brandon - doctoral preliminary exam
TRANSCRIPT
Steven Brandon 31 July 2004
1
WRITTEN PRELIMINARY DOCTORAL EXAMINATION FOR STEVEN BRANDON
Dr. Paul L. Dawson - July 21, 2004
NNNAAANNNOOO---SSSCCCAAALLLEEE SSSPPPAAARRRKKKIIINNNGGG DDDEEEVVVIIICCCEEE
PhD examination for Steve Brandon (7-21-04). You have 10 days to finish
this so you can have plenty of time to think about the solution.
1. Nanotechnology is currently a focus of scientific research. I had an
idea that might be considered nanotechnology. The idea is related to the
generation of nanoscale electrical “sparks” on surfaces to inhibit or kill bacteria. The questions I have related to this idea are:
a. What is the minimum distance (gap) between two points that can
produce an electrical spark? b. What type of particles could be embedded on a surface that when
the whole surface was charged with an electrical impulse would generate sparks?
c. What is the minimum electrical charge needed to generate a spark between our nanoparticle gaps?
2. The big follow-up question is develop a detailed description of how you
would develop and test a 3 x 3 inch surface that when charged with a suggested amount of electricity, would produce numerous sparks across the
whole surface.
Since I know what detail you go into on all of your work this will be the only question I have for you. Also, if we can develop this I would like to include you in an invention disclosure, so we can keep this confidential until we see
if it is feasible.
Steven Brandon 31 July 2004
2
Question 1a What is the minimum distance (gap) between two points that can produce an electrical spark?
Sparks
A spark gap consists of an arrangement of two conducting electrodes
separated by a gap filled with some gas (usually air). When a suitable
voltage is supplied, a spark forms, ionizing the gas and drastically reducing its electrical resistance. An electric current then flows until the path of
ionized gas is broken. This happens usually when the voltage drops, but in
some cases when the heated gas rises, stretching out and then breaking the
filament of ionized gas. Usually the action of ionizing the gas is violent and disruptive, often leading to sound (ranging from a snap for a spark plug to
thunder for a lightning discharge), light and heat.1 (See Figure 1, below.)
Figure 1: A spark gap2
Plasma
Refer to Appendix Section A1, Sparks and Lightning, for an entertaining
primer on the physics of sparks and lightning. This article explains that
ionized gas produced by sparks is plasma, the fourth state of matter as distinguished from solids, liquids and gas. Figure 2, below, illustrates this
concept.
Steven Brandon 31 July 2004
3
Figure 2: Four states of matter3
Plasmas consist of freely moving charged particles, i.e., electrons and ions.
Formed at high temperatures when electrons are stripped from neutral atoms, plasmas are common in nature. For instance, stars are predominantly plasma. Plasmas are the referred to as "Fourth State of
Matter" because of their unique physical properties, distinct from solids,
liquids and gases. Plasma densities and temperatures vary widely.4 Figure 3, below, illustrates the relationship between lightning (and sparks) and
other forms of plasma.
Steven Brandon 31 July 2004
4
Figure 3: Various kinds of plasma. 5
Destruction of bacteria and other micro-organisms by plasmas has been
reported. Plasma exposure, while being lethal to both Gram-positive and Gram-negative bacterial classes, also produced gross structural damage in
the Gram-negative E. coli while none was observed in the more structurally
robust Gram-positive Bacillus subtilis.6 A one-atmosphere, uniform-glow discharge plasma (OAUGDP), is capable of operating at atmospheric
pressure in air and providing antimicrobially active species at room
temperature. OAUGDP exposures have reduced log numbers of bacteria (Escherichia coli, Staphylococcus aureus, and Pseudomonas aeruginosa),
bacterial endospores (Bacillus subtilis and Bacillus pumilus), and various
yeast and bacterial viruses on a variety of surfaces. These surfaces included
polypropylene, filter paper, paper strips, solid culture media, and glass. Experimental results showed at least a 5 log10 colony forming units (CFU)
reduction in bacteria within a range of 15 to 90 s of exposure.7 A device
using electrostatic charge to remove airborne bacteria from poultry plants is described in Section A8, Zapping Airborne Salmonella and Dust.8 Another
paper presents the results of a study involving plasma arcs submerged in
water.9
Steven Brandon 31 July 2004
5
Paschen Curve
The article in Appendix Section A2, Electric Sparks, gives a technical
explanation of the physics of sparks and introduces Paschen’s Law, which
states that the voltage difference across a gap between oppositely charged
electrodes, V, is a function of the product of the gas pressure in the gap, p, and the distance separating the two electrodes, the “spark gap,” d, i.e., V =
f(pd). The article in Section A3, Paschen’s Law, provides formulas relating
breakdown voltage to the gas pressure-spark gap product, pd. These
formulas indicate a roughly linear relationship between breakdown voltage, which must be overcome to initiate a spark discharge, and product pd. See
Figure 4, below.
Figure 4: The Paschen Curve10
The linear range of pd values, for which these formulas are valid, occurs at
pd values greater than about 25 mBar-mm (1.9 torr-cm). At atmospheric
pressure in air, this corresponds to spark gap, d, of about 25 µm. As the pd product is reduced below about 25 mBar-mm, either by reducing pressure or
by reducing the spark gap, the breakdown voltage begins to level off
reaching a minimum near 7.5 mBar-mm (0.57 torr-cm). This corresponds to
a spark gap of about 7 µm in air at atmospheric pressure (at which point the breakdown voltage has fallen to its minimum value in atmospheric pressure
air, 327 V)11. Below this minimum, the voltage required to initiate a spark
increases rapidly.12 (See Section A6, The Paschen Curve, for a more detailed
description of this phenomenon.)
Steven Brandon 31 July 2004
6
Most commonly, low pd values have been obtained by reducing gas
pressure. This is the approach employed in a device known as a pseudo-spark switch which is used in high voltage applications. Since, according to
the Paschen Law, breakdown voltage is a function of the product pd, the
phenomenon described above should apply equally well to atmospheric
pressure applications with microscopic spark gaps (d < 25 µm). This means
that, while it is possible to produce sparks in atmospheric pressure air with
microscopic spark gaps, the voltage which must be overcome to initiate the spark may become so large as to become impractical.
However, Wallash and Levit13, who report the results of their study of electrical breakdown behavior for devices with micron and sub-micron gaps
between conductors, came to a different conclusion. They indicate that the
conventional Paschen curve does not adequately describe behavior in these
small spark gaps. Instead they explain that a Modified Paschen Curve should be employed for air at 1 atmosphere pressure and spark gaps smaller than 5 µm. See Figure 5, below.
Figure 5: Paschen curve and Modified Paschen Curve
The modified Paschen curve is explained in more detail in another
reference.14
Wallash and Levit conclude “that breakdown in air at atmospheric pressure can occur well below the Paschen curve minimum of 360V and should be
considered in the processing, handling and operation of devices with micron
and sub-micron gaps.” This paper also presents the following table
describing the relevant physics that dominate over various spark gap scales down to spark gaps below 2 nm, which is well into the scale of organic
macromolecules.
Steven Brandon 31 July 2004
7
Figure 6: Spark current mechanism table
From this, I can surmise that there is no practical lower limit to the minimum distance between two points that can produce an electrical spark.
Question 1b What type of particles could be embedded on a surface that when the whole surface was charged with an electrical impulse would generate sparks?
Conductors
A conductor is a substance or body that offers small resistance to the passage of an electric current.15 The article in Section A7, Static Electricity
Sparks and Lightning, gives a good explanation of the necessity for conductors to form sparks. Sparks need conductors, so that the electrons
can freely move about and gather enough charges together to be able to
jump from one material to another. Static electricity is formed and gathered
on the surface of a non-conductor, but it must be then transferred to a
conductor to cause a spark. When a conductor—like a metal rod—is brought
near a charged non-conductor, the free electrons in the conductor will move
to one end or the other of the rod, depending on whether the non-conductor
surface is positive or negative.
Steven Brandon 31 July 2004
8
Figure 7: Opposite charges in conductor move toward non-conductor
When the conductor is brought into contact with the non-conductor, the
electrical charges on the surface of the non-conductor are "sucked" into the
conductor. In other words, if negative charges are on the surface of the non-conductor, these electrons will move into the conductor. If positively charged
atoms are on the surface, electrons from the metal or conductor will neutralize those atoms, resulting in an excess of positive charges in the conductor. If another conductor is brought near the first conductor, the
same thing will happen. Since electrons can move so freely in a conductor, many may collect near the surface and actually jump across the air gap as a
spark.16
In science and engineering, conductors are materials that contain movable
charges of electricity. When an electric potential difference is impressed at separate points on a conductor, an electric current appears in accordance
with Ohm's law. While many conductors are metallic, there are many non-metallic conductors as well. Under normal conditions, all materials offer some resistance to flowing charges, which generates heat. The motion of
charges also creates an electromagnetic field around the conductor that exerts a mechanical force on the conductor. Consequently, a conductor of a
given material and volume (length x cross-sectional area) has a limit to the
current it can carry without being destroyed thermally or mechanically. This effect is especially critical in printed circuits, where conductors are relatively
small.17
For these reasons, it is desirable to select a material with a high electrical
conductivity. Table 1, below lists physical properties, including electrical conductivity, of a variety of conductive materials. I have sorted this table in
order of descending electrical conductivity and added the properties of 316L
stainless steel (SS) from another source18 to compare this metal, commonly
used in food processing applications, to the others materials listed in Table
1.
Steven Brandon 31 July 2004
9
Table 1: Physical properties of various materials19
Material
Melting
Pt.
Boiling
Pt. Density
Thermal
Conductivity
Electrical
Conductivity
- ºC ºC g/cm³ W/cm-K 106/ohm-cm
silver 962 2212 10.5 4.29 0.63
copper 1083 2567 8.96 4.01 0.596
gold 1064 2807 19.3 3.17 0.452
aluminum 660 2467 2.70 2.37 0.377
magnesium 649 1090 1.74 1.56 0.226
rhodium 1966 3727 12.4 1.5 0.211
iridium 2410 4130 22.5 1.47 0.197
tungsten 3410 5660 19.3 1.74 0.189
molybdenum 2617 4612 10.2 1.38 0.187
cobalt 1495 2870 8.90 1.00 0.172
zinc 420 907 7.14 1.16 0.166
nickel 1453 2732 8.90 0.907 0.143
ruthenium 2310 3900 12.20 1.17 0.137
iron 1535 2750 7.86 0.802 0.0993
platinum 1772 3827 21.4 0.716 0.0966
palladium 1552 3140 12.0 0.718 0.0950
tin 232 2270 7.30 0.666 0.0917
chromium 1857 2672 7.19 0.937 0.0774
tantalum 2996 5425 16.6 0.575 0.0761
niobium 2468 4742 8.55 0.537 0.0693
thorium 1750 4790 11.7 0.540 0.0653
rhenium 3180 5627 21.0 0.479 0.0542
vanadium 1890 3380 5.80 0.307 0.0489
lead 328 1740 11.4 0.353 0.0481
uranium 1132 3818 18.9 0.276 0.0380
hafnium 2227 4602 13.1 0.230 0.0312
zirconium 1852 4377 6.49 0.227 0.0236
titanium 1660 3287 4.50 0.219 0.0234
scandium 1539 2832 3.00 0.158 0.0177
yttrium 1523 3337 4.50 0.172 0.0166
316L SS 1397 7.95 0.134 0.0135
manganese 1244 1962 7.43 0.078 0.00695
graphite 3550 4827 2.62 1.29 0.00061
It is clear that while, 316L SS has the highly desirable property of resistance
to corrosion, which makes it a good choice for food applications, it is not a
particularly good electrical conductor. This means that it would exhibit large voltage loss between the power supply and the spark gap. This lost energy
would be dissipated as heat resulting in high temperature in the metal,
which could damage the structure of the electrode assembly.
