Impact of the electricity mix and use prole in the life-cycle assessmentof electric vehiclesRicardo Fariaa,n, Pedro Marquesb, Pedro Mouraa,nn, Fausto Freireb, Joaquim Delgadoa,Anbal T. de AlmeidaaaInstitute of Systems and Robotics, Department of Electrical and Computer Engineering, University of Coimbra, 3030-290, PortugalbADAI-LAETA, Department of Mechanical Engineering, University of Coimbra, 3030-290 Coimbra, Portugala rti cle in foArticle history:Received 30 November 2012Received in revised form6 March 2013Accepted 15 March 2013Available online 16 April 2013Keywords:Battery electric vehiclesPlug-in hybrid electric vehiclesLife-cycle assessmentGreenhouse gas emissionsWell-to-wheel balancesTotal costs of ownershipabstractThispaperpresentsanenvironmentalandaneconomicLife-CycleAssessment(LCA)forconventionalandelectric vehicle technologies, focusing mainly onthe primary energy source andthe vehicleoperationphase Greenhouse Gas (GHG) emissions. Adetailedanalysis of the electricity mix wasperformed, based on the contribution of each type of primary energy source and their variation alonga year. Three mixes were considered, with different life cycle GHG intensity: one mainly based in fossilsources, a second one with a large contribution from nuclear and a third one with a signicant share ofrenewableenergysources. TheconventionalvehicletechnologyisrepresentedbygasolineanddieselInternational Combustion Engine Vehicles (ICEVs), while the electric technology is represented by Plug-in Hybrid Electric Vehicles (PHEVs) and Battery Electric Vehicles (BEVs). Real world tests were performedforrepresentativecompact andsub-compact EVs. Theuseproleof thevehiclewasbasedondataacquired by a real time data acquisition system installed in the vehicles. The results show that a mix witha large contribution from Renewable Energy Sources (RESs) does not always translate directly into lowGHG emissions for EVs due to the high variability of these sources. The driving prole under differentscenarios was also analyzed, showing that an aggressive style can increase the energy consumption by47%. The tests also showed that the use of climate control can increase the energy consumption between24 and 60%. Compared with other technologies, EVs can be more sustainable from an environmental andeconomic perspective; however, three main factors are required: improvement of battery technology, aneco-driving attitude and an environmental friendly electricity mix.& 2013 Elsevier Ltd. All rights reserved.Contents1. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2722. Life-cycle model and system boundary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2722.1. Vehicle characteristics and assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2732.2. Use phase and main factors that contribute to EVs energy consumption. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2752.3. Energy sources: emissions assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2782.3.1. Crude oil extraction and rening. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2782.3.2. Electricity mix. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278Contents lists available at SciVerse ScienceDirectjournal homepage: www.elsevier.com/locate/rserRenewable and Sustainable Energy Reviews1364-0321/$ - see front matter& 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.rser.2013.03.063Abbreviations: AC, AlternateCurrent;AFV, AlternativeFuel Vehicle;BEV, BatteryElectricVehicle;BMS, BatteryManagementSystem;CNG, CompressedNatural Gas;DOD, Depth of Discharge; ESS, Energy Storage System; EC, European Commission; EU, European Union; EV, Electric Vehicle; FC, Fuel Cell; FCEV, Fuel Cell Electric Vehicle;FTP-75, Federal Test Procedure 75; GWP, Global Warming Potential; GHG, Greenhouse Gas; HEV, Hybrid Electric Vehicle; HFCV, Hydrogen Fuel Cell Vehicle; ICE, InternalCombustion Engine; ICEV, Internal Combustion Engine Vehicle; IMU, Inertial Measurement Unit; IEA, International Energy Agency; LCA, Life-Cycle Assessment; LNG,Liqueed Natural Gas; LPG, Liqueed Petroleum Gas; MFI, Maintenance, Fuel, Insurance; MRT, Maintenance, Repair, Insurance and Taxes; NEDC, New European Drive Cycle;PEM, Power Electronics Module; PHEV, Plug-in Hybrid Electric Vehicle; PM, Particulate Matter; PT, Powertrain; RES, Renewable Energy Source; SFTP, Supplemental FederalTest Procedures; SOC, State of Charge; TCO, Total Cost of Ownership; TTW, Tank-to-Wheel; WTT, Well-to-Tank; WTW, Well-to-WheelnPrincipal corresponding author. Tel.: +351 965 513 899.nnCorresponding author. Tel.: +351 239 796 250.E-mail addresses: rfaria@isr.uc.pt (R. Faria), pedro.marques@dem.uc.pt (P. Marques), pmoura@isr.uc.pt (P. Moura), fausto.freire@dem.uc.pt (F. Freire),jdelgado@elect.estv.ipv.pt (J. Delgado), adealmeida@isr.uc.pt (A.T. de Almeida).Renewable and Sustainable Energy Reviews 24 (2013) 2712872.3.3. Contribution of the RESs to the reduction of GHG emissions over the next decade in the European Union (EU) . . . . . . . . . . . 2783. Life-cycle GHG emissions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2793.0.4. Impact of the vehicle charging prole on the overall GHG emissions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2803.0.5. Impact of the vehicle driving prole on the overall GHG emissions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2804. Life-cycle ownership costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2825. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2861. IntroductionClimate change ismore than ever a driving force ondifferentaspects of today's society. People are becoming increasingly awareof the real impacts of this change and want to play an active partin the mitigation of this issue. The search for more energy efcientandenvironmentallyfriendlyproductscanalsotranslateintoareduction of the energy bill.The transport sector is a major contributor of carbondioxideemissions and EVs are becoming more relevant on the future of thetransport system, for the Governments around the world, due to theirpotential to reduce GHG emissions and increase energy security [1].Withaccess toa more accessiblemobility, the demandfor newvehicles is rising rapidly and vehicle numbers are estimated to morethandoublebefore2050, withhighestgrowthratesindevelopingcountries [24]. In the absence of new legislation or policies that mayaffect the energy market, the transport energy use and associated GHGemissions, are projected to increase by 46% in 2035, compared to 2005values, accounting for 82% of the total increase in liquid fuelconsumption, while the use of liquid fuels for electric power genera-tiondeclines [5]. Tolimit theriseof theglobal longtermmeantemperature by 2 1C, a 5085% reduction of GHG emissions, comparedto 2000 levels, has been proposed [6].To meet the increasing demand for transport and at the sametime reduce GHG emissions and improve air quality, the paradigmof personal transportationhastochange. Thischangeembracesalternativevehicleandfuel technologyaswell asmarterinfra-structure, through the electrication of the powertrain, usingbatteriesor fuel cells, andtheuseof alternativefuels, suchasbiofuels, natural gas and hydrogen [79].AlternativeFuel Vehicles (AFVs), suchas thosepoweredbyCompressedNaturalGas(CNG), LiqueedNaturalGas(LNG)andLiqueedPetroleumGas(LPG)arealsootherviablelow-carbonalternatives.These are in hand with conventionalICE vehiclesintermsofenvironmental impacts, associatedwiththeproductionanddisposal. Theadditional impactsduringtheproductionanddisposal phase come from the additional material required for thegasreservoir. Intermsof investment, thesevehiclesare510%more expensive than the conventional vehicles but present loweroperatingcosts, duetothelower cost of thefuel. Duringtheoperation phase, these vehicles have lower GHG emissions whencompared with conventional ICEVs, with a signicant reduction inparticle emissions [1012]. However, some studies also shown anincreaseinNOxandCOdependingonthedrivingcycle[1315].AkeyaspectthatshouldalsobetakenintoaccountistheGHGupstream emissions, which vary with the fuel production processand geographic location, which could offset the GHG savings.HydrogenFuel Cell Vehicles (HFCVs) aresimilar toBEVs interms of havingahighefciencypowertrain. Akeydifferencebetween both technologies is the upstream GHG emissions asso-ciated with the generation and delivery of the energy carrier [16].Two key aspects that affect the viability of HFCVs are related withthe hydrogen storage and production [17]. In [18,19] severalhydrogenproductionandstoragetechnologiesarestudiedfroma