Dusty Plasmas

Having a dusty plasma in the lab does not necessarily mean that your vacuum pump filter needs changing. On the contrary, producing a dusty plasma often requires a pristine system to which the dust particles of interest are added. Let's first take a brief look at an example of a dusty plasma in interplanetary space.

Saturn's rings

Still considered a planet, Saturn boasts a ring system spanning 175,000 miles or so with a thickness measured in feet. Both Voyager I and II spacecraft detected radially aligned "spokes" that floated above the rings yet doggedly hovered over and under them as the spokes rotated in synchrony with the planet's magnetic field. Spokes are probably microscopic dust particles and, considering their environment, are examples of dusty plasmas. Saturn's rings, especially its E ring, as well as those rings found around some of the other planets in our solar systems and elsewhere, possess many of the characteristics of dusty plasmas.

In an online article in Spaceflight Now (see References), Professor Mihaly Horanyi (a dust plasma physicist at the Laboratory for Atmospheric and Space Physics) states that the Hubble Space Telescope has occasionally spotted the spokes. He associates the spokes with the direction of the plane of Saturn's rings relative to the sun. Horanyi adds that the plasma environment of the rings enables electrostatic charges picked up by micron-size dust particles composed of ice and rock debris to levitate the particles some 50 miles or so above the rings.

At right, this photo supplied by NASA/VOYAGER shows several spokes levitated over the planet's rings.

NASA/VOYAGER PHOTO

During its final approach to Saturn in 2004, Cassini failed to detect the elusive spokes. Depending on the direction of lighting relative to the plane of the rings, the spokes can appear dark or light. Considerable work is being done to try to predict their future appearances as well as to determine their exact mode of creation.

What is dust?

Let's imagine that you open the dust bag from your vacuum cleaner and examine its contents with a microscope. You would probably find insect parts, carpet fibers, cellulose, spores, fragments of salt crystals, micrometeorites, carbon granules, finely divided slag from industrial plants, Saharan desert sand, skin cells, soil, volcanic lava flakes, pollen, starch from last week's pancakes, and just plain dirt. Among the larger, visible fragments in the bag will be found pulverized powder with particle sizes measured in microns and nanometers. That's dust. But surely that's not the kind of dust that comprises dusty plasmas? Part and parcel of electrical discharges on our planet (e.g., lightning, ball lightning, piezoelectric effects from earthquakes, aurorae) are plasmas. And our atmosphere is just loaded with all kinds of dust. Little surprise then that dusty plasmas abound. Arguably, even humble candle flames are dusty plasmas comprised of carbon soot.

Many particulates, including smoke and metallic aerosols, can associate with plasmas. These selected particulates are what dusty plasma researchers work with in the lab. When the foreign dust particles pick up electrostatic charges from colliding plasma ions and electrons, they are rendered susceptible to electric and magnetic fields (in addition to gravity). The dust particles take on the behavior of the immersing plasma. In other words, they go with the flow.

Spectral lamps as dust generators

The study of dusty (also called complex) plasmas brings together researchers from diverse fields, such as atmospheric science, astrophysics, material science, semiconductor manufacturing, plasma physics, and the biological sciences. As expected, the required equipment can be both extensive and expensive. I wanted to examine some of the behavior associated with dusty plasmas. Vacuum sputtering can produce a dispersal of sub-micron dust. With that in mind, I looked for convenient and low-cost sources of dusty plasmas.

A while back, I purchased some spectral tubes for my work in spectroscopy. The glass or quartz tubes used in atomic absorption spectroscopy (AAS) are often available online at low cost. The condition of the tubes that I wound up with ranged from being used a lot to new-in-the-box, but all were many years old. It was pot luck. The tubes or lamps usually require several hundred volts DC to operate. Because of their age, some of the tubes functioned poorly or not at all. Since AAS tubes work by sputtering the element of interest with gas—usually neon—ions, I decided to try operating them using a neon power supply to excite the "dead" tubes. Neon power supplies generate high voltage AC at frequencies of 30 kHz and above. Their stated, nominal output voltages will vary considerably based on tube load characteristics, such as capacitance. Some of my tubes required 5kV supplies, while others worked well with a small, 1 kV neon-power supply.

WARNING: DO NOT ATTEMPT TO USE A NEON TUBE TRANSFORMER FOR THIS APPLICATION. IT WON'T WORK AND IT IS INCREDIBLY DANGEROUS! NEON POWER SUPPLIES ALSO CAN DELIVER A DEADLY SHOCK. DO NOT USE UNLESS YOU ARE VERY FAMILIAR WITH HANDLING HIGH VOLTAGE EQUIPMENT!

