The Solar Duck Curve
and Sustainable Storage
Options: A Policy
Recommendation
DEVON ROOD
SPRING 2018

Abstract
Before the rise in renewable energy, few people considered the consequences of adding large
amounts of intermittent power onto the grid. As renewable energy has become more prevalent,
utility companies must adapt their business practices to accommodate these unique sources of
power. This is leading to challenges on how best to manage a grid with large amounts of
renewable power.
Arizona Public Service (APS), the largest electricity provider in the state of Arizona, has more
than 70,000 distributed solar customers on their grid and the number of solar customers increases
every day. With this increase in distributed solar customers comes the solar duck curve—the
phenomenon whereby solar produces energy during times of low demand. However, with the use
of storage, the duck curve problem may be mitigated.
This project examines the sustainability of three storage options: pumped hydro energy storage,
compressed air energy storage, and lithium-ion batteries. Using several sustainability indicators,
this project makes a policy recommendation to APS on the most sustainable choice for largescale energy storage.
This project found that compressed air energy storage was the most sustainable option for APS.
This considered the impacts of compressed air on the environment, communities, and the costs of
this storage option. One important aspect to acknowledge regarding this technology is that in its
current form, it does emit some carbon emissions. However, the carbon emissions may have less
of an impact if this storage facility can allow APS to use its renewable energy assets most
efficiently and continue to use energy from Palo Verde, the nuclear facility in Arizona.

1

Table of Contents
Abstract ........................................................................................................................................... 1
Introduction ..................................................................................................................................... 3
Problem Context in Arizona ........................................................................................................... 5
Storage and Renewable Energy ...................................................................................................... 6
Sustainability Implications.............................................................................................................. 7
The Three Storage Options ............................................................................................................. 8
Pumped Hydro Energy Storage ................................................................................................... 8
Compressed Air Energy Storage ............................................................................................... 10
Lithium-Ion Batteries ................................................................................................................ 13
Seasonal Storage Capability ...................................................................................................... 16
Current Storage Projects in the Project Area of Arizona .............................................................. 17
Pumped Hydro Energy Storage ................................................................................................. 17
CAES ......................................................................................................................................... 17
APS Pilot Projects ..................................................................................................................... 18
Sustainability Matrix: Comparing the Three Options ................................................................... 19
Planet/Environment ................................................................................................................... 19
Profit/Costs ................................................................................................................................ 20
People/Communities ................................................................................................................. 21
Summary ................................................................................................................................... 22
Limitations ................................................................................................................................ 24
Policy Recommendation ............................................................................................................... 25
Future Steps .................................................................................................................................. 26
Conclusion .................................................................................................................................... 27
References ..................................................................................................................................... 28

2

Introduction
Electricity generation is a key contributor to carbon emissions, accounting for almost 30% of all
greenhouse gas emissions in the U.S. (Sources of Greenhouse Gas Emissions, 2015; Ameli &
Brandt, 2015). If the world wishes to minimize the impacts of climate change, the electricity
sector will need to rapidly transition to low- and zero-carbon sources of energy. Utility
companies are not the only entities that can choose low-carbon energy options. Businesses,
nonprofit organizations, schools, and homeowners can all play a role in this transition through
options like the installation of rooftop solar systems.
Due to the significant drop in the price of solar panels in recent years, there are now 1.3 million
photovoltaic solar systems in the U.S. (Munsell, 2017). The dramatic drop in the price of solar,
along with creative financing options and an attractive tax credit, have enabled homeowners to
install PV solar systems on their homes. This means that homeowners can now provide some of
their own power instead of simply relying on their utility company for all their electricity needs.
While there are few, if any, reasonable and factual arguments against electricity generated from
solar panels, the fact that homeowners can now generate electricity from their roofs is changing
the business model for the utility company. Previously, electricity generation was a one-way
street. Electricity flowed from the utility company’s power plants to the customer’s home. For
every kilowatt hour consumed at the home, the utility company charged a set rate and received
an expected payment from the customer each month. Now that customers can generate some of
their own electricity, the one-way street has become a two-way street. Customers with solar can
now generate their own energy, thereby reducing the amount of energy purchased while selling
electricity back to their own utility company for use by other customers. The utility companies’
100-year-old business model must adapt to these changes. These changes also offer significant
opportunities for the utility company, customers, and planet to collectively benefit.
As distributed solar generation became more popular, the excess generation that comes from
solar panels during the middle of the day became a challenge for utility companies. Solar
generation occurs during the day when electricity demand is low. When people start returning
home after work, demand for electricity starts to increase, especially in the summer months in
Phoenix due to the heavy reliance on air conditioners to cope with extreme heat. Simultaneously,
solar generation sharply decreases because the sun is much lower in the sky. This mismatch
between the demand for electricity and solar generation is often called the duck curve.

3

Figure 1 (APS, 2018)
Distributed generation is more of a technical challenge for utility companies than large, utilityscale wind or solar farms, because distributed resources are not controlled by the utility. Since
utility companies have no control over these sources of energy, they cannot turn these energy
sources off when there is too much energy on the grid. When there is too much energy, the utility
company needs to look at the resources it owns or controls to determine the best option for
energy curtailment.
The duck curve is becoming a threat to the future of baseload power resources owned or operated
by Arizona Public Service (APS), which stands at about 3,000 MW. There is so much excess
energy generation in the middle of the day during winter months that there is an incentive for
APS to reduce baseload power generation assets. These are usually large power plants, like Palo
Verde, that run all day, every day. Another option for dealing with the problem of excess energy
is energy storage which would allow APS to continue to use its baseload powerplants while also
using energy from renewable sources during times of high demand.
This project will focus on creating a policy recommendation of the most sustainable option for
energy storage. It will examine pumped hydro energy storage, compressed air energy storage,
and lithium-ion batteries and evaluate which ones will be best for APS, customers of APS, and
the environment. Instead of a typical cost-benefit analysis, the storage options will be evaluated
through the lens of sustainability. The impacts on the environment, costs, and communities will
be examined as they apply to the storage options. For environmental impacts, the project will
examine the carbon emissions, impact on the surrounding environment, and the materials
4

needed. The cost analysis will look at the estimated construction/maintenance costs, lifetime, and
reliability. Finally, the impacts on communities will be determined by the potential for
displacement, change to the natural area, and the need for new transmission lines. The
information provided in this matrix will then be used to determine what would best serve APS
and its customers.

Problem Context in Arizona
APS is the largest electricity provider in Arizona and has different needs than other utility
companies around the country. The weather in many parts of APS’ territory greatly changes the
electricity demanded from APS. There are two seasons in many parts of Arizona; summer is
extremely hot, and the fall, winter, and spring are much milder. These two seasons require vastly
different energy resources, with especially high demands in the summer for air conditioning
homes and other buildings. According to estimates from APS, the electricity demand in their
territory goes from about 3,000 megawatts (MW) in non-summer months to about 7,500 MW in
summer months, specifically June through September. APS needs to manage its energy resources
to provide electricity for the high demand in the summer months.
In addition to the unique energy needs due to Arizona’s weather, APS has many more distributed
solar systems compared to other utility companies in the U.S. APS has over 70,000 solar
customers in its territory with rooftop solar systems generating electricity. These 70,000 systems
total about 700 MW in total nameplated capacity—that is, the total capacity available from these
distributed systems. However, during the middle of the summer, solar panels only provide about
235 MW of energy during the hours of 12 p.m. – 2 p.m. Yet the high summer demand occurs
much later in the evening, between 6:00 p.m. – 7:00 p.m. These distributed solar systems are
simply not helping to provide APS power when it needs power the most.
The low energy demand in Phoenix due to the mild weather in spring, fall, and winter means that
APS does not need to have all of its available power plants producing electricity. It only needs
baseload power plants running to meet the low demand. However, the distributed generation
systems on the grid are still producing energy during this time of low demand, thus creating an
excess of energy on the grid. This excess energy threatens the future of some baseload power
plants
Excess energy for a utility company sounds like a beneficial thing considering they are
occasionally worried about having enough supply to meet customers’ demand for electricity.
However, excess energy does not benefit the grid, and it has the potential to increase carbon
emissions for APS. Because APS has so many distributed solar systems that are producing
energy at times of low demand, APS is worried about the future of the Palo Verde Generating
Station, the largest power producer in the United States (Arizona Emergency Information
Network, 2015). Palo Verde provides APS with baseload power that is running constantly. This
is because nuclear power plants are not easily ramped up or down. They operate at relatively the
same levels unless there is a scheduled maintenance or another reason for a temporary shutdown.
Also, despite the criticism of nuclear power, it is a low-carbon source of energy. With a goal to
reduce the amount of carbon emissions worldwide, nuclear is a good option for electricity
5

generation, especially since it operates very similarly to fossil fuel plants in that it does not rely
on nature to operate. If this low-carbon source of energy for Phoenix is forced offline and
replaced by natural gas peaking plants to help smooth the fluctuating nature of renewable energy,
specifically solar here in Arizona, then carbon emissions will likely increase. This is precisely
the opposite intended outcome of increasing the amount of renewable energy on the grid.
Therefore, APS is interested in looking at storage options to help with the excess energy on the
grid while maintaining the benefits of the baseload power coming from Palo Verde.