Well-to-Tank (WTT) perspective. Despite some of themcontributetolower Well-to-Wheel (WTW) GHGemissions, theaddedcostassociatedwithdoesnotenablethemassicationofthe technology. Due to the energy intensive process of producinghydrogenbyhydrolysis, aHFCVwill requireuptothreetimesmore energy and will emit three time more GHG than a BEV [11].However, the use of HFCVs ina scenariowhere electricity isobtainedfromRESscouldhaveadvantagesoverBEVstoachievea higher autonomy.ImprovementsinconventionalInternalCombustionEngine(ICE)technology can reduce petroleum demand, however efciency alonecannot reduce GHG emissions to levels 80% below 1990 for the lightvehicleseet. The cost and availability of low GHG fuels will imposelimits to their use. To achieve signicant reductions in GHG emissionsthe future transport systems will require optimized combinations ofadvanced fuels and vehicle technologies [11].Powertrainelectricationisseenhasthesolutiontoamoresustainable transport system that can contribute to a reduction ofGHG emissions and the import of crude oil, due to a much higherefciency,from 2830% of an ICEV to 7485% of an EV, and zerotailpipe emissions [20], however due to a higher initial cost whencomparedwithconventional technology, takingseveral yearstooffset the investment, their penetration on the market is expectedbe slow [21].The presence of an energy storage device allows the EVs to be usedas a exible load and to support large scale renewable energygenerationthroughsmart rechargingmethodologies, however themassive electrication of the vehicles may bring additional problemsto the electric grid if not correctly managed [22,23]. Despite the use ofelectricity as the energy source for EVs and zero tailpipe emissions, it isfundamental toassessthesevehiclesintermsofeconomic, energyconsumptionandGHGemissions, takingintoaccountnotonlythevehicle LCA but also the electricity generation [24].The contribution of EVs to reduce GHG emissions must be assessedby using a cradle-to-grave perspective for the vehicles and WTW forenergy carries. In this context, a comparative LCA can be performed forICEVs and EVs. This type of assessment allows a detailed comparisonbetween several vehicle technologies and the identication of oppor-tunitiesoftechnologicaldevelopmentandimprovementinthelife-cycle phase of a vehicle. In this LCA, the vehicle technology analyzedwas the conventional motorization, represented by both gasoline anddiesel ICEVs, andelectricvehicletechnologyrepresentedbyPHEVsand BEVs. Fuel Cell Electric Vehicles (FCEVs) were not considered inthisstudyduetohighcostandtechnical issueswiththistypeoftechnology that still need to be addressed, such as hydrogen capture,storage and distribution, limiting their widespread adoption [7,25,26].Since the EVs environmental impact is directly related to the electricitygeneration mix, several scenarios from the European Union (EU) wereconsidered based on the renewable, nuclear and fossil fuel share.2. Life-cycle model and system boundaryThe model andrespective system boundaries for each vehicletechnology (ICEV, PHEV and BEV) are presented on Fig. 1. A globalR. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 272perspectiveisconsideredoverthevehicleslife-cycleandtheirkeycomponents, suchasthebattery, aswelltheelectricity generation.The manufacturing phase was considered common to all vehicles andcalculated based on the vehicle weight. An additional burden, due tothe use of a battery, was added to the EVs based on battery weight.Several LCA studies show that the most critical phase in termsof GHG emissions is typically the operation phase, whether it be aconventional orelectricvehicle[27]. Theexceptionisfor BEVscharged with electricity in a mix with low GHG emissions, wherethe production phase is the most critical. Gasoline and diesel ICEVshave the higher emissions during the operation phase followed byPHEVs and BEVs while the emissions during the production phaseare similar to all vehicle technologies, excluding the batteryproduction. Duringthevehicleproduction, thebatteryusedinPHEVs and BEVs is the most critical component in terms of GHGemissions, contributingwith3050%of total emissions, mainlydue to the materials and quantities required for the batteryproduction [2830,27].During the vehicle use phase, for the BEV and for the PHEV thevehicle Tank-to-Wheel (TTW) efciency was taken into account aswell real worlddrivingproles. GHGemissionsassociatedwiththe infrastructure for crude oil processing, transport and distribu-tionaswell forelectricitygenerationwerealsoconsidered. Theproductionand end of life phases for the considered vehicletechnologywerebasedon[31], whilethebatteryLCAinventorywas based on [29,32]. The LCA database for the consideredvehicleswascreatedusingtheEcoinventdataset[33]. ThisLCAintends to provide a comparison between EVs and ICEVs over theirentire life-cycle and also assess the impact due to the use prole ofthevehicleandenergysource. Thistypeof analyzeallowstheidentifyof themost critical aspects intermsof emissions andwhere actions can be taken in order to reduce them.2.1. Vehicle characteristics and assumptionsThevehicletechnologyaddressedwereICEVs, PHEVsandBEVs.ICEVs (both diesel and gasoline) were used as a baseline for the impactassessment of theother vehicletechnologyconsidered. Basedonmarketrelevance, theICEVisrepresentedbytheVolkswagenGolf(best seller in the European market), both for the gasoline and dieselversion, the PHEV and BEV are represented by the Chevrolet Volt andby the Nissan Leaf respectively. Since subcompact vehicles arebecoming more popular in a urban environment, two electric vehiclesthat t in this category were also considered, represented by the SmartED and Peugeot iOn (also sold in Europe re-branded as the Mitsubishii-MiEVandCitronC-Zero). SincetheSmart EDtsinadifferentcategorywhencomparedwiththeothervehiclesconsidered, bothdiesel and gasoline versions were included to be used as a comparisonbase. The vehicle main characteristics are listed on Table 1, based onmanufacturer data and average cost in the European market [34].Fuel Extraction andRefiningMaintenanceBattery RecyclingBattery DisposalVehicle Use Phase End of LifeICEV and PHEV specific motorization processesBEV and PHEV specific motorization processesElectricityGenerationBattery ProductionVehicle ProductionFig. 1. Model and system boundaries for the three vehicle technologies.Table 1Characteristics of the vehicles considered for each category.Characteristics ICEV PHEV BEVVW Golf 1.6 TDI VW Golf 1.4 TSI Smart CDI Smart Chevrolet Volt Nissan Leaf Smart ED Peugeot iOnEmissions gCO2=km 118 144 98 86 0(ED) 160(GD) 91(MD) Fuel consumption (l/100 km) 4.2 6.2 3.3 4.2 -(ED) 6.9(GD) 3.9(MD) Electricity consumption (Wh/km) 140 140 110 120Combustion engine (cc) 1600 1400 800 1000 1400 Electric motor (kW) 111 80 30 47Battery capacity (kW h) 16 24 16.5 16Battery weight (kg) 197 300 140 200Battery type Li-Ion Li-Ion Li-Ion Li-IonRange (km) 700+ 700+ 500+ 500+ 580 (80ED + 500ER) 160 135 120Curb weight (kg) 1240 1290 770 750 1715 1521 870 1080Note: For the PHEV were considered three driving scenarios: only gasoline drive (GD), only electric drive (ED) and mixed drive (MD). The fuel consumption is given for thevehicle running with the battery depleted.R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 273Tocorrectlyassess the impact of vehicle interms of GHGemissions during the use phase, the energy losses in a vehicle needto be characterized. For ICEVs these losses are translated directlyto fuel consumption, however since BEVs and PHEVs requireenergy from the grid to charge the batteries, the additional lossesin the transmission and distribution system must be accounted for,usually around 910%[3537]. Forinstance, tochargea 16 kW hbattery connected to a power grid with an efciency of 90% it willrequire a generation of 17.6 kW h.InaEVs, energylosses occur intheenergystoragesystem(usuallyabattery), at thedrivetrain(thegroupof mechanicalcomponents responsible to transmit the power from the motor tothewheels topropel thevehicle) andinthepower electronicmodule(responsibleforthemotorcontrol, regenerativebrakingand charging) [38,39].Theenergystoragesystemcanbeconstitutedbybatteries,supercapacitorsor bythecombinationof them. Nowadays, thebattery is the main component of the Energy Storage System (ESS)dueto alower costwhencompared withother storage technol-ogies [4043]. Lithium-Ion batteries are the most common in EVsduetoaspecic energyupto400 Wh/kgandahighspecicpower (up to 10 