JackEllisMOI

1

Author:JackEllisSupervisor:AurèliaEstradaMañé

MásterOficialenInternacionalización:Aspectosempresariales,económicosypolitico-jurídicos

JackEllisMOI

2

AFutureofDecentralisedEnergy?

AnInvestigationintotheImpactofBatteryStorageTechnology

JackEllisMOI

3

TABLEOFCONTENTSListofFigures....................................................................................................................................................4

ListofTables.....................................................................................................................................................4

Introduction......................................................................................................................................................5

Objective..........................................................................................................................................................11

Methodology...................................................................................................................................................13

TrendsinEnergyFinance..........................................................................................................................14

Chapter1:TechnicalAspectsofBatteryStorageTechnologies....................................................22BatteryStorageTechnologies.............................................................................................................22Lead-acid....................................................................................................................................................................22Lithium-ion...............................................................................................................................................................24FlowBatteries..........................................................................................................................................................25Sodium/MoltenSalt..............................................................................................................................................25Summarytableofkeyfactors...........................................................................................................................26

Applicationsofbatterystoragetechnology...................................................................................26BatteryStorage–Islandsandoff-gridapplications................................................................................27HouseholdsolarPV...............................................................................................................................................28Variablerenewableenergysmoothingandsupplyshift......................................................................30Fastregulationingridswithhighvariablerenewableenergyshares............................................32

Summary....................................................................................................................................................33

Chapter2:Theeconomicviabilityofbatterystoragetechnology...............................................35ResidentialSolarPV...............................................................................................................................35OptimalSolarPVSystemSize...........................................................................................................................37OptimalStorageSize.............................................................................................................................................39StorageProfitability..............................................................................................................................................40Summary....................................................................................................................................................................42

Pricedevelopmentoflithium-iontechnology...............................................................................43Marketdevelopmentforbatterystoragetechnology.................................................................46Utility,residentialandnon-residentialmarketsegments...................................................................47Internationalmarkets..........................................................................................................................................52U.S.A.......................................................................................................................................................................53Japan......................................................................................................................................................................54Germany...............................................................................................................................................................56China......................................................................................................................................................................57

Summary....................................................................................................................................................................59

Chapter3:Theimpactofbatterystoragetechnologyinperspectives.......................................61Environmentalperspective.................................................................................................................61Production.................................................................................................................................................................61Use................................................................................................................................................................................63Disposal......................................................................................................................................................................64

Internationalizationperspective.......................................................................................................67Public/Utilityperspective....................................................................................................................69Implicationsforenergy-systemmarketparticipantsandotherstakeholders...........................72Futuretrajectories.................................................................................................................................................73Democratisationofenergy.................................................................................................................................77

Conclusion.......................................................................................................................................................80

Bibliography...................................................................................................................................................83

JackEllisMOI

4

ListofFiguresFigure1:CO2Concentrationinatmosphere.............................................................................................................6Figure2:TheDuckcurve..................................................................................................................................................8Figure3:Evolutionofpolysiliconprices.................................................................................................................15Figure4:Siliconphotovoltaicvaluechain..............................................................................................................16Figure5:Globalaveragemoduleprice.....................................................................................................................17Figure6:EvolutionPVprice($/Watt).....................................................................................................................18Figure7:GlobalAnnualPVProduction....................................................................................................................19Figure8:ElectricityOutput(GWh)fromSolarPVInstallationsinMajorEconomies.........................20Figure9:ElectricityOutput(GWh)fromSolarPVinstallationsinOECDcountries.............................20Figure10:Servicesprovidedbyenergystorage..................................................................................................26Figure11(left):noenergystorage............................................................................................................................27Figure12(right):withlead-acidbatterystorage................................................................................................27Figure13:SolarPVandbatterystorage..................................................................................................................29Figure14:Smoothingfrombatterypowerstorage............................................................................................31Figure15:Energysupplyshift.....................................................................................................................................32Figure16:100MWBatterystorage(left)versus100MWgasturbine(right)........................................33Figure17:Averageendconsumerpriceforinstalledrooftopsystems.....................................................35Figure18:OptimalSolarPVplantsizeS1-S5........................................................................................................37Figure19:OptimalPVplantsizeS6-S8....................................................................................................................39Figure20:OptimalStorageSizeS1-S5.....................................................................................................................39Figure21:OptimalStorageSizeS1-S5.....................................................................................................................40Figure22:StorageprofitabilityS1-S5......................................................................................................................41Figure23:StorageprofitabilityS6-S8......................................................................................................................42Figure24:CostofLithium-ionbatterypacksinbatteryelectricvehicles................................................44Figure25:Estimatedinstalledbatterycapacityandcommissionsinpowersector,2014...............47Figure26:Worldwideforecastofbatterystoragecapacity(MW)andannualrevenue(USD)for

utilityscaleapplications.....................................................................................................................................48Figure27:AnnualUSsolarPVinstallationsfrom2000-2015.......................................................................50Figure28:USPVinstallationforecast2010-2021&USinstallationforecastbysegment................51Figure29:Possiblepathsforelectricitygridevolution....................................................................................75Figure30:OwnershipofGermanrenewablesin2012.....................................................................................78Figure31:NumberofenergycooperativesinGermany2001-2013..........................................................79

ListofTablesTable1:EvolutionofChinesePVcellproduction................................................................................................14Table2:Electricitypricescenariosusedinmodelsimulations...................................................................37Table3:Applicationsforbatteryuse........................................................................................................................48Table4:Installationtargetsofmajormarketsandoverallshareofrenewablegeneration.............52Table5:Characterisedimpactperkgofbatteryproduction..........................................................................62Table6:CharacteriseddatarangeforbatteryproductionperMJcapacity.............................................63Table7:Typologyofenvironmentrelatedrisks..................................................................................................74

JackEllisMOI

5

IntroductionInthesecondhalfofthe20thcenturyscientistsbecameincreasinglyawareoftheeffectsthat gases such as carbon dioxide and methane were having on the planet’senvironment.Initially,concernsaboutthewarmingeffectsofgasseswerecounteredbysuggestionsthathumanactivitycouldhavecoolingeffectsontheenvironmentthroughthe use of aerosols. However, by the 1970’s scientific consensus had veered towardswarming rather than cooling. In 1975, Wallace Broecker published a paper titled,Climate Change: Are we on the Brink of a Pronounced GlobalWarming?This paper isoftencreditedwithcoiningtheterm‘globalwarming.’Nowadays,one ismore likely tohearof ‘climatechange’ rather than ‘globalwarming’.Whereas ‘global warming’ refers simply to the earth’s rising temperature, ‘climatechange’ isamoreall-encompassingconceptthatreferstotheearth’swarmingand theside effects of that warming - melting glaciers, rising sea levels and increasingdesertificationforexample.The first international treaty to combat climate change was the United NationsFrameworkConventiononClimateChange(UNFCCC).ThiswasnegotiatedattheEarthSummit in Rio de Janeiro in 1992 and came into force on the 21st ofMarch 1994. Itsobjectivewastostabilizethegreenhousegasconcentrationsintheatmosphereatalevelthatwould, ‘prevent dangerous anthropogenic interferencewith the climate system.’1This treatywas extended upon in the Kyoto Protocol by committing states to reducegreenhouse gas emissions on the premise that 1) globalwarming exists and 2) it hasbeencausedbymanmadecarbondioxideemissions.

1UNITEDNATIONS. (1992)UnitedNations Framework Convention on Climate Change [Online] Availablefrom:https://unfccc.int/resource/docs/convkp/conveng.pdf

JackEllisMOI

6

Figure1:CO2Concentrationinatmosphere2

Figure1clearlyshowstheeffectthatmanhashadonthelevelofcarbondioxideintheatmosphere.Forthepast650,000yearsCO2hadneverbeenhigherthanapproximately300partspermillion.AtthetimeofwritingthecurrentlevelofCO2intheatmospherestandsatapproximately403partspermillion.3The latest agreement within the framework of the UNFCCC is the Paris Agreement,which was drafted in late 2015 at the 21st Conference of Parties in Paris. The ParisAgreementisyettoenterintoforce,butaswithpreviousUNFCCCagreementsthereareconcerns regarding the lack of binding targets as well as an effective enforcementmechanism. The impact that this agreement will have on future carbon emissionsremainstobeseen.SuchistheconcernoverclimatechangethattheGlobalRisksReport2016publishedbytheWorldEconomicForumrates, ‘failureofclimatechangemitigationandadaptation’as the greatest risk in terms of impact.4Climate change is linked to risks of food andwatercrises,profoundsocial instability,extremeweatherevents,biodiversity lossandecosystem collapse, and large-scale involuntary migration. The threats posed by thefailuretomitigatetheeffectsofclimatechangearesevere.Recognizing the need to provide energy, without adding to the concentrations ofgreenhousegasseswithintheearth’satmosphere,therehasbeenasurgeinrenewableenergy technology during the 21st century. However, the most accessible forms ofrenewableenergy–solarandwindhavemajorproblemsregardingtheirreliabilityandstability.Inotherwords,thefailureofthesuntoshineandthewindtoblow24hoursaday means that the proliferation of clean energy sources has been hindered. Theintermittent nature of renewable energy is a problem that must be solved if theirdevelopment is to continue. The followingquote by thedistinguishedCzech-CanadianscientistVaclavSmil:‘If electric utilities had an inexpensive way to store massive amounts of excess powergenerated bywind and solarwhen demand is low,which could later be tapped tomeetpeakdemand,thenthenewrenewableswouldexpandmuchmorequickly.Unfortunately,decadesofdevelopmenthaveprovidedonlyonegood,large-scalesolution:pumpingwaterup toanelevatedreservoir so it can flowback througha turbine togenerateelectricity.Notmanylocalitieshavetheelevationchangeorspacetomakethiswork,andtheprocess

2GLOBALCLIMATE.(2015).GlobalClimateChange:Evidenceandcauses.[Online]Availablefrom:http://globalclimate.ucr.edu/resources.html3NASA.(2015).GlobalClimateChange:Vitalsignsoftheplanet.[Online]Availablefrom:http://climate.nasa.gov4LEVITT,T.(2016)ClimateChangeFailstotopListofThreatsforBusinessLeadersatDavos.TheGuardian.[Online]Availablefrom:http://www.theguardian.com/sustainablebusiness/2016/jan/20/climate-change-threats-business-leaders-davos-surveyIt is interesting tonote that climatechangewasnot considered tobeahighconcern to1,400CEOs fromaroundtheworld.AlistprovidedbyPricewaterhouseCoopers(PwC)showedthatoverregulationwastheirgreatest concern (79%), followedby geopolitical uncertainty (74%). Climate changewasmentioned as athreat tobusinessbyonly50%ofCEO’s.Both thePwCsurveyanda separate surveyof13,000businessleaderscarriedoutbytheWorldEconomicForum(WEF)showedarelativeabsenceofconcernoverclimatechangeandenvironmentalrisk.

JackEllisMOI

7

entailsnetenergyloss.’5

The ability to store electrical energy in a safe and economically viablemannerwouldserve tocombat the limitations thatarisedue to the intermittentnatureof renewableenergysources.Thus,electricalstoragetechnologycouldlikelybeasignificantfactorinthe development of renewable energy systems and consequently a factor in climatechangemitigation.

Smart grids, too, have the potential to combat the limitations of renewables. Throughthe use of integrated communications as well as sensing and measuring technology,system reliability can be optimized. Distributed power flow control is a noteworthyexample.Inthiscase,smartwirescontroltheflowofpowerwithinexistingtransmissionlinesandasaresultmorerenewableenergyissupportedonthegrid.Arpa-E,abranchoftheDepartmentofEnergyintheUnitedStatessaysthefollowingonthesubject:

‘Smartwirescouldsupportgreateruseofrenewableenergybyprovidingmoreconsistentcontrol over how that energy is routedwithin the grid on a real-time basis. Thiswouldlessentheconcernssurroundingthegrid'sinabilitytoeffectivelystoreintermittentenergyfromrenewablesforlateruse.’6

Althoughof critical importance in the futureof energy systems, this investigationwillfocusonbatterystoragetechnologyratherthansmartgrids,withregardstotheirabilitytoutilizeintermittentenergyfromrenewables.

Atpresent,oneofthegreatestchallengesfacingthedevelopmentofsolarpoweristheeffectthat ithasonthedailydemandforutilityelectric.Figure2belowshowsthenetsupply and demand of power on California’s electric grid during a 24-hour period in2012-13.

5SMIL,V.(2014)TheLongSlowRiseofSolarandWind.ScientificAmerican.6ARPA-E.(2011)DistributedPowerFlowControl.[Online]Availablefrom:http://arpa-e.energy.gov/?q=slick-sheet-project/distributed-power-flow-control

JackEllisMOI

8

Figure2:TheDuckcurve7

Theabovefigureiscommonlyreferredtoas‘theduckcurve’onaccountofthefactthatyearonyearthegraphrepresentstheshapeofaduck’sheadandbody.Astheamountofinstalled solar has increased, the demand on the grid has fallen. Before 2012 energydemandwassaidtoresembleacamelwithtwohumps.Asonecanseefromthegraphtherewerepeaksinboththemorningandintheearlyevening.Thisenergywasmostlysuppliedbyutilityoperatedpowerplants.However, increasingly thisenergyhasbeensubstituted by local solar power, which by meeting local demand greatly affects thedemandforenergyfromthegridduringtheday.Theorangecamelistransformingintoagreenduck.

Thegraphhasbecomealmostasymbolicpictureofutilitycomplaints.Utilitycompaniesregard the growth of distributed solar as a major technical problem, rather than aneconomicone.Ofparticularconcernforutilitycompaniesisonepartofthegraph–theramp up period in the late afternoon. This is the time period in which the energyproduced from solar is decreasingwhile energy demand increases. According to JohnFarrell of Clean Technica, in traditional grid operating models, accommodating thisramp-up inenergyrequires, ‘lotsof standbypower fromexpensive tooperate, rapid-responsepowerplants.’8

Asa result, therehavebeenvarioussuggestionsattempting to ‘flatten theduck’.Theyarelistedbelow:

▪ Targetenergyefficiencymeasuresforthe“rampup”period▪ Orientsolarpanelstothewesttocatchmorelateeveningsun▪ SubstitutesomesolarthermalwithstorageforsolarPV▪ Allowthegridoperatormoredemandmanagementviaelectricwaterheating

[alreadydoneextensivelybyruralcooperativesinMinnesota]▪ Requirelargenewairconditionerstohavetwohoursofthermalstorage

accessibletotheutility▪ Retireinflexiblegeneratingplants(coalandnuclear)thatneedtorunconstantly

inoff-peakperiods▪ Concentrateutilitydemandchargesontherampupperiod▪ Deployelectricitystorageintotargetedareas,includingelectricvehicle-to-grid▪ Implementaggressivedemandresponseprograms(subscribingmore

businessesandhomesintoprogramstoshedtheirenergydemandatkeyperiods)

▪ Useinter-regionalpowertransactions▪ Selectivelycurtailasmallportionofsolarpowergeneration

Inlightofthisitcanbesuggestedthatthetechnicalchallengesofthe‘duckproblem’aremanageable, in the most part with existing technology. Battery storage linked to PVsystemswouldalsobeabletocontributetoasolution.Excesssupplystoredintheday

7FARRELL,J.CleanTechnica[Online]Availablefrom:http://cleantechnica.com/2014/07/21/utilities-cry-fowl-over-duck-chart-and-distributed-solar-powercrying-fowl-or-crying-wolf-open-season-on-the-utilitys-solar-duck-chart/8ibid.

JackEllisMOI

9

wouldbeabletoplayapart intheramp-upperiodasenergydemandincreases intheeveningaspeoplereturnhomefromwork.

Itislikelythateconomicproblemsexistforutilities,andthequestionoftowhatextenttheirbusinessmodelmaybeoutdatedisaninterestingone.Asmoresolarcomesontothegrid,utilityownedgasplantswhichmeetpeakloadwillbeoutbid.Furthermore,inthe near future solar production will be sufficient to cut into the “baseload” power,usuallyprovidedbycoalpowerplants,whichareonlyeconomicallyviableoperating24hoursaday.

Thustheelectricitygridisundoubtedlychanging.The20thcenturysystemofinflexibleandcentralisedpowerplantswith longdistance transmission lines is transforming, insomecases,todistributedrenewableenergy.Theduckgraphhighlightsthelimitationsofusinga20thcenturygridmodel fora21st centurysystem.Toutilitycompanies theduck graph serves as evidence of the technical problems caused by the continueddevelopmentor renewableenergy sources.However, aswehave seen it is theenergydemand that is displaced by solar that represents the real problem for utilities. If anincreasingnumberof customers seekmore control over their energy consumptionbymoving to solar, themarket shareof utilities is likely todecrease.Analysis fromPWCnotesthatthegainsinrenewable/distributedpowersystemshave,‘alteredthebusinessequation sufficiently that the customer is rapidly becoming the dominant force.’9It isclear that the industry is undergoing a huge change as firms have traditionally led itwithvirtualmonopoliesovercustomers.Hencetheproblemfacingutilitycompaniesismore economic in nature than technical. Furthermore, an abundance of renewablepowerchangestheprofitabilityofbaseloadandpeakingpowerplants.AsFarrellwrites:

“Economically,anabundanceoflow-costrenewableenergywillchangetheprofitabilityofbaseload and peaking power plants. Baseload power plants will suffer from a drop inwholesale electricity prices, as has happened in Germany. Fast-response power plantoperators will also struggle, because while peak energy prices may remain high, moresolar energy on the grid will shorten periods of peak energy demand for these powerplants.”10

Althoughsomewhatinflexible,likecoalandnuclear,renewableenergyhasnofuelcostsand little operation costs. Therefore, in theory it should be the first power a utilitycompanywouldwanttouseonthegrid.Furthermore,thoughithasbeensuggestedthat‘baseloadisnotcompatiblewitharenewableenergyfuture,’11itisclearthatquickandflexible response to electricity supply will become increasingly important in arenewable energy future. As a result, energy storage technologies may provide animportantsolution. Itshould furtherbenotedthateven if therewere improvement tothe grid with the use of smart grid technology, energy storage would still reduce itscapacityrequirements.

9PWC.(2015).2015utilitiestrends.[Online]Accessedon4/4/2015from:http://www.strategyand.pwc.com/perspectives/2015-utilities-trends10FARRELL,J.(2014)EnergyStorage:TheNextChargeforDistributedEnergy.TheInstituteforlocalSelf-Reliancep.6.11HOPE,M.(2013).TheEnergiewendeandenergyprices:PublicsupportandGermany’slongtermvision.TheCarbonBrief.[Online]Accessedon3/3/2015from:http://tinyurl.commk6qnp2

JackEllisMOI

10

Energy storage and renewables are a powerful combination thatwill ultimately allowforamorethoroughadoptionofrenewableenergy,somethingthatiscritical ineffortstomitigateclimatechangeandacceleratetheenergytransitiontosustainablesources.Itislikely,althoughnotcertain,thatthecombinationalsohastheaddedbonusofgreaterlocal control of the energy system. Thus, the 21st century dynamic offers us thepossibility of a dramatically different energy future to that of the previous century:Millions ofwidely dispersed renewable energy plants and storage systems tied into asmartgrid.Theeraoflargeanddistantcentralisedpowerplantscouldbeconsignedtothe past. The 21st century grid could be a democratized network of independentlyowned andwidely dispersed renewable energy generation,with economic benefits ofelectricitygenerationdispersedaswidelyastheownership.

Dr.NorbertRottgen,theGermanFederalMinisterfortheenvironment,believesthatthetimeisapproachingwhencountrieswillhavetomakeadecisiononthefutureoftheirenergysystem:

‘It is economically nonsensical to pursue two strategies at the same time, for both acentralised and decentralized energy supply system, since both strategies would involveenormous investment requirements. I am convinced that the investment in renewableenergies is the more economically promising project. But we will have to make up ourminds.Wecantgodownbothpathsatthesametime.’12Theroleofelectricalenergystoragesystems,theirimpact,andtheextenttowhichtheyareabletoacceleratethetransitiontodecentralizedenergysystems,isthefocusofthispaper.

12FARRELL,J.(2011).DemocratizingtheElectricitySystem.TheNewRulesProject.p.27.

JackEllisMOI

11

ObjectiveRenewabletechnologieshavegainedeconomiccompetitivenessintherecentpast.Thisis mostly true in the case of solar energy as a result of falling PV cell costs. Despiterenewable energy’s increasing competitiveness it still possesses problems that hinderwidespread deployment. One major obstacle is the intermittent nature of renewableenergy.Thispaperseekstounderstandtherolethatbatterystoragetechnologyisabletoplayinthedisseminationof renewableenergysourcesanda21st centurygridsystem.This isprincipally achieved by counteracting the key disadvantage of intermittency. In orderfor this tobe carriedout, both the technical andeconomic aspectsof electrical energystorage are to be investigated. In addition, this paper aims to showwhat impact theconclusions drawn from the investigation will have. The paper shows the impact ofbattery storage technology through several different perspectives and ultimately askswhattheimpactofbatterystoragetechnologywillbeoncentralisedenergysystems.This project is carried out from the perspective of Internationalisation. Tointernationalise is commonlydefinedas toputunder internationalcontrolor tomakeinternational in character. In economic terms Internationalisation is the process ofincreasing involvement of enterprises in international markets, and the spread ofeconomicactivitiesacrossnationalboundaries.13Internationalizationisevidentthroughtheincreasinglevelsofinvestmentinrenewableenergy.Inordertounderstandtheroleofbatterystoragetechnology,itisnecessarytounderstand the current trends in renewable energy finance. Renewable energy hasundergone sizable changes in the recent past and before investigating the technicalaspects of energy storage technologies, it is important to understand what keydevelopments there have been in the recent past. This is done in an effort tocontextualizetheissueofbatterystorage,andtounderstandhowithasbecomesuchacriticalissueforthecontinueddevelopmentofrenewableenergysources.Consequently,aminor objective of the paper is to understand recent trends in the development ofrenewableenergysources.Once the recent trends have been understood chapter onewill focus on the technicalaspects of battery storage technology. The various battery storage technologies thatcurrently exist will be compared and the various different applications for thesetechnologieswillbeinvestigated.Chapter two investigates battery storage technology specifically and will have aneconomicrather than technical focus.The current status and the economic viability ofthetechnologyareunderstood,anditsfuturehypothesized.Thispartofthepaperalsoseekstounderstandtheobstaclesthatbatterystoragetechnologyfaces.

13GEREFFIetal.(2001).Introduction,Globalisation,ValueChainsandDevelopment.InstituteofDevelopmentStudies.[Online]Accessedon3/3/2015from:https://www.ids.ac.uk/files/dmfile/gereffietal323.pdf

JackEllisMOI

12

Chapter three aims to assess the impact of battery storage technology. This is to beachieved by evaluating its impact from several different perspectives:Internationalization – howwill the conclusions drawn from parts one and two affectstates and the large energy corporations?Environmental –what is the environmentalcost of energy storage solutions? Public – To what extent is the public accepting ofenergystorageandtheconsequencesofitsintegrationintheenergysystem?Finally, theconclusionwill sumup the findingsof thepaper,offeravenues for furtherresearchandsuggestlimitationsofthepaper.

JackEllisMOI

13

MethodologyThemethodology of the paper will begin with the collection of relevant information.Fundamentally, this investigationisoneintothefutureofenergysystemsandtherolethat energy storage technology will play in their future. Thus although the paper isprincipally concerned with battery storage technology, it is also important tounderstand the problems that renewable energy faces. Firstly, an explorativeinvestigationintoproblemsfacedbyrenewableenergyisimportant.Theseissueshavebeen discussed in the introduction and serve as the foundation for investigation intostoragetechnology.‘Trendsinenergyfinance’willshowrecentdevelopmentsinthepriceofsolarPVpanelsandtheimpactthatithashadontheadoptionofrenewabletechnology.Documents, peer-reviewed papers and data that relates to storage technology energywill be collected and investigated in order to have a complete understanding of thetechnologyanditsapplications.ThisservesasChapteroneandtechnicalinitsapproach.Toeffectivelycarryoutparttwooftheinvestigation,whichfocusesontheeconomicsofbattery storage technology several steps are necessary. Firstly, the economics areinvestigatedat amicro level, in an effort tounderstandwhetherornot investment inresidential battery systems is economically viable. Other parts of the chapter seek tounderstandthepricedevelopmentof lithium-iontechnologyandits impactonvariousassociatedmarkets.Thefinalpartoftheinvestigationwilluseawiderangeofdocumentsinordertoassessthe impact of battery storage from different perspectives. Documents from theEuropeanCommission,willhaverelevancewhilescientificpaperswillbenecessary inunderstandingtheimpactthatbatterieshaveontheenvironment.

JackEllisMOI

14

TrendsinEnergyFinanceSince 2004 China has greatly increased investment in the various parts of the solarmanufacturing value chain: poly-silicon feedstock,wafers, cells andmodules. In 2008Chinesefirmsbeganreapinglargebenefitsfromeconomiesofscaleintheproductionofpurified silicon. According to Fischer, until that point soaring global prices due to anoligopolisticmarketstructurehadhamperedtheexpansionofthesector inChina.14ByJuneof2008thecountryhadover700PVmanufacturingcompanies,15andbythistimeChina had become the largest PVmanufacturer in theworldwith 98% of its productshippedoverseas.16By 2009 the Chinese government had recognized solar manufacturing as a strategicindustry. Attempts were made to speed up the growth of the industry, ‘principallythroughacombinationoflow-costdebtandsubsidy.’17Inthefollowingyearfourofthetop10solarPVmanufacturersintheworldwereChinese.18Duringthefinancialcrisisduring2008-09thepaceofPVinstallationsinEuropeslowed.Inresponse,Chinastimulateddomesticdemandfortheirmanufacturedsolarproducts.As the table below shows, Chinese annual solar installation increased over 100 timesduring the period from 2007 to 2012. What is noticeable from the data is how theannualinstallationsinChinaaremuchlowerthantheannualoutputofthecountry.AsPuttaswamy and Sahil Ali note, ‘Chinese solarmanufacturing policywas driven by itsexport potential rather than concerns about supporting domestic deployment, whichweresatisfiedbydefault.’19

Table1:EvolutionofChinesePVcellproduction20

14PEGELS.A.(Ed.).(2014).GreenIndustrialPolicyinEmergingCountries.[Online]Accessedon3/3/2015from:https://books.google.co.in/books?id=C4n8AgAAQBAJ&pg=PA77&lpg=PA77&dq=evolution+of+china%25+27s+solar+PV+manufacturing+industry&source=bl&ots=l5o1-LSY2O&hl=en&sa=X&ei=NpAFVLeXNs248gWR7ILQCg#v=onepage&q=evolutio%2C%202014.&f=false15PAN,J.,MA,H.andZHANG,Y.(2011).GreenEconomyandGreenJobsinChina.WorldWatchInstitute.[Online]Accessedon3/3/2015from:http://www.worldwatch.org/system/files/185%20Green%20China.pdf16SEMI.ChinaMarketGrowthFuelledbyGovernmentSpendingDuringIndustryDownturn[Online]Accessedon3/3/2015from:http://www.semi.org/en/MarketInfo/ctr_02759617PUTTASWAMY,NandSAHILALI,M.(2015).HowDidChinaBecometheLargestSolarPVManufacturingCountry?Step.p.2.18PAN,J.,MA,H.andZHANG,Y.(2011)GreenEconomyandGreenJobsinChina.WorldWatchInstitute.[Online]Accessedon3/3/2015from:http://www.worldwatch.org/system/files/185%20Green%20China.pdf19PUTTASWAMY,NandSAHILALI,M.(2015).HowDidChinaBecometheLargestSolarPVManufacturingCountry?Step.p.2.20ibid.

CSTEP-Note-2015-02

© CSTEP www.cstep.in 2

The Decade of Chinese March Globally increasing demand for solar gear and a domestic thrust on solar manufacturing have propelled China to the top position in terms of Solar Photovoltaic (PV) manufacturing countries. Since 2004, China’s production march on all fronts of the solar manufacturing value chain began- poly-silicon feedstock, wafers, cells and modules. By 2008, the growth of solar industry became formidable as the Chinese firms started reaping economies of scale in the production of purified silicon. By then, China had become the largest PV manufacturer in the world, with 98% of its product shipped overseas (1).

In 2009, the government identified solar manufacturing as a strategic industry and attempted to accelerate its growth principally through a combination of low cost debt and subsidy. By 2010, China accounted for about half of the global production of solar gear, and four out of the global top 10 solar PV cell manufacturers were Chinese (2)(3).

China’s own domestic market for PV installations gathered steam much later. Towards the end of 2007, China’s cumulative installation was only about 100 MW, representing only 1% of the total global PV installations(4). So when the financial crisis of 2008-09 struck Europe and slowed down the pace of PV installations, the Chinese government stimulated their domestic demand for solar gear. As Table 1 shows, Chinese annual solar installation grew by over 100 times between 2007 and 2012 to 3.6 GW. However, this was still not comparable with the annual output of PV cells from China (~20 GW). Chinese solar manufacturing policy therefore was driven by its export potential rather than concerns about supporting domestic deployment, which were satisfied by default.