Storage and Renewable Energy
While renewable energy has had tremendous growth, there is overwhelming consensus that
without the ability to store energy, renewable energy cannot be the main source of energy for the
grid. That is because renewable energy is often intermittent. Wind and solar, the two major
renewable energy sources, only work when the wind is blowing, or the sun is shining.
One of the biggest challenges of deployment of large-scale renewable energy is storage. Better,
more advanced renewable energy technologies will likely happen. However, this is not nearly as
important as storage (Bullough, Gatzen, Jakiel, Koller, Nowi, & Zunft, 2004). The intermittency
of wind, sun, and other similar sources renewable energy, causes problems for the electrical grid.
The grid’s design depends on the supply of electricity always matching demand (Bullough et al.,
2004).
The intermittent nature of renewables causes problems and uncertainty for the grid and grid
operators (Beaudin, Zareipour, Schellenberglabe, & Rosehart, 2010). With higher percentages of
renewables on the grid, there is less predictability and lower control over the supply of energy
(Bullough et al., 2004; Swider, 2007). Additionally, intermittency interferes with energy
generation and load balancing (Luo, Wang, Dooner, & Clarke, 2015). Renewable energy sources
“can rarely provide immediate response to demand as these sources do not deliver a regular
supply easily adjustable to consumption needs,” (Ibrahim, Ilinca, and Perron, 2008).
Additionally, intermittent sources of power are not flexible and therefore cannot be adjusted to
meet differing power demands (Yang & Jackson, 2010). Because of the inflexibility of these
sources of power, along with their unpredictable nature, renewable energy plants are simply
turned off when there is too much energy on the grid. When the traditional baseload sources of
power are operating close to or at their minimum load, wind and solar plants are curtailed. Since
traditional power plants are not easily ramped up or down quickly, it is the wind and solar power
plants that are taken offline when there is excess energy (Foley, Leahy, Li, McKeogh, and
Morrison, 2015).
Wind, solar, and other renewables will only be able to reduce carbon emissions from electricity
generation in significant amounts and make the electricity grid more sustainable if these
generation sources can replace baseload power plants. In order to accomplish this, storage for
these renewable sources is critical (Succar & Williams, 2008; Evans et al., 2012; Hadjipaschalis,
Poullikkas, & Efthimiou, 2009). There are a variety of storage options available to utility
companies now, and the unique needs of each utility company will determine the storage options
pursued. In addition to the individual needs of APS, there are other decision factors when it
6

comes to choosing a storage option, like the sustainability of each of these choices, that should
be incorporated into any decision made by APS.

Sustainability Implications
While sustainability is becoming more common in everyday advertising, many people still do not
have a working definition of the word. Because sustainability is a complex topic, it encompasses
a variety of disciplines with several acceptable definitions for sustainability. For this project, the
best definition is the original definition of the word - “sustainable development is development
that meets the needs of the present without compromising the ability of future generations to
meet their own needs” (Brundtland, 1987). As most of the world is increasingly concerned about
climate change, and there are many initiatives to improve the health and wellness of all
communities, sustainability is more prevalent and important. Incorporating sustainability into
both small, every day choices as well as larger, long-term choices is imperative to preserving the
planet for future generations.
Sustainability is not simply about the environment. While the impacts on the environment are
pertinent when it comes to sustainability, outcomes on communities and economies due to an
activity or decision are imperative, too. These three topics, environment, communities, and
economies, which are sometimes referred to as the pillars of sustainability, must be carefully
evaluated and analyzed before declaring whether something is truly sustainable.
Often, businesses and companies use a traditional cost-benefit analysis to determine the best
option. However, simply looking at the costs and benefits leaves out important aspects, namely
the impacts on the environment and communities. This is problematic because ignoring the
consequences of a decision on these two areas does not consider the true cost of a product or
decision. Using sustainability as a factor allows for a more holistic decision-making process.
The issue of the duck curve, especially here in Arizona, is a sustainability problem in itself.
While excess energy on the grid and the difficulty of quickly ramping up electricity generation
are experienced by other utility companies with significant solar PV resources, Arizona may be
unique in that it is home to Palo Verde. If the duck curve becomes significant enough to force the
retirement of this powerplant, there will be a few negative sustainability outcomes.
If Palo Verde is taken offline, then the energy must be replaced with another resource, most
likely a carbon resource. This would lead to increases in carbon emissions from the generation of
electricity in Arizona. The irony of this would be that renewable energy leads to increased
carbon emissions. Carbon emission increases are not the only problem if Palo Verde is
shutdown. Palo Verde employs 2,600 people regularly, and an additional 800 people twice a year
during re-fueling (APS, 2017). With the closure of Palo Verde, all these employees would be
without a job which would negatively impact thousands of families around the state. And finally,
the state would lose a large source of economic activity with the plant closure. Through taxes,
salaries to employees, and services it uses, Palo Verde’s annual economic output is about $1.8
billion (APS, 2017). Additionally, it is the state’s largest taxpayer, paying around $50 million in
taxes annually (APS, 2017). The closure of Palo Verde could then be used by those who are
against the deployment of renewable energy to demonstrate that renewables should not be
pursued because they cause economic problems for the state.
7

The Three Storage Options
Pumped Hydro Energy Storage
Pumped hydro energy storage (PHES) is one of two viable options when it comes to large-scale
utility energy storage. PHES works in a very similar way to traditional hydroelectric power
plants. When there is excess energy on the grid or when energy prices are low, water is pumped
uphill to a reservoir where the water is stored. When energy is needed, like during peak demand
hours, water is released from the reservoir and passes through a turbine to create electricity. This
water is then stored in a reservoir downstream (Evans et al., 2012; Ibrahim et al., 2008; Yang &
Jackson, 2010; Deane et al., 2009; Steffen, 2012; Rehman et al., 2015; Uria, Martinez,
O’Connor, Johnson, 2015). The cycle repeats whenever there is excess energy on the grid.
PHES is not a new technology; PHES has been in use around the world for almost 100 years
(Foley et al., 2015; Evans et al., 2012; Ma et al., 2014). In fact, Germany’s first PHES was
completed in the 1920s (Steffens, 2012). Since PHES has been around for many decades, it
should come as no surprise that it is not only a predictable technology, but it is also the most
common form of energy storage technology in the world (Deane et al., 2009; Steffen, 2012;
Yang & Jackson, 2010; Dunn et al., 2011; Evans et al., 2012; Foley et al., 2015; Ma et al., 2014).
Worldwide, there are 127,000 MW of PHES capacity compared to compressed air energy
storage, the second most common form of energy storage, which only has 440 MW of capacity
(Dunn et al., 2011).
PHES has a variety of benefits, both in terms of energy storage as well as services it can provide
to the electricity grid. One of the biggest benefits of PHES is that due to its long and proven
history, it is both a stable and reliable form of energy storage (Evans et al., 2012; Ibrahim et al.,
2008). In addition to its reliability it also has a long lifespan. PHES plants are expected to last
anywhere between 50-100 years (Dunn et al., 2011; Evans et al., 2012; Deane et al., 2009).
PHES has a fast ramp time which is another one of the reasons utility companies find it an
attractive option for energy storage (Yang, Makarov, Desteese, Viswanathan, Nyeng, McManus,
& Pease, 2008; Deane et al., 2009; Evans et al., 2012; Rehman et al., 2015). PHES can also be
turned on and off quickly, help stabilize voltages on the grid, balance supply and demand, and
shift loads (Rehman et al., 2015; Evans et al., 2012; Deane et al., 2009; Uria et al.,
2015). Because of its flexibility, it has comparable benefits to other sources of energy that are
used as back-up power such as peaker plants or gas turbines (Deane et al., 2009). Peaker plants
are resources used by utility companies when demand is particularly high, and they can be turned
on and off relatively quickly. The price for this electricity is often much higher than baseload
power because it comes from resources that are only used at times of high demand.
PHES also has a high efficiency. It operates somewhere between 70-85% efficiency (Evans et
al., 2012; Deane et al., 2009). While this efficiency is important, it means that PHES is also a net
consumer of energy (Evans et al., 2012). Therefore, it is imperative that the PHES system drives
water uphill at times of low demand and then releases the water at times of high demand to make
the plant economically feasible (Evans et al., 2012). A PHES system addresses the duck curve
problem by using the excess energy in the middle of the day, a period of low demand, to move
the water uphill to the higher reservoir to be used later at times of high demand.
8

