HI, I think you are going to keep having problems if you are trying to read very small DC values.
My approach would be to modulate the light source and use a optical semiconductor detector. The receiver detects the modulated light, you filter and measure its amplitude.
That way you filter out ambient light and electrical noise that can enter the system.
Most systems use scatter methods; https://www.fondriest.com/environmental-measurements/measurements/measuring-water-quality/turbidity-sensors-meters-and-methods/
Surely you have resources there at the University that has a chemistry or analysis department that has one of these devices or can supply info about them.