H2 Hokum

We would all like to be able to imagine a solution to our energy dependency problem.  From the time fire was discovered by burning plant cellulose, we have increasingly used fuel with higher and higher ratios of hydrogen to carbon atoms: coal, oil and natural gas.  Graphing the H:C ratio vs. time suggests that a 2:0 ratio could be next.  But simply extrapolating trend lines can also result in rather poor policy.  Our Republican President and the Democrats have granted some $1.5 billion over five years to the Department of Energy's "FreedomCAR" program.  But, as Matt Wald points out in the May, 2004 issue of Scientific American and in an interview with NPR, today's children may be receiving social security benefits before they get their first hydrogen cars.

These sentiments echo those of Former Acting Assistant Secretary of Energy Joseph Romm, MIT PhD author of The Hype about Hydrogen.  For starters, on-board hydrogen storage is highly problematic.  Subsidized prototypes built by the major car companies either stored liquid hydrogen in a thermos at -423°F or as highly pressurized gas.  The insulated liquid "burped off" completely within a week so the thermos gas tank would only be useful for delivery vehicles that run all day on the boiling hydrogen.  Liquefaction of one kilo of hydrogen also entails some 30% in additional energy expenditure.  Compressed hydrogen is more practical and energy efficient but, since a tank of gasoline contains about 3000 times as much energy as the same volume of hydrogen gas at room temperature and pressure, the gas would have to be compressed to approximately 1000 times atmospheric pressure.  Such filling stations would be cumbersome, slow and generate waste heat equivalent to at least 15% of the energy stored in the hydrogen.  They would also pose a significant safety challenge because hydrogen ignites extremely easily and burns invisibly.  Metal hydride storage is more stable but such storage solutions weigh around 100kg per gallon of gasoline equivalent and require long refueling times with oven-like temperatures for the charge/recharge cycle.

Such problems led to the conclusion of the report issued by the National Academy of Science in 2/2004 (NB When this link was created, it opened the .pdf to the page containing the following quote) that both liquid and compressed storage have “little promise of long-term practicality for light-duty vehicles”.  The NAS even recommended that DOE halt research in both areas. 

Powering Electric Motors from the Grid is More Efficient than On-Board Fuel Cells

While storing hydrogen on-board cars is expensive and difficult, the bottom line is that all of the three commercial methods of producing hydrogen to generate electricity for the drivetrain are uneconomic compared with directly using electricity off the power grid.

1) Using electricity to electrolyze water and get the electricity back from fuel cells is about 50% efficient vs. at least 85% for sending the electricity over the electric grid, and still more energy is needed to store the hydrogen on the vehicle.  This is a particularly bad approach unless electricity is cheap.

2) Reforming natural gas with catalysts produces hydrogen with 30% of the energy going to waste heat.  With expensive overhauls of home heating systems, the heat may be used warm residential air & water.  Fuel cells may then attain 60% efficiency on average for a best case of about 42% plus whatever is recovered for home heating.  Though electric plants fueled with natural gas are more expensive to operate than those fueled by coal (or uranium), they are now about 55% efficient; with transmission losses of 85%, the overall efficiency is 47%.  Use of the waste heat could tip the balance in favor of fuel cells but they are still rather more likely to end in your basement than in your car because of the on-board hydrogen storage problems.

3) Steam can be passed over coal to produce hydrogen and carbon dioxide.  The Department of Energy's FutureGen project aims to co-generate electricity and hydrogen with underground sequestration of the carbon dioxide with a target date of 2020.  How efficient will that be?  We don't have the numbers yet but simply burning coal generates more energy and, combined with the on-board hydrogen storage losses, it is a fair bet that this approach will not be the most energy efficient way to power surface travel either.

e-Guideways and Dualmode Cars

At least three serious entrepreneurs propose building guideways and electrifying them. These could be powered by a variety of approaches without the strategic uncertainties of using oil.  These include new generations of meltdown proof and disposal facilitating pebble based uranium reactors, "clean coal", wind and solar.  

Separation from cross traffic, people, and animals would make it possible to greatly reduce the potential for accidents so that long, aerodynamic car-trains could be dynamically assembled under computer control.  Some guideway vehicles may, like public transit, only go station-to-station located at guideway exits.  However, people generally demand door-to-door service so --- assuming that they can be made highly breakdown resistant --- most cars would be "dualmode" and able to travel on the road as well.  Most trips are within and between metropolitan areas so many users would be well served by dualmode cars that could be recharged directly from the guideway and would never be far from one.  For trips more than 5-10 miles from the guideway, hybrid cars will be needed.  But a sufficiently complete network of e-guideways would make all-electric cars practical for most trips in the US.  Because these cars would not require heavy internal combustion motors, and the collision prevention system described below would support the guideways' huge improvement in safety, lighter cars should become more acceptable to consumers for daily trips.  Combining digitally precise (and fault tolerant) merges with shorter (about 10') commuter oriented vehicles could push guideway throughput into the vicinity of 500 vehicles/direction/minute, about fifteen times more than a single freeway lane.

