigations on possible optimization measures in the case of thermal heat recovery. Natural gas is required for two installed heating boilers (hot water supply) and for aluminium melting and its maintenance in the molten state. The 5 MW heating boiler is used as peak load boiler, as the company site had been originally planned much bigger and an additional production line had been provided. With the 3 MW and 5 MW heating boiler a hot water recirculation pipe with a flow temperature of 80e90 C is supplied. Directly connected to the hot water recirculation pipe, seven cleaning processes are the main consumers of the generated hot water with a total process heat demand of 5800 MWh/y (basis 2009). Five gas-fired furnaces (4 furnaces to maintain the molten aluminium in the molten state and 1 melting furnace) build the further main consumers of natural gas with a total demand of around 6500 MWh/y (basis 2009). Aluminium is completely delivered to the company in the molten state in specific thermal containers and kept molten in the company before used. The melting furnace is used to recover the aluminium from rejected pieces and aluminium residues. The hot exhaust gases of the five gas-fired furnaces are emitted via shared collector stack. As the performance of the installed exhaust fan was designed for the needs of the holding furnaces, an air damper had to be installed in the exhaust pipe of the melting furnace to avoid low pressure conditions in the burning chamber. These low pressure conditions led in the past to a strong suction of air and thereby to ineffective burning conditions in the burning chamber. Within the system hardening four hardening furnaces with integrated cleaning processes are the main consumers of thermal energy. The cleaning processes are supplied with hot water via the mentioned
150
P. Enderle et al. / Journal of Cleaner Production 34 (2012) 146e152
recirculation pipe. The hardening furnaces are directly supplied with gas for the hardening process. To determine potential options to improve the energy efficiency and to reduce the primary energy consumption, reduction options were analyzed in detail and limiting factors for a comprehensive and successful implementation were defined and evaluated. The reduction options clarified below have been worked out in detail taking into account limiting factors identified within the system analysis. One reduction option, regarding the heat recovery within the die casting line, provides the integration of a heat exchanger to raise the return temperature of the recirculation pipe of the company’s hot water supply system. As no potential heat sink within the system “die casting” could be identified, this solution enables the recovery of 184 MWh/y (Table 1) waste heat within the expanded system “die casting”. To recover the usable waste heat from the melting and holding furnaces emitted via shared collector stack, the exhaust gas has to be cooled down from 140 C to 110 C and used to raise the return temperature of the recirculation pipe (the recirculation pipe is situated nearby). By cooling down the exhaust gas to 110 C, an appropriate distance to the dew-point temperature will be given and therefore the condensation of water and corrosion problems due to the formation of acids can be avoided. As the performance of the heat exchanger can be influenced negatively by impurities in the exhaust gas, the planning phase of such heat recovery measures should also include the evaluation of the used additives in the melting process. This includes in particular the consideration of the possible usage of additives to remove impurities from the molten aluminium or to remove the slag when cleaning the melting furnace. However, negative side effects to the durability of the heat exchanger could be ruled out in the investigated case, as additives are not used for the melting process within the company. Nevertheless, to avoid negative side effects in the case of cleaning and maintenance works, an option to by-pass the heat exchanger was included in the concept. To recover the useable waste heat of the central heating system, the reduction option provides the integration of a heat exchanger (economizer) into the collector stack to raise the return temperature of the recirculation pipe. The possible reduction of natural gas was calculated on the basis of cooling down the exhaust gas to 100 C taking into consideration the reduced gas demand (reduced usable waste heat) when realizing the waste heat integration measure within the system die casting. In both cases, a certain amount of the available waste heat from the melting/holding furnaces and the two installed heating boilers can be used indirectly to supply the component cleaning facilities in an economical way. All in all approx. 40,000 m3/y (basis 2009) of natural gas and approx. 100 t CO2/y (specific CO2 equivalent based on GEMIS database) could be saved if these optimization measures are implemented within the investigated company. Due to technical and economical reasons, the possibility to reuse the available waste heat from the hardening process was rejected, as an integration of a heat exchanger would make the modification of the exhaust system within the hardening line necessary.
Table 1 Available and usable waste heat within the expanded system die casting. System
Available waste heat
Usable waste heat
Natural gas reduction
Die casting Hardening Central heating Total
576 MWh/y 517 MWh/y 181 MWh/y 1274 MWh/y
184 MWh/y 0 MWh/y 181 MWh/y 365 MWh/y
19,913 0 19,598 39,511
m3/y m3/y m3/y m3/y
3.2. Water-related considerations of the company site Accruing wastewater is treated via vacuum evaporator with a design capacity of 23 m3/d. The vacuum evaporator is operated under contracting arrangements by an external partner. The energy supply is carried out by 1/3 with waste heat of the company’s air compressor units and by 2/3 with electrical energy. Due to capacity problems nearly 30% of the yearly wastewater flow has to be discharged externally. The distillate is discharged to the public sewage system. To verify the yearly wastewater effluent from the HPDC facilities, measurements were carried out and projected on the basis of the product output and on the specific fresh water demand per shot (produced die cast component). The permanent dies are cooled and sprayed with release agents prior to casting, which means before the molten aluminium is injected at high speed and high pressure into the die. The conventional technique for the use of these release agents is to spray a mixture of release agent and water by means of several arranged nozzles on the hot die in one go (EIPPCB, 2005). In the investigated case this mixture was based on a release agent to water ratio of 1e80, whereby 8 L of this mixture was used per shot/ per produced casting. Nearly 40% from this mixture evaporates, cooling the die and leaving the release agent in place, while most of the rest runs off the die. In the investigated case, this excess release agent with a total yearly wastewater flow of 4800 m3/y (nearly 65% of the total yearly wastewater flow) is collected in the machine bed (collection tank) and transported via internal wastewater pipe system to a storage tank before treated at the mentioned vacuum evaporator. Component cleaning facilities are responsible for about 20% of the yearly wastewater flow, the leftover can be allocated to old coolants and lubricants used at CNC machining centres. 3.2.1. Production line die casting Regarding the possibility to treat and reuse the accruing excess release agent as solvent for the preparation of the water soluble release agent, a solution was worked out which includes the integration of an ultra filtration plant. The pilot tests over the test period of 30 days (net test period 220 h) showed an average COD reduction of 77%. Problems like scaling and blocking caused by additives of the release agent (e.g. polysiloxane, synthetic polymers, etc.) could be ruled out. The remaining COD load is mainly caused by leaked liquids from the hydraulic system and therefore by the contamination with glycolic substances. These impurities cannot be removed by the technology of ultra filtration, as the molecular weight of glycol requires a cut-off size < 0.001 mm. In general, ethylene glycol itself is biologically degradable (Biological Oxygen Demand ¼ BOD5 ¼ 0.6 g O2/g) (Jehle et al., 1995), other applicable treatment techniques could be distillation or vacuum evaporation (EIPPCB, 2005). Although glycolic substances have antifreeze properties, promising laboratory test by freeze concentration were made to reach a further reduction of the remaining COD load. The results showed an average reduction of the remaining organic carbon from the generated filtrate after UF of 73% with an ice production rate of 0.14 g/m2$s1. Therefore, a total reduction of organic carbon of about 94% could be reached by using the technologies of ultra filtration and freeze concentration. Based on these results, the following limiting factors were defined and taken into account when working out the concept: - Remaining COD load (glycol, diethylene glycol) - Corrosion behaviour - Formation of organic layers (i.e. biofouling) The reuse without an adequate further treatment step might lead to a discolored surface and to an increased porosity of the
