Department of Energy Argonne National Laboratory Office of Science NEWTON's Homepage NEWTON's Homepage
NEWTON, Ask A Scientist!
NEWTON Home Page NEWTON Teachers Visit Our Archives Ask A Question How To Ask A Question Question of the Week Our Expert Scientists Volunteer at NEWTON! Frequently Asked Questions Referencing NEWTON About NEWTON About Ask A Scientist Education At Argonne Antifreeze and High Temperature
Name: Clint
Status: other
Grade: other
Location: ND
Country: N/A
Date: August 2007

I read your response about Deionize Water Misconceptions and you said that glycol was needed for its high boiling point and water was added to reduce cost and to reduce the boiling point to just over the running temperature of the engine. Being a scientist I am sure you know that as the pressure goes up so does the boiling point of water, the coolant system in an auto is closed so straight water can be used and will not boil over. The glycol is used to reduce the freezing point thus the name antifreeze. How can it increase the boiling point?


If two liquids that have vapor pressures are combined and they form an ideal solution, then the boiling point of the solution as a function of composition is a straight line between the boiling points of the pure liquids. So, imagine a graph with percent of glycol on the x-axis and temperature on the y-axis, then a line would go from 100degC (0%-glycol) to 197degC (100%-glycol). This is to say that for ideal solutions, the boiling point is just the weighted average of the boiling points as a function of composition.

(In reality, the glycol-water solution is not an ideal solution and forms a curved graph. See: for an idea of what this looks like. The reasons for the non-ideality should be saved for another time.)

To reason this out, I like using the definition of boiling point as the temperature at which the rate in which molecules go from liquid state to vapor state is equal to the rate in which the molecules go from vapor state to liquid state. The vapor-to-liquid rate would depend on the external pressure (how many particles the vapor molecules would hit on the way out and force the molecules back into the liquid) - the higher the pressure, the higher the return rate and therefore more heat would be needed to make the liquid-to-vapor rate match this increased vapor-to-liquid rate. One could then imagine that as more and more of the less volatile substance (in this case, the glycol) is introduced into the solution (at constant pressure), there would be fewer molecules escaping into the vapor phase. Thus, more heat, higher temperatures, would be needed in order to increase the rate of liquid-to-vapor rate.

This is further compounded by the fact that the vapor is going to contain more of the more volatile liquid, which means that the solution is going to have a higher concentration of the less volatile liquid - and further push the boiling temperature up. But the concept remains the same.

Greg (Roberto Gregorius)

Anti-freeze (ethylene glycol) has several functions as a heat exchange medium in an automobile engine. One is, as you point out, to reduce the freezing point of the heat exchange fluid. But the freezing point of pure ethylene glycol is -13 C. (which is only 8.6 F.) so at first this does not seem like a very likely candidate for an "anti-freeze". But the process is more subtle. Mixtures of ethylene glycol and water freeze at temperatures less than either component. In addition, both ethylene glycol and ethylene glycol / water mixtures tend not to freeze but rather form a viscous glass instead. As an aside, glycerol has a melting point of 17.8 C., but few people have ever seen crystals of glycerol, because it too tends to form a glass rather than crystallizing as the temperature is lowered.

The boiling point of ethylene glycol is about 198 C. -- much higher than the boiling point of water -- so mixtures of the two can be used as a heat exchange fluid at temperatures significantly greater than 100 C. at significantly lower pressures than water alone (in a closed system). In principle, water alone could be used as a heat exchange fluid, but its vapor pressure doubles from 1 to 2 atmospheres at a temperature just over 120 C. So the problem becomes one of engineering hoses, fittings, etc. that could withstand high pressures if water alone were used as a heat exchange fluid.

When the term "boiling over" is used in the context of a heat exchange medium what is meant, as I am sure you know, is that the pressure of the fluid is greater than atmospheric pressure and the fluid evaporates until the pressures are equal.

In the "real world" there are a number of other factors that need to be taken into account, such as corrosion, and viscosity.

Vince Calder


The coolant loop of a car recirculates, but it is not completely "closed" (see the overflow reservoir). Water could work if we fully closed the system and beefed up plumbing for higher pressure, but pure hot water is a little too corrosive to metals. So at least half glycol is needed, and some additives. (Also I suggest you look up the "Thermostat" valve in the car's cooling loop.)

Pure glycol is both: lower-freezing and higher-boiling, than pure water. Yes we need the lower-freezing effect. But we have also adapted to depend on that boost in boiling temperature. In pushing cars' performance to the max, most available advantages get used. A hotter radiator can be smaller, for the same amount of heat-flow being dumped into the air. And the engine gets to run a little hotter, which helps efficiency and maybe reduces carbon build-up.

Boiling point elevation:

Water boils at 100C - H-OH

Ethylene glycol boils at about 198 C. - H-HCOH-HCOH-H

There are other liquids that boil as high as 300C.

Glycerin is one of them - H-HCOH-HCOH-HCOH-H

Mix glycerin in water and virtually all the vapor is water vapor (say below 200C). The water vapor pressure that remains is reduced compared to pure water for two reasons:

1) there is less water present in the fluid, by a modest percentage. Reducing concentration of water in the fluid, a factor of two (from 100% to 50%), reduces the vapor pressure at any given temperature by that same factor of two. "Boiling point" means the temperature at which the vapor pressure passes one atmosphere, so the boiling point is elevated by something like 20 degrees C, or whatever increment it takes to double the vapor pressure.

2) a given water molecule likes being mixed with glycols. It can do so in the liquid, but it cannot in the vapor state. This favors the liquid state, so it may elevate the boiling point another several degrees.

Freezing point and boiling point are set by rather different things, so it is quite possible for one solute to push them in opposite directions. Liquid-vs-Gas is about the average cohesive attraction between molecules when in close contact.