Table 1: Annual Output of PV Cells and Share of Global Cell Production,(5)(6)

2004 2005 2006 2007 2008 2009 2010 2011 2012

Annual output of PV cells (MWp) 50

200 370 1,087 2,589

4,676 13,018 20,000 23,000

Share of global cell production (%) 4.0 11.0

14.6 27.2 32.9 37.6 47.8 57.3 71.4

Annual PV installations in China (MW)

10 8.9 10 20 40 210 559 2200 3,567

Global Glut in Solar Gear Market During the financial crisis, China strengthened its grip on the export markets. It also contributed to the global glut in the supply of solar gear, which led to further decline prices. The price decline cannot be attributed to productivity gains alone, but also in large part to the supply-demand mismatch.

By 2011, wafer prices had dropped to about 70%, solar cells by about 60%, and module prices halved(7). As the global production in 2010 reached 20.5 GW (or 160,000 metric tons), the prices of components had fallen from $4.5/Watt in year 2000 to $1.7/Watt in 2010(8). Production of solar cells from China alone was around 10 GW, accounting for 50% of the total global production, and more than 90% of solar cells were exported (5) (9).

JackEllisMOI

15

AsTable1shows,duringthetimeofthefinancialcrisisChinawasimprovingitspositionintheexportmarkets.In2004thecountryproduced50MWpofPVcells,afifthofwhichwere installed domestically. This output only represented 4% of the total global cellproductionatthattime.Overthefollowing8yearstheoutputofPVincreasedyear-on-year.In2012thetotaloutputhadreached23,000MWpandtheshareoftheglobalcellproductionhadrisento71.4%.Since2004thepricesofPVcellshavefallensignificantly.Thatsaid,thereisanelementofdebateas towhy thishasbeen thecase.Regarding the totalcostofproduction, thematerialinputconstitutesahighpercentage.AreportbytheInternationalTradeCentrefound75%ofaChinesesolarPVmanufacturer’stotalcostofproductionwasspentonmaterial inputs.21Although this studydidnotexamineawiderangeofmanufacturers,their findingswould suggest that if thenecessarymaterialswere todecrease inprice,thenitwouldaffectthepriceofsolarPVcells.In 2010 the global production of solar PV cells was 20.5 GW and the prices ofcomponents had fallen from $4.5/Watt in 2000, down to $1.7/Watt.22Figure 3 belowshows the development in recent years of the price of imported and domesticpolysilicon in china. This is significant, as polysilicon is the first step in the siliconphotovoltaicvaluechain.(Figure4).

Figure3:Evolutionofpolysiliconprices23

21InternationalTradeCentre.(2015).Servicesinglobalvaluechains:SolarpanelmanufacturinginChinap.vii.22ibid.23FU,R.,JAMES,T.,WOODHOUSE,M.(2015).EconomicMeasurementsofPolysiliconforthePhotovoltaicIndustry:MarketCompetitionandManufacturingCompetitiveness.IEEEJournalofPhotovoltaics.Vol.5.No.2.p.516.

JackEllisMOI

16

Figure4:Siliconphotovoltaicvaluechain

Itcannotbeignoredthatas75%ofthecompany’stotalproductioncostsareinputcosts,thestrongdeclineinpricesafter2010islikelytohavesignificantlyaffectedthepriceofthe solar PV cells produced. IRENA, the International Renewable Energy Agency,calculated that polysilicon prices accounted for 45% of the reduction of solar PVmodules from Q4 2010 to Q4 2012, while other material costs accounted for 19%.24Therefore,in2008,whenthefirstChinesefirmshadmasteredthetechnologiesneededfor the large-scale production of purified silicon, the Chinese solar PV industry grewrapidly.AlthoughChinawasalreadythelargestPVmanufacturerintheworldin2008,itwasafterthistechnologicalbreakthroughthatinvestmentintheproductionofpurifiedsiliconrosesharply.25Theincreasedproductioncontributedtoa‘glut’insupplythatledtofurtherdeclinesinprices.PuttaswamyandSahilAlinotethatthedeclineinpricescannotbeattributedtoproductivitygainsalone,butrathertoasupply-demandmismatch.26The financial crisis did not prove to be damaging to manufacturing. In fact, it issuggestedbyPegels thatmany installers, particularly inEurope and theUSA, actuallybecamemore interested in the cheaper products from china. She cites the Europeananticipationoffurtheradjustmentstofeed-in-tariffs(FITs)asareason.27Estimatesvary,butatendof2012,theworldwideannualsolarPVinstalledcapacityhadreachedabout31-36GW,whileglobalproductioncapacitywasat least60GW,ofwhichChinaaloneconstituted40-55GW.2829IRENA notes that in 2012, all-time-low prices of solar PV modules “overshot” theexpectedlearningcurvewhichwastheresultoftherebeing‘significantovercapacityinmodulemanufacturingandcut-throatcompetition.’30Figure5showshowfarbelowthelearningcurve thepricesofPVmoduleshavebeensince2012,witha learningrateof18-22%.

24IRENA.(2015).RenewablePowerGenerationCostsin2014.p82.25PUTTASWAMY,NandSAHILALI,M.(2015).HowDidChinaBecometheLargestSolarPVManufacturingCountry?Step.26ibid.27PEGELS,A.GreenIndustrialPolicyinEmergingCountries.[Online]Accessedon4/3/2015from:https://books.google.co.in/books?id=C4n8AgAAQBAJ&pg=PA77&lpg=PA77&dq=evolution+of+china%25+27s+solar+PV+manufacturing+industry&source=bl&ots=l5o1-LSY2O&hl=en&sa=X&ei=NpAFVLeXNs248gWR7ILQCg#v=onepage&q=evolutio%2C%202014.&f=false28PUTTASWAMY,NandSAHILALI,M.(2015).HowDidChinaBecometheLargestSolarPVManufacturingCountry?Step.p.2.29EUROPEANCOMISSION.Memo:EUimposesprovisionalanti-dumpingdutiesonChinesesolarpanels[Online]Accessedon4/4/2015from:http://europa.eu/rapid/press-release_MEMO-13-497_en.htm30IRENA.(2015).RenewablePowerGenerationCostsin2014.p82.

3

Si Solar Manufacturing Supply Chain

Also… • Capital equipment • Raw materials • Intermediate products

Poly Si Ingot Wafer Cell Module System

JackEllisMOI

17

Figure5:Globalaveragemoduleprice31

It should be mentioned that both the United States and the European Union haveapplied anti-dumping measures against Chinese manufacturers. Consequently, profitsfor Chinese firms moved from 30% to less than 10%.32In 2012 the EU commissionimposedananti-dumpingdutytobesetatanaverageof47%.33Chineseexportersalsohadtoacceptquantitativerestrictionsonimports.PuttaswamyandSahilAlibelievethatthattheoversupplyandtheensuingpricewararealsofactorsthathavecontributedtoChina’s emergence as a solar manufacturing giant, as even when solar panels havecontinuedtofallthesolarmanufacturersdidnotsufferlosses.ConverselyinEurope,theEU Commission calculated that between 2009 - July 2012, about 40 EU producersdeclared insolvency,6stoppedproduction,2quit thesolarbusinessand4weretakenoverbyChineseinvestors.34In addition to supply-demand asymmetries, China has become the world’s dominantmanufacturer of solar panels due to economies of scale. China is traditionally seen ashaving absolute advantage in labour costs in the manufacturing industry, whencomparedwithindustrialeconomies.35However,Chandlernotesthefollowing:

‘WhileChinadoesindeedhaveasmalladvantageinlabourcosts…thathasrelativelylittleimpactonpricesbecausesolar-panelmanufacturing ishighlyautomated.The lowercostof labour inChinaprovidesanadvantageof7centsperwatt,relativetoa factory inthe31ibid.32PEGELS,A.GreenIndustrialPolicyinEmergingCountries.[Online]Accessedon4/3/2015from:https://books.google.co.in/books?id=C4n8AgAAQBAJ&pg=PA77&lpg=PA77&dq=evolution+of+china%25+27s+solar+PV+manufacturing+industry&source=bl&ots=l5o1-LSY2O&hl=en&sa=X&ei=NpAFVLeXNs248gWR7ILQCg#v=onepage&q=evolutio%2C%202014.&f=false33EUROPEANCOMISSION.Memo:EUimposesprovisionalanti-dumpingdutiesonChinesesolarpanels[Online]Accessedon4/4/2015from:http://europa.eu/rapid/press-release_MEMO-13-497_en.htm34ibid.35PEGELS,A.GreenIndustrialPolicyinEmergingCountries.[Online]Accessedon4/3/2015from:https://books.google.co.in/books?id=C4n8AgAAQBAJ&pg=PA77&lpg=PA77&dq=evolution+of+china%25+27s+solar+PV+manufacturing+industry&source=bl&ots=l5o1-LSY2O&hl=en&sa=X&ei=NpAFVLeXNs248gWR7ILQCg#v=onepage&q=evolutio%2C%202014.&f=false

82

accounting for almost half of the reduction, was a decline in polysilicon prices (45%), followed by other material costs (19%), greater economies of scale in module manufacturing (11%) and technology advancements (10%), while all other factors contributed a total of 16% (GTM Research, 2014).

With prices of solar PV modules at all-time lows, prices in 2012 significantly overshot the expected learning curve (Figure 5.4). This was the result of significant overcapacity in module manufacturing and cut-throat competition that saw many module transactions occur at cash-cost, or in some cases even lower, as financially stressed manufacturers tried to maintain cash flows. In 2013, despite record solar PV installations of around 39  GW, global PV manufacturing capacity, including c-Si and thin-film, exceeded 63  GW (Photon Consulting, 2014). An additional 10  GW of new module production capacity may have been added in 2014 (GTM Research, 2014). The competitive pressures in the solar PV module manufacturing industry are therefore likely to remain intense, although – unlike in recent years – profitability for the major manufacturers has improved and is now on a more sustainable footing.

The rapid decline in c-Si PV module prices due to manufacturing overcapacity has reduced the price

advantage of thin-film PV module manufacturers. This has led to considerable consolidation in the thin-film industry, which should put the remaining manufacturers on a more secure financial footing. However, it remains to be seen whether the specific technological advantages – such as better performance in low-light conditions or hot climates – are sufficient for thin-film modules to substantially increase their share of new installations from current levels.

Despite the pause in reductions in average module selling prices in 2014, current prices are still significantly below the learning curve. They are also now so low that continued cost reductions, based on learning rates of 18% to 22%, will not yield large absolute cost reductions, as in the past. This means – in most countries – that BoS costs, and in particular the soft costs, will provide the largest opportunity for future cost reductions in absolute terms and represent the next great challenge for the solar PV industry.

BALANCE OF SYSTEM COSTS

BoS costs include all the cost components required for a solar PV system, excluding the module costs and includes the hardware costs (e.g. inverters, electrical cabling, racking, etc.) and the soft costs

1 10 100 1 000 10 000 100 000 1 000 000

100.0

10.0

1.0

0.1

1

Cumulative production volume (MW)

%22doubling of cumulative volume%22doubling of cumulative volume

2010

2004

1998

Polysilicon price

Other material

costs

Technology advances

Economies of scale: manufacturing

Other

Module price reduction breakdown bysource

Global average module price (2014 USD/W)

22% price reduction for each cumulative volume

c-SiCdTe

2006 c-Si price increase due to polysilicon shortage

1992

2002

2012

Sources: Based on data from EPIA and the Photovoltaic Technology Platform, 2011; GlobalData, 2014; GTM Research, 2014; Liebreich, 2011; pvXchange, 2014 and IRENA analysis.

FIGURE 5.4: SOLAR PV CRYSTALLINE SILICON AND THIN-FILM MODULE COST LEARNING CURVE

JackEllisMOI

18

United States, but that amount is countered by other country-specific factors, such ashigherinflation.’36

RatheritisthefactthatatypicalChinesePVfactoryisfourtimeslargerthanthoseintheUS. It is this thatmakesacriticaldifferenceaseconomiescanbe found inotherways,suchasbynegotiatingbettercontractswithclientsandmoreeffectiveuseofequipment.

Of further consideration is the efficiency of siliconPVmodules.Not only are they themostefficienttechnology,buttheyhavealsoenjoyedthegreatestriseinefficiencyfrom15%in2003,to21%in2012.37WithsolarPVmodules,astheirefficiencyincreaseslesssurfaceareaisrequiredtomakeamoduleofacertainwattage,thusthepriceperkWhisreduced.Efficiency improvementsreducetheLevelizedCostofEnergy(LCOE)ofsolarPV.AnalysisbyLazardfoundthattheLCOEofrooftopsolarPVwasexpectedtodeclinein the coming years. This is expected to be a result of more efficient installationtechniques,lowercostsofcapitalandimprovedsupplychains.38Duringtheperiod2010-2014 the global average for utility-scale solar PV is estimated to have declined byaroundhalf.Movingfrom$0.32/kWhto0.16/kWh.39

ThecombinationofthevariousfactorsoutlinedabovehasincreasedChineseproductionof solar PV technology and has made the technology itself increasingly economicallyattractive. Swanson’s Law is the observation that the price of PV modules tends todecrease 20% for every doubling of cumulative shipped volume. Themethod used ismorecommonlyreferredtoasalearningcurve.Figure6showstheSwansoneffectoveranextendedperiodoftime.Figure7showsthedevelopmentofglobalPVproduction.

Figure6:EvolutionPVprice($/Watt)40

36MITNEWS.(2013).Solar-cellmanufacturingcosts:innovationcouldlevelthefield.[Online]Accessedon5/3/2015from:http://news.mit.edu/2013/solar-cell-manufacturing-costs-090537IRENA.(2015).RenewablePowerGenerationCostsin2014.p83.38LAZARD.(2015).LazardLevelizedCostofEnergyAnalysis9.0–Keyfindings.p.1.39IRENA.(2015).RenewablePowerGenerationCostsin2014.p94.40ECONOMIST.(2012).PricingSunshine.[Online]Accessedon5/3/2015from:http://www.economist.com/blogs/graphicdetail/2012/12/daily-chart-19

JackEllisMOI

19

0

5000

10000

15000

20000

25000

30000

35000

40000

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

ElectricityOutput(GW

h)

MajorEconomies

Australia France Germany Japan

Spain UnitedKingdom UnitedStates

Figure7:GlobalAnnualPVProduction41

Asevident fromfigure6 thepriceofsiliconPVcellshas fallen from$76.67 in the late1970’s to just $0.74 in2013.According to energytrend.comat the timeofwriting theaverage current cost of a Chinese Multi-Si Cell stands at $0.32. This dramatic fall inprices is responsible for solar PV installations becoming increasingly economicallyviable.Figures8and9showtheincreasingamountofelectricityfromsolarPVinstallations.

41WARMUTH,W.(2016).PhotovoltaicsReport.FraunhoferInstituteforSolarEnergySystemsISE.[Online]Availablefromhttps://www.ise.fraunhofer.de/de/downloads/pdf-files/aktuelles/photovoltaics-report-in-englischer-sprache.pdf

JackEllisMOI

20

Figure8:ElectricityOutput(GWh)fromSolarPVInstallationsinMajorEconomies42

Figure9:ElectricityOutput(GWh)fromSolarPVinstallationsinOECDcountries43

Both figures 8 and 9 show that electricity output has risen dramatically in manycountriesoverthepastseventoeightyears.Whenfigure9iscomparedwithfigure7,itisevidentthat increasedglobalannualproductionhasresulted in increasedelectricityoutputfromsolarPVinstallations.Althoughthisisafairlystraightforwardconclusion,itmaybenotedthat figure9showsanaccelerated levelofelectricityoutputsince2010.Furthermore, figure 8 shows that while Germany’s increasing electricity output fromsolar PV has slowed slightly, the country remains significantly ahead of other majoreconomies.TheUK,Spain,FranceandAustraliaenjoyverymodestgrowthofelectricityoutput whereas the United States and Japan have experienced accelerated levels ofgrowthinrecentyears.DuetoGermany’sslowdown,boththeUnitedStatesandJapanappear on course to overtake it. The reason for Germany’s position is due to itsEnergiewende.This is theGermanword for transitionof thecountry toanenergymixdominated by renewable sources of energy that has been in place since the GermanRenewableEnergyActof2000.Itislikelythattheselevelsofgrowthwillcontinueasin2015renewablescontinuedtobreakrecords.According toBloombergNewEnergyFinance,new investment in cleanenergy reached$329Bn in2015,breaking theprevious recordof$318Bnset in2011.2015alsosetanewrecordintheamountofcapacityadded–121GW.44Thesefigurescanberegardedasverypositive,especiallywhenoneconsidershowthepriceofoil, coalandgashaveevolvedsincemid-2014.45Despite this, the figuresmaskthe truth about European investment in clean energy. In 2015 European investment42INTERNATIONALENERGYAGENCY.[Online]Accessedon8/3/2015from:http://wds.iea.org/WDS/Common/Login/login.aspx43ibid.44BLOOMBERGNEWENERGYFINANCE.(2016)CleanEnergyInvestment:Q42015.45Brentoiltradedat$114inJune2014andatthetimeofwritingstandsatbelow$40.

0

20000

40000

60000

80000

100000

120000

140000

160000

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Electricityoutput(GW

h)

OECDTotal

JackEllisMOI

21

only accounted for $72Bn of the total $329Bn, its lowest investment in clean energysince2006.AsinvestmentfromtheAmericashasremainedrelativelyconsistentinthepastsixyears,(fluctuatinginarangeof$66Bn-$85Bn),EuropehasbeensupplantedbyAsia, and in particular China, as the major driving force for renewables. Chinesespendingonrenewableenergyinfrastructurereached$111Bnin2015,a17%increasefromthepreviousyearandalmostasmuchastheUnitedStatesandEuropecombined.Perhapsthemostsignificanttrendinrenewableenergyfinanceisthefactthattheworldisnowaddingmorecapacityforrenewablesourcesthanforcoal,oilandgascombined.Thiswasfirstrealizedin2013whenthe143GWofrenewablecapacitywasaddedand141GWofcapacityfromfossilfuelsources.BloombergNewEnergyFinancestatedinitsNewEnergyOutlook2015publicationthefollowing:

‘By2040,theworld’spower-generatingcapacitymixwillhavetransformed:fromtoday’ssystem composed of two-thirds fossil fuels to one with 56% from zero-emission energysources. Renewables will command just under 60% of the 9,786GW of new generatingcapacity installed over the next 25 years, and two-thirds of the $12.2 trillion ofinvestment.’46

Inlightoftheevidencepresentedinthissection,itisclearthataseismicshifttowardsrenewableenergyisalreadyunderway.Letusnowturnourattentiontobatterystoragetechnologiesandtheirapplications.

46BloombergNewEnergyFinance.(2015).NewEnergyOutlook2015.

JackEllisMOI

22

Chapter1:TechnicalAspectsofBatteryStorageTechnologiesAsstated,chapteroneaimstoprovideacomprehensiveoverviewofthedifferenttypesof battery storage. It further aims to show various technical applications of batterystoragetechnologyincombinationwithrenewableenergy.Thisisinordertoshowtheservicesthatitisabletoperformontheelectricitysystem.As noted in a 2015 report by the International Renewable Energy Agency (IRENA),‘Energystorageconsistsofa suiteof technologiesatvariousstagesofdevelopment.’47Ontheoppositeendsofthespectrumarepumpedhydropowerandbatterystorage.Theformerrepresents99%ofthestorageinuseandiseconomicallyandtechnicallyproventhroughout the world.48The latter on the other hand, is new market development.Despitethelimitedcurrentdeploymentofbatterystorage,itiscriticalforthepurposeofthisinvestigationthatitsvarioustechnicalaspectsareunderstood.

BatteryStorageTechnologiesThe inventionof thebattery isaccredited toAlessandroVolta,an Italianphysicistandchemist who in 1799 invented the first operational battery. His voltaic pile, whichconsistedofcoinsofcopperandzincseparatedbycardboardsoakedinsaltwater,wasnotrechargeable. Itwasn’tuntil60years laterthat theFrenchphysicistGastonPlantéinvented the world’s first rechargeable battery.49The type of battery invented was alead-acidbatteryandeventhoughtheconceptisover150yearsoldthebatteryisstillknown for its cost effectiveness today. In this section lead-acid, lithium-ion aswell asotherbatterytypeswillbeinvestigated.

Lead-acidAccording to IRENA, lead-acid batteries are already extensively deployed to supportrenewable development.50An example of this is that fact that between 1995-2009Morocco deployed approximately 50,000 solar home systems coupled to batteries inorder to provide rural electrification, while in Bangladesh there are 3.5million solarhomesystems,eachcoupledtoabattery.Typically,theyarefoundintransportvehicles.

Pros

• Easyandcheaptoproduce• Maturetechnology(150yearsofdevelopment)• Highsurgetoweighratio(makingthemsuitableforvehicles)• Easytorecycle

47IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook48ibid.49EUROPEANCOMMISSION.DGENERWorkingPaper:ThefutureroleandchallengesofEnergyStorage.50IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.P.41.

JackEllisMOI

23

Cons

• largeandheavy• Shortlifespan• Environmentalconcerns(Leadisahighlytoxicelement)• Corrosion(chemicalreactions)

Inadditiontotheconsoutlined,manylead-acidbatteriescansufferfromlowDepthofDischarge,(theamountofthebattery’scapacitythathasbeenutilizedandexpressedasa percentage of thebattery’s full energy capacity). This canbe lower than20%.51Thebatteries can also have low cycle numbers (<500) and a limited lifetime of 3-4 years.Theirenergydensityis50Wh/kg.ResearchbyGarciahasshownthatmorerecentversionsofthetechnologycanachieve2,800cyclesataDepthofDischargeof50%,insuringaservicelifeofupto17yearsforindustrial systems.52Of final consideration is that ambient temperature may affectbatteryperformance.This isdue to the fact thathigh temperaturescancause internalreactionstooccur,thusmanybatteriescanlosecapacityinhotterclimates.Conversely,inverycoldclimatesreactionsmaybeslowandcouldevenstopaltogether.Lead-acidconsequently, is a battery technology that ‘may require integrated temperaturemanagementinthebatteryinstallationforoptimalperformanceandsafety.’53Oberhofer,writingonbehalfof theGlobalEnergyNetwork Institute (GENI),expressesthe belief that lead-acid battery technology has reached an end point in terms of itsdevelopment. Hewrites, ‘It is clear that no significant improvements can bemade incapacity, density or weight. Therefore, resources on future development shouldconcentrateonotherbatterytechnologieswithhigherpotentials.’54Althoughtheircosteffectivenesshasmadethemanimportantpartofmanytechnologysystems,Oberhofernotesthatthebatteriesareunlikelytomakeanimpactongridstorage,Hewrites:‘These batteries are not capable of storing huge amount of energy compared to othersystemslikeaPumpedStorageHydroelectricityplant(PSH),whilestayingcosteffectiveastheenergydensityisjusttoolow.Itispossibletointegratebatterybanksforfewsmallerdecentralizedsystems(likephotovoltaic[PV]systemsonrooftops);but,itcantbeusedasadefinitesolution,justforthesimplereasonthattheamountofresourcesarenotavailablefortherequiredcapacityscale.’55Despite this criticism it is important to note that ‘Advanced lead-acid’ batteries alsoexist. The “Ultrabattery”, developed by the Commonwealth Scientific and IndustrialResearchOrganization of Australia, uses an ultracapacitor that enables the battery tooperate longer and more effectively in partial state of charge applications than51ibid.52GARCIA,R.A.(2013).Off-GridCommercialMicro-gridSystemProvidesEnergyStorageforResortinIndia.TrojanBatteryCompany.IntersolarEurope,Munich.53IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.7.54OBERHOFER,A.(2012).EnergyStorageTechnologies&TheirRoleinRenewableRegeneration.GlobalEnergyNetworkInstitute.p.12.55ibid.

JackEllisMOI

24

traditionalleadacidbatteries.56Thisbatteryhasbeentestedforhybridvehiclesbuthasalsobeenproposedanddemonstratedforpowersectorapplicationssuchasfrequencyresponse and smoothing.57Investigation into the applications of battery storage willfollowshortly.

Lithium-ionDuetothefactthatLithiumisthelightestmetallithium-ionbatterieshaveahighenergydensity. Inaddition, theyalsopossessahighpowerdensity (the rateatwhichenergychanges) when compared to other batteries. The combination of these twocharacteristicsallowsthemtotakeupaminimumofphysicalspacewhileprovidinghighlevelsofpower(kW)andenergy(kWh).Furthermore,theirperformance,bothintermsofenergyandpower,continuestoimprove.Thisisonereasonwhytheyaresopopularinconsumerelectronicandpowersectorapplications.58

Pros

• Highlyefficient(typically80-90%)• Highpower(typically3.7Vcomparedto2.0Vforleadacid)• Highenergydensity• Lowenergyloss• Materials available in large amounts (lithium, available in seawater and

obtainablethroughtechnicalmethodsandgraphite)

Cons

• Expensive• Cellscanbecomedamagedwithcompletedischarges• Deteriorateifunused• Safetyconcerns

Thecombinationofhighpowerandenergydensitymakelithium-ionbatteriesanidealtechnology for frequency regulation and other applications that require a relativelyshortdischargeandhighpowerperformance.According to IRENA,oneof thegreatestobstacles facing the technology is safety: Lithium is highly reactive element and iscombustible.Thiscombinedwiththehighenergydensityofthecellsmeanthattheycanoverheatandcatchfire.59Althoughsafetyisanobstacletobeovercome,theircoststoohindertheapplicationofthetechnology.Parttwowillshowthedevelopmentinprices.Thiswill be of importance because asOberhofer notes, ‘lithium-ion batteries have anincrediblyhugepotential.’60

56IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.42.57ibid.58ibid.p.4359ibid.p.43.Thiscanleadtoasituationknownasthermalrunawaywhenneighbouringcellsalsooverheat.Thisleadstoleaks,smoke,gasventingand/orthecellpackcomingalight.60OBERHOFER,A.(2012).EnergyStorageTechnologies&TheirRoleinRenewableRegeneration.GlobalEnergyNetworkInstitute.p.14.

JackEllisMOI

25

Currently there are various companies that are seeking to realize this potential. Theyinclude:

• IBM–TheBattery500Project• Tesla–Powerwallandcars(Roadster,ModelS,Xand3)• Panesonic–Energylogicsystem• Sonnen–SonnenBatterie• RedbackTechnologies–OuijaBoard• Arpa-E61

FlowBatteriesFlowbatteriesaresimilartoothertypesexceptthattheelectrolytescanbeexchanged.Thismeans thatas thebattery isdischarged the fluidsare replacedwith loadedones.Thismakesthemlessaffectedbyoverchargeordischargeandmeansthat theycanbeused without a significant degradation of performance. It is relatively easy to addcapacitytothem,astheirpowerisafunctionofthenumberofcellsthatarestacked.62Theyenjoyalonglifespanbuttheirmajorlimitationisregardingtheirenergydensity,whichat around35Wh/Kg isof a similar level to lead-acidbatteries.Althoughworthmentioning,flowbatterieswillnotbeexploredwithinthispaper.

Sodium/MoltenSaltSodium batteries are a further technology under development, but they are alreadyoperationalinsomecountries.NGKofJapanhasmadesodiumbatteriesforgridstoragefor anumberof years andapproximately250MWof thebatterieshavebeen installedwithin the country.63The technology possesses several advantages: They have a highenergydensity(240Wh/kg),alonglifespanof10-15yearsandahighefficiencyrating.ThebatteriescontainnorareelementsandarefullyrecyclableTheir major limitation however, is that they typically need to be operated attemperaturesapproaching350oCinorderforthesodiumtobeliquid.Thismakesthemdifficult and expensive to operate but most of all dangerous as liquid sodium reactseasilywithwaterintheatmosphere.64

61Arpa-EistechnicallyabranchoftheUSDepartmentofEnergy.Itwasfoundedin2009underObamaseconomicrecoveryplantofundearlystageresearchintothegenerationandstorageofenergy.Projectsaresaidtorelyonmaterials‘beyondcurrentlithium-ionbatteries.’GUARDIAN[Online]Accessedon4/3/2015from:http://www.theguardian.com/environment/2016/mar/03/us-agency-says-has-beaten-elon-musk-gates-to-holy-grail-battery-storage62IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.44.63GREENTECHMEDIA.(2011).IsSodiumtheFutureFormulaforEnergyStorage?[Online]Accessedon4/3/2015from:http://www.greentechmedia.com/articles/read/is-sodium-the-future-formula-for-energy-storage64OBERHOFER,A.(2012).EnergyStorageTechnologies&TheirRoleinRenewableRegeneration.GlobalEnergyNetworkInstitute.p.16.

JackEllisMOI

26

Summarytableofkeyfactors

Source:IRENA.

ApplicationsofbatterystoragetechnologyIt is unquestionable that battery storage in the power sector can be deployed in avarietyofwaysandovermultipletimeperiods.A2013reportbyEPRIandDOEdescribe14servicesunder5umbrellagroupsthatcanbeprovidedbyenergystorage.65Belowisasummaryofthegroups:

Figure10:Servicesprovidedbyenergystorage66

Within this section, four application areas will be discussed that are most directlyrelated to solar PV power integration. As shown in figure 7 in ‘Trends in EnergyFinance’,theelectricityoutputfromsolarPVinstallationshasrisensharplysince2010andisforecastedtoincreaseinthefuture.Throughthisapproachtheservicesthatare65IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.11.66ibid.