The first tube I selected to try was a mercury lamp. This choice proved fortuitous as none of the other tubes (for elements such as sodium, calcium, and magnesium) have generated a dusty plasma—at least not yet. All the mercury lamps I've experimented with do work, though the dusty plasmas behave differently depending on the geometry of the tube.

Here are several of the mercury lamps used in the dusty plasma experiments described below. Typically, old AAS tubes available at auction were manufactured by Varian and Perkin-Elmer. Mercury tubes are usually made from quartz or special UV glass, especially at the front port. Under power, and assuming the tubes haven't cracked with use, the tubes will glow with an orange light characteristic of neon gas.

In order to operate the tubes using neon power supplies, I had to rewire their power cable ends. Some tubes have bare pins that make it easy to attach leads using a matching vacuum tube socket. To efficiently produce a dusty plasma, it is important to connect one power lead to the tube's cathode pin. This pin can easily be distinguished form the tube's anode by examining the base and the internal connections to the electrodes.

CAUTION: Mercury tubes can produce considerable amounts of UV radiation in operation. This also applies to old tubes even if they are being powered by high-frequency AC as in these experiments. Wear protective glasses.

In order to see the dusty plasma, I set up a simple laser light sheet. This is a tried-and-true method for detecting hard-to-see particle motion. The laser beam from a powerful 532 nm DPSS laser passes through a cylindrical lens arranged to produce a tall, thin beam transversely dissecting the mercury tube. The light sheet can be moved longitudinally along the axis of the tube by sliding a flat reflecting mirror along its track on an optical bench. The photo below shows my setup. Mouse over the photo to display an overlay with component locations.

The CCD camera records the dusty plasma behavior. The green filter is a narrow-band, 532 nm rectangular optical element. Its purpose is to block most of the neon plasma background, enhancing the dusty plasma's visibility. The function generator is used to modulate the DPSS laser using its TTL output. For some plasma particle motions, the modulated laser acts as a stroboscope to permit closer scrutiny of particle motion. Most of the time strobing is not needed. For video studies, it suffices to use both the normal video frame speed and a selected shutter speed to slow down rapid motions. The TTL-driven laser stroboscope may help in selecting a good shutter speed.

Energized by connecting to a 1 kV neon power supply, the mercury lamp at right emits a telltale neon glow, especially at the cathode and anode. For this picture, the laser light sheet was reduced in intensity in order to allow photographing the normal plasma. At left in the photo is a wooden support that acts as an insulator to prevent distorting the dusty plasma (not seen here). The support also isolates the lamp from cool surfaces that may crack it. To its right is the intensely glowing cathode. Beyond the green light ring is an externally mounted copper strip that serves as the anode. At the extreme right is the quartz viewport.

Not all neon power supplies will operate these tubes under the conditions shown. Finding the right unit that does not shut itself off right away may require some trial-and-error experiments. In operation, the tube is being excited by low-frequency RF (30kHz and above). The copper strip anode should be snuggly mounted to the tube's surface to prevent overheating at points of inadequate contact. Depending on the power input, the area around the anode may get very hot in operation and could cause the tube to fail or crack.

ALWAYS WEAR PROTECTIVE GOGGLES SUITABLE FOR INTENSE LASER LIGHT AND FLYING GLASS. BE EXTREMELY CAUTIOUS WITH THE DEADLY HIGH VOLTAGES PRESENT ON THE TUBE!

I refer to the tube's cathode and anode even though the device is operating off alternating current. In some power supplies, the AC waveform is similar to the discharge from a Tesla coil or an induction coil. In these instances, the AC is fairly unidirectional, and the meaning is unambiguous. If this is not the case for some power supplies you may use, one arbitrarily selected lead from the power supply should be connected to the AAS tube's cathode while the other lead is hooked up to the copper strip anode. If the results are not satisfactory, shut off the power and reverse the leads. Always allow the tube to cool down to room temperature before starting it up again. Don't run the tube for too long at one time as the dusty plasma rings will be unsatisfactory. Experiment with different tubes and different anode positions.

In practice, the lower the voltage that you use while still producing a bright, stable plasma, the better. But, use sufficient voltage to produce a satisfactory dusty plasma. Try to avoid neon plasma oscillations from excessive voltages. The oscillations destabilize the dusty plasma and probably harm the tube as well.

Videos of Hg tubes in action

I set up a video camera to record images of the mercury tubes as dusty plasmas were being generated. For most of the recordings, I used a variable zoom telephoto lens pointed down the tube's axis through the front port. The highest magnification shots were taken using a long-distance microscope attached to a video camera. (Double click the video at right to view it.)

