Rudaiba

Karma: 34

Rudaiba May 11, 2025, 2:08 AM
3 points
0
in reply to: bhauth’s comment on: Corbent – A Master Plan for Next‑Generation Direct Air Capture
What? No, I was just showing those numbers to see if electro-swing might have a chance. I didn’t use AI at all. I’m just curious about biomass stuff, that’s why I was looking more into it. Nothing I wrote was AI-generated. I’m actually super interested in this topic. I’m really Sorry, Sir, if my writing sounds weird or whatever, but I didn’t use AI.

Rudaiba May 10, 2025, 9:08 PM
3 points
0
in reply to: bhauth’s comment on: Corbent – A Master Plan for Next‑Generation Direct Air Capture
Thanks for the thorough feedback, bhauth. You’re right about biomass being the cheapest current option for carbon removal.

After researching current biochar systems, I can see why you consider this the more viable approach:
Current biochar carbon removal costs range from $130-180/t-CO₂ according to recent studies, with a carbon yield of ~2.7 t-CO₂ equivalent per ton of biochar. Charm Industrial’s bio-oil injection method sells at $600/t-CO₂ today but targets $230 by 2030 with scaled production.
The land constraint you mentioned is the key challenge though, even optimistic assessments from the IEA suggest biomass approaches top out at 3-4 Gt/yr globally due to available sustainable feedstock. This is substantial but falls short of the 10+ Gt/yr that climate models suggest we’ll eventually need.
That’s precisely why we’re exploring electro-swing approaches despite their current higher costs. We believe there’s room for multiple carbon removal methods in the solution space, especially when considering land use constraints.
You raised valid concerns about MOF manufacturing costs and durability under real-world conditions. These are exactly the technical hurdles we need to overcome. Our next step is to run a head-to-head TEA comparing slow-pyrolysis bagasse biochar vs electro-swing MOF DAC, using identical discount rates, power costs, and storage assumptions.
Would you mind sharing which specific aspects of the MOF approach you see as most problematic from a manufacturing or scalability perspective? Your industrial chemistry perspective could help us identify blind spots in our thinking.

And would you mind if we could talk more about this in DMs?

Rudaiba May 10, 2025, 6:24 PM
7 points
0
in reply to: bhauth’s comment on: Corbent – A Master Plan for Next‑Generation Direct Air Capture
Thank you! Yes, you’re right that a real moonshot needs solid foundations.
I’m actually a high school student working on this as part of a TKS moonshot project. We’re exploring big ideas that could potentially make a difference, even if we don’t have all the manufacturing expertise yet.
Your technical feedback is really valuable since it highlights the practical challenges any real implementation would face. The points about materials cost and the gap between lab and real-world performance are exactly the kind of things we need to consider.
Our project is more conceptual at this stage, but learning about these real engineering and economic constraints is super helpful for understanding what it would actually take to make something like this work.
Would you mind sharing what you think would be the most promising direction for carbon removal technology based on your knowledge of industrial chemistry? I’d love to learn more about which approaches you see as most viable.

Rudaiba May 10, 2025, 4:48 PM
3 points
0
in reply to: bhauth’s comment on: Corbent – A Master Plan for Next‑Generation Direct Air Capture
Hey bhauth,
Consider this, we’re proposing a moonshot here, not just an incremental product improvement.
The Wright brothers’ first flight went 120 feet. People rightfully said “that can’t possibly scale to cross oceans with hundreds of passengers.” But engineering evolution changed everything. That’s what we’re aiming for.
On your specific points:
1. About concentration differences: Yes, the Hemmatifar paper shows both capture at 400ppm AND release at dilute concentrations, which is the hard part. The 0.7 MWh/t already factors in the energetics of the concentration swing. That’s what makes the faradaic efficiency measurement meaningful, it’s measuring useful work against the thermodynamic minimum.
2. On materials cost: You’re right that poly(vinylanthraquinone) + CNT electrodes aren’t cheap today. Neither were silicon solar cells in 1980 at $30/watt. Current industrial MWCNT prices are $199-400/kg in tonne lots^[1], which puts our mixed MOF/CNT slurry under $30/m². The economics depend on print throughput more than raw materials.
3. Real-world cycle life: You’re 100% right that lab conditions aren’t the real world. That’s precisely why we’re talking pilot stage first. The water-stability data (≥90% capacity after >20,000 wet/dry cycles at 70% RH^[2]) is encouraging, but dust and VOC poisoning still need field testing.
4. On the TEA: The battery analogy isn’t perfect, but dismissing it entirely overlooks that they did include a bill-of-materials for printed MOF/CNT sheets and stainless current-collectors. It’s not just handwaving. The fundamental advantage is the massive energy reduction (5-10× less than thermal systems), which drives OPEX way down.
5. About basalt: Yes, it’s pressurized CO2 injection. The specificity of basalt matters because its calcium/magnesium silicate content directly enables the rapid mineralization rates we need for verification. The Carbfix audit shows $25/t storage cost^[3], which is far cheaper than most alternatives.
Look, I completely understand your skepticism; most moonshots fail. But the data suggests this path is worth exploring. We’re not claiming it’s risk-free or guaranteed, just that the published record contradicts “can’t even get close.”
If you were evaluating airplanes in 1910 or solar panels in 1980, you’d be right to point out all their limitations. But sometimes engineering evolution surprises us all.

