...and Two Steps Back for Cleantech

• The Better Place bankruptcy ends an interesting effort to circumvent some big impediments to the wider adoption of electric vehicles.
• DESERTEC's original concept would have matched European solar investment with superior North African solar resources, but was no match for European politics.

Within the last week two of the previous decade's Big Ideas for accelerating the shift from fossil fuels to renewable energy--or at least to electricity generated from a variety of cleaner sources--have come up short.  On May 26th electric-vehicle-battery-swapping firm Better Place filed for bankruptcy liquidation in Israel, and just a few days later the DESERTEC Foundation reportedly "abandoned its strategy to export solar power generated from the Sahara to Europe".  Both of these concepts originally looked promising, and I take no satisfaction in their apparent failure.  However, these events must be telling us something.

Better Place was aimed squarely at two of the largest perceived barriers to wider acceptance of electric vehicles (EVs): the limited range of today's EV batteries and the relatively long times required to recharge them, compared to a typical three-minute fill-up at the gas pump.  Better Place's big idea involved the standardization of EV battery packs on a design that could be quickly removed from the vehicle and robotically replaced with a fully charged battery. This required large up-front investments in facilities and hardware, but the firm didn't fail for lack of capitalization. 

Despite having raised around $800 million since its founding in 2008, and convincing French carmaker Renault to produce vehicles designed to work with their technology, Better Place failed to standardize the emerging EV battery market.  Tesla used a different battery configuration from the start and has focused on its own fast-charging technology, while even Renault's global alliance partner Nissan didn't make compatibility with Better Place a standard feature of its Leaf EV in markets like the US or Australia. That led Better Place to invest in building more-conventional EV recharging networks to accommodate other EVs, diluting both its capital and its concept. 

I see two lessons here. First, EVs and related services are still a niche market, and in spite of its aspirations Better Place became a niche within this niche, largely dependent on the success of EV manufacturers at growing their potential market.  That's a poor place from which to launch a business that ultimately depends on achieving high volumes.  The other lesson is that when you can't make sense of a company's revenue and working-capital model, there's probably a good reason.  At this stage in their development, EV battery packs are apparently still too expensive to sit idle in large numbers, waiting for a swap, when the hardware to exchange them requires the same retail footprint as a car-repair bay--all this to support a service arguably only worth a few hundred dollars per year to an EV owner, compared to the normal cost of recharging.

DESERTEC's big idea was even simpler than Better Place's.  A well-sited solar array in North Africa would inherently generate at least twice as much electricity per year as the same array in Germany, the Netherlands, or Belgium.  All else being equal, it would make more sense to invest in solar where the sun shines brightly for more than 6 hours a day, on average, and to send it by wire to the cloudy, northern countries that want more green power.  Of course physics can't always trump politics, and I suspect that this has more to do with DESERTEC's withdrawal from its basic concept than the cited concerns about transmission capacity and grid congestion across Spain and France. 

Politics enter the story in two main ways.  Renewable energy in the EU is deeply entangled with industrial policy and green jobs. From that standpoint, it's even better if a PV panel in Germany produces half the output as one in Morocco, because you can sell twice as many, all installed by local firms and workers. Then there's the interaction between the EU's generous solar subsidies and the solar manufacturing incentives in Asia and elsewhere, resulting in enormous overcapacity, relative to demand, and a now-global wave of solar bankruptcies and defaults.  This has pushed PV module prices down to a level at which the other costs of solar energy, including installation and transmission, begin to outweigh the module costs. That erodes North Africa's solar advantage relative to its northern neighbors. Throw in the lingering effects of the financial crisis, and a once-big idea looks like an unworkable dead end, at least for now.

Neither the failure of Better Place, which might yet find a bargain-hunting savior, nor the retreat of DESERTEC looks like a mortal blow to the long energy transition now underway.  However, they do suggest that the timeline is a little less likely to be shortened by the kinds of big leaps they offered.  EVs will have to gain market share the hard way, with better, cheaper batteries and ample recharging infrastructure--plus continued taxpayer subsidies--while inefficient solar subsidies continue to divert investment away from some of the world's best renewable energy resources, keeping the technology's global contribution smaller for longer.    

