Epistemic status: Honestly, I'm still only getting into this. I am trying to track my time spent on projects better to be more accountable, so I installed the 'Time Things' Obsidian plugin to measure how long I wrote this document for. Right now, I've spent 5000 seconds writing this document, not including reading research (which I am trying to figure out how to measure better). I just want to get quick feedback on whether this is a decent concept first. ## Background reading and context ### Does Far UVC Work? - [[Blueprint Biosecurity, Blueprint for Far UVC]] - [[Eadie, E., Hiwar, W., Fletcher, L. et al. Far-UVC (222 nm) efficiently inactivates an airborne pathogen in a room-sized chamber. Sci Rep 12, 4373 (2022).]] - [[Eadie E, Barnard IMR, Ibbotson SH, Wood K. Extreme Exposure to Filtered Far-UVC A Case Study†. Photochem Photobiol. 2021 May;97(3)527-531. doi 10.1111php.13385. Epub 2021 Feb 3. PMID 33471372; PMCID PMC8638665.]] - [[Blatchley, E. R., III, Brenner, D. J., Claus, H., Cowan, T. E., Linden, K. G., Liu, Y., … Sliney, D. H. (2022). Critical Reviews in Environmental Science and Technology, 53(6), 733–753. doi.org 10.1080 10643389.2022.2084315]] ### Modelling Far UVC - [[C.J. Noakes, L.A. Fletcher, C.B. Beggs, P.A. Sleigh, K.G. Kerr, Journal of Aerosol Science, Volume 35, Issue 4, 2004]] - [[Yue Pan, Tongling Xia, Kangqi Guo, Yuting An, Chun Chen, Predicting spatial distribution of ultraviolet irradiance and disinfection of exhaled bioaerosols with a modified irradiance model, Building and Environment, Volume 228, 2023,]] ## TL; DR We need to do more trials, but right now I have some reasons to believe that far UVC light seems to kill pathogens, whilst being overall safe to humans in controlled doses. See [[Eadie, E., Hiwar, W., Fletcher, L. et al. Far-UVC (222 nm) efficiently inactivates an airborne pathogen in a room-sized chamber. Sci Rep 12, 4373 (2022).]] [[Blueprint Biosecurity, Blueprint for Far UVC]] When more studies confirm its safety, there will be challenges in implementing it for the wider world. I want to write more about this in future posts. As a first step I decided to try and do a first principles model myself of Far UVC in a ventilated room. My goal right now is to figure out - how I might safely install one in my house if I could, or in a general space, and then - create some sort of tool that makes it easy to do this. I did this as an exercise to force me to think about how this could work. It also biases me a bit more towards creation, which I think is important for understanding. Also, actually doing it helps me figure out what better 'tools' we might need to make this whole process easier, and where the current tools are bad. I am working on a system which uses some CFD to efficiently calculate where to place Far UVC lamps for a given room, and given ventilation space. The dream solution aims to be able to run on any desktop computer or phone. ## The Problem It seems to be currently hard to model the optimal installation of far UVC, given an environment with different factors. The modelling literature and quality itself doesn't seem to be very developed yet. And even if it was, I couldn't find easily available web apps that allow you to do this. My attention was brought to these concerns from [[Blueprint Biosecurity, Blueprint for Far UVC]], where one of the research priorities is the design of practical guidance in different environments. > Sophisticated modeling, such as computational fluid dynamics combined with lighting simulation, is needed to translate experimental findings into practical guidance for the safe and effective deployment in diverse environments. As these research methods can be inaccessible to end users, the goal of this research has to be practical heuristics that can be endorsed by public health agencies and implemented by practitioners at scale (see Recommendation 5). There are a lot of degrees of freedom that would need to go into a good model - Where the ventilation points in the room are. - How fast is the ventilation happening. - On average, how many people in the room do we think are infected. - Where and how high do we place the Far UVC lamps. - Where and how high do we place the diffusers (if at all). - How intense should the Far UVC lamps emit given the room size. - How do we model the humidity. - Does the intensity change depending on which pathogen is in the room (z values) And right now, I think that a good model which is - can be run on a computer / server cheaply - accessible to many - reproducible - very easy to use to the point that it just works on your phone, like airBnB just doesn't exist. Imagine ## Academic Modelling and Solutions So far, the literature on this has been quite light, which is as sign in itself that there is more work to be done. So far I've only found a paper which does some computational modelling on Far UVC itself - [[Yue Pan, Tongling Xia, Kangqi Guo, Yuting An, Chun Chen, Predicting spatial distribution of ultraviolet irradiance and disinfection of exhaled bioaerosols with a modified irradiance model, Building and Environment, Volume 228, 2023,]]. This paper seems to be the most comprehensive that I've found so far. Some 'points' on reading this paper - There doesn't seem to be any available code to model the CFD. - There is a lot of dependence on the shape of a far UVC lamp and its irradiance model. - It appears the CFD provider is a commercial one, and so the results are hard to replicate. - The exact method to link the irradiance model, with the flow model, and therefore the bioaerosol concentration model is unclear. Caveat this with the fact that I just might be reading this wrong since I am a beginner. Their 'all in simulation' is a time evolved map of virus concentration with and without Far UVC. Basically, this is what I want to replicate: ![[Screenshot 2025-04-30 at 10.43.44.png]] There are other papers as well which I think focus on the modelling which are older, and not specifically on Far UVC but the principles should be very similar, like [[C.J. Noakes, L.A. Fletcher, C.B. Beggs, P.A. Sleigh, K.G. Kerr, Journal of Aerosol Science, Volume 35, Issue 4, 2004]] ### The base case This is the simplest base case to test my model against, as demonstrated in [[Eadie, E., Hiwar, W., Fletcher, L. et al. Far-UVC (222 nm) efficiently inactivates an airborne pathogen in a room-sized chamber. Sci Rep 12, 4373 (2022).]], which claims around 90% efficiency at their highest level of intensity. ![[Screenshot 2025-04-30 at 11.00.50.png]] ## What is Painful, and in Theory Solvable? So