Temporal and Spatial Evolution of Laser Ablated Carbon Plasma in Ambient Gas and Magnetic Field
Amit Neogi
Department of Physics, Indian Institute of Technology, Kanpur, Kanpur 208016
A detailed investigation of laser ablated carbon plasma expanding in an ambient gas and nonuniform magnetic field using emission spectroscopy, fast photography and Langmuir probe is presented. The nonuniform magnetic field and ambient gas control the plasma behavior in very many ways which may be important for numerous technological applications and studying astrophysical processes
The carbon plasma was produced by focusing an Nd:YAG laser (DCR-4G, Spectra Physics) pulse (E = 800 mJ/pulse , l = 1.06 m m, pulse width = 8 ns) by a spherical lens on to the surface of a rotating graphite target kept inside the vacuum chamber. The chamber can be evacuated to a pressure better than 10-4 Torr. It has a provision for introduction of gas into it in a controlled manner. The plasma formed was imaged on to the slit of a monochromator (HRS-2, Jobin Yvon) which can be tuned by a microprocessor scanning system. The output from the monochomator was detected by a photo multiplier tube (PMT) (IP28, Hamamatsu). PMT signal was recorded on the chart recorder and oscilloscope and was finally stored in the personal computer. The two dimensional image of the plasma plume were recorded using intensified charged coupled device (ICCD-576G/2, Princeton). The delay of the opening of the camera shutter with respect to ablating pulse can be varied in steps of 5 nsec. The detector consists of a microchannel plate (MCP) with a spectral response in the region 200-820 nm and 384´ 576 CCD array. The microchannel plate along with the CCD makes ICCD a highly sensitive device. The system essentially is an opto-electronic device where the images are stored electronically and analysed by the image processing software. A langmuir probe kept at a distance of 4 mm from the target surface at an angle of 45o monitored the evolution of the plasma.
An one dimensional Lagrangian Code MEDUSA which simulates the temporal and spatial evolution of laser ablated plasma in vacuum [1] is modified for the plasma expansion in high pressure ambient atmosphere [2]. Plasma expanding in high pressure ambient gas gives rise to shock waves in the ambient gas. Rankine-Hugoniot relations for the shock wave [3] are used to calculate the boundary conditions of the plasma front. Due to the shear stress the plasma plume experiences viscous drag which decelerates the flow. Prandtl’s mixing length theory [4] is used to calculate the total shear stress or the viscous drag experienced by the plasma plume. The external gas diffuses into the plasma plume when the pressure outside the plasma front is almost equal to the pressure inside the plume which gives rise to a diffusive force opposing the motion of the plasma. This diffusive force is proportional to the ratio of internal to external pressure and density of the gas. Finally, the Navier-Stokes equation is modified by incorporating the viscous drag term and diffusive force term. The modified MEDUSA (with the changed boundary conditions and modified Navier-Stokes equation) was used to simulate the carbon plasma expanding in high pressure (300 Torr) ambient gas (He, Ar, Ne and Xe). The simulated temporal variation of the plasma front for all the four gases matches well with the experimental observations. The plasma parameters like velocity, temperature, pressure and electron density of the plasma front can be simulated for any gas at any higher pressure with the help of modified code.
Click here for Figure 1
The temporal profiles of CI transition 2p3 3D0 - 8f F(5/2) at 399.7 nm, CII transition 3d 2D - 4f 2F0 at 426.7 nm, CIII transition 3s 3S - 3p 3P0 at 465.0 nm and CIV transition at 3s 2S - 3p 2P0 at 580.1 nm were recorded at different distances from the target surface at various pressures of air (0.1, 5 and 100 mTorr). The intensity and nature of the temporal profile is studied for different species and for different pressures [5]. The velocity of all these species was estimated from the delay of the peak of the signal with respect to the ablating pulse. In vacuum, it is observed that the species with higher charge have higher velocities (CII ~ 1.24´ 106 cm/sec, CIII ~ 1.76´ 106 cm/sec, CIV ~ 1.90´ 106 cm/sec at a distance of 4 mm from the target). The intensity of the species is observed to increase in presence of ambient gas. The ICCD photographs at 5 mTorr at various delay times show stratification of plasma plume into two components, fast and slow at later time as shown in Fig.l(b) [5]. The velocity of the two components is estimated to be 1.0´ 106 cm/sec and 7.5 ´ 105 cm/sec and hence termed as fast and slow components respectively. The ICCD images show that the fast component is bulky at earlier time (~ 800 ns) whereas slow component becomes bulky at later time (~1500 ns). In vacuum the temporal profiles show a single peak structure for all the species. In 5 mTorr and 100 mTorr, double peak structure is observed for CI, CII and CIII transitions whereas no double peak is observed for CIV transition. The appearance of double peak in the temporal profiles is attributed to the stratification of the plasma into slow and fast components as observed in ICCD photographs. At 100 mTorr the dimension of the plume gets reduced due to increase in pressure from all sides so stratification of the plasma plume into two separate components are not observed. The stratification of the plasma plume is attributed to the diffusion of ionized air particles into the plasma plume and collisions with the different charged species of the plasma plume. The collided species give rise to slow component where as uncollided species give rise to fast component resulting in stratification.
