nickel chloride battery

Journal of Cleaner Production xxx (2013) 1e10

Contents lists available at ScienceDirect

Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro

Life cycle assessment of storage systems: the case study of a sodium/nickel chloride battery Sonia Longo a, *, Vincenzo Antonucci b, Maurizio Cellura a, Marco Ferraro b a Dipartimento di Energia, Ingegneria dell’Informazione e Modelli Matematici, Università degli Studi di Palermo, Viale delle Scienze Ed. 9 - 90128 Palermo, Italy b Consiglio Nazionale delle Ricerche, Istituto di Tecnologie Avanzate per l’Energia “Nicola Giordano”, salita S. Lucia sopra Contesse, 5 e 98126 Messina, Italy

a r t i c l e i n f o

a b s t r a c t

Article history: Received 9 January 2013 Received in revised form 27 September 2013 Accepted 1 October 2013 Available online xxx

This study assesses the energy and environmental impacts of sodium/nickel chloride batteries, one of the emerging battery technologies for energy storage and smart grids. The analysis was conducted using the Life Cycle Assessment methodology according to the standards of the ISO 14040 series. The study system was one sodium/nickel cell battery providing electric storage for a photovoltaic system, and the manufacturing, operation, and end-of-life steps were analysed. The results indicated that the operation step has the greatest energy impact (55e70% of the total), with the manufacturing step, particularly cell manufacturing, contributing the greatest environmental impact (>60% of the total). This paper makes two original contributions: 1) it presents one of the first LCA analyses of sodium/ nickel chloride batteries with the aim of identifying the energy and environmental impacts of this technology; 2) it provides a set of energy and environmental outcomes identifying the “hot spots” of the selected technology that must be carefully considered to upgrade the current efficiency and sustainability of electric storage device standards. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Life cycle assessment Energy storage systems Na/NiCl2 battery ZEBRA Environmental impacts

1. Introduction European energy policy has three objectives: fighting climate change, limiting Europe’s dependence on imported hydrocarbons, and providing secure and affordable energy to consumers (Commission of the European Communities, 2007a). Electricity storage technologies play an important role in facilitating more secure, efficient and sustainable energy sources and forms of energy use (Naish et al., 2008) and in the development of the European power system by increasing the market share of renewable energy and distributed energy generation (European Commission, 2012a). The European Commission stressed the importance of future research on electricity storage (Commission of the European Communities, 2007b; European Commission, 2012b). In particular, the development of electricity storage systems is included in the electricity grid initiative launched by the European Commission (Entsoe, 2010) and in a list of strategic energy technologies that need to be developed and given higher priority in future research (European Commission, 2012b). * Corresponding author. Tel.: þ39 091 23861977; fax: þ39 091 484425. E-mail address: [email protected] (S. Longo).

Electrical energy storage is a priority due to the intermittent or variable nature of most forms of low carbon energy generation (Beccali et al., 2008), in contrast with the traditional fossil fuel dominated electricity network. Electrical energy storage offers the potential to store generated electricity and subsequently match supply with demand as required (Naish et al., 2008). Electricity storage technologies will prepare the electricity grid at all voltage levels for the massive increase in small-scale decentralised and large-scale centralised renewable electricity (European Commission, 2010a) and can help to achieve more sustainable forms of goods production, energy use and mobility. Batteries allow for reductions in energy consumption and carbon dioxide emissions when used for partially and fully electrified vehicles, and represent a storage option for energy generated during off-peak periods by photovoltaic systems and wind turbines (Sullivan and Gaines, 2010; McManus, 2012). The battery market is already large and has developed to serve mobile and stationary applications. The European and Middle Eastern electricity storage market is approaching 1 billion euro annually (Eurobat, 2012). Improving performance standards will make market demand relevant in the coming decades. The sodium/nickel chloride battery or ZEBRA (Zero Emission Battery Research Activities) battery (Parkhided, 2006) is an

0959-6526/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jclepro.2013.10.004

Please cite this article in press as: Longo, S., et al., Life cycle assessment of storage systems: the case study of a sodium/nickel chloride battery, Journal of Cleaner Production (2013), http://dx.doi.org/10.1016/j.jclepro.2013.10.004

2

S. Longo et al. / Journal of Cleaner Production xxx (2013) 1e10

Nomenclature Ac BMI CV DC FE GER FU GWP HT LCA LU

acidification battery management interface coefficient of variation direct current freshwater eutrophication global energy requirement functional unit global warming potential human toxicity life cycle assessment land use

innovative energy storage system with applications in electric cars, vans, buses and hybrid vehicles, and marine technologies. Recently, ZEBRA batteries have been used as storage devices in electrical networks and power grids, as telecommunications back-up power, in direct current (DC) supply from photovoltaic and wind generators, and for load levelling (Manzoni et al., 2008; Sudworth, 2001). The raw electrode materials for ZEBRA batteries are plain salt and nickel, in combination with a ceramic electrolyte and a molten salt. Batteries consist of individual cells enclosed in a thermally insulating package. During the cycling of the battery, internal resistive losses allow for an average operating temperature of 270  C. When the battery stands idle for prolonged periods (exceeding 24 h), additional heating is required to keep the battery warm (Matheys et al., 2004). The characteristics of ZEBRA batteries are almost independent of ambient temperature, and there is effectively no lower temperature limit for battery operation. In addition, as ZEBRA batteries operate at an average temperature of 270  C, their use at extremely cold or hot ambient temperatures does not lead to any detrimental effects. This is an advantage with respect to conventional battery systems such as lead-acid batteries, which require more elaborate thermal management at extreme temperatures to avoid reduced battery performance (Parkhided, 2006). The assessment of the real advantages of using ZEBRA batteries for energy storage must include an analysis of the energy and environmental impacts during the life cycle of these systems. Even if these batteries have no direct emissions during the operations step and are charged with electricity produced by renewable energy technologies, they cannot be considered totally clean. In fact, batteries consume energy and cause environmental impacts during their life cycle that cannot be neglected. The life cycle thinking approach allows the resource use (raw materials and energy) and environmental burdens related to the full life cycle of the technology to be taken into account (Beccali et al., 2012). This paper applied the Life Cycle Assessment (LCA) methodology to assess the energy and environmental impacts related to the life cycle of ZEBRA batteries. This paper makes two original contributions: 1) it presents one of the first applications of LCA to ZEBRA batteries with the aim of identifying the energy and environmental impacts of these technologies; 2) it provides a set of energy and environmental outcomes and identifies the “hot spots” of this technology that must be carefully considered to improve upon the current efficiency and sustainability of electric storage device standards. 2. LCA of batteries: state-of-the-art To the authors’ knowledge, the only LCA of sodium/nickel chloride batteries was carried out within the SUBAT project

MDV ME MV NRE ODP POF PV RE SD TCB TE WRD ZEBRA

median value marine eutrophication medium value non-renewable energy consumption ozone depletion potential photochemical ozone formation photovoltaic panel renewable energy consumption standard deviation thermo-compression-bonded terrestrial eutrophication water resource depletion zero emission battery research activities