kW/kg) when compared with lead-acid batteries,whichhaveaspecicenergydensitybetween20and30 Wh/kgandaspecicpowerupto400 W/kg[44]. ThebatteriesusedinEVshaveanenergy capacityintherangeof1085 kW handanefciencyofabout70%95%[45]. Internalresistance, typicallyinthe range of milliohms, which increases both with cycling and age,is one parameter that contributes tothe batteryefciencybycausing a voltage drop under load and by reducing the maximumoutput current affecting the charge/discharge rate [4648].Battery life and capacity are key aspects for the wide adoptionofelectricvehicles. To estimatethepotential savingsin terms ofGHGemissionsandalsotheTotal CostofOwnership(TCO)ofavehicle it is fundamental to estimate the life of a battery under realworld operation. The battery state of health is greatly inuencedby the load and environmental conditions [49,50]. Depending onthe lithium-ion cell chemistry, both high and low State of Charge(SOC)contributeto thedeteriorationofthebattery performanceand lifetime. Overcharge, over-discharge, high Depth of Discharge(DOD)s and high temperatures also inuence the fast decay of thebattery life and low temperatures can also have a negative impact,mainly during the charging phase [46,51].Modern electric vehicles employ Battery Management System(BMS) that monitor the battery state of health and avoid workingpointsthatcontributetoacceleratedaging, forinstancebycon-trollingthebatterypacktemperatureusingactiveheatingandcooling systems, as well as by avoiding the battery overcharge andover-discharge. Depth of discharge is also managed by limiting theamount of energy that can be drawn (from a 24 kW h pack, only20 kW h or less are used). It is expected, that the battery, managedby the BMS, working within specied boundaries will last, with ahigh degree of probability, the life time of a vehicle. The user canalso contribute to the mitigationofthe aging effectsby avoidingcompletebattery charge/dischargecyclesandbyminimizingtheload applied to the battery due to fast accelerations, by adoptingan eco-driving prole.The powertrain includes the Internal ICE and/or electric motor,transmission, drive shaft, differential and drive wheels. In the caseof diesel ICEs, the efciency is around 3035% in the ideal speedrange, decliningoutsidethisrange, whilegasolineICEshaveanefciency of 1825%, meaning that around 80% of available energyis lost as heat.In EVs, the most common types of electric motors are perma-nent magnet motors andinductionmotors withanefciency,depending on the type of motor, that can go from 85% up to 95%from a wide speed range [52].In the drivetrain, responsible by the transmission of themechanical energy generated by the engine to the road, the energylosses occur in the differential andnal drive. In an ICEV, the oilpumpisresponsiblefor3040%of total powerloss, theclutchcontributeswithadditional 2025%andgearmeshing, bearings,bushings and drag on the gears caused by the gear oil areresponsiblefortheadditional losses. Inthecaseofrearandall-wheel drive vehicles the power losses in the differential tend to belarger due to a 901 turn in the torque path and are usually 610%,while losses from the drive shaft are 0.51% of total losses [53,36].The total losses in the drivetrain account around 58%.One characteristic of electric motors is the ability to supply themaximumamount of torqueoveraverylargerangeof speeds,eliminating the need for gear shifting in EVs and giving the vehicleasmoother accelerationandbraking. Themechanical power isWall SocketL2 Slow Charger (95-97%)DC Fast Charger (91-94%)Battery(90-95%)Inverter(90-98%)ElectricMotor(85-96%)Drivetrain(87-93%)DrivingCharging Elec. Gen. Reg. BrakeGenerationTransmission(98-99%)Distribution(91-93%)Wall SocketFig. 2. Range of efciency of the different components in the energy path of an EV.Table 2WTW systemefciency for the electric powertrain, consideringthe power lossesalongtheelectricitypath, usingaDCfastcharger(DC)andastandard240VACcharger (L2), with Lithium-Ion batteries as energy storage. It should be noted thatfor the overall system efciency the battery efciency was accounted twice due tothe charge and discharge cycles.System components Global system efciency (%)Minimum MaximumTransmission 98 99Distribution 91 93L2 Charger (L2) 95 97DC Fast Charge (DC) 91 94Battery 90 95Inverter 90 98Electric Motor 85 96Drivetrain 87 93WTT (w/L2) 76.2 84.8WTT (w/DC) 73.0 82.2TTW 59.9 83.1R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 274delivered directly, or through a simple gear reduction step to themain-shaft of the transmission, so the only loss sources arewindage, frictionanddrag, resultingintotalat-the-wheellossesas low as 1.5% to 2%. The Power Electronics Module (PEM) controlsthe energyow from the battery to the motor and vice versa. Themain component is an inverter with an efciency of 90-98%. Othercomponent tobetakenintoaccount, bothforPHEVsandBEVsefciency, isthechargertypicallywith9197%efciency. Fig. 2shows the conversionpathof energy, fromgenerationtothewheels.Table2summarizestheefciencyof eachsystemalongtheenergy path to power an EV. WTT and TTWefciency wascalculated using (1) and (2).WTT trans: dist: charger batt:1TTW batt: inv: electricMotor trans:22.2. Use phase and main factors that contribute to EVs energyconsumptionThe phase during the vehicle life-cycle that dominates theoverall GHG emissions is generally the use phase, regardless of thetypeof vehicle, beingthegasolineICEVtheonewithahigherenvironmental impact, according to several LCA studies [28,54,55],thereforeitscharacterizationwithahighdegreeof accuracyisdesirable. Usually the emissions from motor vehicles are based inspecicdrivecyclesapprovedbyGovernmental entities, suchasNew European Drive Cycle (NEDC) in Europe or FTP-75 in USA, toassess vehicle fuel consumption, performance and emissions [56].These type of cycles are important since they compare and check ifa specic vehicle meet Government requirements. However theydo not take into account some factors that contribute heavily to avehicle energy consumption, such as the path elevation prole andthe driving prole. Fig. 3. Data acquisition system installed on the EVs [55].0.000.050.100.150.200.250.300.350.40010000200003000040000500000 10 20 30 40 50 60 70 80 90 100 110 120 130 140Energy Consumption [kWh/km]Power [W]Speed [km/h]Leaf [kW] Volt [kW] Smart [kW] iOn [kW]Leaf_er [kW] Volt_er [kW] Smart_er [kW] iOn_er [kW]Leaf [kWh/km] Volt [kWh/km] iOn [kWh/km] Smart [kWh/km]0.000.100.200.300.400.500.600.700.800.901.00010000200003000040000500000 10 20 30 40 50 60 70 80 90 100 110Energy Consumption [kWh/km]Power [W]Speed [km/h]Leaf [kW] Volt [kW] Smart [kW] iOn [kW]Leaf_er [kW] Volt_er [kW] Smart_er [kW] iOn_er [kW]Leaf [kWh/km] Volt [kWh/km] iOn [kWh/km] Smart [kWh/km]Fig. 4. Power and energy consumption for the Nissan Leaf, Smart ED and Chevrolet Volt for a 01 (top) and 7.21 (bottom) slope. Some data points from experimental runs areidentied with the subscript_er with an error of less than 5% when compared with the mathematical model from Eq. (7).R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 275Inthecaseof EVs, sincetheyareequippedwithanelectricmotor that can work as a generator, they can recharge the batterywhengoingdownhillorduringthebrakingphase, reducingtheoverall energy consumption, aspect that is not taken in to accountin drive cycles suitable for ICEVs.In order to assess the energy consumption, and associated GHGemissions, a data acquisition system was developed and installedin the EVs considered in the study. The system was constituted bya6degreesof freedomXSensMTI-gInertial MeasurementUnit(IMU), with a three axis accelerometer, gyroscope, magnetometer,GPS and barometric sensor for instantaneous velocity and positionmeasurements; a Fluke i410 current clamp and a Metex M3640Dmultimeter for instantaneous power measurements, bothcon-nected to a computer, through RS-232 for data logging up to 10 Hz(Fig. 3).Using this data acquisition system it is possible to correlate theenergyconsumptionwiththeroadproleandwiththedrivingprole. Thissystemalsoallowsthebreakdownof theinstanta-neouspower consumptionintothepowerrequiredtoovercometheforces actuatingonthevehicle. Amoving vehicle, atagivenspeed, requires power toovercomethefollowingforces: aero-dynamicdrag,rollingresistance, gravitywhenascendingaslope0200400600800100012000%10%20%30%40%50%60%70%80%90%100%Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecgCO2e/kWhPolish Electricity Mix: Share and GHG Emissions by source for 2011Coal CHP Fuel Gas Hidro Wind gCO2e/kWh01002003004005006000%10%20%30%40%50%60%70%80%90%100%Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecgCO2e/kWhPortuguese Electricity Mix : Share and GHG Emissions by source for 2011Coal CHP Fuel Gas Hydro Wind PV gCO2e/kWh0204060801001201401600%10%20%30%40%50%60%70%80%90%100%Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecgCO2e/kWhFrench Electricity Mix: Share and GHG Emissions by source for 2011Nuclear Coal Fuel Gas Hydro Wind gCO2e/kWhFig. 