Running Efficiency and Independence

The smooth guideways would need only a small gap between vehicle and guideway which, together with the major effect of streamlining many cars into continuous trains, will decrease the effective coefficient of drag to the range 0.05 - 0.20 as compared with around 0.25 for best that can be done for regular cars.  The guideway smoothness will also allow harder wheels having low rolling resistance.  These reductions in friction imply 2 to 4X greater running efficiency than without guideways.  This means that 100MPH at 80MPG equivalent energy consumption seems fairly attainable in minivan sized vehicles.  (If cars platoon on the road at less than one car length --- which is much less than in the federally supported 1997 Automated Highway System trial in San Diego --- the guideway efficiency advantage may be only 1.5 to 2.5X.  But it may never be safe enough to use AHS at high speeds with cars in such close proximity.  Using ultra high strength permanent magnets to suspend the cars (Maglev) has the potential to improve economy on guideways only slightly more because aerodynamic drag dominates at high speeds.  However, partly because of the cost of the magnet suspension, most Maglev car designs have been captive to the guideway.)

The "well-to-wheel" efficiency of electrified guideways is also much better.  Though typical cars convert gasoline to motion with 15% efficiency and hybrid-electric cars attain around 28%, the efficiency of new gas electric power with transmission is 47% as stated above, and there is no energy loss from oil refining and gasoline transporting.  So it will still be at least 50% more efficient to run vehicles from an electrified guideway if the power plants are gas fired.  But there are less expensive ways to produce electricity and these estimates do not account for the additional energy losses from making and distributing gasoline.

Combining the energy and running efficiencies yields overall efficiency gains of three to six times.  At the residential retail rate of $0.11 per kWh, the cost to travel 100 miles at 100 MPH in a mid-size vehicle would likely fall in the range of $2 to $4 using relatively plentiful supplies of coal that now supply one-half of US electricity.  Whether greenhouse gas curtailment requires carbon dioxide sequestration and how that cost-benefit tradeoff compares with nuclear or even wind or solar power remain open questions.  However, several of these sources are likely to become preferable to oil and gas given the likelihood of limited future production. 

Platooning and Collision Prevention

The huge traffic throughput of guideways suggests a need for an overhaul of the operational régime on city streets. So, in order to handle ramps that could easily disgorge some 50 cars/minute, the e-cars may be made to form short trains on the avenues. These closely packed clumps of cars known as platoons can be achieved using car-following cruise control (CCC) and would be useful immediately on HOV/HOT/CCC lanes for doubling their capacity, particularly at bottlenecks created by accidents and traffic lights. While this kind of cruise control is available on high end cars today, it would need to be adapted to permit following distances of about one foot up to 35 MPH. But, in order to make it safe for platoons – which would involve multiple cars in collisions – the most dangerous crashes at intersections must be rigorously prevented.

So, to prevent intersection collisions, we will need a traffic alert and collision avoidance for cars that is similar to the TCAS used by jets. For example, beacons on city streets could augment Global Positioning Systems so that each car can continuously calculate and broadcast its position, velocity and the driver's signaled intentions. Priority may then be dynamically negotiated according to speed and traffic volume to optimize flow.  Traffic would group into pods that move faster as they grow, like drops of rain on a windowpane, but which also anticipate cross traffic and negotiate speeds well in advance of intersections so as to maintain relatively constant speed. 

Mandatory but inexpensive retrofits of the TCAS for older cars would merely provide speed recommendations and warnings of impending collisions, but the newer platooning e-cars could actually take over braking if there happened to be a vehicle around the corner heading through a red light, or if you happened to be the errant driver.  It is likely that the cost/benefit of inexpensive electronics for TCAS would be far better than air bags and would justify their mandatory use within a decade.  

Note that cars do not need guidance systems or steer-by-wire in order to platoon.  Humans can manage to keep a car within a lane although they are not very good at opening spaces for other cars needing to change lanes.  A little software will be able to do that just fine.