P. Enderle et al. / Journal of Cleaner Production 34 (2012) 146e152
produced die castings, but also to the formation of deposits on the die (Mueller, 1998). Furthermore, a formation of unsaturated organic compounds can arise by heat degradation of certain ingredients of the release agent. Depending on the structure of these ingredients, the formation of organic oxidation products (e.g. organic acids) may occur, which can lead to corrosion problems on the die casting tools (e.g. casting dies) (Reynaud, 2010; Mueller, 1998; Lugscheider et al., 1998). Due to this, the water quality (diluent) for the preparation of the water soluble release agent plays an important role to avoid product quality problems and negative side effects to the die casting tools. The influence of the water quality is quite understandable when considering the mixture ratio of water soluble release agents (in the investigated case 1:80) and the percentage of approx. 99% of water (diluent) used for their preparation. Therefore, a maximum tolerable concentration has to be defined, in order to rule out negative side effects caused by the implementation of excess release agent reuse measures. On the one hand, this maximum tolerable concentration can be reached by an extensive treatment of the accruing excess release agent before reused. On the other hand, there is the possibility to dilute the generated filtrate in order to thin out the remaining COD concentration caused by the mentioned glycolic substances. The formation of organic layers, i.e. biofouling, mainly occurs in the case of direct reuse measures without adequate treatment steps, as additives like biocides (additive of the release agent) evaporate to a higher extent than other ingredients of the release agent when applied to the hot die casting die. Therefore, in the present case a solution was worked out, which includes the treatment and reuse of 18% of the accruing excess release agent through ultra filtration. Thus, a maximum tolerable amount of 25% of the required water demand for the preparation of the water soluble release agent was defined, taking into account the mentioned limiting factors in the case of excess release agent reuse. The economic analysis showed a reduction potential of 18% of the annual costs for the internal water management and a payback period for the required investments of two years. The economical advantage is mainly caused by the reduction of environmental relevant costs in connection with the existing end-of-pipe technology (vacuum evaporator) and wastewater discharge system. 3.2.2. Production line mechanical processing and hardening As evaluated within the laboratory and pilot test, the adjustment between cleaning systems and membranes represents the main factor for an economic implementation of membrane filtration for the treatment of cleaning baths. As builder substances like borates, silicates, alkalis, phosphates and complexing agents are partly retained by membranes, the choice of cleaning systems has to be made on the avoidance of silicates and other builder substances in order to avoid negative effects on the performance of the membrane (scaling, etc.). When considering these aspects, builder substances will pass through the membrane by 90e95% (Brunn, 2001). As the consumption of surfactants at cleaning processes is higher than the consumption of builder substances (due to the partial fixing of surfactants at the removed oil), the use of modular cleaning systems has to be preferred. This enables a consumption-controlled dosage of surfactants and builder substances. Generally, surfactants used for component cleaning lower the surface tension between oil and water, whereby the oil contracts and forms drops. Impurities are absorbed by these drops and removed together with the oil (Gruen, 1999). Modern cleaning chemicals include surfactants emulsifying only slightly, which makes the recycling of cleaning chemicals and the treatment of cleaning baths over mechanical treatment processes possible.
151
However, additives of coolants and lubricants can influence the demulsifying properties of these cleaning systems negatively, resulting in high emulsion stability. Due to additional requirements of coolants and lubricants (e.g. corrosion protection, stability, etc.), they also include a wide variety of additives. These additives interact with the cleaning chemicals when brought in over the product which has to be cleaned, resulting in the mentioned side effects. Conventional cleaning systems combine builder substances and surfactants in one product. By the combination of surfactants and builder substances (e.g. phosphates, borates, sodium hydroxide, etc.) the cleaning performance of surfactants is enhanced. However, by a higher consumption of surfactants than builder substances, the performance of conventional cleaners is accompanied by an increase of builder substances. This can mainly be explained by the partial fixing of surfactants at the formed oil drops, whereby the builder substances are only consumed by dragouts. Modern cleaning systems allow the dosage of surfactants and builder substances individually. This enables the dosing according to the real needs of the cleaning bath resulting in the possibility to use membranes for recycling and treatment measures (Brunn, 2001). In the investigated case, conventional cleaning systems as well as modular cleaning systems are used for the cleaning processes, mainly based on spray cleaning and running at working temperatures from 50 to 70 C. The bath heating is done partly by heating rods and therefore by electricity. Seven components cleaning facilities are supplied by hot water over integrated heat exchanger connected to the recirculation pipe (Fig. 1). The average bath lifetime of degreasing baths was five weeks, whereby a wastewater flow of approximately 500 m3/y was caused by direct bath exchange (basis 2009). Beside this, a certain amount of wastewater also accrued from the intermediate release of cleaning fluids in the case of decreasing cleaning performance. The demand for surfactants is monitored daily by measuring the dynamic surface tension by the bubble pressure principle. The concentration of builder substances is monitored by means of acid-base titration. The monitoring is done discontinuously, a continuous dosing of cleaning chemicals was not provided in the past. Generally, the bubble pressure method could also be used continuously by online measuring, which enables the regulation/control of dosing pumps and therefore the adjustment of constant surfactant concentrations of the cleaning system (Gruen, 2009). Within the investigated case, the integration of a membrane filtration plant for a discontinuous operation at the central wastewater treatment plant represents one possible option. Due to the existing piping system and the possibility of using an intermediate storage tank, the cleaning baths of all component cleaning facilities could be treated discontinuously over a central membrane filtration plant and reused at the individual component cleaning facility. The further treatment of the retentate can subsequently be carried out at the existing vacuum evaporator. During the project, a coarse filter was integrated as a first optimization step at the central wastewater treatment plant in order to remove impurities like tiny metal filings, paper, paint particles, etc. from the cleaning fluids during general cleaning works at the component cleaning facilities. By this low cost investment, the filtered cleaning bath is now piped back to the individual component cleaning facility and reused for the cleaning process. In the past, the cleaning baths were rejected during general cleaning works although the cleaning performance would have been sufficient for its further use. If this system has proved its worth, a downstream integration of a central membrane filtration plant as a holistic concept would be conceivable. An important prerequisite for this would be the consideration of all component cleaning facilities taking into account the following points:
