How can I measure the caffeine content in drinks?
Caffeine is a naturally occurring molecule found in various plants worldwide, and found in beverages such as tea, coffee and soft drinks.

The amount of caffeine present in drinks can be measured by UV spectroscopy.
The full method below, as originally outlined by the Jenway team at Bibby Scientific, can be used to measure caffeine in drinks for quality control, comparison across different brands or as an educational introduction to working with UV-Vis spectroscopy.
Apparatus required;
- Analytical balance
- Weighing boats
- Spatula
- 200ml volumetric flask x1
- Micropipette – ideally 1ml to 10ml size
- Compatible Pipette tips
- 50ml volumetric flasks x6
- Hotplate stirrer
- Stirrer bar
- 250ml beakers (one per hot drink sample to be tested)
- 100ml measuring cylinder
- Separating funnel
- Retort stand with separating funnel ring
- 250ml Conical flasks (one per sample to be tested)
- Dropper pipettes
- Quartz cuvette with 10mm path length
- UV range Spectrophotometer (with concentration mode preferred)
Reagents required;
- Caffeine (pure) for standard preparation
- Purified water
- Dichloromethane
PPE;
Method;
- Prepare a standard 1,000 ppm stock solution of caffeine; use an analytical balance to weigh 198.2mg of caffeine and make this up to 200ml using purified water in a volumetric flask.
- Prepare calibration standards; using a pipette add 25ml, 12.5ml, 10ml, 7.5ml, 5ml and 2.5ml to each of 6x 50ml volumetric flasks. Make the standards up to the 50ml volume using purified water. These amounts will create 100ppm, 50ppm, 40ppm, 30ppm, 20ppm and 10ppm calibration standards respectively (ppm = mg/L)
- Prepare the samples;
- For instant coffee; add 2g of granules to a 250ml beaker and add 200ml boiling purified water. Stir at 500 rpm on a magnetic stirrer for 30 seconds, then leave to cool to room temperature without further stirring.
- For tea; add 3.2g of dried tea leaves to a 250ml beaker and add 200ml boiling purified water. Stir at 500 rpm on a magnetic stirrer for 30 seconds, then leave to cool to room temperature without further stirring.
- For pre made soft drinks (e.g. cola, energy drinks) no preparation is required, skip ahead to step 4
- Extract caffeine from the samples; take 50ml of the calibration standard or sample and it to a separating funnel. Use the measuring cylinder to add 25ml of dichloromethane.Invert the separating funnel 3 times, then vent to avoid pressure build-up. Put the funnel in a stand and allow the layers to separate, before removing the dichloromethane layer to a labelled conical flask.Return the calibration standard or sample to the separating funnel and repeat twice more, until 3x 25ml dichloromethane layers have been combined in the conical flask.
- Measure the calibration curve; use a dropper pipette to add the calibration standards to the quartz cuvette for measurement.First measure purified water only as a blank, then measure each of the calibration standards in increasing order of concentration.Tabulate the results of caffeine concentration in ppm vs. absorbance at 260nm.
- Calculate the calibration curve; use a spreadsheet to create a line graph of the calibration curve results. Find the linear regression equation of the calibration curve, y = mx +c (where y = absorbance and x = concentration)
- Measure the samples; use a dropper pipette to add your first sample to the (cleaned and dried) cuvette. Take a measurement and record the absorbance at 260nm. Repeat for each sample, taking care to clean and dry the cuvette carefully between samples.
- Calculate the results; using the y = mx + c equation from your calibration graph, you can calculate the caffeine concentration of your samples. Substitute Y for the absorbance value recorded for that sample, keep M and C constant, and rearrange to solve for X.For some models of spectrophotometer, such as the Jenway 7305, a concentration mode is available which allows the instrument to do this calculation for you based on your y = mx + c equation, so the readout on the spectrophotometer will be in ppm directly.
