Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Debunking the hydrogen economy

  1. Aug 25, 2008 #1
    The only way a hydrogen economy would make sense is if it produced energy (e.g. via fusion).

    Without getting energy from hydrogen, fuel cell vehicles and etc are basically overly complicated batteries.
     
  2. jcsd
  3. Aug 25, 2008 #2

    Ygggdrasil

    User Avatar
    Science Advisor
    2015 Award

    However, batteries currently are not very good at storing energy and electric vehicle cannot travel distances comparable to gas powered vehicles. Using hydrogen as an intermediary for energy storage could allow them to act as a better "battery" with a greater range than electric cars.

    However, I do agree that a lot more work is needed on hydrogen in order to determine whether it will actually be feasible. As you correctly mentioned, an environmentally friendly hydrogen economy relies on the development of a renewable energy infrastructure as hydrogen only stores energy. Furthermore, its not necessarily clear that battery technology won't improve, and it may also be more advantageous to change our driving habits to support the use of electric vehicles.
     
  4. Aug 25, 2008 #3

    mgb_phys

    User Avatar
    Science Advisor
    Homework Helper

    It makes sense for transferring large amounts of energy large distancess
    If you have geothermal (iceland) or solar (n. africa) it is easier to make hydrogren and ship that to LA than run a power line or ship recharged nicad batteries.
    The infrastructure is more easily modified, lots of natural gas and LPG technology.

    It's not clear if hydrogen powered cars make more sense than electric. Power densities are similair and both need to improve if they are going to be more than around-town subcompacts. Hydrogen fits the gas-station fillup / oil company model but electric lets you recharge at home overnight.
     
  5. Aug 25, 2008 #4
    Also, even using oil or coal power plants to get electricity and store it in hydrogen is more efficient than using gasoline automobiles, because it's easier to stay efficient on a large-scale than to have every driver take necessary steps to make their car more efficient.
     
  6. Aug 25, 2008 #5

    russ_watters

    User Avatar

    Staff: Mentor

    The way you worded it, it isn't true. Batteries store and recover energy at an efficiency of better than 90%. Fuel cells, depending on how the hydrogen is produced, only around 50%.
    Now the way that is worded, it could be true. Reglar batteries have some drawbacks, like weight, cost, nasty materials, and charging time. Fuel cells might be better in those respects.
     
  7. Aug 25, 2008 #6

    russ_watters

    User Avatar

    Staff: Mentor

    That is true of battery-powered cars, but not true of fuel cell cars. Assuming probably too high estimates of 70% efficiency for both the electrolysis and the fuel cell, 60% efficiency for the power plant, and 90% for the motor, that gives you about 27% efficiency. A battery powered car loses about 7% in the power lines and the batteries, charger, and motors are all better than 90% efficiency. That gives a conservatively low estimate of 40% overall efficiency.
     
  8. Aug 25, 2008 #7
    The point of hydrogen in the "hydrogen economy" is as a common medium for energy transport and storage. If you didn't understand this already then you don't understand the "hydrogen economy" enough to debunk it...

    That 7% sounds very optimistic, are you assuming the power was generated locally? What if the power was generated a long way away and was transported over long distance power lines, or was generated at some point in the past and stored? One assumes hydrogen, if it's treated like normal substance fuels, would be used this way, I don't know whether electricity can be transported and stored as easily or not...
     
  9. Aug 25, 2008 #8

    russ_watters

    User Avatar

    Staff: Mentor

    I'm talking about the transmission lines themselves and it is about 7%:
    http://en.wikipedia.org/wiki/Electric_power_transmission#Losses

    And I factored that into the batteries but not the hydrogen because I assumed the hydrogen would be produced at the power plant. Electric cars will almost certainly be recharged at home.
     
  10. Aug 27, 2008 #9
    Didn't MIT just have a "break through" by using a new catalyst and cathode to have electrolysis operate at efficiencies of around 90%? Current PEM fuel cells operate around 50% efficiency although I think Honda's (FCX) is approaching 60%. I can see this number being 70% in ten years. And although it isn't hear yet solar thermolysis shows promise of some extremely high efficiencies.

