London’s Red Double-Deckers Go Greener
New System for Efficient Cooling of Hybrid Drivetrains

A Compact Brayton-Cycle Engine and Biomass Process for Mixed-Alcohol Fuels

The StarRotor compact Brayton cycle engine.

A Texas A&M chemical engineering professor has developed a process to convert biomass to a mixed alcohol fuel that contains more energy than fuel ethanol. He has also developed a compact Brayton-cycle engine (the same thermodynamic cycle employed by jet engines) capable of being powered by any type of fuel—including his MixAlco mixed alcohol fuel.

Prof. Mark Holtzapple projects that his StarRotor engine, which is being developed by a company of the same name, could deliver efficiencies of 49–55% applied in a passenger car—about 2.5 to 3 times more efficient than a conventioanl gasoline engine.

The StarRotor engine. In the classic Brayton-cycle engine, ambient air is pressurized in a compressor, passed to a mixing chamber where fuel is added, and then ignited in an expansion chamber. It then expands through a piston/cylinder.

The Brayton cycle.

As applied to gas turbines, the Brayton engine has a compressor, a burner and an expansion turbine. Ambient air is compressed and passed through a heat exchanger for pre-heating. The pre-heated charge goes to a combustor where fuel is ignited, and the hot compressed air then flows to an expander where the thermal energy is converted to shaft work. The hot exhaust gases from the expander are sent to the heat exchanger where they are cooled and then discharged.

Brayton cycle engines have a high power density (hence their use in jet aircraft), compared to the lower power density of Otto (spark ignition) and Diesel engines.

The major challenge in implementing Brayton cycle engines, according to an analysis done for the Defense Advanced Research Projects Agency by Holtzapple, is to find a means to process large volumes of air to achieve a desired power output.

Traditionally, this is accomplished using dynamic (i.e., axial or centrifugal) compressors and expanders. The devices, however, require very high speeds—e.g., 100,000 rpm for a 30kW unit—to develop the desired pressure and flow. They also operate efficiently at only one speed, and are affected by changes in air density.

The patented StarRotor Brayton cycle engine uses gerotors for both the compressor and expander. (A gerotor is a positive displacement pump mechanism that delivers a known, predetermined quantity of fluid in proportion to speed.)

The StarRotor compressor has an inner gerotor with n teeth and an outer gerotor with n + 1 teeth. As the gerotors rotate, the void that opens draws air in through the inlet port. As the rotation continues, the void closes and compresses the air. When the air is compressed enough, the compressed air exhausts through the outlet port.

Because the void opens n + 1 times per revolution of the outer gerotor, the gerotor compressor is able to process enormous volumes of gas in a very compact size. The expander operates similarly to the compressor, except in reverse.

The StarRotor applied in a vehicle could yield efficiencies of 49 to 55% and fuel economy of 75 to 100 mpg, according to Holtzapple.

The gerotor teeth must be dry—lubricants are not compatible with the high temperatures. To prevent wear and friction, there must be no physical contact between the teeth of the inner and outer gerotors. StarRotor employs an inexpensive surface treatment to minimize gas leakage through the small gap, and an external synchronization mechanism ensures proper motion of the inner and outer gerotors.

The StarRotor, according to Holtzapple, can offer power ranges from 50W to 50,000kW. Lower-power versions employ a single stage that compresses air from 1 to 6 atm. The medium-power engines employ a second stage that compresses air from 6 to 36 atm. The high-power engines employ a third stage that compresses air from 36 to 216 atm. The power density is improved by using small-diameter rotors that rotate rapidly.

Energy Content of Fuels
Fuel MJ/L Btu/gallon
Gasoline 34.9 125,000
MixAlco Blend 1 29.0 104,000
MixAlco Blend 2 26.5 95,000
Ethanol 23.4 84,300

MixAlco. The MixAlco process converts biomass into organic chemicals and alcohols with a multi-stage process that includes lime pretreatment, non-sterile acidogenic digestion, product concentration, thermal conversion to ketones and their subsequent hydrogenation to create mixed alcohol end products.

We can use anything that biodegrades. If you put it outside and it rots, we can use it. So we can use trees, grass, manure, sewage sludge or garbage.

