Eco-efficiency of Laser Welding Applications

Stefan Kaierle*, Martin Dahmen, Okan Gdkkurt

Fraunhofer-Institut for Laser Technology, Steinbachstrasse 15, D-52074 Aachen, Germany

Abstract As widely known laser materials processing has some advantages regarding local heat input and controllability. In many fields applications were developed which are not accessible for conventional thermal processing. In other fields laser-supported manufacturing techniques are a valuable alternative. On the one hand laser techniques enable increased processing speed and less post-processing, leading to an increased productivity. On the other hand low efficiencies in the energy conversion seem to be a major drawback and apparently limit the range of applications. In the frame of conventional processing schemes laser beam welding requires a high utilization in order to run economically. Main advantages lie in the reduced consumption of material and the reduced efforts in post processing. Because of the locally concentrated heat input process emissions are lower which reduces energy and material consumption in the auxiliary chain. To make full use of the often-conjured flexibility a multitude of manufacturing schemes had been developed and adapted. In order to appraise the versatility of laser driven processing techniques a cost and benefit analysis based on a life-cycle approach is conducted including both, economics and ecology. Eco-efficiency is rated by a variation of the BASF method. Taking into account the reduced consumption of consumables, reduced effort for preparation and post-processing, and focusing on specific application ranges a positive environmental impact can be proven.
Keywrds: Laser beam welding, Energetic efficiency, Ecologic appraisal, BASF method, Green Supply Chain, Eco-efficiency

1. Introduction In the 21st century not only in the Europe but also in the USA and other countries the government and the public became more and more aware of environmental problems and started to demand more effort from the companies to improve their sustainability, to reduce their energy and material consumption or to reduce waste of their manufacturing processes and supply chain. The new trend is called The Green Supply Chain and it combines the traditional supply chain management with the environmental issues and concerns [1]. Figure 1 illustrates the green supply chain model and points out the various stages where energy and material are used, wastes occur and opportunities exist to limit waste by reuse, recycling, remanufacturing and also by deciding for a more innovative process. The manufacturing is the best possible stage where the environmentally friendly innovations like implementation of laser applications may be utilized. The reason is that the organization can more directly see the impact and the benefits of implementing environmentally friendly processes [2]. The manufacturing industry is aware that they are requested to operate economically. In terms of sustainability this means to make responsible use of resources according to the economic principle. Instead of maximum profit from minimum capital the term should be superseded by maximum results (economically and ecologically) from minimum resources. The necessity of efficient application of resources advances further into the focus of corporative tasks in economy, research and politics (Fig. 2). Within the framework of manufacturing the question has to be raised which possibilities exist to reduce the input of resources as well as the emissions [5].

* Corresponding author. Tel.: +49 241 8906-212; fax: +49 241 8906-121. E-mail address: kaierle@ilt.fraunhofer.de.

Figure 1. Functional model of an organizational supply chain with environmentally influential practices [2]

Against the background of statements like . . . lasers are your best friend for a greener planet! [7] the ecological impact of lasers in materials processing will be investigated. Indeed laser materials processing fulfils most of the conditions for resource and energy efficient manufacturing due to its specific characteristics like concentrated energy input and high thermal efficiencies.
Added value at preserved assets

Business case
Eco efficiency Eco effectiveness Socio efficiency

Sustainable Development

Socioeffectiveness

Natural case
Conservation of Vital critical natural goods Sufficiency

Societal case
Ecological equity Preservation/ conservation of action potential

Figure 2. The tree cases of mercantilism and production and their interlinks [3]

In order to identify the actual position a basic appraisal of laser beam welding techniques was conducted. The approach is based mainly on the energy consumption of the joining process. Autogenous laser beam welding is compared with arc welding techniques like MAG or plasma welding applied on the same task. The comparison includes pre-fabrication steps and their material-intensity. The results reported show a detail of a complete/holistic review which is currently conducted and will follow later. The well known business case tries to monitor and evaluate transfer and transformation of the factors of production in units of money flow. Evaluation methods in trade and industry are matured over a couple of hundreds of years and became more and more reliable. Assessment of the natural case was introduced recently and is characterized by a great physical complexity because of different entities have to be included [8]. An indicator system needs to be developed in order to make the different entities comparable. Numerous methods have been developed with specific emphasis [9]. The social case is investigated on a micro to meso level [10] assessing the balance of the impact on workers and society. Assessment methods for the investigation of the social impact still need to be developed especially with regard to the interfaces to the other assessment areas.