(Matheys et al., 2004, 2007, 2008; Van den Bossche et al., 2006). This project investigated the environmental impacts (expressed in eco-indicator points) of five different battery technologies used for electric vehicles, including lead-acid, nickelecadmium, nickele metal hydride, lithium-ion, and sodium/nickel chloride batteries. The authors selected as functional unit (FU) “a battery enabling the vehicle to cover a specific range (60 km) when driving up to 80% of the depth-of-discharge of the battery”, and followed a “cradle to grave” approach, including the manufacturing of the battery, the operation (energy losses due to the battery mass and energy efficiency), and recycling. The results showed that for all studied systems, the manufacturing step has the greatest impact. The manufacturing of lead-acid battery has the highest impact (1091 eco-points), followed by nickelemetal hydride (945 ecopoints), nickelecadmium (861 eco-points), sodium/nickel chloride (368 eco-points), and lithium-ion batteries (361 eco-points). Considering the entire life-cycle and including the negative environmental impact of the recycling process, the nickelecadmium battery has the highest impact (108 eco-points), followed by leadacid (100 eco-points), nickelemetal hydride (97.7 eco-points), lithium-ion (55.2 eco-points), and sodium/nickel chloride batteries (46.5 eco-points). Sullivan and Gaines (2010) presented an interesting literature review of LCA studies of batteries. The authors examined cradleto-grave studies on lead-acid, nickelecadmium, nickelemetal hydride, sodiumesulphur, and lithium-ion battery technologies. The results reveal considerable variation in the primary energy consumption for each battery technology due to location effects, dated and missing information, the compilation of data from numerous sources, the effects of different battery applications, and uncertain material requirements and manufacturing processes. The magnitude of primary energy consumption and greenhouse gas emissions increase in the following order: lead-acid (intermediate value: approximately 25 MJ/kg and 91 kg CO2eq/kg), nickelecadmium (intermediate value: approximately 100 MJ/kg and 240 kg CO2eq/kg), lithium-ion (intermediate value: approximately 170 MJ/kg and 357 kg CO2eq/kg), sodiumesulphur (intermediate value: approximately 180 MJ/kg and 566 kg CO2eq/kg), and nickelemetal hydride (intermediate value: approximately 200 MJ/kg and 524 kg CO2eq/kg). Other LCA studies have been performed that were not included in the review by Sullivan and Gaines (Majeau-Bettez et al., 2011; Schexnayder et al., 2001;Samaras and Meisterling, 2008; Notter et al., 2010; Zackrisson et al., 2010; McManus, 2012). The results of these studies in terms of global warming potential (GWP), global primary energy requirement (GER) and non-renewable primary energy requirement (NRE) are provided in Table 1.

Please cite this article in press as: Longo, S., et al., Life cycle assessment of storage systems: the case study of a sodium/nickel chloride battery, Journal of Cleaner Production (2013), http://dx.doi.org/10.1016/j.jclepro.2013.10.004

S. Longo et al. / Journal of Cleaner Production xxx (2013) 1e10

3. Case study: LCA of a sodium/nickel chloride battery

3

Table 1 LCA studies on batteries.

3.1. Goal and scope definition

Battery typology

System boundaries

Impacts per kg of battery

References

3.1.1. Goal of the study The main goals of this study are to assess the energy and environmental impacts of a sodium/nickel chloride battery (ZEBRA) considering two different operational scenarios; to evaluate the contribution of each life-cycle step to the total impact; to assess the uncertainty of the final results by a Monte Carlo analysis; to demonstrate the variation in the impact of the operation step caused by different methods of electricity production; to assess the energy and environmental benefits of using electricity produced by renewable energy technologies during the operation step instead of electricity obtained from the grid, which is generated by renewable and non-renewable energy sources. The results of this analysis will help battery manufacturers to identify the “hot spots” for further investigation to improve the energy and environmental performance of these batteries. This study applies the LCA methodology as regulated by the international standards of series ISO 14040 (ISO 14040, 2006; ISO 14044, 2006), and follows an attributional approach in which the life cycle is modelled by depicting the existing supply-chain of the product and including the input and output flows of all system processes as they occur (European Commission, 2010b). The attributional approach is a static approach, and the results describe the system over a short period without accounting for possible future variations in electricity production or improvements in the characteristics and performance of renewable energy technologies.

LiFePO4

Manufacturing

Nickel cobalt manganese Li-ion Li-ion

Manufacturing

GWP: 22 kg CO2eq NRE: 192.6 MJ GWP: 22 kg CO2eq NRE: 196.8 MJ

Majeau-Bettez et al. (2011) Majeau-Bettez et al. (2011)

Manufacturing

GWP: 12 kg CO2eq GER: 170 MJ

Manufacturing

GWP: 12.5 kg CO2eq GER: 90 MJ

Samaras and Meisterling (2008) McManus (2012)

Manufacturing

GWP: 4.4 kg CO2eq GER: 88 MJ

McManus (2012)

Manufacturing

GWP: 0.9 kg CO2eq GER: 17 MJ GWP: 6 kg CO2eq NRE: 104 MJ GWP: 41.04 kg CO2eq

McManus (2012)

Manufacturing and operation

GWP: 31.71 kg CO2eq

Zackrisson et al. (2010)

Manufacturing, operation and end-of-life Manufacturing, operation and end-of-life

GWP: 54.6 kg CO2eq GER: 732.7 MJ

Schexnayder et al. (2001)

GWP: 40.5 kg CO2eq GER: 566.3 MJ

Schexnayder et al. (2001)

3.1.2. Functional unit and system boundaries The selected FU is a 48-TL-200 ZEBRA battery (Fig. 1), including the Battery Management Interface (BMI). The technical characteristics of the battery are shown in Table 2. The battery is used to store energy generated by a multi-Si photovoltaic system installed on a flat roof. The system boundaries, including the processes examined in the study, are as follows (Fig. 2): - battery manufacturing step, including raw material supply, manufacturing/assembly of the main components and final waste treatment, with the waste representing the raw material packaging; - operation step, including the production of electricity used by the battery during its useful life and the manufacturing of the photovoltaic system; - end-of-life step. The transportation of the battery to the end user and the transportation of packaging waste to the disposal site were not taken into account as their energy and environmental impact can be assumed to be negligible (Ishihara et al., 2006; Rydh and Sandèn, 2005; Zackrisson et al., 2010). ZEBRA batteries do not require maintenance. Consequently, the maintenance step does not cause any energy or environmental impact and was excluded from the analysis. At the end-of-life, all battery components can be recycled. The stainless steel case and the glass wool can be recycled in established processes (Dustmann, 2004; Parkhided, 2006). The nickel, the salt and the ceramic contained in the cells are used in steel melting in stainless steel manufacturing (Dustmann, 2004; Parkhided, 2006). Due to the lack of data on the eco-profiles of the recycling of sodium/nickel chloride batteries (Ardente et al., 2003), the end-of-life step was accounted for considering average

Li-ion (NMP solvent) Lion (water solvent) Pb-acid LiMnO4 LiFePO4 (NMP solvent) LiFePO4 (water solvent) Nickelemetal hydride Li-ion

Manufacturing and end-of-life Manufacturing and operation

Notter et al. (2010) Zackrisson et al. (2010)

data for a European recycling process (Frischknecht et al., 2007a) that represents a combination of the recycling processes for lithium-ion batteries (pyrometallurgical and hydrometallurgical processes) and nickelemetal hydride batteries (pyrometallurgical process). 3.1.3. Impact assessment methodology and impact categories The following energy and environmental indexes were selected to describe the performance of the investigated system: -

Global energy requirement (GER); Non-renewable energy requirement (NRE); Renewable energy requirement (RE); Global warming potential (GWP); Ozone depletion potential (ODP);

Fig. 1. ZEBRA battery.

