3
$\begingroup$

As part of a project on decaffeination methods, I'm looking to see how activated carbon filters can selectively adsorb caffeine in tea and coffee solutions, and avoid adsorbing other taste-giving components.

The technique here is water decaffeination, and involves soaking tea leaves/ coffee beans in hot water, and using a carbon filter to adsorb the caffeine, thereby decaffeinating the solution. However, I can't seem to find any documents relating to the selectivity of this process : I don't see why the activated carbon wouldn't adsorb any flavourful compounds too.

The only mention to this problem I've found is on The science behind decaf, which states :

The problem is that the extract also contains hundreds of other compounds that are critical to coffee flavor, many of which would also be removed by the carbon filter. But there is a way around this. Preload the filter with chemicals that are unlike caffeine in molecular structure, but similar to other flavour and colour compounds found in coffee. Sugar and formic acid are the chemicals used. So obviously the process isn’t “chemical-free.”. Since the activated carbon’s adsorption sites for such chemicals are now occupied, the filter will not remove coffee components other than caffeine.

Sadly, the webpage doesn't cite any sources, leaving me wondering why formic acid and sugar can increase adorption selectivity.

I'm wondering if any sympathetic coffee enthusiast can explain this to me, or cite some documents or studies that delve into this problem, how to make activated carbon only adorb caffeine ? Any link or website would be appreciated.

Improve this question
$\endgroup$
4
  • 3
    $\begingroup$ The problem is, coffee and tea contains many compounds less polar than caffeine, that would adsorb themselves better. Activated carbon IMHO is not the way to go. $\endgroup$
    – Poutnik
    Commented Jun 9, 2021 at 20:43
  • $\begingroup$ My best guess is that activated charcoal has holes of different sizes and the various solutions used to block everything but caffeine actually occupy the holes that don't interact with caffeine. $\endgroup$
    – Andie
    Commented Nov 10, 2021 at 15:25
  • $\begingroup$ While the idea of using activated carbon might be possible the issue is whether it is better and cheaper than alternative methods which are widely used and seem to work well (both methylene chloride and supercritical carbon dioxide have been used in processes that selectively extract caffeine). Solutions involving activated carbon seem more elaborate and expensive to me. $\endgroup$
    – matt_black
    Commented Nov 12, 2021 at 17:15
  • $\begingroup$ The method is currently described in (nearly) complete detail in the current Wikipedia blog for "Decaffeination" (January 2024). A "flat" "filter" is saturated with other substances, allegedly preventing absorption of non-caffiene molecules. $\endgroup$
    – Raven
    Commented Jan 27 at 22:00

1 Answer 1

Reset to default
1
$\begingroup$

The Swiss Water process uses hot aqueous solution (see below) to extract caffeine from green beans, and activated charcoal to remove the caffeine from the extract. This decaffeinated solution is used again for the next extraction (minimizing the leaching out of flavor components). See e.g. the "SWISS WATER® Process animation video" https://youtu.be/tAEQ4G-1jTQ

This only works if the flavor compounds are not absorbed by the activated charcoal (otherwise decaffeinated coffee would taste really bad, not just somewhat bad). There is a patent[1] that suggests that pretreating the activated charcoal can suppress flavor absorption, https://patents.google.com/patent/US5208056A/en. Patents are often written in a way that includes the method actually used, but also includes other possible methods (that might not work). This is to protect intellectual property without giving a specific recipe others could use after the patent runs out. This patent talks of using sucrose for pretreating the active charcoal. I did not find any information on using formic acid in a similar manner.

[1] U.S. Patent #5,208,056 issued May 4, 1993 to Fischer and Kummer (Chocolat Suchard Societe)

Improve this answer
$\endgroup$
0

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.