PHYSICS OF PLASMAS

VOLUME 8, NUMBER 5

MAY 2001

A survey of dusty plasma physics*

P. K. Shukla
r Theoretische Physik IV, Fakulta t fu r Physik und Astronomie, Ruhr-Universita t Bochum, Institut fu University, S-90187 Umea , Sweden D-44780 Bochum, Germany and Department of Plasma Physics, Umea

Received 16 October 2000; accepted 7 November 2000 Two omnipresent ingredients of the Universe are plasmas and charged dust. The interplay between these two has opened up a new and fascinating research area, that of dusty plasmas, which are ubiquitous in different parts of our solar system, namely planetary rings, circumsolar dust rings, the interplanetary medium, cometary comae and tails, as well as in interstellar molecular clouds, etc. Dusty plasmas also occur in noctilucent clouds in the arctic troposphere and mesosphere, cloud-to-ground lightening in thunderstorms containing smoke-contaminated air over the United States, in the ame of a humble candle, as well as in microelectronic processing devices, in low-temperature laboratory discharges, and in tokamaks. Dusty plasma physics has appeared as one of the most rapidly growing elds of science, besides the eld of the BoseEinstein condensate, as demonstrated by the number of published papers in scientic journals and conference proceedings. In fact, it is a truly interdisciplinary science because it has many potential applications in astrophysics viz. in understanding the formation of dust clusters and structures, instabilities of interstellar molecular clouds and star formation, decoupling of magnetic elds from plasmas, etc. as well as in the planetary magnetospheres of our solar system viz. Saturn particularly, the physics of spokes and braids in the B and F rings, Jupiter, Uranus, Neptune, and Mars and in strongly coupled laboratory dusty plasmas. Since a dusty plasma system involves the charging and dynamics of massive charged dust grains, it can be characterized as a complex plasma system providing new physics insights. In this paper, the basic physics of dusty plasmas as well as numerous collective processes are discussed. The focus will be on theoretical and experimental observations of charging processes, waves and instabilities, associated forces, the dynamics of rotating and elongated dust grains, and some nonlinear structures such as dust ion-acoustic shocks, Mach cones, dust voids, vortices, etc. The latter are typical in astrophysical settings and in several laboratory experiments. It appears that collective processes in a complex dusty plasma would have excellent future perspectives in the twenty-rst century, because they have not only potential applications in interplanetary space environments, or in understanding the physics of our universe, but also in advancing our scientic knowledge in multidisciplinary areas of science. 2001 American Institute of Physics. DOI: 10.1063/1.1343087

I. INTRODUCTION

The interplay between plasmas and charged dust grains has opened up a new and fascinating research area, that of a dusty or complex plasma. A dusty or complex plasma is a normal electron-ion plasma with an additional charged component of small micron-sized particulates. This extra component, which increases the complexity of the system even further, is responsible for the name complex plasma. Dusty complex plasmas are ubiquitous in different parts of our cosmic environment,15 namely, in planetary rings, in circumsolar and the Phobos dust rings, in the interplanetary medium, in cometary comae and tails, and in interstellar molecular clouds. In fact, the dark bands of dust, which block parts of the Orion, Lagoon, Coalsack, Horsehead, and Eagle nebulae, indicate that dust must have been abundant in the nebulae that coalesced to form the Sun, planets, and other
*Paper HI1 1, Bull. Am. Phys. Soc. 45, 156 2000. Invited speaker.
1070-664X/2001/8(5)/1791/13/$18.00 1791

stars. On the other hand, during the Voyager 1 and 2 ybys of the outer planets and the ICE yby of comet Giaobini Zinner, it has been demonstrated that the plasma wave instrument can detect small dust particles striking the spacecraft.6 Complex dusty plasmas also occur in the ame of a humble candle, in the zodiacal light, in cloud-to-ground lightening in thunderstorms containing smoke-contaminated air over the United States, in volcanic eruptions, and in ball lightening. A recent investigation7 suggests that ball lightening is caused by oxidation of nanoparticle networks from normal lightening strikes on soil. Furthermore, meteoritic dust is thought to be present in the Earths mesosphere at altitudes of 8095 km. It has been conjectured that in the cold summer mesopause, ice particles can form around meteoritic dust particles, with the icy dust particles possibly inuencing the charge balance of the region.8,9 On the other hand, the presence of charged dust particles in the polar summer mesopause has been invoked to explain aspects of the very strong polar summer radar echoes referred to as polar mesosphere summer echoes PMSE,
2001 American Institute of Physics

Downloaded 04 Jul 2006 to 210.212.158.131. Redistribution subject to AIP license or copyright, see http://pop.aip.org/pop/copyright.jsp

1792

Phys. Plasmas, Vol. 8, No. 5, May 2001

P. K. Shukla

which occur at altitudes of 8093 km. Recently, the presence of charged dust in the mesosphere has been detected by direct rocket probe measurements, and both negatively and positively charged dust grains have been reported.10,11 The role of charged dust in mesospheric electric elds is recognized by Zadorozhny.12 The formation of an articial dusty plasma in the ionosphere was also revealed during the Spacelab 2 mission when the space shuttle orbital maneuver system engines were red.13 Since dust particles are a main element of interest in the solar system and in the interstellar medium, there are a number of future missions viz. the European Space AgencyROSETTA mission for detecting dust on comet 46P/ Wirtanen in 2012, the Cassini spacecraft mission arriving at Saturn in 2004 for exploring in detail the possible dust size, dust charge, dust dynamics, as well as other collective processes involving charged dust grains that will provide in detail the properties and global dynamics of charged dust grains. It is anticipated that future rocket campaigns in northern Scandinavia will provide more information regarding the mesospheric dust, while experiments on the International Space Station will determine the dusty plasma behavior under microgravity conditions. Lately, the physics of dusty complex plasmas has appeared as one of the most rapidly growing elds of science, besides the eld of BoseEinstein condensates, as demonstrated by the number of published papers in scientic journals and conference proceedings. It has a tremendous impact in astrophysics and low-temperature laboratory discharges including processing plasmas in the semiconductor industry. While in the latter one wants to clean up charged dusts which are anathema to microchips, charged dust grains are also deliberately created in low-temperature radio frequency and glow discharges to understand the basic physical processes associated with the presence of those grains. In laboratory discharges, one is able to study the growth of dust grains under gas densities and temperatures typical of the nebula from which the solar system was formed. The particulates look like tiny cauliowers14 pressed together in irregular stringsa growth pattern that offers clues to the rate at which dust particles in interstellar space turned into the clumps of matter, which are large enough to assemble into planets due to gravity. Irregular structures of charged dust particulates also appear in tokamaks.15 In fact, major boosts to dusty plasma research came after the discovery of the dust acoustic wave DAW,16 the dust ion-acoustic wave DIAW,17 the dusty plasma crystal DPC,1820 and the dust lattice wave DLW.21 We note that the idea of the dust acoustic wave DAW was put forward by the present author in the Capri Meeting on Dusty Plasmas in July 1989, while Ikezi22 theoretically predicted the Coulomb crystallization of charged dust grains in a strongly coupled dusty plasma system when the ratio between the Coulomb interaction and the dust thermal energies exceeds 170. A number of laboratory experiments have spectacularly veried the theoretical predictions of the DAWs,2329 the DIAWs,30,31 and the DLWs.3234 Dusty complex plasmas are fully or partially ionized low-temperature gases comprising neutral gas molecules,

electrons, ions, and extremely massive charged submicron and micron sized dust grains. The latter, which are a billion times heavier than the ions, acquire several thousands of electron charges. The dust grain charging occurs due to a variety of physical processes3543 including the collection of the background plasma electrons and ions by dust grains, the photoelectron emission, secondary electron emission, and thermionic emission, etc. Dust grains can be charged both negatively and positively. The grains act as a source of electrons when they are charged positively due to the irradiation of ultraviolet UV radiation. Both the positive and negative dust grains can coexist in laboratory and space plasmas. It appears that the dust grain charging is a new physical process in a dusty plasma, which marks a distinction between the latter and the usual multicomponent electron-ion plasma containing two ion species. In this paper, we present the basic physics of dusty plasmas and describe the progress that has been made in the area of collective processes in dusty plasmas during the last decade. Specically, we shall discuss the properties of a dusty plasma, illustrate various charging processes, and focus on waves, instabilities, and coherent nonlinear structures. The manuscript is organized in the following fashion. In Sec. II, we briey describe the general properties of a dusty plasma. In Sec. III, we discuss various charging processes. Section IV contains dusty plasma waves in an unmagnetized case. The excitation of dusty plasma waves are considered in Sec. V. Section VI points out the importance of collective effects with regard to the generation of a wakeeld in dusty plasmas. In Sec. VII, we discuss theories for dust ion-acoustic shocks and holes, and present experimental observations of those nonlinear structures. Section VIII presents the wave spectra in a magnetized dusty plasma. Finally, Sec. IX contains a summary of our investigation.
II. PROPERTIES OF DUSTY PLASMAS

The constituents of dusty plasmas are neutral gas molecules, electrons, ions, and massive compared to the ions charged dust grains. There are three characteristic length scales for such a combined dust and plasma mixture. These are the dust grain radius R, the dusty plasma Debye radius D , and an average intergrain distance d. The latter is related to the dust number density n d by n d d 3 1. The dusty plasma Debye radius D is given by44 1
2 D

1
2 De

1
2 Di

where De ( T e /4 n e 0 e 2 ) 1/2 Di ( T i /4 n i 0 e 2 ) 1/2 is the electron ion Debye radius, where T e ( T i ) is the electron ion temperature, n e 0 ( n i 0 ) is the unperturbed electron ion number density, and e is the magnitude of the electron charge. When T e T i and n e 0 n i 0 , we have De Di , while for T e T i and n i 0 n e 0 we have D Di De . In dusty plasmas, we typically have R D . One can treat the dust from a particle dynamics point of view when R D d , and in that case we have a plasma containing isolated screened dust grains, or a dust-ion plasma. On the other hand, collective effects of charged dust grains become im-

