Sciencemadness Discussion Board

X-PRIZE Foundation: Energy and Environment

bereal511 - 16-11-2008 at 12:08

A few days ago, I received a notice that my video submission to the X-Prize "crazy green idea" video competition was chosen as one of top three submissions! I don't want to sound like I am trying to solicit votes (feel free to vote on the other submissions if you think they are worthy of being the next X-Prize) but I just wanted to share this great news.

Here's the link:

http://www.xprize.org/crazy-green-idea

The X-Prize Foundation wanted to hear what others would like to propose as the next X-Prize in Energy and the Environment. My team and I put together "The Capacitor Challenge" to address the problems of electrochemical cells as they pertain to the environment. It's really just me toying with the idea of having electric vehicles that can recharge instantaneously after reading about ultra-capacitors developed at MIT (and knowing that regular EVs take forever to recharge):

http://www.popularmechanics.com/science/research/4252623.htm...

Tell me what you guys think.

12AX7 - 16-11-2008 at 15:01

Congratulations on the video.

Capacitors, huh? Nah, cells are still the way to go. Build them small (like that silicon nanowire stuff they've been taking about) and you get incredible surface area = high current density, and fractal structure = low resistance. A capacitor, which uses conductively bound electrons and the insulating ability of the dielectric, inherently cannot store as much energy as a cell, which depends on ionically bound electrons (or the absence of them). Down to the bottom of it, a cell can be 100% pure charge (although only a fraction of this can be realized practically), whereas a capacitor will fail exactly when it tries to become a cell: when the electric potential ionizes atoms.

Good news about the supercapacitors though. With better power density than most cells (we'll have to wait and see what comes of new technologies), that makes them well suited to high power applications. For instance, you could fill the trunk of your car with ultracapacitors (and an inverter to drive a ~20HP DC motor from them) and get to highway speed in a couple of seconds. Caveat: present prices run that around $30k. Ditto a battery pack (of any chemistry) big enough for a long haul (300 mile range is typically quoted).

Incidentially, EVs take about 2 hours to charge (at full charging power) due to chemistry; faster-charging chemistries may allow as little as 15 minutes (nearly tolerable as far as "filling up the tank"), but 150MJ is far too much to charge in that time, even on a moderate commercial power line (think 480V three phase at >150A). A 120V 15A circuit could handle a full charge in about a day.

Tim

watson.fawkes - 16-11-2008 at 16:20

Quote:
Originally posted by 12AX7
Capacitors, huh? Nah, cells are still the way to go. [...] Down to the bottom of it, a cell can be 100% pure charge (although only a fraction of this can be realized practically), whereas a capacitor will fail exactly when it tries to become a cell: when the electric potential ionizes atoms.
I have to say that don't find this argument immediately persuasive. Implicit in this argument is the claim that a capacitor would stop being a capacitor when its substrate starts ionizing. This doesn't seem right.

Now as to the present discussion, energy density is the concern. Certainly a capacitor in a state of ionizing internally with increased potential is not going to act like a linear device. You have another energy term active, ionization energy in addition to charge separation. Yet linearity isn't the question. At some point, all the outer electrons have been stripped off and further ionization can't happen until another plateau of electric field is reached. In this regime, the capacitor can continue to accept charge, at least hypothetically. In summary, breakdown (which is what internal ionization is) doesn't need to be total, permanent, or cause a short.

After this first breakdown (assuming the device is still a device), there will be another breakdown potential at a higher level, and this level is conceptually much higher, since it involves rearrangement of inner shell electrons, that is, "non-chemical" ones, and as such could store more energy than a chemical cell.

12AX7 - 16-11-2008 at 23:29

Inner shell electrons?

No.

As soon as you start speaking of ionized atoms (or similarly, chemical bonds formed), you're dealing with chemistry and not capacitance. A capacitor does not operate with ions. Although a net charge on any given plate seems to imply some amount of ionization, it is delocalized over an electrically conductive substrate. If there is an ion, then in typical delocalization fashion, the entire electrode must be considered the ion.