Please cite this article in press as: Longo, S., et al., Life cycle assessment of storage systems: the case study of a sodium/nickel chloride battery, Journal of Cleaner Production (2013), http://dx.doi.org/10.1016/j.jclepro.2013.10.004

S. Longo et al. / Journal of Cleaner Production xxx (2013) 1e10

Unit of measure

Value

Nominal voltage Open circuit voltage Nominal capacity Nominal energy Gravimetric energy density Thermal loss in operation Operating temperature range Mass Dimensions

V V Ah Wh Wh/kg W  C kg mm

48 51.6 200 9600 91 105 20 to þ60 105 558  496  320

-

Human toxicity (HT); Photochemical ozone formation (POF); Acidification (Ac); Terrestrial eutrophication (TE); Freshwater eutrophication (FE); Marine eutrophication (ME); Land use (LU); Water resource depletion (WRD).

The characterisation models used for the impact calculations are the Cumulative Energy Demand method (Frischknecht et al., 2007b; Prè, 2012), which allows for the assessment of renewable and non-renewable primary energy consumption, and the ILCD 2011 Midpoint impact assessment method, elaborated by Prè (2012) according to the European Commission (2012c). 3.1.4. Data quality The eco-profiles of materials and energy sources used to produce the battery and the impacts related to the transportation step and to the end-of-life processes of packaging materials were based on the Ecoinvent database (Frischknecht et al., 2007a). The ecoprofile of the BMI was taken from Majeau-Bettez et al. (2011). The eco-profiles of electricity and natural gas used in the manufacturing process as well as the eco-profiles of raw materials are referred to the European context, with the exception of glass wool, which referred to the Swiss context, and of nickel, battery cables and the integrated circuits of the BMI, which referred to the worldwide context (Frischknecht et al., 2007a). 3.2. Life cycle inventory analysis 3.2.1. Data collection The description of the data collection is critical to ensure the reliability of the results. For this reason, LCA studies should clearly indicate the data sources and data collection procedures (Ardente et al., 2004; Cellura et al., 2011). Therefore, the authors describe the examined product, the data collection process, and the assumptions made in the manufacturing and operation steps. The collected data were elaborated to calculate the energy and raw material consumption, the emissions to air, water and soil, and waste production. 3.2.1.1. The examined product. The examined battery is made up of 100 sodium/nickel chloride cells that are assembled in a stainless steel battery case, connected by a nickel inter-cell jumper and insulated by mica sheets. The thermal transmittance of the battery case is very low due to the use of glass fibres as insulation material, providing very low heat conductivity. An ohmic heater inside the battery case maintains the operation temperature and is controlled by the BMI (Galloway and Haslam, 1999; Manzoni et al., 2008; Tilley and Galloway, 2001).

2NaCl þ Ni/NiCl2 þ 2Na:

(1)

In the discharge step, the process is spontaneously inverted without generating undesired byproducts:

NiCl2 þ 2Na/2NaCl þ Ni:

BATTERY MANUFACTURING

Characteristics

The main component of the battery is the ZEBRA cell (Fig. 3) (Sudworth, 2001; Parkhided, 2006; Tannaz, 2012; Thompson and Tilley, 2002), which is enclosed in a steel case coated with nickel. One of the main components of the cell is the b-alumina ceramic electrolyte tube, which separates the anode and the cathode while acting as a sodium ion conductor. The negative electrode (sodium (Na) metal formed by the reduction of sodium chloride (NaCl)) is located between the ceramic electrolyte and the steel case. Na is a liquid at the typical battery operating temperature and is both volatile and reactive; therefore, the cell is hermetically sealed by laser welded nickel rings that are thermo-compression-bonded (TCB) to an a-alumina collar that is glass brazed to the b-alumina tube. The positive electrode (cathode) consists of a nickel chloride (NiCl2). It is located inside the ceramic electrolyte tube together with the nickel (Ni) current collector that is required for longitudinal conduction to the cell terminal. As the reactive components of the positive electrode are in the solid state at operating temperature, a liquid electrolyte is used to facilitate the movement of Na ions within the cathode. The sodium aluminium chloride (NaAlCl4) electrolyte is obtained by combining NaCl and anhydrous aluminium chloride (AlCl3). Both NiCl2 and NaAlCl4 are hygroscopic and very difficult to handle, so the standard practise is to load the cell in the discharged state by adding a granulated mixture of Ni powder and NaCl and filling the voids with liquid electrolyte. Thus, the main charge reaction in the sodium/metal chloride cells involves the chlorination of high surface area Ni powder with NaCl to form the positive NiCl2 electrode and Na:

BATTERY OPERATION

Table 2 Technical characteristics of the battery.

BATTERY END-OF-LIFE

4

(2)

Raw material production

Raw material transport Production/assembly of the battery components

Battery transport to final users

Battery use

Battery transport to final users

Battery recycling Step included in the system boundaries Step not included in the system boundaries Fig. 2. System boundaries.

Please cite this article in press as: Longo, S., et al., Life cycle assessment of storage systems: the case study of a sodium/nickel chloride battery, Journal of Cleaner Production (2013), http://dx.doi.org/10.1016/j.jclepro.2013.10.004

S. Longo et al. / Journal of Cleaner Production xxx (2013) 1e10

5

Table 3 Main components of battery.

Current collector (+ pole)

Nickel chloride + sodium aluminium chloride

Ceramic electrolyte

Sodium

Cell case (-pole)

Component

Material

Mass (kg)

Battery case Sodium/nickel chloride cells Thermal insulation Ohmic heater Insulation among cells BMI Electric cables Cells inter-connection Other

Stainless steel e Glass wool Silicon Mica e Nickel alloy Nickel e

11.00 80.50 10.00 0.35 0.35 0.70 0.20 0.36 1.54

Table 4 Main inputs used for the manufacturing of a single cell. Material

Mass (kg)

Hilumina (cell case) Boehmite (ceramic tube) Sodium carbonate (ceramic tube) Nickel (TCB seal) a-Alumina (TCB seal) Nickel-plated steel Aluminium chloride anhydrous Sodium chloride Nickel powder Nickel sheet Water

0.070 0.218 0.038 0.018 0.006 0.080 0.080 0.140 0.120 0.035 2.94

a Hilumin: high grade, cold-rolled steel strip with an electro-deposited layer of pure nickel on both sides.

Fig. 3. ZEBRA cell.

3.2.1.2. The battery manufacturing process. The primary data related to the battery manufacturing process were collected by distributing a questionnaire to the manufacturer and directly analysing the battery manufacturing process. In particular, the following data were collected: - materials and masses of the main components of the battery (Table 3); - materials used for the manufacture of cells and relative masses (Table 4); - energy consumption needed for the cell manufacture: 7.13 MJ of electricity and 116 MJ of natural gas per cell; - packaging (polyethylene film, polypropylene, pallets, metal cans) used for the raw materials and the final product; - information about waste disposal (the waste represents the raw material packaging); - information about the transportation by truck of the battery raw materials and components (Table 5). 3.2.1.3. The battery operation step. As outlined in the previous paragraphs, during the operation step the battery is used as a storage system for the electricity produced by a multi-Si photovoltaic system installed on a flat roof, which is the most common residential photovoltaic system (Cellura et al., 2012). Information about the electrical power needs of the battery during the operation step was assessed in laboratory tests by directly monitoring the system under different working conditions (Restello et al., 2011). In detail, the power needs are related to: - BMI (power: 8 W); - ohmic heater for stand-by operation (power: 90 W).