—Mark Holtzapple (The Eagle)

The MixAlco process consumes about 90% of the raw material substrate, and the process recycles all of its water and primary reagents. It can be tuned to produce the chemicals most in demand at a given time.

MixAlco Chemical flowchart. Click to enlarge.



Rafael Seidl

The engine is like a jet engine in that it relies on continous isobaric combustion using compressed air delivered by a turboshaft. The only - though significant - difference is the gerotor mechanism of the compressor and turbine. This brings down the size at which the design is competitive, but you'd still need 10,000 RPM or more to get adequate specific power for a vehicle engine -> noisy.

In this setup, the turbine is continously exposed to high temperatures. In large jet engines, some of the compressed air is forced through the shaft and turbine vanes to create a shroud of relatively cool air such that the combustion can occur at very high temperatures for better thermodynamic efficiency.

Small (radial) turbines cannot be cooled this way due to manufacturing limitations. It's unclear if the inner and outer parts of a gerotor turbine design rated at e.g. 50kW could protected by air shrouds. If not, maximum process temperatures are limited by the turbine material and thermodynamic efficiency will be rather less than advertised above.

The other issue is that the outer part of the turbine gerotor is a hot drum spinning at high RPM. The entails non-trivial air friction and radiation losses.


The fuel process may be more interesting. A 90% conversion rate for arbitrary biomass would be great. Higher energy density (i.e. longer hydrocarbon chains) are very desireable for increased range and the ability to blend in higher fractions into regular gasoline.

John W.

This engine is super. I mentioned it a month or two ago on another post. I'm sure the outer casing can be cooled with both external vanes and efficient internal liquid cooling. The insides would be more challenging, but I'm sure any or most of the possible difficulties the previous post mentioned could be overcome. High temp, high rpm bearings are already in use.

It's still relatively new: with time it will be improved further & refined. Hard to believe how much they have refined the "primitive" 4-stroke gas engine. This engine, once they put it all together (see their website), is mechanical poetry in motion.

I'd love to see a high efficiency baby version of this engine, hooked up to a small generator for a backup for long range driving on an electric car: a series hybrid. (Or a small rand-cam engine if not this). Add Raser's motor technology and a good battery pack, and a plug-in can handle long drives very efficiently as well.

I'm telling you, in a few years when all these things are available to us it will be the home tinkerer's paradise. Expect to see lots of start-up companies. It will only be a question of cost, but that will go down with time as well, especially with all the competition in our increasingly energy-starved economy. This news makes me just as giddy as when I heard Microsoft is going to release Halo 2 for PC! Who-hoah! I think the warthog runs on this system, actually. Wart wart wart.


Kinda reminds me of a gang of MGB oil pumps plumbed together.

Barry R. Guthrie

This engine would not work for transient operation applications. It might work for standard gas turbine related applications.

The noise would be ridiculously high from both the combustor and the StarRotor compressor and expander. Anyone who has worked with superchargers will tell you just how noisy they are. With the star tooth design here it could be really bad and separating them into a compressor and expander will generate twice the ear shattering clangs. Especially at 100,000 rpm.

Rafael Seidl

Barry -

the 100.000 RPM number referred to radial turbines such as are used in turbochargers. The gerotors proposed here would deliver similar power at 1/10 that speed but the components feature larger moments of inertia. Torque flux is moment of inertia (constant) times angular acceleration, i.e. transient response might be no better than a single-stage radial compressor-turbine combo.

On a related note: in a car, torque demand rises with speed, which is coupled to engine RPM via the transmission (incl. serial hybrids). A single-stage Brayton engine also feature a torque curve that rises with RPM. This is a recipe for poor RPM stability, even instability in certain scenarios. What you want is an engine with a negative torque-RPM slope for a given fuel flow. This can be achieved by adding a load turbine, at significant cost and reduced fuel efficiency.



I think Barry cuts to one of the main points. The brayton cycle tends not to do too well at variable speeds. It's kind of meant to start up, and stay up. Most turbines run this way. I can see how a very small one could be used in conjunction with a hybrid to constantly recharge the battery or power the electric motor, but It still seems like a significant technical challenge. Others have conjectured about using turbines before in automobiles, without much sucess. Eh. who knows, I wish 'em luck. Maybe they can pull it off!



