Debug: Database connection successful Thermal Energy Storage / Science, Technology, and Astronomy / New Mars Forums

New Mars Forums

Official discussion forum of The Mars Society and MarsNews.com

You are not logged in.

Announcement

Announcement: This forum has successfully made it through the upgraded. Please login.

#1 2021-09-24 21:47:04

tahanson43206
Moderator
Registered: 2018-04-27
Posts: 19,585

Thermal Energy Storage

For SpaceNut ... there was not a pre-existing topic containing these three words.

This topic is offered for those who might wish to have a consolidated repository for posts about a variety of thermal energy storage concepts.

Over recent times, members have contributed posts about interesting thermal energy storage methods that may well turn out to be worth considering for Mars (as well as the Earth).

This topic will (hopefully) attract updates about various concepts.

Update 2024/04/18 ... Index to posted contributed by NewMars members:

http://newmars.com/forums/viewtopic.php … 75#p221975
Post by kbd512 ... this is a long detailed project plan

Update 2024/05/12 .... this post by SpaceNut shows a crushed rock energy storage system:
https://newmars.com/forums/viewtopic.ph … 91#p222991

(th)

Offline

Like button can go here

#2 2021-09-25 16:05:22

SpaceNut
Administrator
From: New Hampshire
Registered: 2004-07-22
Posts: 29,436

Re: Thermal Energy Storage

This boils down to energy quality of temperature source hours of input -- minus energy used to circulation or flow through the storage media capture that is =  saved as an isolated volume of stored temperature.

Then to do work requires again a minus to create flow through the convertor which in turn makes the energy type we desire.

Offline

Like button can go here

#3 2021-09-25 18:43:48

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,889

Re: Thermal Energy Storage

The greatest advantage to Thermal Energy Storage (TES) systems is longevity, durability, and simplicity, well beyond what any Electrical Energy Storage (EES) systems have demonstrated, with super capacitors being the sole exception.  The energy density of commercial super caps are well below both Lead-acid / Lithium-ion / TES / Mechanical Energy Storage (MES) systems (flywheels / weights moved around with cables and pulleys and gravity / pumped hydro), when Watt-hours per kilogram is taken into consideration.

A well-built TES can last for many decades.  They do require periodic maintenance, but components can be repaired or refurbished, rather than entirely replaced.  Corrosion tends to be the most significant problem.  Certain types of low energy density batteries can be refurbished if they're built in such a way that disassembly is possible, and some commercial Lead-acid batteries fit this description.

A TES can be scaled to any size required, whereas all EES have practical limits due to electrical resistance, complex series-parallel wiring schemes, and the requirement for electronic control, which describes all types of Lithium-ion batteries, which require very strict voltage and current regulation, only achievable with integrated circuits, to prevent cell damage or catastrophic failure.

For example, the Lithium-ion batteries used to power a cordless drill or laptop can use air or conduction cooling and a single chip for voltage and current regulation, whereas any practical battery pack for an electric car requires liquid cooling and radiators, as well as hundreds of microchips and wiring runs for voltage and current regulation.  As the voltage / current / energy capacity of the battery increases in size, that increase in overall solution complexity scales linearly with the output demanded of the battery pack.

A TES, in contrast, need only scale up in physical dimensions.  For example, NREL's Supercrititical CO2 (SCO2) gas turbine that produces 200kWe of output looks like a miniature version of the 2MWe version, with the 2MWe version a miniature version of the 200MWe version.  The Printed Circuit Heat Exchanger (PCRE) for all versions is the exact same technology, with an increase in dimensions as thermal power transfer requirement increases.  Ditto for the hot and cold reservoirs and piping.  You'll see more valves and piping, the larger the solution becomes, in order to control power output level, but the total solution is not fundamentally different.

Air vs liquid cooling and 1 microcontroller vs 100s to 1,000s of microcontrollers communicating with each other to "wear-level" every cell in the battery pack, as close to identically as they can, are markedly different technologies with markedly different levels of complexity.  It doesn't matter what something "looks like" upon casual observation, either, it only matters how it actually works, or doesn't work.

The energy cost and therefore monetary cost of the materials used is also dramatically better for TES than other forms of energy storage.  TES is concrete and steel, with few exceptions.  To date, all types of batteries are considerably more energy-intensive to produce for a given total energy storage requirement.  All synthesized gaseous and liquid hydrocarbon fuels are also far more energy-intensive, despite having gravimetric and volumetric energy densities at least an order of magnitude above EES and TES.

TES can be seamlessly paired with solar thermal and nuclear thermal and combustion technologies to maximize total output and efficiency.  In every nuclear power plant, at least 50% to 60% of the waste thermal energy is rejected to the atmosphere or to a cooling pond.  If this waste heat was captured in a TES, then a nuclear power plant could provide both base-load power and load-following capabilities to supply ramp-up / ramp-down power during the daytime, negating the need for additional load-following gas turbines burning natural gas or coal slurry.  The same is true of a solar thermal power plant.  The Sun can't be throttled and a large nuclear reactor needs to operate at maximum output continuously for sake of efficiency and longevity, yet all that waste thermal power output can be funneled into a TES for dramatic cost savings.  For countries that operate coal-fired or gas-fired boilers, again,  all that waste heat can be captured and stored.  At the scale we're talking about, this is an enormous amount of energy that would otherwise be wasted.

Lastly, when it comes to maintenance, a good mechanic aided by a good training program, with a keen eye for detail, can easily spot TES components that are beginning to fail.  Nobody can eyeball a battery and determine if it's still acceptable to operate, until it's blatantly obvious that it's been severely damaged or destroyed.

I'm in favor of using EES / MES / TES where appropriate, but for powering an entire civilization, to date that has only been achieved using thermal power systems in conjunction with other types of energy production and storage systems, mostly based upon combustion.  I do not intend to assert that it's impossible to run an entire society using electricity alone, merely that there are no actual societies anywhere on Earth that solely use electrical power and energy storage, and that the impending demise of thermal power has been greatly exaggerated.  All attempts to force the use of electricity for every conceivable energy production and storage requirement are an utter waste of time, money, and brain power.  When our electrical energy technology set is truly ready to power human civilization in its entirety, then the corporations that provide power will switch without any artificial external forces dictating technology selections to them.  Both economists and engineers abhor inefficiency, but the intelligent ones accept some inefficiency when no other practical solutions exist, which is where we're at right now, in terms of technological capabilities.

The amount of money that's been spent on electrical technology development, especially photovoltaics and batteries, often without any usable result, could've easily paid for conversion to thermal energy production and storage technologies based upon solar thermal and nuclear thermal power plants with attached thermal energy storage.

Germany spent enough money on photovoltaics over the past 20 years to switch to 100% solar or nuclear thermal power.  Instead, they spent enormous sums of money on photovoltaics and wind turbines, with the net result that their CO2 emissions are exactly the same as they were 20 years ago when they started that madness, because they're burning lignite coal to produce power when photovoltaics and wind turbines don't produce any usable power.  20 years later, they now have to replace most of those photovoltaics and wind turbines with brand new photovoltaics and wind turbines.  20 years from now, since apparently nobody in Germany knows how to count, they'll do the same thing all over again- spending more money, extracting more resources, burning more coal and gas, all in a vain attempt to counteract what all of us know so well about batteries and electronics, namely that electronics don't last very long because we don't know how to make a non-trivial complexity electronic device last for more than about 10 years or so, 20 years tops in the case of photovoltaics, before significant degredation or outright failure occurs.