LeadAcid AdvancedLead Lithium-ion Sodium-Sulphur Flow

PowerRange(MW) 1-50 1-50 <100 5-100 1-100StorageDuration 2-4h 1min-8h 1min-8h 1min-8h 1-5h

EnergyDensity(Wh/kg) 30-40 50 110-160 240 35

Cycles1,000-5,000

4,500-10,000 500-1000+ 2,500-4,500 >10,000

Operatinglife(years) 3-15 5-15 5-15 5-15 15-20PriceperkWh($) 50-140 140+ 240-2,200 400 N/AEfficiency% 70-90 90-94 85-98 80-90 65-85Responsetime <Seconds <Seconds <Seconds <Seconds <Seconds

JackEllisMOI

27

highlightedinredabovewillbefocusedupon.Specifically,andinlinewiththeobjectiveof this investigation, theapplicationscompensate for thevariablenatureofrenewableenergy. It is also important to note that a single battery installation is able to servemultiple uses. As noted by IRENA, ‘a combination of value streams may benefit theeconomicsofaninstallation.’67

BatteryStorage–Islandsandoff-gridapplicationsManyislandsandoff-gridareasusedieselgeneratorsassourcesofpower.Thelocationis usually remote and the lack of infrastructure means that diesel imports are oftencostly. In addition to being expensive, this type of electrification has high levels ofemissions and has problems regarding security of supply. Despite these issues, dieselhastraditionallybeenusedasthe‘mostaccessibleandcosteffectivesolution.’68Many islands have a lack of flexible sources and as a result would benefit from theapplicationofbatterystorage,asitwouldhelptoreliablyintegratesignificantamountsofrenewableenergyfromsolarorwindandtherebyreducetherelianceondieselorgasgeneration. Figures 11 and 12 below show the increased integration of renewableenergy when combining utility scale wind, diesel powered electricity generation andlead-acidbatteries.Figure11(left):noenergystorage

Figure12(right):withlead-acidbatterystorage.69

Thetwofiguresdemonstratetheabilityofbatterystoragetoincreasethepenetrationofrenewable energyanddecreasebothdiesel andpeakgasuse. In figure11, renewable

67ibid.68ibid.p.12.69BALZA,L.etal.(2014),PotentialforEnergyStorageinCombinationwithRenewableEnergyinLatinAmericaandtheCaribbean,InterAmericaDevelopmentbank

35

tons daily. The figure below shows how wind power and conventional generation are

used for covering the system’s daily load.

Figure 4.4: Impact of Optimizing RE in a Small Island Country

0

20

40

60

80

100

120

140

160

180

MW

327 MWhWind Energy

2.4 GWhBase load

Diesel Gen.

460 MWhShoulder

Diesel Gen.

63 MWhPeak

Gas Turbine

Source: Own calculations

In the figure above, utility scale wind power is displacing some peak generation, but is

constrained by the inability to exceed 15 percent of demand at any given time. Therefore,

there are significant unrealized benefits in this scenario.

When assuming a capital cost for solar PV of US$1,700 per kW and limiting the amount

of wind power capacity to a maximum of 10MW, the model determines that the optimal

amount of solar PV capacity that should be integrated in the electric grid is 29.5MW.

Figure 4.5 below shows how solar PV, wind power, and conventional generation are used

for supply electricity throughout the day.

37

Figure 4.6: Impact of Optimizing RE with ES in a Small Island Country

0

20

40

60

80

100

120

140

160

180

MW

1.4 GWhWind Energy

134 MWhLead-Acid

Battery

1.6 GWhBase load

Diesel Gen.

31.8 MWhShoulder

Diesel Gen.

Source: Own calculations

In the figure above, ES increases the amount of RE used daily by 1GWh and

reduces carbon dioxide emissions by 1,423 tons daily. The figure shows that by adding

ES, all of the expensive peak generation is displaced by lower-cost RE and ES. In

addition, a sizable amount of shoulder and base load capacity is replaced by lower-cost

RE; however, as the figure shows, at a certain point it becomes too expensive to pay for

more ES to back up RE replacing shoulder and base load capacity. For that reason, the

model does not determine that lower-cost RE should replace all conventional generation.

In the RE with ES scenario that constrains the amount of wind power capacity to

10MW and uses aggressive assumptions for solar PV, the model determines that given

the availability of ES, the optimal amount of solar PV capacity that should be integrated

in the electric grid is 49.5MW (in addition to 10MW of wind power capacity). The model

also determines that the electric grid should integrate a 5.4MWh lead-acid battery ES

JackEllisMOI

28

energyfromwindcontributestothepeakdemand.Itspenetrationhowever,isrestrictedby‘theinabilitytoexceed15%ofdemandatanygiventime.’70Balzaetal.assumethistobethemaximumrenewablepenetrationwithoutstorageandconsequentlybelievetheretobesignificantunrealizedbenefitstothescenario.Theirmodeldeterminesthat theelectricitygridshould integratea75.6MWhlead-acidbattery system. Figure 12 shows the impact that combining renewable energy withbattery storagewould have on the daily load profile of the island. (The island underconsiderationisBarbados).Theuseofbatterystoragewasable to increase theamountof renewableenergyuseddailyby1GWhandreducescarbondioxideemissionsby1,423tonesdaily.71Thiswouldmean that emissions of over half amillion tones of CO2would be avoided every yearthroughtheincorporationofa75.6MWhbatterysystem.Theexpensivepeakgenerationwould be displaced by lower cost renewable energy and storage,while the base loadwouldalsobereplaced.Itisimportanttonotethatthereisapointatwhichitbecomestooexpensiveformorestoragetoreplaceshoulderandbaseloadcapacity.AsBelzaetal. write, ‘for that reason, the model does not determine that lower cost renewableenergyshouldreplaceallgeneration.’72Insummarythescenarioencompassesalloftheserviceshighlightedinredinfigure10.It isaneconomicallyviableoptionforachievingmuchgreaterutilizationofrenewableenergybydisplacingdieselandgasgeneration.

HouseholdsolarPVBattery storage at a household level represents perhaps one the most interestingavenuesof investigation.This isprincipallybecause it allows for a far greater levelofself-consumptionofelectricityproducedbysolarPV.Batterystoragehastheability toalign the electricity demand of the user with solar production. It also has the addedbenefitofrelievinglocalgridcapacityconstraints.73

70ibid.71ibid.72ibid.73IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.14.

JackEllisMOI

29

Figure13:SolarPVandbatterystorage74

Figure 13 shows the difference between conventional storage and grid-optimizedstorage. The former, effectively, has uncontrolled battery chargingwhereas the latterhaschargingandPVproductionthattakesgriddemandintoaccount.Thefigureshowsthat optimizing self-consumption of solar-PV is just one aspect of battery storage. IfsolarPVandstoragearetoprovebeneficialtoboththeuserandthelocalgrid,thenthelocalgridareademandmustbetakenintoaccount.75A simple explanation is as follows: If the output from solar PV and batteries are notcontrolled then the batterywill begin to charge in themorning andwill become fullychargedassoonaspossible.Suchaconfigurationmaymeanthat,‘peaksolarproductionisexportedtothegridduringitsmaximumoutput.’76It ishighly likelythatthisexportwill not correspond to periods of peak demand and subsequently would result in anoversupply of renewable energy in relation to demand. The consequences would bevoltagesthatexceedsupportablelimits.AsIRENAconcedesinitstechnologyoutlookforbattery storage, ‘If a large number of distributed solar PV systems are running in aspecific area, this practicemay also limit renewable energy deployment.’77Therefore,integratingbatterystorageontothegridcanleadtothedevelopmentofmorerenewableenergysources.CalculationscarriedoutbytheFraunhofer Institute inGermanyhaveshownthat66%more solarPV canbe installed in a given areawherepeak solarPVproduction is not

74ibid.p.15.75ibid.76ibid.77ibid.

BATTERY STORAGE FOR RENEWABLES: MARKET STATUS AND TECHNOLOGY OUTLOOK 15

with peak demand periods (Fraunhofer ISE, 2013a; BSW, 2013). The application discussed here is similar to the supply shift discussed below, but at a smaller scale for household solar PV.

Figure 9 demonstrates the difference between uncon-trolled battery charging and charging and solar PV production that takes grid demand into account. As fig-ure 9 below demonstrates, optimising self-consumption of solar PV is just one aspect of household battery storage. Local area grid demand must also be taken into account if solar PV and storage are to benefit both the user and local grid. Peak solar radiation usually occurs at or just before noon. If solar PV output and battery charging profiles are not controlled, the battery will charge in the morning and become fully charged. This may mean peak solar power production is exported to the grid during its maximum output. This export may not correspond to grid peak demand periods. This

results in an oversupply of renewable energy in relation to demand, especially in distribution networks, which can potentially leading to voltages that exceed tolerable limits and curtailment of renewable energy resources. If a large number of distributed solar PV systems are running in a specific area, this practice may also limit renewable energy deployment.

The Fraunhofer Institute in Germany calculates that up to 66% more solar PV can be installed in a given area under circumstances where peak solar PV production is not exported to the grid. This is possible when solar feed-in to the grid is restrained, and battery supply matched to household demand. This means self-consumption of solar power can at least double depending on the size of the solar PV installation and battery. For example, a five kilowatt peak (kWp) PV installation with a 4 kWh battery can increase the household’s consumption of PV power from 30% to 60% (Fraunhofer, 2013a; Fraunhofer,

Figure 9: Solar PV and battery storage

charging during periods of high electricityproduction

maximum feed-into grid

charging untilbattery is full

consumption of stored solar power

midday production peakis fed into the grid

reduced feed-into grid

decreased feed-in capacityincreases local grid capacity

by 66%

6 pmdirect consumption reduces

evening peak load

12 pm6 am

6 pm12 pm6 am

Grid optimized storage

Conventional storage

Source: Bundesverband Solarwirtschaft, 2014

JackEllisMOI

30

exportedtothegrid.78Controllingelectricityfeed-intothegridfromsolarPViscrucial,asismatchingthebatterysupplytohouseholddemand.Fraunhoferstatesthattheself-consumption of solar power can double, depending on the size of the solar PVinstallation and battery. The example given is that the percentage of a household’sconsumption of PV power can rise from30% to 60%using the combination of a fivekilowattpeak(kWp)PVinstallationwitha4KWhbattery.79Thisapplicationofbatterystorage then, increaseshousehold solarpowerpenetration, allowsmore solarPV in agivenareaandcontributestogridstability.Offinalconsiderationisthattheattractivenessofresidentialbatterystoragedependson‘the correspondenceofpeak solarproductionwithpeak systemdemand.’80The closerthesetwotimesexistthenthemorebeneficialtheapplicationis.Somehouseholdsmayexperiencepeakdemandduringtheday,butitislikelythatthisismorethecaseingulfcountriesassolarcorrespondswithdemandforair-conditioning.Ratherthewiderpointisthat ‘theoptimalchargingalgorithmswillvaryaccordingtotheparticularelectricitysystemandarea,householdandtimeofyear.’81

VariablerenewableenergysmoothingandsupplyshiftLetusseparatebothapplications:Smoothing:Occursasbatterystorageelectricity is fedontothegridwhile, ‘smoothing’thevariableproductionofcentralisedwindandsolargeneration.Figure14providesaclearillustration.Energyshiftsupply:Excessproductionfromrenewableenergycanbestoredandthenmatched with periods of higher demand. “Shifting” the supply to take advantage ofdifferentprices.Thisisillustratedinfigure15.Bothapplicationsdistinguish themselvesasdifferent fromregulationas theyoccurontheproductionside.Theenergythatisabletobestoredisdirectlygeneratedfromthespecific renewable energy resource. Battery storage regulation on the other hand,operatesatthegridlevel.Letusconsiderthefollowingwithregardsto‘smoothing’:A cloud blocking the sun can cause output to fall by 90% ‘almost instantly’.82(Unexpecteddecreasesinoutputoccurwithwindtoo,butareregardedasslower.)Thisis liable to cause problems for system voltage levels in both the distribution networkandtheoverallstabilityofthesystem.Naturallythiswouldbedependentonanumberof factors such as the size of the system, and its ability to deal with unforeseen

78Dr.BRUNO,B.(2016)FraunhoferInstituteforSolarEnergyISE–PowergenerationfromrenewableenergyinGermany–Assessmentof2015.79ibid.p.3.80IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.15.81ibid.p.16.82ibid.

JackEllisMOI

31

supply/demand changes. Smoothing helps to retain system reliability and voltageconcerns.83As figure14belowshows,smoothingmitigates theshort-termfluctuationsofsolarandwind.

Figure14:Smoothingfrombatterypowerstorage84

The above figure represents the output of centralised PV production of Hawaii. Thebatterypowerisusedtobothchargeanddischarge,helpingtosmooththeoutput.Theconsequencesofthisapplicationarethatoutputfromsolarandwindcanbeintegratedeasierintotheelectricitynetworkandgridstabilitycanbeoptimized.Thisapplicationisespecially important in island andoff-grid systems thatmight have todealwith largefluctuationsinrenewableenergyfeed-in.With supply shifting excess renewable energy is stored for period of higher demand.System stability is one consequence of the application but moreover it allows theintegrationofmorerenewableenergy.Thisdecreasesrelianceonfossilfuelsandavoidsdecliningpowerfromrenewables.Thisapplicationcanmakesoundeconomicsensebychargingthebatterieswhenpricesarelowanddischargingduringpeakdemandwhenpricesarehigher. Italycurrentlyemploysanetmeteringscheme–ScambioSulPosto–which,‘provideseconomiccompensationforfeedinginrenewableenergydependingonthetimeofdayanddemand.’85

83ibid.p17.84ibid.p18.85ibid.p.17.BATTERY STORAGE FOR RENEWABLES: MARKET STATUS AND TECHNOLOGY OUTLOOK18

Figure 10: Illustration of battery storage power smoothing

Example of Power Smoothing

Remaining Battery Storage

12:15 PMBatt

ery

Stor

age

(kW

h)

12:30 PM

AC P

ower

(kW

)

1000

80

600

400

200

0

-200

251

250.5

250

249.5

1200 AC PV PowerAC Power with BatteryBattery Power

Source: Johnson et al. (2011)

by the Italian Energy Regulator in November 2014, battery storage systems could be used to regulate the amount of electricity consumed and fed into the grid, depending on the capacity of the plant and on the applying supporting scheme (Toxiri, 2014). Can’t be existing and planned. One of the existing battery storage systems in Italy uses energy time shift methods to make electricity supply from renewable energy sources more predictable. The plan is for NGK Insulators to provide a sodium-sulphur battery (1-7  MWh) to correct forecast errors for a 30 MW wind park, allowing wind power supply to become more dispatchable. With an estimated investment of EUR 4.5 million, however, the project may not be economic even with the savings arising from avoided prediction penalties (Mazzochi, 2014).

In figure 11, night-time refers to a period of low demand and relatively low prices. Renewable energy is stored at this time and released when demand is higher, corresponding to relatively higher prices and greater stress on the system. Energy supply shift may also occur over a smaller time frame (i.e. 15 minutes to one hour) to avoid curtailment. Over this shorter period, it may display similar economic characteristics.

Battery storage may be applied to grid-level storage as well as renewable energy production. In this case, several variable renewable energy and other generators supply the battery with an electrical charge during periods of low demand. At that point, when prices and demand are higher, electricity is released from the storage system.

The smoothing application is well suited to battery storage, given the need for rapid, quick charging and discharging (see figure 4). Many storage technologies can provide energy time shift. In fact, shifting renewable energy production over a period of time is unique to energy storage. No other flexibility measure can provide this facility.

4.4 Fast regulation in grids with high variable renewable energy shares

Along with other storage technologies, battery storage is well suited to a range of ancillary services. These are defined here as facilities that enhance the security and reliability of the electricity system as well as servicing

JackEllisMOI

32

Figure15:Energysupplyshift86

The services provided by these applications are found under ‘customer energymanagementservices’.Bothaffectpowerreliability,andenergyshiftingisclearlylinkedto the ‘retail electricenergy timeshift’.Manystorage technologiesareable toprovidethisservice,moreovershiftingrenewableproductionisanapplicationthatis‘uniquetoenergy storage’.87On the other hand, the smoothing application is particularly wellsuited tobattery storage, as it requiresveryquickcharginganddischarging,whichashaspreviouslybeenshown,isacharacteristicoflithium-ionbatteries.

Fastregulationingridswithhighvariablerenewableenergyshares.As shown in figure 10, this application falls under the category of ancillary services.IRENAdefinestheseservicesas‘facilitiesthanenhancethesecurityandreliabilityoftheelectricity system as well as servicing the normal production and consumption ofelectricity.'88Inorder tounderstandtheseservices letusconsider threedifferent timeperiodsandcontrol:

• Primarycontrol–10-60seconds(frequencyresponse)• Secondarycontrol–upto10minutes(regulation)• Tertiarycontrol–over10minutestoseveralhours(imbalances/reserves)

Batterypowerhastheabilitytoprovidebalancingpoweratallofthetimeframesabove.Asmentioneditcanchargeanddischarge inseconds,and is ‘fasterandmoreaccuratethan thermal power plants.’89The advantages of battery storage in regulation can besummarizedasfollows:

86ibid.p.19.87ibid.p.17.88ibid.p.19.89ibid.p.21.

BATTERY STORAGE FOR RENEWABLES: MARKET STATUS AND TECHNOLOGY OUTLOOK 19

the normal production and consumption of electricity. This application is also useful in islands - see box 1 for an example.

This report focuses on balancing or controlling power meant to solve short-term active power imbalances (over seconds to hours) that cause the system frequency

to diverge from its target (Hirth and Ziegenhagen, 2013). These services are generally categorised accord-ing to the time frame for which power is provided or taken away. The exact definition varies among systems. The discussion in this section focuses on short-term regulation or frequency/primary response (in seconds). This is an important ancillary service needed in systems

Box 3: Case study: Doha, Qatar, frequency response and other ancillary servicesAs a climate change conference got underway in late 2012, BYD launched its lithium-ion phosphate battery solution at the Qatar Science and Technology Park. The battery is 500 kWh and is charged by an adjacent solar PV installation and diesel generator.

The project provides an interesting hybrid system with benefits for both on and off-grid applications. These include voltage/reactive power support, frequency regulation and black start capabilities. Black start is when a power station restarts without the external electricity grid due to a total or partial shutdown of the transmission system. The project represents an interesting bundle of ancillary services. Battery storage can take over at least some of these, in addition to frequency response.

Some of the most important project criteria were ambient conditions and temperature management in a desert environment location and bundling of services. Performance requirements were another, including the ability to deliver power quickly and stay highly charged in order to supply black start capability. See figures 4 and 5 for an overview of factors significant for battery selection.

Case study 12 in the addendum to this report provides additional information.

Sources: BYD Energy (2012; 2014).

Figure 11: Illustration of energy supply shift

Energy supply

Load curve

Power generation

Energy storage

21 0 63 9 12 15 18 21

21 0 63 9 12 15 18 21

Night time

Day time

Night time

Day time

Energy supply

Energy storage

Carnegie, et al. (2013)

JackEllisMOI

33

• Short term output variations from renewable energy sources can be quicklycompensated

• Fullnegativeandpositivecapacityforregulation.(Seefigure14)• Fasterramp-upperiodsthanfossilfuelpowerplants• Reducestheneedtokeepcombustionturbinesonline.(Reductionofgreenhouse

gasemissions)• Reduction of maintenance cost, as frequency regulation provided by

conventional power plants may ‘accelerate equipment degradation due torampingrequirementsoffrequencyregulation.’90

Figure16:100MWBatterystorage(left)versus100MWgasturbine(right)91

Batterystorageclearlyhasmanypositives inensuringgridstability. It is likelythat itsapplicationwill become evenmore compelling in the future as increasing amounts ofrenewableenergyareintegratedintotheelectricitysystem.Itisworthmentioningthatbatteries used in this applicationwill be subject tomultiple charges/discharge cycles,andwillberequiredtoprovideahighlevelofpoweroverashorttimeperiod.

SummaryChapteronehasachievedtwoimportantstepsof the investigation.Firstly, thevariousdifferent typesofbatterystorage technologyareunderstood:Lead-acidbatterieshavebeenusedtosupportrenewableenergyproduction.However,althoughstilluseful,thebattery suffers from a number of limitations and is unlikely to undergo significantfurther development in the future. Not disregarding alternatives, lithium-ion enjoysadvantages in terms of power, energy and efficiency and is regarded as the mostpromisingintermsofpotential.

90ibid.p.22.91ibid.p.21.

BATTERY STORAGE FOR RENEWABLES: MARKET STATUS AND TECHNOLOGY OUTLOOK 21

charge and discharge energy in seconds or less, faster and more accurately than thermal power plants.

An electricity system benefits in several ways from the fast, accurate ramping provided by battery storage. The battery can quickly and accurately compensate for short-term output deviations from variable renewable energy generators in order to maintain system fre-quency. This concept is illustrated in figure 12. It shows wind output in the Texas system and corresponding regulation employed to maintain a frequency of 60 Hz., the system target in the U.S.

Figure 12 illustrates the concept of frequency response in an interconnected system. It shows significant downward ramps of wind in the Electric Reliability Council of Texas (ERCOT) system. These start at around 9 am and 1.30 pm. It also shows the required deployment of rapid response reserves or primary frequency response. The red line depicts system frequency. This is seen to fall as the slope of aggregated wind generation resource output decreases, indicated by the blue line. The purple line illustrates primary frequency response deployed (rapid response reserve) due to the decline in frequency from the 60 Hz target. The need for this short-term regulation is probably related to errors in forecasting peaks and troughs in the wind resource. Though the rapid response reserve depicted

here is probably not battery storage, batteries are well suited to this application. Other types of regulation are also being deployed by fossil fuel plants. This is the aggregate amount indicated by the green line. This application occurs at the aggregated grid level and is therefore distinguished from variable renewable energy smoothing (as displayed in figure 10) which is employed at the renewable energy installation site. Texas has around 12  GW of wind capacity according to the Energy Information Administration (EIA, 2014a) and peak demand of 50-70 GW (winter and summer) (ERCOT, 2014).

Battery storage offers its full negative and positive capacity for regulation, as well as a faster ramp rate than fossil fuel power plants. By contrast, a fossil fuel plant is constrained by a minimum operating level requirement below which operation and maintenance costs would suffer (see figure 13).

The battery resource needs less capacity than its fossil fuel regulation equivalent due to its positive regulation attributes. This is because battery storage is faster, more accurate and able to provide its full capacity8 for positive

8 The actual economic range of operation provided by a battery depends on the technology (cycle life, DoD limitations etc.) and how remuneration for these services is calculated.

Figure 13: 100 MW Battery storage (left) versus 100 MW gas turbine (right)

-100 MW

+100 MW-25 MW

+25 MW

+50 MW

50 MW flexible range

Minimum operating level

Source: Vassallo, A. (2013)

JackEllisMOI

34

Secondly, thissectionhasshowntheservices thatcanbeprovidedbybatterystorage.Different applications of battery storage have shown a technology able to mitigateagainsttheintermittentnatureofrenewableenergyinanumberofdifferentwaysandindifferentareas.Thishasincludedbothproductionandgridlevelapplicationsaswellas residential/utility scales. Inachieving the fiveserviceshighlighted in figure10,onecanseehowoneinstallationcanservemultipleuses.Theeffectofthesevaluestreamson the economics of the installation will be investigated in the following section. Inassessing theeconomicviabilityofbattery storage, it is additionallynecessary togainunderstanding of howbattery storage technologieswill evolve, both in terms of theirtechnicalspecificationsandprice.Priortochapter2isalistofbatterystoragebenefits:

• Consumercontrol–customersenjoygreatercontroloftheirbillsbyshiftingenergyuse.

• Abletosupplycapacityandbackuppower(atacheaperratethanquickresponsefossilfuelplants)

• Ancillaryservices–keepingelectricitysupplyanddemandinbalance.Helpstomaintainthevoltageandfrequencyoftheelectricitysystem

o Avoidsdamagetoelectronics/motorso Avoidspowercuts.

• Beneficialtoinfrastructure-powerlinesandgridinfrastructurewearquickeroperatingatpeakcapacity.Astheenergycanbeshiftedinfrastructureinvestmentcanbereduced.

• Renewableenergysupport–Variableinnature,renewableenergycanbedifficulttoaccommodateforutilitypowerplants.Batterystoragerespondsquicklytovariationsinoutputresultinginhigherpenetrationsofrenewableenergyontothegrid.

• Qualityandreliability–especiallytrueinareasofweakinterconnectionsuchasislands.

• Promotesjobcreationintechnologicalandserviceindustries.• Reducessizingofdistributedgenerationsystems

JackEllisMOI

35

Chapter2:TheeconomicviabilityofbatterystoragetechnologyAlthoughthecombinationofbatterystorageandrenewablesbringsbenefits,thereisadegreeofuncertaintyunderwhichconditionsbatterystoragecanbeoperatedwithoutpolicysupport.This sectionof thepaper investigates theeconomicviabilityofbatterystorage technology. The secondpart of the chapter provides a brief investigation intothe price development of lithium-ion technology. The final part looks at the differentmarketsforenergystorage.Itaskshowmarketsarelikelytodevelopandwhatbarriersstoragetechnologiesneedtoovercome.

ResidentialSolarPV

As shown in ‘Trends in Energy Finance’, the price of solar PV technology has fallengreatlyoverthepast40years.ThishasbeenakeyfactorintheincreasedinstallationofsolarPVofmanycountries.Ofthemajoreconomies,Germanyhasachievedthegreatestelectrical output from solar PV installations. A paper published by the FraunhoferInstituteexplainswhythishasbeenthecase:‘Thanks to technological progress, the learning curve and economies of scale theinvestment costs for solar PV plants, whichmake up the greatest outlay, have fallen anaverageof14%peryear–inall,almost75%since2006.’92Figure 17 below shows the price evolution since the second quarter of 2006. Thepercentageinorangerepresentthesolarmodulecostsoftheinstallation.

Figure17:Averageendconsumerpriceforinstalledrooftopsystems.93

92Dr.WIRTH,H.(2015).RecentFactsaboutPhotovoltaicsinGermany.FraunhoferInstituteforSolarEnergySystemsISE.

2015_Jan_21_Recent_Facts_about_PV_in_Germany.docx25.01.16 9 (89)

Figure 3: Average end customer price (net system price) for installed rooftop systems with rated

nominal power from 10 - 100 kWp, data from BSW, plotted by PSE AG.

Module costs are responsible for about fifty percent of the total investment costs for PV power plants of this size. This percentage increases for larger power plants. The price development of PV modules follows a so-called “price learning curve,” in which dou-bling the total capacity installed causes prices to fall by a constant percentage. Figure 4 shows the global prices adjusted for inflation and calculated in euros in line with the 2013 exchange. At the end of 2014, the cumulative installed PV capacity worldwide reached approximately 180 GW. Provided that significant progress can continue to be made in product development and manufacturing processes, prices are expected to keep dropping in accordance with this rule.

JackEllisMOI

36

AlthoughaddingbatterystoragetoresidentialsolarPVsystemscanenhancethevalueoftheelectricityproduced,italsoincreasestheoverallinvestmentcost.Forthisreason,inGermanythereexistprogrammesthatsubsidizetheuseofenergystorage.ApaperbyHoppmann et al. aims to investigate the profitability of storage for PV systems anddetermine the optimal size of the system. In doing this they seek to overcome twolimitationsofexisting studies:Firstly,due to theuncertainties surrounding tariffsandtheexpectationthattheywillbephased-outatsomepointinthefuture,profitabilityisinvestigated without ‘demand side subsidies.’ 94 Secondly, especially under theassumption of no additional incentives, the sizes of the solar PV system and batterystoragegreatlyaffecttheviabilityoftheintegratedbattery-PVsystem.Thereasonthisisthecaseisbecausetheeconomicviabilityofthesystemis,‘stronglydrivenbythedegreetowhichelectricityproducedbythePVsystemisself-consumed.’95Theextenttowhichtheelectricity is self-consumed isdependentuponsizeofboth thePVsystemand thebattery storage. That is to say that currently there exists uncertainty as to wheninvestmentsinbatterystoragewillbeeconomicallyviable,forahouseholdthatwishestooptimizeboththesolarPVandbatterysystems.In order to answer the question ofwhen andunderwhich conditions battery storagewill be economically viable with solar PV, without subsidies, they created a techno-economic model that is able to calculate profitability. The authors’ investigation iscarriedouton8differentelectricitypricescenariosandtheirresultsshowtheoutcomesthroughthreesections:

1. OptimalPVsystemsize2. Optimalstoragesize3. Profitabilityofstorage

Table2showstheelectricitypricescenariosusedinthemodelsimulations.Scenarios1-5 all assume that the household has unlimited access to the wholesale market.Conversely,scenarios6-8allassumethatthehouseholdhasnoaccesstothewholesalemarket.Theoutcomesofthemodelareillustratedforeachsectionontheassumptionornot of access to the wholesale electricity market. All of the electricity wholesale andretailpricescenariosareperyear.

93ibid.p.9.94Aresultofthisapproachisthatchangestothewholesaleandretailpricesofelectricitywillstronglyaffectstorageprofitability.HOPPMANJ.,VOLLAND,J.,SCHMIDT,T.S.,HOFFMANN,V.H.(2014).TheEconomicViabilityofBatteryStorageforResidentialSolarPhotovoltaicSystems–AReviewandSimulationModel,RenewableandSustainableEnergyReviews39,1101-1118.p.3.95ibid.

JackEllisMOI

37

Table2:Electricitypricescenariosusedinmodelsimulations96

Themajorpointsofinterestfromtheresultsareoutlinedbelow.