With any storage option, cost must be examined closely. Utility companies want “cost-effective”
storage options because storage has the potential to help alleviate problems faced by the
electricity sector. However, these storage technologies come at a cost that must be paid by the
utility company. For PHES, there is no set industry cost per kilowatt (kW) stored because PHES
systems are extremely site specific (Deane et al., 2009; Foley et al.; Evans et al., 2012).
However, there are some estimates for an average PHES plant. Estimates vary anywhere
between $600-$3,300 per kW depending on the date of the study and the location (Evans et al.,
2012; Foley et al., 2015; Deane et al., 2009). This large price discrepancy demonstrates that for
an accurate price estimate, an assessment of the proposed PHES site must be done by a company
with knowledge of this type of energy storage. While PHES can be expensive to build, there are
low operation and maintenance costs for the plant (Deane et al., 2009; Dunn et al., 2011).
Since PHES is a proven technology and has been used globally for decades, why is it not
utilized? To begin, storing energy is a difficult task which requires equipment that is bulky,
expensive, or both (Ibrahim et al., 2008). As a result, utility companies opt to build more
generation instead of investing in energy storage technologies (Dunn et al., 2011). Also, PHES
has some specific characteristics that make it infeasible for many locations.
First, PHES requires particular site conditions (Evans et al., 2012; Ma et al., 2014; Yang et al.,
2008; Deane et al., 2009). There need to be two locations to hold water and reliable access to
water (Deane et al., 2009). Additionally, the reservoirs must be large enough to accommodate the
amount of water that will need to travel uphill and downhill and they must also be at different
altitudes. These requirements are important as PHES has low energy density (Rehman et al.,
2015).
The site-specific requirements of PHES also mean that long transmission lines may need to be
built as the site may be located far from centers of high demand (Evans et al., 2012). Building
new transmission lines have their own problems, including high cost, long construction time,
incredible size, and the fact that they impact wildlife areas (Personal interview, Marc Romito,
February 21, 2018). Additionally, these lines usually must pass through both public and private
land which requires cooperation between many different entities. In Arizona, the issue is
compounded even more by the fact that a great deal of land that would be needed for the
transmission lines is on Native American Reservations (Personal interview, Marc Romito,
February 21, 2018).
PHES gives rise to environmental concerns, similar to regular hydroelectric power plants (Evans
et al., 2012; Yang, et al., 2008; Ma et al., 2014). Regular hydro power negatively impacts
riparian ecosystems and fishing resources by permanently changing the river system (Kosnik,
2010). These environmental concerns apply to PHES as well. Many experts point to these
environmental concerns as barriers to the more widespread adoption of PHES both in the U.S.
and globally. For example, Yang and Jackson explored several proposed PHES projects in the
U.S. and found that opposition by environmental groups played a significant role in the
9

termination of most of the proposed projects before they started (2010). However, they also note
that engaging with all stakeholders, including environmental groups, increased the ease of
proposing and developing PHES projects (Yang & Jackson, 2010).
Historical events, both nationally and globally, have had a negative impact on PHES
development over the last 30 years. PHES was a very popular option for new energy
development in the 1960s during the height of the nuclear energy era (Yang & Jackson, 2010;
Deane et al., 2009; Uria et al., 2015). The oil crisis during this period, especially in the U.S.,
increased the development of nuclear energy. With the increase in nuclear power, there was also
an uptick in the development of PHES facilities as a complementary component to nuclear
energy. Since nuclear power cannot easily or quickly ramp up or down to meet demand, PHES
was often paired with nuclear facilities to provide the ability to quickly provide power during
times of high demand (Yang & Jackson, 2010; Deane et al., 2009).
However, the slowdown in nuclear power development in the 1980s lead to a concomitant
decline in PHES development (Yang & Jackson, 2010; Deane et al., 2009). As natural gas
became much cheaper in the 1980s, utility companies chose natural gas instead of PHES to be
used as peaking power plants (Yang & Jackson, 2010; Deane et al., 2009; Uria et al., 2015). This
is because PHES competes directly with natural gas turbines as they provide very similar
benefits to the electrical grid.
Finally, the electricity sector in the U.S. underwent a deregulation transformation in the 1980s
and 1990s. The utility industry transitioned to one where generators of electricity separated from
those who owned the transmission lines used to transport energy. Since energy storage does not
neatly fit into either of those categories, development of PHES slowed (Yang & Jackson, 2010;
Deane et al., 2009).
With the difficulties of storage outlined above, and the vastly different price estimates for PHES
plants, there is nonetheless increased demand for PHES because of the new goals for renewable
energy deployment in both the U.S. as well as other countries. With more renewable energy
mandates, there is an increased need for some way to store the energy from these low-carbon
sources of energy. Additionally, there is an increased need for energy and peak power, both of
which PHES can also provide (Yang & Jackson, 2010). Much of this proposed new development
of PHES in the U.S. is occurring in the western part of the country where many of the states have
higher renewable portfolio standards (Deane et al., 2009). Germany is another place where there
is a new interest in PHES due to their renewable energy targets (Steffens, 2012).

Compressed Air Energy Storage
Compressed air energy storage (CAES) is the other major viable large-scale energy storage
technology today (Chen, Zhang, Liu, & Tan, 2013; Chen et al., 2009; Succar & Williams, 2008;
Abbaspour, Satkin, Mohammadi, Ivatloo, Cotfi, and Noorollahi, 2013). The two plants in the
world using this technology are both diabatic CAES, or known as D-CAES, plants. There are
other theories about how CAES can work, but since these technologies have not been deployed
by a utility company anywhere in the world, there is no information on their viability. This
10

project will therefore discuss D-CAES and its benefits and concerns. The mechanism of CAES is
not complex. During off-peak periods, electricity is turned into potential energy by compressors
which pressurize air. This air is then stored underground in a cavern. When energy is needed,
especially during times of high demand, the pressurized air is released and expanded in a turbine.
To expand the air, a fuel source like natural gas, is used to heat the air. The turbine converts the
stored energy into electricity (Budt, Wolf, Span, and Yan, 2016; Bullough et al., 2004;
Greenblatt, Succar, Denkenberger, Williams, and Socolow, 2007; Simmons, Barnhart, Reynolds,
and Son, 2010; Swider, 2007; Succar & Williams, 2008; Fertig & Apt, 2011; Beaudin et al.,
2010; Chen et al., 2013; Ibrahim et al., 2008; Lund, Salgi, Elmendgaard, and Andersen, 2008;
Abbaspour et al., 2012; Luo et al., 2015; Chen et al., 2008; Lund & Salgi, 2009; Fthenakis,
Mason, and Zweibel, 2008).
CAES is a relatively new technology. Research into compressed air as a form of storage began in
the late 1960s, and the first plant was built in 1978 (Budt et al., 2016). The motivation behind
CAES was similar to the increase in PHES deployment. In the 1960s and 1970s, due to a variety
of factors including the oil crisis and an increase in nuclear energy, interest in storage options
started to boom (Succar & Williams, 2008). Similar to PHES, CAES was explored as a potential
form of storage due to the prevalence of baseload power, mainly from coal and nuclear power
(Budt et al., 2016). With the baseload power provided by coal and nuclear power, utility
companies have an incentive to store energy from off-peak times for later use during peak times
(Budt et al., 2016). Many viewed storage, including CAES, as economically useful to utility
companies because it could improve the use of expensive baseload power plants (Sharma, Chiu,
Ahrens, Ahluwalia, & Ragsdell, 1979).
In the 1970s, northern Germany needed storage but PHES was not possible in this part of the
country due to the geography in the area. However, CAES was a possibility in the region, and the
first CAES plant in the world was built near Huntorf, Germany in 1978 (Budt et al., 2016). The
Huntorf plant spurred interest in CAES research in the U.S., but the first and only CAES plant in
the U.S. was not built until 1991 (Budt et al., 2016).
This first CAES plant was built with a capacity of 290 MW and about 2 hours of storage using a
salt dome as its gas reservoir (Budt et al., 2016; Bullough et al., 2004; Greenblatt et al., 2006;
Succar & Williams, 2008; Fertig & Apt, 2011; Beaudin et al., 2010; Ibrahim et al., 2008; Chen et
al., 2013; Abbaspour et al., 2013; Lund et al., 2008; Luo et al., 2015; Chen et al., 2008; Simmons
et al., 2010; Fthenakis et al., 2008). The Huntorf plant is currently used for leveling the variable
energy on the grid in Germany, even though it was not specifically designed with that function in
mind (Greenblatt et al., 2006). The plant helps with peak shaving and assists with filling the gaps
left by plants that are slow to start, like coal and nuclear (Succar & Williams, 2008; Fertig &
Apt, 2011). Additionally, Huntorf has the ability to blackstart, which means that there is no wait
time for the plant to provide power (Budt et al., 2016; Luo et al., 2015). The Huntorf plant also
helps with variable wind power in Germany, an increasing source of energy (Fertig & Apt,
2011). There have been some repairs and adjustments to the original plant, but the facility has
been extremely reliable (Budt et al., 2016; Chen et al., 2008; Beaudin et al., 2010).

