Renewable Energy 28 (2003) 11211127

www.elsevier.com/locate/renene

Technical note

Feasibility of tapping atmospheric charge as a

power source
M.L. Breuer
University of Massachusetts, Hasbrouck Laboratory Department of Physics, Room 411 North Pleasant
Street, Amherst, MA 01003, USA
Received 1 May 2002; received in revised form 19 May 2002; accepted 2 October 2002

Abstract
Mechanisms which generate atmospheric charge are examined in order to estimate the rate
of field regeneration. Energy density contained in the atmospheric electric field is examined.
Charge-harvesting methods are suggested and a practical use of electrostatic motors examined.
Comparisons to projected energy consumption, and comparison to other energy sources are
made, and viability of tapping atmospheric charge is assessed.
2002 Elsevier Science Ltd. All rights reserved.
Keywords: Atmospheric charge; Atmospheric current density; Atmospheric electric field; Atmospheric
voltage gradient; Charge separation; Charge transport; Cosmic ray; Electrometry; Electrostatic motors;
Lightning; Thunderstorm

1. Introduction
We have been aware of the presence of atmospheric electrical energy, usually in
the form of lightning, throughout history. The quantitative study of this charge began
as early as the mid 1700s with the experiments of Benjamin Franklin, for instance
(References to Franklins work in 1752 may be found in [1], p. 507). The most
popular awareness of Franklins work with atmospheric electricity has to do with
his experiments with kites during thunderstorms. However, he also constructed a
detector of atmospheric charge which consisted of two bells, one attached to an
antenna, the other to ground, and between which was suspended a pendulum. The

Tel.: +1-413-218-1713.
E-mail address: [email protected] (M.L. Breuer).

0960-1481/03/$ - see front matter 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 9 6 0 - 1 4 8 1 ( 0 2 ) 0 0 2 1 0 - 0

1122

M.L. Breuer / Renewable Energy 28 (2003) 11211127

Nomenclature
C
Jz
Q
r
V
W

capacitance (Farads)
current density (Amps / meter2, or Coulombs / (second meter2)
charge (Coulombs)
radius (meter)
voltage (Volts, or Joules / Coulomb)
work, energy (Joules)

Greek symbols
p
!E

3.14159
energy density (Joules / meter3)

pendulum rang the bells by being alternately attracted to the charge of each. The
frequency was directly proportional to the atmospheric electric charge and indicated
the approach of storms and thus atmospheric electrification. This also served as a
charge transport device and has been the subject of contemporary research [2,3].
Since then, atmospheric electrometry has become far more sophisticated and relies,
among other things, upon ground-based electronics, electrometers and field
mills[4,5], upon other instruments borne aloft by balloons, and upon optical instruments aboard satellites circling the earth in various orbits[6].
2. Mechanisms of atmospheric charge production
More than one mechanism is responsible for this atmospheric charge. It is somewhat generally agreed that the principal source of charge separation in the atmosphere
results from thunderstorms generally located about the equator of the earth. It is
estimated that approximately 2000 thunderstorms are active around the earth at any
given time with maximum intensity occurring at mid-afternoon local time and
cover an area of about 5 x 1011m2, or 1/1000 of the earths surface [6, p. 21]. Values
for the current of charge separation produced by these tropical thunderstorms range
from 1000 to 2000 amps [8,9]. Since the conductivity of the atmosphere increases
exponentially as a function of altitude, but the atmospheric electric field decreases
exponentially as a function of altitude (and the two factors scale equally) the atmospheric current density Jz in the earthatmosphere electric circuit (Fig. 1.) is essentially
constant [10].
Cosmic rays produce ionization of atmospheric gases at an intensity that decreases
at a negatively exponential rate, as a function of penetration of the atmosphere [10],
but the primary factor in the separation of the ionized particles is the electric field

M.L. Breuer / Renewable Energy 28 (2003) 11211127

Fig. 1.