Temperature Extremes and EV Battery Trade-offs

The first production-model Nissan Leaf electric vehicle is scheduled to be delivered to a customer in the San Francisco Bay Area tomorrow. I know if I were on the receiving end, I'd be as excited as a kid on Christmas morning, particularly in a place where having the first Leaf will score its owner many green points. However, if the assessment by MIT's Technology Review of Nissan's choices concerning the temperature control of the Leaf's battery pack is accurate, then it's probably just as well that the first one is going to a location with such a benevolent climate, instead of the Midwest, upstate New York, or the desert Southwest. Batteries are sensitive to external temperature, in terms of both performance and longevity, and Nissan appears to be betting that making the battery simpler to replace is a higher priority than optimizing its condition at all times, as GM has done for the battery pack in the Chevrolet Volt.

It's easy to forget that batteries are fundamentally chemical, rather than just electronic devices. The chemical reactions in a battery absorb or release heat during the charge/discharge cycle, and the capacity of the battery's environment to accommodate those heat flows can affect these reactions. For a battery pack storing and delivering as much energy as required to run a car, these interactions are significant, and early adopters of EVs are already learning that the range of EVs becomes more limited in hot or cold weather. It's not as clear that they understand the degree to which extreme temperatures can degrade battery life. The economics of an EV could look very different if a battery pack only lasted six or seven years, instead of ten.

As the article explains, GM chose a liquid cooling system for the battery pack in its Volt range-extended EV. This system cools or heats all of the battery's cells, as necessary, and sometimes draws power for this purpose even when the vehicle is parked, as I learned when I test-drove one with the Volt's Vehicle Line Director last winter. According to him, GM's design team knew it had to go to extraordinary lengths to ensure the battery would perform reliably and last the expected ten years or 150,000 miles. Nissan appears to have taken a different path to battery management, providing a cooling fan for the battery pack and an optional battery heater--an option reportedly not available on the first Leafs. You don't have to be an expert in heat transfer to guess that air won't move heat around the battery pack's cells as well as liquid can, and that as a result, at least part of the Leaf's battery could potentially be exposed to more heat and cold--and possibly suffer more performance impact from them--than the Volt's.

That trade-off might reflect a different vision for how the battery will be used. Nissan (with its alliance partner Renault) is the main carmaker working with Better Place, Shai Agassi's EV battery recharging-and-exchanging start-up. A battery pack with only electrical connections to the car will be much easier and neater to swap in and out than one with liquid hoses running to a radiator and heater. This situation wouldn't even be a consideration for the Volt, which has an onboard generator to take over when the battery's charge falls too low. But for battery-only EVs, battery-swapping is as close as they can get to replicating the convenience of refueling a gasoline or diesel car in a few minutes. If EVs catch on via a business model like Better Place's, in which consumers routinely exchange their flat batteries for fully-charged ones (and might not even own the battery pack, but instead rent it by the month or the mile) any shortcomings from Nissan's less robust battery-conditioning strategy would fall on someone other than the consumer, as a statistical cost of doing business.

From my perspective this is just one of the uncertainties concerning the operation and consumer acceptance of EVs about which we'll learn more as their numbers climb from the low thousands to the hundreds of thousands and millions. However, I find it interesting that few journalists have picked up on an issue that could have far more impact on the EV ownership experience than the tempest in a teapot that some stirred up when they found out that the Volt's wheels are occasionally driven partly by the engine-generator, rather than entirely electrically. If I were buying one of these cars, I'd be a lot more interested in how far its expensive battery pack will carry me and how long it will last, than in whether the car is truly a range-extended EV or just a plug-in hybrid.

