Blog Post

It's Time to Rethink the All-Glass Building

The highly glazed CORE Building on 7th Avenue in the Chelsea neighborhood of New York City. Upper floors are residential. Photo: Alex Wilson Click on image to enlarge.

The July 2010 issue of Environmental Building News asks whether we should end our love affair with all-glass buildings. The short answer is "yes." With most large commercial buildings there is an energy penalty associated with increasing the glazing area beyond 20 to 30 percent. Given today's glazing technologies, it does not make sense to create highly glazed buildings. That is especially the case with "green" buildings, where extra effort is being made to reduce energy consumption and environmental impact.

The energy penalty of more glazing

In researching and writing this article, "Rethinking the All-Glass Building," I got energy modeling support from Fiona Cousins, P.E., Scott Bondi, P.E., Ph.D., and Cameron Talbot from the New York City office of Arup, one of the world's leading engineering firms. I used that data to make the case that highly glazed façades don't make sense from an energy standpoint. As covered in detail in the full feature article, the Arup engineers modeled a ten-story, 100,000 square-foot commercial building in three cities, New York, Miami, and San Francisco with four different glazing types (single-glazed clear, double-glazed clear, double-glazed low-e, and triple-glazed low-e), three different building footprints (square, moderately elongated, and highly elongated), and four different glazing fractions (20%, 40%, 60%, and 80%), measuring the impact of these variables on annual energy consumption as well as peak heat and cooling demand.

Summary results of that modeling are shown in the graphs below. Neither exterior nor interior shading is assumed in this study, but lighting energy use and plug loads are factored in (i.e., highly glazed facades reduce lighting energy use).

As can be seen in the graphs, as the glazing area is increased, energy consumption and peak energy demand increases in all climates and with all glazing configurations. In colder climates, the energy penalty of increasing the glazing area is greater, and with higher-performance glazings the energy penalty of increasing the glazing area is lessened.

Annual energy consumption is compared for a square building in three different cities and with four glazing types as the glazing areas is increased from 20% to 80%. The steepest line (Type 1) is for single glazing. Graphs created by Amie Walter based on Arup data. Click on image to enlarge.

The increase in peak heating and cooling demand or load (a measure of how large the heating and cooling equipment has to be to maintain comfort) with increasing glazing area is even greater than the increase in annual energy consumption as the glazing area in increased. With a square building footprint and standard double glazing in New York, for example, going from 40% glazing to 80% glazing increases the peak heating demand by 48% and the peak cooling demand by 39%, while the same change increases the annual energy consumption by only 26%.


BuildingGreen relies on our premium members, not on advertisers. Help make our work possible.

See membership options »
These graphs show the impact of different glazing types and glazing percentages on peak cooling load (in tons) and peak heating load (in million Btus per hour) for New York and Miami. A square building footprint is assumed. Type 1 is single glazing. Graphs created by Amie Walter based on Arup data. Click on image to enlarge.

Arup also examined the impact of building shape (footprint) and orientation on energy consumption and peak loads as the glazing area is increased. While those graphs aren't shown here, they show that increasing the glazing fraction of an elongated building (with greater facade area) has a greater impact on energy consumption than it has with a square building. While the difference between a north-south orientation and an east-west orientation was not as great as I had expected, the north-south orientation (with the long sides facing east and west) results in greater annual energy consumption.

Making all-glass buildings work

The article devotes considerable attention to how we can minimize the energy penalty associated with increased glazing area. These strategies include:

    • Substitute insulated spandrel panels for glazing (this can retain the all-glass look).
    • Use the best spectrally selective, low-e glazings, and (where possible) specify different glazings for different orientations of the building.
    • Increase the number of layers of glazing.
    • Provide fixed exterior shading to control solar heat gain and reduce cooling energy use.
    • Provide exterior roller blinds or shades to control both solar heat gain and heat loss.
    • Provide automated, interior blinds to control solar heat gain (this is not as effective as blocking that solar gain on the outside of the glazing).
    • Use lightshelves and other features to bring daylighting deeper into buildings and keep that solar heat gain further from the façade zone.
  • Install dynamic glazing whose properties can change to regulate energy flow.

