I have a household fridge with a normal freezer section.

I have two identical bottles of inexpensive white wine at room temperature.

I have a mixing bowl, I put a liter or so of tap water in it.

I open the door once fairly briefly and

With bottle A I do this,

enter image description here

With bottle B I do this,

enter image description here

Later - say 10 to 20 minutes later (long before anything has frozen) - I pour the wine in to glasses.

Which wine got colder more quickly?

Part B (a completely different problem):

That was room temperature water. Say I have a bowl of water that is already about as cold as the air in the freezer (is that possible? how cold is air in a freezer?) In that case, which one wins?

(ie, which "transmits cold" to the bottle faster, air or water at same temperature?)

Please note that as it says above

long before anything has frozen

Ice is not involved here.

Issues include...

Cooling the wine is tricky: you have to cool it "through" the bottle. How do the thermal properties of glass affect things? For example, is it even "worth" cooling the glass quickly?

What, really, cools things in a freezer? (Say, the bag of vegetables there.) In fact, is it the air? Or in reality is it just the other small surfaces (the racks, etc.?) If it was floating in a perfect vacuum, would it cool at all?

  • $\begingroup$ Guesses: A: The one in water (mixing bowl) cooled faster B: It is possible but I would think the water would be frozen. The one in the bowl still "wins" or cools quicker. $\endgroup$
    – chackerian
    Commented May 11, 2017 at 18:50
  • 1
    $\begingroup$ Water has greater thermal conductivity than air so it is more effective to refrigerate something by letting it be in contact with cold water. In the first experiment I believe the water in the bowl cools down quicker because of the free surface in contact with cold air. Hence I believe bottle B gets colder slightly faster than bottle A. In experiment B, bottle B cools down much quicker than bottle A. $\endgroup$
    – Diracology
    Commented May 11, 2017 at 18:54
  • $\begingroup$ Hi Dirac, that seems to be the most useful info here. Thanks. $\endgroup$
    – Fattie
    Commented May 12, 2017 at 13:42
  • $\begingroup$ You have photos of what you are asking about for Part A, so haven't you already done the the experiment and know the answer? If not, is there a reason why haven't you spent the half-hour needed to do the experiment? For Part B, if the water is at the same temperature as the air in the freezer (ideally -18C), it will be ice, which is hard to insert a bottle into. $\endgroup$ Commented Jun 29 at 0:33
  • $\begingroup$ Aside: I have never seen French put wine in a freezer - usually on the bottom shelf of the fridge. But then, in France "freezer" literally means -15C and lower (not just below zero) whereas wine should be chilled, not cold/frozen. $\endgroup$
    – Roger V.
    Commented Jul 1 at 12:41

5 Answers 5


In part A, the bottle in the water will cool slower because you must remove the heat from the water and the wine. The bottle on the shelf will cool faster because all you have to cool is the wine.

In part B, if you have water with a touch of salt, it will not freeze at zero Celsius and the water will remove the heat from the wine faster, the larger surface area of the water coupled with some evaporation that cools the water will cause it to cool significantly faster than the dry bottle on the shelf.

In a third scenario you didn't consider, which I believe is the fastest way to chill a bottle of wine, you wrap a bottle in a very damp rag and place it in the freezer. The water in the rag will begin to evaporate and draw the heat out of the bottle quite quickly between evaporation and convection.

The evaporation concept is utilized all over the world in cooling towers and water chillers. Water is heated by cooling something else, then the hot water is pumped to the cooling tower where it falls through a cascade medium that breaks it up into small droplets which loose heat through evaporation. They want small droplets because it increases the surface area of the water volume and causes more evaporation. More evaporation, more heat lost through the phase change. Efficient cooling towers are capable of chilling water below ambient air temperatures.

Drew K

  • $\begingroup$ "the larger surface area of the water" ..... ??? $\endgroup$
    – Fattie
    Commented May 11, 2017 at 20:25
  • $\begingroup$ I'm ... just not certain if "evaporation" comes in to play in paragraph 3? $\endgroup$
    – Fattie
    Commented May 11, 2017 at 20:29
  • $\begingroup$ The air inside a refrigerator is dry due to the cooling process, this causes the water to evaporate faster. The Larger surface area of the water is because there will be heat loss through the plastic tub and the surface of the water. The surface area of 1.5 l of water plus the part of the bottle sticking out of the water is greater than the surface of the bottle alone. These things combined will increase the cooling of the bottle. If you were to graph it the initial cooling of B would be fast, then it would level off more than A then they will form a similar slopes. $\endgroup$ Commented May 11, 2017 at 21:01
  • $\begingroup$ Good thoughtful answers $\endgroup$
    – Lambda
    Commented May 11, 2017 at 22:07
  • $\begingroup$ The temperature in a common freezer is around -20 C. Water with a touch of salt will probably still freeze. $\endgroup$
    – nasu
    Commented May 12, 2017 at 15:17

Liquid water has a few interesting properties that we need to discuss before answering your questions.