It is interesting to note that graphite is reported to have relatively high
thermal conductivity and low electrical conductivity. Actually, the properties
of graphite are highly anisotropic due to its planar crystalline structure, i.e.,
its properties vary greatly depending on whether the property is measured
Steven Brandon 31 July 2004
10
parallel to, or normal to the orientation of the crystalline structure. A form
of carbon that is currently of great interest is the carbon nanotube, which has been reported to have very high electrical conductivity along the length
of the tube. However, there is considerable controversy about these
numbers and there is still much development to be done to be able to
assemble the nanotubes into useful, larger-scale structures. Section A9,
Carbon Nanotubes, provides background information.
It is probably wise to consider an electrode material with high conductivity
while retaining the high degree of corrosion resistance required for food
processing applications. The two materials with the highest conductivity listed in Table 1, silver and copper, are both subject to formation of oxides
on their surfaces in food processing environments. Such an oxide layer
would act as an insulator blinding off the surface of the electrode and
inhibiting spark formation. The fourth material in the table, aluminum is resistant to acids, but is quickly attacked by the alkaline cleaning solutions commonly used in food plants. The third material listed in the table, gold,
combines the excellent corrosion resistance with high electrical conductivity, making it an excellent choice for our particle-electrode material. The obvious shortcoming of gold is its high cost. However, depending of the
design of the spark producing apparatus, it is possible that little gold would be required, thereby holding costs to reasonable levels. For these reasons,
my choice for material for the particles could be embedded on the surface of the spark apparatus is gold. Figure 8, below, shows roughly 15-nm gold nanoparticles.
Figure 8: Gold Nanoparticles20
Question 1c What is the minimum electrical charge needed to generate a spark between
our nanoparticle gaps?
According to Wallash and Levit13, for gaps larger than 2 nm but less than
5000 nm, spark discharge is by field emission of electrons from metals. The
Steven Brandon 31 July 2004
11
field emission current, I, is described by the Fowler-Nordheim (F-N) equation
,
where E is the electric field, Φ is the work function, and a and b’ are
constants. If current is due to field emission, a plot of 1/E vs. ln(I/E²) -- or
1/V vs. ln(I/V²) -- should yield a straight line with a negative slope that is proportional to the work function of metal. This plot is known as the Fowler-
Nordheim (F-N) plot and can be used to determine whether the current flow
is due to field emitted electrons.
To obtain some realistic values for the work function, and the constants a
and b’, I will use data presented by Wallash and Levit for a 0.9-µm gap device in 1 atmosphere air. See Figure 9, below.
Figure 9: Fowler-Nordheim plot for 0.9-µm gap device near breakdown
The Fowler-Nordheim formula can be re-arranged as follows:
E
b
aeE
I
2
3
'
2 .
Taking the natural logarithm of both sides of this equation yields,
Steven Brandon 31 July 2004
12
E
ba
E
I 2
3
2
'lnln
which can, in turn, be rearranged to
aE
bE
Iln
1'ln 2
3
2
A least-squares linear curve fit of 1/E vs. ln(I/E²) data reported by Wallash and Levit in Figure 8, above was found to have a intercept of -24.832, the
constant a = 1.6429 x 10-11. Similarly, since the slope was found to be -0.9117, the term b’Φ3/2 = -0.9117 (µm-Vx10-3). Since the x-axis was plotted
as 1/E (µm-Vx10-3), the slope can be multiplied by 1000 to get b’Φ3/2 = -911.7 (µm-V). Armed with these numbers, I can return to the F-N formula
and solve for I, current.
E
b
aeEI
2
3
2
Substituting the values determined for a and for the term b’Φ3/2 yields,
E
Vm
eEI
7.911
112 106429.1
Since E = V/d and d = 0.9-µm (in this case), E=1.11V (µm-1). So, I =
(1.11V µm-1)2 x (1.6429 x 10-11) x exp[-911.7 (µm-V)/1.11V (µm-1)]. I
used Excel to produce the following table showing calculated current for
voltages from 120 V to 150 V.
Steven Brandon 31 July 2004
13
Table 2: Calculated Current
Gap Voltage E E2
a b’Φ3/2
(b’Φ3/2
/E) exp(-) I I
µm V V-µm (V-µm)2
µm-V A nA
0.9 120 133.3 17778 1.6429E-11 -911.7 -6.8378 0.0011 3.1324E-10 0.31
0.9 122 135.6 18375 1.6429E-11 -910.7 -6.7183 0.0012 3.6486E-10 0.36
0.9 124 137.8 18983 1.6429E-11 -909.7 -6.6027 0.0014 4.2311E-10 0.42
0.9 126 140.0 19600 1.6429E-11 -908.7 -6.4907 0.0015 4.8862E-10 0.49
0.9 128 142.2 20227 1.6429E-11 -907.7 -6.3823 0.0017 5.6202E-10 0.56
0.9 130 144.4 20864 1.6429E-11 -906.7 -6.2772 0.0019 6.4397E-10 0.64
0.9 132 146.7 21511 1.6429E-11 -905.7 -6.1752 0.0021 7.3518E-10 0.74
0.9 134 148.9 22168 1.6429E-11 -904.7 -6.0763 0.0023 8.3638E-10 0.84
0.9 136 151.1 22835 1.6429E-11 -903.7 -5.9804 0.0025 9.4831E-10 0.95
0.9 138 153.3 23511 1.6429E-11 -902.7 -5.8872 0.0028 1.0718E-09 1.07
0.9 140 155.6 24198 1.6429E-11 -901.7 -5.7966 0.0030 1.2076E-09 1.21
0.9 142 157.8 24894 1.6429E-11 -900.7 -5.7087 0.0033 1.3566E-09 1.36
0.9 144 160.0 25600 1.6429E-11 -899.7 -5.6231 0.0036 1.5197E-09 1.52
0.9 146 162.2 26316 1.6429E-11 -898.7 -5.5399 0.0039 1.6977E-09 1.70
0.9 148 164.4 27042 1.6429E-11 -897.7 -5.4590 0.0043 1.8916E-09 1.89
0.9 150 166.7 27778 1.6429E-11 -896.7 -5.3802 0.0046 2.1024E-09 2.10
Figure 10, below, is a plot of voltage vs. current from Table 2.
Current vs. Voltage
0.00
0.50
1.00
1.50
2.00
2.50
110 120 130 140 150 160
Voltage, V
Cu
rren
t, n
A
Figure 10: Voltage vs. Current plot based on my calculated current values
This compares favorably with the same data presented by Wallash and Levit (see Figure 11, below).
Figure 11: Voltage vs. Current plot reported by Wallash and Levit
Steven Brandon 31 July 2004
14
In their tests, they began with zero volts across the electrodes and then
gradually increased the voltage while monitoring current. They report that the discharge occurred in their test rig with a 0.9-µm gap at 151 V and
about 2.1 nA (nanoamperes) of electrical current.
The precise current required to produce sparks in the device under
consideration would depend on the design details of the device. However, it
is reasonable, based on the discussion above, to expect that the current requirement would be very low.
Question 2 Develop a detailed description of how you would develop and test a 3 x 3
inch surface that when charged with a suggested amount of electricity, would produce numerous sparks across the whole surface.
Development
My approach would be to use gold (for reasons described above) electrodes supported by a non-conducting substrate, such as soda glass. The basic
concept is illustrated in Figure 12, below. Thin layers of gold are arranged in parallel lines alternating between anodic and cathodic electrodes. Electrical
connections are made to each side of the array inducing a voltage difference between the parallel gold bands. These spaces serve as the spark gaps
permitting the formation of transient plasma discharges between conducting
gold bands at numerous locations over the surface of the device.
Figure 12. Conceptual diagram of spark producing device
In the actual device, the parallel bands of gold would be much smaller than
shown in the conceptual illustration in Figure 12, above. The actual device
would have thousands of microscopic, parallel bands of gold with narrow
gaps separating the oppositely charged, alternating bands. The dimension of
the spark gap should be selected to produce sparks of a scale sufficient to be
effective in killing bacterial cells, which typically range from about 0.2 to 2
µm (200 to 2000 nm). For this reason, I believe the optimal spark gap for
Steven Brandon 31 July 2004
15
this application would be in the range of 1 µm, rather than the nanometer
scales (1 to 100 nm).
Methods for forming metallic nanoparticle monolayers on glass are described
by Shipway, et al. One method, as shown in Figure 13, below, involves
adsorption of the nanoparticle onto a thin film of polymerized siloxane.
Figure 13. Construction of Au-nanoparticle monolayers on glass21
Another approach, also described by Shipway, et al., is photolithography.
This is illustrated in Figure 14, below.
Steven Brandon 31 July 2004
16
Figure 14: Photo-lithographically deposited gold monolayer structures22
Microphotograph C, in Figure 14, above, is particularly interesting, since it shows a structure that is similar to the one I am proposing for the sparking
device. Photolithography, a well established technology, is described in detail in Section A10, Photolithography.23
Testing
A very simple spark gap setup is presented in Section A8, A Simple Spark
Gap Test Rig. The test setup for the proposed spark device should be similar to that used by Wallash and Levit in their study of sub-micron spark gaps as
shown in Figure 15, below.
Figure 15. Experimental setup used by Wallash and Levit.24
Steven Brandon 31 July 2004
17
This test rig consists of a Keithley 2400 power source-measurement unit to
supply voltage and measure current to the spark gap device with sufficient sensitivity to measure nano-ampere currents. A Tektronix CT-6 current
probe and a LeCroy 9362 digital oscilloscope were used to measure the
current transient at the point of breakdown. (This may not be useful in a
device in which numerous sparks are planned.) This system was connected
to a computer equipped with LabView software to automate data acquisition.
The test procedure used by Wallash and Levit was to make electrical contact
with the terminals on each side of the spark gap device and then ramp the
voltage from zero volts, in small steps, while measuring the current, until breakdown occurred. The proposed setup would use a similar procedure,
however, rather than stopping with the first spark is observed, the voltage
would be set to a level that produces numerous small sparks.
The effectiveness of the device would be measured by exposing a challenge organism (preferably a non-pathogen, for safety reasons) at a know
concentration to the device and then comparing the reduction in cell counts before and after exposure to a prescribed spark treatment of set voltage and duration. This would be replicated several times at several selected voltage
levels and durations.
The bacterial cell concentration before and after exposure to the device could be measured by filtering a specified volume of the test gas (which
would generally, or always, be air) and then either microscopically examining the filter disk (direct count method) or by transfering the exposed
filter disk to a growth medium and allowing the viable cells to grow into easily observable colonies (plate count method). Such a method is
described in detail by Kelly-Wintenberg, et al.25
Steven Brandon 31 July 2004
18
Appendix
Section A1: Sparks and Lightning 26
While attempting to explain sparks and lightning to some friends, I realized that I
didn't have a good gut-level understanding of them myself. As usual, my lack of
understanding was an attractive irritant, like a pimple that one can't help picking at.
