CO2 Stripper Postmortem Thoughts

Diffractor30 Nov 2019 21:20 UTC
113 points
37 comments8 min readLW link

[EDIT: A crucial consideration was pointed out in the comments. For all the designs I’ve looked at, it’s cheaper to just get a heat exchanger and ventilation fans, and blow the air outside/​pull it inside and eat the extra heating costs/​throw on an extra layer of clothing, than it is to buy a CO2 stripper. There’s still an application niche for poorly ventilated rooms without windows, but that describes a lot fewer occasions than my previous dreams of commercial use.]


So, I have finally completed building a CO2 stripper that removes CO2 from the air to (hopefully) improve cognition in environments with high CO2 levels. In California, the weather is pretty good so it’s easy to just crack a window at any point during the year, but other areas get quite cold during the winter or quite warm during summer and it’s infeasible to open a window unless you want to spend an awful lot of money on heating or cooling bills. It didn’t work quite as well as the math indicated at first, but the whole thing is built, and basically functional. The rest of this post will be a reflection on the lessons learned while doing so.


1: In hardware, ideas are cheap, execution is expensive

So, the fundamental idea is extremely simple once you have some basic knowledge of chemistry. The goal is to get CO2 into some form that isn’t the gas form, via some sort of chemical reaction.

Submarines and CO2 capture from flue gas use a reversible reaction with ethanolamines, where they absorb CO2 at high temperatures and release it at low temperatures. Reversible reactions are good for making waste, but heating up and cooling down large quantities of liquid takes an awful lot of energy. Submarines have nuclear reactors onboard, and flue gas is hot, but we don’t necessarily have the energy required. Also ethanolamines are toxic and hard to get a hold of for a civilian and really stinky, being the major component of “submarine smell”.

Adsorption onto zeolites is also plausible, but the issue is that it requires alternately exposing the zeolites to high air pressure and low air pressure, and high airflow is required. The combination of high pressure and high airflow means that again, you’re using a lot of energy. The basic math is as follows: One human produces about 1 kg of CO2 in 24 hours. We can idealize a perfect CO2 stripper as a magic box that inhales air and spits it out at 0 ppm. If you want a steady-state concentration of 500 ppm for 2 people, then we can see how much air-flow is required to lock up 2 kg of CO2 in 24 hours. This comes out to about 100 cubic feet per minute. This is the bare minimum air flow for any CO2 stripper, but in this particular case, it corresponds to a 25 horsepower air compressor, which is 18 kilowatts. This is equivalent to running 5 electric dryers at once. So that one is out too, especially since we were assuming 100% efficiency at eliminating CO2.

What about irreversible reactions? Just lock the CO2 up as a solid waste? Well, to begin with, this is going to produce quite a waste stream, and consume quite a bit of chemicals, you’d better hope it’s safe and that the feed chemical is cheap. The reaction used on space missions used lithium hydroxide. The basic idea is that lithium hydroxide makes a very basic solution. Carbon dioxide is slightly acidic, so it dissolves very fast into basic solutions. Then you get precipitation of lithium carbonate which is safe.

The problem is that lithium hydroxide is quite expensive. It was used on space missions because it’s the most mass-efficient way of doing that sort of reaction and every gram counts in space missions, but we want the cheapest way of doing that reaction.

And then we hit upon the perfect solution. Calcium hydroxide. It’s an extremely cheap bulk chemical, 15 bucks for a 50-pound sack of it at a hardware store. It’s fairly mild as far as hydroxides go, being pH 12.4. So instead of giving you horrible chemical burns, it’s safe to handle unless you’re exposed to it for over an hour at a time without washing it off. It’s the alkaline analogue of the difference between 1 M hydrochloric acid, and lemon juice. And when it reacts with CO2, it makes CaCO3, aka limestone, which is totally harmless. In fact, it’s a common laboratory demonstration that breathing onto a solution of this stuff produces a white film/​crust on the top, which is the CO2 in the breath locked up as solid limestone. It’s the obvious choice if you’re trying to remove CO2 via chemical means.

And in fact, in the SSC comment section, someone else independently had the exact same idea! Just lock up CO2 with calcium hydroxide!

The simplicity of an idea in the field of atoms instead of bits doesn’t necessarily mean that anyone on earth has ever done it before, though, or will ever do it, and I’m not worried about anyone scooping the idea, because building novel hardware is hard enough to provide a natural barrier to entry unless it’s a large company that’s interested in the idea. Ideas are cheap, execution is expensive, in both time and money.

2: Only polymaths need apply

If you’re trying to build a novel machine in your garage, and aren’t working as part of an engineering team, you will either need an improbably wide range of knowledge, or the general ability to pick up whatever you need to learn. There’s the basic knowledge of chemistry to spot that this is the obvious reaction to go for, but the full design requires:

Familiarity with wastewater aerators to know what to buy to prevent clogging with solids, knowledge on which materials won’t react with your chemicals, the math of air flow in pipes, the ability to read fan pressure/​airflow curves, the ability to go from “I want a circuit that does this” to building a novel electronic circuit on a breadboard without frying anything important, enough programming knowledge to write some basic arduino code, familiarity with hazardous waste disposal regulations in your state, familiarity with waste dewatering techniques, familiarity with which sort of pumps can pump sludge instead of pure water, some electrical engineering knowledge to work safely with 220V power without frying yourself or anyone else, knowledge of soundproofing, and especially the familiarity with everything at Home Depot that lets you home in on the most efficient and foolproof way of building a thing that does what you want. Probably some other stuff too that I consider obvious but others might not.