2. Methods of appraisal

2.1. A short definition of Eco-Efficiency In a short definition of Eco-Efficiency the term can be expressed as the ratio between economic creation to ecological destruction [3]. Indeed, in literature exist different approaches to this definition. The most common are basically the following three [4]. Firstly, an economic approach which is strongly related to cost efficiency (this is related to the question under which conditions a given economic resource yields the maximum outcome; however, this is not much related to ecology). Secondly, eco-efficiency can be seen just from an environmental view as ecological efficiency. This means in fact the ratio between undesired emissions and the desired products or services as a result by them. Lastly, eco-efficiency can be defined as economic-ecological efficiency. In this regard, eco-efficiency can be interpreted as cross-efficiency linking environmental with economic issues and measuring the environmental impact added caused per monetary unit earned. As the latter seems to be the most widely used definition and is also supported by the World Business Council for Sustainable Development (WBCSD) we will take up this definition for our further study as well. In the following, we first discuss different existing methods and compare them to our problem addressed. 2.2. Factor Four and Factor Five The method, published in 1995, is basically an economic concept aiming at suited technologies and processes using a lower amount of resources to meet the human need for goods and services [12]. Rather than a master plan Factor Four is a rule of thumb enabling the relative assessment of approaches for innovative processes within complex settings. Based on five quantitative and five qualitative indicators best practices are identified. The environmental factor MIPS (Material Input per Service unit) is calculated as a comparative measure against a reference, e.g. a state-of-the-art product. Targets of this method are single products, cooperation processes, and driving sectors. The continued development of the Factor Four method resulted in a method which allows savings of up to 80% compared to nowadays values [13]. New with this method is the inclusion of sufficiency defined as the effects of selfrestriction in the use of goods. 2.3. The BASF method and the LCA methodology Based on the conceptual work of von Weizscker [12] and Fussler [14] a pragmatic approach of eco-efficiency was developed by BASF [15]. In the centre of this analysis stand the specific customer benefits. Efficient solutions are regarded as those that meet the customers requirements closer than others from the viewpoint of financial costs and the environment. A life cycle analysis is conducted, first focused on the entire life cycle, but then concentrated on specific events where the alternatives under consideration differ. This analysis is based on the environmental profile obtained from the data of the plants. Extending this approach a life cycle assessment (LCA) is obtained. Adding additional assessment criteria followed by an economic assessment leads to the eco-efficiency analysis (Fig. 3). The environmental impact is determined on the basis of the main aspects: raw materials consumption energy consumption emissions into air, water and soil (wastes) toxic potential of the substances used and released abuse and risk potential Data acquisition and calculation is carried out according to ISO 14040. The results are represented in individual graphs or the respective category. In a similar manner the total costs over the life cycle are calculated. This comprises the real (actual) costs as well as the subsequent costs expected to occur in the future.

Definitionof customers benefit

Cost assessment in every livecylcle step

Calculation of aggregated LCcosts

Cost scaling

Identification of product/processes

Lifecycle inventory analysis

Quantification of action categories

Compilation of life cycle

Assign LCinventory to action category

Scaling of environmentalloading

public ENISO14040 43

Clusteringof action categories inLCsteps

Nonpublic according to ISO

Economical assessment

Figure 3. Instruments for internal eco balancing

Figure 4 shows an evaluation chart suited for relative appraisal within the framework of EN ISO 14040. This is used to illustrate a product rating described in section 4.2. A more recent version focused on the evaluation in a production environment and includes energy and materials consumption, preparation a post processing effort, and hazardous emissions. This model is illustrated in section 4.1 at the example of manufacture of T-beams. A clear advantage in applying the BASF method lies in the fact that the comparability between different methods and indicators can quickly be visualized and are therefore easy to compare. However, the difficulty is the correct setting of the indicators, i.e. especially their correct order of magnitude. This is important to have the centre of gravity of the footprint also in the centre of the graphs.
Energy consumption Energy source Process efficiency Manipulation Materials consumption Consumables Scrap Debris Emissions Process emissions (Aerosols, gases) Hazardous substances Risk potential Hazardous substances Optical hazards Electrical hazards Risk potential Post processing

