Christopher O’Dea reports on how technology is transforming the energy sector
At a glance
• Battery technology is opening new vistas in transportation and electricity generation.
• Two broad approaches are emerging – an extended ‘lithium value chain’ and a downstream approach focusing on voltage conversion and control of energy flows.
• Better batteries will speed adoption of solar and other renewables.
It wasn’t that long ago that every game and toy carried the warning “batteries not included”, and failure to heed that admonition could spoil a holiday morning.
Today batteries have become the essential component of the digital economy, powering everything from smartphones and laptops to power tools and electric vehicles. Advances in battery science and the technology that controls battery power are opening new frontiers in transportation and electrical generation – and promising to create winners and losers in the process.
Fund managers and investment analysts are recasting their universes of potential investments and research coverage, while scientists and technology researchers continue to push the boundaries of energy storage by raising the density of batteries through innovative configurations of new materials and optimising design of energy cells for commercial uses.
“We have started a new round of a high innovation cycle where within a few years, we will make great progress in energy density in batteries, allowing us to move considerably down the cost curve,” says Thiemo Lang, Zurich-based senior portfolio manager of the RobecoSAM Smart Energy fund.
Lithium will remain as the dominant battery technology as industry consolidation, high capital costs for commercial-scale production and access to emerging science form barriers to entry that create an extended lithium-ion battery value chain across the globe. That’s the forecast by a team of analysts at Joh. Berenberg, Gossler & Co (Berenberg), led by Asad Farid, that contends a sharp decline in battery costs over the next five years will mark a tipping point towards mass adoption of battery-powered cars, buses and utility-scale electrical storage that will transform the legacy grid as solar and wind generation take hold.
But that doesn’t mean investing in opportunities in the new energy storage economy will be as easy as popping a couple of AA-batteries into a toy robot on Christmas day. While there are clear themes that encompass how improved energy storage can be put to use in a variety of industries – both new and emerging – it may be difficult to find ways to capitalise directly on energy technology advances. “It’s not always that obvious to find pure plays on these clean energy technology ideas,” says RobecoSAM’s Lang.
“There are so many things going on to reshape the energy sector of the future, it’s not appropriate to invest in all of them,” adds Lang, who says his firm put its universe together differently for 2016 to bring the new opportunity set better focus. One major change was to drop coverage of upstream natural gas producers, while retaining natural gas distribution in an energy distribution theme with electric networks and equipment suppliers – one of four new ‘clusters’ encompassing the new energy economy.
The other three clusters are renewable energies, energy efficiency and energy management. Energy management, which includes storage and semiconductor power management, has an outsize role in the RobecoSAM portfolio, at 26% of fund assets. Most of that position – 21% – is comprised of companies that design and manufacture semiconductors that specialise in energy management and power consumption control units. “Semiconductors are the enabling technology of the energy system of the future,” Lang says.
For example, California-based Enphase makes micro inverters, a device that is used for efficient solar module-level power conversion, that will in the future also be applied to AC battery storage, enabling them for later use in appliances or for sale into a grid. Some of Enphase’s products are already designed to work with specific grid requirements in US states like Hawaii and California, and the company is rapidly expanding its global footprint. In March it opened a new R&D facility in New Zealand, in preparation for the launch later this year of residential battery storage and energy management systems there and in Australia.
The efficiency of the conversion process mediated by power management chips is critical to the efficiency of energy systems, Lang notes. Specialised chips control each step as electricity moves from generation sources, into batteries and devices like smartphones. The task becomes more important as systems become larger, such as for commercial data systems running on solar energy, like those operated by Apple in Nevada, where the reduction of heat is a major goal – heat represents lost energy, wasted power consumption, and higher costs for cooling computers. “Efficient conversion is a big issue,” Lang concludes.
Power conversion is perhaps the biggest issue in the still-nascent electric vehicle market, which is expected to be the biggest beneficiary of the falling cost of battery storage. Berenberg says battery costs are falling by half every five years as manufacturing scale increases, and the bank expects the global electric vehicle (EV) market to grow 14-fold by 2020, into a $140bn (€123bn) sector as developed countries and China move to reduce transport-related air pollution by encouraging adoption of battery electric vehicles over hybrids. Battery costs will fall 43% by 2020, to $170 per kilowatt hour from $300 today, Berenberg projects, based on the manufacturing plans of Tesla, in partnership with Japan’s Panasonic, Samsung SDI and LG Chem of Korea and China’s BDY, that will increase global automotive lithium ion battery capacity fourfold. In a major study of the global battery market last September, Berenberg says the cost decline will erase the price premium of EVs compared to internal-combustion vehicles.
An intense battle over battery efficiency is playing out in the EV market, illustrating the ongoing competition between underlying technologies, and how the choice of a battery technology interacts with power and temperature control technologies to potentially affect long-run prospects in one of the world’s biggest new markets.
BNP Paribas analyst Pete Yu says the cost of the systems required to turn cells into a battery pack was as much as 50-100% of the cost of the cells for an EV in 2012. The casing and cabling also adds weight, which can reduce range. “Pack overhead costs will shrink to about 20-30% of cell costs in 2017 with improved design and scale benefits,” Yu projects, resulting from large-format batteries. “We expect large-format EV batteries to be the preferred choice for car OEMs [original equipment manufacturers],” Yu writes in a March report on the sector.
