I am looking to build an Arduino (Rev3) circuit that can detect a specific chemical in someone's body by shining a light through their skin to detect the distinct spectral signature of the chemical. I want the circuit to operate in the near-infrared range and the chemical has a peak absorbance at 1445nm.
My idea is to use an IR LED and a sensor photodiode (linked below):
The IR LED would shine through the skin and if the chemical is present, it would absorb the 1445nm wavelength. The IR Photodiode would receive the reflected spectra back and if the 1445nm wavelength is diminished, it would be able to detect the presence of the chemical.
I am relatively new to Arduino, so how can I put these components together into a circuit? What resistors should I use and are there any other components I need? How might the Arduino code look for this project? Also, is there a better way to implement this "chemical-detecting" idea?
What is your plan to make wavelength-dependent measurements? The photodiode you linked responds over the entire range of 900nm to 1700nm, and the LED emission bandwidth is 150 nm.
Researchers have been working hard on related approaches for several decades, in particular to noninvasively measure blood glucose concentration. So far, no practical solution has emerged, although some report success when each person is individually calibrated.
You might find reading about that work and the problems encountered informative, for example, this recent review:
Those parts you have chosen are woefully inadequacy for such a task.
The LED only transmits at one wavelength and the sensor only receives at one wavelength, so it is impossible to use them for what you state you want to do.
Not true. Both are comparatively broad band. See post #2 for the details.
Filters or a diffraction grating are required to resolve the wavelength of interest.
It is true that they are totally useless for the stated intention.
If you want to do this then these three sensors boards are what you need. Each one is expensive but dirt cheap compared to the professional alternatives.
@firsthero45 To save you from wasting your time, note that the sensor recommended and linked in post #5 doesn't cover the wavelengths of interest for your project, by a long shot.
It's very similar to what I want to do but with glucose. I've thought of using filters (similar to what astronomers use for spectroscopy) but they seem to be a bit too expensive. Also, the project linked above seems to have the same setup I'm thinking of which doesn't use any filters. The project above uses a capacitor and an op-amp, but I still don't know how to set these up with the components I listed or if they're even necessary.
Whether a filter or diffraction grating is required depends on what other substances are present in the sample, their concentrations and their absorbance bands.
To determine that for your case, look up reports on studies of near IR transmission through human tissue at 1445 nm. (Hint: the prospects of success are very low).
Use of near IR spectroscopy equipment that is suitable for the project may also required.
@firsthero45
Referring to the paper cited in post #9
For the LED circuit, I would use a 180Ω resistor.
Don't use the photodiode circuit shown in that paper, it will destroy your expensive photodiode.
It is old but it has been cited many times including by The Driver Alcohol Detection System for Safety (DADSS) Research Program which is what gave me the idea for my personal project in the first place. As can be seen in Fig 5, there is less interference from the human skin after the 1100nm mark.
As for the chemical I am trying to detect, it is not naturally occurring in the human body and so I still don't think I'll need diffraction grating.
@TomGeorge
I am a mechanical engineering student in my second-year of college, and this project is a personal exploratory project. I have beginner experience using Arduino for basic projects (LED blink circuit, LED on/off button, etc). I am still having trouble getting from the "finding the parts I want" point to the "actually making a circuit out of them that doesn't wreck the parts" point.
Well when you see that the procedure that you are following includes instructions that will destroy your parts, you should stop and consider the veracity of your source.
You're in a case of a blind man following another blind man.
There's an awful lot of sham science out there to let yourself get stuck on something just because a few pitiful scientists published papers on something.
