(The author is a Reuters market analyst. The views expressed are his own.)
By Gerard Wynn
LONDON, April 24 Hydropower remains one of the most efficient, grid-scale energy storage options, according to a recent Stanford University article, implying more research and development required for chemical battery approaches.
Grid-scale energy storage is a potential game-changer in electricity markets, if it can overcome the present problem where power is generally consumed at the instant of generation.
At present, total electricity generating capacity must be far higher than average consumption to cater for peak demand.
Rising renewable power capacity makes an even more urgent case for energy storage, given that wind and solar power supply are variable like demand, requiring additional load balancing.
Storage options range from pumped hydroelectric storage (PHS), which accounts for the vast majority of grid-scale energy storage capacity worldwide available now, to newer, higher technology chemical approaches including lithium ion and sodium sulphur batteries.
If a full life cycle energy cost approach is taken, PHS is by far the least energy-intensive technology alongside compressed air energy storage (CAES), which is still at a development stage.
That is according to the Stanford University paper, "On the importance of reducing the energetic and material demands of electrical energy storage", which was first published online in January in the journal "Energy & Environmental Science".
PHS uses surplus electricity to pump water to higher elevations, working as a dam in reverse. The electricity is then released as required to drive hydropower turbines.
The limitations of PHS include the availability of hydropower resources and planning hurdles in dam projects.
Intermittent renewable power is a growing feature of electric grids in many European countries including Europe's biggest economy, Germany, where last year wind and solar accounted for 13 percent of power generation.
Renewables will push conventional gas and coal-fired power off the grid when the wind and sun are available, given that they are inflexible and have priority grid access in Europe.
German utilities warn that will cut capacity factors and returns, threatening the operation of wholesale power markets.
Energy storage would smooth the variable supply of renewables and make it more flexible, allowing fossil fuel plants to run at optimum efficiencies.
It could also help to avoid other, potentially costly, load- balancing activities which include: maintaining reserve generating capacity; building out interconnectors; paying for generators to curtail surplus generation; and paying consumers to cut peak demand.
The Stanford paper takes a full life-cycle approach to measuring energy use by batteries and other storage technologies, estimating energy (and other resources) consumed both in their manufacture and during operation.
That follows a similar approach to estimating the resource intensity of electric vehicle batteries, where the aim is to assess whether electric and hybrid cars are greener than conventional internal combustion engines.
The Stanford analysis is rather theoretical, given that grid storage is presently at an early stage.
It should be noted that uptake will also depend on the avoided cost of alternative grid balancing approaches, such as building reserve generating capacity.
And life-cycle energy cost is not the main factor used to parse existing electricity-generating technologies from renewables to coal and gas-fired power.
The paper's central thesis is that the energy and material intensity of manufacture will be increasingly important in the event of large-scale adoption.
"Reducing financial cost is not sufficient for creating a scalable energy storage infrastructure," the authors assert.
The paper found that PHS and CAES were far less energy- intensive than chemical batteries on a life cycle basis.
"Electrochemical storage technologies require 10 to 100 times more embodied energy (in manufacture and operation) for a given energy capacity than geological storage technologies."
"Over their entire life, electrochemical storage technologies only store 2-10 times the amount of energy that was required to build them," they found, compared with more than 200 times for PHS and CAES.
Specific advantages of PHS include a high efficiency of storage at around 85 percent and a very high cycle life, defined as the number of times a technology can be charged and discharged.
PHS and CAES have a cycle life of more than 25,000 times compared with central values of 6,000 for lithium ion and 700 for lead acid batteries, the authors state.
Cycle life is more important in electric grids than in electric vehicle applications, where in the latter there is little point having a battery which out-lives the car.
Cycle life also becomes more important in an analysis which accounts for full energy costs including manufacture, as in this paper, given a higher value reduces the frequency of equipment replacement and therefore cuts the estimated cost.
That illustrates how the scope of an analysis necessarily determines its results.
"Most importantly, for grid integrated storage, cycle life must be improved to improve the scalability of battery technologies," the authors concluded.
"For all electrochemical storage technologies, the up-front energy cost ... dominates the energy budget."
The usefulness of the Stanford paper lies in its originality in examining the full energy and other resource impact of adopting energy storage in electric grids, and in illustrating the present gulf in energy intensity between existing and newer electrochemical approaches.
It also provides an argument for exploring further PHS potential as a near-term option, given the authors' contention that the world has at least 10 times present day pumped hydro capacity. (Editing by Stephen Nisbet)