Thus, under stand-by conditions, the battery requires less than 100 W for heating and BMI operation. Electrical power is needed to feed electronic components under discharge conditions. In detail, 2 W are needed for the BMI, and 6 W are needed to feed the main contactors. Considering that the battery has a round trip efficiency1 of approximately 80e90%, a loss of charge occurs during battery operation, leading to additional energy consumption (10e20% of the charge) (Restello et al., 2011). As the battery’s electricity consumption depends on the daily charge/discharge cycles, the useful life of the system, and the conditions of use of the battery, the authors examined two operating scenarios as shown in Table 6. The two scenarios represent two different realistic conditions (Jenkins et al., 2008) for residential storage systems. Depending on the system design, load profile, and seasonal production of renewable energy sources, the storage system can operate continuously (Scenario A) or with stand-by periods (Scenario B). The electricity used by the battery is produced by a multi-Si photovoltaic system installed on a flat roof. The eco-profile of the electricity includes the impacts related to the manufacturing of the photovoltaic system (photovoltaic panels (PVs), inverter, and other electrical components). 3.3. Life cycle impact assessment: results and discussion The life cycle energy and environmental impacts of the selected FU for the two scenarios are shown in Table 7. The highest impacts are observed for Scenario B, as the battery consumes more electricity during this operation step than in Scenario A.

1 Round trip efficiency is the ratio of the total output of the energy storage system (discharge) divided by the total energy input (charge) as measured at the interconnection point.

Please cite this article in press as: Longo, S., et al., Life cycle assessment of storage systems: the case study of a sodium/nickel chloride battery, Journal of Cleaner Production (2013), http://dx.doi.org/10.1016/j.jclepro.2013.10.004

6

S. Longo et al. / Journal of Cleaner Production xxx (2013) 1e10

Table 5 Transport of the main inputs of the process. Main inputs

Medium distance

Boehmite Sodium carbonate a-Alumina Nickel powder Sodium chloride Aluminium chloride anhydrous Nickel-plated steel Nickel sheet BMI Stainless steel Glass wool Electric cables

600 km 1000 km 600 km 1300 km 1300 km 600 km 600 km 100 km 400 km 200 km 850 km 250 km

Thus, the battery operating conditions significantly affect the final LCA results, with the total number of battery cycles during the useful battery life and the daily charge/discharge cycles exhibiting the greatest effects. A detailed analysis of the results showed that: - The GER is 57.4 GJ in Scenario A and 68.7 GJ in Scenario B. The amount of energy obtained from renewable energy sources varies from 26.5 GJ (46.2% of the total in Scenario A) to 35.5 GJ (51.7% of the total in Scenario B) and is mainly due to the electricity produced by the photovoltaic system and used during the operation step; - the contribution of the manufacturing step to the GER ranges from 35.8% (Scenario B) to 42.8% (Scenario A); the operation step contributes from 55.1% (Scenario A) to 70.5% (Scenario B); and the end-of-life of the battery contributes less than 2%; - the manufacturing step contributes from 60.7% to 92.1% of the environmental impact for Scenario A and from 54.5% to 90.1% for Scenario B (Table 8). A detailed analysis of the manufacturing step shows that a relevant share of GER (approximately 21.7 GJ) is caused by the cell manufacturing (89.3%). The BMI and battery case manufacturing processes contribute 4.8% and 3.4% of GER, respectively. Contributions to GER of less than 2.5% are caused by thermal insulation (2.03%), raw material transport (1.29%), cell interconnections (0.28%), and cables (0.05%). Similar results are observed for the environmental impact (Table 9).

Table 6 Description of the two examined operative scenarios. Operative conditions

Unit of measure

Scenario A

Scenario B

Total cycles during the useful life Daily charge/discharge cycles Useful life Charge time Stand-by period Working period Round trip efficiency (Restello et al., 2011) Depth of discharge Daily energy consumption in stand-by (BMI and heater) Daily energy consumption of electronic components Daily energy consumption during cycling Daily total energy consumption Energy consumption during the useful life

Number of cycles Number of cycles Days h (Hour) h (Hour) h (Hour) %

3000 2 1500 10 0 24 90

3500 1.6 2187 10 8 16 90

a

Table 7 Energy and environmental impacts of the battery life cycle for the two examined configurations e base scenario.

GER (MJ) NRE (MJ) RE (MJ) GWP (kg CO2eq) ODP (kg CFC-11eq) HT (CTUh) POF (kg NMVOCeq) Ac (mol Hþeq) TE (mol Neq) FE (kg Peq) ME (kg Neq) LU (kg C deficit) WRD (m3 watereq)

Scenario A

Scenario B

Percentage difference ((B value  A value)/ B value)

5.7Eþ04 3.1Eþ04 2.7Eþ04 2.0Eþ03 2.7E-04 5.3E-04 7.1Eþ00 3.7Eþ01 1.8Eþ01 1.5Eþ00 2.7Eþ00 2.1Eþ03 2.0Eþ00

6.9Eþ04 3.3Eþ04 3.6Eþ04 2.1Eþ03 3.0E-04 5.9E-04 7.7Eþ00 3.8Eþ01 2.0Eþ01 1.6Eþ00 2.8Eþ00 2.2Eþ03 2.1Eþ00

19.6 7.4 33.9 7.3 10.3 10.3 7.0 2.1 7.3 7.1 5.4 5.4 7.1

Thus, the cell manufacturing is the main process that has to be examined in detail to define actions to reduce the energy and environmental impacts of the selected FU. Comparing the impacts of ZEBRA manufacturing on GER, NRE and GWP per kg of battery with the literature data presented in Section 2 results in the following observations: - The GER (234.3 MJ/kg) and NRE (222.6 MJ/kg) of ZEBRA are higher than the values reported in the literature for other types of batteries; - The GWP of ZEBRA (14.32 kg CO2eq/kg) is lower than the GWP obtained by Majeau-Bettez et al. (2011) and by Sullivan and Gaines (2010), with differences ranging from 53.6% to 3560%, and is higher than the GWP calculated in the other studies reported in Section 2, with differences ranging from 12.7% to 93.7%. The authors compared the impact of the ZEBRA manufacturing step with the literature data, but did not compare the total life cycle impacts due to various assumptions regarding the operation and end-of-life steps. 3.3.1. Energy and environmental impacts of a single cell Cell manufacturing is responsible for the main energy and environmental impact of battery manufacturing. Thus, this step represents a key part of the battery life-cycle and has to be examined in detail. The following paragraph describes the energy and environmental impacts related to the manufacturing of a single cell. The GER for a single cell is approximately 217 MJ, of which 95.5% is derived from non-renewable energy sources. Fig. 4 shows that the energy consumed during the cell manufacturing process Table 8 Environmental impacts of the battery life cycle: percentage incidence of each life cycle step e base scenario. Manufacturing

Operation

End-of-life

Scenario A Scenario B Scenario A Scenario B Scenario A Scenario B % kWh/day

95 0a

95 0.784

kWh/day

0.192

0.128

kWh/day

1.90

1.52

kWh/day kWh

2092 6276

2432 8512

GWP (%) ODP (%) HT (%) POF (%) Ac (%) TE (%) FE (%) ME (%) LU (%) WRD (%)

75.49 65.46 60.75 74.95 92.09 73.87 76.41 81.10 79.47 72.41

70.37 58.71 54.51 69.72 90.14 68.48 71.02 76.69 75.22 67.25

20.44 32.30 32.12 21.05 6.08 22.10 21.30 16.14 15.87 21.52

25.84 39.29 39.09 26.56 8.08 27.78 26.85 20.70 20.37 27.10

4.07 2.24 7.13 4.00 1.83 4.03 2.29 2.77 4.66 6.07

3.79 2.01 6.40 3.72 1.79 3.74 2.13 2.62 4.41 5.64

No stand-by conditions. The battery works for 24 h per day.

