=chemistry =energy =startups
In 2013, I had an idea for making
biofuels.
In 2015, I made this site.
In 2016, I
posted that biofuel
idea here.
In 2018, I found the right search terms to find a
similar process. The MixAlco process was
developed in 1995 by Prof
Mark Holtzapple. He continued working on it for decades.
1995:
Terrabon was formed to commercialize technologies including MixAlco.
2007: Terrabon started trying to deploy its research.
2017: Terrabon
went bankrupt; its MixAlco technology was purchased by Earth Energy
Renewables (EER).
2021: EER went bankrupt and was purchased by Ara
Partners.
The MixAlco process was not quite the same as what I proposed. That used naturally occuring mixed cultures, which produce mainly acetate, and calcium hydroxide because using CaCO3 gave conditions too acidic for good growth. I proposed using engineered microbes that produce mainly butyrate and tolerate lower pH. That would allow using (cheaper) CaCO3, make separation slightly easier, and give (maybe) more valuable products.
I emailed Prof Holtzapple in 2018 and we talked a bit. He said:
EER has changed direction and now extracts the acids from the fermentation
broth, for the following reasons:
- Acids are very valuable as a chemical
product
- Calcium carbonate is a poor buffer
- Pyrolysis loses about
10% of the product
- High-temperature salts are difficult to handle
-
The ketones are contaminated with sulfur and nitrogen, which interfere with
downstream catalysts
- The acids are easily transported to a centralized
facility for further processing
I argued that it's better to use
engineered strains and CaCO3, and his position was that the robustness of
natural mixed cultures is more valuable. In retrospect, I was right about
that, but the
basic concept wasn't economical.
I also said it would be better to
replace some rubber in vehicles and use ketones directly as high-octane
fuels than to try to convert them to hydrocarbons...but in the short term,
what people wanted to fund was small amounts of drop-in replacement
renewable fuels for PR purposes.
Fermentation of cellulosic
biomass to carboxylic acids which are then separated and used as an energy
source is what cows do. Cows are able to digest many grasses in 1-3
days, but fermenting chopped grass in a tank takes much longer, perhaps 20
days. Also, the viscosity of water with cellulosic biomass increases
exponentially with concentration, so concentrations are generally 10-20% if
pumping is required. This means very large tanks are needed.
If those
large tanks aren't stirred, then the biomass will settle at the bottom and
form an un-pumpable sludge. But such large tanks with active stirring would
be expensive.
Calcium butyrate is less soluble in water than calcium
acetate or propionate, but it's still pretty soluble. To precipitate any of
those from
the fermentation broth, almost all of the water would need to be removed. Water
desalination is quite cheap, but this water would have more stuff in it, and
more of the water would need to be removed than the ~50% typical for
desalination. Membranes would get fouled, and
multiple-effect distillation would probably be more than 2x as expensive as
for desalination. (Cows have stomachs with large surface areas that use
proteins to actively transport
stuff. Doing the same thing artificially isn't feasible.)
Let's suppose calcium
butyrate concentration is 3% and water removal is $2/m^3, similar to the
cost of thermal desalination. The cost of just the water removal would then
be ~$150 per ton of oil equivalent. That's very optimistic - $18/m^3 for
filtering solids and water evaporation would be more realistic, which by
itself is slightly too expensive for competitive fuels. Then there's the
cost of the biomass used, the tanks, and processing the precipitated salts.
So, a half-assed idea I had 10 years ago and only spent a few hours on was on the same level as something that:
- was
developed by a smarter-than-average professor 20 years earlier, after years
of research
- got patented and used as a basis for startups that raised
millions of dollars
- worked but wasn't economically viable
That's simultaneously not good
enough and more than the people with influence consider credible from me. (I
seem to have made more progress on the 1st problem than the 2nd, which is
probably the opposite of what you should do.) This isn't a cherry-picked
example, it's pretty average.
As I've
said here before, the biomass conversion process I currently like best is
thermochemical conversion to hydrochar, levulinic acid, and furfural.
(GFBiochemicals is the company doing the most similar thing to that, but I
think large improvements vs their process are possible. They sell levulinate
esters as premium solvents.) That has several advantages over MixAlco - for
example, it's much faster, and there are no microbes that need protection
from acids. But for many countries, there simply isn't enough land to grow
enough biomass to use that for most energy usage, and recent
subsidies have been aimed more at
hydrogen and water electrolysis.