Why Do Mentos React With Soda?




About: I am a sophomore in college studying Computer Science and Electrical Engineering and an alumn of a FIRST robotics team. I also love to tinker on my own, which is mostly inspired by the amazing people on this...

It is well known that when a Mento is dropped into soda, something cool happens. But, what actually happens?

The dissolved CO2 in the soda is constantly looking for a place to accumulate and form a bubble. But, bubbles don’t form spontaneously because the water molecules keep the CO2 molecules trapped in its weak structure (connected by hydrogen bonds).

When a Mento is dropped into the soda, the CO2 molecules gather in small craters on the Mento, which can also be called nucleation sites. When enough CO2 molecules gather in a nucleation site, a bubble forms and is lifted up. Using the balloon, these bubbles can be captured, and the amount of CO2 can be measured.

The following experiment was created to show the effect of the amount of Mentos (hence, more surface area and nucleation sites) on the volume of CO2 released by the soda.

Step 1: Defining the Problem for Investigation

Question:How does the number of Mentos affect the amount/rate of CO2 being released from soda?

Theindependent variablein this experiment is the number of Mentos put into a soda can.

The dependent variable is the amount of CO2 released from the soda and the rate at which it is released, as measured by the width of the balloon containing the CO2.

Hypothesis: If a certain amount of Mentos are dropped into a container of soda, then the container with the most Mentos dropped into it will produce the most CO2 at the fastest rate. This will happen because the dissolved CO2 will provide the most nucleation sites to come out of the solution.


  • Surrounding temperature
  • Type of soda used
  • Amount of soda used per trial
  • Type of Mento used

Step 2: Materials

This experiment should be conducted somewhere where the ground and wall coincide (this allows the metric tape measure to be taped to the wall), and where the temperature is fairly constant.

  • Safety glasses
  • 1 Beaker
  • 1 12-pack (cans) of soda
  • 2 Sticks of Mint Mentos (37.5 g/ 1.35 oz)
  • 3 Balloons
  • 6 Rubber bands
  • Metric tape measure
  • Duct tape
  • Scissors
  • Glass bottle or reusable plastic bottle with a small top
  • Permanent marker
  • Camera (optional)

Step 3: Procedure

  1. Gather materials
  2. Find a location where the floor and a wall meet
  3. Use duct tape to attach the metric tape measure to the wall around 25 cm from the floor
  4. Cut all three balloons 2 cm from the edge (to allow the balloon to fit well over the bottle)
  5. Place the bottle in front of the wall with the tape measure
  6. Measure 100 mL of soda in the beaker (Figure 3.2)
  7. Slowly pour the 100 mL of soda into the bottle and mark the height of the liquid (Figure 3.3/Figure 3.4)
  8. Skip step 9
  9. Pour soda into the bottle up to the line representing 100 mL
  10. Place a mento into the balloon
  11. Optionally begin recording a video to figure out rate
  12. Wrap the balloon around the top of the bottle
  13. Use two rubber bands to secure the balloon to the top of the bottle
  14. Rustle the balloon so the mentos fall into the soda
  15. Wait 20 seconds or until the balloon has stopped inflating, whichever comes first
  16. Record the width of the balloon
  17. Optionally end recording
  18. Dispose of the 100 mL of soda and wash the bottle
  19. Repeat steps 9-18 three times for the different trials
  20. Replace the rubber bands and balloon with new ones
  21. Repeat steps 9-20 with 2 and 3 mentos instead of 1

A video of steps 9-13 can be found above.

Step 4: Data

The Effect of the Number of Mentos on the Amount of CO2 Produced

Trial 1Trial 2Trial 3Average
1 Mento8.0cm6.0cm5.5cm6.5cm*
2 Mentos7.5cm9.0cm9.5cm8.7cm*
3 Mentos9.5cm8.0cm9.5cm9.0cm*

*Note: Averages are in tenths because the data collected is only accurate up to one decimal point

Figure 4.1: Data of biggest width per trial

Figure 4.2: Average of all the trials to show the difference in rate

Figure 4.3: Data table for figuring out rate

Video: All of the trials at the same time

Next two pictures/GIFs: examples from the experiment

Step 5: Conclusion/Validity

It was hypothesized that if a certain amount of Mentos are dropped into a container of soda, then the container with the most Mentos dropped into it will produce the most CO2 at the fastest rate. This hypothesis was supported because 3 Mentos inflated the balloon an average of 9 cm at a rate of .212 cm/sec. This can be compared to 2 Mentos, which only inflated the balloon an average of 8.7 cm at a rate of .202 cm/sec, or 1 Mento, with the balloon inflating an average of 6.5 cm at a rate of .111 cm/sec. It was observed that the difference in rate and width of the balloon was drastically larger between 2 and 3 Mentos than 1 and 2 (as shown in Figure 4.2). This suggests that the amount of dissolved CO2 in the soda is close to being completely used up.