Tim

watson.fawkes - 17-11-2008 at 05:03

Quote:
Originally posted by 12AX7
As soon as you start speaking of ionized atoms (or similarly, chemical bonds formed), you're dealing with chemistry and not capacitance. A capacitor does not operate with ions. Although a net charge on any given plate seems to imply some amount of ionization, it is delocalized over an electrically conductive substrate. If there is an ion, then in typical delocalization fashion, the entire electrode must be considered the ion.
If you'd like me to integrate the ionization energy of every atom in the conductor to make my argument, fine. The point remains that after all those outer shell electrons in the positive electrode are gone, the electrons left are much more tightly bound and not going anywhere soon. Assuming you still have a material at all in this state (the SF assumption here), you've saturated the positive terminal, the capacitance drops in half (roughly), and you're still storing energy in the device with an electrostatic field and so it's still a capacitor. As long as you can continue to pump electrons into the negative electrode (another SF assumption), you've got the possibility of energy densities greater than purely chemical ones.

I've left out lattice binding energy in comparing the sum of ionization energies of atoms and the total ionization of a crystal. It's some fraction of the total atomic ionizations. It doesn't change the final point that if you can continue to push electrons into the negative electrode, you're still getting higher energy densities.

I'll acknowledge that which the positive electrode is undergoing this electron-stripping, it's undergoing a chemical reaction. It would have to. The Hamiltonians for chemical bonds require the presence of binding electrons. When they're removed, what you have is an interaction between screened nuclei with closed orbitals, and you just don't get bonds. So we have three regimes: an unsaturated capacitor, a hybrid where the positive terminal is saturating, and a saturated-hybrid capacitor. In all these regimes, the device is acting like a capacitor. When it's in the saturating regime, it also has some cell-like characteristics.

Such a hypothetical electrode isn't going to be crystalline bulk metal, since that would just fly apart under such severe removal of electrons. It's conceivable for it to be some kind of 2-D monolayer structure in a ceramic material, probably with a heavy element with plenty of available electron structure to work with, such as bismuth or lead.

not_important - 17-11-2008 at 05:46

Focusing on what you can currently get off the shelf, ultracapacitors have an energy storage density of one to two orders of magnitude less than batteries, making batteries a better choice for bulk power storage. There is a company that claims to have an ultracapacitor design that will give power densities equivalent to batteries; however they have yet to deliver such parts and reviews of their patent applications seem to show densities far less than press reports claim.

And as already pointed out, very fast charge times require grid connection far in excess of typical home and light commercial capacity. The power companies would likely be unhappy with large scale use of such charging systems, as they could easily place severe stress on the power grid, say if a sizable fraction of vehicle drivers arrive home after work and start to recharge at about the same time. High speed recharging implies a second fixed capacitor bank for each vehicle needing charging within a time period, the fixed bank is slowly charged from the mains and dumped into a vehicle's capacitors for "refueling" when needed.

Ultracapacitors are better for storing power from regenerative braking, and sourcing power for acceleration, as they can more quickly charge and discharge. Thus practical vehicle design would likely use both, enough ultracapacitors to handle braking and acceleration, and batteries for cruising. When PML Flightlink overhauled a Mini to make it into an electric, that is the technique they used along with a small motor-generator for extended range driving.

Note that while a battery has a fairly constant voltage over its charge/discharge cycle, capacitors show an exponential charge/discharge voltage curve. This makes their interface electronic more complicated and able to handle much higher currents to deliver the same amount of power when the capacitor is somewhat discharged (it's either that of store quite high voltages so the capacitor is always at or above the delivered voltage for its working charged range.

watson.fawkes - 17-11-2008 at 07:02

Quote:
Originally posted by not_important
There is a company that claims to have an ultracapacitor design that will give power densities equivalent to batteries; however they have yet to deliver such parts and reviews of their patent applications seem to show densities far less than press reports claim.
Yep. A friend of mine and I were discussing these earlier this year. The power densities just weren't at all there; something like two or three orders of magnitudes lower than the most rosy interpretations (I'm being generous).

On the other hand, he and I got talking a lot about what was physically possible, and the upshot of our conversation of what might be physically possible is pretty astonishing. For example, as soon as you get energy densities above that supplied by bond strengths, you get rocket fuel.