far, in getting started, I've already come across things that seem hard but are solvable. - I don't know how the exact mechanisms of Far UVC actually kill bacteria. I am also currently uncertain if that would be useful to know. - [[Blueprint Biosecurity, Blueprint for Far UVC]] report doesn't go into the quantum chemistry level detail on how it works, and the references are light. I would love to see a systematic review. - That said, I do have some experiences in protein spectroscopy on diamides, so maybe I should be the one doing this. - Doing CFD and creating meshes is painful - Most CFD software providers you need to pay, which limits access to many people. - I'm currently experimenting with OpenFOAM, which seems to be the only open source CFD provider out there that works for what we want. - But this feels expensive and - There are no 'easy' python packages that you can do this with, unlike the very accessible ecosytem - So far I've manged - Modelling the irradiance of the light source is also painful, and different light sources require different doses. But there seems to be decent guidance on how to do this - Buying a Far UVC lamp is hard - I'm trying to contact Ushio, but it's not clear at all where I can buy one from my home. - I don't know what filter to get for my house, and how to model this. - I don't know how to actually test the intensity of light once I make it. ## Current Progress So far I've managed to set up - A free CFD package, openFOAM on my personal computer with Docker. - This was surprisingly hard to set up, and the documentation was fairly poor in quality. Also from a first glance the standard of documentation seems nowhere near as good compared to the python pip ecosystem of ML packages. - I've managed to get a simple test case working, which is encouraging! - There is a route for me to make this a lot easier to install and model, with a wrapper. That would be a contribution in of itself. - I've also realised that it seems to be possible to run the relevant - Trying to set up a meeting with someone who works on ventilation in openFOAM, trying to get a tutorial on this so that I can be more efficient in getting up to speed. - I'm contacting people in the CFD space via mutual friends from Cambridge to see if anyone can give me quick pointers on any modelling pitfalls. So far I have a lead. ### What would the ideal solution look like Well, something on your phone that would do the following steps: - Take an image of the room - You input the geometry of the lamps that you have - The geometry of the lamps is sort of outlined in [[Yue Pan, Tongling Xia, Kangqi Guo, Yuting An, Chun Chen, Predicting spatial distribution of ultraviolet irradiance and disinfection of exhaled bioaerosols with a modified irradiance model, Building and Environment, Volume 228, 2023,]] - Input where your ventilation points are - Input your average occupancy, and likelihood of someone entering the room that's infected and then... - The program gives some visualisation of the airflow and expected concentration of the virus with and without far UVC, like the images below. - the program calculates the optimal location and intensities of the lamps required for the correct dose of Far UVC. ![[Screenshot 2025-04-29 at 15.26.53.png]] This would be effective for - Hospitals, where staff or architects could just use it direct. - Closed room mobile hospitals, for charitable purposes - Places of worship - note the Malaysian superspreader mosque incident during Covid-19 which basically ruined everything. - Schools - Researchers trying to build rough guidelines for people who won't use this tool - Any other public spaces. But on the other hand, I wonder if it would be easier to just offer rough guidelines for a space. For example, I take a picture of my room And then it models it and creates a ventilation model, and creates something like the above. ## Other Questions and thoughts on related papers ### Takeaways from the Blueprint Biosecurity Paper - I am still unclear on the physics of how far UVC might affect skin proteins. also, they are unclear on how to distribute tech like this in hosptals to make it effective - dumb question - but where do you even put technology like this - still unclear on whether far UVC is actually better vs standard filtration - I think the section on how UVC destroys nucleic acids could be better substantiated, I could just be being a noob though. - Why does the literature seem relatively light on this? Are there not computational chemists working on this problem? Gotta read more on it. - We need to figure out what the energy is to break these peptide bonds. - They cite this paper Kowalski, W. J., Bahnfleth, W. & Hernandez, M. T. A genomic model for predicting the ultraviolet susceptibility of viruses. IUVA news 11, 15–28 (2009). - But this feels a bit unclear, need to do more work on how this actually helps - I would like to see some kind of diagram or wikipedia page - There is also a section on how far UVC can destroy peptide bonds in the proteins of pathogens. If this is the case, we would also expect this to work on mirror bacteria as well. - How does a KrCl lamp actually work? - Electron jumps and energy level and then a wavelength is emitted. - How do LED approaches work - and are they more efficient? #### Action Points - I could make a wiki like page that examines the engineering and the mechanics of how far UVC affects proteins - There is an engineering question on how we can deliver an effective dose, based on computational fluid modelling from the size of the room. What literature is out there right now that we can model this with a fake room? - And how could we turn this into useful guidance for hospitals? - Use relevant techniques, such as computational fluid dynamics modeling and lighting simulation, to model UV dose distribution based on different room geometries, ventilation regimes, occupancy and the location of the infectious source - What is the most impact I could do with this? #### Ideas for creation - A document outlining how to implement far UVC in your room - A program that predicts wavelengths that destroys certain pathogens. - If you look on reddit, people still seem quite worried about it, and there are a lot of heuristics going on. It looks like the zero covid community is checking it out. - Is far UVC safe? As someone who is not a medical expert, how can I get involved in convincing people that it works? - A communication document that clearly communicates the effectiveness of UVC versus just being out in the sun on a sunny day.