Downloaded 04 Jul 2006 to 210.212.158.131. Redistribution subject to AIP license or copyright, see http://pop.aip.org/pop/copyright.jsp

Phys. Plasmas, Vol. 8, No. 5, May 2001

A survey of dusty plasma physics

1793

portant when R d D . Here, charged dust particulates, which are essential ingredients of the total plasma mixture, can be treated as massive point particles similar to multiply charged negative or positive ions in a multispecies plasma. The dusty plasma quasineutrality condition for negatively charged dust grains is n i0 n e0 Z d0n d0 , 2

potential, and pi ( pe ) is the ion electron plasma frequency. The term f d 2 / 1 arises due to dust charge uctuations.4653 The dust grains in a dusty plasma could be either weakly or strongly correlated depending on the strength of the Coulomb coupling parameter Q2 exp , dT d 6

where n j 0 is the unperturbed number density of the particle species j ( j equals e for the electrons, i for singly charged ions, and d for the dust grains and Z d 0 is the number of unperturbed charges residing on the dust grain surface. When most of the electrons from the ambient plasma are attached to the dust grain surface, we may have Z d 0 n d 0 n e 0 . However, it should be noted that the depletion of the electrons cannot be complete e.g., Whipple et al.43, because the minimum value of the ratio between the electron and ion number densities turns out to be ( m e / m i ) 1/2 when the grain surface potential approaches zero, where m e ( m i ) is the electron ion mass. Here, the dusty plasma may be regarded approximately as a two component plasma composed of negatively charged dust grains and ions; the latter shield the dust grains. Such a situation is common in the Saturn rings as well as in low-temperature laboratory discharges. On the other hand, in thermal or UV irradiated dusty plasmas, the grains emit electrons and they are charged positively. The shielding of positive grains comes from the electrons, and at equilibrium we have45 n e 0 Z d 0 n d 0 , since the ion number density is completely depleted. For a spherical geometry, the solution of the linearized Poisson equation 2 d 1 2 d

d 0 ,

ckel or screened Coulomb or where d is the DebyeHu Yukawa potential

d r g r 0

r0 rr0 exp , r d

g ( r 0 ) is the oating potential of the dust particle at the location r 0 , and the effective dusty plasma Debye radius including the dust charge uctuation effects23,46 is
d D
1 f d 2 / 1 1/2

2 R . Furthermore, Here, we have dened f d 4 n d 0 D

1 R / 2 pi / Di pe / De exp e s / T e
is the dust charge relaxation frequency arising from the dust grain surface potential ( s ) changes, and

2 R / 2 pi / Di 1 e s / T i
pe / De exp e s / T e is a frequency associated with changes in the orbit limited motion OLM currents due to the presence of the oscillating

where Q Z d 0 e , Z d 0 is the number of unperturbed charges residing on the dust grain surface, T d is the dust temperature, and d / d . A dusty plasma can be considered as weakly coupled as long as 1. However, when 1 and 1, charged dust microspheres strongly interact with each other, and we have the possibility of forming Coulomb lattices in a strongly coupled dusty plasma. Strongly coupled dusty plasmas are created in low-temperature dusty plasma discharges for studying the formation and dynamics of dusty plasma crystals. They are also found in laser implosion experiments as well as in colloidal systems. There have been arguments that a dusty plasma is similar to a multi-ion plasma. However, this assertion has to be refuted, because a dusty plasma is signicantly different from a multi-ion plasma in that the presence of massive charged dust grains produce new collective phenomena on completely different time and length scales. An example is the DAW16 in which the dust mass provides the inertia, while the restoring force comes from the pressures of the inertialess electrons and ions. In laboratory dusty plasma discharges,2429 the DAW frequency is typically 1020 Hz, and video images of the DAW wavefronts are possible.24,28 Also, the dust charge uctuation dynamics4753 and dust dust interactions5457 give rise to new effects. The dust mass and shape distributions58 as well as their rotation59,60 and the plasma boundary61 also introduce new effects in dusty plasmas. Furthermore, there is a dust lattice wave21,6264 whose counterpart exists only in solids.65 Besides, dusty plasmas support a great variety of nonlinear structures including dust acoustic66 and dust ion-acoustic31,67,68 shock waves as well as Mach cones69 and vortical structures.7073 Finally, in a strongly coupled dusty plasma we have the possibility of new attractive forces viz. the wakeeld,74,75 the dipolar interaction59,7678 etc. as well as phase transition phenomena79,80 in dusty plasma crystals. In contrast to solid state crystals, the latter have many unusual properties. The ion charges in solid state crystals are of order unity, while in dusty plasma crystals charges on dust grains are huge hundreds of thousand of the electron charge and associated interaction energies typically 900 eV are many orders of magnitude larger than in solid state crystals. Also the lattice spacing in dusty plasma crystals is of the order of a mm, in contrast to the 0.1 nm scale in solid state crystals. Therefore, dusty plasma crystals have ordered plasma inhomogeneities due to the shielding. It turns out that knowledge of basic plasma physics, probe theory, statistical mechanics, as well as solid state physics and condensed matter physics is required for understanding numerous collective processes in dusty plasmas.

Downloaded 04 Jul 2006 to 210.212.158.131. Redistribution subject to AIP license or copyright, see http://pop.aip.org/pop/copyright.jsp

1794

Phys. Plasmas, Vol. 8, No. 5, May 2001

P. K. Shukla

III. CHARGING OF DUST GRAINS

Dust particles are charged due to a variety of processes including the bombardment of the dust grain surface by background plasma electrons and ions, photoelectron emission by UV radiation, ion sputtering, secondary electron production, etc. In low-temperature laboratory plasmas, dust particles are mainly negatively charged when any plasma electrons hitting the surface of the dust grains are attached to it and simply lost from the background plasma. Here, the charging process depends on the charging cross sections, which are determined by the impact parameter of the particle that approaches the grain to distances smaller than the particle size. Thus, the charging cross sections for electrons and ions are given by, respectively,

On the other hand, if the ion streaming velocity v 0 is much larger than the ion thermal velocity, then the approximate expression for the ion current is51 I i 0 R 2 en i 0 v 0 1

2eg m iv 2 0

14

For an arbitrary value of v 0 , we have to use a somewhat more complicated expression51 for I i 0 , namely, I i0 R 2e

8Ti mi
1/2

n i0 F 1 u 0 F 2 u 0

eq d 0 , RT i

15

e q d , v R 2 1
and

2 eq d Rm e v 2

i q d , v R 2 1

2 eq d Rm i v
2

where F 1 ( u 0 v 0 / 2 v ti ) ( /4u 0 )(1 2 u 2 0 )erf( u 0 ) ) and F ( u ) ( /2 u )erf( u ) are written in 0.5exp(u2 2 0 0 0 0 u terms of the error function erf( u 0 ) (2/ ) 0 0 exp(2)d. It is easy to show by Taylor expansion around v 0 0 that both the functions F 1 and F 2 approach unity as u 0 approaches zero. For the equilibrium, we can set I e 0 I i 0 0 and obtain, on using Eqs. 12 and 13, the expression
v te exp e g / T e

n i0 eg , v ti 1 n e0 Ti

16

for v 2 2 e q d / Rm e v 2 , whereas for v 2 v 2 we have * * e ( q d , v ) 0; here v v and q d is the dust charge. Clearly, the electrons must have a minimum speed v in order to * arrive at the grain surface. The charging equation is then given by
t vd q d I e I i I d q d ,

where I d q d

se,i

qs

v s q d , v f s v d 3 v

10

is the plasma current through the dust particle surface, vd is the dust velocity, q e e , q i e , and f s ( v) is the velocity distribution of the particle species s. If the dusty plasma is close to equilibrium, then the distribution function f s can be approximated by a Maxwellian distribution ( f s 0 ) with the drift velocity v0 between the plasma and dust. We have f s0 n s0

1
3/2 2 2 v ts

exp

1
2 2 v ts

v v0 2 ,

11

where n s 0 is the unperturbed number density and v ts ( T s / m s ) 1/2 is the thermal speed of the particle specie s. Assuming that the streaming velocities of electrons and ions are much smaller than their respective thermal velocities, we have the following expressions for the equilibrium electron and ion currents,35,51 respectively, I e 0 R 2 e and I i0 R 2e

8Te me
1/2

1/2

n e 0 exp

eq d 0 , RT e

12

8Ti mi

n i0 1

eq d 0 . RT i

13

which determines the surface potential g of isolated dust grains. The electrons are initially collected by a dust grain, due to their higher thermal velocity relative to the ions. Since the grain is electrically oating, it charges to a negatively surface potential, g 0, in order to repel further electron collection and enhance ion collection. A sphere in a thermalized hydrogen plasma oats to g 2.51T / e , where we have assumed that T e T i T and n i 0 n e 0 . The grain mean charge, q d , is related to its surface potential, g 0, by the grain capacitance, C, which for spherical isolated grains is simply R, and thus q d R g . This model for the grain charge applies to the case where the grains are sufciently far apart in comparison with the Debye length D of the dusty plasma. On the other hand, when the spacing between the grains is comparable to or less than D , the dust grains are closely packed. Here, the difference g 0 between the surface potential g and the plasma potential 0 has a smaller magnitude than in the case with d D , and consequently the average charge on a dust grain q d R is smaller than for isolated grains.1 For this case, we replace g in Eq. 16 by and use n i 0 / n e 0 (1 q d n d 0 / en e 0 ) to obtain the variation of against the ambient plasma number density for a xed dust number density n d 0 . Barkan et al.30 have described the results of a laboratory experiment on the charging of dust grains in a fully ionized, steady state, magnetized plasma column. By varying the ratio d / D , they experimentally demonstrated the predicted reduction42,43 in the grain charge for the case of closely packed grains ( d / D 1). Next, we discuss conditions under which the dust grains are charged positively. In an inert gas with dispersed dust grains, the latter are charged positively36,4850 either by UVinduced photoemission from the dust, or by thermionic emission from radiatively heated dust grains. Both of these mechanisms would require dust grains of low work function W. Examples of low work function material include various