EVs and Energy Density

If the new vehicles on display at this year's Detroit Auto Show have you wondering whether 2010 might be the Year of the Electric Car, you're not alone. GM's Volt plug-in hybrid is due out this fall, and purely-electric options like Nissan's Leaf aren't far behind. The global auto industry is investing billions of dollars in developing this technology, and the US government is putting up additional billions in loan guarantees for EV manufacturers and consumer purchase subsidies. No one should dismiss the seriousness of these efforts or their potential to reshape the vehicle and transportation energy markets over the next couple of decades. At the same time, their ultimate success depends on whether a combination of improved technology and significant changes in consumer expectations concerning vehicle performance and characteristics can overcome the core challenge of vehicle electrification: either matching the effective energy density of liquid fuels or giving up the flexibility they provide.

Understanding the practical consequences of energy density, which refers to the amount of energy that can be stored in a given volume or mass of fuel or battery, requires putting electricity and fuels onto a common basis of comparison. Although I've generally tended to do this in terms of gallons, barrels or BTUs, for a change I'd like to consider the fuels we commonly use in terms of their equivalent electrical energy. The units may be less familiar at first, but this should make a side-by-side comparison with the battery capacities of new electric vehicles (EVs) easier.

According to the Department of Energy a typical gallon of gasoline delivers 116,000 BTUs of energy, and a gallon of diesel fuel 128,000 BTUs, based on their lower heating values. Converting to electricity units gives us 34 kilowatt-hours (kWh) per gallon and 37.5 kWh/gal., respectively. Using typical volumetric densities for these fuels, I come up with figures of 5.5 kWh/lb. for gasoline and 5.3 kWh/lb. for diesel. By comparison, the battery for the extended-range GM Volt hybrid, which is rated at 16 kWh, appears to weigh 400 lb., yielding an energy density of just 0.04 kWh/lb., or less than 1% of the energy density of hydrocarbon fuels. If this were the entire story, EVs would look like a hopeless proposition, and we could dismiss them for another generation.

The factor that helps to bridge the enormous gap in energy density between the best batteries and liquid fuels is efficiency. While neither electric motors nor internal combustion engines (ICEs) can turn 100% of that stored energy into motion, the EV motor has an efficiency advantage of roughly 4:1 over ICEs. Even after taking that into account, we're still left with a requirement for roughly 25 lb. of batteries to deliver the same range as a pound of gasoline, with the effective useful capacity of the Volt's entire battery pack storing the equivalent of no more than one gallon of unleaded regular. Plug-in hybrids like the Volt cleverly finesse this limitation by using on-board generators running on liquid fuels to extend their range. Of course this entails big trade-offs of cost and weight, but the designers of such vehicles hope to come up with a mix that will satisfy consumers who are accustomed to cars that can go 300 miles without provoking "range anxiety".

In some respects the bigger concern related to energy density might be the one that proved to be the Achilles' heel of GM's first effort to produce a consumer-friendly electric car, the EV-1. To understand why recharging EVs is such a tough problem, let's take a look at your last visit to the gas pump in terms that would never occur to most people. Gas pumps in the US are limited by EPA regulations to deliver a maximum of 10 gallons per minute. Half that is probably more typical. But even at 5 gallons per minute, the gas pump is "recharging" your car at the power equivalent of 10 megawatts (MW), effectively delivering the entire daily power consumption of the average US household every 12 seconds. Even if you discount that figure by the lower conversion efficiency of an internal combustion engine, it's still the equivalent of a couple of megawatts. Matching that for an EV would require either stupendous voltages or currents well above most designers' comfort level. For example, a car recharger drawing 100 amps would have to operate at 25,000 Volts--more than ten time the voltage of the electric chair--to deliver a comparable charge in the same interval. At the 240 V of your home's appliance circuit, you'd need about 10,000 amps--similar to what a transit train draws from the "third rail." Almost inevitably, the safe recharging of EV batteries must take longer--hours longer--than refueling your gasoline vehicle, or entail clever-but-costly workarounds such as the battery-swapping scheme of Better Place and other firms.