And so over many months I kept noticing concepts that could be applied to
explanations of sparks. Here's what I've come up with.
To get a good understanding of sparks, you need to encounter their behavior in
detail. One way to do this would be to magnify a small spark, but sparks happen so
quickly that interesting behavior can't be seen, so in addition to magnifying it, we'd
have to slow it down somehow. Here's a better idea: speed yourself up instead. Imagine that you've been exposed to Tholosian water from 'Old Trek.' This is the
substance which causes you to live many times faster than normal. (TV-show
science fiction trivia experts will recall the appearance of a similar drug on The Wild
Wild West as well!) And then, instead of magnifying a tiny spark, let's go outside
during a storm and look at the behavior of an already-large spark. Except for its size, the strange behavior of lightning is very similar to the behavior of tiny sparks.
So, we're standing outside in the time-frozen world of a raging thunderstorm
viewed from our 1000X perceptual acceleration. The trees and bushes around us
are thrashing frozenly in the stopped wind, and a few torn shingles flying from the
nearby roof hang in the air nearby. Higher up we see a tangled, branching network of dimly glowing wiggly purple lines which look something like a tree root. And like
a root, all the tips of the branches are lengthening. But this can't be lightning, it's
dim and purple, not bright blue-white.
One branch-tip is about a hundred feet up from where we're standing. We can see that the wiggly line isn't moving, it's only growing at its tip. It takes a tortuous,
kinky path as it lengthens, and occasionally a new branch starts growing from the
side of the main one at a spot where there is a particularly sharp bend. Then we
notice something else: everything on the ground is starting to glow. Bits of dim
purple fire are popping into existence on the tops of bushes, the edges of the roof of the nearby house, the tips of the rooftop TV antenna, on the ends of all the tree
branches, and even on the flying pieces of roof shingles. As the exploring finger of
dim purple lightning comes downward, the purple "fire" on all the objects becomes
more and more intense. If you hold your hand in front of you, the tips of all your
fingers spout dim purple fire as well.
Now the dim purple lightning from above is about thirty feet away, and the
downward growth of its tip seems to be speeding up. Then something really
disturbing happens. One of the purple flames coming from your fingers has
suddenly started growing upward as a narrow wiggly violet line! You pull your hand
down, but it's too late, the streamer of purple stays attached and grows upward fast, it's two feet long by now. You notice that this purple streamer from your hand
isn't the only one, there are now jagged purple lines growing upwards from many
places which formerly had the little "St. Elmo's Fire" flames. There's a ten foot
Steven Brandon 31 July 2004
19
streamer coming from the tree, another from the bush, and a couple from the roof of the house and the TV antenna. They appear to be moving towards the incoming
lightning strike. There are even several coming from the wind-blown shingles, but
some of these are extending downwards towards the ground while others grow
upwards. The one from you're hand isn't winning, it apparently had a late start,
and the streamers coming from the tree and the shingles are really shooting upwards now ahead of all the others. And the downwards-growing streamer from
the shingles has touched the ground and is spreading out into a small disk of purple
rootlets on the surface of the ground.
Finally the upward-growing streamer from the shingles approaches the lightning streamer coming from above. The two growing branch-tips race together, and just
before they meet they split into several separate branches which all connect. And
NOW it suddenly looks like lightning, because the entire streamer from the shingles
is glowing brighter and brighter. The little disk of purple filaments where it touches
the ground is now several feet across and looks like blazing blue-white tree roots.
The whole thing is far too bright to look at, and it's getting brighter still. And something is happening to you. Your fingers hurt, the muscles in your arm are
tensing by themselves, and you feel yourself blacking out. As you lose
consciousness, you note that the short, dead-end streamer from your hand is still
jutting upwards into the air, glowing bright blue, though nowhere near as brightly
as the streamer from the shingles.
What the heck was all that?! Lightning struck an object hanging in the air?! Well,
sort of, since the shingles somehow launched their own lightning. And how could
lightning be coming from objects on the ground, and from your hand? Why were
you knocked unconscious even though you didn't get struck directly by the main bolt? And isn't lightning supposed to travel at the speed of light? 1000 times
speedup is nothing compared to light speed, so why did we see the lightning as a
bunch of slowly-growing filaments?
There are some mental tricks you can use to understand some of what went on
above. Number one: realize that lightning is not made of electricity. "Lightning is electricity" is a false concept which stands in your way of understanding, and you
need to get rid of it before you can figure out what's going on. The long purple
filaments, which extended through the air, are not electricity; they are actually
made of air. They are nitrogen and oxygen which has been converted into plasma.
Plasma is vaguely like fire, but it is not necessarily hot. When air is converted to plasma, the electrons of the gas atoms are knocked off the atoms and become able
to flow along through the air. Plasma is a conductor, so it's not too wrong to think
of purple plasma filaments as being like wires made of conductive air.
Another mental trick: when you take a conductive object, a metal bar for example, and hold it in a strong electric field, flame-like "St. Elmo's Fire" sprouts from the
ends of the bar. The "fire" is nitrogen/oxygen plasma. And plasma itself IS a
conductive object. So, if an electric field is strong enough, and if a tiny bit of air is
somehow converted into plasma, it's as if your conductive rod has grown little
conductive pieces on its ends. And next, the "sharp" parts of the plasma globs will
themselves sprout extra bits of plasma. And so your metal rod has started "lengthening itself" via fingers of air-plasma. The air can "catch fire" with an
Steven Brandon 31 July 2004
20
outbreak of plasma which grows and grows, with more air turning to plasma as the rods of plasma grow more plasma on their tips.
The plasma takes a particular form: long thin twisty rods. This occurs because "St.
Elmo's Fire" always starts on the sharpest part of an object, and the sharpest part
of a rod is the end of the rod. And so a pre- existing rod of plasma will grow more plasma on its tips and lengthen itself. This self-forming plasma conductor is
vaguely like a motorized metal antenna on a car which extends upwards. But the
plasma-antenna can lengthen itself continuously as long as its tip is still in a strong
electrostatic field.
If the twisted plasma rod should make a sharp bend as it grows, the bend can
behave as a sharp point and more plasma fingers can take off from the bend. In
this way a lengthening plasma streamer develops branches as it goes. Growing
plasma doesn't just form twisted rods, it often forms trees, it forms entire
complicated systems of rootlets which advance and spread. Whether it forms trees
or straight unbent paths depends on the shape of the e-field in the space around it. A parallel e-field will allow tree-shapes to grow. A spreading, radial-shaped field
will tend to force one plasma finger to grow faster than all the others, resulting in a
needle-straight spark.
Since plasma is a conductor, what do you think would happen if a piece of air-plasma were to connect itself between two highly-charged objects having opposite
charge? ZAP! The opposite charges would be shorted out. An enormous electric
current would exist for a moment. This is what happens during a lightning strike,
or during the tiniest spark. Long filaments of air-plasma within the clouds extend
and explore downwards towards the ground and upwards into the charged raindrops. A system of fine plasma-rootlets develops which connects most of the
raindrops to the main conductive plasma tree structure. When the conductive
plasma touches the ground, it discharges both the charge on itself and the charge
on the huge number of electrically charged raindrops. The large momentary
electric current makes the dim purple plasma explode with light and sound.
So, what about lightning and the speed of light? Why can we see lightning "strike"
across the clouds, yet light itself moves so fast that we never see moving light
beams? Why can we sometimes see sparks jump from object to object? This is
because the growing motion of lightning and sparks is actually the growth of
plasma filaments. It is not a movement of light. Lightning can "strike" slowly or quickly depending on how fast the plasma filament tips are extending themselves.
In very large Tesla Coil systems, the giant sparks can lengthen VERY slowly, a
human can sometimes outrun them.
In the speed-up story at the top of this page, why were there plasma filaments appearing on the ground and growing upwards? And why did the wind-blown
shingles send plasma filaments both up AND down? This is hard to explain without
going into detail about electric fields and atoms. But here's a similar question:
suppose you squeeze a clod of dirt between your thumb and forefinger until it
cracks. Would you expect the crack to start at your thumb, or at your finger? Or
might it start from a small spot in the dirt and grow outwards in two directions at
Steven Brandon 31 July 2004
21
once? In truth, applying force to the dirtball can cause a crack to start ANYWHERE within the dirt.
Cracks tend to start at defects, and a similar thing is true with lightning and sparks.
An invisible field of electric force, if applied to air, can cause plasma filaments to
burst into existence anywhere in the volume of air where the field exists. When lightning is advancing towards the ground, there is a strong electric field all through
the air around the plasma branch and in the space above the surface of the earth.
This strong field can trigger new plasma filaments to grow anywhere. Of course, its
main effect is to make the main lightning filament lengthen and grow downwards.
But those blowing shingles represented a "defect" in the air, they distort the invisible electrostatic field in the air and strengthened it near the shingles, just as a
bubble in stressed glass can distort the mechanical forces and initiate a crack in the
glass. The electric field, present throughout the air, caused two plasma dendrites
to take off from the shingle and "strike" simultaneously upwards and downwards.
The defect in the air caused the air to "crack" electrically, the crack being made of
3D plasma filaments.
The same thing happens when aircraft fly between oppositely charged parts of a
thunderstorm: the plane acts as a triggering defect in the air, and plasma fingers
launch themselves from two spots on the airplane. Flying a plane near a
thunderstorm is like poking a highly-stressed windowpane with a nail: the cracks start where the nail touches. Yes, that's right, research has shown that aircraft
rarely are struck by lightning, instead the aircraft themselves do the striking, since
the plasma starts on the wingtips and zips outwards, striking the clouds.
Steven Brandon 31 July 2004
22
Section A2: Electric Sparks 27
Electric sparks are a transient form of gaseous conduction. This type of discharge
is difficult to define, and no universally accepted definition exists. It can perhaps
best be thought of as the transition between two more or less stable forms of
gaseous conduction. For example, the transitional breakdown which occurs in the transition from a glow to an arc discharge may be thought of as a spark.
Electric sparks play an important part in many physical effects. Usually these are
harmful and undesirable effects, ranging from the gradual destruction of contacts in
a conventional electrical switch to the large-scale havoc resulting from lightning discharges. Sometimes, however, the spark may be very useful. Examples are its
function in the ignition system of an automobile, its use as an intense short-
duration illumination source in high-speed photography, and its use as a source of
excitation in spectroscopy. In the second case the spark may actually perform the
function of the camera shutter, because its extinction renders the camera
insensitive.
Mechanisms
The spark is probably the must complicated of all forms of gaseous conduction. It
is exceedingly difficult to study, because it is a transient and because there are so
many variables in the system. Some of these variables are the components of the gaseous medium, the gas pressure, the chemical form of the electrodes, the
physical shape of the electrodes, the microscopic physical surface structure, the
surface temperature, the electrode separation, the functional dependence of
potential drop on time, and the presence or absence of external ionizing agents.
One or more of these conditions may change from one spark to the next. Because of the great complexity, it will be impossible to do more than touch on some of the
main features in this article.
The dependence of breakdown, or sparking, potential on pressure p and electrode
separation if is considered first. It was shown, experimentally by F. Paschen and
theoretically by J. S. Townsend, that the sparking potential is a function of the product pd and not of p or d separately (Fig. 1). Further, there is a value of pd for
which the sparking potential is a minimum. Thus, if it is desired to prevent
sparking between two electrodes, the region may be evacuated or raised to a high
pressure. The latter method is used in accelerators of the electrostatic generator
variety. Here the entire apparatus is placed in a pressurized tank.