Now, most of this is pretty easy to pick up given enough starting mental firepower, and the sense of what to google for. Or just having lots of experience with building material things.

Having one of the relevant fields of knowledge manifests itself as knowing ahead of time which approaches will work and which will fail and what solutions past work in the area has already found.

For some of these, missing it will manifest as not knowing that there’s an incoming bullet in a particular area, like not knowing that fine bubble aerators will promptly clog if there’s lots of particulates in the water, or not suspecting that high air flow rates are incompatible with small pipe (I knew the latter one and it still almost got me until I idly decided to work out airflow velocity in the pipe and realized it was around 200 mph)

3: The planning fallacy is huge here.

So, it wound up costing a lot more than I thought and taking a lot longer than I thought. The mechanism of why the planning fallacy hits so hard here is tied in with the design process. What happens is that you start out with a sketchy outline of all the component parts (like, “I need something that automatically dispenses chemical powder”), and as it becomes time to build a part, you drill down further and further in fleshing out the details until eventually you’ve drilled down far enough for your design to Actually Work in reality. While you do this, you will inevitably come across parts that are a lot harder to do than you expected, which you were glossing over on the first pass. The shiny black box of “build a chemical dispenser” looks more tractable than “how the fuck do I build a motor mounting plate with my inadequate tools”, which you didn’t initially suspect you had to do because you weren’t thinking at that level of detail. And also as you address the parts that are easy to do, all that is left is the parts that are hard or annoying or time-consuming to do, which can be somewhat demoralizing.

Same sort of thing goes with cost. You start out with “so here’s the cost for the big parts and everything else that’s left shouldn’t cost that much” (black-box warning on “everything else”!), and then you go to Home Depot and pick up a bunch of 4-inch ABS pipe and black glue and all the 90 degree and T pieces you need for the aeration pipes and look at the cost and it’s 100 bucks. Home Depot trips add up shockingly fast. There’s also all the stuff you buy that you don’t eventually end up using because the design evolves as you actually try to build it, like buying gears when you don’t actually need gears, and all the stuff you didn’t think you had to buy but it turns out that you do need it.

And sometimes you just get hit with some problem you didn’t expect at all and now have to fix, like “my fan is making a screaming noise, what do”

4. Why is there a valley of death?

Universities and the government funds basic research. Then there’s the private sector of business. The gap between the two, where you have to go from basic research to a business selling the new exciting thing is called the “valley of death”. Now, you’d think this is what R&D is for. But a lot of R&D from a business seems to be focused on marginal improvements to existing things that already fall under the scope of what the existing business does, and not so much on building a novel thing that can be the seed of a new business. Building a novel thing requires a wide knowledge base, as discussed before, and inevitably takes a lot more money and time than expected. It’s the sort of thing done by inventors in a garage as a project of love, not the sort of thing you get paid to do.

Further, crossing the valley of death requires both the technical capacity to build the thing, and the business skills to make a new business from scratch. If you have several people with different skills joined together, it can be bridged, but one flaw of doing it alone is that there are a lot more inventors with the ability to build the thing, than inventors with the ability to build the thing and also the ability or willingness to start a business that sells the thing. I’m in the former category. I can build it, but I hate building it and if I have to build all the machines myself to sell, I’d flatly reject it, and I really don’t want to be responsible for running a business selling it, I’d have no idea how to run a business, and it’d eat too much time. My dream is to get a design good enough to sell, patent it, find someone willing to make a business out of it, and just receive a cut of profits without having to be involved in anything more regarding the production or selling of the machines, besides helping out with technical design work. Further, someone with just the business skills won’t necessarily have the technical ability to come up with the machine in the first place, let alone build it. And there’s also the lemon market problem of businesspeople identifying competent non-scam technical people with a viable design, and technical people finding competent non-scam businesspeople.

There are further issues such as designing the new invention such that it is robust and keeps working for a while (not a property that prototypes generally have), and designing it such that it is easy to build and maintain (also not usually a property associated with garage prototypes).

I’ve heard that there’s a company in the UK that takes garage prototypes and updates the design for robustness, easy constructibility, and cost, which seems like an important part of closing the valley.

5. Building alone vs building as part of a team.

In a certain sense, I was blessed on this project, because I had complete control over the entire design. I had to contend with no meetings, and no unexpected changes to parts of the design that were already locked in, and no team decisions that were dumb and couldn’t possibly work. It’s the dream for anyone who dislikes group projects in engineering. All failures are attributable to me alone, as well as all successes. Then again, having someone else to work on the project with me definitely would have sped it up and I could rely on their knowledge of things I was ignorant of, relaxing the polymath requirement. Maybe there’s an optimal design team size? I guess it’d depend on how parallelizable the work is, as well as how decision-making-quality scales with group size.

6. Final diagnosis and where to go from here.

So, it was over-time and over-budget and didn’t work as well as I had hoped, but it does indeed work. Planning fallacy is a huge obstacle here, and I now certainly see why there’s a valley of death for this sort of work.

In order to make a version that’s practical for domestic use, I’d have to redo the design to be a rain-column design, primarily because it only requires high airflow, instead of the combination of high airflow and high pressure, which requires buying an expensive fan from China and the expensive electronic components which provide the appropriate power to operate the fan. A rain column design could use a much cheaper and simpler fan that operates from a wall outlet.

Further, in order for others interested in CO2 reduction to have one of their own, I’d have to team up with someone who could make a small business in assembling and selling these things, preferably involving someone who is not me building the relevant thing. PM me if interested.

Diffractor30 Nov 2019 21:20 UTC
113 points
37 comments8 min readLW link