Conversely, we have solar thermal and nuclear thermal power plants that have been in continuous operation from the 1950s to 1970s to the present day.  The solar thermal power plant that powers NREL has been in continuous operation since before I was born, an I'm 40 years old.  No solar panels made around the time I was born are still operating today, no computers from that era are in use today, and nobody who has a lick of uncommon sense actually believes that any of the computers or other electronic devices in use today will still be operational 20 years from now.  That constant churn of temporary benefit from the latest and greatest "whiz-bang" technology is what drives both energy consumption and ultimately economic costs, inexorably upwards.  That old "you get more for your money" silliness is also wearing a little thin.  If the battery you buy today lasts for 5 years and the battery you bought 5 years ago for half the cost, also lasted for 5 years, then the only variable that measurably changed was that the "latest technology" battery you purchased ended up costing twice as much.  Since no new batteries have doubled in energy storage capacity, each new unit costs more money for less economic benefit.

All well-designed thermal power and storage plants can last for 50 years or more.  A few have been operating over a human lifetime- meaning everyone who built those first nuclear power plants is now dead, yet their life's work is still generating electricity for their grandchildren.  A thermal power plant is therefore a multi-generational asset, other competing power provisioning technologies should be compared in those terms, so that whatever perception exists that thermal power plants are more costly than notional all-electric alternatives is easily identified as the falsehood that it truly is.  During that same period of time, a wind and photovoltaic power provisioning infrastructure would have to be replaced a minimum of 2 times and likely 3 times, which means its true cost was at least double to triple the price of the initial investment, because 25 / 50 / 75 / 100 years from now, humanity will still be using electrical power.

On the go-forward, we should be building power plants and energy storage systems that can truly withstand the test of time.  Only ultimate durability measurably reduces energy demand.  The LED light bulb is one of the few examples of electronic technology that is an actual improvement over what came before it, when durability and longevity are also taken into consideration.  Whereas the old incandescent bulbs would last a maximum of 2,000hrs, but likely closer to 1,000hrs, the latest LED bulbs can easily last 20,000hrs.  Similarly, when a master blacksmith makes a very high quality steel knife once, and that tool then goes on to serve a succession of owners (temporary caretakers of the technology that built human civilization, ultimately) for hundreds of years, then whatever energy was invested into making that knife was a pittance compared to its total utility to multiple generations of users / caretakers.  Disposable appliance mentality / "planned obsolescence" has already proven to be a much greater problem for humanity than energy efficiency.

Offline

Like button can go here

#4 2021-11-21 12:44:51

SpaceNut
Administrator
From: New Hampshire
Registered: 2004-07-22
Posts: 29,436

Re: Thermal Energy Storage

I guess the best place to capture and retain for use would be in those places that are already hot.
Earth's hottest places will surprise youAAQWSam.img?h=416&w=799&m=6&q=60&u=t&o=f&l=f

Dallol, in northern Ethiopia, may not hold the record for the highest ever recorded temperature but some argue its overall average annual temperature of 95°F (35°C) puts it in top (hot) spot as the hottest inhabited place on Earth. Plus, the summer heat pushes 104°F (40°C) and above.

or closer to home
BB19eQ1v.img?h=416&w=799&m=6&q=60&u=t&o=f&l=f

Phoenix had the highest number of days at 90°F (32°C) or above per year, with an incredible 169 of them. It edged into first place ahead of fellow Arizonian, Tucson, which had 147 days annually at 90°F (32°C) or above.

Offline

Like button can go here

#5 2021-11-21 14:44:08

Calliban
Member
From: Northern England, UK
Registered: 2019-08-18
Posts: 3,814

Re: Thermal Energy Storage

Kbd512, that is an impressive post on TES.  I wonder if thermal energy storage could ever be a practical means of powering a vehicle, maybe a large truck?  If rail based transportation can provide freight transportation between regional nodes, then relatively short range trucks can provide transportation to and from nodes and surrounding areas.  The hot source on the vehicle, would be a molten salt phase change material.  The cold source could be a block of ice, or better still, a tank of liquid air, which would be a working fluid in a Brayton cycle gas turbine.

Most TES schemes are about storing heat, in situations where the end use is heat.  For hot water and space heating, water and masonry are the most likely storage materials.  A water tank is a long lived component.  There is no reason why large water tank cannot last a century.

Could we use soda-lime glass as a high temperature phase change material?  The material has a variable melting point, but most commonly around 1000°C.  Heating can be achieved using rotating magnetic fields.  Heat recovery can be accomplished using liquid sodium flowing through steel pipes, transferring heat to S-CO2 power conversion modules.  The containment vessel would be steel or cast iron, surrounded by masonry, and reinforced concrete.

Anhydrous sodium carbonate melts at 850°C.  Some steels still retain reasonable strength at this temperature. But carbon steels have lost 90% of strength.

Last edited by Calliban (2021-11-21 15:20:01)


"Plan and prepare for every possibility, and you will never act. It is nobler to have courage as we stumble into half the things we fear than to analyse every possible obstacle and begin nothing. Great things are achieved by embracing great dangers."

Offline

Like button can go here

#6 2021-11-21 15:14:46

SpaceNut
Administrator
From: New Hampshire
Registered: 2004-07-22
Posts: 29,436

Re: Thermal Energy Storage

High temperature insulation materials will be used between each material transition to contain the heat that we are trying to make use of.

Offline

Like button can go here

#7 2021-11-21 16:02:53

Calliban
Member
From: Northern England, UK
Registered: 2019-08-18
Posts: 3,814

Re: Thermal Energy Storage

SpaceNut wrote:

High temperature insulation materials will be used between each material transition to contain the heat that we are trying to make use of.

Ferritic stainless steel appear to be the preferred material for containment of molten metals and other corrosive materials at these temperatures.
https://www.outokumpu.com/en/expertise/ … ure-grades

These steels have high chromium content and very low carbon, allowing ferrite to remain stable at high temperatures.

For insulation, there are fire bricks that are stable at these temperatures.  There are also mineral wool products, which appear to be based on silica with calcium and aluminium oxide inclusions.
http://fabsrv.com/thermal-insulation/hi … nsulation/

Sand would be a cheap insulating material as well, as would calcium oxide.  Concrete would need shielding from these temperatures, as it starts to take damage at temperatures greater than about 200°C, due to dehydration.  One of the complications in building high temperature nuclear reactors, is the need to keep the concrete (which provides structural support and shielding) cool.  Fail to do that and shrinkage can occur, along with cracking that lowers compressive strength.

The company I linked provides high temperature heating pads as well.  However, a fast rotating magnetic field would directly heat the stainless steel containment vessel, which would then transfer heat by conduction into the sodium carbonate melt.  Heat recovery could be via tubes running through the melt, or tubes running outside the vessel, relying on conduction through the walls.  The second would be a much easier design solution, at the expense of a relatively large temperature drop between the molten sodium carbonate and the sodium coolant running through the tubes.  It could make replacement of failed tubes relatively straightforward.  Tubes would run vertically, through holes cut in aluminium oxide blocks in contact with the vessel.  These would conduct heat out of the vessel, and transfer heat to the sodium carrying pipes through thermal radiation.  Failure of an individual sodium pipe can then be dealt with by draining the sodium, cutting out the faulty pipe and replacing it.

A thermal store like this would be designed to receive intermittent electricity and generate electric power constantly, 24/7/365.  Thermal systems like this do not well tolerate the thermal gradients associated with cooling down and warming up.  Those sorts of transients cause thermal shock, which is differential expansion of materials, leading to stress fractures.  The thermal store operates at close to constant temperatures for most of its life, because any heat input is used to melt the sodium carbonate, whilst heat withdrawal solidifies it.  So long as the store contains the phase change materials in both phases, then temperatures remain constant.  Only occasionally will its temperature drop beneath its melting point, during deep lulls in power generation.  Such a device could easily last 100 years, if thermal shock is avoided.  The power generation loops will need to be replaced eventually.  But if multiple loops are used, one can be removed, whilst the others remain in operation.  The thermal store itself will maintain a constant temperature throughout its lifetime.

