CYAN Project Progress

The goal here is to push forward DIY Micro Household Open-Source Direct Air Carbon Capture. I built such a unit for < $100 and am documenting my progress for those who would like to replicate this.

This alone won’t solve the climate crisis but it is a critical step in the right direction to get folks interested and engaged in Direct Air Carbon Capture. It’s important that people see how easy and possible this is to do, and to support any/all efforts that remove CO2 from our atmosphere.

Picture of the first Cyan

Initial Carbon Capture quantities, will update over this weekend with a better test

Videos on Progress:

Calcium carbonate detection, demonstrating CO2 capture

An informal introduction and Q&A on the initial Cyan unit

1 Like

The conversation that follows from the second video is so OpenAir :smile:

Great weekend collecting and drying calcium carbonate, I think we have something usable.

I repeated the above experiment with a larger calcium hydroxide sorbent quantity and 24 hour runtime. I also ran a third experiment with a vastly simplified system and shorter runtime required to achieve good CO2 capture. Just bubbling air through the sorbent for 6.5 hours works as well as using the file box (water pump + fan) for 24 hours. No complicated system or instructions necessary, I literally had 3 pieces to my stripped-down system (aquarium air pump, air stone+tubing, and plastic cup).

Here is the summary of all three experiments performed so far (the one above and the two I ran this weekend). This shows that for every amount of CO2 captured, twice as much input material must be added. So for a kg of CO2, 2 kg calcium hydroxide.

And here are screenshots of the calculations and notes on experimental conditions for the two new data points:


I recommend the air bubbling approach as it is way simpler. I used a Tetra Whisper 10 aquarium air pump with an air stone connected by a short length of tubing. The only issue is that with the new batch of calcium hydroxide (from Bonide Hydrated Lime), there is a clay-like consistency to part of the output produced that makes it hard to get out of the bottom of the cup when wet.

Since I didn’t have this earlier when mixing my own calcium hydroxide, my guess is this is made up of impurities from the hydrated lime. The MSDS says >90% calcium hydroxide; the label on the back suggests there are quantities of MgOH, CaO and MgO present. The last two convert to hydroxides upon adding water and there weren’t much of these two because I noticed no exothermic effect upon adding water. From my reading, MgOH (magnesium hydroxide) will absorb CO2 as well, though to a lesser extent than Ca(OH)2.

Updating this after a prolonged dry time of 6 more hours: Despite appearing dry, I still had some residual water earlier in my dried CaCO3 so the CO2 calculation was a bit high. Now the weights have stopped decreasing further and I am getting 2.5x Ca(OH)2 per 1x CO2 captured. (Earlier I had a 2:1 ratio.)

Hi Cyan,
Very interesting work. Do you know how many DAC cycles we can do with Ca(OH)2 and MEA?.

Hi, thanks for the question!

Ca(OH)2 and MEA are two very different types of sorbents for CO2; Ca(OH)2 does not have cycles since it is effectively one-way.

It is possible to regenerate the Ca(OH)2 if you heat CaCO3 to drive off its CO2, restoring CaO which can be hydrated with water to form Ca(OH)2 again. Such heating would need to be done at a solar concentrating plant since otherwise the emissions from fossil fuels would greatly subtract from the CO2 sequestered. It is unlikely that a home setup would achieve the temperatures required (>800 C), that is why I mention CO2 absorption with Ca(OH)2 is effectively one way.

MEA, on the other hand, is an amine sorbent that has a weak affinity for CO2 and thus under the right conditions the CO2 can be released. This allows MEA to be used in cycles.

Some thoughts: if you really want to lock up CO2, capturing it in mineral form would seem to be a good strategy. CO2 in the form of CaCO3 is stable, does not require pressurization which takes energy, and CaCO3 is the main constituent of limestone, a natural material.

I’ve been told there is always unused CaO from the cement industry that could be hydrated to form Ca(OH)2. This could be used for starters to supply Ca(OH)2 from a source where it would otherwise go to waste.