Repeatability: This experiment can be repeated easily due to the understandable procedure, demonstrations, the definition of controlled variables, and a list of materials with amounts included.

Precision: The data of this experiment is reasonably precise. The data clearly showed a trend, but was still fairly scattered (see Reliability for explanation).

Accuracy: The data collected in this experiment can’t be compared to a true value because there is not one for this experiment. But, the trend of the data is supported by other studies of the same subject (See sources for more information).

Reliability: The data is reliable because it is supported by other experiments, the constants were kept, and the procedure was followed closely. A human error in this experiment that lowered the reliability was that the balloon did not make an airtight seal with the bottle (see trial 3 with 1 Mento in the video). This can be solved by using a tighter rubber band or by using two rubber bands (see trial 3 of 3 Mentos). Two additional constant variables that should have been kept are the time waited before recording the data, and the type of balloon used. These would both increase the reliability of the data by raising the precision of the data.

Step 6: Further Studies and Sources

Further Studies/Ideas

Alternate Independent variables:

  • Surface area of Mento
  • Mento vs. other substances
  • Different types of mentos
  • Temperature of surroundings

College physics class' version of this experiment: http://tinyurl.com/sodaAndMentos

Wikipedia: http://tinyurl.com/2nylqz

NewScientist: http://tinyurl.com/k2cgsos

These sources are helpful, but the best way to figure anything out is by experiMENTOtion.

Please VOTE for this instructable if you liked it! Also, this instructable was designed by a 15-year-old, but can be replicated by anyone!



    • Remix Contest

      Remix Contest
    • Sweet Treats Challenge

      Sweet Treats Challenge
    • Faux-Real Contest

      Faux-Real Contest

    20 Discussions


    11 months ago

    Because science!


    1 year ago

    this is cool but I'm only ten I don't know what CO2 means so....................

    1 reply

    3 years ago

    Thank you so much for creating this article, it saved my life when science fair rolled around!

    1 reply

    4 years ago on Introduction

    Nice instructable! Thanks so much for sharing your hard work! Have a super weekend!



    4 years ago on Introduction

    Very nice experiment! From one scientist to another, a couple of useful notes: first, the precision that you get with your averages is based on the total number of significant figures (which you may or may not have experience with). A good explanation of sig. figs. can be found here: http://www.usca.edu/chemistry/genchem/sigfig.htm Second, balloons make for a tricky pressure/volume gauge. They're hard to inflate when they're small, but get easier as they get bigger (to a certain point). A more consistent way to measure might be some kind of piston mounted on top of the bottle. A piece of solid acrylic tube with a syringe plunger that is greased (to minimize friction while maintaining a seal) would be one possible example. All in all, very impressive! It's exciting to see someone taking an interest in scientific experimentation at an early age!

    3 replies

    Reply 4 years ago on Introduction

    Thanks for the suggestion for the syringe on top of the bottle! I've definitely used one to measure volume before in school, but I didn't remember it for this experiment. Also, in regard to your second suggestion, should I add .0 to the end of numbers that only have 1 significant digit?


    Reply 4 years ago on Introduction

    No problem! You should report all the numbers you accurately can, including zeroes to the right of the decimal. So, if you used a ruler with millimeter markings, you can report your measurements to the half millimeter. However, there's significant debate as to how much precision you can "eyeball." A lot of chemists recommend the 3-5-7 rule, where you would approximate between the smallest markings to either .3, .5, or .7 (or .00 if it's exactly on the mark). In short, you should be able to add a zero to the values you originally reported as a single digit. Have fun science-ing!


    Reply 4 years ago on Introduction

    Thanks! I'll change what I can (w/the data I have now), and I will keep this in mind for future experiments!


    Reply 4 years ago on Introduction

    I think the best way to respond to this comment is with a horrible pun. I really hope you didn't MENTO be mean, but some people might think you're SO DArn rude if you keep this up!


    4 years ago on Introduction

    Even though the title is a little misleading ("How does the amount of mentos affect the reaktion with coke?" or something like this would be better) I realy like what you did. You'd definitly get my vote if this becomes an entry to the contest. Right now there's no vote button, you know? ;)

    1 reply

    Reply 4 years ago on Introduction

    You should be able to vote now. I didn't want my title to be too long and I've changed it multiple times (as shown by the url). But thanks for the suggestion!