The specific heat of sodium carbonate is 112.3 J/mol-K.  The heat of fusion is 29.7KJ/mol.  Density is 2530 kg/m3 @20°C.  Molecular weight = 0.106kg/mol.
http://www.chemister.ru/Database/proper … .php?id=66

A single cubic metre of sodium carbonate contains 23,868 mols.  Complete melting will absorb some 709MJ.  A tank of 1000m3 would allow generation of 100MWe (at 50% efficiency) for 1 hour.  A tank 24,000m3, would be a right circular cylinder, some 30m in diameter and 30m high.

A thermal store like this would best be built close to the location of a wind farm.  By absorbing intermittent electricity and discharging more dependable energy, it would allow better utilisation of transmission infrastructure.  Power output from a 1000MWe-p wind farm fitted with a thermal store, would rarely drop below 100MWe, because the store would provide an effective baseload.  However, power levels onto the grid would also never exceed 400MWe, because generation above that level would be absorbed by the store.  The result would be a far more steady and dependable power output, at the expense of perhaps 25% of the electricity generated by the wind farm.  During rare deep lulls in wind capacity, gas turbines would generate power.  But these would only be needed for perhaps a week or two over the entire year.  Fuel requirements would therefore be small.

Last edited by Calliban (2021-11-21 17:04:24)


"Plan and prepare for every possibility, and you will never act. It is nobler to have courage as we stumble into half the things we fear than to analyse every possible obstacle and begin nothing. Great things are achieved by embracing great dangers."

Offline

Like button can go here

#8 2021-11-21 18:15:27

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,889

Re: Thermal Energy Storage

Calliban,

My gut feel is that a pure thermal power storage system lacks the volumetric and gravimetric energy density required to power a heavy duty truck in a practical way.  A train only requires 1/9th the power of a road-bound motor vehicle using rubber tires, in order to move a given unit of weight at a given speed.

Molten Silicon: 528Wh/kg and 4.3MJ/L @ 1414C
Lithium-ion batteries: 250Wh/kg and 2.6MJ/L
Solid State Lithium-ion batteries: 500Wh/kg and 5.2MJ/L

I would rather have solid state Lithium-ion batteries and deal with the electrical complexity than safely storing multiple cubic meters of molten metal aboard a high speed vehicle.  1 cubic meter provides 1.23MWh of energy and weighs 2,328kg, just for the molten Silicon.  If an equivalent solid state Lithium-ion battery achieves 250Wh/kg at the pack level, then it's 4,920kg, and I don't have to store something hot enough to melt Inconel.  The only way to store molten Silicon is with an Alumina-based ceramic, which melts at 2,072C.  Alumina Oxide is 3.95g/cm^3- heavier than Aluminum metal.  The tank and its thermal insulation will weigh every bit as much as a solid state Lithium-ion battery.  You need some kind of very thin Tungsten liner on the inside and outside, as well as Inconel hardware to secure the tank to the vehicle.  The exterior tank liner would be hot enough to light a cigarette on at all times, so an overwrap of insulation is required.  You'd also need a way to gradually cool the tank for inspection or de-fueling (removing the molten Silicon for vehicle storage).  Overall, solid state Lithium-ion and molten Silicon solutions are both very impractical at significant scale and require Silicon free of impurities.  For stationary storage, maybe, but at the scale human civilization requires, it's just not very practical.

Compressed Natural Gas: 14.9kWh/kg and 9MJ/L @ 3,600psi
Liquefied Natural Gas: 14.9kWh/kg and 22MJ/L @ CH4's Boiling Point
Liquefied Petroleum Gas: 13.8kWh/kg and 25MJ/L @ 125psi at room temperature
Gasoline: 13.2kWh/kg and 34.2MJ/L
Diesel: 12.7kWh/kg and 38.6MJ/L

Trains powered by molten Silicon are much more practical.  There are far fewer of them (approximately 100,000 locomotives across the entire planet) so one-off fabrication rather than a dedicated assembly line is more feasible to do, their operators receive specialized and ongoing training, and trains are very large masses of rolling steel, which means an accident that cracks a molten Silicon tank doesn't spill molten metal all over a roadway, which would require repair prior to reuse.

A prototypical locomotive stores around 4,000 US gallons of diesel fuel, or roughly 15m^3, so an equivalent volume of molten Silicon is 18.45MWh.  Each US gallon of diesel contains around 40,588Wh of energy, but we only obtain around 13,529Wh of usable energy from it using a turbo diesel engine.  In total, 4,000 gallons of diesel provides about 54MWh of energy.  The average locomotive is 6,000hp to 7,000hp, so if it uses half the horsepower at a given speed, then around 2.250MWh per hour of operation, so roughly 24 hours of operation prior to refill.

If you get around 70% of the stored thermal energy in the molten Silicon, via Thermal Photovoltaic Cells, which is what we can achieve right now with existing TPV technology, then 12.9MWh, and then electrical efficiency is around 96%, so 12.4MWh all said and done.  The weight of the massive diesel engine could be replaced with more molten Silicon, so all said and done you could have around 16MWh of energy for equivalent weight.

To obtain like-kind performance, 44m^3 of molten Silicon would be required, and this equates to 102,432kg.  Current locomotives weigh 220,000kg, so it seems probable that heavier locomotives or a redesign of the locomotive chassis would be required to keep total weight equivalent while providing like-kind performance.  Basically, the locomotive becomes a tanker car, looking somewhat like a traditional coal-fired locomotive, but powered by electric motors and TPV cells inside the molten Silicon tank.

Is this doable?

Short answer is yes.  Long answer is that it still requires a lot of structural design work to assure the tanks of molten Silicon are appropriately protected in the event of a derailment.

Offline

Like button can go here

#9 2021-11-22 03:37:58

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,889

Re: Thermal Energy Storage

Calliban,

The US DOT-117/117R specification tank cars are as follows:

Trains.com - DOT-117 Tank Car Specifications

DOT-117 tanks are 9/16" steel shells, thermal protection material thicknesses as required, followed by 1/8" steel jackets, followed by 1/2" steel heat shields on both ends of the tank.  Maximum speed limit is 50mph.  Any tank car that does not meet the new 117/117R standard is limited to 40mph.  The steel used must be TC-128 Grade B normalized carbon steel, which is a specification defined by the Association of American Railroads.

Sentry Rail - DOT-117 29,000 Gallon Insulated and Jacketed Tank Car

Further Tank Car Specifications:

Railway Supply Institute - Tank Car Standards in North America

AAR TC-128 Carbon Manganese Steel Specifications:

NIST.gov - MECHANICAL PROPERTIES AND FRACTURE TOUGHNESS OF AAR TC128 GRADE B STEEL

AAR TC 128 Grade B/AAR A516, Grade 70

The Association of American Railroads' Manual of Standards and Recommended Practices, Section C-III, Specification for Tank Cars identifies two grades suitable for the construction of railcar tanks. TC 128, Grade B is a high-strength carbon manganese steel while the AAR ASTM A516, Grade 70 is a carbon steel produced in accordance with the supplementary requirements of AAR Specifications for Tank Cars for use in tank car heads, shells, and sumps. Each grade must meet the requirements identified in Appendix M-1002 of the AAR Tank Car Manual.