    One overlooked aspect of batteries is their disposal. They don't last forever, and most high capacity batts don't last longer than 500 cycles. Also you have to consider that all batteries self discharge. BTW, if you compare the Honda FCX to the Tesla Roadster, they both have about the same max distance. But the FCX does cost a hell of a lot more.
     
  11. Aug 27, 2008 #10

    mheslep

    User Avatar
    Gold Member

  12. Aug 27, 2008 #11

    mheslep

    User Avatar
    Gold Member

    That depends a great deal on the level of discharge, the top off charge, and temperature. If one mediates all these, which the Tesla roadster does and your laptop does not, the battery life will be greatly extended.
     
  13. Aug 27, 2008 #12
    My laptop has hardware designed to maximize battery fitness, doesn't yours. :confused:

    Anyway, LiFe(P) has done a lot for battery durability but that comes at a pretty big hit of specific energy which in the long run kills efficiency. The lightest batts, Lipoly not Lion or LiFe(P), have very limited life spans even when properly used. I've never gotten more than 200 cycles out of my Lipos for my RC heli's while I could run NiMH for over a thousand.
     
  14. Aug 27, 2008 #13

    mheslep

    User Avatar
    Gold Member

    All the mfn's design to maximize operation time per single charge, as that's what is advertised and reviewed. Their first priority is to give you non-stop operation on that coast-to-coast air travel, lifecycle is an afterthought. Obviously laptops have no temperature stabilization on the battery. The Chevy Volt for instance will hold in reserve some percentage (10%?) of the battery charge, and they never top off the charge either, neither of which is provided by your laptop. Generally laptop charge circuitry is only tasked with limiting charge rates and Li ion safety issues.
     
  15. Aug 27, 2008 #14

    mheslep

    User Avatar
    Gold Member

    Re: Debunking the hydrogen economy: Renewable H2 filling stations?

    Of the several problems with an H2 economy by far the most intractable to my mind is the transportation and/or storage of H2 as a fuel. With some of the more recent developments I 've been musing that perhaps a solar/wind/grid based local roadside H2 station might start to make sense. Details below. First the problems w/ H2 up until now.

    This chart lays it out nicely:
    http://www.physorg.com/news85074285.html
    To run an H2/fuel cell car, starting with 100kWh electric power one ends up w/ only 23kWh of tractive power delivered by the vehicle:
    AC-DC: 95%
    Electrolysis: 75%
    Compression: 90%
    Transport/transfer: 80%
    Fuel Cell: 50%
    Electric drive train: 90%

    So far the logistics of providing roadside fueling of an H2 car has meant:
    stage 1: a large electrolysis center somewhere w/ with a massive megawatt connection to the grid to make the H2 (or reforming from natural gas but lets drop the fossile source for this line). Then compression, storage, and transport to local road side stations. Note that it takes 15-20 tankers of compressed 3000 to 5000 psi H2 to deliver the energy equivalent of one gasoline tanker, and existing pipelines won't handle H2 at all.
    stage 2: road side H2 station, storage again until delivered to the vehicle.

    Now some musing on how this might be done differently, given recent developments:
    -Nocera's efficient and cheap electrolysis breakthrough (linked above)
    -Improving solar PV efficiency and technology, especially concentrated PV
    -Electric transmission becoming more expensive and difficult.
    These three lead me to the concept of a completely local, self sufficient H2 roadside station. Given: the average existing US gas station pumps 2000 gal/day. At 136Mjoule/gal-gasoline, thats 272000 Mjoule/day, or 76000 kW-hrs/day. Assuming the new H2/fuel cell cars are 3X more efficient than existing ICE cars, we need only a third, or 25300 kW-hrs/day of equivalent electrical energy.

    Now, efficiencies. We need no AC/DC conversion. After 90% efficient Nocera on-site electrolysis and 90% efficient on-site compression we need 31234 kW-hrs/day. To get that from the grid means a 1.3MW average electric service at all 200,000 US filling stations, hard for both the local filling stations and the grid at large to accommodate, so lets try onsite generation.