Please cite this article in press as: Longo, S., et al., Life cycle assessment of storage systems: the case study of a sodium/nickel chloride battery, Journal of Cleaner Production (2013), http://dx.doi.org/10.1016/j.jclepro.2013.10.004

S. Longo et al. / Journal of Cleaner Production xxx (2013) 1e10

7

Table 9 Production step: incidence of each battery component on the environmental impacts. Battery case

Cells inter-connection

Thermal insulation

Cables

Cells

BMI

Raw material transport

GWP (kg CO2eq) ODP (kg CFC-11eq) HT (CTUh) POF (kg NMVOCeq) Ac (mol Hþeq) TE (mol Neq) FE (kg Peq) ME (kg Neq) LU (kg C deficit) WRD (m3 watereq)

5.0Eþ01 2.6E-06 7.3E-05 1.6E-01 3.1E-01 5.5E-01 2.4E-02 5.2E-02 5.6Eþ01 5.3E-02

3.9Eþ00 3.1E-07 1.9E-06 6.8E-02 6.8E-01 1.4E-01 1.4E-02 1.0E-02 9.0Eþ00 1.6E-02

1.5Eþ01 2.5E-06 2.6E-06 4.0E-02 1.0E-01 3.3E-01 5.3E-03 1.3E-02 1.5Eþ01 3.9E-02

4.7E-01 2.1E-08 1.8E-06 4.4E-03 1.7E-02 1.51E-02 5.2E-03 1.3E-03 8.0E-01 9.9E-04

1.3Eþ03 1.6E-04 2.0E-04 4.6Eþ00 3.2Eþ01 1.0Eþ01 7.9E-01 1.9Eþ00 1.5Eþ03 1.2Eþ00

7.3Eþ01 7.7E-06 4.5E-05 2.9E-01 4.7E-01 1.1Eþ00 3.1E-01 1.1E-01 5.6Eþ01 1.3E-01

1.8Eþ01 2.8E-06 2.5E-06 1.7E-01 1.3E-01 6.0E-01 1.9E-03 5.5E-02 4.6Eþ01 1.3E-02

GWP (%) ODP (%) HT (%) POF (%) Ac (%) TE (%) FE (%) ME (%) LU (%) WRD (%)

3.3Eþ00 1.5Eþ00 2.2Eþ01 3.0Eþ00 9.1E-01 4.1Eþ00 2.1Eþ00 2.4Eþ00 3.4Eþ00 3.7Eþ00

2.6E-01 1.8E-01 6.0E-01 1.3Eþ00 2.0Eþ00 1.0Eþ00 1.2Eþ00 4.7E-01 5.5E-01 1.1Eþ00

9.9E-01 1.4Eþ00 8.0E-01 7.5E-01 3.0E-01 2.4Eþ00 4.6E-01 5.8E-01 8.9E-01 2.7Eþ00

3.1E-02 1.2E-02 5.7E-01 8.2E-02 5.0E-02 1.1E-01 4.5E-01 6.0E-02 4.9E-02 6.9E-02

8.9Eþ01 9.1Eþ01 6.1Eþ01 8.6Eþ01 9.5Eþ01 8.0Eþ01 6.8Eþ01 8.9Eþ01 8.9Eþ01 8.2Eþ01

4.9Eþ00 4.4Eþ00 1.4Eþ01 5.3Eþ00 1.4Eþ00 8.4Eþ00 2.7Eþ01 5.2Eþ00 3.4Eþ00 9.3Eþ00

1.2Eþ00 1.9E-07 1.7E-07 1.1E-02 8.7E-03 4.0E-02 1.3E-04 3.7E-03 3.0Eþ00 8.5E-04

(electricity and natural gas) is the main contributor to GER, followed by nickel powder manufacturing. The contributions of the cell components to the environmental impacts are shown in Fig. 5. The energy consumed during the cell manufacturing process is the main factor in the following impacts: ODP (77.4%), GWP (69.6%) and LU (39.1%). The manufacturing of nickel powder is the main contributor to HT (32.7%), TE (42.5%), WRD (48.2%), POF (49.3%), FE (60.5%) and Ac (70.2%). The greatest impact on the ME (58.7%) results from the packaging manufacturing and end-of-life, which effects on the other impacts are variables from 1% (FE) to 30.3% (LU). Impacts lower than 3.5% are observed for the manufacture of a-alumina, boehmite, Na2CO3, NaCl and nickel used in TCB. The main impact caused by AlCl3 is related to ODP (6.9% of the total), while the contribution to the other impacts is less than 3%. Hilumin and nickel-plated sheet contribute less than 2% to each of the impact variables, with the exception of the contribution to HT (10e11% of the total). The contributions of the manufacturing of nickel sheet to the outcomes range from 2% (ODP) to 20.5% (Ac). The above results show that improving energy efficiency using efficient and low energy machinery, and using renewable energy sources to produce the electricity used in the cell manufacturing process can reduce the energy and environmental impacts of the examined FU. 3.3.2. Quantifying the uncertainty of the results through Monte Carlo analysis The authors carried out a Monte Carlo analysis to quantify the uncertainty in the final results caused by the variability and uncertainty of secondary input data, characterised by a log-normal statistical distribution (Prè, 2012). Table 10 presents the estimated distribution of the LCA results, describing the mean value (MV), median value (MDV), standard deviation (SD), coefficient of variation (CV) and 95% confidence interval (the 2.5th percentile and the 97.5th percentile). The highest uncertainties exist for the impact categories “FE” and “HT, cancer effects”, as noted by the CV values greater than 50%. Thus, the final results for the above categories are characterised by a high dispersion and are strongly influenced by the variability of the input data. The other impact categories have CVs lower than 50%, indicating that the distribution of the results is correctly represented by the MV and that the final results are characterised by a low dispersion.

For both scenarios, CV values lower than 8% are observed for the impact categories “RE, wind, solar, geothermal”, “GWP”, and “WRD”. This means that the final results are distributed around the MV and are characterised by very low uncertainty. 4. Scenario analysis: electricity used in the operation step The energy and environmental impacts of the battery operation step also depend on how the electricity is produced. To determine the variation in the impact of the battery operation step resulting from different methods of electricity production and to assess the energy and environmental benefits obtained by using electricity produced by renewable energy technologies instead of electricity from the grid (generated by renewable and non-renewable energy sources), the following electricity production scenarios were examined (Prè, 2012): - Base scenario: production of electricity by a multi-Si photovoltaic system installed on a flat roof; - Scenario 1: production of electricity in Europe by wind turbines: 98% of energy is generated by on-shore plants with a power of 800 kW and the remaining 2% by off-shore plants with a power of 2 MW; - Scenario 2: production of electricity in Italy by hydropower, with 64% of energy obtained by reservoir power plants in alpine regions and 36% by run-of-river power plants.