TC-128 has 50ksi YS and 81ksi-101ksi UTS

Since any type of steel alloy, including something as exotic as Hastelloy C276, would simply melt or have no remaining strength at 1,414C, the melting point for Silicon, we have 3 suitable materials available in sufficient quantities:

1. W-Ni-Fe Tungsten alloy - to hold onto the molten Silicon
2. W-Ni-Cu Tungsten alloy - to collect current from the TPV cells
3. 99.9%+ pure Alumina Oxide ceramic - insulator material between the tank and outer jacket

90W-7Ni-3Fe is 94ksi-123ksi YS, but I'm still hunting for UTS and Elongation % is very similar to TC-128, but 17.1g/cm^3 vs 8.05g/cm^3 for carbon steel.

9/16" x 36" x 36" of TC-128 is ~206.7lbs
9/32" x 36" x 36" of W-Ni-Fe is ~225.5lbs

W-Ni-Fe is approximately twice as strong as TC-128, so we need about half as much for equivalent strength.

Silicon-Carbide ceramics would oxidize, so those are out.  All types of Iron and Titanium alloys are out, because they have near-zero strength left at the required operating temperatures, or are already liquid.

We need 44m^3 / 11,623.6 gallons of molten Silicon to produce equivalent electrical energy as 4,000 gallons of diesel fuel, given the relative efficiency of both energy sources.  The 29,000 gallon capacity of a US DOT-117 tank car is about 2.5X the 11,623.6 gallons of molten Silicon required, so we will have a much shorter / stiffer structure.

DOT-117 specifies tank OD as 119.25", so a 118.96875" tank ID x 244" tank length provides 11,742 gallons of capacity (extra capacity for thermal expansion and in-tank thermal-to-electrical power conversion apparatus).  Bottom line is a 20' 4" (molten Silicon tank length) vs 48' 9.5" (DOT-117 tank length).  We could also afford to use a thicker W-Ni-Fe plate for a nominal increase in weight.

The 44m^3 of molten Silicon weighs 102,432kg + 29,500kg for the tank car, and we're at 131,932kg, and still well below the 220,000kg of a locomotive.  A 5,000hp GE electric motor is 13,612kg, so figure on 15,000kg for the electric motors, so now we're at 146,932kg.  Throw in another 15,000kg for W-Ni-Cu wiring and we're at 161,932kg.  That's right in line with a garden variety diesel electric locomotive, for a complete zero-emissions electric locomotive that provides like-kind capability with respect to a diesel engine powered locomotive.  Regardless, we have more than sufficient fudge-factor to provide equivalent capability at the same or lower weight.  I forgot to add the tank insulation, but it won't weigh that much.  The insulation would have to be Alumina Oxide powder or ceramic blocks.

Now for the expensive part:

W-Ni-Fe is around $55/kg, so if we use 20,000kg, that's $825,000 USD.  Diesel locomotives range in purchase price from $1,200,000 to $3,000,000 USD.  That would mean the cost of the Tungsten alloy tank alone is a significant fraction of the total purchase price.  Add the Alumina Oxide insulation, and we're easily at $1,000,000 USD.

According to the article below, a 6-axle diesel locomotive is 73' in length, weighs more than 200,000kg, contains more than 6 miles of wiring, and can contain as many as 210,000 parts.  Our molten Silicon locomotive would be less than half as long, but probably every bit as heavy, though it would contain far fewer parts and far fewer moving parts.  It is a high technology vehicle, though, and TPV cells are very expensive.

Railway Age - Does Rebuilding Locomotives Beat Buying New?

If the morons presently in charge of the federal government are insistent upon moving Canadian crude via railway, instead of the pipelines that they're firmly against, because they lower crude oil prices, then a low-maintenance locomotive that doesn't burn diesel to move product to market would be a boon to both the railway industry and the American consumer, who is over-burdened with the idiocy of the radical leftists who hate people and don't care about our environment.  Anyway, we've hit upon a feasible way to counter their stupidity and arrogance.  It will be very humorous to watch their opposition to a new railway transport technology that doesn't burn fuel and lowers the cost of crude oil transport from Canada to the oil refineries in Houston.

Offline

Like button can go here

#10 2021-11-22 04:16:04

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,889

Re: Thermal Energy Storage

There are around 25,000 diesel locomotives in the US, so that's the total American market for this technology.  Worldwide, there were 100,472 locomotives of any description in 2017.  That number has since fallen due to COVID.  If you figure that each new locomotive using molten Silicon technology will cost around $4M USD, then the total US market is $100B and the total global market is $400B.  However, US EPA states that railway freight only accounts for 0.5% of US CO2 emissions.  US railroads move 40% of all freight, but only account for 1.9% of the total transportation-related greenhouse gas emissions.  That probably explains why there's no great push to eliminate diesel locomotives.  On average, US railways move 1 ton of freight 480 miles using 1 gallon of fuel.  No other form of fossil fuel based transportation is nearly as efficient.  There are probably much better uses of $100B USD, if the goal is to reduce emissions.  The compressed air powered car is one such technology.

Offline

Like button can go here

#11 2021-11-22 04:48:48

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,889

Re: Thermal Energy Storage

We only get 861,000Wh/m^3 at 70% efficiency, not 1,230,000Wh/m^3, which means we need a lot more molten Silicon.  I messed up that calculation, as so often happens when I have other things going on.  That means we need 63m^3 of molten Silicon.  Now we're over our weight target for a molten Silicon powered locomotive that provides an equivalent of 54MWh worth of energy, because we now need 146,664kg of molten Silicon.  We could raise the temperature of the molten Silicon to achieve higher thermal efficiency, but it would only be a partial offset.  Even so, it doesn't detract from the overall feasibility of the concept since locomotives refuel once or twice per week.  Thus, there's no reason why a mid-week re-heat of the Silicon would materially detract from the usability of the locomotive.  A pair of our smaller molten Silicon locomotives would occupy the same physical space as a single larger diesel powered locomotive, and that has to be worth something.

Offline

Like button can go here

#12 2021-11-22 17:25:08

Calliban
Member
From: Northern England, UK
Registered: 2019-08-18
Posts: 3,814

Re: Thermal Energy Storage

If you can retrieve 861kWh from 1m3 of silicon, weighing 2570kg, then mass energy density is comparable to the best Li-ion batteries.  We have the same problems with charging thermal stores as we do with batteries - fast charging requires enormous power.  Do we end up having to have extra locomotives, so we can charge some whilst the others are in use?  Materials costs will be lower than Li-ion batteries.  The catch is that this is a system only works on large scales.  Freight trains and maybe ships, where the thermal stores could provide ballast.

If electricity is the source of energy, then one advantage with electrification of the railway is to spread those electrical power demands over the entire length of the journey.  But electrification costs a pretty penny.

Could we use biomass to power freight trains?  By this I mean chipped biomass that can be fed onto a gasification burner on the train, that powers the engine?  One the positives here is that 1kg of dry woody material contains up to 20MJ of stored energy, about half as much as diesel.  Things like miscanthus giganteus would give good yields across the southern US and is drought tolerant.  It produces dry, woody stems at harvesting and would be perfect for chipping.  As a feedstock for liquid fuel production, yields per acre would be far superior for miscanthus giganteus than they would be for corn ethanol.  And miscanthus can be grown on more marginal land, with hilly relief, or poorer soil and limited rainfall.

Last edited by Calliban (2021-11-22 17:30:34)


"Plan and prepare for every possibility, and you will never act. It is nobler to have courage as we stumble into half the things we fear than to analyse every possible obstacle and begin nothing. Great things are achieved by embracing great dangers."

