I'm into mixology, and lately I've been looking into experimenting with home-made tonic water, by extracting quinine (quinine salts) from cinchona barks in an acidic aqueous solution. I've read a bit about the potential dangers of quinine overdose, so I want to make sure I know how much the amount of quinine is in my extract.

I know it is fluorescent under UV; I thought maybe there might be a minimum quinine-to-water ratio for fluorescence. If this is the case, I can dilute the solution gradually until the solution is no longer fluorescent. However my knowledge on this subject is fairly limited.

Is there a home-accessible way of measuring the amount of quinine in an aqueous solution?


Without standard solutions and a fluorimeter, you can do a side-by-side comparison using a commercial tonic water and your solution. I just prevailed in getting a poor photograph of the beautiful blue fluorescence from a commercial tonic water (begins with an S), using 405 nm excitation (BluRay laser diode). The photo is below:

Tonic water fluorescence

The two cuvettes contain tonic water (undiluted on the right) and diluted about 5 times (on the left). The laser entered from the right. The bubbles must be removed because they cause significant light scatter that gives red artifacts in the camera. By naked eye, it actually looks lots better: no red artifacts.

Removing the dissolved gas, to avoid bubble formation, is often done done by using an ultrasonic bath, sparging or by simply shaking the bottle repeatedly and waiting. Adding NaCl is a no-go because it quenches quinine fluorescence (a nice Stern-Volmer experiment).

Fluorescence is very sensitive, but the standard joke is that only 10% of all substances fluoresce ... and they are the wrong 10%. But this is a good case!

So, you would simply A-B the solutions: make dilutions of the commercial tonic water and see which one most closely matches your home-made tonic water. Cheers!

Note: you can use near-UV LEDs, i.e., 395 or so nm peak emission wavelength. The absorption max for quinine sulfate in dilute sulfuric acid is around 350 nm, and the peak emission wavelength is around 460 nm. But the photo shows that 405 nm works fine: it is not always necessary to be at the optimum excitation wavelength. Driving an LED is trivially simple: just a DC voltage supply, such as a battery or 'wall wart', and a current limiting resistor. Or use a constant current source that can output 20 or so mA.

One more thing: if you dilute the tonic water sufficiently, you will not be able to see the fluorescence with the naked eye. But a fluorimeter can detect it at even lower concentrations. Eventually, with sufficient dilution, you reach the limit of detection. But that is way below whatever you might want in usable tonic water.

  • 1
    $\begingroup$ Years ago, I bought 100 UV LEDs, with 397 nm nominal peak emission wavelength. I only used one, for a paper I published, so I have 99 left. I am thinking it would be fun to use two more to illustrate how to do the A-B comparison. Stay tuned! $\endgroup$ – Ed V Mar 30 '20 at 3:04
  • $\begingroup$ Amazing images! I am always amazed by fluorescence. If you can find a test for checking olive oils with your UV, that would be great! They fluoresce in red color (the good ones). I wrote something for The Physics Teacher on olive oil fluorescence demos but could not investigate more. $\endgroup$ – M. Farooq Mar 30 '20 at 3:39
  • $\begingroup$ The OP can do ImageJ analysis and construct a calibration curve with intensity after taking pictures like yours. ImageJ is free of cost. This is certainly home doable :-). He just needs one standard of quinine. $\endgroup$ – M. Farooq Mar 30 '20 at 3:41
  • 1
    $\begingroup$ Ah, chlorophyll! Should have guessed. The hole burning was entirely by surprise: red and black balloons simply pop, but blue balloons get drilled and the edges of the tiny hole are essentially ‘cauterized’, so the balloons ‘bleed out’, as it were. $\endgroup$ – Ed V Mar 30 '20 at 3:57
  • 1
    $\begingroup$ You've also mentioned the fluorescence quenching property of NaCl, can't that be also be exploited to measure the amount? Great answer btw. $\endgroup$ – zetaprime Mar 30 '20 at 7:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.