Whether anything like this appears in my lifetime, I have no idea.

JohnWW - 17-11-2008 at 07:09

What would those alleged new ultracapacitors use for their dielectric medium, inserted between the charge storage plates (which themselves must also have a very large area)?

12AX7 - 17-11-2008 at 13:51

Probably the same electrolytic barrier thing they've been using, an electric double layer. Which I guess is produced by simply soaking the material in a suitable electrolyte, which adsorbs onto it, forming the requisite distribution.

Tim

Twospoons - 17-11-2008 at 14:03

The latest generation of Li batteries (see Toshiba SCiB) boast charge times of 5 mins to 90% and life times of 10 years or 10,000 cycles. They also have astonishing power density (= ultracaps) , though energy density is a bit lower than the current generation of Li ION cells. All thanks to new anode and cathode materials (eg Lithium titanate and lithium iron phosphate). I find it hard to see how Ultracaps can compete effectively with that.

JohnWW - 17-11-2008 at 15:10

The capacitance of a simple capacitor is
C = Q/V = €A/d farad,
for flat parallel charged plates of area A separated by distance d with a dielectric medium of dielectric constant € between them.
For a cylindrical capacitor with outer and inner radii b and a and length L, A becomes the logarithmic mean of the inner and outer areas, neglecting end effects,
C = 2¶€L/ln(b/a) ;
for a spherical capacitor A becomes the harmonic mean of the inner and outer areas,
C = 4¶€ab/(b-a) ;
and for a toroidal capacitor with gross radius r (r >b>a) the formula becomes
C = 4¶²€r/ln(b/a) .

Therefore, the only ways of increasing C, and hence the charge Q and energy Q²/2C or ½CV² that can be stored using a given voltage V, are by either increasing the effective charged surface area (which can be done by having many plates arranged in parallel or concentrically, or two long plates rolled into a cylindrical coil), reducing the distance between them (with a greater risk of charge leakage due to a defect), or by changing the dielectric medium with one having a higher €. The last-mentioned is currently the main means of increasing capacitance, by way of research into better dielectrics. The best dielectrics are perovskite-type minerals, like BaTiO3, and similar refractories. While good dielectrics are usually also good insulators, an exception is water, with a high dielectric constant but a poor insulator.

[Edited on 18-11-08 by JohnWW]

not_important - 17-11-2008 at 18:47

Quote:
Originally posted by JohnWW
What would those alleged new ultracapacitors use for their dielectric medium, inserted between the charge storage plates (which themselves must also have a very large area)?



The company is EEStor, they claim that using extremely high purity barium titanate powder with a glassy or Al2O3 coating gives them a permittivity of 18,500. Wiki has a decent overview:

http://en.wikipedia.org/wiki/EEStor

I should note that their projected delivery dates have slipped several times, I think getting close to a couple of years worth now.


[Edited on 18-11-2008 by not_important]

Think before you jump

franklyn - 17-11-2008 at 23:59

And the point of all this is what ?
Annual U.S. electric consumption for 2006 - 4160 terawatt hours
with production at 92 % of capacity.
Annual gasoline consumption in 2006 - 142,421 million gallons
at 36.6 kilowatt hour per US gallon that comes to 5213 terawatt hours , duh
You're a bit short of capacity there too Einstein.

The military could use high energy density capacitors though
to fire hypervelocity projectiles.

.

not_important - 18-11-2008 at 02:09

You can't do a direct comparison like that. Otto and Diesel cycle propelled cars have a series of energy losses. The Otto engines are about 25% efficient, Diesel closer to 30%. The power train - transmission, drive shaft, differential, and all that, is fairly lossy, another 5 to 10 percent of the energy.

A pure battery-electric vehicle gets about 70 to 85 percent efficiency in the battery charge/discharge, and a motor+controller efficiency of 85 to 95 percent. Real world BEVs get 5 to 6 km/kwh, which using you figure of 36.6 kilowatt hour per US gallon would work out to 180-220 km/gallon or 108-132 MPG.

To do a proper comparison you need to do a full well-to-wheel analysis, which includes refinery losses, fuel transportation overhead, electric powerplant efficiency, power transmission losses, and so on.

