building-scienceheader image
0 Helpful?

What Happens When You Put a Plastic Vapor Barrier in Your Wall?

A look at the physics of humid air and cool surfaces

Posted on Apr 30 2014 by Allison A. Bailes III, PhD

A lot of people have heard advice about vapor barriers and vapor retarders. Many of them have walked away confused. A big part of the problem, I think, is that they've been told what to do — "Put it on the warm-in-winter side," or "Never use one" — but they haven't had the physics of what happens explained to them.

In this article, I'm not going to get into the details of vapor barriers or all the possible scenarios of different wall assemblies and moisture loads. I'm simply going to explain what happens in a wall cavity with and without a plastic vapor barrier installed.

Scenario 1: plastic on the inside, hot humid weather

I'm writing this article because one of our HERS raters came across a house in Charleston, South Carolina that had polyethylene under the drywall, on the interior side of the wall assembly. Polyethylene, often shortened to poly, is a Class I vapor retarder, which is usually what’s meant by the term vapor barrier. If you're at all familiar with the climate in Charleston and understand moisture, you know that can't be a good thing.

I was there one day in June a few years ago and saw condensation on the outside of a window... at one o'clock in the afternoon of a sunny day. The dew point of the outdoor air was 78° F. The window had a single pane of glass. They were running the air conditioner, so the indoor temperature was probably 75 or below. Humid air hits cool surface. Condensation results.

Now, imagine that pane of glass is actually a sheet of polyethylene. Next, imagine that a layer of drywall separates the poly from the indoor air. Then build a wood-frame wall outside the poly, complete with claddingMaterials used on the roof and walls to enclose a house, providing protection against weather. and air permeable insulation in the cavities. Will that poly be protected from the outdoor humidity? Or will it, like the window I saw, be dripping with condensation?

If it's a typical wall, chances are good that water vapor in the outdoor air will find its way into the wall cavity, eventually finding the sheet of poly, pressed up against the drywall. If that wall allows outdoor air to infiltrate and the poly is below the dew point, condensation is the likely result. If those conditions last long enough, the condensed water will run down the poly, get the wood framing wet, and begin to rot out the wall.

The truth, however, is that the water vapor in the outdoor air is rarely the source of moisture that rots out a wall. More likely is that moisture from a wet foundation makes its way up into the wall by capillaryForces that lift water or pull it through porous materials, such as concrete. The tendency of a material to wick water due to the surface tension of the water molecules. action, or bulk water from leaks around openings gets into the wall cavity. The presence of an interior vapor barrier makes drying out the cavity harder to do, though.

Without poly beneath the drywall, water vapor hits the drywall and diffuses through to the drier (in summer) indoor air. By installing a sheet of poly there, you cut off that drying mechanism and water that finds its way into walls can stay there longer and do more damage.

Scenario 2: plastic on the inside, cold weather

In cold weather, a sheet of poly on the interior side of a wall probably won't cause any problems. The humid air is indoors, and the dry air is outdoors. The sheet of poly still cuts off drying to the indoors, but it keeps the water vapor in the humid indoor air away from the cold surfaces inside the wall. This is what building scientists proposed as the solution for walls that wouldn't hold paint back in the early days of insulation. It didn't solve the paint problem, though, because water vapor from the indoor air wasn't the main source of moisture.

Scenario 3: plastic on the outside, cold weather

Plastic on the outer surface of a wall in cold weather could cause problems. The humid air is indoors. The cool surface is the sheathingMaterial, usually plywood or oriented strand board (OSB), but sometimes wooden boards, installed on the exterior of wall studs, rafters, or roof trusses; siding or roofing installed on the sheathing—sometimes over strapping to create a rainscreen. , assuming no exterior insulation. If water vapor diffuses or infiltrates into the wall cavity and finds the cool surface, moisture problems can occur.