11

The other CAES plant is located near McIntosh, Alabama. The plant was constructed in 1991
and is slightly smaller than the Huntorf plant at just 110 MW of capacity. This plant is also
located in a salt dome and works in the same way as the Huntorf plant (Bundt et al., 2016;
Bullough et al., 2004; Greenblatt et al., 2007; Simmons et al., 2010; Succar & Williams, 2008;
Fertig & Apt, 2011; Beaudin et al., 2010; Ibrahim et al., 2008; Chen et al., 2013; Abbaspour et
al., 2013; Luo et al., 2015; Chen et al., 2008; Fthenakis et al., 2008). The McIntosh plant can
provide 100MW of energy in 14 minutes and it burns about one-third the natural gas that similar
gas turbines would burn (Power South, n.d.). This plant is also considered to be extremely
reliable source of power (Budt et al., 2016; Beaudin et al., 2010).
Through improvements on the design of CAES, the McIntosh plant is about 25% more efficient
than Huntorf (Budt et al., 2016; Beaudin et al., 2010; Luo et al., 2015; Chen et al., 2008).
Additionally, the McIntosh plant was designed for weekly load shifting by taking cheap baseload
power at night and storing it until peak times when energy is expensive, while Huntorf was
designed for reserve power (Budt et al., 2016; Fertig & Apt, 2011).
These two CAES plants are located underground, though an above ground design is believed to
be possible. There are a number of potential types of underground caverns that are considered
acceptable locations for CAES: mined salt caverns, spent natural gas caverns, hard rock cavities,
and aquifers (Budt et al., 2016; Greenblatt et al., 2007; Simmons et al., 2010; Succar &
Williams, 2008; Sharma et al., 1979; Ibrahim et al., 2008; Chen et al., 2013; Luo et al., 2015;
Chen et al., 2008). However, mined salt caverns are the only option in use today. Natural gas has
been stored in a similar manner in this type of cavern, so there is a greater understanding of how
pressurized gas behaves in that environment (Budt et al., 2016). There is some discussion of
using aquifers as acceptable locations for a CAES cavern; however, it has not been used in a
commercial plant yet (Budt et al., 2016; Succar & Williams, 2008). While above ground is an
option, in theory, it would likely be much more expensive than storing compressed air in an
already developed cavern underground because a container for storing the air would need to be
built (Greenblatt et al., 2007).
The similarities between CAES and PHES explain why they are the only two commercially
viable large-scale energy storage technologies (Chen et al., 2013; Chen et al., 2008; Succar &
Williams, 2008; Abbaspour et al., 2013). These two technologies have high efficiencies, large
capacities, long storage times, and the ability for quick ramp times (Succar & Williams, 2008;
Beaudin et al., 2010; Abbaspour et al., 2013; Fthenakis et al., 2008; Fertig & Apt, 2011; Chen et
al., 2008).
While CAES appears to be a promising storage option, PHES has historically been a much more
popular choice for energy storage. since PHES has been in operation since the late 1920s and is
often seen as a proven technology (Luo et al., 2015). Nonetheless, many developed countries no
longer have any more PHES options available (Bullough et al., 2004).
If large-scale storage is needed, CAES may be the best option when PHES is not possible and
batteries are cost prohibitive (Swider, 2007). CAES does not have the same limitations or severe
environmental concerns like PHES. CAES has similar size capabilities as PHES, and it does not
12

have the geographic restrictions presented by PHES (Bullough et al., 2004). In fact, it is
estimated that 75% of the area in the U.S. is suitable for CAES, including spots in Arizona
(Fthenakis et al., 2008; Beaudin et al., 2010 Simmons et al., 2010). CAES also has fewer
environmental concerns than PHES (Chen et al., 2008). CAES uses little land area as much of it
is underground. Because of the little land necessary for CAES, it has a smaller impact on the
land and surrounding ecology, especially when compared to PHES (Succar & Williams, 2008;
Beaudin et al., 2010).
Similar to PHES, estimating the cost of CAES is difficult as there are only two plants from
which cost data to build such a system can be obtained. The varying cost estimates for CAES
which may be attributed to the available caverns to store the compressed air. Estimates range
from $400-$800 per kW of energy storage capacity (Beaudin et al., 2010; Chen et al., 2013;
Chen et al., 2008). Because CAES can be built in caverns that already exist, capital costs can be
lower compared to other energy storage options (Succar & Williams, 2008).
While CAES may have certain characteristics that make it more attractive or feasible than PHES,
it also has some drawbacks. CAES needs a secondary form of energy when discharging the
stored energy to heat up the air (Budt et al., 2016). With the two existing plants using fossil fuels
to help power CAES, this inevitably means that there are carbon emissions coming from this
technology, which may make CAES a less attractive form of energy storage (Chen et al., 2013).
However, these plants use about one-third the fuel that regular turbines use and thus have onethird the carbon emissions (Power South, n.d.). It is important to note that there is research on
how to reduce, or possibly eliminate, the carbon emissions from CAES which would help
improve the environmental impact of using CAES (Succar & Williams, 2008).
The salt caverns used for both existing CAES plants in the world cause environmental issues as
well. To create the dome or cavern, the salt must be dissolved which produces brine (Simmons et
al., 2010). This process not only requires large amounts of water, but then the briny water must
be disposed as it cannot be released into fresh water.
Finally, there are a lack of new CAES facilities. Many of the available papers discuss future
CAES projects, but all but one project in the U.S. have been abandoned so far. The only potential
project is a project located in Texas. The cancellation of many of these projects is discouraging.
However, CAES may start experiencing a comeback in the U.S. due to the increase in renewable
energy and the subsequent need to balance out these intermittent sources of power (Budt et al.,
2016; Succar & Williams, 2008).

Lithium-Ion Batteries
APS has several pilot projects underway to increase their storage capacity. Many of these
projects involve lithium-ion batteries. To fully compare the sustainability aspects of CAES and
PHES with the APS pilot projects, a review of lithium-ion batteries is necessary. While there are
a few articles outlining how these batteries work, their benefits, and the costs associated with this
technology, many of the articles available now outline the tested improvements to this
technology. Some news stories about using large lithium-ion batteries as storage provided more
information than academic articles, especially for the context of this project.
13

Under the right conditions, batteries for storage can help with the challenges presented by solar
and other renewable energy sources (Hill, Such, Gonzalez, Grady, 2012). Batteries have fast
ramp rates and can provide voltage support and stability to the system. Therefore, batteries can
help with the variability challenges that come from solar generation (Chen et al., 2008; Hill et al.,
2012).
Batteries are the oldest form of energy storage, dating back over a century (Chen et al., 2008).
Due to the changes in the energy sector, as well as the transportation industry, batteries are
quickly becoming a more popular way to store energy. In 2015, 95% of all new energy storage
installed was in the form of batteries (D’Aprile, Newman, Pinner, 2016). This may be due to
flexibility in location of batteries as well as short construction times (Luo et al., 2014). However,
the use of lithium-ion batteries by utility companies for large-scale storage is a new phenomenon
(Chen et al., 2008).
While everyone uses some sort of battery daily, many of these same people may not understand
the mechanics of how batteries operate. There is an anode, the negative electrode, and the
cathode, the positive electrode. There is also an electrolyte in the battery as well which allows for
the transfer of electrons between the anode and cathode (Hadjipaschalis et al., 2009; Luo et al.,
2015; Chen et al., 2008; Dunn et al., 2011). In lithium-ion batteries, often referred to as Li-ion
batteries, a lithium metal oxide is used for the cathode and graphite is the anode. The electrolyte
in the battery is a liquid usually containing lithium salt (Luo et al., 2015; Scrosati & Garche,
2010). In this battery, the lithium-ion moves between the anode and cathode to create electricity
(Evans et al., 2012; Diaz-Gonzalez, Sumper, Gomis-Bellmunt, Villafafila-Robles, 2012).
Li-ion batteries are used for many applications today, but the most popular use is in portable
technology like cell phones, cameras, laptops, and iPads (Diaz-Gonzalez et al., 2012; Dunn et al.,
2011; Beaudin et al., 2010; Hadjipaschalis et al., 2009; Lindley, 2010). In the early 1990s, Sony
was the first to commercially produce Li-ion batteries, and the technology has only improved
since then (Dunn et al., 2011; Beaudin et al., 2008; Chen et al., 2008). In fact, Li-ion batteries
have improved since the publication of these articles. Some of the older articles mention that Liion batteries may be used for electric vehicles or hybrid vehicles. Today, the auto industry has a
variety of electric vehicles on the market that can power the vehicle for a couple hundred miles
on one charge.
There are a variety of characteristics of Li-ion batteries that make them attractive for energy
storage, specifically for utility companies. They have high energy densities (Diaz-Gonzalez et
al., 2012; Dunn et al., 2011; Beaudin et al., 2010; Chen et al., 2008). Li-ion batteries have fast
charge and discharge rates (Diaz-Gonzalez et al., 2012). This means that batteries can either
absorb or dispatch power quickly, which is a benefit to utility companies that need storage to
quickly respond to the needs of the grid. Li-ion batteries also have high efficiency rates, around
95% (Diaz-Gonzalez et al., 2012; Chen et al., 2008). In fact, Li-ion batteries have a higher
energy density and efficiency than lead-acid and nickel batteries (Hadjipaschalis et al., 2009).
Another benefit of Li-ion batteries is that they have low self-discharge rates, that is the amount
of energy lost due to processes inside the battery (Evans et al., 2012; Hadjipaschalis et al., 2009).