Downloaded 04 Jul 2006 to 210.212.158.131. Redistribution subject to AIP license or copyright, see http://pop.aip.org/pop/copyright.jsp

Phys. Plasmas, Vol. 8, No. 5, May 2001

A survey of dusty plasma physics

1795

metals that typically have W 5 eV e.g. Ag, Cu, Al, Ca, and Cs carbides with W 2.18 3.50 eV, borides with W 2.54 2.92 eV, and oxides of metals with work functions ranging from W 1 eV to W 4 eV. We focus on the creation of a dust-electron plasma electrons and positively charged dust when the grains are charged positively due to the photoelectric emission of electrons in the presence of a ux of UV photons with energy W I , where is the Planck constant, is the frequency, and I is the ionization potential of an inert gas. For a unidirectional photon ux, the electron photoemission current for g p 0 is35,36 I ph e R 2 Q absY J exp e p / T pe , 17

attain a positive charge. The mean measured charge is positive with a value of 5.3( 1.6) 104 e for zinc grains, 5.0( 1.0) 104 e for copper grains, and 4.1( 1.0) 104 e for graphite grains.
IV. WAVES IN DUSTY PLASMAS

where Q abs is the efciency of absorption for the UV radiation ( Q abs 1 for 2 R / 1, where is the wavelength of the radiation, J is the UV photon ux, Y is the yield of the photoelectrons, and T pe is their average energy. The exponential factor in Eq. 17 takes into account the fact that the photoelectrons have sufcient energy to overcome the potential barrier of positively charged grains with the surface potential p . When the ratio between the electron mean free path in the gas to the dust radius is large, the orbit-limited current to a positive dust probe is35,36 I pe 8 1/2en e 0 v te R 2 1

ep . Te

18

By setting I ph I pe 0, we then obtain the equilibrium charge Z d Z d 1 bZ d

Q absY J exp bZ d T e / T pe , 8 n d 0 v te

19

where b e 2 / RT e and we have used Q Z d e C p R p and n e 0 Z d n d 0 R p n d 0 / e . Next we suppose that the dust grains are positively charged due to thermionic emission induced by laser heating i.e., infrared or visible or by thermal radiative heating. The RichardsonDushman form for the thermionic emission current is35 I th 4 R
2

The addition of charged particulates in an electron-ion plasma is found to modify or even dominate wave propagation, wave instability, wave scattering, etc. The modications occur owing to the inhomogeneity associated with the random distribution of charged particulates and the departure from the conventional quasineutrality condition in an electron-ion plasma due to the presence of charged dust grains. On the other hand, in multicomponent dusty plasmas there appear new wave modes that are associated with the dust grain dynamics. Here, we discuss the physics of the dust acoustic, dust ion acoustic, and dust lattice waves which have been experimentally observed. During the Capri Dusty Plasma Physics Meeting in July 1989, the author proposed consideration of the wave phenomena taking into account the dust grain dynamics and Boltzmann electron and ion distributions. In such a situation, we can have extremely low phase velocity v p / k in comparison with the electron and ion thermal velocities, lowfrequency ( tens of Hz, which is much smaller than the dust plasma frequency in low-temperature dusty plasma discharges DAWs16 in which the restoring force comes from the electron and ion pressures while the dust mass provides the inertia to maintain the DAWs. The wavelengths (2 / k ) of the latter are 0.6 cm, and they appear on a very long time scale so that their wavefronts can be seen with naked eyes.24 The spectra of dusty plasma waves are obtained by Fourier analyzing the Vlasov, Poisson, and Maxwell equations, supplemented by the dust charging equation. In a dusty plasma, the properties of the electrostatic waves are determined from

, k 0 ,
where is the wave potential,

22 23

A 0T 2 g

Wep exp , Tg

20

, k 1 e i d qe qi

where A 0 4 em e / 3 and T g is the grain temperature. By using I th I pe 0, we obtain36 y 1 y exp y 2 W b A 0T 3 , g exp e v te n d 0 Tg

21

is the dielectric constant, e , i , d are the electron, ion, and dust susceptibilities, respectively, and qs is the linear susceptibility associated with the dust charge uctuation dynamics. In a dusty plasma the electrons and ions are weakly coupled, and for / k v te , v ti , we have

where y e p / T e and T e / T g . The laser intensity required to heat the grains to the temperature T g can be estimated by balancing the heating rate due to a unidirectional photon ux and the blackbody radiative loss rate. Recent experiments on the Russian space station MIR have generated a dusty plasma with positively charged grains under solar radiation in the microgravity environment.39 Furthermore, Sickafoose et al.40 have conducted a laboratory experiment to demonstrate that photoelectric emission UV illumination causes an isolated spherical grain in vacuum to

e
and

1
2 k 2 De

24

1
2 k 2 Di

25

which correspond to Boltzmann distributed electrons and ions; i.e., the corresponding electron and ion number density perturbations are n e 0 e / T e and n i 0 e / T i , respectively.

Downloaded 04 Jul 2006 to 210.212.158.131. Redistribution subject to AIP license or copyright, see http://pop.aip.org/pop/copyright.jsp

1796

Phys. Plasmas, Vol. 8, No. 5, May 2001

P. K. Shukla

When the dust grains are weakly coupled, we obtain for / k v td , the dust susceptibility

2 pd
i d 3 k 2v 2 td

26

where i is the effective ion-dust/neutral collision frequency. Neglecting the dust charge uctuation effect, we obtain the frequency of the DIAWs17 for i

where v td ( T d / m d ) 1/2 is the dust thermal speed, d is the pd effective dust collision frequency, and 2 2 1/2 (4 Z d e n / m ) is the dust plasma frequency. d 0 d 0 For negatively charged dust grains, we have51,52

k De pi
2 1/2 1 k 2 De

32

2 which for k 2 De 1 reduces to

qe qi

f d 2
2 k 2 D 1 i

kC S ,

33

27

For d , k v td 1 , we obtain the dust acoustic wave frequency

k d pd
1/2 1 k 2 2 d

28

2 which for f d 2 / 1 1 and k 2 D 1 reduces to16

kC D .

29

Here, C D Z d ( n d 0 / n i 0 ) 1/2( T i / m d ) 1/2(1 n e 0 T i / n i 0 T e ) 1/2 is the dust acoustic velocity. Using the expression 29, the wave phase velocity v p / k can be estimated81 if one knows the plasma and dust parameters. In the experiment of Chu et al.,23 one has R 1 m, n d 0 m d 1 g/cm3 and thus m d 4 10 12 g, T e 2.6 eV, T i 0.26 eV, and T d 10 2 T e . The charge on each dust grain is Z d 8100. The average intergrain distance is of the order of 300 m, from which n d 0 3.7 104 cm 3 . The plasma density is roughly n e 0 5 109 cm 3 . For this set of parameters, Eq. 29 depicts v p / k 7 cm/s, which is to be compared with the v p f w w 12 Hz 0.5 cm 6 cm/s observed in the experiment. Furthermore, in a we have CD plasma with n e0T i n i0T e , Z d ( T i / m d ) 1/2( n d 0 / n i 0 ) 1/2. For dust grains of 5 m size, m d 10 12 kg, Z d 4 104 , and n d 0 / n i 0 10 8 , we obtain C D 8 cm/s, which is in good agreement with the observations24 which report v p 9 cm/s. Hence, for w 0.6 cm we obtain the DAW frequency f w 15 Hz. The spatial attenuation rate of the DAWs in a collisional dusty plasma is determined from26 k 2 1

where C S ( n i 0 / n e 0 ) 1/2( T e / m i ) 1/2 is the dust-modied ionacoustic speed. We note in Eq. 33 that the DIA wave phase velocity, / k , increases with the relative dust concentration, because in dusty plasmas, we have n i 0 n e 0 . Barkan et al.30 performed an experiment in a dusty plasma device to investigate the propagation and damping of dust ion-acoustic waves. They found that the phase velocity of the latter increases in accordance with Eq. 33. As a consequence of the phase velocity increase, the ion Landau damping rate is signicantly reduced. The observed DIAW frequencies were in the range 35 kHz depending on the value of Z d n d 0 / n i 0 ). In a plasma with positively charged dust grains and electrons, the shielding of the dust grains comes from the electrons. Here, the frequency of low-phase velocity in comparison with the electron thermal velocity DAWs is45 2 1/2 k De pd /(1 k 2 De ) . Inclusion of the dust charge uctuation somewhat modies the shielding length. Collective modes in a strongly coupled dusty plasma were theoretically investigated by several authors.5457 Rosenberg and Kalman55 investigated the effect of strong dust coupling on the DAWs by supposing that charged dust grains interact with each other via a screened Coulomb Yukawa potential

D r

Q2 exp r / D , r

34

with the exponential factor taking into account the screening of the dust charge by the plasma electrons and ions which are weakly correlated. The quasilocalized charge QLC approximation was then used to derive the following dispersion relation for DAWs:55

id

2 2 d pd i d

30

pd

where k k r ik i , k r ( k i ) is the real imaginary part of the wave number, and is real. Next, we consider the dust ion-acoustic waves DIAWs which occur in the frequency domain k v td , k v ti , pd k v te , pi , where pi is the ion plasma frequency. On these time scales, the electrons follow the Boltzmann distribution, while the ions dust grains are inertial immobile. Thus, e is given by Eq. 24, while the ion susceptibility reads31

k 2d 2 k 2d 2 2

D k,

1/2

35

2 pi i i 3 k 2v 2 ti

31

where d / D is a measure of dust charge screening by the plasma, and the term arising from strong coupling is given approximately by D ( k 0) f s k 2 d 2 with f s (4/45) 0.9 2 0.05 2 when 1 and d Z 2 d e / dT d 1. In the regime kd i.e., k d 1), Eq. 35 gives kC D (1 f s 2 ) 1/2. The latter shows that the effect of strong dust coupling is to reduce the DAW phase velocity, since f s 0. The decrease of the phase speed as increases may be related to an increase in the compressibility of the dust uid as the range of the intergrain potential decreases. In the regime kd i.e., k D 1), we obtain from Eq. 35 pd (1 2 1/2 ) , which shows that the effective dust plasma f s 2k 2 D

