TODAY’S STUDY: WHERE AND HOW TO DEPLOY A SOLAR HYBRID

From Plato’s Symposium –

...For our original nature was by no means the same as it is now. In the first place, there were three kinds of human beings, not merely the two sexes, male and female, as at present: there was a third kind as well, which had equal shares of the other two, and whose name survives though the thing itself has vanished. For ‘man-woman’ was then a unity in form no less than name, composed of both sexes and sharing equally in male and female...

...Now, they were of surprising strength and vigor, and so lofty in their notions that they even conspired against the gods...

...Zeus, putting all his wits together, spoke at length and said: ‘Methinks I can contrive that men, without ceasing to exist, shall give over their iniquity through a lessening of their strength. I propose now to slice every one of them in two, so that while making them weaker we shall find them more useful by reason of their multiplication; and they shall walk erect upon two legs...

...Parts he now shifted to the front, to be used for propagating on each other—in the female member by means of the male; so that if in their embracements a man should happen on a woman there might be conception and continuation of their kind; and also, if male met with male they might have satiety of their union and a relief, and so might turn their hands to their labors and their interest to ordinary life. Thus anciently is mutual love ingrained in mankind, reassembling our early estate and endeavoring to combine two in one and heal the human sore...Each of us, then, is but a tally of a man...

...The cause of it all is this, that our original form was as I have described, and we were entire; and the craving and pursuit of that entirety is called Love. Formerly, as I have said, we were one; but now for our sins we are all dispersed...

...This applies to the whole world of men and women—that the way to bring happiness to our race is to give our love its true fulfillment: let every one find his own favorite, and so revert to his primal estate. If this be the best thing of all, the nearest approach to it among all acts open to us now must accordingly be the best to choose; and that is, to find a favorite whose nature is exactly to our mind. Love is the god who brings this about; he fully deserves our hymns...

Specialization is thought to be an essential to success. Photovoltaic (PV) solar panels are specialized for rooftops, concentrating solar power (CSP) mirror technology is specialized for solar power plants and Paris Hilton is specialized for tabloids.

So what is one to make of Concentrating Photovoltaic (CPV) technology? It is neither entirely PV nor entirely CSP yet its success is on the rise.

Perhaps, society is moving toward another Platonic time: This is an age that rewards multitasking. From working people who combine two careers to cars with two drivetrains, versatility has never been more valued. CPV is the Prius Hybrid of solar energy, the image of a humanity whose partners have found one another.

CPV uses CSP technology but instead of concentrating heat, it concentrates the sun’s light (photo) to maximize the electricity output (voltaic) of a PV array. As reported in the study highlighted below, it combines strengths and minimizes weaknesses.

Because CPV concentrates a sunbeam on a very small bit of light-active material, CPV minimizes two of solar energy’s most troublesome negatives. First, it reduces the cost of solar energy-generated electricity because the light-active material is the most expensive part of a solar cell. Second, because those light active materials are the source of the largest portion of greenhouse gas emissions and other environmentally toxic by-products from solar cell production, CPV minimizes the harmful effects of solar panel manufacturing.

The result is a more price-competitive solar product, both in up front and external costs.

Like any hybrid, CPV’s other advantages are relative to the technology it replaces. It is more land efficient than standard PV arrays but less so than CSP plants. It requires less water than CPV plants but no less water per panel than PV – but CPV arrays have fewer panels than PV arrays, making it more water efficient by that measure.

Because the sun falls everywhere, on the brave and the timid, the bright and the dull, the quick and the dead, there probably is no universal solar solution. That answers are contextual is not news. As Albert Einstein famously pointed out, some things are relative.

This good earth offers sun and wind and deep heat and flowing waters for the harvesting. With each opportunity, there are challenges.

There is nothing wrong with specializing to meet one of those challenges and seize the opportunity. Often, however, the solution requires combining existing technologies to come up with something better.

As Plato observed, two heads are better than one.

An Assessment of the Environmental Impacts of Concentrator Photovoltaics and Modeling of Concentrator Photovoltaic Deployment Using the SWITCH Model
Daniel Kammen with James Nelson, Ana Mileva and Josiah Johnson, June 2011 (Renewable and Appropriate Energy Laboratory/CPV Consortium)

INTRODUCTION

The environmental and societal benefits of deploying renewable energy technologies at utility scale must be considered alongside the concomitant costs and alternatives in order to properly evaluate the social return on investment of each technology.