A 1993 Honda Civic VX has a WTW rating of 0.52 km/MJ, a Toyota Camry is rated around 0.28 km/MJ. The hybrid car 2005 Honda Insight rates at 0.64 km/MJ, a Toyota Prius at 0.56 km/MJ. Pure battery-electrics run 1.15 km/MJ for the sporty Tesla on up to 1.5 km/MJ for some prototypes. Thus a BEV is twice as efficient in overall energy consumption as the best ICE commercial automobile, and four times as a midrange ICE.

Not factored into those numbers is regenerative braking, which becomes important in urban driving. Another thing to consider is that BEVs can use power from wind farms or other renewable sources, while ICE vehicles can not directly use that electricity. Even sticking with fossil fuels BEVs and PEHVs could get higher ratings were more power plants using modern designs to increase their efficiency by 1.5 to 2 fold.

Using the 5 km/kwh figure and the approximate U.S. annual vehicle travel distance of 3E12 miles, I get 9.6E11 watt-hours to totally replace private vehicles with BEV.

BTW - both the hydrogen fuel cell powered vehicle, and the compressed air powered car, run at WTW efficiencies lower that current hybrid vehicles.

franklyn - 19-11-2008 at 14:10

Apparently I'm talking to myself. The point I make ( deftly evaded
in a rambling retort ) is the national electric grid is now at near its
maximum generating capacity ( principally because generating
infrastructure costs whether it is utilized or not - it doesn't pay to
have excessive extra capacity ). Vehicles whether combustion driven
or electric, have inertia and mass, never mind how stingy it may use
energy. Efficiency of electric generating plants is not much better
than a diesel engine delivering torque. Where is all this additional
electric supply going to come from ?
There are around 60 nuclear plants and 500 coal plants supplying
about 80 % of electric demand ( The balance hydro and minimally
" alternative " sources such as solar, wind, etcetera ). Building the
extra generating capacity and the mining of coal and ore is the least
of it, far more daunting is augmenting the distribution grid to cope
with the increased current load.

Economics figure into this very prominently. Electricity is already the
most expensive energy produced. The price is driven by demand
so that every one who buys from a utility will then be subsidizing
drivers of electric cars.

.

not_important - 20-11-2008 at 09:37

OK, I guess I have to run more numbers for you.

First off there is no need to materialise this additional source of electricity overnight. It takes at least a decade to replace the private vehicle fleet in the U.S. through normal new vehicle purchasing.

Next is that you missed what I said about motive efficiency. While the raw thermal efficiency of a large automotive diesel engine is about the same as the typical non-CC fossil fuel plant, it is significantly lower when considered as well-to-wheels cost, which includes losses in the power train, refining process, and distribution of product. For $2US per gallon fuel cost, the cost per mile is about $0,0666 (hmm...) and obviously for $4 per gallon twice that. Using the US average residential price of electricity of $0,12 per kWh a pure battery electric vehicle that gets the average 3+ miles per kWh will cost 4 cents per mile, or 2/3 to 1/3 the cost of the conventional automobile using recent pricing. That makes it difficult to call electricity the most expensive energy, when considering the end system costs or what the consumer sees.

Now a rush to simple BEVs would result in pushing up the price of electricity, and in theory a drop in price at the pump for conventional fuels. But going to plug-in hybrid designs allows the driver to choose their energy source for shorter trips, while getting 20 to 30 percent higher WTW efficiency from fossil fuel. Serial hybrid systems instead of the current parallel ones would boost that efficiency somewhat, especially with motor-in-wheel designs.

Back to power production, and the target of adding roughly 1/4 to U.S. electricity production. For the U.S. about 50% of it's power comes from coal fired power plants. Many of these are aging and will need to be replaced, even more when the desire to reduce emissions of SOx, NOx, mercury, radioactives, and other heavy metals, are taken into account.

The older plants run at 35% to 40% efficiency, the most modern plants are in the 45% to 50% range; thus a one to one replacement based on coal feed rate would result in a 10% increase in output per unit of fuel or a 20+ percent in terms of electrical output. If the U.S. could manage to be as efficient as the first world, these plants would be used as CHP plants, and in proximity to petroleum refineries that have uses for the waste heat boosting plant economics, and who produce high carbon content waste that can be used as fuel.

