Here is a video link for the experiment.

In the experiment, cold water is being poured on a closed jar containing hot water and water vapour and we observe that the hot water inside jar starts boiling.

The explanation given is that the vapour condenses as temperature decreases causing the pressure to decrease, which causes boiling point temperature to decrease, so hot water in liquid state starts to boil and form vapour until the temperature-pressure point reaches vaporisation curve.

My question is why would this whole process happen since vapour lost due to condensation is being returned back due to boiling? Why not only either vapour condenses if the pressure has to be decreased or the hot water boils if the pressure has to be increased?

  • 2
    $\begingroup$ have you watched the whole video, it's titled "How ice water makes hot water boil"? $\endgroup$ Commented May 24 at 20:16
  • $\begingroup$ You may also want to watch Charles Marzzacco's original video that inspired the video that you link to. $\endgroup$ Commented May 25 at 14:06
  • $\begingroup$ Why the focus is on the water vapor to the exclusion of the air in the bottle? When cold water is run over the bottle, the water vapor is cooled but so are the other gases (the air). That lowers the pressure and thus drops the boiling point. Is the cooling of the air somehow less important than the cooling of the vapor? $\endgroup$ Commented May 25 at 14:10
  • $\begingroup$ @aaaaasaysreinstateMonica: Have you read Navneet's whole post? The explanation from the video is summarized there, so clearly they watched the video. The questions in the last paragraph are asking for more detail and clarification. $\endgroup$ Commented May 25 at 14:13

2 Answers 2


The reason is that the vapor becomes much cooler than the water in the bottle.

Before the change they had the same temperature, and the vapor pressure was correct for that temperature. After the cold water is applied the vapor region cools much more than the hot water because it has less "thermal mass" (ability to store heat and therefore buffer against temperature change).

So after cooling, the vapor region has cold walls that condense some vapor and also the vapor that exists will be reduced in volume, two reasons why the pressure will drop. At the same time the hot water temperature is barely changed, so it still requires the same pressure as before to avoid boiling. But that pressure isn't there any more, so it boils.

  • $\begingroup$ Can you further explain what makes the water vapour loose temperature much faster than the hot water? $\endgroup$
    – Navneet
    Commented May 23 at 7:33
  • 15
    $\begingroup$ Because a system can be in a state other than equilibrium. The boiling is the mechanism by which this system reestablishes equilibrium. $\endgroup$ Commented May 23 at 10:28
  • 3
    $\begingroup$ @Navneet "At 1atm no water vapour can exist at 99°C and liquid water at 101°C." Sure they can. Supersaturated air and superheated liquid both can and do happen. $\endgroup$
    – Tashus
    Commented May 23 at 16:57
  • 10
    $\begingroup$ @Navneet If water vapour cant exist below 99°C, then how does humidity work? Substances typically have a "vapour pressure" associated with their temperature, and some of the substance will vaporize into the air until that partial pressure is reached for the substance being vaporized $\endgroup$
    – JMac
    Commented May 23 at 16:58
  • 7
    $\begingroup$ I’m sorry, but you deeply misunderstand the concept of boiling, my friend. There is always some equilibrium vapor pressure at every temperature. But at boiling the vapor pressure has reached the surrounding atmosphere’s ambient pressure. $\endgroup$ Commented May 23 at 19:01

What happens in the video is very interesting. Initially when the there was no cold water the liquid and vapour had same temperature and pressure. Then cold water was poured, and you could see that it is poured carefully so that heat exchange occur with the vapour only. A minimum portion of the water flow touches the glass exposed to the liquid. As a result, the vapour is cooled lot more rapidly than the liquid. There are 2 main reasons for this difference in cooling:

  1. Cold water is exposed mostly to the vapour.

  2. Liquid cooled because its vapour cooled, so vapour started cooling first then the liquid and thus vapour was always cooler than the liquid. And also more the conductivity, faster will be the rate of cooling. Gas molecules are in rapid motion, thus gas conducts faster.

Now when the gas is cooling rapidly, the pressure also drops rapidly (some of the cooled vapours condense, and pressure decreases with temperature decrease) and the boiling point of the liquid (it depends on the pressure of the environment on it) also drops. But the rate of decrease in temperature of the liquid turns out to be less than the rate of decrease of boiling point. Thus, the liquid is always above boiling point and boils. As the liquid boils, it releases vapour and increases the pressure, and a time comes when the rate of release of vapour is enough to balance the decrease of pressure due to cooling, and thus boiling point starts increasing again and boiling stops.

  • 2
    $\begingroup$ "Gas molecules are in rapid motion thus gas conducts faster ." Gasses are known to have pretty poor thermal conductivity. In fact, the best insulators often work because they have a lot of air in them. The extremely low density wins over the rapid motion (easily). $\endgroup$ Commented May 24 at 7:58
  • $\begingroup$ Gas conducts very well among itself meaning heat distributes itself very rapidly in gases . I didn't mean to say that it absorbs heat rapidly from the wall. $\endgroup$
    – Users
    Commented May 24 at 8:09
  • 2
    $\begingroup$ gas "conducts" ("distributes" would be better) heat very slowly by conduction/diffusion. It distributes heat rapidly by convection. The temperature differences in the bottle (and the small volume) do not encourage convection over the time scales involved. See Nusselt number for a starting point on the race between convection and conduction for heat transfer. $\endgroup$ Commented May 24 at 20:07

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.