NaCl: 0.21% Nickel powder: 10.36%

Nickel sheet: 3.02%

Nickel-plated steel: 1.03% Hilumin: 0.90% AlCl3: 0.72% Nickel: 0.52% α-alumina: 0.16% Boehmite: 0.01%

Energy: 77.71%

Packaging: 5.02%

Na2CO3: 0.34%

Fig. 4. Single cell production: percentage incidence of each cell component on GER.

Please cite this article in press as: Longo, S., et al., Life cycle assessment of storage systems: the case study of a sodium/nickel chloride battery, Journal of Cleaner Production (2013), http://dx.doi.org/10.1016/j.jclepro.2013.10.004

8

S. Longo et al. / Journal of Cleaner Production xxx (2013) 1e10

Nickel powder

WRD (m3 watereq)

NaCl LU (kg C deficit)

Nickel sheet ME (kg Neq)

Nickel-plated steel Hilumin

FE (kg Peq) TE (mol Neq)

AlCl3

Ac (mol H+eq)

Nickel a-alumina

POF (kg NMVOCeq)

Boehmite

HT (CTUh)

Na2CO3 ODP (kg CFC-11eq)

Packaging GWP (kg CO2eq)

Energy 0%

20%

40%

60%

80%

100%

Fig. 5. Single cell production: percentage incidence of each cell component on the environmental impacts.

- Scenario 3: Italian electricity mix. In detail, 15.1% of the overall energy provided is obtained by coal, 16.1% by crude oil, 45.7% by natural gas, 1.9% by industrial gas, 19.9% by hydropower, 0.7% by wind and 0.6% by biomass and biogas cogeneration. Table 10 Monte Carlo analysis simulation: main results. Impact category Scenario A NRE, biomass (MJ) NRE, nuclear (MJ) NRE, fossil (MJ) RE, biomass (MJ) RE, water (MJ) RE, wind, solar, geothermal (MJ) GWP (kg CO2eq) ODP (kg CFC-11eq) HT, cancer effect (CTUh) HT, non-cancer effect (CTUh) POF (kg NMVOCeq) Ac (mol Hþeq) TE (mol Neq) FE (kg Peq) ME (kg Neq) LU (kg C deficit) WRD (m3 watereq) Scenario B NRE, biomass (MJ) NRE, nuclear (MJ) NRE, fossil (MJ) RE, biomass (MJ) RE, water (MJ) RE, wind, solar, geothermal (MJ) GWP (kg CO2eq) ODP (kg CFC-11eq) HT, cancer effect (CTUh) HT, non-cancer effect (CTUh) POF (kg NMVOCeq) Ac (mol Hþeq) TE (mol Neq) FE (kg Peq) ME (kg Neq) LU (kg C deficit) WRD (m3 watereq)

MV

MDV

SD

CV

2.50%

97.5%

8.5E-02 3590 27,300 358 1970 24,300

7.8E-02 3470 26,800 348 1950 24,200

3.3E-02 698 4250 71 195 1070

38.6% 19.5% 15.6% 19.8% 9.9% 4.42%

4.0E-02 2440 20,300 244 1660 22,300

1.7E-01 5160 36,800 523 2380 26,500

1990 2.7E-04 2.0E-04 3.2E-04

1970 2.6E-04 1.6E-04 3.1E-04

151 5.7E-05 1.8E-04 7.3E-05

7.6% 21.3% 89.8% 22.6%

1740 1.8E-04 1.0E-04 2.2E-04

2330 4.0E-04 6.0E-04 4.9E-04

7.13 36.7 18.3 1.52 2.72 2070 1.98

7.07 35.6 18.2 1.30 2.48 1940 1.96

0.657 7.23 1.83 1.03 0.875 689 0.155

9.21% 19.7% 9.98% 67.7% 32.2% 33.3% 7.85%

5.93 25.2 15.2 0.802 1.79 1100 1.73

8.54 53.6 22.6 3.55 5.17 3800 2.31

1.0E-01 3960 29,100 416 2290 32,800

9.3E-02 3870 28,700 408 2270 32,800

4.1E-02 798 4570 79.4 255 1460

40.3% 20.2% 15.7% 19.1% 11.1% 4.44%

4.8E-02 2650 21,900 287 1860 30,000

2.1E-01 5780 39,200 610 2860 35,800

2140 3.0E-04 2.3E-04 3.7E-04

2130 2.9E-04 1.8E-04 3.6E-04

171 6.1E-05 2.4E-04 8.5E-05

7.99% 20.3% 106% 22.9%

1850 2.0E-04 1.1E-04 2.5E-04

2530 4.4E-04 6.3E-04 5.7E-04

7.65 37.4 19.7 1.64 2.85 2160 2.13

7.61 36.6 19.6 1.4 2.59 2030 2.11

0.72 7.34 1.94 0.89 0.973 681 0.171

9.4% 19.6% 9.84% 54.3% 34.1% 31.5% 8.04%

6.35 25 16.2 0.881 1.92 1220 1.84

9.13 53.3 23.9 4.01 5.52 3830 2.5

Less than 0.001% of the electricity is produced by photovoltaic technology; - Scenario 4: European electricity mix, with 2.5% produced in Belgium, 2.3% in the Czech Republic, 17.4% in Germany, 8.1% in Spain, 16.5% in France, 11.3% in the United Kingdom, 8.7% in Italy, 2.9% in the Netherlands, 3.3% in Norway, 4.3% in Poland, and 4.5% in Sweden. The remaining 18.2% is produced in other European countries, each contributing less than 2%. The results of the scenario analysis, reported in Tables 11 and 12, highlight that producing the electricity with renewable energy sources results in a lower impact than electricity generated with a mix of renewable and non-renewable energy sources. The following detailed considerations can be made: - generating electricity using wind turbines (Scenario 1) or hydropower (Scenario 2) reduces GER with respect to the Base Scenario (photovoltaic electricity production) by approximately 20e24%; - the Base Scenario GER is 112% and 144% lower than the GER for Scenarios 3 (Italian energy mix) and 4 (European energy mix), respectively; - all environmental impacts can be reduced by substituting PVs with wind turbines. The reduction varies from 59.5% (HT) to 95% (ODP). The only exception is LU, which increases by approximately 100%; - the use of electricity produced by hydropower instead of by PVs reduces the environmental impacts with effects ranging from 90% (TE) to 98% (ODP). In addition, producing electricity by hydropower has a negative impact on LU; - substituting the Base Scenario with Scenarios 3 and 4, HT increases between 76.5% (Scenario 3) and 213% (Scenario 4), POF grows between five (Scenario 4) and eight times (Scenario 3), and the other impacts increase by one order of magnitude.

5. Conclusions The aim of this study was to evaluate the energy and environmental performance of one ZEBRA battery, examining the manufacturing, operation and end-of-life steps. Two realistic operation conditions were examined considering two different durations of useful battery life.