Platooning will allow today's roads to handle far more cars but once much of the traffic diverts to the guideway, another possibility appears: dramatically reducing the speed limit in denser areas within about 10 blocks of the guideway.  10 MPH is almost 3 blocks per minute if the pace is maintained by efficiently negotiating the right-of-way at intersections. Densely packed, low speed but relatively short platoons under computer control will allow more frequent light changes while sharply reducing the risk of fatal accidents for pedestrians, bikers and skaters. 

Quieter, Cleaner and Safer Cities

The main impact on cities will come from the diversion of huge throngs of aggressive and honking loud traffic from city thoroughfares to much quieter, cleaner and safer e-guideways.  This portends a revolution in the quality of city life where, for example, conversations can be held at a whisper in sidewalk cafes whose tables spill out into landscaped green-space now claimed by cars. 

Placing guideways along major avenues will however create a major visual intrusion and add to the risk of objects falling on pedestrians.  Though guideways will be much narrower than elevated subway tracks and highways and could be incorporated into landscaping, routes with heavy pedestrian use will suggest underground tunnels, placement through and over buildings, hidden from street sight-lines in back alleys, or inside earthen berms (perhaps with adjacent lanes for bikes that could benefit from the wind generated).  Because the intersections of two, dual-carriage guideways require up to eight separate ramps, city streets will tend to have a one-way guideway only with just two ramps that can fit in the space available in existing intersections.  

Once traffic which formerly traveled on the street moves to the guideways, pedestrians could reclaim rights ceded one century ago to automobiles.  If the average commuter vehicle is also smaller and able to self-propel with the help of passing platoons, the existing parking facilities will be used much more efficiently.  This, and the expected increased use of short-term rentals (because of much less accidents and maintenance, and drop-off locations on every block), suggests that street parking may also be reduced or eliminated so that even more public space can be reclaimed.

Benefits of a National e-Guideway Network

On a 100MPH interstate guideway network, one could go door-to-door from the Washington, DC area to the New York area in about 2.5 hours, as fast as taking taxis from the respective downtowns to a plane.  Yet the guideway trip should cost less than bus fare, which is currently $10 from both cities’ Chinatown districts.  Of course, one could also depart from any place at any time and work, play, eat and sleep comfortably during the trip. Bunk beds in a large van would allow a small family to comfortably cross the USA together in three nights, stopping to see friends and take in national parks along the way for about a dime a mile, total cost. People would still fly for speed but since automobiles are already a direct substitute for air travel up to 1000 miles, we can safely estimate that comfortable and driverless e-cars would substitute for at least 50% of all domestic flights.

There would be many knock-on effects.  Such vehicles would quickly become fully automated for freight, which would also depend on the prevalence of platoons whose formations could be joined using semi-automated or computer vision systems long before such computers can fully take over driving. A guideway network would allow freight to be packed into smaller vehicles for direct delivery to supermarkets, stores and neighborhoods. Internet retailing would be greatly facilitated but, more importantly in this author's mind, California produce could be picked ripe for a smooth one-day, just-in-time delivery on the East Coast.  Guideways would also reduce or eliminate direct expenses for drivers, fuel and transshipment, and indirect costs of repairing road damage caused by heavy trucks.

There would of course be undesirable results too. Cars led to urban decay in the 1950's and 1960's as families moved away from the noise and the smog and into larger suburban homes. Though guideways could greatly improve the quality of city life, the improvements might not offset the increased allure that the mobility gives to bigger plots of land even further from the urban cores.

Projected Evolution of Surface Transportation Technology in the USA

Status:  0 – not feasible  1 – theoretic acceptance  2 – prototype designed  3 – prototype realized  4 – deployment fully supported  5 – self-sustaining

 

e-Cars Facilitate Car Sharing and Public Transportation

Such e-cars under the more positive signaling and control of TCAS would almost never get into accidents. They would be far less prone to maintenance without high speed salt spray and traditional car parts like mufflers, non-regenerative brakes, transmissions, valves, fuel pumps, injection and ignition systems, all of which are prone to failure. e-Cars will be much easier to deliver to the curbside where they are needed via platoon. If cars are re-used in this manner like taxis, they will require less parking. These factors will drive down the cost of short-term car hire relative to car ownership.  Once the vehicles no longer require drivers -- an advance that should be achieved shortly after platooning is realized -- mass transit systems may even operate "out of the fare box" or at least greatly reduce the typical 80% taxpayer subsidy.  Government operated mass transit may simply be rendered obsolete by affordable private e-cars and e-vans offering much more frequent service with far fewer stops. Even the NYC subway system might eventually be bypassed since guideways have comparable capacity and provide non-stop, on-demand service.  Typical subway rides that now require a transfer and take 45 minutes to cover ten miles would instead be completed in about ten minutes station-to-station or twelve minutes door-to-door.