Other than this source, producing Ca(OH)2 in a renewable manner is necessary. Since most CaO comes from CaCO3 (limestone) already, capturing the CO2 given off upon creating CaO is absolutely necessary (can be done with amine sorbents on the required industrial scale). Then the CaO can be hydrated to form Ca(OH)2 which once again can absorb CO2, this time on the consumer scale in individual Cyan units.

Important note: In the above spreadsheet screenshots, I did my calculations assuming 100% absorption of CO2 to form CaCO3. This in all likelihood is not the case based on what I just found in the literature as being expected for ambient CO2 concentrations at near room temperature for a 2 hour time frame (https://www.journal.csj.jp/doi/pdf/10.1246/bcsj.6.319 - see Table 2). Thus the graph above actually represents maximum CO2 that might be captured, not actual CO2 captured.

After lots of thought, I can’t actually think of a good way of determining actual CO2 captured. Might need to wait for OpenAir’s Sorbent Tester or maybe someone has another idea.

Another note for completeness: I had made a copy-paste error for the g/mol of Ca(OH)2 (cells D13 and D10 two posts up), it is 74.09268 rather than 110.984. This did not affect any later calculations since this error was able to cancel itself out.

Linked source: “Studies on the Reactions Between Gas and Solid. Part II. Absorption of CO2 by CaO and Ca(OH)2.” Takeo Aono, 1931. A really great reference.

What all this means is that I know I am getting CO2 captured (there’s a weight increase measured and a positive vinegar test), but I can only calculate the maximum possible absorption and not the actual amount.

My next step is to try to quantify the weight increase per unit time passed and to compare that with the results in the above reference. Hoping to have enough data points soon for a good graph, though it takes at least 24 hours per experiment (runtime and dry time).

I’ve repurposed the insides of my Cyan unit and now I’m getting consistent weight increases with an even simpler setup. Total cost is now under $40 in parts to replicate - file box ($10), air pump ($10), tubing + air stone ($5), ultrasonic fogger ($10), and water container ($2)

In this video you’ll see the 100% humidity from 5 micron sized water droplets generated by a 1.7 MHz ultrasonic fogger. This, combined with inputted air, yields a dry weight increase of 5.1% after 1 hour of treatment time and 16 hours drying time (during which CO2 takeup is also possible). The only reaction that should be occurring here is carbonation, so I’m expecting this weight increase to be due to CO2 capture.

Here is a video of the new setup. The red cup was used in another test and does not serve a function in this experiment. The coffee filter and plastic on either side of the box are to keep the humidity in. The fan was not being run. There is, instead, an air pump to the left of the box that is feeding air through an air stone into the same container where the fog is being generated. There is 10 g of powdered Ca(OH)2 on a coffee filter at the bottom, spread out to no greater than a 5 mm thickness.

Compare this faster performance with the slower carbonation rates in this paper (Carbonation of lime-based materials under ambient conditions for direct air capture - ScienceDirect), also using ambient air but of much lower humidity.

The humid air treatment is important because without it, I see only a 1.64% weight increase after simply soaking the 10 g Ca(OH)2 in water, then drying for 16 hours. If I apply the treatment for 2 hours and dry for 16 hours I get a 6.4% weight increase. One note, at 2 hours I am getting a lot of condensation on the bottom so that the Ca(OH)2 is in a bit of standing water while inside the coffee filter.

I have a 12.5% weight increase to report, which is double the highest I’ve had. I put in 10 g Ca(OH)2 and out came 11.25 g of product. That is 1.25 g of CO2 uptake and I only used 1.5W of electricity for everything (just the air pump). Thus, for every quantity of CO2 you want to capture, you will need to put in 8x the amount of Ca(OH)2 in a flat layer to maximize CO2 absorption. I’m hoping to make further improvements still.

The total time is still 16 hours, but instead of 1-2 hours of Ca(OH)2 treatment and 14-15 hours of dry time as before, I did 13 hours treatment and 3 hours dry time using a fan. The much shorter dry time was enabled by not getting the Ca(OH)2 so wet when humidifying it. To do that, I had it sit in a coffee filter inside a lid on top of an old Sabra hummus container which happens to fit perfectly within this container.