152
P. Enderle et al. / Journal of Cleaner Production 34 (2012) 146e152
- Reduced contamination with impurities of upstream processes by the reduction of drag-outs, - Adaptation of cleaning chemicals (cleaning systems) and adjustments to the required component surface purity (definition of cleaning requirements), - Optimization and adaptation of existing bath maintenance systems (e.g. oil separators, micro filter, etc.), - Optimized dosage of cleaning chemicals and monitoring (continuously) of process stability and cleaning performance, - Optimized interaction between machine revisions/general cleaning work and the treatment of cleaning fluids/cleaning systems and - Consideration of possible negative side effects concerning biological conversion in the piping system. Concerning the adaption of existing bath maintenance systems, all 15 component cleaning facilities and the performance of their bath maintenance systems (mainly oil separators) were evaluated within the project. The oil separator system of one component cleaning facility with the highest throughput, wastewater flow and chemical demand was adapted during the project also including an automatic chemical dosage system to guarantee a constant concentration of surfactants and builder substances. Furthermore, the cleaning formulation was changed to a modular cleaning system. By these optimization measures, the lifetime of the cleaning baths could be increased from 5 to 25 weeks and the yearly operating costs could be decreased by 43% (basis 2010). To assess the total annual costs, the specific operational costs for chemicals, fresh water, demonized water, wastewater treatment and wastewater discharge were considered. 4. Conclusion The industrial water and energy systems consist of several subsystems which are more or less connected or influenced by each other in a more or less intensive way. This strong conjunction of water and heat management at industrial sites requires a systematic approach to design the best possible concept for the simultaneous reduction of water and energy. System boundaries and therefore communication boundaries between different divisions within one company have to be reflected in order to identify the main variables influencing the successful and extensive implementation of optimization measures. Considering the resources energy and water the individual influences between sub-processes were investigated within the potential study and limiting factors for a successful and extensive implementation were worked out. It has been demonstrated that the areas of heat recovery and process water reuse are closely connected and that optimization measures within one area have an immediate effect on the whole system. Therefore an intelligent combination of technology and system optimization is the basic requirement in order to develop practical optimization concepts through a systematic approach within complex production systems. However, when working out optimization measures to retrofit existing processes, the maximum efficiency is influenced through a variety of sub-processes, which limit the economical implementation potential as a whole. Thus, in many cases far-reaching adaptations of sub-processes have to accompany these optimization measures, to use e.g. the maximum potential of the available waste heat for heat recovery measures. Furthermore, the usage of process auxiliaries has far-reaching effects on process water reuse measures and therefore to the possibility to minimize the wastewater flow. Within this potential study, possible optimization measures were worked out, which could have a positive influence to the economical and ecological
performance of the investigated company. When working out the optimization measures, the focus was laid to measures with short payback periods. Acknowledgements The authors would like to thank the Austrian Research Promotion Agency (FFG) for their financial support. This project was supported by the Austrian Climate and Energy Fund in the frame of the research and technology programme “New Energies 2020”. References Adams, H.N., Jelinek, T.W., 1999. Cleaning and Degreasing in the Metal Industry, first ed. Eugen G. Leuze Verlag, Saulgau/Württ, Germany (in German). Alves, C., Ferrao, P.M.C., Silva, A.J., Reis, L.G., Freitas, M., Rodrigues, L.B., Alves, D.E., 2010. Ecodesign of automotive components making use of natural jute fibre composites. Journal of Cleaner Production 18 (4), 313e327. Brunn, K., 2001. Modular Aqueous Cleaning Systems-Different Fields of Application. SurTec Technical Letter 3. Surtec GmbH, Zwingenberg, Germany (in German). Christensen, T.B., 2011. Modularised eco-innovation in the auto industry. Journal of Cleaner Production 19 (2e3), 212e220. Cui, J., Roven, H.J., 2010. Recycling of automotive aluminium. Transactions of Nonferrous Metals Society of China 20 (11), 2057e2063. Deul, A.S., 2002. Systematic approach to water resource management in industry. In: Water Recycling and Resource Recovery in Industry e Analysis, Technologies and Implementation, vol. 252e270. IWA Publishing, London, UK. Duval, D., MacLean, H.L., 2007. The role of product information in automotive plastics recycling: a financial and life cycle assessment. Journal of Cleaner Production 15 (11e12), 1158e1168. EIPPCB (European IPPC Bureau), 2005. Reference Document on Best Available Techniques in the Smitheries and Foundries Industry. European Commission. Granata, G., Moscardini, E., Furlani, G., Pagnanelli, F., 2009. Characterisation of automobile shredder residue. Chemical Engineering Transaction 17, 427e432. Gruen, R., 2009. Cleaning Bath Maintenance and Treatment. ZVO report, special edition. Fachverband Industrieller Beschichter e.V., Hilden, Germany (in German). Gruen, R., 1999. Cleaning as a Part of the Heat Treatment. SurTec Technical Letter 13a. Surtec GmbH, Zwingenberg, Germany. van den Hoed, R., 2007. Source of radical technological innovation: the emergence of fuel cell technology in the automotive industry. Journal of Cleaner Production 15 (11e12), 1014e1021. Jehle, W., Staneff, Th., Wagner, B., Steinwandel, J., 1995. Separation of glycol and water from coolant liquids by evaporation, reverse osmosis and pervaporation. Journal of Membrane Science 102, 9e19. Keudel, L., 2011. Treatment of wastewater from the automotive industry. In: Proceedings DWA Conference Behandlung von Industrie- und Gewerbeabwasser. Kassel, Germany, 2011, pp. 322e345 (in German). Kjaerheim, G., 2005. Cleaner production and sustainability. Journal of Cleaner Production 13 (4), 329e339. Lugscheider, E., Barimani, C., Guerreiro, S., Bobzin, K., 1998. Corrosion tests of PVD coatings with die lubricant used for Al high-pressure die-casting. Surface and Coatings Technology 108e109, 408e412. Mueller, W., 1998. Corrosion at Die Casting Facilities When Using Aqueous Products. Technical Letter. Geiger þ Co. Schmierstoff-Chemie GmbH, Heilbronn, Germany (in German). Neto, B., Kroeze, C., Hordijk, L., Costa, C., Pulles, T., 2009. Strategies to reduce the environmental impact of an aluminium pressure die casting plant: a scenario analysis. Journal of Environmental Management 90 (2), 815e830. Orsato, R.J., Wells, P., 2007. U-turn: the rise and demise of the automotive industry. Journal of Cleaner Production 15 (11e12), 994e1006. Passarini, F., Ciacci, L., Santini, A., Vassura, I., Morselli, L., 2012. Auto shredder residue LCA: implications of ASR composition evolution. Journal of Cleaner Production 23 (1), 28e36. Pusavec, F., Kramar, D., Krajnik, P., Kopac, J., 2010. Transitioning to sustainable production e part II: evaluation of sustainable machining technologies. Journal of Cleaner Production 18 (12), 1211e1221. Reynaud, A., 2010. Corrosion of Cast Irons. Sheir’s Corrosion 3: Corrosion and Degradation of Engineering Materials 1737e1788. Tharumarajah, A., 2008. Benchmarking aluminium die casting operations. Resources, Conservation and Recycling 52, 1185e1189. Telukdarie, A., Buckley, C., Koefoed, M., 2006. The importance of assessment tools in promoting cleaner production in the metal finishing industry. Journal of Cleaner Production 14 (18), 1612e1621. Wyman, H., Mueller, H., 2002. The new 3 face grinding blade design for dry-cutting. Technical Magazine of the Gearing Partners Klingelnberg, Liebherr and Oerlokon, sigma report 12, 21e29. Zapata, C., Nieuwenhuis, P., 2010. Exploring innovation in the automotive industry: new technologies for cleaner cars. Journal of Cleaner Production 18 (1), 14e20.