Changing aesthetic preferences

There are a lot of good reasons we like all-glass buildings: from the speed of construction of curtainwall assemblies to the fact that curtainwall manufacturers end up bearing responsibility (and liability) for those façades. But design and aesthetics are likely the dominant reasons. A woman architect I interviewed for the article even suggested that architects are drawn toward all-glass buildings for the same reason they are attracted to women's lingerie--the façades being "sleek, smooth, sexy, shimmering, simple--and simultaneously transparent and mirroring."

New York City's MetLife Building with the LEED-Platinum-certified Bank of America Building on the right. Photo: Alex Wilson. Click on image to enlarge.

In my conclusions to the article I quote architect Henry Siegel, FAIA, who calls the insistence on transparency "a real failure in leadership and vision in the design community." He calls for "broadening the definition of design excellence to include values other than aesthetics."

I agree, and I'd like to see the green design community champion a shift away from the all-glass building aesthetic--at least until heavily glazed facades can be created without any increase in energy consumption or peak heating and cooling loads.

I invite you to share comments on this blog. Do you agree with the general arguments in this article and, if so, how do we convince clients--especially green clients--that all-glass isn't the best way to go?

Alex Wilson is the executive editor of Environmental Building News and founder of BuildingGreen, LLC. In addition to occasional blogs relating to EBN articles, Alex writes two weekly blogs: Energy Solutions, and Alex's Cool Product of the Week. You can sign up to receive e-mails about BuildingGreen blogs, by filling in your e-mail address in the upper right corner of any blog page. You can also follow Alex's latest articles and musings by signing up for his Twitter feeds.

Published July 1, 2010

(2010, July 1). It's Time to Rethink the All-Glass Building. Retrieved from

Add new comment

To post a comment, you need to register for a BuildingGreen Basic membership (free) or login to your existing profile.


July 22, 2010 - 8:12 pm

The article and the comments are correct given the current limitations of glazing, but we are very close to having photovoltaic (PV) glazing that could allow for highly glazed exteriors which actually help save and create energy.

A system I have designed, Integrated Framing collects and distributes power and data through the curtainwall framing. By significantly reducing wiring to the building core and back, as well as the two-step rough-in and electrical installation, it makes PV glazing cost effective. It creates a "smart grid" on the building exterior, which can collect real-time weather data for fine-tuning the lighting, HVAC and other systems. By leaving the roof free of solar panels, it allows for a green planted roof which can filter and harvest rainwater for use in the plumbing systems. It has the potential to make Net Zero Energy Buildings a near term reality.

We are in the process of doing the energy and cost modeling now, using similar models which were prepared for the Department of Energy. DOE currently recommend using dynamic glazing in highly glazed facades to mitigate the problems that you mentioned. While they currently estimate that PV glazing is ten years out (mostly because they want it to have an r-10 rating), we believe it can be made with suitable vision and insulating characteristics and energy production capacity much sooner than that if the demand is demonstrated. Installations are already occurring in Europe, China and Japan in less temperate areas.

In the very near future, we hope to have the system available so the architect no longer has to choose between the aesthetic and functional qualities of the building.

July 15, 2010 - 6:37 am

If a fully glazed building without any solar protection can exist on paper, I don’t know any in the real life. The biggest glazed surfaces without solar protections are usually atrium (ok I have seen some public area such as mall or library without solar protection but none stands long before the investors came to see us to find a solution or close the space). Of course I consider a tree as a solar protection. A surrounding building casting a shadow is also a solar protection.
So the assumption is quite strong and unfortunately limits the conclusions. I hope they will write another article that takes into account the solar protections this time. True the study is going to be more complex by far. Solar protections are becoming more and more technical. Fabrics have coatings to bounce back a large part of the sun spectrum. Venetian blinds can bounce the light on the ceiling to extend the passive zone. More interesting, automation optimizes the dynamic passive strategies and allows architects and engineers to design genuine bioclimatic façade.
When I lecture on why green buildings are so important and why everything starts with a well-designed façade, I compare the envelope of a building with our outfits. Some people are living in paradise but for most of us we dress according to the seasons, the weather and the activity we have. A building “is living” outside so it has to cope with the same seasons and weather. Moreover, each and every building is used a unique way that is usually periodic but evolves 24/7. So for the same reason we need several clothes, the envelope of the building must have dynamic specs. And that’s the different dynamic parts of the envelope, starting with the solar protection that allows that.
Now coming back to the study, it means that not only different kind of solar protection should be evaluated, but different automation systems also have to be evaluated. Also the surrounding will have an impact. If there’s a manual mode, it can have an even bigger impact. Last but not least the comparison is tricky as comparing trousers and skirt or pullover and shirt!