The ICCD photographs of the plasma expanding in a nonuniform magnetic field show oscillatory behavior of the plasma front at various delay times [6]. The first contraction of the plasma front is seen at 150 ns and second contraction is observed at 500 ns with respect to the ablating pulse as shown in Fig.l(d). The solution of energy, momentum and Ohm's law equations of the plasma plume expanding in magnetic field predicts periodic bouncing of the radius of the plasma plume [7]. The experimental observations are in accordance with the theoretical predictions though the radius of the first bounce observed experimentally (0.43 cm) is less than that predicted theoretically (1.64 cm). At a later time (~ 1000 ns) the plume is observed to break into two lobes which move toward the poles of the magnet. The ICCD images were also used to estimate the plasma parameters. The electron density in presence of magnetic field is found to be higher than that of vacuum case and shows oscillations because of alternate contraction and expansion of the plasmoid at 150 and 500 ns. The typical values of electron density at 150 ns in magnetic field, 100 mTorr air and vacuum are 1.0´ 1017, 5.7´ 1016 and 2.7´ 1016 cm-3, respectively. The temperature of the plasma plume is estimated assuming the plasma plume to behave as an ensemble of black body radiators. The typical values of temperature at 150 ns in magnetic field, 100 mTorr air and vacuum are 3.19, 3.94 and 2.42 eV respectively. The temperature in case of magnetic field increases due to Ohmic heating caused by the surface current [8]. In presence of ambient gas the temperature rises due to enhanced interparticle collisions with the gas particles.
The time integrated spectra of the plasma plume propagating in magnetic field is recorded at a distance of 6 mm from the target surface keeping the target at three different positions in magnetic field (say P, Q, R) [9]. The magnetic field strength at position P, Q, and R is 1.0, 3.0 and 2.7 kG, respectively. When target is kept at position P, the plasma expands in convex magnetic field with positive field gradient, the position Q is similar to P but gradient is less than that in P whereas in position R the field is concave with negative field gradient. It is found that it is not the magnetic field strength but the curvature and the gradient of the magnetic field which controls greatly the plasma behavior. The plasma expanding in convex magnetic field with positive field gradient (position P) shows a maximum rise in ionic line intensity compared to other positions though the magnetic field is lower. Furthermore, C2 molecular Swan band emission, d3p g - a3p u, with D v = - 1, 0, 1 is observed to dominate at this position [9, 10]. The intensity of C2 Swan band is very weak at other positions of the target. At position R where magnetic field is concave and gradient is negative, the intensity is less than that of no magnetic field case. The increase in line intensity is attributed to compression of magnetic field lines ( magnetic Reynold's number ~ 5) . C2 emission in position P of the target is attributed to the movement of electrons due to electromagnetic interaction and that of the ions due to curvature gradient drift which are in the same direction favoring recombination whereas they are in opposite direction in position R of the target.