The Ca(OH)2 treatment consisted of humidification as a result of constant airflow into the water of this tiny container. The container and air pump can also neatly fit inside the file box if desired. The airflow was not a lot, just that produced by the air pump. It is possible I could get more of a percent weight increase from more airflow, but the challenge will be retaining the necessary humidity.

I’ve just published a 5-minute setup and intro video for my Cyan unit. You’ll be able to see where the various parts are, how it is assembled, how it works, and finally what kinds of results can be expected from it.

I have done some additional research into the use of sucrose (table sugar) to improve carbonation, and thus the formation of calcium carbonate. Here are 4 pieces of evidence that sucrose improves carbonation, and that it is possible to remove the sucrose from the calcium carbonate later, yielding a high purity product:

  1. Excerpt from https://www.researchgate.net/publication/265850297_Surfactant-assisted_synthesis_of_pure_calcium_carbonate_nanoparticles_from_Sri_Lankan_dolomite showing a 25x increase in the yield of calcium carbonate in aqueous solution when prepared with sucrose than without.

image

  1. Cheney Lime & Cement Company says at that link “the addition of 35 grams of sugar will increase the solubility of the calcium hydroxide from 0.159 to 13.332 grams per 100 grams of saturated solution at 25C; which is a solubility factor increase of 84.”

  2. Effects of organic additives on calcium hydroxide crystallisation during lime slaking - ScienceDirect provides several papers as references that sucrose increases the rate of carbonation. Do a search on “carbonation” in the paper and you’ll be able to find the sources.

  3. I found a patent on using much larger quantities of sugar to enhance the rate of carbonation of calcium hydroxide: EP2483203A1 - Production of calcium carbonate - Google Patents. The quantities of sugar I am using are much lower since I don’t think it’s reasonable to use 5-10x more sugar than calcium hydroxide. Also the patent is for an aqueous solution whereas I am using moist air.

  4. I said I had 4 pieces of evidence…I’ll add a 5th which are my experimental results from the past few days. Here I am not subtracting the weight of the sucrose from the initial weight because I’m not yet sure of the nature of the complex it forms with calcium ions. From the literature, it is possible to separate the sucrose from the calcium carbonate by dissolving in water, but that is another step that will likely result in some product loss due to filtration. Note that if I did subtract the weight of the sucrose from the initial weight, the weight increases measured at the end would be more than double what they are (column J). Better to underestimate at this early stage than to overestimate.

1 Like

I’m watching this morning’s meeting video now, and I was thinking along the lines of this maximum - wondering if some of the weight gain is from moisture. Are you starting out with a certain moisture level and then drying to get back to that level? I’m not totally clear on the process.

Hi Sue,

Thanks for asking, great question! In short, I am indeed weighing a small amount of additional moisture left over from humidification. However, it is a very small amount which seems to be responsible for continued carbonation.

The material I’m starting with is over 90% calcium hydroxide according to the Bonide Hydrated Lime SDS. It is already hydrated lime (cement that has taken up water). The rest is magnesium hydroxide and sand, according to the bag. There should be no residual calcium oxide to take up water as it is a pretty fast process for calcium oxide to pull moisture from the air.

The humidity here is about 10-15% consistently, too low for carbonation to proceed rapidly without humidification. So if I leave some hydrated lime sitting out I don’t see a weight increase. This shows I am not getting weight increases simply by exposing the material to ambient air.

I also re-weighed some of my earlier samples in the Supporting Data > CO2 Capture Calculations document on the Cyan Google Drive after 9-10 days. These had gained a little more weight, which is consistent with additional carbonation at a much slower rate.

Since there has to be water present for this after-carbonation to occur, that means yes, some of what I’m weighing must be moisture. However, there is also calcium carbonate definitely present because when I add 5% vinegar I get a good burst of bubbling that trails off after a few seconds.

Also when I do my weighing I wait until there is no change after 30 minutes and I count that as the final weight. Thus I didn’t notice the after-carbonation until weighing again days later. The product at this stage is crumbly dry. So there is not much moisture left.