Of course, you can have moisture problems here even without the exterior vapor barrier because of what Bill Rose calls the rule of material wetting. That is, warm materials dry more quickly than cold materials. If humid air from indoors finds cold sheathing and starts accumulating there, the moisture will stay there longer because it’s difficult to dry at low temperatures, even though the outdoor air is dry in terms of absolute humidity.

Scenario 4: plastic on the outside, hot humid weather

A vapor barrier can cause problems if it prevents drying to a drier space. In a building with air conditioning during hot humid weather, the drier space is indoors. The humid air is outdoors. Putting a vapor barrier on the inside, as we saw in Scenario 1 above, can lead to problems because any humid air that gets into the wall cavity is blocked from drying to the inside.

If the vapor barrier is on the outside, it prevents the humid air from diffusing into the wall cavity and finding the cold surface on the other side of the cavity, the back side of the drywall. So, like a vapor barrier on the inner surface in cold weather, putting one on the outer surface in hot weather isn't likely to have a moisture problem due to vapor diffusion. It doesn’t mean you won’t have a moisture problem, though, because moisture can get into the walls by means other than diffusion.

It's not just a climate issue

We can summarize the vapor barrier issue like this:

  • A Class I vapor retarder's job is to keep water vapor in humid air from diffusing through one side of a wall and finding a cool surface inside the wall.
  • When a Class I vapor retarder is on the side of a wall where the dry air is (i.e., outside in winter or inside in summer), moisture problems can occur.
  • A Class I vapor retarder reduces the movement of water vapor by diffusion. Holes in the vapor barrier that allow humid air through may allow a lot more water vapor into an assembly than the vapor barrier is stopping. Because of this, air sealing is more important than vapor retarders.

If you're in a place like Miami, where it'll almost never be colder outdoors than indoors, a Class I vapor retarder on the outer surface of a wall assembly may be OK. If you're in Maine and never use an air conditioner, a Class I vapor retarder on the inner surface may be OK. If you're in a cold climate, however, and do use air conditioning, you need to be careful with interior vapor barriers like polyethylene. You could be creating the kind of problems I described in Scenario 1 above.

Enhancing drying vs. preventing wetness

If designers and builders want buildings to do their jobs properly and not fail prematurely, it's important to understanding moisture. We know now that mid-twentieth century building science incorrectly ascribed magical properties to vapor barriers. Water vapor from indoor air wasn't the source of most moisture problems. Bulk water from deficiencies in drainage planes, flashing, and other moisture management details caused most of the problems.

Building science has progressed since then. We know that vapor barriers can cause problems, but we still have homes like the one in Charleston with poly in the walls. And we have $4 million dollar homes with poly on the walls, too. I saw the one shown in the photo below when Martin Holladay came to Atlanta last year. The photo was taken in the basement, and the kneewalls in the attic also were covered with poly.

Our understanding now is that it's generally more important for wall assemblies to be able to dry than it is to block water vapor with materials like polyethylene. In most cases, you’re fine with the vapor retarding ability of the materials already in your walls, and you don’t need to use Class I vapor retarders.

Here's what Bill Rose wrote in his book, Water in Buildings: "Given the fact that a very small percentage of building problems (1 to 5% at most in the author's experience) are associated with wetting by water vapor diffusion, the argument for enhanced drying potential becomes much stronger."

In other words, the potential for creating problems is greater when you use Class I vapor barriers that prevent drying, so be absolutely certain you need one before going that route.

Allison Bailes of Decatur, Georgia, is a speaker, writer, energy consultant, RESNET-certified trainer, and the author of the Energy Vanguard Blog. You can follow him on Twitter at @EnergyVanguard.

Tags: , ,

Image Credits:

  1. Energy Vanguard

Apr 30, 2014 11:14 AM ET

by Ron Keagle


The original justification for a warm-side vapor barrier was to solve the wall moisture problem that arose with the advent of filling the walls with insulation, which at that time was fiberglass. The insulation lowered the temperature of the inside surface of the sheathing below the dew point of the outward vapor flow. So the vapor barrier was introduced to stop the outward vapor flow.