Energy consumption
1,0

0,5

Materials consumption

Emissions

Figure 4. Indicators and coordinate system for their relative appraisal

3. Efficiencies Although laser beams have excellent properties which make them well suited for a wide range of processes the efficiencies of the conversion of electricity to optical power is limited. The main reason is found in the level distances where a radiative transition is possible. So is the process efficiency of a CO2 laser process according to the term schema

approximately 48% at the emission wavelength of 10.6 m. Losses induced by pumping and cooling lead to an overall efficiency of maximum 15 % for high-power units. Figure 5 gives an overview over the efficiencies by up-to-date laser beam sources used in macro materials processing. Nd:YAG (LPSSL) lasers show the lowest efficiency at 2.5 to 4%. Significant improvement could be obtained by diode pumping and resonator design. The latest improvements were made with the fibre and disk lasers where pumping of small volumes and fitting the pumping wavelength to the characteristics of the active material leads to efficiencies of up to 25%. Because of the direct conversion of electrical energy to optical radiation diode lasers have the highest efficiencies. The limited beam quality which made these lasers suited for mainly hardening and deposition welding had been improved by beam shaping and fibre transfer enabling a wider field of applications, further increases in beam quality can be expected shortly.
35 30 HPDL FC-HPDL Disk Fibre Fibre mono M. CO2 DPSSL

Efficiency e-o/1

25 20 15 10 5

LPSSL 0,3 1 3 10 30 100 300

Beam Parameter Product BPP/mrad

Figure 5. Typical electro-optical efficiencies of lasers used in materials processing dependent on the beam parameter product

Besides the efficiencies of the energy supply the welding processes have their characteristics. MAG welding requires an edge preparation in order to prepare the edge according to welding standards. Thus there is an effort to connect to milling or grinding and also imposes a loss of material. The deficit volume has to be filled with material during the welding process again.
Table 1 Rates of particulate emissions for selected welding processes
Emission e/mg s-1 Stick welding MAG MIG TIG atmospheric EAuBW autogenous LBW Hybrid LBW 3.9 7.6 2.5 30.4 3.03 31,96 0.038 - 3 22 - 32 2.6 - 8 4.6 18,5 0.17 0.28 0.1 0.25 0.4 Particle size dHe/m 0.34 0.45 0.13 0.45 0.32 0.42 Emission p. unit length e/mg m-1 0,33 0 63 0,08 1.01 0.1 1.06 0.002 0.17 0.46 0.67 0.03 0.09 0.04 0.15

In Table 1 the rates of emissions or airborne substances are given for a selection of welding processes. The great variations are caused by the wide range of metals processed [16]. Lowest rates are obtained in welding stainless steels and Nickel base alloys, highest in welding Aluminium and Zinc-coated steel. Particles emitted are found as chains or agglomerates of primary particles with typical diameters in the order of magnitude of 10 nm. The agglomerated particles reach sizes of typical median equivalent diameters between 0.1 and 0.4 m. Because of particles of this size can be deposited in the respiratory tract immission control is needed. Dividing the emission rates by a typical

production speed for each welding technique delivers an estimate of the emission per Meter length welded (table 1 right column).
Table 2 Efficiencies of different welding processes
TIG Transformation u 0.9 el./el. Generation G 0.8 arc Coupling K Total supply ges max. Penetration ts/mm max. aspect ratio 0.85 0.7 4 0.5 MAG 0.750.85 el./el. 0. arc 0.85 0.58 8 0.5 MIG 0.9 el./el 0.8 arc 0.85 0.61 6 0.5 EB 0.7-0.85 el./electron 0.85 Electron beam 0.85-0.95 0.68 30 50 LB 0.04-0.3 El./light 0.96 Light beam 0.75-0.95 0.27 22 15 0.85 Arc + light beam 0.85 0. 2 25 10 LB/MAG 0.58 Plasma 0.8 el./el. 0.85 Plasma jet 0.6-0.85 0.58 12 5