5. Electricity mixes share and associated GHG emissions.R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 276and vehicle inertia when accelerating given by Eqs. (3), (4), (5) and(6) respectively:Pdrag 12 Cd A vvwind33Proll m g v Crr cos 4Pslope m g v sin 5Pacc m a v 6where Crr is the rolling coefcient, Cd is the drag coefcient, isthe air density, m is the vehicle mass, A is the vehicle frontal area,vandvwindcorrespondtothevehicleandwindvelocityrespec-tively, a is the vehicle acceleration, g is the gravity acceleration and is the terrain slope, in degrees.Eq. (7) representsthetotal powerrequiredtoovercometheforces acting over the vehicle considering the motor and transmis-sion efciency and the auxiliary power, denoted by Paux, requiredfor the control equipment, radio, lights etc.Ptotal Pacc Pslope Pdrag Pfrictionmotor trans Paux7On Fig.4 the comparison between the EVs in terms of powerand energy consumption in for aat road prole is presented. Dueto a lighter weight, the Smart ED is vehicle with the lower energyconsumptionfollowedbytheNissanLeaf andbytheChevroletVolt, howeverat higherspeedstheChevrolet volt requireslesspower and consequently less energy due to a better aerodynamiccoefcient(Cd)whencomparedwiththeNissanLeaf (0.28and0.29 respectively).0200400600800100012000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23gCO2e/kWhPolish Electricity Mix: GHG Emissions variation during a dayJan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec01002003004005006007000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23gCO2e/kWhPortuguese Electricity Mix: GHG Emissions variation during a dayJan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec0204060801001201401601800 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23gCO2e/kWhFrench Electricity Mix: GHG Emissions variation during a dayJan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecFig. 6. Average GHG emissions variation during the day, over a year, for the considered mixes.R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 2772.3. Energy sources: emissions assessment2.3.1. Crude oil extraction and reningIn Europe, transport is responsible for almost 20% of the GHGemissions, with 930 MTCO2e in 2010 according the Eurostat latestdata, with passenger cars contributing with almost 12%. EUtransportemissionscurrentlycontributewith3.5%of theglobalCO2emissions. Relative CO2emissions fromtransport havedecreased over the past years, from 2005 to 2010 they decreased3.4% despite the increase of 5% in the motorization rate during thesame period, due to a proposed a legislative framework to reduceGHG emissions from the transport sector [57,58].The International Energy Agency (IEA) predicts that the crudeoil consumptionwill increase27%until 2030, from83 MMbbl/d(million barrels of oil per day), in 2009, to 102 MMbbl/d, in 2030.Crudeoilextraction, transportandrening, accountsinaverage,for about 18% of WTW of GHG emissions [59]. The quanticationof the WTW emissions is divided intove components associatedwiththe petroleumproduction: (i) extraction, (ii) aring andventing, (iii) fugitive emissions, (iv) crude oil transport and(v) rening.InEuropethecrudeoil comes fromalargenumber of oilelds around the world and each of themhas specic GHGemissions, dependingonthetype. Thecarbonintensityofcrudeoil rangesfrom4to50 gCO2e=MJ (gramsof CO2-equivalent permegajoule), with an average of 1372 gCO2e=MJ [59]. Extraction ofcrudeoil fromtar sands, averyenergyintensiveprocess, con-tributingtohighGHGemissions andtheIEAprojects that 8%(8.9 MMbbl/d) of the world crude oil will come from this source in2035 [60].Additional emissionsfromfuel combustioninmotorvehiclesare about 73 g CO2e=MJ 152 gCO2e=km, for gasoline with34.8 MJ/l andaconsumptionof 6 l/100 kmand127 gCO2e=km,for diesel with 38.4 MJ/l and a consumption of 4.5 l/100 km). life-cycleemissionsfordiesel andgasolineduetofuel burningarearound 3200 gCO2e=l and 2900 gCO2e=l respectively.2.3.2. Electricity mixGenerallytheusephasedominatestheoverallimpacts, how-ever, for an electricity mix with a large contribution from RES, theproductionphaseofaBEVistheonethatdominatestheoverallenvironmental impact [29,61]. To assess correctly the impact of aBEV the electricity mix used to charge the vehicle must be knownwith a good degree of certainty.The primary energy source used for electricity generationcontributes directly to the overall GHG emissions of the generationmix, which in turn affect the use phase GHG emissions of an EV.Emissions associated with fuel production are more or lessconstantovertime, mainlyaffectedfromwherethecrudeoil isextracted, andcouldbeconsideredoveralargegeographicareaunlike electricity generation that depends directly from the shareandtypeofpowerplantsonthesystemandcouldvarysigni-cantlyfromcountrytocountry. Other keyaspect totakeintoconsideration, is that theshareof eachtypeof energysourcecontributing to the overall electricity mix varies daily and is alsodependent on the season [62].Most studies consider an average electricity mix over a year fora given geographic region that can lead to results, in terms of GHGemissions during a vehicle life-cycle, that do not reect the reality,sincetheemissionsassociatedtodifferent chargingperiodsarenot constant during the day. One objective of the presented studyis to assess the impact of the electricity mix and the EV chargingproleintheoverall life-cycleemissions. Toassessmoreaccu-ratelythecontributionfromelectricitygenerationtotheoverallLCA, aWTTanalyzeswas performedconsideringthefollowingmixes found in Europe:

A mix with high GHG emissions represented by Poland.

AmixwithverylowGHGemissionsrepresentedbyFrance.item A mix with large contributionfrom RESs represented byPortugal.On Fig. 5 and on Fig. 6 it is easily noticeable that the mix variesconsiderablyovertheyearandalsoduringtheday, duetotheintermittence and share of RESs for the overall electricity genera-tion. This variation is very noticeable in mixes with a largecontributionfromRESsorNuclear, suchasPortugal andFrance,where a drop in the contribution from RESs must be compensatedbyusingfossil poweredplants, leadingtohigher overall GHGemissions [63]. In a mix mainly ruled by fossil fuel powered powerplants, the associated GHG emissions are fairly constant over theyear and over the day. This type of mix has the higher associatedemissions with an average value of 979 gCO2e=kW h over the year,while the Portuguese and French mix had an average value of 376and 103 gCO2e=kW h respectively. It should be noted that for thecaseofPolandduetoahighshareoffossilfuelpoweredpowerplants and low variation in the electricity mix during the year, thedata for a typical month was considered for all year.TheGHGemissionswerecalculatedtakingintoaccount thestandard emissions from Table 3 [64].2.3.3. Contribution of the RESs to the reduction of GHG emissionsover the next decade in the European Union (EU)TheEuropeanCommission(EC) is committedtoreducetheGHGemissionsto8095%bellow1990levelsby2050andtheinvestment in RESs is seen as a key factor for the decarbonizationobjective whileatthesametimereducingthedependencefromfossil fuel sources and ensuring security of energy supply. Accord-ing to the National Renewable Energy Action Plans, in a referencescenario, is estimated that the RESs in EU will contribute with 25%in 2015 and 31% in 2020. In an increased energy efciencyscenario, the contributionfromRES is expectedtobe 26%in2015and34%in2020[65]. Projectionsestimatethat REScon-tribution will increase to 902 TW h, a 32% growth, by 2015 and to1216 TW hby 2020, a 87%growth, fromthe 2010 652 TW h.Between2010and2020, theRESs withthehighest growthinterms of installed capacity will be solar, with a 250% growth, andwind, with a 151% growth. In 2020, hydro and wind will continueto be the sources with the highest share, both in terms of installedcapacity and electricity generation, with 495 TW h and 370 TW hrespectively. This increase in the share of RESs in the totalelectricity generation, to 3040% in 2020, will lead to a reductionintheelectricitymixemissions of 1466 gCO2=kW hfromthe312364 gCO2=kW hin2009. Additional measures topromoteenergy efciency will lead to additional reductions. Due to this factTable 3Typical lifecycle GHG intensity by type of generator [64].Technology Emissions (gCO2e/kW h) ObservationsCoal 1050 Without scrubbingCoal 960 With scrubbingWind 910Hydroelectric 13 Run of riverHydroelectric 10 ReservoirBiomass 1441Solar PV 32 Polycrystalline siliconeNuclear 66Natural gas 443Diesel oil 778CHP 354R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 278the EV is the only vehicle that gets cleaner, due to an increasinglycleaner energy source, during its life-cycle.3. Life-cycle GHG emissionsFig. 