14

In addition to these benefits, Li-ion batteries are easily scalable and are very versatile which
makes them attractive for storage (Larcher & Tarascon, 2015).
Li-ion batteries, unlike the other two forms of storage discussed above, can be placed anywhere
on the grid. There are no geographic limitations for the placement of batteries. This is hugely
beneficial to the utility company because it means that batteries can be used by utilities that need
storage in any location. Also, since batteries can be placed anywhere, construction of new
transmission lines can be avoided.
The cost of large-scale Li-ion batteries has been the biggest barrier to the deployment of more
battery storage. In fact, the cost of these batteries makes it an unattractive option in many
scenarios (Evans et al., 2012; Beaudin et al., 2010; Chen et al., 2008; Lindley, 2010; Scrosati &
Garche, 2010). Currently, Li-ion batteries cost about $500 per kilowatt (Roberts, 2017; Baidwai,
2017). One reason for this is that Li-ion batteries need to be managed by some device or
computer to ensure that they are not fully discharged as this would negatively impact the lifetime
of the battery (Luo et al., 2015; Hadjipaschalis et al., 2009). This monitoring system adds to the
cost of using a large battery for storage.
While there are many benefits to Li-ion batteries, there are some concerns about this type of
energy storage. The biggest concern is safety and the possibility of fires due to the use of lithium
salt as it can become flammable (Diaz-Gonzalez et al., 2012; Beaudin et al., 2010; Lindley,
2010). There are also concerns about available reserves of lithium worldwide as it is a finite
resource (Dunn et al., 2011; Larcher & Tarascon, 2015; Scrosati & Garche, 2010).
Environmental concerns, especially in the manufacturing of the battery, are important to consider
when using Li-ion batteries. Li-ion require huge energy inputs during manufacturing (Larcher &
Tarascon, 2015. This means that unless the batteries are made using electricity from renewable
sources, there are significant carbon emissions just from the manufacturing process alone.
In addition to the available sources of lithium in the world, there are also concerns about the
extraction process of lithium. While lithium mining is not the same as mining for other metals,
like copper, it still causes some environmental impact. The majority of the world’s lithium comes
from Australia and South America. In South America, there are concerns by the indigenous
communities near lithium mines over the large amount of water needed to process lithium
(Merchant, 2017).
There are issues with temperature and Li-ion batteries. High temperatures mean that the Li-ion
batteries age more quickly (Hadjipaschalis et al., 2009). This is important to remember
considering the hot summer temperatures in the Phoenix area as well as other areas within APS
service territory. Additionally, Li-ion batteries have a shorter lifetime than either CAES or
PHES. These batteries have a lifetime varying from 5-16 years (Diaz-Gonzalez et al., 2012;
Evans et al., 2012). This short lifespan means that the utility will need to procure more batteries
in the future when the current batteries are at the end of their useful life.
Recent literature on Li-ion batteries focuses on how to improve batteries, while reputable news
organizations fill in the gap by providing information on new Li-ion battery systems installed
around the world. In the last nine months, there have been two large installations of Li-ion
15

batteries, one in California and one in Australia. In September 2017, the largest Li-ion battery in
the U.S. went live in southern California. This battery, built for San Diego Gas & Electric, can
dispatch 30 MW of power in seconds and was constructed in less than six months (Spector,
2017). This system was built in response to a gas leak in Southern California in 2015 that lasted
more than four months (Spector, 2017). The utility company wanted a technology that could
provide power immediately but also exist in dense urban areas. More traditional options, like gas
peaker plants or hydro plants, were not possible in populated areas, like San Diego (Spector,
2017). Therefore, the utility company decided to utilize Li-ion batteries. This system behaves
like a gas peaker plant, but it does not emit air pollution or greenhouse gas emissions while in
use. Additionally, it is much quicker to build a battery system than a gas plant (Spector, 2017).
This battery system is being used as a test for large-scale Li-ion batteries for the grid (Spector,
2017).
Shortly after the construction of this battery, a much larger Li-ion battery came online in
Australia. In December 2017, a 100 MW battery was built by Tesla in Southern Australia in less
than 100 days (BBC, 2017). This battery system is connected to a wind farm in the area, and it is
used to “support and stabilise (sic) existing electricity supplies,” (BBC, 2017). Tesla said that its
experience with Li-ion batteries for its vehicles helped it build this system, even though the
batteries used for storage are slightly different than those used in their electric vehicles (BBC,
2017).
McKinsey, a large consulting firm in the U.S., sees potential for batteries to become more
prominent in the short-term. McKinsey predicts that battery prices will fall, which is imperative
for more widespread use of this technology. They expect batteries will cost $200 per kilowatt
hour in 2020, which is half the cost of these batteries in 2016, and $165 per kilowatt hour, or
less, by 2025 (D-Aprile et al., 2016). Additionally, researchers continue to work to improve Liion batteries. As batteries improve, in terms of cost, lifespan, and material extraction, there is
hope that renewable energy sources could provide a significant portion of the energy needs in the
U.S.

Seasonal Storage Capability
The weather in many parts of APS service territory gives the utility company unique energy
demands. Because of the extreme heat in the summer, APS must build enough generation to
fulfill this demand. This power is extremely expensive, both for the utility company and the
consumer as the costs are passed on.
The mild fall, winter, and spring means that the excess energy from distributed solar is an even
greater threat to baseload power during these seasons. The utility does not need as many sources
of power because demand is low. The difference in demand in the summer compared to the other
three seasons provides an interesting case for seasonal storage.
Seasonal storage is an idea that has not been put to practical use. However, due to the
characteristics of different storage technologies, some storage options are better suited for the
possibility of seasonal storage. Due to size, storage ability, and other characteristics, both CAES

16

and PHES are possible options for seasonal storage (Beaudin et al., 2010; Luo et al., 2015;
Simmons et al., 2010; Succar and Williams, 2008).

Current Storage Projects in the Project Area of Arizona
These three storage options were chosen because all three are possible in Arizona. Below are
details on potential future projects or research that indicates how best each particular storage
option can be implemented in the state.

Pumped Hydro Energy Storage
Currently, there is an active preliminary permit for a PHES facility in the Chino Valley. This
project was filed in October 2017 with the Federal Energy Regulatory Commission (FERC). The
2,000 MW project proposes pumping water from the Big Chino aquifer, the same aquifer that
supplies water to the residents of parts of northern Arizona as well as the Verde River. In the
filing, this project looks to compliment the growth in renewable energy in the region (Ciampoli,
2017). The proposal states that it will explore options to reduce the impact on regional water use
(Ciampolo, 2017).
Upon the filing of this permit, the town of Clarkdale filed a motion to intervene on the project.
Their main concern is the water that will be used for the project will negatively impact the water
the town uses. In fact, the town of Clarkdale states that the amount of water needed for this
project, estimated to be 19,000-acre feet, is equal to more than 40 years of water use by the
residents and businesses in Clarkdale (Motion to Intervene, Town of Clarkdale, Arizona, 2017).
Additionally, Clarkdale is concerned that this project will negatively impact the Verde River and
the surrounding riparian ecosystem (Motion to Intervene, Town of Clarkdale, Arizona, 2017).
The concerns put forth by Clarkdale, and possibly other cities or towns, are legitimate concerns.
Water is a precious resource, especially here in Arizona. Additionally, there are valid concerns
when it comes to the health of riparian areas where PHES plants exist. However, working with
all stakeholders, including city governments, citizens, and environmental groups may help ease
the worries of the small cities and towns that rely on the Big Chino aquifer for their water.
Without buy-in from the local communities, the proposed Big Chino Valley PHES plant will
likely face an expensive, lengthy, and contentious battle before it even gets approved for
construction.

CAES
There are a couple of studies published that demonstrate that Arizona has potential locations for
CAES around the state (Simmons et al., 2010; Neal & Rauzi, 1996) There are at least two salt
caverns in APS territory, one in west Phoenix and one in Pinal County. These are both large
caverns which means that they have the potential to store large amounts of energy. As discussed
above, the availability of underground caverns means that APS would not need to include the
cost of building a structure to the already high costs of energy storage.

17

Map of known and potential salt caverns in Arizona (Simmons et al., 2010)

APS Pilot Projects
The idea for this project came from investigating what APS was doing to alleviate the duck curve
problem. Right now, the main strategy for APS is to try several different pilot projects and see
which one, if any, can make a significant impact with the problem. In September 2017, APS
announced that they were in the process of developing more programs to help the utility save
energy when it needed to most and helping customers save money at the same time. Some of
these new programs will be available to APS customers beginning in the summer of 2018. These
programs include smart thermostats and smart water heaters. The intent of these programs is help
absorb the excess energy on the grid in the middle of the day from solar.
In addition to the programs listed above, APS will also have two programs that include Li-ion
batteries for residential customers free of charge. These projects are in addition to the other
battery experiments APS has carried out throughout its territory. The first program, the Solar
Innovation Study–75, gives customers a variety of energy equipment, like rooftop solar systems,
pool pumps, load controllers with a home energy management system, and batteries. The
program will study how customers manage their own energy without guidance. There will be 10
homes with a battery designed to charge during the day and then discharge between APS’ peak
hours, 3p.m.-8p.m. They are 2-kilowatt, 5-hour batteries, or 10 kilowatt-hours of storage total.

18

For the homes that have batteries, they will also have rooftop solar systems (Personal interview,
John Pinho, March 16, 2018).
The second program is part of APS’ rewards program. APS will install 50 batteries in homes that
do not have rooftop solar systems. Similar to the Solar Innovation Study – 75, the batteries are 2kilowatt, 5-hour batteries and will come with a critical load panel. These batteries are also
designed to charge during off-peak times and discharge during the peak. APS has specific
locations throughout their territory that have high solar penetrations and want to place the
batteries in these areas. For this program, they have allocated $3 million for the purchase and
installation of the 50 batteries. Applications for this program will state in early summer 2018
(Personal interview, John Pinho, March 16, 2018).