For more information on this test and the equipment needed to perform it, just follow the links above or contact us with any questions;
Dear Sir,
our factory is looking for a lab device to measure the caffeine in coffee after it is brewed. how can i receive the details for your device? how much is the price of UV Spectroscopy?
Hi Samira,
Thanks for your question – I have emailed you with some more details.
Kind regards, Camlab Support Team
Hi – is there such a device as a ‘caffeine meter’ or some simple tool that can measure the amount of caffeine in any drink?
Thank you!
Bob
Hi Bob,
Good question – the method we’ve described here is intended to give high accuracy results in a lab environment.
There are some test strips available on the market which simply state “yes” or “no” but don’t quantify the amount of caffeine present.
There’s a mixed review here; http://www.popularmechanics.com/technology/gadgets/a12771/4309061/ I don’t know of a UK supplier of these but you may be able to
find them online.
Other than this we couldn’t see any quick tests, but you may have better luck in searching!
Camlab Support Team
What other chemicals could interfere at 272nm? Could these give inflated results?
Hi Sebastien,
Good question – the measurement at 272nm is suggested as standard in the literature, which is why we’ve recommended it here.
If you are concerned about interferences some sources get around this with calculation steps (e.g. Belay et al here; http://www.sciencedirect.com/science/article/pii/S0308814607010308 “Gaussian fit was applied to eliminate the possible interference with the caffeine spectra.”) or alternatively consult one of the UV data atlases to see what other compounds would be found at this wavelength, and see if any of them are likely to occur in your samples (list of good sources here; http://webbook.nist.gov/chemistry/uv-vis/ )
I hope this helps, and best of luck with your research!
Kind regards
Camlab Support Team
Dear Sohie
Im starting a coffee brewery…. And we are looking for a lab device to measure the caffeine in coffee after it is brewed.
Can you send me the details for your device? how much is the price of UV Spectroscopy?
Hi Sophie
Can you send me the details for your device? how much is the price of UV Spectroscopy?
Thanks
Gerard
Hi Gerard,
You can see the links to all the products in the above blog post, the Jenway 7305 UV/VIS is currently £2,600 ex VAT which you can see on our website: https://www.camlab.co.uk/jenway-7305-uvvisible-spectrophotometer-p12542.aspx
If you would like any further information please get in touch sales@camlab.co.uk
Kind regards,
Emma
Pretty interesting, Sophie! I wonder, is there perhaps a video on YouTube showing this process? I’d love to see how caffeine is measured using this method here.
Hi Margaret
There are many videos on youTube relating to caffeine determinations.
https://www.youtube.com/watch?v=Xzh-6ZDitQ8
Hope this helps
Camlab tech Support
Hi,
What is the concentration of dichloromethane that you used for this experiment?
Thanks,
Max
Hi Max,
Thank you for commenting on the Camlab Blog, you can see the Dichlorimethane A.R. on our website here: https://www.camlab.co.uk/dichloromethane-ar-p14675.aspx
Kind regards,
Emma
Camlab Tech Support
Hi ,
I just have a few questions about the experiment. Do you choose to do either the calibration sample or the calibration standard or do you do both? Because I don’t get how you use both and if it’s only one I don’t get why you would use the standard over the sample? And do you need a colorimeter to record the absorbance because I don’t get how to record it without one?
Hi,
I would like to conduct this experiment in my school, but they consider dichloromethane dangerous. Is there a safer substance that can be used in its place?
Hi Mary,
Thank you for contacting us. I have had a look at a few papers detailing methods and they all seem to list this as the means to extract the caffeine. Sorry we can’t recommend a safer alternative!
Kind regards,
Camlab Tech Support
Hi, could you by any chance give me the results of the calibration curve at least since i’m not sure i’ve done it correctly.
Hi Farah,
Unfortunately your calibration curve would depend on your method, how good your extraction is etc. We no longer have a copy of ours for you to compare.
kind regards,
Sarah, Tech Support