Contents lists available at SciVerse ScienceDirect
Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro
Potential alternative for water and energy savings in the automotive industry: case study for an Austrian automotive supplier Peter Enderle*, Otto Nowak, Julia Kvas Resources e Institute for Water, Energy and Sustainability, Joanneum Research, Elisabethstrasse 16/I, 8010 Graz, Austria
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 July 2011 Received in revised form 10 November 2011 Accepted 12 November 2011 Available online 20 November 2011
This paper show the alternative possible optimization measures to increase the water and energy efficiency in the automotive industry. Within a potential study the technological system optimization of an automotive supplier was performed by combining the fields of process water reuse and heat recovery. Possible optimization measures to retrofit existing processes with the focus on high pressure die casting and component cleaning were worked out. Furthermore, limiting factors for a successful and extensive implementation in the case of retrofitting existing processes and systems were evaluated. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: Aluminium die casting Component cleaning System optimization Process water reuse Heat recovery
1. Introduction The automotive industry is one of Austria’s most important industrial sectors comprising the complete value chain from engineering services, metal working, metal processing, electronics manufacturing up to the production of die cast components (in particular lightweight components made out of aluminium and magnesium alloys). In contrast to other branches of industry, Original Equipment Manufacturers (OEMs) have outsourced large parts of their production and R&D activities to the supplier industry over the last decades. Thus, automotive suppliers are an integral part of automobile production, whereby their portfolio has changed over the years from the production of single components to complete systems and devices (e.g. driveline systems, engines, etc.), also including in-house R&D activities. Today, a modern automobile comprises up to 10,000 components, whereby nearly 80% of them are produced by external partners. Furthermore, beside automotive suppliers many other branches of industry are directly or indirectly linked to the automotive sector as an integral part of upstream and downstream production steps (e.g. chemical industry, textile industry, recycling industry, etc.). Therefore, economic and environmental challenges within the automobile sector have * Corresponding author. Tel.: þ43 316 876 2420; fax: þ43 316 8769 2420. E-mail addresses: [email protected] (P. Enderle), otto.nowak@ joanneum.at (O. Nowak), [email protected] (J. Kvas). 0959-6526/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.jclepro.2011.11.013
immediate impacts to the development of other industrial sectors (Orsato and Wells, 2007). A number of studies have been published in the recent years mainly focused on special issues regarding ecoinnovations, new technology developments and vehicle recycling (Christensen, 2011; Passarini et al., 2012; Alves et al., 2010; Zapata and Nieuwenhuis, 2010; Cui and Roven, 2010; Granata et al., 2009; van den Hoed, 2007; Duval and MacLean, 2007). These studies underline the dynamic change of market needs with an immediate impact on production and process needs within the automotive sector but also on other branches of industry. Although the automotive sector as a whole can not specified as water and energy intensive branch of industry and considerable improvements to reduce the environmental impact of car production have been achieved within the last decades, there are still certain production areas and production steps with a high potential to increase the resource efficiency, also including good housekeeping practices. Good housekeeping is focused on measures with low investment requirements, e.g. in order to minimize material losses, to reduce the consumption of process chemicals, to prevent unnecessary waste generation (Telukdarie et al., 2006) but also to raise employees’ environmental awareness so that employees adopt a positive attitude to pollution prevention (Kjaerheim, 2005). Regarding the automotive industry, the production areas of die casting, mechanical processing, paint finishing and hardening can be seen as those areas with a high potential of process water reuse and heat recovery measures. The more often applied die casting
P. Enderle et al. / Journal of Cleaner Production 34 (2012) 146e152
process within the automotive sector is High Pressure Die Casting (HPDC), due to the possibility to reach high output rates and to produce complex lightweight components with low wall thicknesses (Tharumarajah, 2008). In general, the demand for lightweight components is growing worldwide, as aluminium and magnesium alloys offer excellent material properties and can contribute measures to reduce the fuel consumption of automobiles (Cui and Roven, 2010; Neto et al., 2009). Especially for larger companies, there are possibilities to integrate waste heat streams from the die casting line into the hot water supply system in a worthwhile way and thus increasing the energy efficiency to a more economical and profitable performance. Water is primarily used for cooling purposes, e.g. to cool the casting forms and for die casting cooling baths in the case of heat treatment. This is primarily done via internal closed cooling cycles, which allows a substantial reduction in the fresh water demand and wastewater formation. Water is also used for cleaning purposes and as carrier medium for water soluble release agents. Generally, these release agents are highly responsible for the casting quality, the surface finish, the ease of cavity fill and the ease of casting ejection. These release agents can also speed up the casting rate, minimize maintenance requirements and reduce the accumulation of material on the die face (EIPPCB, 2005). The reuse of excess release agents without adequate treatment has to be seen critical, as negative side effects to the casting surface as well as to the die surface can not be ruled out due to the presence of leaked liquids from the hydraulic system of the HPDCs. In the case of HPDC mainly hydraulic fluids on water glycolic base (HFC fluids) are used for the reason of fire protection (Keudel, 2011). Therefore, the direct reuse of these excess release agents is not applied in most of the cases and the wastewater stream is treated using end-of-pipe measures. However, if the HPDCs are equipped with two separate collecting tanks, the possibility to reuse excess release agents could be increased, as the risk of contaminating the excess release agent with leaked liquids and therefore with glycolic substances could be minimized (Keudel, 2011). Furthermore, many manufacturing processes require a cleaning step within the process chain for degreasing and cleaning metal surfaces in order to gain an optimal product quality in the following process steps. In many cases, these cleaning processes need a high amount of water, chemicals and energy and in the majority of the
147
cases a classification of the necessary surface cleanliness is not given. In fact, the operating requirements to the cleaning process are defined by the successful realization of the following process steps (e.g. assembling, hardening, etc.). These cleaning steps are not the core business of manufacturing companies and due to their classification as marginal process steps, the specific knowledge on cleaning processes can often be seen as limited within individual companies. Thereby, in many cases a high potential to enhance the operating performance and to reduce the specific costs on these processes is given. Furthermore, these cleaning processes become more and more important as independent process steps within the value added chain, as increasing quality standards require a defined cleanliness of component surfaces. In order to reduce the input of process utilities (e.g. coolants and lubricants) into the cleaning process, an integrated approach also requires an optimization of upstream manufacturing steps and the best possible adjustment when selecting treatment technologies and chemical systems. For example, dry-cutting technologies became more and more common within the production of automotive suppliers in the past few years (Wyman and Mueller, 2002). The difference to wetcutting technologies is the possibility to reduce coolants and lubricants significantly. This has positive effects to the economical and ecological performance of the cutting process as well as to downstream processes like component cleaning steps, as the introduction of impurities (e.g. coolants and lubricants) can be reduced. However, also the parameter “new part materials” has to be considered when talking about process and technology improvements. The use of new materials (e.g. high-temperature alloys) can limit the implementation of dry-cutting technologies again, as their low thermal conductivity leads to extremely high temperatures in the cutting zone (Pusavec et al., 2010), which makes the removing of heat by the use of specific heat carrier media necessary. 2. Materials and methods 2.1. System analysis and data acquisition Starting in December 2008 a comprehensive potential study was carried out by the author to improve the existing performance of an
Fig. 1. Production site and system boundaries including the relevant processes considered within the project.
