=chemistry =renewable energy =polymers
What chemicals can be made from biomass at competitive prices? Are biofuels economically viable?
sugar
The chemical formula for glucose
is C6H12O6. If you remove 2 H2O and 2 CO2 from it, you get C4H8, which has
31% as much mass.
Without subsidies, sugarcane is the cheapest source
of sugar. Prices for it vary, but let's say it's $300/ton. At that price,
oil from sugar can't possibly be less than $960/ton, and conversion is never
100% efficient.
At 2.8 kilograms in a gallon for gasoline, that's an
absolute minimum of $2.73/gallon of gasoline equivalent from sugar - before
transportation and taxes - and considering losses and processing costs,
$4.60 would be more realistic. Using unsubsidized corn, that would
significantly higher.
By the way, burning $960/ton oil in a 60%
efficient combined-cycle gas turbine would cost at least $0.12/kWh, which is
about 3x as expensive as wholesale electricity in the USA.
So you
see, while sugar is cheap by mass, it's not cheap enough by energy content.
So, if you're making chemicals from sugar or starch, you want to make more
valuable things. This isn't exactly a problem, unless you're trying to
substantially reduce global warming, in which case fuel is the only product
with a large enough volume.
Making pesticides, flavorings, and
certain polymer precursors such as butanediol by fermentation is very much
viable. Yields of anything that's not the main product of metabolism tend to
be low, and lots of things are toxic to cells at high concentrations, and
it's usually hard to separate dilute chemicals from fermentation broth, but
despite all that, production of high-value chemicals by fermentation will
continue to increase.
grass
Grass is much cheaper than sugar.
The cost of miscanthus or switchgrass has been estimated at $65 per dry ton,
but as with anything so cheap, that cost varies significantly with distance.
Most papers on miscanthus as a source of biomass have focused on
Miscanthus x giganteus. This produces much more mass per area than
switchgrass, but it doesn't have seeds so it's more expensive to plant, and
it requires more fertilizer. Energy cane (a sugar cane variant bred for
biomass) produces even more, but needs too much fertilizer.
I
actually think it's better to focus on breeding seeded miscanthus, maybe
lutarioriparius. Depending on your usage, you might want to maximize
cellulose, hemicellulose, or lignin content, and those proportions vary
widely between miscanthus and switchgrass species.
biorefineries for fuel
Suppose you feed grass to a
biorefinery, converting cellulose to levulinic acid and hemicellulose to
furfural. You might get 1/4 of the input dry mass out as those products, and
they have about 1/2 the energy per mass of oil, which means about $520 per
ton oil equivalent just in feedstock. Depending on how processing is done,
the net cost per ton oil equivalent could be anywhere from $700 to $1800,
not including processing to gasoline.
I think it's possible to
profitably produce levulinic acid and furfural from grass at an average
selling price of $500/ton. This estimate is much lower than current prices,
and is based on a novel biorefinery design, but I have fairly high
confidence in it. Production from sugarcane bagasse would be cheaper, but
supply is limited by demand for sugar.
It's possible to produce
relatively good gasolines from those products by, for example, hydrogenating
furfural to 2-methylfuran. The cost of that obviously depends partly on the
cost of hydrogen, and some fuel products involve using methanol too. Also,
the economics of combined production of fuel and polymers is better.
If producing only fuel, and using syngas from natural gas, then I'm
estimating that the cost of fuel using this approach is slightly less than
$3 per gallon of gasoline equivalent, which is slightly more than the
current cost of producing gasoline from only natural gas. That's also
similar to the cost of biodiesel when the seed meal is sold as animal feed,
but biodiesel production requires more land - maybe twice as much land, but
this depends on the plant species - and the demand for seed meal would be
saturated if biodiesel replaced oil.
Obviously, ethanol from corn is
only used in the USA because it's subsidized, and biodiesel is only used in
Europe because it's subsidized. And that's...actually, you know what, that's
fine - as the first step. Having subsidized something enough to get an
industry established, all that's left is the second step: ramping down
subsidies to zero to force companies to be economically competitive. China
has the system down, just follow their example.
Using biogas on a
large scale, it would be more, but less than $4. That's far from
civilization-ending, but it is expensive, and doesn't make a very good sales
pitch to investors. Still, I guess it looks a lot better economically than
most of the stuff Breakthrough Energy Ventures is investing in; that whole
group is weird and doesn't make sense in several ways, but I can't really
talk about that.
Production of furfural and
levulinic acid produces hydrochar as a byproduct, which can be buried for
carbon sequestration. If you're only making fuel, the required CO2
mitigation cost including that and avoided oil production would be maybe
$60/ton. Not terrible; full mitigation of US CO2 emissions at that price
would be about 10% of US federal tax revenue, which is politically
questionable but economically very possible. But again, this depends on a
theoretical biorefinery design with several novel elements.
In
theory, you could gasify hydrochar to produce the hydrogen for biofuel
production; it would be about as good for that as lignite coal, but it's
much better to make syngas from methane than from coal, which is part of why
coal is worth approximately nothing these days. It's also, of course,
possible to make hydrogen by electrolysis, which is far from viability as
long as natural gas is available but could potentially end up being
competitive with biogas.
cellulose fermentation
Most US research on fuel from
biomass has focused on cellulosic ethanol. Nothing even close to practical
has resulted from this extensive reseach. The concept is fundamentally
flawed because ethanol disrupts the cellulose-digesting enzymes, and
breaking cellulose into sugars before fermentation is too expensive.
But hey, it's a lot less dumb as an idea than the hydrogen fuel proposals or
coal power with CO2 sequestration.
We know that producing net
exergy by grinding up
grass and fermenting it to butyrate is possible, because cows do that.
Here is a
process I designed using that approach. The resulting dipropyl ketone is
almost ideal as a biofuel:
- good energy
density
- good octane ratings
- low toxicity
- not too strong a
solvent
- somewhat biodegradable
- suitable boiling point
- low
enough melting point
That involves fermenting solids, separating out solid product, and processing it. It's always more expensive to handle solids than liquids, so that process might be too expensive in practice; its costs vary significantly with process design details. But it's plausibly viable, and I think it's still the most promising approach for economically competitive biofuel production today.
biorefineries for polymers
If you want to get investment for
biorefinery construction, and at least get some things going, it's better to
start with something with less marginal economics than fuel production, and
expand from there.
Here's an example of a more profitable process:
furfural ->
methyl furoate ->
dimethyl 2,2’-bifuran-5,5’-dicarboxylate
-> dimethyl biphenyl-4,4'-dicarboxylate -> polyester
With 70%
ethylene glycol / 30% butanediol the result is a thermoplastic with >150 MPa
flexural strength and >260 MPa tensile strength. It can also be used for
self-reinforced composites, and as a matrix for thermoplastic prepreg. This
polymer could potentially be about as cheap as PET, and in some ways, it's a
higher-performance polymer than anything commercially available now, so
production is viable on a small scale even if it's 10x that expensive.
Of course, it's always possible that there's some problem, so I designed
a couple other routes from furfural to polymers, as one does. Anyway, while
renewable plastics are currently expensive, and some questionable approaches
to them are being pursued, they shouldn't be a big long-term problem.