Navigant Research Blog

Sunflower Concentrating Solar: 2,000 Suns You Can Touch

— October 6, 2014

A concentrating solar photovoltaic (PV) design from a Swiss company called D Solar shows a promising blend of multiple technologies that concentrates the sun by a factor of 2,000 but keeps the resulting temperature below the boiling point of water.

Concentrating solar uses mirrors to reflect sunlight onto a small PV chip to create electricity or on a heat collection liquid to create thermal energy.  The D Solar system does both at the same time.  The new design, called Sunflower, merges advanced concrete engineering with low-cost optics and a water cooling system designed by IBM scientists to provide a cheap method of turning sunlight into electricity and hot water.

At the heart of the Sunflower system is a receiver on which the sunlight is concentrated.  Any attempt at concentrating sunlight onto a PV cell faces a fundamental problem: concentrated sunlight gets too hot for the PV chip.  By running water through the chip at a high rate of speed, that heat can be carried away.  But cooling such a system is an extremely complex engineering task that requires space-age ceramics, precise flow control, and sturdy pumps.  IBM has been working on thermal control of computer chips at data centers, and its engineers saw a use for their cooling technology in the concentrating PV space.

Many Mirrors

Another fundamental problem of concentrating PV is that the mirrors or lenses used to concentrate the sunlight are often as expensive as the PV chips themselves.  To get around this, the Sunflower system uses stretched membranes of reflective plastic.  The Sunflower system resembles a large satellite dish, but instead of a sheer dish, the reflective area comprises multiple round mirrors, each consisting of a stretched foil that’s focused by putting it under vacuum pressure.  The pressure of the vacuum can alter the direction in which the foil reflects sunlight.  The entire dish is then covered by a bubble of another thin film of transparent plastic, which keeps dust, birds, and rain off the reflectors.

The sunlight is reflected onto a central receiver that contains the PV chip and the water-immersed ceramic receiver.  The dish is held up on a pylon of low-cost concrete, making all the materials in the device (save for the square inch of high-efficiency PV) very low cost.

Heat and Power

One of the economic attractions of the design is that, in addition to producing electricity from the PV chip, it also produces a significant amount of hot water, which can then be used for space heating, industrial processes, or even desalination.  The value of the electricity and the thermal energy together means more income can be produced by the same device.

While D Solar isn’t providing any cost estimates for the system (the small-scale prototype has not been completed yet), it’s clear that the design has the potential to be an extremely low-cost method of producing solar power.  While there have been many attempts at designing concentrating PV systems, none have quite been as unique and creative as the Sunflower system.


A New Dawn for Lead Batteries

— October 2, 2014

Donald Sadoway, a materials scientist at the Massachusetts Institute of Technology, is considered one of the smartest and most creative battery scientists in the world.  So admired is Sadoway that, when former Microsoft CEO Bill Gates wanted to learn about batteries, he took Sadoway’s course.  Afterwards, he approached Sadoway and the two discussed the topic of how to rethink battery design from a blank page of paper.  That discussion led to the founding of Ambri, a startup company that is based on Sadoway’s ideas for how to build a better battery.  And at the heart of Ambri is Sadoway’s concept of a high-temperature liquid metal battery whose cathode and anode literally float one on top of each other.

Ambri’s first attempt at a prototype involved the metals antimony and magnesium.  The concept worked, but the high melting point of magnesium (650 degrees Celsius) and the relatively high cost of that material made the prototype battery too expensive to compete against lower-cost batteries like lithium ion and lead-acid.  So Sadoway and his research team kept working.  In a paper just published in the journal Nature, the team released the results of their second prototype, which uses an old standby material of the battery industry: lead.