148
P. Enderle et al. / Journal of Cleaner Production 34 (2012) 146e152
automotive supplier with the focus on the technological system optimization of the production line “die casting”. The investigated company produces driveline and chassis control systems. The machinery comprises more than 250 facilities including CNC machining centres, hardening and tempering furnaces, HPDC devices and component cleaning facilities. More than 48,000 single components (basis 2009) are produced per day in a 3-shift operation. Due to organizational and logistical reasons, the production area of approx. 40,000 m2 is divided into three production lines: mechanical processing, hardening and die casting (Fig. 1). Due to the complexity and variety of productions steps within the company, a detailed system analysis was carried out at the beginning of the potential study to define all relevant energy and water related processes. To enhance the possibility for an efficient heat recovery, the system boundaries of the production line die casting were expanded to the whole company and in doing so to the production lines hardening and mechanical processing (Fig. 1). The expression “system” was chosen as synonym for “production line”, as this term much better represents the interaction between single processes and there exchange of material, energy, water and information flows. Within the “expanded system die casting” relevant energy and water streams were defined and interpreted on the basis of quantitative and qualitative criteria. Technical documents and internal data (e.g. operating times, product output, resource input, process specifications, specific costs, etc.) were evaluated and missing data were collected through measurements and analysis. As mentioned by Deul (2002), a stepwise approach allows the review of the project at each stage, whereby the first step should be a plant-wide audit to provide relevant process information and detailed operation data. Thus, depending on the production capacity, the company site and the different production lines/processes a stepwise approach was used orientating on the companies goals: - Data acquisition of all energy and water related data within the company (demand and availability) - Calculation of the overall energy and water balance of the company - Identifying relevant processes and reviewing of possible measures for enhancing the overall performance (technological
improvements, best available technologies, reduction of heat losses and fresh water demand, etc.) - Thermodynamic calculations and design of a heat exchanger network (heat integration) - Laboratory and pilot tests and system optimization of the internal water management (process water reuse) - Economic and ecological analysis
2.2. Process optimization through heat recovery and process water reuse With the focus on heat recovery and process water reuse measures, the following areas were investigated in detail: - Production line die casting with the focus on melting/holding furnaces and external HPDC cooling (water soluble release agent) - Production line mechanical processing with the focus on component cleaning processes - Production line hardening with the focus on component cleaning processes and hardening furnace - Wastewater treatment and internal water management - Central heating with the focus on hot water supply Fig. 2 should give an overview to the considered in- and output flows of the investigated processes within the potential study. The output stream considered within the potential study include the wastewater streams WW I, WW II, WW III and WW IV and waste heat form melting/holding furnaces and central heating. In the case of component cleaning facilities all 15 devices were evaluated in detail taking into account the following criteria: -
Performance of the existing bath maintenance system Bath lifetime, specific chemical and fresh water demand Specific wastewater flow Process throughput, operating time and purity requirement.
Based on the results of the evaluation step focused on the mentioned criteria, relevant optimization measures for each
Fig. 2. Considered in- and output flows within the investigated company.
P. Enderle et al. / Journal of Cleaner Production 34 (2012) 146e152
cleaning machine were worked out and the performance of one cleaning machine with the highest throughput was improved exemplary. As upstream processes (e.g. cutting-technologies) can have an immediate effect to the component cleaning process, the evaluation also included an analysis of used coolants and lubricants for upstream processes to verify there impact to the cleaning systems. 2.3. Laboratory and pilot tests With the aim to reduce the yearly excess release agent flow WW III (Fig. 2) by generating a filtrate which can be reused as solvent for the release agent, ultra filtration (UF) tests were made and critical/ limiting factors for a successful implementation were defined. These limiting factors were defined, as the water quality plays a major role regarding negative side effects to the product quality and the used die casting tools. By using an ultra membrane test facility (tube module, PES membrane; cut-off 100 kDa; surface area 0.82 m2) the principal application of this technology could be demonstrated. For the experimental tests, a cross-flow velocity of 15 m3/h with an average permeate flux of 70 L/h was chosen. The cross-flow velocity and the pressure difference p1 p2 was controlled over control valves. The ultra membrane test facility was controlled via a central control unit and measurement data were recorded online. The sampling of the filtrate (permeate) was taken from a sampling valve, the wastewater sample (feed solution ¼ excess release agent) was taken from an installed preliminary tank. The parameter COD (Chemical Oxygen Demand), TOC (Total Organic Carbon) and TIC (Total Inorganic Carbon) were analyzed to verify the separation efficiency. To reach a further reduction of the remaining COD load after ultra filtration, promising laboratory tests by freeze concentration were made. These tests were made with a laboratory test facility designed and build by Joanneum Research. This test facility is based on the principle of falling film crystallization. i.e. on the crystallization of the solvent (e.g. water) on a cold surface by a direct contact of the fluid to be treated. Generally, the technology of freeze concentration is a thermal process to separate particles and dissolved substances from aqueous systems based on a solideliquid phase separation. Test with different amounts of water and different flow cycles have been carried out. Impurity rations (%), efficiency (%), ice production rate und energy consumption have been measured. Organic carbon (analyzed as TOC) was used to verify the separation efficiency. To increase the economical and ecological performance of component cleaning processes, additional investigations were made within the study, especially following the aims: - improved removing of impurities (oil, grease, metallic dust, etc.) from the cleaning baths (pre-degreasing, degreasing, rinsing) - reduction of the organic load and selective recovery of cleaning chemicals (surfactants) - increased cleaning bath lifetime and reduced demand of cleaning chemicals Besides commonly used oil separators the practical application of membrane filtration technologies to separate impurities from the cleaning baths by recycling surfactants at the same time was evaluated. Therefore, in the course of laboratory tests with ultra filtration membranes the effective removal of oil and grease at the simultaneous recovery of surfactants was tested. The cross-flow ultra filtration tests were undertaken under different flow velocities (1.06e1.86 m/s) and the retention behaviour for oil/grease and surfactants was determined depending on the pressure difference.