From the above it's hard to avoid the conclusion that EVs and plug-ins might not be quite ready for prime time. However, I was struck by a comment from a GM official cited in a New York Times article on the Detroit Auto Show, concerning the need for first-generation EVs to pave the way for an eventual mass market. There's every indication that these cars will shortly be ready for "innovators" and "early adopters." The Volt, Leaf, and cars like them will prove out not just the technology of vehicle electrification--a trend that began with the original Honda Insight and Toyota Prius and still looks like the strongest competitor to the ICE in the long run--but also the response of real drivers who aren't engineers or environmentalists. My own experience with energy density in the more modest realm of battery-powered lawnmowers suggests that this will require adapting our expectations and usage patterns to this new vehicle type, rather than treating it as plug-and-play in our current lifestyles. In the meantime, the automotive mainstream has some very attractive non-plug-in options for getting the most out of the energy density of our current fuels, based on the steadily-growing variety of conventional hybrids, advanced diesels and downsized gasoline cars with direct injection and other innovations.

Regulating EV Recharging

A feature on the New York Times website tipped me off to a debate that's brewing in California concerning whether and how the state's Public Utilities Commission (PUC) should regulate facilities and firms that will recharge the electric vehicles expected to dot California's roads within a few years. From my own experience in attempting to involve my former employer in the recharging infrastructure for the old GM EV-1 in the late 1990s, I knew this wouldn't be a simple matter, but I had little appreciation for the complexities that have emerged in the last decade. How this gets resolved will have enormous implications for automakers and incumbent utilities, as well as for start-ups such as Better Place that some would like to treat as regulated utilities.

The discussion with the PUC hinges on some very thorny questions: Is a company that buys electricity for resale to consumers for the purpose of recharging electric vehicles--which takes in both battery-electric vehicles and plug-in hybrids--more like a utility or a gasoline distributor or retailer? Who should pay for installing recharging facilities, and how--and from whom--should these parties recover their investment? Should a consumer who already uses large quantities of electricity at home and pays at the top rate tier, which can hit $0.40/kWh in some areas, qualify for discounted power to recharge an EV? How should a customer be billed when recharging outside the service area of the utility from which he normally buys power? The list of such questions is long, and looming behind them are larger questions about how best to gauge the effect of EV recharging on greenhouse gas emissions and air quality concerns, and to manage its impact on the regional generating mix, and on grid stability and reliability. Many EV advocates assume that EVs are inherently grid-stabilizing and renewable power-enabling, though it's not hard to construct scenarios in which the opposite could be equally true, if they're not implemented properly.

The emissions aspect becomes even more interesting in light of the views I saw expressed in a PUC filing by Tesla Motors, Inc., a Silicon Valley manufacturer of high-end electric sports cars that recently qualified for a half-billion dollars in low-interest expansion loans from the federal government. Tesla sees the generation of tradable credits under either cap & trade or the state's Low-Carbon Fuel Standard as a significant source of revenue for the owners of EV recharging facilities, and they might be right, though when I converted the federal estimates of emission allowance values under Waxman-Markey of around $15/ton of CO2 to cents per kilowatt-hour, using California's natural gas-dominated average generating mix, I came up with a value of less than a penny per kWh. I have to wonder how excited utilities will be to take on the cost and risk of putting in EV rechargers for such a small reward, if they can't also make a profit selling power to EV drivers.

The whole notion of regulating resellers of electricity to EVs as utilities also raises serious questions about the alternative business models now under consideration by companies such as Better Place. Would offering EV services on a cents-per-mile basis, rather than cents per kWh, be deemed sufficiently transparent, and would they have to negotiate their profit margins and investment recovery with the PUC? That sounds like a great way to make it harder for anyone new to the scene to compete with traditional utilities in this area.

Fairly soon the California PUC will resolve most of these questions and in the process largely define the environment in which EVs will emerge in the biggest early market for them in the US, potentially setting the standards for their use throughout the US and beyond. I don't have a horse in this race, but I will be watching the outcome with great interest.

Plug and Pay

Yesterday's photo-op at an Indiana RV factory for the purpose of announcing more federal assistance for the electric vehicle industry came just a few days after Nissan debuted its Leaf electric car, which might become the first mass-market EV in the world. Cars powered by batteries alone or a combination of batteries and conventional engines look like one of the most promising long-term solutions to the dual problems of energy security and climate change. But precisely because of their potential to have such a large impact, it's vital that the economic arrangements for their energy consumption are put on the right basis from the start. Among other things, that means avoiding the temptation to provide free public recharging for them. If we get this wrong, we risk negating much of the energy and greenhouse gas benefit these cars offer. We could also inadvertently deter the substantial private investment in recharging infrastructure that would be needed to make EVs fully competitive with cars running on liquid fuels.