7presents thelife-cycleGHGemissions for eachvehicletechnology. The emissions for the EVs were calculated based on theWTWefciencyandontheaverageelectricitymixemissionsfor2011 (979 gCO2e=kW hfor the Polish mix, 376 gCO2e=kW hfor thePortuguese mix and 103 gCO2e=kW h for the French mix). The resultsshow that the overall emissions from EVs are highly dependent onthe electricity mix. The Chevrolet Volt shows higher GHG emissionsthan conventional Internal Combustion Engine (ICE) technology. Theadditionalweight of PHEVs due to the two engines has a negativeimpactsincemoreenergyisrequiredtomovethevehicle. Ontheother hand, the Smart ED shows the lower GHG emissions since it isthe lighter of the EVs analyzed. As expected, charging an EV with anelectricity mixwith lower GHG emissions reduces signicantlytheoverall life-cycle emissions. In this case, the PHEV Chevrolet Volt canhave overall emissions similar to a sub-compact conventional vehicle,the Smart diesel, despite their signicant weight difference.Fig. 8 presents the share of the life-cycle emissions by vehiclecategory, for each vehicle technology and electricitymix. For theInternal CombustionEngine(ICE) conventional technology, thedominant phase is by far the operation phase, accounting for 8590% of the global emissions, mainly associated to fuel combustion.ForEVs, thebattery productionaccountsfor alarge share oftheemissions, almost the same as vehicle production, due tothehighly energy intensive processes required to obtain the materialsusedinthe batteryproduction. The emissions fromoperationphase are highly dependent on the electricity mix. In a mix heavilydependent on fossil energy sources, such as the Polish mix, the usephase account for 7080% of the overall emissions. For anelectricitymixwithlower GHGemissions, suchas theFrenchmix where the generation is mainly from nuclear, the productionphase is the most signicant to the overall emissions, accountingfor around 7075%.GHG emissions from maintenance and vehicle disposal repre-sent less than 10% of the overall emissions. Despite EVs having lessmaintenancecomparedtoconventionalvehicles, theshareasso-ciatedwithmaintenancewill behigher sincetheyhaveloweremissions. Additional emissions for thedisposal of thebatteryshouldalsobetakenintoaccount. However, itshouldbenotedthat batteries still retainsomecapacityat theend-of-life, and,thus, canbereusedonotherapplicationssuchasstaticenergystorage, where the requirements are moreexible thus extendingtheir useful life.050100150200250gCO2e/km050100150200250gCO2e/kmFig. 7. GHG intensity per km traveled and by electricity mix considered (top: compact vehicles; bottom subcompact vehicles) category. Note: In the case of the ChevroletVolt, the FE denominates the full electric mode.R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 2793.0.4. Impact of the vehicle charging prole on the overall GHGemissionsAsstatedbeforetheemissionsassociatedwiththeelectricitymixvarysignicantlyduringtheyearandevenduringtheday.These variations are associated with the intermittent nature of therenewable resources. In this case, some conventional powerplants,powered by fossil fuel, must be in standby increasing theoverallemissionsassociatedtoelectricitygeneration. InAutumnandWinterthecontributionfromRESsissignicantduetotherainyandwindynatureof theseseasons. However duetothevariations of theweather conditions betweenyears this is notalways veried.Other aspectthatinuencestheoverall emissionsfor agivenvehicle is when the vehicle is charged on a daily basis. This is morerelevant for electricity mixes with a large share of RESs. For mixeswithalargecontributionfromfossilornuclearsources, suchasPoland or France respectively, the emissions fromelectricitygeneration are more stable. However, in the case where RESrepresentsalargeshareoftheelectricitymixthisisnotalwaystrueduetodifferentlevelsofrenewablegenerationavailableateach hour. Since coal and gas fueled power plants require asignicant start up time, that can go up to 8 h, they must be keptat a minimum level of operation independently of theconsumption. ThisaspectisvisiblewherewascleanertochargeanEVduringthedayratheratnightforseveralmonthsforthePortuguese electricity mix during 2011 (Fig. 10). For a mix were themainshareof RESishydroandwindsuchasPortugal, duringwinter months, charging at night will emit approximately less 2050%whencomparedwithachargeduringtheday. Forsummermonths, this situationis not always true, as for somemonthscharging during the day will emit approximately less 423% whencompared with a night charge.For mixes witha large contributionfromfossil or nuclearpower, theemissions variationduringthedayfromelectricitygenerationarenotusuallysignicant(Figs. 9and11), exceptinwinter months, meaning that charging at night or during the daywillnothaveamajorimpactontheoverallemissions. Howeverthe best time to charge an EVis at night, since the energyconsumedbytheEVwillcontributetoaatterloadproleandalsotomaximizetheamountof timethatpowerplantsremaincloser to their nominal capacity.3.0.5. Impact of the vehicle driving prole on the overall GHGemissionsToassesstheimpactofthedrivingproleseveralrealworlddriving cycles were performed in two predenedroutes, one0%10%20%30%40%50%60%70%80%90%100%0%10%20%30%40%50%60%70%80%90%100%Fig. 8. ConventionalandelectricvehicletechnologiesGHGemissionsbyshareandby electricitymix considered, dividedintocompact(top)andsubcompact(bottom)category. Note: In the case of the Chevrolet Volt, the FE denominates Full Electric mode.R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 280urban and other suburban, under different driving conditions(aggressive, normalandECO)andwithdifferentsettingsfortheclimate control (A/C OFF, A/C in cooling mode and A/C inheating mode).The aggressive driving prolediffersfrom thenormaldrivingprolemainlyintheaccelerationandbrakingphase, wheretheaggressive has fast accelerations and sudden braking to maximizethe energy consumption. The ECO driving prole is characterizedby slow accelerations, lower top speed and the braking is mainlyduetotheregenerativebraking, tominimizetheoverall energyconsumption. For the driving cycles, the climate control was set inmanual modeat 21 1Cwiththefanat mediumspeed, bothforcoolingandheating, beingthebaselinewiththeclimatecontrolOFF. The use of a manual setting for climate control allows a betterunderstandingabout howit canaffect theoverall energycon-sumptionandassociatedemissions. Fig.12showstheelevationprole for the urban and extra urban routes used to assess the EVenergy consumption.The data was captured using the data acquisition system fromFig. 3. OnTable 4some results of the runs performedunderdifferent conditions and with different settings for climate controlarepresented. Inthissection, all thepresentedresultsrefertotrials performed with a Nissan Leaf. The remaining EVs consideredalongthepaperwill haveasimilarresultsbutreferredtotheirspecic energyconsumptionfor thepresentedscenarios, byaweighting factor.0,0020,0040,0060,0080,00100,00120,00Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecgCO2e/kmVolt FE (00:00-08:00) Leaf(00:00-08:00) iOn(00:00-08:00) Smart ED(00:00-08:00)Volt FE (10:00-17:00) Leaf (10:00-17:00) iOn (10:00-17:00) Smart ED (10:00-17:00)Volt FE (19:00-00:00) Leaf (19:00-00:00) iOn(19:00-00:00) Smart ED (19:00-00:00)Fig. 10. GHGemissionsassociatedtotheoperationphase, fortheconsideredEVs, basedonthetimeofdaywhenthevehicleischarged, consideringthePortugueseelectricity mix for 2011. Only the emissions during the specied charging intervals were accounted for, emissions during 2 h in the morning and in the afternoon, when thevehicle is being used for commuting, were not accounted.80,00100,00120,00140,00160,00180,00200,00Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecgCO2e/kmVolt FE (00:00-08:00) Leaf(00:00-08:00) iOn(00:00-08:00) Smart ED(00:00-08:00)Volt FE (10:00-17:00) Leaf (10:00-17:00) iOn (10:00-17:00) Smart ED (10:00-17:00)Volt FE (19:00-00:00) Leaf (19:00-00:00) iOn(19:00-00:00) Smart ED (19:00-00:00)Fig. 9. GHG emissions associated to the operation phase, for the considered EVs, based on the time of day when the vehicle is charged, considering the Polish electricity mixfor 2011. Only the emissions during the specied charging intervals were accounted for, emissions during 2 h in the morning and in the afternoon, when the vehicle is beingused for commuting, were not accounted.R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 281Table 5 summarizes the average energy consumption andestimated range for the considered scenarios while Table 6, basedontheenergyconsumption, showstheassociatedemissionsbyscenarioandelectricitymix. As expected, aneconomicdrivingprole is more efcient than an aggressive one, which can reducethe driving range by 90 km due to the increased energy consump-tion in 47%. The use of climate control also has a signicant impact,increasingtheenergyconsumptionin24%incoolingmodeand61% in heating mode for the ECO driving prole.In Fig. 13 is possible to compare the impacts, per km traveled,duetodifferent drivingproleandtheuseof climatecontrolagainstabaselinescenario, fortheconsideredelectricitymixes.For an electricitymix with low GHGemissions the way an EV isusedwillnotaffectthelife-cycleemissionsinasignicantway.HoweverforanmixwithhighGHGemissionsthisisnolongervalid. AstheelectricitymixassociatedGHGemissionsincrease,more relevant will be the impact of the use prole of the EV. Fig. 14correlatesthevehicleenergyconsumption(inWh/km), directlyrelated with the emissions per km traveled (in gCO2e=km), and theelectricitymixlife-cycleemissions (ingCO2e=kW h) ontheEVoverall life-cycle GHG emissions.4. Life-cycle ownership costsFrom a consumer stand point a key aspect of EVs is how muchwill a given technology cost during its life-cycle. To evaluate it, aneconomic assessmentfor each vehicle technology duringthe fulllife-cycle was performed, intending to estimate the cross point intime which a given vehicle technology takes advantage overanother. For the economic analysis the purchase, operation,depreciation and capital cost was considered based on EU marketand manufacturer data.The operational cost, per year, was calculated based on a totaldriven distance of 20 000 km and the average fuel and electricityprices inEUin2012(1:42for diesel, 1:51for gasolineand0:17=kW h for electricity) [66]. It should be noted that in certaincountries the electricitycost varies withthe hour of theday.5,0010,0015,0020,0025,0030,00Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecgCO2e/kmVolt FE (00:00-08:00) Leaf(00:00-08:00) iOn(00:00-08:00) Smart ED(00:00-08:00)Volt FE (10:00-17:00) Leaf (10:00-17:00) iOn (10:00-17:00) Smart ED (10:00-17:00)Volt FE (19:00-00:00) Leaf (19:00-00:00) iOn(19:00-00:00) Smart ED (19:00-00:00)Fig. 11. GHG emissions associated to the operation phase, for the considered EVs, based on the time of day when the vehicle is charged, considering the French electricitymix for 2011. Only the emissions during the specied charging intervals were accounted for, emissions during 2 h in the morning and in the afternoon, when the vehicle isbeing used for commuting, were not accounted.Fig. 12. Urban (top) and extra urban (bottom) elevation prole for the two routes considered in this study to assess the driving prole impact in the overall EV energyconsumption.R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 282In Portugal during the day (08:00 h22:00 h) the electricity costs0:14=kW h while at night (22:00 h08:00 h) it costs 0:08=kW h,which translates to a 43% reduction in the electricity cost per yearonly by charging the EV at night. Other operational costs such asmaintenance and insurance were also considered for each vehiclebased on market and manufacturer data. An ination and interestrate of 2.5%/year and 5%/year,respectively,were also considered.Theusescenarioconsideredfor thePHEVwas 80%inelectricmode, for commute during the week, and 20% in extended rangemode for longer trips. Table 7 summarizes the costs associated toeach vehicle used to calculate their annual operational costs.Fig. 15shows that EVs havelower costs thanconventionalvehicles, duetoasimplermaintenanceandtheequivalentelec-tricity cost required to travel a given distance is much lower thangasoline and diesel. The PHEV has a higher maintenance cost duetothecombustionengine, howeverlowerthanICEVs. Themaincostsduringthelifetimeof EVsaretheinitial investment andassociatedinterest, accountingaround7080%of thetotal life-cycle costs, while for conventional vehicles they only account upto 50%.From Fig. 16 it is possible to estimate the break even point foreach EVs when compared with a conventional vehicle. Due to thehigher initial cost of an EV the turn around time could be up to theentirelife-cycleof thevehicle(910years), despitetheirloweroperational cost. This is even worst for the PHEV since it has twotypes of drivetrains, a conventional and an electrical one, making itthe most expensive version of all of them. As expected, thecheapest EV is the Smart ED, because of its lower initial cost, butit should be noted that this vehicle is in a different category fromthe rest in terms of capacity.5. ConclusionWith vehicle emissions becoming stricter over the next decadeduetogovernmental requirements tomitigateglobal warmingandlocal pollution, aswellastherequirementstoimprovefueleconomy, manufacturers are continuously improving ICE technol-ogy and are also developing new technologies such as BEV, HybridElectricVehicle(HEV), PHEVandHFCV. Themarket viabilityofBEV and PHEV is highly dependent on the TCO and how the costsaredistributedalongthevehiclelifetime. Forthesevehicles, themajor cost is the acquisition cost in contrast with the lower costsof ICE vehicles, leading to a slow market penetration.For diesel, gasoline LPG and CNG vehicles, better control overfuel injection, spark timing and a better fuel/air mixture could leadto additional reduction in emissions, by adjusting the fuel burningparameters depending on the vehicle conditions. Advanced gaso-lineanddiesel vehicles, withstop-star capabilities, downsizedTable 4ResultsfortheNissanLeaf real worlddrivingcycles, forurbanandextraurbanroutes, under different driving proles and climate control settings.Type Length(km)Energy consumption(Wh/km)Speed (km/h)ACOFFAC ONCool.AC ONHeat.Max./MedianDrivingproleUrban 16 155.4 177.7 213.4 86/43 AggressiveUrban 16 126.6 148.1 182.3 86/47 NormalUrban 16 95.5 122.2 164.8 61/42 ECOUrban 17 135.1 153.7 183.2 96/58 NormalUrban 17 103.9 128.5 167.8 82/45 ECOUrban 16 114.9 135.6 168.6 71/42 ECOExtraUrban16 157.2 172.7 197.4 115/63 AggressiveExtraUrban16 143.0 157.3 180.1 100/67 NormalExtraUrban16 138.1 154.5 180.6 85/70 NormalExtraUrban16 132.8 148.3 173.0 100/77 ECOExtraUrban16 129.3 145.0 170.0 100/60 ECOTable 5Energy consumption and estimated range for the Nissan Leaf based on the drivingprole and climate control settings based on data acquired from several runs in aurban and suburban environment.Driving style AC OFF AC ON Cool. AC ON Heat.Wh/km km Wh/km km Wh/km kmAggressive 155.4 129 177.7 113 213.4 94Normal 131.0 153 151.0 132 182.8 109ECO 104.7 191 129.0 155 167.1 120Table 6EstimatedGHGemissionsfortheNissanLeaf, perkmtraveled fortheoperation,basedonthe driving prole andclimate control settings, for the consideredelectricity mixes. The emissions considered, for each mix, were the average valuesfor 2011 (979 gCO2e=kW h for the Polish mix, 103 gCO2e=kW h for the French mixand 376 gCO2e=kW h for the Portuguese mix).DrivingstyleAC OFF AC ON Cool. AC ON Heat.PLMixPTMixFRMixPLMixPTMixFRMixPLMixPTMixFRMixAggressive 177 68 19 202 78 21 243 93 26Normal 149 57 16 172 66 18 208 80 22ECO 119 46 13 147 56 15 190 73 200,0050,00100,00150,00200,00250,00300,00350,00Nissan Leaf PT2011 Mix Nissan Leaf FR2011 Mix Nissan Leaf PL2011 MixgCO2e/kmFig. 