In order to compare the technical aspects of welding techniques the efficiencies of various welding techniques are compiled in Table 2. The total efficiency of a welding process is defined by the product of the efficiencies of transformation, i.e. the generation of the processing power from the net current, generation, i.e. the production of the primary energy carrier, and coupling into the work piece to be welded. For several arc welding processes the latter two are defined as the thermal efficiency for which standardized values exist [15]. The values given in SEW088, basically used for calculating the cooling time, are relative values compared against the thermal efficiency of submerged arc welding which was set to 1. More recent studies [16] reveal that the absolute efficiencies deviate from these estimate values and are depending on the operation conditions. For assessing the energy flow in weld fabrication absolute values are required to ensure the maximum accuracy of the result. 4. Examples The assessment is illustrated at two example processes, the welding of T-beams and the cutting and welding for production of flanges (Fig. 6). Material in both groups of parts is stainless steel of different grades or Nickel based alloys for some flanges. Autogenous laser beam welding with a high-power CO2 laser at power of up to 12 kW had been applied to join the parts. The T-beams have a stringer thickness between 10 and 30 mm, a stringer thickness was set for the calculated case. Sheet thickness of the cooling ring lid is 2.5 mm. In both cases the half stuff comes as cold rolled sheet material. Figure 6. Illustration of the examples left: welding process and macro sections for welding of T-beams right: ring cutting and welding of flanges

With regard to the manufacturing sequence the investigation starts after delivery of the plate material and ends before the final finishing operation, because those both are identical for al routes.

4.1. Welding of T-beams T-beams used for tray suspension in rectification columns. In order to avoid damage by hydrogen a connection over the full cross section is required. Manufacture using two welding processes is assessed: the conventional route using MAG welding (case 1) and the laser route using laser beam welding as joining process (case 2). The processing sequence investigated starts with cutting the stringer from the plate material. Case 1 uses plasma cutting affording joint edge finishing either by grinding or milling as a HX preparation with 30 bevel angle is required. Case 2 uses laser beam cutting. For the calculation a 2 mm wide seam cut is foreseen which might be used to compensate deformations caused by residual stresses in the material. As a square joint is preferred for autogenous laser beam welding slight leveling or the roughness by grinding may be applied as the final step in edge preparation. MAG welding is done with a self-consuming electrode adding again material to and filling the grove. Material consumption is more than doubled taking into account a throat thickness of 8 mm. Autogenous laser beam welding used no consumables, hence, the material consumption is restricted to the edge preparation. After normalising the results are displayed in the pentagram as shown in figure 7. In general case 1 leaves the bigger footprint. Regarding the energy consumption MAG welding and laser welding are very close to each other despite the fact that the efficiency of laser beam sources is much worse than that of MAG sources. This almost equal result reflects that laser welding takes place in a single pass where as MAG welding needs multi pass processing.
Table 3 Evaluation criteria and characteristic values for the manufacture of T-beams with 20 mm stringers
Unit Edge preparation Material loss Energy for processing Energy consumption Welding Lead time Material consumption Energy for welding Energy consumption Total mass Total energy h kg/m kJ/m kJ/m kg/m kJ/m kg/m kJ/m kJ/m Case 1 PC/milling 0,453 5040 6300 MAG 0,47 0,954 2671 7604 1,407 13904 Case 2 LBC /(grinding) 0,408 857 6121 LBW 0,061 0 1070 7636 0,408 13757

