Public Lab Research note

Extractions from Grapes with Alcohols

by iamkat | October 06, 2016 15:58 06 Oct 15:58 | #13529 | #13529

What I want to do

Attempting to recreate this protocol (from UC Davis precis of American Journal of Enology and Viticulture. 58(4):451-460. 2007), which looks at extraction of polyphenols from grapes. Aim to test identification and quantification using UV spectra.

My attempt and results

Test: lab-grade ethanol 99% and Boots surgical spirit (95% ethanol, 5% caster oil)

NB: difference from linked protocol - after fiddling with acidification and finding it difficult to raise the pH, I went ahead without acetic acid addition. Solvent 50% solution ethanol or surgical spirit.

  • Mash red grapes in pestle and mortar (seeded grapes unavailable, seedless used)
  • Measure 1:1 grape pulp: solvent
  • Leave stirring for 5 minutes
  • Filter through grade 1 filter paper
  • take solid and repeat with new solvent (keep extract liquids separate)
  • run spectral analysis immediately as the extracts degrade


First Wash GrapesW1Eth1.jpg


Second Wash GrapesW2Eth1.jpg


The outputs look like anthocyanin peaks at around 500nm and perhaps some baicalein at the lower end of the spectrum. The Desktop Spectrometer V3 spectrum doesn't register the lower peaks.

Surgical Spirit

At first the spectrum didn't work well and had a lot of noise, probably due to high opacity due to colloid from the caster oil. I centrifuged for 5 minutes and removed the top layer to run the spectra again.


The lab UV spectrum falls down at the lower nm due to some interference. Probably from the caster oil contamination. This region is lost in the Desktop Spectrometer V3 spectrum below.


Questions and next steps

  • Can I purify the surgical spirit safely so that the UV spectrum is clearer?
  • Can I use a UV light to measure the spectrum below 300 nm? Is there a facility for this on

Why I'm interested

To generate reliable protocols that can be used to answer questions about the lycopene content in food with valid results, outside of a lab.

I'm developing these protocols as part of a much larger project looking at a multiple knowledges / subversive knowledge production approach to food, supported by the London Creative Network and UCL. The first output will be an undergraduate course in January 2017 that tests these approaches with a London community.


The sensors in consumer cameras have very low sensitivity at wavelengths shorter that 400 nm. Some cameras are slightly sensitive down to about 380 nm when the IR cut filter is removed. Most UV light is blocked by standard glass, so that contributes to poor camera sensitivity in the UV.


Reply to this comment...

Also - I noticed you're using a compact fluorescent lamp as a light source - but are you measuring absorption? In that case you probably want a broad spectrum light source. May I ask what kind of spectrometer(s) you're using, and could you share more about your experimental setup?

Is this a question? Click here to post it to the Questions page.

Reply to this comment...

Login to comment.