    Solar is a good fit here in a sunny climate because we're along side the highway, and because we don't care about variability. One just buffers enough H2 to stay ahead of demand. At a year round average of 7kW-hrs/m^2/day we need 4500m^2 or 1.1 acres (with 100% collection): 16 acres of 7% thin film, 7.5 acres of traditional 12% Si PV, 5 acres of advanced 22% PV, 3 acres concentrator panels of triple-junction 37% PV cells (w/ tracking). Big interstate roadside stations, like this one, have that kind of land to spare.

    Cost? Solarbuzz says 22 cents/kWh for traditional PV in sunny climates, or $6900/day for this scenario, plus the electrolysis, compression, and H2 storage gear. That's against $8000 per day for the gasoline station this H2 station replaces, so we're getting close. On the other hand, the triple-junction company Sunrgi claims they'll be doing 7 cents/kWh in a couple years. :uhh:
     
    Last edited: Aug 27, 2008
  16. Aug 27, 2008 #15

    Ygggdrasil

    User Avatar
    Science Advisor
    2015 Award

    Today's New York Times (http://www.nytimes.com/2008/08/27/business/27grid.html) had an interesting article about wind power that goes along the lines of mheslep's comment. Many wind farms have to shut down when the wind is too strong because they're generating too much energy for the power grid. This highlights a major problem with two of the primary sources of renewable energy (wind and solar): their outputs are highly variable and may not match peak times for energy demand. This is where hydrogen, as a means to store this energy that would otherwise go unharvested, comes in. This hydrogen could either be used as a buffer for days where the solar/wind generators can't meet demand and/or could be transported away for use as fuel.

    Of course, there are still problems with transport and storage, but it seems like generating hydrogen may not be so much of an issue now. Personally, I think storing hydrogen in the solid state (e.g. as metal hydrides or adsorbed in nanomaterials) shows the most promise, but these solutions are still in the R&D stage.
     
  17. Aug 27, 2008 #16

    mheslep

    User Avatar
    Gold Member

    Yes efficient production of H2 would help the variability of renewable power like solar and wind in many applications but it doesn't by itself enable a nationwide solution. As that NYT piece shows, the power has to be eventually shipped over transmission lines from the wind belt or the solar belt to the demand areas - except for onsite bufferable problems like a fueling station.
     
  18. Aug 28, 2008 #17

    vanesch

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Re: Debunking the hydrogen economy: Renewable H2 filling stations?

    You mean your average highway station only has about 100-200 customers a day ?
     
  19. Aug 28, 2008 #18
    Non-baseload power generation is the biggest flaw with alternative power sources. I believe it was once estimated that wind power could never account for more than 7% of total power on the grid because of variability. The current bandaid for this is the use of flywheels, for example Beacon Power's flywheels: http://www.reuters.com/article/pressRelease/idUS174731+19-May-2008+BW20080519

    I believe the only answer in the end will by an H2 economy.

    mheslep, that article makes a strong argument but it includes a lot of unnecessary processes. H2 would obviously be created directly at the plant and then shipped, in which case the fast majority of the losses would be just to do to the transportation vehicle itself which given a few years of R&D will become pretty close to current electric vehicle efficiencies.
     
  20. Aug 28, 2008 #19

    mheslep

    User Avatar
    Gold Member

    Re: Debunking the hydrogen economy: Renewable H2 filling stations?

    That sound reasonable, I actually pumped gas as a kid, but I don't know. I used US daily 390 million gallons of gasoline / 200000 US gas stations.
    www.eia.doe.gov/basics/quickoil.html
     
  21. Aug 28, 2008 #20

    mheslep

    User Avatar
    Gold Member

    20%.
    www.20percentwind.org

    No doubt there are many storage methods possible. Pump storage and hydro has been the traditional load levelling go-to for the industry, geography/terrain permitting. At the moment Compressed Air seems to be the darling of DoE (one plant in operation), perhaps electrolysis-H2-gas turbine at NREL (prototype).
    Distributed Energy Program: Compressed Air Energy Storage
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?