Hydrogen: How Shall We Store It?
By Bill Moore
Storing hydrogen is really pretty simple. Like any gas, you can compress it or cryogenically liquify it. There are even certain combinations of metals that act like sponges that will soak it up and when you apply some heat, they release it. There are also chemicals that can be stimulated to release pure hydrogen that fuel cell membranes love. And someday it might actually be possible to create artificial carbon sponges that could be both cheap and relatively light.
As was demonstrated at the SAE Hydrogen and Fuel Cell workshop in Sacramento Feb 18-19th, 2004, we don't want to solutions. Instead, what we want for is the perfect solution, which for the moment doesn't exist. All have their shortcomings, the principle one being the incontrovertible fact of physics that hydrogen, by volume, has only a fraction of the energy density of gasoline, the standard by which all alternative fuels are measured. Yet, progress is being made and here's what EV World learned.
Four presenters discussed each of the major hydrogen storage mediums in development today. John Williams with Quantum Technologies discussed the status of compressed hydrogen tank storage. Rex Luzader -- who also did double duty as a session moderator, in addition to his role as representing Millennium Cell -- talked about chemical hydride storage. Benjamin Chao, who works for Texaco Ovonics Hydrogen Systems, focused on metal hydride storage. And perhaps most intriguing presentation of all was NREL's Philip Parilla on the perplexing science behind carbon nanotubes, that-much-hoped-for breakthrough storage system that remains so frustratingly elusive.
As you might expect, with the exception of Parilla, each presentation tended to focus on the efforts of each sponsoring company, which is understandable, although SAE encouraged each to keep their company 'pitches' to a minimum, a rule that was somewhat loosely applied. Still, the information presented for informative and valuable, as was the opportunity to ask follow up questions, some occasionally "hardballs."
Compressed Hydrogen Storage
With rare exception, nearly all current prototype fuel and hydrogen-fueled ICE R&D vehicles use compressed hydrogen storage technology. It is the most familiar and currently the cheapest storage method. The auto industry has used compressed natural gas, LPG and propane for decades. Honda continues to offer to consumers a CNG version of its popular Civic, a vehicle rated at by the EPA and others as one of the cleanest production vehicles overall, despite burning a fossil fuel. So, compressing hydrogen for automotive use is a natural next step in the evolution.
A compressed hydrogen tank superficially resembles its fossil fuel counterparts, being typically cylindrical in shape. It is here, however, that similarities end. Where a CNG or propane tank can be made relatively cheaply from steel, modern hydrogen tanks are a sophisticated amalgam of high-tech carbon fibers wound about a hydrogen-impervious cylinder of aluminum or carbon composite with a high molecular weight polymer liner, as in Quantum's current generation of high-pressure tanks.
Why the difference? It's the energy density issue again.
According to a recent article in The Industrial Physicist, "One kilogram of hydrogen stored in common laboratory gas cylinders at 2,200 psi occupies 91.2 liter (1.6 MJ/l, hydrogen basis—the effective energy density in a storage system will be substantially lower). For comparison, a mere 8.2 liter of gasoline carries the same energy. Hydrogen tanks of 5,000 and 10,000 psi are being developed, but even at 10,000 psi, the volume of hydrogen is 27 l/kg (5.3 MJ/l, hydrogen basis)."
Th above article also contains a very illustrative graph that plots the volumetric to gravimetric density for gasoline and the major hydrogen storage technologies. It dramatically demonstrates the challenge confronting the hydrogen economy and why there is the on-going push to increase tank pressures, now at 10,000 psi.
John William’s presentation offered another way to look at the challenge. To store 5 kg of hydrogen, which is roughly equivalent to 5 gallons of gasoline, a Type I steel tank would have a mass of nearly 400 kg. A modern Type III tank with aluminum liner and carbon fiber wrap would have a mass of 150 kg. Quantum’s latest generation of 10,000 psi tanks drops this to about 65 kg. A new "Space" tank currently in development at Quantum would cut this to about 40 kg, one-tenth the mass of the Type I steel tank.