Yu has a buy rating on the four main players in the market, Panasonic, LG, SKI and Samsung SGI. Demand for longer-range EVs, means “the bar has been raised”, he says. “Major EV battery makers will increasingly gain more bargaining power over car makers as only a handful of EV battery companies can supply high energy density batteries with a safety track record.”
Whichever way the technologies are eventually sorted out, scientists and investment analysts say the goal isn’t one technology or another, but the combination that enables the production of an EV with significant range that can be sold for $20,000-30,000 per vehicle.
Battery ground zero
While manufacturers work to cut costs though greater scale, scientists are working on next-generation battery chemistry to pack more of a punch into each cell. And while battery manufacturing takes place in Asia, ground zero for battery science is the Joint Center for Energy Storage Research, JCESR, a partnership of academic, industrial and government researchers based at Argonne National Laboratory, a US Department of Energy facility managed by the University of Chicago.
The most promising of several new approaches to battery storage is one using lithium sulphur, says George Crabtree, an Argonne Distinguished Fellow and Director of JCESR. A battery stores electricity in chemicals, comprised of an electrolyte substance that connects the different chemicals in the anode, which releases electrons, and cathode, which accepts electrons. The reactions occur at the same time, creating electric current from the battery. While lithium will remain the fundamental element, Crabtree says researchers are working on a new solid-state battery that will rely on actual chemical bonds – in which the sharing of electron pairs between atoms enables storage of much more energy per lithium ion than current techniques – to boost the energy density of each cell.
It would also make the batteries safer by replacing liquid electrolyte with a solid. While it will take time to bring such a battery into commercial production, Crabtree estimates that in about two years the major components of the new technology will have been developed and proven at the lab.
There will be ample demand for whatever new storage techniques emerge from JCESR. In addition to electric cars and mass-transit, Berenberg estimates utilities are poised to start adding storage capacity at a brisk pace that will create a $14bn market by 2020. Battery storage will help utilities cope with increasing fluctuations in electricity supply that result from the increased use of renewable generation methods. Grid-level battery storage requirements have been rising globally, Berenberg says, led by the US, with lithium ion batteries expected to remain the dominant technology in the high-power sector as well.
Distributed storage for home and small business solar systems will develop more slowly, but be worth $8bn by 2026, says Cosmin Laslau, senior analyst at New York-based technology consultancy Lux Research. Partnerships between Stem and SunPower, Green Charge Networks and SunEdison, and Sonnen and Sungevity illustrate the convergence of solar generation and storage companies. And software is a key differentiator, Laslau says. SolarCity offers demand management software that integrates storage, Sunverge links to smart devices and electric vehicles, while Sonnenbatterie analyses weather data to optimise solar consumption and storage.
The focus on system-level applications is the next step for power semiconductor companies like Enphase, says RobecoSAM’s Lang. In March the company launched a new grid optimisation service, using its installed base of microinverters to give utilities and grid operators information about voltage and power flow in solar-rich circuits on a grid. Enphase has piloted the service with Hawaiian Electric, and in a sign that the new energy economy has arrived, the company has partnered with leading utility software vendors Esri and OSIsoft to integrate its solar intelligence into a commercial product for grid control.
At Argonne, JCESR’s Crabtree is keeping an eye on the grid as well. In addition to stabilising fluctuations resulting from the intermittent nature of solar and wind power, JCESR scientists are looking at applications like time shifting, in which utilities store large amounts of energy generated when it’s not needed, for instance on windy nights, to be released quickly during morning hours. Both are storage tasks, says Crabtree, “but you need two different kinds of batteries.”
Bigger may be better for electric vehicle batteries
There is more than one way to power up your electric vehicle (EV). Tesla uses a cylindrical battery cell like those in laptop computers, which currently has the highest energy density and lowest cost per kilowatt hour.
The battery for an EV is actually a number of individual cells. Managing the charge level, voltage, temperature and other technical factors in each cell is the role of the battery management system (BMS), and the more cells, the more complex the BMS and the tasks it must perform, and that adds to the cost of each EV. Cylindrical battery packs in Teslas use more than 7,100 cells, many more than next-generation large-format batteries.
BMW, for example, uses prismatic cells, which have a lower energy density at the cell level, but more efficiently transmit energy so that when its 96 cells are combined into a car battery pack the energy level is about the same as Tesla’s cylindrical pack. Another type, called the pouch, is used by GM and Nissan; some analysts estimate that the latest versions, which use between 190 and 290 cells, are about as efficient as the prismatic cells.
Larger format cells also have larger surface areas, which can help heat dissipate, reducing the risk of fires, lessening wear and tear on the battery itself, and limiting the complexity and cost of the thermal management systems integrated into battery packs to control temperature. Cylindrical batteries must be cooled by a special liquid glycol system, while prismatic cells can be cooled with the same refrigerant used in vehicle air-conditioning systems.