The benefit of evaluating the environmental impact of a technology before large-scale deployment cannot be stressed enough. The United States wind industry has learned difficult lessons from its deployment of wind turbines at the Altamont pass in California, where windmills have been found to kill at least one bird per year per turbine (Ritter 2005).

Had there been a proper environmental impact study of the area, 4000 turbines would not have been sited in an important bird migration route, and the wind industry would not have received negative press surrounding the harmful environmental impacts of a prominent green technology. Mitigation efforts for new wind projects such as using radar to detect flocks of birds and furl turbine blades are now underway (Iberdrola 2009), but this type of technology could have been used from the inception of wind deployment.

The first part of this report touches on important environmental areas that must be considered when deploying Concentrator Photovoltaics (CPV). It does not attempt to evaluate the best sites for CPV development on an environmental basis. Rather, CPV is compared to other solar technologies and more broadly, to other electric power generating technologies with respect to key life cycle environmental metrics.

In the second part of this report, the possible future deployment of CPV is investigated using UC Berkeley’s SWITCH model, and the emissions benefits of including CPV in the future Western United States electric power system are discussed.

click to enlarge

AN ASSESSMENT OF THE ENVIRONMENTAL IMPACTS OF CPV

To accurately portray the environmental impact of any technology, all impacts from inception to retirement must be taken into account. Life Cycle Assessment (LCA) methodology considers three distinct phases in the life cycle of CPV: (1) fabrication of CPV modules and deployment in the field on two-axis tracking systems (2) energy production (3) recycling and disposal at end of life. Here, four LCA environmental impact metrics are discussed in the context of CPV: energy, emissions, water use and land use.

click to enlarge

Embodied Energy and Emissions

The production of photovoltaics is an energy-intensive process. As most current forms of energy-intensive processes use greenhouse gas (GHG) intensive fuels, it is important to quantify the effect of the production of photovoltaics on our energy supply and on the stock of GHG in our atmosphere. The LCA community refers to the energy used and GHGs emitted in the production and disposal of a product as ‘embodied energy’ and ‘embodied emissions’ respectively.

By concentrating sunlight on highly efficient photovoltaic material, CPV systems minimize the amount of active photovoltaic material that must be mined, refined and purified into the final device. However, additional components related to light concentration and sun tracking must be included in CPV systems, thereby making the net embodied energy and emissions of light concentration in photovoltaic devices uncertain. Here we review LCA literature on CPV embodied energy and emissions and compare the results to other electric power generators.

A dominant LCA energy metric is the Energy Payback Time (EPBT), which denotes the time in years it takes for a technology to produce as much energy (net) as it takes to create and dispose of the device. EPBT is a measure of energy efficacy - for an energy technology to be a worthwhile investment from an energy production perspective, the EPBT should be much less than the lifetime of the device. In the past, the fast-paced solar industry has been plagued with outdated literature values of EPBT in the range of 3 - 11 years for a technology with a lifetime of 20 - 30 years (Alsema 1998, Alsema 2007), leading to fallacious conclusions that solar energy doesn’t warrant deployment due to large energy demands in production. Figure 1.1 shows recent EPBT values for a range of solar technologies, all of which are less than or equal to two years.

click to enlarge

EPBTs are calculated (eq. 1) by adding up all energy used in fabrication and installation of an electric power device, as well as disposal/recycling at the end of life, and then dividing this Cumulative Energy Demand (CED) by the yearly net energy during operation. The yearly net energy during operation is expressed in units of primary energy per year, thereby giving the EPBT in years.

The conversion from yearly net electricity generated by the device PGeneratedNet to primary energy terms is accomplished by dividing PGeneratedNet by the efficiency of electric power grid at converting primary energy into electricity at the site of deployment of the device. This conversion represents the input energy that would have been used to create a unit of electricity from other electric power generators, had the device in question not been installed. The primary energy used in operations and maintenance PO&M is subtracted from the denominator to obtain the yearly operational net energy.