Please cite this article in press as: Longo, S., et al., Life cycle assessment of storage systems: the case study of a sodium/nickel chloride battery, Journal of Cleaner Production (2013), http://dx.doi.org/10.1016/j.jclepro.2013.10.004

S. Longo et al. / Journal of Cleaner Production xxx (2013) 1e10 Table 11 Scenario A: comparison of environmental impacts related on different scenarios of energy production: operation step. Base scenario Scenario 1 Scenario 2 Scenario 3 Scenario 4 GER (MJ) NRE (MJ) RE (MJ) GWP (kg CO2eq) ODP (kg CFC-11eq) HT (CTUh) POF (kg NMVOCeq) Ac (mol Hþeq) TE (mol Neq) FE (kg Peq) ME (kg Neq) LU (kg C deficit) WRD (m3 watereq)

3.2Eþ04 6.4Eþ03 2.5Eþ04 4.1Eþ02 8.7E-05 1.7E-04 1.5Eþ00 2.2Eþ00 4.1Eþ00 3.2E-01 4.3E-01 3.3Eþ02 4.3E-01

2.5Eþ04 1.1Eþ03 2.4Eþ04 7.1Eþ01 4.3E-06 6.9E-05 2.1E-01 3.9E-01 6.9E-01 4.5E-02 7.8E-02 6.8Eþ02 1.1E-01

2.4Eþ04 3.3Eþ02 2.4Eþ04 3.1Eþ01 1.8E-06 1.1E-05 1.1E-01 1.3E-01 4.2E-01 7.7E-03 3.8E-02 9.8Eþ01 6.6E-02

6.7Eþ04 6.2Eþ04 4.9Eþ03 4.5Eþ03 4.0E-04 3.0E-04 1.2Eþ01 2.5Eþ01 4.0Eþ01 1.0Eþ00 3.7Eþ00 5.0Eþ03 1.9Eþ00

7.7Eþ04 7.1Eþ04 6.4Eþ03 3.5Eþ03 1.7E-04 5.3E-04 7.4Eþ00 1.9Eþ01 2.6Eþ01 3.2Eþ00 2.9Eþ00 1.7Eþ03 4.5Eþ00

Table 12 Scenario B: comparison of environmental impacts related on different scenarios of energy production: operation step. Base scenario Scenario 1 Scenario 2 Scenario 3 Scenario 4 GER (MJ) NRE (MJ) RE (MJ) GWP (kg CO2eq) ODP (kg CFC-11eq) HT (CTUh) POF (kg NMVOCeq) Ac (mol Hþeq) TE (mol Neq) FE (kg Peq) ME (kg Neq) LU (kg C deficit) WRD (m3 watereq)

4.3Eþ04 8.7Eþ03 3.4Eþ04 5.5Eþ02 1.2E-04 2.3E-04 2.0Eþ00 3.0Eþ00 5.5Eþ00 4.4E-01 5.9E-01 4.4Eþ02 5.8E-01

3.4Eþ04 1.4Eþ03 3.3Eþ04 9.6Eþ01 5.8E-06 9.4E-05 2.8E-01 5.3E-01 9.3E-01 6.1E-02 1.1E-01 9.3Eþ02 1.5E-01

3.3Eþ04 4.4Eþ02 3.2Eþ04 4.2Eþ01 2.4E-06 1.5E-05 1.6E-01 1.7E-01 5.6E-01 1.0E-02 5.2E-02 1.3Eþ02 9.0E-02

9.1Eþ04 8.4Eþ04 6.7Eþ03 6.1Eþ03 5.4E-04 4.1E-04 1.6Eþ01 3.4Eþ01 5.4Eþ01 1.4Eþ00 5.1Eþ00 6.8Eþ03 2.6Eþ00

1.0Eþ05 9.6Eþ04 8.7Eþ03 4.8Eþ03 2.3E-04 7.3E-04 1.0Eþ01 2.6Eþ01 3.5Eþ01 4.3Eþ00 4.0Eþ00 2.3Eþ03 6.1Eþ00

The obtained results showed that the operation step has the highest energy impact, while the manufacturing step has the highest environmental impact, with the cell manufacturing as the biggest contributor to this impact. The use of energy (electricity and natural gas) in the cell manufacturing process causes relevant impacts. An important finding of this study is that the life-cycle impact of the battery can be decreased by increasing the efficiency and decreasing the carbon emissions of the manufacturing process. This goal can be achieved by reducing the direct energy consumption or by using energy from renewable sources. To evaluate how electricity production methods may impact the eco-profile of these batteries, different scenarios of electricity production were analysed. This comparison allowed for an assessment of the energy and environmental benefits related to the production of electricity with renewable sources. Thus, the energy and environmental performances of electricity storage systems can be improved using low carbon electricity during the manufacturing and operation steps and by improving the energy efficiency of the manufacturing processes. Acknowledgements The authors wish to thank FIAMM Company for its essential contribution to the data collection and processing for the case study. References Ardente, F., Beccali, G., Cellura, M., 2003. Eco-sustainable energy and environmental strategies in design for recycling: the software “ENDLESS”. Ecol. Model. 163 (1e 2), 101e118.