Downloaded 04 Jul 2006 to 210.212.158.131. Redistribution subject to AIP license or copyright, see http://pop.aip.org/pop/copyright.jsp

Phys. Plasmas, Vol. 8, No. 5, May 2001

A survey of dusty plasma physics

1797

frequency is reduced due to a decrease of the effective dust charge with stronger screening. The DAW dispersion relation of Murillo56 in the strong coupling limit is 2 1/2 kC D /(1 k 2 D ) (1 k 2 d 2 /16) 1/2, which does not depend explicitly on . On the other hand, Kaw and Sen57 employed a generalized hydrodynamic GH description, which incorporates the nonlocal viscoelasticity with memory effects arising from the strong correlation among dust particles, and obtained the dust susceptibility

2x n t
2

xn Fc , t md

39

where d is the dust-neutral collision frequency ( 2 2 n g R 2 c g with the Epstein drag law; n g is the neutral gas density, R m mean free path of the gas molecules, and c g is the thermal velocity of the gas molecule and21 F c
2 Z2 de

d3

2 2 2 exp x n 1 2 x n x n 1 ,

2 2 d dk 2v 2 td i k * / 1 i m where d is the adiabatic index, d 1 U ( ) is the compressibility, U ( ) E c / n d 0 T d is the so-called excess internal energy, E c is the correlation energy, (4 s /3) * b / m d n d 0 and m (4 s /3) b / n d 0 T d (1 d d ) 4 U ( ) , and s , b are the coefcients for the shear and bulk viscosity, respectively. For longitudinal low-frequency waves ( k v te , k v ti ), the linear dielectric responses of weakly coupled electrons and ions which obey the Boltzmann law are given by Eqs. 24 and 25. Thus, from 1 e i d 0, we obtain for m 1

2 pd

40

36

describes the force acting on n th particle due to its interaction between neighboring particles in the presence of the ckel or Yukawa interaction potential D ( x ) DebyeHu ( Q / x )exp(x/D). Following the standard approach65 for longitudinal waves on an innite linear chain, we obtain from Eq. 39 the dispersion relation for the DLWs as
2 2 i d DL ,

41

where
2 DL

2 pd n d0d 3

2 k 2 d dv 2 td

2 2 pd D 2 1 k 2 D

2 2 2 exp sin2 kd /2 ,

42

i k2 ,

37

which depicts the damping of the modied DAWs. On the other hand, for m 1, one obtains

2 k 2C D 2 1 k 2 D

1 d d

2 d
2 D

2 1 1 k 2 D

4 U 15

is the squared dust lattice frequency.21 Dust lattice waves in dust plasma crystals have been observed experimentally33,34 in rf discharges. In the latter f DL DL/2 is typically less than 50 Hz, which is larger than d 4 Hz. The knowledge of the dust lattice frequency is useful in deducing the screening of the particles in the rf sheath.

, 38
V. INSTABILITIES IN DUSTY PLASMAS

where d v td / pd is the dust Debye radius. Kaw and Sen57 have also shown that their GH description also admits a low-frequency shear mode which is 1 , or a either a damped wave i k 2 for m * 1/2 1 . The propagating wave k ( d E c / n d 0 m d ) for m latter is analogous to elastic wave propagation in solids with the correlation energy E c playing the role of the elastic modulus. Transverse waves have been experimentally observed82 in a two-dimensional screened Coulomb crystal. They were excited by applying a chopped laser beam to a 2D dusty plasma. Measurements of the dispersion relation reveal an acoustic, i.e., nondispersive, character over the entire range of measured wave numbers, 0.2 kd / 0.7. Recent works83 have further focused on wave dispersion relations in strongly coupled plasmas. Next, we discuss the properties of DLWs in the strongly coupled plasma crystal in which the particulates interact by means of their mutual shielded Coulomb repulsion. Under these conditions wave type motions of the dust grains relative to each other are possible, and they have been observed in several experiments.3234 In linear particle arrangements as well as in monolayer plasma crystals these waves have been identied as dust lattice waves under strong coupling conditions. The dispersion relation of the DLW is obtained from the equation of motion for a dust particle

In the past, Bharuthram et al.84 and Rosenberg85 discussed the possibility of dusty plasma wave excitation in the presence of equilibrium ion drifts in a uniform dusty plasma. Specically, they theoretically predicted the excitation of the dust acoustic and dust ion-acoustic waves due to the two stream and kinetic instabilities. The appropriate dispersion relation for the DAWs in the kinetic regime is85,86 1 1
2 k 2 D

1
2 k 2 Di

1/2

ku i 0 2 pd 2 0 , k v ti

43

where u i 0 eE 0 / m i i is the streaming ion drift velocity in the presence of a constant electric eld E 0 . Equation 43 admits an oscillatory instability of the DAW when u i 0 2 1/2 r i i / k , where r kC D /(1 k 2 D ) and the growth rate is

1/2

3 r

1
2 k 2 Di

2 pd

u i0 . v ti

44

For n i 0 / n e 0 1 and T e T i , the expression Eq. 44 for the growth rate ( r ) of the ion streaming driven DAWs ( r /2 15 Hz, r / k 9 cm/s, and 2 / k 0.6 cm is consistent with observations24,29 which reports u 0 eE 0 / m i in 2 105 cm/s for E 0 1 V/cm.

Downloaded 04 Jul 2006 to 210.212.158.131. Redistribution subject to AIP license or copyright, see http://pop.aip.org/pop/copyright.jsp

1798

Phys. Plasmas, Vol. 8, No. 5, May 2001

P. K. Shukla

On the other hand, the dispersion relation in the ion-dust two-stream regime is87 1

u 1 x 2 R 1 , d t 1 Q 1 Q 0

50

2 pi
A s ku i 0 2

2 i pd 1i 0 , ku i 0 A s 2

45

where A s 1 ( k De ) 2 . In the absence of ion-dust collisions, Eq. 45 for ku i 0 gives r i 1/3 ( pi 2 pd ) / A s , with a maximum growth rate at ku i 0 pi / A s . Furthermore, in a collisional dusty plasma with ku i 0 pi / A s , i , Eq. 45 has the approximate solution87

pd

1 i pi 2 i

1/2

1
3/4 As

2 2 (D ) is the square of the Dewhere d t t u 0 x and k D bye wave number. We notice that Eqs. 48 and 50 contain / m d ) Q 1 and Q 0 u 1 x , which are associated with the terms ( 0 perturbed electrostatic forces involving the unperturbed sheath electric eld and the convection of the equilibrium dust charge gradient, respectively. These two forces are responsible for the novel absolute instabilities, as discussed below. The local dispersion relation can be obtained from Eqs. 47 to 50 by supposing that the perturbed quantities are 2 1 we have89 proportional to exp(ikyit). For k 2 / k D 2 D

46 1

which admits a dissipative instability. However, in a nonuniform dusty plasma sheath, there a dust charge gradient appears.88 Shukla89 has shown that free energy stored in the latter can be coupled to the DAWs. Consequently, the dusty plasma sheath becomes unstable.88 In order to understand the physics of an unstable dusty plasma sheath, we rst consider its equilibrium properties u 0u 0 which are governed by n d 0 u 0 constant, g x 0, u 0 Q 0 I e 0 I i 0 I 0 , and e ( n i 0 ( Q 0 / m d) 0 n e 0 ) Q 0 n d 0 0, where u 0 is the component of the equi u 0 / x , librium dust uid velocity along the x axis, u 0 Q 0 ( g / R ) is the unperturbed dust charge, g is the un 0 / x E 0 x is the unperperturbed grain potential, 0 turbed sheath electric eld, m d g x is the x component of the gravity force, and m d is the dust mass. The unperturbed OLM currents are I e 0 R 2 en e ( 0 ) 1/2 and I i 0 R 2 en i ( 0 ) (8/ ) v te exp(eQ0 /RTe) 1/2 (8/ ) v ti (1 eQ 0 / RT i ). It follows that in the absence of the equilibrium dust uid velocity, we have Q 0 E 0 x m d g x and I e 0 I i 0 0. The latter determines the equilibrium charge on the dust grain surface, while the former dictates that the balance between the sheath electric and gravity forces is responsible for the levitation of the dust grains. On the other hand, in the presence of a uniform dust ow, there appears a dust charge gradient Q 0 I 0 / u 0 , which can be ex2 1/2 pressed as Q 0 R e (8/ u 2 0 ) N e v te exp(eQ0 /RTe) Nivti(1eQ0 /RTi), where N e n e 0 exp(e0 /Te) and N i n i 0 exp(e0 /Ti). In order to study the instability of our equilibrium state, as described above, we let the number density n d n d 0 n 1 , u1 , the potential the dust uid velocity ud u 0 x 0 ( x ) 1 , and the dust charge q d Q 0 ( x ) Q 1 , where n d 0 and u 0 are uniform, and n 1 , u1 , 1 , and Q 1 are small perturbations of their equilibrium values. The relevant equations for the perturbed quantities associated with the DAW in a nonuniform dusty plasma are89 d t n 1 n d 0 u1 0, d t u1 Q 0 / m d 1 x 0 / m d Q 1 0,
2 1 4 Q 0n 1 n d0Q 1 , 2 1 k D