But safely compressing sufficient quantities of hydrogen into higher pressure tanks is only part of the challenge. Whereas gasoline can be stored in cheap-to-press steel tanks, the material costs of a compressed hydrogen tank are significant, though Williams didn’t talk about specific costs, preferring instead to say Quantum was not interested in selling individual components to garage hobbyists and experimenters. The company takes the view that what it has to offer OEMs is an integrated development system that includes storage, safety, regulation and mounting systems.
He did note, however, that carbon fiber represents the lion’s share of the costs for compressed hydrogen tank, something on the order of between 40-80 percent, depending on the class of fiber used. Fully 90 percent of the cost of a compressed hydrogen tank is its carbon fiber and stainless steel fittings.
Safety testing constitutes another major cost of compressed hydrogen storage. According to Williams, each type of tank the firm develops must be subjected to 24 different types of tests including: hydrotatic burst, extreme temperature cycle, ambient cycle, acid environment, bonfire, gunfire penetration, flaw tolerance, accelerated stress, drop test, etc., etc. Not only must 24 tanks be destroyed for each tank the company wants to build, but similar tests must be conducted for seven different world certification bodies. For this reason, Quantum -- and presumably other tank manufacturers -- would prefer the industry settle on a common standard in tank size and shape, rather than building custom tanks to fit a specific vehicle.
Williams presented an interesting slide that depicted DOE hydrogen storage goals which included improving the usable energy density (kw hr/L) from 1.2 in 2005 to 2.7 by 2015; reducing the costs in dollars/kw hr from $6 to $2; improving tank cycle life from 500 fueling cycles to 1,500 cycles and to shorten the refueling time from 0.5 kg H2/minute to 2 kg H2/minute, all by 2015. All of these are daunting challenges, but Williams was confident that Quantum is on tack to meet the 2005 milestones, noting that in 2002, for example, the company had successfully demonstrated a 3-minute 10,000 psi "fast-fill." In 2003, it produced a 5,000 psi tank aerospace with a 13.36 wt% tank efficiency, as well as achieving greater than 45,000 cycle fatigue life and a 30% cost-reduction in the design and process of its 10,000 psi tanks.
In concluding, he was confident Quantum could meet the DOE’s 2005 goals. Going beyond 2005, however, its uncertain how the DOE’s ambitious goals will be achieved using compression technology. It’s an acceptable short-term approach to hydrogen storage for handfuls of prototype vehicles with ranges between 100 and 200 miles. But to meet the 2015 goals, we’ll probably have to look elsewhere.
Chemical Hydride Storage
Which brings us to sodium borohydride, an ingenious method for storing hydrogen in a chemical form that can be released on demand. Rex Luzader, Vice President of Sales for Millennium Cell, the current leader in this technology, used his presentation to explain the principle behind the company’s chemical storage system, which has been installed in a DaimlerChrysler minivan. The company is also working with Peugeot/Citroen on integrating their hydrogen fuel storage system into prototype vehicles in Europe.
Unlike other hydrogen storage systems, Millennium Cell’s sodium borohydride (NaBH4) is a closed-loop system where a liquid solution is passed through a catalyst that releases eight hydrogen atoms, four from the sodium borohydride and four from the water portion of the solution. The "waste" borax (NaBO2) is stored on-board the vehicle and can later be recycled, while the hydrogen is used to power the vehicle’s fuel cell. The water generated by the fuel cell is recycled though the catalyst where more hydrogen is pulled off. It is a very neat system that offers a way around the energy density bottle neck of compressed hydrogen.
One of Luzader’s graphs compared the volumetric storage efficiency of a sodium borohydride system compared to compressed and liquid hydrogen storage. At 30 wt%, NaBH4 equates to about 63 grams of H2 per liter, just 8 grams less than cryogenic liquid hydrogen at 71g H2/L. By contrast 10,000 psi hydrogen equals 39g H2/L and 5,000 psi translates into a mere 23g H2/L.