149
As feed solution the degreasing bath of one component cleaning facility within the investigated company (modular cleaning system) was used, whereby different UF membranes were tested (regenerated cellulose membrane: cut-off 100 kDa; polyethersulfone membrane: cut-off 150 kDa; polysulfone membrane: cut-off 100 kDa). The regenerated cellulose membrane showed the best filtration and retention characteristics, whereby an average COD reduction of 75% and a selective recovery of surfactants (nonionic) of on average 30% could be reached. Based on these investigations, pilot tests were made with the pre-degreasing bath of one selected component cleaning facility (modular cleaning system) within the investigated company. Using the above mentioned ultra membrane test facility (tube module, PES membrane; cut-off 100 kDa; surface area 0.82 m2) a cross-flow velocity of 15 m3/h was chosen for the experimental tests, whereby the retentate was recycled to the preliminary tank and the filtrate (permeate) was collected in a downstream collection tank. The filtrate showed an average COD reduction of 80% and a remaining concentration of surfactants (nonionic) of 1100 mg/L. This would be sufficient to guarantee a good cleaning performance when reused for the pre-degreasing bath in the investigated case, although a trend towards a lower recovery rate of surfactants than to builder components could be observed. This correlates with literature and can be described by the partial fixing of surfactants at the removed oil (Brunn, 2001; Adams and Jelinek, 1999). Furthermore, the bath maintenance system (oil separator) of the component cleaning facility with the highest product throughput was adapted also including the continuous addition of the necessary cleaning chemicals. 3. Results and discussion 3.1. Energy-related considerations of the company site The specific gas demand of the main consumers within the expanded system die casting formed the basis for the further investigations on possible optimization measures in the case of thermal heat recovery. Natural gas is required for two installed heating boilers (hot water supply) and for aluminium melting and its maintenance in the molten state. The 5 MW heating boiler is used as peak load boiler, as the company site had been originally planned much bigger and an additional production line had been provided. With the 3 MW and 5 MW heating boiler a hot water recirculation pipe with a flow temperature of 80e90 C is supplied. Directly connected to the hot water recirculation pipe, seven cleaning processes are the main consumers of the generated hot water with a total process heat demand of 5800 MWh/y (basis 2009). Five gas-fired furnaces (4 furnaces to maintain the molten aluminium in the molten state and 1 melting furnace) build the further main consumers of natural gas with a total demand of around 6500 MWh/y (basis 2009). Aluminium is completely delivered to the company in the molten state in specific thermal containers and kept molten in the company before used. The melting furnace is used to recover the aluminium from rejected pieces and aluminium residues. The hot exhaust gases of the five gas-fired furnaces are emitted via shared collector stack. As the performance of the installed exhaust fan was designed for the needs of the holding furnaces, an air damper had to be installed in the exhaust pipe of the melting furnace to avoid low pressure conditions in the burning chamber. These low pressure conditions led in the past to a strong suction of air and thereby to ineffective burning conditions in the burning chamber. Within the system hardening four hardening furnaces with integrated cleaning processes are the main consumers of thermal energy. The cleaning processes are supplied with hot water via the mentioned
150
P. Enderle et al. / Journal of Cleaner Production 34 (2012) 146e152
recirculation pipe. The hardening furnaces are directly supplied with gas for the hardening process. To determine potential options to improve the energy efficiency and to reduce the primary energy consumption, reduction options were analyzed in detail and limiting factors for a comprehensive and successful implementation were defined and evaluated. The reduction options clarified below have been worked out in detail taking into account limiting factors identified within the system analysis. One reduction option, regarding the heat recovery within the die casting line, provides the integration of a heat exchanger to raise the return temperature of the recirculation pipe of the company’s hot water supply system. As no potential heat sink within the system “die casting” could be identified, this solution enables the recovery of 184 MWh/y (Table 1) waste heat within the expanded system “die casting”. To recover the usable waste heat from the melting and holding furnaces emitted via shared collector stack, the exhaust gas has to be cooled down from 140 C to 110 C and used to raise the return temperature of the recirculation pipe (the recirculation pipe is situated nearby). By cooling down the exhaust gas to 110 C, an appropriate distance to the dew-point temperature will be given and therefore the condensation of water and corrosion problems due to the formation of acids can be avoided. As the performance of the heat exchanger can be influenced negatively by impurities in the exhaust gas, the planning phase of such heat recovery measures should also include the evaluation of the used additives in the melting process. This includes in particular the consideration of the possible usage of additives to remove impurities from the molten aluminium or to remove the slag when cleaning the melting furnace. However, negative side effects to the durability of the heat exchanger could be ruled out in the investigated case, as additives are not used for the melting process within the company. Nevertheless, to avoid negative side effects in the case of cleaning and maintenance works, an option to by-pass the heat exchanger was included in the concept. To recover the useable waste heat of the central heating system, the reduction option provides the integration of a heat exchanger (economizer) into the collector stack to raise the return temperature of the recirculation pipe. The possible reduction of natural gas was calculated on the basis of cooling down the exhaust gas to 100 C taking into consideration the reduced gas demand (reduced usable waste heat) when realizing the waste heat integration measure within the system die casting. In both cases, a certain amount of the available waste heat from the melting/holding furnaces and the two installed heating boilers can be used indirectly to supply the component cleaning facilities in an economical way. All in all approx. 40,000 m3/y (basis 2009) of natural gas and approx. 100 t CO2/y (specific CO2 equivalent based on GEMIS database) could be saved if these optimization measures are implemented within the investigated company. Due to technical and economical reasons, the possibility to reuse the available waste heat from the hardening process was rejected, as an integration of a heat exchanger would make the modification of the exhaust system within the hardening line necessary.
Table 1 Available and usable waste heat within the expanded system die casting. System
Available waste heat
Usable waste heat
Natural gas reduction
Die casting Hardening Central heating Total
576 MWh/y 517 MWh/y 181 MWh/y 1274 MWh/y
184 MWh/y 0 MWh/y 181 MWh/y 365 MWh/y
19,913 0 19,598 39,511
m3/y m3/y m3/y m3/y
3.2. Water-related considerations of the company site Accruing wastewater is treated via vacuum evaporator with a design capacity of 23 m3/d. The vacuum evaporator is operated under contracting arrangements by an external partner. The energy supply is carried out by 1/3 with waste heat of the company’s air compressor units and by 2/3 with electrical energy. Due to capacity problems nearly 30% of the yearly wastewater flow has to be discharged externally. The distillate is discharged to the public sewage system. To verify the yearly wastewater effluent from the HPDC facilities, measurements were carried out and projected on the basis of the product output and on the specific fresh water demand per shot (produced die cast component). The permanent dies are cooled and sprayed with release agents prior to casting, which means before the molten aluminium is injected at high speed and high pressure into the die. The conventional technique for the use of these release agents is to spray a mixture of release agent and water by means of several arranged nozzles on the hot die in one go (EIPPCB, 2005). In the investigated case this mixture was based on a release agent to water ratio of 1e80, whereby 8 L of this mixture was used per shot/ per produced casting. Nearly 40% from this mixture evaporates, cooling the die and leaving the release agent in place, while most of the rest runs off the die. In the investigated case, this excess release agent with a total yearly wastewater flow of 4800 m3/y (nearly 65% of the total yearly wastewater flow) is collected in the machine bed (collection tank) and transported via internal wastewater pipe system to a storage tank before treated at the mentioned vacuum evaporator. Component cleaning facilities are responsible for about 20% of the yearly wastewater flow, the leftover can be allocated to old coolants and lubricants used at CNC machining centres. 