Offline

Like button can go here

#13 2021-11-22 22:29:01

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,889

Re: Thermal Energy Storage

Calliban,

Unlike any type of battery except a flow battery, you can replenish thermal energy using supplies of molten Silicon stored at railway yards, in much the same way that they store diesel fuel, except you need a lot less molten Silicon than diesel fuel.  Doing that requires heating and it requires specialized pumping equipment made from refractory metal alloys or ceramics, but it can be done.  You can't as easily swap Lithium-ion batteries because they are not primarily liquids that can be pumped between holding tanks and they have electrical contacts subject to corrosion.  You don't need extra locomotives.  You need additional supplies of "recharged" / "heated back up to operating temperature" molten Silicon kept at railway stations, using electrical power from the grid.  Note how often a single train requires replenishment.  It refuels once per week.  That is doable without requiring a massive discharge of electrical power over the span of minutes to hours.

Think about how this actually works.  Assume that 1/2 of all US trains (12,500 locomotives) are operating at any given time, and using 2,500hp to maintain a modest speed.  This is objective reality (real world train usage), but let's play with the numbers a bit more.

12,500 trains * 2,500hp = 31,250,000hp = 23,303,505Watts = 23.3MW

Assume all 25,000 trains were operating at the same time = 46.6MW

Assume all 25,000 trains were using all 5,000hp (even though most locomotives don't have 5,000hp) = 93.2MW

Can we supply enough electrical power over several days to assure that half of those trains can consume a fresh tank of molten Silicon at a railway yard?

We can do that easily, without adding a single Watt of grid capacity.  It puts nowhere near the strain on the electrical grid that fast charging millions of battery powered motor vehicles does.  The total power that all locomotives in America (1/4 of all trains in the world) can consume at any given time, if all of them are using 100% of available horsepower, is equivalent to 1 GE-90 turbofan at takeoff power.  If we doubled the number of trains to carry more freight, then it's equivalent to a twin GE-90 powered 787 takeoff, so we can still do it.

Lest we forget, conversion of electrical to thermal heat is nearly 100% efficient.  Long distance transmission of electricity is the real problem, because the internal resistance of commercial solar photovoltaic cells is so bad that 20% to 25% of the photon energy collected is lost as heat at the solar farm, before a single Watt of electrical power is transmitted anywhere outside the solar farm.  This is a major reason why "less efficient" solar thermal power systems end up being as efficient or more efficient than photovoltaics.  Thermal photovoltaics are a different kind of technology, and the efficiency of these cells can vary wildly dependent upon type and implementation.

Millions of tons of batteries, square miles of photovoltaic semiconductors, the wiring maze, and sophisticated computer control networks is what costs a pretty penny.  Steel is cheap.  Unrefined Silicon is cheap.  Salt is cheap.  Concrete is cheap.  Even Tungsten, by way of comparison to rare Earth metals ($55USD/kg for Tungsten vs $628USD/kg), is quite cheap.

Why would we burn wood or corn ethanol or coal to power trains?

We simply don't need to.  We already have a number of "better ways" of powering trains.  If we wanted to power trains off of biomass, then we'd use algae to produce synthetic diesel fuel, because that's the most efficient method by far.  Trees are too scarce to burn to power trains.  We want more trees, not less.  Algae requires zero land use and doesn't care about drought at all since it'd be grown in the ocean.

Unlike Lithium-ion batteries, a molten Silicon based solution is another one-time investment that lasts for a human lifetime.  When you're done using the locomotive, you drain off the Silicon, and then you recycle the Tungsten and Iron-based alloys into a new locomotive.  We can't easily do that with current locomotives, nor can we do it with Lithium-ion batteries and photovoltaic panels.  Only monolithic metal and ceramic products are easily recycled.

The major advantage provided molten Silicon is that we go from 200,000+ parts to perhaps 20,000 parts, and a literal handful of moving parts.  Since there are zero emissions, US EPA rules about when you can refurbish vs scrap a locomotive are no longer applicable.

Offline

Like button can go here

#14 2021-11-22 22:52:50

tahanson43206
Moderator
Registered: 2018-04-27
Posts: 19,585

Re: Thermal Energy Storage

For kbd512 re molten silicon branch of Thermal Energy Storage ....

The (to me surprising) development of this topic is sparking my interest, and I hope you will carry your vision a bit further.

I did a quick check with Google and found paper after paper are available on the subject of molten silicon, because it is so central to the semiconductor industry.

If you've already covered what I'm about to ask, please just point me to the posts ... I don't want to ask you to repeat anything...

Questions:

1) I agree that delivery of thermal energy from the grid to a thermal energy store is nearly 100% efficient.

2) How would you pull that thermal energy out of the store on a train or other consumer of energy?

I can imagine using water as a medium, for example, but that's fairly inefficient.  Perhaps there is a more efficient way of converting thermal energy into motion?

3) What kind of storage containers are needed to hold molten silicon?  How is heat flow from the hot silicon to the surroundings reduced?

4) is there a risk management component to system design?

Again, if you've already covered this, please just point to the posts ... no need to repeat.

(th)

Offline

Like button can go here

#15 2021-11-23 21:56:16

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,889

Re: Thermal Energy Storage

tahanson43206,

The tank of molten Silicon contains thermophotovoltaic (TPV) cells.  These are specialized extremely high temperature solar panels that are either bathing in the molten Silicon that provides IR photons for them to work, or as close to the photon source (molten Silicon) in the container, as is feasible.  Unlike normal / non-concentrator photovoltaics, TPVs have power densities measure in tens of kilowatts per square meter.

The particular TPV technology I had in mind provides a little over 77.7kWe/m^2 at 1,500K (1,226.85C), but we're going with 75kWe/m^2.  To produce 5,000hp (746W/hp * 5000 = 3.73MW; 3,730,000 / 75,000 = 49.7333m^2), so call it 50m^2.  We're going with the interior dimensions of the tank car I undersized due to my TPV efficiency calculation error (don't try to do math and handle wife + kids at the same time), but even that has 58.8m^2 of lateral surface area.  However, simply coating the interior walls of the tank in TPV is the wrong way to do this.  You need a bunch of little TPV pipes evenly spaced inside the much larger molten Silicon tank, similar to the heat pipes inside the boiler on a ship or a train.

The reason I oversized the tank by 100 gallons or so, was to accommodate interior ALON pipes containing TPV coated interior walls with W-Ni-Cu (an electrically conductive Tungsten alloy) current collectors to extract the electrical power from the TPV tubes.  The interior of our Tungsten alloy molten Silicon containment tank is a "solar electric boiler", filled with much smaller tubes containing TPV cells and electrical conductors, very similar to boilers containing lots of much smaller Copper or Iron tubes filled with water, in order to produce steam.  In our case, we transfer IR photons to TPV to directly produce electricity that can be consumed by the electric motors of a prototypical diesel-electric locomotive.

Try to imagine a cross between a spark plug and a long fluorescent light bulb.  The glass is the ALON material.  The phosphor coating on the inside of the bulb is replaced by TPV cells.  Rather than a gas, the center of the "bulb" contains W-Ni-Cu current collectors (like a spark plug).  Now imagine dozens of these things running the length of the molten Silicon tank.  Each end of the TPV rod will be capped with an Alumina Oxide / ceramic insulator (like a spark plug) and a pair of electrodes (like a spark plug) to collect electrical power (complete an electric circuit).  Several interior W-Ni-Fe plates (like a drain plug with a bunch of circular holes in it- to support the weight of the rods, spaced every meter or so) will hold these "thermal electric light bulbs" in place.

Storage containers at the "recharging station" / railway station will be relatively thick Alumina Oxide tanks with very thin W-Ni-Fe liners and embedded W-Ni-Cu heating elements.  The storage tanks will be below ground and surrounded by concrete, in order to prevent a tank rupture from spilling molten Silicon onto the ground.  The locomotives will pass by the tanks, discharge their "cooler" molten Silicon, and replace it with fresh / "hot" molten Silicon using special pumping equipment (mostly ceramic and W-Ni-Fe alloy) designed to handle the intense heat.