OptimalSolarPVSystemSizeFigure18:OptimalSolarPVplantsizeS1-S597

96Takenfrom:HOPPMANJ.,VOLLAND,J.,SCHMIDT,T.S.,HOFFMANN,V.H.(2014).TheEconomicViabilityofBatteryStorageforResidentialSolarPhotovoltaicSystems–AReviewandSimulationModel,RenewableandSustainableEnergyReviews39,1101-1118.p.14.97ibid.p.19.

Scenario AssumptionElectricityWholesalePrice

ScenarioElectricityRetailPrice

Scenario

S1 Unlimitedaccessofhouseholdto

wholesalemarket

High:+3% High:2%S2 Low:-1%

S3 Medium:1.5% Medium:+1%

S4 High:+3% Low:+0%S5 Low:-1%

S6 Noaccessofhouseholdto

wholesalemarket

Constant:0EUR/kWh

High:2%

S7 Medium:+1%

S8 Low:+0%

19

Figure 7: Optimal PV plant size under electricity price scenarios S1 to S5

Figure 8: Optimal PV plant size under the assumption that the household cannot sell electricity on the wholesale market (electricity price scenarios S6 to S8)

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14 3.5

2.5

3.0

2.0

1.5

1.0

0.5

0.02022202120202018201720162015 20192013 2014

S5: Low retail, low wholesale price scenarioS4: Low retail, high wholesale price scenarioS3: Medium retail, medium wholesale price scenarioS2: High retail, low wholesale price scenarioS1: High retail, high wholesale price scenario

OptimalPV system sizein kWp

Year ofinvestment

Electricity generation/electricity consumption

ratio

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

2016 201920182017201520142013 20212020 2022

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0

OptimalPV system sizein kWp

Year ofinvestment

S8: Low retail price scenario, wholesale price = 0 EUR/kWhS7: Medium retail price scenario, wholesale price = 0 EUR/kWhS6: High retail price scenario, wholesale price = 0 EUR/kWh

Electricity generation/electricity consumption

ratio

JackEllisMOI

38

• AllinvestmentsinsolarPVsystemsareprofitableundertheinvestigation,asnoscenariohastheoptimumsizeof0.(ThisistruewithS6-S8infigure19aswell).

• WhenlookingatS3(mediumretailprice,mediumwholesaleprice),theoptimalsizeofthesolarPVsystemsignificantlyincreasesovertimefrom<3kWpto7.

• Until 2016, the optimal size of the generation consumption ratio is <1. Thismeans that the household generates less electricity than it consumes. Theauthorsoutlinetheirreasoningastowhythisisthecase:‘Thisisduetofactthatinvestment costs for both the PV and storage systems are relatively high,requiringthehouseholdtohaveahighrateofdirectselfconsumptionwhichcanonlybeachievedwhenchoosingasmallPVsystemsize’.98Itshouldbenotedthatasinvestmentcostsfalltheoptimalgeneration/consumptionratioincreases.

• The difference between S1 (high, high) and S5 (low, low) increases over theyears and by 2022 is radically different. This would suggest that the optimalsolar PV plant size is very sensitive to future retail and wholesale electricityprices. As an analysis, the authors write: ‘stronger increases in retail pricesfavourlargerPVplantsizesastheyenhancethevalueoftheelectricityproducedbythePVsystem–whichsubstituteselectricitypurchasedfromthegrid.’99

• Withwholesaleprices,optimalsystemsizeisgreaterwhenwholesalepricesarehigh (S1, S4).This isbecause it increases thevalueof excesselectricity,whichcanbesoldonthemarketatahigherprice.

• Initially, it istheretailpricethathasgreaterinfluenceoverthePVsystemsize,howeverthischangesinthelaterperiodandthewholesalepricebecomesmoreimportant.Onthissubject,Hoppmannetal.write:‘Thiscanbeexplainedbythefactthatwithfallingtechnologycosts,thesizeofPVplantsrisesovertimewhichleads to a situation where households, despite using storage, need to sell anincreasingamountoftheirelectricityonthewholesalemarket.’100ThefactthatS1needstosell2.5timesitstotalconsumptionin2022,underpinstheinfluencethatthewholesalepricescanhaveontheoptimalPVsystemsize.

98ibid.p.19.99ibid.p.20.100ibid

JackEllisMOI

39

Figure19:OptimalPVplantsizeS6-S8101

• WiththeassumptionthatthehouseholdhasnoaccesstothewholesalemarkettheoptimumPVsizeissignificantlysmallerwhencomparedtoS1-S5.Thisisduetothe fact thatanyextraelectricitygeneratedcannotbesoldonthewholesalemarket,thusthehouseholdwillnaturallychoosethePVsystemsizethat limitstheproductionofextraandinconsumableelectricity.

OptimalStorageSizeFigure20:OptimalStorageSizeS1-S5102

101ibid.p.19.

19

Figure 7: Optimal PV plant size under electricity price scenarios S1 to S5

Figure 8: Optimal PV plant size under the assumption that the household cannot sell electricity on the wholesale market (electricity price scenarios S6 to S8)

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14 3.5

2.5

3.0

2.0

1.5

1.0

0.5

0.02022202120202018201720162015 20192013 2014

S5: Low retail, low wholesale price scenarioS4: Low retail, high wholesale price scenarioS3: Medium retail, medium wholesale price scenarioS2: High retail, low wholesale price scenarioS1: High retail, high wholesale price scenario

OptimalPV system sizein kWp

Year ofinvestment

Electricity generation/electricity consumption

ratio

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

2016 201920182017201520142013 20212020 2022

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0

OptimalPV system sizein kWp

Year ofinvestment

S8: Low retail price scenario, wholesale price = 0 EUR/kWhS7: Medium retail price scenario, wholesale price = 0 EUR/kWhS6: High retail price scenario, wholesale price = 0 EUR/kWh

Electricity generation/electricity consumption

ratio

21

Figure 9: Optimal storage size under electricity price scenarios S1 to S5

Figure 10: Optimal storage size under the assumption that the household cannot sell electricity on the wholesale market (electricity price scenarios S6 to S8)

0.00.51.01.52.02.53.03.54.04.55.05.56.06.57.07.58.08.5

202120202019201820172016201520142013 2022

Optimalstorage sizein kWh

Year ofinvestment

S1: High retail, high wholesale price scenarioS2: High retail, low wholesale price scenarioS3: Medium retail, medium wholesale price scenario

S5: Low retail, low wholesale price scenarioS4: Low retail, high wholesale price scenario

0.00.51.01.52.02.53.03.54.04.55.05.56.06.57.07.58.08.5

2022202120202019201820172016201520142013

Optimalstorage sizein kWh

Year ofinvestment

S8: Low retail price scenario, wholesale price = 0 EUR/kWhS7: Medium retail price scenario, wholesale price = 0 EUR/kWhS6: High retail price scenario, wholesale price = 0 EUR/kWh

JackEllisMOI

40

• AswithsolarPVplantsize,theoptimalstoragegenerallyincreasesoverthetimeperiod.InthecaseofS3,theoptimalstoragesizemovesfrom4.5kWhin2013to>7in2021.103

• When considering a particular retail price, either high or low, the optimalstoragesize ismarginally larger inscenarios thatassumeagreater increase inwholesale prices.104For example, when considering high retail prices, S1 isgenerallylargerthanS2,andwithlowretailpricesS4isgenerallylargerthanS5.As shown in the previous section, higher wholesale prices lead to larger PVplants.Itislikelythatthis,inturn,raisestheoptimalstoragecapacity.

Figure21:OptimalStorageSizeS1-S5105

• Generally speaking, the impact of wholesale prices on optimal storage size islow.ThisisdemonstratedbythefactthatS6-S8infigure19areverysimilartoamediumretailpricescenarioinfigure18.

StorageProfitability

• Asshowninfigures20and21investmentsinstoragewerealreadyprofitablein2013underallofthepricescenarios.

• The profitability of storage continuously rises over time in an almost linearfashion.ThismeansthatthevalueofstorageperEuroinvestedinstoragemovesinS3from<0.5to>2.5within10years.

102ibid.p.21.103ThereasonS3levelsoffpost2021isduetothefactthatthemodelincludesaconstraintformaximumsolarPVsystemsize.Thislessensthesizeofstorageinstalledundertheeconomicconsiderations.104ibid.p.22.105ibid.p.21.

21

Figure 9: Optimal storage size under electricity price scenarios S1 to S5

Figure 10: Optimal storage size under the assumption that the household cannot sell electricity on the wholesale market (electricity price scenarios S6 to S8)

0.00.51.01.52.02.53.03.54.04.55.05.56.06.57.07.58.08.5

202120202019201820172016201520142013 2022

Optimalstorage sizein kWh

Year ofinvestment

S1: High retail, high wholesale price scenarioS2: High retail, low wholesale price scenarioS3: Medium retail, medium wholesale price scenario

S5: Low retail, low wholesale price scenarioS4: Low retail, high wholesale price scenario

0.00.51.01.52.02.53.03.54.04.55.05.56.06.57.07.58.08.5

2022202120202019201820172016201520142013

Optimalstorage sizein kWh

Year ofinvestment

S8: Low retail price scenario, wholesale price = 0 EUR/kWhS7: Medium retail price scenario, wholesale price = 0 EUR/kWhS6: High retail price scenario, wholesale price = 0 EUR/kWh

JackEllisMOI

41

• Aswith optimal PV system size and optimal storage size, storage profitabilitydependsgreatlyonretailprices.Asisclearlyvisible,higherretailscenarios(S1andS2)raiseprofitability.

• Inlateryears,particularlyfrom2018,lowerwholesaleelectricitypricesincreasetheprofitabilityofstorageinvestments.ThisiswhenthePVsystemsarelargerandhouseholdssellahighershareoftheirelectricityonthemarket.

• The result of this is that storage investments are profitablewith no access towholesalemarkets(S1-S8)

Figure22:StorageprofitabilityS1-S5106

106ibid.p.23.

23

Figure 11: Storage profitability under electricity price scenarios S1 to S5

Figure 12: Storage profitability under the assumption that the household cannot sell electricity on the wholesale market (electricity price scenarios S6 to S8)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

2022202120202019201820172016201520142013

Net present value of storage per EUR invested in storage

Year ofinvestment

S2: High retail, low wholesale price scenarioS1: High retail, high wholesale price scenario

S5: Low retail, low wholesale price scenarioS4: Low retail, high wholesale price scenarioS3: Medium retail, medium wholesale price scenario

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

2022202120202019201820172016201520142013

Net present value of storage per EUR invested in storage

Year ofinvestment

S8: Low retail price scenario, wholesale price = 0 EUR/kWhS7: Medium retail price scenario, wholesale price = 0 EUR/kWhS6: High retail price scenario, wholesale price = 0 EUR/kWh

23

Figure 11: Storage profitability under electricity price scenarios S1 to S5

Figure 12: Storage profitability under the assumption that the household cannot sell electricity on the wholesale market (electricity price scenarios S6 to S8)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

2022202120202019201820172016201520142013

Net present value of storage per EUR invested in storage

Year ofinvestment

S2: High retail, low wholesale price scenarioS1: High retail, high wholesale price scenario

S5: Low retail, low wholesale price scenarioS4: Low retail, high wholesale price scenarioS3: Medium retail, medium wholesale price scenario

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

2022202120202019201820172016201520142013

Net present value of storage per EUR invested in storage

Year ofinvestment

S8: Low retail price scenario, wholesale price = 0 EUR/kWhS7: Medium retail price scenario, wholesale price = 0 EUR/kWhS6: High retail price scenario, wholesale price = 0 EUR/kWh

JackEllisMOI

42

Figure23:StorageprofitabilityS6-S8107

SummaryTheresultsof the investigationcarriedoutbyHoppmannetal.havebeenusedwithinthispaperas theyserve itwell.Theyprovideacomprehensiveassessmentofhowthesizes of the solar PV system and battery storage change over time using a number ofdifferentscenarios.Furthermore,themodellingwascarriedoutwithoutsubsidies.Theresultsdemonstratethatbatterystorageiseconomicallyviableunderallofthedifferentpricescenariosthatwereconsidered.Theprofitabilityofstorageisforeseentoincreasegreatestwithadecrease inwholesaleelectricitypricesandasimultaneous increase inretailprices.Infact,storageprofitabilityisnotunderminedevenifthehouseholdhasnoaccesstothewholesalemarkets.It is worth reiterating that investment in battery storage is already profitable forresidential solar PV systems. As the authors conclude, the optimal size of both thesystem and storage rises over time and has the consequence of making, ‘householdsbecomenetenergyproducersbetween2015and2021iftheyareprovidedaccesstotheelectricitywholesalemarket.’108Theeconomicviabilityofstorageiscontributedtowithanydevelopmentthata)increasesretailpricesorb)decreaseswholesaleprices.Despitetheusefulnessofthestudyitdoeshaveseverallimitations.AstheinvestigationwasonlycarriedoutforamodelhouseholdinGermanyitwouldbebeneficialtorepeattheanalysisindifferentlocations.Thisistrueastechnologycosts,solarirradiation,andelectricity prices and consumption patterns differ in countries.109Awider criticism oftheinvestigationisthattheanalysisfocusesonretailandwholesaleprices.Aswehaveseen inchapter1,batterystorage isable togeneratevalue throughotherapplicationssuch as frequency regulation and energy shifting (arbitrage). This would lead one tosuggest that using a combination of different applicationswould further increase theeconomicviabilityofbatterystorage.In‘RecentFactsaboutPhotovoltaicsinGermany’,thefollowingiswritten:

‘ThepredominantlydecentralizedwayinwhichPVisfedintothedistributiongridincloseproximitytoconsumersreducesgridoperatingcostsandinparticularthoserelatingtothetransmissiongrid.AfurtheradvantageoffeedinginPVisthatinadditiontofeedinginrealpower, PV plants are in principle able to offer extra grid services (e.g. local voltageregulation)atcost-effectiveprices.’110

CombinedsolarPV-batterysystemsarenotonlyeconomicallyviableforconsumers,butare also able to lessen transmission costs and provide other services. As a result, theeconomicviabilityofstorageis likelytobeconsiderablygreaterthanoutlinedthroughtheHoffmannpaper.

107ibid.108ibid.p.29.109ibid.p.28.110Dr.WIRTH,H.(2015).RecentFactsaboutPhotovoltaicsinGermany.FraunhoferInstituteforSolarEnergySystemsISE.P.39.

JackEllisMOI

43

Of final consideration is the fact that lead-acid batteries were used throughout theinvestigation.Asshowninchapter1,thistechnology,althoughwidelyusedandeffectivedoesnotconstitutethebestoptionasitisunlikelytodevelopgreatlyinthefuture.Themostpromisingbatterytechnologyislithium-ion,which,amongotheradvantages,hasahigherpowerandenergydensitythanlead-acid.ThefollowingsectionwillexplorethedevelopmentsinpriceofLithium-iontechnology.

Pricedevelopmentoflithium-iontechnologyItissafetosaythatthepriceoflithium-iontechnologyisdecreasing.Thatsaid,thereareuncertaintiesoverhowmuch ithasdecreased in thepastandtherateatwhich itwillcontinuetodecreaseinthefuture.Obtaininganaccurateassessmentofthesekeyfactorsconstitutes an important part of this investigation and is critical in assessing theeconomic viability of battery storage technology. One reason for the difficulty inobtaininganaccurateassessment isdue to the fact that the industry is secretivewithsensitive information. It is possible for example that costs are overestimated in anattempttohidetheactualcostsorbatteriessubsidizedinordertogainmarketshare.111ArecentstudypublishedinNatureClimateChangeanalysesover80differentestimatesreported during the period 2007–2014. It describes itself as presenting a ‘first-of-its-kindsystematicreviewofthecostofbatterypacks,’112andtracesthecostoflithium-ionbatteriesthroughthisperiod.Itisimportanttostatethatthebatteriesanalysedwereforelectricvehicleapplication.However,batterypacksforresidential/businessusearetheexactsametechnology.Thebatterypacksarecomposedofthemodulesandthemodulescomposed of the cells. The application may be different but the technology isfundamentally the same. The only area where lithium-ion technology differs is withregards to the materials that companies use for the cathode and anode within theindividualcellsofthebattery.Thepaperincludes,‘costestimatesofallvariantsofLi-iontechnologyusesforBEV(BatteryElectricVehicles),astheaimistotracktheprogressofBEV technology in general and data is too scarce for individual Li-ion cell chemistryvariants.’113Thus, the review allows us to gain insight in to the price development oflithium-iontechnology.Figure24belowshowsthedevelopmentofpriceandthereareseveralpertinentpointsthatmustbemade.Themostobvious,butalsoimportantobservationisthatthepriceoflithium-ion batteries per kWh has decreased significantly. As of 2014 there are twoprice estimates, (calculated as themean), that are significant. The first, shown by theblack line, represents the industry wide cost. This is $410, which represents a 14%declineannuallysince2007.Thesecond,shownwithadashedgreenline,representsthemarket leaders.Theircost issignificantly lowerat$300perkWhandsince2007theyhaveenjoyedan8%annualdecrease.NykvistandNilssonpointoutthatthisis, ‘ofthe

111NYKVIST,B.&NILSSON,M.(2015)Rapidlyfallingcostsofbatterypacksforelectricvehicles.NatureClimateChange.Vol.5.April2015.p.329.112ibid.113ibid.

JackEllisMOI

44

order of two to four times lower than many recent-peer reviewed papers havesuggested.’114Perhapswhat ismoresignificant is thatby2014thecostswere inmanycases below the average projected costs for 2020, which are shown by the yellowtriangles.The costof lithium-ion technology is clearly indeclineand furthermore, thecostsofbatterypacksamongmarketleadersaremuchlowerthanpreviouslyreported.Figure24:CostofLithium-ionbatterypacksinbatteryelectricvehicles115

Regarding learning rates, the cost reduction following a cumulative doubling ofproduction, it is true that the authors, ‘recognize the huge uncertainty in aggregatingdifferent estimates.’116The authors state that thedatahas toomuchuncertainty tobeused directly together with data on cumulative capacity to estimate learning rates.However,byusing‘modelledaveragecosts’,theauthorscalculatealearningrateof9%fortheindustryasawholeand6%formarketleadingactors.117Howthislearningratedevelopsinthefutureisofgreatsignificanceinunderstandinghoweconomicallyviablebatterystoragetechnologyislikelytobeinthefuture.Catenaccietal.suggestthatthereare still research and development improvements to be made in anode and cathodematerials, separator stability and thickness and electrolyte composition.118This isconsideredbyNykvistandNilsson,whowritethattogetherwithimprovementsduetoeconomies of scale, ‘a 12-14% learning rate is conceivable.’119In light of this it seems

114ibid.115ibid.p.330.116IEA.(2015).EVbatteries:howquicklyarepricesfalling?[Online]Accessedon25/3/2016from:https://www.iea.org/newsroomandevents/pressreleasesnews/3-questions/ev-batteries-how-quickly-are-costs-falling.html117NYKVIST,B.&NILSSON,M.(2015)Rapidlyfallingcostsofbatterypacksforelectricvehicles.NatureClimateChange.Vol.5.April2015.p.329.118CATENACCI,M.,VERDOLINI,E.,BOSETTI,V.&FIORESE.(2013).G.Goingelectric:Expertsurveyonthefutureofbatterytechnologiesforelectricvehicles.EnergyPolicy61,403-413119NYKVIST,B.&NILSSON,M.(2015)Rapidlyfallingcostsofbatterypacksforelectricvehicles.NatureClimateChange.Vol.5.April2015.p.330.

LETTERS NATURE CLIMATE CHANGE DOI: 10.1038/NCLIMATE2564

0100200300400500600700800900

1,0001,1001,2001,3001,4001,5001,6001,7001,8001,9002,000

2005 2010 2015 2020Year

2025 2030

2014

US$

per

kW

h

95% conf interval whole industry95% conf interval market leadersPublications, reports and journals

News items with expert statementsLog fit of news, reports, and journals: 12 ± 6% decline

Additional cost estimates without clear methodMarket leader, Nissan Motors, Leaf

Market leader, Tesla Motors, Model SOther battery electric vehicles

Log fit of market leaders only: 8 ± 8% declineLog fit of all estimates: 14 ± 6% declineFuture costs estimated in publications

<US$150 per kWh goal for commercialization

Figure 1 | Cost of Li-ion battery packs in BEV. Data are from multiple types of sources and trace both reported cost for the industry and costs formarket-leading manufactures. If costs reach US$150 per kWh this is commonly considered as the point of commercialization of BEV.

100

1,000

10 100 1,000 10,000 100,000Cumulative MWh

Whole industry, LR = 9 ± 1.1% Log−log fit R = 0.99; p = 0.006Market leaders, LR = 6 ± 0.6% Log−log fit R = 0.99; p = 0.004Other industry, LR = 6 ± 1.6% Log−log fit R = 0.94; p = 0.031

2014

US$

per

kW

h

Figure 2 | Modelled experience curves for battery packs. Learning rate isbased on modelled cost data and estimated cumulative capacity for thewhole industry, market leaders, and other industry with market leaderssubtracted. Underlying uncertainty in cost data must be taken into accountwhen interpreting results.

most important, and costs as low as US$300 per kWh due to suchimprovements have been discussed2. Together with improvementsdue to economies of scale, a 12–14% learning rate is conceivable.A techno-economic explanation for the identified rapid decline incost is that the period since 2007 represents the earliest stage ofsales growth for BEVs. The estimates for the industry as a wholethus reflect a wide range of Li-ion battery variants at initiallylow production volumes, as well as necessarily immature batterypack production techniques among BEV manufacturers. A rapidlydeveloping and restructuring industry in its early phase could yieldhigh learning rates at pack level. However, the learning rate forNiMH batteries in hybrid vehicle applications have historicallybeen 9% (ref. 2), much closer to the modelled learning rates inthis paper. Hence, we believe that the 8% annual cost decline formarket-leading actors is more likely to represent the probable futurecost improvement for Li-ion battery packs in BEV, whereas the14% decline for the industry as a whole to some degree representsa correction of earlier, overestimated costs. It is likely that themanufacturers with the highest car sales at present will have the

most competitive battery pack costs and that these represent a morerealistic long-term learning rate. With a cost level of approximatelyUS$300 per kWh these market-leading actors now set the de factocurrent costs for state-of-the-art battery packs.

It can be expected that the cost gap between market leadersand the industry as a whole will narrow over the coming years. Insuch a scenario2, assuming continued sales growth of the order of100%, and using learning rates and cost declines identified in thispaper, there is a convergence of estimates of battery cost for thewhole industry and costs for market-leading car manufacturers in2017–2018 at around US$230 per kWh. This is significantly lowerthan what is otherwise recognized in peer-reviewed literature, andon par with the most optimistic future estimate among analystsoutside academia (by McKinsey), which stated in 2012 that US$200per kWh can be reached in 2020, and US$160 per kWh in 2025(ref. 15). From US$230 per kWh, costs need to fall a further third toreach US$150 per kWh, at which BEVs are commonly understoodas becoming cost competitive with internal combustion vehicles5.More recent academic studies find similar target costs16, but analystsof, for example, the US market suggest that competitiveness withinternal combustion vehicles is reached already at US$400 per kWhfor fuel cost of US$6 per gallon, and US$250 per kWh at US$3–4.5per gallon11,15, the latter range reflecting current conditions. TheInternational Energy Agency (IEA) estimates that parity withinternal combustion cars in general is reached at US$300 per kWh(ref. 17). However, there are large uncertainties in these typesof scenarios, and recent empirical research has found no clearcorrelation between fuel prices and actual BEV uptake18. BEV salesare taking o� at today’s cost of US$300 per kWh, but BEVs are still aniche product among early adopters. As well as lower battery costs,important explanatory factors behind this take-o� include publicincentive schemes, and the local or regional presence of charginginfrastructure and national manufacturers18, because each of thesecontribute to alleviating cognitive barriers10. However, if costs reachas low as US$150 per kWh this means that electric vehicles willprobably move beyond niche applications and begin to penetratethe market widely, leading to a potential paradigm shift in vehicletechnology.However, it should be noted that factors such as resourceavailability and environmental impacts from a life-cycle perspective

330 NATURE CLIMATE CHANGE | VOL 5 | APRIL 2015 | www.nature.com/natureclimatechange

© 2015 Macmillan Publishers Limited. All rights reserved

JackEllisMOI

45

reasonable to suggest that learning rates could increase in the near future. That said,MunueraandCazzolawritethefollowinginregardstolearningrates:‘Thedegreetowhichthesetrendscanbeextrapolatedintothefutureisnotclear–wehavecome to understand over the past few years that that learning rates for energytechnologiesareveryrarelyconstantovertime,andtheyhavenotbeenforLi-ion.’120Thiswouldsuggestthatevenifalearningrateof12-14%wereachieved,itisunlikelytobemaintainedoveranextendedperiodof time.Despitethis theauthorsalsostate thefollowing:‘Having said that, when excluding quoted figures or estimates, which are often notcomparable,butinsteadatthefundamentalsofthetechnologyaswellasdevelopmentsinthepipeline,westillbelievethatthereisasignificantpotentialforcostreductionwithLi-ionbatteries’121Whetherornotthis,“significantpotential”translatesintoahigherlearningrateenjoyedover a long period of time remains to be seen. Some authors have noted that manyadvancementsatcell levelhavealreadybeenrealized,122andothersthatacommercialbreakthrough of the next generation of lithium air-based batteries is still far in thefuture.123Suchperspectivesgiveweighttotheargumentthatthelearningratewillslowinthecomingyears.Whatisclearthoughisthatthelearningrateenjoyedbytechnologyhas leadtodecreasingcosts foranumberofyearsnowandforthetimebeingat leastthesecostswillcontinuetofall.NykvistandNilssonstatetheirbeliefthatthe8%annualcostdeclineenjoyedbymarketleaders is likely torepresent the, ‘probable futurecost improvement forLi-ionbatterypacks,whereasthe14%declinefortheindustryasawholetosomedegreerepresentsacorrectionofearlier,over-estimatedcosts.’124Thisanalysiswouldsuggestthatthecostlevelof$300perkWhthemarketleadersnowsetsisthede-factocurrentcostforstateoftheartbatterypacks.Whatisparticularlynoteworthywithintheirstudyisregardingtheconvergenceofbatterycostsforthewholeindustry.Theypredictthatthisissettohappen in 2017-2018 at around $230 per kWh.125Thus it appears that costs will bepushedtowards$200perkWh.Thisislikelytobethecasewithoutmajorchangestocellchemistry, but perhaps is to some extent dependent on the success of large-scaleproduction facilities such as Tesla’s Gigafactory, in order to achieve the necessaryeconomies of scale. Speaking at the unveiling of the Tesla Model 3 electric car, ElonMusk, theCEOofteslamotorssaidthattheGigafactorywould, ‘producemore lithium-

120IEA.(2015).EVbatteries:howquicklyarepricesfalling?[Online]Accessedon25/3/2016from:https://www.iea.org/newsroomandevents/pressreleasesnews/3-questions/ev-batteries-how-quickly-are-costs-falling.html121ibid.122CAIRNS,E.J.&ALBERTUS,P.(2010)Batteriesforelectricandhybrid-electricvehicles.Annu.Rev.Chem.Biomol.Eng.1,299-320.123MAYER,T.,KREYENBERG,D.,WIND,J&BRAUN,F.FEASIBILITYstudyof2020targetcostsforPEMfuelcellsandlithium-ionbatteries:Atwofactorexperience.124NYKVIST,B.&NILSSON,M.(2015)Rapidlyfallingcostsofbatterypacksforelectricvehicles.NatureClimateChange.Vol.5.April2015.p.330.125ibid.

JackEllisMOI

46

ionbatteriesthanallotherfactoriesintheworldcombined.’126Hefurtherstatedthatitwouldhave50GWperyearofproduction,withthemostadvanced lithium-ionbatterybeingproduced.BeforemovingontothefinalpartofthechapteritshouldbementionedthatNavigant,aresearch group, forecasts thatpriceswill reach$200per kWh.However, according totheir forecasts thepriceswill reach a floor at $200or slightlybelow.They state theirbeliefthatpriceswillbebelow$200perkWhinthenext5yearsbutsaythat,‘beingabletogobeyondthat,to$150or$100ispotentiallyimpossiblebutdefiantlyveryveryhardtodowithLithium-ionchemistries.’127Ultimatelytimewilltellhowlowpriceswillfall.

MarketdevelopmentforbatterystoragetechnologyThe finalpartof chapter2 aims todevelopourunderstandingofhow themarkethasdevelopedforbatterystoragetechnology.Aswehaveseen,thecostsofbatterysystemsand integrated solar PV-battery systems have fallen in recent years. There have beenincreased levels of deployment and greater interest in the use of battery storage forrenewable energy integration. A clear example of this is the fact that in Germany thepricesforbatterystoragesystemsconnectedtoasolarPVsystemfellby25%in2014.128In2014themostinstalledbatterytypewassodiumsulphur.Thiswillnotbethecaseformuch longer as themarket is shifting towards lithium-ion batteries and also utilizingadvanced lead-acid technology. As shown in chapter 1, lithium-ion has proved itselfpreferabletootherchemistriesinregardstoenergyandpowerdensity, lifecyclesandcost.Sodiumsulphurbatteriesremainanimportantbatterytype,butintermsofmarketdevelopment theshift isundoubtedly towards lithium-ion.Figure25belowshows theestimated installed battery capacity and commissions in the power sector by batterytype.