1 Like

Progress on Upgrading Cyan’s Carbon Capture Capability

I had ordered some parts a few days ago to prepare for increasing my unit’s carbon capture. I used 5 EZSTAX plastic dividers (they come in packs of 12), cut the 5 in half, and cut the tabs off to make these fit within the file box. I then used some foam pipe insulation I happened to have, cut it into 1/2" wide chunks, and slipped these pieces in as spacers between the plastic dividers. Lastly, I used electrical tape to compress the stack and hold it in place. Two pieces on the sides facing the front and three in the rear did the job. Here’s the result:

The stack is resting on a plastic container that used to hold turkey slices. Inside that plastic container rests an ultrasonic fogger I had purchased for $9.99, it is 24V 1A so that’s 24W of electricity used whenever it is on. However, it does an excellent job at producing quick humidification and it will only be on for short bursts of time.

Today I put in an order for a $42 humidistat that will supply power to the ultrasonic fogger when the humidity is low and turn it off when the humidity reaches a set level. When that happens, another outlet will turn on the fan to its lowest speed which will serve to replenish the air with CO2 and cause the humidity to fall. The optimum humidity for carbonation of calcium hydroxide is between 50-70% so I really don’t need condensing levels of humidity. The cycling will help lower energy use. And to validate the energy consumption, I ordered a $12 smart plug from Emporia Energy; it will beam stats via Wi-Fi to my phone so I can see just how much energy my Cyan is using.

When these items arrive on Tuesday, I’ll be able to test how much of an increase in CO2 capture I get from using the stack. Will post an update once I do the test.

Progress on DIY Low-Carbon Calcium Hydroxide from Waste Cement

This is a research item that I just started today. There is a seemingly endless supply of waste cement from concrete plants distributed everywhere, and if this could serve as the raw material for Cyan rather than buying bags of hydrated lime, this would be of great benefit.

To quickly simulate waste cement to see how I could possibly get calcium hydroxide from it, I went to the hardware store and bought a 47 lb bag of Portland cement for $6.50. It was a rather small bag for its weight, and the small cost really, really does not take into account all the environmental impacts of making cement, transporting it to the hardware store, etc.

Dried Portland cement is composed of calcium silicate hydrate (C-S-H) and calcium hydroxide. The calcium from C-S-H can also be leached out to add to the calcium available to make calcium hydroxide, leaving behind a silicate hydrate.

Here is my Portland cement drying so that tomorrow I can hopefully break these up for further testing. I used a 1:1 ratio of cement (20 g) to water (20 mL) for all but two of these.

First Look at Cyan 2:

Here it is, fully operational. This video also shows the inside of Cyan 1, working away to the left of Cyan 2. Cyan 1 is carbonating Mg(OH)2 instead of Ca(OH)2 to see how it performs. I have 100 g Ca(OH)2 loaded onto the different stacked layers of Cyan 2. I am using a $42 humidistat to regulate the humidity within a high but non-condensing range. I am also using a $10.99 smart outlet from Emporia Energy to determine energy consumption, important for determining the carbon footprint relative to my CO2 uptake.

Will update us soon on the performance of the Mg(OH)2. As for Cyan 2, I have to repeat the experiment again due to spillage of the Ca(OH)2. A better design would be to have very shallow trays instead of the open ends.

My waste cement experiment was on hold since I found a promising low-carbon magnesium hydroxide supplier, and now I have to see how well mag hydroxide performs. If it works well, we will probably use that instead of calcium hydroxide.

Magnesium Hydroxide Results

The previous design was not optimal because there were no lips on the sides of each layer preventing the material from falling off. Since the goal was to test the performance of magnesium hydroxide as well as whether a stack of layers would improve the CO2 removed, I did the following two experiments.

The first experiment in the spreadsheet below is with a single layer of Mg(OH)2 on a coffee filter in Cyan 1. It shows that the CO2 absorption is on par with calcium hydroxide. We get 2.15 g CO2 captured for 10 g of input material.

The second experiment below is with 11 layers, also done within Cyan 1. The CO2 absorption per layer is not as good (1.84 g CO2) potentially because there was low humid airflow to the rear of the stack. A longer dry time was provided to ensure dryness given the greater density of material in the stacked layers.