Embodied GHG emissions are calculated by adding up all GHGs emitted throughout the life cycle of an electric power device and then dividing by the total electricity produced by the device, giving units gCO2-eq/kWh. EPBT and embodied GHGs are plotted in Figure 1.1 for a variety of solar energy technologies. Figure 1.2 puts LCA GHG emissions from solar technologies in the broader context of other electric power generators.

click to enlarge

…[T]he EPBTs of CPV systems are comparable with those of non-concentrator PV systems. CPV systems have EPBTs of 0.7 to 1.9 years, whereas non-concentrator PV systems have EPBTs of 0.8 to 1.8 years. This broad range of EPBTs for both non-concentrator and CPV technologies reflect differences in methodology and scope of each LCA study; consequently comparisons as in Figure 1.1 should be taken with a note of caution. One of the largest sources of inconsistency between studies is the site at which each solar system is assumed to be installed, causing the incident insolation and hence power production to vary by up to 30 % from study to study. The power conversion efficiency from primary energy to grid electricity also varies significantly from site to site.

The embodied emissions of CPV systems are higher than those of most non-concentrator PV systems. This is primarily due to the tracking system necessary for CPV technology, which contains large amounts of GHG-intensive steel. Redesigning tracking mechanisms to reduce wind susceptibility and thus the need for steel could help to reduce the GHG footprint of CPV.

CPV modules have made large gains in efficiency in the past three years, relative to other solar technologies (Hartsoch 2011). Many of the studies cited in Figure 1.1 were performed in the 2008-2010 timeframe, and therefore the efficiency of CPV systems in these reports is lower than is found today. While a complete life cycle assessment with using these new efficiency values is out of the scope of this work, it is clear that these efficiency gains will translate into reduced EPBTs and GHG emissions for CPV.

click to enlarge

Owing to the steel tracking systems, CPV systems are consequently heavier than non-concentrator PV systems and therefore require more energy to transport, incurring more GHG emissions along the way. Transportation can account for up to 20 % of GHG emissions along the CPV supply chain (Reich-Weiser et al. 2008), a value that has recently been reduced by collocating CPV component manufacturing sites with areas of high direct normal insolation (Phoenix Business Journal 2009, Amonix 2010).

Changing the electricity supply mix during CPV system manufacturing could also significantly decrease the embodied energy and emissions of CPV. Roughly one quarter of life cycle GHG emissions originate from electricity used in CPV module production (Reich-Weiser et al. 2008). If instead of manufacturing CPV modules using the average emissions of the electric power system – a GHG emissions-heavy system which in the United States is fueled in large part by coal and natural gas – the CPV panels were instead manufactured using electricity generated by existing solar power plants, the emissions attributed to electricity used in module production could be reduced to a fraction of the current value. This concept is known as the PV breeder concept (Fthenakis et al. 2008).

The embodied energy of solar thermal Concentrating Solar Power (CSP) systems has received much less attention than that of photovoltaics, but recent estimates put the EPBT of a parabolic trough system with six hours of thermal storage at 1.06 years (Burkhardt et al. 2010). While both the EPBT and GHG emissions values for solar thermal (without natural gas backup) are within the range of estimates for CPV, other environmental factors such as water use must be taken into account when considering solar thermal systems. Using dry cooling technology in the above solar thermal plant raises the CED by 8% and the GHG emissions by 7%. A comparison of water use of these plants can be found below.

click to enlarge

Other Embodied Emissions

LCA studies of solar technologies are in the nascent phase of incorporating emissions of substances other than GHGs (e.g. cadmium) into their sustainability metrics (Fthenakis et al. 2008). A recent study compares CPV to multicrystalline PV using one unified metric that includes fossil fuel depletion, global warming potential, water and air pollution, acid rain, etc (Nishimura et al. 2010). It was found that CPV has roughly double the environmental impact of multicrystalline PV, with most of the added environmental stress coming from embodied pollutants in the tracking system. Strategies for reducing the environmental intensity of the CPV tracking system are discussed above.

click to enlarge

Water Use

Water use represents an important environmental impact of electricity generation, especially in the context of growing demand for water and ever more limited supplies (Fthenakis and Kim 2010B). Currently 41 % of all water withdrawals in the United States come from electricity generation (Burkhardt et al. 2010). Water is consumed prior to power plant operation during the energy-intensive manufacturing of power plant components and is also used by thermal power plants during power production for cooling. As manufacturing and power production may not be collocated, the local effect of water withdrawal may differ substantially between these two phases.