126YOUTUBE.(2016).ElonMuskUnveilsTeslaModel3.[Online]Accessedon5/4/2016.Availablefromhttps://www.youtube.com/watch?v=7HOhOeKwe30127NAVIGANT.(2016).BeyondLithium-ion.[Online]Accessedon5/4/2016.Availablefromhttps://www.navigantresearch.com/webinarvideos/webinar-replay-beyond-lithium-ion128IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.27.

JackEllisMOI

47

Figure25:Estimatedinstalledbatterycapacityandcommissionsinpowersector,2014129

Although the estimates are slightly old, the figure gives a clear indication of how theinstalled capacity in 2013-2014 for lithium-ion far outstrips other technologies. Thisdevelopmentisunlikelytoslow.Asshownintheprevioussection,thepriceoflithium-ion technology has decreased greatly in the recent past and will continue to do so.IRENA notes that, ‘the most dramatic cost developments have been for lithium-ionchemistries,drivenbypoliciestodeploythetechnologyintheelectricitysectorandtheelectric vehicle market.’ 130 As lithium-ion technology permits a wide range ofapplications(chapter1),therearemultiplebenefitsfortheelectricitysector.Inadditiontobatteriesbecomingincreasinglycompetitiveinthemarket,thankstocostreductions,itisworthmentioningthatregulationsarebeginningtomoveawayfromanapproach to grid services centredon fossil fuel.131Furthermore, continued research iscausingmore barriers to be overcome, and increased knowledge of how installationswork will make residential, commercial and utility adopters more comfortable withtheirutilization.

Utility,residentialandnon-residentialmarketsegmentsThe utility market is expected to grow considerably in the coming years. Figure 26belowshowstheworldwideforecastofbatterystoragecapacityandannualrevenueofutilityscaleapplications.

129IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.26.130ibid.p.28.131ibid.p.24.

BATTERY STORAGE FOR RENEWABLES: MARKET STATUS AND TECHNOLOGY OUTLOOK26

sustainability of the batteries themselves must also be considered. This is highlighted by the recycle/reuse phase of figure 16. Box 4 gives an overview of how to assess battery sustainability.13

Figure 17 shows some of the companies providing advanced battery storage technologies for utility-scale applications. NGK insulators is the market leader with almost 350 MW of installed battery storage systems, of which around 200 MW in utility-scale applications. This market has changed dramatically in recent years. Com-mon in new and developing markets, some companies have gone bankrupt. Several of these have re-emerged as a different company or through acquisition. These include A123 Systems (which sold its grid storage unit to NEC), Xtreme Power (acquired by Younicos), Valence Technology, Exide and others.

13 The data has been verified and augmented with data from the DOE global database for capacity installed to date (lithium-ion, vana-dium redox flow, zinc bromine redox flow and nickel-cadmium) for operational projects.

Recent developments for battery types present in the power sector are described in section 5.2, which includes cost trends and statistics.

5.2 Analysis of battery types

A few years in the energy sector is usually considered a blink of an eye. This makes the rapid transformation of the battery storage market in recent years even more remarkable. The battery storage landscape in the elec-tricity sector is moving away from the former market concentration of sodium-sulphur batteries provided by NGK Insulators, still the world’s only provider of this type. It has shifted towards lithium-ion batteries, as well as advanced lead-acid. This is depicted in figure 18. For many applications, lithium-ion has proved preferable to other chemistries with respect to energy and power density, cycle and calendar life, and cost.

The capacity estimates above do not include small solar PV installations coupled with battery storage at the household level. They also exclude a recent 35 MW

Figure 18: Estimated installed battery capacity and commissions (MW) in the power sector by type, 201413

0

100

200

300

400

Otherbatteries

Nickel-cadmium

Redox flowbattery

Advanced lead-acid

Lithium-ion

Sodiumsulphur

MW

Existing capacity in 2012

Installed capacity in 2013-2014

Source: Navigant Research (Dehmana, et al. 2014)

JackEllisMOI

48

Figure26:Worldwideforecastofbatterystoragecapacity(MW)andannualrevenue(USD)forutilityscaleapplications132

Therevenuefromapplicationsissettoincreasefromaround$200min2014to$18Bnin2023. Thestoragecapacitywillrisefrom360MWto14GWoverthesameperiod.Itshould be highlighted that these estimates are for utility only and do not includebatteriesandsolarPVsystemsthatare“behindthemeter”,suchashouseholdsolarPVinstallations.Thesetooareundoubtedlyasignificantmarketopportunity.Regardingbatteryuse,Navigantexpectsittobecomprisedofthefollowing:

Table3:Applicationsforbatteryuse.133

Themultipleapplicationsforbatteriesareakeyreasonwhythereappearstobeaverystrongupwardtrend,asshowninfigure15.Itmustbesaidthatpredictingenergysectordevelopmentisverydifficultandasaresultthefuturerevenueisalmostcertaintodifferfrom the estimates. The complexity is increased when one considers that there areseveralstepswithinthebatterymarketsupplychain.Manydifferentindustrialplayersare involved in order for the batteries to bemanufactured and sold.134That said, theSolarEnergy IndustryAssociation of theUSA, (SEIA) predicts strong growth in utility

132ibid.133ibid.134Batterymarketsupplychain:chemicalsupply>batteryproduction>integration>use>recycle/re-use

BATTERY STORAGE FOR RENEWABLES: MARKET STATUS AND TECHNOLOGY OUTLOOK24

the market. Further, regulations are beginning to move away from an approach to grid services centred on fossil fuels. All these factors will continue to drive the use of battery storage in the electricity grid to unprecedented levels, albeit starting from a very low baseline.

The battery storage market in the power sector has seen significant growth in recent years. For utility-scale applications (excluding battery storage installed behind-the-meter), global 2014 revenue was around USD 220 million, according to Navigant research. Asia Pacific, Europe and North America are first movers in the market. A country analysis, including drivers, is presented in section 5.3.

This market is expected to grow in coming years. Figure 14 shows Navigant’s worldwide sales estimate for cells used in utility-scale projects (data excludes batteries with solar PV system behind the meter). The annual revenue for all applications is expected to increase from USD 220 million in 2014 to USD 18 billion in 2023. An-nual battery storage capacity will rise from 360 MW to 14 GW over the same period. For utility-scale projects, Navigant expects battery use for renewables integration

in 2014 to comprise 29% of the total. This is followed by peak shaving (20%),10 load shift (18%, similar to the energy supply shift application discussed in this report), ancillary services (17%) and other applications (16%). Renewables integration is expected to remain a primary application in 2023, providing 40% of cell-based rev-enue. This will be followed by load shifting application (37%), peak shaving (15%), ancillary services (3%) and others (5%) (Jaffe and Adamson, 2014). However, these numbers do not include household solar PV installa-tions, which represents a significant market opportunity.

The strong upward trend is noteworthy, although future revenue will differ from figure 15 estimates above due to the complexity of predicting energy sector develop-ment. Several steps are needed to manufacture and sell the batteries. This involves many industry players, as illustrated in figure 16.11 12

10 Peak shaving refers to utilisation of battery storage to reduce a facility’s peak demand. The application is primarily for commercial and industrial customers in developed grids who face high peak demand charges.

11

12

Figure 15: Worldwide forecast of battery storage capacity (MW) and annual revenue (USD) for utility-scale applications

0

3000

6000

9000

12000

15000

18000

0

2000

4000

6000

8000

10000

12000

14000

16000

2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

Ann

ual r

even

ue (U

SD t

hous

ands

)

MW

Annual Power Capacity (MW) Annual Revenue (USD Thousands)

Source: Data from Navigant Research (Jaffe and Adamson, 2014)

Application 2014 2023Renewablesintegration 29% 40%

Reducingpeakdemand(shaving) 20% 37%Energysupplyshift 18% 15%Ancillaryservices 17% 3%

Other 16% 5%

JackEllisMOI

49

scalePVprojects,andforecaststhat12GWofnewutilityscalePVgenerationwillcomeonline in 2016, almost triple that of 2015. In addition to this, SEIA also noted that19.8GW of utility-scale are ‘contracted for or under construction.’135Bloomberg NewEnergy Finance forecasts that solar will, ‘boom worldwide, accounting for 35%(3,429GW)ofcapacityadditionsandnearlyathird($3.7trillion)ofglobal investment,splitevenlybetweensmallandutility-scaleinstallations.’136Theirforecastingwasupto2040. In light of this itwould appear that there is very strong growth in utility-scalesolarPV,andthatitissetforaverybrightfuture.Howmuchinstalledbatterycapacitytherewillbeisverydifficulttopredict,buttheopportunityforintegratedbattery-solarPV systems is certainly growing at a rapid rate. The different value streams offeredthrough different applications of battery storage only serve to emphasize utility andeconomic advantages. This is especially the case with integrating a higher level ofrenewableenergy into theenergysystem, theuseofwhich is forecasted torise in thefollowingyears.In regards to residential and small-scale solar PV, Bloomberg New Energy Financewrites the following: ‘The real solar revolution will be on rooftops, driven by highresidential and high commercial power prices, and the availability of storage.’137Thepublicationgoeson to say thatby2040, justunder13%ofglobalgenerating capacitywillbesmall-scalePV.138Thisishighlysignificant.Aswehavealreadyseeninchapter2,solar PV-battery storage systems are economically viable under a range of differentpricescenarios. Itwasalso foundthat theoptimalsizeofboththesystemandstoragerisesovertime.Aslithium-ionbatteriesfallincostitislikelythatanincreasingnumberofpeopleinstallbattery-PVsystems,andreducetheamountofelectricitytheyconsumefromthegrid.BeforeturningattentiontotheinternationalmarketletusbrieflyinvestigatethestatusoftheUSsolarmarketsegmentsattheendof2015.Figure27belowshowstheannualUSsolarPVinstallationsfrom2000-2015.

135SOLARENERGYINDUSTRIESASSOCIATION.(2016)Demandforutility-scalesolarstilldominating.[Online]Accessedon5/4/2016.Availablefrom:http://www.seia.org/blog/demand-utility-scale-solar-still-dominating136BloombergNewEnergyFinance.(2015).NewEnergyOutlook2015.137Ibid.138ibid.

JackEllisMOI

50

Figure27:AnnualUSsolarPVinstallationsfrom2000-2015139

• Utility solar PV installations remain the largest segment by capacity adding4,150MWofcapacity.Thisrepresentsa6%increaseover2014and57%ofthetotal capacity installed in 2015. As noted, SEIA expects installations in thissegmenttotripleto12GWin2016.

• Theresidential segment installed2,099MWin2015,whichrepresenteda66%

growthover2014.Themarketgrewat its largestannualgrowth rate in2015.This is particularly impressive as it was the fourth year in which growth hasexceeded50%.

• Non-residential PV installations dropped 5% from 2014 levels with 1,011MW

installed. The segment has suffered from flat-demand, but installations areexpectedtorisein2015withtheinstallationof‘atriple-digit-megawattpipelineofcommunitysolarprojects.’140

As figure27shows, the solarPVmarket in theUShasexperiencedenormousgrowth.Regarding individualsegments,utilityscaleremainsthe largestandis likelytoremainso.Residentialhasbeenthefastestgrowing,buttheoverallgrowthininstallationshasbeensogreatthatin2015forthefirsttimesolarPV(29.4%)beatnaturalgas(29%)incapacityadditions.141Figure28belowshowstheUSPVinstallationforecastuntil2021aswellastheforecastsby segment. The reason 2016 has such a high level of additional capacity is due to afederal InvestmentTaxCredit,whichwasdue to runout in2016.TheSEIAdescribedcongress’ decision of a multi-year extension of the tax credit in December 2015 as,

139GMTRESEARCH&SEIA,(2016).U.S.SolarMarketInsight,ExecutiveSummary.GMTResearch&SEIA.p.5.140ibid.,p.6.141ibid.

Introduction Error*

U.S. Solar Market Insight March 2016 │ 5

1. INTRODUCTION 2015 was a momentous year for solar power in the United States. Solar PV deployments reached an all-time high of 7,260 megawatts direct current (MWdc), up 16% over 2014 and 8.5 times the amount installed five years earlier. Total operating solar PV capacity reached 25.6 GWdc by the end of the year, with over 900,000 individual projects delivering power each day. By the the time this report is published in Q1 2016, the U.S. will be approaching its millionth solar PV installation.

Figure 1.1 Annual U.S. Solar PV Installations, 2000-2015

When accounting for all projects (both distributed and centralized), solar accounted for 29.4% of new electric generating capacity installed in the U.S. in 2015, exceeding the total for natural gas for the first time.

4 11 23 45 58 79 105 160 298 385852

1,925

3,372

4,761

6,247

7,260

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

PV In

stal

latio

ns (M

Wdc

)

Residential Non-Residential Utility Total Installations

JackEllisMOI

51

‘withoutquestion…themostimportantdevelopmentforU.Ssolarinalmostadecade.’142Consequently, a considerable amount of capacity sought to take advantage of the taxcredit,whichexplainsthelargeamountofinstalledcapacityforecastedfor2016.What is the significance of this for the battery market? Well, it is likely to be verysignificantindeedasitgreatlyincreasestheopportunityofintegratedsolarPV-batterysystems. In addition to the recordbreaking levels of solarPV capacitybeing added in2015andforecastedfor2016,thereisanotherfactortoconsider:In2015,39%ofnewcapacityadditionsintheUSweremadeinWind.143Thatmeansapproximately17GWofrenewable energy was installed in the US in 2015. As has been shown in thisinvestigation, battery storage technology is able to support intermittent renewableenergythroughanumberofvalue-addingapplications,andultimatelyisabletoachievehigherpenetrationsofrenewableenergyontothegrid.Forthesereasons,themarketforbattery storage technology appears set to increase, especially in regards to the utilityandresidentialscalesolarPVsegments,nottomentionlargeapplicationopportunitieswithregardstowindpower.Figure28:USPVinstallationforecast2010-2021&USinstallationforecastbysegment144

Before turning to the internationalmarkets, it should bementioned that 2015 saw alarge amount of debate regarding solar regulations. The Public Utilities Commission(PUC)ofCalifornia,reachedadecisiononnetmeteringwhichhasbeenlargelyviewedas favourable for solar. (Net metering is when utilities are required to buy excesselectricity generatedbyhomeowners’ solarpanels.)On theotherhand, inNevada thePUChas issued anorder that increases customer fixed charges.Thisnot onlymade itmoreexpensivetousesolar,butalsomadeituneconomical forthosethathadalreadysigned up.145Although NV Energy, has proposed a ‘grandfathering proposal’, whichallowstheoldratestructuresforexistingconsumers,thedebacleservesanexamplefortheattitudeofutilitiestowardssolar.RegardingthisCarlPopewritesthefollowing:

142ibid.p.16.143ibid.p.6.144ibid.p.17.145BUHAYAR,N.(2016)WhoOwnsTheSun?BloombergBusinessWeek[Online]Accessedon7/4/2016.Availablefrom:http://www.bloomberg.com/features/2016-solar-power-buffett-vs-musk/

Photovoltaics Error*

U.S. Solar Market Insight March 2016 │ 17

Looking ahead, the federal ITC will remain at 30% through 2019, and then step down to 26% in 2020 and 22% in 2021. In 2022, it will step down to 10% for third-party-owned residential, non-residential, and utility PV projects, while expiring entirely for direct-owned residential PV. Equally important, projects that commence construction but do not interconnect in years 2019, 2020, and 2021 can qualify for correspondingly larger tax credits if they come on-line by the end of 2023.

Given the timing of the federal ITC extension, however, the wheels are already in motion for U.S. solar to benefit in 2016 from a double-digit-gigawatt pipeline of late-stage utility PV projects that rushed through development last year. In turn, we expect another record year for the U.S. PV market in 2016, with installations reaching 16 GWdc, a 119% increase over 2015. In 2017, while the residential and non-residential PV markets are both expected to grow year-over-year, the U.S. solar market is still expected to drop on annual basis due to the aforementioned pull-in of utility PV demand in 2016.

In 2018, U.S. solar is expected to resume year-over-year growth across all market segments. And by 2021, more than half of all states in the U.S. will be 100+ MWdc annual solar markets, bringing cumulative U.S. solar installations above the 100 GWdc mark.

Forecast details by state (34 states plus Washington, D.C.) and market segment through 2021 are available in the full report.

Figure 2.7 U.S. PV Installation Forecast, 2010-2021E Figure 2.8 U.S. PV Installation Forecast by Segment

Source: GTM Research

0

5,000

10,000

15,000

20,000

25,000

2010 2011 2012 2013 2014 2015 2016E 2017E 2018E 2019E 2020E 2021E

Inst

alle

d Ca

paci

ty (M

Wdc

)

Residential PV Non-Residential PV Utility PV

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2010 2011 2012 2013 2014 2015 2016E 2017E 2018E 2019E 2020E 2021E

Inst

alle

d Ca

paci

ty (M

Wdc

)

Residential PV Non-Residential Utility

JackEllisMOI

52

‘Utilitiesdon’tmindthatsolar isrenewable,carbonfreeandenjoys free fuel–as longastheyownit.Butsolarisalsomodularanddecentralized,whichtheyhate.Theydon’twantto competewith their own customers. Rooftop threatens the sunk utility investments incentralisedfossilpowerplantsandtheirrigid“bigtosmall–guaranteedreturnoncapital”business model. Above a certain scale, rooftop solar will force utilities both to retireexpensivecentralstationpowerplantstheywantintheirbaserateandtotransformtheirbusinessmodeltoaccommodatesmallgeneratortolargegridelectronflowwhichrooftopsolarenables.146The themes that arise from the quote abovewill be focusedupon in greater depth inChapter 3. For now, let us draw the conclusion that the economic impact of thepropagationofrooftopsolarisunlikelytobeinthebestinterestsofutilities.

InternationalmarketsTable 4 below shows the installation targets of major markets for utility scale PVsystems.

Market/Country InstallationTarget ShareofRenewableGenerationChina 150GWby2020 20%by2030Japan 64GWby2030 22-24%by2030US - 20%by2030

Germany 66GWby2030 50%by2030UK 22GWby2020 15%by2020India 100GWby2022 40%by2030Taiwan 8.7GWby2030 13.3by2030

Table4:Installationtargetsofmajormarketsandoverallshareofrenewablegeneration.147

Achievingtheseinstallationtargetswillpututility-scalesolarPVonthepathtobecomethenumberonesectorintermsofcapacityadditionsoverthenext25years,whichitisforecastedtobecome.148According to an extensive study into energy storage carried out by AECOM, the keycharacteristicsthatinfluenceinvestmentresponseininternationalmarketsincludedthefollowing:

• Highpenetrationsofrooftopsolar• Highpenetrationsofutilityscalewindandsolar149

146POPE,C.(2016).RooftopSolarWarsContinue.EcoWatch.[Online]Accessedon7/4/2016.Availablefrom:http://ecowatch.com/2016/01/29/rooftop-solar-wars/147ENERGYTREND.(2016).InstalledcostofutilityscalePVsystemintheUSdown17percent.[Online]Accessedon5/4/2016.Availablefrom:http://pv.energytrend.com/research/Installed_Cost_of_Utility_Scale_PV_System_in_the_US_Down_17percent_YoY_in_3Q15.html148BLOOMBERGNEWENERGYFINANCE.(2015).NewEnergyOutlook2015.p.4.149AECOM.(2015).EnergyStorageStudy.AustralianRenewableEnergyAgency.p.54

JackEllisMOI

53

Thissuggeststhattheinstallationtargetsforutility-scalePV,systemsshownintable4,will have the effect of increasing the investment in battery storage technology. It ishighlylikelythatthiswillalsobethecasewithrooftopsolar;thattheincreasingshareofrenewablegenerationasshowninfigure28willleadtoincreasedinvestmentinbatterystorage.

U.S.A.Currently theUS is themarket leader inbattery storage implementation,withgrowthbeing driven by the country’s 2009 federal stimulus package, the American Recoveryand Reinvestment act (ARRA).150ARRA provided $100m for power sector batterystorage projects. This amount was matched by private funds which made a total of$222mtowardsbatterystorageimplementation.Asshownintheintroductionwith‘theduck curve’, the integration of variable renewable energy has created grid reliabilityissues.AccordingtoIRENA,thishas‘drawnattentiontotheneedtoleveltheregulatoryplayingfieldandcompensatenon-traditionalmeasuresforthebenefitstheyprovide.’151In 2007 the Federal Energy Regulatory Commission (FERC) passed order 890. Arequirementoftheorderwasthatenergystoragebegantobeconsideredforancillaryandgridservices.Essentiallytheorderraisedthepossibilityofbatterystoragenotonlyprovidinggridservices,butgettingpaidforthemtoo.In California the Renewables Portfolio Standard requires all utilities in the state tosource a third of electricity sales from renewable sources by 2020.152The state alsosubsidizes battery installations by around $1.6/Watt through its self-generationincentive programme.153The program is one of the longest running distributedgeneration programs in the United States. It is expected that as the penetration ofrenewables increases therewill be a growing demand for frequency control ancillaryservices. Traditionally these have been provided by slow responding fossil fuelgeneratorssuchasgasturbinesandcoalpoweredgenerators.Batterytechnologieshavefaster response times for frequency regulation services and are able to do so withgreateraccuracywhencompared to traditional services. In2011FERCorder755wasissued, which meant that the superior delivery of frequency regulation services wasfinancially compensated.Wholesalemarketswere forced to pay for the actual qualityand accuracy of the services provided. It is likely that this will make energy storagetechnologiesmorecompetitiveinthefrequencyregulationmarket.The California Public Utilities Commission has identified barriers which hinder theadoptionofbatterystoragetechnology.Theseincludethefollowing:

• Lackofcohesiveregulatoryframework• Lackofcosttransparency• Lackofcommercialoperatingexperience

150IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.32.151ibid.152AECOM.(2015).EnergyStorageStudy.AustralianRenewableEnergyAgency.p.34153IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.32.

JackEllisMOI

54

• Lackofdefinitiveoperationalneeds154The lack of operating experience will improve over time as will the definitiveoperationalneeds.Regardingcosttransparency,FERCorder784wasissuedinordertoclarify accounting and reporting rules for energy storage in ancillary markets. Thusalthoughbarriers still exist, there is clear evidence that they arebeing addressedandare likelytodecrease inthe future,makingtheadoptionofbatterystoragetechnologyincreasinglyviable.On the east coast the state of New York has moved assertively to promote batterystorage.On the26thofFebruary2015, theNewYorkStatePublic ServiceCommissionissued an order adopting a regulatory framework and implementation plan for‘ReformingtheEnergyVision’(REV).Theaimis toencourageandrewardparticipants(utilities/customers/servicecompanies)foractivitiesthatcontributetogridefficiency.Theplanhasastrongemphasisinreducingbothpeakdemandandcapitalinvestmentinnetwork infrastructure.155The New York Battery and Energy Storage TechnologyConsortium has been created and is introducing incentiveswhich include, ‘a planned$2100/kW battery storage incentive for 50% of the project cost for summer peakdemand reduction.’156In light of this evidence, it would appear that the regulatorybarriers hindering the adoption of battery storage technology are also diminishingwithintheUnitedStates.Overallthepolicylandscapeisactiveassince2011atleast10stateshaveintroduced14bills related to energy storage, although it should be noted that not all have passed.DespitethisthefutureprospectsfortheUnitedStateslookverypositive,asaswehaveseen installationsacrossallmarket segmentsare forecasted to increase in thecomingyears.Therehasclearlybeena focusonutilityor ‘frontof themeter’ storagerecently(figure 28), but GMT research forecasts that by 2019 ‘behind the meter’ storage,(residential/commercial/governmental), will account for 45% of the overall market,and will be led in markets such as California and New York where customers areexposed to high energy and peak demand charge tariffs.157As a result of thesedevelopments, the energy storage market in the US is forecasted to reach 861MWannuallyandbevaluedat$1.5Bn.158

JapanHistorically Japan has used energy storage to reduce the demand variability from itsnucleargenerators. Japanboasts thecompanyNipponGaishiKaisha(NGK),which isaworld leader in sodium-sulphur batteries. However, the demand for this type isbecoming overshadowed by the growth in demand for lithium-ion technology.159As a154AECOM.(2015).EnergyStorageStudy.AustralianRenewableEnergyAgency.p.34155ibid.35.156IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.33.157BLUME,S(ed).(2015)GlobalEnergyStorageMarketOverview&RegionalReportSummary.EnergyStorageCouncil.[Online]Accessedon10/4/2016.Availablefrom:http://neca.asn.au/sites/default/files//media/state_nsw/News%20&%20Views/ESC%20Global%20Energy%20Storage%20Report_2015.pdf158ibid.159AECOM.(2015).EnergyStorageStudy.AustralianRenewableEnergyAgency.p.42.

JackEllisMOI

55

resultofthe2011FukushimaearthquakeJapanlost60GWofnuclearcapacityandhasturnedtorenewableenergyinordertobecomelessdependentonimportedfossilfuels.The environmentminister has even been quoted saying 30% of the country’s energyshouldcomefromrenewablesourcesby2030.160TheincreasedemphasisonrenewableenergyhasbeenincentivizedthroughFITs,whichpayahighfixedpriceover10or20years.ForsolarPV installationsof<10kWthe incentive iscurrently$0.37/kWh,whilethehouseholdretailelectricitypriceis$0.21/kWh.ThishasgivenrisetorapidgrowthinsolarPV,whichattheendof2014stoodatmorethan10GW.161Regardingbatterystorage,itsimplementationatahouseholdlevelhasfollowedtheriseofrenewablesinspiteofthehighFITs.AlthoughJapanrecognizesthatstoragecanleadto higher levels of renewable penetration, analysis by IRENA suggests that the theincreasing implementation of battery storage is, ‘fuelled by a desire for security ofelectricity supplygiven therecentnuclear shutdown.Othermotivating factors includegovernment subsidies and the avoidance of high retail electricity rates by increasingsolarselfconsumption.’162The subsidy program mentioned supports the installation of stationary lithium-ionbatteriesbyindividualsandbusinesses.Itsupportsuptotwo-thirdsofthecostpaymentsystem and is paid for by theMinistry of Economy, Trade and Industry (METI) withbudget of $98.3m.163According to BNEF, the number of applications received hadalready exceeded the allocation of the budget before the end of 2014.164The subsidyprogramhasclearlybeenasuccessintermsofhomeownerandbusinessparticipation.Rather than encouraging implementation, perhaps a greater challenge for Japanregarding batteries will be producing them. The Japanese government is aiming forJapanesecompaniestoachievehalfoftheworld’sbatterystoragemarketshareby2020.Thisistobeachievedinanattempttoseehowmassproductionwillaidself-sufficiencyforthecountrybydrivingthecostofbatterieslower.165TheMETIannounceditspolicyon battery storage in 2012. If it is to be successful in realizing its objective, then itcalculatesthattheJapanesemanufacturerswillmovefromamarketvalueof lessthanonetrillionyenin2011totentrillionyenin2020.166Itremainstobeseenwhetherornotthiswillbeachieved.However,whatiscertainisthatJapan,liketheUS,hasaveryactivepolicyapproachtoenergystorage.Itseekstoimproveuponbothresidentialandindustrialpolicies,createfuturemarketsandtargetsglobalblockstotheadaptationofstoragetechnology.167

160ibid.43.161IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.34.162ibid.163AECOM.(2015).EnergyStorageStudy.AustralianRenewableEnergyAgency.p.42.Paymentsarecappedatapproximately$10,000forindividualsand$980,000forbusinessesinstallingbatterysystemswithacapacityof1kWhormore.164IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.34.165AECOM.(2015).EnergyStorageStudy.AustralianRenewableEnergyAgency.p.42.166IRENA.(2014).PoliciesandRegulationsforEnergyStorageinJapan.InstituteofEnergyEconomicsJapan.[Online]Accessedon10/4/2016.Availablefrom:http://www.irena.org/DocumentDownloads/events/2014/March/6_Tomita.pdf167BLUME,S(ed).(2015)GlobalEnergyStorageMarketOverview&RegionalReportSummary.EnergyStorageCouncil.[Online]Accessedon10/4/2016.Availablefrom:

JackEllisMOI

56

GermanyGermanyisaworldleaderinrenewableenergyimplementationandhastheambitioustargetsofgenerating50%ofitselectricityfromrenewablesourcesby2030,168and80%by2050.169Energystorageisviewedasanimportantenablingtechnologyinachievingthesegoals.Inresponsetothe2011nuclearmeltdowninJapan,Germanyplanstophaseoutitsremainingnuclearpowerplantsby2023.Thiswillhaveprofoundconsequencesfortheenergyinfrastructureofthecountry.Asincreasingamountsofrenewableenergyare integratedonto thegridmajor investmentwillbeneeded in the transmissiongridandstoragesolutions.This isemphasizedby the fact that thegeographicalmakeupofthecountrymakespumpedhydrostorageverydifficult.170Storagetechnologiesaswellas smart grids will play crucial roles in integrating variable renewable energy andensuring grid stability. To this end the German FederalMinistry for Economic AffairsandEnergyand theGermanyFederalMinistryofEducationandResearchare fundingapproximately200projectsonenergystoragewithatotalof€202m.171Despite there being utility-scale battery storage applications, the current driver is athouseholdlevelimplementedwithsolarPV.Germanyisunderstoodtobeaworldleaderinthisareaandin2013thecountryhadthemostsolarPVinstalledbothonatotalandper capita basis.172Around 20% of new solar PV installations now include batterypacks,173andcurrenteconomictrendsarelikelytogiverisetogreaterimplementationofsolarPVwithbatterystorage.Thereareseveralreasonsforthis:Firstly,retailratesareabove€0.30/kWhandhavebeenrising.Secondly, integratingstorage ispresentedwith an opportunity cost, as FITs for solar generation are falling. And finally, batterycostsaredeclining.Thetrendslistedaboveareacceleratedbysubsidiesgiventobatteries.FromMay2013theGermangovernmenthasprovidedagrantof30%thebatterycostaswellasalow-interest loan program for solar PV ownerswishing to add under 30kW of storage totheirsystem.174TheoverallaimistoencouragetheadoptionofbatterystoragewithPVsystems. It has been successful as by October 2014, around 6,500 battery-solar PVsystemshadbeeninstalledasaresultofthesubsidy.175Inadditiontothisapproximately4,000 systemshavebeen installedwithout government subsidy,whichmeans that bytheendof2014around12%ofsolarPVsystemswereinstalledwithabatterysystem.176http://neca.asn.au/sites/default/files//media/state_nsw/News%20&%20Views/ESC%20Global%20Energy%20Storage%20Report_2015.pdf168ibid.169IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.35.170BLUME,S(ed).(2015)GlobalEnergyStorageMarketOverview&RegionalReportSummary.EnergyStorageCouncil.[Online]Accessedon10/4/2016.Availablefrom:http://neca.asn.au/sites/default/files//media/state_nsw/News%20&%20Views/ESC%20Global%20Energy%20Storage%20Report_2015.pdf171ibid.172IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.35.173BLUME,S(ed).(2015)GlobalEnergyStorageMarketOverview&RegionalReportSummary.EnergyStorageCouncil.[Online]Accessedon10/4/2016.Availablefrom:http://neca.asn.au/sites/default/files//media/state_nsw/News%20&%20Views/ESC%20Global%20Energy%20Storage%20Report_2015.pdf174ibid.175IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.36.176ibid.