Melting Point

The battery consists of three basic inputs: lithium salts, lead, and antimony.  The lithium serves as the anode, or negative electrode, which holds the energy in storage while the battery is being charged.  Alloyed together, the lead and the antimony form the cathode, or positive electrode, which releases electrons during the discharge of the battery.  Once the battery is heated so that the alloy mixture and the metallic lithium melt into liquids (which requires a temperature of 253 degrees Celsius), the battery can start cycling through charges and discharges.  The lower temperature means that there are fewer parasitic losses during cycling, which makes the battery more efficient (the paper claims a 73% round trip efficiency, which is similar to the efficiency of many flow battery technologies).

More interestingly, Sadoway’s team calculates that the cost of input materials for the battery would be a mere $68 per kWh, which compares favorably to almost every other battery chemistry.  Finally, the Nature paper shows that accelerated testing of the battery predicts that, after 10 years of daily 100% cycling, the battery will still have a usable capacity above 85% of the capacity the battery had when it went through its first charge/discharge cycle.  In that regard, it compares to accelerated testing of other high quality batteries.

Lead Leader

Will Ambri’s new battery take over the market share of the other incumbent battery technologies?  It’s not likely.  Because the battery needs to be kept at a high temperature, it won’t function well in situations that require maximum flexibility and uncertainty.  However, it will be an excellent choice for any application that requires a long-duration and highly consistent charge/discharge cycle.  Although that’s a niche of the overall stationary energy storage industry, it could eventually be a large one.  Decades from now, when people talk about lead batteries, they might just be referring to Ambri’s molten battery, not their car starter batteries.


Unknowns Narrow for Tesla’s Gigafactory

— July 31, 2014

Tesla Motors announced today that it has started civil engineering work at a site in Nevada for the eventual construction of its Gigafactory – a battery manufacturing plant that will produce 50 GWh of batteries a year.  Broadcast in a shareholder letter that accompanied Tesla’s quarterly earnings results, the announcement confirmed some rumors but was still extremely short on specifics.   A lot of uncertainties remain about how, where, and by when the Gigafactory will be built.

The first issue is site location.  Tesla has said in the past that it will build the factory in one of five states: California, Nevada, Arizona, New Mexico, or Texas.  It has also said that it will begin development work on more than one site, choosing the eventual location from multiple contenders upon which initial civil engineering work has already been done.  Now we know that the Nevada site outside of Reno is one of those finalists.  Where is/are the other/s?  No word from Tesla on that, but it is pretty easy to identify the five top contenders.  That’s because the Gigafactory will need to be on a main rail line that connects with the company’s Fremont, California automobile factory.  It will also need to be near a large population center in one of those five states.  That leaves the following contenders:

  • Central Valley, California
  • Tucson, Arizona
  • Albuquerque, New Mexico
  • El Paso, Texas
  • Austin/San Antonio, Texas

You can expect the second site to be in one of those areas.  There is still one potential curveball that might come  from Tesla – the possibility that the Gigafactory will be composed of multiple sites: maybe a separator factory in one state, an electrolyte factory in another locale, and a final assembly plant in another.

More to Come

The other piece of interesting information still to be determined is exactly how the Gigafactory will be structured.  No blueprint exists for how to design a factory that is owned by multiple parties; it’s a unique concept that has never been tried before.  One day earlier, Tesla said that Panasonic will definitely be the manufacturing partner for the Gigafactory.  Now the questions are: How will the ownership of the site and its equipment be divided, and who will be the other component manufacturing partners? Expect a number of announcements on that end to come out over the next several months.  Among the potential other manufacturing partners that Navigant Research expects to be chosen are a cathode material supplier (such as Nippon Denko or Umicore), a graphite supplier (Northern Graphite or Alabama Graphite), an electrolyte manufacturer (Ube, Sumitomo, or Nichia), and a separator manufacturer (Celgard, Ube, or Toray).  Other materials needed for the batteries, such as lithium carbonate, copper foil, and aluminum casings, will probably be made offsite and delivered by rail.