The time resolved spectra of carbon plasma were recorded for all the species CI, CII, CIII and CIV [11]. As the plasma expands in magnetic field, the field gradient develops at the plume boundary which gives rise to surface current J. The current J inside the plume is in a direction opposite to that at the front. The current J interacts with the magnetic field and gives rise to J ´ B force which decelerates the bulk of the plasma. But - J ´ B accelerates the boundary [8]. Consequently, we observed three components of the plume viz; fast, intermediate and slow. The ICCD images show that the plasma expanding in nonuniform magnetic field gives rise to two lobes, each of which has three components. The observed temporal profile of the species are fitted with Gaussian functions for further investigations. It is found that the species at the outermost boundary, show a single peak structure at earlier distance from the target surface whereas with increasing distance multiple peaks (6 peaks) are observed. They are because of two fast, two intermediate and two slow components arising from two asymmetric lobes. The multiple peaks and oscillations are attributed due to edge oscillations in magnetic field. The temporal profiles of the species show interesting features
when the plasma expands in combined environment of ambient gas and magnetic field. The multiple peaks vanish and only the double peak structure is observed, similar to the one observed when plasma expands in only gas. Another interesting feature observed is a sudden delay in the second peak for CIII and CII species but not for CI [12]. This delay is not observed when the plasma expands in either only gas or only magnetic field. In order to investigate the delay the target is kept at 6 different positions in magnetic field and it is observed that delay occurs only when the plasma expands in concave magnetic field. As the target is moved further into the magnetic field a distinct triple peak structure is observed just before the onset distance of sudden delay in second peak. This behavior may be due to Rayleigh-Taylor instability which occurs when a plasma expands in concave magnetic field lines. The plasma (heavy fluid) expanding against magnetic field (light fluid) undergoes Rayleigh-Taylor instability when the plasma particles move along the concave magnetic feld and the centrifugal force drives this instability [13]. The Rayleigh-Taylor instability along with the diffusion of air may be responsible for the sudden delay in second peak The growth time of the instability is estimated to be 1000 ns. The effective deceleration and density gradient scale length of the plasma was estimated from the ICCD photographs. Since the delay is not observed for neutral species the possibility of recombination or charge exchange is significantly low.
The fast ablated plasma flow into a stationary ambient medium in presence of static magnetic field as discussed earlier can be used to study many magnetospheric and astrophysical processes [14]. The plasma behavior in the curved magnetic field can be used to study the motion of the plasma in curved magnetic field of the earth like generation of ring current in the equatorial plane [15]. Because of cosmic abundance of carbon in the interstellar space, spectroscopic study of carbon species is of great significance. The knowledge of plasma behavior in different ambient conditions elucidates our understanding regarding thin film deposition, surgery and many technological applications.
ACKNOWLEDGEMENT
The author would like to take the opportunity to thank his thesis supervisor Prof. R. K. Thareja for his guidance and encouragement during the course of this work. The work was partly supported by Department of Atomic Energy, Govt. of India.
REFERENCES
[1] J. P. Christiansen, DETF Ashby and K. V. Roberts, Comp. Phys. Comm. 7, 271 (1974).
[2] A. Neogi and R. K. Thareja, Pramana 85, 63 (1998)
[3] James E. A. John, Gas Dynamics, Second edition, (Allwyn & Bacon, INC , Boston, 1984).
[4] H. Slitching, Boundary layer Theory (Pergamon Press, London, 1955).
[5] A. Neogi, A. rvIishra and R. K. Thareja, J. Appl. Phys. 83, 2831 (1998).
[6] A. Neogi and R. K. Thareja, Phys. of Plasmas, 6, 365 (1999).
[7] D. K. Bhadra, Phys. Fluids, 11, 234 (1968).
[8] L. Dirnberger, P E. Dyer, S. R. Farrar and P. H. Key, Appl. Phys. A 59, 311 (1994).
[9] A. Neogi and R. K. Thareja, Phys. Letts A. (to appear). [10] Fumino Kokai, Jpn. J. Appl. Phys. 36, 3504 (1997).
[11] A. Neogi and R. K. Thareja, J. Appl. Phys. 85, 1131 (1999)
[12] A. Neogi and R. K. Thareja, Phys. of Plasmas. (to appear).
[13] Francis F. Chen, Introduction to Plasma Physics and Controlled Fusion Second edition, vol. 1 (Plenum Press, New York, 1984).
[14] P. Bernhardt, et. al. Geophys. Res. 92, 5777 (1987).[15] J. A. Bittencourt, Fundamentals of Plasma Physics (Pergamon Press, Oxford, 1986).