1) It has a high specific heat compared to most things on Earth. Specific heat is a measure of how much energy it takes to heat or cool a fixed amount of the material and has SI units of $\frac{J}{gK}$. Water's specific heat is about $4.2\frac{J}{gK}$ while wine's, at $14\%$ alcohol by volume, is about $4\frac{J}{gK}$.

2) It also has a pretty high thermal conductivity. Thermal conductivity is a measure of how fast energy moves from the warm side to the cold side of an object, and it has units of $\frac{W}{mK}$ (remember, Watts are Joules per second). It turns out this is a super complicated calculation in reality (materials have different conductivities in different directions), but this is a good approximation for this problem.

3) Thermal contact between objects is also quite complicated. Broadly, objects thermally interact by radiating, conducting and convecting (forced and natural). 3a) Radiation happens all the time, and it's how the Sun gets heat to the Earth - all objects emit photons in every direction at a wavelength consistent with the their temperature. Radiation is related to surface area. 3b) Conduction is when two objects are in physical contact and they share energy at their interface. Conduction is also related to the surface area. 3c) Convection is when a fluid moves over the surface of a solid (or when a gas moves over the surface of a liquid). If a solid has a vertical interface with a fluid, the boundary layer of the fluid will move up if the solid is hot and down if it's cold which is called natural convection. Forced convection is when currents in the fluid move over the surface (as with a fan, a breeze or a flow). Convection scales with the surface area of the interface.

All three of these are pretty complicated which is why you've received so many answers that feel like educated guesses. Last, until the whole bottle reaches thermal equilibrium, the freezer will be colder than the glass which is colder than the wine.

A) Wine bottle on rack vs. wine bottle in water: With the tap water at room temperature, the free bottle will be cooler because water has a very high specific heat that insulates the wet bottle. If you got a thermally conductive container with less water, but more of the bottle's surface in contact with the water, you would cool the wet bottle faster, but the geometry of these things is really important, and that is the wrong bowl.

B) Getting water colder than freezing typically changes it to ice. You can lower the freezing point a little by adding alcohol, or a lot by adding table salt. The thermal conductivity of ice is half that of water, so I'll assume that we salted the bowl, and then the wet wine would cool much faster because while the interfaces are the same, the water can accept more heat without changing temperature as much as the air.

C) "Cooling wine is tricky": Not sure what you mean by tricky - you put it in thermal contact with something colder. Glass insulates the wine, but it has a much lower heat capacity and it's a not that thermally conductive. The only way to get the wine cooler than the glass is to take the chilled bottle out of the freezer and leave it in a warmer environment.

D) How freezers work: Freezers pressurize a fluid, expose it to the room temperature and then allow the fluid to de-pressurize it in thermal contact with the inside of the freezer. Feel a bike tire while you let a lot of the air out, and you'll see what I mean. There's typically a fan that pushes air over the pipes/radiator and into the main chamber because it cools the contents more quickly. I leave it as an exercise to decide which of the mechanisms for heat transfer apply here.

  • $\begingroup$ Next time, please just say what you want directly. $\endgroup$
    – user121330
    Commented Jul 2 at 17:59

Some Important Stuff:

  1. Fourier's law of heat conduction tells us that the flow of heat energy per unit area per unit time is directly proportional to the thermal conductivity and the temperature gradient.
  2. Thermal conductivity of water increases with temperature upto about 130 degrees Celsius. It decreases after that.
  3. Water conducts heat much faster than air, even when they're at the same temperature.
  4. Ice conducts heat faster than water. Solids have systematic arrangement of atoms, where the intermolecular forces are very much stronger than in liquid form. While liquids are characterised by quite weaker intermolecular forces. The systematic arrangement of atoms is the reason for high thermal conductivity of ice. Conduction through solids, liquids and gases is by collision of molecules. When the molecules are arranged in a systematic position, it is easier to conduct heat from one lattice to another, they are close to each other while in liquids due to weak intermolecular forces of molecules, molecules are far apart, thus making it difficult to transfer heat from high energy molecules to low energy ones.

In the second case, I assume that the water indeed remains water for the 10-20 minutes that you're talking about and doesn't freeze. A pinch of salt might prevent it from freezing. The temperature gradient is same for both water and air but comparing the conductivities leads us to conclude that the chardonnay in the bowl will cool much faster. If the water did freeze to ice, it would still cool faster than when it was in air. But you would need some quantitative data to find out whether ice or water at same temperature cools down the bottle faster. Though I did mention that ice conducts heat faster, you must realise that the area of the bottle in contact with ice is less than when it was in contact with water. Crushed ice might solve the problem though.

In the first case, there are opposing factors.

1) The water is at room temperature. So, the temperature gradient of the wine-water system is quite low, compared to the wine-air system.