JackEllisMOI

57

There is undoubtedly strong growth in integrated battery-solar PV systems forhomeowners in Germany. It is logical that that this trendwill only increase in futureyears as battery storage becomes increasingly financially attractive due to the trendsoutlinedabove.According toDeutscheBank, theGermanmarket forelectrical storagedevicesisexpectedto ‘at leastdoublebetween2012and2025.’177LargelyduetosolarPVowners and theopportunities they arepresentedwith, battery storage is set for abrightfutureinGermany.

ChinaChina is theworlds largestproducerandconsumerof energyandandalsoboasts themost installed generational capacity. Consequently, the country plays a key role in allglobalenergymarkets.Thepowersectorofthecountryhastraditionallyfocusedoncoalandother fossil fuels,but inrecentyears thishasshifted.Growingelectricitydemand,concerns over security of supply, energy source diversification and environmentalconcerns are all contributing factors. IRENA notes that China’s renewable energycapacityhasgrownexponentiallyinrecentyears,andhassomeofthelargestwindandsolar farms in theworld.178On the otherhand, the integrationof renewables has alsobeenobstructedbyalackoftransmissioninfrastructure.Forexample,in2012thetotalinstalledwind capacitywas 75GW, but only 61GW could be utilized.179Other sourcessuggestthatsomewindfarmsloseupto40%oftheirgeneration.180Duetothis,thereisanewpolicydirectiveinChinathatrequiresintermittentgeneratorstobepairedwithenergystorageasagridconnectionrequirement.181Thepolicyissaidtoapplyatall levels fromrooftopsolar toutilitywind.The implicationsof thispolicy,especially given the rise in renewable generation, is described as “staggering” in onereport.182State Grid, the government owned transmission and distribution company,has recommended that the storage systems are sized at 14% of the installedgenerationalcapacity.Inlightofthis,thebatterymarketinChinaappearsinitsinfancy.Research forecasts an $8.5bn energy storagemarket by 2025. Many current projectsfocus on EV applications which are forecasted to play a large role in the increasingdemand.Transportapplicationsareexpected to takean85%shareof therevenuesofthequotedfigure.183$1.3bnislikelytobemadefromstationaryapplications.184Thiswillberealizedthroughintegratinghigherlevelsofrenewableenergywithfirmsfocusedon

177BLUME,S(ed).(2015)GlobalEnergyStorageMarketOverview&RegionalReportSummary.EnergyStorageCouncil.[Online]Accessedon10/4/2016.Availablefrom:http://neca.asn.au/sites/default/files//media/state_nsw/News%20&%20Views/ESC%20Global%20Energy%20Storage%20Report_2015.pdf178IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook.p.36.179ibid.180AECOM.(2015).EnergyStorageStudy.AustralianRenewableEnergyAgency.p.41181ibid.182ibid.183BLUME,S(ed).(2015)GlobalEnergyStorageMarketOverview&RegionalReportSummary.EnergyStorageCouncil.[Online]Accessedon10/4/2016.Availablefrom:http://neca.asn.au/sites/default/files//media/state_nsw/News%20&%20Views/ESC%20Global%20Energy%20Storage%20Report_2015.pdf184ibid.

JackEllisMOI

58

grid-scaleenergystorage.However,inlightoftheoutlinedpolicydirectiveandgrowthinrenewablegeneration,thisfigurecouldlikelybehigher.Renewableintegration(27%)andEV’s(13%)areexpectedtomakeupthesecondandthirdlargestapplications,whileuser-side applications are expected to account for 50% of China’s energy storagemarket.185It is alsonoteworthy that74% theelectrical storagemarket inChinauses lithium-iontechnology.186This high percentage is understandable as it is the typemost suited toEVs. It is also in clear contrast to Germany, where many of the batteries used forintegratedsolarPV-batterysystemsarebothadifferenttype(leadacid),andadifferentapplication (residential). That said, lead-acid batteries have proven to be a popularchoiceforsolarintegrationprojectsonislandsthatwouldotherwisebeforcedtorelyonimporteddieselforpowergeneration.187China’senergystorage industryhasbeengrowingata rapidpace.By theendof2014Chinahad84.4MWof installed capacityon thegrid (not includingpumpedhydroandthermal storage). This was an increase of 31MW and a growth rate of 58%, aconsiderable increase from 14% in 2013.188In addition to being the world’s largestproducerandconsumerofenergy,Chinaisquitepossiblythemostambitious:Itplanstoincrease its wind capacity from 100GW in 2015 to 200GW by 2020. Solar is to beincreased from 35GW to 50GW in the same time period.189The combination of thesetargets, backed up by a series ofwide ranging reforms that started being released in2015, suggest that the energy storage industry is set for further growth in the future.Thereformsareacombinationof13policies,thelastofwhichwillbereleasedin2016.The Chinese Energy Storage Alliance (CNESA) is able to provide information on theenergysectorreformsandhowtheywilladdresstheexistingbarriers.Theseinclude:

• Ancillaryservicecompensationtoincrease• Peakshiftingcompensationtoincrease• Compensationforpairingofwindfarmswithcoalfiredgenerationtoincrease• Demand-sidemanagementtoincrease(withencouragement)• Distributedgenerationisbeingencouraged190

Withtheintroductionofsomanyreformsitwilltaketimetounderstandfullytheeffectsthey have and the success they enjoy in overcoming barriers. Overall there is anincreasinglevelofflexibilityintheenergysector.Newpoliciesarefavouringdistributedgeneration and encouraging private investment, which will open the market toincreased levels of competition andnew sources of capital.Moreover, supply-demandreflexivepricingistobeintroducedby2020.Aspricesbecomemoredynamic,(beingsetbymarket forces, rather thandrivendrivenbypolicy), therewillbegreater flexibilityandincreasedresponsivenessfromthemarket.

185ibid.186ibid.187ibid.188ibid.189ibid.190ibid.

JackEllisMOI

59

SummaryInvestigationintothemarketsofkeycountrieshasconfirmedthefindingsoftheAECOMstudy,thathighpenetrationsofrooftopsolarandhighpenetrationsofutilityscalewindand solar, influence investment response in energy storage. Of the limited number ofcountries included in this investigationGermany and theUS appear to have themostadvanced energy storage programs, with the investment focus spread across allapplications.Ingeneral,thecountriesshareanumberofthemes.GermanyandChinaareseeking to improve their transmission infrastructure and all of the countries haveprovided incentives for those seeking to utilize battery storage. In Germany thedeclining FITs for renewable energy is a factor that will likely contribute to greaterimplementation of energy storage. As support for renewables declines, theimplementationofbatterystorage is likely to increase.As the firstpartof thechaptershows,storagesystemsaremoreprofitablewhereretailpricesarehigh.Regulatorychangesareanothercommonthemeandaremadeinanefforttocounterakey barrier to storage implementation – the lack of a cohesive regulatory framework.ThemostextensivereformsareinChina,butthosetakingplaceintheUSareextremelynoteworthy as there is an increasing trend for services being provided by batterystorage to receivecompensationbasedon theirqualityandaccuracy.Asnoted, this islikelytomakecertainmarketsrelatedtoenergystoragemorecompetitive.Creatinganorganised and interrelated framework, where compensation is available for the widerange of applications that energy storage can provide, while increasing competition,would logically have the effect of driving storage implementation. Not only wouldbattery storage be able to seek compensation by providing services through anincreasing number of applications, but rather a stronger regulatory frameworkmightinspiregreaterconfidenceinfurtherinvestment.Aworking paper published by the European Commission identifies technological andregulatoryissuesaskeychallengesforstorage.Inadditiontothese,astrategicchallengeis noted: The Commission recognises the need to develop a, ‘systemic or holisticapproachtostorage,bridgingtechnical,regulatory,marketandpoliticalaspects.’191Thedevelopmentofsuchanapproachwillbeofparamountimportancetoenergystorageinthefuture.AstheCommissionconcludes,‘themainchallengeforenergydevelopmentiseconomic.’192Creatingthenecessaryenvironmentforinvestmentiscrucial.Thechapterhasshownthatenergystorageand itsassociatedmarketswillgrowforanumber of reasons. Residential solar PV-battery systems have been shown to beeconomically viable under a number of different scenarios. For the sake of thisinvestigation, it should be highlighted that the optimal size of these systems is set toincreasewhichwillcausean increasingnumberofhomeownerstobecomenetenergyproducers.Duetothefallingcostoflithium-ionbatteries,utilitiesandhomeownerswillbe able to integrate higher percentages of renewable energymore efficiently. Energystoragewillalsogivebusinessesandindividualsmorecontrolovertheirenergysystem.

191EUROPEANCOMMISSION.WorkingPaper:ThefutureroleandchallengesofEnergyStorage.192ibid.

JackEllisMOI

60

The use of lithium-ion technology in electric cars is a further reasonwhy the energystorage market is set to grow. China is forecasted to earn $7.4bn in revenues fromtransportapplicationsby2025,andhasatargetof10mEVsdeployedby2020.193IntheUS on the 31st of March 2016, the company Tesla unveiled its Model 3 car. This istechnicallythefourthEVthecompanyhasreleased,butrepresentsthecompany’sfirstmassmarketEV.Writingontheirblogonthe7thofApril,theTeslateamsaidthattheyhad received 325,000 reservations. This corresponds to, ‘about $14 billion in impliedfuturesales,makingthisthesinglebiggestone-weeklaunchofanyproductever.’194Bymeansofcomparison,Apple’slargestlaunchachieved$8.5bninsaleswhentheiPhone6S sold 13m units on its opening weekend.195(A device that also uses a lithium-ionbattery).Thosecustomerswhodeposited therequired$1,000areable toberefundeduntil the final saleof thecar,withdeliveriesdueat theendof2017.The fact that thecompany did not advertise theModel 3 only serves to highlight the public desire forsustainabletransportation.Thus, although other types of battery are used, lithium-ion appears the technology ofchoice in most markets. As shown in chapter 1, the technology enjoys advantagesregardingefficiencyand lifecycle. It has also shown tobe cost-competitive andenjoys‘perceivedsafety’accordingtoonereport.196Thesafetyofthetechnology,aswellasitsenvironmentalimpactsareundoubtedlyimportantandshallbeinvestigatedaccordinglyinthefinalchapter.

193BLUME,S(ed).(2015)GlobalEnergyStorageMarketOverview&RegionalReportSummary.EnergyStorageCouncil.[Online]Accessedon10/4/2016.Availablefrom:http://neca.asn.au/sites/default/files//media/state_nsw/News%20&%20Views/ESC%20Global%20Energy%20Storage%20Report_2015.pdf194TESLA.(2016).Theweektheelectricvehicleswentmainstream.[Online]Accessedon13/4/2016.Availablefrom:https://www.teslamotors.com/blog/the-week-electric-vehicles-went-mainstream?utm_campaign=Blog_Model3_040716&utm_source=Twitter&utm_medium=social195CNNMONEY.(2016).TeslaModel3’sfirstweek:325,000orders.[Online]Accessedon13/4/2016.Availablefrom:http://money.cnn.com/2016/04/07/autos/tesla-model-3-orders/196AECOM.(2015).EnergyStorageStudy.AustralianRenewableEnergyAgency.p.54.

JackEllisMOI

61

Chapter3:TheimpactofbatterystoragetechnologyinperspectivesChapters 1 and 2 have provided insight into both technical and economic aspects ofbatterystoragetechnology.Thefinalchapterofthisinvestigationseekstogaingreaterunderstandingofbatterystorageandevaluatetheimpactthatitsincreasingdeploymentmay have. This is to be achieved by examining it through a series of differentperspectivesandaskswhattheimplicationsofincreasedbatterystorageusewillbe.

EnvironmentalperspectiveThisreporthasmostlyfocusedonsolarPVastherenewablesourceofenergyintegratedwith battery storage.When compared to other sources of energy such as coal or gaspoweredplants,solarPVisunderstoodtogenerateconsiderablylessemissionsoveritslife-cycle.197Regarding battery storage, it is a sensible assumption that the increasingprofitability of storage systemswill result in their increasing deployment. This raisesenvironmentalconcernsoverthelevelsofpollutionthatsuchdiffusionmayhave.Ashasbeen shown there are various different types of battery. The different types all havedifferent environmental impacts. As this investigation has focused on lithium-ion andleadacidbatteries,itisimportantthattheimpactoftheirproduction,useanddisposalisunderstood.

ProductionTheproductionoflithium-ionbatteriesisunderstoodtohavesevereimpacts.Normallylithiumisfoundinthebrineofsaltflats.Brineispumpedtothesurfaceandevaporatesinponds,allowinglithiumcarbonatetobeextractedthroughachemicalprocess.Thereare other lithium deposits, such as sedimentary and igneous rock, but as McManusnotes,‘muchofthegloballithiumissuppliedthroughbrinedepositsasitistheclosesttothe surface.’198These salt flats are located in arid territories, with the largest minesbeing in theAtacamadesert in SouthAmerica.Mines also exist in China andTibet.199RegardingtheselocationsZacunewritesthefollowing:‘Intheseplaces,access towater iskey forthe localcommunitiesandtheir livelihoods,aswell as the local flora and fauna. In Chile’s Atacama salt flats, mining consumes,contaminates and diverts scarce water resources away from local communities. Theextractionof lithiumhascausedwater-relatedconflictswithdifferentcommunities, suchasthecommunityofToconao inthenorthofChile. InArgentina’sSaladeHobreMuerto,

197HOPPMANJ.,VOLLAND,J.,SCHMIDT,T.S.,HOFFMANN,V.H.(2014).TheEconomicViabilityofBatteryStorageforResidentialSolarPhotovoltaicSystems–AReviewandSimulationModel,RenewableandSustainableEnergyReviews39,1101-1118.p.25.198McMANUS.(2012).Environmentalconsequencesoftheuseofbatteriesinlowcarbonsystems:Theimpactofbatteryproduction.AppliedEnergy93p.289.199ibid.

JackEllisMOI

62

local communities claim that lithium operations have contaminated streams used forhumans,livestockandcropirrigation.’200Thus, largely due to water depletion and pollution, the lithium extraction process isunderstoodtohavesignificantenvironmentalandsocial impacts.This issupportedbythe work of McManus, who notes that lithium batteries have an impact on humantoxicityandconcedesthatthe,‘miningoflithiumcancausesignificanthumanhealthandsocial impacts.’201However, these environmental impacts are not only confined tolithium-ionbatteries.Leadacidbatteriescontainsulphuricacidinadditiontotoxiclead,whichgeneratescarbonemissionsduringtheminingprocess.Althoughasweshallsee,the environmental impact of lead-acid battery types can be reduced when correctlyrecycled.In terms of which batteries are the most energy intensive to produce, here againlithium-ionfairspoorlyastable5showsbelow.

Table5:Characterisedimpactperkgofbatteryproduction202

The table shows that the lithium-ion batteries and nickelmetal hydride are themostenergy intensive batteries to produce. Lead acid and sodium sulphur are the leastenergydemanding.Intermsofmetaldepletion,thelithiumvarietiesarebyfarthemostdemanding. They are also the most demanding technology, along with nickel metalhydride,forfossilfueldepletion.Inspiteof this,comparing the impactof theproductionofbatteriesbyweightmaybeunfair.203As has been shown in chapter 1, some batteries have superior performanceper unit of weight during their life-cycle, as energy density differs between differenttypes.Becauseofthedifferencesinlife-spanandcharge/dischargecyclesthatdifferentbatteries enjoy, McManus believes that to understand the true impact of production,batteriesmustbeexaminedonanenergybasis.Table6belowshowsacomparisonofbatteriesonaperenergybasis.

200ZACUNE,J,(2013).Lithium.FriendsofEarthEurope.[Online]Accessedon18/4/2016.Availablefrom:https://www.foeeurope.org/sites/default/files/publications/13_factsheet-lithium-gb.pdf201McMANUS.(2012).Environmentalconsequencesoftheuseofbatteriesinlowcarbonsystems:Theimpactofbatteryproduction.AppliedEnergy93p.289.202ibid.203ibid.

JackEllisMOI

63

Table6:CharacteriseddatarangeforbatteryproductionperMJcapacity204

Under the impact categories of climate change (greenhouse gas emissions) andmetaldepletion, lithium-ion is the worst performer. This is true both per kg of batteryproduction(table5),andperenergycapacity (table6).Leadacidandsodiumsulphurare thebest performers. Thus, despite the advantages that lithium-ionhas over otherbatterytypesintermsofenergydensityandpower,ithascleardisadvantagestoo.Thethemeofresourcedepletionisonethat isverysignificantwithlithium-iontechnology,and as demand for lithium increases it will become increasingly so. The demand forlithiumhas increaseddramatically in thepastand this trend looksset to continue. Itsuseinrechargeablebatterieshasincreasedfrom0%ofthemarketsharein1991to80%in2007.205More relevant for this investigation is the increasinguseof lithium inEVs.This has lead to concerns amongmanufactures that consumer demandmay overtakesupplyby2020,andmanycompanieshavetakenmeasuresinordertosecureaccesstolithium deposits206Despite their concerns, McManus estimates that there is enoughlithium tomeet the demand this century.207Research carried out byNavigant doesn’tcontradict this, although it does raise some interesting points: In particular theirresearch indicated that by around 2020 the annual market will cross the 150GWhthreshold. The implications of this being that pricingwill begin to be affected by thedemand for batteries.208Other necessary materials such as cobalt also face scarcitychallenges.Navigantdescribesthesourcingoftheelementas‘veryveryconstrained’asitcomesfromconflictzonesandthushashumanrightsimplicationsinitsuse.209Inlightoftheabovetheenvironmentalcostsofsourcingthematerialstomeetthisdemandwillbeconsiderable.Althoughtheymaynotappearsignificantfactorstoday,astheindustrygrows and demand formaterialsmultiplies it is likely that theywill only increase intheirseverity.

UseTheuseoflithium-ionbatteriesbringswithittwokeyhazards.Thesecanbeclassifiedasthefollowing:

204ibid.205ZACUNE,J,(2013).Lithium.FriendsofEarthEurope.[Online]Accessedon18/4/2016.Availablefrom:https://www.foeeurope.org/sites/default/files/publications/13_factsheet-lithium-gb.pdf206ibid.207McMANUS.(2012).Environmentalconsequencesoftheuseofbatteriesinlowcarbonsystems:Theimpactofbatteryproduction.AppliedEnergy93p.289.208NAVIGANTRESEARCH.(2014).WebinarReplay–BeyondLithiumIon.[Online]Accessedon19/4/2016.Availablefrom:https://www.navigantresearch.com/webinarvideos/webinar-replay-beyond-lithium-ion209ibid.

JackEllisMOI

64

• Chemical – This refers to the accidental exposure of chemicals within the

battery.This couldbe a spillageor a gas emission. Inboth cases thehazard islinkedtothecorrosiveandflammablepropertiesofthechemicals.

• Electrical – The current flowwithin a battery creates heat through the Jouleeffect.210It is important that theheatgeneratedbycharginganddischarging iscontrolledbya thermalmanagement system.Protectionagainsthighelectricalcurrents and short circuits are also important. In addition to the current flow,thestateofchargemustalsobecontrolled.Thetemperatureofthebatterycanincrease through overcharge and over-discharge as the reactions can bemoreexothermicthannormal.211

The potential cumulative effects of the hazards outlined above is known as “thermalrun-away.” This occurs when the temperature increases, (usually in a kind ofuncontrolledpositivefeedback),toapointthatisdestructivetothebattery.ARechargereportonthesafetyoflithium-ionbatteriesgivesanexample:

‘In caseof a short circuit, the Joule effectwill increase the cell temperature to thepointwheretheorganicsolventleavesthecellviathevent.Atthistimeanyhotspotmayinducea fire.Thepossible consequencesof this cumulative effectare…fire, toxic orharmful gasemissionandtheejectionofparts’212Itisclearthattheuseoflithium-iontechnologyisnotwithoutitsrisks.Thebatteryitselfisanarticlewithnointendedreleaseofitssubstances.213Ifthecontentsarereleasedinanaccidentthenanemergencyresponsewillbeprompted.Itisnoteworthythoughthattheyareequippedwithelectronicprotections inordertoavoidtheirmisuse, thattheyareshockresistant,andcanbesafelyoperatedinawidetemperaturerange.Ultimatelya trulyvastamountofelectronicequipment ispoweredby thesebatteries throughouttheworldeveryday.Thisconfirmsthefactthatthesafetyoflithium-ionbatteriesiswellmanaged, which has lead the technology to be regarded as one of the safest energystoragesystems.214

DisposalIn September 2006 the European Parliament issued directive 2006/66/EC – OnBatteries and Accumulators and Waste Batteries and Accumulators. It is commonlyknown as “the battery directive”, and regulates the manufacture and disposal ofbatteries within the European Union. It has the aim of ‘improving environmentalperformanceof batteries and accumulators.’215According toMcManus, the regulations

210RECHARGE–THEEUROPEANASSOCIATIONFORADVANCEDRECHARGEABLEBATTERIES.(2013).Safetyoflithium-ionbatteries.Belgium.211ibid.212ibid.213RECHARGEBATTERIES.ORG.(2013).Batteryinformationfactsheet:Lithium-ionbatteries.214RECHARGE–THEEUROPEANASSOCIATIONFORADVANCEDRECHARGEABLEBATTERIES.(2013).Safetyoflithium-ionbatteries.Belgium.215DIRECTIVE2006/66/ECOFTHEEUROPEANPARLIAMENTANDOFTHECOUNCILof6September2006onbatteriesandaccumulatorsandwastebatteriesandaccumulatorsandrepealingDirective91/157/EECp.5.

JackEllisMOI

65

‘form part of the producer responsibility suite of regulations and requires batteryproducerstotakecareoftheirwaste.’216HenotesthatproducersthatplacemorethanonetonneofportablebatteriesintheUKmarketarerequiredtopayforthecollection,treatment,recyclinganddisposalofwastebatteriesinproportiontotheirmarketshare.Thisisnormallydonethroughacomplianceschemewhichalsoregisterstheproducerswith the appropriate environmental agency.217It should also be mentioned thatproducers placing less than one tonne are still required to register with anenvironmentalagency.The legislation is expected to increase thenumberofbatteries recycled,meaning thatthe impact of resource depletion should be reduced. Gaines writes that in an idealsystem, lithium-ion batteries would be sent for responsible recycling and not beexported todeveloping countrieswith less stringent environmental, health and safetyregulations.218It is important to stress the different ways in which recycling isimportantfortheenvironment.Notonlyisresourcedepletionmitigated,butasGainespointsout if theyarerecycledthentheycannotbetransportedtocountrieswhichareunlikely to have the necessary recycling infrastructure to depose of the batteriescorrectly. In such countries the battery is likely to end up aswastewith both humantoxicity potential and ecotoxicity potential.219She also points out that the recycledproductneedstobeofhighenoughqualitytofindamarketforitsoriginalpurpose,oritmustfindanalternativemarket.220Althoughlithiumisafiniteresource,itisunlikelytorunoutinthenearfutureduetotheuseofbatteries.221Rather,as itsmininghasbeenshowntohaveahighenvironmentalcost, appropriate measures should be taken regarding its recycling process and therecoveryofmaterials.Ifthebatteriesenduponalandfillsite,thenthematerialswithinthebatteryarenotabletoberecovered.Theresultofthisistheincreasedextractionofraw materials, which uses large amounts of energy and creates emissions that areharmfultotheenvironment. Inaddition, ifbatteriesendupona landfillsitethentheymayleachtoxicchemicalsintothesoil.Therepairandre-useofbatteriesisnotpossible.Recyclingistheonlywayofrecoveringthevaluablematerials.Kumarstatesthatthebatteryrecyclingmarketislargelypricedriven.222Itisimportanttorememberthatrecyclingcompaniesarebusinessesandaimtomaximiseprofitfromthesaleofrecoveredmaterials.Inlithium-ionbatteriesthemostvaluablematerialsare

216McMANUS.(2012).Environmentalconsequencesoftheuseofbatteriesinlowcarbonsystems:Theimpactofbatteryproduction.AppliedEnergy93p.289.217ibid.218GAINES,L.(2014)Thefutureofautomotivelithium-ionbatteryrecycling:Chartingasustainablecourse.SustainableMaterialsandTechnologies.1-22-7.219BOYEN,ANNA.(2014)Theenvironmentalimpactsofrecyclingportablelithium-ionbatteries.AustralianNationalUniversity.220ibid.221McMANUS.(2012).Environmentalconsequencesoftheuseofbatteriesinlowcarbonsystems:Theimpactofbatteryproduction.AppliedEnergy93p.289.222WASTEMANAGEMENTWORLD.(2011).TheLithiumBatteryRecyclingChallenge.[Online]Accessedon19/4/2016.Availablefrom:https://waste-management-world.com/a/1-the-lithium-battery-recycling-challenge

JackEllisMOI

66

cobalt, nickel and copper.223The aforementioned battery directive of the EuropeanCommission states that the recycling processes must achieve the minimum recyclingefficiency of, ‘50% by average weight of other waste batteries and accumulators.’224Lithium-ion is within this category. Clearly there is no obligation to recycle certainmaterialsandasaresultthemostvaluablematerialsarerecovered.However,basedonatypicallithium-ionbatterytherecoveryofcobalt,nickelandcopperwouldachieveanefficiencyofabout30%.225ThereforeinordertomeetthetargetefficiencyasstatedbytheEuropeanCommission,morematerialsarerequiredtoberecovered. Ifaluminium,steel, lithiumandmanganesewererecoveredtoo, the targetwouldbeachievedas thematerialsconstituteabout57%ofthebattery’scomposition.226Recycledlithiumitselfisupto5timesmoreexpensivethanobtainedbythecheapestprocess.227Thiscurrentlymakesitsrecoverynoteconomicallyviable.AstheUShasnobatterydirectiveandthetheprocessisentirelycostdriven,itcouldbethecasethatonlyafractionoflithiumisrecycledinthecomingyears.Regarding Lead-acid batteries, they are said to be recycled more than any otherconsumerproduct,andintheUSapproximately99%arerecycled.228Whenrecyclingtheleadtheenvironmentalimpactofthebatteriesissignificantlyreduced.Furthermore,therecyclingoperationitselfisprofitableastherecycledleadisofhighquality.Gainesnotesthat because of this there is, ‘little incentive to export to places with less-stringentregulations.’229Onemajoradvantageofrecyclinglead-acidbatteriesisthattheprocessis relatively simple and the materials homogeneous. Despite this, Hopperman writesthat the use of lead-acid batteries ismore problematic in countrieswhich do not yetpossess a, ‘working recycling infrastructure.’ 230 Having a functioning recyclinginfrastructure is of paramount importance in limiting the environmental effects ofbatteries.Aslithiumdemandissettogrowinthefutureandresourcesincreasinglyfinite,itwouldsuggest that recovering the materials from used batteries will become increasinglyeconomically viable. It is not only the environmental impact of mining the requiredmaterialsandtheirimproperdisposalwhicharereasonsforrecycling.Rather,thepricedriven recyclingmarketmay find increasing returns in recovering valuablematerials.Despite this the current trend in lithium-ion technology is towards cathodematerialsthatdonotcontaincobalt.231Thisthreatenstheprofitabilityofrecycling.Itmaybethe

223BOYEN,ANNA.(2014)Theenvironmentalimpactsofrecyclingportablelithium-ionbatteries.AustralianNationalUniversity.224DIRECTIVE2006/66/ECOFTHEEUROPEANPARLIAMENTANDOFTHECOUNCILof6September2006onbatteriesandaccumulatorsandwastebatteriesandaccumulatorsandrepealingDirective91/157/EECp.27.225BOYEN,ANNA.(2014)Theenvironmentalimpactsofrecyclingportablelithium-ionbatteries.AustralianNationalUniversity.226ibid.227ibid.228ibid.229ibid.230HOPPMANJ.,VOLLAND,J.,SCHMIDT,T.S.,HOFFMANN,V.H..(2014).TheEconomicViabilityofBatteryStorageforResidentialSolarPhotovoltaicSystems–AReviewandSimulationModel,RenewableandSustainableEnergyReviews39,1101-1118.p.25.231BOYEN,ANNA.(2014)Theenvironmentalimpactsofrecyclingportablelithium-ionbatteries.AustralianNationalUniversity.