More recently, however, that original explanation of the moisture problem has been replaced by a new explanation involving the fact that cold sheathing holds more water than warm sheathing. So we have explanation #1 being replaced by explanation #2.

Above, you say that explanation #2 is ultimately caused by bulk water intrusion. So, If one were to eliminate all bulk water intrusion, would that eliminate the wetting of cold sheathing which can hold more moisture than warm sheathing?

Apr 30, 2014 5:26 PM ET

Ron Keagle's Question
by Bill Burke

Bulk water is a problem. But so is moisture carried by warm air, which comes from the interior in winter. That is why Alison speaks of the importance of air sealing. Air carries moisture in its gaseous state, which condenses when it reaches a cold surface. See

Apr 30, 2014 7:57 PM ET

Edited Apr 30, 2014 7:57 PM ET.

All in all ...
by flitch plate

All in all, Class 1 vapor barriers are a bad idea. I would even venture to say that Class II (at least on the low end < 1.0 perm) VDR’s have limited if any useful application nowadays. Seen too many places using, heard too many builders proclaiming: vapor barriers for infiltration/exfiltration control … that is, as a cheap and easy air-seal. Bring out the dehumidifiers, defrost the HRV's.

Apr 30, 2014 8:12 PM ET

Edited Apr 30, 2014 8:12 PM ET.

Reply to Bill Burke
by Ron Keagle


I understand your point, however, Referring to the erroneous conclusion of earlier building scientists, Allison said this, “Water vapor from indoor air wasn't the source of most moisture problems. Bulk water from deficiencies in drainage planes, flashing, and other moisture management details caused most of the problems.”

And he said this: “In cold weather, a sheet of poly on the interior side of a wall probably won't cause any problems. The humid air is indoors, and the dry air is outdoors. The sheet of poly still cuts off drying to the indoors, but it keeps the water vapor in the humid indoor air away from the cold surfaces inside the wall. This is what building scientists proposed as the solution for walls that wouldn't hold paint back in the early days of insulation. It didn't solve the paint problem, though, because water vapor from the indoor air wasn't the main source of moisture.”

I don’t understand how poly can have the ability to “keep the water vapor in the humid indoor air away from the cold surfaces inside of the wall,” and yet be incapable of solving the paint adhesion problem because for that problem, “water vapor from the indoor air is not the main source of moisture.”

Apr 30, 2014 8:39 PM ET

Edited May 1, 2014 10:59 AM ET.

Poly has its place.
by Malcolm Taylor

In some climates poly still has a place as an air barrier. I agree in the long term its use should be phased out but for now it is a viable option in some places.
Here in British Columbia both the code and inspectors became very serious about air sealing over twenty years ago. The poly had to be sealed at all sills and penetrations. The advantage to this approach were two fold: It was done as a separate task and subject to visual inspection. This had a knock on effect of making insulators take their jobs seriously and we never had the poor batt installations I see are common elsewhere.
So BC and much of the rest of Canada has a system of insuring tight houses in which a small group might because of their summer cooling have some vulnerability in the poly being present. The alternatives, an airtight drywall approach, or in more sophisticated building envelopes burying the air sealing in another part of the enclosure, while fine in high end houses, presents real problems for most housing production. It can't be inspected until the problems have occurred (by a blower door test) and disrupts what is for now a system in which the building trades have been educated and understand.
As poly is abandoned in many parts of the States, builders are interpreting the problems caused by it's vapour impermeability as a reason to just abandon air sealing altogether, rather than adopting another strategy. In northern climates I don't think it is a good idea to get rid of a system that in the main works, until an alternative method becomes widely understood and disseminated.

May 1, 2014 1:40 PM ET

on the contrary
by Matt Dirksen

"As poly is abandoned in many parts of the States, builders are interpreting the problems caused by it's vapour impermeability as a reason to just abandon air sealing altogether"

I don't know. I believe that the builders that are abandoning the use of poly are now becoming more educated on the fundamental difference between air sealing and vapor sealing. Besides, prescriptive air sealing has been in the code books (and has been enforced) now for many years down here. Of course I live in Maryland, where we immediately adopt the latest version of the IRC and IECC. Although, perhaps I live in a bubble (and one likely made of poly.)

