Liquid based color matching demo #89
Replies: 7 comments 2 replies
-
Related Examples
Theory
The demo could be used to predict higher-fidelity UV-Vis spectra. |
Beta Was this translation helpful? Give feedback.
-
Bill of Materials (WIP)To be updated on main README once fleshed out.
|
Beta Was this translation helpful? Give feedback.
-
|
BOM being updated at https://github.com/sparks-baird/self-driving-lab-demo/blob/main/reports/clslab-liquid-bom.xlsx CSV (not necessarily always up-to-date) at https://github.com/sparks-baird/self-driving-lab-demo/blob/main/reports/clslab-liquid-bom.csv |
Beta Was this translation helpful? Give feedback.
-
Chemically resistant replacements/modifications
Adding more PWM channels (i.e. more than 16)Some options:
|
Beta Was this translation helpful? Give feedback.
-
TaskSome thoughts after running an optimization campaign with the liquid color-matching demo:
|
Beta Was this translation helpful? Give feedback.
-
|
This project is awesome again. It made me think about how cementitious materials are a lot like food dye - hear me out 🤣 It is a material that is usually characterized by spectroscopic methods such as XRF and XRF - It defies a complete characterization which is why experimental validation of the final product is always required. Here lies an optimization potential. The difficulty with cementitious materials is that amorphous and crystalline structures governs short-term and long-term behavior differently and are difficult to quantify distinctively - making validations an absolute must. Maybe you can introduce some sort of aging process in the color mixing 🤔. In addition, the composition of alternative supplementary cementitious materials such as calcined clays or waste slags may be constantly changing - just as food dyes may not always have the same color when mixed. They also present an optimization challenge that is highly dependent on (fully interchangeable) constituent information such as location, available volumes, eco-footprint, etc. Thus, almost endless optimization scenarios can be created. I see many scenarios where food dye could act as some sort of surrogate material to develop clever optimization strategies for cement. However, unlike cement food dye has the great advantage of not being in powder form - one of the main drawbacks in the automated synthesis of cement. I am thrilled to see how this project develops. |
Beta Was this translation helpful? Give feedback.
-
|
@BAMcvoelker thanks for the really interesting comments! This is good food for thought for me. I like the idea of incorporating aging in the color-mixing process. It's also good to know that the demo can capture some additional real-world application scenarios. Thanks for sharing! |
Beta Was this translation helpful? Give feedback.
-
|
Note: I made a working prototype, though it probably needs a 3D printed housing for the cuvette and could certainly be cleaned up regarding wires and connections.
https://github.com/sparks-baird/self-driving-lab-demo#clslabliquid |
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Follow-up to #77.
Idea Summary
In principle, this is very similar to Atinary SDLabs Demo and a few other color-matching demos, but with a focus on spectrophotometry. Here is the visual summary:
Drip bags
For the drip bags, polyolefin is probably the right chemical-resistant material, a combination of HDPE and PP.
Cuvettes
https://www.amazon.com/BrandTech-759071D-Polystyrene-2-5-4-5ml-Spectrophotometry/dp/B003UTUPPI
https://www.amazon.com/dp/B0829CD8SN/ref=emc_b_5_t
https://www.amazon.com/dp/B07XZMMNT6/ref=emc_b_5_t
https://www.amazon.com/Quartz-Cuvette-Pack-Spectrophotometer-Transparent/dp/B086GVNK7X/
https://ecuvettes.com/product/qa25-silicone-cuvette-lid-with-center-hole/
Jugs
https://smartbottleinc.com/product/5-3-gallon/Collapsible jug
https://www.uline.com/Product/Detail/S-15648/Jugs/Carboy-2-1-2-Gallon
Food pouches
Peristaltic pumps
Motor Drivers
See CytronTechnologies/MAKER-PI-PICO#13 for full suggestions.
LEDs
Drip Manifold
A drip manifold will combine multiple inlets and have them go to a single outlet; thus, only two ports are needed for the cuvette (see below). There needs to be at minimum 4 inlets, but preferably 5+ so that one can be used for gas (either open atmosphere or pulling from a sealed system). If using a sealed, atmospheric pressure system (i.e. the IV drip bag filled with air or gas), then an additional peristaltic pump is needed. If using a sealed, pressurized system (e.g. gas cylinder), then a solenoid valve and a constant-flow meter/controller (there are some cheap, manually adjustable ones) can be used, but then there needs to be an air outlet (e.g. bubbler) at some point downstream in the system.
Also called flow splitters.
some possible options (need to pick one and ensure appropriate connectors)
A more expensive option would be a multi-channel gradient valve that I think would take the place of the manifold drip and the peristaltic pumps.
Related: https://www.brechtel.com/product/flow-splitters/
Theory
Cuvette Ports
A flow cell cuvette would be ideal, e.g. one from Starna, but these are expensive. I could picture a formlabs 3D printed version using the resin I used for Devanathan electrochemical cells. See this stack exchange post for context. The stopper cuvette seems viable; I could either modify the stopper to add two ports, or purchase a stopper with two ports already. It seems a bit unreasonable to expect having all 5 ports that would be necessary (3-colors, water rinse, waste removal). Mixing the solution can be done by using the waste pump cyclically in short spurts (simpler/easier than a stirring bar with a magnetic stirrer). Reminiscent of AM-ARES' use of a helical path insertion into a sponge to clean a needle tip (see one of the figures in SI of https://doi.org/10.1557/s43577-021-00051-1).
Probably some great things to be learned from C3PU by Cronin group:
(see https://twitter.com/leecronin/status/1578042411061846021).
Maybe worth also asking Huat from the Guzik group (I'm thinking of the electrochemistry setup).
Beta Was this translation helpful? Give feedback.
All reactions