Total: 22.5 g CO2 captured!

This is the stack for the second experiment. It is composed of 12 business cards held in place by nuts mounted on 4 screws. These create 11 layers (the top layer is unused). Business cards also make up the back and two sides to prevent material from falling out.

Here is the stack, loaded with material, showing the distance from the air stone. The air stone would eject droplets of water that slowly wet the business cards over the course of several hours.

Bonus image for today…my first try mixing a 1:1:1 mixture of calcium carbonate product, biochar, and Portland cement, and pouring them into cube molds to form different sized cubes. Everything but the Portland cement is a form of CO2 storage, that’s 2/3 of each block by volume. Biochar keeps air-derived CO2 converted to biomass from being released as CO2 upon plant decomposition, and the calcium carbonate also contains air-derived CO2.

1 Like

A Vibration Platform for Cyan

In an effort to boost carbon capture efficiency, I set up this vibration system to intermittently disturb the input material. Tests by University of Michigan students showed that just disturbing the surface of the hydroxide powder with a card once halfway through the experimental run doubled the amount of CO2 captured, so I figured something like this could help.

This system consists of one 6.5" speaker (I bought 2 for $44.95), one amplifier ($34.79), and one cell phone (my iPhone, though an Android phone with an equivalent free function generator program would also suffice), along with an RCA to headphone audio cable, and one headphone to lightning audio adapter, the latter not needed with Android phones. The speaker is resting in a plastic bowl I had around the house.

Here is a screenshot of the free function generator from my iPhone. It allows 1 Hz alterations in frequency and many different kinds of waveforms. I found that 20 Hz seemed promising, and that is what is shown in the video which is in real time.

And here is the power consumption of the speaker and amplifier when in operation. The cell phone was on battery so I don’t have its power consumption. It stayed pretty constant at 3.63 watts, but it probably only needs to be on once an hour for a few seconds. I took this screenshot also from my cell phone; the Emporia Energy app receives data from the smart outlet I had purchased previously to monitor my energy consumption.

For next steps, I will likely put plastic wrap over the speaker to waterproof it and pre-wet the powder beforehand so that it dries on the filter paper on a lid above the speaker. It may close off the pore structure of the powder grains with water, but hopefully the extra excitation will yield improved carbon removal.

60% greater carbon removal with vibration

For this experiment I had two plastic bowls, each with a coffee filter holding about 10 g Ca(OH)2. I slowly added 30 mL water to each bowl from either side, between the coffee filter and the bowl so as to not disturb the material too much and to allow the water to naturally wet the material. Both bowls were kept in the same area of the room. For the next four hours, I intermittently turned on the speaker once per hour for 5 minutes on one of the bowls; the other served as a control and did not get vibrated. The 14 Hz vibration was quite vigorous and was chosen due to yielding the highest amplitude of standing waves within the wet material. However, not all the material participated in the motion - some of it remained stuck to itself.

After 4 hours, both bowls were allowed to dry naturally without the help of a fan. Due to the additional water added to each bowl at the beginning, it took a while to dry. The bowls were left for 3 days to dry fully. The vibrated bowl had a lot of material that went through the filter paper and ended up between that and the bowl; the non-vibrated bowl did not.

The vibrated bowl had an 11% weight increase while the non-vibrated bowl had a 7% weight increase. In carbon captured, the vibrated bowl captured 1.81 g CO2 while the non-vibrated bowl captured 1.13 g CO2. In terms of % increase, (1.81-1.13)/1.13 * 100 = 60% greater carbon removal was seen with the vibrated bowl.

In terms of energy consumption, the setup consumed 2 watts on standby and 6 watts when in motion. I ordered a timer outlet to not only allow automatic vibration once per hour, but also to eliminate the standby energy consumption.

1 Like

Encouraging results! There are a number of options available for reducing the power consumption. A “Class D” amplifier can be very efficient, and can have require very little standby power, especially if you’re not aiming for audiophile fidelity :grinning:. Here are some example breakout boards that look like possibilities, depending on how much wiggle you need: Adafruit Industries, Unique & fun DIY electronics and kits.

1 Like