1123

Earthatmosphere electric circuit.

resulting from tropical thunderstorms. Ionization due to cosmic rays fluctuates as a

result of diurnal exposure to the solar wind and to the suns coronal mass ejections
whose matter can partially shield the earth from the influx of these rays. Other
sources of ionization in the atmosphere include dust and sand storms, volcanic
activity, man-made radioactive contamination and earthquakes (through the release
of ionizing radon) [11].
The mechanism of charge separation provided by thunderstorm activity produces
a net positive charge above the earth, and a net negative charge on the earths surface.
Thus, the electric field is oriented downward, normal to the surface. It is estimated
that if all thunderstorm activity would cease lowering negative charge to the surface
all the charge would be conducted up into the atmosphere on a time scale of less
than one hour [9].
3. Energy density characterization
Although atmospheric charge is distributed vertically as a function of a number
of factors, we will now look at an ideal case in order to examine some order of
magnitude approximations. The main electric field gradient from charge separation
is produced across a distance of approximately 30 miles from the surface upward,
or the point where the electrosphere begins. The potential between these two layers
is approximated to be 3 x 105 volts, and is maintained by a charge in the neighborhood of 1 x 106 Coulombs at the earths surface [9]. In an ideal calculation, via
the familiar
C " Q/V

(1)

for capacitors, and the two values just cited, we determine the capacitance to be
3.33 Farad.
Next, using the also familiar
W " .5QV

(2)

1124

M.L. Breuer / Renewable Energy 28 (2003) 11211127

with the two values cited, we determine the energy required to produce this charge
separation to be 1.5 x 1011 Joules. Realizing that the charge is not located on two
parallel, flat plates of a capacitor, but is distributed across the earthelectrosphere
layer and therefore would require less energy for the charge separation we
down-scale this value to an approximate 1.0 x 1011 Joules. A determination of the
energy density of this spherical shell volume is then given by
!E "

.5QV

4 3 3
p(r #r )
3 b a

"

(3)

where rb and ra are the radius of the electrosphere layer and the radius of the earths
surface, respectively, from the earths center. This equation returns a value of 5.67
x 108 J/m3.
In comparison to the energy density of fossil fuels, such as 3.3 x 1010 J/m3 for
crude oil, 4.0 x 107 J/m3 for natural gas, 3.9 x 1010 J/m3 for coal, or other sources,
such as 1.42 x 1018 J/m3 for nuclear fission fuels 1, the atmospheric energy density
is 15 to 26 orders of magnitude smaller. It may reasonably be argued, however, that
means of mining and transporting atmospheric charge is far simpler: drive a stake
in the ground, erect an antenna and begin to tap the charge immediately; the thunderstorms will take care of the rest. In short, no energy is expended in the collection
of the fuel. Naturally, the very first question is: but is this feasible?
4. Charge harvesting and potential uses
The primary difficulty of using an antenna to harvest charge is its limitation to
a fixed location. Alternatives include the use of wire-trailing rockets, as used in
lightning research: these directly contact a pocket of charge residing in thunderstorm
clouds but they are subject to the demands of maintenance and retrieval, as well as
to the difficulties of launch in inclement weather (not to mention potential dangers
to aircraft). It simply is not realistic on any regular basis. A third alternative that is
not without precedent, however, is the use of an ionizing laser beam to create a
conducting path for charge to flow along [12,13]. Since researchers cited in the
references regarding laser-induced atmospheric electric discharge had other things
in mind, questions of energy required to create the path vs. energy retrieved need
to be worked out.
Other suggestions for efficient means of harvesting charge include pinpoint
antennae whose tips are coated with radioactive elements. Radioactivity ionizes
atmospheric gases and increases the conduction of charge to the metal.
The atmospheric voltage gradient at the earths surface may vary widely as a result
of local weather activity: from 100 V/m to 900 V/m in warm front rain, and by
as much as 2000 V/m in rain associated with a cold front [14]. Short of harvesting
1

Based on the density of uranium.