Sustainability Matrix: Comparing the Three Options
The storage options detailed above can all help APS in terms of storing excess energy from solar.
However, the different characteristics of each storage option impact the sustainability of each of
the technologies. The sustainability matrix below helps detail the sustainability assessment of
each storage option.
The three pillars of sustainability are often referred to as planet, people, and profit. For each of
the pillars, there are three aspects that will help determine the sustainability score for each
storage option. For the planet section, CO2 emissions, impacts on surrounding environment, and
materials needed for constructed are evaluated. The cost section examines the cost to build
and/or maintain, lifetime of the storage option, and reliability. Finally, the last pillar, people or
communities, will evaluate the risk of displacement, transmission line construction, and negative
impacts to natural space enjoyment.
The metrics are evaluated quantitatively using the information provided in both the literature
review as well as discussions with professionals and academics involved in the energy storage
industry. The matrix has a minus sign (-) showing negative aspects of the technology, a plus sign
(+) representing positive aspects, and no symbol indicates the impact is neutral.

Planet/Environment
Impacts on the environment must be carefully evaluated for any sustainability decision. Any
storage option will have some impact on the environment. However, energy storage is necessary
to continue to expand the role of renewable energy technologies into the APS generation
portfolio.
The use of PHES does not have any CO2 emissions, which is a major benefit of this type of
storage. However, PHES, like regular dams, can cause issues on the surrounding riparian
ecosystems. Dams change the nature of the river and can cause harm to both wildlife as well as
plants in the area (Kosnick, 2010). PHES also requires a lot of materials, and these construction
materials can have their own impact on the environment. For example, concrete is a major
contributor to carbon dioxide emissions globally.

19

CAES, while it can be used for large-scale storage like PHES, it differs greatly from PHES in
terms of environmental issues. For one, CAES does create carbon emissions. In a world that is
looking to greatly reduce carbon emissions as quickly as possible, the carbon emissions from this
type of energy storage are problematic. However, CAES has little impact on the surrounding
environment. CAES facilities are located underground and can use existing caverns. There is a
possibility that creating these underground caverns may create other environmental issues. If the
salt must be removed from the caverns, briny water is produced. The disposal of this water can
cause other environmental problems. However, the salt from the caverns may be extracted and
processed by salt companies. If the briny water is disposed of properly or if the salt can be used
in a commercial application, the environmental issues with the salt may not be a problem.
Li-ion batteries also impact the environment. Like PHES, Li-ion batteries do not emit carbon
emissions while in use. Li-ion batteries do not necessarily negatively impact the surrounding
environment either. However, for Li-ion batteries to provide enough storage capacity, they must
be large battery installations. This means that undeveloped desert landscape must be transformed
to accommodate a large battery installation. The construction and subsequent battery system
could disrupt this ecosystem. And finally, the materials needed for Li-ion batteries are
problematic due to the mining of precious metals necessary for the batteries. Additionally, Li-ion
lifespans are somewhere between 5-16 years (Diaz-Gonzalez et al., 2012; Evans et al., 2012).
Therefore, the disposal of these batteries must be considered when analyzing the environmental
sustainability of batteries.

Profit/Costs
The cost of any of the storage options is an important aspect to consider because APS ratepayers
will ultimately pay for any of these options, much like they would pay for new generation plants.
Externalities must also be part of the cost discussion for any new technology. Externalities are
not explicitly discussed in this section; however, the externalities of the storage technologies are
encapsulated in the entire sustainability matrix.
Determining a potential cost for PHES may be difficult due to the extreme variability in costs for
this type of project. As outlined in the literature review, costs vary from $600-$3,300 per kW.
(Foley et al., 2015; Evans et al., 2012). This includes the construction as well as the maintenance
of the system. Compared to the other options, PHES is one of the most expensive storage
options. With PHES, a major benefit of this storage technology is that the operation and
maintenance costs are very low. And finally, PHES has a long lifetime. Conservative estimates
put the lifetime at 40-50 years, but other sources say these plants last up to 100 years (Dunn et
al., 2011; Evans et al., 2012; Deane et al., 2009). This is important because unlike batteries, this
system can be used for decades without the need to be completely replaced. PHES is an
extremely reliable storage technology. It is not only the most common form of energy storage
globally, but it also has set the requirements for energy storage (Larcher & Tarascon, 2015).
The costs for CAES can vary greatly as well because of the specifics of the sites to be used for
the plant. However, for this exercise, we will assume that the plant will use caverns underground
as Arizona has caverns available. This is a slightly cheaper option as a structure above ground
that can tolerate compressed air will not need to be constructed. Costs for CAES range from
$400-$800 per kW (Beaudin et al., 2010; Chen et al., 2013; Chen et al., 2008). CAES is also a
20

long-term asset. The plant in Germany was built in 1978 and is still in use 40 years later. The
plant in Alabama is a bit newer, but it has been in operation for almost 30 years. These two
plants are also extremely reliable (Budt et al., 2016; Chen et al., 2008; Beaudin et al., 2010). One
thing to consider with CAES is the need to purchase fossil fuels, most likely natural gas, to run.
The fuel is used to heat the gas after it has been depressurized and needed to create electricity.
This is an on-going cost for any CAES plant.
Cost is one of the most talked about problems with batteries, though these batteries are cheaper
than CAES and PHES. The cost of batteries is more straightforward than either PHES or CAES.
Right now, batteries cost about $500 per kW (Roberts, 2017; Baidawi, 2017). There is also
optimism that battery costs will fall as technology either changes or improves. The increasing
popularity of electric cars, which use a similar Li-ion battery, may help reduce the costs of Li-ion
batteries for storage in the future. Unlike, PHES or CAES, Li-ion batteries are not a long-term
asset as they must be replaced. Utility companies would need to consider the purchase of
batteries in the future as an additional cost in terms of battery deployment. While Li-ion batteries
are reliable in small portable electronics, using these batteries as large-scale storage is a new
application of this technology. Reliability of this type of storage will need to be evaluated over
the next several years as batteries become more popular.

People/Communities
The impact on communities from any storage option needs to be considered as well. Often, the
impacts to communities are not fully considered when decisions are made by large companies or
corporations. However, all communities must be acknowledged when a decision may impact
them, either directly or indirectly.
With PHES, there is the issue of displacement of communities due to the construction of the two
reservoirs. The most convenient location for PHES may be in locations where people live. In the
case of the Big Chino Valley PHES proposal, communities may not be displaced due to the
location of the reservoirs. However, there is concern that the amount of fresh water needed for
the storage could negatively impact the livelihoods of people living in communities near this
project. PHES could also negatively impact people’s enjoyment of natural areas. Constructing a
large PHES facility would alter the natural environmental significantly. The characteristics of
rivers, streams, and small ponds are likely to change, and there would be the addition of the
mechanisms that move the water from one reservoir to another. These would all alter the
landscape in ways that may impact the abilities of communities to enjoy nature. Finally, due to
the geographic specifications of PHES, there would likely be new transmission lines needed. As
described above, these transmission lines are large, unsightly, and can impact the natural
environment.
The location of the available CAES sites in Arizona do not necessarily have the same
displacement impact communities compared to PHES. One of the sites considered in Phoenix is
located near Luke Air Force Base. Construction of a plant on this property would not require
communities to move. Due to the geographic limitations of CAES, this storage option is like
PHES in that it may require the construction of new transmission lines. However, because CAES
locations may be closer to areas of large energy demand, new transmission lines are not a given.
And finally, both CAES plants in the world use underground caverns. There are parts of the plant
21

do require space aboveground, which may impact the enjoyment of the natural area around the
site depending on the location of the plant. Although it is less dramatic than PHES, this is a point
to consider for this storage technology.

Picture of McIntosh CAES plant (www.wired.com)
Since batteries can be located anywhere, the chances of displacement are extremely small.
Batteries can be placed adjacent to renewable energy farms, at the substation, or even at
distribution feeders. This means that the location is already developed, and communities would
not need to be moved from these locations. The flexibility of location also indicates that
transmission lines do not necessarily need to be built for this storage option. The downside with
batteries in terms of communities is the size of batteries needed for the duck curve in Arizona. A
battery system would have to be large and this could negatively impact enjoyment of the natural
environment. These batteries are large square boxes and would not improve the natural aesthetic
of an area.