After turning on the neon power supply, the tubes usually need a few minutes to produce a substantial dusty plasma at a distance from the cathode. The distance of the dusty plasma increases with longer operation, though the highly visible neon-gas plasma appears fairly constant in extent. (The warm-up period apparently permits extremely minute mercury droplets to form. This produces a floating cloud held in place by electromagnetic and electrostatic forces.) However, each tube should be observed as soon as power is applied. In one case, a substantial ring formed within a minute. That tube is the first one seen in the video. Had I waited longer, the remarkable ring collapse shown would not have been observed.

A close examination of the hovering droplets revealed extremely minute, highly reflective specks that resembled ultramicroscopic particles viewed under a darkfield microscope. While no hint of structure could be spotted under 35x magnification, I had the impression that neither 350x nor 3500x would have helped. It seemed that the higher the magnification I used, the smaller the mercury droplets appeared. It was like looking at tiny, floating, specularly lit, chromium-plated, stainless steel balls.

As far as the conditions being experienced by those levitated droplets, based on the applied voltage of 1 kV, the electron temperature in the neon plasma is on the order of tens of thousands of degrees (>>10000 K). Ion temperatures are considerably lower than that. And the temperatures experienced by the mercury dust particles should be even less. The boiling point of mercury is approximately 357°C. The temperature of the mercury droplets must be lower than their boiling point in order to exist for any length of time.

In addition to the behavior of the suspended particles whose speed and direction of motion were related to their proximity to the cathode, interesting things would happen when the power supply was turned off. The change in behavior happened quickly on shutoff and vanished almost as quickly, depending on the tube, its length of operation during the experiment, etc. This behavior is shown in the video.

The dusty plasma responded to grounded objects (e.g., my finger) placed nearby. The mercury droplet plasma approached the object and would move back to its usual position when the object was removed. This is also seen in the video.

In the video, only a few of many possible experiments using simple mercury-tube, dusty-plasma generators are shown. In effect, we have a conveniently produced collection of charged microparticles available for easy manipulation by electrostatic fields. One possible next step would be to determine the size of the particles using dynamic light scattering or some such technique. This is much easier said than done, however.

Next time, we will explore another area in plasma science.