Steven Brandon 31 July 2004
23
Qualitatively, one of the aspects of a spark is that the entire path between
electrodes is ionized. It is the photon emission from recombination and decay of
excited states which gives rise to the light from the spark. Further, if the spark
leads to a stable conduction state, the cathode must be capable of supplying the needed secondary electrons, and the conduction state produced must permit the
discharge of the inter-electrode capacitance at the very minimum.
In a consideration of the mechanism involved in the spark, the time required for the
breakdown of the gas in a gap is an important element. L. B. Loeb pointed out that
this time is often less than that required for an electron to traverse the gap completely. This implies that there must be some means of ionization present
other than electron impact and that the velocity of propagation of this ionizing
agent or mechanism must be much greater than the electron velocity. It seems
definitely established that this additional method must be photo-ionization. In the
intense electric field which is necessary for the spark, the initial electron will produce a heavy avalanche of cumulative ionization. Light resulting from the decay
processes will produce ionization throughout the gas and electrons at the surfaces
by the photoelectric effect (Fig. 2). The electrons resulting from this will in turn
produce further avalanches through the entire region, so that in a time of the order
of 10-8 sec the entire path becomes conducting. If the pressure is approximately atmospheric, the spark will be confined to a relatively narrow region, so that the
conducting path, while not straight, will be a well-defined line. If the external
circuit can supply the necessary current, the spark will result in an arc discharge.
At lower pressure, the path becomes more diffuse, and the discharge takes on
either a glow or arc characteristic.
Steven Brandon 31 July 2004
24
Figure 2 shows:
A, the electron multiplication of electrons by the cumulative ionization of a single
electron liberated from the cathode by a photon;
B, a secondary electron emitted from the cathode by a positively charged ion;
C, the development and structure of an avalanche, with positively charged ions
behind electrons at the tip;
D, the avalanche crossing the gap and spreading by diffusion; and F, an older avalanche when electrons have disappeared into the anode.
A positive space-charge boss appears on the cathode at F. Ion pairs, outside the
trail, indicate the appearance of photoelectric ion pairs in the gas produced by
photons from the avalanche. E shows a photoelectron from the surface of the cathode produced by the avalanche.
Theory
Mathematically, the theory of Townsend predicts that the current in a self-sustained
discharge of the glow variety will follow Eq. (1),
ax
ax
eI
eII
0
Eq. (1)
Steven Brandon 31 July 2004
25
where I is the current with a given plate separation x, I0 is the current when x approaches zero, and a and γ are constants associated with the Townsend
coefficients. This equation represents the case where the electrode separation is
varied while the ratio of electric field to pressure is held constant. The condition for
a spark is that the denominator approach zero, which may be stated as in Eq. (2).
axe Eq. (2)
Loeb indicated that this criterion must be handled carefully. Townsend's equation
really represents a steady-state situation, and it is here being used to explain a
transient effect. If the processes which are involved are examined more carefully,
it appears that there should be a dependence on I0 as well.
References: 1. J. Beyon, Conduction of Electricity through Gases, 1972.
2. J. D. Cobine, Gaseous Conductors, 1941. 3. L. B. Loeb, Statistical factors in spark discharge mechanisms. Rev. Mod Phys 20:151-
160, 1948. 4. D. Roller and D. H. C. Roller, The Development of the Concept of Electric Charge,
1954.
Steven Brandon 31 July 2004
26
Section A3: Paschen's Law 28
In 1889, F. Paschen published a paper (Wied. Ann., 37, 69) which set out
what has become known as Paschen's Law. The law essentially states that
the breakdown characteristics of a gap are a function, which is (generally
not linear) of the product of the gas pressure and the gap length, usually written as V = f(pd), where p is the pressure and d is the gap distance. In
actuality, the pressure should be replaced by the gas density.
For air, and gaps on the order of a millimeter, the breakdown is roughly a
linear function of the gap length: V = 30pd + 1.35 kV, where d is in
centimeters, and p is in atmospheres.
Much research has been done since then to provide a theoretical basis for the law and to develop a greater understanding of the mechanisms of breakdown. Some of this will be described in the rest of this section, but it
should be realized that there are many, many factors which have an effect on the breakdown of a gap, such as radiation, dust, surface irregularities. Excessive theoretical analysis might help understanding why a gap breaks
down, but won't necessarily provide a more accurate value for the
breakdown voltage in any given situation.
Paschen's Law reflects the Townsend breakdown mechanism in gases, that
is, a cascading of secondary electrons emitted by collisions in the gap. The
significant parameter is pd, the product of the gap distance and the pressure. Typically, the Townsend mechanism, and by extension Paschen's
law, apply at pd products less than 1000 torr cm, or gaps around a centimeter at one atmosphere. Furthermore, some modifications are
necessary for highly electronegative gases because they recombine the secondary electrons very quickly.
In general, an equation for breakdown is derived and suitable parameters
are chosen by fitting to empirical data. Here are three equations:
Breakdown voltage:
VBREAKDOWN = B x p x d / [C + ln(p x d)]
Breakdown field strength:
EBREAKDOWN = B x p / [C + ln(p x d)]
where:
C =ln[A / ln(1 + 1 / gamma)]
Steven Brandon 31 July 2004
27
where: gamma is the (poorly known) secondary ionization coefficient.
For air (data from Bazelyan, p.32):
A = 15 cm-1torr -1
B = 365 Vcm-1 torr-1
and gamma = 10-2
so C = 1.18
Minimum sparking potential for various gases (data from Naidu, p.27):
Gas VS min
(V)
pd at VS
min (torr cm)
Air 327 0.567
Ar 137 0.9
H2 273 1.15
He 156 4.0
CO2 420 0.51
N2 251 0.67
N2O 418 0.50
O2 450 0.7
SO2 457 0.33
H2S 414 0.6
Note that the sparking voltage is affected by the electrode material, with
cathodes of Barium and Magnesium having higher voltages than Aluminum,
for example.
Temperature dependence
Paschen's law, V = f(pd), should really be stated as V = f(Nd) where N is the
density of gas molecules, which is, of course, affected by the temperature as
well as the pressure of the gas (n/V = p/RT). An empirical formula for air,
considering it as an ideal gas, is:
VBREAKDOWN = 24.22 x + 6.08 SQRT(x),
where x = 293 pd / (760 T)
Steven Brandon 31 July 2004
28
p = pressure in torr (mm Hg),
d = distance in cm, T = Absolute Temperature in Kelvins
VBREAKDOWN in kV
Humidity dependence
In air, increasing humidity increases the breakdown voltage. The effect is
most noticeable in uniform field, and less important in non-uniform gaps
(such as sphere gaps where the gap is a large fraction of the sphere
diameter, or in rod or needle gaps).
Gamma - Townsend's secondary ionization coefficient
Gamma is the net number of secondary electrons produced per incident
positive ion, photon, excited or meta-stable particle. It is a function of gas
pressure and E/p. Electronegative gases (SF6, Freon, oxygen, CO2) re-attach the electrons very quickly, so they have low gammas.
For nitrogen, gamma ranges between 10-3 and 10-2 for E/p of 100 to 700 V
cm-1 torr-1. Insulating gases like SF6 or Freon have gammas of 10-4 or even less.
References:
Naidu, M.S. and Kamaraju, V., High Voltage Engineering, 2nd ed., McGraw Hill, 1995.
Bazelyan, E.M. and Raizer, Yu. P., Spark Discharge, CRC Press, Boca Raton, 1998.
Steven Brandon 31 July 2004
29
Section A4: About common sparks29
Questions about Science and Technology
Alan Wallwork asked: How powerful is a common static electric charge? And can a common charge be measured? (common; as in the type of charge you get when you get out of a car or when you drag your feet across the carpet.)
When you walk across a carpeted floor, friction between your shoes and the carpet causes electrons to be rubbed off the carpet onto your body: your body now has an excess of electrons and as a result has become electrically charged. A split second before you touch a conductor (anything made of metal, like a door knob, will do), the electrons jump across the gap between your fingers and the conductor.
This flow of electricity ionizes (forms charged particles) the air. If the voltage across the gap is great enough, an electric discharge is created. At normal pressure and temperature, that electric discharge is in the form of a spark. (Editor's Note: For a more technical description, refer to Miscellaneous Notes, below.)
Upon formation of the discharge the ionized air produces heat, causing the air to rapidly expand. That expansion of air produces a compression wave, heard as a crackling sound. The heat energy causes the air molecules to emit light and a flash is seen (if you want to see this dramatically illustrated, take off your sweater in a dark closet).
There are all kinds of sparks of course, from lightning to the type created by spark plugs in car engines. To answer the question of what voltage is seen when a "common" spark is formed, we need to look at Paschen's Law.
Paschen's Law states that the voltage at which a spark occurs (known as breakdown voltage) is dependent on the product of air pressure (p) and the separation between the electrodes (d). A graph of Paschen's Law as it applies to air (different gases create sparks differently) is seen below.
Steven Brandon 31 July 2004
30
For a common spark, then, we assume the distance between the two electrodes (your hand and the door knob in this example) is about 1 mm, with one atmosphere of pressure (760 torr, sorry for the ancient units). From the graph, therefore, the voltage of a common spark works out to be around 2000 V.
Miscellaneous Notes
For a more technical explanation of spark formation, refer to the graph of Gaseous Conduction, below, and the following explanation.
As the electrons begin to flow from one electrode to the other, the voltage between the electrodes rises from 0 to V1 and ionization of the air occurs, causing a small current to flow between the electrodes. This stage is known as primary ionization.
Steven Brandon 31 July 2004
31
Between V1 and V2 the current is such that the ions are conveyed to the electrodes as fast as they are produced. At this stage, as the voltage increases, the current remains the same and saturation is said to have been reached.
As the voltage increases beyond V2, the ions created by primary ionization have gained enough energy from the electric field to cause further ionization by collision. This increases the current further. More and more ions are accelerated in this manner, known as secondary ionization.
At V3, breakdown voltage occurs, the current increases indefinitely and the air is no longer able to allow the flow of electrons without electric discharge (or spark formation).
Once the discharge has formed, the current strength drops and remains constant.
If you've ever driven behind a gasoline truck you'll know that they drag a short chain on the ground behind them. This is for safety, because friction between the tires and the road causes the truck to become charged; the chain allows the charge to drain back onto the road. If this wasn't done, formation of a spark might ignite the gasoline vapors.
Lightning is a spark with a length of 150 m to 3 km, a width of 1 cm to 30 cm, a temperature of up to 30,000°C and a lifespan of 0.002 s to 1.6 s.
Townsend Criteria for Formation of a Spark: gamma ead = 1, where d is the gap length, a is the number of electrons produced by a single electron traveling 1 cm in direction of the electric field, and gamma is the secondary ionization coefficient (the number of secondary electrons produced per ionizing collision in the gas).
At a given pressure, the breakdown voltage is a direct linear function of the length of the gap between the electrodes.
When the pressure is low, the electric discharge is in the form of cathode rays. When it's very low, the discharge is in the form of x-rays.
Glossary and Related Terms
Arc Discharge: A luminous electrical gas discharge characterized by very high current. The intense ionization necessary to maintain the large current is provided mainly by the evaporation of the electrodes, which are raised to incandescence by the discharge (the enormous heat produced is utilized in arc-welding).