JackEllisMOI

67

case that the threat is greater outside Europe where there are no laws targetingrecyclingefficiencies.Thiswouldsuggestthatlegislationsimilartothebatterydirectivemayberequired.Acostcomparisonofrecyclingcompaniesshowedthatcompanieswillmost commonly buy batteries with cobalt, and may even charge a fee if they don’tcontainit.232Ingeneral,therecyclingprocessesaremostcommonlyaimedatincreasingthenumberof materials recovered, as well as improving environmental effects of emissions andenergy consumption. Regarding lithium-ion varieties, there is a trend for theseprocesses tobecome increasinglyspecialised inorder toachieve thebestresults,withdifferent batteries having a specific recycling process, each dedicated to the specificchemistry.233Although not the direct concern of this paper, investigation into howincreasedbatteryusewouldaffecttheUSrecyclingindustrywouldbeinteresting.

InternationalizationperspectiveThis investigation has defined internationalization as the process of achievinginternationalcontrolortheprocessofmakingsomethinginternational incharacter. Ineconomic terms it was described as the spread of economic activity over nationalboundaries. Following on from environmental perspectives, let us find an immediateexampleofinternationalisation.According to Kumar, the concern for reliable sources of lithium has lead to inter-governmental partnerships, and also partnerships between original equipmentmanufacturers (OEMs) and governments. For example, EV manufacturers such asMitsubishi have forgedpartnershipswith lithium-exploration companies and investedlargesumstodeveloplithiumdepositsinArgentina.Thishasbeendoneinanefforttosafeguardandsecurethelithiumresourcestofulfiltheirneeds.Furthermore,JapanhasforgedapartnershipwiththeBoliviangovernment.ThisagreementbindsJapantooffer,‘comprehensive economic aid in exchange of supplies of lithium and other raremetals.’234Developments such as the ones outlined above are a clear example ofinternationalisation. Companies and governments are seeking increasing control overthematerials required tomanufacture lithium-ionbatteries. This is done as requiringthe materials constitutes the “upstream” part of the global value chain. Increasedinternationalisation, through participation in global value chains, provides theopportunity to achieve economies of scale, expand market share and increaseproductivity. Moreover, participation in value streams and cooperation with otherpartnersislikelytoenhancetheflowofinformationandlearning.Itcanalsointroducenew business practices and more advanced technology which could lead to greatergrowth and revenues. The partnerships outlined serve as an example of increasing

232ibid.233ibid.234WASTEMANAGEMENTWORLD.(2011).TheLithiumBatteryRecyclingChallenge.[Online]Accessedon19/4/2016. Available from: https://waste-management-world.com/a/1-the-lithium-battery-recycling-challenge

JackEllisMOI

68

stakeholders in the lithium supply chain. As governments and enterprises seek toinvolve themselves in economic activity over national boundaries in order to gaincontrolovermaterials,internationalisationisclearlypresent.Internationalisationisalsoevidentthroughincreasedparticipationinthe“downstream”of globalvalue chains. In the recentpast therehasbeenan increase in thenumberofsolar PV suppliers entering into the US market. This is not only true of solar PVequipment but also battery systems too. To pick an example, in 2015 the Japanesecompany Tabuchi presented a 10kWh EneTelus Intelligent Battery System to the U.S.market for the first time.235It is noteworthy that the battery is designed for theresidentialstoragemarket.Regardingsolarpower,downstreamcostsareforecastedtodecrease.In‘TrendsinEnergyFinance’itwasshownthatupstreamcostshaddecreasedsignificantlyoverrecentyears.AnalysisbyMcKinsey&Companyhasshownthateventhoughmodulecostsshouldcontinuetofall,evenbiggeropportunitiesareavailableinthe downstream costs associated with installation and service. Their research showsthat half the expense of installing residential solar systems in the US is comprised ofdownstreamcosts,andthatastheybecomecheapertheoverallcoststoconsumerswillfall from $2.30 per watt in 2015 to $1.60 by 2020.236In light of this, not only isinternationalisationpresent throughout the value chains of both solar PV andbatterystorage,butitmayalsohavetheeffectofloweringcosts.Thisshallbecommenteduponinthefollowingsection.As an important part of this investigation is into the impact of battery storage oncentralisedenergy systems, it is important that thepotential consequences toutilitiesarewell understood.The following and final sectionof the investigationwill focusonbattery storage technology from both the public and utility perspective, and theirrelationshipwith eachother. Itwill investigate thedynamicbetween them, and showhow this is set to change as a result of what been presented throughout this report.Beforehand, and in order to conclude the perspective of internationalisation, let usreturnbrieflyto‘trendsinenergyfinance’.‘Trendsinenergyfinance’alsoshowedincreasinglevelsofinternationalisationastherewas an increasing involvement of enterprises, mostly Chinese, across nationalboundaries. In addition, different parts of the solar PV value chains were present indifferent countries. Solar cellsmanufactured in Chinawere commonly integrated intosystemsinEurope.Itshouldbenotedthatalthoughinternationalisationhasbeenshowntolowercosts,‘trendsinenergyfinance’showedthatthelowpricesofChinesePVcellsimpacted European PV manufacturers badly, with many going out of business. Thus,although internationalisation is present regarding Solar PV technology and batterystorage technology, the low prices that brings can negatively affect businesses.Ultimatelyhowever,increasinginternationalisationhasbeenshowntoplayakeyroleintheenergytransitiontowardsasystembasedonrenewableenergysources.

235PVMAGAZING.(2015).IntersolarNorthAmerica.Tabuchi’smarketentrypointstoincreasinginternationalisation.[Online]Accessedon19/4/2016.Availablefrom:http://www.pv-magazine.com/news/details/beitrag/intersolar-north-america--tabuchis-market-entry-points-to-increasing-internationalization_100020228/#axzz46Rokh8cm236McKINSEY&COMPANY.(2016)Thedisruptivepotentialofsolarpower

JackEllisMOI

69

Public/UtilityperspectiveTheprevioussectionshowedhowincreasing levelsof internationalisationcontributedto falling costs. The falling cost of renewable energy is of great interest to thisinvestigation.Asitbecomesincreasinglycompetitivewithtraditionalpowergenerationtechnologies, such as coal, gas and nuclear, it has the potential to disrupt the utilitysector. In2015theUSEnergyInformationAdministrationcalculatedthatsolarenergyaccounted for 0.6%of total electricity generation.Although this number is very smallMcKinsey&Companywritethefollowing:‘Thebusinessmodelforutilitiesdependsnotsomuchonthecurrentgenerationbaseasoninstallationsofnewcapacity.Solarcouldseriouslythreatenthelatterbecauseitsgrowthundermines the utilities ability to count on capturing all the new demand whichhistoricallyhasfuelledalargeshareofannualrevenuegrowth.’237Intermsoftheimpactofsolaronutilities,thefactthatthecurrentpercentagegeneratedthroughsolar is so low issomewhatmisleading.What ismorerelevant is thedemandfor new installations.McKinsey&Company suggest that new solar installations couldaccountforuptohalfofnewconsumption,andthatbyalteringthedemandsideoftheequation, solar isable toaffect theamountsof capital thatutilitiesareable todeploy.Thereasonforthisisthattheyhaveapredeterminedreturnonequity.238Thatistosaythatsolar,inspiteofitslowelectricitygenerationpercentage,hastheabilitytohaveanoversized effect on the economics of utilities, and consequently the future of theindustry.As has been shown in Chapter 2, the optimal solar PV system sizes rise over time. Inscenarioswhereexcessenergycouldbesoldonthewholesalemarkets,theincreaseinsizewasmoresignificantthanwherethiswasnotpossible.Ifsystemsizesincrease,thiswill raise the amount of energy households produce themselves. The use of batterystoragesupportsthistrendbyallowinghouseholdstoconsumealargeramountoftheselfproducedelectricity.Thishastheconsequenceofreducingtheamountofelectricitybought fromutilities. In addition, if householdsareable to sell their electricityon thewholesalemarketinthefuturethenanincreasingnumberofhouseholdswillmovefrombeing electricity consumers to electricity producers.239As noted by Hopperman et al.,‘thistrendhasthepotentialoffundamentallyalteringtheexistingmarketstructure.’240Intheiropinionthefactthatelectricutilitiesarelikelytobeconfrontedwithagrowingnumber of households producing and selling their own electricity, ‘fundamentallyunderminestheirbusinessmodel.’241It is clear then that increased deployment of solar PV-battery systems is a threat toutilities.Notonlydoesitreducetheamountofelectricityboughtfromutilities,butalso

237ibid.238ibid.239HOPPMANJ.,VOLLAND,J.,SCHMIDT,T.S.,HOFFMANN,V.H.(2014).TheEconomicViabilityofBatteryStorageforResidentialSolarPhotovoltaicSystems–AReviewandSimulationModel,RenewableandSustainableEnergyReviews39,1101-1118.p.26.240ibid.241ibid.

JackEllisMOI

70

affects the amount of capital that they are able to deploy in search of the return oninvestment that they have previously enjoyed. It can be argued that the effects arealready being felt in Europe. A Citi Group report on German utilities makes severalpertinent points. It describes E.ON as the latest utility to concede that businessconditionsare‘irreversiblychanging.’242Theelectricityutilityhassplititsbusinessintotwo,withnewbusinesscalledNewCo.Despiteitsnamethereisnothing“new”abouttheassetsofthecompany,whichincludeallpowergenerationapartfromrenewables.Thenew business includes what are regarded by Citi Group as “riskier” assets, such asnuclearpowerand thermalgeneration.243AlthoughCitiGroupregards thesplitas ‘therightthingtodo’,itstatesitsbeliefthatitwasdonefromapositionofweakness,‘giventhe drastic changes in E.ON’s operating environment and one that avoids furtherdestructionof valueas conditionsdeteriorate rather than creatingvalue fromcurrentlevels.’244Suchanalysisleaveslittledoubtoftheeffectthattheincreasingdeploymentofrenewableenergyhasonutilitycompanies.ItshouldbehighlightedonceagainthatasintegratingbatterystoragewithsolarPVincreasestheindependenceofhouseholds,itsincreaseddeploymentwillbringwithitfurthernegativeconsequencestoutilities.Fromthepublicperspective,manymembersofthepublicwillfaceachoiceinthefuture.On the one hand, they can remain a traditional customer, using the electricity grid asthey have always done. On the other hand, customers may find it increasinglyeconomicallyviabletodeflectfrombothutilitiesandtheelectricitygrid.Theymaywishtocost-effectivelyself-generate,supplyingthemselveswithpowerfromintegratedsolarPV-battery systems. Research from the RockyMountain Institute (RMI) indicates thatthe number of people thatwould actually deflect from the grid is small. Rather, theyregard a more likely scenario as one in which customers invest in solar PV-batterysystemsthatareconnectedtothegrid.Insuchascenariothesystemswouldbeabletobenefitfromthegridandcouldbeoptimallysized,thussmallerandlessexpensive.Thiswould have the knock-on effect of being economic for more customers sooner, andwouldleadtofasteradoption.The increaseddecentralisationbroughtaboutbyadoptionofsolarPV-batterysystemswill reduce theneed for transmission. Chapter2has shown that lithium-ionbatteriesare forecasted to decrease in the future. As customers installmore andmore batterycapacity due to its falling prices thiswould logically have the effect of reducing theirpurchaseofelectricityfromthegriduntilthegridtakesaback-uprole.245TheRMIstatesthefollowing:‘In Westchester County, NY, our analysis shows the grid’s contribution shrinking from100%todayforcommercialcustomersto~25%byaround2030tolessthan5%by2050.Inversely,solarPV’scontributionrisessignificantlytomakeupthedifference.’246

242CITIGROUP.(2014).GermanUtilities:Letthesurvivalgamebeginasthelostdecadetakeshold.CitiResearch.243ibid.244ibid.245ROCKYMOUNTAININSTITUTE.(2015).Theeconomicsofloaddefection:Howgrid-connectedsolar-plus-batterysystemswillcompetewithtraditionalelectricservice,whyitmatters,andpossiblepathsforward.RockyMountainInstitute.Colorado.246ibid.

JackEllisMOI

71

Such analysis supports conclusions drawn by Roland Berger (Strategy Consultants),whosuggestthatsolarPVwillchangetheutilityindustry.Theynotethatutilities’roleofcentralized production will evolve from, ‘delivering volumes to providing access toelectricitycapacity…Maintainingabalanceandafunctionalgridcapableofdealingwithmulti-directionalpower flowsandpower tradewill change thenatureof thenetworkcompanies.’247AkeyproblemidentifiedbytheRMIisthatbetween2010and2030theelectricitygridoftheUSwillrequireuptoanestimated$2trillioninvestment.Thesecosts,whichareroughly$100bnperyear,aretoberecoveredthroughenergysales.Therefore,anylossin energy sales is a loss in investment as the revenue is effectively lost. The authorswritethatthis,‘willlikelyhavealargeimpactonsystemeconomics.’248BelowareexamplesofwhattheRMIcalculatestobethemaximumpossiblekWhsaleserosioncouldbeintheNortheastUSby2030.

• Residentialo ~58millionMWhannually(50%ofutilityresidentialkWhsales)o 9.6millioncustomerso ~$15billioninrevenue

• Commercial

o ~83millionMWh(60%ofutilitycommercialkWhsales)o 1.9millioncustomerso ~$19billioninrevenue249

Clearly, the sales erosion suffered by utilities has the potential to be very large.Nevertheless, it is importanttomentiontheopportunitiesthata largenumberofgrid-connected-solar PV-battery storage systems presents. The fact that the customersmaintain a grid connection means that their systems have the potential to providebenefits and services back to the grid. This is the casewithmany of the applicationsdiscussed in Chapter 1. The chapter was able to show that different applications ofbattery storagewere able to achieve a number of different services. It included bothproduction and grid level applications as well as at residential/utility scales, andshowedhowoneinstallationcouldprovidemultipleservices.Therangeofvaluecreatingapplicationsthatbatterystorageisabletoperformprovidesis significant. It means that although there could be a substantial load loss, thecustomers’ grid connected systems can add value by providing services. As the RMIreportnotes,thiswillespeciallybethecaseifthevalueflowsaremonetizedwithnew

247ROLANDBERGERSTRATEGYCONSULTANTS.(2015).SolarPVwillchangetheutilitylandscape.RolandBergerStrategyConsultants.Paris248ROCKYMOUNTAININSTITUTE.(2015).Theeconomicsofloaddefection:Howgrid-connectedsolar-plus-batterysystemswillcompetewithtraditionalelectricservice,whyitmatters,andpossiblepathsforward.RockyMountainInstitute.Colorado.249Ibid.

JackEllisMOI

72

ratestructures,businessmodelsandregulatoryframeworks.250Belowisabreakdownofthepotentialimpactontheelectricity-systemparticipants.

Implicationsforenergy-systemmarketparticipantsandotherstakeholders

CustomersthatinvestinsolarPV-batterysystemsAnalysis by RMI suggests that with intelligent investments customers could see peakpricingemerge.Thiswouldallowcustomerstoinsulatethemselvesfromrisingpricesofgrid-supplied electricity. The traditional grid-supplied customers, as well as thosecompletelyoff-grid,weresubjecttohighercosts.Thiswasduetorisingpricesforretailcustomersandmoreexpensivestand-alonesolarPV-batterysystemsforthoseoff-grid.

DistributiongridoperatorsTheimplicationsforgridoperatorsareseenaspositive.Thisisbecausecustomerswithdistributed systems shouldbe able toprovidevalue to the grid.According to theRMIthese include, ‘upgrade deferrals, congestion relief, and ancillary services.’251Despitethis,theRMIdoesconcedethatnewpricing,regulatoryandbusinessmodeswillhavetoemergeandmatureinordertofullycapitalizeontheopportunities.252

CentralgenerationandtransmissionSolar PV-battery systems are foreseen to hasten the decline of sales from centralgeneration utilities. Furthermore, they are also predicted to reduce peak spikes inderegulatedmarketsandalsoencroachonmarkets forancillary services.253It is likelythat as an increasing number of people use solar PV-battery systems the assets willserveadiminishingload.Inorderforcoststobecoveredbythisdecreasingload,priceincreasesmayberequired. Inturn, thismayaccelerate furtherdecline insaleasmorecustomers invest in solar PV-battery systems, seeking to reduce costs. The RMI alsonotes that assets in the planning stages will not be able to see the future demandrequiredtojustifytheircapacityandgenerationaloutput.254

Vertically-integratedutilitiesAdjustments in businessmodelswill be necessary in order to capitalize on the risingadoptionofsolarPVandbatteries.Ifnoactionistaken,thentheymayfindthemselvesunderstrainfromthesystemsoutlined.

250ibid.251ibid.252ibid.253ibid.254ibid.

JackEllisMOI

73

FuturetrajectoriesIt appears that the electricity system, that is the current centralized energy system,standsatsomethingofacrossroads.TheRMIsummarisesbothpaths:‘Downonepatharepricingstructures,businessmodels,andregulatoryenvironmentsthatfavour non-exporting solar and solar-plus-battery-systems. When economic and otherconsiderationsreachtherighttippingpoint,thistrajectoryfavourstruegriddefection.Inthemeantime,anupwardpricespiralbasedonstrandedassetsservingadiminishingloadwillmakesolar-plus-batteryadoptionincreasinglyattractiveforcustomerswhocan,andlead tountenablyhighpricing for customerswhoremainon thegrid, including lowandfixed-income customers who would bear a disproportionate burden of estimated retailelectricitypricing. Inthe future,bothgridandcustomer-sideresourcesareoverbuiltandunderutilized,leavingexcesscapitalonbothsidesofthemeter’255The pricing structures mentioned in the quote above refer to policies such as theelimination of netmetering.Netmeteringwasmentionedbriefly in chapter 2 (utility,residentialandnon-residentialmarketsegments)andiswhereutilitiesarerequiredtobuyexcesselectricitygeneratedbyhomeowners’ solarpanels.As theelectricity isnotexportedinthisscenarioitisnotpurchasedbyutilities.Asthequoteaboveshows,thistrajectoryfavoursdefectingawayfromthegridastherearelimitedadvantagesofbeingconnected. Moreover, as an increasing number of people defect away from the grid,pricesriseanditbecomesmoreeconomicforagreaternumberofpeopletodefect.Thephrase‘pricespiral’isaccurateinthissenseasrisingpricescausemorepeopletodefect,and thus causehigherprices as theassets (whichhave remainedunchanged), serveadecreasingload.Increasingpricesandincreasingdefectionarelockedinaviciouscircle.Regarding net metering, the findings of the RMI were that the elimination of netmetering ‘merely delayed significant load loss.’256The conclusion drawn was that,althoughitmightbegradual,batterysystemswouldultimatelycausea ‘neartotalloadlosseveninnetmetering’sabsence.’257Analternatepolicyoffixedchargesisalsosaidtosimplydelaythe‘ultimateloaddefectionoutcome.’258Inlightoftheabove,itappearsinevitablethattherewillbeasignificantlossofrevenuesfor utilities. This is emphasised buy the fact that key policies only serve to delay theoutcome. On a different topic, it should be highlighted that, “low and fixed incomecustomers pay a disproportionate burdenof estimated electricity pricing.” It could besuggestedthatthepathofgriddefectionisonewhichdemocratisationisleastpresent.This is as lower earners are punished by high prices for not being able to defect tobatterysystems,exhibitingnovoiceintheirenergysystem.Chapter 2 also highlighted how solar PV threatens businessmodels. It was shown tothreatenthe ‘sunkutility investments incentralisedfossilpowerplantsandtheirrigid

255ibid.256ibid.257ibid.258ibid.

JackEllisMOI

74

“big to small – guaranteed return on capital businessmodel.”’259The key concern forutilities is thatdecreasing revenueswill lead to “strandedassets”,which is essentiallywhen an asset becomes a liability. According to the Smith School of Enterprise andEnvironmentattheUniversityofOxford,thereisanumberofrisksthatcancauseassetstobecomestranded,asthetablebelowshows.

Table7:Typologyofenvironmentrelatedrisks260

Itisclearthatmanyrisksexistthatcancontributetocentralisedfossilfuelpowerplantsbecomingstrandedassets.Technologychange,asthekeyfocusofthispaper,representsjust one of the risks. This suggests that the combination of all the factors identifiedabove poses a profound threat to centralised utilities, and that they will find itincreasinglydifficulttoguaranteereturnsinthefuture.Thisisespeciallytrueinthecaseofnon-exportationofsolarPV,asthispathultimatelyleadstogriddefection.Theotherpathisdescribedhere:‘Downanotherpatharepricingstructures,businessmodels,andregulatoryenvironmentsinwhichdistributedenergyresourcessuchassolarPVandbatteries–andtheirinherentbenefitsandcosts–areappropriatelyvaluedaspartofan integratedgrid.SolarPVandbatteriescanpotentiallylowersystem-widecostswhilecontributingtothefoundationofareliable, resilient, affordable, low carbon grid of the future in which customers areempoweredwithchoice.Inthisfuture,gridandcustomer-sideresourcesworktogetheras

259POPE,C.(2016).RooftopSolarWarsContinue.EcoWatch.[Online]Accessedon7/4/2016.Availablefrom:http://ecowatch.com/2016/01/29/rooftop-solar-wars/260SMITHSCHOOLOFENTERPRISEANDTHEENVIRONMENT.(2016).Strandedassetsandthermalcoal–Ananalysisofenvironmental-relatedriskexposure.UniversityofOxford.[Online]Accessedon25/4/2016.Availablefrom:http://www.smithschool.ox.ac.uk/research-programmes/stranded-assets/satc.pdf

27Stranded Assets and Thermal Coal: An analysis of environment-related risk exposure

1 IntroductionThe principal aim of this report is to turn the latest academic and industry research on environment-related risk factors facing thermal coal assets into actionable investment hypotheses for investors. By examining the fundamental drivers of environment-related risk, creating appropriate measures to differentiate the exposure of different assets to these risks, and linking this analysis to company ownership, debt issuance, and capital expenditure plans, our research can help to inform specific actions related to risk management, screening, voting, engagement, and disinvestment. This report contains a thorough and up-to-date assessment of the key environment-related risk factors facing thermal coal assets and may also be of use for policymakers, companies, and civil society. The typology of environment-related risks is described in Table 6. Another aim of this work is for the datasets that underpin our analysis, as well as the analysis itself, to enable new lines of academic research and inquiry.

The vast majority of analyses that concern environment-related risks facing different sectors of the global economy are ‘top down’. They look at company-level reporting and usually focus on measures of carbon intensity or carbon emissions. Even if this company level reporting is accurate and up-to-date (in many cases it is not), this is an overly simplistic approach that attempts to measure a wide range of environment-related risk factors (often with widely varying degrees of correlation) through one proxy metric (carbon). While this might be a useful exercise, we believe that more sophisticated ‘bottom up’ approaches can yield improved insights for asset performance and, if appropriately aggregated, company performance. In this report, we apply this bottom up, asset-specific approach to the thermal coal value chain.

Set Subset

Environmental Change Climate change; natural capital depletion and degradation; biodiversity loss and decreasing species richness; air, land, and water contamination; habitat loss; and freshwater availability.

Resource Landscapes Price and availability of different resources such as oil, gas, coal and other minerals and metals (e.g. shale gas revolution, phosphate availability, and rare earth metals).

Government Regulations Carbon pricing (via taxes and trading schemes); subsidy regimes (e.g. for fossil fuels and renewables); air pollution regulation; voluntary and compulsory disclosure requirements; changing liability regimes and stricter licence conditions for operation; the ‘carbon bubble’ and international climate policy.

Technology Change Falling clean technology costs (e.g. solar PV, onshore wind); disruptive technologies; GMO; and electric vehicles.

Social Norms and Consumer Behaviour

Fossil fuel divestment campaign; product labelling and certification schemes; and changing consumer preferences.

Litigation and Statutory Interpretations

Carbon liability; litigation; damages; and changes in the way existing laws are applied or interpreted.

Table 6: Typology of environment-related risks

JackEllisMOI

75

part of an integrated grid with for more efficient deployment of capital and physicalassets.’261Thefigurebelowshowsthe‘metaphoricalforkintheroad’thattheelectricitysystemfindsitselfat,asdescribedbyRMI.

Figure29:Possiblepathsforelectricitygridevolution262

ThealternatepathoutlinedisoneinwhichsolarPVandbatteriesareanimportantpartof an integrated grid. In contrast to the grid defection path, an integrated grid bringswith it a much greater democratisation for customers as they are “empowered bychoice.” As shown in chapter 1, the technical applications of battery storage arewideranging. The result of this is that if batteries are successfully integrated into the gridsystem, then the system-wide costs are reduced. Ultimately the path towards anintegrated grid is one which both grid and customer resources are able to worksymbiotically,achievingamuchmoreefficientuseofcapitalandassets.Realisingthesebenefitsisunderstoodtorequirereformonthreefronts:263

• Newpricingandratestructures:Theneedsof the21st centurygridarecomplexandpricingshouldevolveindifferentways:

o Locational–allowingcongestionpricingorotherincentiveso Temporal–timeofuseandreal-timepricingo Attributebased– breaking part energy, capacity, ancillary services and

otherservicecomponents261ROCKYMOUNTAININSTITUTE.(2015).Theeconomicsofloaddefection:Howgrid-connectedsolar-plus-batterysystemswillcompetewithtraditionalelectricservice,whyitmatters,andpossiblepathsforward.RockyMountainInstitute.Colorado.262ibid.263COVINGTON&BURLINGLLP.(2015).Theimpactofdistributedgenerationonelectricutilities:Howbig,howlikelyandhowsoon?[Online]Accessedon26/4/2016.Availablefrom:https://www.insideenergyandenvironment.com/2015/04/the-impact-of-distributed-generation-on-electric-utilities-how-big-how-likely-and-how-soon/

JackEllisMOI

76

• Newbusinessmodels–Theneed to evolve from centralised generation and theunidirectional use of the grid. Businessmodels need to become based on gridconnected customers with distributed resources and a two-way flow ofelectricityonthegrid

• Newregulatorymodels–thatprovidethefollowing:o Fairandequalcustomeraccesstodistributedresourceso Recognise, quantify andmonetise the benefits and costs of distributed

resourceso Treatallcustomersequally

Althoughnot explicitlymentioned in the above reforms,Hoppmanet al. note that theshift towards a system of strongly distributed electricity generation will ‘probablyrequiremajoradaptationsinthetechnicalinfrastructureoftheelectricitysystem,suchasdistributiongrids.’264Theconcernregardinginterconnectionisalsopresentinotherreportsondistributedgeneration.AnMITstudynotesthefollowing:‘Theintegrationofdistributedgenerationpresentsnewchallengesfordistributionsystemplanning and operations, principally because the configuration of power lines andprotective relaying inmost existing distribution systemsassumeauni-directional powerflow and are designed and operated on that assumption…While the physical wires andtransformers can carry power flow in the reverse direction, Distributed generationnonethelesscanhaveadverseimpactsonsystemreliability,powerqualityandsafety.’265The above quote appears to give support to the concerns of Hopperman et al. Justbecausebatterystoragehasproventobeeconomicallyviableforhouseholds,itdoesnotimply thatbatterystorageanddistributedgenerationsystemsare ‘beneficial fromtheperspective of overall stability of the electric system.’266Despite the concerns, withinthis investigationtheapplicationsofbatterystoragehavebeenshowntocontributetogridstabilitywhencombinedwithdistributedgenerationsourcessuchassolarPV.ThisissupportedbyHollingeretal.whofoundthatbatterystorageinresidentialPVsystemscanreducetheburdenonelectricitydistributiongridsbyaround40%.267Eventhoughother studieshave foundnopositiveeffectofbattery storagealleviating stresson thedistributiongrid,268itseemshighlyunlikelythatbyintegratingsolarPV-batterysystemsintoasmartgrid,improvementsingridstabilitywouldnotbepossible.Improvements in grid stability emphasise the characteristics of the path towards anintegratedgridasopposedtogriddefection.Withthereformsoutlinedaboveaswellas

264HOPPMANJ.,VOLLAND,J.,SCHMIDT,T.S.,HOFFMANN,V.H.(2014).TheEconomicViabilityofBatteryStorageforResidentialSolarPhotovoltaicSystems–AReviewandSimulationModel,RenewableandSustainableEnergyReviews39,1101-1118.p.26.265MIT.Theimpactofdistributedgenerationandelectricvehicles.[Online]Accessedon27/4/2016.Availablefrom:https://mitei.mit.edu/system/files/Electric_Grid_5_Impact_Distributed_Generation_Electric_Vehicles.pdf266ibid.267HOPPMANJ.,VOLLAND,J.,SCHMIDT,T.S.,HOFFMANN,V.H.(2014).TheEconomicViabilityofBatteryStorageforResidentialSolarPhotovoltaicSystems–AReviewandSimulationModel,RenewableandSustainableEnergyReviews39,1101-1118.p.26.268ibid.