July 15, 2010 - 10:31 am

It's interesting to see the mental hoops folks will jump through to try to convince themselves that the findings of this article don't apply to their work.

In fact, the results are clear and obvious. We use too much glass. There might not be many buildings that are truly "all" glass, but I see new proposals for 60-80% glazed facades almost constantly. And there is a myth that we need all this glass for daylighting. Of course the glass boxes look good on the magazine covers, but it's time to start focusing on what really works.

July 14, 2010 - 3:05 pm

While I agree that "all glass buildings" should be a thing of the past, the analysis you used was overly simplified to understand how to optimize designs. The article states " Neither exterior nor interior shading is assumed in this study" which means that 100% of possible solar gains are impacting every window. This is simply not realistic, given obstructions from other buildngs, self-shading, trees, awnings, and interior operable blinds, which are ubiquitous. This assumption greatly overemphasizes the cooling load impacts of the glass, while also misunderstanding issues of glare from sun penetration and how occupants will respond by closing blinds or compensating with electric lighting under glaring conditions. These types of studies which ignore naturally occuring shading have lead ASHRAE and other code bodies to over value SHGC compared to the visible light transmittance of glass, and are creating institutionized impediments to wider use of daylighting in buildings.

In a recent field study of 61 daylit spaces around the US, we found that on average the windows had 28% of their area shaded by exterior obstructions. We also found that 93% of the windows had operable blinds installed, and 74% seemed to be actively used by the occupants. We are currently working on trying to get better quantification of existing shading conditions for a large population of real office buildngs in California, so we can help others do a better job of modeling typcial building conditions and understanding the trade-offs between heating, cooling and lighting in facade design choices.

I hope you continue to follow this discussion and refine your studies so that we can optimize facade design for realistic conditons, not just worst case conditons.
Thanks for launching the conversation.
Lisa Heschong, architect, Heschong Mahone Group, Sacramento, CA.

July 14, 2010 - 8:10 am

Thank you for an outstanding article on an important topic! The analysis is balanced , the graphics are very helpful, and as someone has already said, this clear, concise article will help support many discussions about thermal performance of buildings.

My one comment is that the article does not state what U-factors were assumed for the glass or the opaque wall. As far as I know, the best triple glazing available can achieve a U-factor of approximately 0.125, which may be increased somewhat if framing is accounted for. From the plots for energy consumption, it would appear that the opaque walls were assumed to have a U-factor not much less than the glass. A U-factor for the opaque wall of, say, 0.10 would be roughly equivalent to R-10, which is not a very high-performing wall assembly. With 3 inches of extruded polystyrene insulation (and minimal thermal bridges), we could achieve R-15, in which case the difference between the glass performance and the opaque wall performance would lead to a steeper plot meaning that there would be more of a penalty for a high percentage of glass.

Could the authors state what U-factors were used in the modeling?

Assuming that what I have said above is correct and relevant, it would be helpful to add an "addendum" to this article in the next issue of EBN, stating the U-factors used in the modeling accompanied by an explanation of how increasing the thermal performance of the opaque wall has an impact on the results of modeling like this.

Thanks again.

July 26, 2010 - 6:34 am


Great news. I'll install it next to my affordable reliable fuel cell, and my Mr. Fusion home generating system.

While we wait patiently for all these miracle products, can't we all agree to use less glass? It's cheaper, it's more efficient, and it works right now. What's so hard?

August 12, 2010 - 8:58 am

This is indeed a very wise article to save the energy of world. This broadly described statistics is very usefull. I never think like that the glazed building could be harmfull for the nature. Its looked so nice that people cant think beyond it that it can harm. In todays world our main target is Go Green. I’m also in the favour of green revolution around the world. Here for the energy saving one thing I can suggest the front of the every office could be south faced. So it can avoid the scorching heat of the sun. It can save huge amount of enery by this process also.