Dusty Plasmas Having a dusty plasma in the lab does not necessarily mean that your vacuum pump filter needs changing. On the contrary, producing a dusty plasma often requires a pristine system to which the dust particles of interest are added. Let's first take a brief look at an example of a dusty plasma in interplanetary space.
Saturn's rings Still considered a planet, Saturn boasts a ring system spanning 175,000 miles or so with a thickness measured in feet. Both Voyager I and II spacecraft detected radially aligned "spokes" that floated above the rings yet doggedly hovered over and under them as the spokes rotated in synchrony with the planet's magnetic field. Spokes are probably microscopic dust particles and, considering their environment, are examples of dusty plasmas. Saturn's rings, especially its E ring, as well as those rings found around some of the other planets in our solar systems and elsewhere, possess many of the characteristics of dusty plasmas. In an online article in Spaceflight Now (see References), Professor Mihaly Horanyi (a dust plasma physicist at the Laboratory for Atmospheric and Space Physics) states that the Hubble Space Telescope has occasionally spotted the spokes. He associates the spokes with the direction of the plane of Saturn's rings relative to the sun. Horanyi adds that the plasma environment of the rings enables electrostatic charges picked up by micron-size dust particles composed of ice and rock debris to levitate the particles some 50 miles or so above the rings. At right, this photo supplied by NASA/VOYAGER shows several spokes levitated over the planet's rings.	NASA/VOYAGER PHOTO
During its final approach to Saturn in 2004, Cassini failed to detect the elusive spokes. Depending on the direction of lighting relative to the plane of the rings, the spokes can appear dark or light. Considerable work is being done to try to predict their future appearances as well as to determine their exact mode of creation.
What is dust? Let's imagine that you open the dust bag from your vacuum cleaner and examine its contents with a microscope. You would probably find insect parts, carpet fibers, cellulose, spores, fragments of salt crystals, micrometeorites, carbon granules, finely divided slag from industrial plants, Saharan desert sand, skin cells, soil, volcanic lava flakes, pollen, starch from last week's pancakes, and just plain dirt. Among the larger, visible fragments in the bag will be found pulverized powder with particle sizes measured in microns and nanometers. That's dust. But surely that's not the kind of dust that comprises dusty plasmas? Part and parcel of electrical discharges on our planet (e.g., lightning, ball lightning, piezoelectric effects from earthquakes, aurorae) are plasmas. And our atmosphere is just loaded with all kinds of dust. Little surprise then that dusty plasmas abound. Arguably, even humble candle flames are dusty plasmas comprised of carbon soot. Many particulates, including smoke and metallic aerosols, can associate with plasmas. These selected particulates are what dusty plasma researchers work with in the lab. When the foreign dust particles pick up electrostatic charges from colliding plasma ions and electrons, they are rendered susceptible to electric and magnetic fields (in addition to gravity). The dust particles take on the behavior of the immersing plasma. In other words, they go with the flow.
Spectral lamps as dust generators The study of dusty (also called complex) plasmas brings together researchers from diverse fields, such as atmospheric science, astrophysics, material science, semiconductor manufacturing, plasma physics, and the biological sciences. As expected, the required equipment can be both extensive and expensive. I wanted to examine some of the behavior associated with dusty plasmas. Vacuum sputtering can produce a dispersal of sub-micron dust. With that in mind, I looked for convenient and low-cost sources of dusty plasmas. A while back, I purchased some spectral tubes for my work in spectroscopy. The glass or quartz tubes used in atomic absorption spectroscopy (AAS) are often available online at low cost. The condition of the tubes that I wound up with ranged from being used a lot to new-in-the-box, but all were many years old. It was pot luck. The tubes or lamps usually require several hundred volts DC to operate. Because of their age, some of the tubes functioned poorly or not at all. Since AAS tubes work by sputtering the element of interest with gas—usually neon—ions, I decided to try operating them using a neon power supply to excite the "dead" tubes. Neon power supplies generate high voltage AC at frequencies of 30 kHz and above. Their stated, nominal output voltages will vary considerably based on tube load characteristics, such as capacitance. Some of my tubes required 5kV supplies, while others worked well with a small, 1 kV neon-power supply. WARNING: DO NOT ATTEMPT TO USE A NEON TUBE TRANSFORMER FOR THIS APPLICATION. IT WON'T WORK AND IT IS INCREDIBLY DANGEROUS! NEON POWER SUPPLIES ALSO CAN DELIVER A DEADLY SHOCK. DO NOT USE UNLESS YOU ARE VERY FAMILIAR WITH HANDLING HIGH VOLTAGE EQUIPMENT! The first tube I selected to try was a mercury lamp. This choice proved fortuitous as none of the other tubes (for elements such as sodium, calcium, and magnesium) have generated a dusty plasma—at least not yet. All the mercury lamps I've experimented with do work, though the dusty plasmas behave differently depending on the geometry of the tube.
Here are several of the mercury lamps used in the dusty plasma experiments described below. Typically, old AAS tubes available at auction were manufactured by Varian and Perkin-Elmer. Mercury tubes are usually made from quartz or special UV glass, especially at the front port. Under power, and assuming the tubes haven't cracked with use, the tubes will glow with an orange light characteristic of neon gas. In order to operate the tubes using neon power supplies, I had to rewire their power cable ends. Some tubes have bare pins that make it easy to attach leads using a matching vacuum tube socket. To efficiently produce a dusty plasma, it is important to connect one power lead to the tube's cathode pin. This pin can easily be distinguished form the tube's anode by examining the base and the internal connections to the electrodes. CAUTION: Mercury tubes can produce considerable amounts of UV radiation in operation. This also applies to old tubes even if they are being powered by high-frequency AC as in these experiments. Wear protective glasses.
In order to see the dusty plasma, I set up a simple laser light sheet. This is a tried-and-true method for detecting hard-to-see particle motion. The laser beam from a powerful 532 nm DPSS laser passes through a cylindrical lens arranged to produce a tall, thin beam transversely dissecting the mercury tube. The light sheet can be moved longitudinally along the axis of the tube by sliding a flat reflecting mirror along its track on an optical bench. The photo below shows my setup. Mouse over the photo to display an overlay with component locations.

