I have adapted the lab experiment described in this J. Chem. Ed. article for the organic chemistry lab course I teach. In the past year, the first year of implementation, I have had 67 failures (100% failure rate) on the extraction part of this experiment.

The experiment involves the isolation/separation of the active components in Excedrin(R): acetylsalicylic acid, acetaminophen, and caffeine.

Acetylsalicylic acid, acetaminophen, and caffeine

The experiment is in two parts. Part 1 involves analysis by TLC and separation by column chromatography. This part works beautifully. Part 2 involves separation by successive extractions with increasingly strong bases. Acetylsalicylic acid was extracted with 10% aqueous Na2CO3. Then, acetaminophen was extracted with 3 M NaOH. Finally, caffeine was recovered by evaporation of the remaining organic layer.

In Part 2 We recovered lots of acetylsalicylic acid and caffeine, but no-one recovered any acetaminophen. Below is a list of things I tried in order to improve the results. None of them worked. I know from TLC analysis of the initial organic solution that acetaminophen was present. What happened to it?

Things I have tried:

  1. Changing the concentration of both bases (up and down).
  2. Using a weaker base for the first extraction (like NaHCO3 or Na3PO4).
  3. Increasing the HCl concentration used to work up the second extraction.
  4. No acetaminophen was extracted with the weak base, nor was any left behind after the second extraction.

Where has my acetaminophen gone?

  • 1
    $\begingroup$ How much of the total quantity of each did you recover? Was it a 90% yield for ASA and Caffeine or something like 40%? $\endgroup$
    – CHM
    Jun 27, 2012 at 23:50
  • $\begingroup$ Since these were undergraduates doing extractions for the first time, the recoveries of ASA ranged from 50-90%. Recovery of caffeine was lower, like 20-50%, but that matches the results from the paper I linked. Acetaminophen was universally 0% recovery. $\endgroup$
    – Ben Norris
    Jun 28, 2012 at 0:09
  • $\begingroup$ I've provided a partial answer below describing a ridiculously easy procedure. It can even be done at home (it's a classic drug addict's/poor man's extraction of opiates from OTC drugs)! $\endgroup$
    – CHM
    Jun 28, 2012 at 0:28
  • 1
    $\begingroup$ they note that you should use KOH and not NaOH. $\endgroup$
    – Chris
    Jun 29, 2012 at 5:01
  • $\begingroup$ @Chris They do note that. I will be trying that variation this coming semester and if it works, I will provide that as an answer to my own question. $\endgroup$
    – Ben Norris
    Aug 16, 2012 at 13:37

3 Answers 3


Did you notice any change of colour (e.g. to brown) and/or the formation of gunk upon adding the 3M NaOH?

I'd assume one would have to boil it, but is the following scenario conceivable under your conditions:

  1. alkaline hydrolysis of the acetaminophen (paracetamol) to p-aminophenol
  2. aerial oxidation to the corresponding quinone imine
  3. partial polymerisation of the latter

You have confirmed that acetaminophen was initially there. A a change in the constitution of Excedrine since the publication of the article can thus be excluded. Isn't a reaction (hydrolysis or whatever) under your extraction not the most likely explanation for the vanished acetaminophen?

  • 1
    $\begingroup$ This is the closest to what I figure is happening. When we switched to KOH, the results improved, but not nearly to what they should be. The potassium salt of acetaminophen is apparently more soluble in water, and thus the extraction is more efficient. The sodium salt comes out of solution as a precipitate before redissolving (presumably due to hydrolysis). $\endgroup$
    – Ben Norris
    Feb 22, 2014 at 13:58

Here's what I suggest:

  1. Mash up the pills in a mortar
  2. Dissolve the powder in cold distilled water, use a beaker
  3. Shake well
  4. Put the beaker in a freezer and wait until right before the water freezes to remove it
  5. Carefully decant the solution
  6. Check what's left both in the water and the powder residue

I don't know exactly how much water is optimal for the separation, that you will have to figure out.

The powder should contain acetaminophen.

  • $\begingroup$ This is a neat technique. However, the only component in Excedrin that has appreciable water solubility is caffeine. Acetaminophen is less soluble, and ASA is basically insoluble in cold water. $\endgroup$
    – Ben Norris
    Jun 28, 2012 at 0:45
  • $\begingroup$ Also, I was more hoping for a way that I could fix or salvage the original experiment. $\endgroup$
    – Ben Norris
    Jun 28, 2012 at 0:57
  • $\begingroup$ @BenNorris yes, ASA and acetaminophen will both stay out of solution. How do you measure the presence of each component? FTIR? $\endgroup$
    – CHM
    Jun 28, 2012 at 1:12
  • $\begingroup$ TLC against standards. FTIR with 16 students competing for the same instrument in one hour is a little tricky. $\endgroup$
    – Ben Norris
    Jun 28, 2012 at 2:05
  • 1
    $\begingroup$ Actually, we have a pretty nice Nicolet Avatar with an ATR module. The limit isn't the speed of the technique, but the speed of an undergraduate student taking his or her second IR spectrum ever. $\endgroup$
    – Ben Norris
    Jun 28, 2012 at 11:02

You could fix/salvage the experiment by first extracting the acetaminophen with DCM, and then perform your base extraction.. although it deviates a bit.

Here's an example lab: http://www.odinity.com/separation-of-the-components-of-an-analgesic-tablet/


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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