Glucose Powered Fuel Cell

MTS

Has anyone gotten a chance to play around with a bio battery? I'm super interested in developing something that could take advantage of using blood sugar as a power source in the future..

bciuser

So it sounds like a cool idea, but I suggest you first do some math to figure out in an ideal world how much energy such a battery could harvest. 

If you're getting energy from glucose, you need to break it down into CO2 and H2O (combustion). This is what your body does with it in a step-by-step process. So to run something small like a microcontroller with some LEDs let's say we need 100mW of power. This means we need to generate 0.1 Joules of energy per second. Since glucose has an enthalpy of combustion (how much energy you get from burning it) of -2805kJ per mol, you would need:

-2805kJ * 1000J/kJ * X mols glucose * 1 second = 0.1 Joules per second

You need approximately 3.6 * 10^-8 mols of glucose per second. Assuming human blood is an 0.01M glucose solution, and your device has some way of filtering glucose from blood, you would need to filter out approximately

3.6 * 10^-8 mols per second / 0.01 mols per liter = 3.6 * 10^-5 Liters per second (0.036mL) if you harvested all the glucose. Or you could take only 1% of the glucose out of 10mL of blood. This seems reasonable. 

However, I should point out for 100mW of power and completely ideal conditions you would use approximately one third of the glucose present in the blood initially over a 24-hour period, so I'm doubtful this could actually be used to power an implant.

ElectricFeel

Bit of a sidetrack, but if you had some mechanism to filter and use glucose from the blood that would be an interesting thing to apply to the needs of diabetics under hyperglycemic stresses.

MTS

@bciuser thanks for your input. You're right, we would be robbing our bodies of some much needed glucose which cant be safe.

@electricfeel hmm, that would be a great application. Having a device that could not only monitor blood sugar levels, but in case of a hyperglycemic emergency, "take glucose out", would be one hell of a safety net.

MTS

@electricfeel I wonder if it would be possible to reverse the process, and use glucose stored in the bio fuel cell to inject glucose back into the bloodstream if blood sugar levels became too low..

chironex

The celll itself isn't very complicated. the problem arises when you notice it requires vertically aligned nanotubes and platinum to work. So have fun sourcing that. oh and you need to be able to fabricate microchips. And you can't just buy one since they'll be ungodly expensive. Also you don't necessarily break glucose into CO2. Most of it runs on just stealing an electron or two from glucose if memory serves

bciuser

@MTS that has already been done by the company Medtronics (albeit they inject insulin not glucose directly).

MTS

@chironex that sounds like one bad trip.

ElectricFeel

I can get access to my school's silicon fab if I create a proposal that looks good enough for the university to let me.

MTS

@bciuser yeah Ive looked into insulin pumps, interesting stuff. I was just wondering if it were possible to recycle glucose, read the blood sugar levels, and if sugar levels are too low, feed the glucose back into the bloodstream and hopefully makes it into the right cells. But kf course this just an idea and not even a well thought through concept

MTS

@electricfeel I think it would be worth a shot, we can start by creating a detailed concept if you want.

MTS

@electricfeel the device would be simpler if it wasnt a glucose fuel cell too so we could make two different concepts:

1. Monitors blood sugar levels, and also:
a. extracts glucose from the blood when levels get too high
b. uses glucose to power the device
c. can release recycled glucose back into the bloodstream when levels get too low

2. Monitors blood sugar levels, and also:
a. extracts glucose from the blood when levels get too high
b. does not use glucose for energy
c. but can still release recycled glucose back into the bloodstream when levels get too low

glims

If you had a switchable surface (changes molecular configuration via ph or other factors, not as hard as it sounds), that locked the glucose in a polymer brush or a hydrogel, and then released it upon switch, that would be your simplest, least fail prone way to do that, imho.

The problem was, that it would only work as long as there was regulation from outside the system. Think about it like if you have a way to hold glucose, your bucket so to speak. Using my method, or any method that follow the parameters that you put forth, there is a finite amount of glucose that you can trap. This means that it only works as long as one expects constant and regular fluctuations from really low to really high. Or am I missing something.

MTS

@glims spot on. the thing is we would probably design it for older humans, being that diabetes gets harder to control the more someone ages. Also, every diabetic has spikes after eating, and even hough they may be small, the device may be able to accumulate a considerable amount of glucose (what ever that may be), over time. But this is more of an emergency device for managing less frequent spikes.

Cassox

The endogenous system for doing is well understood. Is there a reason you prefer a device mediated solution? Look up glycogenin if you get a chance. There are a lot of ways to mediate blood sugar other than insulin alone.

Cassox

A molecule that bonds glucose in a way like hgb binds o2 would be cool. Basically, hold if the local concentration is high. Release if low.

glims

That's basically what im describing,  polymerized and switchable. Polymerized so it is stable and doesn't get broken down, and switchable so that is it binds and releases. If you just had a molecule, it will filter and or pass through your system. And it needs switchability (making up words) to bind and release based on the balance in the system.

So if you took the binding ammino sequence of glycogenin, and made a polymerized protein brush from that... Im not ready to look that up right now, but you add a secondary chain to that you can make it switch... This is your hard part. If your getting ph shifts in blood your fucked, so you need to make it respond to a secondary stimulant that isn't ph.  it's a start. 

Any ChemE undergrads up in here?

The trigger is the hard part....

MTS

@Cassox Ahh, thats very smart. converting glucose to glycogen would be much more efficient.

@Glims I see what both of you are saying.

I'm gonna research some more