The total material loss caused by edge preparation and by filler wire addition amount to 1.407 kg/m for case 1. In case 2 the waste material production is 0,408 kg due to the fact that a seam cut is applied. If the seam cut is taken less wide material consumption is reduced, e.g. 0.126 kg/m for a seam width of 1 mm. In consequence case 2 has a material intensity of approximately 21 % compared to case 2. Due to the lesser mass that has to be removed the work effort is reduced to 30% in the laser case. The emission of hazardous airborne substances is calculated from the data shown in table 3. Small emission rates due to the deep penetration welding process as well as short lead times have a significant decrease of the total emissions as consequence. Further benefits can be added regarding the reduced effort for extraction and filtering. Post processing is arduous in case 1. Semi-finishing requires brushing and grinding before the beam itself has to be stress relief annealed in an oven. After (mostly mechanical) straightening the whole piece has to be postheated again. Case 2 is easier to handle because finishing required only brushing, straightening is not necessary, and if heat treatment becomes necessary it can be applied locally. This leads to a reduction of the post processing effort of about 50%. In this calculation the energy consumption is not yet included.

Figure 7. Comparison of the ecological footprint of welding stainless steel T-Beams

4.2. Manufacture of flanges The second example is manufacturing of flanges. The body of the flange will be produced on a lathe where a 15 mm deep groove for water cooling is sunk into the upper side of the flange. Rings are cut by laser radiation exactly fitting into the actual groove. These rings will be inserted into the grove and welded by two seams. In order to compensate distortion bead-on-plate welds on the flat side of the flange will be produced.
Autogenous LBW Plate cutting Cut edge processing Joint edge prep flange sheet Welding (2 seams) Feed rate Power Material consumption Consumables Gas Aersol emission rate Distortion Longtudinal bend Transverse angular Energy consumption Laser beam cutting n.a. turning 1 pass 2,8 m/min 2,5 kW n.a. Helium 20l/min 0.7 mg/s 0,5 mm 2 0,743 MJ/m TIG welding Water jet n.a. turning milling (bevel) 1 pass 0,7 m/min 4 kW 1,5 m/min (1,2 mm) Argon 18 l/min 1,5 mg/s 2 mm 4 1,19 MJ/m

Figure 8. Rating scheme for laser beam welding vs. TIG welding in flange production

The comparison is made between the technique described above and a combination of water jet cutting and TIG welding with consumables. For laser beam welding a beam source with nominal output power of 2,6 kW was used. Fig. 8 shows the rating scheme in physical units. TIG welding requires bevelling of at least one of the joining edges. After welding distortion caused by the great arc energies makes mechanical straightening and post-weld heat treatment necessary. No such steps are required in the case of laser beam welding. The assessment graph shown in Fig. 9 indicates a clear advantage of laser-based manufacturing in all categories.
Energy consumption
1,0

0,5

Materials consumption

Emissions

Laser beam welding TIG welding Risk potential Postprocessing Plasma welding (square butt joint)

Figure 9. Comparison of the ecological footprint of welding stainless steel flanges

The effort of post-processing is reduced by the localised heat input on the one side and the integrated straightening step. Risk potential is minimised due to the fact, that the process runs in an encapsulated environment which becomes necessary for reasons of radiation shielding. Materials consumption is minimum for plasma and laser beam welding because only material of the joining partners is used for the joining process. Energy consumption is low with the better efficiency of the low powered beam source. Also integrating both manufacturing steps on one machine contributes to the positive balance. Emissions are lower according to the data from table 1.

5. Discussion The examples discussed show the potential of laser materials processing to contribute to sustainable manufacturing. Dematerialisation is possible by material-conserving edge preparation and the limited use of consumables. Localised heat input at low energies per unit length leads to less effort for post-processing. Considerable savings of energy and costs for heat treatment are possible. Using fiber or diode lasers instead of CO2 lasers will decrease the energy consumption. Also by the use of hybrid welding techniques significant increase in output and as a consequence reduced energy savings per piece is possible. Improvements by use of different laser systems in terms of increased efficiency, as using fibre or disk laser instead. Furthermore the intrinsic properties enable an efficient process control. Via an in-situ one-hundred percent supervision a production with zero rejects comes into reach. By this a considerable contribution to resource-conservation can be obtained. With enhanced understanding of the processes in conjunction with process control closed loop control of manufacturing becomes possible. With the emergence of self-regulating systems not only the set-up times can be reduced but also the safety of operations and processes will be improved.