Against the backdrop of $2.4 billion in new subsidies for EV and battery manufacturers and federal electric vehicle tax credits ranging up to $7,500 per car, my concerns about collecting for the electricity actually used by the first few mass-production EVs might seem disproportionate or even eccentric. After all, how much juice can a few battery cars use, compared to our factories, office buildings, and billions of home appliances? Initially, very little and eventually still less than you might imagine. If every vehicle-mile traveled in the US were driven in an EV averaging 3 miles per kilowatt-hour (kWh), US electricity consumption would only increase by about 27%. The impact on emissions is much harder to assess, however, since it depends heavily on which generating technologies deliver the power used by EVs, and that in turn depends to a large degree on the time of day when they are recharged. Charge up at 3 AM, and you might be getting zero-emission wind power that would otherwise go to waste. Charge up at 3 PM, and you are almost certainly going to be drawing on a gas turbine somewhere--probably a fairly inefficient "peaking" unit--or a coal power plant. To put that in perspective, let's look at the emissions from two comparable cars, under both scenarios.

For our baseline, consider a Prius-type hybrid that gets all of its energy from the fuel that goes into its tank. At 50 mpg, its emissions from gasoline amount to roughly 40 lb. of CO2 per 100 miles. For an EV getting 4 miles per kWh and recharged with wind power, they would be essentially zero. However, the same car recharging during mid-peak or peak electricity demand would trigger power plant emissions between 35 lb. ("peaker" turbine @ 12,000 BTU/kWh on natural gas) and 53 lb. (average US coal plant) for every 100 miles. In other words, while the hidden emissions from an EV would in the worst case still be lower than those of the average car in America today (around 80 lb. CO2/100 mi.), they could be substantially higher than from an ordinary hybrid that never plugs in. So if we want EVs to repay the substantial national investment we're making in them by reducing our fossil fuel consumption and greenhouse gas emissions, we will want them to recharge as little as possible during daylight hours, particularly in the late afternoon, at least until wind, solar and geothermal power account for a much higher share of our annual electricity generation than the 1.6% they contributed last year.

Paying for the electricity to recharge plug-in electric vehicles involves major cultural and behavioral shifts. The price of gasoline is one of the most visible, ubiquitous and transparent prices in our society. You stand at the pump and see the dollars going into your tank. But when you recharge an EV at home, unless you have a separate electric meter, you're going to have to sift through a power bill with a welter of distribution, fuel and non-fuel supply charges plus various state and local taxes and fees to see what it actually cost. At the current national average rate of around $0.11/kWh, a typical driver might only see an extra $27 a month, a big savings compared to the typical gasoline bill even at the current $2.55/gal. The extra power cost could easily get lost in seasonal usage fluctuations and rate changes. The impact would likely be more noticeable for utility customers in places with sharply graduated rate structures or time-of-use rates. For many people, however, even if they don't charge up using someone else's electricity--their employer's, their town's, or the local Starbucks'--it could look nearly free.

That would have implications for companies that are building vehicle recharging infrastructure that would need to recoup their investment on a per-kWh basis or, like Better Place, charges per mile of usage in a manner similar to cellphone service contracts. Those investments won't happen and the companies involved will go out of business if consumers regard the electricity for their new plug-in vehicles as effectively free and resist paying as they now do for fuel.

How this will all turn out is anyone's guess at this point, and I emphasize "guess." Until there are at least hundreds of thousands of these vehicles on the road, in the hands of many ordinary consumers and not just unrepresentative deep-green or "gear-head" early adopters, we can only make assumptions about how they will really be used. Still, it seems safe to predict that recharging that was free or regarded as free would get used more, resulting in more trips, more miles traveled, and eventually more energy consumption and emissions.