The Power and Transportation Future - Hydrogen and Fuel Cells: Advances in Transportation and Power - page 275

also help the launch problem, but these are not a well developed technol-
ogy.
A typical solar panel mass is 20-kg-per kilowatt, then without con-
sidering the mass of the support structure, antenna or any significant mass
reduction of focusing mirrors, a 4-GW power station would weigh about
80,000 metric tons. Although, this is huge, a space solar-panel would not
need to support its own weight and would not be subject to earth's corro-
sive atmosphere. Some very lightweight designs could allow 1-kg/kW, or
4000 metric tons for a 4-GW station. This could require 40 to 800 launches
to send the material to low earth orbit, where it would be turned into sub-
assembly solar arrays and then ion-engine style rockets would move them
to final orbit.
With the cost for a shuttle-based launch at $500 million to $800 mil-
lion, total launch costs would range between $20 billion for low weight
panels to $320 billion for heavy panels. This does not include the cost of
assembly or materials and manufacturing. For each gigawatt rating, a SPS
system can generate 8.75 terawatt hours of electricity per year. At $0.11 per
kWh one gigawatt is worth about $1 billion per year. A 4-GW SSPS could
generate over $40 billion each decade.
Prices for electricity fluctuate depending on the time of day. England
has electricity costs of 22 cents per kilowatt hour and is further north than
most inhabited parts of Canada and receives limited solar radiation over
much of the year. This makes conventional solar power not very compet-
itive at grid delivered costs. However, kilowatt hour photovoltaic costs
have been in an exponential decline for decades, with a 20-fold decrease
from 1975 to 2001.
In order to be competitive, SPS must cost no more than existing sup-
pliers, this may be difficult, especially if it is deployed to North America.
Either it must cost less to deploy, or it must operate for a very long period
of time. Many proponents suggest that the lifetime is effectively infinite,
but normal maintenance and replacement of less durable components
makes this unlikely. Satellites do not, in our extensive experience, last for-
ever.
One concept is to build the SPS in orbit with materials from the
moon. Launch costs from the moon are about 100 times lower than from
Earth, due to the lower gravity. This concept will work if the number of
satellites to be built is near several hundred. Otherwise, the costs of setting
up the production in space and mining facilities on the moon are as high
as launching from Earth.