For solar thermal generation technologies, water usage for cooling purposes during electricity production may be an important limiting factor to tapping high quality solar resources, as these tend to be located in arid areas with severe water shortages…

Upstream water use…Water use during power plant operation…

click to enlarge

Land Use

Solar power has long been criticized for using vast amounts of land relative to conventional generation sources. When mining and transportation and disposal of non-renewable, conventional fuels are taken into account, the land requirements for solar are comparable to those of non-renewable fuels. Life cycle land use is well covered…

The decreased land demand for CPV systems with respect to non-concentrator, ground mounted PV systems is due to the high power conversion efficiency of CPV. CPV systems require more land per square meter of module area than non-concentrator PV due to spacing requirements imposed by the two-axis tracking system, but this effect is more than offset by CPV’s high power conversion efficiency, leading to lower overall land use…

click to enlarge

MODELING OF CPV DEPLOYMENT USING THE SWITCH MODEL

Overview

• SWITCH is an electric power system capacity expansion model of Western North America that plans long-term grid investments while minimizing the cost of electricity in a given policy context.

• We used the SWITCH model to project how CPV could be integrated into the grid. SWITCH is a good tool for this evaluation because it considers many factors necessary for integrating renewable energy sources. These factors include:

- Matching hourly intermittent power output with hourly load

- Optimizing the location of renewable energy sites with respect to the grid

- Building traditional generators to “firm up” intermittent power output

- Building new transmission to move renewable power to loads

- Planning grid operations to fully use available intermittent energy

• Our results show that:

- It would be economical to install between 12 and 43 GW of CPV by 2030 in the United States Desert Southwest

- Including CPV allows for deeper CO2 reductions in the electric power system

- CPV displaces natural gas generation on the margin

- Strong carbon policy increases the deployment of CPV…

click to enlarge

SWITCH CPV Results

CPV is included in the optimal power system at any cost on carbon emissions investigated here, starting in 2022 and continuing through 2029. This demonstrates the economic viability of CPV as a power generation technology in the WECC, subject to CPV achieving future capital cost targets. CPV outcompetes rooftop and central station PV...to achieve deployment in the United States Desert Southwest, with 12 GW installed by 2026 absent a price on carbon.

The removal of new coal generation from the optimal power mix at $40/tCO2 represents a large opportunity for CPV, as the amount of CPV depolyed increases to 43 GW by 2026, generating 12 % of the WECC’s electricity between 2026 and 2029. CPV is included in the optimal power mix to serve load throughout Nevada, Arizona, New Mexico and Northern Baja Mexico, as well as to meet Califorina’s high demand for renewable resources via imports from the surrounding states (Figure 2.4).

click to enlarge

...[H]our by hour generator dispatch [can be] optimized by SWITCH. As carbon policy strengthens – represented by an increasing carbon cost – CPV supplants natural gas primarily in the hours of peak load, which tend to coincide with the maximum CPV output...[F]uel switching occurs in the Desert Southwest where the highest quality CPV resources is located.

The reduction in carbon emissions from the inclusion of CPV in the power system are small if no carbon policy is in effect (Figure 2.5), as CPV substitues for traditional PV...both of which are low carbon generation sources. The reduction in carbon emissions effected by the inclusion of CPV becomes significant at a carbon price of $40/tCO2, as CPV substitutes for gas-fired generation on the margin. This change represents a WECC-wide reduction in carbon intensity of 32 gCO2/kWh. When this difference is attributed to CPV (as the only thing changed to induce this reduction in carbon emissions was inclusion of CPV in the optimization), this translates to a reduction in carbon intensity of 260 gCO2 per each kWh produced by CPV. The ~ 35 gCO2/kWh of emissions incurred to produce CPV (Figure 1.1) is thereby more than offset by the cost-effective displacement of fossil-fueled generation by CPV.

click to enlarge

Conclusions

The cost of including CPV in the optimal power mix is minimal. For all investment periods and carbon prices, not just for 2026-2029 as depicted in Figure 2.5, the power cost difference between the CPV excluded and CPV included scenarios is less than 1 %, with CPV lowering the cost of delievered power in many cases. Should CPV meet its cost targets, it is poised to reduce the carbon intensity of the electric power system at no added cost.