13. Impact of the operation phase on the Nissan Leaf on the overall life-cycle emissions per km traveled, by electricity mix. The baseline scenario is based on the rangefor a full charge provided by the manufacturer (122 gCO2e=kW h for the Portuguese mix, 83 gCO2e=kW h for the French mix and 210 gCO2e=kW h for the Polish mix).R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 283turbo-charged engines and with some level of hybridization due tothe use of an electric motor coupled with small lithium-ion batterieswill contribute to additional reduction of vehicles emissions withoutgreatly increasing the cost of ICE vehicles. The use of alternative fuels,such as bio-diesel, ethanol and hydrogen, may also lead to lower lifecycle environmental impacts, due to the improvements not only invehicle technology but also in the fuel production cycle.Thereductionof vehicleweightbyusinglightermaterialsisanotherstrategytoreduceenergyconsumptionandGHGemis-sions which can be implemented independently of the powertraintechnology. Abreakthroughincarbon-ber cost wouldbeveryvaluable, mainly for body parts.The future of the transport systemwill pass through theincreasingelectricationof thepowertrainsinceEVs aremoreefcient, canbecleanerthanICEVsandsignicantlyreducethedependence from fossil fuels. Although the payback period couldtakeupto912yearsduetothehigherinitial costofaBEVorPHEV, the long term EV ownership costs is lower when comparedwithaconventional vehiclefromthesamecategory. Improve-ments in lithium-ion batteries and super-capacitor technology byusing advanced materials will increase their lifetime and efciencycontributing to a longer life cycle andoptimizing the batteryproduction will reduce their cost, making EVs more affordable. Thesecond life of EVs batteries, for use in grid storage for instance, willalsocontribute, toacertainextent, toreduceoverall environ-mental impacts associated to EVs.Also, theelectricitypriceisalsomorestablewhencomparedwith gasoline or diesel, and with the constant increase in the costofcrudeoilthepaybackperiodofanEVmaybereducedinthefuture. With improvements in the battery technology, the cost ofthe EVs will likely decrease, due to the reduction of massproduction costs and increased efciency, making the EV aincreasingly attractive choice from the consumer stand point.For all vehicle technologies analyzed, the operation phase is theone that most contribute to the GHG emissions over the life-cycleof a vehicle, contributing with 8590% for a conventionalvehicle while for an EV this is highly dependent on the electricitymix. For a mix dominated by fossil sources, the operation phase ofan EVwill be dominant representing more than 75% of thevehiclelife-cycleemissions. For amixwithaheavy contributionfromnuclear or renewableenergysourcethedominant phasewill be the vehicle andbatteryproduction, withat least 50%of the life-cycle emissions, largely dependent on the batterycapacity. Otherrelevantaspect isthatanelectricitymixwithalarge contribution fromRESs can have signicant emissions,duringspecicperiods(hoursordays). Theintermittentnatureof renewable sources requires fossil power to be in standby to takeover thegenerationinthecaseof failurefromtherenewablesources.Duringtheoperationphasebesidesthedirectimpactof theelectricity mix in the overall emissions of EVs, the way the vehicleisoperatedisakey aspect. A veryaggressivedrivingstyle, with10337697950100150200250300350105125129131151155167178183213gCO2e/kWhgCO2e/kmWh/kmFig. 14. Variation of life-cycle GHG emissions, per km traveled (based on electricity mix and energy consumption for the Nissan Leaf). The emissions per km traveled takeinto account the emissions associated with vehicle production.Table 7Economicperformanceforthevehiclesconsideredperyear. Theconsideredcostsweretheinvestmentandownershipcostsperyear(whichincludefuel/electricity,maintenance, repair and taxes). All the calculations were based on a driving distance of 20,000 km/year and the average fuel and electricity prices for 2011 in EU (1:42=l fordiesel, 1:51=l for gasoline and 0:17=kW h for electricity).Costs ICEV PHEV BEVGolf 1.6 TDI Golf 1.4 TSI Smart CDI Smart Volt ED Volt GD Volt MD Nissan Leaf Smart ED Peugeot iOnInvestment 22,200 20,300 12,600 9800 42,000 35,000 19,000 30,000Insurance =year 350 350 350 350 350 350 350 350Battery leasing =year 720 Fuel =year 1192.8 1872.4 937.2 1268.4 1932.8 386 Electricity =year 552.5 442 425 340 368Maintenance =year 420 380 330 300 335 303 240 280Operational cost =year 1962.8 2602.4 1617.2 1918.4 1237.5 2617.8 1513.6 1078 1650 998.3Note: For the PHEV were considered three scenarios: electric drive (ED) powered only by the battery; gasoline drive (GD) powered only by the ICE and mixed drive (MD)powered 80% of the time by battery and the remaining 20% by the ICE.R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 284fast accelerations andhighspeed, is directlytranslatedintoareducedrange per charge and therefore translated into moreemissions due to a higher energy consumption. By increasing theconsumer awareness to a more efcient driving style it is possibletoachieve, onlybyitself, signicant impact inthereductionofGHG emissions.During the production phase the impacts can bereduced by using recycled materials and replacing certain alloys bycomposite materials.0100002000030000400005000060000700008000090000Golf 1.6 TDI Golf 1.4 TSI Smart CDI Smart Leaf Smart ED iOn Volt ED Volt GD Volt MDLife Cycle Costs []050010001500200025003000Golf 1.6 TDI Golf 1.4 TSI Smart CDI Smart Leaf Smart ED iOn Volt ED Volt GD Volt MDAnnual Cost of Operation []Fuel/Electricity Maintenace and Insurance Battery LeasingInvestment Interest OperationFig. 15. Total life-cycle costs (top) for the considered vehicles and annual operation cost (bottom). It should be noted that the annual ownership cost does not take intoaccount the ination rate, however it is considered in the life-cycle costs.10000150002000025000300003500040000450005000055000600006500070000750001 2 3 4 5 6 7 8 9 10Cost []YearGolf 1.6 TDI Golf 1.4 TSI Volt GD Smart Smart CDILeaf Volt ED Volt MD Smart ED iOnFig. 16. Cumulative costs of ownership for the considered vehicles during their life-cycle. Only the costs directly associated with the use of the vehicle were considered,depreciation and interests were not taken into account.R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 285AcknowledgmentsThis work was supported by Foundation for Science and Technol-ogy(FCT)undertheprojectsMIT/MCA/0066/2009(EconomicandEnvironmentalSustainabilityofElectricVehicleSystems), MIT/SET/0014/2009(Capturing Uncertainty inBiofuels for Transportation.Resolving Environmental Performance and Enabling Improved Use)and PTDC/SEN-TRA/117251/2010 (Extended well-to-wheels assess-ment of biodiesel for heavytransport vehicles). This researchisframed under the Energy for Sustainability Initiative of the Universityof Coimbra and was supported by the FCT, through the MITPORTUGAL program.References[1] SmithWJ. Canev(electricvehicles) addressIreland's CO2emissions fromtransport? Energy 2010;35(12):451421.[2] Dargay J, Gately D, Sommer M. Vehicle ownership and income growth,worldwide: 19602030. The Energy Journal 2007;0(4):14370.[3] Meyer I, Leimbach M, Jaeger C. International passenger transport and climatechange: a sector analysis in car demand and associated emissions from 2000to 2050. Energy Policy 2007;35(12):633245.[4] HaoH, WangH, Yi R. Hybridmodelingof China's vehicleownershipandprojection through 2050. Energy 2011;36(2):135161.[5] Energy Information Administration, International energy outlook 2011, Report(September 2011).[6] Forster P, Ramaswamy V, Artaxo P, Berntsen T, Betts R, Fahey D, et al. Changesinatmosphericconstituentsandinradiativeforcing. In:SolomonS, QinD,ManningM, ChenZ, Marquis M, Averyt KB, Tignor M, Miller HL, editors,Climatechange2007: thephysical sciencebasis. contributionof workinggroupI tothefourthassessmentreportof theintergovernmental panel onclimate change. Cambridge University Press; 2007.[7] FayazH, SaidurR, Razali N, AnuarF, SalemanA, IslamM. Anoverviewofhydrogen as a vehicle fuel. Renewable and Sustainable Energy Reviews2012;16(8):551128.[8] Bradley TH, Frank AA. Design demonstrations and sustainability impactassessmentsforplug-inhybridelectricvehicles. RenewableandSustainableEnergy Reviews 2009;13(1):11528.[9] Streimikiene D, Baleentis T, Baleentiene L. Comparative assessment of roadtransporttechnologies. RenewableandSustainableEnergyReviews2013;20(0):6118.[10] LauJ, HungW, CheungC. On-boardgaseous emissions of LPGtaxis andestimation of taxieet emissions. Science of the Total Environment 2011;409(24):5292300.[11] ReichmuthDS, LutzAE, ManleyDK, KellerJO. Comparisonof thetechnicalpotentialfor hydrogen,battery electric, and conventional light-duty vehiclesto reduce greenhouse gas emissions and petroleum consumption in the UnitedStates. International Journal of Hydrogen Energy 2013;38(2):12008.[12] AdamT, AstorgaC, ClairotteM, DuaneM, ElsasserM, KrasenbrinkA, etal.Chemical analysisandozoneformationpotential ofexhaustfromdual-fuel(liqueed petroleum gas/gasoline)light duty vehicles. Atmospheric Environ-ment 2011;45(17):28428.