2 D

f d 2 0, 1 i

51

where D k pd / k D kC D is the dust acoustic frequency,16 2 2 / Q 0 , f d 4 n d0R / k D D Q 0k Q , and 0/ m d , k Q Q 0 2 D / . Equation 51 is a cubic polynomial in , which can be analyzed numerically. However, some useful analytical results, which exhibit novel instabilities, can be derived from Eq. 51 in several limiting cases. We consider that the wave frequency is much smaller than the dust charge relaxation frequency 1 , which is typically the case in low temperature laboratory dusty plasma discharges. Here, Eq. 51 takes the form
2 2 2 / 2 D 1 i D f d 2 0. 2 D

52

Several useful results follow from Eq. 52. First, for D 0 which ensures that there are no dc electric eld and the dust charge gradient, we have from Eq. 52 the modied DAW frequency D /(1 f d 2 / 1 ) 0 . Second, for D i , where D , D , Eq. 52 gives the growth rate

2 2 2 D f d 2 D D 2 4 2 2 1 D D

53

47 48 49

and

2 0. The latter is fullled of a novel DAW instability for D 0. Clearly, the dc electric eld and the dust if E 0 x Q 0 charge gradient must oppose each other for the dusty plasma to become unstable. As an illustration, we mention that in laboratory experiments,90 we typically have n e 0 n i 0 108 cm 3 , T e 10 T i 1 eV, R 1 10 m, D 102 103 m. Accordingly, for f d 2 / 1 1, 1 103 s 1 , D 60 s 1 , and D 10 s 1 the growth time, deduced from Eq. 52, is a fraction of a second. This is consistent with observations.88 Finally, we discuss the instability of the dust acoustic waves in a dusty plasma which contains nonspherical elongated rotating dust grains.60 Here, the dipole moments of the dust grains are nonzero. The motion of a charged dust grain in the electromagnetic elds ( E c 1 t A and B A, where and A are the scalar and vector potentials, respectively, and c is the speed of light in vacuum is described by the Lagrangian

Downloaded 04 Jul 2006 to 210.212.158.131. Redistribution subject to AIP license or copyright, see http://pop.aip.org/pop/copyright.jsp

Phys. Plasmas, Vol. 8, No. 5, May 2001

A survey of dusty plasma physics

1799

m d v2 1 q I v A r, t q m B 2 2 c 1 d 2

v q i ri E B , c

Letting 0 i in Eq. 56, where 0 , we obtain the 2 2 2 2 1/2 r / k pd ) 0 growth rate for DA (1 k 54

3 2

1/2 4/3

2 2 k r

1/3

k2 2 pd

0 .

59

where m d i m i and q i q i are the total mass and the charge of the grain, I i m i ( ri ) 2 ( ri ) ( ri ) is the tensor of the moment of inertia, d i q i ri is the dipole moment of the elongated grain, and m (1/2c ) i q i ( r Ui )( Ui ri is the rotational velocity of the grain is the magnetic moment of the grain. Furthermore, m i and q i are the mass and the charge of the ith part of the grain, respectively, ri r ri and ui v ri are its coordinate and the velocity ( v and r are the velocity and position of the center of mass, ri is the coordinate of the i th part of the grain relative to the center of mass, and is the angular velocity of the dust grain. We note that Eq. 54 has been derived by assuming that the scale of inhomogeneity of the electromagnetic eld is much larger than the grain size R, as well as by using the relation d t d d. Mahmoodi et al.60 have derived the kinetic equation for the dust grain distribution in the presence of the Lagrangian Eq. 54. Subsequently, the perturbed dust distribution function is obtained for the case where the wave phase velocity exceeds the thermal velocity of the dust particles. By choosing the unperturbed distribution function of the form f d 0 n d 0 2 m d T d 3/2 2 IT d 1/2 exp

Equation 59 exhibits that the growth rate of the instability 2/3 0 . Thus, energy of the dust rotation is proportional to r can ow into dust acoustic oscillations, driving them at nonthermal level.
VI. WAKEFIELD

ckel and far-eld potentials,75 Besides the DebyeHu there also appears an oscillatory wake potential74,75 in a dusty plasma. The wakeeld, which may be responsible for the attraction of charged dust grains of like polarity, arises due to the resonance interaction of a test dust charge which moves with a velocity close to the modied ion acoustic or dust acoustic speeds. The physics of the charged dust grain attraction is similar to the electron attraction in superconductors in which Cooper pairs are formed due to collective interaction involving phonons. In dusty plasmas, the latter are replaced by the DIAWs and DAWs. Negatively charged dust grains feel an attractive force in the negative part of the oscillatory potential,74,75 where the positive ions are focused. The wake potential of a test charge in the presence of the DAW in an unmagnetized plasma is75

w 0, t , t

p2 p p 0 2 , 2 m dT d 2 IT d

qt cos t / L , t

60

55

where T d is the dust temperature, p 0 I 0 , and 0 is the preferred angular frequency of the rotating dust grains. Mahmoodi et al.60 derived the dielectric tensor for the dusty plasma following the standard method. For the longitudinal waves ( kc ) the modied dispersion relation for the DAW reads60 1 1
2 k 2 D

2 pd 2

2 k

k2 0

2 r

2 r

02

where t z v t t , q t is the charge of the test particle, L 2 1/2 D ( v t V 0) 2 C D / C D is the lattice spacing, v t is the test particle velocity, V 0 is the equilibrium ion streaming velocity, and z are the radial and axial coordinates in a cylindrical geometry. For v t V 0 30 cm/s, D 300 m, and C D 6 cm/s, we nd that L 1 mm, which is in agreement with observations.23 The concept of the wakeeld and subsequent attraction of negatively charged dust grains in a linear chain has been veried both by computer simulations90,91 and in dusty plasma experiments.92,93
VII. NONLINEAR STRUCTURES

0,

56

where r (4 d 2 n d 0 /4I ) 1/2. From Eq. 56 it follows that the dust grain rotation gives a contribution only for waves 2 0. For 0 0 that contribution is expressed in the with k change of the dust acoustic frequency

DA 1

2 2 k r

k2 2 pd

1/2

57

2 1/2 ) . However, in the preswhere DA k D pd /(1 k 2 D ence of the dust grain rotation, Eq. 56 admits complex solutions for any rotation frequency 0 , satisfying the condition

0 DA 1

2 2 k r

1/3 3/2

k2 2 pd

58

In this section, we discuss possible nonlinear structures in dusty plasmas. Specically, we focus on dust ion-acoustic DIA shocks and DIA holes. The latter are due to the trapped vortex-like ion distribution. Dust ion-acoustic shocks in an unmagnetized dusty plasma may arise when there is a balance between the nonlinearity associated with the harmonic generation and the kinematic viscosity introduced by the dust-ion drag. The formation of shock structures in dusty plasmas has been observed by Nakamura et al.31 and Luo et al.67 Luo et al.67 have presented an experimental investigation of the effect of negatively charged dust grains on the ion acoustic shock formation in a Q machine. Luo et al.67 observed that ion acoustic compressional pulses steepened as they traveled through a dusty plasma if the percentage of the negative charges in the plasma on the dust grains was 75%. On the other hand,

Downloaded 04 Jul 2006 to 210.212.158.131. Redistribution subject to AIP license or copyright, see http://pop.aip.org/pop/copyright.jsp

1800

Phys. Plasmas, Vol. 8, No. 5, May 2001

P. K. Shukla

Nakamura et al.31 found that in the linear regime, the phase velocity of the DIA waves increases and the wave suffers heavy damping when the dust number density in an electronion plasma is considerably increased. Furthermore, Nakamura et al31 found that an oscillatory ion-acoustic shock wave in a usual argon plasma transforms into a monotonic shock front when it travels through the dusty plasma column. The formation of the shock structure is due to a balance between the harmonic nonlinearity and a kinematic viscosity that is produced by ion-dust collisions. The dynamics of the shock structure associated with DIA waves is modelled by the Kortwegde Vries KdV-Burgers equation68
2 0 3 2 0 0 0, 1/2

The Mach number and the maximum amplitude are related by 1/2 4 b /5F . The above discussions of the formation of DIA shock waves and DIA holes are limited to a weakly coupled dusty plasma. However, a recent laboratory experiment by Samsonov et al.69 has shown the formation of Mach cone shocks, or V-shaped disturbances created by superdust acoustic objects, in two-dimensional Coulomb crystals. Mach cones were double, rst compressive then rarefactive, due to the strongly coupled crystalline state. There are also indications of probe-induced particle circulation Law et al.70 as well as vortex formation7173 in strongly coupled plasma systems.
VIII. COLLECTIVE PROCESSES IN A DUSTY MAGNETOPLASMA

61

where ( 3 ) , n i 0 / n e 0 , T i / T e , 0 id / pi , 2 0 d / pi De , (3 1 12 )/ , e / T e , with id being the ion-dust collision frequency, pi is the ion plasma frequency, and d is the kinematic viscosity of the dusty plasma. We note that Eq. 61 is derived from the hydrodynamic comprising the Boltzmann electron density distribution, the continuity and momentum equations for the ions and Poisson equations, by employing the standard reductive perturbation method and the stretched variables 1/2( x t ) and 3/2t , where is a smallness parameter. Equation 61 without the 0 term admits an oscillatory shock structure. On the other hand, when the kinematic viscosity overwhelms the dispersion, Eq. 61 assumes the form of a Burgers equation for 0 0. The Burgers equation, viz. 3 terms, admits a monoEq. 61 without the 0 and tonic shock prole, as observed by Nakamura et al.31 Nakamura et al.31 also observed ion holes, which are described in terms of the nonlinear DIA involving the trapped vortex-like ion distribution. The latter might appear as a nonlinear saturated state of the two-stream instability in a dusty plasma. In the presence of trapped ions, the ion number density perturbation for small amplitudes viz. 1) is68 n i n i 0 1 a b 3/2 O 2 , 62 where a 1 2 M W ( M / 2), b (4/3 )(1 M 2 )exp(M 2/2) for 0, and the Dawson integral is dey dw exp(w2). Furthermore, noted by W ( y ) exp(y2)0 e / T i , M is the Mach number the speed of the nonlinear structure/the ion thermal speed v ti ), and the electric potential is assumed to be negative, restricted by 0, where plays the role of the normalized by T i / e ) amplitude. We note that is a parameter which determines the number of trapped ions. A plateau in the resonant region is given by 0, and 0 corresponds to a vortex-like excavated trapped ion distribution. Substituting n e n e 0 (1 2 /2) and Eq. 62 into Poissons equation, we readily obtain68
2 F b 3/2 0,