It depends on what stuff, what molecular sub-groups, show on the surface of each molecule. Some groups are "stickier" than others. But Liquid-vs-Solid is not about getting closer, it is about getting locked into fixed relative positions. It depends in great detail on the shapes of the molecules and how well they fit together. It depends on how well an infinitely stacked-up 3-D crystal can be built with that shape. And imagine two of the same molecule in contact, sliding past each other. Bumps in cohesive force will be felt, like dragging chains of magnet-balls past each other. Stronger bumpiness translates into higher freezing point. These dependencies are difficult to estimate in one's head, or even by computer. So we just "ask nature": we measure the freezing point. Boiling points are noticeably more calculable.

Jim Swenson


Auto-mechanics and Chemistry are two subjects that involve a lot of detail! Since I am inclined in both, I will answer more than just your question so that you can understand how the entire system works.

A cooling system works by circulating a liquid (antifreeze) around the outside of the cylinders and throughout other important parts of the engine block and head so that these parts do not superheat and end up failing. Antifreeze, as you so well know is a mixture of water and glycol--specifically either propylene or less commonly today, ethylene because it is much more toxic. Propylene glycol has a melting/freezing point of -59C and a boiling point of 189C. Often times, the temperature of coolant has to exceed 200C (my Honda Civic has an operating temp of 205C). Propylene glycol on its own cannot obtain this temperature, so it has to be mixed with water to increase the boiling point. Water has a boiling point of 100C, obviously. You bring up a point that under pressure, the boiling point of a liquid can be increased. While this is true, in practice for an auto, the engine can only take so much pressure before something blows. Your radiator cap is designed to be the first and most inexpensive thing to go--and it also lets you know that something is wrong! It is a good thing that the radiator cap blows off because if your head gasket blows, then you have a very expensive repair bill on your hands!

Pure water would not do you any good because it freezes in the winter and will build up way too much pressure on its own. The reason that you can add straight water to your coolant reservoir and not blow up your radiator is because when your engine is cold, the thermostat closes and shuts off access to coolant that is in your engine block. So even if you drain the coolant and replace it with water, there is still a lot of coolant still inside your engine. Next time you turn it on and the thermostat opens up, the pure water will mix with the coolant and the fp/bp will be changed.

Pure glycol, on the other hand, still is not a great choice because the temperature still is not a high enough range for all vehicles. And the less pressure you put on your engine, the better. So a mixture must be used, but why does the mixture of these two liquids cause the fp/bp range to become much broader? The reason is due to hydrogen bonding. You can see the structure of glycol here:

Boiling is obtained when the surface pressure of the liquid is equal to the atmospheric pressure. Because of the strength of the hydrogen bonding holding the molecules of water and glycol together, the temperature needed to obtain the same pressure is drastically increased.

One thing that has not been considered here is thermal expansion. When a gas is heated, it expands and rises, right? When a liquid is heated, it too will expand until it reaches its boiling point and turns into a gas. While pure water does not expand (and cannot be compressed very much either) upon heating, organic solvents tend to expand quite a bit and can be compressed into smaller volumes with pressure. So even if the antifreeze is not boiling, the mixture will expand quite a bit and hence create a lot of extra pressure. This is handled by the overflow container that is connected near the radiator cap. This overflow container is in the closed system with the cooling system, but there is simply air in that part. This air is at the highest part of the cooling system, so it will not get into the engine and cause trouble. It will, however, catch all of the expanding coolant when the fluid heats up. When the coolant cools back down, the liquid will contract and create a vacuum (via a hose in the overflow container), which sucks the overflow back into the radiator to top it off.

Matt Voss

Glycol, or any additive, will increase the boiling temperature of water (or any volatile solvent) simply by lowering the concentration of the water in the liquid.

Think of it this way: The boiling temperature is the temperature at which a component of a liquid is in equilibrium with the vapor at a vapor pressure of 1 atmosphere. Since the vapor pressure of the glycol component will be much lower than the vapor pressure of the water component, we can safely approximate that the vapor is entirely water vapor with very little loss of accuracy.

What does it mean for the vapor and liquid to be in equilibrium? It means that the rate of molecules of vapor condensing to liquid exactly equals the rate of molecules of liquid evaporating to the vapor. What determines these rates? Well, the higher the temperature, the less likely it is for vapor to condense to liquid and the more likely it is for liquid to evaporate to vapor. Also, the higher the vapor density is (that is, the more molecules of vapor that are in a given volume), the more likely it is for vapor molecules to condense. The same holds true for the liquid. The greater the concentration of molecules in the liquid, the more likely it is for some of them to dissociate and join the vapor phase. (If, say, 1% can be expected to dissociate in a given time, 1% of a large number is more than 1% of a small number.)

So, what happens if the water exists as a mixture with glycol instead of by itself? Well, the water in the liquid phase is less concentrated than it would be if there were no glycol, and fewer water molecules will join the vapor. Thus, it will be necessary to heat the liquid to a higher temperature to bring the vapor to atmospheric pressure, that is, to make the liquid boil.

You might notice that most jugs of antifreeze identify themselves as "anti-freeze anti-boil" and not just as antifreeze.

Richard Barrans
Department of Physics and Astronomy
University of Wyoming

Click here to return to the Chemistry Archives

NEWTON is an electronic community for Science, Math, and Computer Science K-12 Educators, sponsored and operated by Argonne National Laboratory's Educational Programs, Andrew Skipor, Ph.D., Head of Educational Programs.

For assistance with NEWTON contact a System Operator (, or at Argonne's Educational Programs

Educational Programs
Building 360
9700 S. Cass Ave.
Argonne, Illinois
60439-4845, USA
Update: June 2012
Weclome To Newton

Argonne National Laboratory