9

Ardente, F., Beccali, M., Cellura, M., 2004. F.A.L.C.A.D.E.: a fuzzy software for the energy and environmental balances of products. Ecol. Model. 176 (3e4), 359e 379. Beccali, M., Brunone, S., Cellura, M., Franzitta, V., 2008. Energy, economic and environmental analysis on RET-hydrogen systems in residential buildings. Renew. Energy 33, 366e382. Beccali, M., Cellura, M., Longo, S., Nocke, B., Finocchiaro, P., 2012. LCA of a solar heating and cooling system equipped with a small watereammonia absorption chiller. Solar Energy 86 (5), 1491e1503. Cellura, M., Longo, S., Mistretta, M., 2011. Sensitivity analysis to quantify uncertainty in life cycle assessment: the case study of an Italian tile. Renew. Sustain. Energy Rev. 15, 4697e4705. Cellura, M., Di Gangi, A., Longo, S., Orioli, A., 2012. Photovoltaic electricity scenario analysis in urban contests: an Italian case study. Renew. Sustain. Energy Rev. 16, 2041e2052. Commission of the European Communities, 2007a. COM (2007) 1 Final Communication from the Commission to the European Council and the European Parliament e An Energy Policy for Europe SEC(2007) 12. Brussels, 10.1.2007. Commission of the European Communities, 2007b. COM(2007) 723 Final Communication from the Commission to the Council, the European Parliament, the European Economic and Social Committee and the Committee of the Regions. Towards a Low Carbon Future. Brussels, 22.11.2007. Dustmann, C.H., 2004. Advances in ZEBRA batteries. J. Power Sources 127, 85e92. Entsoe e European Network of Transmission System Operators for Electricity, 2010. The European Electricity Grid Initiative (EEGI) Roadmap 2010e18 and Detailed Implementation Plan 2010e12. EDSO for smart grids.. Available on: www.ec. europa.eu. Eurobat, 2012. Battery Economic Outlook e Industrial Battery Market Europe Middle East and Africa (EMEA). Available in: www.eurobat.org. European Commission, 2010a. COM(2010) 639 Final Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions e Energy 2020 a Strategy for Competitive, Sustainable and Secure Energy (SEC(2010) 1346). Brussels, 10.11.2010. European Commission, Joint Research Centre e Institute for Environment and Sustainability, 2010b. ILCD Handbook International Reference Life Cycle Data System. General Guide for Life Cycle Assessment e Provisions and Action Steps. European Commission, 2012a. Commission Staff Working Document Accompanying the Document Council and the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions e Renewable Energy: A Major Player in the European Energy Market SWD (2012) 164 Final. Brussels, 6.6.2012. European Commission, 2012b. Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions e Renewable Energy: A Major Player in the European Energy Market COM (2012) 271 Final. Brussels, 6.6.2012. European Commission, DG Joint Research Centre e Institute for Environment and Sustainability, 2012c. Characterization Factors of the ILCD Recommended Life Cycle Impact Assessment Methods. Database and Supporting Information. EUR 25167, first ed. Publications Office of the European Union, Luxembourg. Frischknecht, R., Jungbluth, N., Althaus, H.J., Doka, G., Dones, R., Heck, T., Hellweg, S., Hischier, R., Nemecek, T., Rebitzer, G., Spielmann, M., 2007a. Overview and Methodology. Ecoinvent Report No. 1, ver.2.0. Swiss Centre for Life Cycle Inventories, Dübendorf, CH. Frischknecht, R., Jungbluth, N., Althaus, H.J., Bauer, C., Doka, G., Dones, R., Hischier, R., Hellweg, S., Humbert, S., Köllner, T., Loerincik, Y., Margni, M., Nemecek, T., 2007b. Implementation of Life Cycle Impact Assessment Methods. Ecoinvent report No. 3, v2.0. Swiss Centre for Life Cycle Inventories, Dübendorf. Galloway, R.C., Haslam, S., 1999. The ZEBRA electric vehicle battery: power and energy improvements. J. Power Sources 80, 164e170. Ishihara, K., Kihira, N., Terada, N., Iwahori, T., 2006. Environmental Burdens of Large Lithium-ion Batteries Developed in a Japanese National Project. Available on: http://www.electrochem.org/dl. ISO 14040, 2006. Environmental Management e Life Cycle Assessment e Principles and Framework. ISO 14044, 2006. Environmental Management e Life Cycle Assessment e Requirements and Guidelines. Jenkins, D.P., Fletcher, J., Kane, D., 2008. Model for evaluating impact of battery storage on microgeneration systems in dwellings. Energy Convers. Manage. 49, 2413e2424. Manzoni, R., Metzger, M., Crugnola, G., November 2008. ZEBRA Electric Energy Storage System: From R&D to Market. HTE hi.tech.expo, pp. 25e58. Milan. Majeau-Bettez, G., Hawkins, T.R., Hammer Strømman, A., 2011. Life cycle environmental assessment of lithium-ion and nickel metal hydride batteries for plug-in hybrid and battery electric vehicles. Environ. Sci. Technol. 45 (10), 4548e4554. Matheys, J., Van Autenboer, W., Van Mierlo, J., 2004. SUBAT: Sustainable Batteries e Work Package 5: Overall Assessment. Final Public Report. Vrije Universiteit Brussel e ETEC. Matheys, J., Van Autenboer, W., Timmermans, J.M., Van Mierlo, J., Van den Bossche, P., Maggetto, G., 2007. Influence of functional unit on the life cycle assessment of traction batteries. Int. J. Life Cycle Assess. 12 (3), 191e196. Matheys, J., Van Mierlo, J., Timmermans, J.M., 2011. Life-cycle assessment of batteries in the context of the EU directive on end-of-life vehicles. Int. J. Vehicle Des. 46 (2), 189e203.

Please cite this article in press as: Longo, S., et al., Life cycle assessment of storage systems: the case study of a sodium/nickel chloride battery, Journal of Cleaner Production (2013), http://dx.doi.org/10.1016/j.jclepro.2013.10.004

10

S. Longo et al. / Journal of Cleaner Production xxx (2013) 1e10

McManus, M.C., 2012. Environmental consequences of the use of batteries in low carbon systems: the impact of battery production. Appl. Energy 93, 288e295. Naish, C., McCubbin, I., Edberg, O., Harfoot, M., 2008. Outlook of Energy Storage Technologies. European Parliament, Policy Department Economic and Scientific Policy. Notter, D.A., Gauch, M., Widmer, R., Wager, P., Stamp, A., Zah, R., Althaus, H.J., 2010. Contribution of Li-ion batteries to the environmental impact of electric vehicles. Environ. Sci. Technol. 44 (17), 6550e6556. Parkhided, B., 2006. Project: Storage Technologies for Hybrid Electric Buses, Subject: ZEBRA Battery. Second revision. Electrochemical Energy Conversion and Storage Systems Group RWTH Aachen University, Germany. Available on: http://www.euromatic.no/ZEBRA_Aug17.pdf. Prè-Product Ecology Consultants, 2012. Software SimaPro7. Restello, S., Lodi, G., Paolin, E., 2011. Sodiumenickel batteries for telecom applications: design, performance and field operational overview. Telecommunications Energy Conference (INTELEC), 2011 IEEE 33rd International, 1e6. Rydh, C.J., Sandèn, B.A., 2005. Energy analysis of batteries in photovoltaic systems. Part I: performance and energy requirements. Energy Convers. Manage. 46, 1957e1979. Samaras, C., Meisterling, K., 2008. Life cycle assessment of greenhouse gas emissions from plug-in vehicles: implications for policy. Environ. Sci. Technol. 42, 3170e3176.

Schexnayder, S.M., Das, S., Dhingra, R., Overly, J.G., Tonn, B.E., Peretz, J.H., Waidley, G., Davis, G.A., 2001. Environmental Evaluation of New Generation Vehicles and Vehicle Components. ORNL/TM-2001-266. Available on: http:// wwwcta.ornl.gov. Sudworth, J.L., 2001. The sodium/nickel chloride (ZEBRA) battery. J. Power Sources 100, 149e163. Sullivan, J.L., Gaines, L., 2010. A Review of Battery Life-cycle Analysis: State of Knowledge and Critical Needs. ANL/ESD/10-7. Argonne National Laboratory e Energy Systems Division. Available on: http://www.osti.gov. Tannaz, J.D., 2012. Microstructure and Conductivity of the Sodium Nickel Chloride (Zebra) Battery Cathode. Open Access Dissertations and Theses. Paper 7441. Tilley, R., Galloway, R., 2001. Design Changes to Improve the Specific Energy of the ZEBRA Battery. EVS 18, Berlin, October 2001. Thompson, R.W., Tilley, R., 2002, Advantages of ZEBRA high temperature batteries for military hybrid applications. AECV Conference, Noordwijkerhout, the Netherlands. Van den Bossche, P., Vergels, F., Van Mierlo, J., Matheys, J., Van Autenboer, W., 2006. SUBAT: an assessment of sustainable battery technology. J. Power Sources 162, 913e919. Zackrisson, M., Avellàn, L., Orlenius, J., 2010. Life cycle assessment of lithium-ion batteries for plug-in hybrid electric vehicles e critical issues. J. Clean. Prod. 18, 1519e1529.

Please cite this article in press as: Longo, S., et al., Life cycle assessment of storage systems: the case study of a sodium/nickel chloride battery, Journal of Cleaner Production (2013), http://dx.doi.org/10.1016/j.jclepro.2013.10.004