Summary
Using the information above, CAES is deemed the most sustainable storage option for APS even
though it has some carbon emissions. This is because it outweighs the other two options in the
three pillars of sustainability. CAES it has less of an impact on the environment than batteries or
PHES, has moderate costs compared to PHES, and its impact on communities is not as
drastically negative as PHES. Additionally, it is one of only two options that has the potential to
be used for seasonal storage which makes this option even more attractive to APS. This is not to
say that the other two options do not have their own benefits. Li-ion batteries came in a close
second to CAES as a sustainable storage option for APS. These batteries have much less of a
potential impact on communities, produce no carbon emissions while in use, and the costs for the
batteries are falling. PHES came in third even though it is a zero-carbon source of energy. This is
because of the negative impacts it poses to both riparian ecosystems and surrounding
communities.
22

Storage Technology
PHES* possible for
seasonal storage
application

Environment
+ Zero-carbon energy storage
system
- Damage to riparian ecosystems
- Many materials needed for
construction

Economics
- High cost ($600-$3,300 per
kW)
+ Long lifetime (50-100 years)
+ Low maintenance costs

CAES* possible for
seasonal storage
application

- Emits some carbon emissions
(about one-third those of a regular
gas turbine)
+ Little impact on surrounding
environment
Briny water: disposal of water may
be issue; however, salt may be
extracted and used (neutral)
+ Zero-carbon energy storage
system
- Large, aboveground installations
- Concern over precious metals
used in battery

+ Moderate costs ($400-$800
per kW)
+ Long lifetime
- Moderate maintenance costs
(must purchase natural gas)

Li-ion batteries

+ Moderate price ($500 per
kW) with prices expected to
fall
- Must be replaced due to short
lifetime
Potential reliability issues
(neutral)

Communities
- Potential displacement of
communities for construction of
reservoirs
- Change in how communities can
enjoy the natural area
- Need for new transmission lines
+Potential CAES sites would not
require displacement of
communities
- Change in how communities can
enjoy the natural area
- Potential for new transmission
lines
+ Negligible change of
displacement
+ Need for new transmission lines
unlikely
- Change in how communities can
enjoy natural area

23

Limitations
With any decision, and especially with a large decision like energy storage, there are inherent
risks with each possible choice. While risks were not included in the sustainability matrix above,
they should be considered when making a decision for the best storage option. There are both
internal and external risks for the three storage options. Some internal risks that could have a
negative impact on the storage options include issues with supply chains for Li-ion batteries,
intense legal battles for PHES, and unsuitability of cavern sites for CAES. Additionally, there are
external risks, like climate change, complete change in utility structure, or a radically new energy
storage technology, that could also negatively impact any one of the storage options above. It is
difficult to predict which, if any, of these risks may occur. However, understanding risks and
creating a risk management tool will assist APS to prepare for these potential risks.

24

Policy Recommendation
APS’ vision is to “create a sustainable energy future for Arizona.” With sustainability a part of
APS vision, decisions regarding new energy sources and the long-term plans for their energy
portfolio must include aspects of sustainability. APS is the largest electric provider in the state of
Arizona with resources that other utility companies may not have access to in their territories.
This allows APS to be a leader when it comes to renewable energy and storage projects.
A recent proposal put forth by Andrew Tobin, an Arizona Corporation Commissioner, calls for
Arizona to significantly increase its renewable energy portfolio and storage capacity. The
announcement suggested a plan for Arizona to have 80% clean energy by 2050 and 3,000 MW
of storage by 2030 (Spector, 2018). This plan also seeks to improve the reliability of the grid in
Arizona and bring down consumer costs for electricity (Spector, 2018). Tobin’s plan was big
news, not just for energy storage but for Arizona as well. If the Arizona Corporation
Commission (ACC) approves this proposal, Arizona would be ahead of any other state in terms
of energy storage.
To meet these targets, especially the storage target, utility companies like APS will need to
quickly develop more storage options. While batteries for energy storage are in use today, they
tend to be smaller in terms of capacity. If Arizona plans to have 3,000 MW of energy storage in
12 years, large-scale storage must be examined. For APS to pursue its own vision, it must
consider the sustainability aspects of storage technologies as well as new renewable energy
power plants.
APS’ business plan for 2018 incorporates solutions to the duck curve. The seriousness of this
problem requires APS to put significant resources towards a solution. The changing landscape of
the utility business is not only occurring in Arizona. In fact, the duck curve is an issue in many
other parts of the country that have experienced a boom in renewable energy. Many states, like
California and New York, are placing an emphasis on storage to complement their renewable
energy assets. Storage not only helps renewables compete with baseload power, but as shown
above, it provides other benefits to the grid.
The information from the literature reviews, along with discussions with experts in the field of
energy storage, demonstrate CAES to be the best sustainability option for dealing with the duck
curve. It is large enough to be a utility-scale storage option, it is feasible in the state of Arizona,
it has a variety of benefits including grid support, and is a reliable technology. Also, it can be
constructed relatively quickly, does not cause the same environmental problems as PHES, and is
a long-term asset for the utility.
This is not to say that CAES is a perfect solution with no drawbacks. As a society concerned
about CO2 emissions, any technology that emits carbon emissions should be carefully vetted.
However, CAES would enable APS to use its renewable sources of energy to their fullest
potential. The ability to store and then use energy from zero-carbon sources would help reduce
carbon emissions from the production of electricity in Arizona. Also, it would allow APS to
pursue more renewable energy power plants as they would now have a tool to store energy from
these intermittent energy resources. Perhaps most importantly, a large-scale storage solution
25

would allow Palo Verde to continue to provide Arizona with zero-carbon energy for many years
to come.
CAES does come at a cost to APS, a cost that will be passed onto customers of APS. However,
this kind of project could help APS in managing its grid better and has the potential to improve
the image of APS and their commitment to renewable energy. Committing to a serious and
innovative storage technology will allow APS to be a role model for other utility companies
around the country by demonstrating that large-scale storage can benefit the environment, its
customers, and the company.

Future Steps
As is widely recognized, sustainability is an incredibly complex topic. There is no one answer
that will completely transform the electricity grid into a sustainable energy grid overnight. As
outlined above, all choices for storage presented in this project have benefits and costs to the
environment and the community. In addition, these storage options are not free; in fact, they are
quite expensive. However, if the utility company wants to use its renewable assets to the fullest
and continue to reduce carbon emissions from the electricity sector, storage is imperative. Using
sustainability factors, the utility can make the most sustainable storage decision, even if that
option is not perfect.
Since storage is becoming more popular with the increase in renewables, there are a variety of
future projects that can use this current project as a starting point. Potential future projects are as
follows:
•

•

•

•

•

This project only analyzes three storage options. However, literature regarding storage
indicates that there are many other options when it comes to energy storage. Two
innovative and interesting storage concepts are desalination and electrolysis. For
desalination, excess energy is used to create drinking water from salt water. This is
particularly interesting since one of the biggest problems with desalination is the amount
of energy needed. Electrolysis also requires large amount of energy to create hydrogen.
The excess energy during the middle of the day could be used to create hydrogen which
can be used as a fuel or an additive to gasoline. A future report could analyze the
feasibility of both in Arizona.
A student could analyze the carbon emissions from CAES versus the carbon emissions if
Palo Verde were shutdown. It would be interesting to see which would cause more
carbon emissions for the state.
APS has pilot projects that include both batteries as well as other technologies. Some of
these technologies require customer participation. A future student could put together an
informational brochure on best practices for including customers in utility programs.
Another student could also work with APS on some of their pilot projects to analyze how
the programs are operating and then propose best practices on how to improve customers’
participation in such programs.
An analysis of Andrew Tobin’s proposal in terms of its sustainability is another project
option for a future student.

26

Conclusion
The issue of storing electricity is one of the main obstacles renewable energy must overcome to
provide more electricity to the grid. However, the recent uptick in research into different storage
options may allow renewables to provide most of our electricity. If these storage ideas can be put
to use in real-world applications and help renewable energy provide more of our nation’s energy,
then our grid will be part of the solution to create a more sustainable future.
If APS is serious about both solving the duck curve and its commitment to sustainability, then
CAES provides the best option when it comes to energy storage. Using CAES will not only help
APS in improving its grid; CAES provides both indirect and direct benefits to the community
and the environment. Also, APS has the potential to become a pioneer in the utility industry by
using sustainability in their decision making regarding energy storage for renewable energy.
With the deployment of CAES, Arizona can become a leader in both renewable energy and
storage.

27

References
Abbaspour, M., Satkin, M., Mohammadi-Ivatloo, B., Lotfi, F. H., & Noorollahi, Y. (2013).
Optimal operation scheduling of wind power integrated with compressed air energy
storage (CAES). Renewable Energy, 51, 53-59.
Ameli, N., & Brandt, N. (2015). Determinants of households’ investment in energy efficiency
and renewables: evidence from the OECD survey on household environmental behaviour
and attitudes. Environmental Research Letters, 10(4), 044015.
APS - Palo Verde Clean Energy Performance Highlighted as Industry Gathers in Ariz. (2017,
May 19). Retrieved March 13, 2018, from
https://www.aps.com/en/ourcompany/news/latestnews/Pages/palo-verde-clean-energyperformance-highlighted-as-industry-gathers-in-ariz.aspx
Arizona Emergency Information Network. (2015). Palo Verde- Facts About the Nation’s Largest
Power Producer. https://ein.az.gov/sites/default/files/palo_verde_facts.pdf
Baidawi, A. (2017, November 30). Australia Powers Up the World's Biggest Battery - Courtesy
of Elon Musk. Retrieved April 19, 2018, from
https://www.nytimes.com/2017/11/30/world/australia/elon-musk-south-australiabattery.html
BBC. Tesla mega-battery in Australia activated. (2017, December 01). Retrieved March 01,
2018, from http://www.bbc.com/news/world-australia-42190358
Beaudin, M., Zareipour, H., Schellenberglabe, A., & Rosehart, W. (2010). Energy storage for
mitigating the variability of renewable electricity sources: An updated review. Energy for
sustainable development, 14(4), 302-314.
Brundtland, G. H. (1987). Report of the World Commission on environment and development:"
our common future.". United Nations.
Budt, M., Wolf, D., Span, R., & Yan, J. (2016). A review on compressed air energy storage:
Basic principles, past milestones and recent developments. Applied Energy, 170, 250268.
Bullough, C., Gatzen, C., Jakiel, C., Koller, M., Nowi, A., & Zunft, S. (2004, November).
Advanced adiabatic compressed air energy storage for the integration of wind energy. In
Proceedings of the European Wind Energy Conference, EWEC (Vol. 22, p. 25).
Ciampoli, P. (2017, October 26). Proposal calls for 2,000-MW pumped storage facility in
Arizona. Retrieved January 26, 2018, from
https://www.publicpower.org/periodical/article/proposal-calls-2000-mw-pumped-storagefacility-arizona