Breakdown: A sudden electric discharge.
Breakdown Voltage: The voltage at which a sudden electric discharge occurs.
Charge: A property of some elementary particles (e.g., protons and electrons) that causes them to exert forces on one another. Like charges repel and unlike charges attract each other. The charge of an object depends on the relation of the number of the object's protons and electrons. If the object has an excess of electrons, the object is said to have a negative charge. Similarly, if the object has an excess of protons, the object has a positive charge.
Conductor: A substance or body that offers small resistance to the passage of an electric current.
Corona Discharge: A type of electric discharge that results when partial breakdown of the surrounding gas takes place.
Steven Brandon 31 July 2004
32
Current: The rate of flow of electricity, measured in units of amperes. A current of 1 ampere is equivalent to the flow of about 1,000,000,000,000,000,000 electrons per second.
Dielectric: A substance that is capable of sustaining an electrical stress, i.e., an insulator.
Discharge: The removal or reduction of an electric charge of a body. It's also the passage of an electric current or charge through a dielectric, usually accompanied by luminous effects. There are many kinds of discharge, among them spark discharge, glow discharge, arc discharge and corona discharge.
Electric Field: The space surrounding an electric charge within which it is capable of exerting a perceptible force on another electric charge.
Electrode: A device for emitting electric charge carriers, e.g. a wire that leads current into or out of a dielectric.
Electron: An elementary particle carrying a negative charge. They are constituents of all atoms and as free electrons they are primarily responsible for electrical conduction in most substances.
Flashover Voltage: The voltage at which an electric discharge occurs between two electrodes that are separated by an insulator (depends on the wetness of the insulator surface).
Glow Discharge: An electric discharge through a gas, usually at a low pressure, in which the gas becomes luminous. Neon signs utilize glow discharge.
Insulator: A substance that provides very high resistance to the passage of an electric current.
Ionization: The process of forming electrically charged atoms.
Paschen's Law: The breakdown voltage for a discharge between electrodes in gases is a function of the product of pressure and distance.
Potential Difference: The difference between electric field strength at two points.
Spark: A visible discharge of electricity between two places. Preceded by ionization of the path, there is a rapid heating effect of the air through which the spark passes, which creates a sharp crackling noise.
Spark Discharge: Formation of a spark; this type of discharge appears around atmospheric pressure.
Sparking Potential: The lowest voltage at which a spark will be created.
Townsend Criteria: A mathematical formula that illustrates when a spark will be formed.
Volt: The unit of electric potential, defined as the potential difference between two points on a conductor carrying a current of one ampere when the power dissipated is one watt.
Voltage: The potential difference between two specified points in a circuit or device (measured in units of volts).
References and Further Reading
Definitions come from The Penguin Dictionary of Physics, edited by Valerie H. Pitt, published by Penguin Books, Ltd., 1983
Steven Brandon 31 July 2004
33
Section A5: Does an electrical arc spark need air to happen? 30 Date: Fri Jul 25 18:08:37 2003 Posted By: Aaron J. Redd, Post-doc/Fellow, Plasma Physics and Controlled Nuclear Fusion, University of Washington Area of science: Chemistry ID: 1050069563.Ch -------------------------------------------------------------------------------- Message: The short answers: yes, a spark can occur in a vacuum; and, no, the spark isn't really a "fire", but such a spark can start a fire. So it is still a safety hazard. The spark or electrical arc is not a fire, in the sense that it is actually a superheated gas, also known as a plasma. The plasma of the arc does not burn, in the sense that it is sustained by the electrical current traveling through the arc -- just as a wire will heat up when electrical current travels through it, the plasma stays hot because of the current in the arc. In theory, two metal surfaces with a vacuum gap between them can safely hold off an arbitrarily high voltage between them. In the real world, though, there are two important ways for the spark to happen: (1) No vacuum is perfect, so there will be some small amount of gas present. As shown by Paschen in the early 20th century, there is a maximum voltage that can be sustained between two (metal) electrodes, determined by the density of the trace gases in the area and (speaking loosely) the distance between the metal plates. If the voltage is above the so-called Paschen voltage, then there will be an arc between the metal surfaces. Also, the gas doesn't need to be air: helium, neon, hydrogen, carbon dioxide -- any of these will show the same Paschen breakdown. (2) Real surfaces (such as metal surfaces) aren't perfect either, and if the electric field at the surface is too high, then the material will sputter and/or vaporize, creating some gas which can then become an arc plasma. Once the arc is occurring, more material can be sputtered off from the surface, adding more gas to the plasma arc and pitting the metal surface. This is part of the reason why electrical fires are so hard to fight: the spark that causes the fire won't stop until the electrical current stops, so whoever is fighting the fire first has to shut off the current, and then try to put out the fire. As for this pitting on the electrode surfaces, it is quite noticeable -- electrodes tend to look very "beat up" after being exposed to arcs. If your light switches (or any other switches!) generate visible sparks when they are switched on or off, then they need to be replaced immediately. Like I said above, the spark isn't a fire itself, but it can start a fire.
Steven Brandon 31 July 2004
34
Section A6: The Paschen Curve 31
The Paschen curve is a plot of breakdown voltage for gases as a function of the
product of gas pressure and electrode separation. When pressure is given in mBar
(thousandths of an atmosphere) and electrode separation in mm the product is
mBar-mm. For pressure-distance values above about 25 mBar-mm, the breakdown voltage increases with increasing pressure and or separation. For those of us who
have played around with spark gaps, this is not surprising. Below 25 mBar-mm the
curve begins to flatten out and as the pressure-distance product drops below about
7 mBar-mm the breakdown voltage begins to rise very rapidly!
Why does this happen? Gasses become conductive when they are converted into plasma. Plasma, also referred to as the fourth state of matter, is a fluid composed
not of molecules but of positive ions (cations) and electrons. Plasmas are formed
when energy is put into a collection of molecules faster than energy is lost by
radiation when the cations and electrons recombine. Recombination of the cations
and electrons causes energy to be released as light, this is how fluorescent lights
and neon signs work.
In a simple atmospheric pressure spark gap, as the potential difference (voltage)
between the electrodes is increased (or as the distance between the electrodes is
decreased), electrons begin to be emitted by the cathode (negative electrode) and
travel to the anode (positive electrode). As they travel, some of the electrons will
collide with gas molecules, knocking electrons loose and forming cations and more
Steven Brandon 31 July 2004
35
free electrons. Near the electrodes, where the concentration of traveling electrons is highest, a faint glow caused by the recombination of ions and electrons will
become visible. This glow is called a corona discharge and removes energy from
the ionized gas at a high enough rate to prevent the formation of plasma. At
higher voltages (or smaller gaps) the energy put into the molecules by electron
collision exceeds the ability of the corona discharge to dissipate the energy and a plasma is formed. The electric field between the electrodes will then separate the
cations and electrons. The electrons will flow towards the anode, while the cations
will flow towards the cathode. When the cations impact the cathode, they
recombine with electrons from the surface of the cathode, completing the electric
circuit. In an atmospheric pressure spark gap, the flow of cations and electrons tends to be confined to a fairly narrow channel which is called a spark. At the point
on the cathode where the spark connects, a large amount of heat is generated by
the impacting cations, damaging the electrode surface (more current = more
damage). Electrons impacting the anode surface do not cause much damage, since
they are thousands of times lighter than the cations and thus have much less
kinetic energy.
Before we can go further, we need to understand gasses better, particularly how
the rate of collision between gas molecules changes as the pressure changes. The
rate at which molecules collide depends on the concentration of molecules (how
many there are in a given volume) and the average velocity of the molecules. The
temperature of a gas is actually a measure of the average kinetic energy of the molecules which make up the gas. The velocity of the molecules then depends on
temperature and the mass of the individual molecules. The pressure of a gas is the
amount of force the gas exerts against a unit of area. This is a combination of the
kinetic energy of the individual molecules and the total number of molecules in a
unit of volume. (This is the Kinetic Theory of Gasses in a nut shell.) Now, if we consider a sample of gas at constant temperature and volume, it is clear that a
reduction in pressure means that there are fewer molecules and fewer collisions.
Another way to think of this is that the distance a molecule can travel (remember
we have not changed the temperature, so the average velocity of the molecules has
not changed) without colliding with another molecule has increased. This distance
is called the "mean free path". In air at one atmosphere of pressure and a temperature of 0°C (standard temperature and pressure or STP) the mean free
path is 1 X 10-7 meters (or 1 X 10-4 millimeters, 0.1 micrometers, 100 nanometers).
If we go back to the atmospheric pressure spark gap for a moment, we now realize
that one electron, traveling a centimeter or so from the cathode to the anode, will
collide with about 100,000 gas molecules. Whether or not a particular molecule will be ionized depends on how much kinetic energy the electron has when it hits the
molecule. If the energy of the incoming electron is greater than the binding force
of the electrons in the molecule (the "ionization potential") then one or more
electrons can be knocked loose. Unlike the gas molecules, whose kinetic energy
depends on the gas temperature, the kinetic energy of the electrons is determined by the acceleration caused by the electric field between the electrodes and the
distance over which this accelerating force acts. The distance that the electron
"falls" in the electric field (and hence its kinetic energy) is limited by the mean free
path of the gas. Every time the electron collides with a gas molecule, it loses most
of its kinetic energy. If the distance through which the electron accelerates is very
Steven Brandon 31 July 2004
36
small, it will never have enough kinetic energy to ionize the gas molecules. This is why higher pressure requires a greater voltage before a plasma can form. Longer
gaps reduce the electric field strength and consequently reduce the kinetic energy
of the electrons which is why longer gaps require higher voltages to initiate
plasmas.
So what is happening on the "left side of the Paschen curve?” As the pressure drops, the mean free path of the gas increases and the kinetic energy of the
electrons will also increase, meaning that a collision with a gas molecule will be
more likely to result in ionization. Now, remember that the x-axis of the Paschen
curve is not pressure, but the product of pressure and distance. What has
happened is that the distance between the electrodes is now smaller than the mean free path of the gas. The electrons have plenty of kinetic energy, but they are no
longer colliding with any gas molecules, so no ionization occurs.
The author continues by describing pseudo-spark switches, which use a spark gap
in a rarified gas to function as a high-voltage switch. Presumably, since the
determining factor is the product of pressure and spark gap, it is also possible to
reach the “pseudo-spark” range, indicated on the accompanying Paschen curve plot, by reducing the spark gap below the 5 to 50 µm range while maintaining
atmospheric pressure. The shape of the Paschen curve, presented here, means
that employing spark gaps below about 5 µm will require much higher voltage to
produce a spark.
Steven Brandon 31 July 2004
37
Section A7: Static Electricity Sparks and Lightning32
by Ron Kurtus (revised 28 December 2002)
A spark is a stream of electrons jumping across an air gap, heating the air until it glows and
expands. Certain conditions can cause enough static electricity buildup to cause a spark or lightning. A spark often requires both a conductor and non-conductor. Lightning is an
extreme example of a spark.
Conditions for sparks
Sparks do not happen easily. They are violent occurrences that require special conditions.
They need both non-conductors and conductors to occur. The way this happens can get complex. These conditions include walking on a carpet on a very dry day or the rapid
movement of tiny water particles in a summer storm.
In the home
The Triboelectric Series shows that when certain materials are in contact, they can cause a
great increase in electrical charges on the surfaces of those materials. This is typically the case in for sparks that people personally experience.