e-Guideway Construction Cost Estimated at $1 Trillion and Up

The cost is considerable.  It consists of 1) constructing the guideways and 2) nearly doubling the US electricity generating capacity (now 3.2 billion mWh), half of which would likely be due to increased use from today's 20,000 miles/user/year to 40,000 miles/user/year. The latter can be paid off by the revenues for electricity.  Excess power demand can be reduced to supply by slowing the guideways.  A reduction in speed from 100MPH to 70MPH would cut power consumption by about 50% if needed, for example, on hot days for air conditioning.

We can only estimate the guideway cost at this time. Sextupling the estimates of the two leading entrepreneurs (RUF International and MegaRail Transportation Systems) to reach $40 million per customized metro mile, of which some 20,000 miles would be needed in addition to 40,000 miles paralleling the remaining Interstate, which is estimated closer to $5 million per mile, leads to a total of $1 trillion.  However, this construction expense of $100 billion per year over the first 10 years fits in the context of transportation related goods and services that contributed 1.047 trillion dollars to the GDP in 2001 and averaged $7800 for 109 million households ($850 billion).  It can also be compared with a 2002 analysis by Argonne National Laboratory that, "infrastructure to serve 40% of the light duty fleet is likely to cost over $500 billion  It seems likely that at least another 100,000 miles would be strongly desired to serve areas now reached by state highways, which could push the total investment towards $2 trillion.  It would best begin along critically congested corridors like the 101/I-80 San Jose to San Francisco to Berkeley, or the I-95 Washington to New York to Boston with public minivan sized vehicles providing continuous station-to-station service.  The whole system would take 50 years to grow fully so the annual expenses need not exceed $100 billion.  

However, the value of e-guideway benefits suggests that a much higher rate of investment might be prudent.  Estimates should include the reduced cost of freeway accidents, which - if not incalculable due to the lives saved and permanent injuries avoided - would be about $80 billion, gained productivity of time formerly spent driving (circa $140B), reduction of time lost to congestion ($125B), the costs saved by car sharing (minimum of $50B), reduced maintenance and increased vehicle longevity ($200B), improved productivity of shipping (at least $30billion) yields approximately $600 billion per year in costs saved and value added.   Additionally, people will pay for the value of extra mobility just as they pay for toll roads today.  Although a two year payback cannot be expected because many people will wait a long time to trade in their old cars, the payback may be made attractive to private investors, particularly along the critical corridors. 

Ten Years and $10-$20 Billion to Design e-Guideways

A system to be built on a national - indeed continental and even global - scale should be designed with the utmost care. The structure of that effort ought to include adequate incentives for pioneers like Palle Jensen of RUF International and Kirston Henderson of MegaRail to enter their patents into a series of engineering bake-offs. With adequate investments and commitments from foreign, federal, state and local entities, the first fully functioning prototype might be designed and tested in a theme park before committing to a metropolitan build.  The first real system might be built on Oahu – an island with significant commuter problems but where design errors need not propagate.  The system will also need to be extensively tested in an area with regular icing conditions, a major issue for guideways.  Most people would experience the system for the first time in an automated taxi car or van sized vehicle providing direct service with few or no stops. These public services will provide utility and revenue immediately after construction and help to get over the chicken and egg problem of having guideways without dualmode vehicles or dualmode vehicles without guideways.  

Though legislative support will be needed for TCAS, the technology can be developed with private funds and suggests the possibility of a start-up.  Cruise control for tight formation platooning is a marketing rather than technical problem though it also will need state house actions to create HOV/HOT/CCC lanes.  

Original versions by Bruce A. McHenry.

Thanks to Francis Reynolds, Palle Jensen, Kirston Henderson, Randy Leong, Jesse Ausubel and Secretary Rodney Slater for their suggestions. Special thanks to Professor Jerry Schneider, University of Washington for his Innovative Transportation Technologies web site and mentoring, and to Harriet Taber for her continuing support.

This column was originally published March 9, 2004 in the MIT alumni publication whatmatters (with images from RUF International), copied to discussIT.org and revised on 4/28/04, 5/23/04, 6/5/04, 10/22/05 (table merged into this article), 3/16/06 (change noted in link to National Academy report on hydrogen prospects)````````````````````````````````````````````````````````````````````````````````````f.

Images courtesy of RUF International.

 © 2004-2005 discussIT.org