The foregoing investigations have dealt with charging and numerous collective processes in an unmagnetized dusty plasma. However, laboratory and space plasmas are embed B 0 , where z is the unit ded in an external magnetic eld z vector along the z direction and B 0 is the strength of the magnetic eld. While the charging of dust grains in an external magnetic eld is not fully explored at the present stage, there have been several attempts to examine the plasma wave spectra,94105 wakeelds,106,107 and nonenvelope solitons108 in a magnetized dusty plasma, by ignoring the dust charge uctuation dynamics. The presence of an external magnetic eld can signicantly affect the velocities of the plasma species. For example, when the wave frequency is much smaller than the electron gyrofrequency ce eB 0 / m e c , we can have modied lower-hybrid LH waves ( ci , cd ce ) 94 and electrostatic ioncyclotron EIC waves ( ci ce , k z v te ) 97 whose frequencies are, respectively,

and

pi
1 e 1/2

2 m in e0k z 2 m e n i 0 k

1/2

65

ci

2 2 cs n i 0 k 1 n e0 2 ci

1/2

66

63

where , x / Di , and F a T i / T e . A possible solution of Eq. 63 is a potential hole

25F 2 16b 2

sech4 Fx /4 Di .

64

2 2 2 2 2 where e 2 pe k / ce k k / k z , and k z and k are the direction and parallel to the external magnetic eld B 0 z k z perpendicular components of the wave vector k( z 1/2 k ), c s ( T e / m i ) is the ion acoustic speed, ci eB 0 / m i c , and cd Z d eB 0 / m d c . Since n e 0 n i 0 in dusty plasmas, the LH wave frequency is somewhat reduced. On the other hand, we see that the phase velocity of the usual EIC waves is increased when a dust component is added in an electron-ion plasma. The LH waves can be excited in the presence of streaming with the velocity V d 0 , which is produced by the equilibrium electric eld E0 ) dust grains. For ( k z / k ) 2 ( m e / m i ) n i 0 / n e 0 the maximum growth rate is102 1/3 max(3/24/3) LH( Z 2 where LH dn d0m i / n i0m d) , pi / 1 k y V d 0 . We note that the growth rate is valid for ( pd / pi ) 2/3 ci / LH . Using the parameters of Ref. 13, namely V d 0 4 km/s, B 0 0.35 G, pi / ci 500, pe / ce 2.5, we nd that LH 6 7 104 s 1 . The con-

Downloaded 04 Jul 2006 to 210.212.158.131. Redistribution subject to AIP license or copyright, see http://pop.aip.org/pop/copyright.jsp

Phys. Plasmas, Vol. 8, No. 5, May 2001

A survey of dusty plasma physics

1801

dition for the maximum growth rate then gives k y 15 18 m 1 , which is approximately twice the wave number for the Arecibo 430 MHz radar. Since the radar scatters with wave numbers k s 2 k 0 , this LH instability, whose growth time is typically 0.25 ms, is relevant for the observations reported in Ref. 13. Furthermore, we note that streaming electrons can drive EIC waves, which have been observed in a dusty plasma laboratory experiment96 that contains an external magnetic eld ( B 0 0.4 T. The critical electron drift for excitation of the current driven EIC instability is V cr v ti (6 12n e 0 T i / n i 0 T e ), which is lower than the case with no dust component. The measured perpendicular wavelength turned out to be 2.5 cm. There are several ultralow-frequency ( ci ) electrostatic waves in a uniform dusty magnetoplasma. First, the frequency ( cd ci , ce k z / k , k z v te ) of the DAW in a magnetized plasma is100

of the magnetic eld may be important on the dust motion. Merlino et al.24 have suggested that using relatively small dust grains ( 0.01 m, it may be possible to investigate processes involving magnetized dust e.g., for B 0 0.5 T, T d 0.025 eV, the mass density of the grain material 2 103 kg/m3 , we have the dust gyroradius 1 cm and the dust gyrofrequency f cd cd /2 10 Hz; hence video imaging of dust cyclotron waves should be possible as well. In a magnetized dusty plasma the Coloumbian and oscillatory wake potentials are signicantly modied. In the following, we briey discuss the formation of wakeelds in the presence of EIC waves in a uniform dusty magnetoplasma. Here, the wake potential for 0 z ut De , s C s / ci is107

k De
2 2 2 1/2 k s 1 k 2 De

2 2 pi k z

2 pd

2 q t De cos 0 / L 0 , 2 0 s

72

1/2

67

where s De pi / ci . Equation 68 shows that for k z / k pd / pi , we have

k sC S . 2 2 2 1/2 1 k s k 2 De

68

On the other hand, in the opposite limit, viz. k z / k pd / pi , Eq. 68 gives100

k De pd . 2 2 2 1/2 1 k s k 2 De
2

where L 0 2 u /3 ci is the effective attraction length, and u is the z component of the test charge velocity. Equation 68 also holds in the presence of magnetic eld aligned ion streaming with the velocity V i 0 ), in which case u is replaced by V i 0 in 0 and L 0 . We see that the magnetic eld signicantly affects the amplitude of the wake potential as well as the attraction length. Finally, we mention that the theory of wake potentials in a nonuniform dusty magnetoplasma is presented by Salimullah and Shukla.111 They found that both the Coulombian and oscillatory wake potentials are signicantly affected by the presence of a dc magnetic eld and the density inhomogeneity in a dusty plasma.
IX. DISCUSSION AND CONCLUSIONS

69 In this paper, we have described various collective processes in dusty plasmas. Starting from the occurrence of dusty plasmas, we have described the properties of dusty plasmas as well as discussed charging of dust grains and new waves, new instabilities, and some coherent nonlinear structures that are observed in an unmagnetized dusty plasma. The focus was on the dust acoustic, dust ion acoustic, and dust lattice waves. The low-frequency dust acoustic waves are excited due to the streaming and kinetic instabilities in a uniform plasma, while there appears a new class of instability in a nonuniform dusty plasma sheath which contains a dc electric eld and an equilibrium dust charge gradient. The latter is maintained by the equilibrium electron and ion currents that reach the dust grain surface, as well as by a nite equilibrium dust ow that is driven by the dc electric eld. It is found that when the forces associated with the dc electric eld and the dust charge gradient oppose each other, a dusty plasma is subjected to an absolute instability. This instability has been observed by Nunomura et al.88 near the sheath boundary in a dusty plasma. Physically, instabilities arise, because the dc sheath electric eld does work on the dust grains to create dust charge uctuations which cannot keep in phase with the potential of electrostatic disturbances in a nonuniform dusty plasma with a dust charge gradient. Thus, free energy stored in the latter is coupled to unstable electrostatic waves when the dc electric eld in association with the dust charge uctuation produces a charge imbalance in

Second, for pe pi k z k pd ci k , we have the modied convective cells

n e0m i n i0m e

1/2

kz ci , k ci

70

. which propagate almost perpendicular to z In a nonuniform dusty magnetoplasma, the E B0 current remains nite. This leads to the appearance of a new low-frequency in comparison with the ion gyrofrequency ci ) electrostatic wave98

k yd
2 k

ci ,

71

1 which is ute-like. Here, d n i 0 ( Z d n d 0 )/ x . On the other hand, the frequency of ES drift waves is95 C 2 S k y i / ci (1 b s ), where i ln ni0 /x and b s 2 2 2 k C S / ci . Shukla and Varma98 and Shukla et al.95 have discussed the properties of vortices associated with modied convective cells given by Eq. 72 and ES drift waves. The above discussions of wave motions in a magnetized dusty plasma assume immobile dust grains. The inclusion of n,94 dust the dust grain dynamics will give rise to dust-Alfve 51 104 waves. The frequency cyclotron, and dust whistler 2 2 1/2 2 k ) c cd / 2 ( cd ci ) of the latter is k z ( k z pd . One should also comment on the possibility of performing laboratory experiments in dusty plasmas where the effect

Downloaded 04 Jul 2006 to 210.212.158.131. Redistribution subject to AIP license or copyright, see http://pop.aip.org/pop/copyright.jsp

1802

Phys. Plasmas, Vol. 8, No. 5, May 2001

P. K. Shukla

the dusty plasma. Furthermore, we have discussed the dispersion properties of a dusty plasma containing elongated and rotating dust grains. It is found that the rotational energy of the dust grains can be coupled to plasma oscillations when the wave frequency is close to the rotational angular frequency of the dust grain. The instability of the longitudinal waves occurs only in the case when the wave vector lies in the plane of the dust grains. Also pointed out is the importance of collective effects with regard to the generation of wakeelds in dusty plasmas. The concept of wakeelds and associated charged dust attraction in a linear chain has been veried by computer simulations and in laboratory experiments. Finally, we have illuminated the physical properties of DIA shocks and DIA holes that were observed in a laboratory dusty plasma device by Nakamura et al.31 Attempts have been made to correlate several theoretical predictions with laboratory observations and computer simulations in a dusty plasma. Due to the limitation of pages here, we are able to discuss only a selected class of wave phenomena in a weakly coupled dusty magnetoplasma. We have neither discussed ionization nor ion-drag related instabilities109 which seem to be observed experimentally.110 Several laboratory experiments112114 have demonstrated the effects of poloidal ion ows associated with E B0 drift on dust particle behavior in magnetized plasmas. However, we do hope that this paper shall stimulate further studies of collective processes linear waves viz. DLWs, shear waves, etc. and their instabilities as well as numerous nonlinear structures viz. solitons, shocks, voids, and vortices in a strongly coupled dusty magnetoplasma. In closing, we stress that the results of the present investigation should also be useful in understanding the salient features of low-frequency uctuations and associated nonlinear structures viz. shock waves in planetary rings, in interstellar dust-molecular clouds, in supernova explosions, in cosmic particle acceleration, and in cometary plasmas where massive charged particulates are common.
ACKNOWLEDGMENTS

