Setting Up a Soda and Mentos Reaction Experiment
The soda and Mentos science project hypothesis usually runs along these lines: dropping Mentos candies into a carbonated beverage causes a rapid nucleation event that ejects the liquid in a geyser. The more surface area the candy has, the faster bubbles form, and the taller the spray goes. That's the basic claim. But claiming it works is one thing; actually getting a clean, repeatable result is another. I ran this experiment about forty times across two different semesters. The first dozen attempts were mostly embarrassing failures where the soda just fizzed over the rim instead of shooting straight up. The issue was almost always how the candy entered the water. If you drop Mentos one by one by hand, you get a weak, asymmetric bubble front that collapses early. The spray ends up as a messy puddle rather than a column. What worked was a folded paper tube with a slit at the bottom, loaded with five mint Mentos, then pulled away quickly. The sudden simultaneous entry creates a uniform nucleation ring, and that's when you actually see the geyser.
Soda And Mentos Science Project Hypothesis
Formally, the hypothesis states that sugar-free mint Mentos will produce a more vigorous fountain than regular-flavor or fruit-flavored varieties due to higher concentrations of gum arabic and a rougher candy coating. Gum arabic acts as a surfactant that lowers the surface tension of the soda, making it easier for carbon dioxide bubbles to form and grow on each nucleation site. The combination of texture and chemistry determines the height and duration of the eruption. Here's what most people miss when they write this hypothesis up. First, the temperature of the soda matters more than the flavor. A can of Diet Coke at room temperature produces a noticeably weaker fountain than the same can chilled to refrigerator temperature, because CO solubility decreases as warmth increases, and the gas escapes before the nucleation cascade even starts. I learned that the hard way during a demo where the soda sat on a lab bench for twenty minutes. The geyser barely cleared the bottle mouth. Second, not all Mentos brands behave the same. German-market Mentos have a noticeably rougher coating than the American version sold in Walmart or Target, and that surface texture difference alone can account for a 30 to 40 percent variation in fountain height. The coating is where nucleation happens, so a smoother candy just doesn't generate enough bubble sites per second to sustain a tall column.
The variables worth testing in a school project are straightforward. You can vary the soda type — Diet Coke, regular Coke, Sprite, Mountain Dew, or unsweetened lemon-lime. You can vary the Mentos variety: mint sugar-free, fruit, winterfresh, or the smaller Mini version. You can vary the bottle neck size, though that requires buying different bottles or using adapters, which most student projects don't do. You can also vary the number of Mentos dropped at once, which is the easiest independent variable to control. What I found after timing the eruptions was that the peak flow rate hits within the first three seconds and then drops off sharply. A full 2-liter bottle of Diet Coke with five mint Mentos produces its maximum spray between second one and second two, and the fountain is usually dead by second eight. Measuring height is tricky because the spray is asymmetrical. The tallest individual stream can be two meters while the average splash zone reaches only about one meter. I used a measuring stick held vertically next to the bottle and recorded the highest visible droplet path, not the average. That's a defensible measurement approach if your teacher asks. There are real limitations to this experiment, and they're worth stating upfront. The reaction is inherently chaotic. Two runs with the same conditions can produce visibly different heights because the exact orientation of each candy as it enters the water is impossible to control precisely. Even the angle at which you pull the paper tube away changes the result. This isn't a precision science — it's a demonstration of nucleation kinetics, and the data will have high variance. If your goal is a clean line graph with low error bars, you'll be disappointed. If your goal is showing that a hypothesis can be tested and refined through repeated trials, it works fine.
Get the Full Details

Another pitfall is using regular sugar Mentos instead of sugar-free. The sugar variant is coated with a different layer that dissolves too quickly, and the resulting foam is thin and short-lived. I tried it once thinking it would be a useful control group. The fountain lasted about four seconds and reached maybe thirty centimeters. Sugar-free mint is the only reliable variant for visible results. If you want a downloadable setup guide, there isn't a single authoritative source. The American Chemical Society has a brief page on nucleation demos, and a few university outreach sites host student worksheets, but nothing comprehensive. What I'd recommend is writing your own procedure sheet with these steps: chill all soda cans to four degrees Celsius, use 2-liter narrow-neck bottles, prepare five mint Mentos in a paper delivery tube, measure the bottle on a flat surface outdoors, drop the candies, measure the highest spray with a stick, record the duration until flow stops, repeat three times per condition, and calculate the average height. That's a complete method for a middle school or early high school project. The hypothesis itself is simple enough that you can state it in one sentence: sugar-free mint Mentos dropped simultaneously into cold carbonated soda will produce a taller and longer fountain than other Mentos varieties or warm soda, due to enhanced nucleation from gum arabic and surface texture. Whether your data supports it depends entirely on how carefully you control temperature and delivery method.