click to enlarge
This diagrams the temperature profile between a triple-glazed IGU and a double-glazed IGU with a room-side low-e coating. The IGU with the room-side low-e coating has a much lower surface temperature and this can create an issue with downdraft. Image: Payette  


Windows are an invaluable architectural feature to lab design. They provide daylight and views to the exterior. Many studies support the benefits of windows on occupant mood and productivity. However, windows are five to eight times less resistant to heat transfer than a solid wall, and while the insulative values for windows are improving, they still pale in comparison to those for a solid wall. This poses a two-fold problem—energy loss and the effect of thermal comfort on the occupant. The standard measure to compare a window’s thermal performance is U-value or U-factor. This value measures the heat transmittance through a window. A lower U-value equates to a lower amount of heat transferred through the window.

A single pane of glass provides almost no resistance to heat transfer while a double-glazed insulated glazing unit (IGU), consisting of an air gap sandwiched between two panes of glass, is much more resistant to heat transmittance. Owners and architects now routinely consider triple-glazed IGUs especially in cold climates; however, the additional pane of glass increases the cost and weight of the unit. In the current market, a double-glazed IGU with a low-emissivity coating on the inner surface is becoming more common and maintains the same U-value as a triple-glazed IGU. Do equivalent U-values create comparable windows in regards to thermal comfort?

Heat transfer and thermal comfort
There are two modes of heat transfer pertinent to the discussion here: radiation and convection. They are the basis for thermal comfort of an occupant within a space. While you sit and read this article, you radiate heat to your surroundings, if they are colder than you are. Now let’s place you next to a window with a poor U-value on a winter’s day when it is 15 degrees outside, you will lose more radiant heat to this window than to a window that has a better (for example, lower) U-value and does not transmit as much heat.

Not all surfaces are created equal in terms of radiation, the factor that differentiates them is known as emissivity. Some materials have high emissivities and absorb and re-emit most of the radiant energy received. Uncoated glass falls into this category and it re-emits 90% of the radiant energy it receives from an interior space. On the opposite end of the spectrum are metals, which can absorb and re-emit only 3% of such radiation. The low emissivity of aluminum foil is the reason we wrap food in it to keep it warm.

It is easy to see the benefit of applying a thin film of a material that has a low emissivity to a glass surface. Typically these ‘low-e’ coatings are metal oxides that allow visible light to transmit through but emit the radiant heat back into the space. These coatings tended to be unstable to outside air and were enclosed within the IGU in order for them to remain effective over the window’s lifetime. However, in recent years with better scratch resistance of the coating, it is becoming more common see an additional low-e coating in the exposed interior surface of the glass, known as a room-side low-e coating.  

The second mode of heat transfer affecting thermal comfort is air convection. As air temperature drops, its density is increased, making it fall. In looking at the window example on a winter’s day, when warm interior air hits the cold surface of the glass, it is cooled, causing convective air currents to cascade along the glass surface and onto the floor or writing surface also causing occupant discomfort.

The challenge of room-side low-e coatings
Recently, a project team at Payette decided between a triple-glazed IGU versus a double-glazed IGU with a room-side low-e coating. The team specified a triple-glazed IGU for the design of a lab building, and as a cost reduction strategy, a double-glazed IGU with a room-side low-e was presented as a comparable alternative. Since both U-values were almost equal, was the double-glazed IGU with room-side low-e an acceptable substitution?



Center of Glass U-Value (Btu / (hr*ft2*F)

Double-Glazed (Argon filled) with room-side low-e


Triple-Glazed (Argon filled)



To answer this question, we revisited the different modes of heat transfer. The room-side low-e coating is effectively like placing a piece of aluminum foil over the window. An occupant will not lose radiant heat to this glass surface, and from a radiative comfort standpoint, the double-glazed IGU with room-side low-e is the clear winner.


click to enlarge
This is a computational fluid dynamic (CFD) study completed for an office in a lab building and shows the increased air speed and lower air temperature along the window’s interior surface. Image: Payette  


We also looked at the issue of convection. Because the low-e coating is effectively reducing the heat transfer between the interior and the glass IGU, the surface temperature of the room-side low-e coated glass is lower than a piece of uncoated glass that is more effective in absorbing radiant heat from the space. To illustrate this point (Figure 1), we mapped the temperature profile through the two different types of IGU’s setting the exterior temperature at 15 F and the interior design temperature at 72 F. The interior surface temperature of the glass with room-side low-e is almost 10 degrees cooler than the triple-glazed IGU.

This lower surface temperature creates a larger potential of downdraft occurring along the glass surface. A computational fluid dynamic (CFD) study illustrates this fact (Figure 2) showing the increased air speed and lowered air temperature along the glass pane and floor surface, which can cause occupant discomfort.


click to enlarge
This shows three different typical glazing scenarios and how design parameters and an addition of a room-side low-e coating may affect thermal comfort of occupants. Image: Payette  


Another cause of concern is the risk of condensation. In a humidified environment at 70 F/50% RH, the dew point is 51 F. Depending on the climate, the interior glass temperature of the room-side low-e coating could be close to or below the dew point. Therefore, when designing a humidified environment, a room-side low-e coating should be carefully evaluated due to potential condensation.

What began as a simple question of switching from one glazing type to another creates an interrelated and complicated set of issues. The following case studies demonstrate how the addition of room-side low-e might affect the design of a space.  

Payette’s office (Figure 3a) has punched windows that are 4’-0” x 7’-0” tall. Since desks are located adjacent to the exterior wall, downdraft creates cool convective air currents that flow onto the desk surface causing people’s hands to feel cold. In an office that has floor-to-ceiling glazing but where the desk is located far away from the window (Figure 3b), limiting the radiative heat transfer with a room-side low-e coating could be beneficial, since thermal discomfort due to downdraft is less of a concern. In a larger gathering space (Figure 3c), a room-side low-e coating would increase the risk of discomfort due to downdraft since the cool air currents have a greater distance they can travel. One potential design solution is to create internal horizontal obstructions that break up the cascading air currents.  

We are currently developing a tool that will take different design parameters, such as window height and distance of an occupant from a window, and calculate which issue, either radiation or convection, will be the over-riding thermal comfort issue. This can help define a set of design parameters for the architect, engineer and owner to work within. The U-value does not tell the whole story in understanding the thermal comfort issues surrounding a window and investigating additional design considerations are required to select the best glazing product.

Extra: Can sustainable design be cost effective?