3.2.1. Production line die casting Regarding the possibility to treat and reuse the accruing excess release agent as solvent for the preparation of the water soluble release agent, a solution was worked out which includes the integration of an ultra filtration plant. The pilot tests over the test period of 30 days (net test period 220 h) showed an average COD reduction of 77%. Problems like scaling and blocking caused by additives of the release agent (e.g. polysiloxane, synthetic polymers, etc.) could be ruled out. The remaining COD load is mainly caused by leaked liquids from the hydraulic system and therefore by the contamination with glycolic substances. These impurities cannot be removed by the technology of ultra filtration, as the molecular weight of glycol requires a cut-off size < 0.001 mm. In general, ethylene glycol itself is biologically degradable (Biological Oxygen Demand ¼ BOD5 ¼ 0.6 g O2/g) (Jehle et al., 1995), other applicable treatment techniques could be distillation or vacuum evaporation (EIPPCB, 2005). Although glycolic substances have antifreeze properties, promising laboratory test by freeze concentration were made to reach a further reduction of the remaining COD load. The results showed an average reduction of the remaining organic carbon from the generated filtrate after UF of 73% with an ice production rate of 0.14 g/m2$s1. Therefore, a total reduction of organic carbon of about 94% could be reached by using the technologies of ultra filtration and freeze concentration. Based on these results, the following limiting factors were defined and taken into account when working out the concept: - Remaining COD load (glycol, diethylene glycol) - Corrosion behaviour - Formation of organic layers (i.e. biofouling) The reuse without an adequate further treatment step might lead to a discolored surface and to an increased porosity of the
P. Enderle et al. / Journal of Cleaner Production 34 (2012) 146e152
produced die castings, but also to the formation of deposits on the die (Mueller, 1998). Furthermore, a formation of unsaturated organic compounds can arise by heat degradation of certain ingredients of the release agent. Depending on the structure of these ingredients, the formation of organic oxidation products (e.g. organic acids) may occur, which can lead to corrosion problems on the die casting tools (e.g. casting dies) (Reynaud, 2010; Mueller, 1998; Lugscheider et al., 1998). Due to this, the water quality (diluent) for the preparation of the water soluble release agent plays an important role to avoid product quality problems and negative side effects to the die casting tools. The influence of the water quality is quite understandable when considering the mixture ratio of water soluble release agents (in the investigated case 1:80) and the percentage of approx. 99% of water (diluent) used for their preparation. Therefore, a maximum tolerable concentration has to be defined, in order to rule out negative side effects caused by the implementation of excess release agent reuse measures. On the one hand, this maximum tolerable concentration can be reached by an extensive treatment of the accruing excess release agent before reused. On the other hand, there is the possibility to dilute the generated filtrate in order to thin out the remaining COD concentration caused by the mentioned glycolic substances. The formation of organic layers, i.e. biofouling, mainly occurs in the case of direct reuse measures without adequate treatment steps, as additives like biocides (additive of the release agent) evaporate to a higher extent than other ingredients of the release agent when applied to the hot die casting die. Therefore, in the present case a solution was worked out, which includes the treatment and reuse of 18% of the accruing excess release agent through ultra filtration. Thus, a maximum tolerable amount of 25% of the required water demand for the preparation of the water soluble release agent was defined, taking into account the mentioned limiting factors in the case of excess release agent reuse. The economic analysis showed a reduction potential of 18% of the annual costs for the internal water management and a payback period for the required investments of two years. The economical advantage is mainly caused by the reduction of environmental relevant costs in connection with the existing end-of-pipe technology (vacuum evaporator) and wastewater discharge system. 3.2.2. Production line mechanical processing and hardening As evaluated within the laboratory and pilot test, the adjustment between cleaning systems and membranes represents the main factor for an economic implementation of membrane filtration for the treatment of cleaning baths. As builder substances like borates, silicates, alkalis, phosphates and complexing agents are partly retained by membranes, the choice of cleaning systems has to be made on the avoidance of silicates and other builder substances in order to avoid negative effects on the performance of the membrane (scaling, etc.). When considering these aspects, builder substances will pass through the membrane by 90e95% (Brunn, 2001). As the consumption of surfactants at cleaning processes is higher than the consumption of builder substances (due to the partial fixing of surfactants at the removed oil), the use of modular cleaning systems has to be preferred. This enables a consumption-controlled dosage of surfactants and builder substances. Generally, surfactants used for component cleaning lower the surface tension between oil and water, whereby the oil contracts and forms drops. Impurities are absorbed by these drops and removed together with the oil (Gruen, 1999). Modern cleaning chemicals include surfactants emulsifying only slightly, which makes the recycling of cleaning chemicals and the treatment of cleaning baths over mechanical treatment processes possible.
151
However, additives of coolants and lubricants can influence the demulsifying properties of these cleaning systems negatively, resulting in high emulsion stability. Due to additional requirements of coolants and lubricants (e.g. corrosion protection, stability, etc.), they also include a wide variety of additives. These additives interact with the cleaning chemicals when brought in over the product which has to be cleaned, resulting in the mentioned side effects. Conventional cleaning systems combine builder substances and surfactants in one product. By the combination of surfactants and builder substances (e.g. phosphates, borates, sodium hydroxide, etc.) the cleaning performance of surfactants is enhanced. However, by a higher consumption of surfactants than builder substances, the performance of conventional cleaners is accompanied by an increase of builder substances. This can mainly be explained by the partial fixing of surfactants at the formed oil drops, whereby the builder substances are only consumed by dragouts. Modern cleaning systems allow the dosage of surfactants and builder substances individually. This enables the dosing according to the real needs of the cleaning bath resulting in the possibility to use membranes for recycling and treatment measures (Brunn, 2001). In the investigated case, conventional cleaning systems as well as modular cleaning systems are used for the cleaning processes, mainly based on spray cleaning and running at working temperatures from 50 to 70 C. The bath heating is done partly by heating rods and therefore by electricity. Seven components cleaning facilities are supplied by hot water over integrated heat exchanger connected to the recirculation pipe (Fig. 1). The average bath lifetime of degreasing baths was five weeks, whereby a wastewater flow of approximately 500 m3/y was caused by direct bath exchange (basis 2009). Beside this, a certain amount of wastewater also accrued from the intermediate release of cleaning fluids in the case of decreasing cleaning performance. The demand for surfactants is monitored daily by measuring the dynamic surface tension by the bubble pressure principle. The concentration of builder substances is monitored by means of acid-base titration. The monitoring is done discontinuously, a continuous dosing of cleaning chemicals was not provided in the past. Generally, the bubble pressure method could also be used continuously by online measuring, which enables the regulation/control of dosing pumps and therefore the adjustment of constant surfactant concentrations of the cleaning system (Gruen, 2009). Within the investigated case, the integration of a membrane filtration plant for a discontinuous operation at the central wastewater treatment plant represents one possible option. Due to the existing piping system and the possibility of using an intermediate storage tank, the cleaning baths of all component cleaning facilities could be treated discontinuously over a central membrane filtration plant and reused at the individual component cleaning facility. The further treatment of the retentate can subsequently be carried out at the existing vacuum evaporator. During the project, a coarse filter was integrated as a first optimization step at the central wastewater treatment plant in order to remove impurities like tiny metal filings, paper, paint particles, etc. from the cleaning fluids during general cleaning works at the component cleaning facilities. By this low cost investment, the filtered cleaning bath is now piped back to the individual component cleaning facility and reused for the cleaning process. In the past, the cleaning baths were rejected during general cleaning works although the cleaning performance would have been sufficient for its further use. If this system has proved its worth, a downstream integration of a central membrane filtration plant as a holistic concept would be conceivable. An important prerequisite for this would be the consideration of all component cleaning facilities taking into account the following points:
152
P. Enderle et al. / Journal of Cleaner Production 34 (2012) 146e152
- Reduced contamination with impurities of upstream processes by the reduction of drag-outs, - Adaptation of cleaning chemicals (cleaning systems) and adjustments to the required component surface purity (definition of cleaning requirements), - Optimization and adaptation of existing bath maintenance systems (e.g. oil separators, micro filter, etc.), - Optimized dosage of cleaning chemicals and monitoring (continuously) of process stability and cleaning performance, - Optimized interaction between machine revisions/general cleaning work and the treatment of cleaning fluids/cleaning systems and - Consideration of possible negative side effects concerning biological conversion in the piping system. Concerning the adaption of existing bath maintenance systems, all 15 component cleaning facilities and the performance of their bath maintenance systems (mainly oil separators) were evaluated within the project. The oil separator system of one component cleaning facility with the highest throughput, wastewater flow and chemical demand was adapted during the project also including an automatic chemical dosage system to guarantee a constant concentration of surfactants and builder substances. Furthermore, the cleaning formulation was changed to a modular cleaning system. By these optimization measures, the lifetime of the cleaning baths could be increased from 5 to 25 weeks and the yearly operating costs could be decreased by 43% (basis 2010). To assess the total annual costs, the specific operational costs for chemicals, fresh water, demonized water, wastewater treatment and wastewater discharge were considered. 4. Conclusion The industrial water and energy systems consist of several subsystems which are more or less connected or influenced by each other in a more or less intensive way. This strong conjunction of water and heat management at industrial sites requires a systematic approach to design the best possible concept for the simultaneous reduction of water and energy. System boundaries and therefore communication boundaries between different divisions within one company have to be reflected in order to identify the main variables influencing the successful and extensive implementation of optimization measures. Considering the resources energy and water the individual influences between sub-processes were investigated within the potential study and limiting factors for a successful and extensive implementation were worked out. It has been demonstrated that the areas of heat recovery and process water reuse are closely connected and that optimization measures within one area have an immediate effect on the whole system. Therefore an intelligent combination of technology and system optimization is the basic requirement in order to develop practical optimization concepts through a systematic approach within complex production systems. However, when working out optimization measures to retrofit existing processes, the maximum efficiency is influenced through a variety of sub-processes, which limit the economical implementation potential as a whole. Thus, in many cases far-reaching adaptations of sub-processes have to accompany these optimization measures, to use e.g. the maximum potential of the available waste heat for heat recovery measures. Furthermore, the usage of process auxiliaries has far-reaching effects on process water reuse measures and therefore to the possibility to minimize the wastewater flow. Within this potential study, possible optimization measures were worked out, which could have a positive influence to the economical and ecological
performance of the investigated company. When working out the optimization measures, the focus was laid to measures with short payback periods. Acknowledgements The authors would like to thank the Austrian Research Promotion Agency (FFG) for their financial support. This project was supported by the Austrian Climate and Energy Fund in the frame of the research and technology programme “New Energies 2020”. References Adams, H.N., Jelinek, T.W., 1999. Cleaning and Degreasing in the Metal Industry, first ed. Eugen G. Leuze Verlag, Saulgau/Württ, Germany (in German). Alves, C., Ferrao, P.M.C., Silva, A.J., Reis, L.G., Freitas, M., Rodrigues, L.B., Alves, D.E., 2010. Ecodesign of automotive components making use of natural jute fibre composites. Journal of Cleaner Production 18 (4), 313e327. Brunn, K., 2001. Modular Aqueous Cleaning Systems-Different Fields of Application. SurTec Technical Letter 3. Surtec GmbH, Zwingenberg, Germany (in German). Christensen, T.B., 2011. Modularised eco-innovation in the auto industry. Journal of Cleaner Production 19 (2e3), 212e220. Cui, J., Roven, H.J., 2010. Recycling of automotive aluminium. Transactions of Nonferrous Metals Society of China 20 (11), 2057e2063. Deul, A.S., 2002. Systematic approach to water resource management in industry. In: Water Recycling and Resource Recovery in Industry e Analysis, Technologies and Implementation, vol. 252e270. IWA Publishing, London, UK. Duval, D., MacLean, H.L., 2007. The role of product information in automotive plastics recycling: a financial and life cycle assessment. Journal of Cleaner Production 15 (11e12), 1158e1168. EIPPCB (European IPPC Bureau), 2005. Reference Document on Best Available Techniques in the Smitheries and Foundries Industry. European Commission. Granata, G., Moscardini, E., Furlani, G., Pagnanelli, F., 2009. Characterisation of automobile shredder residue. Chemical Engineering Transaction 17, 427e432. Gruen, R., 2009. Cleaning Bath Maintenance and Treatment. ZVO report, special edition. Fachverband Industrieller Beschichter e.V., Hilden, Germany (in German). Gruen, R., 1999. Cleaning as a Part of the Heat Treatment. SurTec Technical Letter 13a. Surtec GmbH, Zwingenberg, Germany. van den Hoed, R., 2007. Source of radical technological innovation: the emergence of fuel cell technology in the automotive industry. Journal of Cleaner Production 15 (11e12), 1014e1021. Jehle, W., Staneff, Th., Wagner, B., Steinwandel, J., 1995. Separation of glycol and water from coolant liquids by evaporation, reverse osmosis and pervaporation. Journal of Membrane Science 102, 9e19. Keudel, L., 2011. Treatment of wastewater from the automotive industry. In: Proceedings DWA Conference Behandlung von Industrie- und Gewerbeabwasser. Kassel, Germany, 2011, pp. 322e345 (in German). Kjaerheim, G., 2005. Cleaner production and sustainability. Journal of Cleaner Production 13 (4), 329e339. Lugscheider, E., Barimani, C., Guerreiro, S., Bobzin, K., 1998. Corrosion tests of PVD coatings with die lubricant used for Al high-pressure die-casting. Surface and Coatings Technology 108e109, 408e412. Mueller, W., 1998. Corrosion at Die Casting Facilities When Using Aqueous Products. Technical Letter. Geiger þ Co. Schmierstoff-Chemie GmbH, Heilbronn, Germany (in German). Neto, B., Kroeze, C., Hordijk, L., Costa, C., Pulles, T., 2009. Strategies to reduce the environmental impact of an aluminium pressure die casting plant: a scenario analysis. Journal of Environmental Management 90 (2), 815e830. Orsato, R.J., Wells, P., 2007. U-turn: the rise and demise of the automotive industry. Journal of Cleaner Production 15 (11e12), 994e1006. Passarini, F., Ciacci, L., Santini, A., Vassura, I., Morselli, L., 2012. Auto shredder residue LCA: implications of ASR composition evolution. Journal of Cleaner Production 23 (1), 28e36. Pusavec, F., Kramar, D., Krajnik, P., Kopac, J., 2010. Transitioning to sustainable production e part II: evaluation of sustainable machining technologies. Journal of Cleaner Production 18 (12), 1211e1221. Reynaud, A., 2010. Corrosion of Cast Irons. Sheir’s Corrosion 3: Corrosion and Degradation of Engineering Materials 1737e1788. Tharumarajah, A., 2008. Benchmarking aluminium die casting operations. Resources, Conservation and Recycling 52, 1185e1189. Telukdarie, A., Buckley, C., Koefoed, M., 2006. The importance of assessment tools in promoting cleaner production in the metal finishing industry. Journal of Cleaner Production 14 (18), 1612e1621. Wyman, H., Mueller, H., 2002. The new 3 face grinding blade design for dry-cutting. Technical Magazine of the Gearing Partners Klingelnberg, Liebherr and Oerlokon, sigma report 12, 21e29. Zapata, C., Nieuwenhuis, P., 2010. Exploring innovation in the automotive industry: new technologies for cleaner cars. Journal of Cleaner Production 18 (1), 14e20.