May 1, 2014 2:53 PM ET

Warm wet room. cold frame.
by Roger Anthony

Let's try and add some detail. Wood is hygroscopic. Molecules of water vapor are small enough to enter the pores of wood when it is below dew point and raise its water content, to the point where mold can form and where wood rot can start.

Water vapor doesn't settle inside warm wood, that is to say wood that is above dew point. Therefore, having the insulation on the room side of a wall would seem to be wrong, as the temperature of the frame will be somewhere approximate to the temperature outside and at times condensation will result inside the wood, however, experience shows that where the wood is warmed and dried by the summer sun each year this isn't usually a problem.

The principal problem we need to overcome is, that wood is poor insulation, and heat moving to cold as the first law of thermodynamics, causes heat loss not only through the frame, but in all directions through any solid. Meaning heat is also lost down into the ground and up into the roof and on into the sky.

In winter heating areas, it is desirable to have the insulation on the room side of the frame, thereby stopping our heat reaching the frame and slipping away. The way to do this is to fix sheets of closed cell insulation on the room side of the frame, to cover the insulation with plastic sheet, followed by drywall which is then kept at room temperature, thereby being above dew point at all times and avoiding condensation.

This solution neatly gets round the usual problem, of poorly fitted water vapor barriers, where because water vapor molecules are so tiny they are able to pass through gaps that are too small to see, and they then do untold damage inside the frame.

May 1, 2014 4:22 PM ET

Water vapor and wood
by Derek Roff

I'm not in agreement with Roger Anthony, regarding his statements about water vapor and wood. Water vapor will enter (and leave) wood slowly, but relentlessly, regardless of the dew point. If humidity conditions are fairly stable, the wood will eventually reach equilibrium moisture content (EMC), where water vapor molecules are entering and leaving the wood at equal rates. If wood is drier than the EMC for the average temperature and relative humidity of the air surrounding it, then it will gain moisture, even if the wood and the air are constantly above the dew point.

In the wall of a house, the relative humidity and the temperature will vary with time of day, weather, and season. However, it remains true that staying above the dew point will not prevent moisture uptake by wooden framing members.

May 1, 2014 8:43 PM ET

Edited May 1, 2014 8:46 PM ET.

Poly in a double wall assembly-cold climate
by Matthew Michaud

I am building a double-walled home in northern Maine. I have come up with a wall that includes a dedicated poly air barrier that joins to the poly below the slab. It rests interior to 2" of r-board polyiso which is is found between my two Roxul-filled stud walls. The poly layer lies interior to ~R-35 and exterior to ~R-15, seemingly below the dew point region. Being on the warm side of the wall (more than 2/3 R-value inward), and having an HRV to control humidity levels, am I safe to assume that condensation on the poly won't be an issue? Thought of excluding the poly and just taping the interior polyiso seams to create my air barrier, but thought it would involves more labor, more expense (tape) and the chance for the seal to become compromised with settling of the house.

May 2, 2014 6:24 AM ET

Response to Matthew Michaud
by Martin Holladay

Polyethylene can be used safely the way you intend to use it -- because you live in such a cold climate.

May 3, 2014 2:37 PM ET

Better Building Materials
by George Hawirko

Building with Wood is a real problem we all seem to just tolerate, rather that suggest better materials that don't give us all the grief. For example, EPS Composites, that, Insulate and cladding without all those confusing layers.

May 3, 2014 4:54 PM ET

by Malcolm Taylor

Wood has been successfully used in human habitations for almost as long humans have been on earth. It is abundant, environmentally friendly and biodegrades when its lifespan is over. We are just having some teething pains over very recent changes in the way we insulate. No need to throw the baby out with the bathwater.