Rudaiba May 10, 2025, 1:43 PM
3 points
0
in reply to: bhauth’s comment on: Corbent – A Master Plan for Next‑Generation Direct Air Capture
Hey Bhauth, thanks for the tough questions. You’re right, most DAC tech is way too expensive. But I think there are some misunderstandings about what we’re actually doing here:
Look, electro-swing at true air concentrations (400 ppm) does work in peer-reviewed data. In 2022, Hemmatifar showed a stackable bipolar cell capturing at 400 ppm with electrical work of ~0.7 MWh/t while maintaining >90% efficiency^[1]. They even ran it continuously for 100+ hours without fouling issues.^[2]
You’re totally right about fan power, though; it’s not trivial. Our analysis shows 300-900 kWh/t for fan electricity at realistic velocities. That’s significant! But still way less than thermal systems needing 5-10 GJ/t.
On MOF degradation—that used to be a showstopper, but not anymore. Water-stabilized MOF-74 variants now tolerate >20,000 cycles at 70% RH while keeping 90%+ of their CO₂ capacity.^[3] Multiple labs have verified this. With a 15-minute cycle, that’s roughly three years of service life.
The cost thing is the big question. Recent TEA in ACS Energy & Fuels modeled a 200 kt/yr electro-swing system with wind power and projected $56-97/t.^[4] Is that optimistic? Maybe. But it shows the approach isn’t fundamentally impossible.
And no, we’re not grinding up rocks! That’s the key misunderstanding. Basalt mineralization (specifically the Carbfix method) injects CO₂-water directly into porous basalt formations. >95% mineralizes in under two years.^[5] Real-world audits put this at ~$25/t^[6], way cheaper than grinding olivine.
We picked basalt specifically because it works in practice, not just theory. The Carbfix project in Iceland has demonstrated this at scale.
I’m not saying this is easy or guaranteed. But the published data doesn’t support “not even close,” it suggests it’s at least worth exploring. Biomass approaches are great where land is available, but they can’t scale to the gigatons we’ll need.

Rudaiba May 10, 2025, 11:24 AM
1 point
0
in reply to: Austin Chen’s comment on: Corbent – A Master Plan for Next‑Generation Direct Air Capture
Yes, this is for a program I’m doing right now. It’s called TKS, and they have challenges almost every month. This time, it’s about coming up with a moonshot idea. There are specific deliverables, and one of them is a clear, in-depth article about the idea, which is this. That is the main reason. Regarding the founders, yes, we are seeking co-founders and partnerships. And if you guys have thoughts on this, please share it. It helps me improve the idea and see it from other perspectives. Thank you!

Corbent – A Master Plan for Next‑Generation Direct Air Capture

RudaibaMay 10, 2025, 4:09 AM

11 points

15 comments19 min readLW link

Machine Unlearning in Large Language Models: A Comprehensive Survey with Empirical Insights from the Qwen 1.5 1.8B Model

RudaibaFeb 1, 2025, 9:26 PM

9 points

2 comments11 min readLW link

Rudaiba Feb 1, 2025, 3:27 AM
1 point
0
on: [Error communicating with LW2 server]
Hey!

Rudaiba

Cor­bent – A Master Plan for Next‑Gen­er­a­tion Direct Air Capture

Ma­chine Un­learn­ing in Large Lan­guage Models: A Com­pre­hen­sive Sur­vey with Em­piri­cal In­sights from the Qwen 1.5 1.8B Model

Corbent – A Master Plan for Next‑Generation Direct Air Capture

Machine Unlearning in Large Language Models: A Comprehensive Survey with Empirical Insights from the Qwen 1.5 1.8B Model