In the clouds
Normally water inhibits static electricity, but in the case of thunderstorms, there is so much movement of air and water droplets within the clouds that charges collect on the surface of
the droplets. Enormous amount of charges can collect in the clouds, some positive ( + ) and
some negative ( - ).
Sparks require conductors
You know that static electricity collects on the surface of non-conductors. But you seldom—if
ever—see a spark fly from one non-conductor to another. The reason is that sparks need conductors, so that the electrons can freely move about and gather enough charges
together to be able to jump from one material to another.
If you took a charged piece of plastic and put it next to some metal, there would be no spark. The charges are held on the surface of the plastic, so that they won't jump the air
gap
Another good example of this concerns how you can get shocked with a spark. You are a conductor of electricity—although not as good as a piece of metal. The reason you conduct
electricity is because of the salt in your blood and your cells. Now if you notice, you usually see sparks when you start to touch something metal—like a doorknob—or another person or
animal.
So static electricity is formed and gathered on the surface of a non-conductor, but it must be then transferred to a conductor to cause a spark.
Steven Brandon 31 July 2004
38
Charges move in conductor
When a conductor—like a metal rod—is brought near a charged non-conductor, the free
electrons in the conductor will move to one end or the other of the rod, depending on whether the non-conductor surface is positive or negative.
Opposite charges in conductor move toward non-conductor
When the conductor is brought into contact with the non-conductor, the electrical charges
on the surface of the non-conductor are "sucked" into the conductor. In other words, if negative charges are on the surface of the non-conductor, these electrons will move into the
conductor. If positively charged atoms are on the surface, electrons from the metal or conductor will neutralize those atoms, resulting in an excess of positive charges in the
conductor.
Now, if another conductor is brought near the first conductor, the same thing will happen. Since electrons can move so freely in a conductor, many may collect near the surface and
actually jump across the air gap as a spark.
Anatomy of a spark
Air is a non-conductor of electricity and resists the movement of electrons through it. When the attraction or electrical pressure is great enough between objects with positive ( + ) and
negative ( - ) charges -- or even between a charged object and a neutral one -- some
electrons are able to overcome the resistance and jump the air gap. This electrical pressure is also called potential difference or voltage difference.
Heats up the air
Since air is a non-conductor of electricity, it does not readily let electrons pass through it.
But if the attraction is great enough, some electrons will leave their material and fly to the
other object. While they move through the air, may smash into and bounce off molecules or atoms that are in their way. This heats up the air. (See Heat for an explanation.)
Spark glows white-hot
Steven Brandon 31 July 2004
39
Lower resistance
Now, the hotter the air is, the less resistance it gives the electrons. So as the air gets
heated, more and more electrons start jumping over to the other side. This only heats the air even more, until it actually gets white-hot. That is the spark or bolt of lightning that you
see and feel.
Electrons stop jumping
Once enough electrons have made the jump, the attraction is reduced and the flow stops.
The spark quickly cools down and the air stops glowing. It is all over in a fraction of a
second. Since this happens for such a short time, so you may only feel a slight discomfort from the heat of the spark. But if the spark is a bolt of lightning, it can cause an enormous
amount of damage.
Lightning is a real big spark
Lightning works the same way as a little spark, except that it happens on a massive scale.
Some lightning bolts are several miles long. Compare that to the tiny 1/4 inch or 1
centimeter length of the spark that comes off your finger.
Lightning can be quite dramatic
Lightning is created when water drops are churning around in a thunder cloud. They gather
either positive or negative electrical charges, so that soon one cloud may be positive and another cloud may be negative. Or perhaps some object on the earth may have an excess
of opposite charge.
Has high electrical pressure
The electrical pressure builds up, the same way as it does for a spark. Since the distances
are so much greater between the clouds, the electrical pressure must be extremely high for lightning to start. But once it does and a lightning bolt jumps from one cloud to another, it
is a tremendous spectacle.
Most go from cloud to cloud
Most lightning bolts are from cloud to cloud, but sometimes there are no positive charged
clouds nearby, so the negative cloud, sends its electrons to the ground or any object that
may have a slight positive charge.
Steven Brandon 31 July 2004
40
Thunder
Air expands when it is heated and contract when it cools. Since the spark happens so fast,
the air expands and contracts very rapidly. When it contracts, the air slaps together, just like when you clap your hands or pop a balloon. The noise you hear from a spark is just a
snap, because it is so small.
On the other hand, the noise of thunder is a tremendous crash, because the size of lightning is so large. The snap of a spark and the crash of thunder are caused by the same effect. The
only difference is in the size of the spark.
In conclusion
Static electricity is caused when friction causes electrical charges to build up on a surface of
a non-conduction material. Its explanation comes from the Atomic Theory of Matter. Static electricity can cause sparks and other problems, but it also is useful in pollution control.
Steven Brandon 31 July 2004
41
Section A8: Zapping Airborne Salmonella and Dust 33 Agricultural Research Service scientists have found a way to reduce Salmonella and dust in poultry areas. The technology may sound commonplace, but for many in the poultry industry it's exciting news. The technology uses a negative electrostatic charge to remove dust from the air. Unlike most air cleaners, this device does not require air to move through it for cleaning to occur. Reducing the dust is important, because these particles often give hitchhiking germs a free ride into chicks' lungs and feathers. Bailey W. Mitchell, an agricultural engineer, developed the ionizer system, in cooperation with veterinarian Henry D. Stone. Both researchers work at ARS' Southeast Poultry Research Laboratory in Athens, Georgia. ARS applied for a patent on the technology in July 1998. The first industry trials began in June of 1998, and a commercial product is now available.
Agricultural engineer Bailey Mitchell demonstrates an electrostatic air cleaning system. The hatching cabinet used here is a small version of ones used commercially for hatching chicks.
Early trials in 1994 suggested the process would reduce dust and had the potential to reduce airborne transmission of Newcastle disease virus and other disease organisms such as Salmonella. "When Bailey first started this work, we tested it in a small chick hatcher," says Stone. "He modified it many times. When I saw the consistent reduction in dust particles and bacteria during hatch, I knew it had potential."
Steven Brandon 31 July 2004
42
Mitchell says credit is also due to veterinary medical officer Daniel J. King, physiologist R. Jeff Buhr, and microbiologists Peter S. Holt, Richard K. Gast, K.H. Seo, Mark E. Berrang, S. Stan Bailey, and Nelson A. Cox for their collaboration with this research.
Dust Spreads Disease
Keeping hatching cabinets free of pathogens is especially important, because one infected hatching chick can very quickly spread disease organisms to an entire cabinet of 15,000 tiny birds. One reason: The strong air needed to move warmth throughout the cabinet also moves dust. Currently, chemical sprays are the only effective means of reducing airborne disease transmission in hatching cabinets, but they can be expensive and can damage hatching equipment. This electrostatic system would be safer for poultry and other livestock. It would also keep dust levels down better than existing methods and would continually clean the air of pathogens. The Simco Company of Hatfield, Pennsylvania, is one of the world's largest manufacturers of electrostatic equipment. Mitchell says the company provided electrostatic insights, equipment, and instrumentation under a federal-industry cooperative research and development agreement. "It makes sense that reducing the fluff in the hatching cabinet would reduce bacterial contamination at pipping," says Hank Engster, vice president of technical service for Purdue Farms of Salisbury, Maryland. Pipping is when the chick breaks through its shell during hatching. "We are pursuing a test of the technology at one of our complexes on the Delmarva Peninsula," says Engster. Experiments conducted in a small chamber with agar plates exposed to a continuous Salmonella aerosol showed that high levels of charge can, on average, reduce airborne Salmonella levels from over 1,000 per plate to near 0 in what appears to be an instantaneous sterilizing effect.
Steven Brandon 31 July 2004
43
The petri dishes below show sterilization effects of negative air ionization on a chamber
aerosolized with Salmonella enteritidis. The left sample is untreated; the right, treated.
The electrostatic technology consistently reduced Salmonella transmission between chicks by 98 percent and reduced Salmonella in air samples by 95 percent in a room with Salmonella-infected egg-laying hens. In other tests, Mitchell built up hatching cabinet dust levels to 40 times above normal. The device reduced airborne particles by 99 percent in 60 seconds. The system was tested on a hatching cabinet with a few infected fertile eggs interspersed among healthy ones. Salmonella counts in the guts of 7-day-old chicks in the cabinet with the device were reduced by a factor of 1,000- to 10,000-fold, when compared to counts in chicks in a hatching cabinet without the device.
Producers May Flock to Air Cleaners
"We are mainly interested in the technology for food safety--but also for improved growth and productivity in our flocks," says Purdue's Engster. "We sent a group down to Athens, Georgia, to assess how well the technology would meet our needs. Bailey showed us a system installed at Seaboard Farms." Seaboard Farms in Athens supplies poultry for many fast-food companies. The company, with four hatcheries, produces over 5 million chicks a week. Installing the ARS ionizer costs about $2,500 per hatching cabinet. Seaboard Farms hopes to install it in all of the cabinets in one hatchery. "We tested the technology at our hatcheries," says Steve Bolden, vice president of live production at Seaboard Farms. "We found it reduced bacteria in three out of five tests and consistently kept dust levels down. We have negotiated with ARS to license the technology."
Steven Brandon 31 July 2004
44
In addition, hatchability--the percentage of eggs that produce live chicks--increased as much as 2.7 percent in tests of the system, thanks to reduced pathogens, Mitchell says. "Multiply this seemingly modest increase by the millions of hatching eggs farmers sell and you can see the potential." The technology has also interested turkey producers. Wampler Foods of Harrisonburg, Virginia, the seventh largest U.S. broiler chicken producer and third largest turkey producer, invited Mitchell to demonstrate the technology. Wampler is interested and would like to install units when commercially available, according to Tom Knapp, manager of Wampler's turkey breeding operations. He says the company is also planning on model modifications to better fit their hatching cabinets. "Initial tests in poultry production look promising in terms of improved vitality and health of flocks," says Knapp. "If we can verify reduced levels of bacteria, we think the technology would be a vital component to our overall live production health programs." According to Mitchell, numerous simple ionizer systems have been developed and marketed for air-cleaning applications with little or no research. Although many of these devices had potential in small spaces with light dust loads, they require air to pass through them and are not able to handle the larger space and higher dust levels of a typical hatching cabinet. The supercharged ionizer/dust collection system developed by ARS appears able to do the job. The process is likely to have applications outside agriculture, Mitchell says. In tests, the researcher removed smoke from a 3,300-cubic-foot room with 95-percent efficiency. Many other companies, he adds, are asking to review the technology for environmental and other air-cleaning applications. This research is part of Animal Health, an ARS National Program (#103) described on the World Wide Web at http://www.nps.ars. usda.gov/programs/appvs.htm. Bailey W. Mitchell and Henry D. Stone are with the USDA-ARS Southeast Poultry Research Laboratory, 934 College Station Rd., Athens, GA 30605; phone (706) 546-3443 [Mitchell], (706) 546-3431 [Stone], fax (706) 546-3161, e-mail
Steven Brandon 31 July 2004
45
Section A8: A Simple Spark Gap Test Rig
This simple test rig was built by two undergraduate physics students at MIT,
Igor Sylvester and his friend Harry. Their used the following equipment
required to produce sparks: a transformer to convert 120-volt alternating
current (AC) power to direct current (DC); a rectifier to convert direct current to high frequency alternating current; and a capacitor to store
electrical charge until it is discharged by the spark gaps. Their spark gaps
consist of three pairs of nails positioned with the tip of one nail pointed
toward the other nail in each pair leaving a small gap between.