It's important to keep Oxygen out of the Silicon.  I need to think a bit more about the best way to do that.

Everything I've laid out here is intended to keep the Silicon well-insulated and to prevent accidental spills.  The Silicon is hot enough to melt any kind of steel, including all super alloys like Hastelloy, so containing it properly is of the utmost importance, similar to the fissile material and fission products inside a nuclear reactor.

Nothing is ever risk-free, but we know how to contain molten metal.  There's no danger of a thermal-runaway, as there is with a battery.  It's not flammable or explosive, either, as is the case with all Hydrogen-based fuels in oxidizing environments (Earth's atmosphere).

Offline

Like button can go here

#16 2021-11-24 01:27:32

Calliban
Member
From: Northern England, UK
Registered: 2019-08-18
Posts: 3,814

Re: Thermal Energy Storage

Kbd512, 1HP = 746 watts.  To power all 25,000 US trains (assuming they run at 100% capacity) would require 46.6GW of constant power.  But your conclusion is still valid: this is not an unrealistic burden on the US generating system.  It would add maybe 10% to US baseload power generation.  This isn't a bad deal, given that rail is actually the back bone of the US freight transportation system.

I think the way this works may have to be a little different to what you envisage.  Silicon is a phase change material in this context.  The energy stored will be a mixture of sensible heat and latent heat.  We are heating silicon past its melting point.  As it melts, it absorbs large amounts of energy breaking the inter-atomic bonds that would bind individual atoms into a crystal structure up until it melts.  In much the same way, boiling water or melting ice, consumes a huge amount of energy without increasing temperature.  This is the energy transition that we exploit in a phase change material.  It has the advantages of high energy storage density and a constant temperature, which makes it easier to design things like heat exchangers and turbomachinery with constant thermal gradients, constant pressure ratios and constant vapour quality.  This is why boilers and nuclear reactors are designed to operate at one temperature, which is usually as hot as possible.  Changing operating temperature would cause all sorts of problems.

So you're tank of molten silicon (or silicon based alloy?) will probably operate in a narrow temperature range.  As you withdraw heat, it will freeze molten material within the tank.  You recharge it by melting it at your recharging station.  This means that you probably won't be transferring molten silicon.  Most likely, you will have a large tank of phase change material at the filling station, with a temperature somewhat higher than the operating temperature of the silicon based material on your locomotive.  You will transfer heat between them using intermediate heat exchangers: probably using an inert gas as the heat transfer fluid.  Nitrogen would be an option; argon would be the best option; carbon dioxide would probably be too reactive at ~1400°C.

At the temperatures we are talking about, the large tank in the rail yard will probably be heated using magnetic fields that heat the material via eddy currents in the same way that electric furnaces are heated.  Direct contact resistance heating would be more difficult at those temperatures due to materials compatability problems.  The silicon could be doped with iron to improve its electrical conductivity.  We would dope the silicon in the locomotive to reduce its melting point compared to the tank in the yard.  We need it to melt at a lower temperature to allow heat transfer during recharging.  An alternative would be to store molten silicon dioxide in the yard tank and molten silicon in the loco tank.  That would give a 200°C temperature difference between the two.

Presumably, any energy storage system capable of powering a freight loco could also power a ship.  The molten silicon tanks could in fact provide ballast if we put them along the keel.  A PV power generation system would generate electric power that can then power a propulsor of some kind.  It would be interesting to explore this option for powering a cargo ship.

In terms of risks inherent to this strategy: The most obvious one is fire.  At these temperatures, combustible materials do not need to come into contact with molten silicon or hot surfaces to ignite.  Thermal radiation from anything that hot will autoignite just about any combustible material.  The way around that is to insulate all hot surfaces.  Thermal shock is a big concern.  Rapid uncontrolled cooling will lead to thermal stresses that will fracture metal vessels.  It is a big problem at much lower temperatures than we are talking about here and becomes more of a problem as operating temperatures increase.  Good encapsulation of hot surfaces, to prevent water ingress, is essential.  At the temperatures we are discussing here, water ingress could result in steam explosions.  None of these hazards are problems that cannot be managed.  The designers need to be aware of them.  And the plant maintenance schedule needs to have strict procedures in place for maintaining ingress protected lagging and preventing water ingress.  In all sectors of engineering, people's lives depend upon getting these sorts of things right.

The PV thermal energy power generation system is fascinating.  Conventional PV has efficiency limitations as large amounts of photon energy are wasted if photon energy is greater than or less than the band gap of the semiconductor.  But in a thermal system, this limitation does not exist, because wasted energy is reradiated back into the thermal store.  But it will effect power output.  One limitation of this technology: the photon flux received by the PV surface is a strong function of temperature, it scales with the fourth power of temperature in accordance with the Stefan-Boltzman equation.  So a 10% reduction in temperature leads to a 33% drop in flux.  With a 20% drop, flux goes down to just 41%.

Last edited by Calliban (2021-11-24 04:18:54)


"Plan and prepare for every possibility, and you will never act. It is nobler to have courage as we stumble into half the things we fear than to analyse every possible obstacle and begin nothing. Great things are achieved by embracing great dangers."

Offline

Like button can go here

#17 2021-11-24 04:12:27

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,889

Re: Thermal Energy Storage

Calliban,

You're right.  I did it again.  I have a constant stream of distractions lately.  I must have hit the divide key instead of the multiply key.  Again, that's what I get for not checking.

BTW, it looks like your figure of 46.6GW is for 2,500hp, not 5,000hp.

(25,000 trains * 5,000hp * 746W/hp) = 93,250,000,000W = 93.25GW

I respond to the rest later.

Offline

Like button can go here

#18 2021-11-24 05:28:39

Calliban
Member
From: Northern England, UK
Registered: 2019-08-18
Posts: 3,814

Re: Thermal Energy Storage

This combination of technologies: a high temperature and high energy density thermal energy storage material, coupled with efficient solid state power converters (heat to DC current), has the potential to be revolutionary.  This is especially the case if charging can be carried out using resistance heaters that are reliable over decades and a stable, high performance insulation solution can be developed.  The components of the system are all relatively cheap, face none of the resource limitations of Li-ion batteries and like a battery, there are no moving parts.  Another big advantage is that resistance heating is a resistive rather than an inductive load.  It can respond quickly and receive very large energy fluxes without taking damage.  This allows a system that is high energy density, can absorb large amounts of power and is cheap and fundamentally simple.  It does not degrade appreciably during charging, so can be topped up with power almost continuously.

If high temperature aerogel insulation can be developed, something like this might power a truck.  Discharge rate is a function of temperature and the surface area to volume ratio of the phase change material.  We could enhance that by embedding the PV material into the phase change material itself, or pumping the PCM through power converters.  But avoiding moving parts that have to be exposed to superheated material is a big selling point of this concept.

These units employed in stationary applications could function like capacitors.  They would absorb excess power and release it slowly at a steady discharge rate.  That discharge rate could be tailored to the application.  For grid storage, we might decide that we want a week of storage capacity.  In that case, build a bigger tank with the PV modules around the edges.

It is less likely that this system could scale down small enough for automobiles.  Maybe a house sized unit would ultimately be possible.  But large buildings could certainly use a system like this.  Any thermal losses can feed into the building heating system, so overall efficiency is close to 100% and exergy efficiency would be around 80%.

Molten silicon technology would work equally well in situations where high quality heat is needed as an end use.  Large parts of industrial energy use is in the form of heat.