The MixAlco link took too long to download on my slow internet service. However I'd bet the hydrogenation process reduces EROEI despite the high conversion of raw material. There must be a trade-off between the cost of hydrogen and the convenience of fuels that can be used in existing engines.


One point about value of hydrogen is perhaps the reduced wear on mechanical parts and subsequent design changes.
It seems some type of constant velocity/variable speed torque converter would address the high speed issue, though probably at some loss of efficiency.
Anyone recall Andy Granitelli's turbine car at the Indy 500?

Barry R. Guthrie

The StarRotor compressor and expander looks to operate pretty nice, even at lower rpm <=10,000. However, a big problem with this design is not only the noise, but thermal expansion of the housing and inner star rotor. This will result in thermal deformations that could result in wear, damage, and further increase the noise levels. This will also result in sealing losses on the axial rotor sides between the rotor and housing that will result in high pressure losses. Even with this thermal expansion you will still get some air flow that is under pressure. However, the pressure would be unstable and unpredictably degrading with use, resulting in A/F ratio control problems.

Roger Pham

49-55% efficiency is pure fantasy in such a small-sized quasi gas turbine. Even the very large GE power generating gas turbines with all kinds of efficiency enhancement and air-cooled shaft and turbine blades with very high exhaust inlet temperature don't even deliver that kind of efficiency. This kind of far-fetched claim tends to reduce greatly the credibility of the project. If the Gerotor engine won't tear itself apart or melt down during test run, expect no more than ~30% thermal efficiency.


I'd still like to know how an inexpensive surface treatment is suppose to prevent pressurized gases from leaking past the "non-contacting" gerotors. Sealing issues were the bane of the Wankel's development and the challenge of balancing friction with losses continues to challenge all engine designers to this day.

Robert Schwartz

IIRC efficency of heat engines is limited by Carnot's Law. I am under the impression that big electric plants using coal fired steam turbines have pushed that about as far as they can and are less than 50% efficent.

Adrian Akau

I don't care how efficient this motor would be. If a rotar became free at 100,000 rpm, it would slice through several inches of steel like butter. Just ask the air force which has had experience with high speed gyros.


Even if this ever becomes workable, expect a big company to come in and buy up all the patents and use them as they see fit. I think the better news of this is the MixAlco. Could mean the end of landfills if it could be ported to other type engines. BFI might be paying you for trash pick up if this is workable.


Countless attempts to invert compressor to work as combustion engine invariably failed, Wankel rotary being only one limited exception. This one does not seem to address any old problem of such designs. And 10 000 RPM engine for family sedan?!

P.S. John W.: I am fascinated with Razer too.

Roger Arnold

Most of the comments about the StarRotor engine strike me as valid. However, the real significance of this work has nothing to do with the engine. That seems to have been thrown in, perhaps to add some glamor.

The real significance, IMO, is in the work to develop and characterize the MixAlco process. That process looks very promising. The fuel doesn't need a new type of engine; modifications to run a conventional engine on it should be less than required to run on E85. And what this study shows is that it really is feasible to replace imported oil with bio-fuels.

As to the hydrogenation step reducing the ERoEI, that depends on how one does the accounting. If you include as "energy in" all the sunlight that fell on the ground from which the biomass grew, then obviously any biofuel is going to come out with a ridiculously small ERoEI. In the MixAlco process, more energy comes from sources other than sunlight on the growing biomass. Some of it might come from electricity to produce hydrogen. Yet the total energy input is probably less, simply because photocells can convert sunlight to energy much more efficiently than photosynthetic plants. Hence, the total land area required to support a given level of fuel production would be less.

allen zheng

"And 10 000 RPM engine for family sedan?!"
With the reliability of a Formula-1 car, 300 miles and.... KABOOM!!
Roger Arnold:
Hell yea, now to find a cheap, high quality source of lime that uses little energy as possible, and is eco-conscious.
Adrian Akau:
Another analogy would be centrifuges used in concentration of substances (liquid, gases) due to different densities /atomic masses. The Iranians have found out spinning those tubes at ~50,000 rpm ain't so easy. Minor imperfections, and metal goes flying everywhere at high subsonic speeds.

john galt

What about STEAM? Multi fuel capability, cleaner burn of an ECE versus an ICE. Has anyone heard of current research or prototypes?