The CCD camera records the dusty plasma behavior. The green filter is a narrow-band, 532 nm rectangular optical element. Its purpose is to block most of the neon plasma background, enhancing the dusty plasma's visibility. The function generator is used to modulate the DPSS laser using its TTL output. For some plasma particle motions, the modulated laser acts as a stroboscope to permit closer scrutiny of particle motion. Most of the time strobing is not needed. For video studies, it suffices to use both the normal video frame speed and a selected shutter speed to slow down rapid motions. The TTL-driven laser stroboscope may help in selecting a good shutter speed.
Energized by connecting to a 1 kV neon power supply, the mercury lamp at right emits a telltale neon glow, especially at the cathode and anode. For this picture, the laser light sheet was reduced in intensity in order to allow photographing the normal plasma. At left in the photo is a wooden support that acts as an insulator to prevent distorting the dusty plasma (not seen here). The support also isolates the lamp from cool surfaces that may crack it. To its right is the intensely glowing cathode. Beyond the green light ring is an externally mounted copper strip that serves as the anode. At the extreme right is the quartz viewport. Not all neon power supplies will operate these tubes under the conditions shown. Finding the right unit that does not shut itself off right away may require some trial-and-error experiments. In operation, the tube is being excited by low-frequency RF (30kHz and above). The copper strip anode should be snuggly mounted to the tube's surface to prevent overheating at points of inadequate contact. Depending on the power input, the area around the anode may get very hot in operation and could cause the tube to fail or crack.
ALWAYS WEAR PROTECTIVE GOGGLES SUITABLE FOR INTENSE LASER LIGHT AND FLYING GLASS. BE EXTREMELY CAUTIOUS WITH THE DEADLY HIGH VOLTAGES PRESENT ON THE TUBE!
I refer to the tube's cathode and anode even though the device is operating off alternating current. In some power supplies, the AC waveform is similar to the discharge from a Tesla coil or an induction coil. In these instances, the AC is fairly unidirectional, and the meaning is unambiguous. If this is not the case for some power supplies you may use, one arbitrarily selected lead from the power supply should be connected to the AAS tube's cathode while the other lead is hooked up to the copper strip anode. If the results are not satisfactory, shut off the power and reverse the leads. Always allow the tube to cool down to room temperature before starting it up again. Don't run the tube for too long at one time as the dusty plasma rings will be unsatisfactory. Experiment with different tubes and different anode positions. In practice, the lower the voltage that you use while still producing a bright, stable plasma, the better. But, use sufficient voltage to produce a satisfactory dusty plasma. Try to avoid neon plasma oscillations from excessive voltages. The oscillations destabilize the dusty plasma and probably harm the tube as well.
Videos of Hg tubes in action I set up a video camera to record images of the mercury tubes as dusty plasmas were being generated. For most of the recordings, I used a variable zoom telephoto lens pointed down the tube's axis through the front port. The highest magnification shots were taken using a long-distance microscope attached to a video camera. (Double click the video at right to view it.) After turning on the neon power supply, the tubes usually need a few minutes to produce a substantial dusty plasma at a distance from the cathode. The distance of the dusty plasma increases with longer operation, though the highly visible neon-gas plasma appears fairly constant in extent. (The warm-up period apparently permits extremely minute mercury droplets to form. This produces a floating cloud held in place by electromagnetic and electrostatic forces.) However, each tube should be observed as soon as power is applied. In one case, a substantial ring formed within a minute. That tube is the first one seen in the video. Had I waited longer, the remarkable ring collapse shown would not have been observed.
A close examination of the hovering droplets revealed extremely minute, highly reflective specks that resembled ultramicroscopic particles viewed under a darkfield microscope. While no hint of structure could be spotted under 35x magnification, I had the impression that neither 350x nor 3500x would have helped. It seemed that the higher the magnification I used, the smaller the mercury droplets appeared. It was like looking at tiny, floating, specularly lit, chromium-plated, stainless steel balls. As far as the conditions being experienced by those levitated droplets, based on the applied voltage of 1 kV, the electron temperature in the neon plasma is on the order of tens of thousands of degrees (>>10000 K). Ion temperatures are considerably lower than that. And the temperatures experienced by the mercury dust particles should be even less. The boiling point of mercury is approximately 357°C. The temperature of the mercury droplets must be lower than their boiling point in order to exist for any length of time. In addition to the behavior of the suspended particles whose speed and direction of motion were related to their proximity to the cathode, interesting things would happen when the power supply was turned off. The change in behavior happened quickly on shutoff and vanished almost as quickly, depending on the tube, its length of operation during the experiment, etc. This behavior is shown in the video. The dusty plasma responded to grounded objects (e.g., my finger) placed nearby. The mercury droplet plasma approached the object and would move back to its usual position when the object was removed. This is also seen in the video. In the video, only a few of many possible experiments using simple mercury-tube, dusty-plasma generators are shown. In effect, we have a conveniently produced collection of charged microparticles available for easy manipulation by electrostatic fields. One possible next step would be to determine the size of the particles using dynamic light scattering or some such technique. This is much easier said than done, however.
Next time, we will explore another area in plasma science.

Views From Science ©2008-2009
7