The recovered energy density of a molten silicon PV system, is around 1MJ/kg.  This is still about an order of magnitude lower than the recovered energy from burning a chemical fuel.  So this system is less suitable for powering an aircraft.  There may be some advantages in airframe shape that we could exploit to partially close the gap.  An electric aeroplane does not need the huge nacelles and air intake of a jet engine, which is a massive component of drag.  But the plane would probably need superconducting electric motors to keep engine weight to a minimum.  A thermal battery allows fast recharging in a way that isn't practical for an electrochemical system.

Last edited by Calliban (2021-11-24 05:55:42)


"Plan and prepare for every possibility, and you will never act. It is nobler to have courage as we stumble into half the things we fear than to analyse every possible obstacle and begin nothing. Great things are achieved by embracing great dangers."

Offline

Like button can go here

#19 2021-11-24 07:57:46

Calliban
Member
From: Northern England, UK
Registered: 2019-08-18
Posts: 3,814

Re: Thermal Energy Storage

Darn!  You know the old adage about something that seems too good to be true?
https://en.m.wikipedia.org/wiki/Thermophotovoltaic

I wasn't familiar with thermal PV, but assuming the wiki article is accurate, there are two things of note:
1) Realistic efficiency is nowhere near 70%, it is closer to 10%;
2) There apparently must be a temperature difference between the hot source and the PV cells.  It should have been obvious, but without that, the cells would radiate as much heat as they receive and efficiency would be zero.

High temperature TES could still be a useful technology if coupled with an S-CO2 power generation loop.  But for high efficiency, the power recovery system must be thermodynamic in nature.  That means moving parts like turbines, blowers and generators.  And of course heat exchangers.  So the train would need to carry all of this.  An S-CO2 power generation system would have better P/W than a steam engine.


"Plan and prepare for every possibility, and you will never act. It is nobler to have courage as we stumble into half the things we fear than to analyse every possible obstacle and begin nothing. Great things are achieved by embracing great dangers."

Offline

Like button can go here

#20 2021-11-24 09:30:16

tahanson43206
Moderator
Registered: 2018-04-27
Posts: 19,585

Re: Thermal Energy Storage

For Calliban and kbd512 ... thank you for continuing to vigorously explore this topic!

In post #19, we seem to have encountered a hurdle, but I don't think it needs to be insurmountable.

kbd512 has introduced the concept of molten silicon as an energy carrier.  Calliban has added numerous related facts for integration into a concept that might yield a workable method of converting hot matter into useful work (ie, transportation).

It seems to me that charging a mass of silicon with electric current would be a reasonable way to heat a home, if the flow of heat energy from the store can be controlled. 

I'm thinking here along the lines of comparing an all-electric home with a hot air furnace heated home.

Most homes in the US (and I gather in many other countries) are NOT heated with electric devices in floors and baseboards.

Some homes ** are ** heated with hot fluid that is piped through pipes that pass through concrete floors.  The examples that come to mind first are water heated homes.  I've seen them in Europe, and have watched video showing installation of such systems in the US.

Returning to transportation, because that is where Calliban left off in #18

As a suggestion for how to proceed from here ...

Take 0 Centigrade/Celsius as the lower bound for a temperature scale.  I'm thinking of winter on Earth, or anytime on Mars.

Charge up your silicon energy store to it's liquid phase, just above the phase change temperature.

Compute the degrees of temperature difference.

It is that number that would be available for any heat engine solutions discovered over the centuries.

I'm thinking of the Seebeck effect .... it's been a while since I studied that method of generating current, so all I remember with confidence is that the efficiency of the conversion is directly related to the difference of temperature between hot and cold leads to the junction.

It should be possible to arrive at a reasonable level of confidence for some efficiency for production of electricity.

The rest of the thermal energy might be consumed in useful ways, with a caveat.

What I do NOT recall (and hope someone can clarify) is whether the Seebeck effect requires ALL the thermal energy to flow into the heat sink, in order to achieve whatever efficiency is available.

Where I am headed with this is that it ** might ** turn out that heating water to make steam (in a closed system) might turn out to be more efficient than the Seebeck effect.

The distinct advantage of Seebeck is elimination of moving parts, except for highly efficient electric motors.

Edit at 10:54 local time:

I asked Google about efficiency of Seebeck, and it came up with this (to me surprising) citation:

The new technology, known as the spin Seebeck effect, has conversion efficiency 10 times higher than the conventional method. Thermoelectric conversion technology that converts energy abandoned as waste heat back to electric power could potentially save energy and reduce greenhouse gas emissions.
Apr 25, 2016
New spin Seebeck thermoelectric device with higher conversion ...
www.sciencedaily.com › releases › 2016/04
About Featured Snippets

An efficiency of 10 times greater might mean increase from 3% (which I vaguely recall) to 30%, which would be interesting.

For comparison, I asked Google about Diesel engines:

What is the maximum efficiency an engine can achieve? The diesel engine has a theoretical system efficiency of between 55-60%. For reference, the best power stations operate at 50-55% efficiency, and fuel cells are also around 50%+ efficient – so diesel engines can be incredible efficient.
Oct 30, 2018
Fuel use - how low can you go? - Volvo Construction Equipment
www.volvoce.com › News and events › News and stories
About Featured Snippets

The system that kbd512 has proposed for discussion would NOT produce CO2 and noxious particles as a byproduct.

Edit a bit later:

MIT Physicists Develop 5 Times More Efficient Thermoelectric Material
interestingengineering.com › mit-physicists-develop-5-times-more-efficient...
May 27, 2018 · Thermoelectricity also called the Peltier-Seebeck effect, is a two-way process that consists of the direct conversion of temperature ...

Edit later still: from Wikipedia via Google ... the figure of 5-8% is consistent with what I remember from earlier reading.

Efficiency

The typical efficiency of TEGs is around 5–8%. Older devices used bimetallic junctions and were bulky. More recent devices use highly doped semiconductors made from bismuth telluride (Bi2Te3), lead telluride (PbTe),[21] calcium manganese oxide (Ca2Mn3O8),[22][23] or combinations thereof,[24] depending on temperature. These are solid-state devices and unlike dynamos have no moving parts, with the occasional exception of a fan or pump.

in scanning a variety of citations on the Seebeck effect, I came away with the impression that there is a challenge facing any scientist or engineer working in this area ... the flow of heat and the flow of electricity tend to occur simultaneously in most materials.

The idea solution would have 100% electron flow and 0% heat flow, but so far (from what I've read so far) 5-8% is about what might be achieved in practice.

There was one hint from a laboratory paper that a greater efficiency might be possible, but nothing further showed up from that 2016 paper.

For the purposes of this topic, could a molten silicon container provide sufficient thermal energy to produce sufficient electrical power for a vehicle, using the Seebeck effect?  The waste heat (ie, 92% to 95%) would have to go into a heat sink, and that could be a steam engine, or something along those lines.

The Wikipedia article mentioned harvesting heat from components of automobiles, to deliver a small amount of electrical power.

(th)

Offline

Like button can go here

#21 2021-11-24 09:56:10

SpaceNut
Administrator
From: New Hampshire
Registered: 2004-07-22
Posts: 29,436

Re: Thermal Energy Storage

update for above post on Seebeck effect.
Found out that only 5% of the thermal energy is translated into electrical power.

TPV systems usually attempt to match the optical properties of thermal emission (wavelength, polarization, direction) with the most efficient absorption characteristics of the photovoltaic cell, since unconverted thermal emission is a major source of inefficiency. Most groups focus on gallium antimonide (GaSb) cells. Germanium (Ge) is also suitable.[1] Much research and development concerns methods for controlling the emitter's properties.