6. Conclusions A first step toward an enumerated approach of the eco-friendliness of laser material processes is presented. It was found that the material intensity of laser beam welding is lower than for arc welding techniques. Also the occupational risks are modest. Depending on the specific task a considerable increase of energy consumption was found. At least the relative assessment shows that the energy take-up can be reduced by using modern laser beam sources. Thorough process development in combination with process control will lead to further savings. Next stages in research of the present topic will be the development of general dimensionless indicators, integrating economical and social aspects, and finally to develop a closed assessment model. The results will show new directions of research on beam sources, application, processes, and approaches to integration into manufacturing. Tapping out the full potential of the technique will finally lead to the Factor Four doubling wealth, halving resource use.

References
[1] Gilbert, S. Greening supply chain: Enhancing competitiveness through green productivity. Tokyo: Asian Productivity Organization, 2000. [2] Sarkis, J. A Strategic Decision Framework for Green Supply Chain Management. Journal of Cleaner Production, 2003, 22, 397-409. [3] K. Hungerbhler, J. Ranke, T. Mettier: Chemische Produkte und Prozesse. Springer Verlag Berlin, ISBN 3-540-64854-2, 1999 [4] A. Kicherer, S. Schaltegger, H. Tschochohei, B. Ferreira Pozo: Eco-Efficiency. Combining Life Cycle Assessment and Life Cycle Costs via Normalization. Int J LCA 12 (7) 537543 (2007) [5] R. Neugebauer: Energieeffizienz in der Produktion, Fraunhofer-Gesellschaft, Mnchen 2008 [6] L. Lachnit, C. Lange, M. Pallocks (eds.), Zukunftsfhiges Controlling, Mnchen 1998 [7] P. Baker: Lasers for a greener planet, LIA Today, December 2008 [8] C. Lang-Koetz, D. Heubach: Umweltcontrolling umsetzen Erstellung von Kennzahlen fr Stoff- und Energiestrme und deren Integration in die betriebliche IT, Fraunhofer IRB Verlag, Stuttgart 2007 [9] T. Jger, C. Karger: Instrumente zur Nachhaltigkeitsbewertung, Forschungszentrum Jlich, Jlich 2006 [10] H.A. Becker: Social impact assessment, European Journal of Operational Research 129, 311-321 (2001) [11] ISO Standard fo Life Cycle Assessments (ISO 14040), 1997 [12] E.U. v. Weizscker, A.B. Lovins, L. Hunter, Factor Four. Doubling Wealth - Halving Resource Use, Earthscan, London 1997 [13] E.U. v. Weizscker, K. Hargroves, M.H. Smith, C. Desha, P. Stasinopoulos: Factor Five: Transforming the Global Economy through 80% Improvments in Resource Productivity, Earthscan, London 2009 [14] C. Fussler, P.James: Driving eco innovation, Pitman Publishers, London 1996 [15] P. Saling, A. Kicherer, B. Dittrich-Krmer, R. Wittlinger, W. Zombik, I. Schmidt, W. Schrott, S. Schmidt: Eco-efficiency analysis by BASF: The Method, Int. J. LCA 2002, [16] V.E. Spiegel-Ciobanu: Vergleichende Untersuchungen bezglich der Charakterisierung der ultrafeinen Partikel beim Schweien und bei verwandten Verfahren Kurzfassung des Abschluberichtes, Berufsgenossenschaft Metall Nord Sd, Hannover 2008 [17] SEW 088: Schweigeeignete Feinkornbausthle; Richtlinien fr die Verarbeitung, besonders fr das Schmelzschweien, Beuth Verlag, Berlin 1993 [18] IFS, TU Chemnitz: Bestimmung von Wirkungsgraden moderner Schutzgasschweiverfahren; Statusbericht, Chemnitz 2010