[13] AslamM, Masjuki H, KalamM, AbdesselamH, MahliaT, AmalinaM. Anexperimental investigationof CNGas analternativefuel for a retrottedgasoline vehicle. Fuel 2006;85(56):71724.[14] Jayaratne E, Ristovski Z, Meyer N, Morawska L. Particle and gaseous emissionsfrom compressed natural gas and ultralow sulphur diesel-fuelled buses at foursteady engine loads. Science of the Total Environment 2009;407(8):284552.[15] Myung C-L, Choi K, Kim J, Lim Y, Lee J, Park S. Comparative study of regulatedandunregulatedtoxicemissionscharacteristicsfromasparkignitiondirectinjection light-duty vehicle fueled with gasoline and liquid phase LPG(liqueed petroleum gas). Energy 2012;44(1):18996.[16] Panwar N, Kaushik S, Kothari S. Role of renewable energy sources inenvironmental protection: a review. Renewable and Sustainable EnergyReviews 2011;15(3):151324.[17] Palencia JCG, Furubayashi T, Nakata T. Energy use and CO2 emissions reductionpotential in passenger careet using zero emission vehicles and lightweightmaterials. Energy 2012;48(1):54865.[18] HwangJ-J. Sustainabilitystudyof hydrogenpathways for fuel cell vehicleapplications. Renewable and Sustainable Energy Reviews 2013;19(0):2209.[19] PasterM, AhluwaliaR, BerryG, ElgowainyA, LasherS, McKenneyK, etal.Hydrogenstoragetechnologyoptions for fuel cell vehicles: well-to-wheelcosts, energy efciencies, and greenhouse gas emissions. International Journalof Hydrogen Energy 2011;36(22):1453451.[20] Larminie J, Lowry J. Electric vehicle technology explained. England: John Wiley& Sons Ltd.; 2003.[21] Wang G. Advanced vehicles: costs, energy use, and macroeconomic impacts.Journal of Power Sources 2011;196(1):53040.[22] Kempton W, Tomic J. Vehicle-to-grid power implementation: from stabilizingthe grid to supporting large-scale renewable energy. Journal of Power Sources2005;144(1):28094.[23] II RCG, WangL, AlamM. Theimpactof plug-inhybridelectricvehiclesondistributionnetworks: a reviewandoutlook. Renewable andSustainableEnergy Reviews 2011;15(1):54453.[24] Freire F, Marques P. Electric vehicles in Portugal: an integratedenergy, greenhouse gas and cost life-cycle analysis. In: 2012 IEEEinternational symposiumon sustainable systems and technology (ISSST).2012, p. 16.[25] Chan C. The state of the art of electric, hybrid, and fuel cell vehicles.Proceedings of the IEEE 2007;95(4):70418.[26] PolletBG, Staffell I, ShangJL. Currentstatusofhybrid, batteryandfuel cellelectricvehicles:fromelectrochemistrytomarketprospects. ElectrochimicaActa 2012;0.[27] Hawkins TR, Singh B, Majeau-Bettez G, Strmman AH. Comparative environ-mental life cycle assessment of conventional and electric vehicles. Journal ofIndustrial Ecology 17(1).[28] Notter DA, Gauch M, Widmer R, Wager P, Stamp A, Zah R, et al. Contributionofli-ionbatteriesto theenvironmentalimpact ofelectricvehicles.Environ-mental Science and Technology 2009;44(17):65506.[29] ZackrissonM, AvellnL, Orlenius J. Life cycle assessment of lithium-ionbatteries for plug-in hybrid electric vehicles e critical issues. Journal of CleanerProduction 2010;18:151929.[30] Majeau-Bettez G, Hawkins TR, Strmman AH. Life cycle environmentalassessment of lithium-ionandnickel metal hydride batteries for plug-inhybridandbattery electricvehicles. EnvironmentalScienceandTechnology2011;45(10):454854.[31] SpielmannM, DonesR, BauerC. Lifecycleinventoriesoftransportservices.Final report ecoinvent vol 2.0, No. 14.[32] SullivanJ, Gaines L. Status of life cycle inventories for batteries. EnergyConversion and Management 2012;58(0):13448.[33] Swiss Centre for Life Cycle Inventories. Ecoinvent Centre (July 2011). http://www.ecoinvent.org/database/.[34] European Commission. Car prices within the European union, Report (January2011).[35] aatay K, Bayindir MAGA. Comprehensive overviewof hybrid electricvehicle:powertraincongurations, powertraincontroltechniquesandelec-tronic control units. Energy Conversion and Management.[36] Gabriel M. Innovations in automotive transmission engineering. SAE Interna-tional; 2004.[37] Campanari S, Manzolini G, de la Iglesia FG. Energy analysis of electric vehiclesusing batteries or fuel cells through well-to-wheel driving cycle simulations.Journal of Power Sources 2009;186(2):46477.[38] Trummel M, BurkeA. Developmenthistoryof thehybridtestvehicle. IEEETransactions on Vehicular Technology 1983;32(1):714.[39] ArgonneNational Laboratory, Comparisonof powertraincongurationforplug-inHEVsfromafuel economyperspectiveefciencyandcost. Report,2008.[40] Eaves S,Eaves J. A cost comparisonof fuel-celland battery electric vehicles.Journal of Power Sources 2004;130(12):20812.[41] Cericola D, Ruch P, Ktz R, Novk P, Wokaun A. Simulation of a supercapacitor/li-ion battery hybrid for pulsed applications. Journal of Power Sources2010;195(9):27316.[42] CooperA, FurakawaJ, LamL, KellawayM. Theultrabatteryanewbatterydesign for a new beginning in hybrid electric vehicle energy storage. Journal ofPower Sources 2009;188(2):6429.[43] Doucette RT, McCulloch MD. A comparison of high-speedywheels, batteries,andultracapacitorsonthebases of cost andfuel economyas theenergystoragesysteminafuelcellbasedhybridelectricvehicle. JournalofPowerSources 2011;196(3):116370.[44] IbrahimaH, IlincaA, PerronbJ. Energystoragesystemscharacteristicsandcomparisons. Renewable and Sustainable Energy Reviews 2008;12:122150.[45] Divya K, stergaard J. Battery energy storage technology for power systemsan overview. Electric Power Systems Research 2009;79:51120.[46] VetterJ, NovkP, WagnerM, VeitC, MollerK-C, BesenhardJ, etal. Ageingmechanismsinlithium-ionbatteries. JournalofPowerSources2005;147(12):26981.[47] SmithK, WangC-Y. Power andthermal characterizationof alithium-ionbattery pack for hybrid-electric vehicles. Journal of Power Sources2006;160:66273.[48] Zhanga Y, Wanga C-Y, Tangb X. Cycling degradation of an automotive lifepo4lithium-ion battery. Journal of Power Sources 2011;196:151320.[49] Sarre G,Blanchard P, Broussely M.Aging oflithium-ionbatteries.Journal ofPower Sources 2004;127(12):6571.[50] Nuhic A, Terzimehic T, Soczka-GuthT, Buchholz M, Dietmayer K. Healthdiagnosis and remaining useful life prognostics of lithium-ion batteries usingdata-driven methods. Journal of Power Sources 2012;0.[51] Ecker M, Gerschler JB, Vogel J, Kbitz S, Hust F, Dechent P, et al. Development ofa lifetime predictionmodel for lithium-ion batteries based on extendedaccelerated aging test data. Journal of Power Sources 2012;215(0):24857.[52] de Almeida AT, Ferreira FJTE, Fong J, Fonseca P. Eup lot 11 motorsnal report.Report (February 2008).[53] GmbH RB. Bosch automotive handbook. Wiley-Blackwell; 2007.[54] Leuenberger M, Frischknecht R, E. services Ltd., Life cycle assessment ofbattery electric vehicles and concept cars. Report (June 2010).[55] FariaR, MouraP, DelgadoJ, deAlmeidaAT. Asustainabilityassessmentofelectric vehicles as a personal mobility system. Energy Conversion andManagement 2012;61(0):1930.R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 286[56] UnitedNationsEnvironmentProgramme, Internationaltestcyclesforemis-sions and fuel economy; 2011. http://www.unep.org/transport.[57] EuropeanUnion, CommissionplanslegislativeframeworktoensuretheEUmeets its target for cutting CO2emissions from cars. Press Release (February2007).[58] Eurostat, Eurostat energy andenvironment indicators (2011). http://epp.eurostat.ec.europa.eu.[59] International Council on Clean Transportation, Carbon intensity of crude oil inEurope. Report (November 2010).[60] InternationalEnergyAgency, Worldenergyoutlook2009factsheet. Report(November 2009).[61] Samaras C, Meisterling K. Life cycle assessment of greenhouse gas emissionsfrom plug-in hybrid vehicles: implications for policy. Environmental Scienceand Technology 2008;42(9):31706.[62] van Vliet O, Brouwer AS, Kuramochi T, van den Broek M, Faaij A. Energy use,costandCO2emissionsofelectriccars. Journal ofPowerSources2011;196(4):2298310.[63] Moura PS, de Almeida AT. Multi-objective optimization of a mixed renewablesystemwithdemand-sidemanagement. RenewableandSustainableEnergyReviews 2010;14(5):14618.[64] Sovacool BK. Valuingthegreenhousegasemissionsfromnuclearpower:acritical survey. Energy Policy 2008;36(8):295063.[65] Beurskens L, Hekkenberg M. Renewable energy projections as published in thenational renewable energy actionplans of the Europeanmember states.Report, Energy Research Centre of the Netherlands (2011).[66] Europe Energy Portal, Average prices for fuels andelectricity inthe EUmember states (June 2012). http://www.energy.eu.R. Faria et al. / Renewable and Sustainable Energy Reviews 24 (2013) 271287 287