In the DaimlerChrysler "Natrium" minivan, first unveiled in 2001, the company was able to replace the equivalent of three, bulky 10,000 psi compressed hydrogen tanks that would have taken up all of the vehicle’s available cargo space with a single Millennium Cell tank mounted under the vehicle where the gasoline tank usually goes. The system stores 40 gallons of solution, containing 10 kilograms of hydrogen. The result is a vehicle that not only has all the space people buy minivans for, but also has a range of 300 miles, comparable to its gasoline competitor.
Not only does the Millenium Cell chemical storage system offer a good compromise for achieving acceptable consumer range, it is also safe; the sodium borohydride solution is non-flamable and under no pressure, a concern that will always linger with compressed hydrogen. In addition, Luzader announced that the company is researching the development of a solid sodium borohydride system for both mobile and stationary power applications. A solid NaBH4 system solves a number of concerns including heat instabilities in the system and excessive water vapor. Shifting to a solid system will also reduce the volume of the storage system, improving overall efficiency.
So, what the drawback? The biggest problems is what to do with the "waste" borax that is the by-product of the catalytic reaction. It’s a valuable resource that can be recycled and reprocessed into more NaBH4, so you don’t want to simply "exhaust" it into the environment. Instead it is stored on the vehicle in a displacement storage system that replaces depleted NaBH4 with spent NaBO2. Theoretically, it can be reclaimed during the refueling cycle and reprocessed either on-site or at a central processing center.
It is this part of the cycle that is the Achilles heel of this system. With rare exceptions, very few recycling programs have been successful enough to compete with our "through away" economic system that depends cheap resources, cheap energy and lots of landfills. For the Natrium system to work -- and it is possible -- it means not just a new infrastructure to dispense, collect and reprocess the fuel, which itself takes energy, it also requires fundamental changes in our social patterns and economic system.
Assuming we as a culture can adapt to life in what Bill McDonough calls a "cradle-to-cradle" manufacturing system, and we learn to recycle a fuel like sodium borohydride, the next question is how do you recycle the fuel?
According to Luzader, NaBH4 currently is processed using large energy inputs from natural gas. Millennium Cell will be studying, under a DOE grant, electrochemical methods of reprocessing that can make use of electricity from either conventional or renewable sources to remanufacture the spent borax back into fuel. The goal is $1 for a kilogram of sodium borohydride, which equates to $5 per kilogram of hydrogen.
Despite on-going development programs with DaimlerChrysler and Peugeot/Citroen, Millennium Cell now recognizes it has to look for other business opportunities that will fit our current "cradle-to-grave" culture. So, the company is developing portable power modules for the telecomm industry that can provide backup power for remote communications towers and cellphone systems. They are also looking at micro-fuel cell solutions for powering laptop computers that could replace lithium ion batteries, giving the user 8-10 hrs operating time. Luzader envisions someday supplying the telecomm industry with Millennium Cell solid fuel modules that can be "hot-swapped" to provide uninterrupted power, but to do so, the company needs to partner with micro and small-scale fuel cell manufacturers, a search that apparently is still on going. The target here is to provide a 5 kW standby power fuel cell with between 18 to 40 hours continuous run time compared to battery backups that provide only about 4-6 hours of standby power.
Luzader also showed a graph that compared the 10 year lifetime costs of a battery backup system compared to a fuel cell/Millennium Cell combination that showed a savings of close to $8,000US. While the initial purchase price of the fuel cell system was about $5,000 higher than the battery UPS, the replacement costs of batteries every three years meant that after four and half years, the fuel cell system was cost competitive and by year 10, had saved substantially. So, it’s a model that could make sense economically, which is were the "rubber" always meets the road with any new technology.