HPRN-CT-2000-00140. The author also acknowledges the support of the International Space Science Institute ISSI, Bern Switzerland for the international team Dust Plasma Interaction in Space.
C. K. Goertz, Rev. Geophys. 27, 271 1989. P. Bliokh, V. Sinitsin, and V. Yaroshenko, Dusty and Self-Gravitational Plasmas in Space Kluwer Academic, Dordrecht, 1995. 3 See, for example, P. K. Shukla, D. A. Mendis, and V. W. Chow, The Physics of Dusty Plasmas World Scientic, Singapore, 1996. 4 D. A. Mendis and M. Rosenberg, IEEE Trans. Plasma Sci. 20, 929 1992; Annu. Rev. Astrophys. 32, 419 1994; D. A. Mendis, in Advances in Dusty Plasmas, edited by P. K. Shukla, D. A. Mendis, and T. Desai World Scientic, Singapore, 1997, pp. 319. 5 F. Verheest, Waves in Dusty Space Plasmas Kluwer Academic, Dordrecht, 2000. 6 n, D. Gallagher et al., Icarus 53, 236 1983; D. D. A. Gurnett, E. Gru Tsintikidis, D. A. Gurnett, W. S. Kurth et al., J. Geophys. Res. 23, 997 1996; D. A. Gurnett, J. A. Ansher, W. S. Kurth et al., Geophys. Res. nyi, Phys. Plasmas 7, 3847 2000. Lett. 24, 3125 1997; M. Hora 7 J. Abrahamson and J. Dinniss, Nature London 403, 519 2000. 8 J. Y. N. Cho, Rev. Geophys. 31, 243 1993. 9 Q. H. Zao and M. C. Kelley, J. Atmos. Sol.-Terr. Phys. 59, 739 1997; M. C. Kelley, C. Alcala, and J. Y. N. Cho, ibid. 60, 359 1998. 10 O. Havnes, L. I. Nsheim, T. W. Hartquist et al., Planet. Space Sci. 44, 1191 1996. 11 O. Havnes, J. Trim, T. Blix et al., J. Geophys. Res. 101, 10839 1996. 12 A. M. Zadorozhny, Geophys. Res. Lett. 27, 493 2000. 13 P. A. Bernhardt, G. Ganguli, M. C. Kelley, and E. Swartz, J. Geophys. Res. 100, 23811 1995. 14 G. Prabhuram and J. A. Goree, Astrophys. J. 441, 830 1995; P. Holland, A. Garscadden, and B. Ganguly, in Physics of Dusty Plasmas, edited by P. K. Shukla, D. A. Mendis, and V. W. Chow World Scientic, Singapore, 1996, pp. 171176. 15 J. Winter, Plasma Phys. Controlled Fusion 40, 1201 1998; Phys. Plasmas 7, 3862 2000; B. N. Kolbasov, A. B. Kukushkin, V. A. Rantsev-Kartinov et al., Phys. Lett. A 269, 363 2000. 16 N. N. Rao, P. K. Shukla, and M. Y. Yu, Planet. Space Sci. 38, 543 1990. 17 P. K. Shukla and V. P. Silin, Phys. Scr. 45, 508 1992. 18 J. H. Chu and L. I, Phys. Rev. Lett. 72, 4009 1994. 19 H. Thomas, G. E. Morll, V. Demmel et al., Phys. Rev. Lett. 73, 652 1994. 20 Y. Hayashi and K. Tachibana, Jpn. J. Appl. Phys., Part 2 33, L804 1994; Y. Hayashi, Phys. Rev. Lett. 83, 4764 1999. 21 F. Melands, Phys. Plasmas 3, 3890 1996. 22 H. Ikezi, Phys. Fluids 29, 1765 1986. 23 J. Chu, J. B. Du, and Lin I, J. Phys. D 27, 296 1994. 24 A. Barkan, R. L. Merlino, and N. DAngelo, Phys. Plasmas 2, 3563 1995; R. L. Merlino, A. Barkan, C. Thompson et al., ibid. 5, 1607 1998. 25 H. R. Prabhakara and V. L. Tanna, Phys. Plasmas 3, 3176 1996. 26 J. B. Pieper and J. Goree, Phys. Rev. Lett. 77, 3137 1996. 27 E. Thomas, Jr. and M. Watson, Phys. Plasmas 6, 4111 1999. 28 C. Thompson, A. Barkan, R. L. Merlino et al., IEEE Trans. Plasma Sci. 27, 146 1999. 29 V. I. Molotkov, A. P. Nefedov, V. M. Torchinskii et al., Sov. Phys. JETP 89, 477 1999; V. E. Fortov, A. G. Kharpak, S. A. Kharpak et al., Phys. Plasmas 7, 1374 2000. 30 A. Barkan, N. DAngelo, and R. L. Merlino, Planet. Space Sci. 44, 239 1996. 31 Y. Nakamura, H. Bailung, and P. K. Shukla, Phys. Rev. Lett. 83, 1602 1999. 32 G. E. Morll, H. M. Thomas, and M. Zuzic, in Advances in Dusty Plasmas, edited by P. K. Shukla, D. A. Mendis, and T. Desai World Scientic, Singapore, 1997, pp. 99142. 33 H. M. Thomas, J. R. Jokipii, G. E. Morll et al., in Strongly Coupled Coulomb Systems, edited by G. Kalman et al. Plenum, New York, 1998, p. 187. 34 A. Homann, A. Melzer, S. Peters et al., Phys. Lett. A 242, 173 1998. 35 H. Mott-Smith and I. Langmuir, Phys. Rev. 28, 727 1926; M. S. Sodha and S. Guha, in Advances in Plasma Physics, edited by A. Simon and W. B. Thompson Interscience, New York, 1971, Vol. 4, pp. 219309. 36 M. Rosenberg and D. A. Mendis, IEEE Trans. Plasma Sci. 23, 177 1995;
1 2

The author dedicates this paper to D. Asoka Mendis on the occasion of his 65th birthday, as he regards Asoka as a great scholar who has signicantly contributed to the advancement of the knowledge in cosmic dusty plasmas. He also thanks Horst Fichtner, A. A. Mamun, and Lennart Steno for reading the manuscript and for valuable advices. Furthermore, the benet of useful collaboration with Tito Mendonc a, Greg Morll, David Resendes, Marlene Rosenberg, and Davy Tskhakaya is gratefully acknowledged. This work was partially supported by the Max-Planck r Extraterrestrische Physik at Garching Germany Institut fu and the Deutsche Forschungsgemeinschaft Bonn, Germany through the Sonderforschungsbereich 191 as well as by the NATO Project entitled Studies of Collective Processes in Dusty Plasmas through Grant No. SAPST.CGL9747335066, and by the European Union through the Human Potential Research and Training Networks for carrying out the project entitled Complex Plasmas: The Science of Laboratory Colloidal Plasmas and Mesospheric Charged Dust Aerosols through Contract No.

Downloaded 04 Jul 2006 to 210.212.158.131. Redistribution subject to AIP license or copyright, see http://pop.aip.org/pop/copyright.jsp

