Arsenic Prototype 2

From BioDesign for the Real-World
Jump to: navigation, search


This is the beginnings of the Arsenic detector prototypes - Arsenic Prototype 2 within the full Chronology of the prototypes. This version was developed at EPFL in 2013, and is part of their 2013 Final Report.



To quantify the fluorescence of a sample, it is enough to excite it with the specific absorption wavelength in a dark room, filter the light emitted by the excited sample and then quantify it.

The idea is to use a LED generating light around 488 [nm] which is the eGFP absorption wavelength. The excited eGFP then emits light around 509 [nm]. To assure that the only light which is then quantified is the eGFP one, a filter which stops the 488 [nm] wavelength but let the 509 [nm] wavelength pass is inserted between the sample and the light quantifier.
This principle is shown in this scheme : The device works with an excitatory LED which light at 488[nm] excits the sample's eGFP, which light is filtered and quantified with the photoresistance


A lens could be necessary if the light is not enough intense to be quantified by the light quantifier. It only serves to concentrate the light in one point and so on the light quantifier device.

Another way to do it is to use two filters. One for the light filtration and another dichroic filter. As in the following figure, the dichroic is able to reflects certain wavelength and let pass others. So the idea is to reflects at 45° the excitation wavelength and then let the emitted light pass, filter a second time and then quantify the light.


The light quantifier could be a simple photo-resistance. The system we used was very simple: a photo-resistance was used as the R1 place of a voltage divider so that we can measure the voltage Vout through the arduino's microcontroller’s analogic pin. The photo-resistance, as its name indicates, is a simple electric resistance which resistance change proportionally to the light it receives. More the light increases, more the resistance decreases and more Vout tends to equal Vin. Inversely, more the light decreases, more the resistance increases and more Vout tends to be null.

Representation of the voltage divider connections and the calculation of Vout :


Version 2.0

The first version is build with a photo-resistance.
Cable connexions :

But, the device prototype build like this wasn't able to differentiate samples with different arsenic concentration because the photo-resistance was not enough sensitive to detect the light emitted by the dextran samples whereas they emit more light than the transformed bacteria. So we tried to change the light sensor and use the light to frequency component TSL235 in the version 2.1.

Version 2.1

All the device should then change because of both the code and the cable connexion wasn't the same for this component.
Cable connexions :

The device has been tested with different concentration of the fluorescent molecule dextran. The concentration was always diluted by two and the device always displayed a value which was the half of the previous value. So with the dextran's high fluorescence, we get the first results which confirm that this kind of device could work and which could distinguish different fluorescent intensities.

Another difficulty met, was to find the different components which prices aren't accessible to everyone (a filter can easy be about 300 CHF or more).

Testing and Results

Testing and results can be found here: 2013_Final_Report#Data_and_Analysis