Metal Hydride Storage
If Millennium Cell is looking to replace batteries, battery maker Ovonics is looking to store hydrogen using its long established expertise in metal hydrides. It was Energy Conversion Devices, the parent company of Texaco Ovonic Hydrogen Systems (2H2), whose patented advanced nickel metal hydride (NiMH) battery technology revolutionized the portable power market, making possible modern laptop computers, digital cameras and cellular telephones. But the market for larger, automotive batteries simply never materialized for the company, with GM eventually selling its share of GM Ovonics to Texaco.
Now a spin-off of that partnership is looking to exploit the company’s pioneering efforts in hydrogen storage in metal "sponges." Where chemical hydride storage presents recycling challenges, metal hydrides don’t. 2H2’s current generation of metal hydride storage operates at relatively modest -- as these things go -- 1,500 psi. But the beauty is since the hydrogen is locked in an atomic bond with the metal lattice inside the tank, even if the tank is ruptured, the hydrogen won’t explode or burn, making it incredibly safe.
Benjamin Chao heads the company’s development of metal alloys looking for just the right blend of materials to store the most hydrogen at the lowest possible cost. His presentation, originally developed by Dr. Rosa Young, looked at the current state of metal hydride technology. His talk began with a brief discussion of the shortcomings of compressed, liquid and solid (chemical hydride) hydrogen storage. He noted that metal hydride’s chief strength is its reversibility, unlike Millennium Cell’s borax waste problem. Hydrogen can be pumped into the tank and drawn back out over and over again, with the application of a little waste heat from the fuel cell.
He acknowledged early on, however, that weight is a considerable challenge. In one slide he showed two comparable-looking hydrogen storage cylinders. The 5,000 psi carbon-wound tank weighed between 25-30 kg but held just 0.78 kg of hydrogen. The 190 kilogram Ovonic tank (1,500 psi) stored 3.0 kg of hydrogen, or about the equivalent of 3 gallons of gasoline. In a fuel cell vehicle that gets 45 miles per kg of hydrogen, that translates into a modest 135 miles range, not overly impressive, but remember this is just one tank and most prototype fuel cell vehicles have at least three or more compressed hydrogen tanks. Multiply this by three and the range moves to a more acceptable 405 miles.
Besides weight -- and certainly cost -- another issue to be overcome is refueling time. According to Chao, it takes 10 minutes and up to 1,750 psi to add 2.7 kg of their tanks. That works out to be a little more than thirty minutes to refuel that 400 mile-range fuel cell SUV or sedan. He added, though, that the company is aiming to reduce this to 5 minutes for every 2.7 kg.
There is another issue related to refueling that Chao’s presentation raised and that’s the necessity to cryogenically cool the tank while the vehicle is being fueled. In order for the current generation of hydrides to do their magic absorbing hydrogen, they need to be super-cooled, so there are three connectors instead of one on the company’s modified 2002 Toyota Prius. The main connector is for the hydrogen, while the other two are for the coolant inlet and outlet. EV World wasn’t able to determine coolant temperature, other than a reference to -20C on the company web site. This implies that the 2H2 tank is a pressurized thermos design, adding some cost.
In order to extract hydrogen from the 2H2 tank, waste heat from the ICE engine or fuel cell is forced through an internal heat exchanger that warms the hydrides, which in turn release the stored hydrogen. Basically, we’re talking about a rather complex, heavy storage system, though that’s not prevented the company from actively promoting it by converting a Toyota Prius to run on hydrogen. It showed up at the 2003 Michelin Challenge Bibendum with the car. While the tank occupied most of the car’s trunk space, the conversion did end up comparing favorably to the car’s original performance while dramatically reducing its overall emissions.