M.L. Breuer / Renewable Energy 28 (2003) 11211127

1125

concentrations of charge, say in thunderclouds, it has been demonstrated that local

atmospheric fine weather electric fields, on the order of 100 V/m [15] have been
sufficient to operate electrostatic motors producing 0.1 horsepower (or nearly 75
watts) from nothing more than an antenna and a ground wire [16]. Recent work,
which is greatly increasing the power density, and therefore efficiency, of electrostatic motors through the employment of thin-film technology [17,18,19,20,21], has
achieved values of 230 W/kg, which is nearly the same as that of electromagnetic
motors. These new devices have yet to be tested with an atmospheric field source.
As an example of a practical application however, such systems could conceivably
be employed in remote field instruments requiring some small motive power but
where periodic servicing of batteries is not realistic.
5. Comparison to US energy consumption, and to other power sources
We now move to the topic of our current and projected energy needs. A recent
report released by the US Department of Energy estimates the current total US energy
consumption will increase from 99.3 to 130.9 quadrillion Btu between the years 2002
and 2020, an average increase of about 1.4% annually [22]. To obtain a comparison
with atmospheric power available we may take the round value of 100 quadrillion,
or 1.0 x 1017 Btu, or 1.06 x 1020 J for the estimated annual energy consumption.
Our previous value of 1.0 x 1011 J for the energy contained in atmospheric charge
represents only a billionth of the current annual energy consumption. This means
the entire energy of this charge would have to be harvested 1.06 x 109 times annually,
or 2 x 103 times a minute! Using the estimate of less than an hour for existing earth
charge to be conducted away into the atmosphere as our time scale for enough current
to flow for the field to regenerate, we immediately see the slow restoration by thunderstorms of charge artificially tapped by any sort of industry is untenable. And the
attendant effects on the global weather system, were such tapping feasible, are currently unknown [23], but could be considerable. Lastly, if we were to drain only
half the energy in atmospheric charge once an hour (assuming a collection system
having an unrealistic 100% efficiency) this would meet only 4.0 x 104 % of current
energy consumption. Such considerations effectively eliminate this potential source
of energy for any but small, specialized power needs.
In contrast, the most enduring planetary influx of energy is provided by insolation.
At the radius of the earths orbit, areal power density is on the order of 1.0 x 1010
erg/(m2-s), or 1.0 x 103J/(m2-s). Using a value of 6.38 x 106m for the earths radius
yields a value of 4.0 x 1024 J annual influx due to insolation upon the area presented
by the profile of the earths disk. If we use the value of 8 hours to estimate the
effective time any solar collection device is exposed to sunlight, this yields a value
of 2.88 x 107 J/(day-m2), or 1.05 x 1010 J/(year-m2). Assuming no atmospheric losses,
and a collection/conversion, panel system that is 100% efficient, this means an area,
somewhere in the US, equaling 1.0 x 1010m2 would need to be covered to meet all
our current energy needs. Taking into account the 30% efficiency of photovoltaic
elements essentially increases this figure by a factor of 3 bringing the area to 3.0 x

1126

M.L. Breuer / Renewable Energy 28 (2003) 11211127

1010m2, or 3.0 x 104km2, or 1.16 x 104miles2. This amounts to about 9/10 the area
of the state of Texas. If, on the other hand, we design a system using these
calculations to meet only our electrical consumption (which amounts to about
one-third the annual energy consumption) this would require a land area equal to
about 3/10 that of Texas, or about 3860 miles2. Obviously, design of such systems
will have to take into account sites where ecological impact is minimized, where
weather patterns provide a maximal insolation, and where transmission losses
through the existing power grid are minimized. That would place such systems in
the least populated areas of the country.
As a side note, we do have enormous power densities in fission and fusion fuels,
as partly noted above. However, comparison to these two sources is not being made
here because the first has proven to be too serious a long-term ecological and biological contaminant, and the second is currently in too developmental a stage to obtain
any information on working efficiencies.
6. Conclusion
In conclusion, it has been demonstrated that low-power requirements for a limited
number of devices such as electrostatic motors can be met by tapping atmospheric
electricity. However, barring experimental measurement of any factors that significantly alter our understanding of the earthatmosphere electric circuit, it is not likely
that we can expect the energy contained in the atmospheric electric field to provide
a viable, large-scale source of power. But it is also becoming increasingly clear that
the near future will force us to examine other systems and consider alternatives as
the exhaustion of our dwindling reserves of coal and oil inevitably approaches.