Dan Dubya

My interest in any small engine is to act as a generator which continually recharges a bank of batteries in an electrically-driven vehicle. The batteries are the primary power source for the electric motor, capable of variable output based on demand. Is it possible to hook up a generator that provides some fixed power level to recharge the batteries? The battery bank essentially acts as a capacitor for those times when demand is higher than the gas/electric generator is putting out. If the batteries are fully charged (as during extended-idle traffic jams in a crowded urban environment), the generator is idled or shut down until charging is required again.

If you can wire/control this correctly, then having an engine with poor transient response is no longer a problem, because the motor runs at its peak efficiency all the time, and the transients are handled by the batteries. High power is only required of an automobile when accelerating. Cruise doesn't require much power at all (provided the aerodynamic traits of the vehicle are reasonable), so it wouldn't take much of a generator to meet the RMS power demand.

BTW, I agree the most intriguing aspect of this article is the high-power-density biomass-derived fuel. Let's turn our landfills into an energy source! Obviously they wouldn't provide anywhere near all of our energy demand, but they would be doing more good than just sitting there looking ugly. Any way to turn waste into energy is cool. ;-)

Jack Bayless

Search where my engine is #1 Biomass Engine and on search Biomass Engine and it is #3.

jack bayless

Well my own comment knocked my engine off the #1 spot on yahoo. Well it is still #2 which is real good. No big company wants to develop my engine because there is no money in it for them. People can collect their own biomass and won't have to pay gasoline tax. Ethanol is the big thing, but can you imagine there will have to be an ethanol plant in every county in the country to make enough ethanol. Total biomass is much more efficient, probably 10 times as much. Some people may be concerned that my Biomass Engine would not fire evenly. But the one hot combustion chamber could be designed to even that out. And using fine dust solids is done all the time in other industries. The patent link on my web site does not work due to a change at the patent office. However, the patent office search engine comes up and you can click on patent number and enter my patent 4949688 which is lapsed and free to anybody. My engine is perfectly balanced both in rotation and in intertia. So if you are a machinist or know one, build it and test it. We have to do something to stop the high price of gasoline and diesel. My email is on my web site.


JHello this webbsite is cool



Steam is like any other heat engine. Carnot tells you what the efficiency is. Whether ICE, brayton, rankine, stirling or any other heat engine, the efficiency is based on the cold to hot difference. Fuel cells are not heat engines and go from chemical to electrical directly. They would be more efficient than steam.

Sean Holt

Hi All:

The starRoter is an interesting design. However, it's
need for relatively stable, high RPM and fuel
requirement greatly limit it's usefulness.

A far superior engine in every way that is measurable
is the Quasiturbine zero vibration rotary engine,
developed by Dr. SaintHilaire of Montreal Canada and
patented in 1996. This engine has the capability to use
detonation OR combustion of the A/F
mixture. This means that it is the only IC engine that
can effectively use hydrogen fuel. Any rpm is fine so
long as it's turning. It developes 50% of it's torque
in only 10 deg of revolution. This means no flywheel
and no transmission required. The engine is symetrical
so it can run in reverse with additional porting. Just
about any liquid or gas that will go bang can be used
as fuel.
The quastiturbine can replace both the piston and
turbine IC engines in every application where they are
presently used from the very small to the very large
and be used as a pump as well.

Additionally, The U.S. Air Force is very interested in
the Nutating Disk engine, patented by Mr. L. Meyer in
1993. This engine has only one moving part and
apparently is self- supercharging. The preferred fuel is
said to be heavy diesel. The engine is also size
scalable and has variable RPM although to a lesser
degree then the piston or quasiturbine IC engines.

For more info go to and/or
"google" the key words.




Posted by: Cosmo | Apr 28, 2006 1:03:48 PM

The MixAlco link took too long to download on my slow internet service. However I'd bet the hydrogenation process reduces EROEI despite the high conversion of raw material. There must be a trade-off between the cost of hydrogen and the convenience of fuels that can be used in existing engines.

Cosmo, if you get your H2 from gasification of the undigested product, then the EROEI (defined as Energy return as liquid fuel product on energy invested from fossil fuels, thus ignoring any extra electricity produced) is actually INCREASED, because you are putting some of the energy of the undigested product into your liquid fuel product, which would otherwise be simply used to produce electricity, which more than likely would be in excess anyway and you would need to sell it to the grid.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)