But on with straight power production. Needed is an increase of 1/4, simply replacing coal plants as they reach end of life can give a boost of 1/10 to 1/5, meaning an additional 5% to 15% is needed.

In the U.S. utility scale wind power is roughly the same cost as electricity from coal fired plants. Each of the States of North Dakota, Texas, Kansas, South Dakota, and Montana, has wind potential in excess of 1000 TWh, meaning about the same as the 10E15 WH needed for pure BEV. And remember this doesn't need to be added all at once, you've one to two decades to do it in. The bulk price of this power is about 40% the retail currently, and expected to drop to 2 to 4 cents per kWh.

Something is going to be needed to provide motive power for vehicles in the future. The price of petroleum is currently down, but the world is in the leading edge of the largest economic downturn in 70 years; the Baltic Dry Index has dropped 98% in a half year. The Mideast petroleum producing countries are investing significantly in solar and to a lesser degree other alternative sources of energy, and they are sitting on much of the known supply; this would suggest they see a declining petroleum supply in the future. One way or another alternatives are going to be needed to be brought online, meaning new construction of something.

As for economics, consider what the U.S. is spending on playing Hulk Smash in Iraq and in maintaining military bases in the region. This comes from taxes, and in effect is subsidising the cost of petroleum derived fuels not just in the U.S. but around the world. That $100B to $150B per year could build a lot of new power plants, then toss in the interest to be payed because much of the costs are on Uncle Sam's credit card; grand total costs are expected to reach at least two trillion dollars.

More flies in the ointment

franklyn - 12-12-2008 at 01:32

Excerpted from here _
http://www.forbes.com/forbes/2008/1124/034.html?boxes=popsto...

Calculations from Argonne National Laboratory show that a mass-market electric car will require
124 pounds of lithium carbonate.
Only 102,000 tons of lithium carbonate were produced in 2007, and only about a quarter went
into batteries of any kind. Sociedad Química y Minera de Chile S.A., or SQM, the Chilean fertilizer
and mining company that produces nearly a third of the world's lithium carbonate, forecasts that
global production will rise to 176,000 tons by 2018, 10% of which will be for automobiles, enough
for 284,000 electric-only vehicles. That doesn't even account for demand from future hybrids
using lithium batteries. The Freedonia Group, a market research firm, has forecast that worldwide
hybrid sales will hit 4.5 million vehicles in 2013. There's quite a gap here.
William Tahil, the founder of Meridian International Research, a technology consultancy in
Martainville, France, has argued that there simply isn't enough economically recoverable lithium
on the planet to support the auto industry's ambitious plans. Tahil estimated that only 4.4 million
tons of the world's lithium resources can be extracted without prohibitive cost, a supply he
believes will be quickly exhausted if lithium-ion batteries become a staple of next-generation cars.



[Edited on 12-12-2008 by franklyn]

JohnWW - 12-12-2008 at 04:00

The "conventional" mining of lithium is from insoluble Li-containing minerals found in aluminosilicate rocks associated with granite, such as lepidolite (potassium lithium aluminium silicate) and spodumene, LiAlSi2O6. From these, the Li has to be liberated using something like HF or NaOH or KOH, or at any rate something that can dissolve silicate minerals to yield water-soluble products, with the Li then being chemically separated from the Na or K. (Similarly with rare-earth metals in pegmatite, and associated granitic minerals).

An alternative source of Li, much greater in extent, and involving much simpler extractive processes, would be the sea, in which Li occurs to the extent of about 1 ppm. To extract it from sea-water, an ion-exchange resin, or a membrane, would have to be developed that specifically either binds or admits Li, but not Na, K, Rb, Cs, Mg, Ca, Sr, Zn, Fe, Mn, Cu, or other common metals in sea-water. One wonders what progress has been made on this, in view of the pending greatly increased demand for Li for rechargeable batteries. It should be very feasible, due to the very small ionic radius of Li+, compared to that of other aqueous cations, even Na+ and Mg++.