Acknowledgements
The author is grateful for stimulating discussions with Dr. Robert Guyer and Dr.
Robert Krotkov, both of the University of Massachusetts at Amherst, and with Dr.
Earle Williams of the Massachusetts Institute of Technology, in the preparation of
this paper.

References
[1] Benjamin P. The Intellectual Rise in Electricity. New York: Appleton, 1895.
[2] Tuominen MT, Krotkov RV, Breuer ML. Stepwise and hysteretic transport behavior of an electromechanical charge shuttle. Phys. Rev. Lett. 1999;83:3025.
[3] Krotkov RV., Tuominen MT., Breuer ML., Franklins bells and charge transport as a undergraduate lab, Am. J Phys, 69, 55
[4] Uman M. Lightning. New York: Dover, 1969.
[5] Israel H., Atmospheric Electricity. volume I, (National Science Foundation, 1970), p. 197.
[6] Volland H. Atmospheric Electrodynamics. Heidelberg: Springer-Verlag, 1984.

M.L. Breuer / Renewable Energy 28 (2003) 11211127

1127

[8] Tinsley BA. Influence of solar wind on the global electric circuit, and inferred effects on cloud
microphysics, temperature, and dynamics in the troposphere. Space Science Reviews 2000;00:4.
[9] Uman MA. All About Lightning. New York: Dover Publications, 1986.
[10] Williams E, MIT Lincoln Labs, private communication. January 2002.
[11] Pulinets SA., Khegai VV., Boyarchuk KA., Lomonosov AM. The atmospheric electric field as a
source of variability in the ionosphere. Physics - Uspekhi 41 (5), 516.
[12] Raizer YP. Gas Discharge Physics. Heidelberg: Springer-Verlag, 1997.
[13] Wang D, Kawasaki ZI, Matsuura K, Shimada Y, Uchida S, Yamanaka C, Fujiwara E, Izawa Y,
Simokura N, Sonoi Y. A preliminary study on laser-triggered lightning. J. Geophys. Res.
1994;99:1690712.
[14] Chalmers JA. Atmospheric Electricity. London: Pergamon Press, 1967.
[15] Webb W. Earths Electrical Structure. White Sands Missile Range, New Mexico: Atmospheric
Sciences Laboratory, 1975.
[16] Jefimenko O. Electrostatic Motors, Their History, Types and Principles of Operation. Star City, VA:
Electret Scientific Company, 1973.
[17] Niino T., Egawa S., Kimura H., Higuchi T. Electrostatic artificial muscle: compact, high-power
linear actuators with multiple-layer structures. Proc. IEEE Micro Electro Mechanical Systems Workshop 94, pp. 130-135 (1994).
[18] Niino T., Higuchi T., Egawa S. Dual excitation multiphase electrostatic drive. Conference Record
of the 1995 IEEE Industry Applications Society Annual Meeting, pp. 1318-1325 (1995).
[19] Yamamoto A., Niino T., Higuchi T. High power electrostatic actuator with novel electrode design,
Proc. IEEE Micro Electro Mechanical Systems Workshop 98, pp. 408-413 (1998).
[20] Yamamoto A, Niino T, Higuchi T. High precision positioning control using high-power electrostatic
linear drive. Journal of the Japan Society for Precision Engineering 1998;64(9):13859 In Japanese.
[21] Yamamoto A., Niino T., Higuchi T. A high-power electrostatic linear servo motor. Proc. Linear
Drives for Industry Applications 98, pp. 212-215 (1998).
[22] U.S. Department of Energy, Annual Energy Outlook 2002 with Projections to 2020, Report #
DOE/EIA-0383 (2002), December 21, 2001.
[23] Williams E. MIT Lincoln Labs. private communication. April 2002.