28

Chen, H., Cong, T. N., Yang, W., Tan, C., Li, Y., & Ding, Y. (2009). Progress in electrical
energy storage system: A critical review. Progress in Natural Science, 19(3), 291-312.
Chen, H., Zhang, X., Liu, J., & Tan, C. (2013). Compressed air energy storage. In Energy
Storage-Technologies and Applications. InTech.
D’Aprile, P., Newman, J., & Pinner, D. (2016, August). The new economics of energy storage.
Retrieved March 01, 2018, from https://www.mckinsey.com/businessfunctions/sustainability-and-resource-productivity/our-insights/the-new-economics-ofenergy-storage
Deane, J. P., Gallachóir, B. Ó., & McKeogh, E. J. (2010). Techno-economic review of existing
and new pumped hydro energy storage plant. Renewable and Sustainable Energy
Reviews, 14(4), 1293-1302.
Díaz-González, F., Sumper, A., Gomis-Bellmunt, O., & Villafáfila-Robles, R. (2012). A review
of energy storage technologies for wind power applications. Renewable and sustainable
energy reviews, 16(4), 2154-2171.
Dunn, B., Kamath, H., & Tarascon, J. M. (2011). Electrical energy storage for the grid: a battery
of choices. Science, 334(6058), 928-935.
Evans, A., Strezov, V., & Evans, T. J. (2012). Assessment of utility energy storage options for
increased renewable energy penetration. Renewable and Sustainable Energy Reviews,
16(6), 4141-4147.
Fertig, E., & Apt, J. (2011). Economics of compressed air energy storage to integrate wind
power: A case study in ERCOT. Energy Policy, 39(5), 2330-2342.
Foley, A. M., Leahy, P. G., Li, K., McKeogh, E. J., & Morrison, A. P. (2015). A long-term
analysis of pumped hydro storage to firm wind power. Applied Energy, 137, 638-648.
Fthenakis, V., Mason, J. E., & Zweibel, K. (2009). The technical, geographical, and economic
feasibility for solar energy to supply the energy needs of the US. Energy Policy, 37(2),
387-399.
Greenblatt, J. B., Succar, S., Denkenberger, D. C., Williams, R. H., & Socolow, R. H. (2007).
Baseload wind energy: modeling the competition between gas turbines and compressed
air energy storage for supplemental generation. Energy Policy, 35(3), 1474-1492.
Hadjipaschalis, I., Poullikkas, A., & Efthimiou, V. (2009). Overview of current and future
energy storage technologies for electric power applications. Renewable and sustainable
energy reviews, 13(6-7), 1513-1522.

29

Hill, C. A., Such, M. C., Chen, D., Gonzalez, J., & Grady, W. M. (2012). Battery energy storage
for enabling integration of distributed solar power generation. IEEE Transactions on
smart grid, 3(2), 850-857.
Ibrahim, H., Ilinca, A., & Perron, J. (2008). Energy storage systems—characteristics and
comparisons. Renewable and sustainable energy reviews, 12(5), 1221-1250.
Kosnik, L. (2010). The potential for small scale hydropower development in the US. Energy
Policy, 38(10), 5512-5519.
Larcher, D., & Tarascon, J. M. (2015). Towards greener and more sustainable batteries for
electrical energy storage. Nature chemistry, 7(1), 19.
Lindley, D. (2010). Smart grids: The energy storage problem. Nature News, 463(7277), 18-20.
Lund, H., & Salgi, G. (2009). The role of compressed air energy storage (CAES) in future
sustainable energy systems. Energy Conversion and Management, 50(5), 1172-1179.
Lund, H., Salgi, G., Elmegaard, B., & Andersen, A. N. (2009). Optimal operation strategies of
compressed air energy storage (CAES) on electricity spot markets with fluctuating prices.
Applied thermal engineering, 29(5-6), 799-806.
Luo, X., Wang, J., Dooner, M., & Clarke, J. (2015). Overview of current development in
electrical energy storage technologies and the application potential in power system
operation. Applied Energy, 137, 511-536.
Ma, T., Yang, H., & Lu, L. (2014). Feasibility study and economic analysis of pumped hydro
storage and battery storage for a renewable energy powered island. Energy Conversion
and Management, 79, 387-397.
Merchant, E. F. (2017, September 20). Lithium-Ion Battery Production Is Surging, but at What
Cost? Retrieved March 27, 2018, from
https://www.greentechmedia.com/articles/read/lithium-ion-battery-production-is-surgingbut-at-what-cost#gs.tMldG9I
Motion to Intervene, Town of Clarkdale, Arizona. 2017. United States of America. Federal
Energy Regulatory Commission.
Munsell, M. (2017, February 15). US Solar Market Grows 95% in 2016, Smashes Records.
Retrieved September 5, 2017, from https://www.greentechmedia.com/articles/read/ussolar-market-grows-95-in-2016-smashes-records
Neal, J. T., & Rauzi, S. L. (1996). Storage opportunities in Arizona bedded evaporites (No.
SAND--96-1679C; CONF-9610163--2). Sandia National Labs., Albuquerque, NM
(United States).

30

Personal interview, John Pinho, March 16, 2018.
Personal interview, Marc Romito, February 21, 2018.
Power South. (n.d.). Compressed Air Energy Storage McIntosh Power Plant.
Rehman, S., Al-Hadhrami, L. M., & Alam, M. M. (2015). Pumped hydro energy storage system:
A technological review. Renewable and Sustainable Energy Reviews, 44, 586-598.
Roberts, D. (2017, November 28). Elon Musk bet that Tesla could build the world's biggest
battery in 100 days. He won. Retrieved April 19, 2018, from
https://www.vox.com/energy-and-environment/2017/11/28/16709036/elon-musk-biggestbattery-100-days
Sharma, A., Chiu, H. H., Ahrens, F., Ahluwalia, R. K., & Ragsdell, K. (1979). Design of
optimum compressed air energy-storage systems. Energy, 4(2), 201-216.
Simmons, J., Barnhart, A., Reynolds, S., & Son, Y. J. (2010). Study of Compressed Air Energy
Storage with grid and photovoltaic energy generation. The Arizona Research Institute for
Solar Energy (AzRISE)-APS Final Draft Report, Compressed Air Energy Storage and
Photovoltaics Study, University of Arizona.
Steffen, B. (2012). Prospects for pumped-hydro storage in Germany. Energy Policy, 45, 420429.
Scrosati, B., & Garche, J. (2010). Lithium batteries: Status, prospects and future. Journal of
Power Sources, 195(9), 2419-2430.
Spector, J. (2017, September 15). California's big battery experiment: a turning point for energy
storage? Retrieved March 01, 2018, from https://www.theguardian.com/sustainablebusiness/2017/sep/15/californias-big-battery-experiment-a-turning-point-for-energystorage
Spector, J. (2018, January 30). Arizona Regulator Proposes Biggest Storage and Clean Energy
Target Yet. Retrieved February 28, 2018, from
https://www.greentechmedia.com/articles/read/arizona-regulator-proposes-sweepingclean-energy-plan#gs.qx_7kXY
Succar, S., & Williams, R. H. (2008). Compressed air energy storage: theory, resources, and
applications for wind power. Princeton environmental institute report, 8.
Swider, D. J. (2007). Compressed air energy storage in an electricity system with significant
wind power generation. IEEE transactions on energy conversion, 22(1), 95-102.
Uria-Martinez, R., O'Connor, P. W., & Johnson, M. M. (2015). 2014 hydropower market report
(No. DOE/EE-1195). Oak Ridge National Laboratory.

31

Yang, B., Makarov, Y., Desteese, J., Viswanathan, V., Nyeng, P., McManus, B., & Pease, J.
(2008, May). On the use of energy storage technologies for regulation services in electric
power systems with significant penetration of wind energy. In Electricity Market, 2008.
EEM 2008. 5th International Conference on European (pp. 1-6). IEEE.
Yang, C. J., & Jackson, R. B. (2011). Opportunities and barriers to pumped-hydro energy storage
in the United States. Renewable and Sustainable Energy Reviews, 15(1), 839-844.
U.S. EPA. Sources of Greenhouse Gas Emissions. (n.d.). Retrieved September 5, 2017, from
https://www.epa.gov/ghgemissions/sources-greenhouse-gas-emissions

32