JackEllisMOI

77

thenecessaryinvestmentinthetechnicalinfrastructureofthegrid,theadvantagesofapathtowardsgridintegrationaresignificant.Theseincludethefollowing:

• Lowcarbon• Lowerconsumercosts• Lowersystem-widecosts• Improved grid stability (due to power smoothing and ancillary services in

chapter1)• Moreefficientdeploymentofresources/assets• Customerempowerment

It remains to be seenwhich path is taken. But as noted by Leia Guccione, one of theauthorsoftheRMIreport,‘thereisarealcostindoingnothing…Intheabsenceofmorecustomerchoices,customerswilltakemattersintotheirownhands.Andthat’sgoingtolead to sub-optimal outcomes that we see in grid defection – overinvestment andunderutilisedcapital.’269Thepathtowardsanintegratedgridsystemhaswithoutdoubtbeenshowntobethesuperioroption.However,whatisparticularlynoteworthyhereistheroleofthecustomer,whichinregardstoenergysystemshasundergoneaprofoundchange.

DemocratisationofenergyIn the UK only 18% of people believe that energy prices should be decided by thoseproviding the service.270This suggests customers’ dissatisfaction with the fact thatprivate energy companies are the ones that set the prices. In fact, over two thirds ofthose surveyedbelieve that energy companies shouldbe run in thepublic sector.271Ifcustomershave the ability to take control over theirownenergyprices then theyarelikelytodoso.TheEnergiewendeinGermanyprovidesthebestevidenceforthis.

269COVINGTON&BURLINGLLP.(2015).Theimpactofdistributedgenerationonelectricutilities:Howbig,howlikelyandhowsoon?[Online]Accessedon26/4/2016.Availablefrom:https://www.insideenergyandenvironment.com/2015/04/the-impact-of-distributed-generation-on-electric-utilities-how-big-how-likely-and-how-soon/270DAHLGREEN,W.Nationaliseenergyandrailcompaniessaypublic.YouGov.[Online]Accessedon27/4/2016.Availablefrom:https://yougov.co.uk/news/2013/11/04/nationalise-energy-and-rail-companies-say-public/271ibid.

JackEllisMOI

78

Figure30:OwnershipofGermanrenewablesin2012272

The figure above shows that the energy transition is, or at least has shown to be inGermany,ademocraticmovementthatempowersthepublic,allowingthemnotmerelytobeconsumersbutalso“prosumers”-simultaneouslyproducersandconsumers.Somecountries,suchastheUK,promoterenewableswithpoliciessuchas‘quotasystems.’273Inthesecases,targetsaresetforutilitiestoreachandthereisageneralfocusoncost.Inthis system many proposals are rejected. In Germany it is the decision of the localgovernmenttodecidewhererenewablesourceswillbebuiltandifaFITsystemisused.Utilities are under no obligation to increase their share of renewables. The officialEnergiewendewebsitehighlightsthedifferencesofthetwosystems:‘Overall, the difference between the two approaches – feed-in tariffs versus quotas – isstriking. Under quotas, only the least expensive systems go up after time-consumingreviews, and they remain in the hands of corporations; under feed-in tariffs, everythingworthwhilegoesupquickly,andownershipofpowersupplyrapidlytransferstocitizenry.Inotherwords,Germanyisdemocratizingitsenergysector.’274It is estimated that energy cooperatives in Germany, community-owned renewableprojects, leveragedmore than€1.2bn in investments from130,000private citizens in2013.275Although critics may say that only the wealthy are able to invest in suchprojects, this is not true. A single share costs less than €500 in over two thirds ofcooperativeswhilesomeareascheapas€100.276AsnotedbyGuccione,customersaretakingmatters into theirownhands.The increasingnumberofcooperativesshown inthefigurebelowonlyservestohighlightthis.

272ENERGYTRANSITION.(2012).Infographics.[Online]Accessedon27/4/2016.Availablefrom:http://energytransition.de/2014/12/infographs/273ENERGYTRANSITION.(2012).Energybythepeopleforthepeople.[Online]Accessedon27/4/2016.Availablefrom:http://energytransition.de/2012/10/energy-by-the-people/274ibid.275ibid.276ibid.

Energy suppliers12%

Institutionaland strategic investors41%

Totalinstalled

capacity 2012

73 GW Citizens and coops47%

German energy transition is a democratic movementOwnership of renewables in 2012Source: AEE, www.unendlich-viel-energie.de

ccenergy transition.de

JackEllisMOI

79

Figure31:NumberofenergycooperativesinGermany2001-2013277

The increasing number of co-operatives in Germany suggests that the public is veryacceptingofthemovementtowardsdecentralizedenergy.JustasthestatisticsfromtheUKsuggestapublicdissatisfiedwith the levelof control that itpossessesoverenergyprices. The increasing levels of energy democratisationwitnessed through Germany’sEnergiewende are quite extraordinary when compared to the 20th century model ofcentralisedenergy.InitiativessuchascooperativesplayavitalroleinGermany’senergytransition and it should not be forgotten that the transition itself is a democraticmovement,withtheabilitytoempowerthepublic.As the head of Germany’s Solar Industry Association (BSW-Solar) puts it, “EnergycooperativesdemocratizeenergysupplyinGermanyandalloweveryonetobenefitfromtheenergytransitioneveniftheydonotowntheirownhome.”278

277 ENERGY TRANSITION. (2012). Infographics. [Online] Accessed on 27/4/2016. Available from:http://energytransition.de/2014/12/infographs/278ENERGYTRANSITION.(2012).Energybythepeopleforthepeople.[Online]Accessedon27/4/2016.Availablefrom:http://energytransition.de/2012/10/energy-by-the-people/

Number of cooperatives

0

300

600

900

Citizens form cooperatives to drive the energy transitionNumber of energy cooperatives in Germany, 2001–2013Source: www.unendlich-viel-energie.de

ccenergy transition.de

2013201220112010200920082007200620052004200320022001

136

239

398

600

746

888

101867774707066

JackEllisMOI

80

ConclusionThis paper has sought to provide a comprehensive investigation into the impact ofbattery storage technologies. Firstly, and in order to contextualise the issue at hand,recenttrendsinenergyfinancewereinvestigated.ItwasfoundthatthefallingpriceofpolysiliconhasplayedalargeroleinthedecreasingpricesofPVcells.Thiswasduetothe fact thatmaterial inputs constituted sucha largepercentageof themanufacturerscostofproduction.Largescaleproduction,intensecompetitionandimprovedefficiencyledthetechnologyto become increasingly economically viable. This was demonstrated by the Swansoneffect. As a result, major economies, most notably Germany, have increased theirelectricity output from solar PV installations significantly over the past decade. Thistrend has reached the point at which the world is now adding more capacity forrenewableenergythanforcoal,oilandgascombined.Althoughthereisacleartrendtowardsrenewableenergysources,VaclavSmilbelievedthat the trend would be accelerated through the inexpensive application of energystorage.Thiswasduetotheintermittentnatureofrenewables.Thisinvestigationhasshownthatbatterystoragehastheabilitytoplayasignificantrolein countering intermittency. In addition, the investigation has shown that batterystoragehasmanymoreservices,andbringswithitmuchmoresignificantdynamics.Chapter one focused on technical aspects of battery storage technologies andhighlighted the advantages that lithium-ion technology enjoyed over other types. Thetechnology was shown to have advantages in terms of energy density, efficiency andpower.Byinvestigatingthepracticalandreal-worldusesofbatterystorage,thechapterdescribedthevariousservices thatbatterystorage isable toprovide.Asshown in thechapter,energy time-shifting,ancillaryservicesandenergysmoothingwereproven tobeparticularlyusefulAs shown in the summaryof chapter one, the services arenumerous andbringmanybenefits. Power reliability, benefits to existing infrastructure, the ability to providebackuppowerandrenewableenergysupportarealsoimportant.Theinvestigationhasshown that battery storage solves the intermittency problem through energy time-shifting.AccordingtoSmil,thiswillpromotetheexpansionofrenewables.Whatismore,the investigation also found other applications that further increase renewableintegration,aswellasbringingotherbenefitsBenefits are valued in economic terms. The investigation showed that in order forbattery storage to become increasingly utilised, the services it providesmust receiveadequatecompensation.Chapterfocusedontheeconomicaspectsofbatterystorage.TheinvestigationfoundthatintegratedsolarPV-battery systemswereprofitableundera rangeofdifferentpricingscenarios.Theprofitabilityofstoragewasforecastedtoincreaseovertime.Thisfactwas

JackEllisMOI

81

further emphasised in the following section which showed the decreasing price oflithium-ionbatteries.Thechapteralsoidentifiedthelackofcosttransparency,andofacohesiveregulatoryframeworkaskeybarrierstobatterystorage.Theinvestigationdidfind thatmany countries had an active policy environment regarding battery storage.MostnoteworthywasthetrendshownbytheFederalEnergyRegulatoryCommissionin2011.Order755madeitpossiblethatthesuperiorperformanceoffrequencyregulationserviceswasadequatelycompensated.Thistrendisbeingdrivenbytheneedtocreatefairregulationsandcompensatenon-traditionalmeasuresforthebenefitstheyprovide.These aspects are critical in facilitating and promoting investment in battery storage,andconsequentlyplayimportantrolesintheenergytransition.Theinvestigationfoundanincreasingtrendforbatteriestobeusedinancillaryservicesandfrequencyregulation,wheretheycreatedcreatedvaluelargelyduetotheirsuperiorperformance.Inaddition,itwasalsofoundthatthereisanincreasingtrendformarketsto pay for the quality and accuracy of the service provided. These trends complimenteachotherandwill likely lead to thegreaterapplicationofbattery storagedue to theservicesitisabletoprovide.Internationalisation, which has been present in value chains throughout theinvestigation, has undoubtedly contributed to this increased economic viability ofbatterystorage.Marketshavebeenshowntobegrowingatstrongrates,andthereareincreasingopportunities for internationalisationandinvestment.Furthermore, if thereis a trend for services provided by battery storage to receive compensation based ontheir frequency and accuracy, energy storagemarketswill becomemore competitive,and investorsmore confident. Thus, the investigation has shown that battery storagemarketwillbothgrowandincreaseincompetitioninthefuture.Thisisasthetrendofadequatecompensationprovidedbyregulationwillonlyservetostrengthenthebatterystoragemarket.Aswellasbarriers,thereistheaforementionedneedtodevelopacommonapproach.Anapproach that encompasses technical, regulatory, market and political aspects washighlighted. In order to achieve this “holistic” approach to storage, greater levels ofinternationalcooperationarerequired.Althoughinternationalisationhascontributedtoincreased competition in markets, the lack of a holistic approach towards batterystorage suggests that greater internationalisation is necessary. Theremust be greaterpoliticalandregulatoryinitiativesshownbytheinternationalcommunity.Thisinordertoachieveacommonapproachthathasthepotentialtofacilitatetheenergytransition.Thelackofregulatoryandpoliticalaspectsinthecommonapproachisunderpinnedbythefactthatthetechnicalandmarketaspectsofbatterystoragehavebeenproventobeso successful. In light of this, when noted that the EU commission believed themainchallengetobeeconomic,itisnotthatbatterystorageisnotyeteconomicallyviable,butratherfacilitatingtheconditionsforinvestmentthatisthekeychallenge.Of particular significance, the investigationhas shown that as support for renewablesdeclines(FITs),theimplementationofstorageislikelytoincreaseovertime.Thisisduetothe fact that integratedsolarPV-batterystoragesystemswereshownto increase insizeandprofitabilityovertime.Batterystorageisnotonlyalreadyeconomicallyviable

JackEllisMOI

82

under many scenarios, but also drives an increasing number of people to be“prosumers”. This trend is leading energy to become cleaner, safer and increasinglydecentralised.The final section of the paper, regarding the relationship between the public andutilities, showed that the public is becoming increasingly active, participating in thedevelopment of renewable energy sources. As battery storage has proven to beeconomicallyviableforresidentialsystems,andsettoincreaseinprofitability,thereislikelytobeanincreasingtrendawayfromthetraditionalconsumer-utilitydynamic.Theinvestigation highlighted possible paths for the future, of which the integrated gridsystem was shown to be superior. This empowered consumers, as it was the moredemocraticoption,andallowedforamoreefficientuseofcapital.Theinvestigationhasfoundthatbatterystoragetechnologywillhaveaprofoundimpactontraditionalcentralisedutilities. Ifpeoplechoosetodefect fromthegridcompletely,then battery storage allows them to become completely self-sufficient. However, theintegratedgridisthesuperiorpath,andasaresultitiscriticalthatthedecisionsmadebygovernmentsandtheinternationalcommunityreflectthis,andpromotetherequiredinvestment. Although the investigation has shown that utilitieswill be effected by anintegrated grid, and lose revenues, thepath of integrationoffers benefits for both theconsumerandutilitycompanies.Thecustomershavetheability toexport theirexcessenergy, increasingtheirsay in theenergysystem.Theutilitiesontheotherhandhavelessriskofstrandedassetsandareabletoworkalongsideintegratedbatterysystems,withbothprovidingvalue.The fact that the investigation foundthatseeminglyutility-friendlypolicies,suchastheabolitionofnetmetering,onlyhadtheeffectofdelayingtheinevitableoutcomeofgriddefection,givessupporttotheargumentthatthecreationofanintegratedgridisbestpathforthefuture.Duetotheabove,onecanbeforgivenforthinkingthattheelectricityindustryissetformajor changes in the future. The truth is that change is likely to be more gradual.Regardingtheeffectofdistributedgenerationontheelectricityindustry,Costellowritesthat,‘itseemsinevitablethatchangewillcomebutitseffectontheelectricityindustryisstill influxandunknown.’279This investigation, inunderstandingtheimpactofbatterystoragetechnology,hasultimatelyshownwhattheeffectontheelectricityindustrywillbe. Centralised generation is under threat from the increasing use of decentralisedenergy. Battery storage has been proven to accelerate this trend. This trendwill onlyincrease in the future, driven by the public desire to take control of their energy andfallingbatteryprices.Utilitiesmustadapttosurvive,astheyhavealreadybeguntodo.Theintegratedgridoffersthesuperiorpathforthefuture.Itistechnicallypossibleandeconomicallyviable.However,itwillrequireregulatorychanges,transparentcostsandpoliticalwilltomakesurethatthefutureelectricitysystemistheoptimalone.

279COVINGTON&BURLINGLLP.(2015).Theimpactofdistributedgenerationonelectricutilities:Howbig,howlikelyandhowsoon?[Online]Accessedon26/4/2016.Availablefrom:https://www.insideenergyandenvironment.com/2015/04/the-impact-of-distributed-generation-on-electric-utilities-how-big-how-likely-and-how-soon/

JackEllisMOI

83

Bibliography

• AECOM.(2015).EnergyStorageStudy.AustralianRenewableEnergyAgency

• ARPA-E.(2011)DistributedPowerFlowControl.[Online]Availablefrom:http://arpa-

e.energy.gov/?q=slick-sheet-project/distributed-power-flow-control

• BALZA,L.etal.(2014),PotentialforEnergyStorageinCombinationwithRenewable

EnergyinLatinAmericaandtheCaribbean,InterAmericaDevelopmentbank

• BLOOMBERGNEWENERGYFINANCE.(2015).NewEnergyOutlook2015

• BLOOMBERGNEWENERGYFINANCE.(2016)CleanEnergyInvestment:Q42015.

• BLUME,S(ed).(2015)GlobalEnergyStorageMarketOverview&RegionalReport

Summary.EnergyStorageCouncil.[Online]Accessedon10/4/2016.Availablefrom:

http://neca.asn.au/sites/default/files//media/state_nsw/News%20&%20Views/ESC%

20Global%20Energy%20Storage%20Report_2015.pdf

• BOYEN,ANNA.(2014)Theenvironmentalimpactsofrecyclingportablelithium-ion

batteries.AustralianNationalUniversity.

• BRUNO,B.(Dr.)(2016)FraunhoferInstituteforSolarEnergyISE–Powergeneration

fromrenewableenergyinGermany–Assessmentof2015.

• BUHAYAR,N.(2016)WhoOwnsTheSun?BloombergBusinessWeek[Online]Accessed

on7/4/2016.Availablefrom:http://www.bloomberg.com/features/2016-solar-power-

buffett-vs-musk/

• CAIRNS,E.J.&ALBERTUS,P.(2010)Batteriesforelectricandhybrid-electricvehicles.

Annu.Rev.Chem.Biomol.Eng.1,299-320.

• CATENACCI,M.,VERDOLINI,E.,BOSETTI,V.&FIORESE.(2013)G.Goingelectric:Expert

surveyonthefutureofbatterytechnologiesforelectricvehicles.EnergyPolicy61,403-

413

• CITIGROUP.(2014).GermanUtilities:Letthesurvivalgamebeginasthelostdecade

takeshold.CitiResearch.

• CNNMONEY.(2016).TeslaModel3’sfirstweek:325,000orders.[Online]Accessedon

13/4/2016.Availablefrom:http://money.cnn.com/2016/04/07/autos/tesla-model-3-

orders/

• COVINGTON&BURLINGLLP.(2015).Theimpactofdistributedgenerationonelectric

utilities:Howbig,howlikelyandhowsoon?[Online]Accessedon26/4/2016.Available

from:https://www.insideenergyandenvironment.com/2015/04/the-impact-of-

distributed-generation-on-electric-utilities-how-big-how-likely-and-how-soon/

• DAHLGREEN,W.Nationaliseenergyandrailcompaniessaypublic.YouGov.[Online]

Accessedon27/4/2016.Availablefrom:

https://yougov.co.uk/news/2013/11/04/nationalise-energy-and-rail-companies-say-

public/

JackEllisMOI

84

• DIRECTIVE2006/66/ECOFTHEEUROPEANPARLIAMENTANDOFTHECOUNCILof6

September2006onbatteriesandaccumulatorsandwastebatteriesandaccumulators

andrepealingDirective91/157/EECp.27.

• ECONOMIST.(2012).PricingSunshine.[Online]Accessedon5/3/2015from:

http://www.economist.com/blogs/graphicdetail/2012/12/daily-chart-19

• ENERGYTRANSITION.(2012).Energybythepeopleforthepeople.[Online]Accessedon

27/4/2016.Availablefrom:http://energytransition.de/2012/10/energy-by-the-

people/

• ENERGYTRANSITION.(2012).Infographics.[Online]Accessedon27/4/2016.Available

from:http://energytransition.de/2014/12/infographs/

• ENERGYTREND.(2016).InstalledcostofutilityscalePVsystemintheUSdown17

percent.[Online]Accessedon5/4/2016.Availablefrom:

http://pv.energytrend.com/research/Installed_Cost_of_Utility_Scale_PV_System_in_the_

US_Down_17percent_YoY_in_3Q15.html

• EUROPEANCOMISSION.Memo:EUimposesprovisionalanti-dumpingdutiesonChinese

solarpanels[Online]Accessedon4/4/2015from:http://europa.eu/rapid/press-

release_MEMO-13-497_en.htm

• EUROPEANCOMMISSION.DGENERWorkingPaper:Thefutureroleandchallengesof

EnergyStorage.

• FARRELL,J.(2011).DemocratizingtheElectricitySystem.TheNewRulesProject.

• FARRELL,J.(2014)EnergyStorage:TheNextChargeforDistributedEnergy.The

InstituteforlocalSelf-Reliance

• FARRELL,J.CleanTechnica[Online]Availablefrom:

http://cleantechnica.com/2014/07/21/utilities-cry-fowl-over-duck-chart-and-

distributed-solar-powercrying-fowl-or-crying-wolf-open-season-on-the-utilitys-solar-

duck-chart/

• FU,R.,JAMES,T.,WOODHOUSE,M.(2015).EconomicMeasurementsofPolysiliconfor

thePhotovoltaicIndustry:MarketCompetitionandManufacturingCompetitiveness.

IEEEJournalofPhotovoltaics.Vol.5.No.2.

• GAINES,L.(2014)Thefutureofautomotivelithium-ionbatteryrecycling:Chartinga

sustainablecourse.SustainableMaterialsandTechnologies.

• GARCIA,R.A.(2013).Off-GridCommercialMicro-gridSystemProvidesEnergyStorage

forResortinIndia.TrojanBatteryCompany.IntersolarEurope,Munich.

• GEREFFIetal.(2001).Introduction,Globalisation,ValueChainsandDevelopment.

InstituteofDevelopmentStudies.[Online]Accessedon3/3/2015from:

https://www.ids.ac.uk/files/dmfile/gereffietal323.pdf

• GLOBALCLIMATE.(2015).GlobalClimateChange:Evidenceandcauses.[Online]

Availablefrom:http://globalclimate.ucr.edu/resources.html

JackEllisMOI

85

• GMTRESEARCH&SEIA,(2016).U.S.SolarMarketInsight,ExecutiveSummary.GMT

Research&SEIA.

• GREENTECHMEDIA.(2011).IsSodiumtheFutureFormulaforEnergyStorage?[Online]

Accessedon4/3/2015from:http://www.greentechmedia.com/articles/read/is-

sodium-the-future-formula-for-energy-storage

• HOPE,M.(2013).TheEnergiewendeandenergyprices:PublicsupportandGermany’s

longtermvision.TheCarbonBrief.[Online]Accessedon3/3/2015from:

http://tinyurl.commk6qnp2

• HOPPMANJ.,VOLLAND,J.,SCHMIDT,T.S.,HOFFMANN,V.H.(2014).TheEconomic

ViabilityofBatteryStorageforResidentialSolarPhotovoltaicSystems–AReviewand

SimulationModel,RenewableandSustainableEnergyReviews39,1101-1118.

• IEA.(2015).EVbatteries:howquicklyarepricesfalling?[Online]Accessedon

25/3/2016from:https://www.iea.org/newsroomandevents/pressreleasesnews/3-

questions/ev-batteries-how-quickly-are-costs-falling.html

• INTERNATIONALENERGYAGENCY.[Online]Accessedon8/3/2015from:

http://wds.iea.org/WDS/Common/Login/login.aspx(downloadeddataduring

internshipwithTheEnergyCharter,Brussels,Belgium.

• InternationalTradeCentre.(2015).Servicesinglobalvaluechains:Solarpanel

manufacturinginChinap.vii.

• IRENA.(2015).Batterystorageforrenewables:MarketStatusandTechnologyOutlook

• IRENA.(2015).RenewablePowerGenerationCostsin2014

• IRENA.(2014).PoliciesandRegulationsforEnergyStorageinJapan.InstituteofEnergy

EconomicsJapan.[Online]Accessedon10/4/2016.Availablefrom:

http://www.irena.org/DocumentDownloads/events/2014/March/6_Tomita.pdf

• LAZARD.(2015).LazardLevelizedCostofEnergyAnalysis9.0–Keyfindings.

• LEVITT,T.(2016)ClimateChangeFailstotopListofThreatsforBusinessLeadersat

Davos.TheGuardian.[Online]Availablefrom:

http://www.theguardian.com/sustainablebusiness/2016/jan/20/climate-change-

threats-business-leaders-davos-survey

• MAYER,T.,KREYENBERG,D.,WIND,J&BRAUN,F.FEASIBILITYstudyof2020target

costsforPEMfuelcellsandlithium-ionbatteries:Atwofactorexperience.

• McKINSEY&COMPANY.(2016)Thedisruptivepotentialofsolarpower

• McMANUS.(2012).Environmentalconsequencesoftheuseofbatteriesinlowcarbon

systems:Theimpactofbatteryproduction.AppliedEnergy93

• MIT.Theimpactofdistributedgenerationandelectricvehicles.[Online]Accessedon

27/4/2016.Availablefrom:

https://mitei.mit.edu/system/files/Electric_Grid_5_Impact_Distributed_Generation_Elec

tric_Vehicles.pdf

JackEllisMOI

86

• MITNEWS.(2013).Solar-cellmanufacturingcosts:innovationcouldlevelthefield.

[Online]Accessedon5/3/2015from:http://news.mit.edu/2013/solar-cell-

manufacturing-costs-0905

• NAVIGANTRESEARCH.(2014).WebinarReplay–BeyondLithiumIon.[Online]Accessed

on19/4/2016.Availablefrom:

https://www.navigantresearch.com/webinarvideos/webinar-replay-beyond-lithium-

ion

• NYKVIST,B.&NILSSON,M.(2015)Rapidlyfallingcostsofbatterypacksforelectric

vehicles.NatureClimateChange.Vol.5.April2015.

• OBERHOFER,A.(2012).EnergyStorageTechnologies&TheirRoleinRenewable

Regeneration.GlobalEnergyNetworkInstitute.

• PAN,J.,MA,H.andZHANG,Y.(2011)GreenEconomyandGreenJobsinChina.

WorldWatchInstitute.[Online]Accessedon3/3/2015from:

http://www.worldwatch.org/system/files/185%20Green%20China.pdf

• PEGELS,A.GreenIndustrialPolicyinEmergingCountries.[Online]Accessedon

4/3/2015from:

https://books.google.co.in/books?id=C4n8AgAAQBAJ&pg=PA77&lpg=PA77&dq=evoluti

on+of+china%25+27s+solar+PV+manufacturing+industry&source=bl&ots=l5o1-

LSY2O&hl=en&sa=X&ei=NpAFVLeXNs248gWR7ILQCg#v=onepage&q=evolutio%2C%20

2014.&f=false

• POPE,C.(2016).RooftopSolarWarsContinue.EcoWatch.[Online]Accessedon

7/4/2016.Availablefrom:http://ecowatch.com/2016/01/29/rooftop-solar-wars/

• PUTTASWAMY,NandSAHILALI,M.(2015).HowDidChinaBecometheLargestSolarPV

ManufacturingCountry?Step.

• PVMAGAZING.(2015).IntersolarNorthAmerica.Tabuchi’smarketentrypointsto

increasinginternationalisation.[Online]Accessedon19/4/2016.Availablefrom:

http://www.pv-magazine.com/news/details/beitrag/intersolar-north-america--

tabuchis-market-entry-points-to-increasing-

internationalization_100020228/#axzz46Rokh8cm

• PWC.(2015).2015utilitiestrends.[Online]Accessedon4/4/2015from:

http://www.strategyand.pwc.com/perspectives/2015-utilities-trends

• RECHARGE–THEEUROPEANASSOCIATIONFORADVANCEDRECHARGEABLE

BATTERIES.(2013).Safetyoflithium-ionbatteries.Belgium.

• RECHARGEBATTERIES.ORG.(2013).Batteryinformationfactsheet:Lithium-ion

batteries.

• ROCKYMOUNTAININSTITUTE.(2015).Theeconomicsofloaddefection:Howgrid-

connectedsolar-plus-batterysystemswillcompetewithtraditionalelectricservice,why

itmatters,andpossiblepathsforward.RockyMountainInstitute.Colorado.

JackEllisMOI

87

• SEMI.ChinaMarketGrowthFuelledbyGovernmentSpendingDuringIndustryDownturn

[Online]Accessedon3/3/2015from:http://www.semi.org/en/MarketInfo/ctr_027596

• SMIL,V.(2014)TheLongSlowRiseofSolarandWind.ScientificAmerican.

• SMITHSCHOOLOFENTERPRISEANDTHEENVIRONMENT.(2016).Strandedassetsand

thermalcoal–Ananalysisofenvironmental-relatedriskexposure.UniversityofOxford.

[Online]Accessedon25/4/2016.Availablefrom:

http://www.smithschool.ox.ac.uk/research-programmes/stranded-assets/satc.pdf

• SOLARENERGYINDUSTRIESASSOCIATION.(2016)Demandforutility-scalesolarstill

dominating.[Online]Accessedon5/4/2016.Availablefrom:

http://www.seia.org/blog/demand-utility-scale-solar-still-dominating

• TESLA.(2016).Theweektheelectricvehicleswentmainstream.[Online]Accessedon

13/4/2016.Availablefrom:https://www.teslamotors.com/blog/the-week-electric-

vehicles-went-

mainstream?utm_campaign=Blog_Model3_040716&utm_source=Twitter&utm_medium=

social

• UNITEDNATIONS.(1992)UnitedNationsFrameworkConventiononClimateChange

[Online]Availablefrom:https://unfccc.int/resource/docs/convkp/conveng.pdf

• WASTEMANAGEMENTWORLD.(2011).TheLithiumBatteryRecyclingChallenge.

[Online]Accessedon19/4/2016.Availablefrom:https://waste-management-

world.com/a/1-the-lithium-battery-recycling-challenge

• WARMUTH,W.(2016).PhotovoltaicsReport.FraunhoferInstituteforSolarEnergy

SystemsISE.[Online]Availablefrom

https://www.ise.fraunhofer.de/de/downloads/pdf-files/aktuelles/photovoltaics-

report-in-englischer-sprache.pdf

• WIRTH,H.(Dr.)(2015).RecentFactsaboutPhotovoltaicsinGermany.Fraunhofer

InstituteforSolarEnergySystemsISE.

• YOUTUBE.(2016).ElonMuskUnveilsTeslaModel3.[Online]Accessedon5/4/2016.

Availablefromhttps://www.youtube.com/watch?v=7HOhOeKwe30

• ZACUNE,J,(2013).Lithium.FriendsofEarthEurope.[Online]Accessedon18/4/2016.

Availablefrom:

https://www.foeeurope.org/sites/default/files/publications/13_factsheet-lithium-

gb.pdf