This is the triple cell junction design.

Yes thermal can approach the higher efficiency but come from tuning of materials with a greater product mass and expense.

As for the S-CO2 power generation loop its closed loop like an air conditioner or refrigeration system.

Molten silicon storage enough to power city, says MIT

The technology uses two large 10-meter wide graphite tanks, which are heavily insulated and filled with liquid silicon. One tank stores silicon at a temperature of 1926°C. The “cold” tank is connected via a bank of tubes and heating elements to a “hot” tank in which liquid silicon at a temperature of 2370°C is stored.

Excess energy from an adjacent PV system, for example, is used to generate heat, via Joule heating – a process by which an electric current passes through a heating element – to bring up the temperature of the “cold” silicon and move it to the hot tank.

When electricity is needed, the molten white-glowing liquid silicon is pumped through an array of tubes that emit light. The tubes are routed past high-efficiency solar cells, called multi-junction photovoltaics, with the light from the molten silicon then being turned back into electricity. Through that process the silicon cools down and flows back into the “cold” tank, to be used again.

I saw even higher temperatures approaching 5000 ouch

Offline

Like button can go here

#22 2021-11-24 12:06:13

tahanson43206
Moderator
Registered: 2018-04-27
Posts: 19,585

Re: Thermal Energy Storage

For kbd512 and Calliban, with a nod to SpaceNut for that link to the MIT silicon storage power a city link!

May I try to interest you (all three plus anyone not yet engaged) in a practical smaller scale version of the basic idea?

What I have in mind is an energy storage device to deliver electrical power to a device used for city block residential neighborhood maintenance.

There are (most likely) thousands of city blocks identical to the one where I live, in the US, and possibly elsewhere, although the regimented layout of US cities is NOT typical of (much older) European ones.

The need that is coming up is snow removal.  In the current period, the appeal of electric tools is greater than that of gasoline powered ones, even though gasoline powered ones are superior in every respect except (a) noise and (b) CO2 emissions.  And noise can be reduced by expending a small amount of efficiency on the exhaust system baffle structure.

So! Given an interest in electric snow blowers, the consumer has choices of inexpensive wired ones that have plenty of power.  That would be an excellent solution for a home owner in a city block, if the scope of the use of the tool is limited to the immediate lot. A 100 foot cord delivers robust power to all points on the lots in the local neighborhood. 

However, ** I ** would like to be able to clear snow for the entire block, and ** that ** requires a robust tool that can operate independently of a power source.  There are battery powered machines that might do the job, but they are pricey!

Could you (NewMars forum members) design a power cart that would tag along behind an electric power tool for an hour or more?

The power level I'm seeing advertised is 15 amps at 120 volts AC. 

By "tag along" I'm thinking of a modest computer controller able to perform rudimentary navigation and stay within some desired distance of the working tool.

The silicon energy store would be inside a (presumably four wheeled) cart, but it could be as large as needed for the application.

How such a cart would compare in cost to a "traditional" Lithium battery powered system is open to conjecture at this point.

However, the ** longevity ** of the system ** ought ** to be greater.  A near infinite charge cycle capability would seem (to me at least) possible.

Edit: If you want to go for the Gold Ring, make the unit integrated with a Teleoperation capability. so the vehicle operator could direct and supervise the operation of the machine from a wireless connection at home.

(th)

Offline

Like button can go here

#23 2021-11-24 13:17:46

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,889

Re: Thermal Energy Storage

Calliban,

Solar thermophotovoltaics: Progress, challenges, and opportunities

In practice, it is hard to reach the temperature of maximum efficiency as high as 2544 K. As a matter of fact, the conduction and convection of air, as well as the conduction of the sample holder, induces a huge heat loss from the absorber and emitter surfaces, resulting in a low operating temperature. Fortunately, 70% efficiency can be achieved at a moderate temperature of 1000 K, which is nearly twice of the Shockley-Queisser limit. Despite the high temperature requirements for the absorber and emitter, the PV cell needs to be kept cool to maintain a high performance. When the temperature rises, both the efficiency and power of the cell decrease (0.4%–0.5%/K for silicon solar cells).5 Besides, various optically qualified materials are prone to oxidization at high-temperature, leading to the necessity for a vacuum or inertia environment.

The Status of Thermophotovoltaic Energy Conversion Technology at Lockheed Martin Corp.

High-efficiency thermophotovoltaic energy conversion enabled by a metamaterial selective emitter

Lockheed achieved higher than 10% efficiency more than 15 years ago.  Achievable TPV efficiencies have increased dramatically since then.

Thermophotovoltaics: a potential pathway to high efficiency concentrated solar power

Someone has already quadrupled that 10% efficiency figure, over a wide temperature range:

Thermophotovoltaic Efficiency of 40%

Here we report TPV efficiency measurements of > 40%, measured directly through calorimetry. This is a record high TPV efficiency and the first experimental demonstration of high-bandgap tandem TPV cells. This is also the first demonstration of a solid-state heat engine (terrestrial heat source) with efficiency higher than the average heat engine efficiency in the U.S., which is < 35% based on primary energy inputs and electricity output. An efficiency of 40% is also higher than most steam cycles, and is in the same range as simple cycle gas turbines. Thus, 40% represents a  major  step  forward  (see  Fig.  1b), as this is the first time another type of heat engine has the potential to compete with turbines by exhibiting comparable efficiency and potentially even lower CPP e.g., < $0.25/W.

Offline

Like button can go here

#24 2021-11-24 14:04:14

SpaceNut
Administrator
From: New Hampshire
Registered: 2004-07-22
Posts: 29,436

Re: Thermal Energy Storage

The molten battery would be to large and heavy for the portable application since it requires such a heavy tank and layers of isolation from that heat. Then add in the power generation mass.

An application for such can be done with a couple of items to give a target for design. Maybe the cart part would change as well as the power pack but basically this is the desired outcome.

61o-i-kiqHL.__AC_SX300_SY300_QL70_FMwebp_.jpg

71dLPsScCjS._AC_SX425_.jpg

Offline

Like button can go here

#25 2021-11-30 10:01:46

tahanson43206
Moderator
Registered: 2018-04-27
Posts: 19,585

Re: Thermal Energy Storage

For SpaceNut re #24

Your post gave me pause ..... I'm back with a couple of observations and a request of kbd512 ...

The size of a molten silicon storage device is not at all clear.   Is there a reason why it has to be larger than the devices you showed in post #24?

I can't see a reason, but perhaps you can.

Observation 1: Vacuum is an excellent insulator.

Observation 2: Radiation from a hot body is described as "black body" radiation ....

Speculation: a mass of molten silicon in a container capable of holding it would produce black body radiation at a steadily decreasing rate.

If the mass were held in a cavity able to hold vacuum, and if the inside walls of the cavity were lined with photoelectric receptors tuned to a wide range of spectrum (multiple layers?) then I would ** assume ** the thermal energy of the mass, converted to photons, would yield a current flow.

How large that current flow might be is something that a member might be able to estimate.

Unknowns that I'm aware of:

1) Material to hold molten silicon
2) Method of supporting container of molten silicon inside vacuum cavity
3) Selection of photoelectric receptors to cover black body radiation (as much as possible)
4) Method of managing heat generated in the wall of the container due to failed reception

I'm wondering if a quart of molten silicon might yield a 15 AMP flow at 120 VAC for an hour.

That would be a consumer product, and there is a very large market for such a device.

The silicon mass would be heated by simple resistance heating from a utility power source.

This concept would (presumably) be infinitely re-usable, with a lifetime for components measured in decades.

(th)

Offline

Like button can go here

Board footer

Powered by FluxBB