July 4, 2010 - 3:11 am

Thank-you for a long-overdue reminder... I very much agree with the comment that this is a failure of vision and leadership in the design (and I would add, real-estate) communities. Its clear to me that all-glass facades are relics of an other era, when energy use was without cost and mechanical conditioning was the golden solution to every problem and desire...

I would like to further underscore the importance of exterior sunshading, tuned to facade. Adding sunshading to the elongated south-facing tests would have made them the best performing of all. Further making the sunshading into a lightshelf, or other light-redirecting device, would reduce lighting loads, thereby reducing cooling loads, and would capture warmth in winter without glare... I would refer all the architects to the work of William Lam in the early '80s – a tremendous resource.

I think its also worth emphasizing for my fellow architects who might apply this article to smaller buildings: the tests were done on an internal-load dominated case, which dampens some of the negative energy impacts compared to a skin-load-dominated building. These negative impacts will be more dramatic for a small building like a single-family residence.

Thanks as always for a tremendous resource-

Ryan Enschede, architect, NYC

July 1, 2010 - 12:26 pm

Thank you for finally taking on this issue. I've lost this argument with architects many times in the last 20 years.

July 1, 2010 - 7:10 pm

"In colder climates, the energy penalty of increasing the glazing area is greater, and with higher-performance glazings the energy penalty of increasing the glazing area is lessened."

Pardon me, but this is quite self-evident. Arep does great work, and EBN is a great publication, but there isn't anything stated here that hasn't been common knowledge for decades. The real issue isn't that better glazing improves performance -duh - but that poorly designed building are being built anyway.

Why and what to do about it might be more salient treatment.

While correct, those models were simplistic even by residential standards. When Arep models a commercial project, each glazing orientation can be spec'd for SHGC, U value, Visible Light, etc, and it all can be correlated not only with heat and cooling loads, but with lighting loads, daylighting, etc. They are really good at it.

So I suggest what matters would to show a properly designed building, and it's HVAC loads as a monthly cost, vs the same size and shape building poorly specified, and the difference in monthly costs shown as a percentage on likely rental gross profit based on the size.

That is what a commercial developer cares about. The architect catches their eye with a pretty picture, but if the engineer fattens their wallet, things will change.

July 2, 2010 - 7:48 am

the north-south orientation (with the long sides facing east and west) results in greater annual energy consumption. This does not make sense - the short sides should face east and west

Also the reason why architects like glass buildings is that they are no longer capable of elegantly articulating a facade - they arent taught how to do it or event the language of how to do it - most of them wouldnt have a clue how to go about it, which is why so many modern buildings, even if they do have proper windows and walls dont look pretty.So please teach articulation of facades in design courses - then we might get some.

July 2, 2010 - 10:58 am

On the orientation issue, I think we're saying the same thing. I argued that orienting the building on a north-south axis with the long sides facing east and west results in higher energy consumption--a bad thing. We agree that, instead, the short sides should be facing east and west to minimize energy consumption (and reduce the negative impact of a higher glazing fraction).

July 2, 2010 - 10:51 pm

Susan and Alex:
North-South orientation does increase heat gain in Saudi Arabia, where I live and work. Midday high-angle sun isn't much of a problem - it's easily taken care of by good roof insulation and exterior shading; but dealing with heat gain from low-angle sun through acres of east and west-facing glazing is a real pain.
In winter, in Northern climates, an E-W orientation maximizes midday heat gain (on the South side) to reduce heating energy.
However, for millions of years the ants of tropical Northern Australia have been building "magnetic" anthills aligned N-S for temperature control reasons: taking more heat from the sides in the morning and afternoon and less from the midday sun allows the single-zone anthill to maintain a steadier temperature throughout the day, whereas E-W orientation would produce a big temperature fluctuation. Termites also use blocking and unblocking of ventilation "chimneys" to help.
For us, building E-W is better because our buildings are typically not one zone but many and separate; and instead of living with solar gain as the ants have had to, we fight it by all means available in order to satisfy - in many cases - Architectural aspirations.
Sorry - I'm a Mechanical Engineer and, like Christopher, have had many arguments with Architects over the years.