Spark Gap Test Rig Description:
It takes DC current and converts it to high frequency (~200 kHz) AC current. It acts
like a high frequency switch. It's made out of three pairs of nails connected in series. The separations between the nails provide a minimum voltage threshold. When this
threshold is reached (while charging the capacitor) the current "jumps" from one nail to its counterpart and air plasma is formed for a very short period of time. This plasma
lets current flow until the voltage is not high enough to support the plasma (while discharging the capacitor). As a consequence, the plasma is destroyed and the cycle
repeats again, waiting for the voltage to reach the threshold point.34
The spark gap. It takes DC current and transforms it into high frequency AC
current.
The spark gap, capacitor, rectifier and
transformer.
Steven Brandon 31 July 2004
46
Section A9: Carbon Nanotubes Carbon nanotubes, long, thin cylinders of carbon, were discovered in 1991 by
Sumio Iijima.35 These are large macromolecules that are unique for their size,
shape, and remarkable physical properties. They can be thought of as a sheet of
graphite (a hexagonal lattice of carbon) rolled into a cylinder. These intriguing structures have sparked much excitement in the recent years and a large amount
of research has been dedicated to their understanding. Currently, the physical
properties are still being discovered and disputed. What makes it so difficult is that
nanotubes have a very broad range of electronic, thermal, and structural properties
that change depending on the different kinds of nanotube (defined by its diameter,
length, and chirality, or twist). To make things more interesting, besides having a single cylindrical wall (SWNTs), nanotubes can have multiple walls (MWNTs)--
cylinders inside the other cylinders.36 The electrical transport properties of SWNTs
has been recently studied has raised some controversy. The conductance of a tube
is quantized, and a nanotube acts as a ballistic conductor. Nanotubes also have a
constant resistivity, and a tolerance for very high current density.37
An electronic device known as a diode can be formed by joining two nanoscale
carbon tubes with different electronic properties.
Carbon nanotubes are tubular carbon molecules with properties that make them potentially useful in extremely small scale electronic and mechanical applications.
They exhibit unusual strength and unique electrical properties, and are extremely
efficient conductors of heat.
A nanotube has a structure similar to a fullerene, but where a fullerene's carbon atoms form a sphere, a nanotube is cylindrical and each end is capped with half a
fullerene molecule. Their name derives from their size; nanotubes are on the order
of only a few nanometers wide (on the order of one ten-thousandth the width of a
human hair), and their length can be millions of times greater than their width.
Nanotubes are composed entirely of sp² bonds, similar to graphite. Stronger than the sp3 bonds found in diamond, this bonding structure provides them with their
unique strength. Nanotubes naturally align themselves into "ropes" held together
by van der Waals force. Under high pressure, nanotubes can merge together,
trading some sp2 bonds for sp3 bonds, giving great possibility for producing strong,
unlimited-length wires through high-pressure nanotube linking.
Steven Brandon 31 July 2004
47
There are two main categories of nanotubes: single-walled (SWNT) and multi-
walled (MWNT). Additionally, there are a large variety of forms of each of these,
identified by a two-digit sequence. The first digit indicates how many carbon atoms
around the tube is. The second digit determines the offset of where the nanotubes
wrap around to. If the second digit is a zero, the nanotubes are called "armchair". If both digits are the same, the nanotubes are called "zigzag". Otherwise, they are
called "chiral".
The structure of the nanotube, as described above, strongly affects its electrical
conducting properties. For example, (6,0), (6,6), (9,0), and (9,9) nanotubes are all excellent conductors. However, electron holes arise in (7,0), (8,0), (6,2), and (7,5)
nanotubes, making them semiconductors. In theory, nanotubes which conduct can
have an electrical current density more than 1,000 times stronger than metals such
as silver and copper. All nanotubes are expected to be very good thermal
conductors along the tube, but good insulators laterally to the tube.
While it has long been known that carbon fibers can be produced with a carbon arc,
and patents were issued for the process, it was not until 1991 that Sumio Iijima, a
researcher with the NEC Laboratory in Tsukuba, Japan, observed that these fibers
were hollow. This feature of nanotubes is of great interest to physicists because it
permits experiments in one-dimensional quantum physics. Techniques have been developed to produce nanotubes in sizeable quantities, but their cost still prohibits
any large scale use of them.
Fullerenes and carbon nanotubes are not necessarily products of high-tech
laboratories, and are also formed in such mundane places as candle flames. However, these naturally occurring varieties are highly irregular in size and quality,
and attempting to ensure the high degree of uniformity necessary to meet the
needs of research and industry is impossible in such an uncontrolled environment.
Nanotubes can be opened and filled with materials such as biological molecules,
raising the possibility of applications in biotechnology. They can be used to dissipate heat from tiny computer chips.
The strength and flexibility of carbon nanotubes makes them of potential use in
controlling other nanoscale structures, which suggests they will have an important
role in nanotechnology engineering. The highest tensile strength an individual SWNT has been tested to is 63 GPa.
In Earth's upper atmosphere, atomic oxygen erodes the carbon nanotubes, but
other applications rarely need protection of the carbon nanotube surface. Though it
is debatable if nanotube materials can ever be made with a tensile strength approaching that of individual tubes, composites may still yield incredible strengths
potentially sufficient to allow the building of such things as space elevators, artificial
muscles, ultrahigh-speed flywheels, and more. MIT is working on combat jackets
utilizing carbon nanotubes for ultra-strong fibers and for monitoring its wearer's
condition.
Steven Brandon 31 July 2004
48
Carbon nanotubes additionally can also be used to produce nano-wires of other chemicals, such as gold or zinc oxide. These nano-wires in turn can be used to cast
nanotubes of other chemicals, such as gallium nitride. These can have very
different properties from CNTs - for example, gallium nitride nanotubes are
hydrophilic, while CNTs are hydrophobic, giving them possible uses in organic
chemistry that CNTs could not be used for.
One use for nanotubes that has already been developed is as extremely fine
electron guns, which could be used as miniature cathode ray tubes in thin high-
brightness low-energy low-weight displays. This type of display would consist of a
group of many tiny CRTs, each providing the electrons to hit the phosphor of one pixel, instead of having one giant CRT whose electrons are aimed using electric and
magnetic fields. These displays are known as Field Emission Displays (FEDs) A
nanotube formed by joining nanotubes of two different diameters end to end can
act as a diode, suggesting the possibility of constructing electronic computer
circuits entirely out of nanotubes. Nanotubes have been shown to be
superconducting at low temperatures.
Current progress
One application for nanotubes that is currently being researched is high tensile
strength fibers. Two methods are currently being tested for the manufacture of
such fibers. A French team has developed a liquid spun system that involves pulling a fiber of nanotubes from a bath which yields a product that is approximately 60%
nanotubes. The other method, which is simpler but produces weaker fibers uses
traditional melt-drawn polymer fiber techniques with nanotubes mixed in the
polymer. After drawing, the fibers can have the polymer burned out of them to
make them purely nanotube or they can be left as they are.
Scientists working at the University of Texas at Dallas produced the current
toughest material known in mid-2003 by spinning fibers of single wall carbon
nanotubes with polyvinyl alcohol. Beating the previous contender, spider silk, by a
factor of four, the fibers require 600J/g to break. In comparison, the bullet-resistant fiber Kevlar is 27-33J/g.
In 2004 Alan Windle's group of scientists at the Cambridge-MIT Institute developed
a way to make carbon nanotube fiber continuously at the speed of several
centimeters per second just as nanotubes are produced. One thread of carbon nanotubes was more than 100 meters long. The resulting fibers are electrically
conductive and as strong as ordinary textile threads.
High purity (80%) nanotubes with metallic properties can be extracted with
electrophoretic techniques. In April of 2001, IBM announced it had developed a
technique for automatically developing pure semiconductor surfaces from nanotubes. On September 19, 2003, NEC Corporation, Japan, announced stable
fabrication technology of carbon nanotube transistors. In June 2004 scientists from
China's Tsinghua University and Louisiana State University demonstrated the use of
nanotubes in incandescent lamps, replacing a tungsten filament in a light bulb with
a carbon nanotube one.
Steven Brandon 31 July 2004
49
Nano-mechanical computer storage devices using nanotubes are currently in the prototype stages. Both high speed non-volatile memory which can be used to
replace nearly all solid state memory in computers today, and high density storage
that may replace hard drives, are being developed. Major limiting factors in the
development of nanotubes include their cost and difficulties in orienting the
nanotubes, which tend to tangle because of their length.
As of 2003, nanotubes cost upwards from 20 euro per gram to 1000 euro per gram,
depending on purity, composition (single-wall, double-wall, multi-wall) and other
characteristics.
Carbon nanotubes in electrical circuits
Carbon nanotubes have many properties--from their unique dimensions to an
unusual current conduction mechanism--that make them ideal components of
electrical circuits, and it is exciting to envision, or even to implement, novel
transistors, MEMS devices, interconnects, and other circuit elements.
The major hurdles that must be jumped for carbon nanotubes to find prominent
places in circuits relate to fabrication difficulties. The carbon nanotube production
processes are very different from the traditional IC fabrication process. The IC
fabrication process is somewhat like sculpture--films are deposited onto a wafer
and pattern-etched away. Carbon nanotubes are fundamentally different from films; they are like atomic-level spaghetti (and every bit as sticky).
Today, there is no reliable way to arrange carbon nanotubes into a circuit.
Researchers sometimes resort to manipulating nanotubes one-by-one with the tip
of an atomic force microscope in a painstaking, time-consuming process. Perhaps the best hope is that carbon nanotubes can be grown through a chemical vapor
deposition process from patterned catalyst material on a wafer. Though such a CVD
process has been shown to allow a circuit designer to locate one end of a nanotube,
there is no obvious way to control where the other end goes as the nanotube grows
out of the catalyst.
Even if nanotubes could be precisely positioned, there remains the problem that, to
this date, engineers have been unable to control the types of nanotubes--metallic,
semiconducting, single-walled, multi-walled--produced. This is a problem that
chemical engineers must solve if nanotubes are to find a place in commercial circuits.38
Steven Brandon 31 July 2004
50
Section A10: Photolithography 39
Photolithography is a process used in semiconductor device fabrication to transfer a pattern from a photomask (also called reticle) to the surface of a wafer or substrate. It bears a similarity to the conventional lithography used in printing. Lithography involves a combination of etching, chemical deposition, and chemical treatments in repeated steps on an initially flat substrate. A part of a typical silicon lithography procedure would begin by depositing a layer of conductive metal several nanometers thick on the substrate. A layer of photoresist -- a chemical that hardens when exposed to light -- is applied on top of the metal layer. The photoresist is selectively hardened by illuminating it in specific places. For this purpose a transparent plate with patterns printed on it, called a mask, is used together with an illumination source to shine light on specific parts of the photoresist. Then, the photoresist that was not exposed to light and the metal underneath is etched away with a chemical treatment. Finally, the hardened photoresist is etched using a different chemical treatment, and all that remains is a layer of metal in the same shape as the mask. Lithography is used because it affords exact control over the shape and size of the objects it creates, and because it can create patterns over an entire surface simultaneously. Its main disadvantages are that it requires a substrate to start with, it is not very effective at creating shapes that are not flat, and it can require extremely clean operating conditions. In a complex integrated circuit, a wafer will go through the photolithographic area on the order of 20 to 30 times.