Besides rechargeable batteries, Li salts are also used in medicine as antidepressants; and LiOH solution is used on board spacecraft (on account of its much lighter weight per OH- ion compared to NaOH) to remove CO2 from air.

One also wonders whether any research has been done into the possibility of substituting Na, which is much more abundant, for Li in rechargeable batteries. However, because of the greater ionic radius and atomic weight of Na+, its mobility and ability to permeate membranes or porous media would be much less than for Li+; and its greater atomic weight would result in significantly heavier batteries even if it could somehow be used.

Li is a very much rarer element in the universe than Na in spite of its atomic number of only 3 and stable isotopes of mass numbers 6 and 7 (compared to Z=11 and M=23 for Na). This fact arises, in spite of its theoretically ready synthesis in main-sequence stars by fusion, due firstly to the difficulty of nucleosynthesis getting past mass number 5 (of which, like 8, there are no stable or long-lived isotopes of any element, this also explaining why Be and B are also relatively rare for their atomic numbers), and then its subsequent destruction by the intense flux of neutrons and other particles that later occurs in supernova explosions.

[Edited on 13-12-08 by JohnWW]

12AX7 - 12-12-2008 at 07:21

Quote:
Originally posted by JohnWW
The "conventional" mining of lithium is from insoluble Li-containing minerals found in aluminosilicate rocks associated with granite, such as lepidolite (potassium lithium aluminium silicate) and spodumene, LiAlSi2O6. From these the Li has to be liberated using something like HF.


No, I read sulfuric acid leaching.

I have some ground spodumene at home that I should try that with some time. Too bad that, although I have plenty on hand (10 pounds? I forget), it's not much lithium, even at 100% yield. As you say, LiAlSi2O6 is only 3.7%wt lithium, maybe 10% as the carbonate or sulfate.
Quote:
An alternative source of Li, much greater in extent, and involving much simpler extractive processes, would be the sea, in which Li occurs to the extent of about 1 ppm. To extract it from sea-water, an ion-exchange resin would have to be developed that specifically binds Li, but not Na, K, Rb, Cs, Mg, Ca, Sr, Zn, Fe, Mn, Cu, or other common metals in sea-water.


Or hell, go right ahead and bind all those ions (except Na, K and Mg). There's very little Fe in the ocean, for instance, and Mn and Cu might even be sold as a byproduct rather than discarded as a contaminant. In any case, given the specific size of lithium ions, such a resin should be pretty easy to create, no?

Tim

not_important - 12-12-2008 at 07:38

Plenty of other battery chemistries around. Recent improvements in lead-acid design by FireFly have improved that chemistry, they believe they can equal NiMH in storage density with a much longer life.

The Zebra battery has about the same storage density as lithium-ion batteries, it's based on sodium, NaAlCl4, Ni, and NiCl2. Applying similar methods to those being commercialised in Li-ion batteries should enhance the zebra battery's performance further.

There is a variant of Li-ion chemistry that can use sodium : http://www.nature.com/nmat/journal/v6/n10/abs/nmat2007.html


And there's the chain of solid-state ammonia synthesis using N2+steam+electricity, storing the NH3 as alkaline earth halide complexes, and NH3 fuel cells.

I like that Forbes story
Quote:
There the solution is purified and dried until all that remains are crystals of lithium carbonate. These crystals are then granulated into the finished product coveted by battery manufacturers, a fine white powder resembling cocaine.


always can tell when a businessman writes something.

bereal511 - 5-2-2009 at 14:29

Update: I won the competition! I will now spend that $25k improving my laboratory!! Finally, a decent distillation set, new chemicals, and lots lots more!

chemrox - 9-2-2009 at 02:10

You mean all you had to do to win this was make a video about a proposed competition? And all you had to do was record guys talking about it with a rolling campus and Eucalyptus trees in the background? It's like a really good tip at the track! I hope you buy a good FT IR. Tom Black Service Center has a nice one for about 3K and they're near you.

@Franklyn & JohnWW Ashland Oregon has a spring that spouts forth water rich in lithium carbonate and is called Lithia Springs. It is possibly responsible for the mellow and tolerant atmosphere that prevails there. Polverone should require his mods to drink those waters.

[Edited on 9-2-2009 by chemrox]