Phys. Plasmas, Vol. 8, No. 5, May 2001 M. Rosenberg, D. A. Mendis, and D. P. Sheehan, ibid. 24, 1422 1996; ibid. 27, 239 1999; M. Rosenberg and P. K. Shukla, ibid. to be published. 37 A. Barkan, N. DAngelo, and R. L. Merlino, Phys. Rev. Lett. 73, 3093 1994. 38 nyi, and S. Robertson, IEEE Trans. Plasma Sci. 22, 97 B. Walch, M. Hora 1994; Phys. Rev. Lett. 75, 838 1995. 39 nyi, and S. Robertson, Phys. Rev. A. A. Sikafoose, J. E. Colwell, M. Hora Lett. 84, 6034 2000. 40 V. E. Fortov, A. P. Nefedov, O. S. Valuna et al., Sov. Phys. JETP 87, 1087 1998. 41 J. P. Boeuf and C. Punset, in Dusty Plasmas, edited by A. Bouchoule Wiley, New York, 1999, Chap. 1, pp. 176. 42 C. Goertz and W. H. Ip, Geophys. Res. Lett. 11, 349 1984. 43 E. C. Whipple, T. G. Northrop, and D. A. Mendis, J. Geophys. Res. 90, 7405 1985. 44 P. K. Shukla, Phys. Plasmas 1, 1362 1994. 45 P. K. Shukla, Phys. Rev. E 61, 7249 2000. 46 L. Steno, P. K. Shukla, and M. Y. Yu, Phys. Plasmas 7, 2731 2000. 47 R. K. Varma, P. K. Shukla, and V. Krishan, Phys. Rev. E 47, 3612 1993; N. N. Rao and P. K. Shukla, Planet. Space Sci. 42, 221 1994. 48 F. Melands, T. K. Aslaksen, and O. Havnes, Planet. Space Sci. 41, 321 1993. 49 M. R. Jana, A. Sen, and P. K. Kaw, Phys. Rev. E 48, 3930 1993. 50 J. X. Ma and P. K. Shukla, Phys. Plasmas 2, 1506 1995. 51 P. K. Shukla, in The Physics of Dusty Plasmas, edited by P. K. Shukla, D. A. Mendis, and V. W. Chow World Scientic, Singapore, 1996, pp. 107121. 52 P. K. Shukla and D. Resendes, Phys. Plasmas 7, 1614 2000. 53 G. Morll, A. V. Ivlev, and J. R. Jokipii, Phys. Rev. Lett. 83, 971 1999; A. V. Ivlev, U. Konopka, and G. Morll, Phys. Rev. E 62, 2739 2000. 54 U. de Angelis and P. K. Shukla, Phys. Lett. A 244, 557 1998; Phys. Scr. 60, 69 1999. 55 M. Rosenberg and G. Kalman, Phys. Rev. E 56, 7166 1997. 56 M. S. Murillo, Phys. Plasmas 5, 3116 1998; D. Winske, M. Rosenberg, and M. S. Murillo, Phys. Rev. E 59, 2263 1999. 57 P. K. Kaw and A. Sen, Phys. Plasmas 5, 3552 1998. 58 O. Havnes, T. K. Aanesen, and F. Melands, J. Geophys. Res. 95, 6581 1990; A. Brattli, O. Havnes, and F. Melands, J. Plasma Phys. 58, 691 1997. 59 U. Mohideen, H. U. Rahman, M. A. Smith, M. Rosenberg, and D. A. Mendis, Phys. Rev. Lett. 81, 349 1998. 60 J. Mahmoodi, P. K. Shukla, N. L. Tsintsadze, and D. D. Tskhakaya, Phys. Rev. Lett. 84, 2626 2000. 61 P. K. Shukla and M. Rosenberg, Phys. Plasmas 6, 1038 1999; A. A. Mamun and P. K. Shukla, ibid. 7, 3499 2000. 62 B. Farokhi, P. K. Shukla, N. L. Tsintsadze, and D. D. Tskhakaya, Phys. Lett. A 264, 318 1999. 63 B. Farokhi, P. K. Shukla, N. L. Tsintsadze, and D. D. Tskhakaya, Phys. Plasmas 7, 814 2000. 64 P. K. Shukla, Phys. Rev. Lett. 84, 5328 2000. 65 C. Kittel, Introduction to Solid State Physics Wiley, New York, 1956, Chap. 11. 66 F. Melands and P. K. Shukla, Planet. Space Sci. 43, 635 1995. 67 Q. Z. Luo, N. DAngelo, and R. L. Merlino, Phys. Plasmas 6, 3455 1999; ibid. 7, 2370 2000. 68 P. K. Shukla, Phys. Plasmas 7, 1044 2000. 69 O. Havnes, F. Li, and F. Melands, J. Vac. Sci. Technol. A 14, 525 1996; D. Samsonov, J. Goree, Z. W. Ma et al., Phys. Rev. Lett. 83, 3649 1999; D. H. Dubin, Phys. Plasmas 7, 3895 2000. 70 D. A. Law, W. H. Steel, B. M. Annaratone et al., Phys. Rev. Lett. 80, 4189 1998. 71 G. E. Morll, H. M. Thomas, U. Konopka et al., Phys. Rev. Lett. 83, 1598 1999. 72 H. Fujiyama, S. C. Yang, Y. Maemura et al., in Double Layers: Potential Formation and Related Nonlinear Phenomena in Plasmas, edited by Sendai Plasma Forum World Scientic, Singapore, 1997, pp. 321327. 73 S. Iizuka, G. Uchida, T. Kamimura, and N. Sato, in Physics of Dusty nyi et al. The American Institute of Physics, Plasmas, edited by M. Hora Woodbury, NY, 1998, pp. 175178; N. Sato, G. Uchida, R. Ozaki et al., ibid., pp. 239246; N. Sato, G. Uchida, R. Ozaki et al., in Frontiers in Dusty Plasmas, edited by Y. Nakamura, T. Yokota, and P. K. Shukla Elsevier Science, Amsterdam, 2000, pp. 329336.
74

A survey of dusty plasma physics

1803

M. Nambu, S. V. Vladimirov, and P. K. Shukla, Phys. Lett. A 203, 225 1995. 75 P. K. Shukla and N. N. Rao, Phys. Plasmas 3, 1760 1996. 76 H. C. Lee, D. Y. Chen, and R. Rosenstein, Phys. Rev. E 56, 4596 1997. 77 G. Lapenta, Phys. Rev. Lett. 75, 4409 1995; G. Lapenta and J. U. Brackbill, Phys. Scr., T 75, 264 1998; G. Lapenta, Phys. Plasmas 6, 1442 1999. 78 D. P. Resendes, J. T. Mendonc a, and P. K. Shukla, Phys. Lett. A 239, 181 1998; D. P. Resendes, Phys. Rev. E 61, 793 2000. 79 K. Avinash and P. K. Shukla, Phys. Lett. A 225, 82 1999; 258, 195 1999. 80 G. E. Morll, H. M. Thomas, U. Konopka et al., Phys. Plasmas 6, 1769 1999. 81 N. DAngelo, J. Phys. D 28, 1009 1995. 82 S. Nunomura, D. Samsonov, and J. Goree, Phys. Rev. Lett. 84, 5141 2000. 83 H. Ohta and S. Hamaguchi, Phys. Rev. Lett. 84, 6026 2000; G. Kalman, M. Rosenberg, and H. E. DeWitt, ibid. 84, 6030 2000; M. S. Murillo, ibid. 85, 2514 2000; A. A. Mamun, P. K. Shukla, and T. Farid, Phys. Plasmas 7, 2329 2000. 84 R. Bharuthram, H. Saleem, and P. K. Shukla, Phys. Scr. 45, 512 1992. 85 M. Rosenberg, Planet. Space Sci. 41, 229 1993. 86 P. K. Shukla, in Frontiers in Dusty Plasmas, edited by Y. Nakamura, Y. Yokota, and P. K. Shukla Elsevier Science, Amsterdam, 2000, pp. 312. 87 M. Rosenberg, J. Vac. Sci. Technol. 14, 631 1996; D. Winske and M. Rosenberg, IEEE Trans. Plasma Sci. 26, 92 1998. 88 S. Nunomura, T. Misawa, N. Ohno et al., Phys. Rev. Lett. 83, 1970 1999. 89 P. K. Shukla, Phys. Lett. A 268, 100 2000. 90 F. Melands and J. Goree, Phys. Rev. E 52, 5312 1995. 91 D. S. Lemons, M. S. Murillo, W. Daughton et al., Phys. Plasmas 7, 2306 2000; D. Winske, W. Daughton, D. S. Lemons et al., ibid. 7, 2320 2000; M. Lampe, G. Joyce, G. Ganguli et al., ibid. 7, 3851 2000; G. Lapenta, Phys. Rev. E 62, 1175 2000. 92 K. Takahashi, T. Oishi, K. Shimomai et al., Phys. Rev. E 58, 7805 1998. 93 A. Melzer, V. A. Schweigert, and A. Piel, Phys. Rev. Lett. 83, 3194 1999. 94 P. K. Shukla, Phys. Scr. 45, 404 1992. 95 P. K. Shukla, M. Y. Yu, and R. Bharuthram, J. Geophys. Res. 96, 21343 1991. 96 A. Barkan, N. DAngelo, and R. L. Merlino, Planet. Space Sci. 43, 905 1995. 97 V. W. Chow and M. Rosenberg, Planet. Space Sci. 44, 465 1996. 98 P. K. Shukla and R. K. Varma, Phys. Fluids B 5, 236 1993; P. K. Shukla, Phys. Plasmas 7, 3882 2000. 99 P. K. Shukla and H. U. Rahman, Phys. Plasmas 3, 430 1996; G. T. Birk, A. Kopp, and P. K. Shukla, ibid. 3, 3564 1996; N. N. Rao, J. Plasma Phys. 53, 317 1995. 100 P. K. Shukla, Astrophys. Space Sci. 264, 235 1999. 101 R. Bharuthram and T. Pather, Planet. Space Sci. 44, 137 1996. 102 M. Rosenberg, M. Salimullah, and R. Bharuthram, Planet. Space Sci. 47, 1517 1999. 103 O. A. Pokhotelov, O. G. Onishchenko, P. K. Shukla, and L. Steno, J. Geophys. Res. 104, 19797 1999. 104 M. Rosenberg and D. A. Mendis, IEEE Trans. Plasma Sci. 20, 929 1992; P. K. Shukla, Phys. Lett. A 252, 340 1999. 105 A. A. Mamun and P. K. Shukla, Phys. Plasmas 7, 3499 2000. 106 P. K. Shukla and M. Salimullah, Phys. Plasmas 3, 3858 1996. 107 M. Salimullah and M. Nambu, J. Phys. Soc. Jpn. 69, 1868 2000. 108 N. Ya. Kotsarenko, S. V. Koshevaya, G. A. Stewart et al., Planet. Space Sci. 46, 429 1998. 109 P. K. Shukla, M. R. Amin, and G. Morll, Phys. Scr., T 75, 253 1998; P. K. Shukla and G. Morll, ibid. 82, 119 1999. 110 J. M. Liu and Lin I, Phys. Rev. E 62, 5571 2000. 111 M. Salimullah and P. K. Shukla, Phys. Plasmas 5, 4205 1998. 112 H. Fujiyama, H. Kawasaki, S. C. Yang et al., Jpn. J. Appl. Phys., Part 1 33, 4216 1994. 113 S. Nunomura, N. Ohno, and S. Takamura, Jpn. J. Appl. Phys., Part 1 36, 877 1997. 114 U. Konopka, D. Samsonov, A. V. Ivlev et al., Phys. Rev. E 61, 1890 2000.

Downloaded 04 Jul 2006 to 210.212.158.131. Redistribution subject to AIP license or copyright, see http://pop.aip.org/pop/copyright.jsp