Technology
A wafer is introduced onto an automated "wafertrack" system. This track consists of handling robots, bake/cool plates, and coat/develop units. The robots are used to transfer wafers from one module to another. The wafer is initially heated to a temperature sufficient to drive off any moisture that may be present on the wafer surface. Hexa-methyl-disilizane (HMDS) is applied in either liquid or vapor form in order to promote better adhesion of the photo-sensitive polymeric material, called photoresist. Photoresist is dispensed in a liquid form onto the wafer as it undergoes rotation. The speed and acceleration of this rotation are important parameters in determining the resulting thickness of the applied photoresist. The photoresist-coated wafer is then transferred to a hot plate, where a "soft bake" is applied to drive off excess solvent before the wafer is introduced into the exposure system. The desired pattern is then projected onto the wafer in either a machine called a stepper or scanner. The stepper/scanner functions similarly to a slide projector. Light from a mercury arc lamp or excimer laser is focused through a complex system of lenses onto a "mask" (also called a reticle), containing the desired image. The light passes through the mask and is then focused to produce the desired image on the wafer through a reduction lens system. The reduction of the system can vary depending on design, but is typically on the order of 4X-5X in magnitude. When the image is projected onto the wafer, the photoresist material undergoes some wavelength-specific radiation-sensitive chemical reactions, which cause the regions exposed to light to be either more or less acidic. If the exposed regions become more acidic, the material is called a positive photoresist, while if it becomes less susceptible it is a negative photoresist. The resist is then "developed" by exposing it to an alkaline solution that removes either the exposed (positive photoresist) or the unexposed (negative photoresist). This process takes place after the wafer is transferred from the exposure system back to the wafertrack.
Steven Brandon 31 July 2004
51
Developers originally often contained sodium hydroxide (NaOH). However, sodium is considered an extremely undesirable contaminant in MOSFET fabrication because it degrades the insulating properties of gate oxides. Metal-ion-free developers such as tetramethyl ammonium hydroxide (TMAH) are now used. A post-exposure bake is performed before developing, typically to help reduce standing wave phenomena caused by the destructive and constructive interference patterns of the incident light. The develop chemistry is delivered in a similar fashion to how the photoresist was applied. The resulting wafer is then "hardbaked" on a bake plate at high temperature in order to solidify the remaining photoresist, to better serve as a protecting layer in future ion implantation, wet chemical etching, or plasma etching. The ability to project a clear image of a very small feature onto the wafer is limited by the wavelength of the light that is used and the ability of the reduction lens system to capture enough diffraction orders off of the illuminated mask. Current state-of-the-art photolithography tools use Deep Ultraviolet (DUV) light with wavelengths of 248 and 193 nm, which allow minimum feature sizes on the order of 130 to 90 nm. There are indications that 193-nm lithography can be extended to feature sizes below 45-nm using liquid immersion techniques. Also termed immersion lithography, this enables the use of optics with numerical apertures exceeding 1.0. The liquid used is typically ultra-pure, deionised water, which provides for a refractive index above that of the usual air gap between the lens and the wafer surface. This is continually circulated to eliminate thermally-induced distortions. Dopants may be added to the water to give a higher refractive index. Also in development are tools that will use the 157-nm wavelength DUV in a manner similar to current exposure systems. In addition, Extreme Ultraviolet (EUV) radiation lithography systems are currently under development which will use 13-nm wavelengths, approaching the regime of x-rays. Currently, there is uncertainty as to the technology succession in lithography. 157-nm equipment, once targeted to succeed 193 nm at the 65 nm feature size node, has now been pushed back to at least the 45-nm node. This has been due to persistent technical problems with the 157-nm technology and economic considerations that provided strong incentives for the continued use of 193-nm technology. At the 45 nm node and beyond however, 157-nm technology may face competition from both 193-nm technology and EUV lithography. The image for the mask is originated from a computerized data file. This data file is converted to a series of polygons and written onto a square fused quartz substrate covered with a layer of chrome using a photolithographic process. A beam of electrons is used to expose the pattern defined in the data file and travels over the surface of the substrate in either a vector or raster scan manner. Where the photoresist on the mask is exposed, the chrome can be etched away, leaving a clear path for the light in the stepper/scanner systems to travel through. Work is in progress on an optical maskless lithography tool. This uses a digital micro-mirror array to directly manipulate reflected light without the need for an intervening mask. Throughput is inherently low, but the elimination of mask-related production costs - which are rising exponentially with every technology generation - means that such a system would be far more cost-effective for small-scale manufacturing applications.
Steven Brandon 31 July 2004
52
1 http://en.wikipedia.org/wiki/Spark_gap 2 http://www.nology.com/dealer_info/spark_hotwire.htm 3 http://www.plasmas.org/rot-plasmas.htm 4 Jason Edson and Hannah Cohen,
http://fusedweb.pppl.gov/CPEP/Chart_Pages/5.Plasma4StateMatter.html 5 Jason Edson and Hannah Cohen,
http://fusedweb.pppl.gov/CPEP/Chart_Pages/5.Plasma4StateMatter.html 6 Plasma interaction with microbes, M. Laroussi, D. A. Mendis and M. Rosenberg, New
Journal of Physics 5 (2003) 41.1-41.10, http://sys.lib.clemson.edu:2270/EJ/article/-
search=6721490.1/1367-2630/5/1/341/nj3141.html 7 Use of a one atmosphere uniform glow discharge plasma to kill a broad spectrum of
microorganisms, K. Kelly-Wintenberg, Amanda Hodge, T. C. Montie, Liliana Deleanu,
Daniel Sherman, J. Reece Roth, Peter Tsai and Larry Wadsworth, Journal of Vacuum Science & Technology A: Vacuum, Surfaces, and Films -- July 1999 -- Volume 17, Issue 4,
pp. 1539-1544, http://sys.lib.clemson.edu:2573/getabs/servlet/GetabsServlet?prog=normal&id=JVTAD60
00017000004001539000001&idtype=cvips&gifs=yes 8 Zapping Airborne Salmonella and Dust, Jill Lee, Agricultural Research, March 2000 v48 i3
p20, http://www.ars.usda.gov/is/AR/archive/mar00/salm0300.htm 9 Production of photons in the bactericidal effect of transient electric arcs in aqueous
systems, L. Edebo, Applied Microbiology, January 1969, pp. 48 – 53. 10 Pseudospark Switches, Robert E. LaPointe,
http://members.tm.net/lapointe/Pseudospark_Switches.html 11 Paschen’s Law, Jim Lux, 9 Feb 2004, http://home.earthlink.net/~jimlux/hv/paschen.htm 12 Pseudospark Switches, Robert E. LaPointe,
http://members.tm.net/lapointe/Pseudospark_Switches.html 13 Electrical breakdown and ESD phenomena for devices with nanometer-to-micron gaps, Al
Wallash and Larry Levit, www.wallash.com/spie.pdf 14 Electrical breakdown limits for MEMS, ECE234/434 Handout, Thomas B. Jones, 12/12/02,
www.ee.rochester.edu/courses/ECE234/MEMS_ESD.pdf 15 Questions about Science and Technology, Science Web,
http://www.scienceweb.org/questions/q5.html 16 Static Electricity Sparks and Lightning, Ron Kurtus, 28 December 2002,
http://www.school-for-champions.com/science/staticspark.htm 17 http://en.wikipedia.org/wiki/Conductor_%28material%29 18 http://www.matweb.com/search/SpecificMaterial.asp?bassnum=NCAR50 19 Pseudospark Switches, Robert E. LaPointe, http://members.tm.net/lapointe/Metals.htm 20 Nanoparticle arrays on surfaces for electronic, optical and sensor applications, Andrew N.
Shipway, Eugenii Katz, Itamar Willnar, ChemPhysChem 2000, 1, p. 21, http://www.wiley-
vch.de/contents/jc_2267/2000/18_a.pdf 21 Nanoparticle arrays on surfaces for electronic, optical and sensor applications, Andrew N.
Shipway, Eugenii Katz, Itamar Willnar, ChemPhysChem 2000, 1, p. 22, http://www.wiley-vch.de/contents/jc_2267/2000/18_a.pdf
22 Nanoparticle arrays on surfaces for electronic, optical and sensor applications, Andrew N. Shipway, Eugenii Katz, Itamar Willnar, ChemPhysChem 2000, 1, p. 27, http://www.wiley-
vch.de/contents/jc_2267/2000/18_a.pdf 23 http://en.wikipedia.org/wiki/Photolithography 24 Electrical breakdown and ESD phenomena for devices with nanometer-to-micron gaps, Al
Wallash and Larry Levit, www.wallash.com/spie.pdf 25 Use of a one atmosphere uniform glow discharge plasma to kill a broad spectrum of
microorganisms, K. Kelly-Wintenberg, Amanda Hodge, T. C. Montie, Liliana Deleanu, Daniel Sherman, J. Reece Roth, Peter Tsai and Larry Wadsworth, Journal of Vacuum
Science & Technology A: Vacuum, Surfaces, and Films -- July 1999 -- Volume 17, Issue 4, pp. 1539-1544,
Steven Brandon 31 July 2004
53
http://sys.lib.clemson.edu:2573/getabs/servlet/GetabsServlet?prog=normal&id=JVTAD6000017000004001539000001&idtype=cvips&gifs=yes
26 Sparks and Lightning, William Beaty, 1996,
http://www.eskimo.com/~billb/tesla/spark.html 27 McGraw-Hill Encyclopedia of Science and Technology, 5th ed., Electric Spark, Glenn H.
Miller, McGraw-Hill Book Company, New York, NY, 1982, Vol.4, pp. 640-641. 28 Paschen’s Law, Jim Lux, 9 Feb 2004, http://home.earthlink.net/~jimlux/hv/paschen.htm 29 Questions about Science and Technology, Science Web,
http://www.scienceweb.org/questions/q5.html 30 Does an electrical arc spark need air to happen?, Aaron J. Redd, Plasma Physics and
Controlled Nuclear Fusion, University of Washington, July 25 2003,
http://www.madsci.org/posts/archives/Sep2003/1060117419.Ch.r.html 31 Pseudospark Switches, Robert E. LaPointe,
http://members.tm.net/lapointe/Pseudospark_Switches.html 32 Static Electricity Sparks and Lightning, Ron Kurtus, 28 December 2002,
http://www.school-for-champions.com/science/staticspark.htm 33 Zapping Airborne Salmonella and Dust, Jill Lee, Agricultural Research, March 2000 v48 i3
p20, http://www.ars.usda.gov/is/AR/archive/mar00/salm0300.htm 34 Tesla Coils, Igor Sylvester, January 02, 2004, http://web.mit.edu/ias/www/physics/tesla/ 35 Helical microtubules of graphitic carbon, Sumio Iijima, Nature, Nov 7, 1991 v354 n6348,
pp. 56-8. 36 Physical Properties of Carbon Nanotubes, Thomas A. Adams II,
http://www.pa.msu.edu/cmp/csc/ntproperties/main.html#intro 37 Physical Properties of Carbon Nanotubes, Thomas A. Adams II,
http://www.pa.msu.edu/cmp/csc/ntproperties/electricaltransport.html 38 http://en.wikipedia.org/wiki/Nanotubes 39 http://en.wikipedia.org/wiki/Photolithography