May 5, 2014 3:40 PM ET

The Case For Poly
by Ron Keagle

Building science recognizes the following methods for moisture accumulating in wall cavities:

1) Air leaks that carry vapor in the airflow.
2) Diffusion equalizing vapor pressure across a boundary.
3) Bulk water leaks from the exterior.
4) Cold sheathing adsorption.
5) Summertime reverse vapor drive.

In much of the discussion on this topic, items #1 and #3 are treated as though they are inevitable. Particularly item #1 is often framed as though it is nearly impossible to prevent. Therefore, due to the inevitability of wetting the wall cavity from these two moisture sources, the suggested remedy is to allow the wall cavity with all of the drying potential that is possible.

I prefer to build without air leaks or water leaks from bulk water intrusion. Building a wall cavity that can dry out bulk water intrusion seems like furnishing the interior with outdoor carpet and furniture in case the roof leaks. Superinsulated houses demand attention to details and high quality workmanship. Therefore it does not seem unreasonable to expect the flashing details to keep the water out.

Assuming that more vapor is likely to enter a wall cavity when carried airflow through holes than will enter by diffusion, we are told that preventing holes is more important than providing a vapor barrier that stops diffusion. But why assume that the two have to be mutually exclusive?

The often cited claim that diffusion is a minor vapor transfer mechanism is based on diffusion through a material. Whereas a hole through material allows diffusion through air or free equalization of differing vapor pressures. So a hole in a vapor barrier partially defeats the vapor barrier just as a hole in an air barrier partially defeats an air barrier. A hole in either barrier lets vapor pass while carried by airflow, and by movement through diffusion. Therefore, it is essential that either type of barrier be free of holes.

An air barrier that is free of holes stops vapor that is carried by airflow, but does not stop vapor transferred by diffusion. But a vapor barrier that is free of holes stops both airflow and diffusion.

I like the certainty of a poly vapor barrier. If installed properly, it stops air and diffusion. Couple it with proper flashing and exterior details, and the wall cavity stays dry. And just for backup, the cavity can dry to the exterior.

I agree that there is even more backup if the wall can dry to the interior, but the interior is a big source of moisture that wants to move outward much of the time. Rather than letting this moisture into the wall and countering that problem with the remedy that the moisture can dry back out the way it came in, I would prefer to keep it out of the wall in the first place. Otherwise, it strikes me as putting holes in the Titanic so any water that happens to get in can get back out.

So my approach is to use a poly vapor barrier to stop airflow, and diffusion; and to build without any opportunity for bulk water intrusion. I won’t have any holes for air, diffusion, or bulk water to enter the insulation cavity. That leaves only items #4 and #5 to worry about. If items #1-3 are eliminated, I don’t see why item #4 would be a problem. To be a problem, it needs a significant source of wintertime moisture, and without items #1-3, there is no such significant source.

That leaves only item #5 to be concerned about. Certainly, this has the potential to deposit moisture inside of the wall cavity if the outdoor dew point is high enough and the interior air conditioning is cold enough. Clearly, the hottest, humid summertime climates can provide the conditions for item #5 to become a problem. The only question is where the climate dividing line is that separates the areas where the problem is likely from the areas where it is not. Generally, it is said to be a problem everywhere except for the far northern regions of Canada.

However, there are other factors that influence whether item #5 will be a problem. I have no problem with it in central Minnesota. Part of the reason is that I do not cool the interior lower than around 75 degrees. If I liked air conditioning at 68 degrees, there might be some condensation inside the insulation cavities. But the dew point rarely gets as high as 75 degrees, and if it does, it is only for part of a day on a few days during the summer. Another point to consider is that there is some R-value between the warm-side vapor barrier and the conditioned living space. So a 75 degree air temperature in the living space will not cool the vapor barrier as low as 75 degrees. A service cavity between the living space and the vapor barrier will also add some R-value, further separating the vapor barrier temperature from the dew point. So I don’t have a problem with item #5.

Register for a free account and join the conversation

Get a free account and join the conversation!
Become a GBA PRO!