2) Water conducts heat faster.

I'm not entirely sure but Dirac might be right, because water is too good a conductor compared to air.

Aside: Glass is a very good thermal insulator. So it will slow down the heat transfer in both cases, though it doesn't really affect the answers. Also, I believe it's the air which cools down the stuff in refrigerators. If stuff were to float in a perfect vacuum, heat conduction would be zero. Perfect vacuum means there are no molecules to keep the heat flowing. Perfect vacuum has zero thermal conductivity. I hope this answers your question.


Let's break down the two parts of your problem step by step, considering the physics of heat transfer.

Part A: Comparing Cooling of Wine Bottles in Freezer Air vs. Water

1. Cooling Mechanisms

  • Bottle A (in air):

    • Cools primarily through convection with the cold air inside the freezer.
    • Air has a relatively low thermal conductivity.
  • Bottle B (in water):

    • Cools through conduction with the water and convection within the water.
    • Water has a higher thermal conductivity compared to air and can transfer heat more effectively.

2. Thermal Conductivity

  • Water:

    • Thermal conductivity: ~0.6 W/(m·K)
    • Specific heat capacity: ~4184 J/(kg·K)
    • Density: ~1000 kg/m³
  • Air:

    • Thermal conductivity: ~0.025 W/(m·K)
    • Specific heat capacity: ~1005 J/(kg·K)
    • Density: ~1.225 kg/m³ (at room temperature)

3. Heat Transfer Coefficients

  • Convective heat transfer coefficient:
    • Air: ~10-100 W/(m²·K)
    • Water: ~500-10,000 W/(m²·K)

Water has a much higher convective heat transfer coefficient compared to air, meaning it can remove heat from the wine bottle much more effectively.

4. Conclusion for Part A

  • Bottle B (in water) will get colder more quickly than Bottle A (in air). This is because water can transfer heat away from the bottle more efficiently due to its higher thermal conductivity and convective heat transfer coefficient.

Part B: Cooling with Water at Freezer Temperature

1. Freezer Temperature

  • Typical household freezers operate at around -18°C (0°F).

2. Comparing Heat Transfer

  • Bottle A (in air at -18°C):

    • Same convective cooling as before.
  • Bottle B (in water at -18°C):

    • Water at the same temperature as the freezer air will still have a higher heat transfer rate due to better conduction and convection properties.

3. Specific Heat of Glass and Thermal Resistance

  • Glass has a specific heat capacity of about 840 J/(kg·K) and a thermal conductivity of about 1.05 W/(m·K).
  • The thermal resistance of the glass bottle will affect the overall heat transfer, but since both bottles are identical, this resistance is the same for both scenarios.

4. Vacuum Consideration

  • If an object is in a vacuum, it can only lose heat through radiation, which is significantly less efficient than convection and conduction.

5. Conclusion for Part B

  • Bottle B (in water at freezer temperature) will still cool faster than Bottle A (in air at freezer temperature). Water's superior heat transfer properties continue to outperform air, even when both are at the same temperature.


  1. Bottle B in tap water will cool faster than Bottle A in freezer air due to water's higher thermal conductivity and convective heat transfer coefficient.
  2. Bottle B in cold water (at freezer temperature) will still cool faster than Bottle A in freezer air for the same reasons.

In a freezer, the primary cooling mechanism for items is indeed the air, which is cooled by the evaporator coils. The solid surfaces and racks also play a role in heat transfer, but their contribution is secondary compared to the convective heat transfer from the air. If an object were floating in a perfect vacuum inside the freezer, it would primarily lose heat through radiation, which is much less efficient than convective cooling by air or water.

  • $\begingroup$ What an amazing answer. So much better. Sending a bounty! TY $\endgroup$
    – Fattie
    Commented Jun 28 at 16:41

A problem here is that different parts of the fridge are at different temperatures (so it is recommended you store different foods in different parts of the fridge).

The back of a fridge is colder than the front. Cool air sinks so the lower area would be cooler in that sense.

But obviously a bottle in contact with cold water will cool quicker than a bottle that relies on cold air to get cold.

So I would guess bottle B.

Tip. Put the wine glasses in the fridge so you don't pour the cold wine into a room temperature glass.

  • $\begingroup$ The bottle in contact with water will not necessarily cool faster than the one in contact with cold air if the water is warmer than the air. $\endgroup$ Commented May 11, 2017 at 20:21
  • $\begingroup$ This answer makes very little sense. $\endgroup$
    – Fattie
    Commented May 11, 2017 at 20:26
  • $\begingroup$ A closed modern fridge should have fairly even temperature. The fan that kicks on to cool it will mix it quite well. A fridge that is opened often will have regions that stay colder due to air restriction and materials there staying cold during brief times the door is open. $\endgroup$ Commented May 11, 2017 at 21:05

Not the answer you're looking for? Browse other questions tagged or ask your own question.