The final questions are when the Gigafactory will go online and when it will reach full capacity.  Tesla has already said that it hopes that those dates will be 2017 and 2020, respectively, but exactly how the ramp rate works will be interesting to see.  Panasonic has clearly stated that it will invest in the equipment for the factory in a staggered, conservative fashion.  That could lead to a much slower build-up to full capacity than the 3 years that Tesla is claiming.  Regardless of the details of the how, when, and where of the facility, Navigant Research believes strongly that the Gigafactory will be built and will be a successful, potentially revolutionary, manufacturing venture.


With Its E-Motorcycle, Harley-Davidson Outdoes the Automakers

— June 20, 2014

This week Harley-Davidson unveiled a battery-powered electric motorcycle concept, called the LiveWire, that has surprised the transportation world.  Crankshafts, engine oil, and the smell of gasoline are all integral parts of the Harley-Davidson identity, so most expected the company to be dragged kicking and screaming to the electric vehicle (EV) party.  Instead, Harley revealed a concept vehicle that’s better looking, better designed, and might one day be better selling than most automotive company attempts at EVs.

First, a personal detour: I’ve never been a fan of Harley-Davidson.  Its motorcycles are beautiful to behold and intimidating to hear.  But I’ve always disliked the design ethos of a company that decides to sacrifice performance in order to cosmetically alter the sound that a machine makes.  Harley-Davidson’s notorious engine rumble comes from the single pin engine design that can’t match other performance bikes on speed and torque.  Additionally, Harleys are tuned to have a low idle speed in order to give them a more distinct popping sound when not moving.  This, in turn, leads to excessive wear and tear on the engine, leading to lower expected lifetimes of Harley-Davidson engines compared to their peers.

Lag Gone

Which is why I was surprised to view the videos of the electric concept bike.  Instead of trying to market a machine for environmentalist two-wheelers, the company designed a bike for its core market: people who love motorcycles and the power and freedom they represent.  The drivetrain design provides 52 foot-pounds of torque, which for a vehicle that weighs less than 500 pounds is like strapping a Scud missile to a Smart Car.  The resulting power allows the machine to reach 60 miles per hour in less than 4 seconds.

Motorcycle reviewers note the other element of EV drivetrains that can’t be matched by internal combustion engines: instantaneous torque.  An electric motor responds to the driver’s command immediately, with zero torque lag.  Even the most expensive Ferrari or Lotus has a noticeable buildup to full torque, which is inherent in the nature of how combustion drivetrains work.  I’ve noticed instant torque when driving the Chevrolet Volt and Nissan LEAF – albeit much less dramatically than it probably feels on the Harley LiveWire or the Tesla Roadster.

Sound and Fury

Harley-Davidson has done something that no other incumbent vehicle manufacturer (we would expect EV-only companies like Tesla Motors, Brammo, and Zero to get these things right) has tried to do: it has built an EV around the strengths of the drivetrain.  There are, of course, weaknesses too (most notably a 30- to 60-mile range and a 3-hour recharge time), but in the end people like their motorcycles (and cars) for what they can do and are willing to live with the compromises that have to be made for what they can’t do.  Here’s hoping that the design executives from the major automakers take the LiveWire for a test ride and get inspired to make the next generation of muscle cars and hot rods – with electric drivetrains.

And how about the signature Harley-Davidson sound (which the company once unsuccessfully tried to trademark)? Well, this thing sure sounds different.  Thanks to the high frequency electric motor, it sounds more like a jet engine than a traditional Harley.  But it still makes you instantly respect and appreciate the power and fury that the bike is capable of.


Blog Articles

Most Recent

By Date


Clean Transportation, Electric Vehicles, Policy & Regulation, Renewable Energy, Smart Energy Practice, Smart Energy Program, Smart Grid Practice, Smart Transportation Practice, Smart Transportation Program, Utility Innovations

By Author

{"userID":"","pageName":"Sam Jaffe","path":"\/author\/sjaffe","date":"10\/23\/2014"}