[devindotcom]

The processes involved are orders of magnitude below the scale needed to produce usable light. And in the plants it's been tested in, it's also fatal after a short time. These aren't trivial problems, and they're not going to be solved by a designer photoshopping a concept.

[roywiggins]

Reminds me of a "design concept" for a little Nest-shaped box to stick on your window and magically cancel out all street noise. I'm pretty sure this is actually impossible, and is the equivalent of drawing a pegasus and saying "I have this great concept for a novel form of transportation!" i.e. more or less not worth anyone's time, since it's fantasy.

[chillacy]

Possible in a controlled lab environment if you can destructively interfere the sound, but it's gonna depend on the sounds as well as knowing the location of the listener

[roywiggins]

Here it is, for reference. I could imagine it possible in the lab, and I'm not an audio engineer but the whole thing is completely implausible.

And sites like Gizmodo don't bring any critical thinking and just repeat the claims verbatim.

http://gizmodo.com/a-stick-on-speaker-that-uses-your-windows...

[teekert]

Exactly, I once did the calculation, they use the protein luciferase, which requires 1 molecule of Lucferin and 1 molecule of ATP per photon produced! That means if you were to produce say 200 lm (about the output of a 40W incandescent light bulb) you'd need kilo's of ATP and Luciferin a week. Don't expect useful light.

[saalweachter]

Doing plausibility calculations from another direction--

Let's say that we want to have a glowing forest. The glowing trees will absorb sunlight from a certain area and emit it over the same area with a certain efficiency.

Photosynthesis has an efficiency of between 0.1% and 1%. I don't know what the efficiency of glowing is, so let's assume it's 10%. So your plants will illuminate the area with about 0.01%-0.1% of the intensity of the light which they absorb.

Full daylight is about 100,000 lux, so your forest will be illuminated to somewhere around 10-100 lux, which is somewhere between "twilight" and "a poorly lit room".

It gets even worse if you want your plants to illuminate a larger area than they themselves cover: if you have a plant illuminating (say) 10x more area than it covers (a tree in the middle of an open field, or next to a road), you're down to 1-10 lux, optimistically.

[bduerst]

Well, 1 kg of ATP is about 2 moles, and since this is ATP->AMP (10.9 kcal/mol) that's about 21.8 kcal, ignoring other upkeep.

Basically 1kg ATP is 21.8 kilocalories

[Dylan16807]

Only kilos? Compared to bamboo that actually sounds feasible.

[DrJosiah]

(warning: I am not a biologist)

The problem is that generating ATP requires sun or sugar in plants (depending on the specific metabolic process involved), and that the total amount of sugar produced during the day as part of photosynthesis is tiny relative to the amount of ATP necessary to fuel the light at night.

Consider that humans burn 100-150 kilos of ATP daily (according to Wikipedia). This is only possible because we are recycling ATP continuously, fueling the recycling process with sugar and oxygen that we consume at rates several orders of magnitude higher than what a 20 year old tree could produce in the same period.

I can't find numbers for peak sugar production in plants, but considering how many maple trees it takes to make a single small container of maple syrup, I think this is a reasonable statement to make for now. I will stand corrected if someone has good numbers. :)

[Dylan16807]

Well apparently a tree can permanently sequester "up to 48 pounds" of carbon per year (numbers are unclear, some mix up carbon and CO2, bamboo is more, but it makes a good ballpark max). Let's assume a tree that glows instead of growing for simplicity. 48 pounds, times 30/6 to get the number of ATP produced per carbon atom, times 507/12 to account for how much heavier ATP is than carbon, comes out to 12.6 kilograms of recycled ATP per day.

[DrJosiah]

One order of magnitude, I stand corrected :)

[titanomachy]

Wouldn't it be simpler to just genetically engineer humans to have better night vision? :P