A Concrete Problem For Concentrating Solar

An article on Solar Reserve's Crescent Dunes solar project, now under construction in the desert outside Tonopah, Nevada, offers an object lesson in what the engineers and climate scientists refer to as "embedded carbon" -- the amount of CO₂ produced by building renewable energy power sources that must be "paid back" for the facility to achieve carbon neutral status.

The opportunity to examine the embedded carbon issue came with Herman Trabish's article on his visit to the Crescent Dunes project, which visit included an opportunity to climb the project's power tower. That tower, still under construction, will be the world's tallest such solar power tower when completed at 640 feet -- at least until BrightSource gets some of the 750-foot towers now on its drawing board built.

One thing is quite plain in Trabish's coverage of Crescent Dunes: the plant is using a fair amount of concrete, which carries with it a significant CO₂ burden.

Crescent Dunes' site manager Brian Painter told Trabish that the power tower holds 9,100 cubic yards of concrete. (A cubic yard, for those of you who've never bought construction materials in bulk, is a volume equivalent to a cube three feet on an edge. The unit is usually shortened to "yard" in informal use.)

9,100 yards of concrete is quite a bit. But it's nothing compared to what may well turn out to be holding up the heliostats.

Solar Reserve's plan is notable for the concrete pads holding up its large heliostats, each of which spans 34 by 37 feet and weighs three tons. Holding up that heavy a mirror takes a sturdy foundation, and each of the project's 10,300 heliostats will sit atop a triangular pad which supports a cylindrical pedestal, both made of concrete.

It's hard to find specific specs on Crescent Dunes' heliostat mounts: the specific design wasn't nailed down as late as 2011. Given Trabish's photos it seems safe to estimate the triangular pads are nine feet on an edge, and the pedestal about three feet in diameter and ten feet tall. Assuming the pad's depth is around three feet, which would seem likely for a pad holding up three tons of flat mirror subject to strong winds, that works out to around 7.5 yards of concrete per heliostat -- admittedly, a rough estimate.

We welcome Solar Reserve's corrections here, if they have them to give. But if that ballpark figuring is close, mounts for all 10,300 of Crescent Dunes' heliostats would use around 77,250 yards of concrete. With the power tower that's 86,350 yards.

Concrete contains a number of materials including sand, gravel, water, and cement. each has its own CO₂ footprint, but cement's is the highest in terms of CO₂ per pound -- according to the EPA, making a pound of cement creates a pound of CO₂, give or take ten percent. Not only must heavy cement be transported, but the process of making it involves heating calcium carbonate and other minerals in kilns to temperatures of over 2700°F. Cement kilns are major contributors to the atmosphere's CO₂ burden.

The National Ready Mixed Concrete Association (NRMCA) points out that concrete's overall CO₂ burden depends on the proportion of cement used in the mix: that proportion can range from 7-15 percent. Depending on that percentage, says NRMCA, concrete's CO₂ contribution can range from 170 to 500 pounds per cubic yard used.

If Crescent Dunes' concrete runs toward the 170 pounds of CO₂ per yard end of things, that would add up to 14.7 million pounds of CO₂ contributed to the atmosphere. If it's more at the 500 pounds per yard, that's 43.1 million pounds of CO₂.

Of course the whole point of moving to solar energy is to displace energy sources that put CO₂ into the atmosphere. When building a solar energy plant puts CO₂ into the atmosphere, it takes some time to "pay down" that CO₂ "debt" by putting out carbon-free energy before it can be considered a net benefit to the climate.

As it happens, Blue Sky Model has calculated that across the U.S., generating a typical kilowatt-hour of electrical power contributes an average of around a pound of CO₂ to the atmosphere. That certainly makes the math easier. Crescent Dunes' concrete will (by ReWire's estimation) have contributed between 14.7 and 43.1 million pounds of CO₂ to the atmosphere, and so the solar plant will have to create between 14.7 and 43.1 million kilowatt-hours of electrical power to the grid before it pays that debt off.

Crescent Dunes will have a capacity of 110 megawatts. Even with its molten salt storage component, which is intended to allow the plant to produce power while the sun's down, estimates of the plant's actual output over time as a percentage of that capacity -- its "capacity factor" run around 52 percent. For our purposes, that makes the plant the equivalent of a 57-megawatt plant that runs 24/7.

A megawatt is a thousand kilowatts. 14.7 million kilowatt-hours is 14.7 thousand megawatt-hours. Crescent Dunes will, on average, generate 57 megawatt-hours of power every hour. That works out to about ten days and 18 hours worth of production to pay off the CO₂ debt from the cement in the plant's concrete at the optimistic end of the CO₂ spectrum, and 31 days and 12 hours -- a long month -- to pay off the more pessimistic end, assuming the plant works exactly as efficiently as its designers hope it will.

A month to pay back the CO₂ from the cement in the plant's concrete doesn't seem like a whole lot held up against a 25- or 30-year operating lifespan. Of course, the cement in the concrete is just one of the ways in which building the plant is putting CO₂ into the atmosphere. Steel is a major contributor to most construction projects' carbon footprint, putting about two pounds of CO₂ into the air for every pound of steel used. Glass (as in mirrors) is a bit lighter on the carbon footprint than steel: about .6 pounds of CO2 for each pound of glass produced. Transportation of materials and crew is another contributor, as is construction of transmission lines to serve the remote plants. All told, the construction carbon footprint of many large desert solar plants can take a surprisingly long time to "amortize" with the resulting carbon-free power.

That's especially true considering that that pound of CO₂ per kilowatt hour is a national average, and some utilities -- including the one that serves most of California -- come in substantially lower that that. Pacific Gas And Electric (PG&E), according to Blue Sky Model, averages one twentieth of a pound of CO₂ produced per kilowatt-hour generated. PG&E will be buying power from Solar Reserve's planned Rice Solar Energy Project east of Joshua Tree National Park: if Rice Solar's cement-related CO₂ is similar to that of Crescent Dunes, it could take close to two years for the plant to pay down that CO₂ debt, as it'll be replacing PG&E's less-carbon-intensive power.

Which is not in and of itself a slam-dunk reason not to build remote solar in the desert: after all, as we've reported recently, rooftop solar has its own built-in CO₂ footprint.

But it's certainly something that ought to be taken into account openly when setting our solar course over the next few decades.