The turbo-charged, hydrogen-burning Prius turned in an EPA city economy rating of 42.3mi/kg compared to 42.3mpg on gasoline. EPA highway milage was 46.1mi/kg. EPA combined mileage was 43.9 on hydrogen and 42.8 on gasoline, while fuel economy at a steady 55 mph came in at an impressive 50.9mi/kg. The range of the vehicle at that speed and with 2.69 kg of H2 works out to be 137 miles. The weight of the car increased from 2800 pounds to 3250 pounds.
Even more encouraging, emissions of hydrocarbons dropped from 0.010 g/mile on the stock Prius to just 0.001 g/mile. Carbon monoxide went from 0.386 g/mi to 0.001. NOx did increase slightly from 0.004 to 0.018, but CO2 went from 222.8 grams per mile to a mere 2.5 g/mile.
So, clearly, the technology does offer some positive features, Yet, even as the blend of alloys continues to evolve in the search for just the right mixture, the system will continue to be relatively heavy for the foreseeable future. As with the other presenters, no prices were discussed, though a comparable system being developed by a small New Mexico firm suggests a possible range of between $8,000 and $15,000 per tank, though mass production could certainly bring that down.
Carbon Nanotube Storage
If there is a Holy Grail of hydrogen storage it is the potential of carbon nanotubes, those mysterious microscopic particles that many hope will make light-weight, compact, affordable hydrogen storage a reality. But like the Holy Grail, there appears to be more myth, rumor and controversy surrounding this material than reality.
We were fortunate to have Dr. Philip Parilla, one of the world’s leading experts on single-wall carbon nanotubes (SWNT), present this fascinating medium. A senior scientist at the National Renewable Energy Lab, Parilla was a key figure in the discovery of the first true inorganic fullerene and hydrogen storage on carbon nanotubes. His discovery was published in Nature, one of the world’s most respected science journals.
What quickly surfaced from his presentation, that last one on hydrogen storage, was that exploitation of carbon nanotubes may be years, if not decades away, mainly because we simply haven’t the understanding nor the technology to manufacture large, uniform quantities of the material. How hydrogen is stored -- even if it is stored -- on SWNT is still open to scientific debate, though we appear to be getting a better grasp of what’s happening at the molecular level and why.
Carbon nanotubes are a special type of carbon that inexplicably forms into fairly uniform tubes that appear to be ideal for storing hydrogen atoms. Up to this point, only very small samples of the material can be produced and often not consistently, which may account for why test results have varied so dramatically. Some researchers has reported small volumes hydrogen stored in the material, while others have seen none. Parilla’s work, and that of others, has been to determine why this is so, eventually leading to a 2005-2006 decision by the federal government as to whether or not nanotubes offer a viable storage pathway in the future.
Perhaps the most promising development has been the introduction of laser vaporization, which appears to make higher portions of nanotubes than the original arc-discharge method. Parilla and his colleagues estimate that while arc-discharge creates tubes that can store less than 1 wt% SWNT, the laser method can create 25-30 wt% SWNTs.
What they also discovered is that the key to increasing the percentage of hydrogen storage appears to be the controlled "contamination" of the material by various metals including titanium, creating what appears to be a hybrid material, though that’s not entirely certain at this point. By varying the metal alloys and their ratios, it appears you can increase or decrease the hydrogen storage capacity.
According to Parilla, the interaction of metal alloys and carbon is quickly emerging as a promising new field of investigation. It is evident from his research and others that this alloy/carbon interaction appears to be the key to this technology, but it is not well understood and needs further study.
So, of the four storage mediums, carbon nanotubes are the furthest out and most uncertain. If we can understand it and learn to control it, it could offer the potential of storing up to 8 wt% in a relatively bountiful material, carbon.
But for the time being, the current lead "horse" in this race appears to be compressed hydrogen, which suggests that future fuel cell car designers are going to have to find ways to build their cars around bulky tanks instead of cast iron engines as in decades past. It also means, earlier adapters will have to live with shorter ranges and longer refueling times, not to mention significantly higher costs.
Still, technology has a way of eventually solving these problems or coming up with even better solutions.
blog comments powered by Disqus